Issue Highlights: Editorials: Original Articles: Arrhythmia

Transcript

Volume 112, Issue 13; September 27, 2005
Issue Highlights:
Issue Highlights
Circulation 2005 112: 1917
Editorials:
Gaining More From Gamma Globulins
Karen Y. Stokes and D. Neil Granger
Circulation 2005 112: 1918 - 1920
Recurrent Pericarditis: Recent Advances and Remaining Questions
Ralph Shabetai
Circulation 2005 112: 1921 - 1923
Another Chromosomal Locus for Mitral Valve Prolapse: Close but No Cigar
Robert Roberts
Circulation 2005 112: 1924 - 1926
Original Articles:
Arrhythmia/Electrophysiology:
Impaired Impulse Propagation in Scn5a-Knockout Mice: Combined
Contribution of Excitability, Connexin Expression, and Tissue Architecture in
Relation to Aging
Toon A.B. van Veen, Mera Stein, Anne Royer, Khaï Le Quang, Flavien
Charpentier, William H. Colledge, Christopher L.-H. Huang, Ronald Wilders,
Andrew A. Grace, Denis Escande, Jacques M.T. de Bakker, and Harold V.M. van
Rijen
Circulation 2005 112: 1927 – 1935
Functional Roles of Cav1.3( 1D) Calcium Channels in Atria: Insights Gained
From Gene-Targeted Null Mutant Mice
Zhao Zhang, Yuxia He, Dipika Tuteja, Danyan Xu, Valeriy Timofeyev, Qian
Zhang, Kathryn A. Glatter, Yanfang Xu, Hee-Sup Shin, Reginald Low, and
Nipavan Chiamvimonvat
Circulation 2005 112: 1936 - 1944
Effect of Fish Oil on Heart Rate in Humans: A Meta-Analysis of Randomized
Controlled Trials
Dariush Mozaffarian, Anouk Geelen, Ingeborg A. Brouwer, Johanna M.
Geleijnse, Peter L. Zock, and Martijn B. Katan
Circulation 2005 112: 1945 - 1952
Congenital Heart Disease:
Sinus Venosus Atrial Septal Defect: Long-Term Postoperative Outcome for 115
Patients
Christine H. Attenhofer Jost, Heidi M. Connolly, Gordon K. Danielson, Kent R.
Bailey, Hartzell V. Schaff, Win-Kuang Shen, Carole A. Warnes, James B.
Seward, Francisco J. Puga, and A. Jamil Tajik
Circulation 2005 112: 1953 - 1958
Coronary Heart Disease:
Rapid Heart Rate Increase at Onset of Exercise Predicts Adverse Cardiac
Events in Patients With Coronary Artery Disease
Colomba Falcone, Maria Paola Buzzi, Catherine Klersy, and Peter J. Schwartz
Circulation 2005 112: 1959 - 1964
Heart Failure:
Viral Persistence in the Myocardium Is Associated With Progressive Cardiac
Dysfunction
Uwe Kühl, Matthias Pauschinger, Bettina Seeberg, Dirk Lassner, Michel
Noutsias, Wolfgang Poller, and Heinz-Peter Schultheiss
Circulation 2005 112: 1965 – 1970
Role of the Protein Kinase C- –Raf-1–MEK-1/2–p44/42 MAPK Signaling
Cascade in the Activation of Signal Transducers and Activators of Transcription 1
and 3 and Induction of Cyclooxygenase-2 After Ischemic Preconditioning
Yu-Ting Xuan, Yiru Guo, Yanqing Zhu, Ou-Li Wang, Gregg Rokosh, Robert O.
Messing, and Roberto Bolli
Circulation 2005 112: 1971 - 1978
Hypertension:
Antihypertensive Effects of Drospirenone With 17ß-Estradiol, a Novel Hormone
Treatment in Postmenopausal Women With Stage 1 Hypertension
William B. White, Bertram Pitt, Richard A. Preston, and Vladimir Hanes
Circulation 2005 112: 1979 - 1984
Imaging:
Is Duplex Surveillance of Value After Leg Vein Bypass Grafting?: Principal
Results of the Vein Graft Surveillance Randomised Trial (VGST)
A.H. Davies, A.J. Hawdon, M.R. Sydes, S.G. Thompson on Behalf of the VGST
Participants
Circulation 2005 112: 1985 - 1991
Interventional Cardiology:
Comparison of Percutaneous Coronary Intervention and Coronary Artery
Bypass Grafting After Acute Myocardial Infarction Complicated by Cardiogenic
Shock: Results From the Should We Emergently Revascularize Occluded
Coronaries for Cardiogenic Shock (SHOCK) Trial
Harvey D. White, Susan F. Assmann, Timothy A. Sanborn, Alice K. Jacobs, John
G. Webb, Lynn A. Sleeper, Cheuk-Kit Wong, James T. Stewart, Philip E.G.
Aylward, Shing-Chiu Wong, and Judith S. Hochman
Circulation 2005 112: 1992 - 2001
Rapamycin, but Not FK-506, Increases Endothelial Tissue Factor Expression:
Implications for Drug-Eluting Stent Design
Jan Steffel, Roberto A. Latini, Alexander Akhmedov, Dorothee Zimmermann,
Pamela Zimmerling, Thomas F. Lüscher, and Felix C. Tanner
Circulation 2005 112: 2002 - 2011
Pericardial Disease:
Colchicine in Addition to Conventional Therapy for Acute Pericarditis: Results
of the COlchicine for acute PEricarditis (COPE) Trial
Massimo Imazio, Marco Bobbio, Enrico Cecchi, Daniela Demarie, Brunella
Demichelis, Franco Pomari, Mauro Moratti, Gianni Gaschino, Massimo
Giammaria, Aldo Ghisio, Riccardo Belli, and Rita Trinchero
Circulation 2005 112: 2012 - 2016
Preventive Cardiology:
Simple Risk Stratification at Admission to Identify Patients With Reduced
Mortality From Primary Angioplasty
Jens Jakob Thune, Dan Eik Hoefsten, Matias Greve Lindholm, Leif Spange
Mortensen, Henning Rud Andersen, Torsten Toftegaard Nielsen, Lars Kober,
Henning Kelbaek for the Danish Multicenter Randomized Study on Fibrinolytic
Therapy Versus Acute Coronary Angioplasty in Acute Myocardial Infarction
(DANAMI)-2 Investigators
Circulation 2005 112: 2017 - 2021
Valvular Heart Disease:
New Locus for Autosomal Dominant Mitral Valve Prolapse on Chromosome 13:
Clinical Insights From Genetic Studies
Francesca Nesta, Maire Leyne, Chaim Yosefy, Charles Simpson, Daisy Dai, Jane
E. Marshall, Judy Hung, Susan A. Slaugenhaupt, and Robert A. Levine
Circulation 2005 112: 2022 - 2030
Vascular Medicine:
Targeting Adhesion Molecules as a Potential Mechanism of Action for
Intravenous Immunoglobulin
Varinder Gill, Christopher Doig, Derrice Knight, Emma Love, and Paul Kubes
Circulation 2005 112: 2031 – 2039
Contemporary Reviews in Cardiovascular Medicine:
Diagnosis and Management of the Cardiac Amyloidoses
Rodney H. Falk
Circulation 2005 112: 2047 - 2060
Special Reports:
Lessons From the Failure and Recall of an Implantable Cardioverter-Defibrillator
Robert G. Hauser and Barry J. Maron
Circulation 2005 112: 2040 - 2042
Report From the Cardiovascular and Renal Drugs Advisory Committee: US Food
and Drug Administration; June 15–16, 2005; Gaithersburg, Md
Steven E. Nissen
Circulation 2005 112: 2043 - 2046
Cardiology Patient Pages:
Cardiac Resynchronization Therapy: A Better and Longer Life for Patients With
Advanced Heart Failure
Srinivas Iyengar and William T. Abraham
Circulation 2005 112: e236 - e237
Images in Cardiovascular Medicine:
Development of a Cardiac Neocavity After Mechanic Double-Valve Replacement:
Evaluation by Cardiac Magnetic Resonance Imaging
Achim Barmeyer, Kai Muellerleile, Gunnar K. Lund, Alexander Stork, Nils
Gosau, Andreas Koops, Sebastian Gehrmann, Thomas Hofmann, Ditmar H.
Koschyk, Claus Nolte-Ernsting, Gerhard Adam, and Thomas Meinertz
Circulation 2005 112: e238 - e239
Periaortic Valve Abscess Presenting as Unstable Angina
Giampaolo Zoffoli and Tiziano Gherli
Circulation 2005 112: e240 - e241
Late Enhancement of a Left Ventricular Cardiac Fibroma Assessed With
Gadolinium-Enhanced Cardiovascular Magnetic Resonance
Francesco De Cobelli, Antonio Esposito, Renata Mellone, Marco Papa, Tiziana
Varisco, Roberto Besana, and Alessandro del Maschio
Circulation 2005 112: e242 - e243
Correspondence:
Letter Regarding Article by Sega et al, "Prognostic Value of Ambulatory and Home
Blood Pressures Compared With Office Blood Pressure in the General Population"
• Response
Tine Willum Hansen, Jørgen Jeppesen, Hans Ibsen, Eamon Dolan, Eoin T.
O’Brien, Jan A. Staessen, Takayoshi Ohkubo, Yutaka Imai, Roberto Sega, Rita
Facchetti, Michele Bombelli, Giancarlo Cesana, Giovanni Corrao, Guido Grassi,
and Giuseppe Mancia
Circulation 2005 112: e244 - e246
AHA Scientific Statements:
Dietary Recommendations for Children and Adolescents: A Guide for
Practitioners: Consensus Statement From the American Heart Association
Endorsed by the American Academy of Pediatrics, Samuel S. Gidding, Barbara
A. Dennison, Leann L. Birch, Stephen R. Daniels, Matthew W. Gilman, Alice H.
Lichtenstein, Karyl Thomas Rattay, Julia Steinberger, Nicolas Stettler, and Linda
Van Horn
Circulation 2005 112: 2061 - 2075
Circulation
JOURNAL
OF THE
AMERICAN HEART ASSOCIATION
Issue Highlights
Vol 112, No 13, September 27, 2005
EFFECT OF FISH OIL ON HEART RATE IN
HUMANS: A META-ANALYSIS OF RANDOMIZED
CONTROLLED TRIALS, by Mozaffarian et al.
COLCHICINE IN ADDITION TO CONVENTIONAL
THERAPY FOR ACUTE PERICARDITIS: RESULTS
OF THE COPE TRIAL, by Imazio et al.
Heart rate is determined by the complex interplay of sympathetic
and parasympathetic stimulation with sinus node automaticity.
Faster heart rates are associated with increasing mortality and
cardiovascular risk, consistent with the observations that increased
sympathetic tone activation is not only a marker of risk but also
has direct adverse effects. ␤-Adrenergic-blocking agents and
physical conditioning slow the heart rate and are beneficial.
Mozaffarian and colleagues show that consumption of fish oil also
slows heart rate. In a meta-analysis of randomized trials including
1678 subjects who ingested fish oil for 4 to 52 weeks, heart rate
slowed by 1.6 beats per minute. Whether this is mediated by an
alteration in cardiac lipid membranes that affects ion channel
function, as has been previously hypothesized, or a reduction in
sympathetic tone through some other mechanism remains to be
established. The findings suggest potential mechanisms for cardiovascular benefits of fish or fish oil consumption. See p 1945.
The treatment of acute pericarditis is based predominantly on observational data. For the problematic group of patients with recurrent
symptoms of pericarditis after the initial episode, some studies have
suggested colchicine to be effective. In this issue of Circulation,
Imazio and colleagues report the results of a prospective, randomized
trial of colchicine in patients with acute pericarditis. Over 18 months
of follow-up, the rate of recurrent symptoms was reduced by 70%.
Moreover, colchicine therapy started during the acute episode resulted
in more rapid resolution of the initial symptoms. These data provide
an evidence base that can inform therapeutic decisions for clinicians in
the treatment of acute pericarditis. See p 2012.
Visit http://www.circ.ahajournals.org:
Cardiology Patient Page
Cardiac Resynchronization Therapy: A Better and Longer
Life for Patients With Advanced Heart Failure. See p e236.
COMPARISON OF PERCUTANEOUS CORONARY
INTERVENTION AND CORONARY ARTERY
BYPASS GRAFTING AFTER ACUTE
MYOCARDIAL INFARCTION COMPLICATED BY
CARDIOGENIC SHOCK: RESULTS FROM THE
SHOULD WE EMERGENTLY REVASCULARIZE
OCCLUDED CORONARIES FOR CARDIOGENIC
SHOCK (SHOCK) TRIAL, by White et al.
Images in Cardiovascular Medicine
Development of a Cardiac Neocavity After Mechanic
Double-Valve Replacement: Evaluation by Cardiac Magnetic Resonance Imaging. See p e238.
Periaortic Valve Abscess Presenting as Unstable Angina.
See p e240.
Despite advances in reperfusion therapies, the incidence of cardiogenic shock has not changed, and it remains the most common
cause of death in patients hospitalized with acute myocardial
infarction. In the SHOCK trial, emergency revascularization, as
compared with medical stabilization, resulted at 1 year in 130 lives
saved per 1000 patients. Furthermore, most survivors had a good
quality of life. To achieve this result, almost 40% of patients had
emergency revascularization surgery. The SHOCK investigation
compared the outcomes according to choice of PCI or CABG,
which was done by site investigators. Patients treated with CABG
had a greater prevalence of diabetes and more advanced coronary
diseases than did patients treated with PCI. However, survival
rates at 12 months were similar. See p 1992.
Late Enhancement of a Left Ventricular Cardiac Fibroma
Assessed With Gadolinium-Enhanced Cardiovascular
Magnetic Resonance. See p e242.
Correspondence
See p e244.
1917
Editorial
Karen Y. Stokes, PhD; D. Neil Granger, PhD
T
he use of intravenous immunoglobulin (IVIg), which
is immunoglobulin G pooled from thousands of
healthy donors, in the treatment of immunodeficient
and autoimmune diseases has grown during the past 2
decades. Although its initial application was largely limited to
replacement therapy in hypogammaglobulinemia, IVIg is
gaining acceptance as therapy for autoimmune thrombocytopenia purpura, and a number of other autoimmune diseases
such as multiple sclerosis.1 Although the exact mechanisms
underlying the protection conferred by IVIg in these immune
disorders remain undefined, several potential molecular and
cellular targets have been proposed. For example, IVIg can
block Fc receptors on macrophages and effector cells to
reduce the phagocytic capacity of these cells. IVIg may also
regulate the immune response by reacting with a number of
membrane receptors on T cells, B cells, and monocytes that
are pertinent to autoreactivity and induction of tolerance to
self.1 Recent work has also revealed a beneficial effect of
IVIg in systemic inflammatory disorders such as sepsis and
asthma. It has been suggested that IVIg may exert its
antiinflammatory effects by attenuating complementmediated attack,2 inducing antiinflammatory cytokines, and
reducing the production of proinflammatory cytokines such
as tumor necrosis factor-␣, interferon-␥ and interleukin-13
(Figure). Many of these mechanistic studies of IVIg effects
on the inflammatory response are based on in vitro models
and in vivo data are lacking.
downregulation of adhesion molecules (P-selectin glycoprotein ligand-1 [PSGL-1], ␤2-integrin) normally expressed on
leukocytes (but not endothelial cells) that mediate the rolling
and (to a lesser extent) firm adhesion steps of leukocyte
recruitment. Another important finding of the Gill study was
that IVIg mediated protection against vascular protein leakage in postischemic feline mesenteric venules. This protection against I/R-induced endothelial barrier dysfunction is
likely to result from the attenuated leukocyte recruitment
rather than a direct action of IVIg on endothelial cell function,
inasmuch as numerous previous studies have demonstrated a
cause-effect relationship between I/R-induced leukocyte-endothelial cell adhesion and increased vascular permeability.5
Nevertheless, a direct action of IVIg on endothelial barrier
function cannot be ruled out because it was recently reported
that IVIg treatment abrogates the gut injury and complement
deposition induced by superior mesenteric artery occlusion in
rats in the absence of any changes in leukocyte infiltration.6
Although these divergent responses to IVIg treatment in the
setting of gut I/R may result from different models and end
points, the benefits noted with IVIg treatment in both studies
highlight the potential utility of IVIg as an inflammationbased therapeutic strategy in ischemic tissue diseases.
Although the findings of Gill and colleagues offer a novel
therapeutic strategy for I/R injury and other inflammationdependent disease processes, the feasibility and utility of IVIg
therapy in the clinical setting remain uncertain and warrant
further consideration. Interference with leukocyte-endothelial
cell adhesion with blocking monoclonal antibodies directed
against P-selectin, PSGL-1, ␤2-integrins, and other adhesion
molecules have proven to be remarkably effective in preventing I/R injury in a number of experimental models.7,8 The
clinical experience with antiadhesion strategies in ischemic
tissue diseases has not yet recapitulated the animal laboratory
experience.9 Duration of the ischemic insult, time of administration of the antibody (before or after the ischemic insult),
adverse activation of cells by the antibodies, and inadequate
design of the clinical trials have been offered as explanations
for the different outcomes in human trials versus animal
models. Whether IVIg can overcome these complications and
limitations of the highly specific and more potent adhesion
molecule– directed antibodies or simply offers more in the
clinical setting because of its multiple sites of action in
the inflammatory cascade (Figure) is unclear. Although the
clinical experience to date predicts a limited potential for
antiadhesion therapy for the treatment of ischemic tissue
diseases, it is difficult to ignore the emerging success of
antiadhesion therapies against other inflammatory conditions
such as multiple sclerosis, asthma, and inflammatory bowel
disease.9
During I/R injury, several cell types are activated, including leukocytes and endothelial cells. Gill et al demonstrated
See p 2031
In a report published in this issue of Circulation, Gill and
coworkers propose a novel target of IVIg action in the
inflammatory cascade (ie, leukocyte-endothelial cell adhesion).4 They report that IVIg prevents leukocyte rolling on
immobilized P- and E-selectin in an in vitro flow chamber
system, and that this effect was mediated on the leukocyte
rather than the selectin substrate. Similar antiadhesive actions
of IVIg were evident on endothelial cell monolayers challenged with histamine and in feline mesenteric postcapillary
venules subjected to ischemia and reperfusion (I/R), with
both models involving P-selectin-dependent leukocyte rolling
and ␤2-integrin-mediated firm adhesion. Their findings with
IVIg were consistent with a mechanism that involves the
The opinions expressed in this article are not necessarily those of the
editors or of the American Heart Association.
From the Department of Molecular and Cellular Physiology, Louisiana
State University Health Sciences Center, Shreveport, La.
Correspondence to D. Neil Granger, PhD, Dept of Molecular and
Cellular Physiology, LSU Health Sciences Center, 1501 E Kings Hwy,
Shreveport, LA 71130-3932. E-mail [email protected]
(Circulation. 2005;112:1918-1920.)
© 2005 American Heart Association, Inc.
Circulation is available at http://www.circulationaha.org
DOI: 10.1161/CIRCULATIONAHA.105.571943
1918
Stokes and Granger
1919
Differential effects of IVIg on several
steps in the inflammatory cascade: Numbers denote potential antiinflammatory
actions of IVIg; letters denote possible
adverse effects of IVIg that may worsen
the inflammatory response. 1, Attenuation of complement deposition on vascular endothelium; 2, reduction of chemokine and cytokine generation; 3,
attenuation of leukocyte recruitment at
least in part via downregulation of adhesion molecules on leukocytes and inhibition of endothelial adhesion molecule
upregulation; 4, abrogation of protein
extravasation/vascular leak; 5, prevention
of NF-␬B activation, which would reduce
gene transcription for many proinflammatory molecules; A, stimulation of superoxide generation from neutrophils; B,
triggering of thrombotic events; C, elevation of leukocyte-platelet aggregate formation and recruitment on vessel wall.
that IVIg does not appear to have a direct effect on endothelial cells; rather, the protective effect appears to be exerted on
the neutrophil.4 This cellular specificity may result from the
study’s focus on the immediate postischemic window, however, because there is evidence that longer exposure to IVIg
induces an antiinflammatory phenotype in cultured endothelial cells. It has been reported that incubation of cytokine-activated endothelial cell monolayers with IVIg prevents the
elevation of mRNA for both cytokines and chemokines.10
Furthermore, IVIg appears capable of preventing the upregulation of endothelial cell adhesion molecules both in vivo and
in vitro.10,11 These effects of IVIg on transcription-dependent
production of cytokines, chemokines, and adhesion molecules may result from an inhibitory action on the transcription
factor NF-␬B.12 Hence, there are several lines of evidence in
the literature that support the possibility that endothelial cells
may also be a major target for the antiinflammatory actions of
IVIg, and the benefit afforded from these actions may be
more evident if the microvascular dysfunction and tissue
injury responses are evaluated later than the immediate
postischemic window.
There is also evidence that the benefits gained from IVIg
treatment for inflammation extend beyond neutrophils, the
major leukocyte population that is recruited into postischemic
mesenteric venules. The same research group has recently
reported the blockade of lymphocyte entry into the brain by
IVIg treatment in a murine model of multiple sclerosis, which
is accompanied by an improved functional recovery.13 Based
on complimentary in vitro data, they proposed that this effect
on lymphocyte recruitment was at least partially caused by
inhibition of ␣4-integrin/vascular cell adhesion molecule-1–
dependent adhesion to endothelial cells. These findings are
consistent with the positive results from a clinical trial in
patients with multiple sclerosis that employed the humanized
monoclonal antibody natalizumab, which saturates ␣4␤1 sites
on T cells.14 The new findings of Gill et al suggest, however,
that downregulation of PSGL-1 may also have played a role
in the lymphocyte recruitment, since rolling of these cells in
venules is primarily dependent on P-selectin–PSGL-1 interactions.15 Their experience with IVIg-mediated inhibition of
lymphocyte recruitment in the brain microcirculation may
also have some bearing on the protective effects of IVIg in
experimental I/R injury. There are several reports that implicate lymphocytes in the pathogenesis of I/R injury. For
example, T lymphocytes appear to mediate the early neutrophil recruitment to sites of I/R injury through the release of
cytokines.16 The well-known ability of IVIg to modulate
immune cell function by binding receptors on T cells
may have important implications in the T cell–mediated
regulation of neutrophil infiltration into the postischemic
microvasculature.
Despite the many purported beneficial actions of IVIg in
modulating the inflammatory cascade, some potentially deleterious actions of IVIg have been described (Figure) that
may limit its application to cardiovascular disease (CVD).
There are reports suggesting that predisposition to cardiovascular risk factors may be enhanced by IVIg therapy. For
example, patients at risk for hypertension may experience
elevated blood pressure as a side effect of IVIg treatment.17
Reactive oxygen species, which have been implicated in the
pathogenesis of several CVDs, including I/R, are produced at
an accelerated rate by neutrophils exposed to IVIg.18
Many CVDs are also associated with platelet activation
and aggregation. There have been several reports describing a
deleterious effect of IVIg on platelet recruitment13 and on
both venous and arterial thrombosis.19 Although the kinetics
of platelet recruitment after vascular injury or inflammation
may differ between tissues, experimental data suggests these
platelets may play an important role in modulating the
leukocyte recruitment and tissue injury responses observed in
different experimental models.20 Inasmuch as P-selectin–
PSGL-1 interactions are known to mediate the platelet–
1920
Circulation
September 27, 2005
leukocyte aggregation and platelet–venular wall interactions
associated with several experimental models of CVD, one
may expect IVIg treatment to inhibit these heterotypic adhesive interactions involving platelets. Although Gill and associates did not monitor platelet adhesion in postischemic
mesenteric venules or the appearance of platelet-leukocyte
aggregates in blood, the same research group has previously
demonstrated that platelet and leukocyte recruitment into the
postischemic cerebral microvasculature is aggravated by IVIg
treatment.13 Because the recruitment of adherent platelets and
leukocytes into the postischemic cerebral microvasculature is
P-selectin-dependent,21 the excessive aggregation and recruitment of platelets and leukocytes induced by IVIg treatment in
this model of ischemic stroke is unexpected in view of the
proposal that IVIg targets PSGL-1; however, it was suggested
that IVIg may bind to Fc receptors that are upregulated on
platelets after stroke, thereby promoting additional leukocyte
and platelet recruitment.13 These potentially detrimental actions of IVIg on platelet function in the setting of CVD raises
the possibility of combining IVIg therapy with antithrombotic
agents such as aspirin, which has been shown to reduce
the incidence of coronary artery lesions in patients with
Kawasaki syndrome.22
The apparent discrepancy between the actions of IVIg in
the microcirculations of the brain13 and gut4 may simply
reflect interorgan differences related to the regulation of
blood cell–vessel interactions. Another notable difference
was the time of IVIg administration, with a preischemic
(before the induction of ischemia) and postischemic (after
2-hour reperfusion) treatment protocol employed in the gut
and brain experiments, respectively. Indeed, the authors
stated that no protection was conferred on the mesenteric
microcirculation when IVIg was administered at the time of
reperfusion.4 Although the need for pretreatment may limit
the therapeutic potential of IVIg for myocardial infarction
and stroke, this strategy may be of benefit during surgical
procedures, in which the time of onset of reperfusion is
known and controlled.
The article by Gill et al in this issue significantly extends
our understanding of the potential mechanisms that underlie
the well-documented benefits of IVIg in a variety of diseases
associated with inflammation. The potency of IVIg as an
antiadhesion agent may well explain its beneficial actions in
diverse clinical conditions and clearly justifies additional
research that is directed toward defining the molecular basis
for the inhibitory action of IVIg on leukocyte-endothelial cell
adhesion. The ability of IVIg to preserve the normal barrier
function of microvascular endothelium has far-reaching implications of therapeutic consequence and is also worthy of
additional study.
References
1. Bayry J, Thirion M, Misra N, Thorenoor N, Delignat S, LacroixDesmazes S, Bellon B, Kaveri S, Kazatchkine MD. Mechanisms of action
of intravenous immunoglobulin in autoimmune and inflammatory
diseases. Neurol Sci. 2003;24:S217–S221.
2. Basta M, Van Goor F, Luccioli S, Billings EM, Vortmeyer AO, Baranyi
L, Szebeni J, Alving CR, Carroll MC, Berkower I, Stojilkovic SS,
Metcalfe DD. F(ab)⬘2-mediated neutralization of C3a and C5a anaphylatoxins: a novel effector function of immunoglobulins. Nat Med. 2003;
9:431– 438.
3. Andersson J, Skansen-Saphir U, Sparrelid E, Andersson U. Intravenous
immune globulin affects cytokine production in T lymphocytes and
monocytes/macrophages. Clin Exp Immunol. 1996;104:10 –20.
4. Gill V, Doig C, Knight D, Love E, Kubes P. Targeting adhesion molecules as a potential mechanism of action for intravenous immunoglobulin. Circulation. 2005;112:2031–2039.
5. Kurose I, Anderson DC, Miyasaka M, Tamatani T, Paulson JC, Todd RF,
Rusche JR, Granger DN. Molecular determinants of reperfusion-induced
leukocyte adhesion and vascular protein leakage. Circ Res. 1994;74:
336 –343.
6. Anderson J, Fleming SD, Rehrig S, Tsokos GC, Basta M, Shea-Donohue
T. Intravenous immunoglobulin attenuates mesenteric ischemiareperfusion injury. Clin Immunol. 2005;114:137–146.
7. Chamoun F, Burne M, O’Donnell M, Rabb H. Pathophysiologic role of
selectins and their ligands in ischemia reperfusion injury. Front Biosci.
2000;5:E103–E109.
8. Ma XL, Tsao PS, Lefer AM. Antibody to CD-18 exerts endothelial and
cardiac protective effects in myocardial ischemia and reperfusion. J Clin
Invest. 1991;88:1237–1243.
9. Yonekawa K, Harlan JM. Targeting leukocyte integrins in human
diseases. J Leukoc Biol. 2005;77:129 –140.
10. Xu C, Poirier B, Van Huyen JP, Lucchiari N, Michel O, Chevalier J,
Kaveri S. Modulation of endothelial cell function by normal polyspecific
human intravenous immunoglobulins: a possible mechanism of action in
vascular diseases. Am J Pathol. 1998;153:1257–1266.
11. Ito Y, Lukita-Atmadja W, Machen NW, Baker GL, McCuskey RS. Effect
of intravenous immunoglobulin G on the TNFalpha-mediated hepatic
microvascular inflammatory response. Shock. 1999;11:291–295.
12. Ichiyama T, Ueno Y, Isumi H, Niimi A, Matsubara T, Furukawa S. An
immunoglobulin agent (IVIG) inhibits NF-kappaB activation in cultured
endothelial cells of coronary arteries in vitro. Inflamm Res. 2004;53:
253–256.
13. Lapointe BM, Herx LM, Gill V, Metz LM, Kubes P. IVIg therapy in brain
inflammation: etiology-dependent differential effects on leucocyte
recruitment. Brain. 2004;127:2649 –2656.
14. Miller DH, Khan OA, Sheremata WA, Blumhardt LD, Rice GP, Libonati
MA, Willmer-Hulme AJ, Dalton CM, Miszkiel KA, O’Connor PW. A
controlled trial of natalizumab for relapsing multiple sclerosis. N Engl
J Med. 2003;348:15–23.
15. Kerfoot SM, Kubes P. Overlapping roles of P-selectin and alpha 4 integrin to
recruit leukocytes to the central nervous system in experimental autoimmune
encephalomyelitis. J Immunol. 2002;169:1000–1006.
16. Horie Y, Wolf R, Chervenak RP, Jennings SR, Granger DN.
T-lymphocytes contribute to hepatic leukostasis and hypoxic stress
induced by gut ischemia-reperfusion. Microcirculation. 1999;6:267–280.
17. Stangel M, Hartung HP, Marx P, Gold R. Side effects of high-dose
intravenous immunoglobulins. Clin Neuropharmacol. 1997;20:385–393.
18. Nemes E, Teichman F, Roos D, Marodi L. Activation of human granulocytes by intravenous immunoglobulin preparations is mediated by
FcgammaRII and FcgammaRIII receptors. Pediatr Res. 2000;47:
357–361.
19. Paran D, Herishanu Y, Elkayam O, Shopin L, Ben-Ami R. Venous and
arterial thrombosis following administration of intravenous immunoglobulins. Blood Coagul Fibrinolysis. 2005;16:313–318.
20. Tailor A, Cooper D, Granger DN. Platelet-vessel wall interactions in the
microcirculation. Microcirculation. 2005;12:275–285.
21. Ishikawa M, Cooper D, Arumugam TV, Zhang JH, Nanda A, Granger
DN. Platelet-leukocyte-endothelial cell interactions after middle cerebral
artery occlusion and reperfusion. J Cereb Blood Flow Metab. 2004;24:
907–915.
22. Abe T, Kawasugi K. Use of intravenous immunoglobulin in various
medical conditions. A Japanese experience. Cancer. 1991;68:1454 –1459.
KEY WORDS: Editorials
molecules 䡲 leukocytes
䡲
endothelium
䡲
ischemia
䡲
cell adhesion
Editorial
Recurrent Pericarditis
Recent Advances and Remaining Questions
Ralph Shabetai, MD
R
not based on large randomized trials; furthermore, when
colchicine was added to the treatment regimen for recurrent
pericarditis, this was done only after a corticosteroid or an
NSAID failed to influence the frequency, severity, and
duration of recurrence.
The COlchicine for acute PEricarditis (COPE) trial published in this issue of Circulation1 is the first large randomized prospective trial of colchicine added to standard treatment of acute pericarditis. Wisely, the authors selected as the
primary end point the ability of colchicine to prevent recurrence and improve the clinical course of every recurrence that
developed during their study. The secondary end point was
the effect of this treatment regimen on the duration of pain
after a first attack of acute pericarditis. The design was not
double-blind placebo-controlled because colchicine is far
from being a new drug, and consequently the study could not
be funded by a pharmaceutical company, again exemplifying
the influence of the pharmaceutical industry on medical
education and publication. The authors did take all necessary
steps to ensure the validity of the results in the absence of
double blinding. Prednisone was reserved for patients in
whom NSAID therapy was either contraindicated or poorly
tolerated.
In the 120 patients studied at 2 centers, the addition of
colchicine reduced the recurrence rate at 18 months from
32.3% to10.7%, a remarkable two thirds reduction. The
secondary end point, persistence of symptoms at 72 hours
after the onset of acute pericarditis, was also significantly
reduced by two thirds. Thus, the addition of colchicine to
standard treatment for acute pericarditis has been placed on
firm footing. As noted by the authors, although it promptly
relieves symptoms, corticosteroid therapy is thought to promote recurrence.2 It is therefore noteworthy that multivariate
analysis of the COPE data confirmed that its prior use is an
independent risk factor for recurrence.
The results of the COPE trial are welcome, coming as they
do in a climate of considerable skepticism regarding the
benefit of colchicine for recurrent pericarditis, fostered by
disappointing experiences in pericarditis with colchicine
noted by many physicians who have included colchicine in
their treatment regimen for recurrent pericarditis. The results
of COPE convincingly correct this anecdotal view of colchicine for enhancing the treatment of recurrent pericarditis.
A common misconception among physicians and fear
among patients is that repeated pericardial inflammation may
lead to constrictive pericarditis or cardiomyopathy but, in the
120 patients in the COPE trial, not a single case of constrictive pericarditis or cardiomyopathy was reported. This confirms that we have solid evidence to back the assurance we
give to patients that, although recurrent pericarditis may
ecurrence is a serious complication of acute pericarditis, characterized by a return of pericardial pain
after recovery from an attack of typical acute pericarditis. Some patients experience only a single recurrence, but
in many less fortunate, pericardial pain returns unexpectedly
at variable intervals during a period that may extend over
many years. The pain and any associated fever and leukocytosis disappear within a day or so of high-dose corticosteroid
administration (eg, 1 to 1.5 mg/kg per day prednisone in most
cases), only to return during tapering to a low dose. Thereafter, the high dose is reinstituted and then maintained for 1
month or 6 weeks, after which prednisone is again slowly
tapered during the next several months. This sequence may
need to be repeated frequently before it becomes possible to
wean the patient from steroidal therapy. This is one of the
reasons recurrent pericarditis is so troublesome to patients
and treating physicians alike.
See p 2012
Until recently, patients in whom recurrences were frequent
and extended for many years often received a massive total
dose of a steroid, usually prednisone or prednisolone, with a
consequently unacceptable incidence of gastric hemorrhage,
aseptic necrosis of the femoral head, and osteoporosis with
spinal compression fracture. Seeking a less toxic treatment,
physicians began to avoid corticosteroids or limited their use
and treated the patients instead with nonsteroidal antiinflammatory drugs (NSAIDs). The change often was not
accomplished easily because patients, and sometimes referring physicians, had been so pleased with the prompt response to high-dose steroid administration. Good progress is
being made in this regard and will continue with further
education of patients about their disease and its treatment.
In the last decade of the 20th century, a number of
investigators published enthusiastic reports of the efficacy of
colchicine as adjuvant treatment of acute pericarditis. Not
surprisingly, because recurrence is the most common major
complication of acute pericarditis, subsequent papers suggested that colchicine should also be used as part of the
treatment regimen for recurrences. Most of these reports were
The opinions expressed in this article are not necessarily those of the
From the University of California, San Diego, and the Veterans
Administration Health Care System, La Jolla, Calif.
Correspondence to Ralph Shabetai, MD, VA Medical Center, Cardiology (111A), 3350 La Jolla Dr, La Jolla, CA 92161. E-mail
[email protected]
(Circulation. 2005;112:1921-1923.)
1921
1922
Circulation
September 27, 2005
sometimes seem to be an endless problem, chronic constrictive pericarditis is a rare sequel, and myocardial damage does
not ensue. Cardiac tamponade is an uncommon complication
that was not seen even once in the COPE trial and would be
promptly recognized in patients being studied for recurrent
pericarditis.
The toxic effects of oral steroid administration are significantly lessened by injecting a nonabsorbable preparation
intrapericardially, as was recommended for pericardial effusion in patients with late-stage renal disease.3 Intrapericardial
administration of triamcinolone is, by the same token, a good
option for recurrent pericarditis and has the added advantage
that the steroid is delivered where it contacts the 2 pericardial
surfaces.4 Using a flexible pericardioscope, the medication
can be delivered into the pericardial space, even in the
absence of effusion.5 Pericardioscopy and the PerDUCER,6
an instrument developed to invade the pericardium when
effusion is not present, are not available in most major
medical centers; although it would be advantageous to establish at least one center in the United States where one of these
techniques or a comparable technique would be used in
clinical practice. At present, for most of us, pericardioscopy is
still an investigational tool.
The cause of acute pericardial disease was investigated in
2 series from Spain, 1 with 231 consecutive patients7 and the
other having 100 consecutive patients,8 with prospective
protocols to determine the cause of pericarditis. The conclusion from these studies was that “diagnostic” pericardial tap
and biopsy seldom yield the cause, whereas paradoxically,
when these procedures were performed for conditions such as
tamponade and suspected purulent infection or neoplastic
disease, the diagnostic yield was much improved. These
important studies have strongly influenced clinical practice
by sharply decreasing the frequency of invasive investigation
and hospitalization for uncomplicated acute pericarditis.
When it appears doubtful that a patient has viral (or idiopathic) pericarditis, or has a complication such as cardiac
tamponade, or fails to respond to standard anti-inflammatory
treatment, hospitalization for treatment such as pericardiocentesis, and comprehensive investigation of causation are mandatory.9 For the COPE trial, Imazio et al selected high-dose
aspirin as the NSAID, as they had in their earlier study of the
management of acute pericarditis in which colchicine was not
included, and found it safe and effective; thus, it is appropriate to use aspirin before moving to an expensive NSAID.
Viral or idiopathic pericarditis is a benign, short-lived condition requiring simple treatment, usually without hospital
admission. When patients present with chest pain but no risk
factors or good evidence for coronary disease, careful clinical
and laboratory evidence of acute pericarditis should be
undertaken before embarking on comprehensive evaluation
for myocardial ischemia or infarction. Markers of myocardial
damage are often slightly elevated, but not to the threshold for
myocardial infarction.10 The mildly elevated markers have no
influence on the outcome of acute pericarditis.
A significant proportion of cases are caused by an autoimmune reaction to an initial episode of pericarditis, itself
frequently caused by a virus; it is a mistake to consider all
recurrence as autoimmune, and definite evidence of autoim-
mune pericarditis is necessary to justify this conclusion.11
Specifically, antisarcolemmal antibodies should be present,
polymerase chain reaction for cardiotropic viruses and other
infectious agents should be negative, and immunoglobulin M
against these agents should not be detectable. In addition,
tissue should be examined after immunocytochemical and
immunohistochemical staining.4,5 For an initial episode of
acute pericarditis and for a first or infrequent recurrence,
because of their high cost and the inevitable invasion of the
pericardium, those studies are difficult to justify, but they
certainly deserve a place for acute pericarditis that fails to
respond to NSAID therapy and frequent recurrence.
Many patients ask why the problem cannot be solved by
simply removing the offending pericardium. In the series of
patients studied for an average of 10 years by Fowler and
Harbin,12 22 were followed for ⱖ5 years and 10 for ⱖ8 years.
The importance of this study is the duration appropriate for a
condition that may persist for years, occasionally even for
decades. Nine had undergone pericardiectomy, but clear
improvement resulted in only 2 patients. This unsatisfactory
outcome differed from an earlier enthusiastic recommendation that recurrent pericarditis should be treated by pericardiectomy. Tuna and Danielson13 reported much better results
that they attributed to virtually complete pericardiectomy.
Worthwhile progress has been made in the treatment of
acute and recurrent pericarditis, but in recurrent pericarditis
many issues require investigation. We need to find reliable
noninvasive methods that will distinguish autoimmune cases
from those caused by reinfection or new infection, and trials
of treatment based on cause. If we can learn how to predict
the outcome of pericardiectomy, then that would be a notable
advance. The riddle of recurrent pain without evident pericarditis remains to be solved, and therefore the place in it for
anti-inflammatory treatment is uncertain. The exact mechanism of the action of colchicines in recurrent pericarditis is in
need of clarification. We lack an animal model of recurrent
pericarditis. Research will include basic and clinical immunology as well as virology and a search for still more effective
drugs.
Successful management requires a lot of patience on the
part of physicians and patients. Patients must be informed
about what is known about the condition and the merits and
problems associated with the various therapeutic options,
including pericardiectomy.
References
1. Imazio M, Bobbio M, Cecchi E, Demarie D, Demichelis B, Pomari F,
Moratti M, Gaschino G, Giammaria M, Ghisio A, Belli R, Trinchero R.
Colchicine in addition to conventional therapy for acute pericarditis:
results of the COPE trial. Circulation. 2005;112:2012–2016.
2. Godeau P, Derrida JP, Bletry O, Herreman G. Pericarditis aiguös recidivantes et cortico-dependance. A propos de 10 observations. Sem Hop
Paris. 1975;51:2393–2400.
3. Buselmeier TJ, Simmons RL, Najarian JS, Mauer SM, Matas AJ,
Kjellstrand CM. Uremic pericardial effusion. Nephron. 1976;16:371–380.
4. Maisch B, Ristic A, Pankuweit S. Intrapericardial treatment of autoreactive pericardial effusion with triamcinolone. The way to avoid side
effects of systemic corticosteroid therapy. Eur Heart J. 2002;23:
15003–15008.
5. Maisch B, Ristic AD, Seferovic PM, Spodick DH. Intrapericardial
treatment of autoreactive myocarditis with triamcinolon. Successful
Shabetai
6.
7.
8.
9.
administration in patients with minimal pericardial effusion. Herz. 2000;
25:781–786.
Maisch B, Ristic AD, Rupp H, Spodick DH. Pericardial access using the
PerDUCER and flexible percutaneous pericardioscopy. Am J Cardiol.
2001;88:1323–1326.
Permanyer-Miralda G, Sagrista-Sauleda J, Soler-Soler J. Primary acute
pericardial disease: a prospective series of 231 consecutive patients.
Am J Cardiol. 1985;56:623.
Zayas R, Anguita M, Torres F, Gimenez D, Bergillos F, Ruiz M, Ciudad
M, Gallardo A, Valles F. Incidence of specific etiology and role of
methods for specific etiologic diagnosis of primary acute pericarditis.
Am J Cardiol. 1995;75:378 –382.
Imazio M, Demichelis B, Parrini I, Giuggia M, Cecchi E, Gaschino G,
Demarie D, Ghisio A, Trinchero R. Day-hospital treatment of acute
10.
11.
12.
13.
Colchicine for Recurrent Pericarditis
1923
pericarditis: a management program for outpatient therapy. J Am Coll
Cardiol. 2004;43:1042–1046.
Imazio M, Demichelis B, Cecchi E, Belli R, Ghisio A, Bobbio M,
Trinchero R. Cardiac troponin I in acute pericarditis. J Am Coll Cardiol.
2003;42:2144 –2148.
Maisch B. Recurrent pericarditis: mysterious or not so mysterious? Eur
Heart J. 2005;26:631– 633.
Fowler NO, Harbin AD III. Recurrent acute pericarditis: follow-up study
of 31 patients. J Am Coll Cardiol. 1986;7:300 –305.
Tuna IC, Danielson GK. Surgical management of pericardial diseases.
Cardiol Clin. 1990;8:683– 696.
䡲
pericarditis
䡲
colchicine
Editorial
Another Chromosomal Locus for Mitral Valve Prolapse
Close but No Cigar
Robert Roberts, MD
T
cantly affected in 97% of individuals. Nevertheless, with a
2.4% prevalence, mitral valve prolapse would be expect to be
present in 7.2 million individuals in the United States and 144
million worldwide. It is also of note that the prevalence is
based primarily on European and North American populations and may not be representative of other ethnic
groups.11,12 Fortunately, the complications of mitral valve
prolapse— heart failure, mitral regurgitation, bacterial endocarditis, thromboembolism, and atrial fibrillation—although
serious, are extremely uncommon and probably affect no
more than 3% of those with mitral valve prolapse.7 It is
perhaps not surprising that mitral valve prolapse is the single
most common cause for surgical repair or replacement of the
mitral valve.7
The dawn of molecular genetics suggested new excitement
in the landscape of this disorder. It has been recognized since
Barlow and Bosman’s description in the 1960s of a family
with this disorder that at least a certain proportion of
individuals with mitral valve prolapse is hereditary.13 In
1999,14 a crack in the armor came when a family with mitral
valve prolapse segregating as an autosomal dominant trait
underwent genetic linkage analysis and a locus was mapped
to chromosome 16p11.2-p12.1. Genetic linkage gave maximum multipoint LOD scores of 5.4 and 5.6, indicating that
this was the responsible locus. This was confirmed by
haplotype analysis showing a chromosomal region of about 5
cm containing the locus (a genetic distance equivalent to 5
million DNA base pairs) was present in all affected individuals. Analysis of this family showed mitral valve prolapse
exhibits age-dependent penetrance and the disease seldom
appears before age 30. The investigators evaluated several
candidate genes but none showed a mutation responsible for
the disease. In 2003, Freed et al15 identified a second locus for
mitral valve prolapse at chromosome 11p15.4. Genetic analysis again confirmed this to be an autosomal dominant
disease with age-dependent penetrance.
In this issue of Circulation, Nesta et al16 from the Levine
laboratory have identified a third chromosomal locus for
mitral valve prolapse on chromosome 13q31.3-q32.1 with a
multipoint LOD score of 3.17. This is a marginal LOD score,
but haplotype analysis does show that a portion of the
chromosome containing the locus is present in all affected
members of the family. This chromosomal segment contains
⬎8 million bases (the postgenome era enables us to refer to
the number of DNA bases rather than a genetic distance) and
thus encloses a region with multiple genes. The disease
exhibits autosomal dominant inheritance with age-dependent
penetrance. Current knowledge indicates that there are at least
16 known genes in the region, several of which would appear
to be good candidates for mitral valve prolapse. The mapping
he mitral valve forms a complex apparatus whose
closing and opening with the heartbeat coordinates the
flow of blood from the left atrium to the left ventricle.
In an average human life span, it does a command performance to the tune of ⬇3 billion heartbeats. During the life
span, the leaflets experience significant wear and tear exhibited by a thickening of the outer silk lining, but performance
in most individuals remains graceful and relatively undeterred. This synchronized dance to the rhythm of the heartbeat, a rhythm which changes in response to emotional,
mental, and physical demands, appears effortless and relentless. Beneath this superficial and necessary display, however,
lurks many opportunities for a misstep. There have been
decades of statistics suggesting that abnormalities in the
mitral valve are common—2% to 16%.1–5 The symptoms,
when present, are vague and include dizziness, palpitations,
syncope, atypical chest pain, and dyspnea. On physical
examination, the presence of a click or murmur often positions the patient for multiple tests, which can be a source of
great emotional concern. Once the observation came that diet
pills were associated with valvular dysfunction, echocardiographic analysis demanded by lawyers from urban billboards
uncovered many patients with mitral valve prolapse and lifted
the diagnosis to a new level of notoriety.6
See p 2022
In the past 10 years, improved technology and community
studies rather than hospital-based studies have ascertained a
prevalence of ⬇2.4%.7,8 This reduced prevalence is in part
the result of better understanding of the 3-dimensional architecture of the mitral valve annulus provided by 3-dimensional
echocardiography. This led to standardized echocardiographic criteria for the diagnosis of mitral valve prolapse.9,10
Classic mitral valve prolapse is diagnosed if upward displacement of the leaflet exceeds 2 mm and maximal thickness is
ⱖ5 mm; nonclassic mitral valve prolapse refers to displacement that exceeds 2 mm but maximal thickness is ⬍5 mm. It
is comforting to know that the mitral valve apparatus is
designed to perpetuate its motion and withstand the wear and
tear for such a long interval without function being signifiThe opinions expressed in this article are not necessarily those of the
From the University of Ottawa Heart Institute, Ottawa, Ontario,
Canada.
Reprint requests to Robert Roberts, MD, President and CEO, University of Ottawa Heart Institute, 40 Ruskin St, Ottawa, ON K1Y 4W7,
Canada. E-mail [email protected]
(Circulation. 2005;112:1924-1926.)
1924
Roberts
Another Chromosomal Locus for Mitral Valve Prolapse
of the location of 3 genes on 3 separate chromosomes bodes
well for chromosomal mapping of genes; however, it is only
the first step in providing us the opportunity to seek the gene
and identify the ultimate defect. After identification of the
defect, usually functional studies are required to elucidate the
pathogenesis of the disease. Given the increased rapidity to
sequence DNA and identify mutations, one can soon expect a
fiesta of genes for mitral valve prolapse. Once the first gene
is identified, it is likely to significantly accelerate the identification of other genes. This expectation is based on the
assumption that the gene belongs to a class of molecules with
a similar function that would be expected to induce the
phenotype of mitral valve prolapse.
It is also of note that several of the individuals in the family
studied inherited the haplotype containing the defective gene
and had minor nonspecific prolapse, which would not satisfy
the established diagnostic criteria for classical or nonclassical
mitral valve prolapse. Does this mean revised criteria are
required? Until the gene and its mutations are identified, it
would be premature because of the issue of having the
defective gene but because of nonpenetrance, it may not be
expressed. If these nonspecific minor mitral valve abnormalities are independent of the mutation, then it should remain in
the benign nonspecific category. If it relates to the mutation,
then a revision of the diagnostic criteria is in order.
The histological hallmark of mitral valve prolapse is
myxomatous degeneration of the leaflet.17 The essence of the
functional defect is redundancy and leaflet elongation, which
leads to a “billowing” of the leaflet with the potential to
prolapse into the atrium and induce mitral regurgitation. An
analysis of the chemical composition of the leaflets should
suggest potential candidates for genetic defects. The normal
mitral valve has 2 leaflets, 1 end attached to the base of the
annulus fibrosis and their free edges to their chordae tendineae. The posterior leaflet is narrower than the anterior and is
often the one most affected in myxomatous mitral valve
prolapse. The leaflets are composed of 4 layers: auricularis,
fibrosa, ventricularis, and spongiosa.17 The auricularis is
composed of collagen and elastic tissue and forms the contact
for the atrial aspect of the leaflet and is continuous with the
endocardium of the left atrium. The fibrosa is the basic
support of the leaflet and is composed of thick collagen. This
layer is continuous with the fibrosis of the annulus and the
chordae tendineae. The ventricularis is a thin layer of collagen and elastic tissue that covers the ventricular aspect and is
continuous with the left ventricular endocardium. The spongiosa or fourth layer is situated between the auricularis and
the fibrosa layers and is formed of delicate myxomatous
connective tissue. One may assume that some defect in the
basic support of the leaflet results from the genetic defect,
which together with normal wear and tear leads to stretching
and elongation of the leaflet and gives rise to the clinical
phenotype. Major components of the normal leaflet are
collagen, elastin fibers, and a high density of proteoglycans.
The primary abnormality in mitral prolapse appears to occur
in the spongiosa layer, which is characterized by deposition
of proteoglycans.18,19 This disposition invades the other
layers, particularly the fibrosa, and disrupts the normal
support of the leaflet, which could enable the hemodynamic
1925
forces to mechanically stretch and derange the leaflet. The
collagen content of the normal mitral valve leaflet is ⬇74%
type I, 24% type 3, and 2% type 5. In mitral valve prolapse,
the collagen is significantly increased, particularly type 3,
which increases up to 53%. A second abnormality consistently observed in mitral valve prolapse is the increase in
proteoglycans, which are thought to play a role in the
assembly of collagen fibrils.
In a study by Rabkin et al,20 previous observations of
excessive collagen degradation, elastin fragmentation, and
proteoglycan accumulation were confirmed. The predominant
resting cell of the leaflet is a fibroblast-like cell that synthesizes collagen, elastin, and proteglycans. In myxomatous
valves, myofibroblasts are present, which in addition to
collagen are known to secrete collagenase (matrix
metalloproteinase-1 [MMP-1] to MMP-13), gelatinase
(MMP-2, MMP-9), cysteine proteases (cathepsin C and M),
and interleukin-1␤, a cytokine that induces secretion of
proteolytic enzymes. These investigators concluded that the
synthesis of collagen is normal, but degradation is increased
with an accumulation of breakdown products that weaken the
fibroskeleton of the leaflets. It was observed in valves
removed from patients with mitral valve prolapse at the time
of surgery that there was increased content of proleoglycans,
primarily chondroitin, dermatan sulfate, and keratan sulfate.
This together with increased accumulation of water gives the
leaflet its myxomatous appearance and also the floppy gelatinous nature so characteristic of the pathology. The genes
encoding for all of these compounds are potential candidates
for mutations leading to this disorder. All attempts to identify
mutations in the collagen genes (most likely candidates) have
failed.
Included in the chromosomal region of the recent locus on
13 are several important candidates. The gene referred to as
ITR is a G protein-coupled receptor that is increased in
intimal thickening and could play a role in the pathogenesis.21
A more exciting group of candidate genes are the so-called
glypican family of genes,22 of which GPC5 and 6 are located
on chromosome 13 in the region of 13q. This family of genes
encode for the cell surface heparin sulfate proteoglycans,
which serve as ligands for adhesion and several growth
factors including fibroblast growth factor, heparin-binding
epidermal growth factor, hepatocyte growth factor, and Wnts.
Identification of several genes will not immediately provide
elucidation of the pathogenesis but will be a significant step
forward.
In the broad picture, why is the search for genes responsible for mitral valve prolapse significant? It is unlikely to
add to our knowledge of the pathogenesis of atherosclerosis,
the number 1 killer of Americans. If one becomes obsessed
with finding the holy grail, however, then few research
studies would in themselves be significant. It is the series of
discoveries, each of which may appear insignificant, that
leads to the big bang. The discovery of the receptor for
cholesterol opened up a world that led to the development of
statin therapy. A major operation today is that of replacement
or repair of the mitral valve, done primarily for mitral valve
prolapse. There is extensive research ongoing to improve on
valves made from tissue. It is highly likely that identifying the
1926
Circulation
September 27, 2005
genes involved with the growth and maintenance of the valve
will help in this quest. In an era in which there is hope the
whole heart can be regenerated, we must also be prepared to
regenerate such structures as valves. Identifying the genes
may be a prerequisite if we hope to generate cardiac valves in
culture. Understanding the factors that control collagen,
elastin, and proteoglycans such as heparan sulfate have
significance not only in terms of valve leaflets but also for
blood vessels and other organs composed of these structures.
Having identified 3 chromosomal loci for mitral valve prolapse does not in itself load the train with genes, but it does
suggest that the caboose is waiting to hitch it. Let us hope that
with our ability today to rapidly sequence DNA and evaluate
new candidate genes we will enable the train to leave the
station soon.
10.
11.
12.
13.
14.
15.
References
1. Leathman A, Brigden W. Mild mitral regurgitation and the mitral
prolapse fiasco. Am Heart J. 1980;99:659 – 664.
2. Warth DC, King ME, Cohen JM, Tesoriero VL, Marcus E, Weyman AE.
Prevalence of mitral valve prolapse in normal children. J Am Coll
Cardiol. 1985;5:1173–1177.
3. Markiewicz W, Stoner J, London E, Hung SA, Popp RL. Mitral valve
prolapse in one hundred presumably healthy young females. Circulation.
1976;53:464 – 468.
4. Procacci PM, Savran SV, Schreiter SL, Bryson AL. Prevalence of clinical
mitral-valve prolapse in 1169 women. N Engl J Med. 1976;294:
1086 –1089.
5. Savage DD, Devereux RB, Garrison RJ, Castelli WP, Anderson SJ, Levy
D, Thomas HE, Kannel WB, Feinleib M. Mitral valve prolapse in the
general population. 2. Clinical features: the Framingham Study. Am
Heart J. 1983;106:577–581.
6. Connolly HM, Crary JL, McGoon MD, Hensrud DD, Edwards BS,
Edwards WD, Schaff HV. Valvular heart disease associated with
fenfluramine-phentermine. N Engl J Med. 1997;337:581–588.
7. Freed LA, Levy D, Levine RA, Larson MG, Evans JC, Fuller DL,
Lehman B, Benjamin EJ. Prevalence and clinical outcome of mitral-valve
prolapse. N Engl J Med. 1999;341:1–7.
8. Freed L, Benjamin EJ, Levy D, Larson MG, Evans JC, Fuller DL, Leham
B, Levine RA. Mitral valve prolapse in the general population: the benign
nature of echocardiographic features in the Framingham Heart Study.
J Am Coll Cardiol. 2002;40:1298 –1304.
9. Levine RA, Stathogiannis E, Newell JB, Harrigan P, Weyman AE.
Reconsideration of echocardiographic standards for mitral valve prolapse:
lack of association between leaflet displacement isolated to the apical four
16.
17.
18.
19.
20.
21.
22.
chamber view and independent echocardiographic evidence of abnormality. J Am Coll Cardiol. 1988;11:1010 –1019.
Levine RA, Handschumacher MD, Sanfilippo AJ, Hagege AA, Harrigan
P, Marshall JE, Weyman AE. Three-dimensional echocardiographic
reconstruction of the mitral valve, with implications for the diagnosis of
mitral valve prolapse. Circulation. 1989;80:589 –598.
Devereux RB, Jones EC, Roman MJ, Howard BV, Fabsitz RR, Liu JE,
Palmieri V, Welty TK, Lee ET. Prevalence and correlates of mitral valve
prolapse in a population-based sample of American Indians: the Strong
Heart Study. Am J Med. 2001;111:679 – 685.
Theal M, Sleik K, Anand S, Yi Q, Yusuf S, Lonn E. Prevalence of mitral
valve prolapse in ethnic groups. Can J Cardiol. 2004;20:511–515.
Barlow JB, Bosman CK. Aneurysmal protrusion of the posterior leaflet of
the mitral valve. An auscultatory-electrocardiographic syndrome. Am
Heart J. 1966;71:166 –178.
Disse S, Abergel E, Berrebi A, Houot AM, LeHeuzey JY, Diebold B,
Guise L, Carpentier A, Corvol P, Jeunemaitre X. Mapping of a first locus
for autosomal dominant myxomatous mitral-valve prolapse to chromosome 16p11.2-p12.1. Am J Hum Genet. 1999;65:1242–1251.
Freed LA, Acierno JS, Dai D, Leyne M, Marshall JE, Nesta F, Levine
RA, Slaugenhaupt SA. A locus for autosomal dominant mitral valve
prolapse on chromosome 11p15.4. Am J Hum Genet. 2003;72:
1551–1559.
Nesta F, Leyne M, Yosefy C, Simpson C, Dai D, Marshall JE, Hung J,
Slaugenhaupt SA, Levine RA. New locus for autosomal dominant mitral
valve prolapse on chromosome 13: clinical insights from genetic studies.
Circulation. 2005;112:2022–2030.
O’Rourke RA, Bailey SR. Mitral valve prolapse syndrome. In: Fuster V,
Alexander R, O’Rourke RA, et al, eds. Hurst The Heart. 11th ed. New
York: McGraw-Hill Professional; 2004:1695–1706.
Tamura K, Fukuhara Y, Ishizaki M, Masuda Y, Yamanaka N, Ferrans VJ.
Abnormalities in elastic fibers and other connective-tissue components of
floppy mitral valve. Am Heart J. 1995;129:1149 –1158.
Grande-Allen K, Griffin BP, Calabro A, Ratliff NB, Cosgrove DM,
Vesely I. Myxomatous mitral valve chordae. II: Selective elevation of
glycosaminoglycan content. J Heart Valve Dis. 2001;10:325–332.
Rabkin E, Aikawa M, Stone JR, Fukumoto Y, Libby P, Schoen FJ.
Activated interstitial myofibroblasts express catabolic enzymes and
mediate matrix remodeling in myxomatous heart valves. Circulation.
2001;104:2525–2532.
Tsukada S, Iwai M, Nishiu J, Itoh M, Tomoike H, Horiuchi M, Nakamura
Y, Tanaka T. Inhibition of experimental intimal thickening in mice
lacking a novel G-protein-coupled receptor. Circulation. 2003;107:
313–319.
Paine-Saunders S, Viviano BL, Saunders S. GPC6, a novel member of the
glypican gene family, encodes a product structurally related to GPC4 and
is colocalized with GPC5 on human chromosome 13. Genomics. 1999;
57:455– 458.
䡲
heart failure
䡲
valves
䡲
genetics
䡲
mitral valve
Arrhythmia/Electrophysiology
Impaired Impulse Propagation in Scn5a-Knockout Mice
Combined Contribution of Excitability, Connexin Expression, and Tissue
Architecture in Relation to Aging
Toon A.B. van Veen, PhD*; Mera Stein, MD*; Anne Royer, PhD; Khaï Le Quang, MS;
Flavien Charpentier, PhD; William H. Colledge, PhD; Christopher L.-H. Huang, MD, PhD;
Ronald Wilders, PhD; Andrew A. Grace, PhD, FRCP; Denis Escande, MD, PhD;
Jacques M.T. de Bakker, PhD; Harold V.M. van Rijen, PhD
Background—The SCN5A sodium channel is a major determinant for cardiac impulse propagation. We used epicardial
mapping of the atria, ventricles, and septae to investigate conduction velocity (CV) in Scn5a heterozygous young and
old mice.
Methods and Results—Mice were divided into 4 groups: (1) young (3 to 4 months) wild-type littermates (WT); (2) young
heterozygous Scn5a-knockout mice (HZ); (3) old (12 to 17 months) WT; and (4) old HZ. In young HZ hearts, CV in
the right but not the left ventricle was reduced in agreement with a rightward rotation in the QRS axes; fibrosis was
virtually absent in both ventricles, and the pattern of connexin43 (Cx43) expression was similar to that of WT mice. In
old WT animals, the right ventricle transversal CV was slightly reduced and was associated with interstitial fibrosis. In
old HZ hearts, right and left ventricle CVs were severely reduced both in the transversal and longitudinal direction;
multiple areas of severe reactive fibrosis invaded the myocardium, accompanied by markedly altered Cx43 expression.
The right and left bundle-branch CVs were comparable to those of WT animals. The atria showed only mild fibrosis,
with heterogeneously disturbed Cx40 and Cx43 expression.
Conclusions—A 50% reduction in Scn5a expression alone or age-related interstitial fibrosis only slightly affects
conduction. In aged HZ mice, reduced Scn5a expression is accompanied by the presence of reactive fibrosis and
disarrangement of gap junctions, which results in profound conduction impairment. (Circulation. 2005;112:1927-1935.)
Key Words: fibrosis 䡲 gap junction 䡲 sodium channel 䡲 aging 䡲 conduction
T
he voltage-gated sodium channel is the key determinant
of cardiac excitability. The amplitude of the sodium
current determines the upstroke velocity of the action potential and, in conjunction with the expression/distribution of
gap junction channels and the structural organization of the
collagenous skeleton, the conduction velocity (CV) of the
electrical impulse. The SCN5A gene encodes the poreforming ␣-subunit of the cardiac sodium channel. Haploinsufficiency in SCN5A has been associated with the inherited
Lenègre disease1 (also called progressive cardiac conduction
defect) and with the Brugada syndrome.2,3 In patients with
inherited Lenègre disease, the conduction of the cardiac
impulse is abnormally slow and becomes progressively
slower with aging, ultimately leading to atrioventricular block
and pacemaker implantation in the elderly.1,4 A comparable
conduction defect has also been associated with the SCN5Arelated Brugada syndrome.5 In both situations, alteration in
conduction largely predominates in the right ventricle (RV).
A mouse model with targeted disruption in Scn5a has been
established.6 At the homozygous state, mice are not viable
and die before birth. In contrast, heterozygous Scn5adeficient mice live and reproduce normally. In preceding
reports, we have shown that Scn5a⫹/⫺ mice have ventricular
conduction slowing and ventricular arrhythmias6 and that
Scn5a⫹/⫺ mice recapitulate many aspects of the inherited
Lenègre disease, including the age-related progressive conduction slowing.7 Surprisingly, we found that the progressive
alteration in conduction was associated with myocardial
Received February 2, 2005; revision received June 17, 2005; accepted June 24, 2005.
From the Heart Lung Center Utrecht, Department of Medical Physiology (T.A.B.v.V., M.S., H.V.M.v.R.) and Department of Cardiology (M.S.,
J.M.T.d.B.), University Medical Center Utrecht, Utrecht, the Netherlands; INSERM U533 (A.R., K.L.Q., F.C., D.E.), l’Institut du Thorax, Faculté de
Médecine, Nantes, France; Departments of Biochemistry and Physiology (W.H.C., C.L.-H.H., A.A.G.), University of Cambridge, Cambridge, United
Kingdom; Department of Physiology (R.W.), Academic Medical Center, University of Amsterdam, Amsterdam, the Netherlands; Interuniversity
Cardiology Institute of the Netherlands (J.M.T.d.B.), Utrecht, the Netherlands; and the Experimental and Molecular Cardiology Group (J.M.T.d.B.),
Cardiovascular Research Institute, Amsterdam, the Netherlands.
*Drs van Veen and Stein contributed equally to this article.
Correspondence to Toon A.B. Van Veen, PhD, Department of Medical Physiology, University Medical Center Utrecht, Yalelaan 50, 3584 CM Utrecht,
The Netherlands. E-mail [email protected]
1927
1928
Circulation
September 27, 2005
rearrangements including extensive fibrosis. In the present
work, we have further characterized the conduction defect in
young and old Scn5a⫹/⫺ mice by making use of epicardial
mapping in association with immunohistochemistry. This
investigation demonstrates the following: (1) The conduction
defect resides in the ventricles, whereas bundle-branch CV is
unaffected; (2) conduction slowing preferentially concerns
the RV, which coincides with the predominant phenotype in
inherited Lenègre disease and Brugada syndrome patients;
and (3) the severity of cardiac sodium channel dysfunction
becomes manifest in the presence of an age-related increase
in collagen deposition accompanied by a disturbed pattern of
expressed gap junctions. The present study provides an
experimental ground to support further evaluation of the
therapeutic potential of drugs that prevent myocardial fibrosis, in the context of channelopathies related to loss-offunction SCN5A mutations.
Methods
Animals
Heterozygous Scn5a-knockout mice (HZ), generated in Cambridge,
United Kingdom, were bred at l’Institut du Thorax, Faculté de
Médecine, Nantes, France. All experiments were performed on adult
sex- and age-matched HZ and wild-type (WT) mice from the same
litters (as controls). Mice were divided in 4 groups depending on age
and heterozygosity for Scn5a: group 1, young WT mice; group 2,
young HZ mice; group 3, old WT mice; and group 4, old HZ mice.
Young mice were 3 to 4 months old, and old mice were 12 to 17
months of age. Animal experiments were performed in accordance
with institutional guidelines for animal use in research.
Preparation of Hearts for Langendorff Perfusion
Mice were anesthetized by an intraperitoneal injection of urethane (2
g/kg body weight). The heart was excised, prepared, and connected
to a Langendorff perfusion setup as described previously.8 –10
Recording of Electrograms During
Langendorff Perfusion
Atrial electrograms of the right (RA) and left (LA) atrium were recorded
with a 168-point multielectrode (12⫻14 grid, spacing 200 ␮m). The
atrium was stimulated with an external electrode at a basic cycle length
(BCL) of 150 ms from a site at the upper edge of the electrode grid.
Ventricular and bundle-branch recordings were made with a 247-point
multielectrode (19⫻13 grid, spacing 300 ␮m). The ventricles were
stimulated from the center of the grid at a BCL of 100 ms for the right
ventricle (RV) and 150 ms for the left ventricle (LV). Bundle-branch
recordings were made during pacing of the RA with a bipolar electrode
and at a BCL of 150 ms. Recordings were made in unipolar mode with
regard to a reference electrode connected to the support of the heart.
Electrograms were acquired with a custom-built 256-channel dataacquisition system. Signals were bandpass filtered (low cutoff 0.16 Hz
[12 dB], high cutoff 1 kHz [6 dB]) and digitized with 16-bit resolution
at a bit step of 2 ␮V and a sampling frequency of 2 or 4 kHz. The input
noise of the system was 4 ␮V (peak-peak).
For septal measurements, the RV and LV free walls were removed,
and the electrode grid was positioned on the interventricular septum just
below the atrioventricular valves. The effective refractory period (ERP),
the longest coupling interval of the premature stimulus that failed to
activate the entire heart, was determined for each site of stimulation
separately. Every sixteenth stimulus was followed by 1 premature
stimulus. Starting at 140 or 90 ms (for BCL 150 or 100 ms, respectively), the coupling interval of the premature stimulus was reduced in
steps of 10 or 5 ms (for BCL 150 or 100 ms, respectively) until ERP.11
Data Analysis
The moment of maximal negative dV/dt in the unipolar electrograms
was determined with custom-written software based on Matlab (The
Mathworks Inc), selected as the time of local activation and
activation maps were constructed. CVs of the ventricles in longitudinal (parallel to the fiber orientation) and transversal (perpendicular
to the fiber orientation) directions and of the atria in the transversal
direction were determined from the paced activation maps. Activation times of at least 4 consecutive electrode terminals along lines
perpendicular to intersecting isochronal lines were used to estimate
CVs. Dispersion of conduction was assessed for the LV and RV.12
Statistical comparisons shown in the Table were performed by 2-way
Electrophysiological Parameters of Young and Old Scn5aⴙ/ⴙ and Scn5aⴙ/ⴚ Mice
ANOVA P
Young WT
Young HZ
Old WT
Old HZ
Age
Genotype
RA CV
30.25⫾2.37 (n⫽8)
22.46⫾1.94 (n⫽5)*
30.4⫾2.81 (n⫽8)
23.6⫾1.19 (n⫽6)
0.796
0.006
LA CV
29.69⫾3.5 (n⫽8)
28.9⫾2.08 (n⫽8)
29.6⫾2.00 (n⫽10)
23.2⫾1.2 (n⫽10)
0.190
0.118
33.2⫾2.9 (n⫽3)
31⫾2.25 (n⫽3)
32.5⫾5.5 (n⫽2)
0.455
0.957
0.663
0.533
0.051
⬍0.001
⬍0.001
⬍0.001
RBB CV
35⫾2.16 (n⫽5)
LBB CV
37.4⫾4.86 (n⫽6)
40.7⫾2.74 (n⫽3)
36.3⫾2.95 (n⫽5)
38.2⫾2.8 (n⫽5)
RV longitudinal CV
32.7⫾1.9 (n⫽10)
26.5⫾1.54 (n⫽7)*
30.1⫾1.99 (n⫽9)
21.4⫾1.76 (n⫽12)*
RV transversal CV
23.2⫾1.3 (n⫽10)
18.3⫾1.32 (n⫽7)*
17.5⫾0.59 (n⫽10)†
11.2⫾1.07 (n⫽12)*†
RV AR
1.43⫾0.1 (n⫽10)
1.47⫾0.09 (n⫽7)
1.76⫾0.11 (n⫽9)
2.02⫾0.20 (n⫽12) †
0.008
0.359
LV longitudinal CV
36.2⫾2.77 (n⫽9)
30.1⫾3.58 (n⫽6)
34.6⫾2.06 (n⫽9)
23.1⫾1.59 (n⫽12)*
0.082
⬍0.001
LV transversal CV
20.3⫾1.22 (n⫽10)
19.7⫾1.22 (n⫽6)
18.6⫾1.42 (n⫽9)
13.2⫾1.09 (n⫽12)*†
0.003
0.032
LV AR
1.81⫾0.1 (n⫽8)
1.5⫾0.12 (n⫽6)
1.92⫾0.14 (n⫽9)
1.89⫾0.21 (n⫽12)
0.168
0.354
RA ERP
40.8⫾6.45 (n⫽12)
55.7⫾3.69 (n⫽7)
46.7⫾4.71 (n⫽9)
60⫾4.47 (n⫽11)
0.363
0.015
LA ERP
48.3⫾7.16 (n⫽12)
62.5⫾7.26 (n⫽8)
69.1⫾6.67 (n⫽11)
0.041
0.381
RV ERP
54.6⫾1.9 (n⫽12)
68.1⫾3.77 (n⫽8)*
68.3⫾2.04 (n⫽9)†
78.5⫾3.27 (n⫽13)*†
⬍0.001
⬍0.001
LV ERP
71.4⫾4.04 (n⫽7)
70⫾4.47 (n⫽6)
80.5⫾3.2 (n⫽10)
94.7⫾4.96 (n⫽15)*†
0.002
0.219
71⫾5.86 (n⫽10)†
RBB indicates right bundle branch; LBB, left bundle branch; and AR, anisotropic ratio.
CV is in centimeters per second; ERP is in milliseconds. Values are mean⫾SEM. n⫽No. of independent experiments.
*P⬍0.05 vs WT.
†P⬍0.05 vs Young.
van Veen et al
Conduction Velocity in Scn5a-Knockout Mice
1929
Figure 1. Representative activation maps of ventricles and atria. Depicted are activation maps of the LV (column A), RV (column B),
LA (column C), and RA (column D). Group 1: young SCN5A WT; group 2: young SCN5A HZ; group 3: old SCN5A WT; and group 4: old
SCN5A HZ. Earliest activation is given in red, latest in blue. Black lines indicate sites of isochronal activation; bold arrows, longitudinal
conduction (VL); dashed arrows, transversal conduction (VT).
ANOVA, with Holm-Sidak post hoc test with SigmaStat 3.11
(Systat). RV and LV ERPs and right and left bundle-branch CVs
were compared by a Mann-Whitney rank sum test. All data are
expressed as mean⫾SEM, and probability values ⬍0.05 were
considered statistically significant.
raised against Cx40 (Alpha Diagnostics). Secondary antibodies
(Texas Red and FITC conjugated whole IgG) were purchased from
Jackson Laboratories.
ECG Recording and QRS Axis Measurement
Ventricular Conduction and Refractoriness
Our method to record mouse ECG (leads I, II, and III) can be found
elsewhere.7 The QRS axis was calculated for 12 young Scn5a⫹/⫺
mice and 19 WT littermates. For each lead, the QRS complex surface
area of an average beat was measured and plotted as a vector on an
Einthoven triangle. The electrical axis of the QRS complex was then
determined as the resultant of the 3 vectors.
Typical activation maps of the LV and RV are illustrated in
Figure 1. Crowding of isochronal lines in both LV and RV
activation maps was most prominent in old HZ animals. The
Table shows the average values for CV and ERP of the atria,
bundle branches, and ventricles. Significant differences for
separate groups are indicated, whereas the 2 rightmost columns indicate the overall effects of age and genotype. In
young mice, longitudinal and transversal CVs of the RV were
significantly reduced in HZ mice, whereas the anisotropic
ratio (longitudinal CV divided by the transversal CV) was
unchanged. In the LV of young mice, both longitudinal and
transversal CVs were unaltered. Predominant alteration of the
RV CV was in agreement with the rightward shift of the QRS
axes as measured in surface ECG recordings (Figure 2). QRS
axes of WT mice clustered in the left inferior quadrant, with
an average value of 75⫾6°. In contrast, most HZ mice had a
rightward deviation of their QRS axis, with an average value
of 121⫾25°. The QRS axis of old HZ mice was much more
Immunohistochemistry and Histology
After excision, hearts were rapidly frozen in liquid nitrogen and
stored at ⫺80°C. For each of the 4 groups, 6 hearts were sectioned
serially to generate sections of 10 ␮m thickness. Sections taken from
different levels were incubated with antibodies as reported previously.9 After immunolabeling, sections were mounted in Vectashield
(Vector Laboratories) and examined with a classic light microscope
with epifluorescence equipment (Nikon Optiphot-2). To evaluate the
presence of fibrosis, sections serial to the ones used for antibody
labeling were fixed with 4% paraformaldehyde (in PBS, 30 minutes
at room temperature) and stained with Pico Sirius red.13
Antibodies
We used mouse monoclonal antibodies raised against connexin (Cx)
43 (Transduction Laboratories) and rabbit polyclonal antibodies
Results
1930
Circulation
September 27, 2005
atrial CV was not significantly different between the 4 groups
except for RA, where a reduction in CV due to genotype was
found (Table). ERP in the LA of old WT mice was increased
significantly compared with young mice (Table). In the RA,
there was an overall increase in ERP due to genotype, but for
the separate groups (young HZ versus young WT and old HZ
versus old WT), statistical significance was not reached.
Ventricular Distribution of Fibrosis and
Expression of Gap Junction Proteins
Figure 2. Individual (thin arrows) and median (thick arrows) QRS
axes of 12 young Scn5a⫹/⫺ (red lines) and 19 WT (black lines)
mice. See text for further comments.
dispersed. The rightward shift in QRS axes shown in HZ mice
suggests an abnormal activation sequence of the ventricles.
The pathophysiological relevance of this anomaly remains
unclear because of the unknown mouse ECG symptomatology and lack of 12-lead recordings.
In old WT mice, RV transversal CV was reduced compared
with young WT animals, whereas conduction in the LV was
similar (Table). In old HZ mice, both longitudinal and
transversal CVs in the RV and LV were markedly decreased.
In these mice, the right but not the left anisotropic ratio was
increased. In the RV, the ERP was significantly increased by
age and genotype (Table). For the LV, ERP was significantly
increased in old HZ mice. In all groups except the young HZ
mice, RV ERP was significantly shorter than LV ERP.
Dispersion of conduction of the RV and LV was not significantly different between the 4 groups (data not shown).
Bundle-Branch Conduction
Figure 3A shows electrograms and activation maps during
septal mapping. Both electrograms show a remote atrial
deflection (a), a bundle-branch deflection (p), and a ventricular deflection (v). The upper activation map shows bundlebranch activation (at 0 to 16 ms) and the lower map shows
that of septal activation (at 16 to 24 ms). Typical examples of
right and left bundle-branch activation in young and old
Scn5a⫹/⫹ and Scn5a⫹/⫺ mice are displayed in Figure 3B. CV
in both bundle branches was not affected by downexpression
of Scn5a, age, or both (Table). In animals from all groups, CV
tended to be slower in the right than in the left bundle branch,
but statistical significance was not reached.
Atrial Conduction and Refractoriness
Typical activation maps of paced LA and RA of the 4 groups
of mice are shown in Figure 1. There were no differences in
the activation pattern among the different groups. Grossly,
Histochemical analysis was performed to reveal the presence
of fibrosis in relation to the expression pattern of Cx43, which
constituted the main conductive gap junction channels in the
ventricles. In young mice, either WT or HZ, ventricular
fibrosis (red staining) was virtually absent (Figure 4A). As
shown in Figure 4B, old WT hearts demonstrated interstitial
fibrosis as tiny strands between the muscle fibers. In contrast,
fibrosis in old HZ hearts was largely increased as compared
with either young HZ or old WT hearts. In addition to
increased interstitial fibrosis (arrowhead), Sirius red staining
showed a different pattern of reactive fibrosis (asterisk),
which was heterogeneously present throughout the LV free
wall, RV free wall, and interventricular septum. This pattern
was found in 6 of 6 old HZ hearts. In contrast, small spots of
reactive fibrosis could be detected in only 1 of 6 old WT
hearts.
Immunolabeling of sections serial to the ones used for
evaluation of fibrosis revealed a regular and comparable
pattern of Cx43 in young WT and HZ hearts (Figure 4A). In
old WT hearts with mild interstitial fibrosis (Figure 4B, upper
left), Cx43 expression was gathered in large plaques in a
regular distribution (Figure 4B, upper right) comparable to
the patterns observed in young WT and HZ hearts. In old HZ
hearts, fibrosis was heterogeneous and locally massive (Figure 4B, lower left). In areas with severe deposition of fibrosis,
Cx43 was downregulated, whereas the remaining Cx43
showed an irregular pattern (Figure 4B, lower right). Because
of the heterogeneous character of fibrotic deposition in the
old HZ hearts (Figure 5A), areas existed with an expression
pattern of Cx43 that was close to normal (Figure 5B),
comparable to the expression pattern found in old WT hearts.
This pattern clearly differed from that found in tissue forming
the borderzone between nonfibrotic and fibrotic tissue. Here,
labeling was more diffuse (Figure 5C) and not gathered in
distinct gap junction plaques that might be indicative for
redistribution (arrow). Central in a fibrotic spot, the remaining viable myocytes still expressed low amounts of Cx43 in
an irregular pattern (Figure 5D, arrow). In a higher magnification (Figure 5E), the deposition of fibrosis aligned with the
absence of ␣-actinin staining in a consecutive section where
Cx43 labeling was restricted to the ␣-actinin–positive cardiomyocytes (Figure 5F).
Distribution of Fibrosis and Expression of Gap
Junction Proteins in the Bundle Branches
To identify myocytes composing the bundle branches, sections serial to the ones used for Sirius red staining were
immunolabeled with antibodies against Cx40. Cx40 is a
known marker for the conduction system and is not expressed
van Veen et al
1931
Figure 3. A, Typical example of electrograms and activation maps during RV
septal mapping of a young WT heart.
The selected electrograms contain a
remote atrial deflection (a), a bundlebranch deflection (p), and a ventricular
deflection (v). The first local activation
time of the bundle branch was defined
as 0 ms. The upper activation map was
constructed from times 0 to 16 ms, displaying activation of the bundle branch,
whereas the lower activation map was
constructed from times 16 to 24 ms,
showing activation of the septal myocardium. B, Typical examples of right (RBB)
and left bundle-branch (LBB) activation
patterns in young and old WT and HZ
hearts.
in adult working ventricular cardiomyocytes.14,15 Figures 6A
and 6B show the results obtained in young and old hearts,
respectively. In young mice (Figure 6A), a low amount of
fibrosis was present in both bundle branches, which were
positively labeled with Cx40 staining. With regard to the
degree of fibrosis, no differences were observed between WT
and HZ hearts or between the left and right bundles. Similar
results were obtained with analysis of old HZ and WT hearts
(Figure 6B). However, the overall degree of fibrosis that
surrounded the myocytes composing the Cx40-positive bundle branches was increased compared with young hearts.
Atrial Distribution of Fibrosis and Expression of
Gap Junction Proteins
Atrial expression patterns of the gap junction proteins Cx40
and Cx43 were analyzed in young (Figure 7A) and old hearts
(Figure 7B) of both genotypes. In young WT and HZ atria,
the expression of Cx40 and of Cx43 was highly comparable.
In old HZ mice, however, atrial expression of both Cx40 and
Cx43 differed from that in old WT hearts. Both isoforms were
regionally downregulated, and the pattern of expression was
more diffuse for HZ than for WT atria. In young mice, atrial
fibrosis was absent in both genotypes (Figure 7A, right;
fibrosis in red). Although the general degree of atrial fibrosis
in old mice was increased compared with young mice, no
difference in degree of fibrosis was found between old WT
and old HZ atria (Figure 7B, right panels) or between the LA
and RA. In addition, reactive fibrosis such as that observed in
old HZ ventricles was not found in the atria of all groups.
Discussion
Our mouse model, in which expression of Scn5a in the heart
is genetically reduced by 50%, has been shown to have
slowed ventricular conduction.6,7 Here, we report that al-
1932
Circulation
September 27, 2005
Figure 4. A, Ventricular fibrosis (in red) in young mice (left) is
virtually absent in WT and HZ hearts, which leaves the expression pattern of Cx43 unaffected (right). B, Ventricular fibrosis in
old WT animals is increased in the interstitium between the
muscle fibers (upper left panel, arrowheads), whereas it is locally
massive in HZ old animals (lower left panel). In those areas,
fibrosis is found both laterally (interstitial, arrowhead) and as
replacement (asterisk). Fibrosis is accompanied by a downregulation and redistribution of Cx43 in HZ old hearts (lower right
panel), whereas Cx43 in old WT hearts appears unaffected
(upper right panel). Bar⫽50 ␮m in pictures showing Cx43 and
100 ␮m in those showing Sirius red. ⫹/⫹ indicates WT; ⫹/⫺,
HZ.
though there is slightly impaired conduction of the electrical
impulse (mainly in the RV) in young heterozygous animals, it
is only in older heterozygous animals that conduction becomes markedly impaired at the ventricular level. In those
hearts, reduced expression of Scn5a was associated with
increased fibrosis and a reorganized expression pattern of gap
junction channels. Our observations indicate that only the
synergism between reduced Scn5a expression, increased
fibrosis, and impaired intercellular coupling leads to markedly decreased CV of the electrical impulse in the ventricles.
Determinants of Impulse Propagation
Intercellular coupling, sodium channel expression, and tissue
architecture mediate propagation of the electrical impulse in
cardiac tissue. Disturbances in one of these determinants may
affect propagation of the electrical impulse and vulnerability
Figure 5. A, Low magnification of heterogeneous fibrosis (in red)
in an old HZ LV. Indicated are the areas B, C, and D, which represent the areas in a serial section in which immunolabeling
against Cx43 is depicted in panels B, C, and D respectively. B,
Normal Cx43 expression in areas without severe fibrosis. This
expression pattern differs markedly from that found at the borderzone of a fibrotic area (C), where labeling showed a diffuse
pattern indicative for redistribution. Within a fibrotic spot, Cx43
is severely reduced, leaving only a few Cx43-positive conductive pathways left (D). E, Higher magnification of an area with
severe fibrosis, with, on a consecutive section (F), the accompanying expression of ␣-actinin (in red) and Cx43 (green). Bar⫽250
␮m in A and 50 ␮m in B through F. ⫹/⫺ indicates HZ; SR,
Sirius red.
for arrhythmias. However, a 50% reduction in expression of
Cx43 per se does not affect impulse propagation in the mouse
heart.11,16 Supported by a theoretical study,17 it has been
shown that to induce electrical disturbances, the reduction has
to be very robust11,18 or highly heterogeneous.19 Our recordings in Scn5a young heterozygous mice in the present study
show that CV is only mildly affected by a 50% reduction in
sodium channel expression, in agreement with previous
observations.6 Finally, an increase of interstitial fibrosis by a
factor of 4, as observed in old WT mice, also has only a mild
effect on conduction. This indicates that CV integrity is
preserved over a wide range of alterations in the determinants
of conduction if only 1 of them is affected.
Location of Conduction Slowing
Widening of the QRS complex in the surface ECG of the
heterozygous mice, as previously observed,6,7 could be due to
impaired conduction in the bundle branches or ventricular
myocardium. The present study shows that CV in the bundle
branches of heterozygous mice is normal, whereas it is
reduced in the ventricular myocardium. In addition, the study
van Veen et al
1933
affected. A 50% reduction in sodium channel expression
alone only affects conduction in the RV significantly, albeit
slightly (15%). Aging in the WT mice in the present study
increased fibrosis by a factor of 4. This slightly reduced CV
in the RV, but only in the transverse direction. This is
compatible with other studies that show that the major effect
of fibrosis on conduction is in the transverse direction.21
Normal transversal CV in the LV, as mentioned before, might
be related to its greater wall thickness and the transmural
rotation of the fiber direction.
The synergistic effects on conduction of decreased excitability and cell-cell coupling, in concert with increased
collagen deposition, may be due to the following factors.
Reactive fibrosis has been shown to give rise to tissue
discontinuities, which may result in conduction delay due to
a mismatch between current supply and demand.22,23 If
demand surpasses supply, less current for excitation is available at the discontinuity, which delays conduction. Delay at
the discontinuity will be further increased if less sodium
current is available because of reduced Scn5a expression.
Finally, the reduced cell-cell coupling caused by disturbed
connexin expression reduces conduction even further.
Conduction Slowing in the Atria
Figure 6. A, B, (Immuno)histochemical staining of the left bundle
branch in young (A) and old (B) mice. Left panels show positive
labeling for Cx40 in the left bundle branch (arrows), leaving the
working myocardium negative. Right panels show the presence
of fibrosis (in red) as marked with Sirius red staining of sections
serial to the ones used for Cx40 labeling. Arrows indicate fibrosis in the left bundle branch on sites with positive Cx40 staining
(compare with left panels). Bar⫽50 ␮m. ⫹/⫹ indicates WT;
⫹/⫺, HZ.
shows that in young heterozygous mice, conduction slowing
occurs preferentially in the RV. In loss-of-function sodium
channel–associated human disease, such as the Brugada
syndrome and Lenègre disease, the RV is also preferentially
affected.5 The higher vulnerability of the RV compared with
the LV was also observed in a conditional Cx43-knockout
mouse in which CV in the RV was more affected than CV in
the LV.11 Differences in wall thickness between the RV and
LV might play a role. In this context, Schalij et al20 demonstrated slower conduction in an epicardial monolayer than in
multilayered tissue. Rotation of the fiber direction from
epicardium to endocardium is probably involved. In old
heterozygous mice in the present study, conduction slowing
was more pronounced and occurred in both the RV and LV
but still predominated in the RV.
Synergism of Impaired Conduction Parameters
The severity of reduced sodium channel expression only
becomes manifest in old mice, which show increased reactive
fibrosis together with a disturbed pattern of Cx43 expression.
If only 1 of these factors is impaired, conduction is slightly
In the atria of old heterozygous animals, a mild increase in
fibrosis was observed similar to that found in old WT hearts.
Atrial expression patterns of Cx40 and Cx43 in old WT hearts
were comparable to those found in young WT animals,
although expression of both isoforms in old Scn5a heterozygous atria was slightly aberrant. Both Cx40 and Cx43 were
regionally reduced and irregularly distributed. Electrophysiological measurements showed a reduction of CV due to
genotype in RA only. In both atria, there was no effect of age
on conduction. The lack of dramatic effects of age on impulse
propagation is likely related to the modest morphological
alterations of the atria from old heterozygous hearts.
Comparison With Lenègre Disease and
Brugada Syndrome
The heterozygous mouse model reveals some characteristics
of Lenègre disease but differs in other aspects. The mouse
model mimics Lenègre disease because the severity of conduction defects in the heterozygous mouse increases with age,
as in Lenègre disease. In patients with inherited Lenègre
disease, conduction slowing predominates in the RV, as in the
mouse. Among a group of 25 Lenègre gene carriers, 9 had
right bundle-branch block and 8 had parietal block, whereas
only 2 had left bundle-branch block.4 The high incidence of
parietal block (33%) suggests that in many patients, CV in the
bundles remains close to normal. The mouse model differs
from Lenègre disease, however, because in young and old
heterozygous mice, CV in the bundle branches remains
normal. In the mouse, fibrosis around the bundle branches
increases with age, but there was no increased deposition of
collagen within the bundle branches that could affect conduction. This opposes pathological observations made by Lenègre
and Moreau.24 This difference might be related to the difference
in size between mouse and human hearts. In the mouse, CV in
the bundles is only slightly faster than in the ventricles (see the
1934
Circulation
September 27, 2005
Figure 7. A, B, Immunohistochemical
staining of the atria in young (A) and
old (B) mice. Left and middle panels
show immunolabeling against Cx40 and
Cx43, respectively, whereas right panels
show the presence of fibrosis (in red) as
marked with Sirius red (SR) staining.
Whereas Cx40 and Cx43 patterns in the
young animals are comparable in WT
and HZ animals, both connexins are
downregulated and redistributed in the
old HZ animals. However, the amount of
fibrosis in the old atria is similar in WT
and HZ, even though it has been
increased by ageing. Bar⫽25 ␮m in the
pictures of immunolabeling and 50 ␮m in
the pictures showing Sirius red staining.
⫹/⫹ indicates WT; ⫹/⫺, HZ.
Table), and the role of the bundles in propagating the impulse is
less prominent than in humans. In the Brugada syndrome,
alterations in conduction also predominate in the RV,5 and an
aspect of right bundle-branch block in the right precordial leads
is a common finding that leads to delayed contraction of the
RV.25 Extensive fibrosis has been observed in an explanted heart
of a patient with Brugada syndrome.26 Whether fibrosis participates in the pathophysiology of the Brugada syndrome remains
to be established. Recently, several loss-of-function mutations in
the Scn5a channel have been linked to triggering the onset of
dilated cardiomyopathy in patients at middle age.27 In the
mouse, the mechanisms that lead to structural changes are still
unclear. We previously reported upregulation of Atf3 in heterozygous mice.7 Atf3, a member of the CREB/ATF family of
transcription factors expressed at very low levels in the normal
heart, has been shown to induce fibrosis and conduction abnormalities when overexpressed.
In conclusion, the present data show that CV is slightly
reduced in young heterozygous SCN5A-knockout mice, in
which only sodium channel expression is affected. In old HZ
mice, reduced expression of cardiac sodium channels is
accompanied by the presence of an age-related increase in
collagen deposition and a disturbed pattern of expressed gap
junctions, which results in pronounced conduction slowing at
the ventricular level. The present study provides experimental
grounds to support further evaluation of the therapeutic
potential of drugs that prevent myocardial fibrosis in the
context of channelopathies related to loss-of-function SCN5A
mutations.
Acknowledgments
This study was supported by the Netherlands Organization for
Scientific Research (grant 916.36.012 to Dr van Veen) and the
Netherlands Heart Foundation (grant No. 2003B128 to Dr Stein).
Additional support was received from the Ministère de la Recherche
(Action Concertée Incitative “Biologie du développement et physiologie intégrative,” to Dr Charpentier), the Groupement d’Intérêt
Scientifique - Institut des Maladies Rares (Dr Charpentier), the
Fondation de France (Dr Escande), and the British Heart Foundation
and the UK Medical Research Council (Drs Grace, Huang, and
Colledge).
References
1. Probst V, Kyndt F, Potet F, Trochu JN, Mialet G, Demolombe S, Schott
JJ, Baró I, Escande D, Le Marec H. Haploinsufficiency in combination
with aging causes SCN5A-linked hereditary Lenègre disease. J Am Coll
Cardiol. 2003;41:643– 652.
2. Clancy CE, Kass RS. Defective cardiac ion channels: from mutations to
clinical syndromes. J Clin Invest. 2002;110:1075–1077.
3. Kyndt F, Probst V, Potet F, Demolombe S, Chevallier JC, Baró I, Moisan
JP, Boisseau P, Schott JJ, Escande D, Le Marec H. Novel SCN5A
mutation leading either to isolated cardiac conduction defect or Brugada
syndrome in a large French family. Circulation. 2001;104:3081–3086.
4. Schott JJ, Alshinawi C, Kyndt F, Probst V, Hoorntje TM, Hulsbeek M,
Wilde AA, Escande D, Mannens MM, Le Marec H. Cardiac conduction
defects associate with mutations in SCN5A. Nat Genet. 1999;23:20 –21.
5. Smits JP, Eckardt L, Probst V, Bezzina CR, Schott JJ, Remme CA,
Haverkamp W, Breithardt G, Escande D, Schulze-Bahr E, LeMarec H,
Wilde AA. Genotype-phenotype relationship in Brugada syndrome: elec-
van Veen et al
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
trocardiographic features differentiate SCN5A-related patients from nonSCN5A-related patients. J Am Coll Cardiol. 2002;40:350 –356.
Papadatos GA, Wallerstein PM, Head CE, Ratcliff R, Brady PA,
Benndorf K, Saumarez RC, Trezise AE, Huang CL, Vandenberg JI,
Colledge WH, Grace AA. Slowed conduction and ventricular tachycardia
after targeted disruption of the cardiac sodium channel gene Scn5a. Proc
Natl Acad Sci U S A. 2002;99:6210 – 6215.
Royer A, van Veen TAB, Le Bouter S, Marionneau C, Griol-Charhbili V,
Léoni AL, Steenman M, van Rijen HVM, Demolombe S, Goddard CA,
Richer C, Escoubet B, Jarry-Guichard T, Colledge WH, Gros D, de
Bakker JMT, Grace AA, Escande D, Charpentier F. A mouse model of
SCN5A-linked hereditary Lenègre’s disease: age-related conduction
slowing and myocardial fibrosis. Circulation. 2005;111:1738 –1746.
Van Rijen HV, van Veen TA, van Kempen MJ, Wilms-Schopman FJ,
Potse M, Krueger O, Willecke K, Opthof T, Jongsma HJ, de Bakker JM.
Impaired conduction in the bundle branches of mouse hearts lacking the
gap junction protein connexin40. Circulation. 2001;103:1591–1598.
van Veen TA, van Rijen HV, Wiegerinck RF, Opthof T, Colbert MC,
Clement S, de Bakker JM, Jongsma HJ. Remodeling of gap junctions in
mouse hearts hypertrophied by forced retinoic acid signaling. J Mol Cell
Cardiol. 2002;34:1411–1423.
Alcoléa S, Jarry-Guichard T, de Bakker J, Gonzàlez D, Lamers W,
Coppen S, Barrio L, Jongsma H, Gros D, van Rijen H. Replacement of
connexin40 by connexin45 in the mouse: impact on cardiac electrical
conduction. Circ Res. 2004;94:100 –109.
van Rijen HV, Eckardt D, Degen J, Theis M, Ott T, Willecke K, Jongsma
HJ, Opthof T, de Bakker JM. Slow conduction and enhanced anisotropy
increase the propensity for ventricular tachyarrhythmias in adult mice
with induced deletion of connexin43. Circulation. 2004;109:1048 –1055.
Lammers WJ, Schalij MJ, Kirchhof CJ, Allessie MA. Quantification of
spatial inhomogeneity in conduction and initiation of reentrant atrial
arrhythmias. Am J Physiol. 1990;259:H1254 –H1263.
Sweat F, Puchtler H, Rosenthal SI. Sirius red F3Ba as a stain for connective tissue. Arch Pathol. 1964;78:69 –72.
Bastide B, Neyses L, Ganten D, Paul M, Willecke K, Traub O. Gap
junction protein connexin40 is preferentially expressed in vascular endothelium and conductive bundles of rat myocardium and is increased under
hypertensive conditions. Circ Res. 1993;73:1138 –1149.
Gros D, Jarry-Guichard T, Ten Velde I, De Mazière AMGL, Van
Kempen MJA, Davoust J, Briand JP, Moorman AFM, Jongsma HJ.
Restricted distribution of connexin40, a gap junctional protein, in mammalian heart. Circ Res. 1994;74:839 – 851.
1935
16. Morley GE, Vaidya D, Samie FH, Lo C, Delmar M, Jalife J. Characterization of conduction in the ventricles of normal and heterozygous Cx43
knockout mice using optical mapping. J Cardiovasc Electrophysiol.
1999;10:1361–1375.
17. Jongsma HJ, Wilders R. Gap junctions in cardiovascular disease. Circ
Res. 2000;86:1193–1197.
18. Gutstein DE, Morley GE, Tamaddon H, Vaidya D, Schneider MD, Chen
J, Chien KR, Stuhlmann H, Fishman GI. Conduction slowing and sudden
arrhythmic death in mice with cardiac-restricted inactivation of connexin43. Circ Res. 2001;88:333–339.
19. Gutstein DE, Morley GE, Vaidya D, Liu F, Chen FL, Stuhlmann H,
Fishman GI. Heterogeneous expression of gap junction channels in the
heart leads to conduction defects and ventricular dysfunction. Circulation. 2001;104:1194 –1199.
20. Schalij MJ, Boersma L, Huijberts M, Allessie MA. Anisotropic reentry in
a perfused 2-dimensional layer of rabbit ventricular myocardium. Circulation. 2000;102:2650 –2658.
21. Koura T, Hara M, Takeuchi S, Ota K, Okada Y, Miyoshi S, Watanabe A,
Shiraiwa K, Mitamura H, Kodama I, Ogawa S. Anisotropic conduction
properties in canine atria analyzed by high-resolution optical mapping:
preferential direction of conduction block changes from longitudinal to
transverse with increasing age. Circulation. 2002;105:2092–2098.
22. Kucera JP, Kléber AG, Rohr S. Slow conduction in cardiac tissue, II:
effects of branching tissue geometry. Circ Res. 1998;83:795– 805.
23. Kawara T, Derksen R, de Groot JR, Coronel R, Tasseron RT, Linnenbank
AC, Hauer RNW, Kirkels H, Janse MJ, de Bakker JMT. Activation delay
after premature stimulation in chronically diseased myocardium relates to
the architecture of interstitial fibrosis. Circulation. 2001;104:3069 –3075.
24. Lenègre J, Moreau P. Le bloc auriculo-ventriculaire chronique: Étude
anatomique, clinique et histologique. Arch Mal Coeur Vaiss. 1963;56:
867– 888.
25. Tukkie R, Sogaard P, Vleugels J, de Groot IK, Wilde AA, Tan HL. Delay
in right ventricular activation contributes to Brugada syndrome. Circulation. 2004;109:1272–1277.
26. Coronel R, De Groot JR, Veldkamp MW, Casini S, Koopmann TT,
Bhuiyan Z, Bezzina CR, Brugada P, De Bakker JMT. Electrophysiologic,
genetic and histopathologic characterization of the explanted heart of a
patient with Brugada syndrome. Heart Rhythm. 2004;1:S6.
27. Olson TM, Michels VV, Ballew JD, Reyna SP, Karst ML, Herron KJ,
Horton SC, Rodeheffer RJ, Anderson JL. Sodium channel mutations and
susceptibility to heart failure and atrial fibrillation. JAMA. 2005;293:
447– 454.
CLINICAL PERSPECTIVE
The cardiac electrical system is characterized by redundancy and substantial safety margins. A 50% reduction in sodium
channel expression in ventricular myocardium slows conduction only marginally. Similarly, a reduction in connexin
expression alone has limited effect on conduction. Thus, the heart has solid conduction reserve. A disease process that
affects only 1 factor may not necessarily compromise conduction of the electrical impulse. However, cardiac diseases that
produce electrical remodeling usually alter not only ion channels but also expression and distribution of gap junction
channels that affect cell-to-cell coupling. In addition, aging is accompanied by reduced cellular coupling. These principles
are well demonstrated in mice that are genetically engineered to have reduced cardiac sodium channels. The young animals
have reduced sodium channels and relatively preserved conduction. With aging, the animals develop fibrosis and
diminished cellular coupling, which is accompanied by a marked slowing of conduction. This model demonstrates how a
cardiac ion channel abnormality can have little effect during youth but can become significant as fibrosis and cellular
uncoupling develop with age or additional electrical remodeling. These findings imply that the clinical effect of a genetic
or pathological process may be reduced by therapies that prevent or reduce structural remodeling and fibrosis.
Functional Roles of Cav1.3(␣1D) Calcium Channels in Atria
Insights Gained From Gene-Targeted Null Mutant Mice
Zhao Zhang, MD, PhD; Yuxia He, MD; Dipika Tuteja, PhD; Danyan Xu, MD;
Valeriy Timofeyev, PhD; Qian Zhang, MD; Kathryn A. Glatter, MD; Yanfang Xu, MD, PhD;
Hee-Sup Shin, PhD; Reginald Low, MD; Nipavan Chiamvimonvat, MD
Background—Previous data suggest that L-type Ca2⫹ channels containing the Cav1.3(␣1D) subunit are expressed mainly in
neurons and neuroendocrine cells, whereas those containing the Cav1.2(␣1C) subunit are found in the brain, vascular
smooth muscle, and cardiac tissue. However, our previous report as well as others have shown that Cav1.3 Ca2⫹
channel– deficient mice (Cav1.3⫺/⫺) demonstrate sinus bradycardia with a prolonged PR interval. In the present study,
we extended our study to examine the role of the Cav1.3(␣1D) Ca2⫹ channel in the atria of Cav1.3⫺/⫺ mice.
Methods and Results—We obtained new evidence to demonstrate that there is significant expression of Cav1.3 Ca2⫹
channels predominantly in the atria compared with ventricular tissues. Whole-cell L-type Ca2⫹ currents (ICa,L) recorded
from single, isolated atrial myocytes from Cav1.3⫺/⫺ mice showed a significant depolarizing shift in voltage-dependent
activation. In contrast, there were no significant differences in the ICa,L recorded from ventricular myocytes from
wild-type and null mutant mice. We previously documented the hyperpolarizing shift in the voltage-dependent
activation of Cav1.3 compared with Cav1.2 Ca2⫹ channel subunits in a heterologous expression system. The lack of
Cav1.3 Ca2⫹ channels in null mutant mice would result in a depolarizing shift in the voltage-dependent activation of ICa,L
in atrial myocytes. In addition, the Cav1.3-null mutant mice showed evidence of atrial arrhythmias, with inducible atrial
flutter and fibrillation. We further confirmed the isoform-specific differential expression of Cav1.3 versus Cav1.2 by in
situ hybridization and immunofluorescence confocal microscopy.
Conclusions—Using gene-targeted deletion of the Cav1.3 Ca2⫹ channel, we established the differential distribution of
Cav1.3 Ca2⫹ channels in atrial myocytes compared with ventricles. Our data represent the first report demonstrating
important functional roles for Cav1.3 Ca2⫹ channel in atrial tissues. (Circulation. 2005;112:1936-1944.)
Key Words: arrhythmias 䡲 ion channels 䡲 atrial fibrillation 䡲 calcium 䡲 atrium
oltage-gated Ca2⫹ channels are heteromultimeric complexes of a pore-forming, transmembrane-spanning ␣1subunit, a disulfide-linked complex of ␣2- and ␦-subunits, and
an intracellular ␤- and ␥-subunit.1,2 The ␣1-subunit is the
largest and incorporates the conduction pore, the voltage
sensor, gating apparatus, and the known sites of channel
regulation by second messengers, drugs, and toxins. Mammalian ␣1-subunits of voltage-gated Ca2⫹ channels are encoded by at least 10 distinct genes.3 Previous data suggest that
L-type Ca2⫹ channels (LTCCs) containing the Cav1.3(␣1D)
subunit (D-LTCC) are expressed mainly in neurons and
neuroendocrine cells, whereas those containing the
Cav1.2(␣1C) subunit (C-LTCCs) are found in the brain, vascular smooth muscle, and cardiac tissue. Recently, we and
others, have shown that the Cav1.3 Ca2⫹ channel is highly
expressed in cardiac pacemaking tissue and plays an impor-
V
tant role in the spontaneous diastolic depolarization and
frequency of beating in sinoatrial (SA) node cells.4 – 6 Specifically, using a mouse model of gene-targeted deletion of
Cav1.3 Ca2⫹ channel, we established a role for the Cav1.3 Ca2⫹
channel in the generation of spontaneous action potential in
SA node cells.4 The Cav1.3-null mutant mouse shows evidence of profound SA and atrioventricular (AV) node dysfunction. We observed that D-LTCCs show a low activation
threshold compared with that of C-LTCCs. The hyperpolarizing shift in the activation threshold of the Cav1.3 Ca2⫹
channel can be directly documented in isolated SA node cells
as well as in a nonexcitable expression system, wherein the
Cav1.3 subunit can be expressed and studied alone.4 This
gating property of D-LTCCs contributes importantly to the
generation of spontaneous action potential and pacemaking
activities within SA node cells.
Received January 31, 2005; revision received June 15, 2005; accepted June 27, 2005.
From the Division of Cardiovascular Medicine (Z.Z., X.H., D.T., D.X., V.T., Q.Z., K.A.G., Y.X., R.L., N.C.), Department of Internal Medicine,
University of California, Davis; the Department of Veterans Affairs (N.C.), Northern California Health Care System, Mather, Calif; the Center for
