EDITOR IN CHIEF

ASSOCIATE EDITORS

William C. Roberts,

MD

Baylor Heart & Vascular Institute Baylor University Medical Center Wadley Tower No. 457 3600 Gaston Avenue Dallas, Texas 75246 (214)826-8252 Fax: (214)826-2855

Vincent E. Friedewald Paul A. Grayburn

ASSISTANT EDITORS

Robert C. Kowal Jeffrey M. Schussler

EDITORIAL BOARD
CARDIOVASCULAR MEDICINE In Adults

Antonio Abbate J. Dawn Abbott George S. Abela Jamil Aboulhosn Joseph S. Alpert Martin A. Alpert Ezra A. Amsterdam Jeffrey L. Anderson Evan Appelbaum Richard W. Asinger Pablo Avanzas Gary John Balady Eric Bates Jeroen J. Bax George A. Beller William E. Boden Monty M. Bodenheimer Robert O. Bonow Jeffrey S. Borer Martial G. Bourassa Eugene Braunwald Jeffrey A. Brinker David L. Brown Michael E. Cain Richard O. Cannon III Bernard R. Chaitman Kanu Chatterjee John S. Child Robert J. Cody Lawrence S. Cohen Marc Cohen C. Richard Conti Michael H. Crawford James P. Daubert Gregory J. Dehmer James A. de Lemos Anthony N. DeMaria Pablo Denes George A. Diamond John P. DiMarco Allen Dollar Michael J. Domanski Gerald Dorros Uri Elkayam Kenneth A. Ellenbogen Myrvin H. Ellestad Stephen G. Ellis Toby R. Engel Andrew E. Epstein N. A. Mark Estes, III Michael Ezekowitz

Rodney H. Falk John A. Farmer David P. Faxon Ted Feldman Jack Ferlinz Jerome L. Fleg Gerald F. Fletcher Joseph A. Franciosa Gary S. Francis W. Bruce Fye William H. Gaasch Julius M. Gardin Bernard J. Gersh S. David Gertz Mihai Gheorghiade D. Luke Glancy Stephen P. Glasser Michael R. Gold Samuel Z. Goldhaber Robert E. Goldstein Sidney Goldstein Steven A. Goldstein J. Anthony Gomes Antonio M. Gotto, Jr. K. Lance Gould Donald C. Harrison Philip D. Henry L. David Hillis David R. Holmes, Jr. Mun K. Hong William G. Hundley Ami S. Iskandrian Allan S. Jaffe Hani Jneid Greg L. Kaluza Joel S. Karliner John A. Kastor Sanjiv Kaul Ellen C. Keeley Kenneth M. Kent Richard E. Kerber Dean J. Kereiakes Morton J. Kern Spencer B. King III Robert E. Kleiger George J. Klein Lloyd W. Klein Paul Kligeld Robert A. Kloner John B. Kostis Charles Landau Richard L. Lange Carl J. Lavie Carl V. Leier B. T. Liang

Joseph Lindsay, Jr. Gregory Y.H. Lip Francisco Lopez-Jimenez Joseph Loscalzo G.B. John Mancini Francis E. Marchlinski Frank I. Marcus Barry J. Maron Martin S. Maron Randolph P. Martin Attilo Maseri Charles Maynard Michael D. McGoon Darren K. McGuire Raymond G. McKay Jawahar L. Mehta Bernard Meier Franz H. Messerli Eric L. Michelson Richard V. Milani Alan B. Miller Wayne L. Miller Gary S. Mintz Fred Morady Lori Mosca Arthur J. Moss James E. Muller Gerald B. Naccarelli Navin C. Nanda Robert A. ORourke Erik Magnus Ohman Richard L. Page Sebastian T. Palmeri Seung-Yung Park Eugene R. Passamani Alan S. Pearlman Carl J. Pepine Joseph K. Perloff Bertram Pitt Philip J. Podrid Don Poldermans Arshed A. Quyyumi Charles E. Rackley C. Venkata Ram Nathaniel Reichek Robert Roberts Jennifer G. Robinson Lynda E. Rosenfeld Melvin M. Scheinman David J. Schneider John S. Schroeder Patrick Washington Serruys Pravin M. Shah Prediman K. Shah Jamshid Shirani

Robert J. Siegel Marc A. Silver Ross J. Simpson, Jr. Steven N. Singh Burton E. Sobel John C. Somberg David H. Spodick Lynne W. Stevenson Gregory W. Stone John R. Stratton Jonathan M. Tobis Eric J. Topol Byron F. Vandenberg Hector O. Ventura George W. Vetrovec Robert A. Vogel Ron Waksman David D. Waters Nanette K. Wenger Robert Wilensky James T. Willerson Clyde W. Yancy Barry L. Zaret Douglas P. Zipes
In Infants and Children

Hugh D. Allen Bruce S. Alpert Stanley J. Goldberg Warren G. Guntheroth Howard P. Gutgesell John D. Kugler James E. Lock John W. Moore Lowell W. Perry David J. Sahn Richard M. Schieken
CARDIOVASCULAR SURGERY

Eugene H. Blackstone Lawrence I. Bonchek Lawrence H. Cohn John A. Elefteriades Hartzel V. Schaff
RELATED SPECIALISTS

L. Maximilian Buja Jean-Pierre Despres Michael Emmett Giovanni Filardo Barry A. Franklin Charles B. Higgins Jeffrey E. Saftz Renu Virmani

A5

CONTENTS
Coronary Artery Disease
An Early and Simple Predictor of Severe Left Main and/or Three-Vessel Disease in Patients With NonST-Segment Elevation Acute Coronary Syndrome ...................................................... 495
Masami Kosuge, Toshiaki Ebina, Kiyoshi Hibi, Satoshi Morita, Mitsuaki Endo, Nobuhiki Maejima, Noriaki Iwahashi, Kozo Okada, Toshiyuki Ishikawa, Satoshi Umemura, and Kazuo Kimura

THE AMERICAN JOURNAL OF CARDIOLOGY VOL. 107, NO. 4 FEBRUARY 15, 2011

Comparison of Morbidity and Mortality in Diabetics Versus Nondiabetics Having Isolated Coronary Bypass Versus Coronary Bypass plus Valve Operations Versus Isolated Valve Operations .... 535
Serenella Castelvecchio, Lorenzo Menicanti, Ekaterina Baryshnikova, Carlo de Vincentiis, Alessandro Frigiola, and Marco Ranucci, for the Surgical and Clinical Outcome Research (SCORE) Group

Reperfusion by Primary Percutaneous Coronary Intervention in Patients With ST-Segment Elevation Myocardial Infarction Within 12 to 24 Hours of the Onset of Symptoms (from a Prospective National Observational Study [PL-ACS]) ......................... 501
Marek Gierlotka, Mariusz Gasior, Krzysztof Wilczek, Michal Hawranek, Janusz Szkodzinski, Piotr Paczek, Andrzej Lekston, Zbigniew Kalarus, Marian Zembala, and Lech Polonski

Heart Failure
Relation of Bundle Branch Block to Long-Term (Four-Year) Mortality in Hospitalized Patients With Systolic Heart Failure ...................................... 540
Alon Barsheshet, Ilan Goldenberg, Moshe Garty, Shmuel Gottlieb, Amir Sandach, Avishag Laish-Farkash, Michael Eldar, and Michael Glikson

Review
Percutaneous Coronary Intervention for Non ST-Elevation Acute Coronary Syndromes: Which, When and How? ............................................ 509
Robert K. Riezebos, Jan G.P. Tijssen, Freek W.A. Verheugt, and Gerrit J. Laarman

Characteristics of Depression Remission and Its Relation With Cardiovascular Outcome Among Patients With Chronic Heart Failure (from the SADHART-CHF Study) ...................................... 545
Wei Jiang, Ranga Krishnan, Maragatha Kuchibhatla, Michael S. Cuffe, Carolyn Martsberger, Rebekka M. Arias, and Christopher M. OConnor, for the SADHART-CHF Investigators

Coronary Artery Disease

Long-Term Follow-Up of Patients With First-Time Chest Pain Having 64-Slice Computed Tomography ................................................... 516
Fabiola B. Sozzi, Filippo Civaia, Philippe Rossi, Jean-Francois Robillon, Stephane Rusek, Frederic Berthier, Francois Bourlon, Laura Iacuzio, Gilles Dreyfus, and Vincent Dor

Warfarin Use and Outcomes in Patients With Advanced Chronic Systolic Heart Failure Without Atrial Fibrillation, Prior Thromboembolic Events, or Prosthetic Valves ............................................. 552
Marjan Mujib, Abu-Ahmed Z. Rahman, Ravi V. Desai, Mustafa I. Ahmed, Margaret A. Feller, Inmaculada Aban, Thomas E. Love, Michel White, Prakash Deedwania, Wilbert S. Aronow, Gregg Fonarow, and Ali Ahmed

Usefulness of Cooling and Coronary Catheterization to Improve Survival in Out-of-Hospital Cardiac Arrest ............................................................ 522
Dion Stub, Christopher Hengel, William Chan, Damon Jackson, Karen Sanders, Anthony M. Dart, Andrew Hilton, Vincent Pellegrino, James A. Shaw, Stephen J. Duffy, Stephen Bernard, and David M. Kaye

Editorial
The Risk of Thromboembolism in Heart Failure: Does It Merit Anticoagulation Therapy? ............ 558
Eduard Shantsila and Gregory Y.H. Lip

Two-Year Safety and Effectiveness of SirolimusEluting Stents (from a Prospective Registry) ....... 528
Bimmer E. Claessen, Roxana Mehran, Martin B. Leon, Eric A. Heller, Giora Weisz, George Syros, Gary S. Mintz, Theresa Franklin-Bond, Irene Apostolidou, Jose P.S. Henriques, Gregg W. Stone, Jeffrey W. Moses, and George D. Dangas

Heart Failure
Trials on the Effect of Cardiac Resynchronization on Arterial Blood Pressure in Patients With Heart Failure ........................................................... 561
Sameer Ather, Sripal Bangalore, Srinath Vemuri, Long B. Cao, Biykem Bozkurt, and Franz H. Messerli

A6 THE AMERICAN JOURNAL OF CARDIOLOGY

VOL. 107

FEBRUARY 15, 2011

Patient Perception Versus Medical Record Entry of Health-Related Conditions Among Patients With Heart Failure .................................................. 569
Adnan S. Malik, Grigorios Giamouzis, Vasiliki V. Georgiopoulou, Lucy V. Fike, Andreas P. Kalogeropoulos, Catherine R. Norton, Dan Sorescu, Sidra Azim, Sonjoy R. Laskar, Andrew L. Smith, Sandra B. Dunbar, and Javed Butler

Cardiomyopathy
Relation of Pulse Pressure to Blood Pressure Response to Exercise in Patients With Hypertrophic Cardiomyopathy ............................................. 600
Kevin S. Heffernan, Martin S. Maron, Eshan A. Patvardhan, Richard H. Karas, and Jeffrey T. Kuvin, the Vascular Function Study Group

Editorial
Effectiveness of Serial Increases in Amino-Terminal ProB-Type Natriuretic Peptide Levels to Indicate the Need for Mechanical Circulatory Support in Children With Acute Decompensated Heart Failure ........................................................... 573
Derek T.H. Wong, Kristen George, Judith Wilson, Cedric Manlhiot, Brian W. McCrindle, Khosrow Adeli, and Paul F. Kantor

Clinical Challenges of Genotype Positive ()Phenotype Negative () Family Members in Hypertrophic Cardiomyopathy ......................... 604
Barry J. Maron, Laura Yeates, and Christopher Semsarian

Miscellaneous
Usefulness of Repeated N-Terminal Pro-B-Type Natriuretic Peptide Measurements as Incremental Predictor for Long-Term Cardiovascular Outcome After Vascular Surgery .................................... 609
Dustin Goei, Jan-Peter van Kuijk, Willem-Jan Flu, Sanne E. Hoeks, Michel Chonchol, Hence J.M. Verhagen, Jeroen J. Bax, and Don Poldermans

Arrhythmias and Conduction Disturbances

Relation of Obesity to Recurrence Rate and Burden of Atrial Fibrillation ......................................... 579
Maya Guglin, Kuldeep Maradia, Ren Chen, and Anne B. Curtis

Roundtable Discussion (CME)

The Editors Roundtable: Implantable CardioverterDebrillators in Primary Prevention of Sudden Cardiac Death and Disparity-Related Barriers to Implementation ............................................... 583
Vincent E. Friedewald, Gregg C. Fonarow, Brian Olshansky, Clyde W. Yancy, and William C. Roberts

Usefulness of At Rest and Exercise Hemodynamics to Detect Subclinical Myocardial Disease in Type 2 Diabetes Mellitus ............................................. 615
Christine L. Jellis, Tony Stanton, Rodel Leano, Jennifer Martin, and Thomas H. Marwick

Specic Characteristics of Sudden Death in a Mediterranean Spanish Population ................... 622

M. Teresa Subirana, Josep O. Juan-Babot, Teresa Puig, Joaqu n Lucena, Antonio Rico, Manuel Salguero, Juan C. Borondo, Jorge Ordo ez, Josep Arimany, Rafael Va zquez, Lina Badimon, Gaetano Thiene, and Antonio Baye s de Luna

Valvular Heart Disease

Comparison of the Effectiveness and Safety of LowMolecular Weight Heparin Versus Unfractionated Heparin Anticoagulation After Heart Valve Surgery .......................................................... 591
Claudia Bucci, William H. Geerts, Andrew Sinclair, and Stephen E. Fremes

Clinical and Prognostic Relevance of Echocardiographic Evaluation of Right Ventricular Geometry in Patients With Idiopathic Pulmonary Arterial Hypertension ...................................... 628
Stefano Ghio, Anna Sara Pazzano, Catherine Klersy, Laura Scelsi, Claudia Raineri, Rita Camporotondo, Andrea DArmini, and Luigi Oltrona Visconti

Congenital Heart Disease

Seeking Optimal Relation Between Oxygen Saturation and Hemoglobin Concentration in Adults With Cyanosis from Congenital Heart Disease ... 595
Craig S. Broberg, Ananda R. Jayaweera, Gerhard P. Diller, Sanjay K. Prasad, Swee Lay Thein, Bridget E. Bax, John Burman, and Michael A. Gatzoulis

Clinically Signicant Incidental Findings Among Human Immunodeciency Virus-Infected Men During Computed Tomography for Determination of Coronary Artery Calcium ................................. 633
Nancy Crum-Cianone, James Stepenosky, Sheila Medina, Dylan Wessman, David Krause, and Gilbert Boswell
CONTENTS

A11

Case Report
Self-Terminated Ventricular Fibrillation and Recurrent Syncope .......................................... 638
Yuval Konstantino, Angela Morello, Peter J. Zimetbaum, and Mark E. Josephson

Long-Term Follow Up of Atrioventricular Block in Transcatheter Aortic Valve Implantation ............ 641 Instructions to Authors can be found at the AJC website: www.AJConline.org Classieds on pages A10, A37

Readers Comments
Comparison of 600 Versus 300-mg Clopidogrel Loading Dose in Patients With ST-Segment Elevation Myocardial Infarction Undergoing Primary Coronary Angioplasty ..................................... 641

Full Text: www.ajconline.org

A18 THE AMERICAN JOURNAL OF CARDIOLOGY

VOL. 107

FEBRUARY 15, 2011

An Early and Simple Predictor of Severe Left Main and/or Three-Vessel Disease in Patients With NonST-Segment Elevation Acute Coronary Syndrome
Masami Kosuge, MD*, Toshiaki Ebina, MD, Kiyoshi Hibi, MD, Satoshi Morita, PhD, Mitsuaki Endo, MD, Nobuhiki Maejima, MD, Noriaki Iwahashi, MD, Kozo Okada, MD, Toshiyuki Ishikawa, MD, Satoshi Umemura, MD, and Kazuo Kimura, MD
Clopidogrel should be initiated as soon as possible in patients with nonST-segment elevation acute coronary syndrome (NSTE-ACS) except those who urgently require coronary artery bypass grafting (CABG). The present study assessed the ability to predict severe left main coronary artery and/or 3-vessel disease (LM/3VD) that would most likely require urgent CABG based on only clinical factors on admission in 572 patients with NSTE-ACS undergoing coronary angiography. Severe LM/3VD was dened as >75% stenosis of LM and/or 3VD with >90% stenosis in >2 proximal lesions of the left anterior descending coronary artery and other major epicardial arteries. Patients were divided into the 3 groups according to angiographic ndings: no LM/3VD (n 460), LM/3VD but not severe LM/3VD (n 57), and severe LM/3VD (n 55). Severe LM/3VD was associated with a higher rate of urgent CABG compared to no LM/3VD and LM/3VD but not severe LM/3VD (46%, 2%, and 2%, p <0.001). On multivariate analysis, degree of ST-segment elevation in lead aVR was the strongest predictor of severe LM/3VD (odds ratio 29.1, p <0.001), followed by positive troponin T level (odds ratio 1.27, p 0.044). ST-segment elevation >1.0 mm in lead aVR best identied severe LM/3VD with 80% sensitivity, 93% specicity, 56% positive predictive value, and 98% negative predictive value. In conclusion, ST-segment elevation >1.0 mm in lead aVR on admission electrocardiogram is highly suggestive of severe LM/3VD in patients with NSTE-ACS. Selected patients with this nding might benet from promptly undergoing angiography, withholding clopidogrel to allow early CABG. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107: 495500) Dual antiplatelet therapy with clopidogrel and aspirin should be initiated as soon as possible in patients with nonST-segment elevation acute coronary syndrome (NSTE-ACS).1,2 However, such combination therapy can increase perioperative bleeding in patients undergoing early coronary artery bypass grafting (CABG).37 Therefore, one might consider with-holding clopidogrel until coronary angiography and denition of the coronary anatomy.8 The proportion of patients with NSTE-ACS who undergo CABG during hospitalization is 9% to 21%.4,5,8 12 CABG can often be deferred for several days, and few patients require urgent CABG. Ideally, clopidogrel should be withheld in the minority of patients who urgently require CABG and should be given to the remaining majority of patients. We previously examined clinical factors related to left main coronary artery and/or 3-vessel disease (LM/3VD) that would most likely lead to CABG in patients with NSTEACS but did not evaluate severity of coronary lesions in that study.13 In the present study, we assessed the ability to predict severe LM/3VD, which would most likely to
Division of Cardiology, Yokohama City University Medical Center, Yokohama, Japan. Manuscript received August 28, 2010; revised manuscript received and accepted October 1, 2010. *Corresponding author: Tel: 81-45-261-5656; fax: 81-45-261-9162. E-mail address: masami-kosuge@pop06.odn.ne.jp (M. Kosuge). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.005

require urgent CABG, using only clinical factors on admission in patients with NSTE-ACS. Methods We studied 572 consecutive patients (mean age 67 11 years, range 30 to 92, 397 men and 175 women) who were admitted to Yokohama City University Medical Center (Yokohama, Japan) coronary care unit and fullled the following criteria: (1) typical chest discomfort attributed to cardiac ischemia, lasting 5 minutes, occurring within 24 hours before hospital admission, and involving an unstable pattern of pain including pain at rest, new onset, severe or frequent angina, or accelerating angina14; (2) no conditions precluding evaluation ST-segment changes on electrocardiogram (ECG) such as left or right bundle branch block, left ventricular hypertrophy, or ventricular pacing; (3) fully assessable ECGs on admission; and (4) fully assessable angiographic data during hospitalization. We excluded patients with nonischemic or atypical pain, persistent new ST-segment elevation in leads other than lead aVR, recent (6 months) percutaneous coronary intervention, or previous CABG. All patients gave informed consent. The study protocol was approved by the internal review board of Yokohama City University Medical Center. Standard 12-lead ECGs were recorded on admission at a
www.ajconline.org

496 Table 1 Clinical characteristics

The American Journal of Cardiology (www.ajconline.org)

LM/3VD No LM/3VD (n 460) Age (years) Men Systolic blood pressure on admission (mm Hg) Heart rate on admission (beats/min) Killip class II on admission Symptom onset 6 hours Previous myocardial infarction Previous percutaneous coronary intervention Risk factors Hypertension Diabetes mellitus Smoking Hyperlipidemia* Family history of coronary artery disease Hemoglobin on admission (g/dl) High-sensitivity C-reactive protein on admission (mg/dl) Positive troponin T on admission Creatine kinase-MB on admission (IU/L) Estimated glomerular ltration rate on admission (ml/min/1.73 m2) Brain natriuretic peptide on admission (pg/ml) Cardiac procedures and outcomes at 30 days Death Myocardial (re)infarction Death/myocardial (re)infarction Urgent percutaneous coronary intervention Urgent coronary artery bypass surgery Urgent revascularization (percutaneous coronary intervention or coronary artery bypass surgery) Cardiac procedures Percutaneous coronary intervention Coronary artery bypass surgery Any revascularization (percutaneous coronary intervention or coronary artery bypass surgery) 66 11 322 (70%) 150 25 76 17 26 (6%) 356 (78%) 86 (19%) 90 (20%) 304 (66%) 136 (30%) 229 (50%) 230 (50%) 120 (26%) 14 2 0.131 (0.0610.323) 135 (29%) 14 16 68 25 67 (26179) (n 297) 1 (0.2%) 14 (3%) 15 (3%) 29 (6%) 7 (2%) 36 (8%) Nonsevere (n 57) 69 10 39 (68%) 150 32 81 20 9 (16%) 43 (75%) 18 (32%) 15 (26%) 42 (74%) 29 (51%) 22 (39%) 25 (44%) 13 (23%) 13 2 0.180 (0.0790.453) 28 (49%) 18 24 58 28 187 (81429) (n 32) 1 (2%) 3 (5%) 4 (7%) 7 (12%) 1 (2%) 8 (14%) Severe (n 55) 68 11 36 (66%) 141 26 89 23 17 (31%) 49 (89%) 12 (22%) 5 (9%) 38 (69%) 30 (55%) 23 (42%) 29 (53%) 16 (29%) 13 2 0.253 (0.0990.801) 33 (60%) 27 36 58 26 230 (67571) (n 31) 2 (4%) 5 (9%) 7 (13%) 5 (9%) 25 (46%) 30 (55%)

p Value

0.06 0.78 0.07 0.001 0.001 0.13 0.07 0.06 0.49 0.001 0.18 0.61 0.75 0.033 0.005 0.001 0.001 0.004 0.001

0.010 0.23 0.004 0.22 0.001 0.001 0.001 0.001 0.001

272 (59%) 27 (6%) 291 (63%)

36 (63%) 13 (23%) 49 (86%)

14 (25%) 40 (73%) 54 (98%)

Data are presented as mean SD, median (interquartile range), or number of patients (percentage). * Fasting total cholesterol concentration 220 mg/dl, fasting triglyceride concentration 150 mg/dl, or use of antihyperlipidemic therapy. Available for 360 patients.

Table 2 Electrocardiographic ndings Variable No LM/3VD (n 460) ST-segment depression 0.5 mm Maximal ST-segment depression (mm) Sum of ST-segment depressions (mm) Number of leads with ST-segment depression 0.5 mm ST-segment elevation 0.5 mm in lead aVR ST-segment elevation in lead aVR (mm) 288 (63%) 0.8 1.0 2.6 3.6 2.5 2.5 68 (15%) 0.1 0.3 Nonsevere (n 57) 53 (93%) 1.7 1.1 6.7 5.1 5.1 2.6 39 (68%) 0.6 0.5 LM/3VD Severe (n 55) 55 (100%) 2.6 1.7 10.5 7.3 6.1 2.2 50 (91%) 1.2 0.7 0.001 0.001 0.001 0.001 0.001 0.001 p Value

Data are presented as mean SD or number of patients (percentage).

paper speed of 25 mm/s and an amplication of 10 mm/mV. All ECGs were examined by a single investigator who was blinded to all other clinical data. ST-segment shifts were measured 80 ms after the J-point for ST-segment depression and 20

ms after this point for ST-segment elevation using the preceding TP segment as a baseline.15 ST-segment deviation was considered present if deviation was 0.5 mm in any lead.14 A qualitative assay for cardiac-specic troponin T (de-

Coronary Artery Disease/Prediction of Severe LM/3VD in NSTE-ACS

497

tection limit 0.1 ng/ml of cardiac-specic troponin T; Roche Diagnostics, Tokyo, Japan) was performed on admission. Troponin T 0.1 ng/ml was dened as positive. Blood samples for measuring hemoglobin, plasma high-sensitivity C-reactive protein levels, and estimated glomerular ltration rate were also taken on admission. Japanese equations were used to calculate estimated glomerular ltration rate from serum creatinine level.16 Brain natriuretic peptide was simultaneously measured in 360 patients. Creatine kinase-MB levels were measured on admission, at 3-hour intervals during the rst 24 hours, and in any patient with suspected reinfarction. All patients underwent cardiac catheterization a median of 3 days after admission. Urgent cardiac catheterization was performed in patients with unstable hemodynamics from ischemic attacks or with ischemic attacks that could not be controlled by intensive drug treatment. Type and timing of revascularization were left to the discretion of the treating physician. All coronary angiograms were evaluated by a single investigator who was blinded to all other clinical data. Stenosis 50% in the diameter of the LM or stenosis of 75% in 1 major epicardial vessel or its main branches was considered clinically signicant. Severe LM/3VD was dened as (1) 75% stenosis of the LM, (2) 3VD with 90% stenosis of the proximal portion of the left anterior descending coronary artery and 90% stenosis of the right coronary artery and/or left circumex coronary artery, and (3) denitions 1 and 2. Patients were categorize according to presence (n 112) or absence (n 460) of LM/3VD, and the former group was subdivided according to severity of coronary lesions: nonsevere LM/3VD (n 57) and severe LM/3VD (n 55). Demographic data, risk factors for coronary artery disease, and data from physical examination on admission were collected. Major adverse events such as death, myocardial (re)infarction, or urgent revascularization were also recorded for all patients. Myocardial infarction was diagnosed according to cardiac enzyme levels or electrocardiographic criteria. Enzymatic evidence of myocardial infarction was dened as an increase of creatine kinase-MB to higher than the upper limit of normal if the previous creatine kinase-MB level was in the normal range or 50% above the previous level if the previous level was above the normal range.17 Electrocardiographic evidence of myocardial infarction was dened as new clinically signicant Q waves in 2 contiguous leads distinct from the enrollment myocardial infarction.17 Patients were followed for 30 days after admission. Results are expressed as mean SD or as frequency (percentage), and high-sensitivity C-reactive protein and brain natriuretic peptide levels are expressed as median and interquartile range. Data were compared by 1-way analysis of variance, Kruskal-Wallis test, and chi-square analysis. Differences were considered statistically signicant at p value 0.05. Multivariate logistic regression analysis was used to identify clinical predictors of severe LM/3VD among the variables associated (p 0.05) with this diagnosis on univariate analysis. Odds ratios and 95% condence intervals were calculated. In addition, sensitivity, specicity, positive predictive value, negative predictive value, and predictive accuracy of predictors of severe LM/3VD iden-

Figure 1. Representative electrocardiogram of a patient with severe left main coronary artery and/or 3-vessel disease. Troponin T was positive on admission. ST-segment elevation in lead aVR was 4.5 mm on admission electrocardiogram. Urgent coronary angiography showed 90% stenosis of the left main trunk.

tied on multivariate analysis were determined. SPSS statistical software (SPSS, Inc., Chicago, Illinois) was used for all analyses. Results Baseline characteristics are listed in Table 1. Patients with LM/3VD, especially severe LM/3VD, had a more

498

The American Journal of Cardiology (www.ajconline.org)

Table 3 Univariate and multivariate predictors of severe left main coronary artery and/or three-vessel disease Variable Odds Ratio (95% CI) Univariate Systolic blood pressure Heart rate Killip class II Previous percutaneous coronary intervention Diabetes mellitus High-sensitivity C-reactive protein Positive troponin T Creatine kinase-MB Estimated glomerular ltration rate Maximal ST-segment depression Sum of ST-segment depressions Number of leads with ST-segment depression 0.5 mm Degree of ST-segment elevation in lead aVR CI condence interval. Table 4 Comparison of ST-segment elevation in lead aVR and positive troponin T for predicting severe left main coronary artery and/or three-vessel disease Sensitivity Specicity ST-segment elevation in lead aVR 0.5 mm 1.0 mm 1.5 mm Positive troponin T PPV NPV Predictive Accuracy 0.020 0.001 0.001 0.045 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 p Value Multivariate 0.07 0.29 0.29 0.80 0.08 0.30 0.044 0.33 0.32 0.053 0.055 0.24 0.001

1.27 (1.102.78)

29.1 (9.5449.8)

and predictive accuracy of ST-segment elevation in lead aVR and positive troponin T for severe LM/3VD are presented in Table 4. ST-segment elevation 1.0 mm in lead aVR best identied severe LM/3VD. Discussion Our study showed that ST-segment elevation 1.0 mm in lead aVR and positive troponin T on admission (especially the former) were highly suggestive of severe LM/ 3VD, and the converse was also true, i.e., absence of these ndings was rarely associated with severe LM/3VD. To our knowledge, this is the rst study to establish a reliable technique for early identication of patients with severe LM/3VD who are most likely to require urgent CABG in patients with NSTE-ACS. Our ndings have important implications for identication of high-risk patients and selection of optimal treatment strategy in the setting of NSTEACS. The standard 12-lead ECG, which is an inexpensive, noninvasive, and readily available clinical tool, has a central role in diagnosis and immediate triage for NSTE-ACS and provides important prognostic information. In particular, presence of ST-segment depression on admission ECG has been recognized to be a strong predictor of adverse outcomes in patients with NSTE-ACS.14,1720 The Global Utilization of Strategies to Open Occluded Arteries in Acute Coronary Syndrome IV (GUSTO-IV ACS) trial of 7,800 patients with NSTE-ACS has highlighted the striking prognostic value of ST-segment depression on admission compared to expanded biomarker proles and traditional risk factors.18 However, most previous studies assessing the clinical signicance of admission ECG in patients with NSTE-ACS have focused on ST-segment deviation in leads other than lead aVR; i.e., clinicians have used an 11-lead ECG, neglecting lead aVR. Several studies have found that analysis of lead aVR is useful for evaluation of NSTE-ACS.13,15,21,22 Gorgels et al21 reported that ST-segment elevation in lead aVR accompanied by ST-segment depression in leads I, II, and V4 to V6 during episodes of angina strongly suggests LM/3VD in

91% 80% 27% 60%*

79% 93% 98% 69%

32% 56% 58% 17%

99% 98% 93% 94%

80% 92% 91% 68%

NPV negative predictive value; PPV positive predictive value. * p 0.05; p 0.01 versus ST-segment elevation 1.0 mm in lead aVR.

rapid heart rate, higher prevalences of Killip class II, diabetes mellitus, positive troponin T, and higher levels of high-sensitivity C-reactive protein, creatine kinase-MB, and brain natriuretic peptide than did patients without LM/3VD. LM/3VD was associated with lower levels of hemoglobin and estimated glomerular ltration rate. There were no signicant differences in other clinical variables among the 3 groups. Urgent CABG was more frequently done in patients with severe LM/3VD (46%). In contrast, urgent CABG was done in only 2% of patients with LM/3VD but not severe LM/ 3VD. Electrocardiographic ndings are presented in Table 2. Compared to patients without LM/3VD, those with LM/ 3VD, especially severe LM/3VD, had a higher prevalence and a larger amount of ST-segment depression, a larger number of leads other than lead aVR with ST-segment depression, and a higher prevalence and greater magnitude of ST-segment elevation in lead aVR. Figure 1 shows a representative ECG of a patient with severe LM/3VD. In multivariate models, degree of ST-segment elevation in lead aVR was the strongest predictor of severe LM/3VD, followed by positive troponin T (Table 3). Sensitivity, specicity, positive predictive value, negative predictive value,