Calcium and Learning (H.-S.S.), Division of Life Sciences, Korea Institute of Science and Technology, Seoul, Korea; and the Department of Physiology
(Z.Z.), Henan Medical University, Zhingzhou, China.
Correspondence to Nipavan Chiamvimonvat, Division of Cardiovascular Medicine, University of California, Davis, One Shields Ave, GBSF 6315,
Davis, CA 95616. E-mail [email protected]
1936
Zhang et al
In the present report, we present new evidence that demonstrates that there is significant expression of Cav1.3 Ca2⫹
channels in atrial but not in ventricular tissue. Specifically,
using in situ hybridization and immunocytochemistry, we
show that there is robust expression of Cav1.3 Ca2⫹ channels
in mouse atrial but not ventricular tissue. Because both
isoforms have similar pharmacological properties, it is difficult to isolate one current from the other with conventional
electrophysiology. The Cav1.3-null mutant mouse model
provides a unique opportunity to directly determine the
contribution of D- versus C-LTCCs in atrial versus ventricular tissues. Here, using in vitro and in vivo electrophysiological recordings, we document for the first time the functional roles of the Cav1.3 Ca2⫹ channel in mouse atrial
myocytes.
Methods
Animals and Protocols
The present investigation conformed to the Guide for the Care and
Use of Laboratory Animals published by the US National Institutes
of Health (NIH publication 85-23, revised 1985) and was performed
in accordance with the guidelines of the Animal Care and Use
Committee of the University of California, Davis. Generation of
Cav1.3-null mutant mice has been previously described.4,7
Electrophysiological Recordings
Single atrial and ventricular myocytes were isolated from Cav1.3⫺/⫺,
Cav1.3⫹/⫺, and wild-type (WT, Cav1.3⫹/⫹) littermates on a C57BL/6J
background, as previously described.8 Whole-cell ICa was recorded at
room temperature with patch-clamp techniques.9,10 The external
solution contained (in mmol/L) N-methyl glucamine (NMG) 140,
CsCl 5, MgCl2 0.5, CaCl2 2, 4-amino pyridine 2, glucose 10, and
HEPES 10, and the internal solution contained NMG 135, tetraethylammonium chloride 20, disodium ATP 4, EGTA 1, and HEPES
10. All chemicals were purchased from Sigma Chemical unless
stated otherwise. Cell capacitance was calculated by integrating the
area under a curve of an uncompensated capacitative transient
elicited by a 20-mV hyperpolarizing pulse from a holding potential
of ⫺40 mV. Whole-cell current records were filtered at 2 kHz and
sampled at 10 kHz. Liquid junction potentials were measured as
previously described,11 and all data were corrected for liquid
junction potentials. Curve fitting and data analysis were performed
with Origin software (MicroCal Inc).
Reverse Transcription–Polymerase Chain Reaction
Total RNA was prepared from the atria and ventricles of wild-type
(WT) C57BL/6J mice with TRIzol Reagent (Invitrogen). cDNA was
synthesized from total RNA samples by oligo(dT)-primed reverse
transcription (RT) (Superscript II RNase H-reverse transcriptase,
Invitrogen). cDNA was then subjected to polymerase chain reaction
(PCR) amplification with HotStarTaq DNA polymerase (Qiagen).
Primers used in the PCR were designed from mutually unique
regions of Cav1.2 and Cav1.3 channels as follows: (1) for Cav1.3,
5⬘-ATGAACCTTCCGACATTTTC-3⬘ (forward) and 5⬘GTGCTCATAGTCTGGGCGGC-3⬘ (reverse), according to the published sequence of mouse Cav1.3 (accession No. NM_028981) and
(2) for Cav1.2, 5⬘-ATGGTCAATGAAAACACGA-3⬘ (forward) and
5⬘- ACTGACGGTAGAGATGGTTG-3⬘ (reverse), according to the
published sequence of mouse Cav1.2 (accession No. NM_009781).
The absence of genomic contamination in the RNA samples was
confirmed by RT-negative controls for each experiment.
In Situ Hybridization
Cav1.2- and Cav1.3-specific cDNA fragments were subcloned into a
TA cloning vector (Invitrogen). All clones were sequenced. The
sense and antisense riboprobes were synthesized in the presence of
Cav1.3(␣1D) Ca2ⴙ Channel in Atria
1937
UTP-digoxigenin label with use of a DIG RNA labeling kit (Roche).
Mouse hearts procured from 8- to 10-week-old WT mice were
dissected and perfused first with Rnase-free phosphate-buffered
saline and later with 4% paraformaldehyde (made in phosphatebuffered saline). Cav1.3⫺/⫺ mouse hearts were also used as negative
controls for the Cav1.3 probe. Perfused hearts were fixed overnight at
4°C in 4% paraformaldehyde. Infiltration of hearts was done with a
mixture of 10% and 30% sucrose in ratios of 2:1, 1:1, and 1:2 at
room temperature for 30 minutes (each) with gentle rotation. Hearts
were transferred to 30% sucrose and allowed to settle at 4°C. Hearts
were then transferred to degassed OCT medium (Tissue-Tek) and
maintained at 4°C overnight with rotation. Embedding was done in
OCT medium, and the samples were frozen on a dry ice/ethanol bath.
Cryosectioning was completed, and sections were laid on gelatincoated slides (Fisher Scientific). After air-drying, in situ hybridization was performed with the anti-sense as well as the corresponding
sense cRNA probes on adjacent sections. After hybridization and
washes, the sections were subjected to immunologic detection with
anti-digoxigenin Fab fragments conjugated to alkaline phosphatase
by using a DIG nucleic acid detection kit (Roche). The signals were
developed with nitro blue tetrazolium and bromochloroindolyl phosphate (Roche) added in alkaline phosphatase buffer in the presence
of levamisole (Sigma) to inhibit endogenous alkaline phosphatase.
The specimens were inspected for development of purple precipitate
by bright-field microscopy (Carl Zeiss Vision). Digitized images
were obtained with AxioVision 4 (Zeiss).
Immunofluorescence Confocal Microscopy
Immunofluorescence labeling was performed as described previously.12 The following primary antibodies were used: (1) anti-Cav1.3
(Santa Cruz Biotechnology, Inc), a polyclonal antibody raised in goat
against a purified peptide corresponding to amino acid residues 859
to 875 of rat Cav1.3 (accession No. P27732)13 and (2) anti-Cav1.2
(Alomone Labs), a polyclonal antibody raised in rabbit against a
glutathione-S-transferase fusion protein with residues 1 to 46 of
rabbit Cav1.2 (accession No. P15381).14 The cells were treated with
anti-Cav1.3 or anti-Cav1.2 antibodies (1:200 dilution for 1 hour).
Immunofluorescence labeling for confocal microscopy was performed by treatment with Texas red– conjugated goat anti-rabbit
antibody or rabbit anti-goat antibody (Calbiochem, 1:500 dilution).
Immunofluorescence-labeled samples were examined with a Pascal
Zeiss confocal laser scanning microscope. Control experiments
performed by incubation with secondary antibody only did not show
positive staining under the same experimental conditions. Identical
settings were used for all specimens.
Transient Transfection of Cav1.2 and Cav1.3 Ca2ⴙ
Channels in HEK Cells
HEK 293 cells were maintained in Dulbecco’s modified Eagle’s
medium containing 10% fetal bovine serum, 2 mmol/L L-glutamine,
and 1% penicillin/streptomycin (Invitrogen) and kept at 37°C in a
5% CO2 incubator. Cells were transiently transfected with the
calcium phosphate precipitation procedure (Invitrogen) as described
previously.4 Channel subunits to be studied were subcloned into
pGW1H, an expression vector with a cytomegalovirus promoter
(British Biotechnology). Cells were transiently transfected with 7.5
␮g of plasmid containing the Cav1.3 Ca2⫹ channel (a gift from Dr S.
Seino, Kobe University, Kobe, Japan) or the Cav1.2 Ca2⫹ channel
and coexpressed with 5 ␮g of plasmid containing the gene that
encodes a ␤1A-subunit (derived from skeletal muscle).
In Vivo Electrophysiological Studies in Mice
In vivo electrophysiological studies were performed as previously
described.15 Standard pacing protocols were used to determine the
electrophysiological parameters, including sinus node recovery time;
atrial, AV nodal, and ventricular refractory periods; and AV nodal
conduction properties. Each animal underwent an identical pacing
and programmed stimulation protocol. The Q-T interval was determined manually by placing cursors on the beginning of the QRS and
the end of the T wave. The rate-corrected QT interval was calculated
1938
Circulation
September 27, 2005
with a modified Bazett’s formula as reported by Mitchell et al,16
whereby the RR interval was first expressed as a unitless ratio (RR
in ms/100 ms). The rate-corrected QT interval was defined as QT
interval (in ms)/(RR/100)1/2.
To induce atrial and ventricular tachycardia and fibrillation,
programmed extrastimulation techniques and burst pacing were
used. Programmed right atrial and right ventricular double and triple
extrastimulation techniques were performed at a 100-ms drive cycle
length, down to a minimum coupling interval of 10 ms. Right atrial
and right ventricular burst pacing was performed as eight 50-ms and
four 30-ms cycle-length train episodes repeated several times, up to
a maximum 1-minute time limit of total stimulation. For comparison
of inducibility, programmed extrastimulation techniques and stimulation duration of atrial and ventricular burst pacing were the same in
all mice. Sustained atrial or ventricular arrhythmias were defined as
atrial arrhythmias lasting ⬎30 seconds. Reproducibility was defined
as ⬎1 episode of induced atrial or ventricular tachycardia.
Statistics
Data are presented as mean⫾SEM. Comparison among the 3
genotypes was performed with SigmaStat with ANOVA and pairwise multiple comparison procedures (Holm-Sidak method). We
analyzed each litter separately and did not find significant outliers
from the different litters. Because each litter was too small to allow
for statistical analysis, we combined the data from all litters for the
final statistical analysis. Comparison of the occurrence of atrial
arrhythmias was performed with Fisher’s exact test.
immunofluorescence confocal microscopy study of single,
isolated, mouse atrial and ventricular myocytes. The specificity of the antibodies and the lack of cross reactivity at the
dilutions used for the 2 different isoforms of the Ca2⫹
channels were first tested in expressed Cav1.2 and Cav1.3
Ca2⫹ channels in HEK 293 cells (Figure 2A, 2B, 2D, and 2E)
compared with nontransfected cells (Figure 2G and 2H).
Figure 2C and 2F represent negative controls treated with
secondary antibodies only.
Figure 3A and 3C shows specific labeling with anti-Cav1.2
antibody in isolated mouse atrial and ventricular myocytes,
respectively. In contrast, only atrial myocytes show specific
labeling with the anti-Cav1.3 antibody (Figure 3B). Only a
low level of staining with the anti-Cav1.3 antibody was
observed in ventricular myocytes (Figure 3D). We further
documented the specificity of the anti-Cav1.3 antibody in
isolated atrial myocytes from Cav1.3⫺/⫺ mutant mice (Figure
3E and 3F). Whereas atrial myocytes from Cav1.3⫺/⫺ mutant
mice showed specific labeling with the anti-Cav1.2 antibody
(Figure 3E), no staining was observed with the anti-Cav1.3
antibody (Figure 3F). Figure 3G shows additional images of
negative controls with secondary antibody only.
Functional Roles of the Cav1.3 Ca2ⴙ Channel in
the Heart Assessed in Cav1.3ⴚ/ⴚ Mutant Mice
Results
Cav1.3 Transcripts Are Highly Expressed in
Mouse Atria Compared With Ventricles
We directly probed for the existence of the Cav1.3 Ca2⫹
channel in mouse cardiac myocytes by RT-PCR. Figure 1A
shows representative RT-PCR–amplified products with primers specific for Cav1.2 versus Cav1.3 Ca2⫹ channels and
primers specific for glyceraldehyde 3-phosphate dehydrogenase as a positive control from total RNA from mouse right
atria, left atria, right ventricles, left ventricles, septum, and
brain. Primers designed from mutually unique regions of
Cav1.2 and Cav1.3 channels in the N-termini are shown in
Figure 1B. Whereas Cav1.2 transcripts are expressed throughout the different regions in atria and ventricles, Cav1.3
transcripts are highly expressed in the atria compared with the
ventricles. The signals obtained from the right and left
ventricles and interventricular septum were very low compared with the atria.
To further confirm that the RT-PCR products generated
with primers specific to the Cav1.3 Ca2⫹ channel were indeed
amplified from cardiac myocytes and not other cell types in
the cardiac homogenate (eg, vascular smooth muscle cells),
we further generated sense and antisense riboprobes in the
presence of UTP-digoxigenin label for in situ hybridization.
Figure 1C is a photomicrograph comparing the distribution of
Cav1.2 versus Cav1.3 transcripts in mouse atria and ventricles.
Whereas Cav1.2 transcripts are present in both the atria and
ventricles, Cav1.3 transcripts are present mainly in the atria.
Sense riboprobes were used as negative controls from consecutive sections (labeled as sense).
2ⴙ
Immunodetection of Cav1.2 Versus Cav1.3 Ca
Channels in Dissociated Mouse Atrial and
Ventricular Myocytes
To further examine the regional distribution of the Cav1.2 and
Cav1.3 Ca2⫹ channels at the protein level, we performed an
Our data on the regional localization of the Cav1.3 Ca2⫹
channel transcript and protein with in situ hybridization and
immunofluorescence confocal microscopy are consistent with
expression of the Cav1.3 Ca2⫹ channel mainly in the atria.
However, the functional roles of the differential expression of
the Cav1.3 Ca2⫹ channel are unknown. Because Cav1.2 and
Cav1.3 Ca2⫹ channels have similar pharmacological properties, we reasoned that Cav1.3⫺/⫺ mutant mice would be an
ideal model to study the functional role of Cav1.3 Ca2⫹
channels in the atria. We undertook in vivo electrophysiological studies comparing mutant mice with heterozygous and
WT animals. All mutant mice showed evidence of SA and
AV nodes dysfunction, as assessed by sinus cycle length,
sinus node recovery time, PR interval, and Wenckebach cycle
length (see the Table). Furthermore, atrial arrhythmias,
mainly atrial fibrillation, were induced in all mutant mice and
a small number of heterozygous littermates. In contrast, atrial
arrhythmias were induced in none of the WT littermates
(P⬍0.01 comparing WT and mutant animals by Fisher’s
exact test). Indeed, previous studies of the same background
mouse model have shown that WT mice are not inducible for
atrial arrhythmias in the absence of carbachol.17 Figure 4
shows examples of atrial fibrillation and atrial flutter that
were induced in a Cav1.3⫺/⫺-null mutant mouse. In contrast,
ventricular arrhythmias were not induced in either the WT,
heterozygous or the homozygous mutant mice. The in vivo
electrophysiological parameters are summarized in the Table.
Whole-Cell ICa,L Recorded From Cav1.3ⴚ/ⴚ Atrial
and Ventricular Myocytes Compared With Those
From WT Littermates
To further corroborate the findings from the in vivo functional studies described earlier, we directly recorded ICa,L from
atrial and ventricular myocytes from Cav1.3⫺/⫺ and compared
Zhang et al
1939
Figure 1. A, Representative agarose gels of RT-PCR–amplified products with the use of primers specific for Cav1.2 vs Cav1.3 Ca2⫹
channels and primers specific for glyceraldehyde 3-phosphate dehydrogenase as a positive control from total RNA from mouse RA
(right atria), LA (left atria), RV (right ventricles), LV (left ventricles), S (septa), and B (brains). ⫺ve refers to a negative control (PCRamplified product without RT to ensure that there was no genomic contamination of the RNA samples). Lane 1 is the HI-LO DNA markers (Bioscience, Inc). B, Nucleotide sequence alignment (ClustalW) of the N-termini of mouse Cav1.2 and Cav1.3 Ca2⫹ channels. Highlighted regions refer to the primers used for the PCRs to generate the sense and antisense riboprobes. *Conserved nucleotide between
the 2 isoforms. Nucleotide sequence numbers are given on the right. C, Photomicrographs comparing the distribution of Cav1.2 vs
Cav1.3 transcripts in mouse atria and ventricles. Sections obtained with the corresponding sense riboprobes are shown to the right as
the negative controls. RA and LV refer to right atrium and left ventricle, respectively.
1940
Circulation
September 27, 2005
Figure 2. Confocal photomicrographs of HEK 293 cells expressing Cav1.2 (A–C) vs Cav1.3 (D–F) Ca2⫹ channels. Anti-Cav1.2 (A and D)
and anti-Cav1.3 (B and E) antibodies were used. Immunofluorescence labeling was done by treatment with secondary antibodies (Texas
red– conjugated anti-rabbit antibodies). C and F show samples treated with secondary antibodies only as negative controls. G and H
included nontransfected cells treated with anti-Cav1.2 vs anti-Cav1.3, respectively. Corresponding differential interference contrast (DIC)
images are shown in the right panels. The scale bar is 20 ␮m.
them with those of heterozygous and WT littermates. Wholecell ICa,L was recorded at a holding potential of ⫺55 mV.
Figure 5A shows examples of ICa,L current traces elicited at
the various step potentials from atrial myocytes isolated from
Cav1.3⫹/⫹ and Cav1.3⫹/⫺ littermates compared with Cav1.3⫺/⫺.
The current-voltage relations are summarized in Figure 5B.
Even though there were no significant differences in current
density among the 3 different groups of animals, there was a
depolarizing shift in the voltage-dependent activation of the
current when we compared Cav1.3⫹/⫹ to Cav1.3⫹/⫺ and
Cav1.3⫹/⫺ with Cav1.3⫺/⫺ mice. We further confirmed this
initial impression by generating activation curves from WT,
heterozygous, and mutant animals (Figure 5C). ICa,L recorded
from Cav1.3⫹/⫺ atrial myocytes showed an ⬇5-mV depolarizing shift at the midpoint of activation compared with WT
animals, whereas current from Cav1.3⫺/⫺ mice showed a
further depolarizing shift of ⬇7 mV compared with the
heterozygous animals. Figure 5D shows data obtained with a
2-pulse protocol to examine the voltage- and Ca2⫹-dependent
inactivation of ICa,L in WT, heterozygous, and mutant animals.
The curves appear nearly superimposed, with no significant
differences in the half-inactivation voltages. In addition, the
curves show the typical U-shape configuration for Ca2⫹dependent inactivation of L-type Ca2⫹ current. Typical traces
elicited with the test pulse are shown in the insert. Prepulses
more positive than ⫹20 mV elicited a progressively smaller
inward current as the command voltages approach the reversal potential, leading to partial recovery of the L-type Ca2⫹
current elicited with the test pulse, owing to a decrease in
Ca2⫹-dependent inactivation. There were no significant differences in voltage dependence of the inactivation profile
among the 3 groups of animals.
Figure 6 shows the same set of experiments obtained from
free-wall left ventricular myocytes, comparing the 3 groups
of animals. In contrast with the data obtained from atrial
myocytes, there were no significant differences in ICa,L recorded from the left ventricular myocytes among the 3 groups
of animals. These functional data are consistent with our in
situ hybridization and immunofluorescence studies showing
that the Cav1.3 transcript and protein are present predominantly in atrial tissues.
Previous data provide important clues that the differences
in biophysical properties of Cav1.2 versus Cav1.3 Ca2⫹ channels may be directly responsible for the observed findings in
the atria.18,19 ICa,L recorded from atrial myocytes isolated from
Cav1.3⫺/⫺ mutant animals were activated at more depolarizing
potentials compared with those from Cav1.3⫹/⫹ or Cav1.3⫹/⫺,
which expressed both Cav1.2 and Cav1.3 Ca2⫹ channels
(Figure 5C). Indeed, we have previously documented in a
heterologous expression system that there is a significant
Zhang et al
1941
Figure 3. Confocal photomicrographs from freshly isolated WT mouse atrial (A and B) and ventricular (C and D) myocytes treated with
anti-Cav1.2 (A and C) vs anti-Cav1.3 (B and D) antibodies. Immunofluorescence labeling was done by treatment with secondary antibodies (Texas red– conjugated antibodies). E and F represent mouse atrial myocytes isolated from Cav1.3⫺/⫺ mutant mice. G, Mouse
ventricular myocytes isolated from WT mice treated with secondary antibodies only as negative controls. Corresponding DIC images
are shown in the right panels. The scale bar is 20 ␮m.
depolarizing shift in steady-state activation in Cav1.2 compared with Cav1.3 Ca2⫹ currents, consistent with findings in
the Cav1.3⫺/⫺ mice, which express only the Cav1.2 subunit.4
Discussion
In this study, we directly tested the role of the Cav1.3 Ca2⫹
channel in atrial myocytes in Cav1.3-null mutant mice.
In Vivo Electrophysiological Studies in Cav1.3ⴚ/ⴚ and Cav1.3ⴙ/ⴚ
Mice Compared With WT Littermates
Sinus cycle length
PR interval
QTc interval
Sinus node recovery time
Wenckebach cycle length
AV node ERP
Atrial ERP
Ventricular ERP
Atrial arrhythmias
Ventricular arrhythmias
Cav1.3⫹/⫹
(n⫽9)
Cav1.3⫹/⫺
(n⫽11)
Cav1.3⫺/⫺
(n⫽8)
149.3⫾5.6
44.4⫾1.5
32.6⫾1.3
201.9⫾17
85.6⫾2.4
74.4⫾2.9
23.2⫾2.4
33.3⫾2.9
0/9
0/9
138.6⫾8.8
42.0⫾2.5
36.0⫾1.8
205.8⫾14.9
85.0⫾3.5
69.1⫾2.5
28.8⫾3.0
37.8⫾4.9
2/11
0/11
287.9⫾38.1*
61.0⫾4.7*
33.2⫾2.1
329.6⫾37.7†
123.8⫾9.1*
85.0⫾3.3†
21.8⫾2.3
35.0⫾3.9
8/8*
0/8
Data shown are mean⫾SEM. Measurement of AV node, atrial, and
ventricular effective refractory periods (ERPs) were performed with a basic
cycle length of 100 ms. n refers to the No. of animals in the studies.
*P⬍0.01, †P⬍0.05, Cav1.3⫺/⫺ vs Cav1.3⫹/⫺ and Cav1.3⫹/⫹.
Whole-cell ICa,L recordings from atrial myocytes isolated from
the null mutant mice showed a depolarizing shift in the
voltage-dependent activation compared with WT. In contrast,
there were no significant differences in whole-cell ICa,L
recorded from ventricular myocytes from WT or null mutant
mice. Consistent with these findings, we previously documented the hyperpolarizing shift in voltage-dependent activation of Cav1.3 compared with Cav1.2 Ca2⫹ channel subunits
in a heterologous expression system.4 The lack of Cav1.3 Ca2⫹
channels in the null mutant mice would result in a depolarizing shift in the voltage-dependent activation of ICa,L in atrial
myocytes. We further confirmed the isoform-specific differential expression of Cav1.3 in atrial myocytes by in situ
hybridization and immunofluorescence confocal microscopy.
Voltage-Gated Ca2ⴙ Channel Subtypes in the Heart
The molecular basis for ICa in the heart has previously been
investigated. By in situ hybridization, it was found that the
most prominently expressed low-voltage activated Ca2⫹ channel in the SA node was Cav3.1(␣1G), whereas Cav3.2(␣1H) is
present at moderate levels.20 In addition, we and others have
previously documented the critical role of Cav1.3 in the SA
node by using mutant mouse models.4 – 6 The dominant
high-voltage activated Ca2⫹ channel was Cav1.2, whereas
only a small amount of Cav1.3 mRNA was detected in SA
node myocytes of mice.20 The existence the Cav1.3 Ca2⫹
1942
Circulation
September 27, 2005
Figure 4. In vivo electrophysiological studies in Cav1.3-null mutant mice, showing evidence of inducible atrial fibrillation (A) and atrial
flutter (B) after atrial extrastimuli. Upper tracings are surface ECG (leads I and II). Lower traces are intracardiac electrograms showing
atrial and ventricular electrograms, with induced sustained rapid atrial fibrillation and atrial flutter with relatively slow ventricular
response.
channel in different regions of the heart has been further
documented in a recent study by Marionneau et al21 and is
consistent with our findings: Cav1.3 was found to be more
prominently expressed in the atria compared with the ventricles. However, the functional roles of Cav1.3 in the atria or
ventricular tissues have never been documented.
Role of Cav1.3 Ca2ⴙ Channels in Atrial Myocytes
Here, using gene-targeted deletion of the Cav1.3 isoform, we
were able to document that genetic ablation of the Cav1.3
isoform results in the occurrence of atrial arrhythmias. Indeed, this mutant mouse model represents one of the few
genetic models of atrial fibrillation. Even though Cav1.3
represents only a small amount of the LTCC transcript in the
atria, owing to the significant differences in biophysical
properties of the Cav1.3 isoform, the channel contributes
significantly to the overall function in the atria. Our in vivo
electrophysiological data as well as patch-clamp recordings
are consistent with the notion that Cav1.3 LTCCs are expressed and contribute functionally to atrial cardiac myocytes
in contrast to ventricular myocytes.
Atrial Fibrillation
Atrial fibrillation is the most common clinical arrhythmia and
is associated with a significant increase in morbidity and
mortality.22 The underlying mechanisms of atrial fibrillation
are very heterogeneous and are often related to underlying
heart or pulmonary diseases. However, more recently, several
studies have identified mutations in ion channels as possible
causes of inherited atrial fibrillation.23–26 The first gene for an
inherited form of atrial fibrillation was identified in a family
with autosomal-dominant transmission.24 A mutation was
found in the K⫹ channel gene KCNQ1, resulting in a
gain-of-function mutation. This is in contrast with the reduction in current density seen with mutations in other residues
in this gene causing long QT syndrome type 1. The gain-offunction mutation is consistent with the decrease in action
potential duration and effective refractory period, which are
thought to be the mechanisms of atrial fibrillation. On the
other hand, a number of patients with the mutation also had a
prolonged QT interval,24 emphasizing the fact that our understanding of repolarization is incomplete. In addition, recent
data also suggest a role for genetic modifiers, or incompletely
penetrant disease genes, as mechanisms for the development
of atrial fibrillation. Our data showing the development of
atrial fibrillation in null mutant mice suggest an important
functional role for Cav1.3 in the atria. Ablation of the Cav1.3
Ca2⫹ channel did not alter the atrial effective refractory period
but might nonetheless alter atrial action potential duration or
Ca2⫹-activated repolarizing currents. Direct measurements of
action potential duration are required to establish this. Additional studies are also needed to further examine the effects of
the Cav1.3 Ca2⫹ channel on Ca2⫹ transients. Finally, the
relevance of this model to human atrial fibrillation remains
only speculative at this time.
Compensatory Changes in Mutant Mice
Because the relative contribution of the Cav1.3 Ca2⫹ channel
to total ICa,L in atrial myocytes is unknown, we directly
compared the maximum ICa,L density between mutant mice
and their WT littermates (Figure 5B). The current was
normalized to cell capacity. Cell capacitance of single,
isolated, atrial myocytes from the 3 groups of animals was
53.9⫾2.2, 41.9⫾2.4, and 52.3⫾5.0 pF for Cav1.3⫹/⫹,
Cav1.3⫹/⫺, and Cav1.3⫺/⫺ mice, respectively (n⫽8, P⫽NS).
There was a ⬇12-mV depolarizating shift in the peak ICa,L in
the Cav1.3⫺/⫺ mice compared with their WT littermates;
however, the current density was not significantly different
between the WT and homozygous mutant animals. This may
represent a compensatory change, with upregulation of
Cav1.2 ICa,L in the mutant animals.
In summary, using gene-targeted deletion of the Cav1.3
Ca2⫹ channel, we established the important functional roles of
Cav1.3 Ca2⫹ channels in atrial myocytes in addition to its
Zhang et al
1943
multiple Ca2⫹ channel subtypes appears to be important in
coordinating the different physiological functions in atrial
myocytes in addition to cardiac pacemaking cells. Taken
together, our data represent the first report on the functional
roles for Cav1.3 Ca2⫹ channels in atrial cardiomyocytes.
Importantly, the differential expression of the 2 different
isoforms of the LTCC, with predominant expression of the
Cav1.3 channel in the atria compared with the ventricles, may
offer a unique therapeutic opportunity to directly modify the
atrial cells without interfering with ventricular myocytes.
Acknowledgments
This study was supported by NIH/NHLBI grants (RO1, HL67737,
and HL75274) and the Nora Ecceles Treadwell Foundation Award
Figure 5. A, Example of whole-cell ICa,L recorded from a holding
potential of ⫺55 mV elicited at various step potentials (⫹10,
⫹20, and ⫹30 mV) from atrial myocytes isolated from Cav1.3⫹/⫹
and littermates Cav1.3⫹/⫺ compared with Cav1.3⫺/⫺ mice. The
test potentials used are shown to the left of the current traces.
The current-voltage relations are summarized in B (n⫽10 for
each group). C, Voltage-dependent activation curves showing
the normalized conductances (g/gmax) from WT, heterozygous,
and mutant animals. The solid lines represent fits to the Boltzmann function yielding half-activation voltages (V1/2) of
0.95⫾0.40, 6.30⫾0.40, and 13.75⫾0.38 mV for Cav1.3⫹/⫹,
Cav1.3⫹/⫺, Cav1.3⫺/⫺, and respectively (P⬍0.05 comparing
Cav1.3⫺/⫺ with Cav1.3⫹/⫹, Cav1.3⫺/⫺ with Cav1.3⫹/⫺, and Cav1.3⫹/⫺
with Cav1.3⫹/⫹) and slope factors of 6.8, 7.2, and 7.6 mV
(P⫽NS) for Cav1.3⫹/⫹, Cav1.3⫹/⫺, and Cav1.3⫺/⫺, respectively
(n⫽9 for each group). D, Voltage-dependent inactivation according to a 2-pulse protocol to examine the voltage- and Ca2⫹dependent inactivation of ICa,L in WT, heterozygous, and mutant
animals. Normalized current traces at a test potential of ⫹20
mV, comparing ICa,L recorded from Cav1.3⫹/⫹, Cav1.3⫹/⫺, and
Cav1.3⫺/⫺ atrial myocytes. Examples of traces obtained from the
test pulse are shown in the insert. Prepulses more positive than
⫹20 mV elicited a progressively smaller inward current as the
command voltages approach the reversal potential, leading to
partial recovery of the LTCC elicited with the test pulse owing to
a decrease in the Ca2⫹-dependent inactivation.
previously documented role in pacemaking cells. The hyperpolarizing shift in the activation threshold of the Cav1.3 Ca2⫹
channel can be directly documented by gene-targeted deletion
in the Cav1.3 mutant mouse model. Our in vivo electrophysiological data support important roles for Cav1.3 in atrial
myocytes; the Cav1.2 subunit cannot functionally substitute
for the Cav1.3 subunit. Important phenotypes of atrial fibrillation and atrial flutter were documented in the null mutant
mice. Similar to that in neuronal systems, the expression of
Figure 6. A, Example of whole-cell ICa,L recorded from a holding
potential of ⫺55 mV in the same set of experiments as in Figure
5 from free-wall left ventricular myocytes, comparing the 3
groups of animals. In contrast to the data obtained from the
atrial myocytes, there were no significant differences in ICa,L recorded from the left ventricular myocytes among the 3 groups of
animals. The current-voltage relations are summarized in B
(n⫽10 for each group). C shows the corresponding activation
curves and the normalized conductances (g/gmax) from
Cav1.3⫹/⫹, Cav1.3⫹/⫺, and Cav1.3⫺/⫺ ventricular myocytes. The
solid lines represent fits to the Boltzmann function yielding halfactivation voltages (V1/2) of 9.22⫾0.42, 9.41⫾0.60, and 9.5⫾0.48
mV (P⫽NS) and slope factors of 5.9⫾0.38, 6.1⫾0.54, and
6.4⫾0.43 mV for Cav1.3⫹/⫹, Cav1.3⫹/⫺, and Cav1.3⫺/⫺, respectively (P⫽NS, n⫽9 to 11 for each group). D, Voltage-dependent
inactivation according to a 2-pulse protocol to examine the
voltage- and Ca2⫹-dependent inactivation of ICa,L in WT, heterozygous, and mutant animals. Normalized current traces at a
test potential of ⫹20 mV, comparing ICa,L recorded from
Cav1.3⫹/⫹, Cav1.3⫹/⫺, and Cav1.3⫺/⫺ ventricular myocytes. Examples of traces obtained from the test pulse are shown in the
insert.
1944
Circulation
September 27, 2005
(Dr Chiamvimonvat). The authors thank Dr E.N. Yamoah for helpful
suggestions and comments and the UC, Davis, Health System
Confocal Microscopy Facility.
References
1. Catterall WA. Structure and function of voltage-gated ion channels. Annu
Rev Biochem. 1995;64:493–531.
2. Walker D, De Waard M. Subunit interaction sites in voltage-dependent
Ca2⫹ channels: role in channel function. Trends Neurosci. 1998;21:
148 –154.
3. Ertel EA, Campbell KP, Harpold MM, Hofmann F, Mori Y, Perez-Reyes
E, Schwartz A, Snutch TP, Tanabe T, Birnbaumer L, Tsien RW, Catterall
WA. Nomenclature of voltage-gated calcium channels. Neuron. 2000;25:
533–535.
4. Zhang Z, Xu Y, Song H, Rodriguez J, Tuteja D, Namkung Y, Shin HS,
Chiamvimonvat N. Functional roles of Cav1.3 (␣1D) calcium channel in
sinoatrial nodes: insight gained using gene-targeted null mutant mice.
Circ Res. 2002;90:981–987.
5. Platzer J, Engel J, Schrott-Fischer A, Stephan K, Bova S, Chen H, Zheng
H, Striessnig J. Congenital deafness and sinoatrial node dysfunction in
mice lacking class D L-type Ca2⫹ channels. Cell. 2000;102:89 –97.
6. Mangoni ME, Couette B, Bourinet E, Platzer J, Reimer D, Striessnig J,
Nargeot J. Functional role of L-type Cav1.3 Ca2⫹ channels in cardiac
pacemaker activity. Proc Natl Acad Sci U S A. 2003;100:5543–5548.
7. Namkung Y, Skrypnyk N, Jeong MJ, Lee T, Lee MS, Kim HL, Chin H,
Suh PG, Kim SS, Shin HS. Requirement for the L-type Ca2⫹ channel ␣1D
subunit in postnatal pancreatic ␤-cell generation. J Clin Invest. 2001;108:
1015–1022.
8. Xu Y, Dong PH, Zhang Z, Ahmmed GU, Chiamvimonvat N. Presence of
a calcium-activated chloride current in mouse ventricular myocytes.
Am J Physiol Heart Circ Physiol. 2002;283:H302–H314.
9. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved
patch-clamp techniques for high-resolution current recording from cells
and cell-free membrane patches. Pflugers Arch. 1981;391:85–100.
10. Chiamvimonvat N, O’Rourke B, Kamp TJ, Kallen RG, Hofmann F,
Flockerzi V, Marban E. Functional consequences of sulfhydryl modification in the pore-forming subunits of cardiovascular Ca2⫹ and Na⫹
channels. Circ Res. 1995;76:325–334.
11. Neher E. Correction for liquid junction potentials in patch clamp experiments. Methods Enzymol. 1992;207:123–131.
12. Xu Y, Tuteja D, Zhang Z, Xu D, Zhang Y, Rodriguez J, Nie L, Tuxson
HR, Young JN, Glatter KA, Vazquez AE, Yamoah EN, Chiamvimonvat
N. Molecular identification and functional roles of a Ca2⫹-activated K⫹
channel in human and mouse hearts. J Biol Chem. 2003;278:
49085– 49094.
13. Ihara Y, Yamada Y, Fujii Y, Gonoi T, Yano H, Yasuda K, Inagaki N,
Seino Y, Seino S. Molecular diversity and functional characterization of
voltage-dependent calcium channels (CACN4) expressed in pancreatic
␤-cells. Mol Endocrinol. 1995;9:121–130.
14. Mikami A, Imoto K, Tanabe T, Niidome T, Mori Y, Takeshima H,
Narumiya S, Numa S. Primary structure and functional expression of the
cardiac dihydropyridine-sensitive calcium channel. Nature. 1989;340:
230 –233.
15. Berul CI, Aronovitz MJ, Wang PJ, Mendelsohn ME. In vivo cardiac
electrophysiology studies in the mouse. Circulation. 1996;94:2641–2648.
16. Mitchell GF, Jeron A, Koren G. Measurement of heart rate and Q-T
interval in the conscious mouse. Am J Physiol. 1998;274:H747–H751.
17. Kovoor P, Wickman K, Maguire CT, Pu W, Gehrmann J, Berul CI,
Clapham DE. Evaluation of the role of IK,ACh in atrial fibrillation using a
mouse knockout model. J Am Coll Cardiol. 2001;37:2136 –2143.
18. Koschak A, Reimer D, Huber I, Grabner M, Glossmann H, Engel J,
Striessnig J. ␣1D (Cav1.3) subunits can form L-type Ca2⫹ channels
activating at negative voltages. J Biol Chem. 2001;276:22100 –22106.
19. Bell DC, Butcher AJ, Berrow NS, Page KM, Brust PF, Nesterova A,
Stauderman KA, Seabrook GR, Nurnberg B, Dolphin AC. Biophysical
properties, pharmacology, and modulation of human, neuronal L-type
(␣1D, CaV1.3) voltage-dependent calcium currents. J Neurophysiol. 2001;
85:816 – 827.
20. Bohn G, Moosmang S, Conrad H, Ludwig A, Hofmann F, Klugbauer N.
Expression of T- and L-type calcium channel mRNA in murine sinoatrial
node. FEBS Lett. 2000;481:73–76.
21. Marionneau C, Couette B, Liu J, Li H, Mangoni ME, Nargeot J, Lei M,
Escande D, Demolombe S. Specific pattern of ionic channel gene
expression associated with pacemaker activity in the mouse heart.
J Physiol. 2005;562:223–234.
22. Chugh SS, Blackshear JL, Shen WK, Hammill SC, Gersh BJ. Epidemiology and natural history of atrial fibrillation: clinical implications. J Am
Coll Cardiol. 2001;37:371–378.
23. Brugada R, Tapscott T, Czernuszewicz GZ, Marian AJ, Iglesias A, Mont
L, Brugada J, Girona J, Domingo A, Bachinski LL, Roberts R. Identification of a genetic locus for familial atrial fibrillation. N Engl J Med.
1997;336:905–911.
24. Chen YH, Xu SJ, Bendahhou S, Wang XL, Wang Y, Xu WY, Jin HW,
Sun H, Su XY, Zhuang QN, Yang YQ, Li YB, Liu Y, Xu HJ, Li XF, Ma
N, Mou CP, Chen Z, Barhanin J, Huang W. KCNQ1 gain-of-function
mutation in familial atrial fibrillation. Science. 2003;299:251–254.
25. Ellinor PT, Macrae CA. The genetics of atrial fibrillation. J Cardiovasc
Electrophysiol. 2003;14:1007–1009.
26. Hong K, Bjerregaard P, Gussak I, Brugada R. Short QT syndrome and
atrial fibrillation caused by mutation in KCNH2. J Cardiovasc Electrophysiol. 2005;16:394 –396.
Atrial fibrillation is the most common atrial arrhythmia affecting the American population and is associated with a
significant risk of embolism and stroke. The problem is further exacerbated by the fact that treatment strategies have proven
largely inadequate. An alteration in ion channel expression (electrical remodeling) has been implicated in the maintenance
of the arrhythmias. Importantly, a decrease in Ca2⫹ current density in atrial myocytes of patients with persistent atrial
fibrillation has been documented. Previous data suggest that L-type Ca2⫹ channels containing the Cav1.3(␣1D) subunit are
expressed mainly in neurons and neuroendocrine cells, whereas those containing the Cav1.2(␣1C) subunit are found in the
brain, vascular smooth muscle, and cardiac tissue. We and others have shown that Cav1.3 Ca2⫹ channel– deficient mice
(Cav1.3⫺/⫺) demonstrate sinus bradycardia with a prolonged PR interval. This study examined the role of the Cav1.3 (␣1D)
Ca2⫹ channel in the atria of Cav1.3⫺/⫺ mice. Here, we demonstrate that there is significant expression of Cav1.3 Ca2⫹
channels predominantly in atrial compared with ventricular tissues. In addition, Cav1.3-null mutant mice have an increased
susceptibility to atrial arrhythmias, indicated by inducible atrial flutter and fibrillation. The relevance of this model to
human atrial fibrillation is speculative at this time, but the important functional role of the Cav1.3 Ca2⫹ channel in atrial
tissues in this model warrants additional study to assess its role in humans.
Effect of Fish Oil on Heart Rate in Humans
A Meta-Analysis of Randomized Controlled Trials
Dariush Mozaffarian, MD, MPH; Anouk Geelen, PhD; Ingeborg A. Brouwer, PhD;
Johanna M. Geleijnse, PhD; Peter L. Zock, PhD; Martijn B. Katan, PhD
Background—The effect of fish oil on heart rate (HR), a major risk factor for sudden death, is not well established. We
calculated this effect in a meta-analysis of randomized, double-blind, placebo-controlled trials in humans.
Methods and Results—Randomized trials of fish oil that evaluated HR were identified through MEDLINE (1966 through
January 2005), hand-searching of references, and contact with investigators for unpublished results. Two investigators
independently extracted trial data. A pooled estimate was calculated from random-effects meta-analysis. Predefined
stratified meta-analyses and meta-regression were used to explore potential heterogeneity. Of 197 identified articles, 30
met inclusion criteria. Evidence for publication bias was not present. In the overall pooled estimate, fish oil decreased
HR by 1.6 bpm (95% CI, 0.6 to 2.5; P⫽0.002) compared with placebo. Between-trial heterogeneity was evident (Q test,
P⬍0.001). Fish oil reduced HR by 2.5 bpm (P⬍0.001) in trials with baseline HR ⱖ69 bpm (median) but had little effect
(0.04-bpm reduction; P⫽0.56) in trials with baseline HR ⬍69 bpm (P for interaction⫽0.03). Fish oil reduced HR by
2.5 bpm (P⬍0.001) in trials with duration ⱖ12 weeks but had less effect (0.7-bpm reduction; P⫽0.27) in trials with
duration ⬍12 weeks (P for interaction⫽0.07). HR reduction with fish oil intake did not significantly vary by fish oil
dose (range, 0.81 to 15 g/d), type of HR measure, population age, population health, parallel versus crossover design,
type of control oil, or study quality by Delphi criteria (P for interaction ⬎0.25 for each).
Conclusions—In randomized controlled trials in humans, fish oil reduces HR, particularly in those with higher baseline
HR or longer treatment duration. These findings provide firm evidence that fish oil consumption directly or indirectly
affects cardiac electrophysiology in humans. Potential mechanisms such as effects on the sinus node, ventricular
efficiency, or autonomic function deserve further investigation. (Circulation. 2005;112:1945-1952.)
Key Words: heart rate 䡲 fatty acids, omega-3 䡲 fish oil 䡲 meta-analysis 䡲 randomized controlled trials
F
atty fish and fish oil intake is associated with lower risk
of cardiac arrhythmias, including sudden death, arrhythmic coronary heart disease death, and atrial fibrillation.1– 8
Experimental studies in isolated rat myocytes, exercising
dogs, and nonhuman primates suggest that fish oil has direct
cardiac electrophysiological effects, including slowing of the
heart rate (HR).9 –11 However, such effects are not well
established in humans. Because higher HR is a major independent risk factor for cardiovascular death, particularly
sudden death,12–18 an effect of fish oil on HR would both
confirm an influence on cardiac electrophysiology in humans
and indicate a plausible potential mechanism for observed
relations between fish intake and arrhythmic events. We
therefore performed a meta-analysis of randomized placebocontrolled clinical trials to determine the effect of fish oil
consumption on HR in humans.
Methods
Selection of Randomized Trials
We followed the Quality of Reporting of Meta-Analyses
(QUOROM) standards19 during all phases of the design and implementation of this analysis. Randomized clinical trials of fish oil that
included evaluation of HR were identified through MEDLINE (1966
through February 2005), including fish oil trials designed primarily
to evaluate other outcomes such as blood pressure or coronary
restenosis,* hand-searching of reference lists of obtained articles,
and contacting investigators for unreported HR data in published
trials or for HR data from unpublished trials. To minimize publication bias, we attempted to identify all fish oil trials that may have
measured and reported HR data, rather than limiting our search to
trials designed primarily to evaluate HR. English-language trials in
human subjects ⬎18 years of age were included if oral fish oil
supplementation was randomized and changes in HR or baseline and
follow-up HR were measured; trials with organ transplant subjects,
cointerventions that could not be separated from fish oil treatment,
Received April 19, 2005; revision received July 4, 2005; accepted July 8, 2005.
From the Channing Laboratory (D.M.), Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, and the Departments
of Epidemiology and Nutrition (D.M.), Harvard School of Public Health, Boston, Mass, and Division of Human Nutrition (A.G., I.B., J.G., P.Z., M.K.),
Wageningen University, and Wageningen Centre for Food Sciences (A.G., I.B., P.Z., M.K.), Wageningen, the Netherlands.
Guest Editor for this article was Robert H. Eckel, MD.
*Medline search criteria included (heart rate or blood pressure or restenosis) and (fish oil or n-3 fatty acids or omega-3 or eicosapentaenoic or
docosahexaenoic). Limits were adults ⱖ19 years of age, English language, clinical trial, and humans.
Correspondence to Dr Dariush Mozaffarian, 665 Huntington Ave, Bldg 2, Room 315, Boston, MA 02115. E-mail [email protected]
1945
1946
Circulation
September 27, 2005
no placebo control, nonblinding of participants, or duration ⬍2
weeks were excluded.
Trial Review
When potentially relevant trials were identified, abstracts and, if
necessary, original articles were screened for obvious exclusions by
an investigator. Of 197 identified trials, 161 were excluded for not
being a randomized trial of fish oil (n⫽28), for having no available
HR data (n⫽75), for occurring in organ transplant recipients (n⫽12),
for having no placebo control (n⫽29), for having a cointervention
that could not be separated from fish oil treatment (n⫽6), or for
being a duplicate publication from the same study (n⫽11). The
identified trials included 10 published and 2 unpublished trials for
which we contacted the authors to determine whether unreported HR
data might be available. A list of all reviewed trials and reasons for
exclusion is available by request from the investigators. For the
remaining 36 trials not excluded during initial screening, each
original article was independently reviewed for inclusion by 2
investigators. Six of these trials were excluded for no placebo control
(n⫽2), no follow-up HR data (n⫽2), duration ⬍2 weeks (n⫽1), or
being a duplicate publication from the same study (n⫽1). Thirty
trials met inclusion and exclusion criteria, including 2 trials for
which unpublished HR data were obtained from the authors (personal communications, William Harris, February 18, 2005, and
Ingrid Toft, March 4, 2005).20 – 49 Concordance on inclusion and
exclusion decisions was 100%.
Data Extraction
For each of the articles meeting inclusion and exclusion criteria, data
were independently extracted by 2 investigators on study design;
population; sample size and dropout; fish oil type, dose, and
duration; method of HR assessment; change in HR or baseline and
follow-up HR values; and HR variance measures. For studies
reporting RR interval values (duration of 1 heartbeat in milliseconds), HR was calculated and its corresponding variance was
estimated proportionally to the RR interval variance. Study quality
was also independently assessed by 2 investigators according to the
criteria for quality assessment of randomized clinical trials developed by Delphi consensus.50 The 9 criteria (1a, 1b, and 2 through 8)
include, for example, whether a method of randomization was
performed, whether the treatment groups were similar at baseline
with regard to the most important prognostic indicators, and whether
the analysis was of intention-to-treat design. For the last criterion, we
considered analyses as having intention-to-treat design if all subjects
not lost to follow-up were analyzed according to their original
randomization group; exclusions were not made for noncompliance.
We assessed the validity of data extraction by comparing the
independently abstracted results for concordance, and any discrepancies were resolved by discussion and review of the original
manuscript by the 2 investigators who extracted the data or, if
necessary, a committee comprising all the investigators. When
necessary, missing information (type of control oil, mean age of
participants, etc) was obtained by direct contact with the original
authors. We attempted to minimize clinical heterogeneity by excluding studies in children, in organ transplant recipients, or with
duration ⬍2 weeks. Remaining clinical heterogeneity was evaluated
qualitatively by comparing the mean age, gender distribution, and
general health of the study populations; the doses and durations of
fish oil treatment; and the methods of HR assessment. Clinical
heterogeneity was assessed quantitatively in prespecified stratified
analyses (see Statistical Analysis).
Statistical Analysis
Our primary outcome was the change in HR resulting from fish oil
treatment. For parallel-design trials, the HR change from baseline to
study end in the control group was subtracted from the HR change
from baseline to study end in the treatment group. For crossover
design trials, the HR at the end of the control period was subtracted
from the HR at the end of the treatment period. Within-individual
changes were used when available; otherwise, group means were
used. SEs were abstracted or, if not reported, derived from SDs, CIs,
or probability values. The pooled variance for the net HR change
resulting from fish oil treatment was calculated as (1)
SE2T⫹SE2C⫺2(r)(SET)(SEC) for crossover design trials, where SET
and SEC are the SE of the treatment and control period HR values,
respectively, and r is the within-individual correlation between the
treatment and control period HR values, and (2) SE2TG⫹SE2CG for
parallel-design trials, where SETG and SECG are the SE of the HR
change from baseline to study end in the treatment and control
groups, respectively. For parallel-design trials that reported precision
of baseline and final HR values (n⫽18) rather than HR changes,
SETG and SECG were calculated according to the method of Follmann
et al,51 which involves making an assumption for the unreported
within-individual correlation between baseline and final HR values.
On the basis of measured correlations in fish oil trials (Anouk
Geelen, personal communication, January 27, 2005), the withinindividual correlation between HR values was estimated to be 0.60
for trials using a single HR measure, 0.80 for trials using the average
of multiple measures, and 0.85 for trials using a 24-hour measure,
with the higher correlations consistent with less random error in the
HR measurement. Sensitivity analyses were performed assuming a
within-individual correlation of 0.60 for all trials. Data for the
calculation of the change in HR and the variance of this change were
not missing from any trial.
Pooled estimates of the effect of fish oil on HR were calculated
through the use of random-effects meta-analysis, which accounts for
heterogeneity in treatment effects among trials, using the method of
DerSimonian and Laird52 with inverse-variance (SE) weighting.
Because some trials compared multiple intervention groups with a
single control group (n⫽7), we performed sensitivity analyses in
which separate pooled estimates and variances for the effect of fish
oil on HR were calculated using separate meta-analyses for each of
these trials; these trial-specific estimates then were used in a second
meta-analysis evaluating all trials. Heterogeneity between studies
was tested with the DerSimonian and Laird Q statistic.52,53 To assess
publication bias, a funnel plot of the treatment effect versus SE was
visually inspected.54 Potential publication bias was also evaluated
with the Begg adjusted-rank correlation test,55 a statistical analog of
the visual funnel graph, and the regression asymmetry test according
to the method of Egger et al.54
We performed predefined stratified meta-analyses to explore
potential heterogeneity by dose of eicosapentaenoic acid and docasohexaenoic acid (EPA⫹DHA) (at the median), duration of treatment (ⱖ12 weeks versus less), type of HR measure (single measure,
average of multiple resting measures, or 24-hour measure), baseline
HR (at the median), type of control oil (olive oil versus other),
population age (at the median), general health (healthy versus
otherwise), study design (parallel versus crossover), and study
quality (meeting at least 8 Delphi criteria versus fewer). We used
meta-regression to test for heterogeneity of the pooled treatment
effect by these factors,56 testing for significance of the stratifying
variable by using the Wald test in a mixed-effects meta-regression
model. We also performed sensitivity analyses excluding trials with
ⱖ20% dropout of randomized participants at baseline. All analyses
were performed with Stata version 8.2 (Stata Corp). Statistical
significance was defined as 2-tailed ␣ ⬍0.05.
Results
Overview of Trials
Of the 30 trials meeting inclusion and exclusion criteria, 6
had 2 separate intervention groups, and 1 had 3 separate
intervention groups, for a total of 38 intervention groups in
the 30 trials (Table 1). Although single-blind trials were
acceptable, all were double-blind trials. Eight were crossover
design trials, and 22 were parallel-design trials. Median study
size was 30 participants; in total, this meta-analysis included
1678 individuals treated with fish oil or placebo for 27 615
person-weeks. The mean ages of the study populations ranged
Mozaffarian et al
TABLE 1.
Effect of Fish Oil on Heart Rate: A Meta-Analysis
1947
Characteristics of the 38 Intervention Groups (30 Trials) Included in the Meta-Analysis
Study
Design
Mean Age,
y*
Male,
%
General
Health
Fish Oil,
n†
Control,
n†
EPA⫹DHA,
g/d‡
Duration,
wk
Control
Oil
HR
Measure
Dropout,
%
Delphi
Criteria§
Bairati et al,20 1992
Parallel
54
80
CAD
66
59
4.5
26
Olive
Single
39
9
Christensen et al,21
1999
Parallel
38
58
Healthy
20
20
1.7
12
Olive
24-h continuous
0
8
1999 (group 2)
Parallel
38
58
Healthy
20
20
5.9
12
Olive
24-h continuous
0
8
1998
Parallel
52
59
Renal failure
11
6
4.2
12
Olive
24-h continuous
41
8
1996
Parallel
No
data
No
data
CAD, EF ⬍40
26
23
4.3
12
Olive
24-h continuous
11
9
Conquer and Holub24
1999
Parallel
30
100
Healthy
9
10
3.0
6
n-6
Single
5
7
Deslypere,25 1992
Parallel
56
100
Healthy
15
14
1.0
52
Oleic
Multiple average
0
6
Deslypere,25 1992
(group 2)
Parallel
56
100
Healthy
15
14
1.9
52
Oleic
Multiple average
0
6
Deslypere,25 1992
(group 3)
Parallel
56
100
Healthy
14
14
2.9
52
Oleic
Multiple average
0
6
Dyerberget et al,26 2004
Parallel
39
100
Healthy
24
25
3.2
8
Palmitic
24-h continuous
10
8
Geelen et al,27 2003
Parallel
59
49
Healthy
39
35
1.3
12
Oleic
Multiple average
2
9
Geelen et al,28 2005
Parallel
64
60
Frequent PVCs
41
43
1.3
14
Oleic
24-h continuous
9
9
Gray et al,29 1996
Parallel
56
100
HTN
9
10
3.5
8
Corn
Multiple average
10
9
Grimsgaard et al,30
1998
Parallel
44
100
Healthy
72
77
3.8
7
Corn
Multiple average
4
9
Grimsgaard et al,30
1998 (group 2)
Parallel
44
100
Healthy
75
77
3.6
7
Corn
Multiple average
4
9
Landmark et al,31 1993 Crossover
42
100
HTN,
Hyperlipidemia
18
䡠䡠䡠
4.6
4
Olive
Single
0
9
Leaf et al,32 1994
Parallel
63
79
CAD
201
205
6.9
26
Corn
Single
26
9
Levinson et al,33 1990
Parallel
56
81
HTN
8
8
15.0
6
Palm, corn
Multiple average
6
9
McVeigh et al,34 1994
Crossover
53
80
NIDDM
20
䡠䡠䡠
3.0
6
Olive
Single
0
9
Mehta et al,35 1988
Crossover
63
100
CAD
8
5.5
4
No data
Single
0
9
Mills et al,36 1990
Parallel
23
100
Healthy
10
䡠䡠䡠
10
1.3
4
Safflower
Multiple average
9
7
Mills et al,37 1989
Parallel
28
100
Healthy
10
10
2.6
4
Olive
Single
0
7
Miyajima et al,38 2001
Crossover
45
100
HTN
17
4
Linoleic
Multiple average
0
9
Monahan et al,39 2004
Parallel
25
56
Healthy
9
䡠䡠䡠
9
2.7
5.0
4.3
Olive
Single
0
9
Mori et al,40 1999
Parallel
49
100
Overweight,
Hyperlipidemia
19
20
3.8
6
Olive
24-h ambulatory
5
9
Mori et al,40 1999
(group2)
Parallel
49
100
Overweight,
Hyperlipidemia
17
20
3.7
6
Olive
24-h ambulatory
5
9
Nestel et al,41 2002
Parallel
58
55
Hyperlipidemia
12
14
3.0
7
Olive
Single
7
9
Nestel et al,41 2002
(group2)
Parallel
58
55
Hyperlipidemia
12
14
2.8
7
Olive
Single
7
9
O’Keefe et al,42 2005
Crossover
68
100
CAD, EF ⬍40%
18
16
Corn, olive
1-h continuous
44
9
Parallel
56
80
CAD
5
䡠䡠䡠
5
0.8
Solomon et al,43 1990
4.6
12
Olive
Single
0
9
Stark and Holub,44 2004 Crossover
57
0
Healthy
32
2.8
4
Corn, soy
Multiple average
16
8
3.4
16
Corn
Single
10
8
9.0
6
Palm, cottonseed
Single
25
8
Toft et al,45 1995
䡠䡠䡠
39
Parallel
54
64
HTN
37
Crossover
54
63
CAD
6
Vandongen et al,47 1993
Parallel
46
100
Healthy
17
䡠䡠䡠
18
2.2
12
Olive, palm, safflower
Multiple average
13
5
Vandongen et al,47 1993
(group2)
Parallel
46
100
Healthy
16
18
4.3
12
Olive, palm, safflower
Multiple average
13
5
Vacek et al,46 1989
Wing et al,48 1990
Crossover
61
35
HTN
20
8
Olive
Multiple average
17
9
Parallel
61
76
NIDDM
17
䡠䡠䡠
16
4.5
Woodman et al,49 2002
3.8
6
Olive
24-h ambulatory
15
9
Woodman et al,49 2002
(group2)
Parallel
61
76
NIDDM
17
16
3.7
6
Olive
24-hour
ambulatory
15
9
CAD indicates coronary artery disease; EF, ejection fraction; PVC premature ventricular contractions; HTN, hypertension; and NIDDM, non–insulin-dependent
diabetes mellitus.
*When mean age was not specified, the median age or age range midpoint was used.
†Subjects who completed the trial (ie, after dropout).
‡For 2 studies, the dose of EPA and DHA was estimated as 80% of the n-3 polyunsaturated fatty acid dose.
§Number of Delphi criteria met of a total of 9 (1a, 1b, 2 through 8).
1948
Circulation
September 27, 2005
Figure 1. Funnel plot with pseudo–95% CIs of the 38 intervention groups included in the meta-analysis.
resting measures; and 11 used the average of ambulatory or
continuous monitoring. Twenty-five trials (30 intervention
groups) met at least 8 Delphi criteria for study quality; 5 trials
(8 intervention groups) met ⬍8.
Our broad search methods appeared to be successful in
minimizing the effect of publication bias. Among the 30
included trials, 12 reported HR findings in the abstract (7
reporting an effect, 5 reporting the absence of an effect); 10
reported HR findings in the results text but not the abstract (5
reporting an effect, 5 reporting the absence of an effect); 6
presented HR findings in a table only (all 6 showing no
significant effect); and 2 constituted unpublished results.
Little evidence for publication bias was present by visual
inspection of a funnel plot (Figure 1), Begg’s test (P⫽0.87),
or Egger’s test (P⫽0.69).
Effect of Fish Oil on HR
from 23 to 68 years (median, 54 years). Sixteen intervention
groups were made up of generally healthy populations; 22
comprised individuals with ⱖ1 underlying chronic condition.
The median EPA⫹DHA dose was 3.5 g/d (range, 0.81 to 15
g/d), and the median treatment duration was 8 weeks (range,
4 to 52 weeks). Thirteen intervention groups assessed HR
with a single resting measure; 14 used the average of 2 or 3
The individual trial results and the pooled estimate are
presented in Figure 2. In the overall pooled estimate, fish oil
decreased HR by 1.6 bpm (95% CI, 0.6 to 2.5; P⫽0.002)
compared with placebo. Exclusion of trials with ⱖ20%
dropout (n⫽5) had little effect on the pooled estimate, with
fish oil decreasing HR by 1.3 bpm (95% CI, 0.3 to 2.4;
P⫽0.009). Assuming a within-individual HR correlation of
Figure 2. Change in HR resulting from fish oil consumption. Shaded squares indicate the point estimate for each trial, with the size of
the square proportional to the contribution (inverse variance random effects weight) of the study to the overall estimate. The overall
pooled estimate and 95% CI are indicated by the dotted line and clear diamond, respectively.
Mozaffarian et al
TABLE 2. Effect of Fish Oil on HR According to Prespecified
Study Characteristics
Characteristic
Intervention
Groups, n
Effect of Fish Oil on
HR (95% CI)
30
⫺1.4 (⫺2.5–⫺0.3)
8
⫺2.3 (⫺4.0–⫺0.5)
P for
Interaction*
Design
Parallel
Crossover
0.54
Mean age, y†
⬍55
20
⫺1.3 (⫺2.8–0.2)
ⱖ55
17
⫺1.8 (⫺3.1–⫺0.5)
Generally healthy
16
⫺1.4 (⫺3.0–0.3)
Chronic condition‡
22
⫺1.6 (⫺2.7–⫺0.5)
No
30
⫺1.3 (⫺2.4–⫺0.2)
Yes
8
⫺2.7 (⫺4.8–⫺0.6)
0.61
Health
0.78
CAD§
0.26
Baseline HR, bpm
⬍69
19
⫺0.4 (⫺1.9–1.0)
ⱖ69
19
⫺2.5 (⫺3.5–⫺1.4)
⬍3.5
19
⫺1.4 (⫺2.8–0.0)
ⱖ3.5
19
⫺1.7 (⫺3.1–⫺0.3)
⬍12
22
⫺0.7 (⫺2.0–0.6)
ⱖ12
16
⫺2.5 (⫺4.0–⫺1.1)
Single
13
⫺0.8 (⫺2.6–1.0)
Average of 2 or 3
14
⫺1.4 (⫺3.2–0.4)
Ambulatory/continuous
11
⫺2.0 (⫺2.9–⫺1.1)
Olive
17
⫺1.7 (⫺2.9–⫺0.5)
Mixed/other
20
⫺1.4 (⫺2.7–⫺0.0)
ⱖ8
30
⫺1.4 (⫺2.3–⫺0.5)
⬍8
8
0.03
EPA⫹DHA, g/d
0.72
Duration, wk
0.07
HR measure
0.32
Control oil†
1949
receiving ⱖ12 weeks of fish oil treatment (P for interaction⫽0.07), among whom fish oil reduced HR by 2.5 bpm
(95% CI, 1.1 to 4.0; P⫽0.001). Although other differences
related to study characteristics were not statistically significant (Table 2), several findings were consistent with intuition;
eg, the effect of fish oil on HR appeared possibly greater with
increasing precision of the measurement method used (single
versus average of 2 or 3 measures versus ambulatory/
continuous), consistent with reduced measurement error reducing bias toward the null.
Little evidence was present for a dose-response effect.
Stratified at the median dose of fish oil (3.5 g/d), the
reduction in HR was not significantly different at higher
versus lower doses (each compared with placebo) (P for
interaction⫽0.72) (Table 2). Similarly, stratified into quartiles of fish oil dose, HR was reduced by 1.1 (95% CI, ⫺0.9
to 3.1), 1.8 (95% CI, ⫺0.1 to 3.6), 1.9 (95% CI, 0.1 to 3.8),
and 1.5 (95% CI, ⫺0.6 to 3.6) bpm in quartiles 1 through 4,
respectively, compared with placebo (P for ordinal interaction⫽0.72). Evaluated continuously, the dose of fish oil was
not a predictor of treatment effect (P⫽0.63), above and
beyond being on fish oil treatment (yes/no). In the 2 trials
with EPA⫹DHA doses ⱕ1 g/d, HR was reduced by 5.0 bpm
(95% CI, 2.3 to 7.7; P⬍0.001) compared with 1.4 bpm in the
trials with EPA⫹DHA doses ⬎1 g/d (95% CI, 0.4 to 2.3;
P⬍0.001).
When we evaluated different factors simultaneously in the
meta-regression model, there appeared to be potential independent heterogeneity related to both baseline HR (P for
interaction⫽0.04) and treatment duration (P for interaction⫽0.09). Among the 9 trials with mean baseline HR ⱖ69
bpm and treatment duration ⱖ12 weeks, fish oil reduced HR
by 2.9 bpm (95% CI, 1.5 to 4.4; P⬍0.001) compared with
placebo, without significant between-trial heterogeneity (Q
test, P⬎0.05).
0.74
Discussion
Delphi criteria
⫺1.9 (⫺5.6–1.8)
0.56
*Testing for significance of the stratifying variable by using the Wald test in
a mixed-effects meta-regression model.
†One trial was not included in this subgroup analysis because of missing
data on this covariate.
‡Such as coronary artery disease (CAD), diabetes mellitus, hyperlipidemia,
or hypertension.
§Secondary analysis; not prespecified.
0.60 for all trials also had little effect, with fish oil decreasing
HR by 1.5 bpm (95% CI, 0.5 to 2.5; P⫽0.003). The pooled
estimate was also similar in sensitivity analyses accounting
for multiple intervention groups in some trials, with fish oil
decreasing HR by 1.4 bpm (95% CI, 0.4 to 2.5; P⫽0.007).
Between-trial heterogeneity was evident (Q test,
P⬍0.001). We evaluated prespecified study characteristics to
explore reasons for potential heterogeneity (Table 2). The HR
reduction with fish oil consumption was greater in study
populations with a mean baseline HR ⱖ69 bpm (P for
interaction⫽0.03), among whom fish oil reduced HR by 2.5
bpm (95% CI, 1.4 to 3.5; P⬍0.001), and in study populations
In this meta-analysis of randomized, double-blind, placebocontrolled clinical trials, fish oil consumption reduced HR in
humans. Although the overall effect was modest (1.6-bpm
reduction), on a population level, even modest differences in
risk factors can have a significant impact on health. These
findings provide firm evidence for an effect of fish oil
consumption on cardiac electrophysiology in humans.
The regulation of HR is a complex physiological process,
with components related to vagal tone, sympathetic input,
responsiveness of the sinus node, and systolic and diastolic
left ventricular function. The decrease in HR with fish oil
consumption indicates that marine n-3 fatty acids influence at
least 1 of these parameters. The n-3 fatty acids are incorporated into myocyte membranes and may influence ion channel function9,10; this could directly alter the automaticity or
responsiveness of the sinus node. Fish oil also lowers blood
pressure in humans,57 possibly by reducing systemic vascular
resistance.58 In one observational study, such an effect was
apparent at dietary levels of fish intake.58 Such a decrease in
systemic vascular resistance would reduce left ventricular
afterload and improve diastolic function, which could indirectly reduce HR as a result of better ventricular efficiency.
1950
Circulation
September 27, 2005
Experimental studies in nonhuman primates support the
hypothesis that fish oil consumption improves left ventricular
efficiency.59,60 Intake of n-3 fatty acid may also improve
measures of HR variability,21–23,27 suggesting a potential
effect on autonomic tone. Our findings substantiate an electrophysiological effect of fish oil in humans and support the
need for further investigation of these potential mechanisms.
Higher HR is associated with increased cardiovascular risk,
including greater risk of sudden death,12–15,17 coronary heart
disease death,13,14 and cardiovascular death.16 A higher HR
could directly increase cardiovascular risk, eg, by increasing
myocardial vulnerability to ischemia or arrhythmia. On the
basis of work by Jouven et al,17 our finding of a 1.6-bpm HR
reduction with fish oil consumption would correspond to an
⬇5% lower risk of sudden death. Thus, in addition to effects
on HR, other mechanisms are likely to contribute to the
reductions in sudden death risk with fish or fish oil consumption seen in observational studies and randomized trials. A
higher HR may indicate less optimal underlying cardiovascular health as manifested by increased sympathetic tone,
decreased vagal tone, or decreased ventricular efficiency. The
HR reduction with fish oil consumption could therefore
indicate beneficial effects of fish oil on these other physiological parameters that might reduce cardiovascular risk to a
greater extent than that resulting from the change in HR
alone.
Our exploration of heterogeneity revealed several interesting findings. First, the reduction in HR appeared larger in
trials with longer duration of intake (ⱖ12 weeks). This may
relate in part to the time needed for EPA and DHA to be
incorporated into the tissues where they exert their effects and
suggests that regular consumption over time may have greater
effects than short-term intake. Second, HR was reduced to a
greater extent in populations with higher baseline HR. Because fish oil was compared with placebo in each trial, this
result would not be due to regression toward the mean. This
finding suggests that fish oil may have greater effects on HR
in populations with higher intrinsic sinus node automaticity,
greater sympathetic tone, lower vagal tone, or lower ventricular efficiency. Third, although power was insufficient to
prove equivalence of different doses, very high consumption
of fish oil did not appear to have substantially greater effects
than modest consumption. This is consistent with observational studies and randomized trials indicating clinical benefits of fatty fish or fish oil consumption at relatively modest
intake, ⬇1 to 2 servings per week or 500 to 1000 mg/d
EPA⫹DHA, respectively.1– 8 In the present meta-analysis, the
lowest EPA⫹DHA doses were ⬇1 g/d, and it is possible that
a dose-response effect may exist at lower (eg, dietary) levels
of intake, as suggested by one observational analysis.58
Finally, although the differences were not statistically significant, the HR reduction was smaller in trials using a single
resting measure of HR, intermediate in trials using the
average of 2 or 3 resting measures, and greatest in trials using
ambulatory or continuous measures. This is consistent with a
greater degree of misclassification (random measurement error) when only a single or a few resting measures were used,
suggesting that such trials may underestimate the true effect
of fish oil on HR. Alternatively, the results of trials using
ambulatory and continuous monitoring represent the averaged effect of fish oil consumption on both resting and
activity-related HR responses, which may be somewhat
greater than effects on resting HR alone.
Publication bias is a major potential limitation of metaanalyses. Our broad, prespecified search methods and contacting of investigators for unpublished results appeared to be
successful in minimizing the effect of publication bias; in
only a minority of included trials was a significant HR effect
prominently reported, and little evidence was present for
publication bias in the final included studies. Additionally,
given the large number of included trials, it is unlikely that
the results of even several additional studies would greatly
alter the pooled estimate. Between-trial heterogeneity may
limit the generalizability of the overall pooled estimate; we
attempted to account for potential heterogeneity by using a
random-effects model and by assessing factors that may
explain between-trial differences.
In this meta-analysis of randomized, double-blind,
placebo-controlled clinical trials, fish oil reduced HR, particularly with higher baseline HR or longer durations of treatment. These results provide strong evidence that fish oil
consumption directly or indirectly influences cardiac electrophysiology in humans. This effect may directly account for
part of the observed benefits of fish intake on cardiovascular
risk, particularly risk of arrhythmic events, and may indicate
favorable effects on physiological systems such as on autonomic tone, vascular resistance, or ventricular efficiency that
improve cardiovascular health.
Acknowledgments
The Wageningen Centre for Food Sciences is an alliance of major
Dutch food industries, Maastricht University, TNO Nutrition and
Food Research in Zeist, and Wageningen University and Research
Centre, with financial support by the Dutch Government. Dr Mozaffarian was supported by a Mentored Clinical Scientist Award from
the National Heart, Lung, and Blood Institute, National Institutes of
Health (K08-HL-075628) and thanks Drs Eric Rimm, David Siscovick, and David Herrington for their invaluable guidance and
support. The authors thank Drs William Harris and Ingrid Toft for
sharing unpublished results for this analysis.
References
1. Burr ML, Fehily AM, Gilbert JF, Rogers S, Holliday RM, Sweetnam PM,
Elwood PC, Deadman NM. Effects of changes in fat, fish, and fibre
intakes on death and myocardial reinfarction: Diet and Reinfarction Trial.
Lancet. 1989;2:757–761.
2. Siscovick DS, Raghunathan TE, King I, Weinmann S, Wicklund KG,
Albright J, Bovbjerg V, Arbogast P, Smith H, Kushi LH, et al. Dietary
intake and cell membrane levels of long-chain n-3 polyunsaturated fatty
acids and risk of primary cardiac arrest. JAMA. 1995;274:1363–1367.
3. Albert CM, Hennekens CH, O’Donnell CJ, Ajani UA, Carey VJ, Willett
WC, Ruskin JN, Manson JE. Fish consumption and risk of sudden cardiac
death. JAMA. 1998;279:23–28.
4. Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto miocardico. Dietary supplementation with n-3 polyunsaturated fatty acids and
vitamin E after myocardial infarction: GISSI-Prevenzione trial. Lancet.
1999;354:447– 455.
5. Hu FB, Bronner L, Willett WC, Stampfer MJ, Rexrode KM, Albert CM,
Hunter D, Manson JE. Fish and omega-3 fatty acid intake and risk of
coronary heart disease in women. JAMA. 2002;287:1815–1821.
6. Mozaffarian D, Lemaitre RN, Kuller LH, Burke GL, Tracy RP, Siscovick
DS. Cardiac benefits of fish consumption may depend on the type of fish
Mozaffarian et al
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
meal consumed: the Cardiovascular Health Study. Circulation. 2003;107:
1372–1377.
Mozaffarian D, Ascherio A, Hu FB, Stampfer MJ, Willett WC, Siscovick
DS, Rimm EB. Interplay between different polyunsaturated fatty acids
and risk of coronary heart disease in men. Circulation. 2005;111:
157–164.
Mozaffarian D, Psaty BM, Rimm EB, Lemaitre RN, Burke GL, Lyles
MF, Lefkowitz D, Siscovick DS. Fish intake and risk of incident atrial
fibrillation. Circulation. 2004;110:368 –373.
McLennan PL Myocardial membrane fatty acids and the antiarrhythmic
actions of dietary fish oil in animal models. Lipids. 2001;36:111S–114S.
Leaf A, Kang JX, Xiao YF, Billman GE. Clinical prevention of sudden
cardiac death by n-3 polyunsaturated fatty acids and mechanism of
prevention of arrhythmias by n-3 fish oils. Circulation. 2003;107:
2646 –2652.
Billman GE, Kang JX, Leaf A. Prevention of ischemia-induced cardiac
sudden death by n-3 polyunsaturated fatty acids in dogs. Lipids. 1997;
32:1161–1168.
Dyer AR, Persky V, Stamler J, Paul O, Shekelle RB, Berkson DM,
Lepper M, Schoenberger JA, Lindberg HA. Heart rate as a prognostic
factor for coronary heart disease and mortality: findings in three Chicago
epidemiologic studies. Am J Epidemiol. 1980;112:736 –749.
Kannel WB, Kannel C, Paffenbarger RS Jr, Cupples LA. Heart rate and
cardiovascular mortality: the Framingham study. Am Heart J. 1987;113:
1489 –1494.
Shaper AG, Wannamethee G, Macfarlane PW, Walker M. Heart rate,
ischaemic heart disease, and sudden cardiac death in middle-aged British
men. Br Heart J. 1993;70:49 –55.
Wannamethee G, Shaper AG, Macfarlane PW, Walker M. Risk factors
for sudden cardiac death in middle-aged British men. Circulation. 1995;
91:1749 –1756.
Palatini P, Casiglia E, Julius S, Pessina AC. High heart rate: a risk factor
for cardiovascular death in elderly men. Arch Intern Med. 1999;159:
585–592.
Jouven X, Zureik M, Desnos M, Guerot C, Ducimetiere P. Resting heart
rate as a predictive risk factor for sudden death in middle-aged men.
Cardiovasc Res. 2001;50:373–378.
Palatini P, Thijs L, Staessen JA, Fagard RH, Bulpitt CJ, Clement DL, de
Leeuw PW, Jaaskivi M, Leonetti G, Nachev C, O’Brien ET, Parati G,
Rodicio JL, Roman E, Sarti C, Tuomilehto J, for the Systolic Hypertension in Europe (Syst-Eur) Trial Investigators. Predictive value of clinic
and ambulatory heart rate for mortality in elderly subjects with systolic
hypertension. Arch Intern Med. 2002;162:2313–2321.
Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF.
Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement: Quality of Reporting of MetaAnalyses. Lancet. 1999;354:1896 –1900.
Bairati I, Roy L, Meyer F. Effects of a fish oil supplement on blood
pressure and serum lipids in patients treated for coronary artery disease.
Can J Cardiol. 1992;8:41– 46.
Christensen JH, Christensen MS, Dyerberg J, Schmidt EB. Heart rate
variability and fatty acid content of blood cell membranes: a doseresponse study with n-3 fatty acids. Am J Clin Nutr. 1999;70:331–337.
Christensen JH, Aaroe J, Knudsen N, Dideriksen K, Kornerup HJ,
Dyerberg J, Schmidt EB. Heart rate variability and n-3 fatty acids in
patients with chronic renal failure: a pilot study. Clin Nephrol. 1998;49:
102–106.
Christensen JH, Gustenhoff P, Korup E, Aaroe J, Toft E, Moller J,
Rasmussen K, Dyerberg J, Schmidt EB. Effect of fish oil on heart rate
variability in survivors of myocardial infarction: a double blind randomised controlled trial. BMJ. 1996;312:677– 678.
Conquer JA, Holub BJ. Effect of supplementation with different doses of
DHA on the levels of circulating DHA as non-esterified fatty acid in
subjects of Asian Indian background. J Lipid Res. 1998;39:286 –292.
Deslypere JP. Influence of supplementation with N-3 fatty acids on
different coronary risk factors in men: a placebo controlled study. Verh K
Acad Geneeskd Belg. 1992;54:189 –216.
Dyerberg J, Eskesen DC, Andersen PW, Astrup A, Buemann B, Christensen JH, Clausen P, Rasmussen BF, Schmidt EB, Tholstrup T, Toft E,
Toubro S, Stender S. Effects of trans- and n-3 unsaturated fatty acids on
cardiovascular risk markers in healthy males: an 8 weeks dietary intervention study. Eur J Clin Nutr. 2004;58:1062–1070.
Geelen A, Zock PL, Swenne CA, Brouwer IA, Schouten EG, Katan MB.
Effect of n-3 fatty acids on heart rate variability and baroreflex sensitivity
in middle-aged subjects. Am Heart J. 2003;146:E4.
1951
28. Geelen A, Brouwer IA, Schouten EG, Maan AC, Katan MB, Zock PL.
Effects of n-3 fatty acids from fish on premature ventricular complexes
and heart rate in humans. Am J Clin Nutr. 2005;81:416 – 420.
29. Gray DR, Gozzip CG, Eastham JH, Kashyap ML. Fish oil as an adjuvant
in the treatment of hypertension. Pharmacotherapy. 1996;16:295–300.
30. Grimsgaard S, Bonaa KH, Hansen JB, Myhre ES. Effects of highly
purified eicosapentaenoic acid and docosahexaenoic acid on hemodynamics in humans. Am J Clin Nutr. 1998;68:52–59.
31. Landmark K, Thaulow E, Hysing J, Mundal HH, Eritsland J, Hjermann I.
Effects of fish oil, nifedipine and their combination on blood pressure and
lipids in primary hypertension. J Hum Hypertens. 1993;7:25–32.
32. Leaf A, Jorgensen MB, Jacobs AK, Cote G, Schoenfeld DA, Scheer J,
Weiner BH, Slack JD, Kellett MA, Raizner AE, et al. Do fish oils prevent
restenosis after coronary angioplasty? Circulation. 1994;90:2248 –2257.
33. Levinson PD, Iosiphidis AH, Saritelli AL, Herbert PN, Steiner M. Effects
of n-3 fatty acids in essential hypertension. Am J Hypertens. 1990;3:
754 –760.
34. McVeigh GE, Brennan GM, Cohn JN, Finkelstein SM, Hayes RJ,
Johnston GD. Fish oil improves arterial compliance in non-insulindependent diabetes mellitus. Arterioscler Thromb. 1994;14:1425–1429.
35. Mehta JL, Lopez LM, Lawson D, Wargovich TJ, Williams LL. Dietary
supplementation with omega-3 polyunsaturated fatty acids in patients
with stable coronary heart disease: effects on indices of platelet and
neutrophil function and exercise performance. Am J Med. 1988;84:45–52.
36. Mills DE, Mah M, Ward RP, Morris BL, Floras JS. Alteration of
baroreflex control of forearm vascular resistance by dietary fatty acids.
Am J Physiol. 1990;259:R1164 –R1171.
37. Mills DE, Prkachin KM, Harvey KA, Ward RP. Dietary fatty acid
supplementation alters stress reactivity and performance in man. J Hum
Hypertens. 1989;3:111–116.
38. Miyajima T, Tsujino T, Saito K, Yokoyama M. Effects of eicosapentaenoic acid on blood pressure, cell membrane fatty acids, and intracellular
sodium concentration in essential hypertension. Hypertens Res. 2001;24:
537–542.
39. Monahan KD, Wilson TE, Ray CA. Omega-3 fatty acid supplementation
augments sympathetic nerve activity responses to physiological stressors
in humans. Hypertension. 2004;44:732–738.
40. Mori TA, Bao DQ, Burke V, Puddey IB, Beilin LJ. Docosahexaenoic acid
but not eicosapentaenoic acid lowers ambulatory blood pressure and heart
rate in humans. Hypertension. 1999;34:253–260.
41. Nestel P, Shige H, Pomeroy S, Cehun M, Abbey M, Raederstorff D. The
n-3 fatty acids eicosapentaenoic acid and docosahexaenoic acid increase
systemic arterial compliance in humans. Am J Clin Nutr. 2002;76:
326 –330.
42. O’Keefe JH, Abulssa H, Sastre A, Steinhaus D, Harris W AHArecommended intakes of omega-3 fatty acids improve indicators of
cardiac autonomic tone but not lipids or inflammatory markers. Presented
at: American College of Cardiology Scientific Session; March 6 –9, 2005;
Orlando, Fla. Abstract.
43. Solomon SA, Cartwright I, Pockley G, Greaves M, Preston FE, Ramsay
LE, Waller PC. A placebo-controlled, double-blind study of eicosapentaenoic acid-rich fish oil in patients with stable angina pectoris. Curr Med
Res Opin. 1990;12:1–11.
44. Stark KD, Holub BJ. Differential eicosapentaenoic acid elevations and
altered cardiovascular disease risk factor responses after supplementation
with docosahexaenoic acid in postmenopausal women receiving and not
receiving hormone replacement therapy. Am J Clin Nutr. 2004;79:
765–773.
45. Toft I, Bonaa KH, Ingebretsen OC, Nordoy A, Jenssen T. Effects of n-3
polyunsaturated fatty acids on glucose homeostasis and blood pressure in
essential hypertension: a randomized, controlled trial. Ann Intern Med.
1995;123:911–918.
46. Vacek JL, Harris WS, Haffey K. Short-term effects of omega-3 fatty acids
on exercise stress test parameters, angina and lipoproteins. Biomed Pharmacother. 1989;43:375–379.
47. Vandongen R, Mori TA, Burke V, Beilin LJ, Morris J, Ritchie J. Effects
on blood pressure of omega 3 fats in subjects at increased risk of
cardiovascular disease. Hypertension. 1993;22:371–379.
48. Wing LM, Nestel PJ, Chalmers JP, Rouse I, West MJ, Bune AJ, Tonkin
AL, Russell AE. Lack of effect of fish oil supplementation on blood
pressure in treated hypertensives. J Hypertens. 1990;8:339 –343.
49. Woodman RJ, Mori TA, Burke V, Puddey IB, Watts GF, Beilin LJ.
Effects of purified eicosapentaenoic and docosahexaenoic acids on
glycemic control, blood pressure, and serum lipids in type 2 diabetic
patients with treated hypertension. Am J Clin Nutr. 2002;76:1007–1015.
1952
Circulation
September 27, 2005
50. Verhagen AP, de Vet HC, de Bie RA, Kessels AG, Boers M, Bouter LM,
Knipschild PG. The Delphi List: a criteria list for quality assessment of
randomized clinical trials for conducting systematic reviews developed
by Delphi consensus. J Clin Epidemiol. 1998;51:1235–1241.
51. Follmann D, Elliott P, Suh I, Cutler J. Variance imputation for overviews of
clinical trials with continuous response. J Clin Epidemiol. 1992;45:769–773.
52. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin
Trials. 1986;7:177–188.
53. Takkouche B, Cadarso-Suarez C, Spiegelman D. Evaluation of old and
new tests of heterogeneity in epidemiologic meta-analysis. Am J Epidemiol. 1999;150:206 –215.
54. Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis
detected by a simple, graphical test. Br Med J. 1997;315:629 – 634.
55. Begg CB, Mazumdar M. Operating characteristics of a rank correlation
test for publication bias. Biometrics. 1994;50:1088 –1101.
56. Stram DO. Meta-analysis of published data using a linear mixed-effects
model. Biometrics. 1996;52:536 –544.
57. Geleijnse JM, Giltay EJ, Grobbee DE, Donders AR, Kok FJ. Blood
pressure response to fish oil supplementation: metaregression analysis of
randomized trials. J Hypertens. 2002;20:1493–1499.
58. Mozaffarian D, Gottdiener JS, Siscovick DS. Fish intake and cardiac
structure, function, and hemodynamics: the Cardiovascular Health Study.
Am Heart J. In press.
59. Charnock JS, McLennan PL, Abeywardena MY. Dietary modulation of
lipid metabolism and mechanical performance of the heart. Mol Cell
Biochem. 1992;116:19 –25.
60. McLennan PL, Barnden LR, Bridle TM, Abeywardena MY, Charnock JS.
Dietary fat modulation of left ventricular ejection fraction in the
marmoset due to enhanced filling. Cardiovasc Res. 1992;26:871– 877.
Congenital Heart Disease
Sinus Venosus Atrial Septal Defect
Long-Term Postoperative Outcome for 115 Patients
Christine H. Attenhofer Jost, MD; Heidi M. Connolly, MD; Gordon K. Danielson, MD;
Kent R. Bailey, PhD; Hartzell V. Schaff, MD; Win-Kuang Shen, MD; Carole A. Warnes, MD;
James B. Seward, MD; Francisco J. Puga, MD; A. Jamil Tajik, MD
Background—Sinus venosus atrial septal defect (SVASD) differs from secundum atrial septal defect by its atrial septal
location and its association with anomalous pulmonary venous connection (APVC). Data on long-term outcome after
surgical repair are limited.
Methods and Results—We reviewed outcomes of 115 patients (mean age⫾SD 34⫾23 years) with SVASD who had repair
from 1972 through 1996. APVC was present in 112 patients (97%). Early mortality was 0.9%. Complete follow-up was
obtained for 108 patients (95%) at 144⫾99 months. Symptomatic improvement was noted in 83 patients (77%), and
deterioration was noted in 17 patients (16%). At follow-up, 7 (6%) of 108 patients had sinus node dysfunction, a
permanent pacemaker, or both, and 15 (14%) of 108 patients had atrial fibrillation. Older age at repair was predictive
of postoperative atrial fibrillation (P⫽0.033). No reoperations were required during follow-up. Survival was not
different from expected for an age- and sex-matched population. Clinical improvement was more common with older
age at surgery (P⫽0.014). Older age at repair (P⫽0.008) and preoperative New York Heart Association class III or IV
(P⫽0.038) were independent predictors of late mortality.
Conclusions—Operation for SVASD is associated with low morbidity and mortality, and postoperative subjective clinical
improvement occurs irrespective of age at surgery. Postoperative atrial fibrillation appears to be related to older age at
operation. SVASD repair achieves survival similar to that of a matched population and should be considered whenever
repair may impact survival or symptoms. (Circulation. 2005;112:1953-1958.)
Key Words: heart defects, congenital 䡲 heart septal defects 䡲 surgery 䡲 survival
S
inus venosus atrial septal defect (SVASD), originally
described in 1858, encompasses approximately 4% to
11% of atrial septal defects (ASDs).1,2 The typical malformation is an interatrial communication caused by a deficiency of the common wall between the superior vena cava
(SVC) and the right-sided pulmonary veins.2,3 SVASD is
commonly associated with anomalous pulmonary venous
connection (APVC) of some or all of the pulmonary
veins,3,4 which produces additional left-to-right shunting.
The basic principle of repair is redirection of the APVC
through the interatrial communication into the left atrium.
In contrast to operative repair of secundum ASD, the
surgical approach for SVASD is more complex and carries
the risk of stenosis of the SVC or pulmonary veins,
residual shunting, and sinoatrial node dysfunction (SND).4
The present study reviews outcomes for patients who
underwent repair of SVASD at Mayo Clinic (Rochester,
Minn) and focuses on patient survival and development of
arrhythmias.
Methods
Patients
We reviewed 131 consecutive patients who underwent surgical
repair of SVASD at Mayo Clinic between January 1972 and
December 1996; these patients comprised 4.0% of all 3277 patients
having an operation for ASD during this period. We excluded 16
patients with SVASD associated with severe congenital heart disease. Thus, the study cohort included 115 patients (mean age⫾SD
34⫾23 years; range 1.5 to 80 years). Four patients (3%) were older
than 70 years at operation, the oldest being 80 years. Six patients
(5%) were operated on when they were younger than 5 years.
Typical superiorly located SVASD was present in 109 patients, and
6 had an atypical inferior SVASD. Typical SVASD results from a
deficiency in the wall that normally separates the right pulmonary
veins from the superior vena cava and the right atrium.2,3 Rarely,
there is absence of only the posterior or inferior portions of the atrial
septum (or both), and 1 or more of the right pulmonary veins enters
the right atrium anterior to the atrial septum; this is called atypical
inferior SVASD in the present series of patients because it is not
typical SVASD.5 In 111 patients, the operation was the first attempt
at repair. In 4, the operation was for failed repair performed
elsewhere, including failure to divert 1 or more APVCs (n⫽3) and
recurrent SVASD (n⫽1). The study was approved by the Mayo
Received September 23, 2003; de novo received July 20, 2004; revision received June 2, 2005; accepted June 6, 2005.
From the Division of Cardiovascular Diseases (C.H.A.J., H.M.C., W.-K.S., C.A.W., J.B.S., A.J.T.), the Division of Cardiovascular Surgery (G.K.D.,
H.V.S., F.J.P.), and the Division of Biostatistics (K.R.B.), Mayo Clinic, Rochester, Minn.
Reprint requests to Heidi M. Connolly, MD, Division of Cardiovascular Diseases, Mayo Clinic, 200 First St SW, Rochester, MN 55905.
1953
1954
Circulation
September 27, 2005
pulmonary vein.6 – 8 In 6 patients (5%) who had atypical SVASD,
located more inferiorly and posteriorly in the right atrium, a
pericardial patch was sewn to the margins of the atrial septum and
anterior to the anomalous right pulmonary veins to divert their blood
flow to the left atrium.5 In 23 patients (20%), the atriotomy extended
to or crossed the cavo-atrial junction. The mean (⫾SD) cardiopulmonary bypass time was 68⫾28 minutes (range 25 to 160 minutes),
the mean aortic cross-clamp time was 35⫾16 minutes (range 1 to 79
minutes), and the mean duration of hospital stay was 8.9⫾0.8 days
(range 5 to 42 days).
ECG Findings
Figure 1. The most common type of surgical repair of SVASD in
the present series of patients involved placement of a single
pericardial patch without enlarging the superior vena cava. a,
The defect was exposed through the right atrium. b, The defect
was exposed through an incision in the right atrium that
extended across the cavoatrial junction into the superior vena
cava. C and D, After right atrial incision shown in B. The pericardial patch redirected the anomalous pulmonary venous flow
to the left atrium and closed the SVASD. AO indicates aorta;
CS, coronary sinus; FO, foramen ovale; IVC, inferior vena cava;
PT, pulmonary trunk; RLPV, right lower pulmonary vein (which
connects appropriately to the left atrium in this diagram); RMPV,
right middle pulmonary vein; and RUPV, right upper pulmonary
vein. Reproduced with permission from the Mayo Foundation for
Medical Education and Research.
Foundation Institutional Review Board in 1997 for retrospective
review; written consent was obtained from all patients.
All ECGs were reviewed, and SND was defined as persistent sinus
bradycardia (⬍50 bpm), ectopic atrial rhythm, junctional/nodal
rhythm, or a wandering pacemaker (⬍60 bpm); pauses of more than
3 seconds; or evidence of SND on an electrophysiological study.
New SND was defined as SND that was present at hospital dismissal
or at late follow-up when sinus rhythm was present preoperatively.
For patients with no SND at dismissal but no follow-up, the data
were not analyzed and were considered missing.
Follow-Up Data
The date of latest follow-up for retrospective review was July 1,
1997. Postoperative follow-up information was obtained for 108
(95%) of the surviving 114 early survivors from a subsequent visit to
Mayo Clinic before July 1, 1997 (n⫽60); phone call from the survey
research department (n⫽28); visits to other clinics before July 1,
1997 (n⫽20); and information from the National Death Index (n⫽6).
Six patients were lost to follow-up. The functional classification was
assessed by use of the New York Heart Association (NYHA)
classification. For the present analysis, clinical improvement was
defined as subjective improvement reported by the patient. Subjective improvement included any improvement in exercise capacity or
energy level or a decreased frequency of cardiovascular symptoms.
Echocardiographic Data
Echocardiographic data from our institution were available for 88
patients (77%) preoperatively and 67 patients (58%) postoperatively.
Catheterization
Preoperative cardiac catheterization was performed in 45 patients
(39%). Significant coronary artery disease was found in 2 of 27
patients undergoing coronary angiography. After 1990, preoperative
hemodynamic cardiac catheterization was necessary in only 7
patients, because in most patients, transthoracic or transesophageal
echocardiography provided sufficient information to proceed with
the operation. Total pulmonary vascular resistance and pulmonary
arteriolar resistance were calculated and expressed in Wood units.
Significant pulmonary hypertension was defined as pulmonary
vascular resistance of at least 5 U. The pulmonary-to-systemic flow
ratio was also calculated.
Surgical Techniques
Operations were performed through either a median sternotomy or a
right anterolateral thoracotomy. Cardiopulmonary bypass was used
with bicaval and ascending aortic cannulation at moderate systemic
hypothermia (25°C to 32°C) in all cases. Repairs diverted the APVC
through the SVASD into the left atrium. This diversion usually
required the use of a pericardial patch or baffle. To avoid narrowing
the SVC, the caval incision was often closed with a pericardial patch.
Several different surgical techniques were used in the present
series depending on the SVASD anatomy and the surgeon’s preference. In 76 patients (66%), the SVASD was closed with 1 patch
(n⫽74; 64%; Figure 1); this was the most commonly used technique
in this series. Other operative techniques included direct suture
closure (n⫽2; 2%) without enlarging the SVC.4 A 2-patch repair
technique was used in 27 patients (23%). Six patients had SVASD
repair in which a neopulmonary vein was constructed from the
cardiac end of the right SVC by dividing or ligating the cava between
the entrance of the azygos vein and the most superior anomalous
Discrete variables were summarized as percentages, and continuous
variables were summarized as mean⫾SD. Differences between the
characteristics of 3 patient age groups were tested for significance
with the ␹2 test or the Fisher exact test when appropriate for discrete
variables and with the 2-sample t test or rank sum test for continuous
variables. Trends in baseline characteristics with age were assessed
by the 2-sample t test (on age) for binary variables and the Spearman
rank correlation for continuous variables. Simple and multiple
logistic regression analyses were used to assess patient and surgical
factors related to postoperative improvement. Simple and multiple
Cox regression analysis was used to assess predictors of survival.
Survival after the date of operation was estimated with the KaplanMeier method and was compared with the age- and sex-matched
survival for the US population overall, as well as being stratified by
3 age groups: 40 years or younger, 41 to 60 years, and 61 years or
older. All tests of significance were 2-tailed, with P⬍0.05 assumed
to indicate significance. “Early death” was defined as occurring
within 30 days after SVASD repair or during the index hospitalization. Other outcomes, including postoperative improvement in dyspnea and atrial fibrillation, were compared between the 3 age groups.
The rates of these outcomes were compared between the 3 age
groups by the Pearson ␹2 test of independence.
Results
The study group included 61 women (53%). Preoperatively,
50 patients (43%) were in NYHA functional class I. The
diagnosis of ASD was suspected on the basis of a systolic
murmur in the pulmonary area in 111 patients (97%), cardiomegaly, or recurrent pulmonary infections. NYHA functional
class III or IV dyspnea was present in 20% of patients. Apart
from dyspnea, the following symptoms and history were
Jost et al
Follow-Up After SVASD Repair
TABLE 1.
Preoperative Findings for 115 Patients in 3 Age Groups*
Age at
Repair, y
No. of
Patients
Symptoms,
%†
Dyspnea,
%†
AF,
%†
SPAP,
mm Hg‡
PVR,
U§
PAR,
U储
Shunt
Ratio¶
ⱕ40
69
38
26
4
36⫾10
3.0⫾3.3
1.7⫾2.3
2.4⫾1.2
41 to 60
29
76
55
28
46⫾22
5.4⫾2.1
3.5⫾1.9
2.1⫾1.1
ⱖ61
17
100
94
53
52⫾18
6.3⫾4.5
4.1⫾2.7
2.6⫾0.9
1955
AF indicates atrial fibrillation or atrial flutter; SPAP, systolic pulmonary artery pressure; PVR, total pulmonary
vascular resistance; and PAR, pulmonary arteriolar resistance.
*Continuous data are expressed as mean⫾SD.
†P⬍0.0001.
‡Data available for 103 patients; P⫽0.0002.
§Data available for 45 patients; P⫽0.0005.
储Data available for 40 patients; P⫽0.002.
¶Data available for 45 patients; P⫽0.33.
reported: palpitations (31%), angina (19%), history of congestive heart failure (9%), and history of stroke (3%).
Preoperative hemodynamic data are shown in Table 1.
Systolic pulmonary artery pressure (n⫽103) was
⬎50 mm Hg in 21 patients (20%). Total pulmonary vascular
resistance (n⫽45) was ⬎5 U in 12 patients (27%) and ⬎8 U
in 6 of these patients (13%).
Forty-six patients (40%) were older than 40 years at the
time of repair; 17 (15%) were older than 60 years, and 4 (3%)
were older than 70 years. Patients older than 40 years at
operation were significantly more likely to have preoperative
dyspnea (P⬍0.0001), atrial fibrillation or flutter (P⬍0.0001),
and a higher pulmonary vascular resistance (P⬍0.0001) and
pulmonary arteriolar resistance (P⫽0.001).
Operations
The SVASD was classified according to the surgical findings
into the typical superior type (n⫽109) or the atypical inferior
type (n⫽6). In 112 patients, there was associated APVC, with
insertion of pulmonary veins into 1 or more of the following:
SVC, cavoatrial junction, and right atrium. The pulmonary
vein anatomy was not described for 4 patients.
The mean diameter of the SVASD intraoperatively was
22⫾11 mm (range 5 to 60 mm). A persistent left SVC to the
coronary sinus was found in 17 patients (15%). An associated
secundum ASD was present in 10 patients (9%), and a patent
foramen ovale was present in 20 (17%). In addition to
SVASD repair, the following surgical procedures were performed: CABG (n⫽2), tricuspid valve replacement (n⫽1),
tricuspid valve annuloplasty (n⫽5), excision of benign pericardial tumor consisting of mesothelial cells (n⫽1), cryoablation for arrhythmias (n⫽1), and division of the left vertical
vein with anastomosis to the left atrial appendage for repair of
anomalous pulmonary venous return from the left lung
(n⫽1). All secundum ASDs and patent foramina were closed.
Early Complications
There was 1 early death (0.9%) in a 76-year-old woman with
preoperative NYHA class IV who underwent patch closure of
SVASD, suture closure of a patent foramen ovale, and
insertion of a 31-mm Hancock tricuspid valve in 1974. The
preoperative systolic pulmonary arterial pressure was
35 mm Hg. Six days postoperatively, she died of right-sided
heart failure. Serious postoperative morbidity occurred in 2
patients. A 59-year-old man with chronic atrial fibrillation
and a history of multiple strokes preoperatively had a large
nonhemorrhagic stroke on postoperative day 6. A 36-year-old
woman developed an embolic left femoral artery occlusion on
postoperative day 2 while in sinus rhythm; embolectomy was
successful.
ECG Findings
The preoperative ECG was normal in 7 patients (6%). It
demonstrated right bundle-branch block, right ventricular
hypertrophy, or right axis deviation in 82 patients (71%) and
SND in 2 (2%). First-degree atrioventricular block was seen
in 5 patients (5%). Paroxysmal or chronic atrial fibrillation or
flutter was present in 12 (10%). Complete atrioventricular
block was rare and was found in 1 patient. One patient had a
permanent pacemaker implanted preoperatively. New postoperative SND occurred in 6 patients and was not related to the
presence of persistent left superior vena cava or APVC but
was marginally (P⫽0.07 for each) related to age and presence
of NYHA functional class III or IV symptoms.
Sixty patients had predismissal and late ECG assessment.
Four patients in whom SND developed before dismissal had late
follow-up; SND resolved in 2, atrial flutter developed in 1, and
1 patient received a pacemaker 6 years after surgery. Two
patients required early postoperative permanent pacemaker implantation for slow ectopic atrial rhythm, and 3 required late
permanent pacemaker implantation. Overall, 6 patients had
permanent pacemaker implantation postoperatively.
Twelve patients had atrial fibrillation preoperatively. Of
the remaining 103 patients, new-onset atrial fibrillation occurred in 7 patients postoperatively. Univariable predictors
for new-onset postoperative atrial fibrillation by Cox regression were older age at repair (P⫽0.033) and preoperative
palpitations (P⫽0.086).
Postoperative Echocardiographic Findings
Postoperative echocardiography was performed at Mayo
Clinic in 67 patients; small persistent defects (residual defect
⬍5 mm by echocardiography) were detected in 5 patients
(7%). Pulmonary vein and SVC stenoses were not identified
by follow-up echocardiography.
Follow-Up
Long-term follow-up was possible for 108 (95%) of the 114
early survivors at an average of 144⫾99 months postopera-
1956
Circulation
September 27, 2005
TABLE 2. Postoperative Long-Term Follow-Up Data for 108
Patients Summarized by Age*
Age at
Repair, y
No. of
Patients
Follow-Up,
mo
Dyspnea,
%†
Clinical
Improvement,
%‡
AF,
%†
ⱕ40
66
149⫾104
21
71
3
41 to 60
26
159⫾105
69
81
31
ⱖ61
16
157⫾105
75
94
31
AF indicates atrial fibrillation or atrial flutter.
*Continuous data are presented as mean⫾SD, categorical data as percentage of patients.
†P⬍0.0001 for association of variable with age.
‡P⫽0.14 for association of variable with age.
tively (median 138 months; range 6.8 to 394 months).
Postoperative follow-up data and outcomes according to age
are shown in Table 2. Improvement in symptoms (ie, decrease in NYHA class or improvement in exercise capacity if
preoperative NYHA class was I) occurred in 83 patients
(77%) who were symptomatic or in NYHA class I preoperatively and was more common with older age at operation
(P⫽0.014), presence of symptoms preoperatively (P⫽0.04),
and higher preoperative pulmonary artery pressure (P⫽0.01).
Despite immediate improvement in postoperative symptoms
in the majority of patients, 17 (16%) demonstrated symptomatic deterioration during long-term follow-up. Deterioration
in functional class occurred in 12 patients with preoperative
NYHA class I, 4 patients with preoperative NYHA class II,
and 1 patient with NYHA class III. In multivariable analysis,
higher preoperative pulmonary artery pressure (P⫽0.022),
but not age, was associated with a higher probability of
postoperative symptomatic improvement. The incidence of
postoperative dyspnea (P⬍0.0001) and atrial fibrillation
(P⬍0.0001) also increased with age. Among 14 patients with
preoperative chronic or paroxysmal atrial fibrillation, 2 maintained sinus rhythm late after SVASD repair. In 7 other
patients (average age at repair 44.7⫾20.9 years), atrial
fibrillation or flutter developed postoperatively. The mean
time from repair to the onset of atrial fibrillation was 9.4⫾9.2
years.
Follow-up ranging from 2 months to 21 years was available for 5 of the 6 patients with preoperative pulmonary
vascular resistance ⬎8 U. All of these patients reported
improvement in symptoms. Estimated pulmonary artery systolic pressure at follow-up was available for 2 patients (41
and 77 mm Hg).
Sixteen patients died late during follow-up (mean age
69⫾19 years). The cause of death was unknown for 9
patients. Five patients died of vehicular accident or carcinoma. Two deaths were possibly related to the SVASD. A
65-year-old man with atrial fibrillation died suddenly 41
months postoperatively. A 60-year-old woman with hypertension had sinus bradycardia 12 months postoperatively and
died suddenly 9 months later. Both patients had normal
coronary arteries preoperatively.
In the forward stepwise multivariable Cox regression
analysis, postoperative mortality (n⫽17) was related to older
age at repair (P⫽0.008) and preoperative NYHA class III or
IV (P⫽0.038). However, survival of these patients after
SVASD repair was not significantly different from the
expected survival for the US white population either overall
or within any of the 3 age strata (P⫽0.31 for overall;
1-sample log-rank test; Figure 2; Table 3).
Among 108 patients with long-term follow-up, reoperation
was not required during the follow-up period. Subsequently,
a patient who had SVASD repair in 1976 underwent reoperation for bidirectional shunting.
Discussion
Patients with SVASD demonstrate unique developmental,
anatomic, and surgical features and are at risk for postoperative complications.3,9 –12 The decision to repair any kind of
ASD is based on clinical and compiled echocardiographic
information, including (1) size and location of the ASD, (2)
Figure 2. Kaplan-Meier survival curves
showing observed survival for patients
after repair of SVASD and expected survival for the US white population. A,
Patients 40 years or younger. B, Patients
41 to 60 years old. C, Patients 61 years
or older. D, Overall survival.
Jost et al
Follow-Up After SVASD Repair
TABLE 3.
Comparison of Outcome in 3 Age Groups
Age Group,
y
Total No. of
Patients
Year of
Repair
Average Age
at Repair,
y
Males,
%
Total No. of
Observed
Events
Total No. of
Expected
Events
1-Sample
Log-Rank
Statistic
P
ⱕ40
67
1982
17.4
50.7
2
1.26279
0.43038
0.51180
41 to 60
26
1982
51.2
46.2
7
5.34114
0.51521
0.47289
ⱖ61
17
1986
69.1
35.3
8
6.67015
0.26514
0.60661
hemodynamic impact of the left-to-right shunt and associated
right-sided cardiac volume overload, and (3) the presence and
degree of pulmonary hypertension. Data from the present
series of patients included early and late results of operation
for SVASD. We found a low rate of perioperative morbidity,
mortality, and need for reoperation. In addition, the overall
survival was similar to that of a matched population. However, older age at operation and NYHA functional class III or
IV symptoms were independent predictors of late mortality.
Older age at operation was also the best independent predictor of new-onset postoperative atrial fibrillation. Postoperative symptomatic improvement was noted in the majority of
the patients.
Surgical Considerations
In SVASDs, the complex anatomy with APVCs poses a
challenge to the surgeon. The results of the present study
cannot prove which method of surgical repair is the best.
Current methods of surgical repair, which include a trend
favoring 2-patch repair and incisions away from the
cavoatrial junction at our institution, as well as intraoperative
transesophageal echocardiography, may have been successful
in eliminating venous pathway stenosis and residual ASDs,
and they may affect the incidence of SND.13,14 Late problems
may occur from contraction of the pericardial patches, resulting in stenosis of the venous pathways or recurrent ASD, but
such problems are rare.
Postoperative Symptoms and
Long-Term Follow-Up
Postoperative symptomatic improvement occurred in most
(77%) of the patients, especially older patients. Improvement
also occurred in patients who were in NYHA class I before
surgery. Functional improvement of previously asymptomatic
patients has been reported after secundum ASD closure.15
Lack of preoperative symptoms is not a contraindication to
the current practice of secundum ASD repair and should not
be a contraindication for repair of SVASD, even in adults.
Reports suggest that ASD closure in patients in their 20s is
associated with survival similar to age- and sex-matched
controls, but closure after age 41 years is associated with a
substantial increase in late mortality.16 The present data do
not support these findings. Although ASD closure should be
considered in select symptomatic patients older than 60 years,
in the absence of serious comorbidities or pulmonary hypertension, it is preferable that ASDs be closed as early as
possible. Older patients are reported to deteriorate symptomatically without ASD repair, because the age-related decrease
in left ventricular compliance augments the left-to-right shunt
and because secondary pulmonary hypertension develops.17
1957
The frequency of atrial arrhythmias increases after the fourth
decade, which also contributes to functional deterioration.
Postoperative improvement in functional class in patients
older than 60 years at the time of ASD repair has been
reported from our institution.18 Despite the improvement in
symptoms, patients in the present study who had repair after
age 40 years demonstrated persistent dyspnea on exertion,
possibly due to accelerated late diastolic dysfunction; the
dyspnea was more pronounced than in patients younger than
40 years. This finding underscores the importance of early
surgical intervention.
Sinus Node Dysfunction
Postoperative SND is more common in patients after SVASD
repair than after secundum ASD repair.19 Potential mechanisms for SND in SVASD include anatomic anomaly of the
sinus node (eg, from a persistent left SVC),19 intrinsic SND,
or surgical trauma caused by proximity of the SVASD to the
sinus node, the internodal tracts, and the blood supply to the
sinus node. In the present study, only 6 patients had newly
documented early postoperative SND independent of an
incision across the cavoatrial junction. Because of the small
number of patients with SND, it is difficult to determine
whether any surgical factors play a role in the development of
postoperative SND. Only 5 patients required postoperative
pacemaker implantation. The presence of atrial arrhythmias
decreases the ability to detect SND; thus, the frequency of
SND may be underestimated in the present series of patients.
Atrial Fibrillation or Flutter
Older age at operation was the best independent predictor of
new-onset atrial fibrillation during follow-up in the present
series of patients. Potential mechanisms for occurrence of
atrial fibrillation or flutter in SVASD patients include SND
with bradycardia-dependent atrial arrhythmias, scardependent multiple reentries, and increased atrial size or atrial
fibrosis due to increasing pulmonary venous pressure with
exercise.
In some of the patients in the present series, new atrial
fibrillation developed during long-term follow-up. Other
studies have also demonstrated that late repair of secundum