Coronary Artery Disease/Prediction of Severe LM/3VD in NSTE-ACS

499

patients with angina at rest. Barrabs et al15 demonstrated that presence of ST-segment elevation in lead aVR predicts risk of in-hospital death in patients with a rst nonSTsegment elevation acute myocardial infarction. In that study, ST-segment elevation in lead aVR was also related to LM/3VD; however, coronary angiography was performed in only 56% of subjects within 6 months after infarction. We previously demonstrated that presence of ST-segment elevation 0.5 mm in lead aVR on admission ECG strongly suggested LM/3VD and had a higher prognostic value than ST-segment depression in other leads in patients with NSTE-ACS who underwent coronary angiography in the acute phase.13,22 However, previous studies, including ours, did not consider severity of LM/3VD, which has clinical implications for timing of CABG in relation to dual antiplatelet therapy. An increased risk of perioperative bleeding events due to early clopidogrel administration is clinically problematic in patients with LM/3VD who urgently require CABG. In such patients, postponing CABG for several days might seriously compromise outcomes. Timing of CABG depends on many factors including severity of coronary lesions, risk of ongoing ischemia, general condition of a patient, bleeding risk associated with upstream antithrombotic therapies, and local logistic factors such as collocation of cardiac surgical services and surgical waiting lists. The present study examined predictors of patients with severe LM/3VD likely to require urgent CABG, considering the coronary anatomy. We demonstrated that ST-segment elevation 1.0 mm in lead aVR was the most accurate predictor of severe LM/3VD. However, its positive predictive value was 56%, which was moderate. More importantly, the negative predictive value of ST-segment elevation 1.0 mm in lead aVR for detection of severe LM/3VD was 98%, which was very high. Absence of this nding was rarely associated with severe LM/3VD. If ST-segment elevation 1.0 mm in lead aVR is absent, treatment with upstream clopidogrel is strongly recommended. Lead aVR has a unique position because the positive pole is oriented toward the right upper side of the heart and looks into the left ventricular cavity from the right shoulder in the setting of NSTE-ACS.23 Lead aVR is therefore referred to as a cavity lead, and ST-segment elevation in this lead might reect global subendocardial ischemia.24 In patients with LM/ 3VD, severe extensive ischemia of the subendocardial layer leads to ST-segment elevation in lead aVR and extensive ST-segment depression in leads other than lead aVR. The magnitude of these changes is thought to reect severity of LM/3VD. In the present study, LM/3VD, especially severe LM/3VD, was associated with a greater degree and extent of ST-segment depression and a greater degree of ST-segment elevation in lead aVR. A meta-analysis of 12,030 patients with stable coronary artery disease enrolled in 60 studies demonstrated that amount of ST-segment depression during exercise stress testing is strongly associated with critical coronary artery disease such as LM/3VD.25 Furthermore, a greater degree and extent of ST-segment depression, not only its presence or absence, has been shown to correlate with an increased likelihood of LM/3VD and poor outcomes in patients with NSTE-ACS.17,19,20 The present study demonstrated that the value of ST-segment elevation in lead aVR for detection of severe LM/3VD surpassed that of

ST-segment depression in other leads in patients with NSTE-ACS. Recently approved antiplatelet agents such as prasugrel and ticagrelor, a new reversible agent, have been shown to decrease ischemic events compared to clopidogrel, but the former increased the risk of perioperative bleeding7 and the latter did not decrease the risk of perioperative bleeding.26 Until an antiplatelet agent that decreases ischemic events and decreases perioperative bleeding compared to clopidogrel becomes available, some patients will be exposed to a risk of urgent CABG-related bleeding caused by upstream dual antiplatelet therapy. This study was retrospective, performed at a single center, and included a small number of patients who underwent coronary angiography during hospitalization. However, the proportion of patients undergoing CABG during hospitalization in this study (14%) was similar to that in previous studies.4,10,12 Moreover, because our subjects underwent cardiac catheterization a median of 3 days after admission, our data on clinical outcomes according to angiographic ndings cannot be generalized to hospitals that provide early invasive strategies. Further studies in larger numbers of patients are needed to verify our results.
1. Anderson JL, Adams CD, Antman EM, Bridges CR, Califf RM, Casey DE Jr, Chavey WE II, Fesmire FM, Hochman JS, Levin TN, Lincoff AM, Peterson ED, Theroux P, Wenger NK, Wright RS, Smith SC Jr, Jacobs AK, Halperin JL, Hunt SA, Krumholz HM, Kushner FG, Lytle BW, Nishimura R, Ornato JP, Page RL, Riegel B. ACC/AHA guidelines for the management of patients with unstable angina/nonSTsegment elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (writing committee to revise the 2002 guidelines for the management of patients with unstable angina/non-st-elevation myocardial infarction). Circulation 2007;116:803 877. 2. Bassand JP, Hamm CW, Ardissino D, Boersma E, Budaj A, Fernndez-Avils F, Fox KA, Hasdai D, Ohman EM, Wallentin L, Wijns W. Guidelines for the diagnosis and treatment of non-ST-segment elevation acute coronary syndromes. The task force for the diagnosis and treatment of non-ST-segment elevation acute coronary syndromes of the European Society of Cardiology. Eur Heart J 2007;28:1598 1660. 3. Berger JS, Frye CB, Harshaw Q, Edwards FH, Steinhubl SR, Becker RC. Impact of clopidogrel in patients with acute coronary syndromes requiring coronary artery bypass surgery: a multicenter analysis. J Am Coll Cardiol 2008;52:16931701. 4. Ebrahimi R, Dyke C, Mehran R, Manoukian SV, Feit F, Cox DA, Gersh BJ, Ohman EM, White HD, Moses JW, Ware JH, Lincoff AM, Stone GW. Outcomes following pre-operative clopidogrel administration in patients with acute coronary syndromes undergoing coronary artery bypass surgery: the ACUITY (Acute Catheterization and Urgent Intervention Triage strategY) trial. J Am Coll Cardiol 2009;53:1965 1972. 5. Mehta RH, Roe MT, Mulgund J, Ohman EM, Cannon CP, Gibler WB, Pollack CV Jr, Smith SC Jr, Ferguson TB, Peterson ED. Acute clopidogrel use and outcomes in patients with nonST-segment elevation acute coronary syndromes undergoing coronary artery bypass surgery. J Am Coll Cardiol 2006;48:281286. 6. Fox KA, Mehta SR, Peters R, Zhao F, Lakkis N, Gersh BJ, Yusuf S. Clopidogrel in Unstable Angina to Prevent Recurrent Ischemic Events trial. Benets and risks of the combination of clopidogrel and aspirin in patients undergoing surgical revascularization for non-ST-elevation acute coronary syndrome: the Clopidogrel in Unstable Angina to Prevent Recurrent Ischemic Events (CURE) trial. Circulation 2004; 110:12021208. 7. Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, Gottlieb S, Neumann FJ, Ardissino D, De Servi S, Murphy SA, Riesmeyer J, Weerakkody G, Gibson CM, Antman EM; TRITONTIMI 38 Investigators. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2007;357:20012015.

500

The American Journal of Cardiology (www.ajconline.org) 18. Westerhout CM, Fu Y, Lauer MS, James S, Armstrong PW, Al-Hattab E, Califf RM, Simoons ML, Wallentin L, Boersma E; GUSTO-IV ACS Trial Investigators. Short- and long-term risk stratication in acute coronary syndromes: the added value of quantitative ST-segment depression and multiple biomarkers. J Am Coll Cardiol 2006;48: 939 947. 19. Yan RT, Yan AT, Mahaffey KW, White HD, Pieper K, Sun JL, Pepine CJ, Biasucci LM, Gulba DC, Lopez-Sendon J, Goodman SG; SYNERGY Trial Investigators. Prognostic utility of quantifying evolutionary ST-segment depression on early follow-up electrocardiogram in patients with nonST-segment elevation acute coronary syndromes. Eur Heart J 2010;31:958 966. 20. Holmvang L, Clemmensen P, Lindahl B, Lagerqvist B, Venge P, Wagner G, Wallentin L, Grande P. Quantitative analysis of the admission electrocardiogram identies patients with unstable coronary artery disease who benet the most from early invasive treatment. J Am Coll Cardiol 2003;41:905915. 21. Gorgels AP, Vos MA, Mulleneers R, de Zwaan C, Bar FW, Wellens HJ. Value of the electrocardiogram in diagnosing the number of severely narrowed coronary arteries in rest angina pectoris. Am J Cardiol 1993;72:999 1003. 22. Kosuge M, Kimura K, Ishikawa T, Ebina T, Hibi K, Tsukahara K, Kanna M, Iwahashi N, Okuda J, Nozawa N, Ozaki H, Yano H, Kusama I, Umemura S. Combined prognostic utility of ST segment in lead aVR and troponin T on admission in nonST-segment elevation acute coronary syndromes. Am J Cardiol 2006;97:334 339. 23. Yu PN, Stewart JM. Subendocardial myocardial infarction with special reference to the electrocardiographic changes. Am Heart J 1950;39: 862 880. 24. Kligeld P, Gettes LS, Bailey JJ, Childers R, Deal BJ, Hancock EW, van Herpen G, Kors JA, Macfarlane P, Mirvis DM, Pahlm O, Rautaharju P, Wagner GS. Recommendations for the standardization and interpretation of the electrocardiogram: part I: The electrocardiogram and its technology: a scientic statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Circulation 2007;115:1306 1324. 25. Detrano R, Gianrossi R, Mulvihill D, Lehmann K, Dubach P, Colombo A, Froelicher V. Exercise-induced ST segment depression in the diagnosis of multivessel coronary disease: a meta analysis. J Am Coll Cardiol 1989;14:15011508. 26. Wallentin L, Becker RC, Budaj A, Cannon CP, Emanuelsson H, Held C, Horrow J, Husted S, James S, Katus H, Mahaffey KW, Scirica BM, Skene A, Steg PG, Storey RF, Harrington RA, Freij A, Thorsn M; PLATO Investigators. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2009;361:10451057.

8. Barker CM, Anderson HV. Acute coronary syndromes: dont bypass the clopidogrel. J Am Coll Cardiol 2009;53:19731974. 9. Deyell MW, Ghali WA, Ross DB, Zhang J, Hemmelgarn BR; Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease (APPROACH) Investigators. Timing of nonemergent coronary artery bypass grafting and mortality after nonST elevation acute coronary syndrome. Am Heart J 2010;159:490 496. 10. Sadanandan S, Cannon CP, Gibson CM, Murphy SA, DiBattiste PM, Braunwald E; TIMI Study Group. A risk score to estimate the likelihood of coronary artery bypass surgery during the index hospitalization among patients with unstable angina and nonST-segment elevation myocardial infarction. J Am Coll Cardiol 2004;44:799 803. 11. Chew DP, Mahaffey KW, White HD, Huang Z, Hoekstra JW, Ferguson JJ, Califf RM, Aylward PE. Coronary artery bypass surgery in patients with acute coronary syndromes is difcult to predict. Am Heart J 2008;155:841 847. 12. Mehta RH, Chen AY, Pollack CV Jr, Roe MT, Zalenski RJ, Clements EA, Gibler WB, Ohman EM, Harrington RA, Peterson ED. Challenges in predicting the need for coronary artery bypass grafting at presentation in patients with nonST-segment elevation acute coronary syndromes. Am J Cardiol 2006;98:624 627. 13. Kosuge M, Kimura K, Ishikawa T, Ebina T, Shimizu T, Hibi K, Toda N, Tahara Y, Tsukahara K, Kanna M, Okuda J, Nozawa N, Ozaki H, Yano H, Umemura S. Predictors of left main or three-vessel disease in patients who have acute coronary syndromes with non-ST-segment elevation. Am J Cardiol 2005;95:1366 1369. 14. Cannon CP, McCabe CH, Stone PH, Rogers WJ, Schactman M, Thompson BW, Pearce DJ, Diver DJ, Kells C, Feldman T, Williams M, Gibson RS, Kronenberg MW, Ganz LI, Anderson HV, Braunwald E. The electrocardiogram predicts one-year outcome of patients with unstable angina and non-Q wave myocardial infarction: results of the TIMI III registry ECG ancillary study. J Am Coll Cardiol 1997;30: 133140. 15. Barrabs JA, Figueras J, Moure C, Cortadellas J, Soler-Soler J. Prognostic value of lead aVR in patients with a rst nonST-segment elevation acute myocardial infarction. Circulation 2003;108:814 819. 16. Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, Yamagata K, Tomino Y, Yokoyama H, Hishida A. On behalf of the collaborators developing the Japanese equation for estimated GFR. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53:982992. 17. Savonitto S, Cohen MG, Politi A, Hudson MP, Kong DF, Huang Y, Pieper KS, Mauri F, Wagner GS, Califf RM, Topol EJ, Granger CB. Extent of ST-segment depression and cardiac events in non-ST-segment elevation acute coronary syndromes. Eur Heart J 2005;26:2106 2113.

Reperfusion by Primary Percutaneous Coronary Intervention in Patients With ST-Segment Elevation Myocardial Infarction Within 12 to 24 Hours of the Onset of Symptoms (from a Prospective National Observational Study [PL-ACS])
Marek Gierlotka, MD, PhDa,*, Mariusz Gasior, MD, PhDa, Krzysztof Wilczek, MD, PhDa, Michal Hawranek, MD, PhDa, Janusz Szkodzinski, MD, PhDa, Piotr Paczek, MD, PhDd, Andrzej Lekston, MD, PhDa, Zbigniew Kalarus, MD, PhDb, Marian Zembala, MD, PhDc, and Lech Polonski, MD, PhDa
The aim of the present study was to investigate whether reperfusion by primary percutaneous coronary intervention (PCI) improves 12-month survival in late presenters with ST-segment elevation myocardial infarction (STEMI). We analyzed 2,036 patients with STEMI presenting 12 to 24 hours from onset of symptoms, without cardiogenic shock or pulmonary edema and not reperfused by thrombolysis, of 23,517 patients with STEMI enrolled in the Polish Registry of Acute Coronary Syndromes from June 2005 to August 2006. An invasive approach was chosen in 910 (44.7%) of late presenters and 92% of them underwent reperfusion by PCI. Patients with an invasive approach had lower mortality after 12 months than patients with a conservative approach (9.3% vs 17.9%, p <0.0001). The benet of an invasive approach was also observed after multivariate adjustment with a relative risk 0.73 for 12-month mortality (95% condence interval 0.56 to 0.96) and in a subpopulation of patients selected by a propensity-score matching procedure with an adjusted relative risk 0.73 for 12-month mortality (0.58 to 0.99). In conclusion, almost 1/2 of late presenters with STEMI were considered eligible for reperfusion by primary PCI. These patients had a lower 12-month mortality rate than they would have had if they had been treated conservatively, which supports the idea of late reperfusion in STEMI. However, whether all late presenters with STEMI should be treated invasively remains unanswered. Nevertheless, until a randomized trial is undertaken, late presenters with STEMI could be considered for reperfusion by primary PCI. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:501508) Optimal management for patients with ST-segment elevation myocardial infarction (STEMI) who arrive at a hospital late remains uncertain. A few recently published studies have suggested that reperfusion by percutaneous coronary intervention (PCI) could be benecial to patients if performed just after 12 hours from onset of symptoms.17 In addition, the most recent European guidelines on STEMI recommend PCI for late presenters with ongoing ischemia.8 The aim of our analysis was to assess the current use of invasive treatment and mechanical reperfusion by primary PCI applied 12 to 24 hours from onset of symptoms in patients with STEMI arriving at a hospital 12 to 24 hours from onset of symptoms and to determine the inuence of
a Third and bFirst Departments of Cardiology and cDepartment of Cardiac Surgery and Transplantology, Silesian Center for Heart Diseases, Medical University of Silesia, Zabrze, Poland; dCardiology Department, Polish Medical Group, Sosnowiec, Poland. Manuscript received July 12, 2010; revised manuscript received and accepted October 7, 2010. The Polish Registry of Acute Coronary Syndromes (PL-ACS) is supported by an unrestricted grant from the Polish Ministry of Health, Warsaw, Poland. *Corresponding author: Tel: 48-32-373-3619; fax: 48-32-273-2679. E-mail address: marek.gierlotka@sccs.pl (M. Gierlotka).