ASD does not impact the development of atrial arrhythmias.20
The significant association of older age and postoperative
atrial fibrillation raises the question of whether a maze
procedure should be considered routinely in this subgroup.
Study Limitations
A limitation of the present study is its retrospective design.
We could not assess the natural history of SVASD and do not
have a historical control group for comparison. The natural
1958
Circulation
September 27, 2005
history of unoperated SVASD is unknown but is likely
similar to that in patients with large ASDs. In addition,
postoperative clinical improvement is problematic, because it
is difficult to quantify; however, this subjective information is
the only method to determine clinical outcome in this
retrospective series.
Not all patients had ECG follow-up. Therefore, the frequency of SND or occurrence of long-term atrial arrhythmias
cannot be stated with certainty. The presence of atrial
arrhythmias decreases the ability to detect SND; thus, the
frequency of SND detection may be underestimated. Finally,
the cause of death in the majority of patients is unknown;
therefore, mortality due to SVASD sequelae may be
underestimated.
The decision for operative intervention for SVASD should
be individualized. No definite recommendations about upper
and lower age limits for surgery can be made from these data;
however, operation for SVASD is rarely necessary or advisable in an infant younger than 1 year or in the very elderly.
Recommendations with regard to surgery in asymptomatic
patients are hampered by the fact that during long-term
follow-up, symptomatic deterioration was noted in 24% of
patients who were in functional class I before surgery. We
cannot determine whether this was due to surgery or despite
surgery or whether this was only a normal deterioration
paralleling the long follow-up. In patients who were in
functional class I before surgery, death at last follow-up was
less likely than in the other patients (P⫽0.0005), and there
was no significant difference in the occurrence of sinus node
dysfunction, atrial fibrillation, or pacemaker implantation.
Conclusions
Despite the complexity of the lesion, repair of SVASD with
associated APVC is associated with low morbidity and
mortality even in patients older than 40 years. In our
experience, severe complications are rare, and development
of SND and the need for pacemaker implantation are uncommon. Functional improvement is expected irrespective of age
at repair, but postoperative atrial fibrillation appears to be
related to older age at operation. SVASD repair achieves
survival rates similar to those of a matched population, and
although repair is suggested as early as possible, it should be
considered whenever repair may impact survival or
symptoms.
Disclosure
Dr Shen has received research grants from Medtronic and Guidant.
References
1. Peacock TB. Malformations of the heart. In: Peacock TB, ed. On Malformations, &c., of the Human Heart: With Original Cases. London, UK:
John Churchill; 1858:11–102.
2. Oliver JM, Gallego P, Gonzalez A, Dominguez FJ, Aroca A, Mesa JM.
Sinus venosus syndrome: atrial septal defect or anomalous venous connection? A multiplane transoesophageal approach. Heart. 2002;88:
634 – 638.
3. Van Praagh S, Carrera ME, Sanders SP, Mayer JE, Van Praagh R. Sinus
venosus defects: unroofing of the right pulmonary veins: anatomic and
echocardiographic findings and surgical treatment. Am Heart J. 1994;
128:365–379.
4. Kirklin JW, Barratt-Boyes BG, eds. Cardiac Surgery: Morphology, Diagnostic Criteria, Natural History, Techniques, Results, and Indications.
2nd ed. New York, NY: Churchill Livingstone; 1993;1:609 – 644.
5. Brais MP, Texeira OH. Partial anomalous pulmonary venous connection
of right lung with inferior sinus venosus atrial septal defect. Pediatr
Cardiol. 1984;5:156 –157.
6. DeLeon SY, Freeman JE, Ilbawi MN, Husayni TS, Quinones JA, Ow EP,
Bell TJ, Pifarre R. Surgical techniques in partial anomalous pulmonary
veins to the superior vena cava. Ann Thorac Surg. 1993;55:1222–1226.
7. Warden HE, Gustafson RA, Tarnay TJ, Neal WA. An alternative method
for repair of partial anomalous pulmonary venous connection to the
superior vena cava. Ann Thorac Surg. 1984;38:601– 605.
8. Gustafson RA, Warden HE, Murray GF, Hill RC, Rozar GE. Partial
anomalous pulmonary venous connection to the right side of the heart.
J Thorac Cardiovasc Surg. 1989;98:861– 868.
9. Robicsek F, Daugherty HK, Cook JW, Selle JG. Sinus venosus type of
atrial septal defect with partial anomalous pulmonary venous return.
J Thorac Cardiovasc Surg. 1979;78:559 –562.
10. Pieroni DR, Strife JL, Donahoo JS, Krovetz LJ. Postoperative assessment
of residual defects following cardiac surgery in infants and children, 3:
atrial septal defects. Johns Hopkins Med J. 1973;133:287–294.
11. Friedli B, Guerin R, Davignon A, Fouron JC, Stanley P. Surgical
treatment of partial anomalous pulmonary venous drainage: a long-term
follow-up study. Circulation. 1972;45:159 –170.
12. Rahimtoola SH, Kirklin JW, Burchell HB. Atrial septal defect. Circulation. 1968;38(suppl):2–12.
13. Nicholson IA, Chard RB, Nunn GR, Cartmill TB. Transcaval repair of the
sinus venosus syndrome. J Thorac Cardiovasc Surg. 2000;119:741–744.
14. Walker RE, Mayer JE, Alexander ME, Walsh EP, Berul CI. Paucity of
sinus node dysfunction following repair of sinus venosus defects in
children. Am J Cardiol. 2001;87:1223–1226.
15. Helber U, Baumann R, Seboldt H, Reinhard U, Hoffmeister HM. Atrial
septal defect in adults: cardiopulmonary exercise capacity before and 4
months and 10 years after defect closure. J Am Coll Cardiol. 1997;29:
1345–1350.
16. Ward C. Secundum atrial septal defect: routine surgical treatment is not
of proven benefit. Br Heart J. 1994;71:219 –223.
17. Murphy JG, Gersh BJ, McGoon MD, Mair DD, Porter CJ, Ilstrup DM,
McGoon DC, Puga FJ, Kirklin JW, Danielson GK. Long-term outcome
after surgical repair of isolated atrial septal defect: follow-up at 27 to 32
years. N Engl J Med. 1990;323:1645–1650.
18. John Sutton MG, Tajik AJ, McGoon DC. Atrial septal defect in patients
ages 60 years or older: operative results and long-term postoperative
follow-up. Circulation. 1981;64:402– 409.
19. Arensman FW, Boineau JP, Balfour IC, Flannery DB, Moore HV. Sinus
venosus atrial septal defect and pacemaker requirement in a family.
Am J Cardiol. 1986;57:368 –369.
20. Gatzoulis MA, Freeman MA, Siu SC, Webb GD, Harris L. Atrial arrhythmia after surgical closure of atrial septal defects in adults. N Engl
J Med. 1999;340:839 – 846.
Coronary Heart Disease
Rapid Heart Rate Increase at Onset of Exercise Predicts Adverse
Cardiac Events in Patients With Coronary Artery Disease
Colomba Falcone, MD; Maria Paola Buzzi, MD; Catherine Klersy, MD; Peter J. Schwartz, MD
Background—We previously demonstrated that reduced vagal activity and/or increased sympathetic activity identify
post–myocardial infarction patients at high risk for cardiac mortality. Simple and inexpensive autonomic markers are
necessary to perform autonomic screening in large populations. We tested our hypothesis that abnormally elevated heart
rate (HR) responses at the onset of an exercise stress test, which imply rapid vagal withdrawal immediately preceding
sympathetic activation, might predict adverse cardiac events in patients with documented coronary artery disease.
Methods and Results—The HR increase during the first minute (⌬HR1 minute) of a symptom-limited exercise stress test was
quantified in 458 patients with documented coronary artery disease. During a 6-year (interquartile range 3.7 to 9.0 years)
follow-up, 71 patients experienced adverse cardiac events (21 cardiac deaths, 56 nonfatal myocardial infarctions). In
univariate analysis, ⌬HR1 minute ⱖ12 bpm (above the median value of its distribution) predicted both adverse outcome
and cardiac death with a hazard ratio of 5.0 (95% CI 2.7 to 9.1; P⬍0.0001) and of 15.6 (95% CI 2.0 to 118.7; P⬍0.001),
respectively. After adjustment for potential confounders, ⌬HR1 minute remained predictive for both combined end points
and for cardiac death.
Conclusions—A marked HR increase at the onset of a standard exercise stress test is a novel and easily available parameter
that could be clinically useful as an independent predictor of adverse cardiac events, including death, among patients
with documented coronary artery disease. (Circulation. 2005;112:1959-1964.)
Key Words: exercise 䡲 heart rate 䡲 mortality 䡲 nervous system, autonomic 䡲 risk factors
E
arly identification of individuals at high risk for cardiovascular mortality and morbidity is a cornerstone of
modern medicine. The concept that alterations in the autonomic control of cardiac functions, characterized by augmented sympathetic and reduced vagal activity, play a major
role in cardiovascular mortality1 has had a wide impact.
Indeed, the search for markers of “autonomic imbalance” has
contributed to risk stratification in different patient
populations.
The evidence that reduced ability to reflexly increase vagal
activity, as quantified by baroreflex sensitivity,2 predicts
increased risk for sudden and nonsudden cardiac death after
myocardial infarction3– 6 has stimulated the search for and
testing of other markers of reduced vagal activity. Some of
them, such as heart rate (HR) variability and HR turbulence,
have validity in postinfarction patients.7,8 Others, such as HR
reduction and occurrence of ventricular arrhythmias in the
recovery period of exercise, have focused more on the
autonomic changes induced by exercise testing and have
suggested that this period may provide important prognostic
information.9,10 The latter 2 studies have brought a new
dimension to the inexpensive and frequently used exercise
testing.11 So far, all attempts by ourselves5,6,8 and others7,9,10
have focused on markers of tonic or reflex vagal activity,
searching for a correlation between cardiac events and impaired ability to increase the “protective” vagal activity. We
have now examined a different facet of autonomic regulation,
namely, the rapidity of vagal withdrawal at onset of exercise,
because we postulate that the faster the vagal withdrawal in
response to a stress, the greater will be the deleterious effect
of sympathetic activation unopposed by vagal activity. Accordingly, in a cohort of patients with angiographically
documented coronary artery disease, we tested our hypothesis
that a novel autonomic marker—the rapidity of HR increase
during the first minute of exercise—might predict major
cardiovascular events.
Methods
Patient Population
The study population consisted of 458 consecutive male patients
referred in the 1990s at our center for coronary angiography, who
subsequently underwent an exercise stress test (EST) and were
scheduled for regular follow-up. Patients were included if coronary
arteriography documented significant coronary artery disease
(ⱖ50%) and excluded if they had symptoms or signs of heart failure,
previous evidence of impaired left ventricular ejection fraction, use
of digoxin, valvular or congenital disease, a pacemaker, or a
Received February 23, 2005; revision received July 7, 2005; accepted July 8, 2005.
From the Department of Lung, Blood, and Heart (C.F., M.P.B., P.J.S.), University of Pavia, Pavia, Italy; Department of Cardiology (C.F., M.P.B.,
P.J.S.), IRCCS Policlinico S. Matteo, Pavia, Italy; and Biometry and Clinical Epidemiology Unit (C.K.), IRCCS Policlinico S. Matteo, Pavia, Italy.
Correspondence to Peter J. Schwartz, MD, Professor & Chairman, Department of Cardiology, Policlinico S. Matteo IRCCS, Via le Golgi, 19-27100
Pavia, Italy. E-mail [email protected]
1959
1960
Circulation
September 27, 2005
noninterpretable ECG. Before the EST, a structured interview
gathered data on coronary risk factors, symptoms, medications, and
previous cardiac events. All patients gave informed consent, and the
study was approved by the institutional review board.
Exercise Testing
Stress testing for detection of myocardial ischemia was performed in
accordance with the American College of Cardiology/American
Heart Association guidelines. A multistage symptom-limited EST
was performed on a bicycle ergometer in the semisupine position.
The initial workload was 25 W, with subsequent stepwise increments
of 25 W every 2 minutes at a pedaling rate of 60 rpm; peak workload
was followed by at least a 2-minute cool-down period. Standard
12-lead ECG and blood pressure were recorded in basal conditions,
every minute during exercise, at peak exercise, and every minute
during recovery. Frequent or complex ventricular arrhythmias were
recorded. A positive ECG response was defined as the occurrence of
ST-segment depression ⱖ1 mm compared with the baseline tracing.
The EST was stopped when angina, dyspnea, muscle fatigue,
ST-segment depression ⬎3 mm, or major arrhythmias occurred. The
estimated workload was determined in metabolic equivalents
(METS). Patients who performed their EST in pharmacological
washout stopped use of calcium channel blocking agents and nitrates
48 hours before the EST or gradually reduced ␤-blocker therapy 1
week in advance.
Assessment of HR Response
During EST, HR increases were calculated at 1 minute after the
beginning of exercise (⌬HR1 minute), at the end of each stepwise
increment, and at peak exercise. For the purpose of the analysis,
⌬HR1 minute was dichotomized according to the median value of its
distribution (⬍12 bpm, ⱖ12 bpm). HR recovery was defined as the
difference in HR between the values recorded at the end of exercise
and those recorded 1 minute after termination of exercise. A cutoff
value of 12 bpm or less was considered abnormal.9
Follow-Up
Patients were followed up for a median of 6 years (interquartile range
3.7 to 9.0 years). The end point of the study was a composite of
cardiac death and nonfatal myocardial infarction. Most of the
patients attended our center once or twice per year, according to
clinical conditions; clinical data for those who interrupted their
periodic follow-up were obtained through telephone calls. Out-ofhospital deaths were investigated by means of interview with the
next of kin or patient’s physicians or by analysis of death certificates.
Myocardial infarction was diagnosed on the basis of clinical symptoms, ECG changes, and cardiac enzyme elevations.
Data are presented as mean⫾SD for continuous variables and as
absolute and relative frequencies for categorical variables. Follow-up
time is summarized with median and interquartile range. Comparisons between ⌬HR1 minute groups were performed by means of the
Student t test and Fisher exact test for continuous and categorical
variables, respectively.
Kaplan-Meier estimates of event-free survival were plotted. Time
origin was the time of EST. Patients undergoing revascularization or
dying of other causes were censored at the time of revascularization
when we analyzed the combined event and additionally at the time of
myocardial infarction when we analyzed cardiac death. The event
rate per 100 person-years was computed together with its 95% CI.
Cox proportional hazard model was used to evaluate the role of
⌬HR1 minute dichotomized at 12 bpm as a risk factor for the combined
end point and for cardiac death. To further clarify the role of
⌬HR1 minute in predicting adverse events, we also evaluated the risk of
death and myocardial infarction of our study population according to
the third (⌬HR1 minute from 12 to 18 bpm) and fourth (⌬HR1 minute ⬎18
bpm) quartiles of its distribution. Other known clinical and EST
potential risk factors were also assessed, as was their interaction with
⌬HR1 minute (which was excluded in all cases). The proportional
hazard assumption was tested based on Schoenfeld residuals. No
violation was observed. Hazard ratios and 95% CIs were calculated.
The role of ⌬HR1 minute on a continuous scale was evaluated as well.
Martingale residuals analysis indicated a linear effect. The Cox
model was fitted to compute the hazard ratio for ⌬HR1 minutes, after
adjustment for potential confounders (age, hypertension, hypercholesterolemia, diabetes, obesity, smoking, familial history of coronary
artery disease, chronotropic incompetence, resting HR, abnormal HR
recovery, exercise-induced frequent arrhythmias, exercise-induced
ischemia, exercise-induced change in systolic arterial pressure,
personal history of coronary heart disease, number of diseased
coronary vessels, ␤-blocker therapy, and active drug therapy at the
time of stress test evaluation). Backward stepwise selection was
used, with P-to-remove of 0.2. Finally, subgroup analysis was
performed by fitting Cox models for ⌬HR1 minute within strata of some
relevant patient characteristics.
Stata 8 (StataCorp) was used for computation. All probability
values are 2 sided. Probability values for subgroup analysis are
unadjusted.
Results
The study cohort consisted of 458 male patients (mean age
56⫾8.5 years). At the time of stress test evaluation, 162 patients
(35.4%) reported anginal pain during daily life (49% had
exertion angina, 18% had angina at rest, and 33% had mixed
angina), whereas 296 patients (64.6%) were asymptomatic; 286
patients (62.4%) had a prior MI; and 232 (50.6%) had a prior
coronary revascularization. The EST was performed while
patients were taking ␤-blocker therapy or nondihydropyridine
calcium channel blocking agents in 142 cases (31.0%), whereas
316 patients (69.0%) were in pharmacological washout. An
ischemic response to the EST was observed in 172 patients
(37.5%).
The baseline and stress test characteristics of patients, according to whether their ⌬HR1 minute was ⱖ12 bpm (above the
median) or ⬍12 bpm (equal to or below the median), are shown
in Table 1. These 2 groups were similar for most clinical
features, with no observed differences in the presence of hypertension, hypercholesterolemia, or diabetes; use of ␤-blockers,
calcium channel blocking agents, or nitrates; resting and peak
systolic or diastolic blood pressures; presence of coronary artery
disease; or ejection fraction. Compared with subjects with lower
⌬HR1 minute, those with ⌬HR1 minute ⱖ12 bpm were younger and
had a lower resting HR; during exercise, they reached higher
values of peak HR and were more likely to present abnormal
ST-segment responses. No differences in the percentage of
patients with abnormal HR recovery or exercise-induced frequent or complex arrhythmias were observed in the 2 groups.
Cardiovascular Events
During a median follow-up period of 6 years (interquartile
range 3.7 to 9.0 years), 71 patients (15.5%) had adverse
cardiac events; 15 (3.3%) died, and 56 (12.2%) developed a
nonfatal myocardial infarction, with 6 additional later deaths.
Thus, there were 21 total cardiac deaths (4.6%). No patient
underwent heart transplantation or implantation of an implantable cardioverter defibrillator during follow-up. We
observed 58 adverse events over 1560 person-years among
patients with ⌬HR1 minute ⱖ12 bpm and only 13 over 1370
person-years among patients with ⌬HR1 minute ⬍12. Thus, the
event rate per 100 person-years of those with lower ⌬HR1 minute
was 0.8 (95% CI 0.5 to 1.4), whereas it was 4.2 (95% CI 3.3
Falcone et al
TABLE 1.
Early HR Increase During Exercise Predicts Risk
1961
Clinical and Exercise-Related Characteristics of Patients According to ⌬HR1 minute
⌬HR1 minute ⬍12 bpm
(n⫽244)
⌬HR1 minute ⱖ12 bpm
(n⫽214)
57.4⫾8.6
55.4⫾8.3
0.01
26⫾3.2
25⫾2.7
⬍0.01
Hypertension
116 (47.5)
93 (43.4)
Hypercholesterolemia
148 (60.6)
117 (54.7)
0.19
22 (9.0)
23 (10.7)
0.53
Variable
P
Clinical characteristics
Age, y
Body mass index, kg/m2
Diabetes
Familial history of coronary artery disease
0.38
97 (39.7)
98 (45.8)
0.19
193 (79.1)
152 (71.0)
0.04
45 (18.4)
21 (9.8)
0.008
1-Vessel disease
143 (58.6)
118 (55.1)
0.45
2-Vessel disease
53 (21.7)
58 (27.1)
0.18
3-Vessel disease
48 (19.7)
38 (17.8)
0.60
57⫾10
56⫾10
0.29
75⫾14
70⫾12
⬍0.0001
Resting systolic blood pressure, mm Hg
129⫾18
131⫾18
0.30
Resting diastolic blood pressure, mm Hg
83⫾8
83⫾4
0.62
Peak heart rate, bpm
125⫾21
129⫾20
0.047
Exercise capacity, METS
6.3⫾1
6.1⫾1
0.02
Exercise-induced ST-segment depression ⱖ1 mm
75 (30.7)
97 (45.3)
0.001
Smoking
Prior CABG
Angiographic findings
Ejection fraction
Exercise-related characteristics
Resting heart rate, bpm
Values are mean⫾SD or n (%).
to 5.5) in patients with higher ⌬HR1 minute. The findings show
that ⌬HR1 minute ⱖ12 bpm was strongly predictive of adverse
outcome (hazard ratio [HR] 5.0, 95% CI 2.7 to 9.1;
P⬍0.0001). Event-free survival curves for both groups are
reported in Figure 1A. On a continuous scale, the risk
increased linearly by 40% for each increase in ⌬HR1 minute of
5 bpm (HR 1.4, 95% CI 1.2 to 1.5; P⬍0.001).
The only other predictor of death and myocardial infarction
was hypercholesterolemia (HR 1.6, 95% CI 1.0 to 2.7; P⬍0.05).
Abnormal HR recovery showed only a trend for association with
cardiac events (HR 1.4, 95% CI 0.7 to 2.8; P⫽0.30). The
following variables were nonpredictive for cardiovascular
events: age, hypertension, diabetes, family history of coronary
disease, and exercise-induced arrhythmias.
To further elucidate the prognostic role of ⌬HR1 minute, we
evaluated the event rate of patients with ⌬HR1 minute from 12 to 18
bpm (third quartile) and ⬎18 bpm (fourth quartile) with respect
to those with ⌬HR1 minute ⬍12 bpm. The third and fourth quartiles
were associated with an HR of 3.3 (95% CI 1.7 to 6.6) and 6.3
(95% CI 3.5 to 11.4), respectively (both P⬍0.01); the outcome
was also found to differ between the third and fourth quartiles
(P⫽0.027; Figure 1B). In a backward stepwise multivariate Cox
analysis, ⌬HR1 minute ⱖ12 bpm remained predictive for cardiac
adverse events (adjusted HR 5.8, 95% CI 3.1 to 10.9;
P⬍0.0001) after adjustment for hypertension, hypercholesterolemia, diabetes, obesity, smoking, familial history of coronary
artery disease, chronotropic incompetence, resting HR, abnormal HR recovery, exercise-induced frequent or complex arrhythmias, exercise-induced ischemia, exercise-induced change in
systolic arterial pressure, personal history of coronary heart
disease, number of diseased coronary vessels, coronary revascularization, ␤-blocker therapy, and active drug therapy at the
time of stress test evaluation.
Subgroup Analysis
The subgroup analyses (Table 2), none of which demonstrated a significant interaction, indicated that the effect of
⌬HR1 minute ⱖ12 bpm was present in each subgroup. Of note,
⌬HR1 minute ⱖ12 bpm was predictive of adverse outcome both
in patients taking ␤-blockers or nondihydropyridine calciumchannel– blocking agents (HR 4.9, 95% CI 1.6 to 14.8;
P⬍0.001) and in those not taking these drugs (HR 5.0, 95%
CI 2.4 to 10.2; P⬍0.0001). Nonetheless, patients taking
HR-lowering drugs had lower values for resting HR, peak
HR, and peak systolic and diastolic blood pressures than
patients not taking these drugs at the time of EST evaluation
(all P⬍0.001), whereas ⌬HR1 minute was not significantly
different.
Cardiac Death
⌬HR1 minute ⱖ12 bpm also showed a strong association with
cardiac death (Figure 2A), both with univariate and adjusted
Cox analysis (HR 15.6, 95% CI 2.0 to 118.7, P⬍0.001 and
13.5, 95% CI 1.8 to 103.7, P⬍0.001, respectively). Given the
relatively small number of cardiac deaths, this analysis could
be regarded as exploratory.
Survival curves for the third and fourth quartiles of ⌬HR1 minute
are shown in Figure 2B. With respect to those with ⌬HR1 minute
1962
Circulation
September 27, 2005
⌬HR1 minute has significant advantages. They include universal
in-hospital availability, simplicity, minimal cost, and above all
the fact that valid data are obtained even when patients perform
just the first minute of an EST. These considerations also suggest
that ⌬HR1 minute could be used for autonomic screening in large
populations.
Rapid HR Increase and Risk Stratification
Figure 1. A, Kaplan-Meier event-free (combined for cardiac
death and nonfatal myocardial infarction) survival estimate
according to ⌬HR1 minute ⱖ or ⬍12 bpm. B, Kaplan-Meier eventfree (combined for cardiac death and nonfatal myocardial infarction) survival estimate according to change in ⌬HR1 minute.
⬍12 bpm, HRs for the third and fourth quartiles, respectively,
were 14.7 (95% CI 1.7 to 125.9; P⫽0.014) and 21.7 (95% CI 2.7
to 171.7; P⫽0.004), with no significant difference between the
2 groups (P⫽0.48).
Discussion
The present study demonstrates that a rapid HR increase at
the beginning of a standard EST is a strong and independent
predictor of cardiac death and nonfatal myocardial infarction
in patients with angiographic evidence of coronary artery
disease. This finding has conceptual and practical
implications.
From the point of view of cardiovascular pathophysiology,
the fact that excessive vagal withdrawal is associated with
adverse events contributes to the body of evidence indicating
that autonomic imbalance increases cardiac risk.1,5,6,12–14 It also
raises the intriguing possibility of autonomic modulation, aimed
at increasing vagal activity, as a means to reduce risk.
From the clinical perspective, compared with the complex and
relatively expensive autonomic markers currently available,
The underlying rationale for the assessment of possible
autonomic imbalance is represented by the fact that sympathetic hyperactivity, as well as reduced vagal activity, increases electrical instability, thus enhancing life-threatening
arrhythmias,15,16 and may even predict rapid progression of
coronary artery disease.17 Indeed, after several experimental
studies,3,4 the value of autonomic imbalance in predicting
susceptibility to cardiac death has become evident among
patients with diverse cardiovascular diseases.5–7,18 Recently,
HR variability has been used successfully for risk stratification in a large, prospective clinical trial.19 Measures of
autonomic control, however, are only slowly entering the
process of risk stratification on a routine basis because of the
complexity and cost of most autonomic markers. This mismatch highlights the need for simple tools to allow autonomic
screening in large patient populations.
The present data suggest that ⌬HR1 minute might represent a
novel autonomic marker that could usefully contribute to a
simpler and more accurate identification of high-risk coronary
artery disease patients. ⌬HR1 minute relates to outcome whether
measured as a continuous or a categorical variable. Although the
event rate was only 0.8 per 100 person-years among individuals
with ⌬HR1 minute ⬍12 bpm, it was 4.2 in patients with ⌬HR1 minute
ⱖ12 bpm, with a more than 4-fold increase in risk. Also, the risk
of events increased linearly with increasing values of ⌬HR1 minute.
Along the same lines, patients with ⌬HR1 minute between 12 and
18 bpm and above 18 bpm (third and fourth quartiles, respectively) had a 3- and 6-fold increase in risk for cardiac events
compared with patients with ⌬HR1 minute below 12 bpm. A similar
pattern was observed when we analyzed the risk for cardiac
death. Importantly, the prognostic value of HR response at onset
of exercise was present, with a risk of different magnitude, in all
subgroups of patients with common risk factors for coronary
artery disease (Table 2). Moreover, ⌬HR1 minute was independent
of other EST-related clinical or therapeutic variables, such as
resting HR, abnormal HR recovery, exercise-induced frequent or
complex arrhythmias, hypercholesterolemia, previous myocardial infarction, coronary revascularization, and ␤-blocker
therapy.
The recent report showing that the HR profile during an EST
contains prognostic information about the long-term risk for
sudden death among apparently healthy individuals points to an
important role of the autonomic nervous system in determining
cardiovascular outcomes, as does the present study.20 The main
difference between these 2 studies lies in the present finding that
the essential information can be provided after just the first
minute of exercise, with all the attendant implications for the
many patients unable to perform a complete EST.
Abnormal HR Recovery and Prognosis
A recent series of studies focused on HR recovery after
exercise, which was used as a marker of vagal activation;
Falcone et al
Early HR Increase During Exercise Predicts Risk
1963
TABLE 2. Association Between ⌬HR1 minute >12 bpm and Adverse Events (Cardiac Death and Nonfatal
Myocardial Infarction) in Considered Subgroups and Interaction Analysis
No. of Adverse Events/No. of Patients (%)
Variables
⌬HR1 minute ⬍12 bpm
⌬HR1 minute ⱖ12 bpm
Relative Risk
(95% CI)
P
P for
Interaction
10/196 (5.1)
50/188 (26.6)
5.1 (2.6–10.0)
⬍0.0001
0·79
3/48 (6.2)
8/26 (30.8)
6.5 (1.7–24.6)
⬍0.01
Age
⬍65 y
ⱖ65 y
Previous myocardial infarction
No
6/90 (6.7)
24/82 (29.3)
4.3 (1.7–10.4)
⬍0.001
Yes
7/154 (4.5)
34/132 (25.8)
5.7 (2.5–12.8)
⬍0.0001
No
8/114 (7.0)
29/112 (25.9)
3.2 (1.5–7.0)
⬍0.01
Yes
5/130 (3.8)
29/102 (28.4)
8.3 (3.2–21.3)
⬍0.0001
11/152 (7.2)
35/131 (26.7)
3.5 (1.8–6.9)
⬍0.001
2/92 (2.2)
23/83 (27.7)
13.3 (3.1–56.4)
⬍0.001
10/199 (5.0)
51/193 (26.4)
4.8 (2.4–9.5)
⬍0.0001
3/45 (6.7)
7/21 (33.3)
6.0 (1.5–23.3)
⬍0.01
No
8/126 (6.3)
38/114 (33.3)
5.3 (2.5–11.3)
⬍0.0001
Yes
5/118 (4.2)
20/100 (20.0)
4.7 (1.8–12.6)
⬍0.001
No
9/162 (5.6)
42/154 (27.3)
5.0 (2.4–10.2)
⬍0.0001
Yes
4/82 (4.9)
16/60 (26.7)
4.9 (1.6–14.8)
⬍0.01
0·64
History of coronary revascularization
0·14
Previous percutaneous coronary intervention
No
Yes
0·08
Previous CABG
No
Yes
0·73
Use of ␤-blockers
0·81
EST on therapy
Cole et al9,21 showed that an HR decrease ⱕ12 bpm within
1 minute of recovery after a symptom-limited Bruce protocol
test was a predictor of overall mortality. Patients referred for
an EST with radionuclide testing with an abnormal HR
recovery had a 4 times greater 6-year mortality rate. These
results were confirmed in subsequent studies by the same and
other investigators.9,21–23
In the present study, HR recovery at the end of exercise did
indeed show a trend toward increased risk, which, however, did
not reach statistical significance. This may reflect an insufficient
power of the study or the use of an end point (combined
incidence of infarction and cardiac deaths) that was different
from total mortality. Also, HR recovery is clearly related to other
chronotropic variables (peak HR and percent peak HR, workload), which suggests that it could be an expression of impaired
exercise capacity, which has already been proven to be an
independent risk stratifier.24
Clinical Implications
In the past, several exercise variables have been assessed for
prognostic value, and it has become evident that mortality and
morbidity can be predicted by the evaluation of ST-segment
depression, exercise-induced angina, and exercise capacity.
All these variables are strongly related to and affected by the
clinical status of the patients. Factors such as poor muscle
tone, pulmonary diseases, and self-motivation can reduce
functional capacity and limit the possibility of reaching an
ischemic or angina threshold. A major strength of the present
study lies in the demonstration that ⌬HR1 minute is a useful
0·99
prognostic marker even in the presence of severe limitations
of functional capacity because it requires a very short
duration of exercise.
A limitation of the study is that the relatively small number of
deaths has produced wide CIs, which affect the precision of the
HR estimates without questioning the increased risk associated
with ⌬HR1 minute ⱖ12 bpm. This is not the case for the strong
predictive value of the combined end point (cardiac death and
nonfatal myocardial infarction).
The possibility of identifying a significant interaction
between ⌬HR1 minute and other common risk factors was
limited by the size of the present study population. Nevertheless, ⌬HR1 minute ⱖ12 bpm remained a significant predictor of
adverse events in all subgroups, even if it was associated with
a different degree of risk.
Whether our observations, obtained in a population of patients
with documented coronary artery disease who were eligible for
exercise stress testing, also apply to other populations, such as a
community-based sample, requires further studies. Indeed, the
present observations should be confirmed in a separate data set.
There are 2 main practical implications of the present study.
One is the availability of a simple test, based on a solid
background of experimental pathophysiology and of clinical
evidence, that provides a novel autonomic marker capable of
identifying patients with coronary artery disease at risk of major
events. The other is that the nature of the abnormality unmasked
by the test, autonomic imbalance, allows institution of effective
preventive interventions beyond the obvious consideration for
use of ␤-blockers. Specifically, exercise training titrated to
1964
Circulation
September 27, 2005
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
Figure 2. A, Kaplan-Meier survival estimate according to
⌬HR1 minute. ⱖ or ⬍12 bpm, limited to cardiac deaths. B, KaplanMeier survival estimate according to change in ⌬HR1 minute, limited to cardiac deaths (below the median and 3rd and 4th
quartiles).
effectively increase reflex vagal activity has already shown the
potential of restoring autonomic balance and of reducing subsequent cardiovascular risk.25
18.
19.
Acknowledgment
We are grateful to Pinuccia De Tomasi for expert editorial support.
References
1. Schwartz PJ, La Rovere MT, Vanoli E. Autonomic nervous system and
sudden cardiac death: experimental basis and clinical observations for
post-myocardial infarction risk stratification. Circulation. 1992;85(suppl
I):I-77–I-91.
2. La Rovere MT, Schwartz PJ. Baroreflex sensitivity. In: Zipes DP, Jalife
J, Eds. Cardiac Electrophysiology: From Cell to Bedside. 3rd ed. Philadelphia, Pa: WB Saunders; 2000:771–781.
3. Billman GE, Schwartz PJ, Stone HL. Baroreceptor reflex control of heart
rate: a predictor of sudden cardiac death. Circulation. 1982;66:874 – 880.
4. Schwartz PJ, Vanoli E, Stramba-Badiale M, De Ferrari GM, Billman GE,
Foreman RD. Autonomic mechanisms and sudden death: new insights
from analysis of baroreceptor reflexes in conscious dogs with and without
a myocardial infarction. Circulation. 1988;78:969 –979.
5. La Rovere MT, Bigger JT Jr, Marcus FI, Mortara A, Schwartz PJ, for the
ATRAMI (Autonomic Tone and Reflexes After Myocardial Infarction)
Investigators. Baroreflex sensitivity and heart rate variability in prediction
20.
21.
22.
23.
24.
25.
of total cardiac mortality after myocardial infarction. Lancet. 1998;351:
478 – 484.
La Rovere MT, Pinna GD, Hohnloser SH, Marcus FI, Mortara A, Nohara
R, Bigger JT Jr, Camm AJ, Schwartz PJ, on behalf of the ATRAMI
Investigators. Baroreflex sensitivity and heart rate variability in the identification of patients at risk for life-threatening arrhythmias: implication
for clinical trials. Circulation. 2001;103:2072–2077.
Kleiger RE, Miller JP, Bigger JT Jr, Moss AJ, and the Multicenter
Postinfarction Research Group. Decreased heart rate variability and its
association with increased mortality after acute myocardial infarction.
Am J Cardiol. 1987;59:256 –262.
Ghuran A, Reid F, La Rovere MT, Schmidt G, Bigger JT Jr, Camm AJ,
Schwartz PJ, Malik M, on the behalf of the ATRAMI Investigators. Heart
rate turbulence-based predictors of fatal and nonfatal cardiac arrest (the
Autonomic Tone and Reflexes After Myocardial Infarction Substudy).
Am J Cardiol. 2002;89:184 –190.
Cole CR, Blackstone EH, Pashkow FJ, Snader CE, Lauer MS. Heart rate
recovery immediately after exercise as a predictor of mortality. N Engl
J Med. 1999;341:1351–1357.
Frolkis JP, Pothier CE, Blackstone EH, Lauer MS. Frequent ventricular
ectopy after exercise as a predictor of death. N Engl J Med. 2003;348:
781–790.
Curfman GD, Hills LD. A new look at cardiac exercise testing. N Engl
J Med. 2003;348:775–776.
De Ferrari GM, Vanoli E, Schwartz PJ. Cardiac vagal activity, myocardial ischemia and sudden death. In: Zipes DP, Jalife J, eds. Cardiac
Electrophysiology: From Cell to Bedside. 2nd ed. Philadelphia, Pa: WB
Saunders; 1995:422– 434.
Schwartz PJ. The neural control of heart rate and risk stratification after
myocardial infarction. Eur Heart J. 1999;20(suppl H):H33-H43.
Hull SS Jr, Vanoli E, Adamson PB, Verrier RL, Foreman RD, Schwartz
PJ. Exercise training confers anticipatory protection from sudden death
during acute myocardial ischemia. Circulation. 1994;89:548 –552.
Schwartz PJ, Priori SG. Sympathetic nervous system and cardiac arrhythmias. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell
to Bedside. Philadelphia, Pa: WB Saunders; 1990:330 –343.
Lown B, Verrier RL. Neural activity and ventricular fibrillation. N Engl
J Med. 1976;294:1165–1170.
Huikuri HV, Jokinen V, Syvanne M, Nieminen MS, Airaksinen KE,
Ikaheimo MJ, Koistinen JM, Kauma H, Kesaniemi AY, Majahalme S,
Niemela KO, Frick MH. Heart rate variability and progression of
coronary atherosclerosis. Arterioscler Thromb Vasc Biol. 1999;19:
1979 –1985.
La Rovere MT, Pinna GD, Maestri R, Mortara A, Capomolla S, Febo O,
Ferrari R, Franchini M, Gnemmi M, Opasich C, Riccardi PG, Traversi E,
Cobelli F. Short-term heart rate variability strongly predicts sudden
cardiac death in chronic heart failure patients. Circulation. 2003;107:
565–570.
Camm AJ, Pratt CM, Schwartz PJ, Al-Khalidi HR, Spyt MJ, Holroyde
MJ, Karam R, Sonnenblick EH, Brum JMG, on behalf of the AzimiLide
post Infarct surVival Evaluation (ALIVE) Investigators. Mortality in
patients after a recent myocardial infarction: a randomized, placebocontrolled trial of azimilide using heart rate variability for risk stratification. Circulation. 2004;109:990 –996.
Jouven X, Empana JP, Schwartz PJ, Desnos M, Courbon D, Ducimetière
P. Heart rate profile during exercise as a predictor of sudden death.
N Engl J Med. 2005;352:1951–1958.
Nishime EO, Cole CR, Blackstone EH, Pashkow FJ, Lauer MS. Heart rate
recovery and treadmill exercise score as predictors of mortality in patients
referred for exercise ECG. JAMA. 2000;284:1392–1398.
Shetler K, Marcus R, Froelicher VF, Vora S, Kalisetti D, Prakash M, Do
D, Myers J. Heart rate recovery: validation and methodologic issues. J Am
Coll Cardiol. 2001;38:1980 –1987.
Vivekananthan DP, Blackstone EH, Pothier CE, Lauer MS. Heart rate
recovery after exercise is a predictor of mortality, independent of the
angiographic severity of coronary disease. J Am Coll Cardiol. 2003;42:
831– 838.
Myers J, Prakash M, Froelicher VF, Do D, Partington S, Atwood JE.
Exercise capacity and mortality among men referred for exercise testing.
N Engl J Med. 2002;346:793– 801.
La Rovere MT, Bersano C, Gnemmi M, Specchia G, Schwartz PJ.
Exercise-induced increase in baroreflex sensitivity predicts improved
prognosis after myocardial infarction. Circulation. 2002;106:945–949.
Heart Failure
Viral Persistence in the Myocardium Is Associated With
Progressive Cardiac Dysfunction
Uwe Kühl, PhD; Matthias Pauschinger, MD; Bettina Seeberg, MD;
Dirk Lassner, PhD; Michel Noutsias, MD; Wolfgang Poller, MD; Heinz-Peter Schultheiss, MD
Background—Cardiotropic viral infections have been suspected as one possible cause of myocarditis and dilated
cardiomyopathy. Although adverse outcomes in dilated cardiomyopathy patients have been documented, the natural
course of heart diseases caused by cardiotropic viruses is unknown.
Methods and Results—Consecutive patients (n⫽172) with biopsy-proven viral infection in endomyocardial biopsies
(EMBs) were followed up by reanalysis of EMBs and hemodynamic measurements after a median period of 6.8 months
(range, 5.4 to 11.9). Nested polymerase chain reaction (PCR) and reverse transcription–PCR were performed to analyze
the genomic sequences. Myocardial inflammation was assessed by histology and immunohistology. At baseline, 32.6%
of EMBs in the study group contained enteroviral (EV) RNA, 8.1% adenovirus (ADV) DNA, 36.6% parvovirus B19
(PVB19) DNA, and 10.5% human herpesvirus type 6 (HHV6) DNA. In 12.2% of the samples, dual infection with
PVB19 and HHV6 was present. Follow-up analysis of EMBs by PCR documented spontaneous clearance of viral
genomes in 36.2% (55/151) of all patients with single infections. Virus-specific clearance rates were 50% for EV, 35.7%
for ADV, 22.2% for PVB19, and 44.4% for HHV6. In patients with dual infection with PVB19⫹ and HHV6⫹-, HHV6
was cleared in 42.8% (9/21), whereas PVB19 persisted in all 21 patients. Clearance of viral genomes was associated
with a significant improvement in left ventricular ejection fraction (LVEF), improving from 50.2⫾19.1% to
58.1⫾15.9% (P⬍0.001). In contrast, LV function decreased in patients with persisting viral genomes (LVEF,
54.3⫾16.1% versus 51.4⫾16.1%, P⬍0.01).
Conclusions—In this first biopsy-based analysis of the course of viral heart disease, we show that EV, ADV, PVB19, and
HHV6 persistence detected in the myocardium of patients with LV dysfunction was associated with a progressive
impairment of LVEF, whereas spontaneous viral elimination was associated with a significant improvement in LV
function. (Circulation. 2005;112:1965-1970.)
Key Words: heart diseases 䡲 myocarditis 䡲 cardiomyopathy 䡲 polymerase chain reaction 䡲 immunohistochemistry
C
ardiotropic viral infections are important causative factors in dilated cardiomyopathy (DCM), which appears to
occur as a late sequela of acute viral myocarditis.1– 4 In the
past, mostly enteroviruses (EVs) have been identified5–7 and
are associated with unfavorable clinical and hemodynamic
outcomes.3,8,9 Recently, other viral genomes have been detected in endomyocardial biopsies (EMB) from adults who
presented with the clinical phenotype of acute or chronic
myocarditis and DCM. Among identified viral genomes that
have been reported in EMBs of ⬇67% of patients with this
clinical setting, parvovirus B19 (PVB19) and human herpesvirus type 6 (HHV6) are the most frequently encountered
pathogens.3,4,10 The natural course and possible relevance of
persistent viral infection for improvement, persistence, or
progression of myocardial dysfunction are currently unknown. The present study was a biopsy-based analysis of the
spontaneous course of cardiac infections with various viruses
in follow-up biopsies of patients with regionally or globally
impaired myocardial function.
Methods
Patients
Between July 2001 and September 2004, 841 patients were admitted
to our institution for EMB to further elucidate a possible inflammatory and/or infectious cause of their disease because the clinical
presentation had suggested myocarditis in the past or DCM. In this
study, we enrolled 172 consecutive patients in whom polymerase
chain reaction (PCR) analysis had detected viral genomes in the
biopsy sample at the initial clinical presentation. Patients clinically
presenting with acute myocarditis of recent onset with signs of
myocardial injury (eg, mimicking acute myocardial infarction) were
not included. The majority of enrolled patients (89%) complained of
symptoms of moderate heart failure with fatigue, reduced physical
capacity, or dyspnea on exertion. Most patients were in New York
Heart Association classes II and III (II, 68%; III, 30%; and IV, 2%).
Received March 9, 2005; revision received June 10, 2005; accepted June 24, 2005.
From Charite, Universitätsmedizin Berlin, Campus Benjamin Franklin, Medizinische Klinik II, Abteilung für Kardiologie und Pneumologie, Berlin,
Germany.
Reprints requests to Uwe Kühl, PhD, MD, Charite, Universitätsmedizin Berlin, Campus Benjamin-Franklin, Medizinische Klinik II, Abteilung für
Kardiologie und Pneumologie, Hindenburgdamm 30, 12200 Berlin, Germany. E-mail [email protected]
1965
1966
Circulation
September 27, 2005
TABLE 1.
Baseline Characteristics
No.
Median age (range), y
TABLE 2.
172
46.5 (55.6–34.4)
Clinical Presentation
No.
Fatigue
172
110 (64.0)
Male
90 (54.6)
Angina at rest
62 (36.0)
White
172 (100)
Angina on exertion
32 (18.6)
Median time from onset of symptoms to EMB
(range), mo
5.1 (1.5–24.1)
Dyspnea on exertion
Preceding infection
88 (51.2)
Palpitations
Interval between infection and symptoms, wk
5.5 (1.9, 20.7)
Dizziness
Pericardial effusion, %
12 (7.0)
Syncope
22 (12.8)
Arrhythmias
77 (44.9)
Peripheral edema
Median systolic BP, mm Hg
120 (110, 130)
Median diastolic BP, mm Hg
75.5 (70, 80)
Medication use
Glycosides
ACE inhibitors or ARBs
73 (42.4)
140 (81.4)
␤-Blockers
94 (54.7)
Diuretics
96 (55.8)
Spironolactone
70 (40.7)
Antithrombotic agents
48 (27.9)
Amiodarone
21 (12.2)
ICD/pacemaker
12 (7.0)/8 (4.6)
RV/LV bundle block
13 (7.5)/23 (13.4)
Cardiac function parameters
Median LVEDP, mm Hg
Median EF, %
9 (6, 14)
3.3 (2.7, 4.0)
Median stroke volume index, mL/min
45 (36, 55)
Median PC, mm Hg
8 (6, 11)
Median PAP, mm Hg
14 (11, 18)
Echocardiography
Median left atrial dimension, mm
38 (34, 43)
Median LVEDD, mm
57 (51, 64)
Median LVESD, mm
39 (31, 53)
Median fractional shortening, %
29 (19, 38)
Global wall-motion abnormality
Regional wall-motion abnormality
13 (7.6)
118 (68.6)
30 (17.4)
Sinus tachycardia (⬎100 bpm)
15 (8.7)
Atrial fibrillation
28 (16.3)
Supraventricular extra beats
12 (7.0)
Ventricular extra beats
23 (13.4)
Ventricular tachycardia (nonsustained)
8 (4.7)
Data are presented as No. (%) of subjects.
eters were measured by M-mode echocardiography in the parasternal
long-axis view according to the leading-edge method. Percentage
fractional shortening was calculated in a standardized manner.
Medication use, including angiotensin-converting enzyme inhibitors,
␤-blockers, diuretics, cardiac glycosides, and warfarin, was stable
throughout the follow-up period without significant differences
between patients who developed virus genome persistence and those
who did not.
52.5 (37.2, 66.0)
Median cardiac index, L䡠min⫺1䡠m⫺2 BSA
Median ES, mm
109 (61.6)
9 (3, 17)
112 (65.1)
EMB and Right-Heart Catheterization
Eight EMBs were obtained from the right side of the ventricular
septum of each patient with use of a flexible bioptome (Westmed)
via the femoral vein approach. There were no biopsy-related adverse
events. Two EMBs were used for histological evaluation according
to the Dallas criteria11 and immunohistochemistry,6 whereas the
remaining 4 EMBs were subjected to DNA and RNA extraction for
amplification of the viral genomes. After the EMBs were obtained,
the patients underwent right heart catheterization. Right atrial, right
ventricular, pulmonary arterial, and pulmonary capillary wedge
pressures (all in mm Hg) and cardiac index (L · min⫺1 · m⫺2 body
surface area) were recorded. The protocol was approved by the
Human Research Committee of the Charite, Campus Benjamin
Franklin, Berlin, and all patients gave written, informed consent
before treatment.
60 (34.9)
BP indicates blood pressure; ACE, angiotensin-converting enzyme; ARB,
angiotensin receptor blocker; ICD, implantable cardioverter/defibrillator; BSA,
body surface area; EDP, end-diastolic pressure; PC, pulmonary capillary wedge
pressure; PAP, pulmonary artery pressure; and ES, mitral valve E-point to
septal separation. Data are presented as the median value (range), median
value (25th, 75th percentiles), or No. (%) of subjects.
The demographic and clinical characteristics of patients are shown in
Tables 1 and 2.
Coronary artery disease and other possible causes of myocardial
dysfunction had been excluded by angiography before biopsy in all
patients. At the baseline biopsy, the patients presented with either
regional (35%) or global (65%) wall motion abnormalities. To
determine the course of the viral infection, all patients underwent
follow-up biopsy. Ejection fraction (EF) was determined angiographically. Additionally, standard 2-dimensional and M-mode
echocardiography was performed for all patients in our echocardiographic department 1 day before biopsy (median interval, 6.8
months; range, 5.4 to 11.9). For each echocardiogram, left ventricular (LV) end-diastolic (LVEDD) and end-systolic (LVESD) diam-
Etiologic Investigations
Detection of Viral Genomes by Nested PCR
Detection of viral genomes by nested PCR was carried out as
published recently.3,4,12 In brief, nested PCR/reverse transcription–
PCR was performed on RNA extracted from EMBs for EVs
(including coxsackieviruses and echoviruses) and in DNA for
adenovirus (ADV), PV, and HHV6. As a control for successful
extraction of DNA and RNA from heart muscle tissue, oligonucleotide sequences were chosen from the DNA sequence of the
glyceraldehyde 3-phosphate dehydrogenase gene. Specificity of all
amplification products was confirmed by automatic DNA
sequencing.12–14
Histological and Immunohistological Evaluation
Myocardial inflammation was defined by the detection of infiltrating
lymphocytes (median cell count ⬎7.0 cells/mm2) in association with
enhanced expression of cellular adhesion molecules (HLA-I/II or
CD54) expressed on interstitial or endothelial cells.15,16 Samples
containing low numbers of infiltrating lymphocytes (median cell
count ⬍7.0 cells/mm2), especially those without enhanced cellular
Kühl et al
Natural Course of Virus-Associated Heart Disease
TABLE 3. Distribution of Virus Genomes at Baseline
and Follow-Up
No. of Subjects
(N⫽172)
Virus
Clearance
PVB19
63 (36.6)
14/63 (22.2)
EV
56 (32.6)
28/56) (50.0)
HHV6
18 (10.5)
8/18 (44.4)
ADV
14 (8.1)
5/14 (35.7)
PVB19⫹HHV6
21 (12.1)
9/21 (42.8)
Data are presented as No. (%) of subjects.
adhesion molecule expression, were defined as having no significant
myocardial inflammation.
Statistical analysis was performed with JMP Statistical Discovery
Software, version 3.1.6 (SAS Institute, Inc). All results are presented
as median value (25th, 75th percentile), except when stated otherwise. Follow-up data were analyzed with a paired t test. Qualitative
data were compared by the ␹2 test. A probability value (2 sided)
⬍0.05 was considered statistically significant.
Results
Biopsy Findings
At baseline, 63 (36.6%) of the 172 patients’ EBMs were
positive for PVB19, 56 (32.6%) for EV, 18 (10.5%) for
HHV6, and 14 (8.1%) for ADV. Dual infection with PVB19
and HHV6 was present in 21 (12.2%) biopsy specimens
(Table 3). The spontaneous course of viral infections in these
172 patients was followed up for a median of 6.8 months
(range, 5.4 to 11.9). At the time of the follow-up biopsy,
spontaneous clearance of viral genomes was found in 55 of
151 (36.4%) patients with single infections (EV n⫽28
[50.0%], ADV n⫽5 [35.7%], PVB19 n⫽14 ([22.2%], and
HHV6 n⫽8 [44.4%]) (Figure 1). In patients with PVB19 and
HHV6 dual infections, the HHV6 infection had been cleared
in 42.8% (n⫽9), whereas PVB19 genomes persisted in all 21
cases.
Histological analysis did not detect active or borderline
myocarditis in any of the analyzed samples at baseline or
follow-up. On immunohistological staining, significant CD3⫹
T-lymphocytic infiltrates with a median number of 11.6 (8.4
to 17.2) CD3⫹-positive lymphocytes/mm2 in association with
enhanced cellular adhesion molecule expression16 were detected in 67 patients’ (38.9%) baseline biopsy samples
(versus 2.8 cells/mm2 [1.8 to 5.3] in the remaining 105
patients). At follow-up, enhanced myocardial inflammation
was present in EMBs of 34/172 (19.8%) patients (CD3⫹, 10.9
cells/mm2 [8.4 to 13.5] versus 2.8 cells/mm2 [1.8 to 4.6]), and
increased numbers of inflammatory lymphocytes were detected more frequently in patients who developed virus
persistence (26/108 [24.1%] versus 8/64 [12.5%], P⬍0.05).
Enhanced HLA-I/DR and CD54 expression was significantly
correlated with infiltrating inflammatory cells, but it was
independent of the course of viral infection (data not shown).
1967
Hemodynamic Course
Regardless of the virus involved, complete clearance of viral
genomes (n⫽64) was associated with a significant improvement in LVEF, improving from 50.2⫾19.1% to 58.1⫾15.9%
(P⬍0.001, Figure 1). An increase in systolic LV function was
found to be independent of the infectious agent. In contrast to
the favorable hemodynamic course of patients who eliminated the viral genomes, virus persistence was associated
with a lack of hemodynamic improvement. Between baseline
and follow-up, LVEF significantly decreased from
54.3⫾16.1% to 51.4⫾16.1% (P⬍0.01) in these patients
(n⫽108), despite the relative short follow-up period of 6.8
months. Hemodynamic changes after spontaneous HHV6
clearance but PVB19 persistence were not significant
(P⫽0.49) in patients with PVB19/HHV6 dual infection,
whereas persistence of both viruses was associated with a
mild progression of LV dysfunction (P⫽0.06). Hemodynamic improvement occurred in patients with both mild and
severe LV dysfunction. The improvement was more pronounced in patients with an EF ⬍45% (n⫽51) compared with
patients with an EF ⬎45% (Figure 2). In this subgroup, EF
improved from 29.6⫾7.8% to 44.0⫾13.6% (P⬍0.001,
n⫽24) in association with virus elimination, whereas EF did
not change in patients who developed viral persistence
(32.4⫾8.4% versus 33.9⫾15.8%, P⫽0.57). In patients with
an EF ⬎45% (n⫽121), EF improved from 62.6⫾11.5% to
66.6⫾10.1% (P⬍0.01, n⫽40) or deteriorated from
61.6⫾10.4% to 57.2⫾11.2% (P⬍0.001, n⫽81). The aforementioned hemodynamic changes were independent of the
patients’ medication regimen, which did not differ significantly between the virus-positive and virus-negative cohorts
and that had been kept constant during the follow-up period.
In contrast to the viral course, changes in myocardial inflammatory cells were not additional predictors of the hemodynamic course.
Discussion
Relation Between Clinical and Virological Course
A broad spectrum of viral genomes has been detected in
EMBs from patients with clinically suspected myocarditis in
the past and DCM. So far, EVs have been linked to the
development of myocarditis and its progression to DCM.1,2
Recently, we and others have detected other frequent viral
genomes (eg, ADV, PVB19, and HHV6) in the myocardium
of patients presenting with acute heart failure caused by
myocarditis, with a sudden onset of cardiac symptoms mimicking acute myocardial infarction and with chronic LV
dysfunction diagnosed as “idiopathic” DCM.3–7,17 The natural
course of these viral infections and the prevalence of viral
persistence in these groups have not been investigated yet. To
address this issue and to elucidate the relevance of virus
persistence with respect to LV function, we conducted
follow-up EMBs in 172 consecutive, virus-positive patients
with persistent LV dysfunction.
During follow-up of patients with clinically suspected
myocarditis in the past or with heart failure of unknown
origin, one may observe either “spontaneous” improvement
or progression of ventricular dysfunction despite constant
1968
Circulation
September 27, 2005
Figure 1. Hemodynamic course in 172 patients during a median follow-up of 6.8 months. Spontaneous virus clearance was associated
with improvement in LVEF. A lack of improvement or deterioration in LVEF was observed in patients with viral persistence, with virusspecific differences.
heart failure medication. Our results suggest that these
“spontaneous” changes in cardiac function may actually
reflect the dynamic course of an underlying cardiotropic viral
infection. As shown here, virus clearance was associated with
a spontaneous improvement in LVEF, regardless of the type
of virus involved. It appears that patients with a lower EF
(⬍45%) improve more than do those with milder EF dysfunction. This is reminiscent of a similar phenomenon in a
prior study of interferon-␤–induced virus elimination.12
LVEF did not improve or even deteriorated in patients with
viral persistence. Taking into consideration the slow but
continuous development of LV dysfunction in DCM, the even
mild deterioration of LVEF observed during the short
follow-up period of 6.8 months may progress to substantial
myocardial dysfunction over years.
Molecular Pathomechanisms of Viral
Heart Disease
Figure 2. Comparison of the hemodynamic changes in patients
with an EF below (n⫽51) and above (n⫽121) 45%. Hemodynamic
improvement is more pronounced in the patient group with the
lower EF and spontaneous virus clearance vs virus persistence.
Even small amounts of persistent viral genomes may cause
further progression of the disease, by direct cytopathic effects
of virus-encoded proteins via virus-associated signaling pathways resulting in the release of cytokines,18 –25 by alterations
of the extracellular cardiac matrix or the cytoskeleton,26 –28 or
by chronic myocardial inflammation.15,16,29,30 So far, the
causes of the highly variable natural courses of virusassociated heart disease are unknown but may include
changes in cardiac virus load, as well as the host’s primary
immune responses to the virus.
Kühl et al
Natural Course of Virus-Associated Heart Disease
If the viruses were cleared spontaneously and thus, no
pathogenic agents were detected in the myocardium, diagnostic procedures should result in resolved myocarditis or “idiopathic” DCM. In patients with “resolved” myocarditis, ventricular function may recover completely if the initial
myocardial damage was minor. In other patients, persistence
or further progression of ventricular dysfunction may result
from myocardial remodeling after the initial virus-induced
injury of cardiac tissues.
Myocardial inflammation was detected in 40% of the
virus-positive patients during baseline EMB. This inflammation was significantly reduced at follow-up but still primarily
seen in virus-positive patients (23.9% versus 12.5%, P⬍0.05,
respectively). Further follow-up of virus-negative patients
with inflammation would be required to distinguish an
inflammatory process resolving after virus clearance from
virus-induced persistent inflammation, often referred to as
and indistinguishable from (auto)immune myocarditis or
chronic inflammatory cardiomyopathy.
5.
6.
7.
8.
9.
10.
Conclusions
A broad spectrum of viral genomes has been detected in
patients with de novo wall motion abnormalities or persistent LV dysfunction, clinically often referred to as past
myocarditis or DCM. The influence of this chronic viral
infection on myocardial function is unknown, because
biopsy-based follow-up data on patients infected with
these viruses have never been obtained. By following up a
large cohort of patients with different virus infections, we
could show that spontaneously occurring virus clearance is
associated with spontaneous hemodynamic improvement.
In contrast, LV function deteriorates in patients with virus
persistence, even within a short follow-up period of 6.8
months and despite constant heart failure medication.
These data indicate that persisting cardiac viral infections
may constitute a major cause of progressing LV dysfunction in patients with clinically suspected past myocarditis
or DCM. The data furthermore indicate that only an
EMB-derived virus analysis allows accurate diagnosis in
patients with clinically suspected myocarditis or DCM,
which is mandatory for effective antiviral immunomodulatory treatment of these patients.
Acknowledgment
This work was supported in part by the Deutsche Forschungsgemenschaft through SFB/Transregio 19.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
References
1. Feldmann AM, McNamara D. Myocarditis. N Engl J Med. 2000;343:
1388 –1398.
2. Richardson P, McKenna W, Bristow M, Maisch B, Mautner B, J OC,
Olsen E, Thiene G, Goodwin J, Gyarfas I, Martin I, Nordet P. Report of
the 1995 World Health Organization/International Society and Federation
of Cardiology Task Force on the Definition and Classification of cardiomyopathies. Circulation. 1996;93:841– 842.
3. Kühl U, Pauschinger M, Bock T, Klingel K, Schwimmbeck PL, Seeberg
B, Krautwurm L, Schultheiss HP, Kandolf R. Parvovirus B19 infection
mimicking acute myocardial infarction. Circulation. 2003;108:945–950.
4. Kühl U, Pauschinger M, Michel Noutsias M, Seeberg B, Bock T, Lassner
DPW, Kandolf R, Schultheiss HP. High prevalence of viral genomes and
21.
22.
23.
1969
multiple viral infections in the myocardium of adults with “idiopathic”
left ventricular dysfunction. Circulation. 2005;111:887– 893.
Baboonian C, Treasure T. Meta-analysis of the association of enteroviruses with human heart disease. Heart. 1997;78:539 –543.
Pauschinger M, Doerner A, Kuehl U, Schwimmbeck PL, Poller W,
Kandolf R, Schultheiss HP. Enteroviral RNA replication in the myocardium of patients with left ventricular dysfunction and clinically suspected
myocarditis. Circulation. 1999;99:889 – 895.
Bowles NE, Ni J, Kearney DL, Pauschinger M, Schultheiss HP,
McCarthy R, Hare J, Bricker JT, Bowles KR, Towbin JA. Detection of
viruses in myocardial tissues by polymerase chain reaction: evidence of
adenovirus as a common cause of myocarditis in children and adults.
J Am Coll Cardiol. 2003;42:466 – 472.
Why HJ, Meany BT, Richardson PJ, Olsen EG, Bowles NE, Cunningham L, Freeke CA, Archard LC. Clinical and prognostic significance of detection of enteroviral RNA in the myocardium of patients
with myocarditis or dilated cardiomyopathy. Circulation. 1994;89:
2582–2589.
Frustaci A, Chimenti C, Calabrese F, Pieroni M, Thiene G, Maseri A.
Immunosuppressive therapy for active lymphocytic myocarditis:
virologic and immunologic profile of responders versus nonresponder.
Circulation. 2003;107:857– 863.
Tschope C, Bock CT, Kasner M, Noutsias M, Westermann D,
Schwimmbeck PL, Pauschinger M, Poller WC, Kuhl U, Kandolf R,
Schultheiss HP. High prevalence of cardiac parvovirus B19 infection in
patients with isolated left ventricular diastolic dysfunction. Circulation.
2005;111:879 – 886.
Aretz HT, Billingham ME, Edwards WD, Factor SM, Fallon JT, Fenoglio
JJ Jr, Olsen EG, Schoen FJ. Myocarditis: a histopathologic definition and
classification. Am J Cardiovasc Pathol. 1987;1:3–14.
Kühl U, Pauschinger M, Schwimmbeck PL, Seeberg B, Lober C,
Noutsias M, Poller W, Schultheiss HP. Interferon-␤ treatment eliminates
cardiotropic viruses and improves left ventricular function in patients
with myocardial persistence of viral genomes and left ventricular dysfunction. Circulation. 2003;107:2793–2798.
Smith LM, Sanders JZ, Kaiser RJ, Hughes P, Dodd C, Connell CR.
Flourescence detection in automated DNA sequence analysis. Nature.
1986;321:674 – 679.
Deiss V, Tratschin JD, Weitz M, Siegl G. Cloning of the human parvovirus B19 genome and structural analysis of its palindromic termini.
Virology. 1990;175:247–254.
Kühl U, Noutsias M, Seeberg B, Schultheiss HP. Immunohistological
evidence for a chronic intramyocardial inflammatory process in dilated
cardiomyopathy. Heart. 1996;75:295–300.
Noutsias M, Seeberg B, Schultheiss HP, Kuhl U. Expression of cell
adhesion molecules in dilated cardiomyopathy: evidence for endothelial
activation in inflammatory cardiomyopathy. Circulation. 1999;99:
2124 –2131.
Bueltmann BD, Klingel K, Soltar K, Bock CT, Baba HA, Sauter M,
Kandolf R. Fatal parvovirus B19-associated myocarditis clinically mimicking ischemic heart disease: an endothelial cell-mediated disease.
Human Pathol. 2002;34:92–95.
Matsumori A, Yamada T, Suzuki H, Matoba Y, Sasayama S. Increased
circulating cytokines in patients with myocarditis and cardiomyopathy. Br
Heart J. 1994;72:561–566.
Shioi T, Matsumori A, Sasayama S. Persistent expression of cytokine in
the chronic stage of viral myocarditis in mice. Circulation. 1996;94:
2930 –2937.
Andreoletti L, Hober D, Becquart P, Belaich S, Copin MC, Lambert V,
Wattre P. Experimental CVB3-induced chronic myocarditis in two
murine strains: evidence of interrelationships between virus replication
and myocardial damage in persistent cardiac infection. J Med Virol.
1997;52:206 –214.
Kelly RA, Smith TW. Cytokines and cardiac contractile function. Circulation. 1997;95:778 –781.
Bozkurt B, Kribbs SB, Clubb FJ Jr, Michael LH, Didenko VV, Hornsby
PJ, Seta Y, Oral H, Spinale FG, Mann DL. Pathophysiologically relevant
concentrations of tumor necrosis factor-␣ promote progressive left ventricular dysfunction and remodeling in rats. Circulation. 1998;97:
1382–1391.
Yamamoto N, Shibamori M, Ogura M, Seko Y, Kikuchi M. Effects of
intranasal administration of recombinant murine interferon-␥ on murine
acute myocarditis caused by encephalomyocarditis virus. Circulation.
1998;97:1017–1023.
1970
Circulation
September 27, 2005
24. Horwitz MS, La Cava A, Fine C, Rodriguez E, Ilic A, Sarvetnick N.
Pancreatic expression of interferon-␥ protects mice from lethal coxsackievirus B3 infection and subsequent myocarditis [see comments]. Nat
Med. 2000;6:693– 697.
25. Bevan AL, Zhang H, Li Y, Archard LC. Nitric oxide and Coxsackievirus
B3 myocarditis: differential expression of inducible nitric oxide synthase
in mouse heart after infection with virulent or attenuated virus. J Med
Virol. 2001;64:175–182.
26. Badorff C, Fichtlscherer B, Rhoads RE, Zeiher AM, Muelsch A,
Dimmeler S, Knowlton KU. Nitric oxide inhibits dystrophin proteolysis
by coxsackieviral protease 2A through S-nitrosylation: a protective
mechanism against enteroviral cardiomyopathy. Circulation. 2000;102:
2276 –2281.
27. Pauschinger M, Doerner A, Remppis A, Tannhauser R, Kuhl U, Schultheiss HP. Differential myocardial abundance of collagen type I and type
III mRNA in dilated cardiomyopathy: effects of myocardial inflammation. Cardiovasc Res. 1998;37:123–129.
28. Pauschinger M, Chandrasekharan K, Schultheiss HP. Myocardial
remodeling in viral heart disease: possible interactions between inflammatory mediators and MMP-TIMP system. Heart Fail Rev. 2004;9:
21–31.
29. Klingel K, Kandolf R. The role of enterovirus replication in the
development of acute and chronic heart muscle disease in different
immunocompetent mouse strains. Scand J Infect Dis Suppl. 1993;88:
79 – 85.
30. Badorff C, Lee GH, Lamphear BJ, Martone ME, Campbell KP, Rhoads
RE, Knowlton KU. Enteroviral protease 2A cleaves dystrophin: evidence
of cytoskeletal disruption in an acquired cardiomyopathy [see comments].
Nat Med. 1999;5:320 –326.
Role of the Protein Kinase C-⑀–Raf-1–MEK-1/2–p44/42 MAPK
Signaling Cascade in the Activation of Signal Transducers and
Activators of Transcription 1 and 3 and Induction of
Cyclooxygenase-2 After Ischemic Preconditioning
Yu-Ting Xuan, PhD; Yiru Guo, MD; Yanqing Zhu, MD; Ou-Li Wang, MD; Gregg Rokosh, PhD;
Robert O. Messing, MD; Roberto Bolli, MD
Background—Although Janus kinase (JAK)–mediated Tyr phosphorylation of signal transducers and activators of
transcription (STAT) 1 and 3 is essential for the upregulation of cyclooxygenase-2 (COX-2) and the cardioprotection
of late preconditioning (PC), the role of Ser phosphorylation of STAT1 and STAT3 in late PC and the upstream
signaling mechanisms responsible for mediating Ser phosphorylation of STAT1 and STAT3 remain unknown.
Methods and Results—In mice preconditioned with six 4-minute coronary occlusion/4-minute reperfusion cycles, we
found that (1) ischemic PC activates the Raf1–mitogen-activated protein kinase (MAPK)/extracellular signal–regulated
kinase kinase (MEK) 1/2–p44/42 MAPK signaling pathway, induces phosphorylation of STAT1 and STAT3 on the
Ser-727 residue, and upregulates COX-2 expression; (2) pSer-STAT1 and pSer-STAT3 form complexes with
pTyr-p44/42 MAPKs in preconditioned myocardium, supporting the concept that Ser phosphorylation of these 2 factors
is mediated by activated p44/42 MAPKs; and (3) activation of the Raf-1-MEK-1/2–p44/42 MAPK-pSer-STAT1/3
pathway and induction of COX-2 during ischemic PC are dependent on protein kinase C (PKC)-⑀ activity, as determined
by both pharmacological and genetic inhibition of PKC⑀.
Conclusions—To our knowledge, this is the first study to demonstrate that ischemic PC causes Ser phosphorylation of
STAT1 and STAT3 and that this event is governed by PKC⑀ via a PKC⑀–Raf1-MEK1/2-p44/42 MAPK pathway.
Furthermore, this is the first report that COX-2 expression in the heart is controlled by PKC⑀. Together with our previous
findings, the present study implies that STAT-dependent transcription of the genes responsible for ischemic PC is
modulated by a dual signaling mechanism that involves both JAK1/2 (Tyr phosphorylation) and PKC⑀ (Ser
phosphorylation). (Circulation. 2005;112:1971-1978.)
Key Words: ischemia 䡲 myocardial infarction 䡲 signal transduction
T
he late phase of ischemic preconditioning (PC) is a
delayed protective adaptation whereby brief episodes of
ischemia enhance the resistance of the heart to ischemia/
reperfusion injury 12 to 72 hours later.1 It is now appreciated
that the development of late PC after the ischemic PC
challenge (on day 1) occurs via activation of various signaling molecules, including protein kinase C (PKC)2– 4 and Janus
Tyr kinases (JAKs),5 which in turn activate latent transcription factors, including nuclear factor-␬B and signal transducers and activators of transcription (STATs),1,5,6 leading to the
upregulation of cardioprotective genes such as inducible
nitric oxide synthase and cyclooxygenase-2 (COX-2).5,7–11
Activation of the JAK-STAT pathway is essential for the
development of late PC, as inhibition of this pathway results
in complete loss of protection against infarction.5,10,11 How-
ever, the exact mechanism responsible for the recruitment of
STATs after the PC ischemia remains incompletely understood. Furthermore, little is known about the mechanism by
which ischemic PC regulates COX-2.
Our previous studies have shown that pretreatment with the
JAK inhibitor AG-490 before ischemic PC blocked both the
Tyr phosphorylation and activation of STAT1 and STAT3
and the subsequent upregulation of COX-2 protein, indicating
a necessary role for STAT1 and STAT3 Tyr phosphorylation
in the induction of COX-2.5,10 Tyr phosphorylation of
STAT1/3 is known to result in dimerization, nuclear transport, and transactivation of STAT-responsive genes.12 However, full transcriptional activation of STATs requires not
only Tyr phosphorylation (Tyr-701 in STAT1 and Tyr-705 in
STAT3) but also Ser phosphorylation (Ser-727 in both
Received May 11, 2005; accepted June 10, 2005.
From the Institute of Molecular Cardiology (Y.-T.X., Y.G., Y.Z., O.-L.W., G.R., R.B.), University of Louisville, Louisville, Ky, and the Ernest Gallo
Clinic and Research Center (R.O.M.), Department of Neurology, University of California at San Francisco, Emeryville, Calif.
The online-only Data Supplement, which contains Figures I through VII and Table I, can be found at http://circ.ahajournals.org/cgi/
content/full/CIRCULATIONAHA.105.561522/DC1.
Correspondence to Roberto Bolli, MD, Division of Cardiology, University of Louisville, Louisville, KY 40292. E-mail [email protected]
1971
1972
Circulation
September 27, 2005
STAT1 and STAT3).12–14 At present, nothing is known about
the role of Ser-727 phosphorylation of STAT1/3 in the
upregulation of cardiac COX-2 and in the delayed cardioprotection of late PC. More generally, the role of Ser-727
phosphorylation in STAT-dependent transactivation in the
heart is unknown.
The ⑀ isoform of PKC is known to play a crucial role in the
protective effects of late PC2– 4 and to induce activation of
p44/42 mitogen-activated protein kinases (MAPKs) in conscious rabbits.15 The Ser-727 residue of STAT1/3 is a
recognition site for p44/42 MAPKs,16 suggesting a link
among activation of PKC⑀ and p44/42 MAPKs and Ser-727
phosphorylation of STAT1/3. The Ser/Thr kinase Raf-1 is
known to phosphorylate MAPK/extracellular signal–regulated kinase kinase (MEK) 1/2, leading to activation of
p44/42 MAPKs.17 Because the promoter of the mouse COX-2
gene contains the interferon-␥ activation site (GAS) consensus sequence for the binding of STATs5,18 and because our
previous studies have shown that ischemic PC activates
PKC⑀,2 JAK1/2,5,10 and STAT1/3,5,10 we hypothesized that
ischemic PC upregulates COX-2 protein expression via rapid
activation of PKC⑀, which in turn activates a downstream
pathway that includes Raf-1, MEK-1/2, and p44/42 MAPKs,
leading to Ser phosphorylation of STAT1 and STAT3,
transcription of the COX-2 gene, and cardioprotection.
The overall objective of the present study was to test this
hypothesis. The following specific questions were addressed:
(1) Does ischemic PC induce Ser-727 phosphorylation of
STAT1/3? (2) If so, does ischemic PC activate the Raf-1–
MEK-1/2–p44/42 MAPK pathway? (3) Does pharmacological or genetic inhibition of PKC⑀ prevent activation of the
Raf-1–MEK-1/2–p44/42 MAPK pathway? The results demonstrate, for the first time, that ischemic PC causes Ser
phosphorylation of STAT1/3 and activation of the Raf-1–
MEK-1/2–p44/42 MAPK signaling pathway and that both of
these events, as well as the subsequent induction of COX-2,
are dependent on PKC⑀.
Methods
Animal Care
PKC⑀⫺/⫺ mice and their wild-type (WT) littermates were generated
by intercrossing 129SvJae⫻C57BL/6 hybrid PKC⑀⫹/⫺ mice.19 All
mice were maintained in sterile microisolator cages under pathogenfree conditions. Food and water were autoclaved, and all handling
was done under a laminar-flow hood according to standard procedures for maintaining pathogen-free transgenic mice. The mice were
genotyped by polymerase chain reaction, as previously described,
with DNA prepared from tissue samples taken at the end of the
experiments.9,10
Experimental Preparation
The murine model of late PC has been previously described in
detail.5,9 In brief, mice were anesthetized and ventilated. After
administration of antibiotics, the chest was opened through a midline
sternotomy, and a nontraumatic balloon occluder was implanted
around the mid-left anterior descending coronary artery by using an
8-0 nylon suture. Ischemic PC was elicited by a sequence of six
4-minute coronary occlusion/4-minute reperfusion (O/R) cycles.5,9
To prevent hypotension, blood from a donor mouse was given during
surgery.5,9 Rectal temperature was maintained close to 37°C
throughout the experiment.5,9 The investigation consisted of 2
successive phases (A and B). The objective of phase A was to
determine whether ischemic PC activates the Raf-1–MEK1/2–
p44/42 MAPK signaling pathway and induces Ser phosphorylation
of STAT1/3 and subsequent upregulation of COX-2. The objective
of phase B was to determine whether the activation of this pathway
is dependent on PKC⑀.
Phase A
Mice were assigned to 12 groups (Data Supplement Figure I; see
http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.
105.561522/DC1). Control groups (I, IV, VIII, and X) underwent 1
hour of the open-chest state without coronary occlusion. The PC
groups (II, V, and IX) (PC-30⬘) underwent a sequence of six
4-minute coronary O/R cycles (a protocol that induces late PC5,9).
Groups III and VI (CHE⫹PC-30⬘) received the PKC inhibitor
chelerythrine (0.5 mg/kg IV dissolved in dimethyl sulfoxide) 5
minutes before the first occlusion, whereas group VII (dimethyl
sulfoxide⫹PC-30⬘) received dimethyl sulfoxide 5 minutes before the
first occlusion. All mice in groups I to IX were euthanized 30
minutes after the last reperfusion. Group XI (PC-24 hours) underwent six 4-minute coronary O/R cycles with no treatment, whereas
group XII (CHE⫹PC-24 hours) received chelerythrine 5 minutes
before the first occlusion. Mice in groups X to XII were euthanized
24 hours after the last occlusion or after sham coronary occlusion.
Myocardial samples were rapidly removed from the ischemic/
reperfused region or the left ventricle and frozen in LN2 until used.
Phase B
Mice were assigned to 8 groups (Data Supplement Figure II). Groups
XIV (WT PC-30⬘), XVI (PKC⑀⫺/⫺ PC-30⬘), XVIII (WT PC-24
hours), and XX (PKC⑀⫺/⫺ PC-24 hours) underwent a sequence of six
4-minute coronary O/R cycles, whereas groups XIII (WT control30⬘), XV (PKC⑀⫺/⫺ control-30⬘), XVII (WT control-24 hours), and
XIX (PKC⑀⫺/⫺ control-24) underwent 1 hour of the open-chest state
without coronary occlusion. Mice were euthanized either 30 minutes
(groups XIII–XVI) or 24 hour (groups XVII–XX) after the sham
coronary occlusion or the last reperfusion. In all groups, myocardial
samples were rapidly removed from the ischemic/reperfused region
or the left ventricle and frozen in LN2 until used.
Preparation of Cytosolic, Membranous, and
Nuclear Fractions
Cytosolic, membranous, and nuclear fractions were prepared from
heart samples as previously described.5,9,10
Western Immunoblotting
Western immunoblotting analysis was performed with standard
sodium dodecyl sulfate–polyacrylamide gel electrophoresis Western
immunoblotting techniques as previously described.5,9,10 Details
about the antibodies are provided in Data Supplement Table I. Equal
loading was confirmed by staining with Ponceau-S.5,9,10
Coimmunoprecipitation
Myocardial homogenates (600 ␮g) were incubated with specific
monoclonal anti-pTyr-p44/42 MAPK antibodies for 4 hours, followed by addition of protein G PLUS-Agarose-S (Santa Cruz)
overnight at 4°C as per a previously described method.5,10 After
extensive washing, the anti-pTyr-p44/42 MAPK precipitates were
subjected to immunoblotting with anti-pSer727-STAT1 or antipSer727-STAT3 antibodies.
Electrophoretic Mobility Shift Assays
The DNA binding activity of STAT1/3 was measured with electrophoretic mobility shift assays (EMSAs) as previously described.5 A
synthetic double-stranded probe with the sequence 5⬘GATCAGCTTCATTTCCCGTAAATCCCTA-3⬘ (Gibco) was endlabeled with [␥-32P]ATP (3000 Ci/mmol, Amersham) and T4
polynucleotide kinase and purified on a G-25 Sephadex column
(Pharmacia). This oligonucleotide has the consensus sequence for
GAS elements, as indicated by italics.5,10,20
Xuan et al
Activation of STATs by Preconditioning
1973
Figure 1. Effect of ischemic PC on phosphorylation of Raf-1, MEK-1/2, and p44/42 MAPKs. Cytosolic proteins of myocardial samples
were prepared from control mice that underwent 1 hour of open-chest state without coronary O/R (control group) or from the ischemic/
reperfused region of PC mice that received either no treatment (PC-30⬘) or chelerythrine (CHE⫹PC-30⬘) 5 minutes before 6 cycles of
4-minute coronary O/R. All mice were euthanized 30 minutes after the sham operation or the sixth reperfusion. The figure illustrates
representative immunoblots (A, C, and E) and densitometric analysis of total and Ser-phosphorylated Raf-1 (B), total and Ser-phosphorylated MEK-1/2 (D), and total and Tyr-phosphorylated p44/42 MAPKs (F). Data are mean⫾SEM, n⫽6/group.
Data are reported as mean⫾SEM. Measurements were analyzed by
ANOVA followed by unpaired Student t tests with the Bonferroni
correction. In all Western blot analyses, the content of the specific
protein of interest was expressed as a percentage of the corresponding protein in the anterior left ventricular wall of control mice.5,9,10
Results
A total of 132 mice (20 groups) were used.
Ischemic PC Induces Phosphorylation of Raf-1,
MEK-1/2, and p44/42 MAPKs; Ser
Phosphorylation and DNA Binding of STAT1/3;
and Expression of COX-2
The sequence of six 4-minute coronary O/R cycles resulted in
a marked increase, 30 minutes later (30 minutes after the sixth
reperfusion), in pSer(338)-Raf-1 (190⫾8% of control [Figure
1A and 1B]), pSer-MEK-1/2 (274⫾32% of control [Figure
1C and 1D]), and pThr(202)/Tyr(204)-p44/42 MAPKs, as
detected both with an anti-pTyr(204)-p44/42 MAPK antibody
(393⫾63% of control [Figure 1E and 1F]) and an antipThr(202)/Tyr(204)-p44/42 MAPK antibody that recognizes
the dually phosphorylated form of the kinases (676⫾148% of
control). These data indicate that ischemic PC activates the
Raf-1–MEK-1/2–p44/42 MAPK signaling pathway. The six
4-minute coronary O/R cycles did not change the total levels
of Raf-1, MEK-1/2, and p44/42 MAPKs (Figure 1A through
1F). We could not detect phosphorylation of Raf-1 on
Tyr-341 by either immunoblotting with anti-pTyr(341)-Raf-1
antibodies (Data Supplement Figure III) or immunoprecipitation with anti–Raf-1 antibodies followed by immunoblotting with anti-pTyr(341)-Raf-1 antibodies (Data Supplement
Figure IV). The anti-pSer(338)-Raf-1 antibody (Cell Signaling) did not react with immunoprecipitated A-Raf (Data
Supplement Figure V), indicating that the immunoreactivity
observed in the samples cannot be ascribed to
pSer(338)-A-Raf.
In addition, the six 4-minute coronary O/R cycles caused a
marked increase, 30 minutes later, in the Ser-phosphorylated
forms of STAT1 and STAT3 in the nuclear fraction:
pSer(727)-STAT1, 331⫾37% of control (P⬍0.05 [Figure 2A
and 2B]) and pSer(727)-STAT3, 449⫾64% of control
(P⬍0.05 [Figure 2C and 2D]). The total nuclear levels of
STAT1 (243⫾17% of control, P⬍0.05 [Figure 2A and 2B])
and STAT3 (304⫾25% of control, P⬍0.05 [Figure 2C and
2D]) were also increased 30 minutes after the six 4-minute
coronary O/R cycles, indicating nuclear translocation of these
transcription factors. This translocation was associated with a
striking increase in the STAT1/3-GAS complex in the nuclear
fraction (637⫾30% of control, P⬍0.05) (Figure 2E and 2F),
indicating increased DNA binding activity of these factors,
and with a marked increase in myocardial COX-2 expression
24 hours later (374⫾36% of control, P⬍0.05) (Figure 2G and
2H). In the whole homogenate, ischemic PC increased
pSer(727)-STAT1 (Data Supplement Figure VIA) and
pSer(727)-STAT3 (Data Supplement Figure VIIA) but did
not change total STAT1 or STAT3 content (Data Supplement
Figures VIB and VIIB), confirming increased Ser-727 phosphorylation of STAT1 and STAT3.
Physical Association of p44/42 MAPKs
With STAT1/3
Myocardial cytosolic fractions from control and preconditioned mice were immunoprecipitated with anti-pTyr-p44/42
antibodies, and the resulting immunoprecipitates were immunoblotted with anti-pSer(727)-STAT1 and anti-pSer(727)STAT3 antibodies, respectively. As shown in Figure 3A
1974
Circulation
September 27, 2005
Figure 2. A–D, Effect of ischemic PC on Ser phosphorylation of
STAT1 and STAT3. Nuclear proteins of myocardial samples
were prepared for immunoblotting. E and F, Effects of CHE on
the ischemic PC-induced increase in DNA binding activity of
STAT1/3. Nuclear proteins were subjected to EMSA for analysis
of STAT1/3-DNA binding activity with the 32P-labeled GAS
probe. G and H, Effect of CHE on the ischemic PC-induced
upregulation of COX-2. Myocardial samples were obtained from
mice that underwent sham operation (control) or from the ischemic/reperfused region of PC mice that received no treatment
(PC-24 hours) or CHE 5 minutes before the 6 coronary O/R
cycles (CHE⫹PC-24 hours). Membranous proteins were prepared for determination of COX-2 expression. Data are
mean⫾SEM, n⫽6/group.
through 3D, Tyr-phosphorylated p44/42 MAPKs coprecipitated with pSer(727)-STAT1 (334⫾36% of control [Figure
3A and 3B]) and pSer(727)-STAT3 (310⫾37% of control
[Figure 3C and 3D]), supporting a direct interaction between
phosphorylated (activated) p44/42 MAPKs and Ser-phosphorylated STAT1/3 in the PC myocardium.
Chelerythrine Suppresses Activation of the
Raf-1–MEK-1/2–p44/42 MAPK-pSer-STAT1/3
Pathway and COX-2 Induction
As a first step to interrogate the role of PKC, we used the
broad PKC inhibitor chelerythrine. Pretreatment with chelerythrine 5 minutes before the six 4-minute coronary O/R
cycles blocked the increases in pSer(338)-Raf-1 (Figure 1A
and 1B), pSer-MEK-1/2 (Figure 1C and 1D), and pTyr(204)p44/42 MAPKs (Figure 1E and 1F), suggesting that the
phosphorylation of Raf-1–MEK-1/2-p44/42 MAPKs by ischemic PC is mediated by PKC.
In addition, pretreatment with chelerythrine markedly
blunted the increase in nuclear pSer(727)-STAT1 and
pSer(727)-STAT3 (Figure 2A through 2D), the nuclear translocation of these factors (Figure 2A through 2D), the increase
in STAT1/3-DNA binding activity (Figure 2E and 2F), and
the increase in COX-2 protein (Figure 2G and 2H), suggesting that phosphorylation of STAT1 and STAT3 on Ser-727,
activation of STAT1 and STAT3, and upregulation of COX-2
Figure 3. p44/42 MAPKs interact with Ser-phosphorylated
STAT1 and STAT3 in PC myocardium. Cytosolic proteins of
myocardial samples were prepared 30 minutes after sham operation (control) or 30 minutes after 6 cycles of 4-minute coronary
O/R with no treatment (PC-30⬘) or with prior treatment with CHE
(CHE⫹PC-30⬘). The cytosolic proteins were then immunoprecipitated with anti-phosphorylated p44/42 MAPK antibodies followed by immunoblotting with anti–pSer727-STAT1 or anti–
pSer727-STAT3 antibodies. The figure illustrates representative
immunoblots (A and C) and densitometric analyses of the Serphosphorylated forms of STAT1 and STAT3 (B and D). There
was increased coprecipitation of pSer-STAT1 and pSer-STAT3
with pTyr-p44/42 MAPKs 30 minutes after ischemic PC. Data
are mean⫾SEM, n⫽6/group.
are also PKC dependent. Finally, the formation of complexes
between pTyr-p44/42 and pSer-STAT1/3 was also inhibited
by pretreatment with chelerythrine (Figure 3).
Deletion of PKC⑀ Blunts Activation of the
MEK-1/2–p44/42 MAPK-pSer-STAT1/3 Pathway
and the Upregulation of COX-2
To specifically interrogate the role of the ⑀ isoform of PKC in the
recruitment of the Raf-1–MEK-1/2–p44/42 MAPK signaling
pathway, we examined the effect of deletion of the PKC⑀ gene
(PKC⑀⫺/⫺ mice). Targeted ablation of PKC⑀ blunted the increase
in pSer(338)-Raf-1 (Figure 4A and 4B), pSer-MEK-1/2 (Figure
4A and 4C), and pThr(202)/Tyr(204)-p44/42 MAPK (Figure 4A
and 4D) 30 minutes after ischemic PC.
Deletion of PKC⑀ also inhibited the increase in total STAT1
and pSer(727)-STAT1 (Figure 5A, 5C, and 5E) and in total
STAT3 and pSer(727)-STAT3 (Figure 6A, 6C, and 6E) in the
nuclear fraction. Furthermore, deletion of PKC⑀ inhibited the
increase in STAT1/3-DNA binding activity in the nuclear
fraction 30 minutes after PC (Figure 7A and 7B) and the
subsequent upregulation of COX-2 protein expression 24 hours
later (Figure 7C and 7D). However, deletion of PKC⑀ did not
block the increase in pTyr(701)-STAT1 (Figure 5A and 5D) and
pTyr(705)-STAT3 (Figure 6A and 6D). Taken together, these
data indicate an obligatory role of PKC⑀ in the activation of
Raf-1, MEK-1/2, and p44/42 MAPKs; in the Ser (but not Tyr)
phosphorylation and activation of STAT1/3; and in the induction
of COX-2 after ischemic PC.
Discussion
Inhibition of COX-2 activity during the late phase of PC
results in complete loss of protection against infarction,
Xuan et al
1975
Figure 4. Targeted deletion of the PKC⑀
gene inhibits the phosphorylation of
MEK-1/2 and p44/42 MAPKs by ischemic PC. Myocardial samples were taken
30 minutes after 1 hour of open-chest
state without ischemia (control group) or
30 minutes after ischemic PC in WT and
PKC⑀⫺/⫺ mice. Cytosolic proteins were
used for immunoblotting analysis of the
phosphorylated forms of Raf-1, MEK-1/2,
and p44/42 MAPKs. The anti–pSer(338)Raf-1 antibody was from Cell Signaling.
Western blots (A) and densitometric
analysis (B–D) demonstrate that the ischemic PC-induced increase in
pSer(338)-Raf-1, pSer-MEK-1/2, and
pThr(202)/Tyr(204)-p44/42 MAPKs was
inhibited in PKC⑀⫺/⫺ mice. Ischemic PC
had no effect on Tyr-341 phosphorylation
of Raf-1 (Data Supplement Figures III
and IV). Data are mean⫾SEM,
n⫽5/group.
demonstrating that upregulation of this enzyme after the
initial ischemic stress is necessary for late PC to become
manifest.1,7,8,10,11 However, the signaling mechanism by
which ischemic PC induces the synthesis of COX-2 protein
remains incompletely understood. We have recently found
that activation of STAT1 and STAT3 via JAK-dependent Tyr
phosphorylation is essential for both PC-induced protection5
and PC-induced upregulation of COX-2.10 However, it is
unknown whether ischemic PC also leads to phosphorylation
of STAT1 and STAT3 on the Ser-727 residue (which is
essential for STAT-dependent transcriptional activation in
other systems12–14) and, if so, which upstream signaling
mechanism leads to Ser phosphorylation of these 2 transcription factors and what role Ser phosphorylation of STAT1/3
plays in the upregulation of COX-2 after ischemic PC.
The present study provides new information pertaining to
these issues in an in vivo murine model of myocardial
ischemia and reperfusion. The salient findings can be sum-
Figure 5. Deletion of the PKC⑀ gene (KO)
inhibits the Ser phosphorylation of
STAT1 by ischemic PC. Homogenates
and nuclear extracts were isolated from
myocardial samples taken 30 minutes
after 1 hour of an open-chest state without ischemia (control group) or 30 minutes after ischemic PC in WT and
PKC⑀⫺/⫺ mice. Data are mean⫾SEM,
n⫽5/group.
1976
Circulation
September 27, 2005
Figure 6. Deletion of the PKC⑀ gene blocks
Ser phosphorylation of STAT3 by ischemic
PC. Homogenates and nuclear extracts were
obtained as described in Figure 5. Data are
mean⫾SEM (n⫽5/group).
marized as follows: (1) Ischemic PC activates the Raf-1–
MEK-1/2–p44/42 MAPK signaling pathway, induces phosphorylation of STAT1 and STAT3 on the Ser-727 residue,
and upregulates COX-2; (2) pSer-STAT1 and pSer-STAT3
form complexes with pTyr-p44/42 MAPKs in PC myocardium, supporting the concept that Ser phosphorylation of these
2 factors is mediated by activated p44/42 MAPKs; (3)
activation of the Raf-1–MEK-1/2–p44/42 MAPK pathway,
the Ser phosphorylation of STAT1 and STAT3, and the
upregulation of COX-2 after ischemic PC are suppressed by
broad pharmacological inhibition of the PKC family (chelerythrine) or isoform-specific deletion of PKC⑀ (PKC⑀⫺/⫺
mice), indicating that they are dependent on PKC⑀ activity;
and (4) in contrast, deletion of PKC⑀ has no effect on Tyr
phosphorylation of STAT1/3. Previous studies have documented that Tyr phosphorylation of STAT1 and STAT3 via
JAK activity is a crucial mechanism for the development of
late PC.5,10 To our knowledge, this is the first study to
demonstrate that ischemic PC also causes Ser phosphorylation of STAT1 and STAT3 and that this event is governed by
Figure 7. Deletion of the PKC⑀ gene
blocks the ischemic PC-induced
increase in STAT1/3-DNA binding activity
and COX-2 upregulation. A and B,
Nuclear extracts were obtained as
described in Figure 5 and subjected to
EMSA for analysis of STAT1/3-DNA
binding activity with a 32P-labeled GAS
probe. C and D, Myocardial samples
were obtained from WT and PKC⑀⫺/⫺
mice that underwent sham operation (WT
control and PKC⑀⫺/⫺ control, respectively) or from the ischemic/reperfused
region of WT and PKC⑀⫺/⫺ mice that
were preconditioned with six 4-minute
coronary O/R cycles. All mice were
euthanized 24 hours after sham operation or after the sixth reperfusion. Data
are mean⫾SEM, n⫽5/group.
Xuan et al
PKC⑀. Furthermore, this is the first report that COX-2
expression in the heart is controlled by PKC⑀. Together with our
previous findings,5,10 the present study implies that STATdependent transcription after ischemic PC is modulated by a dual
signaling mechanism that involves both JAK1/2 (Tyr phosphorylation) and PKC⑀ (Ser phosphorylation).
The Ser-Thr kinases p44/42 MAPKs have been reported to
be involved in Ser phosphorylation of STATs14,21,22 and are
required for phosphorylation of STAT3 on Ser-727 in noncardiac cells.14,21 Activation of p44/42 MAPKs, in turn, could
be secondary to PKC activation. PKC, and specifically its ⑀
isoform, is known to play a crucial role in the development of
late PC2– 4 and to activate p44/42 MAPKs in a number of cell
types.23,24 In the heart, ischemic PC has been shown to
activate p44/42 MAPKs and their activators MEK-1/2 via a
PKC⑀-dependent mechanism.15 Accordingly, we postulated
that Ser phosphorylation of STAT1 and STAT3 is mediated
by a PKC⑀–Raf-1–MEK-1/2-p44/42 MAPK signaling cascade and examined each component of this pathway. We
found that six 4-minute coronary O/R cycles led to a rapid
increase in pSer(338)-Raf-1 (Figures 1A, 1B, 4A, and 4B),
pSer-MEK-1/2 (Figure 1C and 1D), pTyr(204)-p44/42
MAPKs (Figure 1E and 1F), and pSer-STAT1/STAT3 (Figure 2A through 2D), all of which are inhibited by pretreatment with chelerythrine (a broad inhibitor of the entire family
of PKCs) (Figure 1A through 1F and 2A–2D) or by targeted
genetic disruption of PKC⑀ (Figure 4A through 4D, 5A, 5C,
6A, and 6C), demonstrating that the entire Raf-1–MEK-1/2–
p44/42 MAPK–pSer-STAT1/3 signaling pathway is PKC⑀
dependent. Our data show that ischemic PC rapidly phosphorylates Raf-1 on Ser-338 but not on Tyr-341 (Figure 4A and
4B and Data Supplement Figures III and IV), demonstrating
that Raf-1 activation by ischemic PC is mediated primarily or
exclusively by phosphorylation on Ser-338. Deletion of
PKC⑀ blocks the Ser-338 phosphorylation of Raf-1 (Figure
4A and 4B), indicating that Ser-338 phosphorylation is PKC⑀
dependent. This is consistent with the previous finding that
PKC⑀ is necessary for Raf-1 activation in mouse C3H10T1/2
fibroblasts.24 Furthermore, we found that the Ser-phosphorylated STAT1 and STAT3 coprecipitated with phosphorylated
(activated) p44/42 MAPKs and that this was also inhibited by
chelerythrine (Figure 3A through 3D). In the aggregate, these
data suggest that PKC⑀ (which is known to be rapidly
activated by ischemic PC2– 4) plays a critical role in the
activation of the downstream Raf-1–MEK-1/2–p44/42
MAPK signaling cascade and in the resulting Ser phosphorylation of STAT1 and STAT3. Of note, deletion of PKC⑀ had
no effect on Tyr phosphorylation of STAT1 or STAT3
(Figures 5A, 5D, 6A, and 6D), indicating that this event is
mediated by a distinct, PKC⑀-independent signaling pathway.
Thus, we propose that the activation of STAT1 and STAT3
after ischemic PC is modulated via 2 parallel pathways,
namely (1) activation of JAK1/2 and subsequent Tyr phosphorylation of STATs5,10 and (2) activation of PKC⑀ and
subsequent recruitment of the Raf-1–MEK-1/2–p44/42
MAPK cascade, leading to Ser phosphorylation of STATs
(Figure 8). We suggest that these 2 pathways are activated
simultaneously, and both of them are necessary for STATdependent transcription after ischemic PC.
1977
Figure 8. Proposed signaling mechanism controlling COX-2 protein expression during ischemic PC. A sublethal ischemic stress
(PC stimulus) induces phosphorylation of STAT1 and STAT3 on
both Tyr and Ser residues. Tyr phosphorylation of STAT1/3 is
mediated by a JAK-dependent mechanism, whereas Ser phosphorylation of these transcription factors is modulated by
p44/42 MAPKs via PKC⑀-dependent activation of the Raf-1–
MEK-1/2–p44/42 MAPK pathway during PC ischemia. On activation, phosphorylated STAT1/3 translocate to the nucleus,
where they promote transcription of the COX-2 gene, leading to
the synthesis of COX-2 protein that is necessary for the development of delayed protection against myocardial infarction.
We found no evidence of Tyr-341 phosphorylation of
Raf-1, as determined by both Western blot analysis (Data
Supplement Figure III) and immunoprecipitation followed by
immunoblotting (Data Supplement Figure IV). Despite the
use of immunoprecipitation, it is still possible that phosphorylation of Tyr-341 occurred after ischemic PC but that its
level was below detection. Several previous studies in various
systems have failed to detect pTyr-341 in active Raf-1.25–28
Activation of Raf-1 is a very complex process, in which
phosphorylation of Ser-338 and Tyr-341 occurs in different
proportions, depending on the stimulus and cell type.28,29
King et al29 have proposed that different degrees of Raf-1
activation are achieved by Ser-338 and/or Tyr-341 phosphorylation occurring individually or in combination.
By documenting a new pathway through which ischemic
PC activates STATs (the PKC⑀–Raf-1–MEK-1/2–p44/42
MAPK axis), the present findings expand our understanding
of the molecular mechanisms whereby these transcription
factors contribute to the upregulation of COX-2, to the
development of late PC, and to the response of the heart to
stress in general. Furthermore, our findings reveal that PKC⑀
controls COX-2 expression in the heart and suggest a new
mechanism whereby this PKC isoenzyme may be involved in
late PC, namely, the modulation of STAT1 and STAT3
activity.
Acknowledgments
This study was supported in part by American Heart Association
grant 0150074N (to Dr Xuan) and by NIH grants R01 HL-65660 (to
Dr Xuan), HL-55757, HL-70897, HL-76794, and HL-78825 (to Dr
Bolli).
References
1. Bolli R. The late phase of preconditioning. Circ Res. 2000;87:972–983.
1978
Circulation
September 27, 2005
2. Ping P, Zhang J, Qiu Y, Tang X-L, Manchikalapudi S, Cao X, Bolli R.
Ischemic preconditioning induces selective translocation of protein kinase
C isoforms ⑀ and ␩ in the heart of conscious rabbits without subcellular
redistribution of total protein kinase C activity. Circ Res. 1997;81:
404 – 414.
3. Ytrehus K, Liu Y, Downey JM. Preconditioning protects ischemic rabbit
heart by protein kinase C activation. Am J Physiol. 1994;266:
H1145–H1152.
4. Cohen MV, Baines CP, Downey JM. Ischemic preconditioning: from
adenosine receptor of KATP channel. Annu Rev Physiol. 2000;62:
79 –109.
5. Xuan YT, Yiru G, Hui H, Yanqing Z, Bolli R. An essential role of the
JAK-STAT pathway in ischemic preconditioning. Proc Natl Acad Sci
U S A. 2001;98:9050 –9055.
6. Xuan YT, Tang XL, Banerjee S, Takano H, Li RC, Han H, Qiu Y, Li JJ,
Bolli R. Nuclear factor-␬B plays an essential role in the late phase of
ischemic preconditioning in conscious rabbits. Circ Res. 1999;84:
1095–1109.
7. Shinmura K, Tang XL, Wang Y, Xuan YT, Liu SQ, Takano H, Bhatnagar
A, Bolli R. Cyclooxygenase-2 mediates the cardioprotective effects of the
late phase of ischemic preconditioning in conscious rabbits. Proc Natl
Acad Sci U S A. 2000;97:10197–10202.
8. Shinmura K, Xuan YT, Tang XL, Kodani E, Han H, Zhu Y, Bolli R.
Inducible nitric oxide synthase modulates cyclooxygenase-2 activity in
the heart of conscious rabbits during the late phase of ischemic preconditioning. Circ Res. 2002;90:602– 608.
9. Guo Y, Jones WK, Xuan YT, Bao W, Wu WJ, Han H, Laubach VE, Ping
P, Yang Z, Qiu Y, Bolli B. The late phase of ischemic preconditioning is
abrogated by targeted disruption of the iNOS gene. Proc Natl Acad Sci
U S A. 1999;96:11507–11512.
10. Xuan YT, Guo Y, Zhu Y, Han H, Langenbach R, Dawn B, Bolli R.
Mechanism of cyclooxygenase-2 upregulation in late preconditioning. J
Mol Cell Cardiol. 2003;35:525–537.
11. Guo Y, Bao W, Wu WJ, Shinmura K, Tang XL, Bolli R. Evidence for an
essential role of cyclooxygenase-2 as a mediator of the late phase of
ischemic preconditioning in mice. Basic Res Cardiol. 2000;95:479 – 484.
12. Levy DE, Darnell JE Jr. STATS: transcriptional control and biological
impact. Nat Rev Mol Cell Biol. 2002;3:651– 662.
13. Wen Z, Zhong Z, Darnell JE Jr. Maximal activation of transcription by
Stat1 and Stat3 requires both tyrosine and serine phosphorylation. Cell.
1995;82:241–250.
14. Wierenga AT, Vogelzang I, Eggen BJ, Vellenga E. Erythropoietininduced serine 727 phosphorylation of STAT3 in erythroid cells is
mediated by a MEK-, ERK-, and MSK1-dependent pathway. Exp
Hematol. 2003;31:398 – 405.
15. Ping P, Zhang J, Cao X, Li RCX, Kong D, Tang XL, Qiu Y,
Manchikalapudi S, Auchampach JA, Black RG, Bolli R. PKC-dependent
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
activation of p44/p42 MAPKs during myocardial ischemia-reperfusion in
conscious rabbits. Am J Physiol. 1999;276:H1468 –H1481.
Shen Y, Schlessinger K, Zhu X, Meffre E, Quimby F, Levy DE, Darnell
JE Jr. Essential role of STAT3 in postnatal survival and growth revealed
by mice lacking STAT3 serine 727 phosphorylation. Mol Cell Biol.
2004;24:407– 419.
King AJ, Sun H, Diaz B, Barnard D, Miao W, Bagrodia S, Marshall MS.
The protein kinase Pak3 positively regulates Raf-1 activity through phosphorylation of serine 338. Nature. 1998;396:180 –183.
Decker T, Kovarik P, Meinke A. GAS elements: a few nucleotides with
a major impact on cytokine-induced gene expression. J Interferon Cytokine Res. 1997;17:121–134.
Khasar SG, Lin YH, Martin A, Dadgar J, McMahon T, Wang D, Hundle
B, Aley KO, Isenberg W, McCarter G, Green PG, Hodge CW, Levine JD,
Messing RO. A novel nociceptor signaling pathway revealed in protein
kinase C ⑀ mutant mice. Neuron. 1999;24:253–260.
Singh K, Balligand JL, Fischer TA, Smith TW, Kelly RA. Regulation of
cytokine-inducible nitric oxide synthase in cardiac myocytes and microvascular endothelial cells: role of extracellular signal-regulated kinases 1
and 2 (ERK1/ERK2) and STAT1␣. J Biol Chem. 1996;271:1111–1117.
Decker T, Kovarik P. Serine phosphorylation of STATs. Oncogene.
2000;19:2628 –2637.
Chung J, Uchida E, Grammer TC, Blenis J. STAT3 serine phosphorylation by ERK-dependent and -independent pathways negatively modulates
its tyrosine phosphorylation. Mol Cell Biol. 1997;17:6508 – 6516.
Traub O, Monia BP, Dean NM, Berk BC. PKC-⑀ is required for mechanosensitive activation of ERK1/2 in endothelial cells. J Biol Chem. 1997;
272:31251–31257.
Hamilton M, Liao J, Cathcart MK, Wolfman A. Constitutive association
of c-N-Ras with c-Raf-1 and protein kinase C-⑀ in latent signaling
modules. J Biol Chem. 2001;276:29079 –9090.
Baccarini M, Sabatini DM, App H, Rapp UR, Stanley ER. Colony
stimulating factor-1 (CSF-1) stimulates temperature dependent phosphorylation and activation of the Raf-1 proto-oncogene product. EMBO J.
1990;9:3649 –3657.
Blackshear PJ, Haupt DM, App H, Rapp UR. Insulin activates the Raf-1
protein kinase. J Biol Chem. 1990;265:12131–12134.
Kovacina KS, Yonezawa K, Brautigan DL, Tonks NK, Rapp UR, Roth
RA. Insulin activates the kinase activity of the Raf-1 proto-oncogene by
increasing its serine phosphorylation. J Biol Chem. 1990;265:
12115–12118.
Mason CS, Springer CJ, Cooper RG, Superti-Furga G, Marshall CJ,
Marais R. Serine and tyrosine phosphorylations cooperate in Raf-1, but
not B-Raf activation. EMBO J. 1999;18:2137–2148.
King AJ, Wireman RS, Hamilton M, Marshall MS. Phosphorylation site
specificity of the Pak-mediated regulation of Raf-1 and cooperativity with
Src. FEBS Lett. 2001;497:6 –14.
Ischemic heart disease is the leading cause of morbidity and mortality in all industrialized nations. As the population grows
older and risk factors become more prevalent, the enormous public health burden caused by ischemic heart disease is likely
to increase even further. Preconditioning (PC) is one of the most powerful cardioprotective interventions identified to date.
It consistently limits infarct size in every animal model and species examined, and evidence suggests that it is effective in
protecting human myocardium as well. Thus, PC represents an attractive strategy for inducing cardioprotection. Because
of its sustained nature, late PC offers the potential to afford long-lasting protection against myocardial cell death and
therefore may have great clinical relevance. The elucidation of the endogenous signaling mechanisms used by this
phenomenon has major pathophysiological and therapeutic implications. The mechanisms regulating ischemic PC are
known to involve many proteins, including protein kinase C and the JAK/signal transducer and activator of transcription
(STAT) pathway, which result in the eventual upregulation of cardioprotective genes such as inducible nitric oxide synthase
and cyclooxygenase-2. However, the exact mechanisms supporting these interactions are unknown. This article establishes
for the first time that ischemic PC causes Ser phosphorylation of STAT1/3 and activation of the Raf-1– extracellular
signal–regulated kinase kinase-1/2–p44/42 mitogen-activated protein kinase signaling pathway and that both of these
events, as well as the subsequent induction of cyclooxygenase-2, are dependent on protein kinase C-⑀. By continuing to
unravel the mechanisms underlying ischemic PC in a clinically relevant murine model of ischemia/reperfusion, we hope
to establish the groundwork for novel therapies that will one day be used for patients suffering from ischemic heart disease.
Hypertension
Antihypertensive Effects of Drospirenone With
17␤-Estradiol, a Novel Hormone Treatment in
Postmenopausal Women With Stage 1 Hypertension
William B. White, MD; Bertram Pitt, MD; Richard A. Preston, MD; Vladimir Hanes, MD
Background—Drospirenone (DRSP) is a novel progestin with antimineralocorticoid activity that has been developed for
hormone therapy in combination with 17␤-estradiol (E2) in postmenopausal women. In prior studies with DRSP in
postmenopausal women that were focused on relief of menopausal symptoms, DRSP/E2 yielded significant reductions
in blood pressure (BP).
Methods and Results—The effects of 3 mg DRSP/1 mg E2 on clinic and 24-hour ambulatory BP as well as potassium
homeostasis were evaluated in postmenopausal women with stage 1 hypertension (systolic, 140 to 159 and/or diastolic,
90 to 99 mm Hg) in a 12-week, multicenter, double-blind, randomized, placebo-controlled study. Clinic BPs were
measured at baseline and at 2, 4, 6, 8, and 12 weeks of therapy, whereas potassium was measured at 2, 6, and 12 weeks
of therapy. Ambulatory BP was performed in a substudy at baseline and at the end of the trial. In the intention-to-treat
population of 213 women, the clinic BP was reduced significantly on DRSP/E2 (clinic BP, ⫺14.1/⫺7.9 for DRSP/E2
versus ⫺7.1/⫺4.3 mm Hg for placebo, P⬍0.0001). In the subgroup of 43 women with ambulatory BP monitoring, the
24-hour BP fell by ⫺8.5/⫺4.2 mm Hg versus ⫺1.8/⫺1.6 mm Hg on placebo (P⫽0.002/0.07). There were no significant
changes from baseline in potassium levels or in the incidence of hyperkalemia (ⱖ5.5 meq/L) on DRSP/E2 compared
with placebo.
Conclusions—Combination therapy with DRSP/E2 significantly lowered both clinic and 24-hour systolic BP in
postmenopausal women with stage 1 systolic hypertension. This characteristic may lead to benefit for cardiovascular
risk reduction in this population. (Circulation. 2005;112:1979-1984.)
Key Words: hormones 䡲 hypertension 䡲 blood pressure 䡲 aldosterone antagonists 䡲 menopause
P
ostmenopausal estrogen deficiency has been associated
with increases in cardiovascular risk, but clinical trials of
standard formulations of hormonal therapy have not confirmed a benefit of hormone therapy in reducing cardiovascular disease in postmenopausal women.1–5 Although the
reasons for these generally negative results are unclear, it is
apparent that innovative, alternative strategies for hormone
therapy in postmenopausal women are warranted.
Several recent experimental and clinical reports have
implicated aldosterone, independent of angiotensin II, in the
pathogenesis of significant cardiovascular and renal disease
and have demonstrated the benefit of aldosterone blockade in
reducing a variety of cardiovascular and renal end points.6 –13
Drospirenone (DRSP) is a novel progestin with antialdosterone and antiandrogenic effects that, in combination with
17␤-estradiol (E2), has been developed for use in postmenopausal women as hormone therapy.14 –16 DRSP/E2 has been
shown to have significant antihypertensive effects in a short-
term study of postmenopausal, hypertensive women treated
with enalapril.17 When compared with other hormone therapies and oral contraceptives, DRSP yields a much greater rise
in plasma aldosterone14 –16 in response to the antimineralocorticoid effect of the compound.
The primary objective of the present study was to determine whether DRSP/E2 treatment has a clinically significant
effect on clinic and 24-hour ambulatory blood pressure (BP)
in hypertensive, postmenopausal women at doses of 3 mg
DSRP and 1 mg E2. In addition, we evaluated the effects of
DSRP/E2 on potassium homeostasis, because aldosterone
blockade has been associated with significant increases in
serum potassium values.14 –16
Methods
Patient Population
Postmenopausal women (aged 45 to 80 years) were included if, in
the untreated condition, their seated clinic systolic BP was 140 to
Received August 19, 2004; revision received July 4, 2005; accepted July 12, 2005.
From the Section of Hypertension and Clinical Pharmacology (W.B.W.), Pat and Jim Calhoun Cardiology Center, University of Connecticut School
of Medicine, Farmington; the Division of Cardiology (B.P.), University of Michigan Medical School, Ann Arbor; the Division of Clinical Pharmacology
(R.P.), University of Miami School of Medicine, Miami, Fla; and Berlex Laboratories (V.H.), Montville, NJ.