this invasive approach on 12-month mortality in clinical practice. Methods We used data from the Polish Registry of Acute Coronary Syndromes (PL-ACS), for which the method and results of the rst 100,193 patients have been described.9 In brief, the PL-ACS is an ongoing, nationwide, multicenter, prospective, observational study of consecutively hospitalized patients with the entire spectrum of ACSs in Poland. The pilot phase of the registry commenced in October 2003 in the Silesia region. In subsequent months, further regions were opened and, since June 2005, all Polish regions collect data for the PL-ACS. Hospitals were invited to enter the registry if they had a coronary care unit, a cardiology unit, a cardiac surgical unit, an internal medicine unit, or an intensive care unit or if they hospitalized 10 patients with ACS per year. A detailed protocol with inclusion and exclusion criteria, methods and logistics, and denitions of all elds in the registry dataset was prepared before the registry was started. However, it has since been revised 2 times. In May 2004, the protocol was adapted to be compatible with the Cardiwww.ajconline.org

0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.008

502

The American Journal of Cardiology (www.ajconline.org)

Table 1 Baseline clinical characteristics of patients assigned to invasive and conservative approaches Variable Age (years), mean SD Women Smokers Hypertension Diabetes mellitus Obesity (body mass index 30 kg/m2) Previous myocardial infarction Previous coronary bypass Previous percutaneous coronary interventions Cardiac arrest before admission Anterior wall infarct location Heart rate on admission (beats/min), mean SD Other than sinus rhythm Systolic blood pressure (mm Hg), mean SD Killip class II on admission Symptom-onset-to-admission time (hours), median (interquartile range) Hospitalization in hospital with percutaneous coronary intervention capability Conservative Approach (n 1,126) 67.2 13.2 44.0% 29.1% 63.1% 23.9% 17.9% 16.6% 5.3% 1.2% 1.0% 41.3% 81 21 10.0% 139 28 24.0% 16 (1421) 26.1% Invasive Approach (n 910) 63.3 11.8 32.4% 42.5% 63.3% 23.4% 18.2% 11.2% 2.9% 1.0% 1.7% 45.2% 78 17 4.8% 136 27 11.8% 15 (1218) 100% p Value 0.0001 0.0001 0.0001 0.94 0.80 0.82 0.0005 0.0058 0.59 0.18 0.080 0.0001 0.0001 0.032 0.0001 0.0001 0.0001

ology Audit and Registration Data Standards (CARDS).10 Nevertheless, the PL-ACS case-report form (CRF) covers only part of the CARDS dataset. A second revision (May 2005) added new elds to the CRF, which included exact dates and times of onset of symptoms, coronary angiography, and PCI procedures. According to the protocol, all admitted patients with suspected ACS were screened for eligibility to enter the registry but were not enrolled until the ACS was conrmed. Patients were then classied as having unstable angina, non-STEMI, or STEMI. STEMI was dened as the presence of (1) ST-segment elevation consistent with infarction of 2 mm in contiguous chest leads and/or ST-segment elevation of 1 mm in 2 standard leads or new left bundle branch block and (2) positive cardiac necrosis markers. If a patient was hospitalized during the same ACS in 1 hospital (transferred patient), all hospitals were required to complete the registry data. These hospitalizations were linked together during data management and were analyzed as 1 ACS. Data were collected by skilled physicians who were in charge of each patient and entered directly into an electronic CRF or temporarily printed onto a CRF before being transferred to an electronic CRF. Internal checks for missing or conicting data and values markedly out of the expected range were implemented by the software. In the Silesian Center for Heart Diseases (Zabrze, Poland) data management and analysis center, further edit checks were applied if necessary. All-cause mortality data with exact dates of deaths were obtained from ofcial mortality records from the National Health Fund. Vital status at 12 months after STEMI was available for all patients enrolled up to August 2006. For the present analysis, we selected patients with STEMI who were enrolled during a consecutive 15-month period, from the time of the second protocol revision (June 2005) to August 2006. In addition, we excluded patients with severe hemodynamic disturbances on admission such as pulmonary edema (Killip class III) and cardiogenic shock

(Killip class IV) for whom the recommended window for invasive treatment exceeded 12 hours. Furthermore, patients initially treated with thrombolysis were excluded. Symptom-onset-to-admission time was calculated for every patient as an absolute difference in minutes between dates and times of hospital admission and onset of symptoms. Only patients with symptom-onset-to-admission times 12 to 24 hours comprised the study population. Two groups of patients were then identied based on whether coronary angiography had been performed and time of the invasive procedure. Patients were included in the invasive-approach group if they had coronary angiography performed 12 to 24 hours from onset of symptoms. All other patients (treated noninvasively or with coronary angiography performed 24 hours from symptom onset) were included in the conservative-approach group. The main outcome measurement was 12-month all-cause mortality. In-hospital outcomes were death from any reason, recurrent MI (dened as an ischemic event that met European Society of Cardiology/American College of Cardiology criteria for reinfarction and was evidently clinically distinct from the index event at time of admission),11 and stroke (dened as an acute neurologic decit that lasted 24 hours and affected the ability to perform daily activities or resulted in death). Continuous variables are reported as mean SD or median and interquartile range as appropriate. Categorical variables are expressed as percentages. Students t test and, when the assumption of normality was violated, MannWhitney U test were used for comparison of continuous variables. Normality of distribution was checked with the Shapiro-Wilk test. To compare categorized variables, chisquare test was used. Follow-up mortality was analyzed using the Kaplan-Meier method, and differences between groups were compared with log-rank test. Multivariate Cox proportional hazard model regression was performed to adjust the inuence of the invasive approach on 12-month mortality, including all parameters listed in Table 1, except

Coronary Artery Disease/Late Reperfusion by PCI in STEMI

503

Figure 1. Analysis scheme of ST-segment elevation myocardial infarction.

Table 2 In-hospital treatment procedures and pharmacotherapy Variable Conservative Approach (n 1,126) 15.0% Invasive Approach (n 910) 100% p Value

Table 3 In-hospital and 12-month outcomes Variable Conservative Approach (n 1,126) 0.6% 5.2% 6.8% 12.2% 10.0% 14.2% 17.9% Invasive Approach (n 910) 0.7% 2.6% 2.6% 5.7% 5.1% 7.5% 9.3% p Value

Coronary angiography Door-to-angiography time (minutes) 90 91120 120 Primary percutaneous coronary intervention Initial Thrombolysis In Myocardial Infarction grade 2 or 3 ow Final Thrombolysis In Myocardial Infarction grade 2 ow Final Thrombolysis In Myocardial Infarction grade 3 ow Stent implantation Coronary bypass Aspirin Thienopyridines Glycoprotein IIb/IIIa receptor inhibitor Heparin Blocker Calcium antagonist Statin Angiotensin-converting enzyme inhibitor Nitrate Diuretic

0.0001

0.6% 1.2% 98.2% 12.1% 22.6%

86.2% 4.6% 9.2% 91.8% 21.8%

0.0001 0.0001 0.85

Stroke in hospital Recurrent myocardial infarction in hospital Death in hospital Combined in-hospital outcome 30-day mortality 6-month mortality 12-month mortality

0.92 0.0041 0.0001 0.0001 0.0001 0.0001 0.0001

6.7%

7.2%

0.84

90.3%

87.3%

0.33

93.4% 0.3% 95.7% 42.0% 3.6% 76.9% 74.7% 5.0% 74.9% 71.1% 53.0% 31.4%

90.2% 1.2% 97.0% 87.8% 22.0% 69.8% 79.0% 3.7% 86.6% 80.8% 27.6% 19.2%

0.23 0.011 0.10 0.0001 0.0001 0.0003 0.022 0.18 0.0001 0.0001 0.0001 0.0001

for place of hospitalization. Hazard ratios and 95% condence intervals were calculated. Due to differences in baseline characteristics between groups, a propensity-score method was used to identify comparable patients treated with invasive and conservative approaches.12 Multivariate logistic regression analysis, including all baseline characteristic parameters listed in Table 1 except for place of hospitalization, was performed to calculate the predicted probability of receiving an invasive approach (propensity score) for every patient. Patients from the conservative group were then matched to patients from the invasive group. Because of substantial differences in baseline parameters, only 68% of patients remained after the matching procedure. The original multivariate Cox proportional hazard model regression was recalculated in the matched groups to further adjust the inuence of an invasive approach on 12-month mortality. Subgroup analysis to identify patients who would potentially benet from an invasive approach was performed with univar-

504

The American Journal of Cardiology (www.ajconline.org)

Figure 2. Adjusted relative risk of 12-month mortality for an invasive approach by multivariate Cox proportional hazard model regression (n 2,036). CABG coronary artery bypass grafting; CI condence interval.

iate Cox proportional hazard model regression in the matched population. A 2-sided p value 0.05 was considered statistically signicant. For all calculations, STATISTICA 7.1 (StatSoft, Inc., Tulsa, Oklahoma) was used. Results In total 23,517 patients with STEMI were hospitalized in 385 hospitals from June 2005 to August 2006. After excluding patients with pulmonary edema and cardiogenic shock on admission and, subsequently, those whose reperfusion was primarily by thrombolysis, there were 19,453 patients, and symptom-onset-to-admission time was calculated. A prehospital delay from 12 to 24 hours was observed in 2,036 patients (10.5%). Criteria for an invasive approach were fullled by 44.7% of them. The remaining patients were assigned to the conservative-approach group (Figure 1). Differences in clinical characteristics between groups presented in Table 1 show that an invasive approach was chosen more frequently in patients with a more favorable baseline risk prole. These patients were on average 4 years younger, more often men, less frequently had a history of MI and previous bypass surgery, and were less often in Killip class II on admission. About 1/4 of patients from the conservative group were treated in hospitals with PCI capability. Most from the invasive group had coronary angiography done within 90 minutes of time of admission (Table 2). Coronary angiography was also performed in 15% of patients assigned to the conservative approach. In almost all these patients, door-to-angiography time exceeded 2 hours, and symptom-onset-to-angiography time was 24 hours. Consequently, PCI was performed in 12% of patients in

conservative group, whereas 92% of patients in the invasive group received primary PCI. Signicant differences were also observed in pharmacotherapy during hospitalization with the invasive group having greater use of thienopyridines, glycoprotein IIb/IIIa inhibitors, blockers, statins, and angiotensin-converting enzyme-inhibitors and less frequent use of nitrates and diuretics. In-hospital and 12-month outcomes are presented in Table 3. Patients in whom an invasive approach was chosen had approximately 1/2 the mortality of the conservative group after 30 days and 12 months. A similar trend was observed for in-hospital reinfarctions. After multivariate adjustment (Figure 2), an invasive approach remained signicantly associated with lower relative risk of 12-month mortality (hazard ratio 0.73, 95% condence interval 0.56 to 0.96). The lower 12-month mortality of an invasive approach was also observed in the subpopulation of patients selected by the propensity-score matching procedure (Figure 3, Table 4). Of note, however, the guidelines recommended pharmacologic treatment was more widely used in patients who received an invasive approach. In addition, we did not nd any specic subgroup of patients for whom an invasive approach could be potentially harmful (Figure 4). Discussion The main nding of our analysis is that current clinical practice of invasive treatment of selected patients with STEMI with late presentation 12 to 24 hours from onset of symptoms is not harmful and leads to lower 12-month mortality compared to a conservative approach. The benet in mortality, primarily large without adjustment, remained signicant after multivariate analysis and propensity-score

Coronary Artery Disease/Late Reperfusion by PCI in STEMI

505

Figure 3. Kaplan-Meier mortality curves for total 12-month mortality in (A) all patients and (B) matched patients.

matching, which were essential due to signicant differences in clinical characteristics that favored patients selected for invasive treatment. Therefore, our study strongly supports the idea of late reperfusion in patients with STEMI, and to our knowledge, this is the rst report on mortality decrease. Nevertheless, there are several points to be discussed when interpreting and comparing the results to other studies. A comment should be made regarding the design of the study. It is a retrospective analysis of a large prospective registry. The denition of an invasive approach as coronary angiography performed in the period of 12 to 24 hours from onset of symptoms was, in our opinion, the best compromise between accuracy and everyday clinical practice. The intension was to include all patients potentially eligible for

primary PCI, making the analysis as close as possible to a randomized setting. As a result, 15% of patients in the conservative-approach group underwent coronary angiography and, subsequently, most underwent PCI. Because doorto-angiography time exceeded 2 hours in almost all of these procedures, we can probably treat them as planned procedures instead of immediate PCI for reperfusion. Furthermore, we did not know whether there was ongoing ischemia on admission, although in those cases the benet of immediate revascularization is much more endorsed. Reasons as to why each patient was treated invasively are also missing. We can assume that, based on the symptoms and signs, these patients were good candidates for late reperfusion for the attending physician. Consequently, the proportion of patients with ongoing ischemia may have been larger with

506

The American Journal of Cardiology (www.ajconline.org)

Table 4 Comparison of clinical characteristics, treatment, and outcomes of patients with invasive approach and patients with conservative approach matched by propensity-score method (n 1,386) Variable Baseline clinical characteristics Age (years), mean SD Women Smokers Hypertension Diabetes mellitus Obesity Previous myocardial infarction Previous coronary bypass Previous percutaneous coronary intervention Cardiac arrest before admission Anterior wall infarct location Heart rate on admission (beats/min), mean SD Other than sinus rhythm Systolic blood pressure (mm Hg), mean SD Killip class II on admission Symptom-onset-to-admission time (hours), median (interquartile range) Hospitalization in hospital with percutaneous coronary intervention capability Treatment Coronary angiography Primary percutaneous coronary intervention Initial Thrombolysis In Myocardial Infarction grade 2 or 3 ow Final Thrombolysis In Myocardial Infarction grade 2 ow Final Thrombolysis In Myocardial Infarction grade 3 ow Stent implantation Coronary bypass Aspirin Thienopyridines Glycoprotein IIb/IIIa receptor inhibitor Heparin Blocker Calcium antagonist Statin Angiotensin-converting enzyme inhibitor Nitrate Diuretic Outcomes Stroke in hospital Recurrent myocardial infarction in hospital Death in hospital Combined in-hospital outcome 30-day mortality 6-month mortality 12-month mortality Conservative Approach (n 693) 64.6 13.2 36.4% 37.2% 64.2% 22.9% 16.7% 11.8% 4.0% 0.9% 1.3% 41.9% 78 17 5.8% 138 27 13.3% 15 (1319) 27.9% 16.5% 13.7% 19.4% 4.3% 93.6% 94.7% 0.3% 95.4% 43.4% 4.5% 74.9% 75.6% 4.9% 73.0% 69.4% 51.5% 24.4% 0.6% 4.9% 5.1% 10.0% 7.7% 11.4% 14.3% Invasive Approach (n 693) 64.7 11.7 38.2% 37.1% 61.6% 23.7% 16.2% 12.3% 3.8% 1.0% 1.3% 41.7% 79 18 5.9% 138 26 14.3% 15 (1319) 100% 100% 91.3% 21.6% 7.9% 85.8% 89.7% 1.6% 96.7% 87.2% 21.7% 69.7% 77.3% 4.0% 86.2% 79.4% 28.4% 20.6% 0.9% 2.9% 2.9% 6.4% 5.8% 8.7% 10.5% p Value

0.89 0.47 0.96 0.32 0.75 0.77 0.80 0.78 0.78 1.0 0.96 0.72 0.90 0.85 0.59 0.49 0.0001 0.0001 0.0001 0.62 0.21 0.036 0.12 0.012 0.22 0.0001 0.0001 0.031 0.45 0.44 0.0001 0.0001 0.0001 0.095 0.53 0.052 0.039 0.014 0.16 0.089 0.034

an invasive approach, which could have biased the results seriously. Interestingly, in current clinical practice in Poland, as many as 45% of late presenters, in the period of 12 to 24 hours, were considered eligible for reperfusion therapy. Of note, it was before the most recent recommendations were published.8 Consequently, it is not surprising that baseline characteristics differed between groups, with a lower risk prole in the group of patients who received an invasive approach. This skewed risk prole is typical for all observational studies comparing PCI to medical therapy in STEMI, including other studies of late reperfusion.1,2 These 2 reported analyses showed a signicant mortality benet of late reperfusion that was not signicant after adjustment in

multivariate analysis1 or showed a strong trend toward lower mortality after propensity-score matching.2 Also, randomized trials performed thus far have not shown any signicant decrease in the number of major adverse clinical events including mortality.37 In fact, the Beyond 12-hours Reperfusion AlternatiVe Evaluation (BRAVE-2) study is the only randomized study comparing primary PCI to medical treatment in a period of potentially successful reperfusion therapy of 12 to 48 hours.3,4 The investigators showed signicantly smaller nal infarct size and improvement in myocardial salvage, which was a primary end point of the study. The large Occluded Artery Trial (OAT) compared coronary interventions to medical treatment in persistent