Reprint requests to William B. White, MD, Section of Hypertension and Clinical Pharmacology, University of Connecticut School of Medicine, 263
Farmington Ave, Farmington, CT 06030-3940. E-mail [email protected]
1979
1980
Circulation
September 27, 2005
159 mm Hg and/or the diastolic BP was 90 to 99 mm Hg. Patients
were excluded from the trial if they had received prior estrogen or
progestin hormone therapy; had sustained a recent myocardial
infarction or unstable angina; had congestive heart failure, clinically
significant liver or renal disease, known secondary hypertension, a
history of stroke or transient ischemic attack, venous thromboembolic disorders, or type 1 diabetes mellitus. Women whose calculated
creatinine clearance was ⬍50 mL/min or whose serum potassium
was abnormal at baseline were also excluded from participation in
the trial.
Study Design
The study was a multicenter (n⫽30 centers), double-blind, randomized, placebo-controlled, parallel-arm trial. The randomization numbers were generated in blocks of 4 by the SAS RANDO macro. The
2 treatments were allocated in a 1:1 ratio. Patients were screened and
evaluated for 3 to 4 weeks to establish the baseline BPs and
laboratory parameters. At randomization, patients received either
placebo or DRSP 3 mg with 1 mg E2 once daily in the morning. The
treatment and placebo groups were continued for 12 weeks. If the
systolic BP was ⬎160 mm Hg or the diastolic BP was ⬎100 mm Hg
on 2 consecutive occasions 1 to 3 days apart at any time during the
trial, the patient was removed from the trial for safety considerations.
In addition, if the patient’s serum potassium level was sustained in
excess of 5.5 meq/L on 2 consecutive occasions 1 to 3 days apart, the
patient was removed from the study and placed on conventional
therapy.
Patients were assessed at 2-week intervals during the trial for BP,
heart rate, adverse events, and concomitant medications. At 10
selected sites, 24-hour ambulatory BP monitoring was performed at
baseline and at 12 weeks of therapy.
Measurements of BP and Heart Rate
The office BP was measured by mercury column sphygmomanometry in triplicate (and averaged) in the seated position at all visits
after a minimum of 5 minutes of rest. These measurements were
performed 22 to 26 hours after dosing with the study medication.
Ambulatory BP and heart rate measurements were obtained with the
SpaceLabs 90207 monitor (Spacelabs Inc) at 10 centers experienced
in the use of ambulatory BP monitoring. Quality criteria used for an
acceptable ambulatory BP recording included a minimum of 80%
valid readings obtained within 24 hours after monitor hookup and a
minimum of 2 valid readings per hour. When these criteria were not
met, the patient was asked to repeat the study within 3 days. If
the repeated study failed to meet the quality control criteria, the
ambulatory BP data were considered nonevaluable. During the
24-hour ambulatory monitoring study, BP and heart rate were
measured every 15 minutes from 6 AM to 10 PM and every 20 minutes
between 10 PM and 6 AM. Monitoring hookup was initiated between
7 and 11 AM, and patients were dosed with study medication at the
time of monitor hookup. Study coordinators recorded times of sleep,
awakening, medication dosing, and monitor hookup in the case
report forms.
Laboratory and Safety Assessments
Serum chemistry values were determined at baseline and after 2, 6,
and 12 weeks of double-blind therapy. An ECG was performed at
baseline and after 4 and 12 weeks of therapy. Adverse event data
were obtained throughout the study by observation and indirect
questioning. Each adverse event was assigned the medical term from
the Hoechst Adverse Reaction Terminology System adverse event
coding manual. Events of special interest in the trial included
hyperkalemia, hypotension, dizziness, palpitations, syncope, and
arrhythmias (including tachycardia and bradycardia). The laboratory
protocol specified that all elevated serum potassium levels (ⱖ5.5
meq/L) were to be checked for hemolysis and repeated within 24
hours for confirmation.
Statistical Analyses
The comparability of patients in the treatment groups was determined from the demographic data and baseline hemodynamic values.
Continuous variables (age, height, BP) were analyzed with an
ANOVA model with factors for treatment, pooled center, and
baseline BP as covariates. Discrete variables were examined with the
Cochran-Mantel-Haenszel test for general association. All analyses
were conducted with SAS 8.2 software. The statistical analyses for
efficacy were performed on an intent-to-treat basis, which included
all patients randomized to the study with a baseline BP assessment
and at least 1 postbaseline assessment during the double-blind dosing
period. The last observed BP values were carried forward for
dropouts. The safety analyses included all patients who received at
least 1 dose of medication during the double-blind treatment phase.
The majority of study centers were small. A small center was
defined as any center with ⬍5 patients with postbaseline data in any
treatment group, resulting in 5 large and 25 small centers. To avoid
loss of information, small centers were pooled from largest to
smallest until the pooled center had ⱖ5 patients in each treatment
group. These centers were grouped into 11 pooled centers for the
purpose of analysis. The pooling algorithm was predetermined
before unblinding the data, and the pooling algorithm was described
in the statistical analysis plan for the study. Considering the
subjective nature of the pooling algorithm, albeit prespecified before
completion of the study, an exploratory analysis was also performed
with actual center as a fixed effect in contrast to pooled centers. This
analysis did not change the probability values up to 4 decimal places
for any of the comparisons between the DRSP/E2 and placebo
groups. The centers in the clinical trial are rarely a random sample of
all possible centers. Therefore, we were in favor of treating the actual
center or pool center as a fixed effect in the model of the analysis.
Analyses in a mixed model with center as a random effect also did
not show any impact on probability values. An exploratory analysis
with treatment-by-center effect in the model was also performed to
investigate the possibility of differential effects across centers. This
interaction was nonsignificant (P⫽0.45, and P⫽0.692 for pooled
and actual centers, respectively), suggesting that the effect of pooled
center in our model of analysis was effective in adjusting the
treatment estimates for center effects.
The primary efficacy end point of the trial was the mean change
from baseline at week 12 in clinic BP for DSRP/E2 and placebo.
Secondary analyses included the changes from baseline in the
24-hour systolic and diastolic BPs and heart rate, as well as other
ambulatory monitoring parameters such as daytime mean and nighttime mean values. In addition, mean changes from baseline were
examined for serum potassium. The incidence of hyperkalemia
(defined as plasma potassium ⱖ5.5 meq/L) was tabulated.
Treatment groups were compared with respect to the change from
baseline in clinic BP with a 2-way ANCOVA, with terms for
treatment, pooled center, and baseline measures as covariates in the
model. Before implementing the final ANCOVA model, the assumption of homogeneity of treatment covariate slopes was tested with an
ANCOVA model that included terms for baseline, treatment, and
treatment-by-baseline interaction.
Adverse events were coded and summarized by treatment group
and tabulated by treatment group and body system. Clinical laboratory data were summarized by treatment group. For each parameter,
the treatments were compared with respect to the mean change from
baseline by ANCOVA. Shifts in baseline laboratory values were
compared between treatment groups.
The planned sample size of 268 subjects (ie, 134 subjects per
treatment group) provided at least 80% power to detect a difference
of 4 mm Hg between active treatment and placebo groups in the
change from baseline in office cuff systolic BP with a 2-sample t test
of the null hypothesis at the 0.05 level of significance. The estimated
sample SD of 11 mm Hg used in the calculation was obtained from
the results of a previous study.17 The sample size calculation was
based on an assumed dropout rate of 10%.
White et al
TABLE 1.
Patient Characteristics at Baseline
Placebo
(n⫽111)
P
56⫾5
57⫾5
0.07
93 (91)
99 (90)
0.70
9 (9)
11 (10)
Body mass index, kg/m2
29⫾4
28⫾4
0.66
Arm circumference, cm
31⫾4
30⫾4
0.32
Smoking history, n (%)
13 (13)
10 (9)
0.35
Clinic systolic BP, mm Hg
145⫾7
146⫾7
0.16
Clinic diastolic BP, mm Hg
89⫾6
89⫾5
0.98
Clinic heart rate, bpm
72⫾9
73⫾8
0.76
24-Hour systolic BP, mm Hg
136⫾16
135⫾13
0.77
24-Hour diastolic BP, mm Hg
83⫾11
82⫾8
0.76
Daytime systolic BP, mm Hg
140⫾17
139⫾13
0.90
Daytime diastolic BP, mm Hg
87⫾11
85⫾8
0.59
Nighttime systolic BP, mm Hg
125⫾16
127⫾15
0.55
Nighttime diastolic BP, mm Hg
73⫾11
73⫾9
0.77
Age, y
The adjusted mean changes in BP in the clinic setting are
shown in Table 2 and Figure 1. After 2 weeks of therapy, the
reductions in systolic BP were significantly greater on
DRSP/E2 compared with placebo. Significant reductions in
diastolic BP occurred after 4 weeks of DRSP/E2 therapy
compared with placebo (Figure 1). At the end of the study, the
mean reductions in clinic BP in the DSRP/E2 group averaged
⫺14.1/⫺7.9 mm Hg, whereas the respective reductions for
the placebo group were ⫺7.1/⫺4.3 mm Hg (P⬍0.001 for
both systolic and diastolic BP). DRSP/E2 also significantly
lowered pulse pressure compared with placebo by
⫺3.5 mm Hg (P⫽0.007). The changes from baseline in heart
rate were similar for DRSP/E2 and placebo (Table 2).
Ethnicity, n (%)
Nonblack
Black
1981
Clinic BP
DRSP/E2
(n⫽102)
Characteristic
Drospirenone and Blood Pressure
Ambulatory BP
The mean changes from baseline in 24-hour ambulatory
systolic and diastolic BPs from the substudy are shown in
Table 2. Significant reductions from baseline in mean 24hour systolic BP (P⫽0.002) were observed in the DRSP/E2
treatment group compared with placebo. The reductions in
ambulatory systolic BP occurred primarily during the daytime. As noted in Table 2, DRSP/E2 induced significant
reductions in both daytime systolic and diastolic BPs compared with placebo, but there were no significant changes
from baseline in nighttime BP. As shown in Figure 2,
DRSP/E2 induced sustained reductions in systolic BP
throughout the 24-hour period compared with baseline and
with placebo treatment. Lesser but significant daytime effects
were observed with changes from baseline in the hourly
diastolic BP (Figure 2). The largest reductions in diastolic BP
were observed during hours 4 to 8 and hours 17 to 21 after
dosing.
Results
Patient Characteristics and Dosing of Drugs
There were 213 patients randomized into the 2 treatment
arms, with similar demographics and baseline clinic and
ambulatory BP values (Table 1). The percentage of patients
who withdrew from the study was 16 (14.4%) in the placebo
group and 10 (9.8%) in the DSRP/E2 group. The main
reasons for withdrawal after randomization were adverse
events and treatment failure. Other reasons included loss to
follow-up, protocol violations, or patient withdrawal of consent. No patient was withdrawn because of hyperkalemia.
TABLE 2. Mean Changes From Baseline in Clinic and Ambulatory BP After 12
Weeks of Therapy With DRSP/E2 vs Placebo (Intent to Treat)
Parameter
DRSP/E2
Placebo
Clinic BP, mm Hg
Differences Between
Treatments (95% CI)
P
n⫽102
n⫽111
Systolic
⫺14.1
⫺7.1
⫺7.0 (⫺9.8, ⫺4.2)
⬍0.0001
Diastolic
⫺7.9
⫺4.3
⫺3.6 (⫺5.3, ⫺2.0)
⬍0.0001
Pulse pressure
⫺6.1
⫺2.7
⫺3.4 (⫺5.8, ⫺1.1)
0.005
Clinic heart rate, bpm
⫺1.1
⫺2.8
1.7 (⫺0.12, 3.7)
0.066
Ambulatory BP, mm Hg
n⫽23
n⫽20
Systolic
⫺8.5
⫺1.8
⫺6.6 (⫺10.6, ⫺2.7)
0.002
Diastolic
⫺4.2
⫺1.6
⫺2.6 (⫺5.5, 0.23)
0.070
Systolic
⫺10.4
⫺2.1
⫺8.3 (⫺12.4, ⫺4.2)
0.0003
Diastolic
⫺5.0
⫺1.6
⫺3.4 (⫺6.7, ⫺0.2)
0.039
Systolic
⫺3.2
⫺0.9
⫺2.3 (⫺7.2, 3.2)
0.41
Diastolic
⫺2.0
⫺1.2
⫺0.9 (⫺4.7, 3.0)
0.65
24-Hour mean BP
Daytime BP
Nighttime BP
CI indicates confidence interval.
1982
Circulation
September 27, 2005
Figure 1. Effects of DRSP/E2 vs placebo on clinic BP during the
12 weeks after randomization. The upper panel shows the systolic BP, and lower panel, diastolic BP. *P⬍0.01, †P⬍0.001,
**P⬍0.0001.
Figure 2. Effects of DRSP/E2 vs placebo on ambulatory BP
after 12 weeks of double-blind therapy. The upper panel shows
the hourly values for systolic BP, and the lower panel, hourly
values for diastolic BP (both at week 12).
Discussion
Adverse Events
Because of the antimineralocorticoid effects of DRSP,
changes in serum potassium were closely monitored. There
were no patients in the DRSP/E2 group who developed a
serum potassium value ⬎5.5 meq/L. In the placebo group, 4
patients (3.6%) had a transient serum potassium value ⬎5.5
meq/ L (P⫽0.122 for DRSP/E2 versus placebo). The patterns
of changes from baseline in serum potassium were quite
similar for DRSP/E2 and placebo (Figure 3). The mean
maximal change from baseline in the DRSP/E2 group was
0.24⫾0.38 meq/L versus 0.16⫾0.43 meq/L for the placebo
group and was not significant (P⫽0.18).
There were no deaths during the course of the study. One
patient randomized to DRSP/E2 sustained an acute myocardial infarction. The incidence of minor, clinically nonsignificant ECG abnormalities was identical for patients randomized to DRSP/E2 (22%) and placebo (22%). There were no
significant differences in the number of patients with selected
cardiovascular events (arrhythmia, bradycardia, dizziness,
palpitations, syncope) on DRSP/E2 versus placebo. The
overall incidence of these adverse events was 7/102 (6.9%) of
patients taking DRSP/E2 versus 3/111 (2.7%) of patients
taking placebo. Dizziness was the most common event (4%
of DRSP patients versus 2% of the placebo patients).
Principal Findings
DRSP (3 mg) with E2 (1 mg), a new hormonal treatment with
selective aldosterone-blocking properties, was effective in
reducing clinic and 24-hour BPs in postmenopausal women
with stage 1 hypertension. Twenty-four-hour ambulatory BP
Figure 3. Effects of DRSP/E2 vs placebo on serum potassium
(meq/L). Shown are the maximal changes for any given individual during the course of the 12-week study.
White et al
may yield a more reliable antihypertensive assessment owing
to the lack of observer bias, the substantially increased
number of values taken over the dosing interval, and the
enhanced statistical reproducibility of ambulatory BP compared with clinical BP measurements.18 –20 Another important
finding in this trial was the lack of evidence for clinically
significant increases in mean serum potassium values with
DRSP/E2 in this population of older women with stage I
hypertension. Furthermore, no patient developed hyperkalemia while taking DRSP/E2.
Clinic and Ambulatory BP
DRSP/E2 lowered both the clinic and daytime ambulatory
BPs significantly compared with placebo; the levels of
ambulatory BP reductions observed in our study are comparable to many other antihypertensive agents, including angiotensin-converting enzyme inhibitors, angiotensin receptor
blockers, calcium channel blockers,21 and the recently approved selective aldosterone blocker eplerenone.11 In fact, in
prior work with eplerenone,11 the mean reduction from
baseline in 24-hour BP was ⬇7/4 mm Hg for the 50-mg dose,
a value similar to that which was observed with 3 mg DRSP
in the present study (Table 2 and Figure 2). Additionally,
Preston et al17 reported that after just 2 weeks of therapy,
DRSP/E2 lowered 24-hour ambulatory BP by 9/5 mm Hg
when the drug was added to enalapril in 12 postmenopausal
women. These reductions in BP were associated with increases in aldosterone of ⬇3 ng/dL (40% above baseline),11
attesting to DRSP’s effect in blocking the mineralocorticoid
receptor.
It is noteworthy that the reductions from baseline in the
clinic BP versus reductions in the daytime ambulatory BP
were somewhat dissimilar for DRSP/E2 (Table 2). This is
often the case in antihypertensive therapy trials, because
typically the mean reduction in ambulatory BPs in clinical
trials is ⬇40% less than the average reduction in clinic BP.18
This phenomenon is due in part to both observer bias and
regression to the mean.18 –20 The reductions in nighttime BPs
were not significantly greater on DRSP/E2 compared with
placebo (Table 2). This is likely to be due to the relatively
normal baseline nighttime BP levels observed in this mildly
hypertensive population rather than a loss of effect at the end
of the dosing period (Table 1). Changes in BP during sleep on
antihypertensive agents are quite dependent on the baseline
level of nocturnal pressure,22 and when baseline values are in
the range of 125/73 mm Hg, as was the case in this population
(Table 1), small declines in sleep BP would be expected
during the treatment period. Although the intention-to-treat
population was smaller than the planned randomization, the
estimates of changes from baseline in BP were larger than
expected, and the statistical power for these changes was
quite high at 99%.
Safety and Tolerability and Laboratory
Assessments
DRSP/E2 was well tolerated in this 213-patient trial, with
adverse-event profiles similar to those of placebo. Most
important, laboratory assessment did not show any clinically
significant changes in serum potassium (Figure 3). Addition-
Drospirenone and Blood Pressure
1983
ally, specific adverse events such as syncope, cardiac arrhythmias, or ECG changes were not observed with DRSP/E2, a
finding that supports its potential advantage in clinical practice in postmenopausal women with hypertension.
Conclusions
Our study demonstrates that DRSP/E2, a new hormone
therapy with mineralocorticoid receptor-blocking activity,
was effective in reducing ambulatory systolic and diastolic
BPs at doses of 3 mg/1 mg daily. The drug was well tolerated,
with no evidence of subjective or objective adverse events.
These findings are clinically relevant, because hormone
therapy for postmenopausal women has been under scrutiny
because of its potential for increasing cardiovascular thrombotic events.1–5 Because reductions in systolic BP have
significant implications for older individuals with hypertension,23–25 especially for the reduction of stroke and congestive
heart failure, DRSP/E2 may have an advantage for the
treatment of menopausal symptoms in older women. In future
antihypertensive studies with DRSP/E2, it will be of interest
to study the effects of E2 alone as well as to compare this
unique progestin to more conventional progestins that lack
antimineralocorticoid effects.
Acknowledgments
This work was supported in part by Berlex Laboratories, Inc
(Montvale, NJ), the Catherine and Patrick Donaghue Medical Research Foundation (Hartford, Conn), and the University of Connecticut Clinical Trials Unit (Farmington, Conn).
Disclosure
This study was funded by a grant from Berlex Laboratories,
Montvale, NJ, the manufacturer of drospirenone, with 17␤-estradiol.
The current work was done in an unrestricted, independent manner
with full access provided to all data. The sponsor was entitled to
comment on manuscripts, and the authors might have considered
these comments, but the rights to publication resided contractually
with the investigators. Dr White has received research grants from
Berlex Laboratories, Astra-Zeneca, Boehringer-Ingelheim, and
Pfizer and has received honoraria to serve on advisory boards of
Berlex and Boehringer Ingelheim during the past 4 years. Dr Pitt has
received honoraria from and been a consultant to Berlex Laboratories
during the past 4 years. Dr Preston has received research grants from
Berlex Laboratories and has served as a paid consultant to Berlex
Laboratories during the past 4 years. Dr Hanes is a full-time
employee in research and development at Berlex Laboratories, Inc.
References
1. Herrington DM, Reboussin DM, Brosnihan KB, Sharp PC, Shumaker SA,
Snyder TE, Furberg CD, Kowalchuk GJ, Stuckey TD, Rogers WJ, Givens
DH, Waters D. Effects of estrogen replacement on the progression of
coronary-artery atherosclerosis. N Engl J Med. 2000;343:522–529.
2. Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B,
Vittinghoff E. Randomized trial of estrogen plus progestin for secondary
prevention of coronary heart disease in postmenopausal women: Heart
and Estrogen/progestin Replacement Study (HERS) Research Group.
JAMA. 1998;280:605– 613.
3. Viscoli CM, Brass LM, Kernan WN, Sarrel PM, Suissa S, Horwitz RI. A
clinical trial of estrogen-replacement therapy after ischemic stroke.
N Engl J Med. 2001;345:1243–1249.
4. Hodis HN, Mack WJ, Azen SP, Lobo RA, Shoupe D, Mahrer PR, Faxon
DP, Cashin-Hemphill L, Sanmarco ME, French WJ, Shook TL, Gaarder
TD, Mehra AO, Rabbani R, Sevanian A, Shil AB, Torres M, Vogelbach
KH, Selzer RH, Women’s Estrogen-Progestin Lipid-Lowering Hormone
Atherosclerosis Regression Trial Research Group. Hormone therapy and
the progression of coronary-artery atherosclerosis in postmenopausal
women. N Engl J Med. 2003;349:535–545.
1984
Circulation
September 27, 2005
5. Manson JE, Hsia J, Johnson KC, Rossouw JE, Assaf AR, Lasser NL,
Trevisan M, Black HR, Heckbert SR, Detrano R, Strickland OL, Wong
ND, Crouse JR, Stein E, Cushman M, the Women’s Health Initiative
Investigators. Estrogen plus progestin and the risk of coronary heart
disease. N Engl J Med. 2003:349:523–534.
6. Duprez D, De Buyzere M, Rietzchel ER, Clement DL. Aldosterone and
vascular damage. Curr Hypertens Rep. 2000;2:327–334.
7. Rocha R, Chander PN, Khanna K, Zuckerman A, Stier CT. Mineralocorticoid blockade reduces vascular injury in stroke-prone hypertensive rats.
Hypertension. 1998;31:451– 458.
8. Farquharson CAJ, Struthers AD. Spironolactone increases nitric oxide
bioactivity, improves endothelial vasodilator dysfunction, and suppresses
vascular angiotensin I/angiotensin II conversion in patients with chronic
heart failure. Circulation. 2000;101:594 –597.
9. Chrysostomou A, Becker G. Spironolactone in addition to ACE inhibition
to reduce proteinuria in patients with chronic renal disease. N Engl J Med.
2001;345:925–926.
10. White WB, Duprez D, St Hillaire R, Krause S, Roniker B, Kuse-Hamilton
J, Weber MA. Effects of the selective aldosterone blocker eplerenone
versus the calcium antagonist amlodipine in systolic hypertension. Hypertension. 2003;41:1021–1026.
11. White WB, Carr AA, Krause S, Jordan R, Roniker B, Oigman W.
Assessment of the novel selective aldosterone blocker eplerenone using
ambulatory and clinical blood pressure in patients with systemic hypertension. Am J Cardiol. 2003;92:38 – 42.
12. Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky
J, Wittes J. The effect of spironolactone on morbidity and mortality in
patients with severe heart failure. N Engl J Med. 1999;341:709 –717.
13. Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, Bittman
R, Hurley S, Kleiman J, Gatlin M. Eplerenone, a selective aldosterone
blocker, in patients with left ventricular dysfunction after myocardial
infarction. N Engl J Med. 2003;348:1309 –1321.
14. Oelkers W, Foidart JM, Dombrovicz N, Welter A, Heithecker R. Effects
of a new oral contraceptive containing an antimineralocorticoid progestogen, drospirenone, on the renin-aldosterone system, body weight,
blood pressure, glucose tolerance, and lipid metabolism. J Clin Endocrinol Metab. 1995;80:1816 –1821.
15. Krattenmacher R. Drospirenone: pharmacology and pharmacokinetics of
a unique progestogen. Contraception. 2000;62:29 –38.
16. Oelkers W. Effects of estrogens and progestogens on the renin-aldosterone system and blood pressure. Steroids. 1996;61:166 –171.
17. Preston RA, Alonso A, Panzitta D, Zhang P, Karara AH. Additive effect
of Drospirenone/17␤-estradiol in hypertensive postmenopausal women
receiving enalapril. Am J Hypertens. 2002;15:816 – 822.
18. White WB. Advances in ambulatory blood pressure monitoring for the
evaluation of antihypertensive therapy in research and practice. In: White
WB ed: Blood Pressure Monitoring in Cardiovascular Medicine and
Therapeutics. Totowa, NJ: Humana Press; 2000:273–298.
19. Mansoor GA, McCabe EJ, White WB. Long-term reproducibility of
ambulatory blood pressure. J Hypertens. 1994;12:703–708.
20. Coats AJS, Radaelli A, Clark SJ, Conway J, Sleight P. The influence of
ambulatory blood pressure monitoring on the design and interpretation of
trials in hypertension. J Hypertens. 1992;10:385–391.
21. White WB, Sica DA, Calhoun D, Mansoor GA, Anders R. Preventing
increases in the early-morning blood pressure, heart rate, and the ratepressure product with controlled-onset extended release verapamil at
bedtime versus enalapril, losartan, and placebo on arising. Am Heart J.
2002;144:657– 665.
22. White WB, Larocca G Improving the utility of the nocturnal hypertension
definition by using absolute sleep blood pressure rather than the ‘dipping’
proportion. Am J Cardiol. 2003;92:1439 –1441.
23. SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results of the Systolic Hypertension in the Elderly Program
(SHEP). JAMA. 1991;265:3255–3264.
24. Staessen JA, Fagard R, Thijs L, Celis H, Arabidze GG, Birkenhager WH,
Bulpitt CJ, de Leeuw PW, Dollery CT, Fletcher AE, Forette F, Leonetti
G, Nachev C, O’Brien ET, Rosenfeld J, Rodicio JL, Tuomilehto J,
Zanchetti A. Randomised double-blind comparison of placebo and active
treatment for older patients with isolated systolic hypertension. Lancet.
1997;350:757–764.
25. White WB. Benefits of antihypertensive therapy in older patients with
hypertension. Arch Intern Med. 2000;160:149 –150.
Imaging
Is Duplex Surveillance of Value After Leg Vein
Bypass Grafting?
Principal Results of the Vein Graft Surveillance Randomised Trial (VGST)
A.H. Davies, MA, DM, FRCS; A.J. Hawdon, PhD; M.R. Sydes, MSc; S.G. Thompson, DSc;
on Behalf of the VGST Participants
Background—The purpose of this study was to assess the benefits of duplex compared with clinical vein graft surveillance
in terms of amputation rates, quality of life, and healthcare costs in patients after femoropopliteal and femorocrural vein
bypass grafts.
Methods and Results—This was a multicenter, prospective, randomized, controlled trial. A total of 594 patients with a
patent vein graft at 30 days after surgery were randomized to either a clinical or duplex follow-up program at 6 weeks,
then 3, 6, 9, 12, and 18 months postoperatively. The clinical and duplex surveillance groups had similar amputation rates
(7% for each group) and vascular mortality rates (3% versus 4%) over 18 months. More patients in the clinical group
had vein graft stenosis at 18 months (19% versus 12%, P⫽0.04), but primary patency, primary assisted patency, and
secondary patency rates, respectively, were similar in the clinical group (69%, 76%, and 80%) and the duplex group
(67%, 76%, and 79%). There were no apparent differences in health-related quality of life, but the average health service
costs incurred by the duplex surveillance program were greater by £495 (95% CI £183 to £807) per patient.
Conclusions—Intensive surveillance with duplex scanning did not show any additional benefit in terms of limb salvage
rates for patients undergoing vein bypass graft operations, but it did incur additional costs. (Circulation. 2005;112:19851991.)
Key Words: imaging 䡲 stenosis 䡲 amputation 䡲 grafting 䡲 occlusion
I
tions performed. Grigg et al13 estimated that if Duplex
surveillance of all vein grafts prevented 5% of patients from
needing an amputation, then the savings would be great
enough to justify the expense of establishing a surveillance
program.
Although amputation is the most clinically relevant measure of graft failure, graft occlusion does not necessarily
result in amputation.12 Unfortunately, the few reports that
have been published10,11,13–15 tend to argue in favor of duplex
surveillance on the basis of patency alone, with no measurement of limb salvage. Golledge et al5 undertook a summation
analysis of infrainguinal vein graft outcomes on those studies
that provided occlusion rates, comparing 2680 duplex surveillance patients with 3969 nonsurveillance patients. The
levels of distal anastomosis and presence of critical ischemia
were found to be similar in both groups. However, only 6 of
17 studies reported amputation rates; only 2 of these were
RCTs, and both of these were small.
In one randomized trial, Lundell et al16 studied both vein
(n⫽106) and synthetic (n⫽50) grafts randomized to either
nfrainguinal vein bypass graft procedures are performed
routinely on patients with lower-limb peripheral arterial
disease; however, vein grafts are prone to develop lesions or
stenoses, which reduce blood flow and can precipitate thrombosis.1,2 Such stenoses are identifiable in 25% to 30% of vein
bypass grafts within the first year.3,4
Duplex ultrasound scanning is currently the best method
for detecting stenotic lesions that threaten graft patency
during follow-up.5 The correction of such lesions may improve graft patency and limb salvage rates.6,7 However, to
date, evidence to support this has been based largely on the
findings of smaller-scale observational studies,8 –11 in the
absence of a large multicenter, randomized, controlled trial
(RCT).
A major consideration within the current healthcare environment is that procedures must be cost-effective.12,13 Duplex
surveillance programs are expensive to establish and maintain, not only with regard to the initial outlay for the machine
but also with regard to the employment of a trained vascular
technologist, as well as funding for the additional interven-
Received November 3, 2004; revision received June 1, 2005; accepted June 6, 2005.
From the Department of Vascular Surgery (A.H.D., A.J.H.), Imperial College London at Charing Cross Hospital, London, United Kingdom; MRC
Clinical Trials Unit (M.R.S.), London, United Kingdom; and MRC Biostatistics Unit (S.G.T.), Institute of Public Health, Cambridge, United Kingdom.
The online-only Data Supplement, which contains information on trial participants and centers, can be found at http://circ.ahajournals.
org/cgi/content/full/112/13/1985/DC1.
Correspondence to Mr Alun H. Davies, Department of Vascular Surgery, Imperial College London at Charing Cross Hospital, Fulham Palace Rd,
London, W6 8RF, United Kingdom. E-mail [email protected]
1985
1986
Circulation
September 27, 2005
“intensive” surveillance of clinical examination, anklebrachial pressure index (ABPI), and duplex scans or “routine”
surveillance of clinical examination and ABPI only. Their
results at 3 years after operation showed that there was an
advantage of duplex scanning when patency rates were
compared but not when amputation rates were compared.
Ihlberg et al17 could not demonstrate any difference in limb
salvage between duplex surveillance and clinical assessment
in a second randomized trial of 185 consecutive vein grafts;
however, they had difficulty obtaining complete follow-up
data on patients.18 Both groups concluded that there was a
need for a large RCT.16 –18
Here, we report on the results of a large-scale RCT of 594
infrainguinal vein graft reconstructions.
Methods
The design of the Vein Graft Surveillance Trial has been reported
previously.19 Patients undergoing femoropopliteal or femorocrural
vein bypasses were recruited between April 1998 and December
2001 from 22 centers within the United Kingdom and 7 from Europe.
Indications for surgical correction included critical ischemia, claudication, or symptomatic popliteal aneurysm. Patients receiving
synthetic grafts such as polytetrafluoroethylene (PTFE) grafts were
excluded from the study. Each center received ethical approval.
Randomization and Follow-Up
Patients from participating centers whose vein graft was patent at 30
days after surgery were randomized at ⬇6 weeks (range 4 to 10
weeks) to either the clinical group (clinical examination with ABPI
measurements) or the duplex group (same as clinical plus a routine
duplex scan). The allocation of patients was performed by a central
computer– based randomization service at the University of York.
This used randomly sized allocation blocks of sizes 4 and 6 (plus a
small number of odd-sized blocks), stratified by center and presenting symptoms (claudication or critical ischemia). Patients then
underwent a surveillance program with follow-up appointments at
the time of recruitment (6 weeks) and then subsequently at 3, 6, 9,
12, and 18 months. All patients received a duplex scan at 18 months;
this was performed in the clinical arm of the trial solely to identify
the incidence of stenoses.
The rationale for scheduling the follow-up to finish at 18 months
was that the majority of stenoses and graft failures occur within the
first year.3,4,20,21 There are 3 time periods of graft failure: early
(within 30 days), which is attributed to technical failure; intermediate
(30 days to 1 year), usually attributed to graft stenosis; and late
failure, usually attributed to progression of disease. Early graft
failures were excluded because an entry criterion for the trial was a
patent graft at 30 days. The length of follow-up in such programs is
controversial; both Idu et al20 and Mills et al21 suggest that as
stenoses occur, early surveillance need only be performed for the
first 6 months, whereas others recommend a longer follow-up.22 The
Transatlantic Consensus states that the optimum length and frequency of follow-up are unknown.23
Duplex Surveillance
The duplex group was scanned along the graft, including the distal
and proximal anastomoses. The inflow and outflow vessels were
scanned for specific structural abnormalities or exceptional flow
characteristics in color-flow images. Graft flow velocity and blood
flow patterns were evaluated at multiple sites along the bypass graft.
A graft at risk of failure was defined as having a slow peak systolic
flow velocity of less than 45 cm/s24 or a ratio of V2 (peak systolic
velocity at the site of the stenosis) to V1 (peak systolic velocity at
any other point within 2 cm at the normal adjacent graft) of ⬎2.8 At
18 months, an abnormal V2/V1 ratio was used to define the presence
of a stenosis. Any other irregularities, such as inflow/outflow
problems, graft dilatation, or arteriovenous fistula, were also noted.
Intervention criteria included clinical signs of a failing graft, such as
onset of disabling claudication, ischemic pain, or ischemic ulcers,
and a decrease in ABPI of ⱖ0.1.25
Outcomes Assessed
The primary outcomes were time to amputation (above knee, below
knee, or through knee) and time to vascular mortality (death due to
myocardial infarction, heart failure, arrythmia, or cerebrovascular
accident). Patency, cost, and quality of life were regarded as
secondary outcome measures. We used the recommended definitions
of patency,25 subdividing primary patency (patency without intervention) from primary assisted patency (patency without intervention plus patency after intervention for graft stenosis) and secondary
patency (patency without intervention plus patency after intervention
for graft stenosis plus patency after intervention for graft occlusion).
Health-related quality-of-life data were collected at 6 and 18
months with the SF-36 (36-item short-form health survey) and
EuroQol questionnaires.26 –28 The SF-36 data were summarized with
the physical and mental subscales and the EuroQol with the derived
EuroQol 5 dimensions (EQ5D) utility measure. Healthcare costs
were obtained for each patient by applying health resource group
costs for the financial year 2002/200329 to the duplex scans,
angiograms, angioplasties, thrombolysis, and surgical interventions
performed.
Statistical Methods
On the basis of anticipated 18-month amputation rates of ⬇10%, the
sample size of 600 patients yields a standard error for the difference
in amputation rates between groups of ⬇2.5%. The original plan was
to recruit 1200 patients,19 but this proved impossible in the time
available because of the increased use of percutaneous endovascular
treatments; the standard error based on 1200 patients would have
been 1.7%. The statistical analysis was conducted according to a
prespecified plan, drawn up before the outcome data were examined,
which used the intention-to-treat principle. The main outcomes of
time to amputation and vascular death were analyzed with Cox
regression; planned adjustment of the resulting hazard ratios for age,
sex, smoking, and diabetes made no material difference, so only the
unadjusted results are presented. Kaplan-Meier estimates of cumulative amputation rates were drawn, with censoring for deaths and
withdrawals. Patency rates over time were estimated with life-table
methods and compared with the log rank test. Quality-of-life scores
were compared between groups with a Mann-Whitney test, whereas
average costs were compared with a t test.30
Results
Of the 594 patients recruited, 290 were randomized to clinical
follow-up and 304 to duplex surveillance. Their baseline data
are shown in Table 1. Preoperative characteristics were
similar in the 2 randomized groups (median age 70 years,
72% male, and median ABPI 0.48). The majority of the
operations were from the common femoral (proximal anastomosis) to the above-knee or below-knee popliteal (distal
anastomosis) and were performed with ipsilateral reversed
leg vein. The most common indication for surgery was
critical ischemia.
The progress of patients throughout the trial is shown in
Figure 1. Apart from deaths, the withdrawal from follow-up
was 12% overall (11% and 13% in the clinical and duplex
groups, respectively). Of the withdrawals, 45% were due to
amputation. Among patients remaining in the trial, the proportion of follow-up appointments attended was 89% in the
clinical group and 90% in the duplex group. At 18 months,
91% of all patients due for follow-up had a duplex scan. The
response rate to the quality-of-life questionnaires was slightly
lower at ⬇80%.
Davies et al
TABLE 1. Patient Preoperative Characteristics and Operation
Details by Randomized Group
Median age (IQR), y
Male
Clinical Follow-Up
(n⫽290)
Duplex Follow-Up
(n⫽304)
70 (61 to 77)
70 (63 to 76)
210 (72)
218 (72)
Smoking
Current
81 (28)
81 (27)
Prior
179 (62)
174 (58)
Diabetes
98 (35)
83 (28)
Median ABPI (IQR)
0.48 (0.34 to 0.62)
0.49 (0.33 to 0.64)
Proximal anastomosis
Common femoral
217 (77)
218 (74)
Superficial femoral
61 (22)
74 (25)
4 (1)
3 (1)
Profunda femoris
Distal anastomosis
Above knee popliteal
82 (28)
97 (34)
Below knee popliteal
107 (37)
106 (37)
99 (34)
94 (31)
Single vessel
Vein used in graft
Ipsilateral
268 (92)
287 (94)
Reversed
192 (66)
200 (67)
13 (4)
11 (4)
Arm
Indication for surgery
Claudication
Critical ischemia
Popliteal aneurysm
92 (32)
90 (30)
190 (66)
202 (66)
8 (3)
12 (4)
Values are given as n (%) of patients unless stated otherwise. IQR indicates
interquartile range.
*For characteristics that were unknown for a few patients, percentages are
of patients with known values.
Some patients had additional radiological or surgical interventions (Table 2). The median time to first intervention was
20 weeks from randomization in the clinical arm and 15
weeks in the duplex arm. Twenty-seven percent of the clinical
group had a duplex scan at some time during the 18-month
follow-up period (owing to a suspicion of a clinical problem
from either history or a fall in ABPI); only 7% of the duplex
group had additional duplex scans beyond those in the
Duplex Surveillance of Leg Vein Grafts
1987
TABLE 2. Patients With Additional Radiological or Surgical
Interventions Over 18 Months’ Follow-Up
Allocated Follow-Up Group
Clinical
Follow-Up
(n⫽290)
Duplex
Follow-Up
(n⫽304)
P*
Additional duplex scan†
77 (27)
20 (7)
䡠䡠䡠
Angiogram
43 (15)
58 (19)
Any diagnostic intervention
90 (31)
66 (22)
䡠䡠䡠
0.01
Angioplasty
28 (10)
41 (13)
䡠䡠䡠
4 (1)
6 (2)
䡠䡠䡠
Surgery
20 (7)
28 (9)
Any therapeutic intervention
46 (16)
66 (22)
䡠䡠䡠
0.07
Intervention
Thrombolysis
Values are given as n (%) of patients.
*Fisher’s exact test.
†In addition to the 211 protocol-planned duplex scans performed in the
clinical follow-up group (at 18 months) and 1589 in the duplex follow-up group.
planned schedule. Angiograms, angioplasty, thrombolysis,
and surgery were each slightly more common in the duplex
group, as might be expected, but none to a very marked
extent. The reported interventional success rate was similar in
the clinical group and duplex group at 90%.
Table 3 indicates the methodology that first raised the
suspicion that a graft was at risk. It does not include the
asymptomatic lesions identified by the 18-month duplex scan
in the clinical arm, because this was used solely to calculate
the incidence of stenoses. Even in the duplex arm of the trial,
49% of patients were deemed to be potentially at risk by
history alone.
The major outcomes in the trial are shown in Table 4.
Amputations, vascular mortality, and overall mortality were
equally distributed between the 2 groups, so the hazard ratios
were close to unity. On the duplex scan at 18 months, the
proportion of patients with a stenosis in the graft (defined in
the protocol as a V2/V1 ratio ⬎2) was greater in the clinical
group.
The cumulative incidence of amputation is shown in Figure
2; there was no difference between the 2 groups. Graft
patency at each follow-up occasion is shown in Figure 3.
Patency diminished over time, primary patency being reTABLE 3. Methodology Whereby Grafts Were Identified as
Being at Risk for the First Time, Excluding Duplex Scans at 18
Months in Clinical Group
Detection Method
History only
ABPI only
Duplex only
History and ABPI
History and duplex
Duplex Group
140 (68)
108 (49)
55 (27)
36 (16)
0 (0)
49 (22)
11 (5)
8 (4)
0 (0)
12 (5)
ABPI and duplex
0 (0)
6 (3)
All methods
0 (0)
0 (0)
206 (100)
219 (100)
Total
Figure 1. CONSORT diagram of patients’ follow-up in the trial.
Clinical Group
Values are given as n (%).
1988
Circulation
September 27, 2005
TABLE 4. Major Outcomes in the Clinical and Duplex
Follow-Up Groups
Clinical
Follow-Up
(n⫽290)
Duplex
Follow-Up
(n⫽304)
Hazard Ratio*
(95% CI)
or P ‡
Clinical outcomes
Amputation
21 (7)
21 (7)
1.01 (0.55 to 1.86)
Vascular death†
10 (3)
12 (4)
1.21 (0.52 to 2.81)
Amputation or
vascular death†
29 (10)
33 (11)
1.15 (0.70 to 1.90)
All deaths
31 (11)
36 (12)
1.22 (0.75 to 1.98)
204
211
䡠䡠䡠
39 (19)
25 (12)
P⫽0.04
Patency outcome
No. of patients with
18-month duplex scan
Stenosis in graft
For clinical outcomes, values are n (%) of patients having amputations or
dying of vascular causes over 18 months’ follow-up with hazard ratio (95%
confidence interval). For patency outcome, values are proportions of patients
with a stenosis in the graft or with V2/V1⬎2 as assessed by duplex scan at 18
months.
*Withdrawals (and deaths or nonvascular deaths as appropriate) censored.
Adjusted hazard ratios were similar (see Methods).
†Deaths known to be of vascular cause.
‡P from ␹2 test.
placed by primary-assisted patency and secondary patency as
expected, but to a similar degree in both groups. The
Kaplan-Meier estimates at 18 months of the proportions with
primary, primary assisted, and secondary patency were 69%,
76%, and 80%, respectively, in the clinical group, and 67%,
76%, and 79%, respectively, in the duplex group. The median
ABPI showed no evidence of a difference between groups
over time.
Results from the quality-of-life assessments (Table 5) gave
no clear indication of a difference between randomized
groups at either 6 or 18 months. However, the average health
service cost per patient was higher in the duplex than in the
clinical follow-up group (mean difference £495, 95% CI
£183 to £807) because of the cost of duplex scans and the
slightly increased rates of intervention in the duplex group.
Discussion
The need for a further, larger RCT was demonstrated by the
results of the 2 small trials16 –18 and reflected in the Transat-
Figure 2. Cumulative incidence of amputation.
Figure 3. Kaplan-Meier plots of patency over time by trial arm.
A, Primary patency; B, primary assisted patency; and C, secondary patency.
lantic Consensus Statement.23 The present study is the largest
multicenter trial to examine the potential benefits of duplex
surveillance in terms of amputation and graft patency.
Overall, the trial has provided conclusive evidence of the
suspicions raised by the summation analysis of Golledge et
al5 and the combined results of the 2 small RCTs that limb
salvage is not improved by duplex surveillance.31 The combined results of the previous small RCTs suggested that
overall patency was worse in patients in the clinical follow-up
arm rather than the duplex group; this trial has shown no
statistically or clinically significant improvement in patency.
Table 6 compares patency rates and limb salvage from the
previous RCTs and the present trial. The 12-month point was
Davies et al
1989
TABLE 5. Quality-of-Life Assessments at 6 and 18 Months* and Health Service Costs
Over 18 Months
No. of Patients
Clinical Follow-Up
Duplex Follow-Up
P†
SF-36 physical score
447
47⫾27
50⫾30
0.19
SF-36 mental score
439
71⫾20
71⫾21
0.93
EQ5D utility score
443
0.59⫾0.30
0.63⫾0.30
0.06
SF-36 physical score
351
48⫾29
50⫾28
0.51
SF-36 mental score
352
71⫾21
74⫾21
0.15
EQ5D utility score
375
0.62⫾0.29
0.64⫾0.29
0.28
594
876⫾2035 (111)
1371⫾1837 (666)
0.002
6-Month outcomes
18-Month outcomes
Health service costs over 18
months
Cost per patient (£) (median)
Values are mean⫾SD except where indicated.
*Higher SF-36 and lower EQ5D scores each represent worse perceived health.
†P values from Mann-Whitney test for quality-of-life scores and from t test for costs.
used for comparison because the studies by Ihlberg et al17,18
only followed up patients to 12 months; we used the former
of the 2 publications, which reports on a larger number of
patients,17 although the latter acknowledges a degree of
difficulty with respect to the number of patients lost to
follow-up.18 The amputation rate in the present study is
comparable to that found in the summation analysis and other
studies. Similarly, the patency rates are comparable to these
other studies. However, the 18-month data with respect to the
incidence of vein graft stenoses are different. These results
are in line with the findings of the Bristol group,32 who
showed that in a cohort of patients who did not receive
treatment for a stenosis or inflow or outflow problems, there
was no difference in terms of patency. Mattos et al33 have also
previously concluded that the majority of stenoses stay patent
whether treated or not. Furthermore, it is well accepted that
the 1-year incidence of stenosis can be as high as 30%. The
present trial has shown a lower prevalence at 18 months
because this figure does not include patients who have had a
previous stenosis corrected.
One of the complicating factors in this area is the indication for determining that a graft is at risk. It is accepted that
duplex scanning is the noninvasive investigation of choice for
identifying an abnormality in a vein graft.6 The following are
the common noninvasive criteria for identifying an at-risk
graft: ABPI fall ⬎0.2, peak systolic velocity ⬍45 cm/s,
increase in peak systolic velocity at the site of the stenosis to
⬎150 cm/s, and peak systolic velocity ratio across a stenosis
⬎2.0.6 However, we used an ABPI fall of 0.1 because this is
TABLE 6.
the figure used by the 1997 standards recommendation of
Rutherford et al.25 Controversy exists as to when one should
intervene; for example, in one series of 46 patients with a
peak systolic velocity ratio ⬎3.0, only 14 grafts were revised,
and only 3 occluded during follow-up.34 Other factors that
may be deemed to be important are graft diameter, outflow,
and location of the distal anastomosis. In devising the present
trial, it was thought to be important to adopt a pragmatic
approach, so that determining exactly when to intervene
should follow local policy. Furthermore, in certain situations,
the exact type of intervention required may be controversial,
for example, whether to perform an endovascular procedure
(angioplasty) or an open revision (such as vein patch angioplasty or interposition graft). Hence, each center was given
the freedom to determine the type of intervention required for
a patient.19
Another issue is the length of follow-up programs. The
highest incidence of developing stenosis is within the first
year, after which there is a very low incidence. Despite this,
the Leicester group22 advocates life-long surveillance,
whereas others suggest that the majority of patients may only
require surveillance for the first 6 months.20,21 The present
trial confirms that the majority of interventions occur within
the first year after implantation.
Patency and limb salvage rates are not the only outcomes
that need consideration: The effect on quality of life and cost
are important. To date, neither has been reported in an RCT.
Improvements in quality of life after bypass surgery have
been well established previously,19 and the data from the
Twelve-Month Comparative Data on Patency and Limb Salvage From 3 Randomized Trials
Clinical Follow-Up
Ihlberg et al17
Lundell et al
Davies et al
16
Duplex Follow-Up
n
PP, %
PAP, %
SP, %
LS, %
n
PP, %
PAP, %
SP, %
LS, %
90
68
74
84
88
95
56
65
71
81
50
NA
82
85
NA
56
NA
74
76
NA
290
73
80
83
93
304
70
78
82
93
n indicates No. in each group; PP, primary patency; PAP, primary assisted patency; SP, secondary patency; LS, limb salvage; and
NA, not available.
1990
Circulation
September 27, 2005
present trial show no evidence of a difference in overall
quality-of-life scores between the 2 types of follow-up used.
Evidence suggests that there is no difference in outcome
between reversed and nonreversed long saphenous vein
grafts37 and that comparable patency can be obtained with
arm vein.38 Because arm vein grafts are often used in difficult
repeat surgery, there may be a mistaken impression of poorer
outcomes. In the present study, the number of arm veins used
was very small, and so no subgroup analysis was performed.
The cost of duplex surveillance is considerable; previous
estimates have suggested that at least a 5% per annum
improvement in limb salvages rates is required to justify a
surveillance program.13 Because primary amputation is more
expensive than successful reconstruction,39 it is tempting to
extrapolate these figures and suggest that interventions to
maintain patency are mandatory. However, not all stenoses
inevitably lead to critical leg ischemia.33 The present study
has confirmed this; however, with the higher incidence of
asymptomatic stenoses in the clinical arm at 18 months, it is
possible that they may have a longer-term impact. Interestingly, an economic study in the United States showed that the
mean costs of reconstruction and a 5-year surveillance program were the same as for primary amputation.40 With the
fact that limbs would be lost irrespective of the surveillance
strategy, the direct comparison with primary amputation is
difficult. In the present study, there was no evidence of a
difference in amputation rates, although there was a higher
intervention rate in the duplex group (some of which therefore could be deemed as unnecessary interventions).
In conclusion, this large RCT has shown no clinical benefit
or quality-of-life improvement in patients participating in a
duplex surveillance program after distal reconstruction despite increased financial costs. Hence, we can no longer
recommend the widespread use of duplex vein graft surveillance in the presence of close clinical follow-up.
Acknowledgments
This study was funded by the British Heart Foundation Project Grant
Number PG/97087. The British Heart Foundation grant applicants
were Mr Alun H. Davies, Dr David Torgerson, Prof Simon G.
Thompson, Mr Michael G. Wyatt, and Prof Roger M. Greenhalgh.
Information about trial participants and centers can be found at
http://circ.ahajournals.org/cgi/content/full/112/13/1985/
DC1.
References
1. Sayers RD, Raptis S, Berce M, Miller JH. Long-term results of femorotibial bypass with vein or polytetrafluoroethylene. Br J Surg. 1998;85:
934 –938.
2. Moody P, DeCossart LM, Douglas HM, Harris PL. Asymptomatic
strictures in femoropopliteal vein grafts. Eur J Vasc Surg. 1989;75:
737–740.
3. Szilagyi DE, Elliott JP, Hageman JH, Smith RF, Dall’olmo CA. Biologic
fate of autogenous vein implants as arterial substitutes: clinical, angiographic and histopathologic observations in femoro-popliteal operations
for atherosclerosis. Ann Surg. 1973; 178:232–246.
4. Mills JL. Mechanisms of vein graft failure: the location, distribution, and
characteristics of lesions that predispose to graft failure. Semin Vasc Surg.
1993;6:78 –91.
5. Golledge J, Beattie DK, Greenhalgh RM, Davies AH. Have the results of
infrainguinal bypass improved with the widespread utilisation of postoperative surveillance? Eur J Vasc Endovasc Surg. 1996;11:388 –392.
6. Davies AH, Magee TR, Tennant SGW, Baird RN, Horrocks M. Criteria
for identification of the “at risk” infrainguinal bypass graft. Eur J Vasc
Surg. 1994;8:315–319.
7. Moody P, Gould DA, Harris PL. Vein graft surveillance improves
patency in femoropopliteal bypass. Eur J Vasc Surg. 1990;4:117–121.
8. Grigg MJ, Nicolaides AN, Wolfe JH. Detection and grading of femorodistal vein graft stenoses: duplex velocity measurements compared with
angiography. J Vasc Surg. 1988;8:661– 666.
9. Wolfe JH, Thomas ML, Jamieson CW, Browse NL, Burnand KG, Rutt
DL. Early diagnosis of femorodistal graft stenoses. Br J Surg. 1987;74:
268 –270.
10. Bandyk DF, Kaebnick HW, Stewart GW, Towne JB. Durability of the in
situ saphenous vein arterial bypass: a comparison of primary and secondary patency. J Vasc Surg. 1987;5:256 –268.
11. Berkowitz HD, Greenstein SM. Improved patency in reversed femoralinfrapopliteal autogenous vein grafts by early detection and treatment of
the failing graft. J Vasc Surg. 1987;5:755–761.
12. Gibbs RGJ, Beattie DK, Greenhalgh RM, Davies AH. Vein graft surveillance: current trends. Br J Surg. 1997;84:63.
13. Grigg M, Nicolaides AN, Wolfe JH. Can postoperative surveillance of
femoro-distal vein grafts be justified? In: Greenhalgh RM, Jamieson CW,
Nicolaides AN, eds. Limb Salvage and Amputation in Vascular Disease.
Philadelphia, Pa: WB Saunders; 1988:259 –270.
14. Nehler MR, Moneta GL, Yeager RA, Edwards JM, Taylor LMJ, Porter
JM. Surgical treatment of threatened reversed infrainguinal vein grafts. J
Vasc Surg. 1994;20:558 –563.
15. Nielsen TG. Natural history of infrainguinal vein bypass stenoses: early
lesions increase the risk of thrombosis. Eur J Vasc Endovasc Surg.
1996;12:60 – 64.
16. Lundell A, Lindblad B, Bergqvist D, Hansen F. Femoropopliteal-crural
graft patency is improved by an intensive surveillance program: a prospective randomized study. J Vasc Surg. 1995;21:26 –33.
17. Ihlberg L, Luther M, Tierala E, Lepantalo M. The utility of duplex
scanning in infra-inguinal vein graft surveillance: results from a randomised controlled study. Eur J Vasc Endovasc Surg. 1998;16:19 –27.
18. Ihlberg L, Luther M, Alback A, Kantonen I, Lepantalo M. Does a
completely accomplished duplex-based surveillance prevent vein-graft
failure? Eur J Vasc Endovasc Surg. 1999;18:395– 400.
19. Kirby PL, Brady AR, Thompson SG, Torgerson D, Davies AH. The Vein
Graft Surveillance Trial: rationale, design and methods. Eur J Vasc
Endovasc Surg. 1999;18:469 – 474.
20. Idu MM, Buth J, Cuypers P, Hop WC, van de Pavoordt ED, Tordoir JM.
Economising vein-graft surveillance programs. Eur J Vasc Endovasc
Surg. 1998;15:432– 438.
21. Mills JL, Bandyk DF, Gathan V, Esses G. The origin of infrainguinal vein
graft stenosis: a prospective randomized trial based on duplex surveillance. J Vasc Surg. 1995;21:16 –25.
22. McCarthy MJ, Olojugba D, Loftus IM, Naylor AR, Bell PR, London NJ.
Lower limb surveillance following autologous vein bypass should be life
long. Br J Surg. 1998;85:1369 –1372.
23. TransAtlantic Inter-Society Consensus. Eur J Vasc Endovasc Surg. 2000;
19:216 –218.
24. Bandyk DF, Cato RF, Towne JB. A low flow velocity predicts failure of
femoropopliteal and femorotibial bypass grafts. Surgery. 1985;98:
799 – 809.
25. Rutherford RB, Baker JD, Ernst C, Johnston KW, Porter JM, Ahn S,
Jones DN. Recommended standards for reports dealing with lower
extremity ischemia: revised version. J Vasc Surg. 1997;26:517–538.
26. Garratt AM, Ruta DA, Abdalla MI, Buckingham JK, Russell IT. The
SF36 health survey questionnaire: an outcome measure suitable for
routine use within the NHS? BMJ. 1993;306:1440 –1444.
27. Brazier J, Jones N, Kind P. Testing the validity of the EuroQol and
comparing it with the SF-36 health survey questionnaire. Qual Life Res.
1993;2:169 –180.
28. van Agt AH, Essink-Bot ML, Krabbe PF, Bonsel GJ. Test-retest reliability of health state valuations collected with the EuroQol questionnaire.
Soc Sci Med. 1994;39:1537–1544.
29. Available at: http://www.dh.gov.uk/PublicationsAndStatistics/Publica
tions/. Accessed September, 2004.
30. Thompson SG, Barber JA. How should cost data in pragmatic randomised
trials be analysed? BMJ. 2000;320:1197–1200.
31. Hawdon AJ, Davies AH. Vein graft surveillance: is the yield worth the
effort? Acta Chi Belg. 2003;103:379 –382.
Davies et al
32. Wilson YG, Davies AH ,Currie IC, Morgan M, McGrath C, Baird RN,
Lamont PM. Vein graft stenosis: incidence and intervention. Eur J Vasc
Endovasc Surg. 1996;11:164 –169.
33. Mattos MA, van BP, Hodgson KJ, Ramsey DE, Barkmeier LD, Sumner
DS. Does correction of stenoses identified with color duplex scanning
improve infrainguinal graft patency? J Vasc Surg. 1993;17:54 – 64.
34. Dougherty MJ, Calligro KD, DeLaurentis DA The natural history of
“failing” arterial bypass grafts in duplex surveillance protocol. Ann Vasc
Surg 1998;12:255–259.
35. Deleted in proof. .
36. Beattie D, Ellis M, Davies AH. Graft Surveillance. In: Beard JD, Murray
S. Pathways of Care in Vascular Surgery. Shrewsbury, UK: TFM Publishing; 2002:107–116.
1991
37. Hozelbein TJ, Pomposelli FB, Miller A, Contreras MA, Gibbons GW,
Campbell DR, Freeman DV, LoGerfo FW. Results of a policy with
arms vein used as the first alternative to an unavailable ipsilateral
great saphenous vein for infrainguinal bypass. J Vasc Surg. 1996;23:
130 –140.
38. Harris PL, Veith FJ, Shanik GD, Nott D, Wengerter KR, Moore DJ.
Prospective randomised trial of in situ and reversed infrapopliteal vein
grafts. Br J Surg. 1993;80:173–176.
39. Singh S, Evans, Datta D, Gaines P, Beard JD. The costs of managing lower
limb-threatening ischaemia. Eur J Vasc Endovasc Surg. 1996;12:359–362.
40. Wixon CL, Mills JL, Westerband A, Hughes JD, Ihnat DM. An economic
appraisal of lower extremity bypass graft maintenance. J Vasc Surg.
2000;32:1–12.
Interventional Cardiology
Comparison of Percutaneous Coronary Intervention and
Coronary Artery Bypass Grafting After Acute Myocardial
Infarction Complicated by Cardiogenic Shock
Results From the Should We Emergently Revascularize Occluded
Coronaries for Cardiogenic Shock (SHOCK) Trial
Harvey D. White, DSc, FCSANZ; Susan F. Assmann, PhD; Timothy A. Sanborn, MD;
Alice K. Jacobs, MD; John G. Webb, MD; Lynn A. Sleeper, ScD;
Cheuk-Kit Wong, MD, FCSANZ; James T. Stewart, MD, FCSANZ;
Philip E.G. Aylward, MD, FCSANZ; Shing-Chiu Wong, MD; Judith S. Hochman, MD
Background—The Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock (SHOCK) trial
demonstrated the survival advantage of emergency revascularization versus initial medical stabilization in patients
developing cardiogenic shock after acute myocardial infarction. The relative merits of coronary artery bypass grafting
(CABG) versus percutaneous coronary intervention (PCI) in patients with shock have not been defined. The objective
of this analysis was to compare the effects of PCI and CABG on 30-day and 1-year survival in the SHOCK trial.
Methods and Results—Of the 302 trial patients, 128 with predominant left ventricular failure had emergency
revascularization. The selection of revascularization procedures was individualized. Eighty-one patients (63.3%) had
PCI, and 47 (36.7%) had CABG. The median time from randomization to intervention was 0.9 hours (interquartile range
[IQR], 0.3 to 2.2 hours) for PCI and 2.7 hours (IQR, 1.3 to 5.5 hours) for CABG. Baseline demographics and
hemodynamics were similar, except that there were more diabetics (48.9% versus 26.9%; P⫽0.02), 3-vessel disease
(80.4% versus 60.3%; P⫽0.03), and left main coronary disease (41.3% versus 13.0%; P⫽0.001) in the CABG group.
In the PCI group, 12.3% had 2-vessel and 2.5% had 3-vessel interventions. In the CABG group, 84.8% received ⱖ2
grafts, 52.2% received ⱖ3 grafts, and 87.2% were deemed completely revascularized. The survival rates were 55.6%
in the PCI group compared with 57.4% in the CABG group at 30 days (P⫽0.86) and 51.9% compared with 46.8%,
respectively, at 1 year (P⫽0.71).
Conclusions—Among SHOCK trial patients randomized to emergency revascularization, those treated with CABG had a
greater prevalence of diabetes and worse coronary disease than those treated with PCI. However, survival rates were
similar. Emergency CABG is an important component of an optimal treatment strategy in patients with cardiogenic
shock, and should be considered a complementary treatment option in patients with extensive coronary disease.
(Circulation. 2005;112:1992-2001.)
Key Words: angioplasty 䡲 mortality 䡲 myocardial infarction 䡲 shock 䡲 surgery
C
and initial medical stabilization, but 1-year survival rates
were higher with emergency revascularization.4,5 Overall,
most survivors had good quality of life.6 The protocol
specified that patients randomized to emergency revascularization should have either percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG)
ardiogenic shock is the commonest cause of death in
patients with acute myocardial infarction (AMI) who
reach hospital alive.1–3 In the international, multicenter
Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock (SHOCK) trial, 6-month survival rates were similar with emergency revascularization
Received February 13, 2005; revision received June 9, 2005; accepted July 5, 2005.
From Green Lane Cardiovascular Service, Auckland City Hospital, Auckland, New Zealand (H.D.W., J.T.S.); Center for Statistical Analysis and
Research, New England Research Institutes Inc, Watertown, Mass (S.F.A., L.A.S.); Division of Cardiology, Evanston Northwestern Healthcare, Evanston,
Ill (T.A.S.); Division of Cardiology and Vascular Medicine, Boston Medical Centre, Boston, Mass (A.K.J.); Division of Cardiology, St Paul’s Hospital,
Vancouver, British Columbia, Canada (J.G.W.); Department of Medical and Surgical Sciences, Otago University, Dunedin, New Zealand (C.-K.W.);
Cardiac Services, Flinders Medical Centre, Adelaide, South Australia (P.E.G.A.); Division of Cardiology, New York Weill Cornell Medical Center, New
York, NY (S.-C.W.); and Cardiovascular Clinical Research Center, New York University School of Medicine, New York, NY (J.S.H.).
Guest Editor for this article was Robert O. Bonow, MD.
Correspondence to Professor Harvey White, Green Lane Cardiovascular Service, Auckland City Hospital, Private Bag 92024, Auckland 1030, New
Zealand. E-mail [email protected]
1992
White et al
Cardiogenic Shock, Revascularization, and Mortality
within 6 hours of randomization and within 18 hours of the
onset of shock. Although recommendations were made
with regard to selection of revascularization procedures,
this decision was made on a case-by-case basis by site
investigators.
PCI and CABG are generally considered complementary
treatment options for patients with chronic stable angina,7 but
the relative merits and survival benefits of these procedures
may differ in patients with cardiogenic shock. PCI has a
lower procedural success rate in patients with shock than in
those without shock, and many patients have complex
3-vessel disease that may be treated better with CABG.8 –12
The benefits of CABG in patients with shock may include
protection of ischemic myocardium with cardioplegia, ventricular unloading during cardiopulmonary bypass, and revascularization of noninfarct zones.
We therefore compared the demographic and survival
differences of patients treated with CABG and those treated
with PCI in a prespecified analysis of patients randomized to
emergency revascularization in the SHOCK trial.
Methods
In the international SHOCK trial, 302 patients who developed
cardiogenic shock within 36 hours of the onset of AMI were
randomized to receive either emergency revascularization or initial
medical stabilization. Eligible patients required either ST-segment
elevation, Q-wave infarction, new left bundle-branch block, or a
posterior infarct with anterior ST-segment depression on the presenting ECG. Cardiogenic shock had to be due to predominant left
ventricular failure, and the trial excluded patients with isolated right
ventricular infarcts and mechanical causes of shock. Shock was
defined according to clinical and hemodynamic criteria including
hypotension (systolic blood pressure ⬍90 mm Hg for ⱖ30 minutes
or need for supportive measures to maintain systolic blood pressure
of ⱖ90 mm Hg), evidence of end-organ hypoperfusion, cardiac
index of ⱕ2.2 L · min⫺1 · m⫺2, and pulmonary capillary wedge
pressure of ⱖ15 mm Hg. Coronary stenosis was defined as a stenosis
of ⱖ50%. PCI was considered successful if Thrombolysis in Myocardial Infarction (TIMI) grade 2 or 3 flow was achieved with ⱖ20%
reduction of the treated stenosis and a residual stenosis of ⱕ50%.13
Complete revascularization with PCI was defined as successful PCI
of all segments with ⱖ50% stenoses in the proximal halves of the
right coronary artery, left circumflex artery, left main coronary
artery, left anterior descending coronary artery, and any previous
grafts. The completeness of revascularization in patients treated with
CABG was judged by the surgeon at the time of surgery, and the
proportion of myocardium at risk was estimated from the coronary
jeopardy score.14 Intra-aortic balloon pumping was recommended
for all patients. Additional details of the study design have been
published previously.4
Protocol Recommendations for
Revascularization Procedures
The protocol recommended that emergency PCI be performed only
on the infarct-related stenosis and only in patients with 1-, 2-, or
3-vessel disease where the stenoses in 2 non–infarct-related arteries
were ⬍90% or were located in arteries supplying small branch
vessels.
Patients with a left main coronary stenosis of ⱖ50%, ⱖ2 total or
subtotal occlusions, stenoses of ⬎90% in 2 non–infarct-related major
arteries, or stenoses unsuitable for PCI were recommended to
undergo CABG, as were patients whose PCI was unsuccessful. It
was also recommended that patients with 3-vessel disease who had
successful PCI of the infarct-related stenosis should be evaluated for
CABG of their remaining stenoses before hospital discharge. The 7
1993
patients who had both PCI and CABG were included in the PCI
group for all analyses reported here.
Selection of Revascularization Procedures
Selection of revascularization procedures was individualized for
each patient by site investigators. No data on how often the
investigator’s choice matched the protocol recommendations were
collected. Each site had designated and approved interventional and
surgical investigators.
Patients who were initially revascularized by PCI were compared
with those initially revascularized by CABG through the use of
Fisher’s exact test for categorical variables, the t test for normally
distributed continuous variables, and the Wilcoxon rank-sum test for
ordinal or skewed continuous variables. Logistic regression was used
to generate adjusted comparisons of mortality and to assess interactions between the revascularization modality and other patient
characteristics. Baseline characteristics included in the adjusted
models were those that differed between treatment groups at the 0.25
significance level, ie, characteristics that were at least moderately
imbalanced between the PCI and CABG groups. These included left
main coronary disease, diabetes, previous AMI, infarct location, time
from AMI to shock, and intra-aortic balloon pumping. Because of the
high correlation between different measures of disease severity, only
1 severity measure was included at a time in each model. The
Hosmer-Lemeshow test was used to assess whether any of the
logistic regression models poorly fitted the data. Values of P⬍0.05
were considered statistically significant. SAS software (SAS Institute Inc) was used for all computations.
Results
Of the 152 patients with cardiogenic shock who were randomized to emergency revascularization, 5 did not have
predominant left ventricular failure, and 5 died before cardiac
catheterization. Of the remaining patients, 14 had no revascularization procedure for various reasons (Figure 1). Two of
these patients died before revascularization could be attempted. Six had no significant stenoses, had TIMI grade 3
flow, or improved without revascularization; their survival
rate was 83.3% at both 30 days and 1 year. Six patients had
stenoses unsuitable for PCI or distal coronary arteries unsuitable for grafting; their survival rates were 33.3% at 30 days
and 16.7% at 1 year.
The subsequent analyses included the 128 patients with
predominant left ventricular failure who had emergency
revascularization. The initial revascularization procedure was
PCI in 81 patients and CABG in 47 patients (Figure 1). Ten
patients (7.8%) had their revascularization procedures (1 PCI
and 9 CABG) beyond the 6-hour time limit stipulated in the
protocol: 5 at 6 to 8 hours and 5 at ⱖ22 hours after
randomization.
The mean ages of the patients were 64.8⫾10.2 years in the
PCI group and 65.3⫾9.8 years in the CABG group (P⫽0.75;
Table 1). The other baseline demographics were also similar,
except diabetes was more prevalent in the CABG group
(48.9% versus 26.9% in the PCI group; P⫽0.02) and the
infarct location differed between the 2 groups (P⬍0.01) in
that patients in the CABG group were more likely to have
AMIs that were nonanterior or noninferior.
The median time from the onset of AMI to revascularization was shorter in patients treated with PCI (11.0 hours;
interquartile range [IQR], 6.1 to 21.4 hours) than in those
treated with CABG (19.1 hours; IQR, 10.4 to 30.5 hours;
1994
Circulation
September 27, 2005
Figure 1. Flow chart of patients treated
with emergency PCI vs emergency CABG in
the SHOCK trial.
P⬍0.001; Table 2). The median time from randomization to
PCI was 0.9 hours (IQR, 0.3 to 2.2 hours), and the median
time to CABG was 2.7 hours (IQR, 1.3 to 5.5 hours;
TABLE 1. Baseline Demographics of Emergency
Revascularization Patients With Cardiogenic Shock Resulting
From Predominant Left Ventricular Failure
PCI
(n⫽81)
CABG
(n⫽47)
64.8⫾10.2
65.3⫾9.8
0.75
Age ⱖ75 y, %
12.3
12.8
1.00
Male, %
63.0
70.2
Age, y*
Race, %
P
0.45
0.43
White
80.2
83.0
Black
4.9
4.3
Asian
6.2
10.6
Unknown
8.6
2.1
Smoker, %
56.5
51.1
0.70
Previous hypertension, %
52.5
51.1
1.00
0.02
Diabetes, %
26.9
48.9
Elevated cholesterol level, %
40.4
40.0
1.00
Peripheral vascular disease, %
13.8
21.2
0.39
4.9
6.5
0.70
Previous renal failure, %
Previous heart failure, %
Previous AMI, %
Previous CABG, %
5.1
2.1
0.65
24.7
36.2
0.22
3.7
0.0
0.30
10.3
6.4
0.53
Anterior
62.0
57.4
Inferior
36.7
27.7
Other
1.3
14.9
Previous PCI, %
⬍0.01
AMI location, %
*Mean⫾SD.
P⬍0.0001). Hemodynamics at the time of randomization,
which were often measured while the patients were on
inotropes and/or intra-aortic balloon pumps, were similar in
the 2 groups.
Seven patients in the PCI group had CABG after their
initial PCI (6 after successful PCI and 1 after unsuccessful
PCI). In 6 of the 7, CABG was performed within 24 hours of
the onset of shock (within 18 hours in 5 and at 19 hours in 1).
In 3 of the 6 PCI patients who subsequently had emergency
CABG, it was done as a planned, staged procedure after
successful PCI on the infarct-related stenosis. The remaining
patient had delayed CABG 15 days after the onset of shock.
Coronary Anatomy and Details of
Revascularization Procedures
Table 3 describes the distribution of coronary disease in the 2
groups. Three-vessel disease (80.4% versus 60.3%; P⫽0.03)
and left main coronary disease (41.3% versus 13.0%;
P⫽0.001) were more prevalent in the CABG group than in
the PCI group. There were also trends for the CABG group to
have more occlusions and more stenoses of ⬎90% in non–
infarct-related arteries. Patients in the CABG group had a
significantly higher mean coronary jeopardy score than those
in the PCI group.
In the PCI group, 69 patients (85.2%) had PCI only on the
infarct-related stenosis (Table 4). PCI was successful in
77.2% of patients, and complete revascularization was
achieved in 23.1%. Stents were used in 30 patients (37.0%),
and their use increased over time from 0% at the start of the
trial to 74.3% in 1997 to 1998. Overall, the use of adjunctive
glycoprotein IIb/IIIa inhibitors was low, with 0 being used at
the start of the trial and abciximab being used in 71.9% of
patients treated in the last 2 years of the trial. Of the 36
patients with 3-vessel disease who had successful PCI on the
White et al
TABLE 2.
1995
Management, Timing, and Baseline Hemodynamics
PCI (n⫽81)
CABG (n⫽47)
P
Eligible for fibrinolytic, %*
95.1
93.6
0.71
Fibrinolytic used, %
48.1
48.9
1.00
Intra-aortic balloon pump, %
88.9
97.9
0.09
Transfer to institution with revascularization
facilities, %
59.3
59.6
1.00
5.0 (2.1–10.5)
6.8 (3.0–14.0)
0.21
54.3
46.8
0.47
11.0 (6.1–21.4)
19.1 (10.4–30.5)
⬍0.001
5.3 (3.0–7.7)
8.9 (5.3–13.8)
⬍0.0001
Time from AMI to shock, h†
Development of shock within 6 h of AMI, %
Time from AMI to revascularization, h†
Time from shock to revascularization, h†
2.7 (1.3–5.5)
⬍0.0001
3142 (1599–6721)
3518 (1402–6204)
0.65
Heart rate, bpm§
101.1⫾22.6
103.6⫾20.2
0.53
Lowest systolic blood pressure, mm Hg§
66.5⫾11.5
65.1⫾18.5
0.68
Systolic blood pressure, mm Hg§
89.9⫾20.9
87.3⫾21.2
0.52
Diastolic blood pressure, mm Hg§
55.1⫾14.2
52.9⫾15.7
0.44
31⫾10.3
28.1⫾11.6
0.28
Time from randomization to revascularization, h†
Highest total creatine kinase level, ␮/L†
0.9 (0.3–2.2)
Hemodynamics‡
Ejection fraction, %§
Cardiac index, L䡠min⫺1䡠m⫺2§
Wedge pressure, mm Hg§
Cardiac power index§
1.8⫾0.7
1.8⫾0.8
0.71
24.6⫾7.6
24.2⫾6.0
0.80
120.9⫾56.6
119.2⫾46.0
0.87
*No absolute contraindication.
†Median (IQR).
‡Values were often measured while patients were receiving support.
§Mean⫾SD.
infarct-related stenosis, 3 patients subsequently had CABG
within 24 hours.
The details of CABG are shown in Table 5. Cardioplegia
was used in 86.1% of the 36 patients who had CABG after
institution of a cardiac surgery data collection form partway
through the trial. The mean number of grafts inserted during
CABG was 2.7⫾1.1. Left internal mammary arterial grafts
were used in 15.2% of patients. Complete revascularization
was achieved in 87.2% of patients, and concomitant valve
procedures were performed in 5.6%. CABG was performed
in similar percentages of patients randomized between midnight and 7:59 AM (31%), between 8 AM and 3:59 PM (38%),
and between 4 PM and 11:59 PM (37%; P⫽0.90).
Survival
The vital status of all patients was ascertained at 1 year. The
96-hour survival rates were 65.4% in the PCI group and
80.9% in the CABG group (P⫽0.07; Table 6 and Figure 2).
Both groups had similar survival rates at 30 days (55.6% with
PCI versus 57.4% with CABG; P⫽0.86) and at 1 year (51.9%
versus 46.8%, respectively; P⫽0.71; Figure 2), and there
were no differences in 30-day or 1-year survival in any
subgroup (Table 7). There was no indication that the association between the revascularization modality and 1-year
mortality changed over the duration of the study (1993 to
1994, 1995 to 1996, 1997 to 1998; P⫽0.69).
Severity of Coronary Disease, Revascularization
Modality, and Survival
There was no association between the revascularization
modality and survival in any category of disease severity
(Table 7). Of 29 patients with left main coronary disease, 10
had PCI and 19 had CABG; their 1-year survival rates were
30.0% and 47.4%, respectively (P⫽0.45). Of 56 patients with
3-vessel disease but no left main coronary disease, 37 had
PCI and 19 had CABG; their 1-year survival rates were
51.4% and 47.4%, respectively (P⫽1.00). Of 39 patients with
1- or 2-vessel disease but no left main coronary disease, 31
had PCI and 8 had CABG; their 1-year survival rates were
61.3% and 50.0%, respectively (P⫽0.69).
Three patients had PCI for a left main coronary stenosis.
Two died within 96 hours, and 1 was still alive at 1 year.
Success of PCI and Survival
Patients who had successful PCI had higher survival rates at 30
days (63.9% versus 22.2%; P⬍0.01) and at 1 year (60.7% versus
22.2%; P⬍0.01) than those who had unsuccessful PCI. The
30-day survival rates were 60.0% in patients who had any
stenosis stented versus 52.9% in patients receiving no stents
(P⫽0.64), and the 1-year survival rates were 56.7% versus
49.0%, respectively (P⫽0.65). Patients with complete revascularization had survival rates of 66.7% at 30 days and 61.1% at 1
year versus 51.7% at 30 days (P⫽0.29) and 50.0% at 1 year
(P⫽0.43) in patients with incomplete revascularization.
Success of CABG and Survival
CABG achieved complete revascularization in 41 patients
(87.2%) and incomplete revascularization in 6 patients
(12.8%). The survival rates were 63.4% versus 16.7%,
respectively, at 30 days (P⫽0.07) and 51.2% versus 16.7%,
respectively, at 1 year (P⫽0.19).
1996
Circulation
September 27, 2005
TABLE 3. Revascularization Modality Shown According to Extent and Severity of
Coronary Disease
PCI (n⫽81), %
CABG (n⫽47), %
P
ⱖ50% Stenosis in left main coronary
artery
13.0
41.3
0.001
3-Vessel disease
60.3
80.4
0.03
Either left main or 3-vessel coronary
disease
60.3
82.6
0.01
1
22.4
3.7
2
23.9
25.9
3
53.7
70.4
No left main coronary disease
Number of diseased vessels
0.08
Number of additional occlusions (other
than infarct-related artery)
0.41
0
70.3
56.0
1
21.9
36.0
2
7.8
8.0
Number of ⬎90% stenoses in
non–infarct-related arteries
0.36
0
64.1
48.0
1
26.6
40.0
2
9.4
12.0
16.7
0.0
⬍0.0001
Coronary jeopardy score*
2
4
6.4
6.5
6
28.2
6.5
8
16.7
8.7
10
19.2
43.5
12
Mean coronary jeopardy score†
12.8
34.8
7.1⫾3.2
9.9⫾2.3
⬍0.0001
*See Methods for definition.
†Mean⫾SD.
Age, Revascularization Modality, and Survival
Among patients ⬍75 years of age, there was no difference in
1-year survival between those treated with PCI and those
treated with CABG (56.3% versus 46.3%, respectively;
P⫽0.33; Table 7). Sixteen patients ⱖ75 years of age had
emergency revascularization, 10 by PCI and 6 by CABG;
their 1-year survival rates were 20.0% and 50.0%, respectively (P⫽0.30).
Multipredictor Model for Survival
Adjusted mortality models showed that there was no difference in 30-day or 1-year survival between patients treated
with PCI and those treated with CABG (Table 8). There was
no significant interaction between the revascularization modality (PCI versus CABG) and the presence of left main
coronary disease. Results were similar when other measures
of disease severity (3-vessel disease, left main and/or 3-vessel
disease, or coronary jeopardy score) were substituted for left
main disease in these models (data not shown). In the
adjusted 30-day mortality model, the area under the curve
was 0.675, and the likelihood ratio probability value for the
whole model was 0.096. In the adjusted 1-year mortality
model, the area under the curve was 0.645, and the likelihood
ratio probability value for the whole model was 0.242.
Discussion
The findings of this study suggest that PCI and CABG are
complementary treatment options for emergency revascularization in patients with cardiogenic shock. Although patients
treated with CABG had more extensive and more severe
coronary disease than patients treated with PCI, they had
similar survival rates at 30 days and 1 year, perhaps because
CABG achieved complete revascularization in a greater
proportion of patients than PCI did. It is interesting that there
was a trend for 96-hour survival to be higher with CABG.
This may have been related to the level of care provided in a
surgical intensive care unit.
Despite the protocol recommendations with regard to
selection of revascularization procedures, this analysis was
not limited by those recommendations because many patients
with 3-vessel disease had PCI. However, the analysis was
probably confounded because patients with 3-vessel disease
who were treated with PCI had less severe coronary disease
than those who were treated with CABG. In an emergency
White et al
TABLE 4.
1997
Details of PCI
Years, %
1993–1994
(n⫽17)
1995–1996
(n⫽29)
1997–1998
(n⫽35)
Successful revascularization*
76.5
67.9
85.3
77.2
Complete revascularization†
17.6
25.9
23.5
23.1
0.0
10.3
22.9
13.6
Multiple vessels treated during index procedure
Total
(n⫽81)
Any stents used
0.0
13.8
74.3
37.0
Stenting in infarct-related artery
0.0
10.3
68.6
33.3
1
100.0
86.2
77.1
85.2
2
0.0
6.9
22.9
12.3
3
0.0
6.9
0.0
2.5
50.0
71.9
69.4
Number of vessels treated
Use of abciximab (n⫽0, 4, 32, 36)‡
䡠䡠䡠
*Defined as a combination of residual stenosis of ⱕ50%, stenosis reduction of ⱖ20%, and achievement of TIMI
grade 2 or 3 flow.
†Complete revascularization was defined as successful PCI of all segments with ⱖ50% stenoses in the proximal
halves of the right coronary artery, left circumflex artery, left main coronary artery, left anterior descending coronary
artery, and any previous grafts (data available on only 78 patients).
‡Most of the data on abciximab use were collected on a stenting data collection form instituted partway through
the trial. If abciximab was used primarily in patients receiving stents, the percentage of patients receiving abciximab
may have been overestimated because patients not receiving stents may have had missing data for this variable.
situation such as cardiogenic shock, investigators may have
opted for emergency PCI on stenoses not ideally suited to
PCI, knowing that although CABG might achieve more
complete revascularization, PCI could be performed more
promptly. PCI may also have been performed on patients with
poor distal vessels. In addition, the further delay in reperfusion by CABG may suggest that more stable patients were
referred for CABG and that very unstable patients were
referred for PCI. However, this concept is not supported by
the hemodynamic data, which were similar in PCI and CABG
patients at baseline.
Diabetes was more prevalent in patients treated with
CABG than in those treated with PCI. The investigators’
preference for CABG in diabetic patients may have been
influenced by the findings of the Bypass Angioplasty Revascularization Investigation (BARI),15 in which diabetic patients with multivessel disease fared better after CABG than
after PCI. However, BARI predated the widespread use of
stenting and glycoprotein IIb/IIIa inhibitors, and a 7-year
follow-up of the BARI registry, in which the revascularization modality was selected by the attending clinician, showed
that diabetic patients had similar survival rates regardless of
whether they were treated with PCI or CABG.16 In the present
study, the outcome of diabetic patients was not predicted by
the revascularization modality. Restenosis has been shown to
be a powerful predictor of long-term mortality in diabetics,17
and new stenting technologies such as drug-eluting stents are
likely to improve the outcome of diabetic patients by reducing restenosis.18
In the current European Society of Cardiology/American
College of Cardiology guidelines,19 AMI with cardiogenic
shock is listed as a class IA indication (ie, a condition for
which there is evidence for and/or general agreement that a
given procedure/treatment is useful and effective) for PCI and
a class IA indication for CABG if the patient has suitable
coronary anatomy. However, emergency CABG is not widely
considered an integral part of contemporary management of
patients with cardiogenic shock. In the SHOCK trial,4 which
provided the evidence base for the guideline recommendations, 36.7% of patients with left ventricular failure who were
assigned to emergency revascularization had CABG as their
initial revascularization procedure. Recent data from the
National Registry of Myocardial Infarction indicate that
CABG is underused in patients with cardiogenic shock, with
only 4.9% having early CABG in 2001.20 This underuse may
reflect the logistical difficulties of arranging emergency
CABG for patients with cardiogenic shock, especially at night
or during weekends.
Findings from the SHOCK trial registry showed that
revascularization was associated with lower in-hospital mortality in patients with cardiogenic shock.21 In this registry of
884 patients with predominant left ventricular pump failure,
276 (31.2%) had PCI and 109 (12.3%) had CABG. The
in-hospital mortality rates were 78.0% in patients treated
medically, 46.4% in those treated with PCI, and 23.9% in
those treated with CABG (P⬍0.001). Patients with singlevessel disease had similar in-hospital mortality rates regardless of whether they were treated with PCI or CABG (32.9%
versus 33.3%). Patients with 2-vessel disease had higher
in-hospital mortality with PCI than with CABG (42.2%
versus 17.7%; P⫽0.025), as did patients with 3-vessel disease
(59.35% versus 29.6%; P⬍0.0001).
Unlike the SHOCK trial, the SHOCK trial registry was
nonrandomized, and patients who had diagnostic angiography
had more favorable hemodynamic findings than those who
did not have angiography, resulting in a selected population.22
No other randomized trials comparing revascularization with
medical treatment in patients with cardiogenic shock have
1998
Circulation
TABLE 5.
Details of CABG
September 27, 2005
Complete surgical revascularization, %*
87.2
No. of grafts,* %
1
15.2
2
32.6
3
32.6
4
10.9
5
8.7
Internal mammary arterial graft, %*
15.2
Data collected on cardiac surgery form (n⫽36)†
Median total perfusion time, min
110 (88–135)
Median total cross-clamp time, min
62 (49–78)
Concomitant valve procedure, %
5.6
Cardioplegia used, %
86.1
Cardioplegia delivery (in 31 patients with known
cardioplegia use), %†
Antegrade only
58.1
Retrograde only
6.4
Both
35.5
Types of cardioplegia (in 31 patients with
known cardioplegia use), %‡
Crystalloid
22.6
Blood
87.1
Additives
54.8
Numbers in parentheses are IQRs.
*These data were judged by the surgeon at the time of CABG.
†Data on perfusion time, cross-clamp time, concomitant valve procedures,
and cardioplegia were collected in only 36 patients after institution of a cardiac
surgery data collection form partway through the trial.
‡More than one type of cardioplegia was used during some procedures.
included CABG as part of the initial revascularization strategy. The Swiss Multicenter Trial of Angioplasty for Shock
(SMASH)23 was terminated early for logistical reasons after
55 patients had been randomized to receive either PCI or
medical treatment, and CABG was performed in only 1
patient.
The SHOCK trial protocol recommended PCI only for the
infarct-related stenosis, and only 13.6% of patients had
emergency PCI on ⬎1 vessel. Given concerns about the
possibility of continuing ischemia in other territories and the
higher procedural success rates now achieved with more
frequent use of stenting and glycoprotein IIb/IIIa inhibitors, it
may be appropriate to also treat additional stenoses during the
index procedure.24 However, PCI of non–infarct-related stenoses may cause deterioration in coronary flow as a result of
embolization of plaque or compromise of side branches in
both the infarct and noninfarct zones, thereby impairing
TABLE 6. Unadjusted Comparisons of 96-Hour, 30-Day, and
1-Year Survival
PCI (n⫽81), %
CABG (n⫽47), %
P
96 h
Survival
65.4
80.9
0.07
30 d
55.6
57.4
0.86
1y
51.9
46.8
0.71
Figure 2. Kaplan-Meier survival estimates at 96 hours (A), 30
days (B), and 1 year (C) in patients treated with emergency PCI
vs emergency CABG.
collateral blood supply to the infarct zone. In the SHOCK
trial, multivessel PCI was associated with a worse outcome
than single-vessel PCI.12 This finding may have been confounded by a treatment selection bias in that some patients
had PCI rather than CABG because they were considered
poor surgical candidates for various reasons.
The procedural success rate of 77.2% in patients who had
emergency PCI12 is similar to rates observed in other studies
of patients with cardiogenic shock.8,11 Although there have
been many advances in PCI (including stenting,12 which
reduces reintervention rates, and adjunctive use of glycoprotein IIb/IIIa inhibitors),25–28 patients with cardiogenic shock
still have a lower likelihood of successful PCI than patients
without shock.4,8 –11 For example, a recent German registry
study of 1333 patients with cardiogenic shock reported that
PCI achieved TIMI 3 flow in only 75.2% of patients.29 In
White et al
TABLE 7. Subgroup Analyses of the PCI and CABG Groups for
30-Day and 1-Year Survival
TABLE 8.
Mortality
1999
Covariate-Adjusted Models for 30-Day and 1-Year
PCI
(n⫽81), %
CABG
(n⫽47), %
P
Odds Ratio (95% CI)
Age ⬍75 y
60.6
58.5
0.84
Left main coronary disease
0.41
Age ⱖ75 y
20.0
50.0
0.30
Diabetes
0.40
0.22
Previous AMI
0.06
Infarct location*
0.04
Survival at 30 d
Mortality at 30 d
Emergency PCI (vs emergency CABG)
Interaction P
30.0
63.2
0.13
59.7
55.6
0.82
Interaction P
0.10
Left main and/or 3-vessel
coronary disease
48.9
57.9
0.51
Neither left main nor 3-vessel
coronary disease, %
64.5
62.5
1.00
Interaction P
0.63
0.13
0.57
Mortality at 1 y
Emergency PCI (vs emergency CABG)
Diabetes
0.81
Previous AMI
0.28
Infarct location*
0.10
Intra-aortic balloon pump
0.13
0.78
Time from AMI to shock
1- or 2-vessel disease with no left
main coronary disease
64.5
62.5
1.00
*2 df.
3-vessel disease with no left main
coronary disease
54.1
52.6
1.00
61.9
60.9
56.1
54.2
Interaction P
1.00
1.00
0.96
Survival at 1 y
Age ⬍75 y
56.3
46.3
0.33
Age ⱖ75 y
20.0
50.0
0.30
Interaction P
0.14
30.0
47.4
0.45
56.7
48.1
0.50
Interaction P
0.25
Left main and/or 3-vessel
coronary disease
46.8
47.4
1.00
Neither left main nor 3-vessel
coronary disease
61.3
50.0
0.69
Interaction P
0.60
30.0
47.4
0.45
1- or 2-vessel disease with no left
main coronary disease
61.3
50.0
0.69
3-vessel disease with no left main
coronary disease
51.4
47.4
1.00
Interaction P
0.55
Diabetic
52.4
47.8
Nondiabetic
54.4
45.8
Interaction P
1.00
0.63
0.84
0.92
0.34
0.13
Nondiabetic
0.95 (0.39–2.31)
63.2
Diabetic
0.49
Time from AMI to shock
30.0
0.32
1.38 (0.55–3.46)
Intra-aortic balloon pump
Interaction P
P
another study of 369 patients with ST-elevation AMI (including 23 patients with cardiogenic shock), drug-eluting stents
were compared with bare metal stents and resulted in similar
postprocedural vessel patency rates and 30-day complication
rates. Patients receiving drug-eluting stents had no increase in
stent thrombosis and were less likely to require reintervention
within 300 days of follow-up.30
In the SHOCK trial, only 37% of patients received stents,
and only 69% received abciximab. A prospective registry
study of patients having PCI for cardiogenic shock at the
Cleveland Clinic27 showed that stenting increased the likelihood of TIMI grade 3 flow and that adjunctive use of
abciximab with stenting increased the survival rate at 1
year.27 In the Abciximab Before Direct Angioplasty and
Stenting in Myocardial Infarction Regarding Acute and
Long-Term Follow-Up (ADMIRAL) study,31 25 patients
presented with cardiogenic shock. Those given adjunctive
abciximab before stenting tended to have a lower 6-month
event rate (death, AMI, or urgent target-vessel revascularization) than those given a placebo (9.1% versus 28.6%;
P⫽0.23). In the Platelet Glycoprotein IIb/IIIa in Unstable
Angina: Receptor Suppression Using Integrilin Therapy
(PURSUIT) trial of eptifibatide in patients with non–STsegment elevation acute coronary syndromes, 237 patients
developed cardiogenic shock after enrollment. Those given
eptifibatide had a lower 30-day mortality rate than those
given a placebo (adjusted odds ratio, 0.51; 95% CI, 0.28 to
0.94).28
In the SHOCK trial, 84.8% of patients treated with CABG
received ⬎1 graft. Left internal mammary arterial grafts were
used in 15.2% of patients. In 87.2% of patients, the surgeons
judged that complete revascularization had been achieved.
Greater use of arterial grafts and advances in cardioplegia and
anesthesia might produce even better results than those
observed in this trial.
2000
Circulation
September 27, 2005
This analysis had a number of limitations, including the
nonrandomized nature of the study and the small number of
patients in each treatment group. The treatment strategies
used and the time to treatment (particularly in the CABG
group) may not be representative of contemporary practice. In
addition, multivessel PCI was performed infrequently, and
the degree of revascularization achieved was less than that
currently recommended.24,32 It is not yet known whether a
strategy of acute stenting of the culprit vessel in AMI with
subsequent elective treatment of nonculprit stenoses is preferable to complete revascularization in the acute phase. The
next logical step would be to perform a randomized comparison of emergency CABG and emergency PCI in patients
with cardiogenic shock, including liberal use of drug-eluting
stents and glycoprotein IIb/IIIa inhibitors.
Conclusions
Among patients randomized to emergency revascularization
in the SHOCK trial, those selected for CABG were more
likely to have diabetes and to have more severe coronary
disease than those selected for PCI. Despite this disparity in
risk factors between the 2 groups, the survival rates at 30 days
and 1 year were similar. In cases in which PCI is unlikely to
achieve complete revascularization or there are associated
mechanical complications or left main or severe 3-vessel
coronary disease, CABG should be performed. Emergency
CABG is an important component of an early invasive
strategy in patients with cardiogenic shock.
Acknowledgments
This work was supported by grants R01-HL-0020-018Z and HL49970 from the National Heart, Lung, and Blood Institute, National
Institutes of Health, Bethesda, Md. Professor White received partial
salary funding from the Green Lane Research and Educational Fund
Board (Auckland, New Zealand). We gratefully acknowledge the
patients and investigators who participated in the SHOCK trial. We
also thank Charlene Nell and Barbara Semb for secretarial assistance
and Anna Breckon, ELS, for editorial assistance.
References
1. Hands ME, Rutherford JD, Muller JE, Davies G, Stone PH, Parker C,
Braunwald E. The in-hospital development of cardiogenic shock after
myocardial infarction: incidence, predictors of occurrence, outcome and
prognostic factors. J Am Coll Cardiol. 1989;14:40 – 46.
2. Holmes DR Jr, Bates ER, Kleiman NS, Sadowski Z, Horgan JH, Morris
DC, Califf RM, Berger PB, Topol EJ. Contemporary reperfusion therapy
for cardiogenic shock: the GUSTO-I trial experience: the GUSTO-I
Investigators: Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries. J Am Coll Cardiol. 1995;
26:668 – 674.
3. Goldberg RJ, Samad NA, Yarzebski J, Gurwitz J, Bigelow C, Gore JM.
Temporal trends in cardiogenic shock complicating acute myocardial
infarction. N Engl J Med. 1999;340:1162–1168.
4. Hochman JS, Sleeper LA, Webb JG, Sanborn TA, White HD, Talley JD,
Buller CE, Jacobs AK, Slater JN, Col J, McKinlay SM, LeJemtel TH,
Picard MH, Menegus MA, Boland J, Dzavik V, Thompson CR, Wong
SC, Steingart R, Forman R, Aylward PE, Godfrey E, Desvigne-Nickens
P, for the SHOCK Investigators. Early revascularization in acute myocardial infarction complicated by cardiogenic shock. N Engl J Med.
1999;341:625– 634.
5. Hochman JS, Sleeper LA, White HD, Dzavik V, Wong SC, Menon V,
Webb JG, Steingart R, Picard MH, Menegus MA, Boland J, Sanborn T,
Buller CE, Modur S, Forman R, Desvigne-Nickens P, Jacobs AK, Slater
JN, LeJemtel TH. One-year survival following early revascularization for
cardiogenic shock. JAMA. 2001;285:190 –192.
6. Sleeper LA, Ramanathan K, Picard MH, LeJerntel TH, White HD,
Dzavik V, Tormey D, Avis NE, Hochman JS. Functional status and
quality of life after emergency revascularization for cardiogenic shock
complicating acute myocardial infarction. J Am Coll Cardiol. 2005;46:
266 –273.
7. White HD. Angioplasty versus bypass surgery. Lancet. 1995;346:
1174 –1175. Commentary.
8. Antoniucci D, Valenti R, Santoro GM, Bolognese L, Trapani M, Moschi
G, Fazzini PF. Systematic direct angioplasty and stent-supported direct
angioplasty therapy for cardiogenic shock complicating acute myocardial
infarction: in-hospital and long-term survival. J Am Coll Cardiol. 1998;
31:294 –300.
9. Moosvi AR, Khaja F, Villanueva L, Gheorghiade M, Douthat L,
Goldstein S. Early revascularization improves survival in cardiogenic
shock complicating acute myocardial infarction. J Am Coll Cardiol.
1992;19:907–914.
10. Hibbard MD, Holmes DR Jr, Bailey KR, Reeder GS, Bresnahan JF, Gersh
BJ. Percutaneous transluminal coronary angioplasty in patients with cardiogenic shock. J Am Coll Cardiol. 1992;19:639 – 646.
11. Gacioch GM, Ellis SG, Lee L, Bates ER, Kirsh M, Walton JA, Topol
EJ. Cardiogenic shock complicating acute myocardial infarction: the use
of coronary angioplasty and the integration of the new support devices
into patient management. J Am Coll Cardiol. 1992;19:647– 653.
12. Webb JG, Lowe AM, Sanborn TA, White HD, Sleeper LA, Carere RG,
Buller CE, Wong SC, Boland J, Dzavik V, Porway M, Pate G, Bergman
G, Hochman JS, for the SHOCK Investigators. Percutaneous coronary
intervention for cardiogenic shock in the SHOCK trial. J Am Coll
Cardiol. 2003;42:1380 –1386.
13. TIMI Study Group. Comparison of invasive and conservative strategies
after treatment with intravenous tissue plasminogen activator in acute
myocardial infarction: results of the Thrombolysis in Myocardial
Infarction (TIMI) Phase II trial. N Engl J Med. 1989;320:618 – 627.
14. Califf RM, Phillips HRI, Hindman MC, Mark DB, Lee KL, Behar VS,
Johnson RA, Pryor DB, Rosati RA, Wagner GS. Prognostic value of a
coronary artery jeopardy score. J Am Coll Cardiol. 1985;5:1055–1063.
15. Bypass Angioplasty Revascularization Investigation (BARI) Investigators. Comparison of coronary bypass surgery with angioplasty in
patients with multivessel disease. N Engl J Med. 1996;335:217–225.
16. Feit F, Brooks MM, Sopko G, Keller NM, Rosen A, Krone R, Berger PB,
Shemin R, Attubato MJ, Williams DO, Frye R, Detre KM. Long-term
clinical outcome in the Bypass Angioplasty Revascularization Investigation Registry: comparison with the randomized trial: BARI Investigators. Circulation. 2000;101:2795–2802.
17. Van Belle E, Ketelers R, Bauters C, Perie M, Abolmaali K, Richard F,
LaBlanche JM, McFadden EP, Bertrand ME. Patency of percutaneous
transluminal coronary angioplasty sites at 6-month angiographic followup: a key determinant of survival in diabetics after coronary balloon
angioplasty. Circulation. 2001;103:1218 –1224.
18. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin M,
Colombo A, Schuler G, Barragan P, Guagliumi G, Molnar F, Falotico R.
A randomized comparison of a sirolimus-eluting stent with a standard
stent for coronary revascularization. N Engl J Med. 2002;346:1773–1780.
19. Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand M,
Hochman JS, Krumholz HM, Kushner FG, Lamas GA, Mullany CJ,
Ornato JP, Pearle DL, Sloan MA, Smith SC Jr, Alpert JS, Anderson JL,
Faxon DP, Fuster V, Gibbons RJ, Gregoratos G, Halperin JL, Hiratzka
LF, Hunt SA, Jacobs AK. ACC/AHA guidelines for the management of
patients with ST-elevation myocardial infarction: executive summary: a
report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise
the 1999 Guidelines for the Management of Patients With Acute Myocardial Infarction). Circulation. 2004;110:588 – 636.
20. Babaev A, Every N, Frederick P, Sichrovsky T, Hochman JS. Trends in
revascularization and mortality in patients with cardiogenic shock complicating acute myocardial infarction: observations from the National
Registry of Myocardial Infarction. Circulation. 2002;106(suppl II):
II-364. Abstract.
21. Webb JG, Sanborn TA, Sleeper LA, Carere RG, Buller CE, Slater JN,
Baran KW, Koller PT, Talley JD, Porway M, Hochman JS. Percutaneous
coronary intervention for cardiogenic shock in the SHOCK trial registry.
Am Heart J. 2001;141:964 –970.
22. Wong SC, Sleeper LA, Monrad ES, Menegus MA, Palazzo A, Dzavik V,
Jacobs A, Jiang X, Hochman JS. Absence of gender differences in clinical
outcomes in patients with cardiogenic shock complicating acute myo-
White et al
23.
24.
25.
26.
27.
cardial infarction: a report from the SHOCK trial registry. J Am Coll
Cardiol. 2001;38:1395–1401.
Urban P, Stauffer JC, Bleed D, Khatchatrian N, Amann W, Bertel O, van
den Brand M, Danchin N, Kauffmann U, Meier B, Machecourt J, Pfisterer
M, for the (S)MASH Investigators. A randomized evaluation of early
revascularization to treat shock complicating acute myocardial infarction:
the (Swiss) Multicenter Trial of Angioplasty for Shock—(S)MASH. Eur
Heart J. 1999;20:1030 –1038.
Montalescot G, Andersen HR, Antoniucci D, Betriu A, de Boer MJ, Grip
L, Neumann FJ, Rothman MT. Recommendations on percutaneous
coronary intervention for the reperfusion of acute ST elevation myocardial infarction. Heart. 2004;90:e37.
Stone GW, Grines CL, Cox DA, Garcia E, Tcheng JE, Griffin JJ, Guagliumi G, Stuckey T, Turco M, Carroll JD, Rutherford BD, Lansky AJ,
for the Controlled Abciximab and Device Investigation to Lower Late
Angioplasty Complications (CADILLAC) Investigators. Comparison of
angioplasty with stenting, with or without abciximab, in acute myocardial
infarction. N Engl J Med. 2002;346:957–966.
Giri S, Mitchel J, Azar RR, Kiernan FJ, Fram DB, McKay RG, Mennett
R, Clive J, Hirst JA. Results of primary percutaneous transluminal
coronary angioplasty plus abciximab with or without stenting for acute
myocardial infarction complicated by cardiogenic shock. Am J Cardiol.
2002;89:126 –131.
Chan AW, Chew DP, Bhatt DL, Moliterno DJ, Topol EJ, Ellis SG.
Long-term mortality benefit with the combination of stents and abciximab
for cardiogenic shock complicating acute myocardial infarction.
Am J Cardiol. 2002;89:132–136.
2001
28. Hasdai D, Harrington RA, Hochman JS, Califf RM, Battler A, Box JW,
Simoons ML, Deckers J, Topol EJ, Holmes DR Jr. Platelet glycoprotein
IIb/IIIa blockade and outcome of cardiogenic shock complicating acute
coronary syndromes without persistent ST-segment elevation. J Am Coll
Cardiol. 2000;36:685– 692.
29. Zeymer U, Vogt A, Zahn R, Weber MA, Tebbe U, Gottwik M, Bonzel T,
Senges J, Neuhaus KL, for the Arbeitsgemeinschaft Leitende Kardiologische Krankenhausärtze (ALKK). Predictors of in-hospital mortality in
1333 patients with acute myocardial infarction complicated by cardiogenic shock treated with primary percutaneous coronary intervention
(PCI): results of the primary PCI registry of the Arbeitsgemeinschaft
Leitende Kardiologische Krankenhausärtze (ALKK). Eur Heart J. 2004;
25:322–328.
30. Lemos PA, Saia F, Hofma SH, Daemen J, Ong AT, Arampatzis CA, Hoye
A, McFadden E, Sianos G, Smits PC, van der Giessen WJ, de Feyter P,
van Domburg RT, Serruys PW. Short- and long-term clinical benefit of
sirolimus-eluting stents compared to conventional bare stents for patients
with acute myocardial infarction. J Am Coll Cardiol. 2004;43:704 –708.
31. Montalescot G, Barragan P, Wittenberg O, Elhadad S, Lefevre T,
Loubeyre C, Lafont A, Zupan M, Paganelli F, Pinton P. Primary stenting
with abciximab in acute myocardial infarction complicated by cardiogenic shock (the ADMIRAL trial). J Am Coll Cardiol. 2002;39(suppl
A):43A. Abstract.
32. Silber S, Albertsson P, Aviles FF, Camici PG, Colombo A, Hamm C,
Jorgensen E, Marco J, Nordrehaug JE, Ruzyllo W, Urban P, Stone GW,
Wijns W. Guidelines for percutaneous coronary interventions: the Task
Force for Percutaneous Coronary Interventions of the European Society
of Cardiology. Eur Heart J. 2005;26:804 – 847.
Rapamycin, but Not FK-506, Increases Endothelial Tissue
Factor Expression
Implications for Drug-Eluting Stent Design
Jan Steffel, MD*; Roberto A. Latini, MD*; Alexander Akhmedov, PhD; Dorothee Zimmermann, BSc;
Pamela Zimmerling, BSc; Thomas F. Lüscher, MD; Felix C. Tanner, MD
Background—Drugs released from stents affect the biology of vascular cells. We examined the effect of rapamycin and
FK-506 on tissue factor (TF) expression in human aortic endothelial cells (HAECs) and vascular smooth muscle cells
(HAVSMCs).
Methods and Results—Rapamycin enhanced thrombin- and tumor necrosis factor (TNF)-␣–induced endothelial TF
expression in a concentration-dependent manner. The maximal increase was 2.5-fold more pronounced than that by
thrombin or TNF-␣ alone and was paralleled by a 1.4-fold higher TF surface activity compared with thrombin alone.
Rapamycin by itself increased basal TF levels by 40%. In HAVSMCs, rapamycin did not affect thrombin- or
TNF-␣–induced TF expression. In contrast to rapamycin, FK-506 did not enhance thrombin- or TNF-␣–induced
endothelial TF expression. Thrombin induced a transient dephosphorylation of the mammalian target of rapamycin
downstream target p70S6 kinase. Rapamycin completely abrogated p70S6 kinase phosphorylation, but FK-506 did not.
FK-506 antagonized the effect of rapamycin on thrombin-induced TF expression. Rapamycin did not alter the pattern
of p38, extracellular signal–regulated kinase, or c-Jun NH2-terminal kinase phosphorylation. Real-time polymerase
chain reaction analysis revealed that rapamycin had no influence on thrombin-induced TF mRNA levels for up to 2
hours but led to an additional increase after 3 and 5 hours.
Conclusions—Rapamycin, but not FK-506, enhances TF expression in HAECs but not in HAVSMCs. This effect requires
binding to FK binding protein-12, is mediated through inhibition of the mammalian target of rapamycin, and partly
occurs at the posttranscriptional level. These findings may be clinically relevant for patients receiving drug-eluting
stents, particularly when antithrombotic drugs are withdrawn or ineffective, and may open novel perspectives for the
design of such stents. (Circulation. 2005;112:2002-2011.)
Key Words: endothelium 䡲 myocardial infarction 䡲 signal transduction 䡲 stents 䡲 thrombosis
P
ercutaneous coronary intervention with stenting of the
culprit lesion is the preferred treatment for patients with
acute coronary syndromes.1–3 Several clinical trials have
demonstrated that drug-eluting stents (DESs) are superior to
bare-metal stents (BMSs) by decreasing the restenosis rates
as well as major adverse cardiac events.4 – 6 Rapamycin
(sirolimus), a macrocyclic lactone, is used on DESs because
the drug inhibits proliferation and migration of vascular
smooth muscle cells (VSMCs).7 FK-506 (tacrolimus), a
macrolide immunosuppressant, is an alternative drug used
with DESs.8,9 Despite reduced restenosis rates, however, stent
thromboses have not decreased with DESs compared with
BMSs.6,10 –12 Indeed, several hundred cases of in-stent thrombosis have been reported with rapamycin-coated stents,13 and
results from a recent multicenter registry imply that throm-
bosis rates with DESs may be higher in “real world” patients
than reported in previous clinical trials.14 The reason for the
discrepancy between reduced restenosis rates and unaltered
or even enhanced thrombosis rates with DESs compared with
BMSs is not known.6,12
Several factors are involved in the pathogenesis of in-stent
thrombosis. These include procedure-related factors such as
mechanical vessel injury or incomplete stent apposition,
patient-related factors such as vessel size or coagulation
activity, and finally, the thrombogenicity of the stent itself.15
It has not yet been explored, however, whether the drugs used
for stent coating could be involved in the development of
in-stent thrombosis.15
Tissue factor (TF), a 263-residue, membrane-bound glycoprotein, is a key enzyme in the initiation of coagulation; it
Received January 7, 2005; de novo received June 15, 2005; accepted July 8, 2005.
From Cardiovascular Research, Physiology Institute, and the Center for Integrative Human Physiology (J.S., R.A.L., A.A., D.Z., P.Z., T.F.L., F.C.T.),
University of Zurich, and the Department of Cardiology (J.S., T.F.L., F.C.T.), Cardiovascular Center, University Hospital Zürich, Zürich, Switzerland.
*The first 2 authors contributed equally to the study.
Guest Editor for this article was James T. Willerson, MD.
Correspondence to Felix C. Tanner, MD, Cardiovascular Research, Physiology Institute, University of Zurich, Winterthurerstrasse 190, CH-8057
Zürich, Switzerland. E-mail [email protected]
2002
Steffel et al
activates factor X (FX) by binding activated factor VII
(FVIIa), which ultimately leads to thrombin formation. Initiation of coagulation is a key event in the pathogenesis of
thrombosis and acute coronary syndromes. Not surprisingly,
atheromatous plaques contain a variety of cells expressing
TF, including endothelial cells (ECs) and VSMCs. Moreover,
TF levels are elevated in the plasma and atherectomy samples
from patients with unstable angina.16 Therefore, TF seems to
be involved in the development of atherosclerosis and restenosis after percutaneous coronary intervention.17–19 TF may
indeed play a major role in stent thrombosis as well. However, the effect of neither rapamycin nor FK-506 on TF
expression has been investigated so far. Moreover, the role of
the mammalian target of rapamycin (mTOR) in regulating TF
expression is also not known. Thus, the present study was
designed to investigate the influence of rapamycin and
FK-506 on TF expression in human aortic endothelial cells
(HAECs) and vascular smooth muscle cells (HAVSMCs).
Methods
Rapamycin Increases TF Expression
2003
minute, and an elongation phase at 72°C for 1 minute. A melting
curve analysis was performed after amplification to verify the
homogeneity of the amplicon. For verification of amplicon size, PCR
products were analyzed on an ethidium bromide–stained 1% agarose
gel. In each real-time PCR run for TF and L28, a calibration curve
generated from serial dilutions of a known TF and L28 standard,
respectively, was included, and for each sample, the target values
were corrected by those for L28.
TF Surface Activity
A colorimetric assay (American Diagnostica) was used to analyze TF
surface activity according to the manufacturer’s recommendations,
with some modifications as described.22,25 Cells were grown in
6-well plates; after stimulation, cells were washed twice with
phosphate-buffered saline and incubated with human FVIIa and FX
at 37°C, resulting in the formation of a TF/FVIIa complex at the cell
surface. The TF/FVIIa complex converted human FX to FXa, which
was subsequently measured by its ability to cleave a chromogenic
substrate. Different concentrations of lipidated human TF were used
as positive controls to confirm that the obtained results were in the
linear range of detection (data not shown).
Proliferation
HAECs and HAVSMCs were cultured as described.20,21 Cells were
grown to confluence in 6-cm culture dishes and rendered quiescent
for 24 hours before stimulation with thrombin or tumor necrosis
factor (TNF)-␣ (Sigma). Rapamycin, wortmannin (both from
Sigma), FK-506 (Alexis), and LY294002 (Cell Signaling) were
added to the dishes 60 minutes before stimulation. Cytotoxicity was
assessed with a colorimetric assay to detect lactate dehydrogenase
release according to the manufacturer’s recommendations (Roche).
To examine the effect of rapamycin and FK-506 on EC proliferation,
HAECs were seeded on 6-cm dishes at 7000 cells/cm2. After 24
hours, when cells had reached ⬇50% confluence, they were serumstarved for 24 hours before incubation with rapamycin (10⫺7 mol/L),
FK-506 (10⫺7 mol/L), or carrier (0.1% dimethyl sulfoxide) in
endothelial basal medium (EBM, Clonetics) containing 10% fetal
calf serum (FCS). At the indicated times, cells were gently
trypsinized and counted in a hemacytometer. Each analysis was
performed in duplicate; results are representative of 3 independent
experiments.
Western Blot Analysis and ELISA
Apoptosis
Protein expression was determined by Western blot analysis as
described.22,23 Cells were lysed in 50 mmol/L Tris buffer, and 30-␮g
samples were loaded and separated by 10% sodium dodecyl sulfatepolyacrylamide gel electrophoresis. Proteins were transferred to a
polyvinylidene difluoride membrane (Millipore) by semidry transfer.
Antibody to human TF (American Diagnostica) was used at 1:2000
dilution; antibodies against the phosphorylated Thr-389 residue of
p70S6 kinase (S6K), phosphorylated p38 mitogen-activated protein
(MAP) kinase (p38), phosphorylated p44/42 MAP kinase (extracellular signal–regulated kinase [ERK]), and phosphorylated c-Jun
NH2-terminal kinase (JNK; all from Cell Signaling) were used at
1:3000, 1:1000, 1:5000, and 1:1000 dilution, respectively. Antibodies against total S6K, total p38, total ERK, and total JNK (all from
Cell Signaling) were used at 1:3000, 1:2000, 1:10 000, and 1:1000
dilution, respectively. All blots were normalized to ␣-tubulin (aT)
expression (1:20 000 dilution, Sigma). Endothelial TF expression
was also measured with a commercially available ELISA (American
Diagnostica) according to the supplier’s recommendations.
To assess induction of apoptosis by rapamycin and FK-506, cells
were cultured in chamber slides (Nunc) at 20 000 cells/well for 24
hours before serum-starvation for 24 hours. Cells were then incubated in EBM with 10% FCS containing rapamycin (10⫺7 mol/L),
FK-506 (10⫺7 mol/L), or carrier (0.1% dimethyl sulfoxide). At the
indicated times, cells were fixed with 4% paraformaldehyde and
processed for terminal deoxynucleotidyl nick end-labeling (TUNEL)
staining with a commercially available kit (Roche) according to the
manufacturer’s recommendations. Afterward, cells were counterstained with 4⬘,6 diamidino-2-phenylindole (DAPI; Vector) and
counted under a fluorescence microscope. Two hundred cells per
time point and condition were counted, and the number of TUNELpositive cells was assessed.
Cell Culture
Real-Time PCR Analysis
RNA was extracted and converted to cDNA as described.22 Realtime polymerase chain reaction (PCR) was performed in an
MX3000P PCR cycler (Stratagene). All PCR experiments were
performed with the SYBR Green JumpStart kit (Sigma). Each
reaction (25 ␮L) contained 2 ␮L cDNA, 1 pmol of each primer, 0.25
␮L of internal reference dye, and 12.5 ␮L of JumpStart Taq
ReadyMix (containing buffer, dNTPs, stabilizers, SYBR Green, Taq
polymerase, and JumpStart Taq antibody). Primers for human TF
were used as described.22,24 Expression of the ribosomal protein L28
(L28) mRNA was used as a loading control; primers for human L28
were designed as follows: sense primer, 5⬘-GCATCTGCAATGGATGGT-3⬘ and antisense primer, 5⬘-TGTTCTTGCGGATCATGTGT-3⬘. The amplification program consisted of 1 cycle
at 95°C for 10 minutes; followed by 40 cycles with a denaturing
phase at 95°C for 30 seconds, an annealing phase at 60°C for 1
Statistics
Data are presented as mean⫾SEM. Unpaired Student t test was used for
statistical analysis. A probability value ⬍0.05 was considered
significant.
Results
Rapamycin Enhances TF Expression in HAECs
but Not HAVSMCs
Stimulation of HAECs with thrombin (1 U/mL) induced TF
expression 23-fold as assessed by Western blotting analysis
(Figure 1A). Incubation with rapamycin (10⫺8 to 10⫺7 mol/L)
before stimulation with thrombin resulted in a concentrationdependent enhancement of TF expression (Figure 1A). The
maximal increase was observed after 5 hours and was
2.3-fold compared with stimulation with thrombin alone and
51-fold compared with the basal level. Similarly, rapamycin
(10⫺8 to 10⫺7 mol/L) enhanced TF expression in response to
TNF-␣ (5 ng/mL); this increase was 2.5-fold, resulting in a
2004
Circulation
September 27, 2005
Figure 1. Rapamycin enhances TF
expression in HAECs. A, Rapamycin
enhances thrombin-induced TF expression in a concentration-dependent manner. Values are given as a percentage of
stimulation with thrombin alone.
*P⬍0.0001, ** P⬍0.001 vs thrombin
alone. B, Rapamycin enhances TNF-␣–
induced TF expression in a
concentration-dependent manner. Values
are given as a percentage of stimulation
with TNF-␣ alone. *P⬍0.02, ** P⬍0.01 vs
TNF-␣ alone. C, Rapamycin increases
basal TF expression. Values are given as
a percentage of unstimulated control.
*P⬍0.01 vs unstimulated control. Values
are representative of at least 3 different
experiments; all blots were normalized to
aT expression. D, TF ELISA confirms that
rapamycin increases both basal (*P⬍0.05
vs unstimulated control) and thrombininduced (**P⬍0.01 vs thrombin alone) TF
expression. E, Rapamycin enhances
thrombin-induced TF surface activity in a
concentration-dependent manner. Values
are given as a percentage of stimulation
with thrombin alone. *P⬍0.01 vs thrombin alone.
35-fold induction compared with the basal level (Figure 1B).
ECs express TF only at very low levels under basal conditions,26
and stimulation of HAECs with rapamycin alone increased basal
TF expression by 40%, as assessed by Western blotting analysis
(Figure 1C), or 25% as assessed by ELISA (Figure 1D).
Expression of TF was 45⫾1 pg per 500 000 cells for control,
57⫾3 pg per 500⬘000 cells for rapamycin (10⫺7 mol/L) alone,
468⫾29 pg per 500 000 cells for thrombin stimulation, and
702⫾30 pg/500 000 cells for thrombin stimulation in the presence of rapamycin (Figure 1D). The rapamycin-enhanced increase in thrombin-induced TF expression was paralleled by an
increase of TF surface activity, which reached 1.4 times the level
induced by thrombin alone (Figure 1E).
Similar to HAECs, thrombin (1 U/mL) and TNF-␣ (5
ng/mL) induced TF expression in HAVSMCs. In contrast to
HAECs, however, rapamycin did not affect TF expression in
response to either mediator in this cell type (Figure 2A and
2B). No cytotoxic effect of rapamycin was observed for any
of the concentrations used (n⫽4, P⫽NS; data not shown).
FK-506 Does Not Affect TF Expression
Incubation with FK-506 (10⫺8 to 10⫺7 mol/L) before stimulation with thrombin (1 U/mL, Figure 3A) or TNF-␣ (5
ng/mL, Figure 3B) did not alter TF expression. No cytotoxic
effect of FK-506 was observed for any of the concentrations
used (n⫽4, P⫽NS; data not shown).
Figure 2. Rapamycin does not affect TF
expression in HAVSMCs. Rapamycin
does not affect thrombin- (A) or TNF-␣–
(B) induced TF expression in HAVSMCs.
Values are given as a percentage of
stimulation with thrombin (1 U/mL) or
TNF-␣ (5 ng/mL) alone. Blots are representative of at least 4 different experiments; all blots were normalized to aT
expression.
Steffel et al
2005
Figure 3. FK-506 does not affect TF
expression in HAECs. FK-506 does not
affect thrombin- (A) or TNF-␣– (B)
induced TF expression in HAECs. Values
are presented as a percentage of stimulation with thrombin (1 U/mL) or TNF-␣ (5
ng/mL) alone. Blots are representative of
at least 3 different experiments; all blots
were normalized to aT expression.
Rapamycin Enhances TF Expression by Inhibiting
mTOR Activity
Phosphorylation of S6K, a downstream target of the mTOR,
is frequently used to assess mTOR inhibition by rapamycin.27,28 When stimulated with thrombin (1 U/mL), S6K
phosphorylation was transiently decreased after 30 minutes to
a minimum of 19% of the basal level (Figure 4A, left).
Rapamycin (10⫺7 mol/L) completely abrogated S6K phosphorylation, in both the presence and absence of thrombin
(Figure 4A, right). Similarly, inhibition of phosphatidyl
inositol 3-kinase with LY294002 or wortmannin almost
completely abrogated S6K phosphorylation, again independent of thrombin stimulation (Figure 4C). In contrast, FK-506
(10⫺7 mol/L) did not affect phosphorylation of S6K in either
the presence or absence of thrombin (Figure 4B).
Rapamycin Enhances TF Expression by Binding
to FKBP-12
Rapamycin and FK-506 bind to the same intracellular receptor, FK binding protein-12 (FKBP-12). When HAECs were
treated with increasing concentrations of FK-506 for 30
minutes before incubation with rapamycin, FK-506 reduced
the effect of rapamycin on thrombin-induced TF expression
(Figure 5). Indeed, when incubated with the highest concentration of FK-506 (10⫺7 mol/L), the increase in TF expression
elicited by rapamycin with respect to stimulation with thrombin alone was reduced by 41% (P⬍0.05).
Effect of Rapamycin on Thrombin-Induced TF
mRNA Levels
Real-time PCR revealed that thrombin induced TF mRNA
expression in a time-dependent manner (Figure 6A). Rapamycin did not alter thrombin-induced mRNA expression
compared with stimulation by thrombin alone after 0.5, 1, and
2 hours. However, after 3 and 5 hours of stimulation,
rapamycin significantly augmented thrombin-induced TF
mRNA levels (Figure 6A and 6B). Rapamycin significantly
increased thrombin-induced TF protein expression after 3, 5,
and 7 hours compared with stimulation by thrombin alone
(Figure 6C).
Rapamycin did not affect the pattern of MAP kinase
activation observed after thrombin stimulation. Indeed, phosphorylation of p38 (Figure 7A), ERK (Figure 7B), and JNK
(Figure 7C) remained unaltered after pretreatment with rapamycin compared with stimulation by thrombin alone.
Rapamycin, but Not FK-506, Inhibits
EC Proliferation
EC proliferation was induced by incubation with EBM
containing 10% FCS (Figure 8A, control). Rapamycin (10⫺7
mol/L) prevented FCS-induced EC proliferation. In contrast,
FK-506 (10⫺7 mol/L) did not significantly inhibit EC proliferation (Figure 8A).
TUNEL staining was used to examine whether rapamycin
(10⫺7 mol/L) or FK-506 (10⫺7 mol/L) induced apoptosis in
HAECs (Figure 8B). Representative sections are shown.
After 24 hours, TUNEL-positive cells accounted for
5.3⫾0.7% of cells in the control group, 4.7⫾1.7% for
rapamycin (P⫽NS versus control), and 4.8⫾1.5% for FK506 (P⫽NS versus control). After 48 hours, 4.2⫾1.7% of
control cells, 3.5⫾0.6% of rapamycin-treated cells (P⫽NS
versus control), and 4.6⫾2.3% of FK-506-treated cells
(P⫽NS versus control) were TUNEL-positive. Cells incubated with H2O2 (1 mmol/L) for 6 hours as well as serum
withdrawal for 48 hours served as positive controls and
resulted in a significant increase in apoptotic cells (data not
shown). Thus, neither rapamycin (10⫺7 mol/L) nor FK-506
(10⫺7 mol/L) led to an increase in apoptotic cells compared
with control conditions.
Discussion
This study demonstrates that rapamycin enhances endothelial
TF expression in response to thrombin and TNF-␣. The
concentrations of rapamycin occurring in vivo compare well
with those used in our study, as maximal systemic concentrations of rapamycin after deployment of 2 sirolimus-eluting
2006
Circulation
September 27, 2005
Figure 4. Rapamycin inhibits endothelial
mTOR activity. A, Thrombin leads to a
transient, time-dependent inhibition of
S6K phosphorylation (left). Rapamycin
(right) completely abrogates S6K phosphorylation in both the presence and
absence of thrombin. Total levels of S6K
remain unchanged. Values are presented
as phosphorylated (Pho) S6K/total (Tot)
S6K. *P⬍0.0001 vs unstimulated conditions. B, FK-506 affects neither basal
phosphorylation levels nor thrombininduced inhibition of S6K phosphorylation. Values are presented as phosphorylated S6K (Pho)/total (Tot) S6K. C,
LY294002 (5⫻10⫺6 mol/L, left) and wortmannin (10⫺7 mol/L, right) almost completely abrogate S6K phosphorylation
(Pho). Total (Tot) levels of S6K remain
unchanged.
stents are reported to be ⬇1 ng/mL (⬇1.15⫻10⫺9 mol/L)29;
moreover, local concentrations, though difficult to assess, are
likely to be significantly higher, partly because of rapamycin’s lipophilic properties, leading to accumulation of the
drug in the vessel wall.11,30 –32 Thus, the concentrations used
in our study may be relevant for patients treated with DESs.
Reendothelialization is initiated soon after vascular injury;
indeed, it has been observed to begin as early as 2 days after
balloon dilation in animal models.33–35 In humans, partial
reendothelialization has been documented 3 weeks after stent
deployment.35–37 Sirolimus-eluting stents are designed in
such a way that ⬇80% of the rapamycin has eluted by 30
Steffel et al
Figure 5. FK-506 antagonizes rapamycin-induced TF expression. Preincubation with FK-506 reduces rapamycin-enhanced
TF expression. Values are presented as a percentage of stimulation with thrombin (1 U/mL) and rapamycin (10⫺7 mol/L).
*P⬍0.05 and **P⬍0.02, compared with thrombin and rapamycin
(10⫺7 mol/L). Blots are representative of at least 3 different
experiments; all blots were normalized to aT expression.
days.4,5 Furthermore, rapamycin easily penetrates cell walls
owing to its lipophilic properties, leading to chronic retention
of the drug in arterial tissue.30 –32 Thus, the time course of
reendothelialization versus the kinetics of rapamycin release
suggests that rapamycin-enhanced endothelial TF expression
may be involved in the pathogenesis of in-stent thrombosis.
In addition, inhibition of endothelial proliferation by rapamycin indicates that rapamycin delays reendothelialization,
which may increase stent thrombogenicity even further.
Several hundred cases of acute and subacute in-stent
thrombosis have been observed after deployment of
rapamycin-eluting stents.13 In addition and in contrast to
BMSs, late thrombosis has been reported after withdrawal of
antithrombotic drugs with DESs.10 Most of these data originated from case reports or were collected in controlled
clinical trials. Recent results from a large-scale, multicenter
registry, however, indicate that in-stent thrombosis is likely
underestimated under these circumstances and that it may
occur at substantially higher rates in real world patients.14 The
pathogenesis of in-stent thrombosis has not yet been fully
explored15; moreover, it is not known whether the pathogenic
events leading to thromboses of DESs are similar to those of
BMSs. Enhanced TF expression in the presence of rapamycin
may indeed favor the development of in-stent thrombosis
after deployment of sirolimus-eluting stents, particularly
when clopidogrel is withdrawn or ineffective because of drug
resistance.38 FK-506, which neither affects endothelial TF
expression nor inhibits EC proliferation, may provide a more
favorable environment for reducing thromboses of DESs. To
assess the implications of these findings in vivo, however,
further studies are needed to examine the degree as well as the
spatiotemporal pattern of TF expression in the arterial wall
after deployment of DESs.
Platelet activation is a crucial event in the pathogenesis of
thrombus formation. Consequently, the use of platelet recep-
2007
tor blockers such as clopidogrel have greatly reduced the
incidence of stent thromboses, whereas withdrawal of antiplatelet therapy favors thrombus formation.10,14 Moreover,
clopidogrel inhibits the release of TF from aggregating
platelets,39 which is of particular interest, as platelet aggregation and secretion are increased in human platelets treated
with rapamycin.40 Thus, effective antiplatelet therapy may
account for the fact that thrombosis rates of sirolimus-eluting
stents are not clearly higher than those of BMSs.
TF induction after deployment of rapamycin-eluting stents
may also have a prothrombotic effect on ECs distal to the
stented site. Indeed, remote effects of rapamycin have been
demonstrated, with pronounced endothelial dysfunction in
coronary arteries distal to sirolimus-eluting stents compared
with BMSs.41 Thus, in addition to the effect on ECs within
the stented region, rapamycin may also increase TF expression in ECs in the distal coronary vasculature. Such an effect
may also contribute to the no-reflow phenomenon after stent
deployment.
Rapamycin did not enhance thrombin- or TNF-␣– driven
TF expression in HAVSMCs, indicating that rapamycin does
not constitute an additional thrombogenic signal to the
VSMC layer. Indeed, a much higher incidence of acute and
subacute stent thromboses would be expected if rapamycin
induced TF expression in VSMCs. Although the rapamycininduced increase in endothelial TF expression may favor
neointima formation via the release of growth factors from
aggregating platelets, the inhibitory effect of rapamycin on
the proliferation and migration of VSMC is very likely to
protect the vessel from such effects.7 Consistent with this
interpretation, sirolimus-eluting stents reduce neointima formation despite inducing a procoagulative state owing to
enhanced endothelial TF expression.
Both thrombin, a coagulation factor, and TNF-␣, an
inflammatory cytokine, are classic inducers of TF expression
in vascular cells. Thrombin induced TF expression 27-fold
when examined by Western blotting analysis and 10.3-fold by
ELISA; similarly, rapamycin enhanced thrombin-induced TF
expression by 2.3-fold in Western blot analysis and 1.5-fold
by ELISA. This difference may be due to a different sensitivity and/or specificity of the 2 assays. In our study,
rapamycin enhanced TF expression in response to both
thrombin and TNF-␣; it may thus upregulate TF expression in
a prothrombotic as well as an inflammatory environment,
both of which are encountered in the coronary vasculature
after stent deployment.
Biologically active TF is located at the cell surface, and
rapamycin-enhanced TF protein expression was indeed paralleled by an increase in TF surface activity. The increase in
activity was not as pronounced as that of protein expression;
this discrepancy has also been observed in response to
thrombin alone.42 The distribution of TF in several cellular
compartments and/or the expression of encrypted TF might
account for this difference.43
The inhibitory role of phosphatidyl inositol 3-kinase on TF
expression is established, as its inhibition enhances TF
expression in response to thrombin.42,44 The mTOR is a
downstream target of phosphatidyl inositol 3-kinase.28 Binding of rapamycin to its intracellular receptor FKBP-12 leads
2008
Circulation
September 27, 2005
Figure 6. Effect of rapamycin on TF mRNA induction. A, Real-time PCR demonstrates a timedependent induction of TF mRNA in response to
thrombin. Rapamycin does not alter this pattern of
induction after stimulation for 0.5, 1, and 2 hours.
Values are given as a percentage of stimulation
with thrombin alone for 2 hours. B, Analysis of
⌬⌬CT values comparing the effect of rapamycin on
thrombin-induced mRNA levels for every time point
reveals that rapamycin significantly increases
thrombin-induced TF mRNA after stimulation for 3
and 5 hours. *P⬍0.0005, **P⬍0.005. All values are
representative of 4 different experiments and were
normalized to L28 mRNA expression. C, Rapamycin enhances thrombin-induced TF protein expression in a time-dependent manner. Values are given
as a percentage of stimulation with thrombin alone
for 3 hours. *P⬍0.0001, **P⬍0.01 vs thrombin
alone. Values are representative of at least 3 different experiments. All blots were normalized to aT
expression.
to formation of the rapamycin–FKBP-12 complex, which in
turn inhibits mTOR activity. Phosphorylation of the downstream target of mTOR, S6K, is routinely used as a readout
for the inhibitory effect of rapamycin on mTOR27,28; indeed,
mTOR-dependent phosphorylation of the Thr-389 residue of
S6K is necessary for its activity.27 In the present study, we
have shown that stimulation with thrombin leads to a transient inhibition of S6K phosphorylation. Rapamycin as well
as the phosphatidyl inositol 3-kinase inhibitors wortmannin
and LY294002 abrogated S6K phosphorylation in both the
presence and absence of thrombin. Because thrombin stimulation as well as preincubation with rapamycin led to inhibition of this pathway, resulting in disinhibition of TF expression, these observations are consistent with the interpretation
that mTOR plays an inhibitory role in TF expression.
FK-506 competitively binds to the same intracellular receptor as rapamycin, ie, FKBP-12.45 In contrast to rapamycin,
however, the FK-506 –FKBP-12 complex inhibits the phosphatase calcineurin and has no effect on mTOR activity.46
Consistently, FK-506 did not alter thrombin- or TNF-␣–
induced TF expression. To assess the specificity of our
observations, we coincubated FK-506 and rapamycin before
thrombin stimulation. The enhancing effect of rapamycin on
thrombin-induced TF expression could indeed be reduced by
FK-506. These findings indicate that binding of rapamycin to
FKBP-12 is necessary for inhibition of mTOR activity and
enhancement of TF expression.
TF expression in response to a variety of stimuli is
mediated by MAP kinase activation, leading to increased
transcription.22,24,42 Indeed, thrombin induced an increase in
p38, ERK, and JNK phosphorylation as well as an increase in
TF transcription. However, rapamycin did not alter the
pattern of thrombin-induced p38, ERK, and JNK activation.
Steffel et al
2009
Figure 7. Rapamycin does not affect MAP kinase activation. Stimulation with thrombin leads to phosphorylation (Pho) of the MAP
kinases p38 (A), ERK (B), and JNK (C). Rapamycin does not alter this pattern of MAP kinase activation. Total (Tot) levels of p38, ERK,
and JNK remain unchanged. Blots are representative of at least 3 different experiments.
Consistent with this observation, thrombin-induced mRNA
levels were unchanged by rapamycin for up to 2 hours after
stimulation. However, after 3 and 5 hours of thrombin
stimulation, rapamycin increased mRNA levels compared
with stimulation by thrombin alone. Taken together, these
data imply that the enhancing effect of rapamycin on
thrombin-induced TF expression initially occurs at the posttranscriptional level and hence, is independent of MAP kinase
Figure 8. Rapamycin, but not FK-506, inhibits EC proliferation. A, Rapamycin completely inhibits EC proliferation induced by 10% FCS;
in contrast, FK-506 does not significantly affect EC proliferation. *P⬍0.05 vs control; **P⬍0.005 vs control; and **P⬍0.002 vs FK-506.
Three different experiments were performed in duplicate for each experimental condition. B, There was no increase in TUNEL-positive
cells after incubation with rapamycin or FK-506 for 24 and 48 hours. Slides show representative TUNEL-positive cells with the corresponding DAPI staining after 24 hours of incubation with carrier (left), rapamycin (middle), and FK-506 (right).
2010
Circulation
September 27, 2005
activation, although a transcriptional effect of rapamycin
cannot be ruled out at later time points. Indeed, mTOR is
known to exert posttranscriptional effects, and TF expression
can be regulated at both the transcriptional and posttranscriptional level.47,48
In summary, our study reveals that rapamycin, but not
FK-506, enhances endothelial TF expression and reduces
HAEC proliferation. These effects may favor the development of thrombus formation after deployment of sirolimuseluting stents, particularly when antithrombotic drugs are
withdrawn or ineffective, and may have interesting implications for the design of DESs.
Acknowledgments
This work was supported by the Swiss National Science Foundation
(grant No. 3200B0-102232/1 to Dr Tanner and grant No. 3100068118.02/1 to Dr Lüscher), the Bonizzi-Theler Foundation, the
Hartmann-Müller Foundation, the Herzog-Egli Foundation, and the
Olga Mayenfisch Foundation. The authors thank Dr F. Eberli and Dr
W. Maier for discussion.
References
1. Schomig A, Kastrati A, Dirschinger J, Mehilli J, Schricke U, Pache J,
Martinoff S, Neumann FJ, Schwaiger M. Coronary stenting plus platelet
glycoprotein IIb/IIIa blockade compared with tissue plasminogen activator in acute myocardial infarction: Stent versus Thrombolysis for
Occluded Coronary Arteries in Patients with Acute Myocardial Infarction
Study Investigators. N Engl J Med. 2000;343:385–391.
2. Grines CL, Cox DA, Stone GW, Garcia E, Mattos LA, Giambartolomei
A, Brodie BR, Madonna O, Eijgelshoven M, Lansky AJ, O’Neill WW,
Morice MC. Coronary angioplasty with or without stent implantation for
acute myocardial infarction: Stent Primary Angioplasty in Myocardial
Infarction Study Group. N Engl J Med. 1999;341:1949 –1956.
3. Lincoff AM, Califf RM, Moliterno DJ, Ellis SG, Ducas J, Kramer JH,
Kleiman NS, Cohen EA, Booth JE, Sapp SK, Cabot CF, Topol EJ. Complementary clinical benefits of coronary-artery stenting and blockade of
platelet glycoprotein IIb/IIIa receptors: Evaluation of Platelet IIb/IIIa
Inhibition in Stenting Investigators. N Engl J Med. 1999;341:319 –327.
4. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin M,
Colombo A, Schuler G, Barragan P, Guagliumi G, Molnar F, Falotico R.
A randomized comparison of a sirolimus-eluting stent with a standard
stent for coronary revascularization. N Engl J Med. 2002;346:1773–1780.
5. Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR,
O’Shaughnessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein PS,
Jaeger JL, Kuntz RE. Sirolimus-eluting stents versus standard stents in
patients with stenosis in a native coronary artery. N Engl J Med. 2003;
349:1315–1323.
6. Babapulle MN, Joseph L, Belisle P, Brophy JM, Eisenberg MJ. A hierarchical Bayesian meta-analysis of randomised clinical trials of drugeluting stents. Lancet. 2004;364:583–591.
7. Gallo R, Padurean A, Jayaraman T, Marx S, Roque M, Adelman S,
Chesebro J, Fallon J, Fuster V, Marks A, Badimon JJ. Inhibition of
intimal thickening after balloon angioplasty in porcine coronary arteries
by targeting regulators of the cell cycle. Circulation. 1999;99:
2164 –2170.
8. Grube E, Buellesfeld L. Rapamycin analogs for stent-based local drug
delivery: everolimus- and tacrolimus-eluting stents. Herz. 2004;29:
162–166.
9. Bartorelli AL, Trabattoni D, Fabbiocchi F, Montorsi P, de Martini S,
Calligaris G, Teruzzi G, Galli S, Ravagnani P. Synergy of passive coating
and targeted drug delivery: the tacrolimus-eluting Janus CarboStent.
J Interv Cardiol. 2003;16:499 –505.
10. McFadden EP, Stabile E, Regar E, Cheneau E, Ong AT, Kinnaird T,
Suddath WO, Weissman NJ, Torguson R, Kent KM, Pichard AD, Satler
LF, Waksman R, Serruys PW. Late thrombosis in drug-eluting coronary
stents after discontinuation of antiplatelet therapy. Lancet. 2004;364:
1519 –1521.
11. Kereiakes DJ, Choo JK, Young JJ, Broderick TM. Thrombosis and
drug-eluting stents: a critical appraisal. Rev Cardiovasc Med. 2004;
5:9 –15.
12. Eisenberg MJ. Drug-eluting stents: some bare facts. Lancet. 2004;364:
1466 –1467.
13. FDA advises physicians of adverse events associated with Cordis Cypher
coronary stents. US Food and Drug Administration Public Health Web
Notification. October 29, 2003:T03-T71.
14. Iakovou I, Schmidt T, Bonizzoni E, Ge L, Sangiorgi GM, Stankovic G,
Airoldi F, Chieffo A, Montorfano M, Carlino M, Michev I, Corvaja N,
Briguori C, Gerckens U, Grube E, Colombo A. Incidence, predictors, and
outcome of thrombosis after successful implantation of drug-eluting
stents. JAMA. 2005;293:2126 –2130.
15. Honda Y, Fitzgerald PJ. Stent thrombosis: an issue revisited in a changing
world. Circulation. 2003;108:2–5.
16. Moons AH, Levi M, Peters RJ. Tissue factor and coronary artery disease.
Cardiovasc Res. 2002;53:313–325.
17. Sato Y, Asada Y, Marutsuka K, Hatakeyama K, Sumiyoshi A. Tissue
factor induces migration of cultured aortic smooth muscle cells. Thromb
Haemost. 1996;75:389 –392.
18. Tutar E, Ozcan M, Kilickap M, Gulec S, Aras O, Pamir G, Oral D,
Dandelet L, Key NS. Elevated whole-blood tissue factor procoagulant
activity as a marker of restenosis after percutaneous transluminal
coronary angioplasty and stent implantation. Circulation. 2003;108:
1581–1584.
19. Marmur JD, Rossikhina M, Guha A, Fyfe B, Friedrich V, Mendlowitz M,
Nemerson Y, Taubman MB. Tissue factor is rapidly induced in arterial
smooth muscle after balloon injury. J Clin Invest. 1993;91:2253–2259.
20. Steffel J, Hermann M, Greutert H, Gay S, Luscher TF, Ruschitzka F,
Tanner FC. Celecoxib decreases endothelial tissue factor expression
through inhibition of c-Jun terminal NH2 kinase phosphorylation. Circulation. 2005;111:1685–1689.
21. Tanner FC, Greutert H, Barandier C, Frischknecht K, Luscher TF. Different cell cycle regulation of vascular smooth muscle in genetic hypertension. Hypertension. 2003;42:184 –188.
22. Steffel J, Akhmedov A, Greutert H, Lüscher TF, Tanner FC. Histamine
induces tissue factor: implications for acute coronary syndromes. Circulation. 2005;112:341–349.
23. Tanner FC, Yang ZY, Duckers E, Gordon D, Nabel GJ, Nabel EG.
Expression of cyclin-dependent kinase inhibitors in vascular disease. Circ
Res. 1998;82:396 – 403.
24. Bogdanov VY, Balasubramanian V, Hathcock J, Vele O, Lieb M,
Nemerson Y. Alternatively spliced human tissue factor: a circulating,
soluble, thrombogenic protein. Nat Med. 2003;9:458 – 462.
25. Cui MZ, Zhao G, Winokur AL, Laag E, Bydash JR, Penn MS, Chisolm
GM, Xu X. Lysophosphatidic acid induction of tissue factor expression in
aortic smooth muscle cells. Arterioscler Thromb Vasc Biol. 2003;23:
224 –230.
26. Bavendiek U, Libby P, Kilbride M, Reynolds R, Mackman N, Schonbeck
U. Induction of tissue factor expression in human endothelial cells by
CD40 ligand is mediated via activator protein 1, nuclear factor-␬B, and
Egr-1. J Biol Chem. 2002;277:25032–25039.
27. Burnett PE, Barrow RK, Cohen NA, Snyder SH, Sabatini DM. RAFT1
phosphorylation of the translational regulators p70 S6 kinase and 4E-BP1.
Proc Natl Acad Sci U S A. 1998;95:1432–1437.
28. Hay N, Sonenberg N. Upstream and downstream of mTOR. Genes Dev.
2004;18:1926 –1945.
29. US Food and Drug Administration, Center for Devices and Radiological
Health. Cypher Sirolimus-eluting coronary stent on RAPTOR overthe-wire delivery system or RAPTORRAIL rapid exchange delivery
system. Available at http://www.fda.gov/cdrh/pdf2/p020026.html.
Accessed December 17, 2004.
30. Gummert JF, Ikonen T, Morris RE. Newer immunosuppressive drugs: a
review. J Am Soc Nephrol. 1999;10:1366 –1380.
31. Klugherz BD, Llanos G, Lieuallen W, Kopia GA, Papandreou G, Narayan
P, Sasseen B, Adelman SJ, Falotico R, Wilensky RL. Twenty-eight-day
efficacy and pharmacokinetics of the sirolimus-eluting stent. Coron
Artery Dis. 2002;13:183–188.
32. Suzuki T, Kopia G, Hayashi S, Bailey LR, Llanos G, Wilensky R,
Klugherz BD, Papandreou G, Narayan P, Leon MB, Yeung AC, Tio F,
Tsao PS, Falotico R, Carter AJ. Stent-based delivery of sirolimus reduces
neointimal formation in a porcine coronary model. Circulation. 2001;
104:1188 –1193.
33. Asahara T, Chen D, Tsurumi Y, Kearney M, Rossow S, Passeri J, Symes
JF, Isner JM. Accelerated restitution of endothelial integrity and endothelium-dependent function after phVEGF165 gene transfer. Circulation.
1996;94:3291–3302.
Steffel et al
34. Tanaka H, Sukhova GK, Swanson SJ, Clinton SK, Ganz P, Cybulsky MI,
Libby P. Sustained activation of vascular cells and leukocytes in the
rabbit aorta after balloon injury. Circulation. 1993;88:1788 –1803.
35. Schwartz RS, Chronos NA, Virmani R. Preclinical restenosis models and
drug-eluting stents: still important, still much to learn. J Am Coll Cardiol.
2004;44:1373–1385.
36. Farb A, Sangiorgi G, Carter AJ, Walley VM, Edwards WD, Schwartz RS,
Virmani R. Pathology of acute and chronic coronary stenting in humans.
Circulation. 1999;99:44 –52.
37. Grewe PH, Deneke T, Machraoui A, Barmeyer J, Muller KM. Acute and
chronic tissue response to coronary stent implantation: pathologic
findings in human specimen. J Am Coll Cardiol. 2000;35:157–163.
38. Matetzky S, Shenkman B, Guetta V, Shechter M, Bienart R, Goldenberg
I, Novikov I, Pres H, Savion N, Varon D, Hod H. Clopidogrel resistance
is associated with increased risk of recurrent atherothrombotic events in
patients with acute myocardial infarction. Circulation. 2004;109:
3171–3175.
39. Leon C, Alex M, Klocke A, Morgenstern E, Moosbauer C, Eckly A,
Spannagl M, Gachet C, Engelmann B. Platelet ADP receptors contribute
to the initiation of intravascular coagulation. Blood. 2004;103:594 – 600.
40. Babinska A, Markell MS, Salifu MO, Akoad M, Ehrlich YH, Kornecki E.
Enhancement of human platelet aggregation and secretion induced by
rapamycin. Nephrol Dial Transplant. 1998;13:3153–3159.
41. Hofma SH, Van der Giessen WJ, Van Dalen B, Lemos PA, Serruys PW
Prolonged endothelial dysfunction late after stenting with sirolimuseluting stent compared to bare metal stent. Eur Heart J. 2004;
25(suppl):525. Abstract.
2011
42. Eto M, Kozai T, Cosentino F, Joch H, Luscher TF. Statin prevents tissue
factor expression in human endothelial cells: role of Rho/Rho-kinase and
Akt pathways. Circulation. 2002;105:1756 –1759.
43. Schecter AD, Spirn B, Rossikhina M, Giesen PL, Bogdanov V, Fallon JT,
Fisher EA, Schnapp LM, Nemerson Y, Taubman MB. Release of active
tissue factor by human arterial smooth muscle cells. Circ Res. 2000;87:
126 –132.
44. Viswambharan H, Ming XF, Zhu S, Hubsch A, Lerch P, Vergeres G,
Rusconi S, Yang Z. Reconstituted high-density lipoprotein inhibits
thrombin-induced endothelial tissue factor expression through inhibition
of RhoA and stimulation of phosphatidylinositol 3-kinase but not Akt/
endothelial nitric oxide synthase. Circ Res. 2004;94:918 –925.
45. Bierer BE, Mattila PS, Standaert RF, Herzenberg LA, Burakoff SJ,
Crabtree G, Schreiber SL. Two distinct signal transmission pathways in T
lymphocytes are inhibited by complexes formed between an immunophilin and either FK506 or rapamycin. Proc Natl Acad Sci U S A.
1990;87:9231–9235.
46. Liu J, Farmer JD Jr, Lane WS, Friedman J, Weissman I, Schreiber SL.
Calcineurin is a common target of cyclophilin-cyclosporin A and
FKBP-FK506 complexes. Cell. 1991;66:807– 815.
47. Deguchi H, Takeya H, Wada H, Gabazza EC, Hayashi N, Urano H,
Suzuki K. Dilazep, an antiplatelet agent, inhibits tissue factor expression
in endothelial cells and monocytes. Blood. 1997;90:2345–2356.
48. Lopez-Pedrera C, Jardi M, Ingles-Esteve J, Munoz-Canoves P, Dorado G,
Velasco F, Felez J. Characterization of tissue factor expression on the
human endothelial cell line ECV304. Am J Hematol. 1997;56:71–78.
Pericardial Disease
Colchicine in Addition to Conventional Therapy
for Acute Pericarditis
Results of the COlchicine for acute PEricarditis (COPE) Trial
Massimo Imazio, MD; Marco Bobbio, MD; Enrico Cecchi, MD; Daniela Demarie, MD;
Brunella Demichelis, MD; Franco Pomari, MD; Mauro Moratti, MD; Gianni Gaschino, MD;
Massimo Giammaria, MD; Aldo Ghisio, MD; Riccardo Belli, MD; Rita Trinchero, MD
Background—Colchicine is effective and safe for the treatment and prevention of recurrent pericarditis and might ultimately
serve as the initial mode of treatment, especially in idiopathic cases. The aim of this work was to verify the safety and efficacy
of colchicine as an adjunct to conventional therapy for the treatment of the first episode of acute pericarditis.
Methods and Results—A prospective, randomized, open-label design was used. A total of 120 patients (mean age
56.9⫾18.8 years, 54 males) with a first episode of acute pericarditis (idiopathic, viral, postpericardiotomy syndromes,
and connective tissue diseases) were randomly assigned to conventional treatment with aspirin (group I) or conventional
treatment plus colchicine 1.0 to 2.0 mg for the first day and then 0.5 to 1.0 mg/d for 3 months (group II). Corticosteroid
therapy was restricted to patients with aspirin contraindications or intolerance. The primary end point was recurrence
rate. During the 2873 patient-month follow-up, colchicine significantly reduced the recurrence rate (recurrence rates at
18 months were, respectively, 10.7% versus 32.3%; P⫽0.004; number needed to treat⫽5) and symptom persistence at
72 hours (respectively, 11.7% versus 36.7%; P⫽0.003). After multivariate analysis, corticosteroid use (OR 4.30, 95%
CI 1.21 to 15.25; P⫽0.024) was an independent risk factor for recurrences. Colchicine was discontinued in 5 cases
(8.3%) because of diarrhea. No serious adverse effects were observed.
Conclusions—Colchicine plus conventional therapy led to a clinically important and statistically significant benefit over
conventional treatment, decreasing the recurrence rate in patients with a first episode of acute pericarditis. Corticosteroid
therapy given in the index attack can favor the occurrence of recurrences. (Circulation. 2005;112:2012-2016.)
Key Words: colchicine 䡲 pericarditis 䡲 survival 䡲 recurrence 䡲 prevention
C
olchicine has been used for hundreds of years as an
antiinflammatory agent for acute arthritis and is the most
specific known treatment for acute attacks of gout.1–3 More
recently, the drug has been used successfully for the prophylaxis of familial Mediterranean fever attacks4,5 and the treatment of recurrent pericarditis.6 –13
has not been evaluated extensively in clinical trials, and randomized trials are lacking to guide the evaluation and management of
acute pericarditis.20,22 A preliminary small French study without
a control group27 tested the use of colchicine in 19 patients with
a first episode of acute pericarditis. After a mean follow-up of 5
months, a recurrence rate of 10.5% was found. To the best of our
knowledge, no prospective, randomized studies have been published to test this intriguing hypothesis.
The aim of the present study was to verify the safety and
efficacy of colchicine as an adjunct to conventional therapy for
treatment of the first episode of acute pericarditis and for
prevention of recurrences.
See p 1921
Recurrent pericarditis is the most troublesome complication of the disease, occurring in from 15% to 50% of
cases.6,14 –22 It is generally accepted that recurrence is an
autoimmune process.6,16,22–24 The optimal management for
preventing recurrences has not been established.6,16,22,24 Colchicine appears to be effective and safe for the treatment and
prevention of recurrent pericarditis6 –13; moreover, it might be a
promising adjunct to the conventional treatment of recurrent
pericarditis and might ultimately serve as the initial mode of
treatment, especially in idiopathic cases.16,24 –26 This indication
Methods
Study Design
A prospective, randomized, open-label, parallel-group study was conducted in 2 Italian centers. Validation of clinical events was ensured by
an ad-hoc committee of expert cardiologists blinded to patients’ treat-
Received February 15, 2005; revision received May 15, 2005; accepted June 10, 2005.
From the Cardiology Department, Maria Vittoria Hospital and Amedeo di Savoia Hospital (M.I., E.C., D.D., B.D., F.P., M.M., G.G., M.G., A.G., R.B.,
R.T.), and Cardiology Medical School (M.B.), University of Turin, Turin, Italy.
Correspondence to Massimo Imazio, MD, Cardiology Department, Maria Vittoria Hospital, Via Cibrario 72, 10141 Torino, Italy. E-mail
[email protected]
2012
Imazio et al
ment assignment. The study was conceived and managed by the
Coordinating Center. Data analyses were performed by an external data
analysis committee, which was blinded to treatment assignment. We
obtained approval of the study protocol by the institutional review
board, and subjects gave informed consent.
Subjects
All consecutive patients with a first episode of acute pericarditis were
enrolled. Eligible patients had no contraindication to colchicine, were
able to provide informed consent, and had no unfavorable short-term
outlook. Inclusion criteria were definite diagnosis of acute pericarditis
(idiopathic, viral, and autoimmune causes, including postpericardiotomy
syndromes and connective tissue diseases), age ⱖ18 years, and provision of informed consent. Exclusion criteria were tuberculous, neoplastic, or purulent causes; known severe liver disease or current transaminases ⬎1.5 times the upper normal limit; current serum creatinine above
2.5 mg/dL; known myopathy or current serum creatine kinase above the
upper normal limit; known blood dyscrasias or gastrointestinal disease;
pregnant and lactating women or women of childbearing potential not
protected by a contraception method; known hypersensitivity to colchicine; and current treatment with colchicine for any indication. Acute
pericarditis was diagnosed when at least 2 of the following criteria were
present: typical chest pain, pericardial friction rub, and widespread
ST-segment elevation on the ECG.14,17,18,25,28
Randomization and Treatment Plan
Patients were randomized to receive a conventional treatment with aspirin
800 mg orally every 6 or 8 hours for 7 to 10 days with gradual tapering over
3 to 4 weeks (group I) or a treatment with aspirin at the same dose combined
with colchicine 1.0 to 2.0 mg for the first day and then a maintenance dose
of 0.5 to 1.0 mg daily for 3 months (group II). The lower dose (initial dose
1.0 mg and maintenance dose 0.5 mg daily) was given to patients who
weighed ⬍70 kg or who were intolerant to the highest dose (initial dose 1.0
mg BID and maintenance dose of 0.5 mg BID). Randomization was based
on permuted blocks, with a block size of 4.
As the preferred nonsteroidal antiinflammatory drug, we used
aspirin according to our previously published experience.17,18 For the
choice for colchicine dose, we considered previous experiences in
the treatment of recurrent pericarditis,2,6 –13 as well as our own
previous experience.13,24 It was decided to assign patients to the
lowest effective dose, thus reducing side effects and improving drug
tolerability. Corticosteroid therapy (prednisone at a dose of 1.0 to 1.5
mg · kg⫺1 · d⫺1 for 2 to 4 weeks with a gradual tapering off) was
restricted to patients with aspirin contraindications (oral anticoagulant therapy, allergy, or history of peptic ulcer or gastrointestinal
bleeding) or intolerance. In every case, a gastroduodenal prophylaxis
was adopted with omeprazole 20 mg/d, also without initial evidence
of gastrointestinal intolerance, as previously published.17,18
We planned the trial procedures to mimic our previous routine care of
acute pericarditis.18 All patients had M-mode, 2D, and Doppler echocardiographic studies performed with a Hewlett-Packard SONOS 2500
or 5500 machine. A clinical and echocardiographic follow-up was
performed at 48 to 72 hours, 10 days, 1 month, 3 months, 6 months, and
1 year and then yearly in uncomplicated cases.
Colchicine for Acute Pericarditis
2013
Patients were considered to have remission when they were free of
symptoms, with disappearance of clinical, ECG, and echocardiographic
signs.10 Conversely, we considered treatment failure to be an unfavorable clinical reaction with persistence of fever, pericardial effusion
appearance or worsening, and general illness lasting more than 7 days or
if the patient showed an incessant or recurrent course.
Safety
During follow-up, monitoring and recording of all adverse events were
performed. A severe adverse event was considered an untoward event
that was fatal, life-threatening, or required hospitalization or that was
significantly or permanently disabling or medically significant (may
jeopardize the patient and may require medical or surgical intervention
to prevent an adverse outcome). A Safety Monitoring Committee
performed 1 interim analysis, blinded to treatment assignment.
A total of 120 patients, 60 in each treatment arm, were needed to detect
a difference in recurrence rates of 32.5% and 10.5% between these 2
treatment arms with a power of 80% using a 2-sided P⫽0.05 level test.
The estimated recurrence rate of 32.5% in the control group was based
on previous studies (recurrence rate from 15% to 50%).6,13,16,18,21,28 The
estimated recurrence rate of 10.5% in the colchicine group was based on
the results of a small preliminary French study (reported recurrence rate
10.5%).27 Analysis was performed by intention to treat.
Data are expressed as mean⫾SD. Comparisons between patient
groups were performed with unpaired t test for continuous variables and
a ␹2 analysis for categorical variables. A probability value ⬍0.05 was
considered to show statistical significance. Time-to-event distributions
were estimated by the Kaplan-Meier method and compared with the
log-rank test. To evaluate possible risk factors for recurrence, a logistic
regression multivariate analysis was performed. All analyses were
performed with the software package SPSS 13.0. The number of
patients needed to treat was estimated with its CI using GraphPad
Software QuickCalcs.
Results
Between January 2002 and August 2004, 120 patients were
randomized. Information on vital status and clinical follow-up
data were available in all patients for a mean follow-up of 24
months (range 8 to 39 months). Sixty patients (mean age
57.2⫾19.6; 26 males) were randomly assigned to aspirin alone
(group I), and 60 patients (mean age 56.5⫾18.2; 28 males) were
assigned to aspirin and colchicine (group II). A detailed trial
profile is reported in Figure 1. Baseline demographic and clinical
characteristics were well balanced across the groups (Table 1).
Corticosteroid therapy was prescribed in 19 patients (15.8%)
because of aspirin contraindication or intolerance, according to
the study protocol. The overall efficacy profile of the 2 treatments is summarized in Table 2. All 60 patients treated by
colchicine responded favorably to therapy.
End Points
Primary End Point
The primary end point was recurrence rate. Criteria for the diagnosis of
recurrence were (1) documented first attack of acute pericarditis according to definite diagnostic criteria and (2) evidence of either recurrence or
continued activity of pericarditis. Recurrence was documented by
recurrent pain and 1 or more of the following signs: fever, pericardial
friction rub, ECG changes, echocardiographic evidence of pericardial
effusion, and elevations in the white blood cell count, erythrocyte
sedimentation rate, or C-reactive protein.6,22,28,29
We included as recurrent pericarditis both the incessant type (patients
in whom discontinuation or attempts to wean from treatment ensured a
recurrence in a period of less than 6 weeks) and the intermittent type
(patients with a symptom-free interval longer than 6 weeks).22,29 The
secondary end point was the rate of symptom persistence at 72 hours
from treatment onset.
During the 2873 patient-months of follow-up, a higher recurrence rate was recorded in patients treated only by aspirin (group
I) than in patients treated with colchicine plus conventional
treatment (group II; respectively, 33.3% versus 11.7%;
P⫽0.009). Nearly all of the recurrences occurred within 18
months. Recurrence rates in group I and group II at 18 months
were 32.3% and 10.7%, respectively (P⫽0.004; number of
patients needed to treat⫽5 [95% CI 3.1 to 10.0]). Patients in
group II had a longer symptom-free interval (22.9⫾10.3 versus
17.2⫾12.3 months; P⫽0.007). Event-free survival in the study
groups is reported in Figure 2. An exploratory analysis was done
by subgroups according to treatment (aspirin, aspirin plus
2014
Circulation
September 27, 2005
Figure 1. Trial profile.
colchicine, prednisone, and prednisone plus colchicine). Recurrence rates at 18 months were 23.5% in the aspirin subgroup,
8.8% in the aspirin plus colchicine subgroup, 86.7% in the
prednisone subgroup, and 11.1% in the prednisone plus colchicine subgroup (log-rank P⬍0.001).
for 5 patients (8.3%) in the colchicine group (group II) and 0
patients in group I. Minor side effects (including abdominal pain
and dyspepsia) were recorded in 4 (6.7%) of 60 cases in group
I without need for drug withdrawal.
Discussion
Secondary End Point and Risk Factors for Recurrences
A lower incidence of symptom persistence at 72 hours was
recorded in group II than in group I (respectively, 11.7% versus
36.7%; P⫽0.003). Baseline clinical features of patients with and
without recurrences during follow-up are reported in Table 3.
Patients with recurrences during follow-up had a higher rate of
corticosteroid use in the index attack (33.3% versus 10.7%;
P⫽0.011). After logistic regression multivariate analysis that
introduced age, gender, pericardial effusion, severe pericardial
effusion, cardiac tamponade, etiology, corticosteroid use, and
colchicine therapy as independent variables, corticosteroid use
remained an independent risk factor for the subsequent development of recurrences (OR 4.30, 95% CI 1.21 to 15.25;
P⫽0.024), whereas the use of colchicine was found to be
protective (OR 0.17, 95% CI 0.05 to 0.53; P⫽0.003).
Safety
Safety profiles of the studied treatments are summarized in
Table 2. Overall drug tolerability was good for aspirin and
colchicine; no serious adverse drug effects were recorded in the
study groups. Colchicine-treated patients had 5 cases of diarrhea
(8.3%), which was promptly reversible after drug withdrawal.
Side effects were reported as a reason for discontinuing therapy
TABLE 1. Baseline Clinical Characteristics of
Randomized Patients
Major Findings
The COPE study provides evidence that colchicine in combination
therapy with aspirin or prednisone is safe and efficacious in the
treatment of the first episode of acute pericarditis, as well as in the
prevention of recurrences. Previous reports6–13,22 have shown that
colchicine is effective and safe as an adjunct for the treatment of
recurrent pericarditis and the prevention of further recurrences after
conventional treatment failure. In these studies, patients treated with
colchicine after previous recurrences showed a reduced recurrence
rate: from 0% to 26%, with a mean value of 14%.16 A small French
study27 in 19 patients with acute pericarditis suggested that colchicine may also be effective in the treatment of the first episode of
acute pericarditis; however, this hypothesis was tested in only 19
patients without a control group. After a mean follow-up of 5
months, a recurrence rate of 10.5% was found, whereas the
recurrence rate may be as high as 15% to 50%16,21 with conventional treatment.
On the basis of cumulative anecdotal evidence and the opinion of
experts, colchicine (0.5 to 0.6 mg BID) is suggested as a possible
therapy for the first episode of acute pericarditis,16,25,26 whereas
nonsteroidal antiinflammatory drugs are the mainstay of treatment.
The threshold of prescription of the drug has been lowered, because
at low doses, the drug is well tolerated, with few side effects;
TABLE 2.
Follow-Up Data of Randomized Patients
Feature
Group I:
No Colchicine
(n⫽60)
Group II:
Colchicine
(n⫽60)
Group I:
No Colchicine
(n⫽60)
P
Feature
Age, y
57.2⫾19.6
56.5⫾18.2
NS
Mean follow-up, mo
Male gender
26 (43.3)
28 (46.7)
NS
Corticosteroid use,* n (%)
10 (16.6)
9 (15.0)
NS
Pericarditic chest pain
60 (100.0)
60 (100.0)
NS
Recurrence, n (%)
20 (33.3)
7 (11.7)
0.009
Pericardial rub
19 (31.7)
21 (35.0)
NS
Recurrence rate at 18 mo, %
ST-segment elevation
53 (88.3)
52 (86.7)
NS
Symptom persistence at 72 h, n (%)
Pericardial effusion
38 (63.3)
41 (68.3)
NS
Cardiac tamponade
1 (1.6)
1 (1.6)
NS
Idiopathic pericarditis
51 (85.0)
50 (83.3)
Autoimmune causes*
9 (15.0)
10 (16.7)
23.7⫾8.8
Group II:
Colchicine
(n⫽60)
24.2⫾8.7
P
NS
32.3
10.7
0.004†
22 (36.7)
7 (11.7)
0.003
Side effects, n (%)
4 (6.7)
5 (8.3)
NS
Severe adverse effects, n (%)
0 (0.0)
0 (0.0)
NS
NS
Cardiac tamponade, n (%)
0 (0.0)
0 (0.0)
NS
NS
Constrictive pericarditis, n (%)
0 (0.0)
0 (0.0)
NS
Values are n (%) or mean⫾SD.
*Autoimmune causes include connective tissue diseases and postpericardiotomy
syndromes.
*Steroid prescribed for the index attack because of aspirin contraindications
or intolerance.
†P value from log-rank test.
Imazio et al
Figure 2. Kaplan-Meier event-free survival curves according to
treatment groups (see text for details).
however, evidence for this use comes from a small study on 19
patients and consensus opinion of the experts. This opinion is
derived mainly from studies in which colchicine was used in the
treatment of patients with recurrences after failure of conventional
treatment but not after a first event. The 2 populations may be quite
different given that it is generally accepted that recurrence is an
autoimmune process,6,16,23 whereas the first episode generally has
an infectious cause (ie, viral).6,14–18,20,28 The use of colchicine in any
acute pericarditis as “primary” prevention of recurrences may
represent an important step in the management of acute pericarditis
if controlled trials confirm the initial positive results.22 However,
there are no clinical studies to guide the evaluation and management
of acute pericarditis, and strategies to prevent recurrences require
further study.20,22
The present study was designed to address whether colchicine
is a useful addition to conventional treatment either in therapy of
the first episode or in prevention of recurrences. Colchicine may
be a way to cope with this complication.
TABLE 3. Baseline Clinical Features of Patients With and
Without Recurrences During Follow-Up
Colchicine for Acute Pericarditis
2015
Colchicine proved useful to control symptoms within 72
hours faster than aspirin or prednisone alone (Table 2). These
data are similar to what has been described in patients with gouty
attack. Most patients who receive colchicine respond within 18
hours, and joint inflammation subsides in 75% to 80% of
patients within 48 hours.1 Moreover, colchicine was able to
reduce the subsequent recurrence rate by ⬇3-fold (recurrence
rates at 18 months were 10.7% versus 32.3% with and without
colchicine, respectively; P⫽0.004), and thus the number of
patients with a first episode of acute pericarditis who need to be
treated to prevent a recurrence is only 5.
The exact mechanism of colchicine action is not fully understood. Most of the pharmacological effects of colchicine on cells
involved in inflammation appear to be related to its capacity to
disrupt microtubules. Colchicine inhibits the process of microtubule
self-assembly by binding ␤-tubulin with the formation of tubulincolchicine complexes. This action takes place either in the mitotic
spindle or in the interphase stage, and thus, colchicine inhibits the
movement of intercellular granules and the secretion of various
substances.1,2 By this mechanism, colchicine is able to inhibit
various leukocyte functions, and this effect should be the most
significant for its antiinflammatory action. Moreover, colchicine
shows a preferential concentration in leukocytes, and the peak
concentration of colchicine may be ⬎16 times the peak concentration in plasma. This appears to be related to its therapeutic effect.1,2
Risk Factors for Recurrences
As already reported,6,14,15,28 in previous studies, no characteristics of the first episode of acute pericarditis were able to predict
the likelihood of recurrences; however, concern has been raised
that treatment of acute pericarditis with prednisone may increase
the risk of recurrence.6,20,22,24,28 –30 The present study appears to
support this fear, because patients with recurrences during
follow-up had a higher rate of previous corticosteroid use in the
index attack (Table 3). After multivariate analysis, prednisone
use was an independent risk factor for the subsequent development of recurrences (OR 4.30, 95% CI 1.21 to 15.25; P⫽0.024).
Animal studies have shown that corticosteroids may exacerbate
virally induced pericardial injury.20 Corticosteroid therapy given in
the index attack can favor the occurrence of recurrences, probably
because of its deleterious effect on viral replication. Corticosteroids
may perpetuate pericardial inflammation instead of resolving it;
moreover, frequent and prolonged administration may lead to
serious complications.6,20,22,28,30 These data argue against the routine
administration of corticosteroids during a first episode of acute
pericarditis.
Feature
Patients With
Recurrence
(n⫽27)
Patients Without
Recurrence
(n⫽93)
P
Age, y
57.3⫾18.8
56.7⫾18.9
NS
Female gender
19 (70.4)
47 (50.5)
NS
Pericardial effusion
20 (74.1)
59 (63.4)
NS
Safety
Severe pericardial effusion
5 (18.5)
5 (5.4)
NS
Cardiac tamponade
1 (3.7)
1 (1.1)
NS
Idiopathic etiology
NS
At doses of 1 to 2 mg per day, colchicine has been found to be
safe even when given continuously over decades.1,2,16 Gastrointestinal side effects are not uncommon, occurring in up to 10%
of cases, although they are generally mild and may resolve with
dose reduction.31,32 In studies in which colchicine was used to
treat recurrences, temporary discontinuation of the drug or a
reduction of its dose was needed in ⬇10% to 14% of cases.16
These side effects may limit its therapeutic applicability.
In the present study, on the basis of previous experiences,13,24
we used the lowest effective dose while also taking into account
the weight of the treated patients. With these doses, we recorded
21 (77.8)
80 (86.0)
Autoimmune causes*
6 (22.2)
13 (14.0)
NS
Corticosteroid use†
9 (33.3)
10 (10.7)
0.011
Colchicine use
7 (3.7)
53 (56.9)
⬍0.001
Values are n (%) or mean⫾SD.
*Autoimmune causes include connective tissue diseases and postpericardiotomy
syndromes.
†Steroid prescribed for the index attack because of aspirin contraindications or
intolerance.
2016
Circulation
September 27, 2005
5 cases of diarrhea (8.3%), which were promptly reversible after
drug withdrawal. Two (40.0%) of these patients experienced
recurrences after drug discontinuation. No serious adverse effects were observed. In the largest prospective multicenter study
on recurrent pericarditis and colchicine,11 the drug (ⱖ1 mg/d)
was discontinued in 39 patients (76.5%), and 14 of them (35.9%)
experienced relapses.
Other concerns are related to bone marrow suppression and
fertility. After a cumulative 15 000 years of follow-up in patients
with familial Mediterranean fever, no interference of colchicine
treatment was recorded with regard to either growth rate or
fertility.33
Study Limitations
A possible study limitation is the open-label design. This work was
designed as a preliminary study to test the hypothesis that early
treatment of the first recurrence with colchicine as an adjunct to
conventional therapy may reduce the subsequent recurrence rate.
Moreover, the measured end points, including symptom status at 72
hours and symptom recurrence over time, are subjectively determined by the patient and physician. These limitations would have
been avoided by the use of a double-blind study design. However,
validation of clinical events was ensured by an ad hoc committee of
expert cardiologists blinded to patients’ treatment assignment,
whereas data analyses were performed by an external data analysis
committee masked to treatment assignment. Moreover, strict adherence to the intention-to-treat principle ensures that the effects seen
correspond closely to what is achievable in clinical practice. At
present, this study is the first randomized trial in this area. The
present study provides good evidence that colchicine as an adjunct
to conventional therapy is safe and effective in treatment of the first
episode of acute pericarditis, and it shows that colchicine plus
conventional therapy might be considered as first-choice treatment
for acute pericarditis.
Appendix
Investigators of the COPE Trial
Coordinating Center: Cardiology Department, Maria Vittoria Hospital,
Turin, Italy. Ad Hoc Committee for the Validation of Clinical Events: E.
Cecchi, D. Demarie. Safety Monitoring Committee: R. Trinchero, M.
Imazio. External Data Analysis Committee: M. Bobbio, A. Brusca (deceased). COPE trial centers: Maria Vittoria Hospital, Amedeo di Savoia
Hospital, Turin, Italy.
References
1. Molad Y. Update on colchicine and its mechanism of action. Curr
Rheumatol Rep. 2002;3:252–256.
2. Lange U, Schumann C, Schmidt KL. Current aspects of colchicine therapy:
classical indications and new therapeutic uses. Eur J Med Res. 2001;6:
150–160.
3. Emmerson BT. The management of gout. N Engl J Med. 1996;334:445–451.
4. Ben-Chetrit E, Levy M. Colchicine prophylaxis in familial Mediterranean
fever: reappraisal after 15 years. Semin Arthritis Rheum. 1991;20:241–246.
5. Kees S, Langevitz P, Zemer D, Padeh S, Pras M, Livneh A. Attacks of
pericarditis as a manifestation of familial Mediterranean fever (FMF). QJM.
1997;90:643–647.
6. Shabetai R, Adler Y. Recurrent pericarditis. In: Rose BD, ed. UptoDate.
Uptodate online. Wellesley, Mass: 2004. Available at: http://patients.uptodate.
com/topic.asp?file⫽myoperic/6168&title⫽Recurrent⫹Pericarditis.
7. Rodriguez de la Serna A, Guido J, Marti V, Bayes de Luna A. Colchicine for
recurrent pericarditis. Lancet. 1987;2:1517. Letter.
8. Guindo J, Rodriguez de la Serna A, Ramio J, de Miguel Diaz MA, Subirana
MT, Perez Ayuso MJ, Cosin J, Bayes de Luna A. Recurrent pericarditis:
relief with colchicine. Circulation. 1990;82:1117–1120.
9. Adler Y, Zandman-Goddard G, Ravid M, Avidan B, Zemer D, Ehrenfeld M,
Shemesh J, Tomer Y, Shoenfeld Y. Usefulness of colchicine in preventing
recurrences of pericarditis. Am J Cardiol. 1994;73:916–917.
10. Millaire A, de Groote P, Decoulx E, Goullard L, Ducloux G. Treatment of
recurrent pericarditis with colchicine. Eur Heart J. 1994;15:120–124.
11. Guindo J, Adler Y, Spodick DH, Rodriguez de la Serna A, Shoenfeld Y,
Daniel-Riesco C, Finkelstein Y, Bayes-Genis A, Miguel de Miguel A,
Subirana M, Scarovsky S, Bayes de Luna A. Colchicine for recurrent pericarditis: 51 patients followed up for 10 years. Circulation. 1997;96(suppl
I):I-29. Abstract.
12. Cacoub P, Sbai A, Wechsler B, Amoura Z, Godeau P, Piette JC. Efficacy of
colchicine in recurrent acute idiopathic pericarditis. Arch Mal Coeur Vaiss.
2000;93:1511–1514.
13. Imazio M, Demichelis B, Cecchi E, Giuggia M, Forno D, Trinchero R.
Recurrent pericarditis: follow-up of 55 cases. Circulation. 2002;106 (suppl
II):II-323. Abstract.
14. Permanyer-Miralda G, Sagrista-Sauleda J, Soler-Soler J. Primary acute pericardial disease: a prospective series of 231 consecutive patients.
Am J Cardiol. 1985;56:623–630.
15. Zayas R, Anguita M, Torres F, Gimenez D, Bergillos F, Ruiz M, Ciudad M,
Gallardo A, Valles F. Incidence of specific etiology and role of methods for
specific etiologic diagnosis of primary acute pericarditis. Am J Cardiol.
1995;75:378–382.
16. Adler Y, Finkelstein Y, Guindo J, Rodriguez de la Serna A, Shoenfeld Y,
Bayes-Genis A, Sagie A, Bayes de Luna A, Spodick DH. Colchicine
treatment for recurrent pericarditis: a decade of experience. Circulation.
1998;97:2183–2185.
17. Imazio M, Demichelis B, Cecchi E, Belli R, Ghisio A, Bobbio M, Trinchero
R. Cardiac troponin I in acute pericarditis. J Am Coll Cardiol. 2003;42:
2144–2148.
18. Imazio M, Demichelis B, Parrini I, Giuggia M, Cecchi E, Gaschino G,
Demarie D, Ghisio A, Trinchero R. Day hospital treatment of acute pericarditis. J Am Coll Cardiol. 2004;43:1042–1046.
19. Troughton R, Asher CR, Klein AL. Pericarditis. Lancet. 2004;363:717–727.
20. Lange RA, Hillis LD. Acute pericarditis. N Engl J Med. 2004;351:
2195–2202.
21. Maisch B, Ristic A. Practical aspects of the management of pericardial
disease. Heart. 2003;89:1096–1103.
22. Soler-Soler J, Sagristà-Sauleda J, Permanyer-Miralda G. Relapsing pericarditis. Heart. 2004;90:1364–1368.
23. Marcolongo R, Russo R, Laveder F, Noventa F, Agostini C. Immunosuppressive therapy prevents recurrent pericarditis. J Am Coll Cardiol. 1995;26:
1276–1279.
24. Imazio M, Trinchero R. Clinical management of acute pericardial disease: a
review of results and outcomes. Ital Heart J. 2004;5:803–817.
25. Spodick DH. Acute pericarditis: current concepts and practice. JAMA. 2003;
289:1150–1153.
26. Maisch B, Seferovic PM, Ristic AD, Erbel R, Rienmuller R, Adler Y,
Tomkowski WZ, Thiene G, Yacoub MH; Task Force on the Diagnosis and
Management of Pericardial Diseases of the European Society of Cardiology.
Guidelines on the diagnosis and management of pericardial diseases. Eur
Heart J. 2004;25: 587–610.
27. Millaire A, Ducloux G. Treatment of acute or recurrent pericarditis with
colchicine. Circulation. 1991;83:1458–1459. Letter.
28. Shabetai R, Imazio M. Evaluation and management of acute pericarditis. In:
Rose BD, ed. UptoDate. Uptodate online. Wellesley, Mass: 2004. Available
at: http://patients.uptodate.com/topic.asp?file⫽myoperic/5163.
29. Spodick DH. The Pericardium: A Comprehensive Textbook. New York, NY.
Marcel Dekker; 1997.
30. Imazio M, Demichelis B, Parrini I, Cecchi E, Pomari F, Demarie D, Gaschino
G, Ghisio A, Belli R, Trinchero R. Recurrent pericarditic pain without
objective evidence of disease in patients with previous acute pericarditis.
Am J Cardiol. 2004;94:973–975.
31. Ben-Chetrit E, Levy M. Colchicine prophylaxis in familial Mediterranean
fever: reappraisal after 15 years. Semin Arthritis Rheum. 1991;20:241–246.
32. Ehrenfeld M, Levy M, Sharon P, Rachmilewitz D, Eliakim M. Gastrointestinal effects of long-term colchicine therapy in patients with recurrent
polyserositis. Dig Dis Sci. 1982;27:723–727.
33. Zemer D, Livneh A, Danon YL, Pras M, Sohar E. Long term colchicine
treatment in children with familial Mediterranean fever. Arthritis Rheum.
1991;34:973–977.
Preventive Cardiology
Simple Risk Stratification at Admission to Identify Patients
With Reduced Mortality From Primary Angioplasty
Jens Jakob Thune, MD; Dan Eik Hoefsten, MD; Matias Greve Lindholm, MD;
Leif Spange Mortensen, MSc; Henning Rud Andersen, MD; Torsten Toftegaard Nielsen, MD;
Lars Kober, MD; Henning Kelbaek, MD; for the Danish Multicenter Randomized Study on Fibrinolytic
Therapy Versus Acute Coronary Angioplasty in Acute Myocardial Infarction
(DANAMI)-2 Investigators
Background—Randomized trials comparing fibrinolysis with primary angioplasty for acute ST-elevation myocardial
infarction have demonstrated a beneficial effect of primary angioplasty on the combined end point of death, reinfarction,
and disabling stroke but not on all-cause death. Identifying a patient group with reduced mortality from an invasive
strategy would be important for early triage. The Thrombolysis in Myocardial Infarction (TIMI) risk score is a simple
validated integer score that makes it possible to identify high-risk patients on admission to hospital. We hypothesized
that a high-risk group might have a reduced mortality with an invasive strategy.
Methods and Results—We classified 1527 patients from the Danish Multicenter Randomized Study on Fibrinolytic
Therapy Versus Acute Coronary Angioplasty in Acute Myocardial Infarction (DANAMI-2) trial with information for
all variables necessary for calculating the TIMI risk score as low risk (TIMI risk score, 0 to 4) or high risk (TIMI risk
score ⱖ5) and investigated the effect of primary angioplasty versus fibrinolysis on mortality and morbidity in the 2
groups. Follow-up was 3 years. We classified 1134 patients as low risk and 393 as high risk. There was a significant
interaction between risk status and effect of primary angioplasty (P⫽0.008). In the low-risk group, there was no
difference in mortality (primary angioplasty, 8.0%; fibrinolysis, 5.6%; P⫽0.11); in the high-risk group, there was a
significant reduction in mortality with primary angioplasty (25.3% versus 36.2%; P⫽0.02).
Conclusions—Risk stratification at admission based on the TIMI risk score identifies a group of high-risk patients who have
a significantly reduced mortality with an invasive strategy of primary angioplasty. (Circulation. 2005;112:2017-2021.)
Key Words: angioplasty 䡲 fibrinolysis 䡲 mortality 䡲 myocardial infarction
T
he initial treatment of patients with acute ST-segment
elevation myocardial infarction is either fibrinolysis or
primary angioplasty.1 Several trials have demonstrated a
superior effect of primary angioplasty over fibrinolysis,
and this observation has been substantiated in meta-analyses.2,3 Consequently, current guidelines recommend primary angioplasty as the treatment of choice whenever
feasible.4 There is still no consensus, however, as to
whether all patients benefit from an invasive strategy when
applied to a community setting, including hospitals without invasive treatment facilities.1,5 Reduced mortality has
been demonstrated with urgent angioplasty for patients
with cardiogenic shock complicating acute myocardial
infarction,6 but no trial that included noninvasive treatment
hospitals has succeeded in documenting an effect on
mortality for a substantial proportion of patients with acute
ST-segment elevation myocardial infarction. The most
recent report on the use of primary angioplasty from the
Global Registry of Acute Coronary Events (GRACE)
showed that 26.7% of patients with acute ST-segment
elevation myocardial infarction were treated with primary
angioplasty and 47% received fibrinolysis.7
It has been shown in a community-based patient sample
that the added benefit of primary angioplasty increases with
higher overall risk.8 Consequently, in high-risk patients,
primary angioplasty might reduce mortality compared with
fibrinolysis. Thus, identifying such high-risk patients would
be desirable. If a full community-wide strategy of invasive
treatment for all patients with ST-elevation myocardial infarction is not feasible, identifying patients most likely to
benefit from the invasive strategy on admission is very
important.
The Thrombolysis in Myocardial Infarction (TIMI) risk
score for ST-segment elevation myocardial infarction is a
Received May 1, 2005; revision received July 6, 2005; accepted July 11, 2005.
From the Department of Cardiology, University Hospital of Copenhagen, Rigshospitalet, Copenhagen (J.J.T., M.G.L., L.K., H.K.); Department of
Medical Research, Funen Hospital, Svendborg (D.E.H.); UNI-C, Danish Information Technology Centre for Education and Research, Aarhus, Denmark
(L.S.M.); and Department of Cardiology, Aarhus University Hospital, Skejby Hospital, Aarhus (H.R.A., T.T.N.), Denmark.
Correspondence to Jens Jakob Thune, MD, Department of Cardiology, B2141, University Hospital of Copenhagen, Rigshospitalet, Blegdamsvej 9,
DK-2100 Copenhagen, Denmark. E-mail [email protected]
2017
2018
Circulation
September 27, 2005
simple arithmetic score based on data easily obtained at
admission. When the TIMI risk score is used at admission,
patients can easily be classified as low or high risk without
any delay in treatment. Analyses of multiple clinical trials of
fibrinolysis and a community-based population that included
patients treated with primary angioplasty have validated that
a higher TIMI risk score is associated with a greater risk of
death from all causes.9,10
Thus, for the present study, we applied the TIMI risk score
to classify patients as low or high risk in the Danish
Multicenter Randomized Study on Fibrinolytic Therapy versus Acute Coronary Angioplasty in Acute Myocardial Infarction (DANAMI-2) and hypothesized that high-risk patients
would have a greater benefit from primary angioplasty than
low-risk patients.
Methods
The design and rationale of the DANAMI-2 trial have been published previously.11 In brief, 1572 patients with acute ST-segment
elevation myocardial infarction were randomized to fibrinolysis with
intravenous alteplase or primary angioplasty. We recruited 1129
patients from 24 referral hospitals without invasive treatment facilities and 443 patients from 5 invasive treatment centers.
For the present study, we used patient data and vital signs obtained
at randomization in the DANAMI-2 trial. For each patient, the TIMI
risk score was calculated as the arithmetic sum of the following
variables obtained at admission: age ⱖ75 years⫽3 points; age 65 to
74 years⫽2 points; systolic blood pressure ⬍100 mm Hg⫽3 points;
heart rate ⬎100 bpm⫽2 points; Killip class 2 to 4⫽2 points; weight
⬍67 kg⫽1 point; anterior ST-segment elevation⫽1 point; time from
symptom onset to treatment ⬎4 hours⫽1 point; and a history of
angina, diabetes, or hypertension⫽1 point, for a possible score of 0
to 14.9 Time to treatment was defined as time to onset of fibrinolysis
or the time to first injection of contrast in the coronary artery.
Patients were classified as low risk if their TIMI score was 0 to 4
and as high risk if their TIMI score was ⱖ5. Survival analyses were
carried out using Kaplan-Meier curves with log-rank tests for
homogeneity. Hazard ratios were analyzed with Cox regression
analysis. Analyses of possible interaction between attributed risk and
treatment strategy were performed by entering an interaction term
into the regression model. For all analyses, a value of P⬍0.05 was
considered statistically significant. Analyses were performed with
SAS version 9.1 (SAS Institute Inc).
The primary end point was time to death from all causes. Our
secondary end point was the composite end point of death, recurrent
myocardial infarction, and disabling stroke. Patient follow-up was 3
years; patients suffering a nonfatal event continued follow-up for the
mortality end point. The present study was not prespecified in the
original DANAMI-2 protocol.
The DANAMI-2 protocol was approved by the local ethics
committee for all participating hospitals. All patients gave written
informed consent.
Results
The results from the DANAMI-2 trial have been published
previously.12 In summary, there was a relative risk reduction
with primary angioplasty for the combined end point of death,
reinfarction, and disabling stroke at 30 days of 42%
(P⬍0.001) but no difference in all-cause mortality (P⫽0.35).
The present study population consisted of 1527 patients for
whom all TIMI risk score variables were available. The
information missing was weight for 19 patients; previous
angina, diabetes, or hypertension for 9 patients; systolic blood
pressure for 4 patients, Killip class for 4 patients; heart rate
for 2 patients; and time to treatment for 22 patients. Results
were unchanged whether the patients with missing data were
included with their highest or lowest possible score. On
average, the 45 patients not included in the analysis were
insignificantly older (mean difference, 4.3 years; P⫽0.06)
and more often female (40% versus 26%; P⫽0.04).
Baseline demographics are shown in the Table. All variables included in the TIMI risk score were more prevalent in
the high-risk group than in the low-risk group. Smoking was
more prevalent in the low-risk group.
The TIMI risk score was distributed as follows: 0 points,
139 patients; 1 point, 265 patients; 2 points, 259 patients; 3
points, 243 patients; 4 points, 228 patients; 5 points, 147
patients; 6 points, 106 patients; 7 points, 82 patients, and ⱖ8
points, 58 patients. No patients had a TIMI risk score ⬎11
points. The TIMI risk score was a significant predictor of
all-cause death (P⬍0.001), with a hazard ratio for each
additional point in the TIMI score of 1.57 (95% CI, 1.48 to
1.68).
Figure 1 displays the Kaplan-Meier curves associated with
the 2 treatment strategies stratified by risk status according to
TIMI risk score. There was a significant interaction between
attributed risk and treatment strategy (P⫽0.008). For patients
classified as low risk, there was no significant difference in
3-year mortality between the 2 treatment arms (primary
angioplasty, 8.0%; fibrinolysis, 5.6%; hazard ratio, 1.44; 95%
CI, 0.91 to 2.27; P⫽0.11), whereas patients classified as high
risk had a significantly lower 3-year mortality rate with the
invasive strategy compared with fibrinolysis (25.3% versus
36.2%; number needed to treat, 9; hazard ratio, 0.66; 95% CI,
0.45 to 0.94; P⫽0.02).
For patients randomized at a referral hospital, there was a
significant reduction in mortality with primary angioplasty
over fibrinolysis for high-risk patients (24.6% versus 36.8%;
number needed to treat, 8; P⫽0.02) but not for low-risk
patients (7.5% versus 6.6%; P⫽0.62). Because of the low
difference in mortality for the low-risk patients, this interaction was not statistically significant (P⫽0.07). Results for
invasive centers were also similar to the overall results but did
not show significance because of low patient numbers and
number of events.
When the composite end point of death, reinfarction, and
disabling stroke is used, there is no difference in effect
between primary angioplasty and fibrinolysis in the low-risk
group (13.7% versus 15.7; P⫽0.30), but there is a significant
reduction in events with primary angioplasty in the high-risk
group (32.3% versus 45.9%; P⫽0.004). The interaction was
not significant (P⫽0.17). Event curves are shown in Figure 2.
There was a significant reduction in number of reinfarctions in the low-risk group with primary angioplasty (6.6%
versus 10.4%; P⫽0.02), whereas in the high-risk group, it
was not significant because of the lower patient numbers
(10.2% versus 13.5%; P⫽0.18). There was no significant
interaction between treatment strategy and risk status
(P⫽0.78). The number of disabling strokes was low in both
the low- and high-risk groups, and there was no significant
difference in effect between treatments in either group (lowrisk group, 1.7% versus 1.6%, P⫽0.87; high-risk group, 5.3%
versus 9.2%, P⫽0.11).
Thune et al
Risk Stratification and Primary Angioplasty
2019
Baseline Demographics
TIMI Score 0 – 4
TIMI Scoreⱖ5
Fx
(n⫽556)
PA
(n⫽578)
All
(n⫽1134)
Fx
(n⫽207)
PA
(n⫽186)
All
(n⫽393)
Total
(n⫽1527)
59⫾11
59⫾11
59⫾11
75⫾9
74⫾10
74⫾10
63⫾13
21
22
21
40
40
40
26
27⫾4
27⫾4
27⫾4
25⫾5
25⫾4
25⫾4
26⫾4
Myocardial infarction*
10
10
10
17
13
16
11
Hypertension*
16
17
17
32
29
30
20
6
6
6
10
11
11
7
Age,* y
Female gender,* %
BMI,* kg/m2
History of
Diabetes*
Smoking*
63
62
63
44
41
43
57
Angina*
23
23
23
40
45
42
28
Randomized at invasive center, %
29
29
29
27
28
27
29
Systolic blood pressure,* mm Hg
135⫾24
138⫾24
137⫾24
132⫾33
131⫾32
131⫾33
135⫾27
Heart rate,* bpm
72⫾16
73⫾16
73⫾16
81⫾24
82⫾25
82⫾24
75⫾20
TIMI risk score components, %
Age ⱖ75 y*
10
6
8
57
54
55
20
Age 65–74 y*
21
24
22
34
37
35
27
Systolic blood pressure ⬍100 mm Hg*
4
2
3
21
22
21
8
Heart rate ⬎100 bpm*
3
4
4
20
21
20
8
Killip class ⬎2–4*†
4
2
3
31
20
26
9
14
13
14
37
37
38
20
Weight ⬍67 kg*
Anterior myocardial infarction*
44
48
46
76
72
74
53
⬎4 h to treatment†
24
31
27
44
61
52
34
Glycoprotein IIb/IIIa inhibitor,† %
0
40
19
0
40
20
20
ACE inhibitor at discharge,* %
34
32
33
50
40
45
36
␤-Blocker at discharge,* %
87
89
88
82
85
83
83
Fx indicates fibrinolysis; PA, primary angioplasty; and BMI, body mass index.
*P⬍0.05 for comparison between the 2 TIMI score groups.
†P⬍0.05 for comparison between the fibrinolysis and primary angioplasty groups within the same TIMI score group.
Discussion
Our results show that stratifying patients with acute STsegment elevation myocardial infarction as low or high risk
by the use of the TIMI risk score identifies a group of
high-risk patients with a lower 3-year mortality rate with
primary angioplasty than with fibrinolysis. To the best of our
knowledge, this is the first time such a substantial and readily
identifiable proportion of patients with acute ST-segment
elevation myocardial infarction has been shown to have
reduced mortality with primary angioplasty compared with
fibrinolysis in a community-setting that included both referral
and invasive treatment hospitals.
The statistical interaction between risk status and treatment
effect was due in part to an inverse effect in the low-risk
group that did not reach statistical significance. This trend
strengthens the conclusion that there indeed is an interaction
between risk status and the effect of treatment and that the
lack of significant effect of primary angioplasty in the
low-risk group is not due merely to a low number of events.
Our results are in concordance with the analysis by Kent
and coworkers,8 who found that an effect on mortality from
primary angioplasty was not likely in patients with an
estimated 30-day mortality rate of ⬇2% or less. The 30-day
mortality rate in the low-risk group treated with fibrinolysis
in the present study was 2.5%, so the low-risk group
corresponds well to the group of patients not likely to obtain
a reduction in mortality from primary angioplasty according
to Kent and coworkers. In contrast to our results, Morrow and
coworkers10 found, when validating the TIMI risk score in the
National Registry of Myocardial Infarction 3, that there was
no difference between the slopes of mortality gradients with
increasing risk scores for primary angioplasty and fibrinolysis. This discordance might be due to differences in demographics because the patients in the National Registry of
Myocardial Infarction 3 were, of course, not randomized.
Our results are also in concordance with the overall 30-day
results from the DANAMI-2 trial. In the present study, there
was a larger reduction in the combined end point in the
high-risk group than in the low-risk group, but this difference
was not significant. Thus, the reduced incidence of the
combined end point for patients randomized to primary
angioplasty was not exclusive to the high-risk group of
patients. The reduced incidence of the combined end point of
death, reinfarction, and disabling stroke with primary angioplasty reported previously for the DANAMI-2 trial applies to
the entire trial population. This results particularly from the
2020
Circulation
September 27, 2005
Figure 1. Mortality rates for low-risk patients treated with fibrinolysis (Fx) (black dashed line) or primary angioplasty (PA) (red
dashed line) and high-risk patients treated with fibrinolysis
(black solid line) or primary angioplasty (red solid line).
Figure 2. Combined event rates of death, reinfarction, or disabling stroke for low-risk patients treated with fibrinolysis (Fx)
(black dashed line) or primary angioplasty (PA) (red dashed line)
and high-risk patients treated with fibrinolysis (black solid line)
or primary angioplasty (red solid line).
markedly reduced rate of reinfarction in the primary angioplasty group and the fact that the effect of treatment does not
interact with risk groups. TIMI risk score was developed to
predict mortality and thus includes parameters not as strongly
related to the risk of reinfarction.
Our finding that primary angioplasty reduces mortality for
high-risk patients admitted to referral hospitals without facilities for primary angioplasty was the hypothesis of the
randomized AIR-PAMI study, which was terminated early as
a result of low inclusion rates.13 The data from AIR-PAMI
showed a nonsignificant trend toward reduced 30-day mortality with primary angioplasty, and we substantiate this
finding because our data demonstrate that high-risk patients
admitted to a referral hospital in the DANAMI-2 trial indeed
did have reduced mortality at 3 years. The definition of
high-risk patients in AIR-PAMI was based on the same
variables as contained in the TIMI risk score, but classification of high risk required only the presence of 1 high-risk
criterion, whereas our definition of high risk requires a
minimum of 2 high-risk criteria present to obtain a TIMI
score of at least 5.
Recently, a paper from the GRACE investigators reported
that there was no benefit in outcomes for patients admitted to
a hospital with invasive facilities compared with patients
admitted to a hospital without invasive facilities.14 This
analysis was based on all patients with acute coronary
syndrome, including patients with unstable angina and non–
ST-segment elevation myocardial infarction, which constitutes a patient group with a much lower overall risk than our
high-risk group. Furthermore, analyses were performed according to whether the hospital had invasive facilities, and not
whether patients were actually treated invasively. Thus, the
analysis of the GRACE registry with its different focus is not
comparable to ours.
The present results could have important implications for
clinical practice. Because previous analyses showed an increased benefit of primary angioplasty in patients at greater
risk, the next step has been to identify such a group of
high-risk patients. Our analysis suggests that the TIMI risk
score might serve as an impetus to perform urgent angioplasty by making it possible to rapidly identify on admission
those patients whose mortality risk would most likely be
reduced by an invasive approach. This will be of particular
importance in communities with limited resources for primary angioplasty where not all patients can be offered this
treatment strategy and where risk stratification would make it
possible to prioritize high-risk patients with particular benefit
of an invasive strategy. It is possible that the added costs of
implementing a community-wide program is due more to the
setup of the infrastructure rather than the individual transfers,
so it might be just as costly to implement a system to transport
only high-risk patients as to transport all patients with
ST-elevation myocardial infarction. However, the TIMI risk
score would still be beneficial in deciding who should be
transported for primary angioplasty in communities with low
capacity.
Because the TIMI risk score for risk stratification of
patients was not published before DANAMI-2 was conducted, the present study constitutes a post hoc analysis not
specified in the original protocol. However, the idea to
investigate using the TIMI risk score to identify high-risk
patients who would possibly benefit more from primary
angioplasty was conceived before any analyses were made.
Our decision to use a TIMI score ⱖ5 to define the high-risk
group is identical to the definition used by the TIMI Study
Group.15 The TIMI Study Group further divided the groups
with TIMI risk scores ⬍5 into a low-risk group and an
intermediate group, but we chose to collapse these 2 groups
into 1 because we considered a high-risk versus low-risk
variable to be more operational and because the 2 groups
showed similar results in our analyses (data not shown).
Although the TIMI risk score assigns 1 point for anterior
myocardial infarction or new left bundle-branch block, patients with left bundle-branch block were excluded from the
Thune et al
DANAMI-2 trial to avoid diagnostic uncertainty, so this
variable is different from that of the original TIMI risk score.
Because there were no patients with left bundle-branch block
in the 2 treatment groups, this should not cause any discrepancy between groups and thus should not affect the analysis.
In summary, although individual trials and a meta-analysis
show a clear benefit of primary angioplasty compared with
fibrinolysis on the combined end point of all-cause death,
reinfarction, and disabling stroke, no benefit on mortality of
a community-wide invasive strategy including invasive and
referral hospitals has been demonstrated. The reason could be
that patients at low risk of death do not obtain a significant
reduction of mortality and that this dilutes the benefit obtained by high-risk patients. Thus, by identifying a group of
easily recognizable high-risk patients, we have shown that
these patients do indeed have a significantly reduced mortality with an invasive strategy. This signifies that not only
patients in cardiogenic shock but also a substantially larger
proportion of patients with acute ST-segment elevation myocardial infarction (26% of the DANAMI-2 population) would
experience a lower mortality with a community-wide implementation of an invasive strategy of primary coronary
angioplasty.
References
1. Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand M,
Hochman JS, Krumholz HM, Kushner FG, Lamas GA, Mullany CJ,
Ornato JP, Pearle DL, Sloan MA, Smith SC Jr, Alpert JS, Anderson JL,
Faxon DP, Fuster V, Gibbons RJ, Gregoratos G, Halperin JL, Hiratzka
LF, Hunt SA, Jacobs AK. ACC/AHA guidelines for the management of
patients with ST-elevation myocardial infarction. Circulation. 2004;110:
e82– e292.
2. Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous
thrombolytic therapy for acute myocardial infarction: a quantitative
review of 23 randomised trials. Lancet. 2003;361:13–20.
3. Dalby M, Bouzamondo A, Lechat P, Montalescot G. Transfer for primary
angioplasty versus immediate thrombolysis in acute myocardial
infarction: a meta-analysis. Circulation. 2003;108:1809 –1814.
4. Van de Werf F, Ardissino D, Betriu A, Cokkinos DV, Falk E, Fox KAA,
Julian D, Lengyel M, Neumann FJ, Ruzyllo W, Thygesen C, Underwood
SR, Vahanian A, Verheugt FWA, Wijns W, for the Task Force on the
Management of Acute Myocardial Infarction of the European Society of
Cardiology. Management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J. 2003;24:28 – 66.
Risk Stratification and Primary Angioplasty
2021
5. Brophy JM, Bogaty P. Primary angioplasty and thrombolysis are both
reasonable options in acute myocardial infarction. Ann Intern Med. 2004;
141:292–297.
6. Hochman JS, Sleeper LA, Webb JG, Sanborn TA, White HD, Talley JD,
Buller CE, Jacobs AK, Slater JN, Col J, McKinlay SM, LeJemtel TH.
Early revascularization in acute myocardial infarction complicated by
cardiogenic shock. N Engl J Med. 1999;341:625– 634.
7. Fox KAA, Goodman SG, Anderson FA Jr, Granger CB, Moscucci M,
Flather MD, Spencer F, Budaj A, Dabbous OH, Gore JM, for the GRACE
Investigators. From guidelines to clinical practice: the impact of hospital
and geographical characteristics on temporal trends in the management of
acute coronary syndromes: the Global Registry of Acute Coronary Events
(GRACE). Eur Heart J. 2003;24:1414 –1424.
8. Kent DM, Schmid CH, Lau J, Selker HP. Is primary angioplasty for some
as good as primary angioplasty for all? Modeling across trials and individual patients. J Gen Intern Med. 2002;17:887– 894.
9. Morrow DA, Antman EM, Charlesworth A, Cairns R, Murphy SA, de
Lemos JA, Giugliano RP, McCabe CH, Braunwald E. TIMI risk score for
ST-elevation myocardial infarction: a convenient, bedside, clinical score
for risk assessment at presentation: an Intravenous nPA for Treatment of
Infarcting Myocardium Early II Trial substudy. Circulation. 2000;102:
2031–2037.
10. Morrow DA, Antman EM, Parsons L, de Lemos JA, Cannon CP,
Giugliano RP, McCabe CH, Barron HV, Braunwald E. Application of the
TIMI risk score for ST-elevation MI in the National Registry of Myocardial Infarction 3. JAMA. 2001;286:1356 –1359.
11. Andersen HR, Nielsen TT, Vesterlund T, Grande P, Abildgaard U,
Thayssen P, Pedersen F, Mortensen LS. Danish multicenter randomized
study on fibrinolytic therapy versus acute coronary angioplasty in acute
myocardial infarction: rationale and design of the Danish Trial in Acute
Myocardial Infarction-2 (DANAMI-2). Am Heart J. 2003;146:234 –241.
12. Andersen HR, Nielsen TT, Rasmussen K, Thuesen L, Kelbaek H,
Thayssen P, Abildgaard U, Pedersen F, Madsen JK, Grande P, Villadsen
AB, Krusell LR, Haghfelt T, Lomholt P, Husted SE, Vigholt E, Kjaergard
HK, Mortensen LS. A comparison of coronary angioplasty with
fibrinolytic therapy in acute myocardial infarction. N Engl J Med. 2003;
349:733–742.
13. Grines CL, Westerhausen J, Grines LL, Hanlon JT, Logemann TL,
Niemela M, Weaver WD, Graham M, Boura J. A randomized trial of
transfer for primary angioplasty versus on-site thrombolysis in patients
with high-risk myocardial infarction: the AIR Primary Angioplasty in
Myocardial Infarction study. J Am Coll Cardiol. 2002;39:1713–1719.
14. Van de Werf F, Gore JM, Avezum A, Gulba DC, Goodman SG, Budaj A,
Brieger D, White K, Fox KAA, Eagle KA, Kennelly BM, for the GRACE
Investigators. Access to catheterisation facilities in patients admitted with
acute coronary syndrome: multinational registry study. BMJ. 2005;
330:441.
15. Karmpaliotis D, Turakhia MP, Kirtane A, Murphy SA, Kosmidou I,
Morrow DA, Giugliano RP, Cannon CP, Antman EM, Braunwald E,
Gibson CM. Sequential risk stratification using TIMI risk score and TIMI
flow grade among patients treated with fibrinolytic therapy for
ST-segment elevation acute myocardial infarction. Am J Cardiol. 2004;
94:1113–1117.
Valvular Heart Disease
New Locus for Autosomal Dominant Mitral Valve Prolapse
on Chromosome 13
Clinical Insights From Genetic Studies
Francesca Nesta, MD*; Maire Leyne, MS*; Chaim Yosefy, MD; Charles Simpson, BS; Daisy Dai, BS;
Jane E. Marshall, RDCS; Judy Hung, MD; Susan A. Slaugenhaupt, PhD†; Robert A. Levine, MD†
Background—Mitral valve prolapse (MVP) is a common disorder associated with mitral regurgitation, endocarditis, heart
failure, and sudden death. To date, 2 MVP loci have been described, but the defective genes have yet to be discovered.
In the present study, we analyzed a large family segregating MVP, and identified a new locus, MMVP3. This study and
others have enabled us to explore mitral valve morphological variations of currently uncertain clinical significance.
Methods and Results—Echocardiograms and blood samples were obtained from 43 individuals who were classified by the
extent and pattern of displacement. Genotypic analyses were performed with polymorphic microsatellite markers.
Evidence of linkage was obtained on chromosome 13q31.3-q32.1, with a peak nonparametric linkage score of 18.41
(P⬍0.0007). Multipoint parametric analysis gave a logarithm of odds score of 3.17 at marker D13S132. Of the 6 related
individuals with mitral valve morphologies not meeting diagnostic criteria but resembling fully developed forms, 5
carried all or part of the haplotype linked to MVP.
Conclusions—The mapping of a new MVP locus to chromosome 13 confirms the observed genetic heterogeneity and
represents an important step toward gene identification. Furthermore, the genetic analysis provides clinical lessons with
regard to previously nondiagnostic morphologies. In the familial context, these may represent early expression in gene
carriers. Early recognition of gene carriers could potentially enhance the clinical evaluation of patients at risk of full
expression, with the ultimate aim of developing interventions to reduce progression. (Circulation. 2005;112:20222030.)
Key Words: echocardiography 䡲 genetics 䡲 mitral valve
M
been related to lack of systematic examination of the entire
human genome and uncertainty of phenotypic diagnosis.
More recently, understanding of mitral valve shape has
improved specificity of echocardiographic diagnosis16 –20 as
the basis for genetic studies.
Accordingly, linkage of myxomatous MVP to chromosome
16 (MMVP1) was reported in 2 of 4 families studied through
the use of current diagnostic criteria and a conservative model
of disease inheritance.21 We have also previously reported
linkage of an MVP locus, MMVP2, on chromosome 11p15.4
in a single large pedigree.22 The current diagnostic approach
has also revealed a X-linked form of MVP.23 Together, these
studies demonstrate the power of the phenotyping and confirm the genetic heterogeneity of this common disorder.
These findings suggest the hypothesis that MVP may be
the final common outcome resulting from one of multiple
itral valve prolapse (MVP) is a common disorder that
exhibits a strong hereditary component. It occurs in
⬇2.4% of the general population.1,2 Patients exhibit fibromyxomatous changes in the mitral leaflet tissue that cause
superior displacement of the leaflets into the left atrium.2– 4
MVP can be associated with significant mitral regurgitation
(MR), bacterial endocarditis, congestive heart failure, and
even sudden death,5– 8 and it is the most common primary
cause of isolated MR requiring surgical repair.9
See p 1924
Although autosomal dominant inheritance has been described for MVP10,11 and MVP occurs in connective tissue
disorders such as Marfan syndrome,12,13 previous studies
have failed to establish linkage of familial MVP with fibrillar
collagen genes.14,15 Prior negative linkage results may have
Received October 26, 2004; revision received June 3, 2005; accepted June 14, 2005.
From the Cardiac Ultrasound Laboratory, Massachusetts General Hospital, Department of Medicine, Harvard Medical School (F.N., C.Y., J.E.M, J.H.,
R.A.L.), and Center for Human Genetic Research, Massachusetts General Hospital, Harvard Medical School (M.L., C.S., D.D., S.A.S.), Boston, Mass.
*Authors Nesta and Leyne contributed equally to the work.
†Senior authorship is acknowledged for Drs Levine and Slaugenhaupt to reflect this cross-disciplinary collaboration.
This work was presented at the 2004 American Heart Association Scientific Sessions, New Orleans, La, November 7–10, 2004, as a finalist for the
Samuel A. Levine Young Clinical Investigator Award, and published in abstract form (Circulation. 2004;110[suppl III]:III-335).
Correspondence to Robert A. Levine, MD, Massachusetts General Hospital, Cardiac Ultrasound Laboratory, 55 Fruit St YWK5068, Boston, MA 02114.
E-mail [email protected]
2022
Nesta et al
genetic defects, analogous to familial hypertrophic cardiomyopathy.24,25 Finding additional loci containing diverse but
functionally related genes could provide helpful clues for
gene identification and increase our understanding of the
pathogenesis, with the ultimate goal of developing targeted
therapies. Therefore, the aim of the present study was to
search for a new MVP locus by genome-wide scanning in a
single large pedigree. Studying familial MVP can also provide unique insights into clinical findings of currently uncertain significance by testing for genetic linkage of individuals
with variations of mitral valve morphology that do not meet
standard criteria but resemble those in fully affected family
members—possible early forms that could ultimately guide
interventions to limit progression.
Methods
Pedigree Collection
This study was carried out on a pedigree of 46 individuals with living
members in 3 generations. Echocardiograms and blood samples were
obtained on 43 members of the pedigree. The proband was a
physician self-referred for family analysis of MVP. This family was
selected because of its substantial size, the number of members with
fully diagnostic MVP, and the absence of Marfan features.
Clinical Evaluation
A detailed medical history was obtained from each family member to
address the following: (1) evidence of Marfan syndrome or other
connective tissue disorders; (2) history of panic attacks with anticipatory anxiety or fear of additional attacks; (3) thoracic cage
deformities; (4) symptoms of chest pain, shortness of breath, and
palpitations; (5) progression of MR, rupture of chordae tendineae,
and surgical valve repair; (6) history of sudden death, with or without
resuscitation; and (7) history of congenital heart disease, cardiomyopathy, or coronary heart disease. Marfan syndrome was defined by
the presence of joint laxity, tall body habitus with long limbs relative
to torso, aortic dissection and aneurysm, scoliosis, and ocular
abnormalities and was excluded by detailed history, visual assessment of body habitus, and echocardiographic evaluation of the aorta.
Data Acquisition
Blood samples were collected on 43 of the 46 family members at the
time of echocardiography. Transformed lymphoblast cell lines were
established for those with confirmed MVP, and DNA was extracted
directly from blood for all others. Complete 2D and Doppler
echocardiograms were recorded with a 2.5- to 5.0-MHz transducer to
optimize resolution. MVP was diagnosed in long-axis views that
contain the highest annular points,16 –20,26 and the medial, central, and
lateral valve scallops were scanned systematically to measure maximal systolic leaflet displacement beyond the annulus. Subjects with
both thickened (⬎5 mm) and relatively thinner leaflets were considered to have fully diagnostic MVP because both occur within the
same pedigrees.21,27 Because the lateral scallop is difficult to
evaluate from long-axis views, its displacement was measured in the
apical 4-chamber view but always confirmed in the long-axis
scans.17,28 Thickness of the leaflet midportion was examined during
diastasis, excluding focal thickening and chordae.17,29 All studies had
Institutional Review Board approval with written informed consent.
Echocardiographic Classification
Echocardiographic classification was performed before any genetic
analysis. On the basis of prior clinical and prognostic studies, classic
MVP is diagnosed if leaflet displacement exceeds 2 mm and
maximal thickness is ⱖ5 mm; MVP is considered nonclassic if
displacement exceeds 2 mm but maximal thickness is
⬍5 mm.17,19,20,26,30 In the genetic studies, both of these fully
diagnostic forms of MVP (classic and nonclassic) were considered
New Locus for MVP on Chromosome 13
2023
affected. Nondiagnostic forms of uncertain clinical importance were
described from the common feature of posterior leaflet asymmetry,
which is frequent in fully diagnostic MVP members. Subjects with
borderline degrees of displacement (ⱕ2 mm involving the posterior
leaflet and not associated in a prior preliminary report with increased
leaflet thickness, MR, left atrial enlargement, or valve-related complications)31 were designated as having “minimal systolic displacement,” and these 6 individuals (12769, 12277, 12188, 13549, 12270,
and 12191) were considered unknown rather than unaffected for the
genetic analysis. This method allows for the possibility that in some
instances minimal displacement may represent a mild form of
expression as opposed to a physiological variant of leaflet position.
We also recognized an interesting prodromal morphology after
reviewing this and other families collected as part of our ongoing
genetic studies of MVP. These subjects do not have diagnostic leaflet
displacement beyond the annulus, but their pattern of leaflet closure
or coaptation resembles that of other family members with fully
expressed MVP. Normally, the leaflets meet posteriorly within the
LV cavity because the posterior leaflet is shorter than the anterior
(Figure 1A). In patients with MVP, coaptation is typically displaced
anteriorly, consistent with elongation of the posterior leaflet, which
can produce excessive leaflet motion not only into the left atrium but
also toward the aortic root. Compare Figure 1C, in which classic
MVP leaflets meet halfway up the dotted annular line, with Figure
1B, which shows a subject with no displacement of leaflets into the
left atrium beyond the annulus but with an anterior shift of the
coaptation point. This shift has been correlated quantitatively with
posterior leaflet length (see Discussion). We therefore suggest that
this pattern may represent an early or prodromal manifestation of
familial MVP without diagnostic leaflet displacement into the atrium
but with 2 salient features: anterior displacement of the coaptation
point ⬎40% anteriorly along the mitral annulus (P/D; Figure 1;
normally within the posterior 25% to 30% of the mitral annulus; see
Discussion)32 and a leaflet coaptation pattern similar to that seen in
fully expressing family members. This pattern of bulging of the
posterior leaflet relative to the anterior, which is seen in all patients
with posterior leaflet prolapse and many with bileaflet MVP (Figure
2), was seen in 2 members of this pedigree. Because of the striking
similarity with fully diagnostic MVP, the prodromal individuals
(12768 and 12278) were coded as affected for the genetic analysis.
Genome Scan and Linkage Analysis
Before beginning the genome scan, we used the SLINK33,34 program
to verify that the pedigree had sufficient power to detect linkage. To
determine whether the family was linked to the previously described
MMVP1 or MMVP2 loci, we genotyped a subset of the family for the
following markers: MMVP1, D16S404-D16S3103-D16S420D16S3133-D16S3068-D16S3080-D16S515; and MMVP2, D11S4046D11S4124-D11S2349-D11S1338-D11S1331-D11S932-D11S4465D11S1349-D11S902-D11S1359-D11S904-D11S914-D11S935D11S905. Because no evidence for linkage was discovered, the
genome scan was performed on 14 family members (identified in
Figure 3) using a panel of 382 genetic markers that span the entire
human genome at approximately 10-cM intervals. The markers make
up the MGH Genomics Core Facility linkage panel, the majority of