Coronary Artery Disease/Late Reperfusion by PCI in STEMI

507

Figure 4. Subgroup analysis of 12-month mortality for conservative and invasive approaches by univariate Cox proportional hazard model regression and multivariate adjustment in matched groups of patients (n 1,386). Multivariate Cox proportional hazard model regression adjusted for age, gender, smoking status, hypertension, diabetes, obesity, previous myocardial infarction, previous coronary artery bypass grafting, previous percutaneous coronary intervention, cardiac arrest before admission, electrocardiographic anterior wall location of infarction, heart rate on admission, other than sinus rhythm on admission, systolic blood pressure on admission, Killip class on admission, and symptom-onset-to-admission time. Abbreviation as in Figure 2.

occluded coronary arteries 3 to 28 days (median 8) after MI, and it reported similar mortality, reinfarction rate, and severe heart failure New York Heart Association class IV after 5 years in the 2 groups.13 However, the median delay of 8 days makes the coronary interventions in this study rather subacute PCI than primary PCI for STEMI. In their recently published study, Busk et al7 reported that substantial myocardial salvage was observed even when the infarctrelated artery was totally occluded at time of primary PCI performed 12 to 72 hours from onset of STEMI. Although nal infarct size after primary PCI was larger in late presenters than in early presenters, the 12-hour limit failed to identify which patients had the potential for myocardial salvage. Together with the results of the BRAVE-2 study, these investigators provide a potential explanation of the results of our study. Other factors that may inuence the effect of an invasive

approach on mortality are signicant differences between groups in the prescribed pharmacotherapy during hospitalization. Patients treated invasively received more complete medical therapy than patients treated conservatively. This nding has also been reported previously2 and cannot be attributed to selection bias, although the differences did not disappear in the propensity-score matched subgroups. A more probable reason is better medical care provided by cardiologists in hospitals with available primary PCI procedures.14 In contrast, this can be an additional potential advantage of choosing a strategy of late reperfusion in patients with STEMI admitted after 12 hours from onset of symptoms. Of note, older age and certain co-morbidities such as diabetes and previous MI were associated with a more evident benet from an invasive approach. Interestingly, in patients with Killip class II on admission, 12month mortality was similar in the invasive and conserva-

508

The American Journal of Cardiology (www.ajconline.org) 3. Schmig A, Mehilli J, Antoniucci D, Ndrepepa G, Markwardt C, Di Pede F, Nekolla SG, Schlotterbeck K, Schhlen H, Pache J, Seyfarth M, Martinoff S, Benzer W, Schmitt C, Dirschinger J, Schwaiger M, Kastrati A. Mechanical reperfusion in patients with acute myocardial infarction presenting more than 12 hours from symptom onset: a randomized controlled trial. JAMA 2005;293:28652872. 4. Parodi G, Ndrepepa G, Kastrati A, Conti A, Mehilli J, Sciagra R, Schwaiger M, Antoniucci D, Schmig A. Ability of mechanical reperfusion to salvage myocardium in patients with acute myocardial infarction presenting beyond 12 hours after onset of symptoms. Am Heart J 2006;152:11331139. 5. Abbate A, Biondi-Zoccai GG, Appleton DL, Erne P, Schoenenberger AW, Lipinski MJ, Agostoni P, Sheiban I, Vetrovec GW. Survival and cardiac remodeling benets in patients undergoing late percutaneous coronary intervention of the infarct-related artery: evidence from a meta-analysis of randomized controlled trials. J Am Coll Cardiol 2008;51:956 964. 6. Silva JC, Rochitte CE, Jnior JS, Tsutsui J, Andrade J, Martinez EE, Moffa PJ, Menegheti JC, Kalil-Filho R, Ramires JF, Nicolau JC. Late coronary artery recanalization effects on left ventricular remodelling and contractility by magnetic resonance imaging. Eur Heart J 2005; 26:36 43. 7. Busk M, Kaltoft A, Nielsen SS, Bttcher M, Rehling M, Thuesen L, Btker HE, Lassen JF, Christiansen EH, Krusell LR, Andersen HR, Nielsen TT, Kristensen SD. Infarct size and myocardial salvage after primary angioplasty in patients presenting with symptoms for 12 h vs. 1272 h. Eur Heart J 2009;30:13221330. 8. Van de Werf F, Bax J, Betriu A, Blomstrom-Lundqvist C, Crea F, Falk V, Filippatos G, Fox K, Huber K, Kastrati A, Rosengren A, Steg PG, Tubaro M, Verheugt F, Weidinger F, Weis M. Management of acute myocardial infarction in patients presenting with persistent ST-segment elevation. Eur Heart J 2008;29:2909 2945. 9. Polonski L, Gasior M, Gierlotka M, Kalarus Z, Cieslin ski A, Dubiel JS, Gil RJ, Ruzyllo W, Trusz-Gluza M, Zembala M, Opolski G. Polish Registry of Acute Coronary Syndromes (PL-ACS). Characteristics treatments and outcomes of patients with acute coronary syndromes in Poland. Kardiol Pol 2007;65:861 872. 10. Flynn MR, Barrett C, Coso FG, Gitt AK, Wallentin L, Kearney P, Lonergan M, Shelley E, Simoons ML. The Cardiology Audit and Registration Data Standards (CARDS), European data standards for clinical cardiology practice. Eur Heart J 2005;26:308 313. 11. The Joint European Society of Cardiology/American College of Cardiology Committee. Myocardial infarction redeneda consensus document of the Joint European Society of Cardiology/American College of Cardiology committee for the redenition of myocardial infarction. Eur Heart J 2000;21:15021513. 12. Rosenbaum PR, Rubin DR. The central role of the propensity score in observational studies for causal effects. Biometrika 1983;70:4155. 13. Hochman JS, Lamas GA, Buller CE, Dzavik V, Reynolds HR, Abramsky SJ, Forman S, Ruzyllo W, Maggioni AP, White H, Sadowski Z, Carvalho AC, Rankin JM, Renkin JP, Steg PG, Mascette AM, Sopko G, Psterer ME, Leor J, Fridrich V, Mark DB, Knatterud GL. Coronary intervention for persistent occlusion after myocardial infarction. N Engl J Med 2006;355:23952407. 14. Casale PN, Jones WL, Wolf FE, Pei Y, Eby LM. Patients treated by cardiologists have a lower in-hospital mortality for acute myocardial infarction. J Am Coll Cardiol 1998;32:885 889. 15. Antman EM, Hand M, Armstrong PW, Bates ER, Green LA, Halasyamani LK, Hochman JS, Krumholz HM, Lamas GA, Mullany CJ, Pearle DL, Sloan MA, Smith SC Jr, Anbe DT, Kushner FG, Ornato JP, Jacobs AK, Adams CD, Anderson JL, Buller CE, Creager MA, Ettinger SM, Halperin JL, Hunt SA, Lytle BW, Nishimura R, Page RL, Riegel B, Tarkington LG, Yancy CW. 2007 Focused update of the ACC/AHA 2004 guidelines for the management of patients with STelevation myocardial infarction. Circulation 2008;117:296 329.

tive groups. This result is somewhat surprising because hemodynamic instability is thought to be 1 of the indications for an invasive approach in this group of patients.15 We do not have an explanation for this nding. There are several limitations to our study, and some were discussed earlier. The PL-ACS is a prospective observational study that includes data from various regions and hospitals throughout Poland. Because patients were treated at different hospitals in different regions, there are inherent uncontrolled differences in clinical practice. In addition, the retrospective nature of our study is a potential weakness. The major limitation is some important measurements such as status of ongoing ischemia on admission and detailed reasons for choosing an invasive or conservative strategy in a particular case, which was discussed earlier. Furthermore, even after propensity-score matching, the groups are likely to be biased by potentially important parameters that are not available in the registry. In addition, all patients from the invasive-approach group were treated in PCI centers compared to only 1/4 of patients from the conservative group. An adjustment for this difference was not possible with the present design of our study. Therefore, the reported significant mortality decrease after adjustment should be interpreted with caution. Acknowledgment: We thank all the physicians and nurses who participated in the PL-ACS Registry, members of the expert committee, regional co-ordinators, and employees of the National Health Fund of Poland for their logistic support. Appendix PL-ACS expert committee: Lech Polonski, MD, PhD (chairman), Mariusz Gasior, MD, PhD (cochairman), Marek Gierlotka, MD, PhD (cochairman), and Zbigniew Kalarus, MD, PhD (cochairman), Zabrze; Andrzej Cieslinski, MD, PhD, Poznan; Jacek Dubiel, MD, PhD, Cracow; Robert Gil, MD, PhD, Grzegorz Opolski, MD, PhD, and Witold Ruzyllo, MD, PhD, Warsaw; Michal Tendera, MD, PhD, Katowice; and Marian Zembala, MD, PhD, Zabrze.
1. Zahn R, Schiele R, Schneider S, Gitt AK, Wienbergen H, Seidl K, Bossaller C, Hauf GF, Gottwik M, Altmann E, Rosahl W, Senges J. Primary angioplasty versus no reperfusion therapy in patients with acute myocardial infarction and a pre-hospital delay of 1224 hours: results from the pooled data of the maximal individual therapy in acute myocardial infarction (MITRA) registry and the myocardial infarction registry (MIR). J Invasive Cardiol 2001;13:367372. 2. Elad Y, French WJ, Shavelle DM, Parsons LS, Sada MJ, Every NR. Primary angioplasty and selection bias in patients presenting late (12 h) after onset of chest pain and ST elevation myocardial infarction. J Am Coll Cardiol 2002;39:826 833.

Percutaneous Coronary Intervention for Non ST-Elevation Acute Coronary Syndromes: Which, When and How?
Robert K. Riezebos, MDa,*, Jan G.P. Tijssen, PhDb, Freek W.A. Verheugt, MD, PhDa, and Gerrit J. Laarman, MD, PhDc
The presentation of patients with suspected non ST-elevation acute coronary syndromes is quite diverse. Therefore, the diagnostic workup and choice of treatment may vary accordingly. Major issues regarding the evaluation are the likelihood of the diagnosis and the risk for adverse events. These factors should guide the choice of diagnostic test. Patients with increased risk for ischemic events and patients with recurrent ischemia are most likely to benet from revascularization. In addition, when percutaneous coronary intervention is considered, evidence suggests that sufcient time should be allowed for pharmacologic stabilization, reducing the possibility of periprocedurally inicted myocardial infarction. However, postponement of intervention may lead to an increase of new spontaneous events, and high-risk patients should apply for revascularization soon after pharmacologic stabilization. The extent of revascularization performed by percutaneous coronary intervention depends predominantly on patient characteristics and anatomy but should be limited to ow-obstructive lesions. In conclusion, patients presenting with nonST elevation acute coronary syndromes constitute a very diverse population; diagnostic workup, treatment, and the timing of a possible intervention should be tailored individually. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:509 515) Patients with chest pain represent a large and increasing proportion of all acute medical presentations worldwide. Of all those presenting for evaluation, only a minority have acute coronary syndromes (ACS). Distinguishing which patients have ACS remains a diagnostic challenge. The principal pathophysiologic mechanism of ACS is myocardial underperfusion, which is caused by atherosclerotic plaque rupture or erosion, with different degrees of superimposed thrombus.1,2 Electrocardiography provides the initial classication. Patients are divided into those with persistent ST-segment elevation and those without persistent ST-segment elevation or non ST-elevation ACS (NSTEACS). In this review, we discusses the diagnostic challenges when NSTEACS are suspected. In addition, we address the role of risk stratication in relation to the choice of treatment strategy. When an invasive approach is preferred, an important issue is the timing of the intervention. The available evidence on this topic is discussed in detail. We conclude with the evidence regarding the type and extent of revascularization in patients with multivessel disease. Diagnostics and Risk Assessment In patients presenting with suspected NSTEACS, 2 major issues must be addressed. The rst challenge is to conrm the diagnosis. Guidelines recommend the use of elementary tools, such as symptoms, risk prole for coronary
a Onze Lieve Vrouwe Gasthuis; bAcademic Medical Center, University of Amsterdam, Amsterdam; and cTweeSteden Ziekenhuis, Tilburg, The Netherlands. Manuscript received August 16, 2010; revised manuscript received and accepted October 5, 2010. *Corresponding author: Tel: 31-20-5993033; fax: 31-20-5994618. E-mail address: r.k.riezebos@xs4all.nl (R.K. Riezebos).

artery disease, electrocardiography, and biomarkers, to estimate the likelihood of disease. In addition, echocardiography in the acute phase can be used to clarify the diagnosis.1,3 However, the diagnosis sometimes remains uncertain. In these cases, the clinical probability of ACS should be assessed.4 Although American College of Cardiology (ACC) and American Heart Association (AHA) as well as European Society of Cardiology (ESC) guidelines do not provide guidance on this topic, the next diagnostic test of choice should depend on the likelihood of disease. In Figure 1, an algorithm is proposed in which the preferred performance of a diagnostic test is related to the estimated probability of NSTEACS. In case of low clinical probability, patients are to be discharged safely, so a diagnostic test should be used with high sensitivity and high negative predictive value. Ischemia testing such as exercise testing with or without an imaging modality is frequently used in the subacute setting. However, such tests are most useful in patients with intermediate probability of ACS. In our opinion, poorly performing tests, such as treadmill or bicycle exercise tests, should be restricted to prognostic purposes only. Despite being not recommended by current ESC guidelines, computed tomographic angiography (CTA) is at present the most accurate noninvasive test to rule out coronary artery disease.1,5 New sophisticated scan protocols, using prospective electrocardiographically gated triggering, substantially reduce radiation exposure (effective dose value approximately 3 mSv), without reducing image quality.6 Extracardiac ndings such as pulmonary tumors, pulmonary embolism, and aortic dissection can also be detected.7 In selected patients with acute chest pain, the diagnostic accuracy of CTA is excellent.8 In addition, this approach is more cost effective and less time-consuming.9
www.ajconline.org

0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.016

510

The American Journal of Cardiology (www.ajconline.org)

Figure 1. Algorithm showing the clinical application of diagnostic tests according to the probability of NSTEACS. ECG electrocardiography.

with suspected intermediate- to high-risk NSTEACS underwent acute coronary angiography.11 The mean TIMI risk score was 3.8. Of these, 78% were diagnosed with NSTEACS, and the remainder had noncardiac chest pain. Of all patients, 55% were treated with PCI, and 10% underwent CABG. The second issue to be addressed in patients with suspected NSTEACS involves risk assessment. Patients with NSTEACS represent a prognostically heterogenous group. Therefore, risk stratication plays a central role in evaluation and management. For this purpose, multiple scoring models have been developed, with the Global Registry of Acute Coronary Events (GRACE) and TIMI risk scores being the most widely used. The 2 models show a strong relation between indicators of the likelihood of NSTEACS and prognosis.12,13 The GRACE risk tool was developed on the basis of data from a large multinational cohort study (GRACE) and validated in subsequent GRACE and Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIb cohorts.12,14 Recently, the GRACE score was prospectively revalidated in a large contemporary cohort.15 The TIMI score was developed using data from the TIMI 11B trial13 and prospectively validated in several cohorts, including that of the Treat Angina With Aggrastat and Determine Cost of Therapy With an Invasive or Conservative Strategy (TACTICS)TIMI 18 trial.16 The GRACE score estimates the risk for death up to 6 months, and the TIMI risk score addresses the 14-day risk for death, recurrent myocardial infarction (MI) or urgent revascularization. This risk estimation, together with individual patient characteristics, should further guide treatment strategy. Indications for Urgent Revascularization A subset of patients with NSTEACS are considered to have such an increased mortality risk that immediate revascularization is recommended.1,2 These include cardiogenic shock, severe left ventricular dysfunction, suspected left main stem disease, recurrent or refractory ischemia at rest despite intensive pharmacologic treatment, mechanical complications such as acute mitral regurgitation, and sustained ventricular tachycardia. This recommendation is based on a single study that suggested better outcomes with revascularization in patients presenting with cardiogenic shock.17 However, most patients can be medically stabilized. These patients should be evaluated for an invasive approach. Routine Invasive Versus Selective Invasive Therapy In the past 2 decades, multiple trials have evaluated different clinical strategies regarding coronary angiography and subsequent revascularization of clinically stabilized patients with NSTEACS. Two general approaches have emerged, the rst being the early invasive or routine invasive strategy, involving routine early coronary angiography followed by revascularization when appropriate. The second is the conservative or selective invasive approach, with initial pharmacologic management and coronary angiography followed by revascularization for recurrent ischemia only. This new ischemia may either be