which are from the ABI Prism Linkage Mapping set, version 2.5
(Perkin-Elmer, Applied Biosystems). The average heterozygosity of
these markers is 0.79. Specific allele frequencies are available at
http://www.appliedbiosystems.com. When additional map resolution
was needed, markers were added from the Cooperative Human Linkage
Center Weber Human Screening Set, version 8 (Research Genetics).
When available, marker distances were obtained from the Marshfield
sex-averaged genetic map (http://research.marshfieldclinic.org). Physical location was used to estimate close genetic distances when markers
were not on the available linkage map. In most instances other than
an X-linked form,23 familial MVP appears to segregate as an
autosomal dominant trait with decreased penetrance.10,11,21,22 However, we acknowledge that the true genetic model of MVP is
unknown; therefore, we initially performed nonparametric linkage
(NPL) analysis using the GENEHUNTER program (Sall scoring
function).35,36 This type of analysis examines allele sharing among
2024
Circulation
September 27, 2005
Figure 1. Echo examples of posteriorly coapting leaflets (anterior leaflet [AL]; posterior leaflet [PL]) in a normal subject (A) vs increased
coaptation height in a family member with a prodromal form and an elongated posterior leaflet (B) and in another family member with
classic MVP of both leaflets into the left atrium (LA) (C). D, Schematics showing projections of anterior (A) and posterior (P) leaflets onto
the mitral annular diameter (D). C indicates projection of the coaptation point onto the LV internal diameter [LVID]; AO, aorta; and RV,
right ventricle.
affected individuals and does not require specification of a genetic
model. Therefore, NPL analysis can demonstrate phenotype-allele
associations that may be missed by parametric analysis performed
with an incorrect model. The GENEHUNTER program limits the
pedigree size by using a specific formula whereby 2n⫺f must be
ⱕ20 (n is nonfounders and f is founders). GENEHUNTER performs
trimming of pedigrees exceeding this size as described in the
program documentation. After analyzing the genome scan data, we
genotyped all individuals for markers on chromosome 13 (Figure 3).
To perform genetic analysis on the entire family, we also calculated
2-point parametric logarithm of odds (LOD) scores between the
disease and individual markers using the MLINK program of
FASTLINK 3.0, a faster version of the original LINKAGE package.37– 41 In addition, multipoint parametric LOD scores for the entire
family were calculated with LINKMAP.42 Because MVP has been
associated with both sex- and age-dependent penetrance,10,11 our
Figure 2. Examples of 2 individuals with prodromal morphology (A, B) and of an individual with posterior leaflet MVP (C). All show
increased coaptation heights and posterior leaflet bulging (arrows) relative to the anterior leaflet, but only the third example (C) shows
fully expressed superior leaflet displacement relative to the mitral annulus (dotted line) into the left atrium.
Nesta et al
2025
Figure 3. MVP pedigree showing chromosome 13 haplotypes. *Individuals used for the genome scan and the GENEHUNTER analysis.
Physical and Transcript Maps
analysis was performed with the model described in the previous
linkage reports.21,22 Briefly, we assumed an autosomal dominant
mode of inheritance with incomplete penetrance and a disease gene
frequency of 0.005, with a phenocopy rate of 1% to account for the
high incidence of sporadic MVP. Penetrance for adults ⬎15 years of
age was set at 95% for female subjects and 63% for male subjects
and at 32% and 21%, respectively, for those ⬍15 years of age. To
prove that our linkage findings were robust given the assumed
genetic model, we also used a stringent model in which we excluded
all unaffected subjects ⬍40 years of age and assumed complete
penetrance of the disease with no phenocopies. Finally, to overcome
the limitations on family size in GENEHUNTER and on the number
of markers used in LINKMAP, we performed SIMWALK analysis,
which permitted evaluation of the entire pedigree using all 31
markers on chromosome 13. Haplotypes across the linked region
were constructed manually and confirmed with both GENEHUNTER and SIMWALK.
We constructed a physical and transcript map for the new MMVP3
locus using data from the UCSC Human Genome Browser (May
2004 freeze) (http://genome.ucsc.edu/cgi-bin/hgGateway).
Results
The complete pedigree used in the present study is shown in
Figure 3. Both founders were of Western European descent.
Blood and echocardiograms were obtained in 43 of 46
subjects (19 male and 24 female subjects; age, 7 to 75 years);
the 2 founders were deceased, and subject 203 did not
participate. The echocardiographic characteristics of the 9
patients meeting full clinical diagnostic criteria for MVP are
provided in the Table. Two of these 9 individuals had
Echocardiogram Characteristics of Pedigree Members With Fully Diagnostic MVP
ID
12183
Age, y
Sex
MVP
Leaflet Thickening
LA, mm
LVIDd, mm
EF, %
MR
77
F
Bileaflet (p⬎a)
Yes
40
46
53
Moderate
12591
75
M
Bileaflet (p⬎a)
Yes
43*
50*
54*
Severe
12766
69
M
Bileaflet (p⬎a)
Yes
35
42
64
Mild
12187
53
M
Bileaflet
Yes
42
46
65
Mild
12189
50
F
Posterior
No
35
36
74
Mild
12166
45
M
Bileaflet
Yes
33
47
60
Mild
12773
43
M
Bileaflet
Yes
35
43
64
Trace
12273
41
F
Bileaflet
Yes
38
46
66
Mild
12184
39
F
Bileaflet (p⬎a)
Yes
33
49
73
Trace
LA indicates left atrial diameter; LVIDd, left ventricular internal diameter (diastolic); EF, ejection fraction; and p⬎a, asymmetric
posterior greater than anterior leaflet displacement. Subject 12591 had mitral valve replacement for bileaflet MVP with severe MR.
*Values are postoperative.
2026
Circulation
September 27, 2005
Figure 4. Nonparametric GENEHUNTER analysis of chromosome 13 after the genome scan.
moderate to severe MR, 1 had ruptured chordae tendineae
requiring surgical intervention, and 0 had a history of endocarditis or sudden death. In the entire pedigree, no extracardiac manifestations of connective tissue abnormalities or
Marfan syndrome were present in any family member. Four
subjects, 2 with MVP and 2 without, had a history of panic
attacks. Three members with fully diagnostic MVP had a
combination of chest pain, shortness of breath, and palpitations; of these, only 1 had an ECG diagnosis of atrial
fibrillation. No individuals had thoracic cage deformities.
One nonaffected individual had a bicuspid aortic valve, and
no family member had a history of cardiomyopathy or
coronary heart disease.
Morphological Heterogeneity
Review and comparison of echo images from multiple family
members revealed a wide spectrum of phenotypic morphologies. Of the 43 individuals in our pedigree, 9 had fully
diagnostic MVP but had varying leaflet involvement, with 5
of the 9 having asymmetric prolapse of the posterior leaflet
beyond the anterior, a common pattern in MVP.17,20,43– 45
Leaflet thickening and degree of MR varied, as first described
within families by Zuppiroli et al.27 Six related individuals
were designated as having forms not meeting current diagnostic criteria: 2 with the prodromal morphology and 4 with
minimal displacement. In addition, 2 spouses in the second
generation also had minimal displacement. All of these
individuals shared an asymmetry of coaptation (posterior
leaflet beyond anterior), as did most of the fully diagnostic
subjects. This asymmetry was often reflected in an eccentric,
anteriorly directed MR jet,43 strikingly similar, for example,
in a fully diagnostic mother (12184) and her daughter (12191)
who had minimal displacement and trace but atypically
eccentric MR. In the prodromals, coaptation was displaced
anteriorly to a point 50% up the annular diameter (versus the
normal posterior location, only 25% up the annulus).
Genome Scan and Linkage Analysis
SLINK analysis performed with our previously described
model22 predicted that the pedigree had ⬇50% power to
detect an LOD score of 2.0, which would provide evidence
suggestive of linkage. The maximum predicted LOD score
obtained in a sample of 500 replicates was 4.62; the average
LOD score was 2.09. Therefore, we performed a genome scan
using 14 individuals, including 9 with fully diagnostic MVPs,
2 prodromals, 1 with minimal systolic displacement, and 2
unaffected individuals (parents of generation 3) (identified in
Figure 3). Inspection of the GENEHUNTER results of the
initial genome scan yielded 4 regions with NPL scores ⬎2.0
and values of P⬍0.05 on chromosomes 4, 11 (58 cM from
MMVP2), 13, and 18. The highest scores were obtained on
chromosome 13 with D13S170 (NPL⫽3.04; P⬍0.01) and
D13S265 (NPL, 6.62; P⬍0.004) (Figure 4). Given that the
best evidence for linkage was on chromosome 13, we
searched the genetic marker maps and genotyped the entire
family using several markers surrounding D13S265. GENEHUNTER analysis of the 14-member pedigree using the
additional markers yielded a peak NPL score of 18.41
(P⬍0.0007) across a 5.0-cM region between D13S886 and
D13S309 (Figure 5), with a corresponding parametric LOD
Nesta et al
2027
Figure 5. NPL scores on chromosome 13 after the addition of several markers to the linked interval.
score of 2.44. The maximum 2-point parametric LOD score
obtained with FASTLINK on the entire family was 2.81 with
the marker D13S1490, and the parametric multipoint LOD
score of 3.17 on the entire family was achieved with LINKMAP with the markers D13S886, D13S129, and D13S132. To
fully use all family and marker information, we then performed SIMWALK analysis for all 31 markers on chromosome 13. This analysis confirms the significance of the
GENEHUNTER and FASTLINK results. The NPL peaks at
the same location as GENEHUNTER, with nearly identical
probability values (P⫽0.0006 versus 0.0007). Similarly,
good agreement was observed in the parametric analysis, with
a FASTLINK score of 3.17 and a SIMWALK score of 2.996.
To confirm that our linkage findings were robust to model
assumptions, we performed parametric and nonparametric
analysis using a stringent model of the disease, as described
in Methods. The maximum 2-point parametric LOD score
obtained with FASTLINK was 2.22 at marker D13S132.
GENEHUNTER analysis of the stringent pedigree yielded an
identical NPL score of 18.41 (P⬍0.0007).
These results, combined with haplotype analysis in this
family, confirm linkage of MMVP3 to an 8.61-cM region on
the long arm of chromosome 13 (Figure 3). All of the fully
diagnostic MVP and prodromal members in this family share
a 12-allele core haplotype for the markers D13S265 through
D13S892. Five unaffected individuals (12772, 14216, 12776,
12775, and 12276) were nonexpressing carriers of the haplotype, 3 of whom were ⬍15 years of age and the other 2
were 30 and 36 years of age. This is consistent with a model
of age-dependent penetrance as observed in the previous
family studies.21,22 Of the 4 related individuals with minimal
systolic displacement, 2 had the complete haplotype (12270
and 12191), and 1 (13549) carried the disease haplotype for
the proximal 3 markers.
A recombination event in individual 12184 between markers D13S794 and D13S265 defines the proximal boundary of
the linked region, whereas a recombination event in individual 12591 between markers D13S892 and D13S786 defines
the distal boundary. The complete disease haplotype and the
locations of the proximal and distal crosses that define the
8.2-Mb candidate interval are shown in Figure 6. Our results
confirm that a third MVP locus, MMVP3, maps between
D13S794 and D13S786 on chromosome 13q31.3-q32.1. The
current transcript map for the 8.2-Mb candidate region
contains 16 genes and shows synteny to mouse chromosome
14 (Figure 7).
Discussion
This analysis demonstrates that a new locus for autosomal
dominant MVP (MMVP3) maps to the long arm of chromosome 13. This finding further confirms the genetic heterogeneity of MVP, previously linked to chromosomes 11 and 16
and the X chromosome.23 In contrast with prior negative
studies, identifying loci on 3 chromosomes demonstrates the
strength of the present approach, combining current diagnostic criteria with systematic genome scanning. Genetic hetero-
2028
Circulation
September 27, 2005
Figure 6. MMVP3 haplotype. All nine fully diagnostic MVP individuals shared the interval between D13S794 and D13S786.
*Noninformative.
geneity provides opportunities to explore relationships between genetic defects and differences in disease expression
and natural history,24,25,46 – 48 as well as providing helpful
clues for gene searches. The genetic analysis has, in turn,
provided important clinical insights, revealing a spectrum of
expression that included valve morphologies previously considered normal variants but now for the first time recognized
as having the same genetic substrate in the familial context.
The clinical lessons learned during this and other recent
genetic studies challenge the concept that MVP has a consis-
tent expression and leaflet thickness within families.21,22,27
Although thick leaflets and MR are associated in individual
patients,20,26 a spectrum of valvular abnormalities, which may
represent variations in disease expression, stage of progression, or modifying factors, occurs within families. This
spectrum also includes family members with minimal displacement or the described prodromal morphology who were
frequently found in this study to carry all or part of the MVP
haplotype (5 of 6 individuals). These may represent either
mild or early gene expression, a distinction that requires
follow-up studies. Recognizing early forms is important
because the disease often manifests clinically in the fifth or
sixth decade of life as a severe cardiac event. Earlier targeted
intervention to reduce leaflet stresses in genetically susceptible individuals,49 as in Marfan syndrome with aortic dilatation,50 could potentially prevent progression to complications
and heart failure.
The recognized prodromal morphology, previously unreported, was also observed in the family linked to the MMVP2
locus on chromosome 11.22 When we reviewed all echocardiograms in that family blinded to haplotype, we discovered
5 individuals with a prodromal morphology who turned out to
be carriers of the haplotype, as did another with minimal
systolic displacement. In the familial context, therefore, the
prodromal finding could acquire diagnostic power. This is
reasonable because the salient feature of this morphology,
anteriorly shifted coaptation, has been associated with increased posterior leaflet length. This association has been
recognized during surgical repair of MVP patients with long
posterior leaflets who are more prone to having their coapted
leaflets shift anteriorly and obstruct the LV outflow tract,51
reducible by Carpentier’s “sliding” of the posterior leaflet
downward.52 Quantitatively, we have found that the height of
coaptation relative to the annulus or LV diameter (P/D or
C/LVID in Figure 1; see legend for abbreviation expansion)
correlated well with the ratio of anterior to posterior leaflet
length (r⫽0.83 to 0.85) in the chromosome 11 family.32
From these findings, minimal displacement can no longer
simply be considered a normal variant in the familial context.
It shares posterior leaflet asymmetry with the prodromal form
and many of those with fully expressed MVP. Posterior
Figure 7. Human transcript map of the MMVP3 candidate region on chromosome 13. The candidate region is within 13q31.3-q32.1,
and all RefSeq genes and their orientation are shown within the 8.2-Mb interval.
Nesta et al
leaflet asymmetry has a recognized role in the mechanism of
MR20,43,44 and the definition of MVP.45 These considerations
support our retention of such individuals as indeterminate as
opposed to unaffected for the genetic analysis. The genetic
studies will therefore be important to provide insights into the
best clinical approach to individuals with such previously
nondiagnostic features.
The association between nondiagnostic forms and MVP
loci cannot be extrapolated beyond the context of familial
MVP, eg, 2 individuals with minimal displacement who
married into the family. However, as in hypertrophic cardiomyopathy where the distinction between normal variation and
pathological hypertrophy must be made in genetic studies, the
familial context permits the use of more sensitive criteria
without sacrificing specificity.25,48 Follow-up studies are
required to determine whether these nondiagnostic forms
progress and what factors correlate with progression.
The current transcript map for the 8.2-Mb MMVP3 candidate region on chromosome 13q31.3-q32.1 contains 16 genes
and shows synteny to mouse chromosome 14. Although we
have only recently started investigating the potential function
of the genes in the region, a few merit consideration as
potential candidates. Intimal thickness–related receptor (ITR)
has been isolated from a heart cDNA library. It contains an
N-terminal signal sequence, 7 transmembrane domains, and a
signature motif found in members of the rhodopsin-like G
protein– coupled receptor superfamily. ITR-null mice suggest
that this gene plays an important role in the regulation of
vascular remodeling.53 Glypican 5 and glypican 6 (GPC5 and
GPC6) are members of a family of cell surface heparan
sulfate proteoglycans that appear to play an important role in
cellular growth control and differentiation. GPC6 has been
localized to mesenchymal tissues in the developing mouse
embryo.54 –56 Interestingly, myxomatous valves are known to
contain significantly more glycosaminoglycans than control
valves.57 These data suggest that these genes should be given
high priority for screening.
In summary, this analysis demonstrates that a third locus
for autosomal dominant MVP maps to an 8.2-Mb region on
chromosome 13. It further confirms the genetic heterogeneity
of MVP and represents an important step toward the identification of MVP genes. Furthermore, in the familial context,
the genetic analysis shows that previously nondiagnostic
morphologies often represent mild or early stages of expression in gene carriers; this early recognition could potentially
enhance our clinical evaluation, with the ultimate aim of
developing interventions to limit progression.
Acknowledgments
This work was funded by grants from the Doris Duke Foundation
and the Aetna Foundation, by an American Heart Association
Postdoctoral Research Fellowship (Dr Nesta), and by NIH grants
R01-HL-38176 and K24-HL-67434. We thank Dr Emelia Benjamin
for referring the proband.
References
1. Freed LA, Benjamin EJ, Levy D, Larson MG, Evans JC, Fuller DL,
Lehman B, Levine RA. Mitral valve prolapse in the general population:
the benign nature of echocardiographic features in the Framingham Heart
Study. J Am Coll Cardiol. 2002;40:1298 –1304.
2029
2. Freed LA, Levy D, Levine RA, Larson MG, Evans JC, Fuller DL,
Lehman B, Benjamin EJ. Prevalence and clinical outcome of mitral-valve
3. Tamura K, Fukuda Y, Ishizaki M, Masuda Y, Yamanaka N, Ferrans VJ.
Abnormalities in elastic fibers and other connective-tissue components of
floppy mitral valve. Am Heart J. 1995;129:1149 –1158.
4. Rabkin E, Aikawa M, Stone JR, Fukumoto Y, Libby P, Schoen FJ.
Activated interstitial myofibroblasts express catabolic enzymes and
mediate matrix remodeling in myxomatous heart valves. Circulation.
2001;104:2525–2532.
5. Devereux RB, Kramer-Fox R, Shear MK, Kligfield P, Pini R, Savage DD.
Diagnosis and classification of severity of mitral valve prolapse: methodologic, biologic, and prognostic considerations. Am Heart J. 1987;113:
1265–1280.
6. Braunwald E. Heart Disease: A Textbook of Cardiovascular Medicine.
Philadelphia, Pa: WB Saunders Co; 1992.
7. Avierinos JF, Gersh BJ, Melton LJ 3rd, Bailey KR, Shub C, Nishimura
RA, Tajik AJ, Enriquez-Sarano M. Natural history of asymptomatic
mitral valve prolapse in the community. Circulation. 2002;106:
1355–1361.
8. Perloff JK, Child JS. Clinical and epidemiologic issues in mitral valve
prolapse: overview and perspective. Am Heart J. 1987;113:1324 –1332.
9. Waller BF, Morrow AG, Maron BJ, Del Negro AA, Kent KM, McGrath
FJ, Wallace RB, McIntosh CL, Roberts WC. Etiology of clinically
isolated, severe, chronic, pure mitral regurgitation: analysis of 97 patients
over 30 years of age having mitral valve replacement. Am Heart J.
1982;104:276 –288.
10. Devereux RB, Brown WT, Kramer-Fox R, Sachs I. Inheritance of mitral
valve prolapse: effect of age and sex on gene expression. Ann Intern Med.
1982;97:826 – 832.
11. Strahan NV, Murphy EA, Fortuin NJ, Come PC, Humphries JO. Inheritance of the mitral valve prolapse syndrome: discussion of a threedimensional penetrance model. Am J Med. 1983;74:967–972.
12. Glesby MJ, Pyeritz RE. Association of mitral valve prolapse and systemic
abnormalities of connective tissue: a phenotypic continuum. JAMA. 1989;
262:523–528.
13. Roman MJ, Devereux RB, Kramer-Fox R, Spitzer MC. Comparison of
cardiovascular and skeletal features of primary mitral valve prolapse and
Marfan syndrome. Am J Cardiol. 1989;63:317–321.
14. Henney AM, Tsipouras P, Schwartz RC, Child AH, Devereux RB, Leech
GJ. Genetic evidence that mutations in the COL1A1, COL1A2, COL3A1,
or COL5A2 collagen genes are not responsible for mitral valve prolapse.
Br Heart J. 1989;61:292–299.
15. Wordsworth P, Ogilvie D, Akhras F, Jackson G, Sykes B. Genetic
segregation analysis of familial mitral valve prolapse shows no linkage to
fibrillar collagen genes. Br Heart J. 1989;61:300 –306.
16. Levine RA, Triulzi MO, Harrigan P, Weyman AE. The relationship of
mitral annular shape to the diagnosis of mitral valve prolapse. Circulation. 1987;75:756 –767.
17. Levine RA, Stathogiannis E, Newell JB, Harrigan P, Weyman AE.
Reconsideration of echocardiographic standards for mitral valve prolapse:
lack of association between leaflet displacement isolated to the apical four
chamber view and independent echocardiographic evidence of abnormality. J Am Coll Cardiol. 1988;11:1010 –1019.
18. Levine RA, Handschumacher MD, Sanfilippo AJ, Hagege AA, Harrigan
P, Marshall JE, Weyman AE. Three-dimensional echocardiographic
reconstruction of the mitral valve, with implications for the diagnosis of
mitral valve prolapse. Circulation. 1989;80:589 –598.
19. Perloff JK, Child JS. Mitral valve prolapse: evolution and refinement of
diagnostic techniques. Circulation. 1989;80:710 –711.
20. Nidorf SM, Weyman AE, Hennessey R, Newell JB, Levine AB. The
relationship between mitral valve morphology and prognosis in patients
with mitral valve prolapse: a prospective echocardiographic study of 568
patients. J Am Soc Echocardiogr. 1993;6:S8. Abstract.
21. Disse S, Abergel E, Berrebi A, Houot AM, Le Heuzey JY, Diebold B,
Guize L, Carpentier A, Corvol P, Jeunemaitre X. Mapping of a first locus
for autosomal dominant myxomatous mitral-valve prolapse to chromosome 16p11.2-p12.1. Am J Hum Genet. 1999;65:1242–1251.
22. Freed LA, Acierno JS Jr, Dai D, Leyne M, Marshall JE, Nesta F, Levine
RA, Slaugenhaupt SA. A locus for autosomal dominant mitral valve
prolapse on chromosome 11p15.4. Am J Hum Genet. 2003;72:
1551–1559.
23. Kyndt F, Schott JJ, Trochu JN, Baranger F, Herbert O, Scott V,
Fressinaud E, David A, Moisan JP, Bouhour JB, Le Marec H, Benichou
2030
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
Circulation
September 27, 2005
B. Mapping of X-linked myxomatous valvular dystrophy to chromosome
Xq28. Am J Hum Genet. 1998;62:627– 632.
Solomon SD, Jarcho JA, McKenna W, Geisterfer-Lowrance A, Germain
R, Salerni R, Seidman JG, Seidman CE. Familial hypertrophic cardiomyopathy is a genetically heterogeneous disease. J Clin Invest. 1990;86:
993–999.
Maron BJ, Moller JH, Seidman CE, Vincent GM, Dietz HC, Moss AJ,
Towbin JA, Sondheimer HM, Pyeritz RE, McGee G, Epstein AE. Impact
of laboratory molecular diagnosis on contemporary diagnostic criteria for
genetically transmitted cardiovascular diseases: hypertrophic cardiomyopathy, long-QT syndrome, and Marfan syndrome: a statement for
healthcare professionals from the Councils on Clinical Cardiology, Cardiovascular Disease in the Young, and Basic Science, American Heart
Association. Circulation. 1998;98:1460 –1471.
Marks AR, Choong CY, Sanfilippo AJ, Ferre M, Weyman AE. Identification of high-risk and low-risk subgroups of patients with mitral-valve
Zuppiroli A, Roman MJ, O’Grady M, Devereux RB. A family study of
anterior mitral leaflet thickness and mitral valve prolapse. Am J Cardiol.
1998;82:823– 826, A10.
Shah PM. Echocardiographic diagnosis of mitral valve prolapse. J Am
Soc Echocardiogr. 1994;7:286 –293.
Weissman NJ, Pini R, Roman MJ, Kramer-Fox R, Andersen HS,
Devereux RB. In vivo mitral valve morphology and motion in mitral
valve prolapse. Am J Cardiol. 1994;73:1080 –1088.
Nishimura RA, McGoon MD, Shub C, Miller FA Jr, Ilstrup DM, Tajik
AJ. Echocardiographically documented mitral-valve prolapse: long-term
follow-up of 237 patients. N Engl J Med. 1985;313:1305–1309.
Vivaldi MT, Sagie A, Adams MS, Levine RA. 10-Year echocardiographic and clinical follow-up of patients with nonclassic mitral valve prolapse:
does it progress? Circulation. 1994;90(suppl I):I-222. Abstract.
Nesta F, Freed L, Leyne M, Marshall J, Slagenhaupt S, Levine R.
Coaptation height: a new marker of mitral leaflet redundancy. J Am Soc
Echocardiogr. 2003;16:538. Abstract.
Ott J. Computer-simulation methods in human linkage analysis. Proc Natl
Acad Sci U S A. 1989;86:4175– 4178.
Weeks DE, Ott J, Lathrop GM. SLINK: a general simulation program for
linkage analysis. Am J Hum Genet. 1990;47(suppl):A204. Abstract.
Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES. Parametric and
nonparametric linkage analysis: a unified multipoint approach. Am J Hum
Genet. 1996;58:1347–1363.
Kong A, Cox NJ. Allele-sharing models: LOD scores and accurate
linkage tests. Am J Hum Genet. 1997;61:1179 –1188.
Cottingham RW Jr, Idury RM, Schaffer AA. Faster sequential genetic
linkage computations. Am J Hum Genet. 1993;53:252–263.
Lathrop GM, Lalouel JM, White RL. Construction of human linkage
maps: likelihood calculations for multilocus linkage analysis. Genet Epidemiol. 1986;3:39 –52.
Shaffer AA, Gupta SK, Shriram K, Cottingham RW Jr. Avoiding recomputation in linkage analysis. Hum Hered. 1994;44:225–237.
Lathrop GM, Lalouel JM, Julier C, Ott J. Strategies for multilocus linkage
analysis in humans. Proc Natl Acad Sci U S A. 1984;81:3443–3446.
Lathrop GM, Lalouel JM. Easy calculations of LOD scores and genetic
risks on small computers. Am J Hum Genet. 1984;36:460 – 465.
Lathrop GM, Lalouel JM, Julier C, Ott J. Multilocus linkage analysis in
humans: detection of linkage and estimation of recombination. Am J Hum
Genet. 1985;37:482– 498.
43. Grayburn PA, Berk MR, Spain MG, Harrison MR, Smith MD, DeMaria
AN. Relation of echocardiographic morphology of the mitral apparatus to
mitral regurgitation in mitral valve prolapse: assessment by Doppler color
flow imaging. Am Heart J. 1990;119:1095–1102.
44. Schwammenthal E, He S, Fontaine AA, Nidorf SM, Vlahakes GJ,
Yoganathan AP, Weyman AE, Levine RA. Impact of leaflet geometry on
the severity of mitral regurgitation in mitral valve prolapse: insights from
an in vitro model. J Am Coll Cardiol. 1994;23:32A. Abstract.
45. Jyo Y, Yoshikawa J, Yoshida K, Kato H, Shakudo M. A new diagnostic
criteria of mitral valve prolapse syndrome. J Cardiol Suppl. 1988;18:
29 – 41.
46. Coonar AS, McKenna WJ. Molecular genetics of familial cardiomyopathies. Adv Genet. 1997;35:285–324.
47. Seidman CE, Seidman JG. Molecular genetic studies of familial hypertrophic cardiomyopathy. Basic Res Cardiol. 1998;93(suppl 3):13–16.
48. Tesson F, Richard P, Charron P, Mathieu B, Cruaud C, Carrier L,
Dubourg O, Lautie N, Desnos M, Millaire A, Isnard R, Hagege AA,
Bouhour JB, Bennaceur M, Hainque B, Guicheney P, Schwartz K,
Komajda M. Genotype-phenotype analysis in four families with
mutations in beta-myosin heavy chain gene responsible for familial
hypertrophic cardiomyopathy. Hum Mutat. 1998;12:385–392.
49. Durbin AD, Gotlieb AI. Advances towards understanding heart valve
response to injury. Cardiovasc Pathol. 2002;11:69 –77.
50. Shores J, Berger KR, Murphy EA, Pyeritz RE. Progression of aortic
dilatation and the benefit of long-term beta-adrenergic blockade in
Marfan’s syndrome. N Engl J Med. 1994;330:1335–1341.
51. Maslow AD, Regan MM, Haering JM, Johnson RG, Levine RA. Echocardiographic predictors of left ventricular outflow tract obstruction and
systolic anterior motion of the mitral valve after mitral valve reconstruction for myxomatous valve disease. J Am Coll Cardiol. 1999;34:
2096 –2104.
52. Jebara VA, Dervanian P, Acar C, Grare P, Mihaileanu S, Chauvaud S,
Fabiani JN, Deloche A, Carpentier A. Mitral valve repair using Carpentier
techniques in patients more than 70 years old: early and late results.
Circulation. 1992;86(suppl II):II-53–II-59.
53. Tsukada S, Iwai M, Nishiu J, Itoh M, Tomoike H, Horiuchi M, Nakamura
Y, Tanaka T. Inhibition of experimental intimal thickening in mice
lacking a novel G-protein– coupled receptor. Circulation. 2003;107:
313–319.
54. Paine-Saunders S, Viviano BL, Saunders S. GPC6, a novel member of the
glypican gene family, encodes a product structurally related to GPC4 and
is colocalized with GPC5 on human chromosome 13. Genomics. 1999;
57:455– 458.
55. Veugelers M, Vermeesch J, Reekmans G, Steinfeld R, Marynen P, David
G. Characterization of glypican-5 and chromosomal localization of
human GPC5, a new member of the glypican gene family. Genomics.
1997;40:24 –30.
56. Veugelers M, De Cat B, Ceulemans H, Bruystens AM, Coomans C, Durr
J, Vermeesch J, Marynen P, David G. Glypican-6, a new member of the
glypican family of cell surface heparan sulfate proteoglycans. J Biol
Chem. 1999;274:26968 –26977.
57. Grande-Allen KJ, Griffin BP, Calabro A, Ratliff NB, Cosgrove DM 3rd,
Vesely I. Myxomatous mitral valve chordae, II: selective elevation of
glycosaminoglycan content. J Heart Valve Dis. 2001;10:325–332; discussion 332–333.
Vascular Medicine
Targeting Adhesion Molecules as a Potential Mechanism of
Action for Intravenous Immunoglobulin
Varinder Gill, MSc; Christopher Doig, MSc, MD; Derrice Knight; Emma Love; Paul Kubes, PhD
Background—Intravenous immunoglobulin (IVIg) therapy has been shown to have therapeutic benefit in more than 50
inflammatory and immune-related diseases; however, the potential benefit of IVIg in cardiovascular disease is more
limited, in part because our understanding of the mechanisms underlying the effects of IVIg in innate immunity is
incomplete.
Methods and Results—In this study, a systematic assessment of the role of IVIg in leukocyte recruitment was completed
with an in vitro flow-chamber system and in vivo intravital microscopy in a feline ischemia-reperfusion model system.
IVIg treatment of blood resulted in a profound decrease in recruitment of either immobilized P-selectin or E-selectin due
to direct effects of IVIg on the leukocyte (not substratum). Similar results were observed on endothelium treated with
histamine, which induces P-selectin– dependent rolling and ␤2-integrin– dependent adhesion. IVIg reduced P-selectin
glycoprotein ligand-1 (PSGL-1) antibody binding to PSGL-1 on leukocytes. Use of a ␤2-integrin– dependent static assay
to bypass selectin-dependent recruitment revealed some inhibitory effectiveness (60%), which suggests that the majority
of the effects of IVIg were due to selectin inhibition, with some inhibition of integrin function. In vivo intravital
microscopy revealed a potent inhibitory effect of IVIg on P-selectin– dependent rolling and ␤2-integrin– dependent
adhesion that led to reduced leukocyte recruitment and vascular dysfunction in postischemic microvessels.
Conclusions—Our data demonstrate that IVIg has direct inhibitory effects on leukocyte recruitment in vitro and in vivo
through inhibition of selectin and integrin function. (Circulation. 2005;112:2031-2039.)
Key Words: endothelium 䡲 ischemia 䡲 reperfusion 䡲 leukocytes 䡲 immunoglobulin
I
ntravenous immunoglobulin (IVIg) is pooled IgG from
thousands of donors. IVIg has been used in the treatment of
many diseases, including a number of primary and secondary
antibody deficiencies, systemic lupus erythematosus, multiple sclerosis, and rheumatoid arthritis.1– 4 It has also been
reported to be beneficial in a number of inflammatory
conditions, including sepsis, systemic vasculitis, transplant
rejection, and Kawasaki’s disease.1–3,5–10 Although the underlying molecular mechanisms of these diseases are quite
different, IVIg appears to provide benefit in many of these
pathologies. To date, our understanding of the role of IVIg in
cardiovascular diseases such as ischemia-reperfusion injury is
lacking. This is not trivial, because many cardiovascular
diseases have an inflammatory component potentially amenable to IVIg treatment. However, not all patients respond
positively to IVIg, and in rare instances, severe complications
do arise.3 Improving our understanding of the mechanisms of
action of IVIg would greatly improve our insights as to which
disease states and which subsets of patients in a particular
disease should be treated with IVIg.
See p 1918
The molecular mechanisms by which IVIg may be effective include the modulation of Fc␥ receptor expression,
interference in the activation of the complement and cytokine
network, provision of anti-idiotypic antibodies, and effects on
the activation, differentiation, and effector functions of T
cells and B cells.1–5 A key feature of each of the aforementioned inflammatory diseases is leukocyte recruitment. Yet to
date, a systematic examination of the role of IVIg in the
cascade of molecular events involved in leukocyte recruitment has not been performed.
Leukocyte recruitment is a multistep process that initially
involves selectins expressed by both leukocytes (L-selectin)
and endothelium (P-selectin and E-selectin) and their respective ligands.11,12 These molecules allow leukocytes to first
tether and then roll along the endothelium, which will permit
the endothelium to present proinflammatory molecules such
as chemokines. Chemokines will cause activation of the
integrins on the leukocytes, which allows for firm adhesion.12
Once adherent, the leukocytes can emigrate from the vascu-
Received February 28, 2005; revision received June 12, 2005; accepted June 17, 2005.
From the Immunology Research Group, Department of Physiology and Biophysics (V.G., D.K., E.L., P.K.), Faculty of Medicine (C.D.), University
of Calgary, Calgary, Alberta, Canada.
The online-only Data Supplement can be found at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.105.546150/DC1.
Correspondence to Paul Kubes, Department of Physiology and Biophysics, Immunology Research Group, University of Calgary, Health Sciences
Centre, 3330 Hospital Dr NW, Calgary, Alberta, Canada T2N 4N1. E-mail [email protected]
2031
2032
Circulation
September 27, 2005
lature via a number of adhesion molecules, including platelet
and endothelial cell adhesion molecule-1 and CD99.11 It has
been shown that inhibition of rolling and adhesion can reduce
both the vascular dysfunction and tissue injury associated
with ischemia reperfusion and the plaque formation associated with atherosclerosis.13,14 Whether IVIg can affect any of
these molecular mechanisms is unclear at the present time.
In this study, we systematically examined the role of IVIg
in an ischemia-reperfusion–induced multistep recruitment
cascade of leukocytes. Intravital microscopy was used in a
feline in vivo model of P-selectin– and integrin-dependent
leukocyte recruitment of ischemia-reperfusion coupled with
an in vitro laminar flow-chamber system. The results demonstrated that IVIg severely impairs P-selectin– dependent
leukocyte rolling in both the in vivo and in vitro system. In
the in vivo ischemia-reperfusion model, both P-selectin–
dependent rolling and integrin-dependent adhesion was inhibited by IVIg, which translated to a very profound reduction in vascular dysfunction similar in degree to the
antiadhesive effects.
Materials
Reagents and Antibodies
IVIg was a generous gift from Bayer Inc (Canada). Recombinant
human P-selectin and E-selectin were purchased from R&D Systems
Inc. Histamine was purchased from Sigma Chemical Co. The
anti-␤2-integrin antibody (IB4) was generously provided by Dr Paul
Naccache (Laval University, Quebec City, Quebec, Canada). The
anti-P-selectin glycoprotein ligand-1 (PSGL-1) antibody (KPL-1)
was purchased from BD Pharmingen. Heparin was purchased from
Organon Ltd. Collagenase A was purchased from Roche. Medium
199 (M199), antibiotic/antimitotic cocktail, glutamine, and trypsinEDTA were all purchased from GIBCO-BRL. Fibronectin was
purchased from Biomedical Technologies, and glass coverslips were
purchased from Fisher Scientific. All other reagents were from
Sigma.
Flow-Chamber Experiment
Endothelium Isolation
Human umbilical vein endothelial cells (HUVECs) were harvested
and cultured from fresh human umbilical cords as described previously.15 Briefly, fresh cords were perfused with sterile PBS. The
cords were filled with collagenase (1 mg/mL) and incubated in warm
PBS for 20 minutes. After the incubation period, the cords were
gently massaged to facilitate the release of endothelial cells from the
vessel walls. The digest from the cords was drained into centrifuge
tubes that contained heat-inactivated fetal bovine serum (FBS), and
the cord was further perfused with M199 that was supplemented with
20% FBS, antibiotic cocktail, and glutamine. The cell lysate was
centrifuged for 8 to 10 minutes at 1100 rpm, and the resulting cell
pellet was resuspended in M199 and seeded in fibronectin-coated
T25 flasks. Once the cells became confluent (3 to 5 days), trypsinEDTA was used to detach the endothelial cells, which were plated
onto fibronectin-coated glass coverslips. All endothelium was from
first-passage HUVECs.
Preparation of Protein-Coated Coverslip
Glass coverslips were coated with the soluble adhesion molecules
P-selectin or E-selectin at 5 ␮g/mL and incubated at 4°C for 18
hours. To inhibit nonspecific interactions with glass, coverslips were
incubated with 1% bovine serum albumin at 37°C for 2 hours.
Flow-Chamber Assay
To study the leukocyte-protein and leukocyte– endothelial cell interactions under shear conditions in vitro, a flow-chamber assay was
used as described previously.16 Glass coverslips plated with soluble
adhesion molecules or confluent endothelial cells were mounted onto
a polycarbonate chamber with parallel plate geometry. The flow
chamber was placed onto the stage of an inverted microscope
(Zeiss), which was enclosed in a warm-air cabinet maintained at
37°C. The substrates were visualized at 200⫻ with the use of
phase-contrast microscopy. A syringe pump (Harvard Apparatus)
was used to draw blood over the substrate. Whole blood was taken
from healthy individuals, and 30 U/mL heparin sodium (1000 U) was
added to prevent coagulation. Heparin has been shown not to affect
leukocyte-endothelium interactions, whereas other anticoagulants,
such as citrate, do affect interactions. The perfusion rate was set at 10
dyne/cm2 for all flow-chamber experiments that involved soluble
adhesion molecules and endothelial monolayers. Experiments were
video recorded via a charge-coupled device camera (Hitachi Denshi)
and a videocassette recorder (Panasonic) attached to the microscope.
Rolling and adherent cell counts were made through playback video
analysis.
Leukocyte recruitment was examined on immobilized E-selectin
or P-selectin and histamine-stimulated endothelial monolayers. For
the immobilized adhesion molecule experiments. Hanks’ balanced
salt solution (HBSS; with Ca2⫹, Mg2⫹, and sodium bicarbonate) was
perfused briefly over the coated coverslip, followed by whole blood,
treated with or without IVIg (20 mg/mL, 30 minutes), and perfusion
at 10 dyne/cm2 for 5 minutes. The coverslip was once again perfused
with HBSS to clear nonattached red blood cells and leukocytes, and
5 fields of view were recorded for 20 seconds each. Rolling and
adhesion were determined by playback analysis as described previously. A leukocyte that remained stationary for at least 10 seconds
was defined as adherent. Experiments without pretreatment of whole
blood with IVIg were also performed wherein IVIg was added
immediately before perfusion of whole blood.
Rapid P-selectin, ␤2-integrin-dependent neutrophil recruitment on
endothelial monolayers (identical to mechanisms in our in vivo
model) was induced with histamine as described previously.17
Confluent endothelial monolayers were perfused with HBSS containing histamine (25 ␮mol/L) for 2.5 minutes. Then, whole blood
was perfused at 10 dyne/cm2 over endothelium for 5 minutes. In
additional experiments, the histamine-treated endothelium was exposed to IVIg (20 mg/mL) for 30 minutes. After 5 minutes of whole
blood perfusion, HBSS was again perfused over the endothelial
monolayer to clear nonattached red blood cells and leukocytes. Three
fields of view were recorded for 20 seconds each to measure rolling,
and a further 2 fields of view were recorded to measure adhesion.
Rolling and adhesion were determined by playback analysis as
described previously. In control experiments, the endothelium was
perfused with HBSS for 2.5 minutes before whole blood was
perfused as described above.
To examine ␤2-integrin– dependent adhesion, leukocytes were first
allowed to adhere under static conditions on histamine-stimulated
endothelium followed by the reintroduction of flow. Whole blood
was untreated or treated with IVIg, and adhesion was determined by
playback analysis as described previously.
Flow Cytometry Measurement
To determine whether the IVIg could block surface expression of
PSGL-1 and ␤2-integrin on leukocytes, we performed fluorescenceactivated cell sorter (FACS) experiments. Briefly, a primary antibody directed against ␤2-integrin (IB4, 2.5 ␮g/mL) or PSGL-1 (2.5
␮g/mL) was added to whole blood. After an antibody incubation of
30 minutes at 4°C, the red blood cells were lysed, and leukocytes
were simultaneously fixed in 1% formalin and then labeled with
FITC-conjugated mouse IgG and measured on a FACScan flow
cytometer (Becton Dickinson). No primary antibody, an isotype, and
no secondary antibody were used as controls for each set of
experiments.
Intravital Microscopic Studies
The experimental preparation used in this study is the same as
described previously.13,14 Briefly, cats (1.2 to 2.4 kg) were fasted for
24 hours and initially anesthetized with ketamine hydrochloride (75
Gill et al
mg IM). The jugular vein was cannulated, and anesthesia was
maintained by the administration of pentobarbital sodium. A tracheotomy was performed to support breathing by artificial ventilation.
Systemic arterial pressure was monitored continuously with a chart
recorder (Grass Instruments) with a Statham P23A (Gould) pressure
transducer connected to a catheter in the left carotid artery. A midline
abdominal incision was made, and a segment of small intestine was
isolated from the ligament of Treitz to the ileocecal valve. The
remainder of the small and large intestines was extirpated. Body
temperature was maintained at 37°C with an infrared heat lamp. All
exposed tissues were moistened with saline-soaked gauze to prevent
evaporation. Heparin sodium (10 000 U) was administered; then, an
arterial circuit was established between the superior mesenteric
arterial and left femoral artery. Superior mesenteric arterial blood
flow was monitored continuously with an electromagnetic flowmeter
(Carolina Medical Electronics).
Cats were placed in a supine position on an adjustable plexiglas
microscope stage, and a segment of midjejunum was exteriorized
through the abdominal incision. The mesentery was prepared for in
vivo microscopic observation as described previously. The mesentery was draped over an optically clear viewing pedestal that allowed
for transillumination of a 30-mm segment of tissue. The temperature
of the pedestal was maintained at 37°C with a constant temperature
circulator (model 80; Fisher Scientific). The exposed bowel was
draped with saline-soaked gauze, whereas the remainder of the
mesentery was covered with Saran Wrap (Dow Corning). The
exposed mesentery was suffused with warmed bicarbonate-buffered
saline (pH 7.4) that was aerated with a mixture of 5% CO2 and 95%
N2. The mesenteric preparation was observed through an intravital
microscope (Optiphot-2; Nikon) with a 25⫻ objective lens (Wetzlar
L25/0.35; E. Leitz) and a 10⫻ eyepiece. The image of the microcirculatory bed (1400⫻ magnification) was recorded with a videocamera (Digital 5100; Panasonic) and a video recorder (NV8950;
Panasonic).
Single unbranched mesenteric venules (25 to 40 ␮m diameter, 250
␮m length) were selected for each study. Venular diameter was
measured either online or offline with a video caliper (Cardiovascular Research Institute, Texas A&M University). The number of
rolling and adherent leukocytes was determined offline during
playback analysis. Rolling leukocytes were defined as white blood
cells that moved at a velocity less than that of erythrocytes in a given
vessel. The number of rolling leukocytes (flux) was counted by
frame-by-frame analysis. To obtain a complete leukocyte rolling
velocity profile, the rolling velocity of all leukocytes entering the
vessel was measured. A leukocyte was defined as adherent to venular
endothelium if it remained stationary for ⬎30 seconds. Adherent
cells were measured at 10-minute intervals and expressed as the
number per 100-␮m length of venule. Red blood cell velocity (VRBC)
was measured with an optical Doppler velocitometer (Cardiovascular
Research Institute, Texas A&M University), and mean velocity
(Vmean) was determined as VRBC/1.6. Wall shear rate was calculated
based on the newtonian definition: shear rate⫽(Vmean/Dv)⫻(8/time
[seconds]), where Dv is the venular diameter.
Experimental Protocol
In Vivo Experiments
Baseline measurements of blood pressure, superior mesenteric arterial blood flow, VRBC, and vessel diameter were obtained. Experiments were performed in untreated animals, IVIg (0.2g/kg)-treated
animals, and, as a positive control, fucoidan-treated (25 mg/kg)
animals in ischemia-reperfusion.
Ischemia-Reperfusion Model
In the first group of animals, the preparation was videotaped for 10
minutes, and then superior mesenteric arterial blood flow was
mechanically reduced (Gaskell clamp) to 20% of control for 1 hour.
The final 10 minutes of the ischemic period were videotaped, and the
clamp was removed to restore intestinal blood flow. Video recordings were made at 10, 30, and 60 minutes of reperfusion. In the other
series of animals, an identical protocol was completed, but the
IVIg and Leukocyte Recruitment
2033
animals received an IVIg pretreatment (0.2 g/kg; Bayer) or a
fucoidan pretreatment (25 mg/kg). The concentration of IVIg used is
at the lower end of the dose administered to humans.18,19 Another
group of animals received 0.2 g/kg human albumin, which served as
a control.
Microvascular Permeability
The degree of microvascular dysfunction was determined by vascular albumin leakage in cat mesenteric venules. Briefly, 25 mg/kg
FITC-labeled bovine albumin was administered intravenously to
animals 15 minutes before the start of the experimental procedure.
Fluorescence intensity (excitation wavelength 420 to 490 nm, emission wavelength 520 nm) was detected with a silicon-intensified
fluorescent camera (model C-2400-08, Hamamatsu Photonics), and
images were recorded for playback analysis with a videocassette
recorder. The fluorescent intensity of FITC-labeled albumin within a
defined area (10⫻50 ␮m) of the venule under study and in the
adjacent perivascular interstitium (20 ␮m from venule) was measured under control conditions at 60 minutes of ischemia and at 10,
30, and 60 minutes of reperfusion. This was accomplished with a
video-capture board (Visionplus AT-OFG, Imaging Technology)
and a computer-assisted digital imaging processor (Optimas, Bioscan). The index of vascular albumin leakage (permeability index)
was determined from the ratio: (interstitial intensity⫺background)/
(venular intensity⫺background), as reported previously.20,21
Statistics
All data are reported as mean⫾SE. A Student t test was used to
compare differences between groups, with a Bonferroni correction
for multiple comparisons. Significance was set at P⬍0.05.
Results
IVIg Can Directly Inhibit Leukocyte Interactions
With Selectins
Figure 1A shows that when whole blood was perfused at 10
dyne/cm2 over P-selectin– coated coverslips, ⬇200 rolling
leukocytes were observed. This rolling was a P-selectin–
specific event, because a P-selectin antibody blocked all
interaction (data not shown). There was no rolling observed
on coverslips that were coated with a nonselectin protein (ie,
BSA). Figure 1A also demonstrates that IVIg inhibited
leukocyte rolling on P-selectin in a dose-dependent manner,
with maximal inhibition at 20 mg/mL, which is within the
range of IVIg concentrations achieved in patients.18,19 Addition of isolated plasma from a single normal human (3 to 10
mg/mL IgG) was insufficient to demonstrate any inhibitory
effects (data not shown), yet IVIg at 1 mg/mL inhibited
recruitment by 60%. Figure 1B demonstrates that addition of
IVIg directly to the P-selectin– coated coverslip had a minimal effect on leukocyte–P-selectin interactions, whereas pretreatment of leukocytes with IVIg eliminated all interactions.
This suggests that IVIg affects the leukocyte rather than the
immobilized P-selectin. In all of the above experiments, we
pretreated the blood or the P-selectin– coated coverslip with
IVIg for 30 minutes. When blood was not pretreated with
IVIg, there was no decrease in P-selectin– dependent leukocyte recruitment (Figure 1C). Similar results were observed
with histamine-dependent rolling and adhesion (data not
shown). Clearly, the 30-minute pretreatment was required.
In a second series of experiments, treatment of whole blood
with IVIg followed by perfusion over immobilized E-selectin
caused approximately an 80% inhibition of leukocyte recruitment, which suggests that IVIg can block interactions with
multiple selectin substrata (Figure 1D). However, rolling on
2034
Circulation
September 27, 2005
Figure 1. IVIg directly inhibits leukocyte
interactions with selectins in vitro. A,
Leukocyte rolling observed on immobilized P-selectin– coated coverslips after
perfusion of whole blood. IVIg treatment
was administered in a dose-dependent
manner, with maximal inhibition occurring at the IVIg concentration of 20
mg/mL (n⫽3). B, Leukocyte rolling on
P-selectin– coated coverslips (open bar)
compared with leukocyte rolling on
P-selectin when either the coverslip or
whole blood was pretreated with IVIg
(n⫽4). C, Leukocyte rolling observed on
immobilized P-selectin when whole
blood was not treated with IVIg (open
bar) or was treated for 0 or 30 minutes
with IVIg (solid bar; n⫽3). D, Leukocyte
rolling observed on immobilized E-selectin– coated coverslips after perfusion of
whole blood with and without IVIg treatment (n⫽4). Mean fluorescence of
PSGL-1 expression on (E) neutrophils
and (F) lymphocytes treated for 30 minutes without or with IVIg (n⫽5).*P⬍0.05
compared with untreated.
vascular cell adhesion molecule-1 was not inhibited by IVIg,
which suggests this is not a nonspecific effect (data not
shown). PSGL-1 is the ligand for P-selectin and E-selectin.
Figure 1E demonstrates that IVIg blocked 50% of the
PSGL-1 expression on neutrophils but failed to block levels
on lymphocytes (Figure 1F)
To study the effect of IVIg on a more complex system, we
used primary passaged human endothelium to observe
whether IVIg could inhibit leukocyte endothelial interactions
in vitro on a physiological substrate. When histamine was
used to induce P-selectin expression on endothelium, ⬇80
leukocytes rolled on the stimulated endothelium (Figure 2A).
When the experiment was repeated with blood treated with an
optimal dose of IVIg, there was almost a complete inhibition
of rolling leukocytes over histamine-stimulated endothelium
(Figure 2A). Furthermore, when untreated blood was perfused over histamine-stimulated endothelium, ⬇80 leukocytes (predominantly neutrophils) adhered, and similarly, the
Figure 2. IVIg inhibits leukocyte– endothelial cell interactions in vitro on a physiological substrate. Leukocyte rolling (A)
and adhesion (B) observed on histaminestimulated endothelium (25 ␮mol/L), with
or without IVIg treatment, of either the
blood or the endothelium (n⫽3). *P⬍0.05
compared with histamine-treated
endothelium
Gill et al
2035
ments with flow cytometry revealed that ⬇30% of ␤2-integrin
expression on both neutrophils (Figure 3B) and lymphocytes
(Figure 3C) was blocked by IVIg.
IVIg Can Directly Inhibit Leukocyte Interactions
and Vascular Dysfunction In Vivo
Figure 3. IVIg partially blocks ␤2-integrin expression on leukocytes in vitro. A, Leukocyte adhesion observed on histaminestimulated endothelium (25 ␮mol/L) under selectin-independent
static conditions with and without IVIg (n⫽5). Mean fluorescence
of ␤2-integrin on (B) neutrophils and (C) lymphocytes treated for
30 minutes without or with IVIg (n⫽6). *P⬍0.05 vs histaminestimulated (no IVIg) group.
number of adherent leukocytes was significantly reduced if
the blood had been pretreated with IVIg (Figure 2B). When
histamine-stimulated endothelium was treated with IVIg followed by perfusion of whole blood, there was no inhibition in
either leukocyte rolling or adhesion, which suggests that IVIg
was having antiadhesive effects on leukocytes rather than the
endothelium. When we removed the coverslips from the flow
chamber, ⬎95% of the adherent cells were neutrophils.
Although IVIg treatment of blood caused a decrease in
both leukocyte rolling and adhesion on histamine-treated
endothelium, this experiment failed to address the question of
whether IVIg was directly inhibiting adhesion (via ␤2integrin) as well as rolling (selectin-dependent). Using a
slightly modified flow-chamber assay in which blood flow is
stopped (allowing for firm adhesion of leukocytes to endothelium independent of selectins), we observed an ⬇60%
decrease in leukocyte adhesion when blood was treated with
IVIg compared with untreated blood (Figure 3A). Experi-
Next, an in vivo model of ischemia-reperfusion, previously
shown to be mediated by both P-selectin and ␤2-integrin–
dependent leukocyte recruitment, was examined (same mechanism as for the histamine-treated endothelium). Under basal
conditions (Figure 4A, top panel), very few cells rolled and
adhered, whereas after ischemia-reperfusion, leukocyte rolling, adhesion, and emigration were increased greatly in the
same vessel (Figure 4A, middle panel). Pretreatment with
IVIg (0.2 g/kg) prevented the accumulation of rolling and
adherent leukocytes at 30 minutes of reperfusion (Figure 4A,
bottom panel; see video in Data Supplement). This concentration of IVIg is within the range that is used in human
patients (0.2 to 2 g/kg).18,19 Quantification of the data is
summarized in Figure 4B through 4D. The flux of rolling
leukocytes under baseline conditions was ⬇30 to 50 cells/
min, and this value remained unchanged during the ischemic
period (data not shown). Administration of IVIg did not affect
this basal leukocyte rolling. During the reperfusion phase, the
flux of rolling cells increased dramatically, ranging from 125
to 175 cells/min in untreated animals. IVIg treatment (0.2
g/kg) prevented the increase in the flux of rolling leukocytes
during the reperfusion phase. In untreated animals, a 10-fold
increase in leukocyte adhesion occurred during the reperfusion phase (Figure 4C). IVIg pretreatment of the animals
reduced the adhesion of leukocytes at 10 minutes of reperfusion and essentially prevented all increases in leukocyte
adhesion at 30 and 60 minutes. In fact, the number of
adherent leukocytes in the IVIg-treated animals during the
reperfusion phase was reduced to near-preischemic values.
Figure 4D shows that in untreated animals, there was a
significant increase in the number of emigrated leukocytes at
60 minutes of reperfusion. In IVIg-treated animals, a much
more subtle increase in emigrated leukocytes was noted.
It has been reported previously that inhibition of leukocyte
adhesion and emigration dramatically reduced the increase in
microvascular dysfunction associated with reperfusion injury.22,23 Moreover, depletion of neutrophils also eliminated the
vascular dysfunction.22,24 FITC-albumin was given intravenously, and the leakage of protein from the mesenteric
microvasculature was determined under control and reperfusion conditions in the same preparation. A very obvious
increase in vascular protein leakage (reperfusion 30 minutes)
could be seen in Figure 5A (middle versus left panel). When
animals were pretreated with IVIg, there was a dramatic
reduction in vascular protein leakage during the reperfusion
phase. Computer-assisted quantification revealed a 6- to
7-fold increase in FITC-albumin leakage from venules during
reperfusion after no IVIg administration or after albumin
administration (control protein), whereas administration of
IVIg to the experimental group increased microvascular
dysfunction by ⬍2-fold during reperfusion (Figure 5B).
Because much of the vascular leakage is due to adhering and
emigrating leukocytes, the decreases in the number of adher-
2036
Circulation
September 27, 2005
Figure 4. IVIg directly inhibits leukocyte
interactions in vivo. A, The microvessel
as visualized under baseline control conditions (top). Ischemia was induced for 1
hour, followed by the reperfusion phase.
The microvessel was visualized at 30
minutes of reperfusion with (middle) and
without (bottom) IVIg treatment. The
quantified data show number of (B) rolling leukocytes, (C) adherent leukocytes,
and (D) emigrated leukocytes during
control period (CON) and 10, 30, and 60
minutes after reperfusion (REP) in
untreated animals or IVIg-treated animals
(n⫽4).*P⬍0.05 relative to control;
⫹P⬍0.05 relative to untreated group.
ent and emigrated cells by IVIg are likely responsible for the
reduced vascular permeability. It has been shown that ⬎80%
of infiltrating leukocytes during ischemia-reperfusion are
neutrophils.23
Nevertheless, we decided to directly compare the effects of
IVIg to antiadhesion (antiselectin) therapy. To inhibit leukocyte recruitment, we used a selectin antagonist, fucoidan, to
inhibit leukocyte rolling, which has an impact on adhesion.
Fucoidan treatment in the ischemia-reperfusion model
showed a dramatic decrease in leukocyte rolling and a 60%
decrease in adhesion at 60 minutes of reperfusion compared
with untreated animals (Figures 6A and 6B). Similarly, there
was also a 60% inhibition of vascular leakage in fucoidantreated animals compared with untreated animals (Figure 6C).
IVIg appeared to be at least as effective if not more effective
(Figure 5 versus Figure 6) at preventing leukocyte adhesion
and subsequent vascular permeability.
Discussion
Recruitment of leukocytes is a hallmark of many of the
diseases in which IVIg has efficacy. Our data suggest that
IVIg may function through interference of leukocyte recruitment. Using a simple system of observing leukocytes under
flow conditions on immobilized protein (P-selectin or
E-selectin), we found that IVIg inhibited leukocyte rolling.
Furthermore, we found that IVIg could interfere in a more
complex in vitro system, such as histamine-treated endothelium, and in an in vivo system in which IVIg interfered with
not only leukocyte rolling but also leukocyte adhesion and
transmigration. Although there have been a number of proposed mechanisms by which IVIg works, there has been no
study that examined directly the role IVIg has on selectinand integrin-dependent leukocyte adhesion under flow conditions. One study to date has showed that IVIg caused a
reduction in the binding of lipopolysaccharide-treated neutrophils to endothelium in a static adhesion assay.25 There is also
indirect evidence that IVIg could reduce the expression of
lymphocyte function-associated antigen-1 (LFA-1) on neutrophils, thereby reducing leukocyte adhesion.26 In vivo
studies using intravital microscopy have demonstrated that
IVIg could reduce leukocyte recruitment into the liver after
systemic administration of tumor necrosis factor-␣ or lipopolysaccharide.27 However, those authors proposed a mechanism that involved macrophage activity. The present study
shows for the first time that IVIg can directly inhibit
selectin-dependent leukocyte rolling in human systems in
vitro and in animal models of ischemia-reperfusion in vivo.
Moreover, the present data suggest that IVIg could be
considered as a form of treatment in certain vascular pathologies associated with ischemia-reperfusion injury.
Gill et al
2037
Figure 5. IVIg can inhibit vascular dysfunction in vivo. The microvessel as visualized when measuring fluorescence under baseline control conditions (left). Ischemia was induced for 1 hour and followed by the reperfusion phase. The microvessel was visualized, with fluorescence measured at 30 minutes of reperfusion with (middle) and without (right) IVIg treatment. The quantified data show the (B) vascular permeability as measured by percent leakage at 10, 30, and 60 minutes after reperfusion (REP) in untreated animals or IVIgtreated animals (n⫽4).*P⬍0.05 relative to control; ⫹P⬍0.05 relative to untreated group.
We have demonstrated that IVIg inhibits leukocyte rolling
on selectins in vitro. It appears that IVIg is targeting the
P-selectin/P-selectin ligand interaction more than the ␤2integrin/intercellular adhesion molecule-1 interaction. Additionally, under in vivo conditions in which recruitment is
largely P-selectin dependent (ischemia-reperfusion model),
IVIg also inhibited leukocyte recruitment. Using the ischemia-reperfusion model, we have previously demonstrated that
when anti-P-selectin therapy is used, rolling must be inhibited
by ⬎90% before any decrease in leukocyte adhesion is
observed.14 The present data showed that in the human in
vitro system, IVIg did block leukocyte rolling by this amount,
which likely explains the dramatic impact on adhesion. By
contrast, there was only approximately a 50% decrease in
leukocyte rolling with IVIg treatment in our ischemiareperfusion in vivo model. Surprisingly, we observed an 80%
decrease in the number of adherent cells. These data suggest
that IVIg in the present in vivo system may be directly
inhibiting both the process of rolling and adhesion of leukocyte recruitment.
Clearly, the present data suggest that in vitro, there is a
more minor effect of IVIg on ␤2-integrin– dependent adhesion, whereas in vivo, the data suggest a very significant
(80%) impact of IVIg on ␤2-integrin– dependent adhesion.
There are a number of possible explanations for this. In vitro,
flow was stopped completely to allow neutrophils to adhere,
and under these conditions, IVIg had only a minor effect. In
vivo, the situation is more dynamic, and in the presence of the
initial shear, it is possible that IVIg can impact more
dramatically on neutrophil adhesion. In other words, the
small (40%) reduction in vitro may translate into a more
profound effect in vivo. Additionally, mast cells and other
cells have been implicated as contributors to neutrophil
recruitment by releasing proinflammatory mediators.28,29
These cells are not present in the in vitro system used in the
present study, and if IVIg affects these cells, then the effect
would only be seen in vivo. Finally, other possible differences include macrovascular endothelium in vitro versus
microvascular endothelium in vivo and potential species
differences.
There was a very profound effect on vascular permeability
with IVIg treatment. Previously, it has been shown that
vascular leakage is entirely dependent on adhering neutrophils.24 Indeed, the prevention of leukocyte adhesion or the
depletion of neutrophils prevents the vascular leakage seen
after ischemia (Figure 6). IVIg treatment inhibited leukocyte
adhesion and subsequent emigration and in turn led to the
decreased vascular permeability observed in treated animals.
Clearly, IVIg could have very beneficial effects in numerous
cardiovascular diseases in which infiltrating leukocytes and
associated edema and vascular dysfunction play a large role
in pathology.
The exact mechanism by which IVIg prevents leukocyte–
endothelial cell interactions remains unknown, but it was
2038
Circulation
September 27, 2005
In summary, the present data clearly demonstrate that IVIg
directly inhibits leukocyte– endothelial cell interactions. The
mechanisms involve direct inhibition of leukocyte interactions with selectins, with a lesser effect with ␤2-integrins.
Furthermore, these data suggest therapeutic potential for IVIg
in the treatment of cardiovascular disease associated with
reperfusion injury.
Acknowledgments
This work was supported by grants from the Bayer Canadian Blood
Services Canadian Institutes of Health Research Partnership Fund.
Dr Kubes is an Alberta Heritage Foundation for Medical Research
Scientist and CRC Chair. V. Gill is a member of the CIHR training
program.
References
Figure 6. Inhibition of rolling and adhesion of leukocytes via
fucoidan treatment in ischemia reperfusion decreases vascular
dysfunction in vivo. Quantified data show the number of (A) rolling leukocytes, (B) adherent leukocytes, and (C) vascular permeability as measured by percent FITC-albumin leakage at control
period (CON) and 60 minutes of reperfusion (REP) in untreated
animals or fucoidan-treated animals (n⫽4). *P⬍0.05 relative to
control; ⫹P⬍0.05 relative to untreated group.
clear that it is the leukocyte that is the target, rather than the
endothelium. Indeed, treatment of leukocytes but not the
endothelium or P-selectin with IVIg blocked the leukocyte
interactions. An additional potential target for IVIg is
PSGL-1, which is the primary ligand for P-selectin.30 It is
possible that IVIg binds to PSGL-1, thus blocking any
interaction with P-selectin. Indeed, we observed a 50%
decrease in PSGL-1antibody binding to PSGL-1 in the
presence of IVIg. Also noteworthy is that PSGL-1 is also an
important ligand for E-selectin, and IVIg blocked this
interaction.
1. Ballow M. Mechanisms of action of intravenous immune serum globulin
in autoimmunity and inflammatory diseases. J Allergy Clin Immunol.
1997;100:151–157.
2. Dwyer JM. Manipulating the immune system with immune globulin.
N Engl J Med. 1992;326:107–116.
3. Bayry J, Thirion M, Misra N, Thorenoor N, Delignat S, LacroixDesmazes S, Bellon B, Kaveri S, Kazatchkine MD. Mechanisms of action
of intravenous immunoglobulin in autoimmune and inflammatory
diseases. Neurol Sci. 2003;24(suppl 4):S217–S221.
4. Crow AR, Song S, Freedman J, Helgason CD, Humphries RK,
Siminovitch KA, Lazarus AH. IVIg-mediated amelioration of murine ITP
via FcgammaRIIB is independent of SHIP1, SHP-1, and Btk activity.
Blood. 2003;102:558 –560.
5. Alejandria MM, Lansang MA, Dans LF, Mantaring JBV. Intravenous
immunoglobulin for treating sepsis and septic shock. In: The Cochrane
Database of Systematic Reviews. 2002, Issue 1, Art. No. CD001090.
6. Oates-Whitehead RM, Baumer JH, Haines L, Love S, Maconochie IK,
Gupta A, Roman K, Dua JS, Flynn I. Intravenous immunoglobulin for the
treatment of Kawasaki disease in children. In: The Cochrane Database of
Systematic Reviews. 2003, Issue 4. Art. No. CD004000.
7. Mok GC, Sung RY, Yam MC, Arifi AA, Lam WW, Fok TF. A child with
Kawasaki disease who survived after rupture of a coronary artery aneurysm. Eur J Pediatr. 2003;162:634 – 636.
8. Terai M, Honda T, Yasukawa K, Higashi K, Hamada H, Kohno Y.
Prognostic impact of vascular leakage in acute Kawasaki disease. Circulation. 2003;108:325–330.
9. Tse SM, Silverman ED, McCrindle BW, Yeung RS. Early treatment with
intravenous immunoglobulin in patients with Kawasaki disease.
J Pediatr. 2002;140:450 – 455.
10. Moran AM, Newburger JW, Sanders SP, Parness IA, Spevak PJ, Burns
JC, Colan SD. Abnormal myocardial mechanics in Kawasaki disease:
rapid response to gamma-globulin. Heart J. 2000;139(pt 1):217–223.
11. Muller WA. The role of PECAM-1 (CD31) in leukocyte emigration:
studies in vitro and in vivo. J Leukoc Biol. 1995;57:523–528.
12. Zimmerman GA, McIntyre TM, Prescott SM. Adhesion and signaling in
vascular cell-cell interactions. J Clin Invest. 1996;98:1699 –1702.
13. Ostrovsky L, Woodman RC, Payne D, Teoh D, Kubes P. Antithrombin III
prevents and rapidly reverses leukocyte recruitment in ischemia/
reperfusion. Circulation. 1997;96:2302–2310.
14. Kubes P, Jutila M, Payne D. Therapeutic potential of inhibiting leukocyte
rolling in ischemia/reperfusion. J Clin Invest. 1995;95:2510 –2519.
15. Ibbotson GC, Doig C, Kaur J, Gill V, Ostrovsky L, Fairhead T, Kubes P.
Functional alpha4-integrin: a newly identified pathway of neutrophil
recruitment in critically ill septic patients. Nat Med. 2001;7:465– 470.
16. Reinhardt PH, Kubes P. Differential leukocyte recruitment from whole
blood via endothelial adhesion molecules under shear conditions. Blood.
1998;92:4691– 4699.
17. Lorant DE, Patel KD, McIntyre TM, McEver RP, Prescott SM,
Zimmerman GA. Coexpression of GMP-140 and PAF by endothelium
stimulated by histamine or thrombin: a juxtacrine system for adhesion and
activation of neutrophils. J Cell Biol. 1991;115:223–234.
18. Levy JB, Pusey CD. Nephrotoxicity of intravenous immunoglobulin. Q
J Med. 2000;93:751–755.
19. Bleeker WK, Teeling JL, Hack CE. Accelerated autoantibody clearance
by intravenous immunoglobulin therapy: studies in experimental models
Gill et al
20.
21.
22.
23.
24.
25.
to determine the magnitude and time course of the effect. Blood. 2001;
98:3136 –3142.
Kurose I, Wolf R, Grisham MB, Aw TY, Specian RD, Granger DN.
Microvascular responses to inhibition of nitric oxide production: role of
active oxidants. Circ Res. 1995;76:30 –39.
Gaboury JP, Niu X-F, Kubes P. Nitric oxide inhibits numerous features of
mast cell-induced inflammation. Circulation. 1996;93:318 –326.
Hernandez LA, Grisham MB, Twohig B, Arfors KE, Harlan JM, Granger
DN. Role of neutrophils in ischemia-reperfusion-induced microvascular
injury. Am J Physiol. 1987;253:H699 –H703.
Oliver MG, Specian RC, Perry MA, Granger DN. Morphologic assessment
of leukocyte-endothelial cell interactions in mesenteric venules subjected to
ischemia and reperfusion. Inflammation. 1991;15:331–346.
Kurose I, Anderson DC, Miyasaka M, Tamatani T, Paulson JC, Todd RF,
Rusche JR, Granger DN. Molecular determinants of reperfusion-induced leukocyte adhesion and vascular protein leakage. Circ Res. 1994;74:336–343.
Jonas E, Dwenger A, Jonas M. Chemiluminescence response and
adherence of neutrophils to cultured endothelial cells: influence of immunoglobulin G. J Biolumin Chemilumin. 1995;10:169 –173.
2039
26. Rigal D, Vermot-Desroches C, Heitz S, Bernaud J, Alfonsi F, Monier JC.
Effects of intravenous immunoglobulins (IVIG) on peripheral blood B,
NK, and T cell subpopulations in women with recurrent spontaneous
abortions: specific effects on LFA-1 and CD56 molecules. Clin Immunol
Immunopathol. 1994;71:309 –314.
27. Ito Y, Lukita-Atmadja W, Machen NW, Baker GL, McCuskey RS. Effect
of intravenous immunoglobulin G on the TNFalpha-mediated hepatic
microvascular inflammatory response. Shock. 1999;11:291–295.
28. Kanwar S, Kubes P. Ischemia/reperfusion-induced granulocyte influx is a
multistep process mediated by mast cells. Microcirculation. 1994;1:
175–182.
29. Kanwar S, Kubes P. Mast cells contribute to ischemia-reperfusioninduced granulocyte infiltration and intestinal dysfunction. Am J Physiol.
1994;267(pt 1):G316 –G321.
30. Asa D, Raycroft L, Ma L, Aeed PA, Kaytes PS, Elhammer AP, Geng JG.
The P-selectin glycoprotein ligand functions as a common human leukocyte ligand for P- and E-selectins. J Biol Chem. 1995;270:
11662–11670.
Special Report
Lessons From the Failure and Recall of an Implantable
Cardioverter-Defibrillator
Robert G. Hauser, MD; Barry J. Maron, MD
M
promptly communicate the detailed nature of this flaw to
physicians and patients. That communication, it was noted,
should also emphasize that no test or monitoring technique could
predict if or when a suspect Prizm 2 DR device may fail. This
recommendation was based on the view that physicians and
patients should have all critical information so that they can
decide whether prophylactic ICD replacement was prudent.
Guidant’s responsibility, in our opinion, was to disclose these
vital data completely and expeditiously.
However, Guidant believed that such a communication was
inadvisable and unnecessary. The company maintained that because
the observed failure rate was very low, physicians could unnecessarily expose patients to the risks of device replacement surgery,
including infection. Guidant’s statistical argument ignored the basic
tenet that patients have a fundamental right to be fully informed
when they are exposed to the risk of death no matter how low that
risk may be perceived. Furthermore, by withholding vital information, Guidant had in effect assumed the primary role of managing
high-risk patients, a responsibility that belongs to physicians. The
prognosis of our young, otherwise healthy patient for a long,
productive life was favorable if sudden death could have been
prevented.5 Certainly, this was the rationale for implanting his
Prizm 2 DR in 2001. If we had known that the Prizm 2 DR was
prone to sudden failure as a result of short circuiting or another
mechanism, his device would have been replaced promptly.
Because Guidant declined to inform patients and their physicians,
we regarded it as our moral and ethical obligation to disclose the
problem to the medical community and the public.3,6 On May 23,
2005, the day before the New York Times reported these events,
Guidant sent a letter to physicians describing the Prizm 2 DR flaw
and suggested the failures were random events; Guidant recommended that physicians continue normal monitoring.7
On June 17, 2005, after alerting the Food and Drug Administration (FDA), Guidant recalled 26 000 Prizm 2 DR devices manufactured before April 2002.4 Simultaneously, Guidant issued 3
additional recalls affecting ⬎50 000 of its cardiac resynchronization
ICDs.8–10 One of these recalls was prompted by 15 Contak Renewal
and Contak Renewal 2 ICD failures8 that were also caused by short
circuiting that Guidant had known about for at least a year. Similar
to its management of the Prizm 2 DR situation, Guidant had
implemented manufacturing changes in August 2004 and did not
disclose the problem until a patient died when a Contak Renewal
ultiple clinical trials have shown that a properly functioning implantable cardioverter-defibrillator (ICD) is
capable of interrupting sudden death caused by ventricular
tachyarrhythmias. Unfortunately, ICDs are complex medical
devices, and they do not always perform as expected. For
example, only 5% of ICD batteries last ⬎7 years, and most
dual-chamber ICD models must be replaced for battery
depletion every 3 to 5 years.1 Normal battery depletion,
however, is reliably predicted by routine follow-up methods
before an ICD fails. In contrast, electronic malfunctions are
unpredictable and may not be detected by standard follow-up
techniques before an ICD is unable to deliver effective
therapy. Thus, sudden cardiac arrest or death may be the first
and only sign that an ICD has failed.2
Recently, we reported the death of a 21-year-old patient who
received a Prizm 2 DR model 1861 ICD pulse generator
(Guidant, Inc) in 2001 to prevent sudden cardiac death resulting
from hypertrophic cardiomyopathy.3 In March 2005, this young
man experienced a witnessed arrest and could not be resuscitated. His ICD was returned to Guidant, which found that the
device had failed during the delivery of a shock. The cause of
failure was massive electronic damage caused by electrical
overstress that occurred when a short circuit developed between
a high-voltage wire and a tube used to test the housing during
manufacturing (see the Figure).
At the time of our patient’s death, Guidant had knowledge
of 25 similar Prizm 2 DR model 1861 failures in patients, 3
of whom had required rescue defibrillation. Indeed, Guidant
had first observed this mode of failure 3 years earlier, in
February 2002, when 2 returned Prizm 2 DR pulse generators
exhibited the same short circuiting that caused our patient’s
device to fail. Guidant was sufficiently concerned about these
failures that manufacturing changes were made in April and
November of 2002, which allegedly prevented short circuiting. Nevertheless, Guidant chose not to inform patients or
physicians about these failures or the manufacturing changes
designed to prevent them. Moreover, Guidant continued to
sell pulse generators that were built before the 2002 manufacturing changes. Unknowingly, therefore, we and other
physicians implanted Prizm 2 DR ICDs in 2002 and 2003 that
Guidant knew were prone to sudden unexpected failure.4
In May 2005, after the death of our patient, we recommended
during meetings with Guidant officials that the company
From the Hypertrophic Cardiomyopathy Center of the Minneapolis Heart Institute Foundation, Minneapolis, Minn.
Correspondence to Robert G. Hauser, MD, Hypertrophic Cardiomyopathy Center, Minneapolis Heart Institute Foundation, 920 E 28th St, Minneapolis, MN 55407.
(Circulation. 2005;112:2040-2042.)
2040
Hauser and Maron
Guidant Prizm 2 DR model 1861 ICD pulse generator. Short circuiting
occurs across the space between the (⫹) backfill tube and the (⫺)
defibrillation (DF) feed-through wire. The wire connects to the defibrillator port in the connector. Reproduced from Gornick et al,3 copyright
2004, with permission from Heart Rhythm Society.
ICD short-circuited during shock delivery on May 30, 2005, while
the FDA was investigating the Prizm 2 DR.
The FDA classified Guidant’s Prizm 2 DR and Contak Renewal
and Contak Renewal 2 notifications as class I recalls, which indicate
that a reasonable probability exists that the use of these devices will
cause serious adverse health consequences or death. In contrast,
Medtronic’s voluntary February 2005 advisory with regard to a
battery short in its Marquis models11 was an FDA class II recall,
which denotes that the probability of serious adverse health consequences such as death are remote. Collectively, these 4 Guidant
recalls, which affected ⬎70 000 implanted devices worldwide,
were among the largest such industry regulatory actions in the past
25 years.
One can only speculate as to what Guidant or the FDA may have
done or the events that may have transpired if the Prizm 2 DR
problem had not been disclosed publicly.12 Nevertheless, there are
important lessons to be learned from the failure and recall of these
ICDs.
Lesson 1: The FDA’s Postmarket ICD Device
Surveillance System Is Broken
According to the FDA, the purpose of postmarket surveillance is to
improve the safety and effectiveness of devices by identifying
serious low-frequency events like those involving the Prizm 2 DR.13
The present experience suggests that the FDA is currently unable to
satisfy its legal responsibility to monitor the safety of market
released medical devices like the Prizm 2 DR.
Guidant reported 2 adverse events to the FDA in 2004 that
described the same Prizm 2 DR defect that caused our patient’s
device to fail and further indicated that the company had made
manufacturing changes to prevent it. The first report14 was received
by the FDA ⬎1 year before to the death of our patient. The second
report15 was received by the FDA in August 2004 after a patient
received an ineffective shock; the second report15 was explicit:
“Analysis confirmed an electrical short between two components,
specifically the feedthrough wire and backfill tube. It was concluded
that the shorted condition involving the header df-1 feed-thru wires
resulted in the clinical observations. Although the occurrence of this
failure has been very low, Guidant implemented manufacturing
enhancements in April and November 2002 to correct this issue.”
Lessons From the Failure and Recall of an ICD
2041
The FDA, therefore, was in possession of important information
about the safety of the Prizm 2 DR many months before the death
of our patient and, to the best of our knowledge, took no action. The
explanation for the FDA’s inaction is unknown, but it may be that
the agency was not prepared for the upsurge in ICD technology and
the extraordinary growth in the number of ICD implantations that
has occurred during the past 5 years.
Fixing the postmarket surveillance system must be a high priority
for the FDA and Congress. The public should have full access to all
of the FDA’s medical device safety and efficacy data. This should
include manufacturers’ annual reports that may contain failure data
and manufacturing changes, which the FDA is presently withholding from public scrutiny presumably because they may reveal a
company’s trade secrets. For government to keep such vital product
safety information from patients or physicians for any reason should
be unlawful.
A physician or patient should be able to learn quickly from the
FDA if a particular ICD or other critical medical device has
exhibited quality issues that may affect the performance of a
product. Moreover, the FDA’s postmarket surveillance system
should alert physicians and user facilities when manufacturers
report the type of life-threatening failure modes exemplified by the
Prizm 2 DR and the Contak Renewal and Contak Renewal 2 ICDs.
Currently, the public’s only source of postmarket product performance data are the FDA’s Manufacturers and User Facility Device
Experience (MAUDE) database. However, MAUDE is not designed to be a routine surveillance tool for physicians or a useful
source of information for patients. Moreover, the MAUDE database
may not be a reliable source of information; eg, it contained only 10
of the 28 Prizm 2 DR failures at the time these devices were
recalled.
Lesson 2: Physicians Do Not Have the Data
Necessary to Assess Device Problems and to
Make Rational Clinical Judgments
Although it is important to identify ICD problems, it is critical to
know their frequency and failure rates over time. Unfortunately, the
medical community has become totally dependent on the ICD
industry to supply failure rate data, but manufacturers can provide
only crude estimates based on their unit sales and returned products.
In its recent Prizm 2 DR recall letter,4 Guidant based its recommendations on 26 000 devices built before April 2002 and 28 failed
devices that were returned for analysis. This approach unavoidably
underestimates the actual number of failures because devices often
are not returned to the manufacturer after death or replacement.4
Consequently, the actual failure rates for the Prizm 2 DR and other
recalled ICDs are unknown.
Without precise failure rate data, physicians and patients cannot
make prudent management decisions. In the aftermath of the Prizm
2 DR and Contak Renewal recalls, physicians and patients have had
to choose between prophylactic replacement or continued followup. Because accurate failure rate data are unavailable for these
devices, management decisions are being made according to the
perceived rather than the actual risk of catastrophic ICD failure.
Caught in this conundrum, and wishing to avoid surgical complications, eg, infection, and to “do no harm,” physicians may be
hesitant to replace these devices. For ICD patients at high risk for
sudden cardiac arrest, however, the low likelihood of a treatable
2042
Circulation
September 27, 2005
infection is acceptable when the alternative is sudden death should
the device malfunction.
Because precise failure rate data are needed, a reformed postmarket surveillance system must include government-mandated
prospective follow-up studies of market-released devices. These
studies should be sufficiently powered to detect low-frequency
device failures and to provide accurate estimates of ICD longevity.
Lesson 3: Quality Standards for ICDs and
Guidelines for Managing Device Recalls
Are Needed
So far, ⬎130 000 ICDs have been recalled or subject to safety alerts
in 2005. ICDs should conform to the highest quality standards for
clinical performance. Yet, remarkably, such standards do not exist.
Standards are needed as the foundation for quality improvement and
for assessing the clinical reliability of ICDs. They can also define
the boundaries for product safety and longevity. Knowledge that a
manufacturer has met or exceeded accepted quality measures would
help patients and physicians select devices for implantation. Furthermore, manufacturers would strengthen their design and quality
processes if they were held accountable for all of the healthcare
costs associated with inferior products.
Additionally, no universally accepted definitions exist for such
critical device events as “random component failure.” Guidant has
stated that short circuiting in the Prizm 2 DR was due to a rare
random component failure and implied that such failures occurred
despite industry’s best efforts to mitigate them.7 To suggest that the
death of our patient or the death of the Contak Renewal patient was
not due to a specific, avoidable failure mode is misleading and
incorrect.
Despite the volume of recalls and advisories and the number of
patients affected by them, the appropriate clinical strategies for
managing ICD recalls and advisories are uncertain. A recent survey
suggested that experienced physicians differ significantly in their
approach to ICD recalls.16 The factor that most strongly influenced
a physician’s decision to replace a suspect device was the manufacturer’s estimated risk of sudden device failure. In our judgment,
the patient’s underlying heart disease and prognosis should be the
deciding factors favoring prophylactic device replacement. For
example, the ICD may be truly life-saving for patients with genetic
heart diseases because, for many of them, the only risk of cardiovascular death is ventricular fibrillation.
Given the large number of patients who have or will receive
ICDs and the inevitability of future device problems, the Heart
Rhythm Society, whose declared mission is to improve the care of
patients by promoting optimal healthcare policies and standards,
should take this unique opportunity to collaborate with other
professional societies to establish realistic quality standards for
ICDs and practical guidelines for managing device recalls.
Conclusions
These unfortunate events underscore the importance of a fundamental principle, namely that patients and their physicians are entitled to
full disclosure of product information that may affect an individual’s
health or safety. This principle is broadly applicable to the healthcare industry, including the manufacturers of medical devices and
drugs, and to regulatory agencies such as the FDA. Successful
application of this principle requires that a completely open, transparent relationship exist between patients, manufacturers, the FDA,
and the medical community.
Although ICDs are highly effective and generally dependable,
the recent Prizm 2 DR model 1861 experience and the recalls of
2005 demand that ICD quality and reliability improve. The Heart
Rhythm Society should lead the development of quality standards
for ICDs and guidelines for managing device recalls and safety
alerts. Congress and the FDA must develop and apply an effective
postmarket surveillance system that improves the safety of medical
devices for all patients. Finally, a major goal of these reforms is to
reassure patients that ICD therapy is reliable and effectively
regulated.
Disclosure
Dr Maron is a grantee from Medtronic, Inc.
References
1. Hauser RG. The growing mismatch between patient longevity and the service life
of implantable cardioverter defibrillators. J Am Coll Cardiol. 2005;45:
2022–2025.
2. Hauser RG, Kallinen L. Deaths associated with implantable cardioverter defibrillator failure and deactivation reported in the United States Food and Drug
Administration Manufacturer and User Facility Device Experience database.
Heart Rhythm. 2004;1:399–405.
3. Gornick CC, Hauser RG, Almquist AK, Maron BJ. Unpredictable implantable
cardioverter-defibrillator pulse generator failure due to electrical overstress
causing sudden death in a young high-risk patient with hypertrophic cardiomyopathy. Heart Rhythm. 2004;2:681–683.
4. Gorsett A. Prizm 2 DR model 1861. Guidant letter to physicians. June 17, 2005.
5. Maron BJ, Shen W-K, Link MS, Epstein AE, Almquist AK, Daubert JP, Bardy
GH, Favale S, Rea RF, Boriani G, Estes NAM III, Casey SA, Stanton MS,
Betocchi S, Spirito P. Efficacy of implantable cardioverter defibrillators for the
prevention of sudden death in patients with hypertrophic cardiomyopathy. N Engl
J Med. 2000;342:365–373.
6. Meier B. Maker of heart device kept flaw from doctors. New York Times. May
24, 2005.
7. Gorsett A. Prizm 2 DR model 1861. Guidant letter to physicians. May 23, 2005.
8. Gorsett A. Contak Renewal (model H135) and Contak Renewal 2 (model H155)
devices manufactured on or before August 26, 2004. Guidant letter to physicians.
June 17, 2005.
9. Gorsett A. Ventak Prizm AVT, Vitality AVT, Contak Renewal AVT. Guidant
letter to physicians. June 17, 2005.07.13.
10. Gorsett A. Contak Renewal 3 and 4, Renewal 3 and 4 AVT, and Renewal RF.
Guidant letter to physicians. June 24, 2005.
11. Myrum S. Marquis family of ICD and ICD-CRT devices having batteries manufactured prior to December 2003. Medtronic letter to physicians. February 2005.
12. Steinbrook R. The controversy over Guidant’s implantable defibrillators. N Engl
J Med. 2005;353:221–224.
13. Feigal DW, Gardner SN, McClellan M. Ensuring safe and effective medical
devices. N Engl J Med. 2003;348:191–192.
14. Guidant, Inc. FDA/CDRH/MAUDE MDR report key 528702. February 27,
2004.
15. Guidant, Inc. FDA/CDRH/MAUDE MDR report key 544031. August 10, 2004.
16. Maisel WH. Physician management of pacemaker and implantable cardioverter
defibrillator advisories. Pacing Clin Electrophysiol. 2004;27:437–442.
KEY WORDS: death, sudden
䡲
defibrillation
䡲
defibrillators, implantable
Special Report
Report From the Cardiovascular and Renal Drugs
Advisory Committee
US Food and Drug Administration; June 15–16, 2005; Gaithersburg, Md
Steven E. Nissen, MD
reduction in the incidence of stroke and myocardial infarction
is the most clearly attributable benefits of blood pressure
reduction. Evidence for reduction in cardiovascular mortality
and renal disease is also clearly attributable to blood pressure
lowering, but the committee felt that the evidence was
somewhat less consistent. The committee agreed that new
drugs seeking a hypertension indication should not automatically receive labeling for all the above benefits. Instead, it
was suggested that language in the “class label” should
describe the “generally expected” benefits for any drug that
lowers blood pressure. The committee recommended that
labeling statements should not conflict with national guidelines such as the Joint National Committee on Prevention,
Detection, Evaluation and Treatment of High Blood Pressure
(JNC7).
A strong consensus emerged that most of the benefits of
antihypertensive agents are related to blood pressure reduction rather than class-specific or agent-specific effects.
However, the committee pointed out that some important
differences between classes exist and agreed that the
benefits are not necessarily identical among members of
any class. For example, better outcomes in preventing
congestive heart failure are evident for ACE inhibitors,
angiotensin receptor blockers (ARBs), and diuretics,
whereas calcium channel blockers appear more useful in
preventing stroke. Committee members felt strongly that
advice regarding specific choices or preferred agents is
best served by guidelines writers rather than drug labels.
The committee discussed whether ACE inhibitors and
ARBs should be considered as a single class. However,
several members pointed out that ARBs lack bradykininmediated biological effects, and there was general agreement that it would be inappropriate to treat these 2
categories as the same class of drugs.
The committee also recommended that specific trial data,
when adequately reviewed by the US Food and Drug Administration, are always relevant and should be included in the
label when appropriate. This approach preserves the incentive
of the industry to perform clinical outcomes trials. The
committee suggested that labeling for each drug should
distinguish whether the specific agent or the drug class
Day 1: Class Labeling for
Antihypertensive Drugs
The Advisory Committee was asked to consider the desirability of class labeling for antihypertensive drugs. Antihypertensive drugs, with few exceptions, have no clinical
outcomes claims in their labeling because approval of
these agents is usually based on demonstration of efficacy
at lowering blood pressure, not reducing morbidity or
mortality. Initial placebo-controlled trials of antihypertensive agents were conducted more than 2 decades ago.
Accordingly, it has been considered ethically inappropriate
to perform placebo-controlled studies of modern antihypertensive agents. These factors have resulted in a clinical
conundrum in which agents believed to be highly effective
at reducing morbidity and/or mortality cannot claim such
benefits. The Advisory Committee meeting was convened
to consider how labeling should address the relationship
between blood pressure and outcome.
The committee felt strongly that not having outcome data
available in antihypertensive drugs labels prevents optimal
education of practitioners regarding the benefits of important
therapeutic agents. Several committee members pointed out
that undertreatment of blood pressure remains a major health
problem that might be addressed by more descriptive class
labeling of antihypertensive agents. The committee agreed
that scientific knowledge about antihypertensive drugs is
extensive and that lowering blood pressure represents one of
the best-studied surrogate outcome measures in clinical medicine. However, the committee expressed some concern
regarding “unintended consequences” of class labeling and
discussed the need for robust preapproval safety studies and
postmarketing surveillance. The industry representative on
the committee cautioned that class labeling might represent a
disincentive to the pharmaceutical industry to engage in
future clinical trials.
Specific Benefits Attributable to Blood
Pressure Reduction
The committee reviewed the effects of blood pressure reduction on specific cardiovascular and renal outcome measures.
After considerable discussion, the committee agreed that
From the Department of Cardiovascular Medicine, Cleveland Clinic Foundation, Cleveland, Ohio.
Correspondence to Steven E. Nissen, MD, Department of Cardiovascular Medicine, Cleveland Clinic Foundation, Cleveland, OH 44195. E-mail
[email protected]
(Circulation. 2005;112:2043-2046.)
2043
2044
Circulation
September 27, 2005
contributed to the available outcome data. Most members of
the committee felt that single trials are inadequately powered
to describe benefits with a high degree of confidence, and
many were comfortable with inclusion of some findings
derived from carefully performed meta-analyses.
Relevance of Epidemiological Data
The committee felt strongly that epidemiological studies are
less reliable and should not contribute significantly to labeling considerations. The committee also suggested that observational data on the relationship between blood pressure and
risk should not be included in drug labels. The committee
agreed that it is important, to a certain extent, to recognize the
convergence of blood pressure with other risk factors (control
of lipids, smoking cessation, weight loss, aerobic exercise,
etc). However, members agreed that, while relevant, the role
of co-existing risks such as hyperlipidemia in determining
optimal blood pressure targets should not be included in drug
labels. The committee recommended explicitly indicating
when a specific agent or class has no available outcome data.
Many committee members felt that not enough information
is known about the importance of dosing intervals to support
specific claims, although this information is relevant to
clinicians, and should be readily accessible in the pharmacokinetic and pharmacodynamic section of the label. There was
additional discussion about differences in the impact of
circadian rhythms among various medications, but the committee reached no conclusion about the relevance of such
information in labels. Some members emphasized that a
single blood pressure measurement does not provide an
adequate assessment of efficacy. There was general agreement that ambulatory blood pressure studies provide more
robust and consistent data and should be strongly encouraged
in drug development. There was considerable discussion
regarding optimal timing for increases in drug doses in initial
hypertension management. Suggestions included increasing
doses at 1-week intervals and including a statement regarding
the likely need for multiple drugs to achieve good control.
However, most committee members suggested that such
details should be left to guideline writers rather than the drug
label.
Optimal Choices for Initial Therapy
Additional discussion focused on the optimal choice of initial
drug therapy. The committee agreed that most data suggest
starting with diuretics; several members emphasized that this
approach is also recommended in the guidelines. However,
there was general agreement that the large Antihypertensive
and Lipid-Lowering Treatment to Prevent Heart Attack Trial
(ALLHAT) study did not support the superiority of any
specific drug class for the primary end point. Thus, all 3 drug
classes (diuretics, calcium channel blockers, and ACE inhibitors) are indistinguishable, and none should be given a
“first-line” preference. Many committee members felt that
this issue generally should be left to the guideline writers
rather than labeling.
The committee discussed the issue of when to add a second
drug, noting that labeling currently recommends starting a
second drug only after a single drug has proven inadequate at
its highest tolerated dose. The committee agreed that there is
no substitute for clinical judgment—some patients reach
dose-limiting toxicity at lower doses than others. Because
clinicians need to make complex decisions on the urgency
necessary to achieve blood pressure control, it would be
difficult to convey this principle in labeling.
The committee considered which labels should include the
results of ALLHAT. The committee agreed that the large size
of ALLHAT contributed greatly to knowledge about blood
pressure management, and a description of those results
should be included in drug labels. Several members suggested
that only the specific agents used in the ALLHAT study
should have this information included in their labels. There
was discussion regarding the uncertainty whether all diuretics
are likely to show the same benefits as chlorthalidone, the
agent used in ALLHAT. Similarly, uncertainty exists regarding the interchangeability of all calcium channel blockers.
The committee discussed statistical methodology and its
applicability based on ALLHAT to support noninferiority
claims for studied agents such as amlodipine.
Pediatric Considerations
The committee discussed issues related to pediatric studies of
blood pressure–lowering therapies. The committee did not
support the concept that the agency should require studies of
antihypertensive drugs in children before approval for use in
adults. Members felt, however, that the agency should promote studies in children by continuing to grant additional
exclusivity (a 6-month patent extension) for assessing the
effects of antihypertensive drugs in children. In discussing the
challenge of placebo-controlled trials in children, some members suggested that 1 to 3 months of treatment with a placebo
would be acceptable in some cases.
The committee considered the issues surrounding drugs
that increase blood pressure. The committee generally agreed
that it is difficult to extrapolate data from blood pressure–
reducing agents to predict the effect on clinical outcomes of
drugs that increase blood pressure. The committee agreed that
the issue is relevant, but concern was expressed about setting
a regulatory standard for labeling, especially without direct
clinical trial evidence. Members further identified the challenge that some drugs that increase blood pressure are
commonly used in patients at lower risk, ie, those with lower
baseline risk. The committee advised against “leaping to
conclusions” about blood pressure–increasing drugs without
solid evidence. However, members suggested that the greater
the magnitude of blood pressure increases and the longer the
duration of contemplated use, the more likely this effect
should appear in a warning.
Day 2: Hydralazine-Nitrate Combination for
Congestive Heart Failure in Blacks
The committee was asked to assess whether clinical trials
adequately support a claim that BiDil (hydralazine 75 mg
plus isorbide dinitrate 40 mg) improves outcome in patients
with congestive heart failure (CHF). The committee reviewed
3 trials, the Veterans Administration Cooperative Vasodilator–Heart Failure Trials (V-HeFT I and II) and the AfricanAmerican Heart Failure Trial (A-HEFT), to provide a recom-
Nissen
CV and Renal Drugs Advisory Committee Report
2045
mendation to the agency regarding approval of BiDil for use
in CHF patients. In considering this recommendation, the
committee reviewed the results of a post hoc analysis of
V-HEFT-I and II for the black subgroup. In committee
discussions, a strong consensus emerged that post hoc analyses of these 2 studies were supportive but alone provided
insufficient evidence for an approval recommendation. Accordingly, the committee considered the A-HEFT trial as the
primary basis for evaluation of efficacy and safety.
committee discussed whether the strength of evidence provided by this probability value was sufficient to support
approval. Typically, 2 trials meeting statistical efficacy are
required. Several committee members pointed out that all 3
components of the composite efficacy score showed benefit,
contributing to the robustness of the results. Others commented that the efforts to study a minority population with a
disproportionate burden of disease also warranted
consideration.
V-HEFT Trials
Questions From the Agency
V-HEFT I compared survival in male veterans with CHF who
were randomized to Hydralazine 75 mg plus isosorbide
dinitrate (ISDN) 40 mg QID, prazocin 5 mg QID, or placebo.
For all patients, there was a trend toward greater survival in
the hydralazine-ISDN group compared with placebo (hazard
ratio [HR], 0.78; log-rank P⫽0.09). However, in a post hoc
analysis, the differences between active treatment and placebo in black patients (n ⫽56) were larger and statistically
significant, showing a mortality rate of 9.7% compared with
17.3% for placebo (P⫽0.04). In V-HeFT II, enalapril and
hydralazine-ISDN were compared. For all patients, there was
a trend toward greater survival in the enalapril arm by the
log-rank test (P⫽0.08). For all patients, 2-year survival was
greater in the enalapril-treated group (18.0% versus 25.0%;
P⫽0.016). However, in black patients, survival was similar
for enalapril and hydralazine-ISDN (HR⫽1.01; P⫽0.96).
Conversely, for white patients, enalapril resulted in a 48%
lower mortality rate compared with hydralazine-ISDN
(P⫽0.02). These findings provided the basis for the A-HEFT
trial, which was designed to compare hydralazine-ISDN
(BiDil) with placebo in black patients receiving contemporary therapies for CHF.
Prior to considering questions from the Agency, the committee listened to 13 speakers at an open public hearing, some
supportive and others critical of the concept of labeling a drug
for use in a specific racial-defined group. Questions from the
Agency included a review of the factors responsible for the
non-significant results for the sponsor’s pre-specified perprotocol analysis (P⫽0.46). The committee discussed this
discrepancy, specifically the per-protocol analysis exclusion
of 60% of the intent-to-treat population. This occurred in
large part because of the early termination of the trial, which
introduced the potential for bias and reduced statistical
power. The committee felt strongly that the intent-to-treat
analysis is more valuable and should constitute the principal
efficacy assessment.
Subjects enrolled before the second interim analysis, when
sample size was reestimated, made up 30% of the total
patients and 42% of the events and showed a nominal 7%
lower risk of death on BiDil. Subjects enrolled after the
second interim analysis showed a 62% lower risk of death on
BiDil. The committee discussed these interim analyses and
decisions made by the DSMB during meetings throughout the
course of the trial. Several committee members warned of the
risk of interpreting statistical strength of evidence when
interim data are used to refine the original study hypothesis.
Specifically, questions were raised about the proper weighting to give to different portions of the sample size, depending
on the timing of the interim analyses. Concern was raised
about the potential bias introduced during the various unblinded examinations of the data and the decision to stop the
trial prematurely.
In interpretation of results, some committee members
thought that recognition should be given for the fact that very
few trials have been done in a solely black population. The
committee discussed the challenges in enrolling an adequate
sample of this restricted population. Some participants argued
that compromises such as those in question, while affecting
the strength of evidence, are necessary in light of the
exigency of doing such a trial in a subpopulation by a small
company with limited resources. Other committee participants disagreed, citing concern about lowering the bar on
strength of evidence. The consultant patient representative
provided an additional patient perspective of the advantages
that such a trial provided to the black population while
expressing appreciation for the complexity of the statistical
analysis.
A-HeFT Trial
The A-HeFT trial randomized 1050 patients to BiDil administered TID or placebo. The primary end point was an unusual
composite score that included all-cause mortality, time to
hospitalization for heart failure, and response to the Minnesota Living With Heart Failure questionnaire. By the sponsor’s and the statistical reviewer’s intent-to-treat analyses,
BiDil was associated with an improved composite risk score
(P⫽0.021 by the agency reviewer). All 3 components of the
composite showed statistically significant benefits of therapy.
All-cause mortality was reduced by 43% (P⫽0.012), time
until hospitalization was prolonged (P⬍0.001), and quality of
life score improved (P⫽0.003). The trial was terminated
early at the recommendation of the Data Safety and Monitoring Committee (DSMB) on the basis of an analysis of
mortality data from the first 1014 patients.
Although the trial met its primary prespecified end point,
the committee spent considerable time discussing the weight
of the statistical evidence and the conditions under which the
trial was terminated. Dr Thomas Fleming, a biostatistician,
discussed at considerable length the optimal methodology to
adjust for interim analyses and early termination and suggested a conservative approach for adjustment of the efficacy
probability value. This adjustment yielded a marginally
significant probability value between 0.04 and 0.05. The
Use of a Composite End Point
There was discussion about the use of a composite end point.
Several committee members suggested that this approach
2046
Circulation
September 27, 2005
might not represent the wisest choice. Most panel members
agreed that the robustness of the data was clearly undermined
by early termination, pointing out that this always reduces
access to valuable data. The committee also discussed the
impact of early termination on the ability to assess the
contribution of each of the components of the composite end
point. The committee discussed the potential toxicity of
hydralazine (drug-induced lupus development). Some members suggested that the threshold of concern is decreased
when considering a nonlethal side effect for a therapy shown
to prolong life. It was pointed out that historical data
regarding safety and efficacy for these components do exist
and can be considered.
Policy Considerations for Combination Drug
There was considerable discussion regarding the regulatory
precedent for development of combination drug products.
Typically, when considering such combinations, the agency
requires demonstration of the effects attributable to each
component. However, the committee expressed comfort in
relaxing this requirement for the BiDil combination in view
of the well-known effects of both components. Several
participants pointed out the ethical challenges associated with
new studies of components of a drug combination shown to
reduce mortality. The committee discussed the importance of
dose information while recognizing that it may be difficult to
study multiple doses in trials with clinical end points. The
committee agreed that this challenge is particularly daunting
as the number of components increases.
Population Likely to Benefit
A-HeFT enrolled only the patients in whom BiDil appeared
to provide optimal benefits in V-HeFT I and II, namely
self-identified blacks. The committee was asked to consider
whether the evidence supported an absence of benefit in
white patients. Furthermore, the committee was asked to
opine about whether labeling should restrict use to black.
Several open public hearing speakers and some committee
members pointed out that the US black population is heterogeneous and suggested that it remains unclear whether these
differences are genetic, social, economic, or health delivery
related. Others commended the agency for requesting this
study in the black population. Many committee members
agreed that labeling should include information about the
A-HEFT study population including only blacks.
There was some committee disagreement about the conclusions supported by A-HeFT for other patient populations.
Several committee members commented that the agency has
a responsibility to precisely describe the population for which
an approved drug has and has not shown benefit. Members
expressed that, when the evidence of effectiveness comes
from a population that we can define, such as self-identified
race, the finding is significant. There was discussion about
the future of genomic-based medicine and the hope that
pharmacogenomic information will be useful in identifying
drug responders in the future. Several members suggested
that self-identified race represents a crude but useful surrogate for genomic-based medicine.
Discussions cited additional information about the white
population available through V-HeFT I and II, both of which
illustrated a different response for white and black populations. From a statistical standpoint, there was discussion
about the unfavorable results in the white population in
V-HeFT, coupled with proactive exclusion of whites in
A-HeFT, leading to the reasonable conclusion that there was
less benefit to risk in this population. There was also
considerable discussion among committee members regarding the evidence for efficacy in subgroups classified by the
severity of CHF. The committee concluded that A-HeFT
demonstrated benefits for BiDil only in patients with New
York Heart Association class III heart failure.
Approval Vote
The committee voted unanimously in favor of approval of
BiDil for the treatment of heart failure. One of the 9 voting
participants thought that the approval should extend to the
general population. A second member expressed concern
about labeling based on self-designation of race and the
heterogeneity among this population. Other members were
not as conflicted, citing the disproportionate burdens of heart
failure in blacks and the challenges of performing clinical
trials in this subgroup. Other comments included the importance of developing effective treatment in the black population in light of disparities in health care. There was additional
discussion about the statistical strength of evidence of
A-HeFT, with general agreement about limitations in interpreting the individual components (heart failure hospitalizations, quality of life, and mortality). Despite these concerns,
the committee felt that the consistency of favorable effects on
components of the end point helped to overcome the limitations of using a single trial for approval. Further recommendations included encouraging the agency to emphasize, in
labeling, that the trial was performed in the context of optimal
use of ACE inhibitors and ␤-blockers, providing clear direction that this treatment should be considered an adjunct to
standard therapy.
Disclosure
Dr Nissen was a consultant to AstraZeneca, Atherogenics, Lipid
Sciences, Wyeth, Novartis, Pfizer, Sankyo, Takeda, Kowa, Isis
Pharmaceuticals, Sanofi-Aventis, Novo-Nordisk, Eli Lilly, Kos
Pharmaceuticals, Glaxo Smith Kline, Forbes Medi-tech, Roche, and
Merck–Schering Plough; has given lectures at the invitation of
AstraZeneca and Pfizer; and has received research support for
clinical trials from AstraZeneca, Eli Lilly, Takeda, Sankyo, SanofiAventis, Pfizer, Atherogenics, and Lipid Sciences. All honoraria
related to these relationships are paid directly to charity by the
sponsors.
KEY WORDS: drugs
䡲
heart failure
䡲
hypertension
CARDIOLOGY PATIENT PAGE
CARDIOLOGY PATIENT PAGE
Cardiac Resynchronization Therapy
A Better and Longer Life for Patients With Advanced Heart Failure
Srinivas Iyengar, MD; William T. Abraham, MD
H
eart failure is a condition in
which the heart muscle cannot
function properly. It can result
from the heart muscle becoming stiff
over time or from a gradual weakening
that results in a decreased ability to
pump blood to the body. The weakening of the heart muscle, called systolic
heart failure, is most commonly a result of processes such as heart attacks,
heart valve abnormalities, uncontrolled
hypertension, and viral illnesses, although a number of other conditions
can cause the heart to weaken as well.
What Is the Treatment for
Heart Failure?
Treating the underlying cause of the
heart failure is the first step of management, whether it is through medicine or surgery to repair a blocked
artery or a damaged heart valve or by
controlling blood pressure. Even if the
cause of the heart failure is unknown
(idiopathic), the medical regimen used
is the same as for the other causes of
heart failure. A sensible diet with salt
and fluid restriction, exercise, and
medications such as angiotensinconverting enzyme (ACE) inhibitors
and beta-blockers, often with diuretics
(sometimes called water pills), are
therapies aimed at relieving the stress
the weakened heart is undergoing.
What Can Be Done for
Patients Who Have
Symptoms of Heart Failure
Despite Treatment?
In cases in which an individual is still
suffering from the symptoms of heart
failure (such as shortness of breath at
rest or with minimal activity, fluid
retention despite diet control and diuretic use, and worsening or nonimproving heart pump function with evidence of electrical abnormalities on
electrocardiogram [ECG]), he or she
might be a candidate for biventricular
pacing or cardiac resynchronization
therapy (CRT).
CRT involves a pacemaker with 3
wires (see the Figure). A CRT pacemaker helps the failing heart pump
blood more effectively. CRT has been
proved over the past 10 years to help
improve a patient’s quality of life,
increase the ability for daily activity,
and even increase the lifespan of people suffering from heart failure.
How Does CRT Work?
The heart has 4 chambers: 2 upper
ones (right and left atria) and 2 lower
ones (right and left ventricles). CRT
works by electrically stimulating both
the left and right sides of the heart,
specifically the left and right ventricles, so that the heart can pump more
effectively. The left ventricle side of
the heart is the larger side and pumps
blood to the brain and body; it is the
side that is usually weakened in heart
failure. Weakening of the left ventricle
eventually leads to an electrical imbalance between the right and left sides of
the heart (as well as an imbalance in
the left ventricle itself). This results in
an inability of the left ventricle to
pump enough blood and in worsening
of the symptoms of heart failure. CRT
corrects this imbalance and helps the
left and right sides of the heart to
resume beating in unison. By doing so,
a greater quantity of blood is pumped
out to the body, and symptoms are
improved.
Who Is a Candidate for
CRT?
Patients who have been diagnosed
with heart failure from a weakened
heart muscle, who have worsening
symptoms despite diet and optimal
medical therapy, and who show elec-
The information contained in this Circulation Cardiology Patient Page is not a substitute for medical advice or treatment, and the American Heart
Association recommends consultation with your doctor or healthcare professional.
From the Division of Cardiovascular Medicine, Ohio State University, Columbus.
Correspondence to Srinivas Iyengar, MD, Division of Cardiovascular Medicine, Ohio State University, 473 W 12th Ave, Room 110P DHLRI,
Columbus, OH 43210-1252. E-mail [email protected]
(Circulation. 2005;112:e236-e237.)
e236
Iyengar and Abraham
CRT. 1, Pacemaker generator; 2, right atrial pacer wire; 3, right ventricular pacer wire;
and 4, coronary sinus (“left ventricular”) pacer wire.
trical abnormalities on their ECG can
be evaluated for CRT (see the Table).
Contemporary medical evidence has
shown that patients with heart failure
from a weakened heart muscle might
also benefit by adding a component to
the CRT pacemaker called a defibrillator (also called an implantable
cardioverter-defibrillator, or ICD). The
ICD component functions by electrically jolting the heart back to a normal
rhythm if any lethal heart rhythms
might occur. Your cardiologist will be
able to decide which device is most
beneficial for you given your medical
condition. (For additional information
Indications for CRT
• Worsening symptoms of heart failure (for
example, shortness of breath, fluid retention,
fatigue) despite optimal medical therapy,
exercise, and diet
• Decreased function of the heart muscle
documented by an imaging test of the heart
(echocardiogram or ventriculogram)
• Evidence of electrical abnormalities in the
heart rhythm (for example, a “widened” QRS
complex seen on an ECG)
on ICDs, please see: Reiffel JA, Dizon
J. The implantable cardioverterdefibrillator: patient perspective. Circulation. 2002;105:1022–1024.)
How Is the CRT
Pacemaker Placed?
A specialized cardiologist, either an
electrophysiologist or a cardiothoracic
surgeon, generally inserts the CRT device. The pacemaker device itself is
the size of a half-dollar coin and is
usually placed under the collarbone in
the left upper shoulder (although it can
be placed on the right side as well).
The device is inserted under local anesthesia with x-ray guidance directly
into veins that lead into the heart
chambers. If it is anticipated that there
may be difficulty in placing the leads
or if another surgery is required (for
example, coronary bypass surgery), a
surgeon can place the leads on top of
the heart in the operating room under
general anesthesia.
What Happens After the CRT
Pacemaker Is Implanted?
The procedure to place the CRT pacemaker usually takes 2 to 3 hours. The
Advanced Heart Failure
e237
patient will stay overnight in the hospital after the procedure is performed.
During the hospitalization, the device
will be tested electronically by taking
readings and performing an echocardiogram to obtain a sonar picture of
the heart. If the device is working
properly and the patient is doing fine,
he or she can go home the next day.
Patients who have had a CRT pacemaker placed should avoid extreme
arm motions and lifting, such as raising the arm on the side in which the
pacemaker was implanted above the
head, shoveling snow, and golfing for
at least 6 weeks after implantation.
Microwave ovens will not damage the
pacemaker, although cell phones
should be used on the opposite side
from where the pacemaker was placed.
At discharge, patients should have a
scheduled appointment with either the
implanting physician or a pacemaker
clinic to check for proper functioning
of the pacemaker.
How Long Until CRT Works?
Two thirds of patients who have undergone CRT implantation have reported an immediate improvement in
their symptoms. It should be stressed,
however, that nearly a third of patients
who undergo implantation of a CRT
pacemaker do not improve. The reasons for this are still being investigated. A majority of patients do derive
some benefit from this form of therapy, and it is instituted only after other
conventional forms of treatment have
already been implemented.
For More Information:
For additional information, visit www.
americanheart.org/chf or www. abouthf.
org.
Disclosure
Dr Abraham has received research grants
from Medtronic and Biotronik; has served
on the speakers’ bureaus of and/or received
honoraria from Medtronic, Guidant, and St.
Jude; and has served as a consultant to or
on the advisory boards of Medtronic and St.
Jude.
Development of a Cardiac Neocavity After Mechanic
Double-Valve Replacement
Evaluation by Cardiac Magnetic Resonance Imaging
Achim Barmeyer, MD; Kai Muellerleile, MD; Gunnar K. Lund, MD; Alexander Stork, MD;
Nils Gosau, MD; Andreas Koops, MD; Sebastian Gehrmann, MD; Thomas Hofmann, MD;
Ditmar H. Koschyk, MD; Claus Nolte-Ernsting, MD; Gerhard Adam, MD; Thomas Meinertz, MD
A
57-year-old woman presented with signs of recurrent
right ventricular congestion and dyspnea. She had a
history of mechanic aortic and mitral valve replacement
resulting from acute endocarditis 7 years ago. At that time,
endocarditis resulted in destruction of the aortic-mitral fibrous curtain and part of the left ventricular (LV) outflow
tract, rendering implantation of prosthetic valves difficult.
The cuffs of both prosthetic valves were sutured together
directly to ensure stable fixation.
Current echocardiographic evaluation revealed good LV
contraction, normal function of the mechanic valves, and
normal diameters of both ventricles. An arterially supplied
neocavity was found between both atria (Figure 1). Transesophageal echocardiography suggested a dehiscence of the
aortic valve sutures, resulting in blood flow between the
aortic root and the neocavity (Movie I). However, further
echocardiographic evaluation was limited by ultrasonic extinction caused by the prosthetic valves.
Cardiac MR cine imaging (1.5 T, Magnetom Vision, Siemens) allowed detailed evaluation of the complex anatomical
findings. The neocavity was located between both atria and
extended from the superior vena cava to the diaphragm in the
craniocaudal orientation (Figure 2). A main finding was that the
neocavity was supplied by a large opening of 1.5 cm2 located
caudal to the aortic and medial to the mitral valve prosthesis. The
systolic increase in size of the neocavity suggested unrestricted
blood flow between the LV and the neocavity (Movies II and
III). Furthermore, a small, predominately diastolic jet was seen
originating from the supravalvular aortic root (Movie III),
confirming the suture dehiscence seen on transesophageal echo-
cardiography. Three-dimensional reconstruction of contrastenhanced T1-weighted FLASH images enabled excellent visualization of the neocavity relative to surrounding cardiovascular
structures (Movie IV). Despite recurrent cardiac failure, the
patient refused surgical repair and was discharged after recompensation and medical therapy.
Figure 1. Transthoracic echocardiography obtained in the subcostal
4-chamber view showed a neocavity between the right and left
atrium (white arrows). Movie I, left, Transesophageal echocardiography obtained at the level of the prosthetic aortic valve (AV) showed
the neocavity (NC) between the left atrium (LA) and right atrium (RA).
Right, Color Doppler imaging revealed blood flow between the aortic root and the neocavity (black arrow), suggesting a dehiscence of
the aortic valve sutures. Detailed echocardiographic evaluation was
limited by ultrasonic extinction caused by the prosthetic valves.
From the Universitäres Herzzentrum (A.B., K.M., G.K.L., N.G., T.H., D.H.K., T.M.), Klinik für Diagnostische und Interventionelle Radiologie (A.S.,
A.K., K.N-E., G.A.), and Institut für Informatik (S.G.), Universitätsklinikum Hamburg-Eppendorf, Hamburg, Germany.
The online-only Data Supplement, which contains Movies I through IV, is available at http://circ.ahajournals.org/cgi/content/full/112/13/e238/DC1.
Correspondence to Gunnar K. Lund, MD, Universitäres Herzzentrum, Klinik und Poliklinik für Kardiologie/Angiologie, Universitätsklinikum
Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. E-mail [email protected]
(Circulation. 2005;112:e238-e239.)
e238
Barmeyer et al
Cardiac Neocavity After Valve Replacement
e239
Figure 2. Cardiac MRI showed the spatial extent of the neocavity (black arrows) in 4- and 2-chamber views. A main finding was that
the neocavity was supplied by a large opening of 1.5 cm2 located caudal to the aortic and medial to the mitral valve prosthesis (asterisk). Movies II and III, Cine MRI showed a systolic increase in the size of the neocavity, indicating unrestricted blood flow between the
LV and neocavity. Furthermore, a small, predominately diastolic jet was seen originating from the supravalvular aortic root (Movie III),
confirming the suture dehiscence seen on transesophageal echocardiography. Movie IV, Three-dimensional reconstruction of contrastenhanced T1-weighted FLASH images enabled excellent visualization of the neocavity relative to surrounding cardiovascular structures.
The right atrium and right ventricle were removed for better visibility of the neocavity (yellow). The left atrium, left ventricle, aorta, and
pulmonary vessels are shown in blue.
Giampaolo Zoffoli, MD; Tiziano Gherli, MD
A
n 82-year-old woman admitted to hospital for endocarditis caused by Staphylococcus aureus presented, after
medical treatment, some episodes of angina. Echocardiography showed a mitroaortic junction abscess communicating
with the left ventricle (Figures 1 and 2).
Angina became unstable, with marked ST-segment depression not responding to treatment. Coronary ischemia in
endocarditis is generally due to preexisting coronary disease
or occasionally is a result of embolism from vegetations.
Coronary angiography showed a severe long stenosis of the
left main trunk in systole in the presence of normal distal
vessels (Figure 3). In diastole, the whole coronary tree was
normal (Figure 4), suggesting an extrinsic coronary
compression.
Emergency surgery confirmed the presence of a periaortic
cavity in the mitroaortic junction (Figure 5), residual of an
abscess, that was closed with a Dacron patch.
Postoperative stay was complicated by acute renal failure;
the patient was discharged after 19 days. A 3-month
follow-up showed a restored good quality of life.
Figure 1. Echocardiography of a mitroaortic junction abscess.
Figure 2. Echocardiography color Doppler shows the mitroaortic
junction abscess communicating with the left ventricle.
From Cattedra e Divisione di Cardiochirurgia, Università degli Studi di Parma, Italy.
The online-only Data Supplement is available at http://circ.ahajournals.org/cgi/content/full/112/13/e240/DC1.
Correspondence to Giampaolo Zoffoli, MD, Cattedra e Divisione di Cardiochirurgia, Università degli Studi di Parma, Italy, Via Gramsci, 14-43100
Parma, Italy. E-mail [email protected]
(Circulation. 2005;112:e240-e241.)
e240
Zoffoli et al
Figure 3. Coronary angiography shows a severe long stenosis
of the left main trunk in systole in the presence of normal distal
vessels.
Figure 5. Periaortic cavity, residual of an abscess, in the
mitroaortic junction. Echocardiography color Doppler (Echo
Cine) shows the pulsatile flow into the mitroaortic junction
abscess.
Figure 4. In diastole, the whole coronary tree is normal.
e241
Late Enhancement of a Left Ventricular Cardiac Fibroma
Assessed With Gadolinium-Enhanced Cardiovascular
Magnetic Resonance
Francesco De Cobelli, MD; Antonio Esposito, MD; Renata Mellone, MD; Marco Papa, MD;
Tiziana Varisco, MD; Roberto Besana, MD; Alessandro del Maschio, MD
A
4-year-old patient was referred to our department and
underwent gadolinium-enhanced cardiovascular MR
(CMR) because of a suspected left ventricular mass. The
infant was referred for an ECG before dental surgery; ECG
revealed negative T waves in leads D1, AVL, V4, V5, and V6.
Echocardiography showed a heterogeneous mass of the left
ventricular lateral wall inside the pericardial space, 4.0⫻2.0
cm in size, with reduced lateral wall motion. The global left
ventricular function was normal.
CMR was performed under sedation with a 1.5-T magnet
(Gyroscan Intera Master, Philips Medical System) with a cardiac
phased-array multicoil. First, “black-blood” multiplanar morphological (Figure 1A) without and with fat suppression and
cine “bright blood” balanced fast field-echo images were obtained (Figure 1B). These images revealed a hypointense mass
arising within the left ventricular free wall, suggestive of cardiac
fibroma. Dynamic contrast-enhanced CMR perfusion images
were acquired during gadolinium-DTPA (0.1 mmol/kg) injection; after 10 minutes, 3D segmented IR-GRE technique to
evaluate the delayed enhancement phase (DECMR) adjusting
inversion time (230 ms) to null normal myocardium was
performed. In early perfusion phase, the tumor demonstrated no
contrast enhancement (Figure 2), but in the delayed phase, the
tumor showed intense enhancement with central hypointensity
compared with normal myocardium (Figure 3A and 3B), suggesting the nature of fibroma. The explanation of this late
hyperenhancement pattern on DECMR is that microscopically
fibromas are a collection of fibroblasts interspersed among large
amounts of collagen. It is known that gadolinium bound to
DTPA diffuses into the interstitial space but not across cell
membranes. In fibromas, there is a great extracellular space for
gadolinium accumulation, and the distribution kinetics are
slower than normal myocardium; these phenomena result in a
delayed and persistently higher relative concentration of gadolinium with late enhancement.
With these CMR findings, the infant was diagnosed as having
left ventricular fibroma, and he did not undergo endomyocardial
biopsy and surgical excision because of the absence of symptoms and the high level of risk for the patient’s life. At a 6-month
follow-up, no changes in clinical symptoms and ECG signs were
found; moreover, at an echocardiography performed at the same
time, no changes of the mass were shown, thus confirming the
benign nature of the lesion.
Figure 1. Black-blood (A) and bright-blood (B) long-axis
4-views show a large hypointense mass into the free wall of
the left ventricle, suggestive of cardiac fibroma.
From the Department of Radiology (F.D.C., A.E., R.M., A.d.M.) and Division of Cardiology (M.P.), San Raffaele Scientific Institute, and Pediatric
and Neonatology Department, Desio Hospital (T.V., R.B.), Milan, Italy.
Correspondence to Francesco De Cobelli, MD, Department of Radiology, San Raffaele Scientific Institute, Via Olgettina 60, Milano 20132, Italy.
(Circulation. 2005;112:e242-e243.)
e242
De Cobelli et al
MR Late Enhancement of a Cardiac Fibroma
e243
Figure 2. Short-axis image in early perfusion phase after gadolinium injection shows the hypoperfused mass encircled by a
thin layer of normal perfused myocardium.
Figure 3. In the long-axis (A) and short-axis (B) postcontrast late
enhancement images, the signal of normal myocardium is
nulled, whereas the mass shows intense enhancement resulting
from the accumulation of gadolinium, with little central hypointense area.
Correspondence
Letter Regarding Article by Sega et al,
“Prognostic Value of Ambulatory and Home Blood
Pressures Compared With Office Blood Pressure
in the General Population”
blood pressure and ambulatory blood pressure: results from a Danish
population survey. J Hypertens. 1998;16:1415–1424.
5. Hansen TW, Jeppesen J, Rasmussen S, Ibsen H, Torp-Pedersen C. Ambulatory blood pressure and mortality: a population-based study. Hypertension. 2005;45:499 –504.
To the Editor:
We read with interest the paper by Sega et al regarding the
prognostic value of ambulatory, home, and office blood pressure
in the PAMELA population.1 However, we find that the main
conclusions of the report may be driven by the lack of adjustment
for confounders. The relationships between level of blood
pressure and risk were not adjusted for age, which may have a
major influence on risk over a long time span. There is indeed a
relation between age and blood pressure,2 and therefore, these
results may be biased. The comparisons of the various blood
pressures were also not adjusted for potential confounders, with
the argument that “no adjustment for age, sex, and other
cardiovascular risk factors was made because comparisons between the predictive value of various blood pressure values
involved the same sample.” However, it has been shown in a
general Belgian population that the within-subject differences
between office and ambulatory blood pressure measurements
increased with older age and greater body mass index.3 In
addition, in the Danish MONICA population, the within-subject
differences between office and ambulatory blood pressure measurements increased with older age, diagnosis of hypertension,
male gender, and presence of diabetes.4 So, to assess the true
prognostic value of office blood pressure versus that of ambulatory blood pressure, it is mandatory to explore whether adjustments for other relevant cardiovascular risk factors would change
the results. Recently, it was shown in the Danish MONICA
population that ambulatory blood pressure was a much better
predictor of all-cause mortality and cardiovascular mortality than
office blood pressure, taking other relevant risk factors into
account.5
Accordingly, to make the results from previous studies comparable to the PAMELA study, we would like to know the results
of adjusted analyses. Until that time, the conclusion that “office,
home, and ambulatory blood pressures are similarly predictive of
the risk of cardiovascular and all-cause death” needs to be
interpreted with caution.
To the Editor:
Sega and coworkers1 recently replicated the work of other
investigators.2,3 In Italians randomly recruited from the population of Monza and enrolled in the Pressioni Arteriose Monitorate
e Loro Associazioni (PAMELA) study,1 they confirmed that
per mm Hg, the risk of all-cause and cardiovascular mortality
increased more with systolic than diastolic blood pressure (BP),
more with nighttime than daytime ambulatory BP, and more with
home or ambulatory than conventional BP measurement.
While confirmatory, the report by Sega et al leaves many
issues unaddressed. First, it deviates from current standards by
not accounting for sex, age, and other cardiovascular risk factors.
The authors argued that comparisons between the various types
of BP measurement involved the same subjects and confounding
factors; however, we previously demonstrated in 2 independent
populations that the parameters of the relations between BP and
age or body mass index significantly differed according to the
type of BP measurement.4 Thus, in Cox regression, the relative
hazard ratios associated with each type of BP measurement
might be substantially different depending on the inclusion of
other explanatory variables. For instance, in Belgian and Irish
subjects, the within-subject differences between office and ambulatory blood pressure measurements increased with age and
body mass index.4 Second, Sega et al presented the likelihood
ratio test statistic only for comparisons of various combinations
of systolic BP measurement in relation to cardiovascular mortality. They did not report these test statistics for comparisons
between the different types of BP measurement, between daytime and nighttime BP, or between systolic and diastolic BP.
Third, over the last decennium, the introduction of invasive
therapies drastically reduced the case-fatality rate of major
cardiovascular complications, in particular those related to the
coronary complications of hypertension. The report by Sega et al
spans 10.9 years of follow-up that ended on December 31, 2003.
This probably explains why the 56 cardiovascular deaths only
represented 30.1% of all-cause mortality.1 Not accounting for
nonfatal events is important for the generalization of the results
of the study by Sega et al. Finally, in Figures 2, 3, and 4, Sega
and coworkers duplicated the unadjusted results already presented in the continuous risk functions given in Figure 1. Because of
the low number of cardiovascular deaths, the vertical scale of the
Kaplan-Meier estimates only spanned 5%. Because these estimates remained unadjusted for confounders, they cannot be
extrapolated to other populations with different age distribution
or cardiovascular risk profiles.
Tine Willum Hansen, MD
Research Center for Prevention and Health
Copenhagen, Denmark
Department of Cardiology
Bispebjerg University Hospital
Copenhagen, Denmark
Jørgen Jeppesen, MD, DMSc
Hans Ibsen, MD, DMSc
Medical Department M
Glostrup University Hospital
Copenhagen, Denmark
Eamon Dolan, MD
Eoin T. O’Brien, MD
Beaumont Hospital
Dublin, Ireland
1. Sega R, Facchetti R, Bombelli M, Cesana G, Corrao G, Grassi G, Mancia
G. Prognostic value of ambulatory and home blood pressures compared
with office blood pressure in the general population: follow-up results
from the Pressioni Arteriose Monitorate e Loro Associazioni (PAMELA)
study. Circulation. 2005;111:1777–1783.
2. Kannel WB. Historic perspectives on the relative contributions of diastolic and systolic blood pressure elevation to cardiovascular risk profile.
Am Heart J. 1999;138:205–210.
3. Staessen J, O’Brien E, Atkins N, Bulpitt CJ, Cox J, Fagard R, O’Malley
K, Thijs L, Amery A. The increase in blood pressure with age and body
mass index is overestimated by conventional sphygmomanometry. Am J
Epidemiol. 1992;136:450 – 459.
4. Rasmussen SL, Torp-Pedersen C, Borch-Johnsen K, Ibsen H. Normal
values for ambulatory blood pressure and differences between casual
Jan A. Staessen, MD
Hypertension Unit
University of Leuven
Leuven, Belgium
1. Sega R, Faccheti R, Bombelli M, Cesana G, Corrao G, Grassi G, Mancia
Study. Circulation. 2005;111:1777–1783.
e244
Correspondence
2. Hansen TW, Jeppesen J, Rasmussen S, Ibsen H, Torp-Pedersen C. Ambulatory blood pressure and mortality: a population-based study. Hypertension. 2005;45:499 –504.
3. Dolan E, Stanton A, Thijs L, Hinedi K, Atkins N, McClory S, Den Hond
E, McCormack P, Staessen JA, O’Brien E. Superiority of ambulatory
over clinic blood pressure measurement in predicting mortality: the
Dublin outcome study. Hypertension. 2005;46:156 –161.
4. Staessen J, O’Brien E, Atkins N, Bulpitt CJ, Cox J, Fagard R, O’Malley
K, Thijs L, Amery A. The increase in blood pressure with age and body
mass index is overestimated by conventional sphygmomanometry. Am J
Epidemiol. 1992;136:450 – 459.
To the Editor:
Sega and colleagues1 compared the prognostic value for
mortality risk of home and ambulatory blood pressure (BP)
measurement with office BP measurement in a Italian general
population using 11-year follow-up data from the Pressioni
Arteriose Monitorate e Loro Associazioni (PAMELA) study.
They reported that the overall ability to predict death was not
greater for home and ambulatory than for office BP
measurement.
In a Japanese general population (the Ohasama study), we
previously reported that the prognostic value of home and
ambulatory BP measurement was superior to office BP measurement for mortality risk.2,3 More recently, using 10-year
follow-up data, we have demonstrated that home and ambulatory
BP measurement is also superior to office BP measurement in
predicting the risk of stroke.4,5
The mostly nonsignificant difference among predictive powers
of the 3 methods of BP measurement in the PAMELA study1
would be attributable to the smaller statistical power due to a
smaller number of events (56 cardiovascular deaths1) compared
with the recent data of the Ohasama study (136 fatal and nonfatal
stroke events4 and 152 composite events of cardiovascular death
and nonfatal stroke5). The less remarkable predictive power of
home BP in the study by Sega et al would also be attributable to
the smaller number of home BP measurements in the PAMELA
population (average of only 2 home BP values obtained in the
morning and in the evening within a day) compared with that
used in our Ohasama study (average of 21 home BP values),
because we demonstrated that the predictive power of home BP
measurement linearly increased with an increase in the number
of home BP measurements taken.4 The authors’ conclusions
from the PAMELA study should be interpreted with caution,
because their results might not be applicable to the predictive
power of home BP obtained by multiple self-measurements.
Comparison of the prognostic value of multiple selfmeasurements of home BP and ambulatory BP awaits further
follow-up results from the Ohasama study.
Takayoshi Ohkubo, MD, PhD
Department of Planning for Drug Development and Clinical
Evaluation
Graduate School of Pharmaceutical Science
Tohoku University
Sendai, Japan
Yutaka Imai, MD, PhD
Department of Clinical Pharmacology and Therapeutics
Graduate School of Pharmaceutical Science and Medicine
Tohoku University
Sendai, Japan
Study. Circulation. 2005;111:1777–1783.
2. Ohkubo T, Imai Y, Tsuji I, Nagai K, Kato J, Kikuchi N, Nishiyama A,
Aihara A, Sekino M, Kikuya M, Ito S, Satoh H, Hisamichi S. Home blood
e245
pressure measurement has a stronger predictive power for mortality than
does screening blood pressure measurement: a population-based observation in Ohasama, Japan. J Hypertens. 1998;16:971–975.
3. Ohkubo T, Imai Y, Tsuji I, Nagai K, Watanabe N, Minami N, Itoh O,
Bando T, Sakuma M, Fukao A, Satoh H, Hisamichi S, Abe K. Prediction
of mortality by ambulatory blood pressure monitoring versus screening
blood pressure measurements: a pilot study in Ohasama. J Hypertens.
1997;15:357–364.
4. Ohkubo T, Asayama K, Kikuya M, Metoki H, Hoshi H, Hashimoto J,
Totsune K, Satoh H, Imai Y. How many times should blood pressure be
measured at home for better prediction of stroke risk? 10-year follow-up
results from the Ohasama study. J Hypertens. 2004;22:1099 –1104.
5. Ohkubo T, Kikuya M, Metoki H, Asayama K, Obara T, Hashimoto J,
Totsune K, Hoshi H, Satoh H, Imai Y. Prognosis of “masked hypertension” and “white-coat” hypertension detected by 24-h ambulatory
blood pressure monitoring: 10-year follow-up from the Ohasama study.
J Am Coll Cardiol. 2005;46:508 –515.
Response
We read with interest the letters commenting on the results of
our study.1 Our reply to the remarks are as follows.
Home blood pressure (BP). Our data do not disagree with the
results of the Ohasama Study2 on the clinical importance of home
BP, because (1) the goodness-of-fit to the model predicting
cardiovascular or all-cause mortality was, if anything, higher
than that of office or 24-hour BP measurement; (2) this was the
case even if only 2 home BP measurements were available,
which did not allow us to fully explore the potentials of this
approach; and (3) combining office with home BP values
improved the predictive ability of the model.
Comparisons of statistical tests. The likelihood ratio test
cannot be used to compare differences in the goodness of fit
between different BPs. The question seems to us somewhat
irrelevant, however, because the goodness-of-fit values were not
lower for office than for ambulatory BP measurements,1 which
justifies our conclusion about the noninferiority of its prognostic
importance.
Adjustment for “confounders.” We decided (see Methods) not
to adjust for other variables because office, home, and ambulatory BPs were obtained at the same time in the same subjects, ie,
we dealt with a within-sample comparison that did not require
adjustments for differences that simply did not exist. Moreover,
in general, we are against the habit of drawing conclusions based
on extensive adjustment of original data because (1) this does not
guarantee that the role played by factors other than that under
study is eliminated, a goal that can be achieved, whenever
possible, by recollection of data devoid of the previously
observed differences, and (2) the statistical attempt to dissociate
the role of factors that are intimately related can be biologically
artificial and in some instances can distort the inherent features
of the phenomenon under study, thus introducing rather than
removing confounders. This applies to the increasing difference
between office and ambulatory or home BP with aging (reported
years ago in a PAMELA report3), which characterizes the overall
relationships and behavior of these pressures and thus must not
be arbitrarily corrected. The above does not exclude that other
factors can differently modify the effects of office, home, or
ambulatory blood pressure, eg, that gender, age, or blood glucose
interacts with office values differently than it does with ambulatory values. It also does not mean that the conclusion reached
for the whole population sample applies to all subsamples, eg,
old versus young subjects, males versus females, hypertensive
versus normotensive individuals; however, these are additional
issues that we did not address, also because of the limited number
of fatal events we could count on.
Novelty of data. It is somewhat contradictory to define the data
as “confirmatory” and at the same time disagree with their
conclusion. Our data confirm some previous results, often
obtained, however, in selected groups of subjects, and none of
e246
Correspondence
which had office, home, and ambulatory values all available. In
addition, the evidence comes from a large population and a very
long follow-up. Finally, much more than previous contributions,
the results emphasize the prognostic importance of office BP and
show the flatter slope of its relationship with events to be the
clearest prognostic difference from the other pressures.
Roberto Sega, MD
Rita Facchetti, PhD
Michele Bombelli, MD
Giancarlo Cesana, MD
Giovanni Corrao, MD
Guido Grassi, MD
Giuseppe Mancia, MD
Clinica Medica
Dipartimento di Medicina Clinica
Prevenzione e Biotecnologie Sanitarie
Universita Milano-Bicocca
Ospedale San Gerardo
Monza, Italy
study. Circulation. 2005;111:1777–1783.
2. Ohkubo T, Imai Y, Tsuji I, Nagai K, Kato J, Kikuchi N, Nishiyama A,
Aihara A, Sekino M, Kikuya M, Ito S, Satoh H, Hisamichi S. Home blood
pressure measurement has a stronger predictive power for mortality than
does screening blood pressure measurement: a population-based observation in Ohasama, Japan. J Hypertens. 1998;16:971–975.
3. Sega R, Cesana G, Milesi C, Grassi G, Zanchetti A, Mancia G. Ambulatory and home blood pressure normality in the elderly: data from the
PAMELA population. Hypertension. 1997;30:1– 6.
Contemporary Reviews in Cardiovascular Medicine
Diagnosis and Management of the Cardiac Amyloidoses
Rodney H. Falk, MD
C
thickening with nondilated ventricles. The ensuing elevation
of pressure in the thin-walled atria is associated with atrial
dilation, despite thickening of the atrial walls by amyloid
deposition.
Because cardiac involvement very frequently coexists with
significant dysfunction of other major organs, the initial
suspicion of cardiac amyloidosis is often triggered by the
recognition that the heart disease is part of a multiorgan
disorder. Conversely, if other organ dysfunction such as
nephrotic syndrome predominates, recognition of a cardiac
problem may be delayed because of the focus on these organ
systems. Because the clinical manifestations and progression
of the disease may vary considerably on the basis of the
amyloid fibril precursor, the various types of amyloid heart
disease are dealt with individually in this review.
ardiac amyloidosis is a manifestation of one of several
systemic diseases known as the amyloidoses.1,2 This
uncommon disease is probably underdiagnosed, and even
when a diagnosis of amyloidosis of the heart is made, the fact
that there are several types of amyloid, each with its unique
features and treatment, is often unrecognized. This can lead to
errors in management and in the information conveyed to the
patient. The purpose of this review is to familiarize the reader
with the clinical features of amyloidosis and to address the
approach to the patient with this disease, focusing on the
various types of amyloidosis, their prognosis and treatment.
The common feature of this group of diseases is the
extracellular deposition of a proteinaceous material that,
when stained with Congo red, demonstrates apple-green
birefringence under polarized light and that has a distinct
color when stained with sulfated Alcian blue (Figure 1).
Viewed with electron microscopy, the amyloid deposits are
seen to be composed of a ␤-sheet fibrillar material (Figure 2).
These nonbranching fibrils have a diameter3 of 7.5 to 10 nm
and are the result of protein misfolding.4,5 Cardiac involvement in amyloidosis may be the predominant feature or may
be found on investigation of a patient presenting with another
major organ involvement. The presence of cardiac amyloidosis and its relative predominance varies with the type of
amyloidosis. Thus, senile systemic amyloidosis and some
forms of transthyretin amyloidosis invariably affect the heart,
whereas cardiac involvement ranges from absent to severe in
amyloidosis derived from a light-chain precursor (AL amyloidosis). Secondary amyloidosis almost never affects the
heart in any clinically significant manner.6 The specific
composition of the fibrils differs in the different types of
amyloid7 and are outlined in the Table. Both on the basis of
common usage and for the sake of simplicity, “cardiac
amyloidosis” is used here to describe involvement of the
heart by amyloid deposition, whether as part of systemic
amyloidosis (as is most commonly the case) or as a localized
phenomenon.
Regardless of the underlying pathogenesis of amyloid
production, cardiac amyloidosis is a myocardial disease
characterized by extracellular amyloid infiltration throughout
the heart.8 Amyloid deposits occur in the ventricles and atria,
as well as perivascularly (particularly in the small vessels)
and in the valves. The conduction system may also be
involved. The infiltrative process results in biventricular wall
AL Amyloidosis
The commonest form of amyloidosis is that associated with a
plasma cell dyscrasia. Amyloid is produced from clonal light
chains, so the disease is referred to as AL amyloidosis. The
commonest plasma cell dyscrasia is multiple myeloma, and
AL amyloidosis overlaps with it. However, only a minority of
myeloma patients develop amyloidosis, and most patients
with AL amyloidosis do not have multiple myeloma. Although AL amyloidosis is considered an uncommon disease,
it has an incidence similar to better-known diseases such as
Hodgkin disease or chronic myelocytic leukemia,9 with an
estimated 2000 to 2500 new cases annually in the United
States. The heart in AL amyloidosis is affected in close to
50% of cases (Figure 3), and congestive heart failure is the
presenting clinical manifestation in about half of these patients.10 Even among patients in whom another organ system
dysfunction predominates, the presence of cardiac amyloidosis is frequently the worst prognostic factor.11 Once congestive heart failure occurs, the median survival is ⬍6 months in
untreated patients10,11; therefore, early recognition of the
disease and prompt initiation of therapy is critical.
Clinical Features
The typical patient with heart failure resulting from AL
amyloidosis frequently presents with rapidly progressive
signs and symptoms. Progressive dyspnea is common, almost
always associated with evidence of elevated right-sided
filling pressure. Peripheral edema may be profound, and in
From the Department of Cardiology, Harvard Vanguard Medical Associates, and Cardiovascular Genetics Center, Brigham and Women’s Hospital,
Boston, Mass.
Correspondence to Rodney H. Falk, MD, Department of Cardiology, Harvard Vanguard Medical Associates, 133 Brookline Ave, Boston, MA 02215.
(Circulation. 2005;112:2047-2060.)
2047
2048
Circulation
September 27, 2005
Figure 2. Myocardial biopsy in cardiac amyloidosis viewed
under electron microscopy. At the lower portion of the figure is
the edge of a myocyte; above it is a mass of amyloid fibrils.
Original magnification ⫻15 000.
Figure 1. A, Endomyocardial biopsy specimen, stained with
hematoxylin and eosin, from a patient with cardiac amyloidosis.
The amyloid stains light pinkish red and is seen as an amorphous material that separates the darker staining myocytes. B,
Staining of the tissue from the same patient using sulfated
Alcian blue. The amyloid stains turquoise green and the myocytes stain yellow, characteristic of amyloid. Courtesy of Dr
Gayle Winters, Brigham and Women’s Hospital. Boston, Mass.
late-stage disease, ascites is not uncommon. Weight loss,
which is common, may represent the effects of the systemic
disease or may be a manifestation of cardiac cachexia.
Patients with cardiac amyloidosis may present with chest
discomfort. Most commonly, this is not typical of angina and
is associated with congestive heart failure, but typical angina
can occur because of involvement of the small vessels of the
heart.12 Imaging studies may be positive, leading to cardiac
catheterization with apparently normal epicardial coronary
arteries on coronary angiography. Myocardial flow reserve in
such patients is impaired13 because of the small vessel
involvement, and a small but persistent elevation in serum
troponin may be present, leading to a misdiagnosis of
non–Q-wave infarction.14 –18 Presumably, the troponin elevation represents ongoing myocyte necrosis, and it has been
shown be a negative prognostic factor.15–17 Small vessel
cardiac amyloid may occur in the absence of wall thickening
on the echocardiogram, although there is often a mild
elevation of left ventricular filling pressure, suggesting diastolic abnormalities of the ventricle. This presentation of
amyloidosis is rare; it is seen in only 1% to 2% of patients
with cardiac involvement.
Although sudden death is common in AL amyloidosis,
ventricular arrhythmias are an uncommon presenting fea-
ture.19 Monitored sudden death in severe cardiac amyloid is
often found to have been due to electromechanical dissociation rather than ventricular arrhythmia; in this, amyloidosis is
similar to other forms of very severe heart disease.20 The
management of syncope is discussed below, but a careful
history may help distinguish arrhythmia-induced syncope
from other sources such as autonomic neuropathy. Sustained
ventricular tachycardia or resuscitation from ventricular fibrillation is a rare presenting manifestation that occurs in
patients with less severe heart failure, presumably because
patients with more advanced disease do not survive an initial
episode.
Fewer than 5% of patients with AL amyloidosis involving
the heart have clinically isolated cardiac disease.10 Complaints of noncardiac symptoms should be sought because
their presence is a clue to the systemic nature of the disease.
The patient should be carefully questioned about dizziness
and syncope with emphasis on the positional nature of any
such symptoms because there are several potential mechanisms of syncope in amyloidosis.20 Dermatological manifestations such as easy bruising and periorbital purpura may
occur21,22; the latter is virtually pathognomonic of the disease.
Macroglossia, characterized by a stiffening and enlargement
of the tongue, often with tooth indentation, is seen in 10% to
20% of patients and sometimes produces dysphonia or
dysgeusia. It may occasionally be profound enough to interfere with eating, swallowing, or breathing. A subtle change in
the voice (particularly hoarseness toward the end of the day),
a quite common complaint, probably represents vocal cord
involvement.23 Neurological symptoms include carpal tunnel
syndrome and peripheral and autonomic neuropathy. Right
upper quadrant discomfort may be due to hepatic congestion
or with amyloid hepatic infiltration.24 Carpal tunnel syndrome often precedes other organ involvement by a few
Falk
Cardiac Amyloidosis
2049
Summary of the Main Forms of Amyloidosis That Affect the Heart
Nomenclature
Precursor of Amyloid Fibril
Organ Involvement
Treatment
Comment
AL
Immunoglobulin light chain
Heart
Kidney
Liver
Chemotherapy
Plasma cell dyscrasia related to (but
usually not associated with) multiple
myeloma
Peripheral/autonomic nerves
Soft tissue
Gastrointestinal system
Heart disease occurs in 1/3 to 1/2 of AL
patients; heart failure tends to progress
rapidly and has a very poor prognosis
ATTR (familial)
Mutant transthyretin
Peripheral/autonomic nerve
Heart
Liver transplantation
? New pharmacological
strategies to stabilize the TTR
Autosomal dominant; amyloid derived
from a mixture of mutant and wild-type
TTR; if present before, cardiac amyloid
may progress despite liver transplantation
AApoA1
Mutant apolipoprotein
Kidney
Heart
? Liver transplantation
Kidney disease is the commonest
presentation; heart involvement rare
Senile systemic amyloid
Wild-type transthyretin
Heart
Supportive
? New pharmacological
strategies to stabilize the TTR.
Almost exclusively found in elderly men;
slowly progressive symptoms
Serum amyloid A
Kidney
Heart (rarely)
Treat underlying inflammatory
process
Heart disease rare and, if present, rarely
clinically significant
Atrial natriuretic peptide
Localized to the atrium
None required
Very common; may increase risk of atrial
fibrillation and/or be deposited in greater
amounts in the fibrillating atrium
AA
AANP
years, and a history of surgical carpal tunnel release is not
uncommon. Although widespread lymphadenopathy is present in only a small minority of patients, submandibular
swelling caused by lymph node and salivary gland infiltration
is not uncommon and often is accompanied by macroglossia.
Nail dystrophy (brittle and slow-growing nails) is sometimes
seen, particular in the hands, and when present is a clue to the
systemic nature of the cardiac disease.25
The cardiovascular physical examination in a patient with
heart failure resulting from amyloidosis usually reveals sinus
rhythm with a normal to low radial pulse volume, although
atrial arrhythmias (most commonly atrial fibrillation) occur in
10% to 15% of patients. When present, atrial fibrillation is
associated with a very high incidence of thromboembolism.
The jugular venous pressure is often markedly elevated, and
the waveform is generally unrevealing, but occasionally, a
prominent X and Y descent is noted.26,27 In contrast to
constrictive pericarditis, with which it may be confused,
Kussmaul’s sign is very rarely present. The apex beat is
frequently impalpable and, when it can be felt, is generally
not displaced. The first and second heart sounds are usually
normal in character. A left ventricular third heart sound is
rarely heard, but in advanced cases, a right ventricular S3,
which often is associated with right ventricular dilation and
dysfunction on the echocardiogram, may be heard best at the
left parasternal edge. Despite the increased stiffness of the left
ventricle, a fourth heart sound is almost never present,
possibly because of atrial dysfunction resulting from amyloid
infiltration.28,29 Blood pressure is often low, even in the
absence of postural hypotension; this may represent decreased cardiac output in conjunction with early autonomic
dysfunction. Blood pressure may fall further on standing,
particularly if autonomic neuropathy is present,30 and should
be measured in the supine, seated, and standing positions both
immediately after standing and after at least 2 minutes
because the systolic pressure may continue to drift down in
the presence of autonomic dysfunction. Hypertension is
unusual, and in patients with a history of hypertension,
“spontaneous” resolution of hypertension over the preceding
few months is common. Examination of the chest may reveal
bilateral pleural effusions, but rales are rarely present, even in
association with advanced heart failure. The pleural effusions
in a patient with AL amyloidosis may simply represent heart
failure, but patients with cardiac amyloid may also have
pleural infiltration with amyloid, resulting in disproportionately large effusions that are diuretic resistant and rapidly
recur after a pleural tap.31
Splenomegaly is rare, whereas hepatomegaly is common
and is due either to congestion from right heart failure or to
amyloid infiltration.32,33 When extensive amyloid infiltration
of the liver is present, the organ is rock-hard and not tender,
often extending several centimeters below the costal margin
and crossing the midline. This contrasts with the firm,
sometimes tender, liver of heart failure. Peripheral edema
may be profound, and if it appears disproportionate to the
degree of heart failure, the possibility of associated nephrotic
syndrome should be considered. In addition to autonomic
dysfunction, amyloidosis may cause a sensory neuropathy,
and the patient may complain of numbness or painful extremities.11 A history of weight loss is common, and proteinuria,
frequently reaching nephrotic range (ⱖ3 g/24 h), coexists
with cardiac disease in 30% to 50% of cases.
Low voltage on the ECG (defined as all limb leads ⬍5 mm
in height) is found in a high proportion of patients and is often
associated with extreme left- or right-axis deviation (Figure
4). Although voltage criteria for left ventricular hypertrophy
have been described in the precordial leads of some patients
with AL amyloidosis, increased limb lead voltage is ex-
2050
Circulation
September 27, 2005
Figure 3. Autopsy specimen of a heart with extensive amyloid
infiltration. Note the nondilated ventricles with biventricular
thickening and the biatrial enlargement with thickening of the
atrial septum. Atrial infiltration leads to atrial failure and can be
associated with atrial thrombi.
tremely uncommon.34 When increased limb or precordial lead
voltage is present, it is frequently a result of an unrelated
coexistent condition such as hypertension. Interestingly, right
bundle-branch block is uncommon, and left bundle-branch
block is very unusual unless it is a preexisting condition.10
The reason for the virtual absence of left bundle-branch block
in patients with AL amyloidosis is unclear. However, because
amyloid deposition affects the heart uniformly, the more
vulnerable right bundle is anticipated to be involved before
the left bundle, so sparing of the right bundle with complete
left bundle-branch block would be very unlikely.
The echocardiographic features of advanced cardiac AL
amyloidosis are distinctive. The initial descriptions concentrated on patients with severe cardiac disease and depicted
nondilated ventricles with concentric left ventricular thickening, right ventricular thickening, prominent valves, and infiltration of the atrial septum. The myocardial texture was
abnormal and described as “granular sparkling.”26,35–37 Subsequent changes in image processing produced a myocardial
appearance that is less “granular” in appearance, but advanced amyloid heart disease still demonstrates an increased
echogenicity of the myocardium and often of the valves
(Figure 5). The classic appearance of a restrictive pattern by
Doppler echocardiography and associated increased echogenicity, biventricular thickening, and valvular infiltration is
limited to patients in the end stage of the disease. More
commonly, the ventricle appears thickened to a degree that is
disproportionate to the degree of current or prior hypertension, and the Doppler features depend on the stage of the
disease, with serial studies demonstrating a progression of
diastolic dysfunction as myocardial infiltration progresses.36
The left ventricular ejection fraction is normal or nearly
normal until late in the disease, and because the left ventricle
does not dilate, a reduced ejection fraction is associated with
a substantially reduced stroke volume. Because the thickening of the ventricle in amyloidosis is due to myocardial
infiltration rather than hypertrophy, the ECG limb lead
voltage tends to decrease as the ventricle thickens. This
results in a decreased ratio of voltage to left ventricular mass,
a finding that strongly suggests an infiltrative cardiomyopathy, of which amyloidosis is the commonest cause.38 In ⬇5%
of patients with cardiac amyloidosis, left ventricular infiltration may mimic hypertrophic cardiomyopathy on the echocardiogram.10,39,40 These patients often have normal or even
mildly hyperdynamic left ventricular function with normal
voltage on the ECG. Associated postural hypotension is
common in these patients, and low afterload may in part
account for the normal to increased ejection fraction. Unlike
true hypertrophic cardiomyopathy, ventricular hypertrophy
on the ECG limb leads is almost never seen and systolic
anterior motion of the mitral valve is uncommon, although
chordal anterior motion may be present with an associated
outflow tract murmur.
Doppler echocardiography is also useful in cardiac amyloidosis. In advanced disease, there is a restrictive transmitral
flow pattern characterized by a short deceleration time of the
E wave and a low-velocity A wave, with associated abnormalities in pulmonary venous flow.26,41,42 The decreased
transmitral A wave in AL amyloidosis is related not only to
late-stage restrictive pathophysiology but also to atrial amyloid infiltration, which results in intrinsic atrial dysfunction28,29,43– 46; thus, a normal deceleration time can be seen in
association with a diminutive A wave. Further insights into
cardiac function in AL amyloidosis can be gained by pulsed
tissue Doppler imaging, which can demonstrate the presence
of diastolic dysfunction more accurately than transmitral and
pulmonary flow and can provide evidence of longitudinal
systolic impairment before the ejection fraction becomes
abnormal.47,48 Strain and strain rate imaging are even more
sensitive than tissue Doppler, demonstrating long-axis dysfunction in early cardiac amyloidosis and often showing
disproportionate impairment of longitudinal contraction despite apparently preserved fractional shortening (Figure 6). In
addition to giving sensitive information about myocardial
function, tissue Doppler and strain and strain rate imaging
may have potential for evaluating the prognosis in AL
amyloidosis.47– 49
Other imaging modalities such as cardiac magnetic resonance show promise for diagnosing cardiac amyloidosis if
echocardiographic features are suspicious.50 Recent descriptions of cardiac MRI in advanced cardiac amyloidosis show
an unusual pattern characterized by global subendocardial
late gadolinium enhancement and associated abnormal myocardial and blood-pool gadolinium kinetics.50 However the
Falk
Cardiac Amyloidosis
2051
Figure 4. Typical appearance of the ECG in AL amyloidosis of the heart. There is low voltage with an abnormal axis and poor R-wave
progression in the precordial leads. The association of low voltage of this degree with thickening of the left ventricle on echocardiogram is highly suggestive of an infiltrative cardiomyopathy.
sensitivity of this technique for detecting early disease is not
known, and the specificity of the described abnormalities is
likely to be low in an unselected population of patients.51
Cardiac Catheterization
The noninvasive imaging features of amyloidosis described
above are usually sufficient to strongly suspect the correct
diagnosis. Thus, cardiac catheterization, other than to obtain
an endomyocardial biopsy, to better assess hemodynamics, or
to evaluate coronary anatomy, currently is of limited value in
the routine evaluation of a patient with suspected amyloidosis. Nevertheless, many patients with an eventual diagnosis
of cardiac amyloidosis undergo cardiac catheterization during
the workup, and if a full hemodynamic study is done, careful
examination of the pressure tracing may provide clues to the
diagnosis. Impaired ventricular filling in advanced cardiac
amyloidosis is associated with an elevated left ventricular
end-diastolic pressure, and the pressure tracings may reveal a
dip-and-plateau waveform52 (Figure 7). It has been suggested
that, unlike constrictive pericarditis, amyloidosis is associated
with a left ventricular end-diastolic pressure that exceeds
right ventricular end-diastolic pressure by at least 7 mm Hg.52
However, this is not always the case, and both disorders may
manifest a dip-and-plateau diastolic pressure tracing with
pressure equalization.53,54 A pulmonary artery systolic pressure ⬎50 mm Hg is rarely seen in “uncomplicated” constrictive pericarditis but may occur in cardiac amyloidosis,55 and
the finding of an inspiratory rise in right ventricular pressure
with an associated fall in left ventricular pressure, representing ventricular interdependence, has been proposed as a
specific sign of constrictive pericarditis that distinguishes it
from restrictive cardiomyopathy.56 However, although certain
hemodynamic clues suggest one diagnosis or the other,
overlap remains, and the diagnosis should not be made on
hemodynamic data alone. In suspected cases of amyloidosis,
clinical examination and review of the echocardiogram are
generally extremely valuable in favoring a diagnosis of
cardiac amyloidosis if present and should never be omitted.
Tissue Diagnosis
The diagnosis of amyloidosis requires a tissue biopsy that
demonstrates apple-green birefringence when stained with
Congo red and viewed under a polarizing microscope. Sulfated Alcian blue is an alternative stain with a high specificity
for amyloid57 (Figure 1). It is not necessary to biopsy the
heart if the echocardiographic appearances are typical for
cardiac amyloidosis, providing that a histological diagnosis
has been made from another tissue. Fine-needle aspiration of
the abdominal fat is a simple procedure that is positive for
amyloid deposits in ⬎70% of patients with AL amyloidosis.58,59 If the diagnosis is not confirmed by biopsy of
another tissue, endomyocardial biopsy is a safe and relatively
simple procedure in skilled hands; it is virtually 100%
sensitive because the amyloid is widely deposited throughout
the heart.60,61 In patients with known amyloid deposits in
other organs and a history of poorly controlled hypertension,
there may be uncertainty as to whether ventricular thickening
represents amyloid infiltration or hypertensive heart disease.
In such cases, endomyocardial biopsy may be helpful to
determine whether the heart is infiltrated with amyloid.
Once a tissue diagnosis of amyloid has been established,
the confirmation that this is AL amyloid requires a search for
the presence of a plasma cell dyscrasia. Serum and urine
immunofixation should be performed rather than serum and
urine electrophoresis because the amount of serum or urine
paraprotein may be small and immunofixation is a much
2052
Circulation
September 27, 2005
Figure 5. Selection of echocardiographic images from a patient with severe AL cardiac amyloidosis. Top left, Parasternal long-axis
view showing concentric left ventricular thickening with a pericardial effusion. The echogenicity of the myocardium is increased, and the
valves are seen unusually clearly suggestive of infiltration. Top right, Apical 4-chamber view showing normal biventricular dimensions
and biatrial enlargement. The atrial septum is thickened, and a pacemaker/ICD lead is seen in the right ventricle. Bottom left, Transmitral Doppler flow showing a restrictive pattern with a short deceleration time and reduced A-wave velocity. Bottom right, Tissue Doppler
recorded from the lateral mitral annulus. There is reduced myocardial velocity throughout systole and both phases of diastole, compatible with a restrictive pathophysiology.
more sensitive test. Even more sensitive is the recently
introduced serum free-light-chain assay, which can detect
circulating free light chains with ⬎10-fold sensitivity than
immunofixation62,63 This is a quantitative test. In AL amyloidosis, free lambda or (less commonly) free kappa levels are
elevated. The normal serum range of kappa free light chains
is 3.3 to 19.4 mg/dL; for lambda, 5.7 to 26.3 mg/dL with a
kappa-to-lambda ratio of 0.26 to 1.65.62,63 It is important to
assess the ratio of kappa to lambda free light chains because
they are renally excreted and renal impairment elevates kappa
and lambda levels without changing the ratio. In AL amyloidosis with renal impairment, elevated levels of both free
lambda and free kappa will be seen because renal impairment
reduces light-chain excretion. However, the kappa-to-lambda
ratio remains abnormal and should always be calculated in
addition to the absolute values. A kappa-to-lambda ratio
⬍0.26 strongly suggests the presence of a population of
plasma cells producing clonal lambda free light chains,
whereas a ratio ⬎1.65 suggests production of clonal kappa
free light chains. In 110 patients with AL amyloidosis, serum
immunofixation was positive in 69%, urine immunofixation
was positive in 83%, and the kappa-to-lambda ratio was
abnormal in 91%. The combination of an abnormal kappa
lambda ratio and a positive serum immunofixation identified
99% of patients with AL amyloidosis.64 A bone marrow
biopsy is mandatory to assess the percentage of plasma cells,
and immunoperoxidase staining will determine whether the
abnormal plasma cells are producing kappa or lambda light
chains.65 Bone marrow biopsy is also required to exclude
myeloma and other less common disorders that can be
associated with AL amyloidosis such as Waldenstrom’s
macroglobulinemia. It is important to recognize that a monoclonal band present on serum immunofixation may be seen as
an apparently incidental finding in 5% to 10% of patients
⬎70 years of age (“monoclonal gammopathy of uncertain
significance”).66 The serum free-light-chain assay is often
normal in such cases,67 but if any doubt exists about the
clinical picture, further testing must be done to exclude
familial or senile forms of amyloid. Such testing includes
either special staining techniques of the amyloid such as
immunogold electron microscopy68 –70 or genetic testing to
rule out familial forms of amyloid.71
Management
Management of cardiac amyloidosis requires a 2-fold approach: management of the cardiac-related symptoms and
treatment of the underlying disease. The mainstay of the
treatment of heart failure in AL amyloidosis is the use of
diuretics; higher doses than anticipated may be required if the
albumin level is low as a result of concomitant nephrotic
syndrome. In a patient with anasarca, intravenous diuresis is
Falk
Cardiac Amyloidosis
2053
Figure 6. Color-coded map of myocardial long-axis strain (percentage change
in length) recorded from the ventricular
septum. Note the top left image recorded in the apical 4-chamber view, with
a bar representing a key to the color
coding. The numbers on the ventricular
septum correspond to the numbers on
the map; the apical septal strain is represented on the top part of the map, and
the base is represented at the bottom. A,
Recording from a normal subject. Immediately after the onset of systole (arrow
indicates R wave), there is a brief light
blue vertical line, representing isovolumic
systole, followed by a broad, uniform
orange/red-coded band representing
ventricular contraction. This is followed
by early relaxation in blue, diastasis in
green, and a late diastolic relaxation represented in a second brief blue area. B,
Strain map from a patient with cardiac
amyloid, heart failure, and a mildly
reduced ejection fraction. The arrow
again represents the onset of the QRS
complex. There is almost no longitudinal
motion in any portion of the septum, with
the large area of green representing
absent motion and the brief patches of
color representing slight elongation in
late diastole (light blue). Systolic strain is
almost absent except for some severely
reduced longitudinal motion at the base
and a brief, reduced contraction in midsystole near the apical septum (yellowish
orange.)
often needed because absorption of diuretics may be impaired. Resistant, large, pleural effusions may indicate the
presence of pleural amyloid.31 They may necessitate recurrent
pleural taps and occasionally require pleurodesis. ACE inhibitors and angiotensin II inhibitors are very poorly tolerated in
subjects with AL amyloidosis; even small doses may result in
profound hypotension. If an attempt is made to introduce
them, it should be done with extreme caution, ideally in a
monitored setting and starting with a very low dose of
captopril, chosen because of its relatively short duration of
action. The extreme hypotensive response seen in some
patients is probably a function of autonomic neuropathy
because angiotensin receptors play a role in the maintenance
of blood pressure and, in the setting of autonomic nervous
system dysfunction, their role may be much greater than
normal. There are no data on the effectiveness of ␤-blockade
on survival in amyloidosis, but the use of ␤-blockers may be
limited because of refractory heart failure or disease-related
severe hypotension. Calcium channel blockers are contraindicated because they often produce a significant negative
inotropic effect.72,73 On occasion, I have been able to maintain patients on oral nitrates for preload reduction, but they
often have only a minor benefit and require cautious introduction and gradual dose escalation. There are no published
data on the use of intravenous inotropic or vasodilator drugs
in patients with severe heart failure resulting from amyloidosis. However, I have found renal-dose dopamine (1 to 3 ␮g
· kg⫺1 · min⫺1) to be a helpful adjunct for the treatment of
anasarca, provided that renal function is unimpaired. Digoxin,
used for its inotropic properties, is of little value in amyloid-
2054
Circulation
September 27, 2005
Figure 7. Simultaneous right and left
ventricular pressure tracings in a patient
with AL amyloidosis and atrial fibrillation.
After a longer R-R interval, several beats
show a dip and plateau morphology of
the diastolic wave form with equalization
of right and left ventricular pressures,
mimicking constrictive pericarditis.
osis, and these patients may be at increased risk of digoxin
toxicity because the drug binds avidly to amyloid fibrils.74 As
a result of the binding to myocardial amyloid, cardiac digoxin
levels may be elevated, and digoxin toxicity can exist even in
the setting of “therapeutic” serum digoxin levels. Nevertheless, when atrial fibrillation with a rapid ventricular response
is present, digoxin (administered cautiously) can be usually
safely and successfully used.
The value of routine anticoagulation in patients with severe
heart failure of any cause is uncertain. However, unless major
contraindications exist, the presence of atrial fibrillation in
AL amyloidosis is a very strong indication for warfarin
anticoagulation because of a very high rate of thromboembolic events. In severe cardiac amyloidosis, the atrium is
infiltrated, and dysfunctional29 and atrial thrombi may be
present even during sinus rhythm.43,45 It is therefore prudent
to anticoagulate patients with AL amyloidosis even if they are
in sinus rhythm if there is a small transmitral A wave seen on
transthoracic echocardiography (ⱕ20 cm/s). Transesophageal
echocardiography may be helpful in selected patients with
apparently poor atrial function45 and, even when the patient in
sinus rhythm, may reveal a left atrial appendage thrombus,
left atrial appendage spontaneous echo contrast, or markedly
decreased atrial appendage Doppler velocities (⬍40 cm/s).
The definitive treatment of AL amyloidosis is antiplasma
cell therapy aimed at stopping the production of the paraprotein responsible for the formation of amyloid.75– 81 A number
of chemotherapeutic regimens exist, but the highest success
rate appears to be with the use of intravenous melphalan, with
a complete hematologic response in ⬇40% of patients who
survive 1 year after chemotherapy.79 Unfortunately, the advanced nature of the cardiac disease in many patients at the
time of diagnosis either renders them unfit for high-dose
chemotherapy with autologous stem cell replacement or
places them at a risk of peritreatment mortality as high as
30%.82 Precise criteria to define a subgroup of patients with
AL amyloidosis who have an acceptably low treatmentrelated mortality in patients with AL cardiac amyloid have
been difficult to define, but the absence of heart failure and
normal ejection fraction and the absence of pleural effusions
appear to augur a better prognosis. In contrast, marked wall
thickening and markedly elevated brain natriuretic peptide or
elevated troponin augur a poorer outcome.15–17 Younger
patients and those without significant involvement of other
organ systems are also more likely to survive chemotherapy,
but unexpected arrhythmias, episodes of electromechanical
dissociation, or worsening of congestive heart failure occur
even in this group.82 An ejection fraction ⬍40% is generally
considered an absolute contraindication to high-dose chemotherapy in a patient with cardiac amyloid, particularly because most of these patients have class III congestive heart
failure and minimal cardiac reserve.
Nevertheless, despite the significant risk of death associated with the use of vigorous chemotherapy in patients with
AL amyloidosis involving the heart, it should be considered
in selected patients because survivors often have a clinical
improvement in congestive heart failure despite an unchanged echocardiographic appearance.83 The improvement
in heart failure may be due to abolition of the production of
freshly produced light chains, which have been shown to be
toxic to myocardial cells, suggesting that AL amyloidosis is
not simply an infiltrative cardiomyopathy but rather a toxic
infiltrative disorder.84,85
Falk
For patients who cannot tolerate high-dose intravenous
melphalan, preliminary data from the UK amyloid group
suggest that a modified intravenous regimen of melphalan,
given monthly, may be better tolerated with a similar response rate, but no direct comparative study has been performed.86 The “standard regimen” of melphalan and prednisone given as a “pulsed” dose for 3 to 5 days every 6 weeks
seems to have little benefit in patients with cardiac amyloidosis, probably because several months are required to see an
effect.87 In addition, the steroid regimen may worsen congestive heart failure. Recently, we have used a low-dose “continuous” melphalan regimen in patients with severe cardiac
amyloid with evidence of hematologic response in 7 of 13
patients.81 Unfortunately, the cardiac disease was often too
severe at the start of treatment to determine whether such a
regimen has any impact on long-term survival. Regimens that
include the use of high-dose dexamethasone such as vincristine, adriamycin, and dexamethasone88 are generally not
tolerated in cardiac amyloidosis because adriamycin, although used in relatively small doses, can produce cardiac
toxicity and dexamethasone may aggravate heart failure.
In highly selected cases, cardiac transplantation may be
considered. Early experience with cardiac transplantation in
AL amyloidosis suggested that short- and medium-term
mortality did not differ from that in other disorders,89 but a
later report of a small series of patients treated at multiple
transplantation centers demonstrated an apparently greater
long-term mortality than expected, usually because of disease
progression in the heart or noncardiac organs.90,91 As a result
of these observations, many transplantation centers consider
AL amyloidosis a contraindication to transplantation. However, with the advent of high-dose chemotherapy and stem
cell transplantation, it is possible to transplant the heart and to
perform chemotherapy 6 to 12 months later to abolish
amyloid production. Potential candidates for this combined
procedure are uncommon because noncardiac organ involvement is a contraindication and cardiac disease is limited
clinically to the heart in ⬍5% of cases. Nevertheless, a
number of patients have been treated successfully with this
combined approach; several have obtained a long-term remission from the disease without evidence of recurrence after 3
to 5 years of follow-up.
Light-Chain Cardiomyopathy
Renal light-chain deposition disease is a well-recognized
entity in which renal failure may occur as a result of the
deposition of light chains either related to multiple myeloma
or as a manifestation of a plasma cell dyscrasia.92 Less well
known, and probably less common, is the cardiac manifestation of light-chain deposition disease. Although not actually a
form of amyloidosis, the rare condition of light-chain cardiomyopathy deserves mention because it may mimic AL
amyloidosis. In this condition, nonfibrillar deposits of light
chains are found in the myocardium in association with either
multiple myeloma or plasma cell dyscrasia.93 The echocardiographic appearance is similar to cardiac amyloidosis, and
heart failure and arrhythmias may occur, but Congo red
staining of the myocardium is negative.94 Kappa light-chain
deposition tends to be more common than lambda. The
Cardiac Amyloidosis
2055
importance of recognition of this entity relates to the occasional patient with evidence of a plasma cell dyscrasia and an
echocardiogram suspicious of amyloidosis in whom no amyloid is seen on endomyocardial biopsy. In such cases, electron
microscopy with antikappa or antilambda immunogold labeling may reveal granular deposits typical of light-chain deposition, thereby confirming the diagnosis.93 Chemotherapy
targeted to the underlying plasma cell dyscrasia may lead to
reversal of the cardiomyopathy.95
Hereditary Amyloidosis
Hereditary amyloidosis exists in a number of forms, but most
cases are due to the production of amyloid from a mutant
transthyretin protein.5 Transthyretin contains 125 pairs of
amino acids, and ⬎70 mutations have been described, most
of which are amyloidogenic. The specific site of an amino
acid substitution determines the phenotype of the disease,
which is transmitted as an autosomal dominant with high
penetrance. The onset occurs from the third decade on, most
commonly after the age of 40. In some forms, peripheral
neuropathy may predominate, with cardiac amyloid being
either absent or limited to the conduction system, most
frequently manifesting as sinus node dysfunction. Other
mutations such as Thr-60-Ala (the substitution of alanine for
threonine at position 60) present with a predominant cardiomyopathy characterized by heart failure and conduction
system disturbances with minimal neuropathy. Renal involvement is generally not a feature of transthyretin-associated
cardiac amyloidosis, and myocardial infiltration may be quite
severe before the onset of heart failure. This results in an
echocardiographic appearance that is very similar to advanced AL cardiac amyloidosis but is associated with less
heart failure and a much better long-term survival.96 Although
strain and strain rate imaging demonstrate subtle differences
in ventricular long-axis function between AL and familial
amyloidosis,97 the difference in survival between these 2
diseases is probably related to the toxic effect of light-chain
deposition on the myocardium in AL amyloidosis,84,85 which
is absent in transthyretin-related amyloidosis.
Among the familial TTR amyloidoses, the mutation characterized by a substitution of isoleucine for valine at position
122 of the transthyretin molecule deserves special mention.98,99 Approximately 4% of the black population in the
United States is heterozygous for this mutation,100 which may
result in a late-onset cardiomyopathy in either sex, manifesting as progressive congestive heart failure. In our initial
experience of 12 cases101 and subsequent personal experience
of a similar number of cases, the disease is found to have
features that are remarkably consistent among patients. The
echocardiogram is similar to that seen in other variants of
TTR amyloidosis, with features of an infiltrative/restrictive
cardiomyopathy. Signs of right-sided heart failure predominate, and peripheral edema and ascites may be profound.
Involvement of the cardiac valves may result in tricuspid
regurgitation, which can further aggravate the right heart
failure. Because of the high prevalence of hypertensive heart
disease in blacks, left ventricular thickening seen on the
echocardiogram may be mistakenly attributed to
hypertension-induced hypertrophy, and the diagnosis of an
2056
Circulation
September 27, 2005
infiltrative cardiomyopathy can be overlooked. The presence
of right ventricular thickening, the absence of left ventricular
hypertrophy on the ECG, and the clinical finding of right
heart failure in a black patient (particularly if a history of
carpal tunnel syndrome is elicited) should strongly suggest
the diagnosis. Unlike AL amyloidosis, the abdominal fat
aspirate frequently stains negative for amyloid, and endomyocardial biopsy may be necessary unless tissue is available for
staining from prior carpal tunnel syndrome surgery. The
penetrance of this disorder is unknown, but many other TTR
mutations have a high penetrance, suggesting that the Ile 122
variant is probably frequently overlooked. Regardless of the
penetrance, the high prevalence of the mutation probably
makes it the commonest familial amyloid cardiomyopathy
and possibly the commonest type of amyloid heart disease.
Unfortunately, the late age of onset and the universal manifestation of heart failure preclude liver transplantation as a
therapy (see below) in the vast majority of patients with this
disorder, although we have treated 1 patient by cardiac
transplantation who had excellent results over the succeeding
several years.
Treatments for ATTR Amyloidosis
Although transthyretin is produced by the liver, it has little
effect on the liver function as liver deposition of amyloid is
minimal or absent.102 Currently, the definitive treatment of
ATTR is liver transplantation, which removes the source of
transthyretin and hence the precursor of amyloid deposition.103 Optimally, liver transplantation should be performed
in a patient with a known mutant transthyretin as soon as
there is clinical evidence of the disease documented by either
deposition of amyloid in fat pad aspirate or clinical evidence
of disease activity.104 Despite significant myocardial infiltration in some patients, the clinical experience has been that
they tolerate the surgical aspect of liver transplantation well.
Because the liver is functionally normal, it has, on occasion,
been removed from an amyloid patient and transplanted into
another patient who requires an urgent liver transplant (domino transplantation).105 To date, only a small number of
domino operations have been done, and it is not known if, or
when, the recipient will develop amyloidosis.
Initial enthusiasm for transplantation as a technique for
arresting progressive cardiomyopathy has been tempered by
the observation that wall thickening progresses in some
patients who have amyloid cardiomyopathy at the time of
liver transplantation.106,107 This is most probably due to the
continued deposition of wild-type transthyretin in the myocardium, a process akin to senile cardiac amyloidosis (SCA).
Occasionally, combined liver and heart transplantation or
heart transplantation alone has been performed for ATTR
amyloid with significant cardiomyopathy.90,108
There is ongoing investigation into the development of
drugs that will stabilize transthyretin and prevent the formation of amyloid.109 In vitro evidence suggests that certain
nonsteroidal agents such as diflunisal can stabilize transthyretin.110 Although currently there is no clinical evidence that
these agents can prevent the progression of TTR amyloidosis,
clinical trials are in the planning stage. However, even if
nonsteroidal agents have some effect on disease progression,
a significant limitation in patients with cardiomyopathy is the
potential for precipitation or aggravation of congestive heart
failure. Thus, other agents without the potential for fluid
retention are actively being sought.5,111
There are other, very rare, causes of familial nontransthyretin amyloid cardiomyopathy. Mutations of the genes encoding apolipoprotein A may be amyloidogenic and can result an
isolated cardiomyopathy that has been successfully treated
with cardiac transplantation.112 Mutations of fibrinogen A
␣-chain and lysozyme can also cause amyloidosis, but deposition is predominantly in organs other than the heart.71
SCA
SCA is the predominant clinical manifestation of senile
systemic amyloidosis. It results from the cardiac deposition of
amyloid derived from wild-type transthyretin (ie, transthyretin with a normal amino acid constitution)113 and invariably
presents as congestive heart failure. The diagnosis requires
the finding of amyloid deposits in the myocardium, in
conjunction with evidence of an infiltrative cardiomyopathy
on echocardiogram. The echocardiographic appearance is
indistinguishable from that found in patients with AL amyloidosis, although the degree of wall thickening may be very
marked despite relatively mild, or easily controllable, heart
failure.114 Once the diagnosis is suspected, confirmation
usually requires an endomyocardial biopsy because noncardiac involvement is rare. However, caution should be exercised in labeling an elderly patient as having senile amyloidosis on the basis of an endomyocardial biopsy if the amyloid
deposits are sparse and the echocardiographic appearance is
not consistent with amyloidosis because small amounts of
amyloid derived from wild-type transthyretin are not uncommon in the very elderly.115,116 Exclusion of a plasma cell
dyscrasia is mandatory, and screening should be performed to
exclude a mutant transthyretin. Although evaluation for a
plasma cell dyscrasia is usually negative, an unrelated benign
monoclonal gammopathy of unknown significance will be
expected to be present by chance in 3% to 5% of patients of
this age when sought by serum protein electrophoresis66,117
and even more commonly if the more sensitive immunofixation is used. If immunofixation is positive, but the clinical
picture is most consistent with SCA other tests to exclude AL
amyloidosis are mandatory. Immunochemistry or immunogold electron microscopy of biopsy tissue staining unequivocally positive for TTR and negative for kappa and lambda
are confirmatory of the transthyretin origin of the amyloid.70
For unclear reasons, SCA is almost exclusively a disorder
of men.114 The disease is rare individuals ⬍70 years of age,
and the median survival from the onset of heart failure is 7.5
years compared with 15 months in patients with AL amyloidosis and a similar degree of LV thickening.114 The clinical
manifestations of senile cardiac amyloidosis are quite similar
among patients.114,118 Progression of heart failure in senile
amyloidosis is insidious but inexorable, and the diagnosis
should be suspected in an elderly man with unexplained
right-sided or biventricular failure and an echocardiogram
showing left ventricular thickening with normal ventricular
cavity size. The disease is not associated with any other major
clinical organ involvement, although carpal tunnel syndrome,
Falk
Cardiac Amyloidosis
2057
Figure 8. Flow diagram outlining the evaluation of a patient with suspected cardiac amyloidosis. Clinical evaluation may reveal clues that
strengthen the likelihood of amyloidosis, but a tissue diagnosis is mandatory. Although special staining of the biopsy may confirm the type of
amyloid, further workup of AL amyloid is required to exclude myeloma and to quantify free light chains. If the biopsy stains positive for transthyretin, further testing is needed to determine whether this is a wild-type or mutant transthyretin. ApoA1 indicates apolipoprotein A1; IFE,
immunofixation electrophoresis; FLC, free-light-chain assay; SSA, senile systemic amyloidosis; and TTR, transthyretin.
often preceding the cardiac disease by a few years, is
common. Bifascicular block on the ECG is common, and
progression to complete AV block occurs not infrequently,
necessitating permanent pacemaker implantation. Implantation of a permanent pacemaker for conduction system disease
may be followed by a worsening of heart failure; this may be
due to the dysynergy produced by right ventricular pacing in
the nondilated, infiltrated ventricle with its small cavity and
reduced ejection fraction. Thus, if conduction system disease
warrants a pacemaker, strong consideration should be given
to biventricular pacing to maximize ventricular stroke volume. Atrial fibrillation is a common arrhythmia in SCA,
presumably because of the combination of atrial infiltration
with amyloid, increased left atrial pressure, and the advanced
age of the patient. Once atrial fibrillation occurs, thromboembolism is common, and warfarin anticoagulation should be
prescribed unless a major contraindication exists. Restoration
and maintenance of sinus rhythm by cardioversion and
antiarrhythmic drugs such as amiodarone may provide some
benefit, but atrial function is usually impaired as a result of
amyloid infiltration. In addition, widespread conduction system disease may worsen in the presence of antiarrhythmic
drugs. Unlike AL amyloidosis, patients with SCA often
tolerate ACE inhibitors, although the mainstay of therapy is
still the judicious use of diuretics. There is no specific
treatment for SCA, but as in familial amyloidosis, drugs that
stabilize the transthyretin molecule hold some promise109 and
are on the verge of clinical trials.
Secondary Amyloidosis
Secondary amyloidosis is increasingly uncommon in the
developed world owing to the eradication of chronic infections. However, it is still seen occasionally in association with
juvenile or adult rheumatoid arthritis and other rheumatic
disorders such as ankylosing spondylitis, as well as with
inflammatory bowel disease. Hepatic and renal amyloid
deposition dominates the clinical picture, and clinical heart
disease related to cardiac amyloid is very rare.6 In the few
cases in which there is echocardiographic evidence of cardiac
amyloidosis resulting from secondary amyloidosis, cardiac
symptoms are usually absent, although we have seen very
occasional cases of mild heart failure with extensive left
ventricular thickening, which was associated with sudden
death in 1 case.
Isolated Atrial Amyloidosis
Atrial amyloid deposition is a common finding at autopsy,
particularly in elderly patients.115,119 Immunohistochemical
evaluation demonstrates its origin from atrial natriuretic
peptide.120 Unlike the other forms of amyloid discussed, atrial
amyloid is a nonsystemic deposition, limited to the atrium.
Until recently, it was believed to be a clinically insignificant
finding that increased in prevalence with increasing age and
with the presence of organic heart disease. Recent data based
on atrial biopsies taken at the time of cardiac surgery suggest
that isolated atrial amyloidosis may be commoner in women
and is more likely to occur in the presence of atrial fibrilla-
2058
Circulation
September 27, 2005
tion.121,122 Interestingly, an inverse relationship with the
presence of atrial fibrosis has been suggested.121,123 The role
of isolated atrial amyloid in the pathogenesis and maintenance of atrial fibrillation remains to be fully elucidated, but
it may be a precipitating factor of atrial fibrillation in some
patients and may be produced as part of the remodeling
associated with this arrhythmia.121
Conclusions
In summary, cardiac amyloidosis, although uncommon, is
characterized by a typical appearance on echocardiography,
the recognition of which should alert the astute clinician to
the probable diagnosis. It is critical to recognize that several
forms of amyloidosis may cause cardiomyopathy and that
treatment and prognosis of these individual cardiomyopathies
differ greatly from each other. A flow diagram to aid in the
diagnosis of the type of amyloid is illustrated in Figure 8. In
AL amyloidosis, chemotherapy may arrest or possibly reverse
the disease, with resultant stabilization or improvement of
symptoms. Thus, early diagnosis is critical because patients
with advanced disease are usually too ill for intensive
chemotherapy. Recognition of non-AL cardiac amyloidosis is
important to avoid unnecessary chemotherapy, to screen
family members, and in the near future, to provide new
medications that will stabilize the amyloidogenic substance in
the blood and prevent the onset of progression of this
important and underdiagnosed condition.
References
1. Merlini G, Bellotti V. Molecular mechanisms of amyloidosis. N Engl
J Med. 2003;349:583–596.
2. Falk RH, Comenzo RL, Skinner M. The systemic amyloidoses. N Engl
J Med. 1997;337:898 –909.
3. Sipe JD, Cohen AS. History of the amyloid fibril. J Struct Biol. 2000;
130:88 –98.
4. Selkoe DJ. Folding proteins in fatal ways. Nature. 2003;426:900 –904.
5. Merlini G, Westermark P. The systemic amyloidoses: clearer understanding of the molecular mechanisms offers hope for more effective
therapies. J Intern Med. 2004;255:159 –178.
6. Dubrey SW, Cha K, Simms RW, Skinner M, Falk RH. Electrocardiography and Doppler echocardiography in secondary (AA) amyloidosis.
Am J Cardiol. 1996;77:313–315.
7. Westermark P, Benson MD, Buxbaum JM, Cohen AS, Frangione B,
Ikeda S, Masters CL, Merlini G, Saraiva MJ, Sipe J. Amyloid protein
fibril nomenclature–2002. Amyloid. 2002;9:197–200.
8. Smith TJ, Kyle RA, Lie JT. Clinical significance of histopathologic
patterns of cardiac amyloidosis. Mayo Clin Proc. 1984;59:547–555.
9. Gertz MA, Lacy MQ, Dispenzieri A. Amyloidosis hematology. Oncol
Clin North Am. 1999;13:1211–1220.
10. Dubrey SW, Cha K, Anderson J, Chamarthi B, Reisinger J, Skinner M,
Falk RH. The clinical features of immunoglobulin light-chain (AL)
amyloidosis with heart involvement. QJM. 1998;91:141–157.
11. Kyle RA, Gertz MA. Primary systemic amyloidosis: clinical and laboratory features in 474 cases. Semin Hematol. 1995;32:45–59.
12. Mueller PS, Edwards WD, Gertz MA. Symptomatic ischemic heart
disease resulting from obstructive intramural coronary amyloidosis.
Am J Med. 2000;109:181–188.
13. Miani D, Rocco M, Alberti E, Spedicato L, Fioretti PM. Amyloidosis of
epicardial and intramural coronary arteries as an unusual cause of myocardial
infarction and refractory angina pectoris. Ital Heart J. 2002;3:479–482.
14. Zabernigg A, Schranzhofer R, Kreczy A, Gattringer K. Continuously
elevated cardiac troponin I in two patients with multiple myeloma and
fatal cardiac amyloidosis. Ann Oncol. 2003;14:1791.
15. Dispenzieri A, Kyle RA, Gertz MA, Therneau TM, Miller WL,
Chandrasekaran K, McConnell JP, Burritt MF, Jaffe AS. Survival in
patients with primary systemic amyloidosis and raised serum cardiac
troponins. Lancet. 2003;361:1787–1789.
16. Dispenzieri A, Gertz MA, Kyle RA, Lacy MQ, Burritt MF, Therneau
TM, McConnell JP, Litzow MR, Gastineau DA, Tefferi A, Inwards DJ,
Micallef IN, Ansell SM, Porrata LF, Elliott MA, Hogan WJ, Rajkumar
SV, Fonseca R, Greipp PR, Witzig TE, Lust JA, Zeldenrust SR, Snow
DS, Hayman SR, McGregor CG, Jaffe AS. Prognostication of survival
using cardiac troponins and N-terminal pro-brain natriuretic peptide in
patients with primary systemic amyloidosis undergoing peripheral blood
stem cell transplantation. Blood. 2004;104:1881–1887.
17. Dispenzieri A, Gertz MA, Kyle RA, Lacy MQ, Burritt MF, Therneau
TM, Greipp PR, Witzig TE, Lust JA, Rajkumar SV, Fonseca R, Zeldenrust
SR, McGregor CG, Jaffe AS. Serum cardiac troponins and N-terminal
pro-brain natriuretic peptide: a staging system for primary systemic amyloidosis. J Clin Oncol. 2004;22:3751–3757.
18. Cantwell RV, Aviles RJ, Bjornsson J, Wright RS, Freeman WK, Oh JK, Hoyer
JD, Markovic S, Jaffe AS. Cardiac amyloidosis presenting with elevations of
cardiac troponin I and angina pectoris. Clin Cardiol. 2002;25:33–37.
19. Falk RH, Rubinow A, Cohen AS. Cardiac arrhythmias in systemic
amyloidosis: correlation with echocardiographic abnormalities. J Am
Coll Cardiol. 1984;3:107–113.
20. Chamarthi B, Dubrey SW, Cha K, Skinner M, Falk RH. Features and
prognosis of exertional syncope in light-chain associated AL cardiac
amyloidosis. Am J Cardiol. 1997;80:1242–1245.
21. Rubinow A, Cohen AS. Skin involvement in generalized amyloidosis: a
study of clinically involved and uninvolved skin in 50 patients with
primary and secondary amyloidosis. Ann Intern Med. 1978;88:781–785.
22. Daoud MS, Lust JA, Kyle RA, Pittelkow MR. Monoclonal gammopathies and associated skin disorders. J Am Acad Dermatol. 1999;
40:507–535; quiz 536 –538.
23. Burroughs EI, Aronson AE, Duffy JR, Kyle RA. Speech disorders in
systemic amyloidosis. Br J Disord Commun. 1991;26:201–206.
24. Park MA, Mueller PS, Kyle RA, Larson DR, Plevak MF, Gertz MA.
Primary (AL) hepatic amyloidosis: clinical features and natural history
in 98 patients. Medicine. 2003;82:291–298.
25. Derrick EK, Price ML. Primary systemic amyloid with nail dystrophy.
J Royal Soc Med. 1995;88:290P–291P.
26. Klein AL, Cohen GI. Doppler echocardiographic assessment of constrictive pericarditis, cardiac amyloidosis, and cardiac tamponade.
Cleveland Clin J Med. 1992;59:278 –290.
27. Tyberg TI, Goodyer AV, Hurst VWD, Alexander J, Langou RA. Left
ventricular filling in differentiating restrictive amyloid cardiomyopathy
and constrictive pericarditis. Am J Cardiol. 1981;47:791–796.
28. Murphy L, Falk RH. Left atrial kinetic energy in AL amyloidosis: can it
detect early dysfunction? Am J Cardiol. 2000;86:244 –246.
29. Modesto KM, Dispenzieri A, Cauduro AS, Lacy MQ, Khandheria BK,
Pellika PA, Belohlavek M, Seward JB, Kyle RA, Tajik AJ, Gertz MA,
Abraham TP. Left atrial myopathy in cardiac amyloidosis: implications
of novel echocardiographic techniques. Eur Heart J. 2005;26:173–179.
30. Bernardi L, Passino C, Porta C, Anesi E, Palladini G, Merlini G.
Widespread cardiovascular autonomic dysfunction in primary amyloidosis: does spontaneous hyperventilation have a compensatory role
against postural hypotension? Heart. 2002;88:615– 621.
31. Berk JL, Keane J, Seldin DC, Sanchorawala V, Koyama J, Dember LM,
Falk RH. Persistent pleural effusions in primary systemic amyloidosis:
etiology and prognosis. Chest. 2003;124:969 –977.
32. Gillmore JD, Lovat LB, Hawkins PN. Amyloidosis and the liver.
J Hepatol. 1999;30(suppl 1):17–33.
33. Chopra S, Rubinow A, Koff RS, Cohen AS. Hepatic amyloidosis: a
histopathologic analysis of primary (AL) and secondary (AA) forms.
Am J Pathol. 1984;115:186 –193.
34. Murtagh B, Hammill SC, Gertz MA, Kyle RA, Tajik AJ, Grogan M.
Electrocardiographic findings in primary systemic amyloidosis and
biopsy-proven cardiac involvement. Am J Cardiol. 2005;95:535–537.
35. Siqueira-Filho AG, Cunha CL, Tajik AJ, Seward JB, Schattenberg TT,
Giuliani ER. M-mode and two-dimensional echocardiographic features
in cardiac amyloidosis. Circulation. 1981;63:188 –196.
36. Klein AL, Hatle LK, Taliercio CP, Taylor CL, Kyle RA, Bailey KR,
Seward JB, Tajik AJ. Serial Doppler echocardiographic follow-up of left
ventricular diastolic function in cardiac amyloidosis. J Am Coll Cardiol.
1990;16:1135–1141.
37. Child JS, Levisman JA, Abbasi AS, MacAlpin RN. Echocardiographic
manifestations of infiltrative cardiomyopathy: a report of seven cases
due to amyloid. Chest. 1976;70:726 –731.
38. Carroll JD, Gaasch WH, McAdam KP. Amyloid cardiomyopathy: characterization by a distinctive voltage/mass relation. Am J Cardiol. 1982;49:9–13.
Falk
39. Sedlis SP, Saffitz JE, Schwob VS, Jaffe AS. Cardiac amyloidosis simulating hypertrophic cardiomyopathy. Am J Cardiol. 1984;53:969 –970.
40. Hemmingson LO, Eriksson P. Cardiac amyloidosis mimicking hypertrophic cardiomyopathy. Acta Med Scand. 1986;219:421– 423.
41. Klein AL, Hatle LK, Burstow DJ, Seward JB, Kyle RA, Bailey KR,
Luscher TF, Gertz MA, Tajik AJ. Doppler characterization of left
ventricular diastolic function in cardiac amyloidosis. J Am Coll Cardiol.
1989;13:1017–1026.
42. Abdalla I, Murray RD, Lee JC, Stewart WJ, Tajik AJ, Klein AL.
Duration of pulmonary venous atrial reversal flow velocity and mitral
inflow a wave: new measure of severity of cardiac amyloidosis. J Am
Soc Echocardiogr. 1998;11:1125–1133.
43. Dubrey S, Pollak A, Skinner M, Falk RH. Atrial thrombi occurring
during sinus rhythm in cardiac amyloidosis: evidence for atrial electromechanical dissociation. Br Heart J. 1995;74:541–544.
44. Plehn JF, Southworth J, Cornwell GG. Brief report: atrial systolic failure
in primary amyloidosis. N Engl J Med. 1992;327:1570 –1573.
45. Santarone M, Corrado G, Tagliagambe LM, Manzillo GF, Tadeo G,
Spata M, Longhi M. Atrial thrombosis in cardiac amyloidosis: diagnostic contribution of transesophageal echocardiography. J Am Soc Echocardiogr. 1999;12:533–536.
46. Moyssakis I, Triposkiadis F, Pantazopoulos NJ, Kyriakidis M,
Nihoyannopoulos P. Left atrial systolic function in primary and familial
amyloidosis: assessment from left atrial volume change. Clin Cardiol.
2004;27:528 –532.
47. Koyama J, Ray-Sequin PA, Davidoff R, Falk RH. Usefulness of pulsed
tissue Doppler imaging for evaluating systolic and diastolic left ventricular function in patients with AL (primary) amyloidosis. Am J Cardiol.
2002;89:1067–1071.
48. Koyama J, Davidoff R, Falk RH. Longitudinal myocardial velocity
gradient derived from pulsed Doppler tissue imaging in AL amyloidosis:
a sensitive indicator of systolic and diastolic dysfunction. J Am Soc
Echocardiogr. 2004;17:36 – 44.
49. Koyama J, Ray-Sequin PA, Falk RH. Longitudinal myocardial function
assessed by tissue velocity, strain, and strain rate tissue Doppler echocardiography in patients with AL (primary) cardiac amyloidosis. Circulation. 2003;107:2446 –2452.
50. Maceira AM, Joshi J, Prasad SK, Moon JC, Perugini E, Harding I, Sheppard
MN, Poole-Wilson PA, Hawkins PN, Pennell DJ. Cardiovascular magnetic
resonance in cardiac amyloidosis. Circulation. 2005;111:186–193.
51. Kwong RY, Falk RH. Cardiovascular magnetic resonance in cardiac
amyloidosis. Circulation. 2005;111:122–124.
52. Swanton RH, Brooksby IA, Davies MJ, Coltart DJ, Jenkins BS, WebbPeploe MM. Systolic and diastolic ventricular function in cardiac amyloidosis: studies in six cases diagnosed with endomyocardial biopsy.
Am J Cardiol. 1977;39:658 – 664.
53. Kern MJ, Lorell BH, Grossman W. Cardiac amyloidosis masquerading
as constrictive pericarditis. Cathet Cardiovasc Diagn. 1982;8:629 – 635.
54. Robbins MA, Pizzarello RA, Stechel RP, Chiaramida SA, Gulotta SJ.
Resting and exercise hemodynamics in constrictive pericarditis and a
case of cardiac amyloidosis mimicking constriction. Cathet Cardiovasc
Diagn. 1983;9:463– 471.
55. Lorell BH, Grossman W. Profiles in constrictive pericarditis, restrictive cardiomyopathy and cardiac tamponade. In: Baim DS, Grossman W, eds. Grossman’s
Cardiac Catheterization, Angiography and Intervention. 6th ed. Philadelphia,
Pa: Lippincott, Williams & Wilkins; 2000:829–850.
56. Hurrell DG, Nishimura RA, Higano ST, Appleton CP, Danielson GK,
Holmes DR Jr, Tajik AJ. Value of dynamic respiratory changes in left
and right ventricular pressures for the diagnosis of constrictive pericarditis. Circulation. 1996;93:2007–2013.
57. Pomerance A, Slavin G, McWatt J. Experience with the sodium
sulphate-Alcian blue stain for amyloid in cardiac pathology. J Clin
Pathol. 1976;29:22–26.
58. Libbey CA, Skinner M, Cohen AS. Use of abdominal fat tissue aspirate in the
diagnosis of systemic amyloidosis. Arch Intern Med. 1983;143:1549–1552.
59. Ansari-Lari MA, Ali SZ. Fine-needle aspiration of abdominal fat pad for
amyloid detection: a clinically useful test? Diagn Cytopathol. 2004;30:
178–181.
60. Ardehali H, Qasim A, Cappola T, Howard D, Hruban R, Hare JM,
Baughman KL, Kasper EK. Endomyocardial biopsy plays a role in
diagnosing patients with unexplained cardiomyopathy. Am Heart J.
2004;147:919 –923.
61. Gertz MA, Grogan M, Kyle RA, Tajik AJ. Endomyocardial biopsyproven light chain amyloidosis (AL) without echocardiographic features
of infiltrative cardiomyopathy. Am J Cardiol. 1997;80:93–95.
Cardiac Amyloidosis
2059
62. Abraham RS, Katzmann JA, Clark RJ, Bradwell AR, Kyle RA, Gertz
MA. Quantitative analysis of serum free light chains: a new marker for
the diagnostic evaluation of primary systemic amyloidosis. Am J Clin
Pathol. 2003;119:274 –278.
63. Katzmann JA, Clark RJ, Roshini S, Abraham RS, Bryant S, Lymp JF,
Bradwell AR, Kyle RA. Serum reference intervals and diagnostic ranges
for free and free immunoglobulin light chains: relative sensitivity for
detection of monoclonal light chains. Clin Chem. 2002;48:1437–1444.
64. Katzmann JA, Abraham RS, Dispenzieri A, Lust JA, Kyle RA. Diagnostic performance of quantitative kappa and lambda free light chain
assays in clinical practice. Clin Chem. 2005;51:878 – 881.
65. Swan N, Skinner M, O’Hara CJ. Bone marrow core biopsy specimens in
AL (primary) amyloidosis: a morphologic and immunohistochemical
study of 100 cases. Am J Clin Pathol. 2003;120:610 – 616.
66. Kyle RA, Therneau TM, Rajkumar SV, Larson DR, Plevak MF, Melton
LJ. Long-term follow-up of 241 patients with monoclonal gammopathy
of undetermined significance: the original Mayo Clinic series 25 years
later. Mayo Clin Proc. 2004;79:859 – 866.
67. Rajkumar SV, Kyle RA, Therneau TM, Clark RJ, Bradwell AR, Melton LJ,
Larson DR, Plevak MF, Katzmann JA. Presence of monoclonal free light
chains in the serum predicts risk of progression in monoclonal gammopathy of
undetermined significance. Br J Haematol. 2004;127:308–310.
68. Anesi E, Palladini G, Perfetti V, Arbustini E, Obici L, Merlini G.
Therapeutic advances demand accurate typing of amyloid deposits.
Am J Med. 2001;111:243–244.
69. Arbustini E, Verga L, Concardi M, Palladini G, Obici L, Merlini G.
Electron and immuno-electron microscopy of abdominal fat identifies
and characterizes amyloid fibrils in suspected cardiac amyloidosis.
Amyloid. 2002;9:108 –114.
70. O’Hara CJ, Falk RH. The diagnosis and typing of cardiac amyloidosis.
Amyloid. 2003;10:127–129.
71. Lachmann HJ, Booth DR, Booth SE, Bybee A, Gilbertson JA, Gillmore
JD, Pepys MB, Hawkins PN. Misdiagnosis of hereditary amyloidosis as
AL (primary) amyloidosis. N Engl J Med. 2002;346:1786 –1791.
72. Gertz MA, Falk RH, Skinner M, Cohen AS, Kyle RA. Worsening of
congestive heart failure in amyloid heart disease treated by calcium
channel– blocking agents. Am J Cardiol. 1985;55:1645.
73. Pollak A, Falk RH. Left ventricular systolic dysfunction precipitated by