In case of a high probability of ACS, patients should be admitted to the hospital for clinical follow-up and treatment. In these patients, false-positive results are more likely to occur. Accordingly, the diagnosis of ACS should be waived only on the basis of tests with high sensitivity and specicity, invasive coronary angiography currently being the gold standard. In this population, coronary angiography is able to exclude coronary artery disease reliably. This should be strived for, because even in the presence of electrocardiographic changes and troponin increase, about a fth of the patients suspected of high-risk NSTEACS show no significant lesions on coronary angiography.10,11 These patients generally are at low risk and should be evaluated for alternative pathologies. Because of the absence of validated scoring systems to estimate the probability of NSTEACS in patients with chest pain, there is limited information on the distribution of the eventual diagnoses across the various levels of suspicion of ACS. A small trial by Goldstein et al9 evaluated the use of CTA in about 200 patients with chest pain and low probability of ACS. The mean Thrombolysis In Myocardial Infarction (TIMI) risk score was 1.2. The number of patients diagnosed with NSTEACS was about 10%, and the remainder had noncardiac chest pain. The percentage of patients who underwent percutaneous coronary intervention (PCI) was 4%, and the percentage requiring coronary artery bypass grafting (CABG) was 2%.9 In the Optimal Timing of PCI in Unstable Angina (OPTIMA) trial, about 250 patients

Review/Invasive Strategies for NSTEACS 36/1,226 (2.9%) 39/1,106 (3.5%) 15/596 (2.5%)

511

spontaneous or provoked by noninvasive stress testing.1,2 Currently, AHA and ACC as well as ESC guidelines support routine invasive management in intermediate- to highrisk patients with NSTEACS.1,2 Four large randomized controlled trials have dominated the debate on the routine performance of invasive diagnostics in NSTEACS. The results, unfortunately, were quite diverse (Table 1). In 1999, the Fragmin and Fast Revascularisation During Instability in Coronary Artery Disease (FRISC) II trial showed a signicant reduction in the combined end point of death and MI with the routine invasive approach.18 The observed difference was driven mainly by an excess in MI in the selective invasive group. The TACTICS TIMI 18 trial, published in 2001, showed similar results: a decrease in MI but no signicant mortality benet.16 In 2003, the Randomized Intervention Trial of Unstable Angina (RITA) 3 trial failed to show any benet for death or MI.19 Ultimately, in 2005, the Invasive Versus Conservative Treatment in Unstable Coronary Syndromes (ICTUS) trial was published.20 This study, with optimal medical treatment in both arms, showed an increased MI risk in the routine invasive arm, with no difference in mortality. Interpretation of the study results is difcult because of important differences in method. Foremost, when the studies are compared, there appears to be a marked variation in the intensity of revascularization between study arms (Table 1). The conservative arm of the ICTUS trial20 showed a revascularization rate similar to the routine invasive arm in RITA 3.19 Also, the denition of MI differed between the trials. The low biomarker threshold used in the ICTUS trial20 may partly explain the higher number of MIs in patients requiring PCI. The improved use of anticoagulants, dual-antiplatelet therapy, statins, and angiotensin-converting enzyme inhibitors may also be part of the assumed demise of the routine invasive treatment benet. This is most clear for the use of statins. In the FRISC II18 and TACTICSTIMI 1816 trials, approximately 1/2 the patients received statins at discharge. In RITA 3,19 this had already increased to 70%, whereas the ICTUS trial20 provided high-dose statin treatment to 92% of patients. Although less sharp, the use angiotensin-converting enzyme inhibitors showed similar patterns. On the basis of the ICTUS trial, the current AHA and ACC guidelines acknowledge the option of a selective invasive strategy with aggressive medical treatment.2 It may not be surprising that long-term (5-year) follow-up of the aforementioned trials showed discordant results.21,22 Remarkably, the initial negative RITA 3 trial suggested a marked 5-year benet in the routine invasive group regarding death and MI (odds ratio 0.78, 95% condence interval 0.61 to 0.99, p 0.04).23 Indeed, a recent meta-analysis based on individual 5-year follow-up patient data from FRISC II, RITA 3, and ICTUS showed a reduction in MI using the routine invasive strategy.24 The relation between treatment effect and patient risk has been evaluated in subanalyses of several trials. Regardless of the risk score used, there appeared to be a consistent treatment benet for the invasive approach in high-risk patients compared to low-risk patients. The FRISC II and TACTICSTIMI 18 trials as well as the 5-year follow-up of RITA 3 showed the greatest benet of the routine invasive

Mortality

23/1,207 (1.9%)

41/895 (4.5%) 44/915 (5%) 16% 28% 36% 57% 34/895 (4%)

36/915 (4%)

SI

Primary Trial Outcome

124/1,226 (10.1%)

76/1,106 (6.9%)

37/114 (3.3%)

MI

94/1,207* (7.8%)

53/1,114 (4.8%)

Table 1 Methodologic differences among trials evaluating routine versus selective invasive approach in non ST-elevation acute coronary syndromes

18%

29%

% Revascularization

SI

37%

35%

42%

42%

% Revascularization

RI

77%

64%

79% ICTUS20 * p 0.03; p 0.002; p 0.005. FU follow-up; RI routine invasive; SI selective invasive. 2448 hours

Timing of Catheterization in the Routine Invasive Approach

448 hours

Time Frame Patient Inclusion

FRISC II

TACTICS TIMI 1816 RITA 319

Trial

18

19972001 (1-year FU) 20012003 (1-year FU)

19961998 (6-month FU) 19971999 (6-month FU)

72 hours

7 days

61%

% PCI

54%

40%

% PCI

90/604 (15%)

RI

59/596 (10%)

SI

15/604 (2.5%)

RI

512

The American Journal of Cardiology (www.ajconline.org)

12/203 (5.9%) 57/1,593 (3.6%) 44/73* (60.3%) 16/175 (9.1%)

approach in high-risk patients.16,18,23 This resulted in a wide acceptance of the routine invasive approach in this subpopulation. The clinical application of the aforementioned TIMI and GRACE risk scores has been evaluated extensively. Remarkably, recent data from the GRACE registry suggest the presence of an inverse relation between patient risk and the rate of PCI.25 In daily practice, angiographic ndings and referral practice may more substantially inuence the decision to proceed to PCI than patients risk status. In conclusion, the different outcomes in the large trials evaluating the invasive approach in NSTEACS mainly reect the changes in study protocols and in pharmacologic treatment. For clinical practice, it seems reasonable to consider a liberal selective invasive approach equivalent to a temperate routine invasive approach. The patients with the highest risks for adverse outcomes are thought to derive the greatest benet from invasive evaluation and revascularization. However, because clinical judgment on risk estimation appears to be challenging, the use of systematic and accurate risk stratication methods seems important. Timing of Percutaneous Coronary Intervention In the past few years, several studies have evaluated the inuence of the timing of intervention in patients with NSTEACS. Once again, comparison of data and interpretation of the results are difcult, mainly because of methodologic differences among the studies (Table 2). Current AHA and ACC as well as ESC guidelines do not give specic recommendations on this topic.1,2 The rst study published evaluating the timing of the routine invasive approach was the Intracoronary Stenting With Antithrombotic Regimen Cooling-Off (ISAR-COOL) trial.26 This trial randomized patients with suspected NSTEACS to an early (24 hours after anginal complaints) or a 3- to 5-day deferred invasive diagnostic strategy. Although there was no difference between groups regarding the individual end points, the combined end point of death and MI occurred signicantly less in the early arm compared to the deferred strategy.26 Recently, the Timing of Intervention in Acute Coronary Syndrome (TIMACS)27 and Angioplasty to Blunt the Rise of Troponin in Acute Coronary Syndromes Randomized for an Immediate or Delayed Intervention (ABOARD)28 trials provided important information on the feasibility of a very early invasive diagnostic routine. TIMACS is clearly the largest study performed, with approximately 1,500 patients in its 2 arms. This studys results were negative with regard to its end points. However, a subanalysis of a high-risk population, dened as having GRACE risk scores 140, suggested a benet in the early arm. The ABOARD trial28 evaluated a primary PCI approach for NSTEACS compared to elective catheterization on the next day. The trial failed to show any benet for this approach. In addition, there appeared to be a trend toward more MIs in the early group. The inuence of timing of PCI remains difcult to determine because the aforementioned trials randomized to the timing of coronary angiography, and only a portion of patients were treated with PCI (Table 2). It is likely that the inuence of timing of coronary angiography is less pronounced in patients who are treated conservatively or who

Mortality

Trial Outcome at 1 Month

MI

Table 2 Methodologic differences among trials evaluating the timing of invasive approach in non ST-elevation acute coronary syndromes

Median Time to PCI (hours)

Early Delayed strategy Median Time to catheterization (hours) Median Time to PCI (hours) Median Time to Catheterization (hours) Time Frame Patient Inclusion % PCI % PCI Early strategy

20002002 20032008 20042007 20062008 ISAR COOL TIMACS27 OPTIMA11 ABOARD28

26

2.4 14 2 1.1 * p 0.005. NA not available.

Trial

64% 60% 100% 80%

NA 16 2.5 NA

86 50 1.8 20.5

70% 55% 99% 70%

NA 52 27 NA

21/207 (10.1%) 59/1,438 (4.1%) 26/69 (37.7%) 8/177 (4.5%)

Delayed

3/207 (1.4%) 46/1,593 (2.9%) 0/73 (0%) 5/175 (2.9%)

Early

0/203 (0.2%) 47/1,438 (3.3%) 0/69 (0%) 2/177 (1.1%)

Delayed

Review/Invasive Strategies for NSTEACS

513

interpretation, which suggests a U-shaped curve for timeevent relation. Using potent antiplatelet and anticoagulation therapy, the early hazard is not as pronounced as in the past.30 In the acute setting, PCI is still most likely to counteract plaque passication by intracoronary manipulation, leading to a higher rate of periprocedural MI. It seems reasonable to want to treat patients with PCI after pharmacokinetic onset of the initiated medication to reduce periprocedural inicted MI. Therefore, sufcient time is needed to allow pharmacologic stabilization, but the postponement of intervention may lead to an increase of new spontaneous events. One may expect that patients at high risk for recurrent events benet most from revascularization soon after pharmacologic stabilization.
Figure 2. The relation between timing of the early intervention and the relative risk (RR) for MI against a delayed strategy at 30-day follow-up.

Revascularization Methods Although numerous clinical trials have compared PCI and CABG, few trials have compared PCI and CABG in a selected population of patients with NSTEACS.31 The Angina With Extremely Serious Operative Mortality Evaluation (AWESOME) trial randomized patients with a NSTEACS to PCI using bare-metal stents or CABG.32 Short- and long-term mortality rates were similar, but PCI was associated with an increase in recurrent ischemia and repeat revascularization. The current guidelines recommend CABG for patients with disease of the left main coronary artery, multivessel disease, and impaired left ventricular function.1,2 Contemporary trials show nevertheless that PCI provides an alternative in patients with fewer complex coronary artery disease.33 Although the use of modern stents and scoring systems aids in the feasibility of PCI in highrisk patient groups, it remains associated with a higher rate of repeat procedures. Patients who present with ACS often show multiple coronary lesions, of which 1 is responsible for the symptoms. These so-called culprit lesions can be identied either by angiographic characteristics or by coronary territory. The latter requires the determination of the localization of ischemia. In patients with ST-elevation MIs, multivessel PCI has been associated with an increased rate of ischemic events compared to PCI of the culprit lesion alone.34 In contrast, in stable patients, no differences in events were observed.35 Although there is a lack of prospective data in patients with NSTEACS, a large registry of patients with NSTEACS treated with PCI showed multivessel revascularization to be equivalent compared with PCI of the culprit lesion alone regarding death or MI. In this registry, multivessel PCI was associated with a lower rate of repeat revascularization.36 In case of multivessel approach, fractional ow reserve guidance should be considered while selective intervention limited to ow obstructive lesions results in a decrease in adverse events.37 Future Directions In the past 20 years, the rise of invasive coronary diagnostics, interventions, and pharmacotherapies has revolutionized modern cardiology. Strategies based on different pathophysiologic assumptions such as plaque sealing38 and

eventually undergo CABG. It is clear that a fast invasive diagnostic approach has diagnostic benets and facilitates the logistics of further treatment planning. However, the question remains: should early angiography always be followed by prompt intervention? The proper answer can be obtained only from a randomized study in which patients are randomized between immediate and delayed PCI, as was done in the OPTIMA trial.11 Although the trial was terminated early because of slow patient recruitment, it suggested the presence of an early hazard. After acute coronary angiography in 251 patients admitted with NSTEACS, this trial randomized 142 acute patients eligible for PCI to immediate (0.5 hours) or deferred (24 hours) PCI. Moreover, OPTIMA used only 1 infarct denition and included all MIs in its end point, including evolving MI at randomization. This was done because with very early PCI, periprocedural MI is hard to distinguish from a spontaneously evolving MI that started before PCI. OPTIMA showed that MI was signicantly more common in patients receiving immediate PCI (Table 2).11 This difference was most likely due to an excess of periprocedural infarctions in the immediately treated group. This seems to contradict with a recently published post hoc analysis of the Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) trial, which suggested a better outcome with urgent revascularization.29 Although this study included a large patient sample, the design of the ACUITY trial was not suited to detect the inuence of timing of PCI. These observational studies are extremely liable to indication bias and should therefore be interpreted with the utmost caution. Are there any clues regarding the optimal timing of intervention that can be distilled from the data provided by the 4 trials on this topic? When it is suggested that the inuence of timing of invasive therapy is the most pronounced just after an acute event, it is likely that the timing of initiation of therapy in the early invasive group will be the most important variable. In this case, a time-event relation can be estimated using the relative risk for MI at 30 days for each trial and plotted against the time of admittance to diagnostic catheterization, the latter being at least remotely related to timing of intervention. Figure 2 shows this