verapamil in cardiac amyloidosis. Chest. 1993;104:618 – 620.
74. Rubinow A, Skinner M, Cohen AS. Digoxin sensitivity in amyloid
cardiomyopathy. Circulation. 1981;63:1285–1288.
75. De Lorenzi E, Giorgetti S, Grossi S, Merlini G, Caccialanza G, Bellotti V.
Pharmaceutical strategies against amyloidosis: old and new drugs in targeting a
“protein misfolding disease.” Curr Med Chem. 2004;11:1065–1084.
76. Sanchorawala V, Wright DG, Seldin DC, Falk RH, Finn KT, Dember
LM, Berk JL, Quillen K, Anderson JJ, Comenzo RL, Skinner M.
High-dose intravenous melphalan and autologous stem cell transplantation as initial therapy or following two cycles of oral chemotherapy for
the treatment of AL amyloidosis: results of a prospective randomized
trial. Bone Marrow Transplant. 2004;33:381–388.
77. Gertz MA, Lacy MQ, Dispenzieri A. Therapy for immunoglobulin light
chain amyloidosis: the new and the old. Blood Reviews. 2004;18:17–37.
78. Palladini G, Perfetti V, Obici L, Caccialanza R, Semino A, Adami F,
Cavallero G, Rustichelli R, Virga G, Merlini G. Association of melphalan and high-dose dexamethasone is effective and well tolerated in
patients with AL (primary) amyloidosis who are ineligible for stem cell
transplantation. Blood. 2004;103:2936 –2938.
79. Skinner M, Sanchorawala V, Seldin DC, Dember LM, Falk RH, Berk
JL, Anderson JJ, O’Hara C, Finn KT, Libbey CA, Wiesman J, Quillen
K, Swan N, Wright DG. High-dose melphalan and autologous stem-cell
transplantation in patients with AL amyloidosis: an 8-year study. Ann
Intern Med. 2004;140:85–93.
80. Seldin DC, Choufani EB, Dember LM, Wiesman JF, Berk JL, Falk RH,
O’Hara C, Fennessey S, Finn KT, Wright DG, Skinner M, Sanchorawala V.
Tolerability and efficacy of thalidomide for the treatment of patients with
light chain-associated (AL) amyloidosis. Clin Lymphoma. 2003;3:241–246.
81. Sanchorawala V, Wright DG, Seldin DC, Falk RH, Berk JL, Dember
LM, Finn KT, Skinner M. Low-dose continuous oral melphalan for the
treatment of primary systemic (AL) amyloidosis. Br J Haematol. 2002;
117:886 – 889.
82. Falk RH, Reisinger J, Dubrey SW, Mendes LA, Sanchorawala V, Ekery
D, Comenzo R, Vosburgh E, Skinner M. The effect of cardiac
involvement on the outcome of intravenous melphalan therapy and
autologous stem cell rescue for AL amyloidosis. In: Kyle RA, Gertz
2060
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
Circulation
September 27, 2005
MA, eds. Amyloid and the Amyloidoses: VIIIIth International Symposium on Amyloidosis. Rochester, Minn: Parthenon; 1998:181–183.
Dubrey S, Mendes L, Skinner M, Falk RH. Resolution of heart failure in
patients with AL amyloidosis. Ann Intern Med. 1996;125:481– 484.
Liao R, Jain M, Teller P, Connors LH, Ngoy S, Skinner M, Falk RH,
Apstein CS. Infusion of light chains from patients with cardiac amyloidosis causes diastolic dysfunction in isolated mouse hearts. Circulation. 2001;104:1594 –1597.
Brenner DA, Jain M, Pimentel DR, Wang B, Connors LH, Skinner M,
Apstein CS, Liao R. Human amyloidogenic light chains directly impair
cardiomyocyte function through an increase in cellular oxidant stress.
Circ Res. 2004;94:1008 –1010.
Lachmann HJ, Gallimore R, Gillmore JD, Carr-Smith HD, Bradwell
AR, Pepys MB, Hawkins PN. Outcome in systemic AL amyloidosis in
relation to changes in concentration of circulating free immunoglobulin
light chains following chemotherapy. Br J Haematol. 2003;122:78 – 84.
Skinner M, Anderson J, Simms R, Falk R, Wang M, Libbey C, Jones
LA, Cohen AS. Treatment of 100 patients with primary amyloidosis: a
randomized trial of melphalan, prednisone, and colchicine versus colchicine only. Am J Med. 1996;100:290 –298.
Wardley AM, Jayson GC, Goldsmith DJ, Venning MC, Ackrill P,
Scarffe JH. The treatment of nephrotic syndrome caused by primary
(light chain) amyloid with vincristine, doxorubicin and dexamethasone.
Br J Cancer. 1998;78:774 –776.
Hosenpud JD, Uretsky BF, Griffith BP, O’Connell JB, Olivari MT,
Valantine HA. Successful intermediate-term outcome for patients with
cardiac amyloidosis undergoing heart transplantation: results of a multicenter survey. J Heart Transplant. 1990;9:346 –350.
Dubrey SW, Burke MM, Hawkins PN, Banner NR. Cardiac transplantation for amyloid heart disease: the United Kingdom experience.
J Heart Lung Transplant. 2004;23:1142–1153.
Hosenpud JD, DeMarco T, Frazier OH, Griffith BP, Uretsky BF, Menkis
AH, O’Connell JB, Olivari MT, Valantine HA. Progression of systemic
disease and reduced long-term survival in patients with cardiac amyloidosis undergoing heart transplantation: follow-up results of a multicenter survey. Circulation. 1991;84:III-338 –II-43.
Pozzi C, D’Amico M, Fogazzi GB, Curioni S, Ferrario F, Pasquali S,
Quattrocchio G, Rollino C, Segagni S, Locatelli F. Light chain deposition disease with renal involvement: clinical characteristics and prognostic factors. Am J Kidney Dis. 2003;42:1154 –1163.
Gallo G, Goni F, Boctor F, Vidal R, Kumar A, Stevens FJ, Frangione B,
Ghiso J. Light chain cardiomyopathy: structural analysis of the light
chain tissue deposits. Am J Pathol. 1996;148:1397–1406.
Buxbaum JN, Genega EN, Lazowski P, Kumar A, Tunick PA, Kronzon
I, Gallo GA. Infiltrative nonamyloidotic monoclonal immunoglobulin
light chain cardiomyopathy: an underappreciated manifestation of
plasma cell dyscrasias. Cardiology. 2000;93:220 –228.
Nakamura M, Satoh M, Kowada S, Satoh H., Tashiro A, Sato F,
Masuda, Hiramori K. Reversible restrictive cardiomyopathy due to
light-chain deposition disease. Mayo Clin Proc. 2002;77:193–196.
Dubrey SW, Cha K, Skinner M, LaValley M, Falk RH. Familial and
primary (AL) cardiac amyloidosis: echocardiographically similar diseases
with distinctly different clinical outcomes. Heart. 1997;78:74–82.
Ogiwara F, Koyama J, Ikeda S, Kinoshita O, Falk RH. Comparison of the
strain Doppler echocardiographic features of familial amyloid polyneuropathy (FAP) and light-chain amyloidosis. Am J Cardiol. 2005;95:538–540.
Falk RH. The neglected entity of familial cardiac amyloidosis in African
Americans. Ethnicity Dis. 2002;12:141–143.
Jacobson DR, Pastore RD, Yaghoubian R, Kane I, Gallo G, Buck FS,
Buxbaum JN. Variant-sequence transthyretin (isoleucine 122) in
late-onset cardiac amyloidosis in black Americans. N Engl J Med.
1997;336:466 – 473.
Jacobson DR, Pastore R, Pool S, Malendowicz S, Kane I, Shivji A,
Embury SH, Ballas SK, Buxbaum JN. Revised transthyretin Ile 122
allele frequency in African-Americans. Hum Genet. 1996;98:236 –238.
Berg A, Falk RH, Connors LH, Theberge R, Skare J, Murakami A,
Skinner M. Transthyretin Ile-122 in a series of black patients with
amyloidosis. In: Kyle RA, Gertz MA, eds. Amyloid and the Amy-
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
loidoses: VIIIIth International Symposium on Amyloidosis. Rochester,
Minn: Parthenon Publishing Group; 1998.
Hamilton JA, Benson MD. Transthyretin: a review from a structural
perspective. Cell Mol Life Sci. 2001;58:1491–1521.
Suhr OB, Herlenius G, Friman S, Ericzon BG. Liver transplantation for
hereditary transthyretin amyloidosis. Liver Transplant. 2000;6:263–276.
Jonsen E, Suhr OB, Tashima K, Athlin E. Early liver transplantation is
essential for familial amyloidotic polyneuropathy patients’ quality of
life. Amyloid. 2001;8:52–57.
Monteiro E, Perdigoto R, Furtado AL. Liver transplantation for familial
amyloid polyneuropathy. Hepato-Gastroenterol. 1998;45:1375–1380.
Dubrey SW, Davidoff R, Skinner M, Bergethon P, Lewis D, Falk RH.
Progression of ventricular wall thickening after liver transplantation for
familial amyloidosis. Transplantation. 1997;64:74 – 80.
Stangou AJ, Hawkins PN, Heaton ND, Rela M, Monaghan M,
Nihoyannopoulos P, O’Grady J, Pepys MB, Williams R. Progressive
cardiac amyloidosis following liver transplantation for familial amyloid
polyneuropathy: implications for amyloid fibrillogenesis. Transplantation. 1998;66:229 –233.
Ruygrok PN, Gane EJ, McCall JL, Chen XZ, Haydock DA, Munn SR.
Combined heart and liver transplantation for familial amyloidosis. Intern
Med J. 2001;31:66 – 67.
Kelly JW. Attacking amyloid. N Engl J Med. 2005;352:722–723.
Miller SR, Sekijima Y, Kelly JW. Native state stabilization by NSAIDs
inhibits transthyretin amyloidogenesis from the most common familial
disease variants. Lab Invest. 2004;84:545–552.
Lachmann HJ, Hawkins PN. Novel pharmacological strategies in amyloidosis. Nephron Clin Pract. 2003;94:c85– c88.
Obici L, Bellotti V, Mangione P, Stoppini M, Arbustini E, Verga L,
Zorzoli I, Anesi E, Zanotti G, Campana C, Vigano M, Merlini G. The
new apolipoprotein A-I variant leu(174)3 Ser causes hereditary cardiac
amyloidosis, and the amyloid fibrils are constituted by the 93-residue
N-terminal polypeptide. Am J Pathol. 1999;155:695–702.
Westermark P, Sletten K, Johansson B, Cornwell GG. Fibril in senile
systemic amyloidosis is derived from normal transthyretin. Proc Natl
Acad Sci U S A. 1990;87:2843–2845.
Ng B, Connors LH, Davidoff R, Skinner M, Falk RH. Senile systemic
amyloidosis presenting with heart failure: a comparison with light chain–
associated (AL) amyloidosis. Arch Intern Med. 2005;165:1425–1429.
Cornwell GG, Murdoch WL, Kyle RA, Westermark P, Pitkanen P.
Frequency and distribution of senile cardiovascular amyloid: a clinicopathologic correlation. Am J Med. 1983;75:618 – 623.
Cornwell GGd, Westermark P. Senile amyloidosis: a protean manifestation of the aging process. J Clin Pathol. 1980;33:1146 –1152.
Anagnostopoulos A, Evangelopoulou A, Sotou D, Gika D, Mitsibounas
D, Dimopoulos MA. Incidence and evolution of monoclonal gammopathy of undetermined significance (MGUS) in Greece. Ann
Hematol. 2002;81:357–361.
Kyle RA, Spittell PC, Gertz MA, Li CY, Edwards WD, Olson LJ,
Thibodeau SN. The premortem recognition of systemic senile amyloidosis with cardiac involvement. Am J Med. 1996;101:395– 400.
Wright JR, Calkins E. Amyloid in the aged heart: frequency and clinical
significance. J Am Geriatr Socy. 1975;23:97–103.
Pucci A, Wharton J, Arbustini E, Grasso M, Diegoli M, Needleman P,
Vigano M, Polak JM. Atrial amyloid deposits in the failing human heart
display both atrial and brain natriuretic peptide-like immunoreactivity.
J Pathol. 1991;165:235–241.
Goette A, Rocken C. Atrial amyloidosis and atrial fibrillation: a genderdependent “arrhythmogenic substrate”? Eur Heart J. 2004;25:1185–1186.
Leone O, Boriani G, Chiappini B, Pacini D, Cenacchi G, Martin Suarez
S, Rapezzi C, Bacchi Reggiani ML, Marinelli G. Amyloid deposition as
a cause of atrial remodelling in persistent valvular atrial fibrillation. Eur
Heart J. 2004;25:1237–1241.
Rocken C, Peters B, Juenemann G, Saeger W, Klein HU, Huth C,
Roessner A, Goette A. Atrial amyloidosis: an arrhythmogenic substrate
for persistent atrial fibrillation. Circulation. 2002;106:2091–2097.
KEY WORDS: amyloid 䡲 cardiomyopathy
diseases 䡲 heart failure
䡲
echocardiography
䡲
heart
AHA Scientific Statement
Dietary Recommendations for Children and Adolescents
A Guide for Practitioners
Consensus Statement From the American Heart Association
Endorsed by the American Academy of Pediatrics
Samuel S. Gidding, MD, Chair; Barbara A. Dennison, MD, Cochair; Leann L. Birch, PhD;
Stephen R. Daniels, MD, PhD; Matthew W. Gilman, MD; Alice H. Lichtenstein, DSc;
Karyl Thomas Rattay, MD; Julia Steinberger, MD; Nicolas Stettler, MD; Linda Van Horn, PhD, RD
Abstract—Since the American Heart Association last presented nutrition guidelines for children, significant changes
have occurred in the prevalence of cardiovascular risk factors and nutrition behaviors in children. Overweight has
increased, whereas saturated fat and cholesterol intake have decreased, at least as percentage of total caloric intake.
Better understanding of children’s cardiovascular risk status and current diet is available from national survey data.
New research on the efficacy of diet intervention in children has been published. Also, increasing attention has
been paid to the importance of nutrition early in life, including the fetal milieu. This scientific statement
summarizes current available information on cardiovascular nutrition in children and makes recommendations for
both primordial and primary prevention of cardiovascular disease beginning at a young age. (Circulation. 2005;
112:2061-2075.)
Key Words: AHA Scientific Statements 䡲 adolescents 䡲 children 䡲 diet 䡲 nutrition
I
t is estimated that 75% to 90% of the cardiovascular
disease epidemic is related to dyslipidemia, hypertension,
diabetes mellitus, tobacco use, physical inactivity, and
obesity; the principal causes of these risk factors are adverse
behaviors, including poor nutrition.1–3 The atherosclerotic
process begins in youth, culminating in the risk factor–related
development of vascular plaque in the third and fourth
decades of life.4 – 6 Good nutrition, a physically active lifestyle, and absence of tobacco use contribute to lower risk
prevalence and either delay or prevent the onset of cardiovascular disease.2,3 These observations have established the
concept of prevention of the development of cardiovascular
risk factors in the first place, now called primordial prevention.7 Education, with the support of the healthcare community, combined with health policy and environmental change
to support optimal nutrition and physical activity, are central
to this health strategy.
This document provides dietary and physical activity
recommendations for healthy children; discusses the current content of children’s diets; reviews the adverse health
consequences of increased intakes of calories (relative to
energy expenditure), saturated and trans fat, and cholesterol; and provides age-specific guidelines for implementation of the recommended diet, including the period from
before birth to 2 years of age. Medical practitioners are the
intended audience, and guidelines to implement recommendations in clinical practice settings are provided.
Public health strategies for improving the quality of
children’s diets are also discussed.
This scientific statement on optimal cardiovascular nutrition for infants, children, and adolescents revises the 1982
document on the same topic and also builds on the recent
consensus statement on optimal nutrition for the prevention
of many chronic diseases of adulthood.8,9 This revision
responds to the obesity epidemic that has emerged since the
publication of the last statement that addressed children’s
nutrition from the American Heart Association (AHA) and
has new focuses on both total caloric intake and eating
behaviors.10,11 This revision strongly conveys the message
that foods and beverages that fulfill nutritional requirements
The American Heart Association makes every effort to avoid any actual or potential conflicts of interest that may arise as a result of an outside
relationship or a personal, professional, or business interest of a member of the writing panel. Specifically, all members of the writing group are required
to complete and submit a Disclosure Questionnaire showing all such relationships that might be perceived as real or potential conflicts of interest.
This statement was approved by the American Heart Association Science Advisory and Coordinating Committee on July 22, 2005. A single reprint
is available by calling 800-242-8721 (US only) or writing the American Heart Association, Public Information, 7272 Greenville Ave, Dallas, TX
75231-4596. Ask for reprint No. 71-0341. To purchase additional reprints: up to 999 copies, call 800-611-6083 (US only) or fax 413-665-2671; 1000
or more copies, call 410-528-4121, fax 410-528-4264, or E-mail [email protected]. To make photocopies for personal or educational use, call the
Copyright Clearance Center, 978-750-8400.
Expert peer review of AHA Scientific Statements is conducted at the AHA National Center. For more on AHA statements and guidelines development,
visit http://www.americanheart.org/presenter.jhtml?identifier⫽3023366.
2061
2062
Circulation
September 27, 2005
TABLE 1. AHA Pediatric Dietary Strategies for Individuals
Aged >2 Years: Recommendations to All Patients and Families
TABLE 2. Tips for Parents to Implement AHA Pediatric
Dietary Guidelines
Balance dietary calories with physical activity to maintain normal growth
Reduce added sugars, including sugar-sweetened drinks and juices
60 Minutes of moderate to vigorous play or physical activity daily
Eat vegetables and fruits daily, limit juice intake
Use canola, soybean, corn oil, safflower oil, or other unsaturated oils in
place of solid fats during food preparation
Use vegetable oils and soft margarines low in saturated fat and trans fatty
acids instead of butter or most other animal fats in the diet
Use recommended portion sizes on food labels when preparing and serving
food
Eat whole grain breads and cereals rather than refined grain products
Use fresh, frozen, and canned vegetables and fruits and serve at every
meal; be careful with added sauces and sugar
Reduce the intake of sugar-sweetened beverages and foods
Use nonfat (skim) or low-fat milk and dairy products daily
Eat more fish, especially oily fish, broiled or baked
Reduce salt intake, including salt from processed foods
are appropriate for growing and developing infants, children,
and adolescents. Calorie-dense foods and beverages with
minimal nutritional content must return to their role as
occasional discretionary items in an otherwise balanced diet.
A critical component of contemporary guidelines is the
strength of the scientific evidence base for recommendations.
Whereas the scientific base for understanding the potential
harm and benefit of current dietary practices and the relationship to risk factors is strong, the scientific base for recommended interventions is weaker for several reasons: limited
number, statistical power, and scope of intervention studies;
limited efficacy of attempted interventions; and lack of
generalizability of studies of feeding behaviors at younger
ages. Historically, most have had small sample size and have
not had ethnic diversity among participants. Nonetheless,
given the current obesity epidemic, sufficient natural history
and prevalence data exist to justify intervention, although
continued evaluation is necessary to identify optimal
strategies.12
Dietary Recommendations
The general dietary recommendations of the AHA for those
aged 2 years and older stress a diet that primarily relies on
fruits and vegetables, whole grains, low-fat and nonfat dairy
products, beans, fish, and lean meat.1,13 These general recommendations echo other recent public health dietary guidelines in emphasizing low intakes of saturated and trans fat,
cholesterol, and added sugar and salt; energy intake and
physical activity appropriate for the maintenance of a normal
weight for height; and adequate intake of micronutrients.14 –16
Tables 1 and 2 provide strategies for implementing healthy
cardiovascular nutrition. The recently published Dietary
Guidelines for Americans (for those 2 years of age and older)
and American Academy of Pediatrics Nutrition Handbook
provide important supporting reference information with
regard to overall diet composition, appropriate caloric intakes
at different ages, macronutrients, micronutrients, portion size,
and food choices.14,17,18 Table 3 provides daily estimated
calorie and serving recommendations for grains, fruits, vegetables, and milk/dairy products by age and gender. Consistent with the Dietary Guidelines for Americans, 2005,14,18
nutrient and energy contributions from each food group are
calculated according to the nutrient-dense forms of foods in
each group (eg, lean meats and fat-free milk), with the
Introduce and regularly serve fish as an entrée
Remove the skin from poultry before eating
Use only lean cuts of meat and reduced-fat meat products
Limit high-calorie sauces such as Alfredo, cream sauces, cheese sauces,
and hollandaise
Eat whole grain breads and cereals rather than refined products; read labels
and ensure that “whole grain” is the first ingredient on the food label of
these products
Eat more legumes (beans) and tofu in place of meat for some entrées
Breads, breakfast cereals, and prepared foods, including soups, may be
high in salt and/or sugar; read food labels for content and choose
high-fiber, low-salt/low-sugar alternatives
exception of the guidelines for 1-year-old children, which
included 2% fat milk. For youth 3 years of age and older,
calorie estimates are based on a sedentary lifestyle. More
physically active children and adolescents will require additional calories.14,17–19 This table is provided as a starting point
for dietary counseling; recommendations will need to be
individualized in clinical practice. Table 4 provides daily
recommended intakes of sodium, potassium, and fiber.18
More complete guidelines for infants, particularly with regard
to the transition from breast/formula-feeding to table foods,
will be discussed below.
Emphases different from the past include the allowance of
a more liberal intake of unsaturated fat and a focus on
ensuring adequate intakes of omega-3 fatty acids. There is an
emphasis on foods that are rich in nutrients and that provide
increased amounts of dietary fiber. The AHA continues to
recommend diets low in saturated and trans fats. Healthy
foods include fruits, vegetables, whole grains, legumes,
low-fat dairy products, fish, poultry, and lean meats. Fruits,
vegetables, and fish are often inadequately consumed by
children and adolescents. Because the major sources of
saturated fat and cholesterol in children’s diets are full-fat
milk and cheese and fatty meats, use of low-fat dairy products
and lean cuts of meat in appropriate portion sizes will be
critical in meeting dietary needs and nutrient requirements.20
Fish is an important food with growing evidence of
potential benefit. However, consumers may have difficulty in
distinguishing among several health messages about fish
consumption. Although strong data associate cardiovascular
disease prevention with increased fish consumption, there are
also concerns about potential polycarbonate phenols (PCBs)
and mercury contamination.21,22 The Food and Drug Administration (FDA) and AHA stress that seafood is an important
part of a healthy diet and advocate consumption of a wide
variety of fish and shellfish. Current FDA recommendations
with regard to limiting fish intake pertain to women who may
Gidding et al
2063
TABLE 3. Daily Estimated Calories and Recommended Servings for Grains, Fruits, Vegetables,
and Milk/Dairy by Age and Gender
4 – 8 Years
9 –13 Years
14 –18 Years
Female
1200 kcal
1600 kcal
1800 kcal
Male
1400 kcal
1800 kcal
2200 kcal
Calories†
Fat
Milk/dairy‡
Lean meat/beans
1 Year
2–3 Years
900 kcal
1000 kcal
30%–40% kcal
30%–35% kcal
25%–35% kcal
25%–35% kcal
25%–35% kcal
2 cups¶
2 cups
2 cups
3 cups
3 cups
1.5 oz
2 oz
5 oz
Female
3 oz
5 oz
Male
4 oz
6 oz
Fruits§
1 cup
1 cup
1.5 cups
1.5 cups
Female
1.5 cups
Male
Vegetables§
2 cups
3/4 cup
1 cup
Female
1 cup
2 cups
2.5 cups
1.5 cup
2.5 cups
3 cups
Female
4 oz
5 oz
6 oz
Male
5oz
6 oz
7 oz
Male
Grains储
2 oz
3 oz
*Calorie estimates are based on a sedentary lifestyle. Increased physical activity will require additional calories: by
0-200 kcal/d if moderately physically active; and by 200 – 400 kcal/d if very physically active.
†For youth 2 years and older; adopted from Table 2, Table 3, and Appendix A-2 of the Dietary Guidelines for
Americans (2005)14; http://www.healthierus.gov/dietaryguidelines. Nutrient and energy contributions from each group
are calculated according to the nutrient-dense forms of food in each group (eg, lean meats and fat-free milk).
‡Milk listed is fat-free (except for children under the age of 2 years). If 1%, 2%, or whole-fat milk is substituted,
this will utilize, for each cup, 19, 39, or 63 kcal of discretionary calories and add 2.6, 5.1, or 9.0 g of total fat, of
which 1.3, 2.6, or 4.6 g are saturated fat.
§Serving sizes are 1/4 cup for 1 year of age, 1/3 cup for 2 to 3 years of age, and 1/2 cup for ⱖ4 years of age.
A variety of vegetables should be selected from each subgroup over the week.
储Half of all grains should be whole grains.
¶For 1-year-old children, calculations are based on 2% fat milk. If 2 cups of whole milk are substituted, 48 kcal
of discretionary calories will be utilized. The American Academy of Pediatrics recommends that low-fat/reduced fat
milk not be started before 2 years of age.
become pregnant or are already pregnant, nursing mothers,
and young children. The FDA recommends that people in
those categories avoid shark, swordfish, king mackerel, and
tilefish because they contain high levels of mercury. Five of
the most commonly eaten varieties of fish are low in mercury
(shrimp, canned light tuna, salmon, pollack, and catfish). The
AHA continues to recommend 2 servings of fish weekly.23
Recent evidence suggests that commercially fried fish products, likely because they are relatively low in omega-3 fatty
acids and high in trans fatty acids (if hydrogenated fat is used
for preparation), do not provide the same benefits as other
sources of fish.24
Discretionary Calories
The obesity epidemic has prioritized consideration of the
complex issue of matching appropriate energy intake to
energy expenditure.10,11 One approach is the concept of
discretionary calories illustrated in Figure 1.14 Total caloric
intake is the sum of essential calories, the total energy intake
necessary to meet recommended nutrient intakes, and discretionary calories, the additional calories necessary to meet
energy demand and for normal growth.18 The figure shows
essential calories and discretionary calories; these increase
with age and increasing levels of physical activity. There is a
large difference in the discretionary calorie allowance
among sedentary, moderately active, and active children,
with more physically active children needing more energy
from food to maintain normal growth. For young sedentary
children, the amount of total energy intake that can come
from foods used purely as a source of energy, ⬇100 to 150
calories, is less than that provided by a usual portion size
of most low-nutrient-dense snacks and beverages. With
increasing activity, this discretionary calorie amount may
increase to 200 to 500 calories, depending on the age and
gender of the child and the level of physical activity. The
message portrayed by Figure 1 is clear: To be sedentary,
have a nutritionally adequate diet, and to avoid excessive
caloric intake in contemporary society is difficult.25 The
challenge to healthcare providers and public health professionals is to translate the complex science-based energy
balance message from Figure 1 into effective practice and
public health policy.25a Consuming diets that include primarily nutrient-dense forms of the foods listed in Table 3,
2064
Circulation
September 27, 2005
TABLE 4. Daily Recommended Intakes of Fiber, Sodium, and
Potassium by Age and Gender
Gender/
Age
Fiber,
g*
Sodium,
mg
Potassium,
mg
19
⬍1500
3000
Female
25
⬍1900
3800
Male
25
⬍1900
3800
Female
26
⬍2200
4500
Male
31
⬍2200
4500
Female
29
⬍2300
4700
Male
38
⬍2300
4700
1–3 y
4–8 y
9–13 y
14–18 y
*Total fiber preferred minimum 14 g/1000 kcal. Read labels to determine
amounts on all packaged foods. Adapted from the report of the 2005 Dietary
Guideline Advisory Committee on Dietary Guidelines for Americans.18
participating in regular moderate to vigorous physical
activity most days of the week for at least 1 hour per day,
and limiting video screen time to less than 2 hours per day
will help accomplish this goal.
Scientific Support for Current
Dietary Recommendations
The importance of dietary saturated and trans fat and cholesterol to the development of elevated cholesterol and subsequent cardiovascular disease as well as other cardiovascular
risk factors has been extensively studied and reviewed.
Pathological evidence demonstrates that as the number of
cardiovascular risk factors increases, so does the evidence of
atherosclerosis in the aorta and coronary arteries beginning in
early childhood.4,5,26 Evidence of increased carotid artery
intima-media thickness and coronary artery calcium mea-
sured by electron beam computed tomography among 29- to
39-year-old young adults who have been monitored from
childhood further documents that the significant precursors of
adult atherosclerosis are obesity, elevated blood pressure, and
dyslipidemia.6,27,28 Epidemiological data from longitudinal
studies provide further evidence that overweight, hypercholesterolemia, and hypertension track over time from childhood into adult life and that lifestyle choices, eg, diet, excess
caloric intake, physical inactivity, and cigarette smoking,
influence these risk factors.29 –34 Intervention studies aimed at
measuring the efficacy and safety of diets reduced in total and
saturated fat and cholesterol have also now contributed
evidence at both the clinical and school-based levels.35–37
A meta-analysis of adult studies of low–saturated fat,
low-cholesterol diets suggested that introduction of the diet
lowers LDL cholesterol an average of 12%, with a 1.93mg/dL decline in LDL cholesterol for every 1% decline in
saturated fat.38 Further restricting saturated fat from 10% of
total energy to 7% (the Therapeutic Lifestyle Change diet)
increased the LDL cholesterol reduction to 16%.38,39 Pediatric
confirmation of adult studies showing safety and efficacy of
a low-cholesterol and low–saturated fat diet has emerged. The
Dietary Intervention Study in Children (DISC) was a randomized trial of a low–saturated fat, low-cholesterol diet conducted over 3 years in US children initially prepubertal and
aged 8 to 11 years.35 The Special Turku Risk Intervention
Program (STRIP) was a randomized dietary intervention trial
begun at weaning (age ⬇7 months) with parental dietary
education continued through the age of 7 years.40 – 42 Both
studies achieved diets in intervention groups consistent with
current recommendations for therapeutic lifestyle changes to
lower elevated cholesterol levels, with total fat ⬍30% of total
calories and cholesterol intake ⬍200 mg/d.39 Saturated fat
intake, although not ⬍7% of total calories, was significantly
less than in children assigned to usual care. Across a wide
array of safety measures, including measures of growth,
Figure 1. Concept of discretionary calories by gender. As daily physical activity increases, more energy is needed for normal growth.
For sedentary children, only small amounts of discretionary calories can be consumed before caloric intake becomes excessive. Discretionary calories for children aged 4 to 8 years are based on 2 servings of dairy per day. Mod Act indicates moderately active. Based
on estimated calorie requirements and discretionary calories published in Dietary Guidelines for Americans (2005).14
Gidding et al
2065
neurological development, metabolic function, and nutrient
adequacy, no adverse effects of the recommended intervention diets were observed.40,43– 45 LDL cholesterol levels were
significantly lower among children receiving dietary intervention in the DISC study and in boys receiving dietary
intervention in the STRIP study compared with controls.35,42
Most importantly, in both studies children receiving dietary
intervention were significantly more likely to make healthy
food choices.25 Three-year follow-up of children with severe
hyperlipidemia who were following recommended therapeutic lifestyle changes showed no adverse effects on growth and
development.46
The relationship between obesity and multiple cardiovascular risk factors, including elevated blood pressure, dyslipidemia, low physical fitness, and insulin resistance/diabetes
mellitus, is well established.10,47– 49 Both excess caloric intake
and physical inactivity are strongly associated with obesity.50
Studies of weight loss in overweight individuals consistently
show improvement in obesity-related comorbidities, particularly when interventions include regular exercise in the
treatment program.51–53 Population-based cross-sectional
studies of secular trends in cardiovascular risk have shown
strong associations between increasing prevalence of obesity
and increasing blood pressure levels but inconsistent trends in
dyslipidemia.47,54,55 Longitudinal studies of secular trends in
children, however, have shown strong relationships between
increases in adiposity and adverse trends in blood pressure
and lipids.56 Maintaining body mass index (BMI) is beneficial, even without weight loss, because this prevents worsening of risk status.57 Maintenance of body weight during
normal growth will improve BMI and cardiovascular risk
status. Although primary prevention trials of reduction of
daily caloric intake in at-risk children are under way, the
evidence for harm from excess caloric intake is sufficient to
support public health efforts for obesity prevention.11,58,59
fruits, 79% meat, and 11% some type of sweetened beverages.61 Sweetened beverages have been consumed by 28% of
the 12- to 14-month-old children, 37% of the 15- to 18month-old children, and 44% of the 19- to 24-month-old
children.61 During the transition from a milk-based diet to
adult foods, the types of vegetables consumed change adversely. Deep yellow vegetables are consumed by 39% of
children at 7 to 8 months and by 13% at 19 to 24 months,
whereas French fries become the most commonly consumed
vegetable by this age.61 Similarly, fruit consumption declines
to the point where one third of 19- to 24-month-old children
consume no fruit, whereas 60% consume baked desserts, 20%
candy, and 44% sweetened beverages on a given day.61
Significant adverse changes have occurred in older children’s food consumption.62 These include a reduction in
regular breakfast consumption, an increase in consumption of
foods prepared away from the home, an increase in the
percentage of total calories from snacks, an increase in
consumption of fried and nutrient-poor foods, a significant
increase in portion size at each meal, and an increase in
consumption of sweetened beverages, whereas dairy product
consumption has decreased, and a shift away from high-fiber
fruits and vegetables as well as a general decline in fruit and
vegetable consumption other than potatoes.62– 67 Fried potatoes make up a substantial portion of the vegetable intake.67
Sugar consumption has increased, particularly in preschool
children.68 With regard to micronutrients, the shift in dietary
patterns has resulted in median intakes below recommended
values of many important nutrients during adolescence.69
Sodium intake is far in excess of recommended levels,
whereas calcium and potassium intakes are below recommended levels.69 –71
What Children Currently Eat
This section reviews age-specific pediatric research on cardiovascular and general nutrition. Although in some areas
there is a reasonable body of work about which to make
useful judgments, in many areas studies have significant
methodological limitations: small sample size, confounding
by a variety of factors (including cultural factors), and
difficulty of using classic randomized trial designs to answer
pertinent research questions. Nevertheless, the current dietary
pattern of contemporary children mandates change. The
recommendations provided herein are based on expert consensus of emerging evidence. Their purpose is to improve the
nutritional quality, amount, and pattern of food consumption
by children and their families. Although the narrative emphasizes nutrition to improve cardiovascular risk, it is recognized
that optimal nutrition for overall health and normal growth is
the preeminent goal.
Recommendations for children’s nutrition consider the
family and cultural milieu.72,73 It has been decades since the
majority of meals were consumed within the home.60 – 67
Sources of nourishment include schools, child-care and afterschool youth programs, restaurants, vending machines, convenience stores, work sites, and foods prepared by industry
designed for minimal preparation time in the household.74
It is important to understand the gap between current dietary
practices and recommended diets for infants, children, and
adolescents. Sufficient population-based data exist to identify
the magnitude of the problem confronting those interested in
improving cardiovascular health in youth. Areas to consider
include appropriateness of total caloric intake, eating patterns, balance of foods/beverages chosen from each food
group, and intake of specific nutrients. Published data evaluate each of these areas with age and gender as important
associated considerations.
For infants, it is encouraging that ⬇76% of mothers have
initiated breast-feeding.60 However, maintenance of breastfeeding for the first 4 to 6 months of life has been less
successful. Only 4% of infants participating in the Special
Supplemental Nutrition Program for Women, Infants, and
Children (WIC) and 17% of the nonparticipants remain
exclusively breast fed at 6 months of age. This suggests a
strong socioeconomic status gradient in breast-feeding behavior. By 4 to 6 months, 66% of infants have received grain
products, 40% vegetables, 42% fruits, 14% meat, and 0.6%
some type of sweetened beverages.61 By 9 to 11 months, 98%
of infants have received grain products, 73% vegetables, 76%
Implementation of Dietary Recommendations
Including Considerations for Specific
Age Groups
2066
Circulation
September 27, 2005
TABLE 5. Parent, Guardian, and Caregiver Responsibilities for
Children’s Nutrition
Choose breast-feeding for first nutrition; try to maintain for 12 months
Control when food is available and when it can be eaten (nutrient quality,
portion size, snacking, regular meals)
Provide social context for eating behavior (family meals, role of food in
social intercourse)
Teach about food and nutrition at the grocery store, when cooking meals
Counteract inaccurate information from the media and other influences
Teach other care providers (eg, daycare, babysitters) about what you want
your children to eat
Serve as role models and lead by example; “do as I do” rather than “do as
I say”
Promote and participate in regular daily physical activity
Common situations affecting food preparation include households in which both parents work, single-parent households,
and work schedules that demand that parents be away from
home at mealtime. Likewise, children have complex schedules that demand frequent meals away from home. Schools
provide less education on food preparation (eg, home economics) than in the past.
Culture-specific dietary practices can influence the diet
both for better and for worse. A specific problem is the folk
belief that a fat baby or chubby toddler is healthy. Popular fad
diets often mix helpful and harmful components in their
educational messages. Layered on top of this is a largely
unregulated media dedicated to selling large quantities of a
wide array of foods and food products of poor nutritional
value. Despite unparalleled availability of nutrition resources,
sifting through the food message bombardment is often the
most difficult task facing a parent interested in providing
proper nutrition for his or her family. Teaching those involved in supervision of children’s diets to consume a
healthful diet themselves and thus provide consistent role
model behavior improves diet quality.73,75 Table 5 provides a
summary of areas in which adult influences are most important with regard to childhood nutrition.76 – 81 Parents choose
the time for meals and snacks and the types of foods and
beverages to be served. Children can then choose how much
to consume. Parents, guardians, and caregivers must provide
appropriate role modeling through their own behavior, that is,
influence children to “do as I do” rather than “do as I say.” A
similar responsibility falls on those who attempt to provide
reliable information to parents and educators in an effort to
counterbalance adverse folk/cultural practices, media influences, and other sources of disinformation. Also critical to
implementation of nutritional change is the social perception
of risk.82 Unless people believe that certain dietary practices
are harmful or food providers believe that their actions
endanger their clients, motivation to change will be limited.
Increasing social pressure for eating properly can counteract
the ubiquitous presence of food and food marketing of
energy-dense, nutrient-poor choices.
Birth to 2 Years
There has been considerable interest in the influence of both the
intrauterine environment and infant nutrition on future cardio-
vascular risk.83,84 It has been hypothesized that “programming”
of future cardiovascular responses is established either in the
womb or in response to feeding exposures early in life. Animal
models support the programming hypothesis, but there are as yet
few human experimental data.85– 87 Lower birth weight, because
of presumed intrauterine malnutrition and association with rapid
postnatal rapid weight gain, is associated with central adiposity,
the metabolic syndrome, diabetes mellitus, and cardiovascular
disease outcomes in adulthood.88 Babies large for gestational
age, probably through consequences of maternal insulin resistance and glucose intolerance, are also at higher risk of future
obesity.89,90
It is important for parents or parents-to-be to obtain a
healthy weight because children whose mothers are obese
early in pregnancy are more likely to be overweight as young
children.91 A similar effect is seen in children whose parents
are or become obese during their childhood.31 To ensure
optimal growth of the fetus, pregnant women must optimize
their nutrition and weight gain during pregnancy, according
to the Institute of Medicine guidelines.92 Excessive maternal
weight gain has been associated with a 2- to 3-fold increased
risk that the mother will be overweight after a pregnancy.93
This may increase subsequent offspring risk during adolescence for obesity, impaired glucose tolerance, impaired insulin secretion, and type 2 diabetes. Studies of maternal
nutrition, for example, assessments of protein and calcium
intake, suggest that maternal diet during pregnancy may
influence offspring’s blood pressure.94,95 However, evidence
is insufficient to make specific recommendations about nutrition during pregnancy based on future cardiovascular
disease.
Human milk is uniquely superior for infant feeding and is
the reference against which other infant feeding strategies
must be measured.96 Breast milk is rich in both saturated fat
and cholesterol but low in sodium. There has been substantial
work on the relationship of breast-feeding to both future
cardiovascular events and cardiovascular risk factors. Although pooling estimates from these studies is difficult
because of differences in exposure and outcome assessment,
recent meta-analyses have suggested no meaningful impact of
breast-feeding on subsequent cardiovascular or all-cause
mortality in adulthood.97 Other systematic reviews, however,
suggest benefits of breast-feeding, particularly in the prevention of future obesity.98,99 Several studies suggest that breastfeeding leads to lower blood pressure later in childhood.100,101
Although breast-feeding is associated with higher blood
cholesterol levels at 1 year of age, it may also result in lower
blood cholesterol levels in adults.102 Rapid weight gain
during the first 4 to 6 months of life is associated with future
risk of overweight103,106; studies suggest that partially breastfed and formula-fed infants consume 20% more total calories
per day than do exclusively breast-fed infants.104,105 Physicians should identify infants who are gaining weight rapidly
and/or whose weight-to-length percentile exceeds the 95th
percentile to help correct overfeeding if present.
At least 2 behavioral benefits of breast-feeding may lead to
reduced cardiovascular risk, but the impact of these has not
been studied in large trials.107–110 The first potential benefit
may be better self-regulation of intake. Compared with
Gidding et al
2067
Figure 2. Dramatic change in food
sources during the first 2 years of life.
Diet is initially based on breast milk at
birth and transitions to conventional
foods by 2 years, although dairy products remain a major source of energy
and nutrition. Reprinted from Lederman
et al126a with permission of Pediatrics.
Adapted from Devaney et al129 and
Skinner et al.130
parents who bottle-feed, mothers who breast-feed appear to
allow the infant to take an active role in controlling intake,
possibly promoting maternal feeding practices that can foster
better self-regulation of energy intake as the child grows
up.108 Children with improved self-regulation may better
withstand the current food surplus environment.111 The second potential benefit relates to taste preference.112–116 Both
amniotic fluid and breast milk provide flavor exposure to the
fetus and infant. These exposures influence taste preference
and food choices after weaning. Thus, exposure to healthier
foods through maternal food consumption during pregnancy
and lactation may improve acceptance of healthy foods after
weaning. Because infant responses to taste are different than
mature taste, these early exposures may be critical in determining food preference later in life.
A critical social problem for mothers interested in breastfeeding in the United States is the lack of a tolerant social
structure.117 Breast-feeding rates decline rapidly between 2
and 3 months, which is when many mothers return to work or
school.118,119 Full-time employment is consistently associated
with shorter periods of breast-feeding.120 –123 In Scandinavian
countries, where women routinely receive paid maternity
benefits (eg, 42 weeks with full pay or 52 weeks at 80% of
salary in Norway), women far surpass the US Healthy People
2010 goals for breast-feeding. In Norway, 97% of women are
breast-feeding when they leave the hospital, 80% are breastfeeding at 3 months, and 36% are breast-feeding at 12
months.124 Most African and Asian countries are highly
supportive of breast-feeding. Policies enacted within the
workplace and public places can also help to overcome
barriers to breast-feeding.117,125
The period from weaning to consumption of a mature diet,
from 4 to 6 months to ⬇2 years of age, represents a radical
shift in pattern of food consumption (Figure 2),17,126,126a,129,130
but there has been very little research on the best methods to
achieve optimal nutritional intakes during this transition.
Infants mature from receiving all nutrition from a milk-based
diet to a diet chosen from the range of adult foods, in part
self-selected and in part provided by caregivers. Transition to
other sources of nutrients should begin at 4 to 6 months of age
to ensure sufficient micronutrients in the diet, but the best
methods for accomplishing this task are essentially unknown.15,126 Current feeding practices and guidelines are
influenced by small-scale studies of infant feeding behavior,
idiosyncratic parental behavior, and popular opinion.17,60,127,128
Food consumption data suggest that infants are currently
exposed to a wide variety of “kid” foods that tend to be high
in fat and sugar, including excess juice, juice-based sweetened beverages, French fries, and nutrient-poor snacks.61
Usual food intakes of infants and young children may exceed
estimated energy requirements. For infants aged 0 to 6
months, reported intakes exceed requirements by 10% to
20%; for children aged 1 to 4 years, intakes exceed requirements by 20% to 35%. Although some of these reports may
reflect overreporting of food intake, these data might also
explain the rise in the prevalence of overweight at very young
ages.129,130
For those participating in public nutrition assistance programs (US Department of Agriculture [USDA] 2002), the
foods supplied for infants and children are limited in variety,
reflecting more closely the nutritional concerns of the 1970s,
when the program was designed (inadequate calories, protein,
vitamin A, vitamin C, and iron), than nutritional concerns
today (excess calories, fat, and sugar and inadequate fruits,
vegetables, and whole grains).131 Moreover, beverages provided to most children are not optimal. Children aged 1 to 5
years enrolled in WIC receive twice the amount of fruit juice
(9.5 fl oz/d) currently recommended, and most participants
also receive or choose whole milk.14,17,131,132 For formula-fed
infants, there may also be a role for clearer prescriptive
feeding advice for parents to understand their infant’s satiety
cues and appropriate energy intake than is currently the norm.
Table 6 provides a number of strategies to improve general
and cardiovascular nutrition during this transitional stage.
When normal growth is present, overfeeding may result from
arbitrarily increasing amounts fed to achieve specific portion
sizes per meal rather than allowing infants and toddlers to
self-regulate. New healthy foods may need to be introduced
TABLE 6.
Improving Nutritional Quality After Weaning
Maintain breast-feeding as the exclusive source of nutrition for the first 4 – 6
months of life
Delay the introduction of 100% juice until at least 6 months of age and limit
to no more than 4 – 6 oz/d; juice should only be fed from a cup
Respond to satiety clues and do not overfeed; infants and young children
can usually self-regulate total caloric intake; do not force children to finish
meals if not hungry as they often vary caloric intake from meal to meal
Introduce healthy foods and continue offering if initially refused; do not
introduce foods without overall nutritional value simply to provide calories
2068
Circulation
September 27, 2005
TABLE 7.
Improving Nutrition in Young Children
Parents choose meal times, not children
Provide a wide variety of nutrient-dense foods such as fruits and vegetables
instead of high-energy-density/nutrient-poor foods such as salty snacks, ice
cream, fried foods, cookies, and sweetened beverages
Pay attention to portion size; serve portions appropriate for the child’s size
and age
Use nonfat or low-fat dairy products as sources of calcium and protein
Limit snacking during sedentary behavior or in response to boredom and
particularly restrict use of sweet/sweetened beverages as snacks (eg, juice,
soda, sports drinks)
Limit sedentary behaviors, with no more than 1 to 2 hours per day of video
screen/television and no television sets in children’s bedrooms
Allow self-regulation of total caloric intake in the presence of normal BMI or
weight for height
Have regular family meals to promote social interaction and role model
food-related behavior
repeatedly, as many as 10 times to establish taste
preferences.133
Age 2 to 6 Years
At this age, recommendations for diet content are similar to
those for older individuals. Challenges here relate to providing quality nutrient intake and avoiding excess caloric intake.
Dairy products are a major source of saturated fat and
cholesterol in this age group, and therefore a transition to
low-fat milk and other dairy products is important.34,134
Sweetened beverages and other sugar-containing snacks are a
major source of caloric intake.135,136 Table 7 provides a list of
strategies for managing nutrition in young children.77,78,137–141
Parents should remember that they are responsible for choosing foods that are eaten and when and where they are eaten.
The child is responsible for whether or not he or she wants to
eat and how much. Two natural parental impulses, pressuring
children to eat and restricting access to specific foods, are not
recommended because they often lead to overeating, dislikes,
and paradoxical interest in forbidden items.142,143
Healthcare providers must provide useful advice to parents,
but they are constrained by time pressures in the typical
health maintenance office visit. In addition to the information
in Table 7, advice on caloric/energy values of food, particularly nutrient-poor foods, can be provided in a relatively short
period of time. At office visits, BMI percentile can be plotted,
the appropriateness of weight gain in the last year can be
assessed from standard growth curves, and recommendations
for optimal weight gain in the next year can be given. Blood
pressure screening and cholesterol measurement, if indicated,
are begun in this age range.7
Ages 6 Years and Above
As children grow up, sources of food and influences on eating
behavior increase. Social constraints on families may necessitate the presence of multiple caregivers, eating out, and
frequent fast food consumption. Many children, because of
parental work schedules, are home alone and prepare their
own snacks and meals. By early adolescence, peer pressure
begins to usurp parental authority, and fad diets may be
initiated. Many meals and snacks are routinely obtained
outside the home, often without supervision. Sites include
schools, friends’ homes, child-care centers, and social events.
Older children have discretionary funds to use for selfselected foods. Current eating patterns do not at all resemble
the “norm” of providing at least breakfast, dinner, and a
single snack at home with lunch carried to school or purchased from a health-conscious cafeteria. For example, current diet studies suggest that many children do not eat
breakfast and get at least one third of calories from snacks.
Sweetened beverage intakes contribute significantly to total
caloric intake.144 Sweetened beverages and naturally sweet
beverages, such as fruit juice, should be limited to 4 to 6 oz
per day for children 1 to 6 years old, and to 8 to 12 oz per day
for children 7 to 18 years old.144,145 These snacks often
contribute to excess consumption of discretionary calories
and/or supplant the intake of foods containing essential
nutrients.25,146
Adolescence is a nutritionally vulnerable developmental
stage because growth rate accelerates. Amplified caloric and
global nutrition needs due to pubertal growth stimulate
appetite. The combination of centrally driven appetite stimulation and an increasingly sedentary lifestyle due to a
decline in recreational sports participation augments obesity.50 Parallel to the psychosocial transition from dependence
on parental authority to independent thought processes, food
choices and purchases are increasingly made by the adolescent. Peer pressure for conformity, in part driven by media
promotion of fast food directly to teens, makes overeating
natural. Currently, adolescents have an increased intake of
sweetened beverages, French fries, pizza, and fast food
entrées, including hamburgers, and a consequent lack of
recommended fruits, vegetables, dairy foods, whole grains,
lean meats, and fish. This change in eating pattern results in
consumption of excess fat, saturated fat, trans fats, and added
sugars along with insufficient consumption of micronutrients
such as calcium, iron, zinc, and potassium, as well as
vitamins A, D, and C and folic acid.146,147
Counseling of older children and adolescents must be
individualized to accommodate the range of contemporary
lifestyles; less success is achieved at older ages. Current
dietary practices and readiness to change must be understood
before family-based intervention is attempted.148 Parental
role modeling is important in establishing children’s food
choices.53,78,132 Depending on their own food choices, parents
can be either positive or negative role models.81
Public Health Issues
Modern life extends the umbrella of social responsibility for
provision of appropriate nutrition and nutrition knowledge
beyond the home to government, the health professions,
schools, the food industry, and the media. It is beyond the
scope of this document to evaluate the large public health
effort related to overweight and nutrition now being undertaken. Some important areas are highlighted below. Because
there is little scientific information to guide current policy
directed at changing eating behaviors, it is strongly recommended that evaluation, safety, and efficacy tools be incorporated into policy implementation.
Gidding et al
TABLE 8.
Strategies for Schools
Identify a “champion” within the school to coordinate healthy nutrition
programs
Establish a multidisciplinary team including student representation to assess
all aspects of the school environment using the School Health Index
(Centers for Disease Control and Prevention) or similar assessment
Identify local, regional, and national nutrition programs; select those proven
effective (http//www.ActionForHealthyKids.org)
Develop policies that promote student health and identify nutrition issues
within the school (http//www.nasbe.org/HealthySchools/healthy_eating.html)
Work to make predominantly healthful foods available at school and school
functions by influencing food and beverage contracts, adapt marketing
techniques to influence students to make healthy choices, and restrict
in-school availability of and marketing of poor food choices
2069
TABLE 9. Types of Legislation Under Consideration to Improve
Children’s Nutrition
Measurement of BMI by school staff for health surveillance and/or to report
information to parents
Restriction of certain types of food and beverages available on school
grounds
Taxation of specific foods or sedentary forms of entertainment
Establishment of local school wellness policies using a multidisciplinary
team of school staff and community volunteers (mandated for schools
participating in federal reimbursable school lunch, breakfast, or milk
programs)
Food labeling regulations, including appropriate descriptions of portion sizes
(eg, a medium-sized sugar-containing drink should be 6 – 8 oz)
Regulation of food advertising directed at children
Maximize opportunities for all physical activity and fitness programs
(competitive and intramural sports); utilize coaches/teachers as role models
Lobby for regulatory changes that improve a school’s ability to serve
nutritious food
Ban food advertising on school campuses
Schools have become a battleground for fighting the
obesity epidemic.149,150 Cafeterias are under attack for serving
unhealthy food, yet the food provided is constrained by
budgetary and regulatory issues largely external to public
health concerns. USDA guidelines require school food programs to provide minimum quantities of specific nutrients
over a 3- to 7-day span but do not address maximum food
amounts. Vending machines and competing nutrient-poor
foods provide excess calories but also provide revenue to
support school programs. Table 8 summarizes some strategies
currently being implemented in many locales.
Nutrition education in schools is considered useful in
improving knowledge about nutrition, but few studies suggest
that it is effective in altering eating behaviors in the absence
of environmental change.150 The largest study, the Child and
Adolescent Trial for Cardiovascular Health (CATCH), was a
multicenter intervention that included nutrition education, a
cafeteria intervention, changes in physical education programs, and parental education. The fat content of school
lunch but not school breakfast was modified, and blood
cholesterol levels and children’s weight status were unchanged.36,151,152 Other school studies have shown improvements in fruit (but not vegetable) intake.153–155 High-quality
foods sold in vending machines will be selected if they are
competitively priced.62 An intervention that included nutrition education, a cafeteria intervention, changes in physical
activity, and a parent component for younger children attending Head Start programs was successful in decreasing children’s blood cholesterol levels but did not affect the prevalence of childhood overweight.156
Physical education programs are often subject to budget
constraints. The percentage of pupils attending physical
education classes has decreased. For example, in a study of a
representative sample of US high schools, participation rates
fell from 79% in ninth grade to 36% in 12th grade.157 In
addition, participation in school-sponsored after-school teams
is frequently limited to elite athletes, limiting the opportunity
for high school–aged students to engage in regular physical
activity.
Media has a pervasive influence on children’s food
choices. Children, including the very young, are heavily
marketed by the food industry. Time spent watching television is directly related to children’s food requests. The most
frequently advertised foods are high-sugar breakfast cereals,
fast food restaurant products, sweetened beverages, frozen
dinners, cookies, and candy. Fruits and vegetables are almost
never advertised. Watching television during meals is associated with increased frequency of poor food choices and
decreased frequency of good choices.77,158 –160 Several European countries now have restrictions on advertising to children as well as school-based marketing.
State and local governments are now becoming active in
the effort to control obesity on a wide variety of fronts. For
example, several states have adopted legislation mandating
school staff report to parents the BMI status of their children.
Changes in food labeling, taxes on certain types of foods,
restrictions on foods provided to children in governmentsponsored programs, and requiring restaurants to provide
nutrition information are examples of regulations under
consideration.11,161 Strategy types are summarized in Table 9.
Given the widening discrepancy between recommended dietary guidelines and current dietary intake, a reevaluation of
federal agricultural policies may be warranted. Strategies for
food subsidies and taxation should reflect health goals. Foods
made available and served through public nutrition programs
must be consistent with current recommendations.
Therapeutic Lifestyle Changes for Treatment
of Hypertension and Hypercholesterolemia
There are currently established consensus guidelines for the
role of diet in the management of children with established
cardiovascular risk factors. Cut points for diagnosing dyslipidemia and hypertension are provided in Table 10.7,39,162,163
The Fourth Pediatric Report of the National High Blood
Pressure Education Program recommends a diet consistent
with the current recommendations for children with hypertension.162 For overweight children, weight loss is the initial
therapeutic strategy. The Dietary Approaches to Stop Hypertension (DASH) study has recently shown that implementation of a diet rich in fruits, vegetables, nonfat dairy products,
2070
Circulation
September 27, 2005
TABLE 10. Consensus Guidelines for Diagnosis of
Hypertension and Dyslipidemia in Children
Hypertension
Guideline
Prehypertension
Systolic or diastolic blood pressure ⬎90th percentile
for age and gender or 120/80 mm Hg, whichever is
less
Stage 1
hypertension
for age and gender on 3 consecutive visits or
140/90 mm Hg, whichever is less
Stage 2
hypertension
⫹ 5 mm Hg for age and gender or 160/110 mm Hg,
whichever is less
Total cholesterol
Borderline
ⱖ170 mg/dL
Abnormal
ⱖ200 mg/dL
LDL cholesterol
Borderline
ⱖ100 mg/dL
Abnormal
ⱖ130 mg/dL
HDL cholesterol
Abnormal
⬍40 mg/dL
Triglycerides
Abnormal
ⱖ200 mg/dL
and whole grains can effectively lower blood pressure in
adults with hypertension.164 Although there are no comparable clinical trial data in children, there is no reason to suspect
that the DASH diet would not be safe to implement in older
children and adolescents as long as protein and calorie needs
are met.165,166
There has not been an update of the Report of the Expert
Panel on Blood Cholesterol Levels in Children and Adolescents published since its publication in 1992, but the National
Cholesterol Education Program (NCEP) generally recommends restriction of saturated fat intake to ⬍7% of total
calories and restriction of cholesterol intake to ⬍200 mg/d for
treatment of elevated LDL cholesterol levels.39,163 There are
now data from randomized trials demonstrating that such
diets are safe in children as young as 7 months of age.40,42,44,45
Efficacy is variable, however, and unless the diet is extremely
high in saturated fat before changes are made, it is unlikely
that diet alone will be sufficient to achieve target levels for
LDL cholesterol in those with genetic dyslipidemias and LDL
cholesterol ⱖ190 mg/dL. Increased intake of soluble fiber is
recommended as an adjunct to the reduced intakes of saturated fatty acids and cholesterol.14,167 Recently plant sterols
and stanols have been used, often in margarines, to lower
LDL cholesterol through inhibition of cholesterol absorption.
Adult studies have shown reductions of 4% to 11% without
adverse events.168 One randomized controlled trial in children
showed that 20 g/d of plant sterol– containing margarine
lowered LDL cholesterol 8%.169 These products may be used,
although caution is recommended with regard to the potential
for decreased absorption of fat-soluble vitamins and betacarotene. Formal recommendation of their use for children
awaits clinical trial data.42,170 –177
Summary
This scientific statement updates nutrition recommendations
for the promotion of cardiovascular health among children.
Recommendations have been made with regard to diet composition, total caloric intake, and physical activity. Implementation requires that children and all other members of their
households actively make the recommended changes. Adverse recent trends in children’s diets have been noted.
Cardiovascular nutrition issues surrounding the first 2 years
of life have been addressed. Strategies to improve implementation of the recommended diet have been presented. A brief
overview of public health issues related to nutrition is
included.
Acknowledgments
The authors acknowledge the contributions of many additional
experts who reviewed portions of the material presented or supplied
additional expertise. These include Julie Mennella, PhD, Gary A.
Emmett, MD, and Penny Kris-Etherton, PhD. Carol Muscar provided
invaluable support in the preparation of the manuscript and its
editing. The AHA thanks the Nemours Health and Prevention
Services group for providing meeting space and organizational
support to the investigators and AHA staff, facilitating the rapid
completion of this report.
Gidding et al
2071
Writing Group Disclosures
Writing Group
Member
Employment
Research Grant
Other Research
Support
Speakers
Bureau/Honoraria
Ownership
Interest
Consultant/Advisory Board
Other
Samuel
Gidding
Barbara
Dennison
Nemours
Foundation
New York
State
Department of
Health
None
None
None
None
None
None
None
None
None
Bassett Healthcare–Research Scientist; Mead Johnson
Nutritionals–Consultant Toddler/Children Panel; American Dairy
Association–Consultant; University of North Carolina–Consultant
None
Stephen
Daniels
Children’s
Hospital
Cincinnati
Northwestern
University–
Feinberg
School of
Medicine
University of
Minnesota
Tufts
University
Pennsylvania
State
University
National Institute of
Diabetes and
Digestive and Kidney
Diseases—grantee
USDA–Improving
Human Nutrition—
grantee
None
None
None
None
None
None
None
None
None
None
Journal of the American Dietetic Association–Editor in Chief
None
None
None
None
None
American Phytotherapy Research Lab–Consultant
None
None
None
None
None
None
None
National Institute of
Child Health & Human
Development–
Grantee; Dairy
Management
Inc–Grantee;
USDA-CSREES–
Grantee
(coinvestigator)
None
None
None
None
Institute of Medicine Committee on Prevention of Childhood
Obesity in Children and Youth–Chair
None
None
None
None
None
None
None
None
None
European Society for Pediatric Research–Member; International
Epidemiological Association–Member; World Heart
Federation–Member; Swiss Pediatric Society–Member; Swiss
Medical Society–Member; American Society for Nutritional
Sciences–Member; American Society for Clinical
Nutrition–Member; North American Association for the Study of
Obesity–Member; Children Are Our Messengers: Changing the
Health Message, International Society on Hypertension in
Blacks–Advisory Committee Member; Community Health Centers,
Child Health Project, US Department of Health and Human
Services–Consultant; Early Origins of Adult Health Working Group,
National Children’s Study–Core Member; Society for Pediatric
Research–Member; Comprehensive School Nutrition Policy Task
Force, US Department of Agriculture/Food–Consultant; National
High Blood Pressure Education Program on Blood Pressure in
Children and Adolescents; Institute of Medicine Committee on
Nutrient Relationships in Seafood
None
None
None
None
None
None
None
None
Linda Van
Horn
Julia
Steinberger
Alice
Lichtenstein
Leann Birch
Nicolas
Stettler
Matthew
Gillman
Karyl Rattay
University of
Pennsylvania
School of
Medicine
Harvard
University
Nemours
Health and
Prevention
Services
CSREES indicates Cooperative State Research, Education, and Extension Service. This table represents the relationships of writing group members that may be perceived as actual
or reasonably perceived conflicts of interest as reported on the Disclosure Questionnaire that all members of the writing group are required to complete and submit. A relationship is
considered “significant” if (1) the person receives ⱖ$10 000 during any 12-month period or ⱖ5% of the person’s gross income or (2) the person owns ⱖ5% of the voting stock or
share of the entity or owns ⱖ$10 000 of the fair market value of the entity. A relationship is considered “modest” if it is less than “significant” under the preceding definition.
*Modest.
†Significant.
Reviewer Disclosures
Reviewer
Frank Greer
Nancy F. Krebs
Kristie Lancaster
William Neal
Theresa A. Nicklas
Employment
Research Grant
Other Research Support
Speakers
Bureau/Honoraria
Ownership
Interest
Consultant/Advisory Board
Other
University of
Wisconsin–Madison
The Children’s Hospital
New York University
West Virginia University
Baylor College of Medicine
None
None
None
None
None
None
None
None
None
USDA; National
Institutes of Health
None
None
None
Dairy Management Inc;
National
Cattlemen’s Beef
Association;
Mars, Inc
None
None
None
National Dairy Council
Speakers Bureau;
National Cattlemen’s Beef
Association Speakers
Bureau
None
None
None
None
None
None
None
Brands Global Advisory
Council on Health,
Nutrition and Fitness; US
Potato Board’s Scientific
Advisory Panel; Cadbury
Scientific Advisory
Committee; Grain Foods
Foundation Scientific
Advisory Board
None
None
None
International
Food
Information
Council
Expert
Resource to
Media; 2005
Dietary
Guidelines
Advisory
Committee
This table represents the relationships of reviewers that may be perceived as actual or reasonably perceived conflicts of interest as reported on the Disclosure
Questionnaire that all reviewers are required to complete and submit.
2072
Circulation
September 27, 2005
References
1. Kavey RE, Daniels SR, Lauer RM, Atkins DL, Hayman LL, Taubert K;
American Heart Association. American Heart Association guidelines for
primary prevention of atherosclerotic cardiovascular disease beginning
in childhood. Circulation. 2003;107:1562–1566.
2. Stamler J, Stamler R, Neaton JD, Wentworth D, Daviglus ML, Garside D,
Dyer AR, Liu K, Greenland P. Low risk-factor profile and long-term
cardiovascular and noncardiovascular mortality and life expectancy:
findings for 5 large cohorts of young adult and middle-aged men and
women. JAMA. 1999;282:2012–2018.
3. Daviglus ML, Stamler J, Pirzada A, Yan LL, Garside DB, Liu K, Wang R,
Dyer AR, Lloyd-Jones DM, Greenland P. Favorable cardiovascular risk
profile in young women and long-term risk of cardiovascular and all-cause
mortality. JAMA. 2004;292:1588–1592.
4. Relationship of atherosclerosis in young men to serum lipoprotein cholesterol concentrations and smoking. A preliminary report from the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) Research Group.
JAMA. 1990;264:3018–3024.
5. Berenson GS, Srinivasan SR, Bao W, Newman WP 3rd, Tracy RE,
Wattigney WA. Association between multiple cardiovascular risk factors
and atherosclerosis in children and young adults. The Bogalusa Heart Study.
N Engl J Med. 1998;338:1650–1656.
6. Mahoney LT, Burns TL, Stanford W, Thompson BH, Witt JD, Rost CA,
Lauer RM. Coronary risk factors measured in childhood and young adult
life are associated with coronary artery calcification in young adults: the
Muscatine Study. J Am Coll Cardiol. 1996;27:277–284.
7. Williams CL, Hayman LL, Daniels SR, Robinson TN, Steinberger J,
Paridon S, Bazzarre T. Cardiovascular health in childhood: A statement for
health professionals from the Committee on Atherosclerosis, Hypertension,
and Obesity in the Young (AHOY) of the Council on Cardiovascular
Disease in the Young, American Heart Association. Circulation. 2002;106:
143–160.
8. Krauss RM, Eckel RH, Howard B, Appel LJ, Daniels SR, Deckelbaum RJ,
Erdman JW Jr, Kris-Etherton P, Goldberg IJ, Kotchen TA, Lichtenstein AH,
Mitch WE, Mullis R, Robinson K, Wylie-Rosett J, St Jeor S, Suttie J,
Tribble DL, Bazzarre TL. AHA Dietary Guidelines: revision 2000: A
statement for healthcare professionals from the Nutrition Committee of the
American Heart Association. Circulation. 2000;102:2284–2299.
9. Grundy SM, Bilheimer D, Blackburn H, Brown WV, Kwiterovich PO Jr,
Mattson F, Schonfeld G, Weidman WH. Rationale of the diet-heart
statement of the American Heart Association. Report of Nutrition Committee. Circulation. 1982;65:839A–854A.
10. Gidding SS, Leibel RL, Daniels S, Rosenbaum M, Van Horn L, Marx GR.
Understanding obesity in youth: A statement for healthcare professionals
from the Committee on Atherosclerosis and Hypertension in the Young of
the Council on Cardiovascular Disease in the Young and the Nutrition
Committee, American Heart Association. Writing Group. Circulation.
1996;94:3383–3387.
11. Institute of Medicine, ed. Preventing Childhood Obesity: Health in the
Balance. Washington, DC: National Academy Press; 2005.
12. Olshansky SJ, Passaro DJ, Hershow RC, Layden J, Carnes BA, Brody J,
Hayflick L, Butler RN, Allison DB, Ludwig DS. A potential decline in life
expectancy in the United States in the 21st century. N Engl J Med. 2005;
352:1138–1145.
13. Fisher EA, Van Horn L, McGill HC Jr. Nutrition and children: A statement
for healthcare professionals from the Nutrition Committee, American Heart
Association. Circulation. 1997;95:2332–2333.
14. US Department of Health and Human Services, US Department of Agriculture. Dietary Guidelines for Americans. 6th ed. Washington, DC: US
Government Printing Office; 2005.
15. US Department of Agriculture. MyPyramid. 2005. Available at: www.
mypyramid.gov. Accessed June 22, 2005.
16. Nicklas T, Johnson R. Position of the American Dietetic Association:
Dietary guidance for healthy children ages 2 to 11 years. J Am Diet Assoc.
2004;104:660–677. Erratum in: J Am Diet Assoc. 2004;104:1075.
17. Kleinman R, ed. Pediatric Nutrition Handbook. 5th ed. Elk Grove, Ill:
American Academy of Pediatrics; 2004.
18. 2005 Dietary Guideline Advisory Committee. Nutrition and your health:
dietary guidelines for Americans. Available at: http://www.health.gov/
dietaryguidelines/dga2005/report/HTML/E_translation.htm. Accessed
January 17, 2005.
19. Institute of Medicine. Dietary Reference Intakes for Energy, Carbohydrates,
Fiber, Fat, Protein, and Amino Acids (Macronutrients). Washington, DC:
The National Academies Press; 2002. Available at: http://www.
nap.edu/books/0309085373/html/. Accessed January 31, 2005.
20. Peterson S, Sigman-Grant M. Impact of adopting lower-fat food choices on
nutrient intake of American children. Pediatrics. 1997;100:E4.
21. He K, Song Y, Daviglus ML, Liu K, Van Horn L, Dyer AR, Greenland P.
Accumulated evidence on fish consumption and coronary heart disease
mortality: a meta-analysis of cohort studies. Circulation. 2004;109:
2705–2711.
22. US Food and Drug Administration, US Environmental Protection Agency.
What you need to know about mercury in fish and shellfish. In: 2004.
23. Kris-Etherton PM, Harris WS, Appel LJ; American Heart Association
Nutrition Committee. Fish consumption, fish oil, omega-3 fatty acids, and
cardiovascular disease. Circulation. 2002;106:2747–2757.
24. Mozaffarian D, Lemaitre RN, Kuller LH, Burke GL, Tracy RP, Siscovick
DS; Cardiovascular Health Study. Cardiac benefits of fish consumption may
depend on the type of fish meal consumed: the Cardiovascular Health Study.
Circulation. 2003;107:1372–1377.
25. Van Horn L, Obarzanek E, Friedman LA, Gernhofer N, Barton B. Children’s adaptations to a fat-reduced diet: The Dietary Intervention Study in
Children (DISC). Pediatrics. 2005;115:1723–1733.
25a.Strong WB, Malina RM, Blimkie CJ, Daniels SR, Dishman RK, Gutin
B, Hergenroeder AC, Must A, Nixon PA, Pivarnik JM, Rowland T,
Trost S, Trudeau F. Evidence based physical activity for school-age
youth. J Pediatr. 2005;146:732–737.
26. McGill HC Jr, McMahan CA, Zieske AW, Malcom GT, Tracy RE, Strong
JP. Effects of nonlipid risk factors on atherosclerosis in youth with a
favorable lipoprotein profile. Circulation. 2001;103:1546–1550.
27. Raitakari OT, Juonala M, Kahonen M, Taittonen L, Laitinen T, MakiTorkko N, Jarvisalo MJ, Uhari M, Jokinen E, Ronnemaa T, Akerblom HK,
Viikari JS. Cardiovascular risk factors in childhood and carotid artery
intima-media thickness in adulthood: the Cardiovascular Risk in Young
Finns Study. JAMA. 2003;290:2277–2283.
28. Li S, Chen W, Srinivasan SR, Bond MG, Tang R, Urbina EM, Berenson
GS. Childhood cardiovascular risk factors and carotid vascular changes in
adulthood: the Bogalusa Heart Study. JAMA. 2003;290:2271–2276.
29. Lauer RM, Lee J, Clarke WR. Factors affecting the relationship between
childhood and adult cholesterol levels: the Muscatine Study. Pediatrics.
1988;82:309–318.
30. Lauer RM, Clarke WR. Childhood risk factors for high adult blood pressure:
the Muscatine Study. Pediatrics. 1989;84:633–641.
31. Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH. Predicting
obesity in young adulthood from childhood and parental obesity. N Engl
J Med. 1997;337:869–873.
32. Janz KF, Dawson JD, Mahoney LT. Tracking physical fitness and physical
activity from childhood to adolescence: the Muscatine study. Med Sci Sports
Exerc. 2000;32:1250–1257.
33. Feunekes GI, de Graaf C, Meyboom S, van Staveren WA. Food choice and
fat intake of adolescents and adults: associations of intakes within social
networks. Prev Med. 1998;27:645–656.
34. Rojas NL, Killen JD, Haydel KF, Robinson TN. Nicotine dependence
among adolescent smokers. Arch Pediatr Adolesc Med. 1998;152:151–156.
35. Obarzanek E, Kimm SY, Barton BA, Van Horn LL, Kwiterovich PO Jr,
Simons-Morton DG, Hunsberger SA, Lasser NL, Robson AM, Franklin FA
Jr, Lauer RM, Stevens VJ, Friedman LA, Dorgan JF, Greenlick MR; DISC
Collaborative Research Group. Long-term safety and efficacy of a cholesterol-lowering diet in children with elevated low-density lipoprotein cholesterol: seven-year results of the Dietary Intervention Study in Children
(DISC). Pediatrics. 2001;107:256–264.
36. Luepker RV, Perry CL, McKinlay SM, Nader PR, Parcel GS, Stone EJ,
Webber LS, Elder JP, Feldman HA, Johnson CC, et al. Outcomes of a field
trial to improve children’s dietary patterns and physical activity. The Child
and Adolescent Trial for Cardiovascular Health. CATCH collaborative
group. JAMA. 1996;275:768–776.
37. Kaitosaari T, Ronnemaa T, Raitakari O, Talvia S, Kallio K, Volanen I,
Leino A, Jokinen E, Valimaki I, Viikari J, Simell O. Effect of 7-year
infancy-onset dietary intervention on serum lipoproteins and lipoprotein
subclasses in healthy children in the prospective, randomized Special Turku
Coronary Risk Factor Intervention Project for Children (STRIP) study.
Circulation. 2003;108:672–677.
38. Yu-Poth S, Zhao G, Etherton T, Naglak M, Jonnalagadda S, Kris-Etherton
PM. Effects of the National Cholesterol Education Program’s Step I and
Step II dietary intervention programs on cardiovascular disease risk factors:
a meta-analysis. Am J Clin Nutr. 1999;69:632–646.
39. Third Report of the National Cholesterol Education Program (NCEP)
Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) Final Report. Circulation.
2002;106:3143–3421.
Gidding et al
40. Lagstrom H, Jokinen E, Seppanen R, Ronnemaa T, Viikari J, Valimaki I,
Venetoklis J, Myyrinmaa A, Niinikoski H, Lapinleimu H, Simell O.
Nutrient intakes by young children in a prospective randomized trial of a
low-saturated fat, low-cholesterol diet. The STRIP Baby Project. Special
Turku Coronary Risk Factor Intervention Project for Babies. Arch Pediatr
Adolesc Med. 1997;151:181–188.
41. Viikari JS, Niinikoski H, Juonala M, Raitakari OT, Lagstrom H, Kaitosaari
T, Ronnemaa T, Simell O. Risk factors for coronary heart disease in
children and young adults. Acta Paediatr Suppl. 2004;93:34–42.
42. Talvia S, Lagstrom H, Rasanen M, Salminen M, Rasanen L, Salo P, Viikari
J, Ronnemaa T, Jokinen E, Vahlberg T, Simell O. A randomized intervention since infancy to reduce intake of saturated fat: calorie (energy) and
nutrient intakes up to the age of 10 years in the Special Turku Coronary Risk
Factor Intervention Project. Arch Pediatr Adolesc Med. 2004;158:41–47.
43. Obarzanek E, Hunsberger SA, Van Horn L, Hartmuller VV, Barton BA,
Stevens VJ, Kwiterovich PO, Franklin FA, Kimm SY, Lasser NL, SimonsMorton DG, Lauer RM. Safety of a fat-reduced diet: the Dietary Intervention Study in Children (DISC). Pediatrics. 1997;100:51–59.
44. Lauer RM, Obarzanek E, Hunsberger SA, Van Horn L, Hartmuller VW,
Barton BA, Stevens VJ, Kwiterovich PO Jr, Franklin FA Jr, Kimm SY,
Lasser NL, Simons-Morton DG. Efficacy and safety of lowering dietary
intake of total fat, saturated fat, and cholesterol in children with elevated
LDL cholesterol: the Dietary Intervention Study in Children. Am J Clin
Nutr. 2000;72:1332S–1342S.
45. Rask-Nissila L, Jokinen E, Terho P, Tammi A, Hakanen M, Ronnemaa T,
Viikari J, Seppanen R, Valimaki I, Helenius H, Simell O. Effects of diet on
the neurologic development of children at 5 years of age: the STRIP project.
J Pediatr. 2002;140:328–333.
46. Jacobson MS, Tomopoulos S, Williams CL, Arden MR, Deckelbaum RJ,
Starc TJ. Normal growth in high-risk hyperlipidemic children and adolescents with dietary intervention. Prev Med. 1998;27:775–780.
47. Cook S, Weitzman M, Auinger P, Nguyen M, Dietz WH. Prevalence of a
metabolic syndrome phenotype in adolescents: findings from the third
National Health and Nutrition Examination Survey, 1988 –1994. Arch
Pediatr Adolesc Med. 2003;157:821–827.
48. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of
overweight to cardiovascular risk factors among children and adolescents:
the Bogalusa Heart Study. Pediatrics. 1999;103:1175–1182.
49. Morrison JA, Sprecher DL, Barton BA, Waclawiw MA, Daniels SR. Overweight, fat patterning, and cardiovascular disease risk factors in black and
white girls: The National Heart, Lung, and Blood Institute Growth and
Health Study. J Pediatr. 1999;135:458–464.
50. Kimm SY, Glynn NW, Kriska AM, Barton BA, Kronsberg SS, Daniels SR,
Crawford PB, Sabry ZI, Liu K. Decline in physical activity in black girls and
white girls during adolescence. N Engl J Med. 2002;347:709–715.
51. Becque MD, Katch VL, Rocchini AP, Marks CR, Moorehead C. Coronary
risk incidence of obese adolescents: reduction by exercise plus diet intervention. Pediatrics. 1988;81:605–612.
52. Rocchini AP, Katch V, Schork A, Kelch RP. Insulin and blood pressure
during weight loss in obese adolescents. Hypertension. 1987;10:267–273.
53. Young EM, Fors SW, Hayes DM. Associations between perceived parent
behaviors and middle school student fruit and vegetable consumption. J
Nutr Educ Behav. 2004;36:2–8.
54. Muntner P, He J, Cutler JA, Wildman RP, Whelton PK. Trends in blood
pressure among children and adolescents. JAMA. 2004;291:2107–2113.
55. Ford ES, Mokdad AH, Ajani UA. Trends in risk factors for cardiovascular
disease among children and adolescents in the United States. Pediatrics.
2004;114:1534–1544.
56. Gidding SS, Bao W, Srinivasan SR, Berenson GS. Effects of secular trends
in obesity on coronary risk factors in children: The Bogalusa Heart Study.
J Pediatr. 1995;127:868–874.
57. Sinaiko AR, Donahue RP, Jacobs DR Jr, Prineas RJ. Relation of weight and
rate of increase in weight during childhood and adolescence to body size,
blood pressure, fasting insulin, and lipids in young adults. The Minneapolis
Children’s Blood Pressure Study. Circulation. 1999;99:1471–1476.
58. Kumanyika SK, Obarzanek E, Robinson TN, Beech BM. Phase 1 of the
Girls health Enrichment Multi-site Studies (GEMS): conclusion. Ethn Dis.
2003;13(suppl 1):S88–S91.
59. Caballero B, Clay T, Davis SM, Ethelbah B, Rock BH, Lohman T, Norman
J, Story M, Stone EJ, Stephenson L, Stevens J; Pathways Study Research
Group. Pathways: a school-based, randomized controlled trial for the prevention of obesity in American Indian schoolchildren. Am J Clin Nutr.
2003;78:1030–1038.
2073
60. Briefel RR, Reidy K, Karwe V, Devaney B. Feeding infants and toddlers
study: Improvements needed in meeting infant feeding recommendations.
J Am Diet Assoc. 2004;104:s31–s37.
61. Fox MK, Pac S, Devaney B, Jankowski L. Feeding infants and toddlers
study: What foods are infants and toddlers eating? J Am Diet Assoc.
2004;104(suppl 1):s22–s30.
62. French SA, Story M, Jeffery RW. Environmental influences on eating and
physical activity. Annu Rev Public Health. 2001;22:309–335.
63. Siega-Riz AM, Popkin BM, Carson T. Trends in breakfast consumption for
children in the United States from 1965–1991. Am J Clin Nutr. 1998;67:
748S–756S.
64. Nielsen SJ, Siega-Riz AM, Popkin BM. Trends in energy intake in U.S.
between 1977 and 1996: similar shifts seen across age groups. Obes Res.
2002;10:370–378.
65. Nielsen SJ, Popkin BM. Patterns and trends in food portion sizes,
1977–1998. JAMA. 2003;289:450–453.
66. Nielsen SJ, Popkin BM. Changes in beverage intake between 1977 and
2001. Am J Prev Med. 2004;27:205–210. Erratum in: Am J Prev Med.
2005;28:413.
67. Cavadini C, Siega-Riz AM, Popkin BM. US adolescent food intake trends
from 1965 to 1996. Arch Dis Child. 2000;83:18–24. Erratum in: Arch Dis
Child. 2002;87:85.
68. Kranz S, Smiciklas-Wright H, Siega-Riz AM, Mitchell D. Adverse effect of
high added sugar consumption on dietary intake in American preschoolers.
J Pediatr. 2005;146:105–111.
69. Wright JD, Wang CY, Kennedy-Stephenson J, Ervin RB. Dietary intake of
ten key nutrients for public health, United States: 1999–2000. Adv Data.
April 2003:1–4.
70. Institute of Medicine Panel on Dietary Reference Intakes for Electrolytes
and Water. Dietary Reference Intakes for Water, Potassium, Sodium,
Chloride, and Sulfate. Washington, DC: National Academies Press; 2004.
71. Ervin RB, Wang CY, Wright JD, Kennedy-Stephenson J. Dietary intake of
selected minerals for the United States population: 1999–2000. Adv Data.
April 2004:1–5.
72. Rozin P. Family resemblance in food and other domains: the family paradox
and the role of parental congruence. Appetite. 1991;16:93–102.
73. Rozin P, Millman L. Family environment, not heredity, accounts for family
resemblances in food preferences and attitudes: a twin study. Appetite.
1987;8:125–134.
74. Cutler D, Glasser E, Shapiro J. Why Have Americans Become More Obese?
The Economics of Obesity Workshop (April 22, 2003). Washington, DC:
Economic Research Service, US Department of Agriculture; May 2004
(E-FAN-04-004).
75. Brown R, Ogden J. Children’s eating attitudes and behaviour: a study of the
modelling and control theories of parental influence. Health Educ Res.
2004;19:261–271.
76. Videon TM, Manning CK. Influences on adolescent eating patterns: the
importance of family meals. J Adolesc Health. 2003;32:365–373.
77. Gillman MW, Rifas-Shiman SL, Frazier AL, Rockett HR, Camargo CA Jr,
Field AE, Berkey CS, Colditz GA. Family dinner and diet quality among
older children and adolescents. Arch Fam Med. 2000;9:235–240.
78. Neumark-Sztainer D, Hannan PJ, Story M, Croll J, Perry C. Family meal
patterns: associations with sociodemographic characteristics and improved
dietary intake among adolescents. J Am Diet Assoc. 2003;103:317–322.
79. Kratt P, Reynolds K, Shewchuk R. The role of availability as a moderator
of family fruit and vegetable consumption. Health Educ Behav. 2000;27:
471–482.
80. Fisher JO, Mitchell DC, Smiciklas-Wright H, Birch LL. Parental influences
on young girls’ fruit and vegetable, micronutrient, and fat intakes. J Am Diet
Assoc. 2002;102:58–64.
81. Fisher J, Mitchell D, Smiciklas-Wright H, Birch L. Maternal milk consumption predicts the tradeoff between milk and soft drinks in young girls’
diets. J Nutr. 2001;131:246–250.
82. World Health Organization. World Health Report 2002: Reducing Risks,
Promoting Healthy Life. Geneva, Switzerland: World Health Organization;
2002. Available at: http://www.who.int/whr/2002/en/whr02_en.pdf.
Accessed April 12, 2005.
83. Rich-Edwards JW, Stampfer MJ, Manson JE, Rosner B, Hankinson SE,
Colditz GA, Willett WC, Hennekens CH. Birth weight and risk of cardiovascular disease in a cohort of women followed up since 1976. BMJ.
1997;315:396–400.
84. Curhan GC, Chertow GM, Willett WC, Spiegelman D, Colditz GA,
Manson JE, Speizer FE, Stampfer MJ. Birth weight and adult hypertension
and obesity in women. Circulation. 1996;94:1310–1315.
2074
Circulation
September 27, 2005
85. Langley-Evans SC. Fetal programming of cardiovascular function through
exposure to maternal undernutrition. Proc Nutr Soc. 2001;60:505–513.
86. Seckl JR. Prenatal glucocorticoids and long-term programming. Eur J
Endocrinol. 2004;151(suppl 3):U49–U62.
87. Edwards LJ, Coulter CL, Symonds ME, McMillen IC. Prenatal undernutrition, glucocorticoids and the programming of adult hypertension. Clin
Exp Pharmacol Physiol. 2001;28:938–941.
88. Eriksson J, Forsen T, Tuomilehto J, Osmond C, Barker D. Size at birth,
childhood growth and obesity in adult life. Int J Obes Relat Metab Disord.
2001;25:735–740.
89. Hediger ML, Overpeck MD, McGlynn A, Kuczmarski RJ, Maurer KR,
Davis WW. Growth and fatness at three to six years of age of children born
small- or large-for-gestational age. Pediatrics. 1999;104:e33.
90. Frisancho AR. Prenatal compared with parental origins of adolescent
fatness. Am J Clin Nutr. 2000;72:1186–1190.
91. Whitaker RC. Predicting preschooler obesity at birth: the role of maternal
obesity in early pregnancy. Pediatrics. 2004;114:e29–e36.
92. Institute of Medicine. Nutrition During Pregnancy, Weight Gain and
Nutrient Supplements: Report of the Subcommittee on Nutritional Status
and Weight Gain During Pregnancy. Washington, DC: National Academy
Press; 1990.
93. Gunderson EP, Abrams B, Selvin S. The relative importance of gestational
gain and maternal characteristics associated with the risk of becoming
overweight after pregnancy. Int J Obes Relat Metab Disord. 2000;24:
1660–1668.
94. Huh SY, Rifas-Shiman SL, Kleinman KP, Rich-Edwards JW, Lipshultz SE,
Gillman MW. Maternal protein intake is not associated with infant blood
pressure. Int J Epidemiol. 2005;34:378–384.
95. Gillman MW, Rifas-Shiman SL, Kleinman KP, Rich-Edwards JW,
Lipshultz SE. Maternal calcium intake and offspring blood pressure. Circulation. 2004;110:1990–1995.
96. Gartner LM, Morton J, Lawrence RA, Naylor AJ, O’Hare D, Schanler RJ,
Eidelman AI; American Academy of Pediatrics Section on Breastfeeding.
Breastfeeding and the use of human milk. Pediatrics. 2005;115:496–506.
97. Martin RM, Davey Smith G, Mangtani P, Tilling K, Frankel S, Gunnell D.
Breastfeeding and cardiovascular mortality: the Boyd Orr cohort and a
systematic review with meta-analysis. Eur Heart J. 2004;25:778–786.
98. Arenz S, Ruckerl R, Koletzko B, von Kries R. Breast-feeding and childhood
obesity–a systematic review. Int J Obes Relat Metab Disord. 2004;28:
1247–1256.
99. Owen CG, Martin RM, Whincup PH, Smith GD, Cook DG. Effect of infant
feeding on the risk of obesity across the life course: a quantitative review of
published evidence. Pediatrics. 2005;115:1367–1377.
100. Martin RM, Gunnell D, Smith GD. Breastfeeding in infancy and blood
pressure in later life: systematic review and meta-analysis. Am J Epidemiol.
2005;161:15–26.
101. Owen CG, Whincup PH, Gilg JA, Cook DG. Effect of breast feeding in
infancy on blood pressure in later life: systematic review and meta-analysis.
BMJ. 2003;327:1189–1195.
102. Owen CG, Whincup PH, Odoki K, Gilg JA, Cook DG. Infant feeding and
blood cholesterol: a study in adolescents and a systematic review. Pediatrics. 2002;110:597–608.
103. Stettler N, Zemel BS, Kumanyika S, Stallings VA. Infant weight gain and
childhood overweight status in a multicenter, cohort study. Pediatrics.
2002;109:194–199.
104. Dewey KG. Nutrition, growth, and complementary feeding of the breastfed
infant. Pediatr Clin North Am. 2001;48:87–104.
105. Haisma H, Coward WA, Albernaz E, Visser GH, Wells JC, Wright A,
Victora CG. Breast milk and energy intake in exclusively, predominantly,
and partially breast-fed infants. Eur J Clin Nutr. 2003;57:1633–1642.
106. Dennison BA, Edmunds LE, Stratton H, Pruzek R. Rapid infant weight gain
associated with childhood overweight at age 4 years. Obes Res. In press.
107. Wright P. Hunger, satiety and feeding behavior in early infancy. In: Boakes
RA, Popplewell DA, Burton MJ, eds. Eating Habits: Food, Physiology and
Learned Behavior. Chichester, UK: Wiley; 1987.
108. Fisher JO, Birch LL, Smiciklas-Wright H, Picciano MF. Breast-feeding
through the first year predicts maternal control in feeding and subsequent
toddler energy intakes. J Am Diet Assoc. 2000;100:641–646.
109. Taveras EM, Scanlon KS, Birch L, Rifas-Shiman SL, Rich-Edwards JW,
Gillman MW. Association of breastfeeding with maternal control of infant
feeding at age 1 year. Pediatrics. 2004;114:e577–e583.
110. Wright P. Patterns of milk intake in breast- and bottle-fed infants. In:
Stratton P, ed. The Psychobiology of the Human Newborn. New York, NY:
Wiley; 1982.
111. Klesges RC, Coates TJ, Brown G, Sturgeon-Tillisch J, MoldenhauerKlesges LM, Holzer B, Woolfrey J, Vollmer J. Parental influences on
children’s eating behavior and relative weight. J Appl Behav Anal. 1983;
16:371–378.
112. Mennella JA, Griffin CE, Beauchamp GK. Flavor programming during
infancy. Pediatrics. 2004;113:840–845.
113. Mennella JA, Jagnow CP, Beauchamp GK. Prenatal and postnatal flavor
learning by human infants. Pediatrics. 2001;107:E88.
114. Mennella JA, Pepino MY, Reed DR. Genetic and environmental determinants of bitter perception and sweet preferences. Pediatrics. 2005;115:
e216–e222.
115. Gerrish CJ, Mennella JA. Flavor variety enhances food acceptance in
formula-fed infants. Am J Clin Nutr. 2001;73:1080–1085.
116. Liem DG, Mennella JA. Heightened sour preferences during childhood.
Chem Senses. 2003;28:173–180.
117. Li R, Hsia J, Fridinger F, Hussain A, Benton-Davis S, Grummer-Strawn L.
Public beliefs about breastfeeding policies in various settings. J Am Diet
Assoc. 2004;104:1162–1168.
118. Visness CM, Kennedy KI. Maternal employment and breast-feeding:
findings from the 1988 National Maternal and Infant Health Survey. Am J
Public Health. 1997;87:945–950.
119. Li R, Zhao Z, Mokdad A, Barker L, Grummer-Strawn L. Prevalence of
breastfeeding in the United States: the 2001 National Immunization Survey.
Pediatrics. 2003;111:1198–1201.
120. Kurinij N, Shiono PH, Ezrine SF, Rhoads GG. Does maternal employment
affect breast-feeding? Am J Public Health. 1989;79:1247–1250.
121. Gielen AC, Faden RR, O’Campo P, Brown CH, Paige DM. Maternal
employment during the early postpartum period: effects on initiation and
continuation of breast-feeding. Pediatrics. 1991;87:298–305.
122. Auerbach KG, Guss E. Maternal employment and breastfeeding. A study of
567 women’s experiences. Am J Dis Child. 1984;138:958–960.
123. Hills-Bonczyk SG, Avery MD, Savik K, Potter S, Duckett LJ. Women’s
experiences with combining breast-feeding and employment. J Nurse Midwifery. 1993;38:257–266.
124. Lande B, Andersen LF, Veierod MB, Baerug A, Johansson L, Trygg KU,
Bjorneboe GE. Breast-feeding at 12 months of age and dietary habits among
breast-fed and non-breast-fed infants. Public Health Nutr. 2004;7:495–503.
125. Ortiz J, McGilligan K, Kelly P. Duration of breast milk expression among
working mothers enrolled in an employer-sponsored lactation program.
Pediatr Nurs. 2004;30:111–119.
126. Kramer MS, Kakuma R. The optimal duration of exclusive breastfeeding: a
systematic review. Adv Exp Med Biol. 2004;554:63–77.
126a.Lederman SA, Akabas SR, Moore BJ, Bentley ME, Devaney B, Gillman
MW, Kramer MS, Mennella JA, Ness A, Wardle J. Summary of the
Presentations at the Conference on Preventing Childhood Obesity,
December 8, 2003. Pediatrics. 2004;114:1146 –1173.
127. Carruth BR, Skinner JD, Houck KS, Moran JD 3rd. Addition of supplementary foods and infant growth (2 to 24 months). J Am Coll Nutr. 2000;
19:405–412.
128. Mehta KC, Specker BL, Bartholmey S, Giddens J, Ho ML. Trial on timing
of introduction to solids and food type on infant growth. Pediatrics. 1998;
102:569–573.
129. Devaney B, Ziegler P, Pac S, Karwe V, Barr SI. Nutrient intakes of infants
and toddlers. J Am Diet Assoc. 2004;104(suppl 1):s14–s21.
130. Skinner JD, Ziegler P, Ponza M. Transitions in infants’ and toddlers’
beverage patterns. J Am Diet Assoc. 2004;104(suppl 1):s45–s50.
131. US Department of Agriculture, Food and Nutrition Services. 2002 CFR title
7 volume 4 chapter II (1-1-02 edition). 246-Special Supplemental Nutrition
Program for Women, Infants and Children (WIC). Subpart D-Participant
benefits. 246.10 Supplemental foods. Available at: http://www.fns.usda.gov/WIC/PDFfiles/WICregulations. Accessed January 31, 2005.
132. Dennison BA, Erb TA, Jenkins PL. Predictors of dietary milk fat intake by
preschool children. Prev Med. 2001;33:536–542.
133. Sullivan SA, Birch LL. Infant dietary experience and acceptance of solid
foods. Pediatrics. 1994;93:271–277.
134. Thompson FE, Dennison BA. Dietary sources of fats and cholesterol in US
children aged 2 through 5 years. Am J Public Health. 1994;84:799–806.
135. Dennison BA, Rockwell HL, Baker SL. Excess fruit juice consumption by
preschool-aged children is associated with short stature and obesity. Pediatrics. 1997;99:15–22.
136. Welsh JA, Cogswell ME, Rogers S, Rockett H, Mei Z, Grummer-Strawn
LM. Overweight among low-income preschool children associated with the
consumption of sweet drinks: Missouri, 1999–2002. Pediatrics. 2005;115:
e223–e229.
137. Drewnowski A. The role of energy density. Lipids. 2003;38:109–115.
Gidding et al
138. Birch LL, Deysher M. Caloric compensation and sensory specific satiety:
evidence for self regulation of food intake by young children. Appetite.
1986;7:323–331.
139. Araya H, Vera G, Alvina M. Effect of the energy density and volume of
high carbohydrate meals on short term satiety in preschool children. Eur
J Clin Nutr. 1999;53:273–276.
140. Bandini LG, Vu D, Must A, Cyr H, Goldberg A, Dietz WH. Comparison of
high-calorie, low-nutrient-dense food consumption among obese and
non-obese adolescents. Obes Res. 1999;7:438–443.
141. Phillips SM, Bandini LG, Naumova EN, Cyr H, Colclough S, Dietz WH,
Must A. Energy-dense snack food intake in adolescence: longitudinal relationship to weight and fatness. Obes Res. 2004;12:461–472.
142. Birch LL, Fisher JO. Development of eating behaviors among children and
adolescents. Pediatrics. 1998;101(pt 2):539–549.
143. Fisher JO, Birch LL. Restricting access to palatable foods affects children’s
behavioral response, food selection, and intake. Am J Clin Nutr. 1999;69:
1264–1272.
144. Ludwig DS, Peterson KE, Gortmaker SL. Relation between consumption of
sugar-sweetened drinks and childhood obesity: a prospective, observational
analysis. Lancet. 2001;357:505–508.
145. Committee on Nutrition. American Academy of Pediatrics: the use and
misuse of fruit juice in pediatrics. Pediatrics. 2001;107:1210–1213.
146. Bowman SA, Gortmaker SL, Ebbeling CB, Pereira MA, Ludwig DS.
Effects of fast-food consumption on energy intake and diet quality among
children in a national household survey. Pediatrics. 2004;113(pt
1):112–118.
147. Putnam J, Allshouse J, Kantor L. U.S. Per Capita Food Supply Trends:
More Calories, Refined Carbohydrates and Fats. Washington, DC:
Economic Research Service, US Department of Agriculture; 2002.
148. Epstein LH, Valoski A, Wing RR, McCurley J. Ten-year follow-up of
behavioral, family-based treatment for obese children. JAMA. 1990;264:
2519–2523.
149. Hayman LL, Williams CL, Daniels SR, Steinberger J, Paridon S, Dennison
BA, McCrindle BW; Committee on Atherosclerosis, Hypertension, and
Obesity in Youth (AHOY) of the Council on Cardiovascular Disease in the
Young, American Heart Association. Cardiovascular health promotion in
the schools: a statement for health and education professionals and child
health advocates from the Committee on Atherosclerosis, Hypertension, and
Obesity in Youth (AHOY) of the Council on Cardiovascular Disease in the
Young, American Heart Association. Circulation. 2004;110:2266–2275.
150. Position of ADA, SNE, and ASFSA: school-based nutrition programs and
services. J Am Diet Assoc. 1995;95:367–369.
151. Osganian SK, Ebzery MK, Montgomery DH, Nicklas TA, Evans MA,
Mitchell PD, Lytle LA, Snyder MP, Stone EJ, Zive MM, Bachman KJ, Rice
R, Parcel GS. Changes in the nutrient content of school lunches: results from
the CATCH Eat Smart Food service Intervention. Prev Med. 1996;25:
400–412.
152. Dwyer JT, Hewes LV, Mitchell PD, Nicklas TA, Montgomery DH, Lytle
LA, Snyder MP, Zive MM, Bachman KJ, Rice R, Parcel GS. Improving
school breakfasts: effects of the CATCH Eat Smart Program on the nutrient
content of school breakfasts. Prev Med. 1996;25:413–422.
153. Perry CL, Bishop DB, Taylor GL, Davis M, Story M, Gray C, Bishop SC,
Mays RA, Lytle LA, Harnack L. A randomized school trial of environmental strategies to encourage fruit and vegetable consumption among
children. Health Educ Behav. 2004;31:65–76.
154. Cullen KW, Ash DM, Warneke C, de Moor C. Intake of soft drinks,
fruit-flavored beverages, and fruits and vegetables by children in grades 4
through 6. Am J Public Health. 2002;92:1475–1478.
155. Birnbaum AS, Lytle LA, Story M, Perry CL, Murray DM. Are differences
in exposure to a multicomponent school-based intervention associated with
varying dietary outcomes in adolescents? Health Educ Behav. 2002;29:
427–443.
156. Williams CL, Strobino BA, Bollella M, Brotanek J. Cardiovascular risk
reduction in preschool children: the “Healthy Start” project. J Am Coll Nutr.
2004;23:117–123.
157. Levin S, Lowry R, Brown DR, Dietz WH. Physical activity and body mass
index among US adolescents: youth risk behavior survey, 1999. Arch
Pediatr Adolesc Med. 2003;157:816–820.
158. Boynton-Jarrett R, Thomas TN, Peterson KE, Wiecha J, Sobol AM,
Gortmaker SL. Impact of television viewing patterns on fruit and vegetable
consumption among adolescents. Pediatrics. 2003;112(pt 1):1321–1326.
2075
159. Dennison BA, Erb TA, Jenkins PL. Television viewing and television in
bedroom associated with overweight risk among low-income preschool
children. Pediatrics. 2002;109:1028–1035.
160. Coon KA, Goldberg J, Rogers BL, Tucker KL. Relationships between use
of television during meals and children’s food consumption patterns. Pediatrics. 2001;107:E7.
161. Institute of Medicine. WIC Food Packages: Time for Change: Executive
Summary. Washington, DC: National Academy of Sciences; 2004.
Available at: http://www.nap.edu/catalog/11280.html. Accessed April 30,
2005.
162. National High Blood Pressure Education Program NIH. The fourth report
on the diagnosis, evaluation, and treatment of high blood pressure in
children and adolescents. Pediatrics. 2004;114:555–576.
163. National Cholesterol Education Program (NCEP): highlights of the report of
the Expert Panel on Blood Cholesterol Levels in Children and Adolescents.
Pediatrics. 1992;89:495–501.
164. Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks FM,
Bray GA, Vogt TM, Cutler JA, Windhauser MM, Lin P-H, Karanja N. A
clinical trial of the effects of dietary patterns on blood pressure. DASH
Collaborative Research Group. N Engl J Med. 1997;336:1117–1124.
165. Simons-Morton DG, Obarzanek E. Diet and blood pressure in children and
adolescents. Pediatr Nephrol. 1997;11:244–249.
166. Simons-Morton DG, Hunsberger SA, Van Horn L, Barton BA, Robson
AM, McMahon RP, Muhonen LE, Kwiterovich PO, Lasser NL, Kimm SY,
Greenlick MR. Nutrient intake and blood pressure in the Dietary Intervention Study in Children. Hypertension. 1997;29:930–936.
167. Van Horn L. Fiber, lipids, and coronary heart disease: a statement for
healthcare professionals from the Nutrition Committee, American Heart
Association. Circulation. 1997;95:2701–2704.
168. Lichtenstein AH, Deckelbaum RJ. AHA Science Advisory. Stanol/sterol
ester-containing foods and blood cholesterol levels. A statement for
healthcare professionals from the Nutrition Committee of the Council on
Nutrition, Physical Activity, and Metabolism of the American Heart Association. Circulation. 2001;103:1177–1179.
169. Tammi A, Ronnemaa T, Miettinen TA, Gylling H, Rask-Nissila L, Viikari
J, Tuominen J, Marniemi J, Simell O. Effects of gender, apolipoprotein E
phenotype and cholesterol-lowering by plant stanol esters in children: the
STRIP study. Special Turku Coronary Risk Factor Intervention Project.
Acta Paediatr. 2002;91:1155–1162.
170. Davidson MH, Maki KC, Umporowicz DM, Ingram KA, Dicklin MR,
Schaefer E, Lane RW, McNamara JR, Ribaya-Mercado JD, Perrone G,
Robins SJ, Franke WC. Safety and tolerability of esterified phytosterols
administered in reduced-fat spread and salad dressing to healthy adult men
and women. J Am Coll Nutr. 2001;20:307–319.
171. Homma Y, Ikeda I, Ishikawa T, Tateno M, Sugano M, Nakamura H.
Decrease in plasma low-density lipoprotein cholesterol, apolipoprotein B,
cholesteryl ester transfer protein, and oxidized low-density lipoprotein by
plant stanol ester-containing spread: a randomized, placebo-controlled trial.
Nutrition. 2003;19:369–374.
172. Ntanios FY, Homma Y, Ushiro S. A spread enriched with plant sterol-esters
lowers blood cholesterol and lipoproteins without affecting vitamins A and
E in normal and hypercholesterolemic Japanese men and women. J Nutr.
2002;132:3650–3655.
173. Volpe R, Niittynen L, Korpela R, Sirtori C, Bucci A, Fraone N, Pazzucconi
F. Effects of yoghurt enriched with plant sterols on serum lipids in patients
with moderate hypercholesterolaemia. Br J Nutr. 2001;86:233–239.
174. Maki KC, Davidson MH, Umporowicz DM, Schaefer EJ, Dicklin MR,
Ingram KA, Chen S, McNamara JR, Gebhart BW, Ribaya-Mercado JD,
Perrone G, Robins SJ, Franke WC. Lipid responses to plant-sterol-enriched
reduced-fat spreads incorporated into a National Cholesterol Education
Program Step I diet. Am J Clin Nutr. 2001;74:33–43.
175. Miettinen TA, Puska P, Gylling H, Vanhanen H, Vartiainen E. Reduction of
serum cholesterol with sitostanol-ester margarine in a mildly hypercholesterolemic population. N Engl J Med. 1995;333:1308–1312.
176. Hendriks HF, Brink EJ, Meijer GW, Princen HM, Ntanios FY. Safety of
long-term consumption of plant sterol esters-enriched spread. Eur J Clin
Nutr. 2003;57:681–692.
177. Nestel P, Cehun M, Pomeroy S, Abbey M, Weldon G. Cholesterol-lowering
effects of plant sterol esters and non-esterified stanols in margarine, butter
and low-fat foods. Eur J Clin Nutr. 2001;55:1084–1090.

Issue Highlights: Editorials: Original Articles: Arrhythmia

Transcript

Documenti analoghi

No. 54 – February - Newspaper Innovation

Wednesday 2 September 2015

Laser Angioplasty Coronary Indications

Scientific Report 2011 Ongoing Research 2012

Part 8: Stabilization of the Patient With Acute Coronary Syndromes

2016 ESC Guidelines for the diagnosis and

Guidelines on myocardial revascularization

Untitled

Council of Europe`s Recommendation 90 (7)

Mendelian randomization analysis supports the causal role of

Intraoperative Flow Measurement - RePub

of Syncope