514

The American Journal of Cardiology (www.ajconline.org) 7. Johnson KM. Extracardiac ndings on cardiac computed tomography. J Am Coll Cardiol 2010;55:1566 1568. 8. Meijboom WB, van Mieghem CA, Mollet NR, Pugliese F, Weustink AC, van Pelt N, Cademartiri F, Nieman K, Boersma E, de Jaegere P, Krestin GP, de Feyter PJ. 64-slice computed tomography coronary angiography in patients with non ST-elevation acute coronary syndrome. Heart 2007;93:1386e92. 9. Goldstein JA, Gallagher MJ, ONeill WW, Ross MA, ONeil BJ, Raff GL. A randomized controlled trial of multi-slice coronary computed tomography for evaluation of acute chest pain. J Am Coll Cardiol 2007;49:863 871. 10. Roe MT, Harrington RA, Prosper DM, Pieper KS, Bhatt DL, Lincoff AM, Simoons ML, Akkerhuis M, Ohman EM, Kitt MM, Vahanian A, Ruzyllo W, Karsch K, Califf RM, Topol EJ; The Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy (PURSUIT) Trial Investigators. Clinical and therapeutic prole of patients presenting with acute coronary syndromes who do not have signicant coronary artery disease. Circulation 2000;102:1101 1106. 11. Riezebos RK, Ronner E, Ter Bals E, Slagboom T, Smits PC, ten Berg JM, Kiemeneij F, Amoroso G, Patterson MS, Suttorp MJ, Tijssen JG, Laarman GJ; OPTIMA trial. Immediate versus deferred coronary angioplasty in non-ST-elevation acute coronary syndromes. Heart 2009; 95:807 812. 12. Fox KA, Dabbous OH, Goldberg RJ, Pieper KS, Eagle KA, Van de Werf F, Avezum A, Goodman SG, Flather MD, Anderson FA Jr, Granger CB. Prediction of risk of death and myocardial infarction in the six months after presentation with acute coronary syndrome: prospective multinational observational study (GRACE). BMJ 2006;333: 1091. 13. Antman EM, Cohen M, Bernink PJ, McCabe CH, Horacek T, Papuchis G, Mautner B, Corbalan R, Radley D, Braunwald E. The TIMI risk score for unstable angina/non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000;284:835 842. 14. Granger CB, Goldberg RJ, Dabbous O, Pieper KS, Eagle KA, Cannon CP, Van De Werf F, Avezum A, Goodman SG, Flather MD, Fox KA; Global Registry of Acute Coronary Events Investigators. Predictors of hospital mortality in the Global Registry of Acute Coronary Events. Arch Intern Med 2003;163:23452353. 15. Pieper KS, Gore JM, FitzGerald G, Granger CB, Goldberg RJ, Steg G, Eagle KA, Anderson FA, Budaj A, Fox KA; Global Registry of Acute Coronary Events (GRACE) Investigators. Validity of a risk-prediction tool for hospital mortality: the Global Registry of Acute Coronary Events. Am Heart J 2009;157:10971105. 16. Cannon CP, Weintraub WS, Demopoulos LA, Robertson DH, Gormley GJ, Braunwald E. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tiroban. N Engl J Med 2001;344: 1879 1887. 17. Hochman JS, Boland J, Sleeper LA, Porway M, Brinker J, Col J, Jacobs A, Slater J, Miller D, Wasserman H. Current spectrum of cardiogenic shock and effect of early revascularisation on mortality: results of an international registry. Circulation 1995;91:873 881. 18. Fragmin and Fast Revascularisation During Instability in Coronary Artery Disease Investigators. Invasive compared with non-invasive treatment in unstable coronary-artery disease: FRISC II prospective randomised multicentre study. Lancet 1999;354;708 715. 19. Fox KA, Poole-Wilson PA, Henderson RA, Clayton TC, Chamberlain DA, Shaw TR, Wheatley DJ, Pocock SJ; Randomized Intervention Trial of Unstable Angina Investigators. Interventional versus conservative treatment for patients with unstable angina or non-ST-elevation myocardial infarction: the British Heart Foundation RITA 3 randomised trial. Lancet 2002;360:743751. 20. de Winter RJ, Windhausen F, Cornel JH, Dunselman PH, Janus CL, Bendermacher PE, Michels HR, Sanders GT, Tijssen JG, Verheugt FW; Invasive Versus Conservative Treatment in Unstable Coronary Syndromes (ICTUS) Investigators. Early invasive versus selectively invasive management for acute coronary syndromes. N Engl J Med 2005;353:10951104. 21. Hirsch A, Windhausen F, Tijssen JG, Verheugt FW, Cornel JH, de Winter RJ; Invasive Versus Conservative Treatment in Unstable Coronary Syndromes (ICTUS) Investigators. Long-term outcome after an early invasive versus selective invasive treatment strategy in patients with

Figure 3. The use of CTA in the initial evaluation of NSTEACS. CTA was performed in a 48-year-old man presenting with chest pain. There was an intermediate probability of ACS. Multiplanar reconstruction of the left anterior descending coronary artery (LAD) showed a moderately severe mixed stenosis in the proximal LAD with evidence of superimposed thrombus (white arrow).

primary PCI11,28 have been considered. Undeniably, coronary revascularization has played a dynamic role. Future research should focus on better identication of those patients with high risk for recurrent unstable disease. Plaque composition and morphology using CTA or optical coherence tomography are being evaluated as promising new techniques.39 41 There is increasing evidence that the use of CTA in patients with suspected NSTEACS can provide important information on the pathophysiology of the acute event by recognizing the vulnerable plaque40 (Figure 3). When an early invasive strategy is preferred, optical coherence tomography is able to identify underlying plaque morphology and detect thrombi of different stages of organization.41 How this new insight will inuence clinical decision making and whether this will alter the choice of therapy will be the subject of debate in the coming years.
1. ESC guidelines on the diagnosis and treatment of non ST-segment elevation acute coronary syndromes. Eur Heart J 2007;28:1598 1660. 2. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2007;116:e148 e304. 3. ACC/AHA/ASE 2003 guideline update for the clinical application of echocardiography: summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASE Committee to Update the 1997 Guidelines for the Clinical Application of Echocardiography). Circulation 2003;108:1146 1162. 4. Scirica BM. Acute coronary syndrome: emerging tools for diagnosis and risk assessment. J Am Coll Cardiol 2010;55:14031415. 5. Schuetz GM, Zacharopoulou NM, Schlattmann P, Dewey M. Metaanalysis: Noninvasive coronary angiography using computed tomography versus magnetic resonance imaging. Ann Intern Med 2010;152: 167177. 6. Consensus document on coronary computed tomographic angiography ACCF/ACR/AHA/NASCI/SAIP/SCAI/SCCT 2010 expert consensus. J Am Coll Cardiol 2010;55:26632699.

Review/Invasive Strategies for NSTEACS non-ST-elevation acute coronary syndrome and elevated cardiac troponin T (the ICTUS trial): a follow-up study. Lancet 2007;369:827 835. Damman P, Hirsch A, Windhausen F, Tijssen JG, de Winter RJ; ICTUS Investigators. A randomised comparison of an early invasive versus selective invasive management in patients with non-ST-elevation acute coronary syndrome. J Am Coll Cardiol 2010;55:858 864. Fox KAA, Poole-Wilson P, Clayton TC. 5-year outcome of an interventional strategy in non-ST elevation acute coronary syndrome: the British Heart Foundation RITA 3 randomised trial. Lancet 2005;366: 914 920. Fox KA, Clayton TC, Damman P, Pocock SJ, de Winter RJ, Tijssen JG, Lagerqvist B, Wallentin L; FIR Collaboration. Long-term outcome of a routine versus selective invasive strategy in patients with nonST-segment elevation acute coronary syndrome a meta-analysis of individual patient data. J Am Coll Cardiol 2010;55:24352445. Fox KA, Anderson FA Jr, Dabbous OH, Steg PG, Lpez-Sendn J, Van de Werf F, Budaj A, Gurnkel EP, Goodman SG, Brieger D; GRACE Investigators. Intervention in acute coronary syndromes: do patients undergo intervention on the basis of their risk characteristics? The Global Registry of Acute Coronary Events (GRACE). Heart 2007;93:177182. Neumann FJ, Kastrati A, Pogatsa-Murray G, Mehilli J, Bollwein H, Bestehorn HP, Schmitt C, Seyfarth M, Dirschinger J, Schmig A. Evaluation of prolonged antithrombotic pretreatment (cooling-off strategy) before intervention in patients with unstable coronary syndromes: a randomized controlled trial. JAMA 2003;290:15931599. Mehta SR, Granger CB, Boden WE, Steg PG, Bassand JP, Faxon DP, Afzal R, Chrolavicius S, Jolly SS, Widimsky P, Avezum A, Rupprecht HJ, Zhu J, Col J, Natarajan MK, Horsman C, Fox KA, Yusuf S; TIMACS Investigators. Early versus delayed intervention in acute coronary syndromes. N Engl J Med 2009;360:21652175. Montalescot G, Cayla G, Collet JP, Elhadad S, Beygui F, Le Breton H, Choussat R, Leclercq F, Silvain J, Duclos F, Aout M, Dubois-Rand JL, Barthlmy O, Ducrocq G, Bellemain-Appaix A, Payot L, Steg PG, Henry P, Spaulding C, Vicaut E; ABOARD Investigators. Immediate vs delayed intervention for acute coronary syndromes: a randomised clinical trial. JAMA 2009;302:947954. Sorajja P, Gersh BJ, Cox DA, McLaughlin MG, Zimetbaum P, Costantini C, Stuckey T, Tcheng JE, Mehran R, Lansky AJ, Grines CL, Stone GW. Impact of delay to angioplasty in patients with acute coronary syndromes undergoing invasive management. J Am Coll Cardiol 2010; 55:1416 1424. de Feyter PJ, Suryapranata H, Serruys PW, Beatt K, van Domburg R, van den Brand M, Tijssen JJ, Azar AJ, Hugenholtz PG. Coronary angioplasty for unstable angina: immediate and late results in 200 consecutive patients with identication of risk factors for unfavorable early and late outcome. J Am Coll Cardiol 1988;12:324 333. Bravata DM, Gienger AL, McDonald KM, Sundaram V, Perez MV, Varghese R, Kapoor JR, Ardehali R, Owens DK, Hlatky MA. Systematic review: the comparative effectiveness of percutaneous coro-

515

22.

32.

23.

24.

33.

25.

34.

26.

35.

27.

36.

28.

37.

29.

38. 39.

30.

40.

31.

41.

nary interventions and coronary artery bypass graft surgery. Ann Intern Med 2007;147:703716. Morrison DA, Sethi G, Sacks J, Henderson W, Grover F, Sedlis S, Esposito R, Ramanathan K, Weiman D, Sucedo J, Antakli T, Paramesh V, Pett S, Vernon S, Birjiniuk V, Welt F, Krucoff M, Wolfe W, Lucke J, Mediratta S, Booth D, Barbiere C, Lewis D. Percutaneous coronary intervention versus coronary artery bypass graft surgery for patients with medically refractory myocardial ischaemia and risk factors for adverse outcomes with bypass: a multicenter, randomised trial. Investigators of the Department of Veterans Affairs Cooperative Study #385, the Angina With Extremely Serious Operative Mortality Evaluation (AWESOME). J Am Coll Cardiol 2001;38:143149. Serruys PW, Morice MC, Kappetein AP, Colombo A, Holmes DR, Mack MJ, Sthle E, Feldman TE, van den Brand M, Bass EJ, Van Dyck N, Leadley K, Dawkins KD, Mohr FW. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med 2009;360:961972. Corpus RA, House JA, Marso SP, Grantham JA, Huber KC Jr, Laster SB, Johnson WL, Daniels WC, Barth CW, Giorgi LV, Rutherford BD. Multivessel percutaneous intervention in patients with multivessel disease and acute myocardial infarction. Am Heart J 2004;148:493 500. Ijsselmuiden AJ, Ezechiels J, Westendorp IC, Tijssen JG, Kiemeneij F, Slagboom T, van der Wieken R, Tangelder G, Serruys PW, Laarman G. Complete versus culprit vessel percutaneous coronary intervention in multivessel disease: a randomised comparison. Am Heart J 2004; 148:467 474. Shishehbor MH, Lauer MS, Singh IM, Chew DP, Karha J, Brener SJ, Moliterno DJ, Ellis SG, Topol EJ, Bhatt DL. In unstable angina or non-ST-segment acute coronary syndrome, should patients with multivessel coronary artery disease undergo multivessel or culprit-only stenting? J Am Coll Cardiol 2007;49:849 854. Tonino PA, De Bruyne B, Pijls NH, Siebert U, Ikeno F, van t Veer M, Klauss V, Manoharan G, Engstrm T, Oldroyd KG, Ver Lee PN, MacCarthy PA, Fearon WF; FAME Study Investigators. Fractional ow reserve versus angiography for guiding percutaneous coronary intervention. N Engl J Med 2009;360:213224. Meier B. Plaque sealing by coronary angioplasty. Heart 2004;90: 13951398. van Velzen JE, Schuijf JD, de Graaf FR, Nucifora G, Pundziute G, Jukema JW, Schalij MJ, Kroft LJ, de Roos A, Reiber JH, van der Wall EE, Bax JJ. Plaque type and composition as evaluated non-invasively by MSCT angiography and invasively by VH IVUS in relation to the degree of stenosis. Heart 2009;95:1990 1996. Russo V, Zavalloni A, Bacchi Reggiani ML, Buttazzi K, Gostoli V, Bartolini S, Fattori R. Incremental prognostic value of coronary CT angiography in patients with suspected coronary artery disease. Circ Cardiovasc Imaging 2010;3:351359. Toutouzas K, Karanasos A, Stefanadis C. Multiple plaque morphologies assessed by optical coherence tomography in a patient with acute coronary syndrome. Heart 2010;96:13351336.

Long-Term Follow-Up of Patients With First-Time Chest Pain Having 64-Slice Computed Tomography
Fabiola B. Sozzi, MD, PhDa,b,*, Filippo Civaia, MDa, Philippe Rossi, MDa, Jean-Francois Robillon, MDa, Stephane Rusek, MSca, Frederic Berthier, MSca, Francois Bourlon, MDa, Laura Iacuzio, MDa, Gilles Dreyfus, MDa, and Vincent Dor, MDa
A paucity of data on outcome of coronary multislice computed tomography (CT) is available. The aim of this study was to assess the long-term follow-up of 64-slice CT in a homogenous patient group. In total 222 patients (136 men, 61%, 59 11 years of age) with chest pain at intermediate risk of coronary artery disease (CAD) and no previous CAD underwent 64-slice CT. Coronary lesions were considered signicant or not based on a threshold of 50% luminal narrowing. Plaques were classied as calcied, noncalcied, and mixed based on type. End point during follow-up was major adverse cardiac events (nonfatal myocardial infarction, unstable angina requiring hospitalization, myocardial revascularization). Coronary plaques were detected in 162 patients (73%). Coronary artery stenosis was signicant in 62 patients. Normal arteries were found in 59 patients (27%). During a mean follow-up of 5 0.5 years, 30 cardiac events occurred. Annualized event rates were 0% in patients with normal coronary arteries, 1.2% in patients with nonsignificant stenosis, and 4.2% in patients with signicant stenosis (p <0.01). Predictors of cardiac events were presence of signicant stenosis, proximal stenosis, and multivessel disease. Noncalcied and mixed plaques had the worse prognosis (p <0.05). In conclusion, 64-CT provides long-term incremental value in patients at intermediate risk of CAD. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:516 521) Coronary multislice computed tomography (CT) is increasingly being used as a tool for noninvasive visualization of coronary arteries.1 The technique provides information on atherosclerotic plaque burden and to some extent on plaque composition.2 4 Accuracy of coronary CT needs to be assessed in management outcome studies in which diagnostic and therapeutic strategies would be decided based on CT alone, without reference to any coronary angiographic results. Therefore, the purpose of the present study was to explore extent, degree, and structure/function of coronary atherosclerosis by 64-CT in a homogenous patient population with chest pain, intermediate probability of coronary artery disease (CAD), and no previous cardiac events and to analyze the impact of these variables on long-term follow-up. Methods The study group was composed of 222 consecutive patients (136 men, 61%, mean age 59 11 years) who underwent 64-CT at the Cardiothoracic Centre of Monaco, Monte Carlo from January to October 2005. All patients included in the study presented with chest pain suspicious for angina. A subgroup of patients had previous equivocal stress test results. Eligibility criteria for this study were suspected but no previously known CAD and intermediate pretest likelihood of CAD (score 9 to 15 points) according
Monaco Cardiothoracic Centre, Monte Carlo, Monaco; bIRCCS Fond Policlinico, Milan, Italy. Manuscript received July 26, 2010; revised manuscript received and accepted October 5, 2010. *Corresponding author: Tel: 39-329-566-2258; fax: 39-023-652-2640. E-mail address: fabiola_sozzi@yahoo.it (F. Sozzi). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.006
a

to a modication of a method by Diamond and Forrester5 as published by Morise et al.6,7 Exclusion criteria for recruitment were previous documented CAD, history of percutaneous transluminal coronary angioplasty or coronary artery bypass grafting, nonchest pain indication for CT, and low and high pretest likelihoods of CAD. A structured interview and clinical history were obtained, and the following cardiac risk factors were assessed before CT: (1) hypertension (dened as blood pressure 140/90 mm Hg or use of antihypertensive agents), (2) hyperlipidemia as dened by lowdensity lipoprotein cholesterol 140 mg/dl, (3) diabetes mellitus (dened as fasting glucose level 120 mg/dl or need for insulin or oral antidiabetic medicines), (4) smoking (current or previous habit), (5) body mass index, (6) family history of CAD in rst-degree relatives, and (7) medication use. All patients gave written informed consent to the study protocol, which was approved by the local ethics committee. All scans were performed with a 64-slice computed tomographic scanner that features a temporal resolution time of 165 ms and a spatial resolution of 0.4 mm3 (Siemens Somatom Sensation 64 Cardiac, Siemens, Forchheim, Germany). If heart rate was 65 beats/min, additional blockers (metoprolol 5 mg intravenously to a maximum dose of 10 mg) were provided. Nitroglycerin (0.4 mg sublingually) was used in all studies unless contraindicated. Patients with known allergy to iodine, signicant arrhythmia (atrial brillation and frequent premature beats) or rapid heart rate (90 beats/min), impaired renal function (serum creatinine 1.3 mg/dl), and contraindications to blockade were not imaged. The following parameters were applied for CT: collimation of 64 0.6 mm, tube rotation time of 330 ms, and tube current of 450 mA at 120 kV. Nonionic contrast mawww.ajconline.org

Coronary Artery Disease/Risk Stratication With 64-Computed Tomography Table 1 Patient characteristics (n 222 patients) Variable Age (years) Men Hypertension Diabetes mellitus Hyperlipidemia* Smoker Body mass index 30 kg/m2 Total 59.2 10.8 136 (61%) 91 (41%) 38 (17%) 78 (35%) 49 (22%) 51 (23%) Normal Arteries (n 59) 53.1 11.2 33 (55%) 16 (27%) 6 (10%) 18 (30%) 6 (10%) 9 (15%) Nonsignicant Stenosis (n 101) 60.6 9.8 62 (61%) 42 (42%) 19 (19%) 36 (36%) 27 (27%) 25 (25%) Signicant Stenosis (n 62) 62.7 9.7 41 (67%) 33 (53%) 13 (21%) 24 (39%) 16 (26%) 17 (27%)

517

p Value 0.001 0.39 0.016 0.27 0.58 0.056 0.32

* Low-density lipoprotein cholesterol 140 mg/dl.

terial was administered in the antecubital vein at 80 to 105 ml, depending on total scan time, and a ow rate of 5 ml/s (Iodixanol 320, GE Healthcare SA, Velizy-Villacoublay, France and Iomeron 400, Bracco Imaging, Courcouronnes, Evry cedex, France) followed by a saline ush of 50 ml at a ow rate of 5 ml/s. The bolus-tracking technique was used to synchronize the arrival of contrast in coronary arteries and initiation of the scan. Automated detection of peak increase in the aortic root was used for timing of the scan. All images were acquired during an inspiratory breath-hold of approximately 10 seconds, with simultaneous registration of a patients electrocardiogram. To evaluate presence of coronary artery plaques, reconstructions in diastole (typically 75% of cardiac cycle) were generated with a slice thickness of 0.75 mm at an increment of 0.6 mm. If motion artifacts were present, additional reconstructions were made at different time points of the RR interval. Axial datasets were transferred to a remote workstation (Syngo, Siemens, Berlin, Germany) for postprocessing and subsequent evaluation. Radiation dose-decreasing techniques, i.e., dose modulation along the z-axis and pulsing algorithm along the RR interval, were employed for all scans. All studies were analyzed by 2 experienced readers blinded to all clinical variables, history, and patient demographics. Coronary arteries were divided into 17 segments according to the modied American Heart Association classication.8 Only segments with a diameter 1.5 mm were included. Each segment was classied as interpretable or not. Patients were excluded from analysis for (1) an uninterpretable proximal or midsegment or (2) 3 uninterpretable segments in general. Then, interpretable segments were evaluated for presence of any atherosclerotic plaque using axial images and multiplanar reconstructions. After visual inspection of volume-rendered images, which depicted the gross coronary artery luminal conguration, coronary artery plaques were carefully inspected on axial images, curved multiplanar reformatted images, and cross-sectional multiplanar reformatted images. Coronary plaques were dened as structures 1 mm2 within and/or adjacent to the coronary artery lumen, which could be clearly distinguished from the vessel lumen and the surrounding pericardial tissue, epicardial fat, or the vessel lumen itself.9 One coronary plaque was assigned per coronary segment. An assessment of plaque composition was also allowed. Differentiation was made among noncalcied plaques (composed exclusively of material having density 130 HU), calcied plaques (composed exclusively of high-density material 130 HU), and

Table 2 Computed tomographic ndings of study population (n 222) Normal coronary circulation Coronary stenosis 50% Coronary stenosis 50% Plaques in left main/proximal left anterior descending coronary artery Plaques in all left anterior descending coronary artery Plaques in left circumex coronary artery Plaques in right coronary artery Number of coronary arteries narrowed 50% 1 2 3 Total number of segments with plaques Total plaque score Calcied plaque (signicant or not) Mixed plaque (signicant or not) Noncalcied plaque (signicant or not) 59 (27%) 101 (45%) 62 (28%) 67 (9%) 280 (39%) 172 (24%) 199 (28%) 43 (19.4%) 21 (9.5%) 21 (9.5%) 718 (19.2%) 3.2 374 (52%) 273 (33%) 109 (15%)

mixed plaques (having components of noncalcied and calcied material). Atherosclerotic lesion was deemed significant if diameter stenosis was 50%. Lesions below this threshold were considered nonsignicant or mild. For each patient, number of diseased coronary segments, number of segments with signicant stenosis, and number of each type of plaque was calculated. Computed tomograms without coronary lesions were considered normal; computed tomograms showing coronary wall irregularities or 1 coronary plaque (signicant or not) were dened as abnormal. Signicant coronary stenosis were further classied as localized in 1 epicardial artery or 2 or 3 epicardial arteries (left anterior descending, left circumex, right coronary arteries) and signicant plaques in the left main and/or proximal left anterior descending coronary artery. Follow-up information was obtained by clinical visits, telephone contact, or questionnaires sent by mail. All reported events were veried by hospital records or direct contacts with the attending physician. The following clinical events were recorded: (1) cardiac death (including death without denitive cause), (2) nonfatal acute myocardial infarction (AMI), (3) unstable angina pectoris requiring hospitalization, and (4) coronary revascularization (by coronary angioplasty or bypass). Coronary revascularization occurring soon after CT was performed as a consequence of its result. Therefore, patients undergoing coronary revascularization procedures sooner than 90 days after CT were excluded

518

The American Journal of Cardiology (www.ajconline.org)

Figure 1. Kaplan-Meier curves for survival free of hard cardiac events in patients with normal arteries and any type of stenosis.

from further analysis. The end point of this study was major adverse cardiac events including nonfatal AMI, unstable angina pectoris requiring hospitalization, and revascularization. Denitions of AMI and unstable angina were previously described.10,11 For multiple events in a given patient, the rst was the event included in the analysis. All patients were followed for a mean period of 5.0 0.5 years (maximum follow-up period 5.5 years). Analyses were performed with SAS 9.1.3 (SAS Institute, Cary, North Carolina). All continuous data are presented as mean SD, and all categorical data are reported as percentage or absolute number. A p value 0.05 was considered statistically signicant. In univariate analysis, unpaired Students t tests or chi-square tests were used to assess differences between groups. Kaplan-Meier survival curves were constructed for computed tomographically diagnosed CAD and were compared with log-rank test. Effect of CAD on hard cardiac events at CT was determined using Cox proportional hazard model. After adjustment for all baseline clinical characteristics, a forward stepwise model was used to determine independent predictors of coronary atherosclerotic variables on computed tomogram (p 0.05). Hazard ratios and condence intervals were calculated. Results Of the 246 patients included in the study, 13 were excluded from analysis because of poor image quality related to cardiac motion artifact and respiratory motion artifact; 11 patients declined to participate in the follow-up study and 30 patients were lost during 5-year follow-up. Replies were obtained from 192 patients (follow-up rate 87%). All pa-

tients presented with chest pain suspicious for angina. A smaller subset of patients (83, 37%) initially underwent stress testing with equivocal ndings and continued symptoms that warranted further evaluation. A complete overview of baseline characteristics of the entire study population is presented in Table 1. Based on coronary computed tomographic results, the study population was divided into 3 subgroups: patients with normal arteries (59, 27%), patients with nonsignicant stenosis (101, 45%), and patients with signicant stenosis (62, 28%). A correlation between clinical risk factors and occurrence and grade of stenosis is presented in Table 1. As presented, signicant stenosis occurred more frequently in patients who were older, had hypertension, and a smoking habit (p 0.05). Computed tomographic characteristics are listed in Table 2. After exclusion of 45 inaccessible segments (1.2%) because of motion artifacts, plaque burden was evaluated in 3,729 segments. According to plaque texture, 28 (45%) were from patients with obstructive calcied plaques, 16 (26%) were from those with obstructive mixed plaques, and 14 (23%) were from those with obstructive noncalcied plaques. A combination of different signicant plaques (calcied and/or noncalcied and/or mixed plaques) was found in 4 patients (6%). Patients with signicant calcied plaques were on average signicantly older then patients with signicant noncalcied and mixed plaques (63 8.2 vs 61 7.1, p 0.01) and more frequently affected by hypertension (18 vs 10, p 0.05). Prognostic analysis was performed in 192 patients. In these patients, 30 major cardiac events occurred (total event rate 15.6%). Nonfatal AMI affected 2 patients (1%) and

Coronary Artery Disease/Risk Stratication With 64-Computed Tomography

519

Figure 2. Kaplan-Meier curves for survival free of hard cardiac events in patients with nonsignicant stenosis and signicant stenosis in 1 vessel and 2 and 3 vessels.

Table 3 Predictors of cardiac events by univariate Cox proportional hazards regression analysis Variables Age (per year) Men Hypertension Diabetes mellitus Hyperlipidemia Smoker Obesity 30 kg/m2 Proximal versus distal plaques Obstructive plaques in left anterior descending artery Obstructive plaques in left circumex artery Obstructive plaques in right coronary artery Obstructive plaques (per segment) 1 segments with signicant stenosis 2 segments with signicant stenosis 2 segments with signicant stenosis Noncalcied plaque Calcied plaque Mixed plaques HR 1.028 1.36 1.87 1.38 1.16 1.81 1.77 3.5 5.74 3.52 1.73 1.46 3.72 8.43 10.47 4.185 2.699 5.04 95% CI 0.981.17 0.523.53 0.734.73 0.454.18 0.453.00 0.684.81 0.654.78 0.9213.2 1.6719.72 1.438.66 0.684.40 1.101.94 0.3441.07 0.9475.42 1.2687.02 1.213.9 0.89.06 1.6215.7 p Value 0.05 0.53 0.19 0.58 0.76 0.24 0.26 0.005 0.006 0.006 0.25 0.009 0.28 0.057 0.03 0.04 0.10 0.005

Table 4 Predictors of events by multivariate Cox proportional hazards regression analysis* Coronary Risk Factor Obstructive plaque Obstructive plaque in left anterior descending coronary artery Obstructive plaque in circumex coronary artery Obstructive plaque in right coronary artery Proximal obstructive plaque Obstructive plaque (per segment) 1 segment with signicant stenosis 2 segments with signicant stenosis 2 segments with signicant stenosis Mixed plaques Noncalcied plaque HR 8.28 5.98 4.84 1.66 3.73 1.69 3.29 17.44 14.13 6.04 5.1 95% CI 2.4827.64 1.2927.61 1.5015.63 0.515.36 0.9614.49 1.132.54 0.1669.71 1.17261.1 1.23162.9 1.3218.7 1.0516.8 p Value 0.0006 0.022 0.008 0.40 0.05 0.011 0.44 0.038 0.034 0.01 0.049

Abbreviations as in Table 3. * Adjusted for age, gender, hypertension, diabetes mellitus, hyperlipidemia, smoking, family history of coronary disease, and obesity.

CI condence interval; HR hazard ratio.

unstable angina 14 patients (7.3%); coronary revascularization was performed in 14 patients (7.3%, 11 with angioplasty and 3 with bypass). No cases of cardiac deaths were recorded. Annual cardiac event rates were 4.2% in patients with signicant stenosis, 1.2% in patients with nonsignicant stenosis, and 0% in patients with normal arteries on 64-slice computed tomogram (Figures 1 and 2). Patients

520

The American Journal of Cardiology (www.ajconline.org)

Figure 3. Kaplan-Meier curves for survival free of hard cardiac events in patients with noncalcied, calcied, and mixed plaques.

with cardiac events were signicantly older. Univariate analysis is presented in Table 3. Predictors of all cardiac events are presented in Table 4. Any obstructive plaque, multivessel distribution, proximal stenosis, and noncalcied and mixed plaques had the worse prognosis (Figures 1 to 3). Discussion To our knowledge, this is the longest follow-up study in the largest homogenous cohort of patients at intermediate risk with no previous CAD who underwent CT to investigate chest pain. Patients with cardiac events had more extensive atherosclerosis on computed tomogram as reected by a large number of segments showing signicant plaques. Cardiac event rate of patients with normal arteries on 64slice computed tomogram was 0%, highlighting an excellent negative predictive value of normal arteries on 64-slice computed tomogram on long-term analysis. These patients may indeed be reassured without need for further testing. Higher event risk was associated with proximal disease severity, particularly within the proximal portion of the left anterior descending and left circumex coronary arteries. Risk of events was considerably higher in patients with signicant stenosis, although patients with nonsignicant stenosis still showed higher risk compared to patients without CAD on 64-slice computed tomogram. Figure 2 shows that after 3-year follow-up patients with nonsignicant stenosis had a similar cardiac event rate compared to patients with signicant 1-vessel stenosis (overlapping of the 2 survival curves after 3-year follow-up). This preliminary observation needs further investigation with larger population studies. A main nding of our study is that plaque compo-

sition represents a long-term predictor of cardiac events. As shown in Figure 3, noncalcied and mixed plaques carried a worse prognosis compared to calcied plaques. Referring to the plaque vulnerability concept, Mann et al12 studied 31 subjects who died suddenly of CAD. They found that lipid core size and minimal cup thickness, 2 major determinants of plaque vulnerability, were not related to absolute plaque size or degree of stenosis. Accordingly, atherosclerotic plaque growth and destabilization are highly variable. Many serial angiographic studies have demonstrated that most AMIs occur from occlusion of coronary arteries that did not previously contain signicant stenosis; furthermore, the coronary artery with the most severe stenosis is usually not the culprit artery.13,14 Thus, plaque progression and clinical outcome are not always closely related, and each is poorly predicted on clinical and angiographic grounds because most plaques that underlie an AMI are 70% stenosed.15 In our study patients with any type of plaque (signicant and nonsignicant) had a worse prognosis compared to patients with normal vessels (p 0.05). In addition, noncalcied and mixed plaques on 64-slice computed tomogram represented an independent predictor of cardiac events. Of interest, these 2 types of plaque composition may represent less advanced and possibly less stabilized atherosclerosis compared to dense calcied lesions. However, further investigations are clearly needed to support these observations. Pundziute et al16 in 2007 published the rst follow-up study of CT in 100 patients with suspected or known CAD, 55 of whom were studied with 16-slice CT. They demonstrated a signicant prognostic value for cardiac events and a very good prognosis for patients without obstructive CAD. Car-

Coronary Artery Disease/Risk Stratication With 64-Computed Tomography

521

rigan et al17 analyzed a group of 227 patients with various estimated pretest probabilities of CAD (low, intermediate, and high). Absence of obstructive CAD was associated with a 99% freedom from cardiac death, AMI, and revascularization during an average of 2.3 years of follow-up. Similar ndings were reported by Hadamitzky et al18 and van Werkhoven et al19 in follow-up studies of 18 months and 621 days, respectively. To date, the prognostic performance of CT has mostly been tested in symptomatic mixed populations of different estimated pretest probabilities of CAD. The prognostic value of CT is strongly dependent on the pretest risk prole. Meijboom et al20 showed that the clinical value of CT is higher in patients with an intermediate pretest probability of CAD. In this study we selected this category of patients. The present study is based on an analysis of a relatively small population. In general, CT was associated with increased radiation exposure, although radiation doses are rapidly decreasing with newer acquisition protocols. Also, no dedicated algorithms that allow quantication of plaque stenosis or volume are available for CT. Diabetes, hyperlipidemia, and hypertension were simply categorized as present or not. Classications could be based on disease severity in future studies. Incidence of incomplete follow-up was rather high. The number of events in this population was quite small compared to the number of variables used for adjustment in the multivariable model. In addition, soft end points were emerging compared to the others. This can be explained by the fact that the population was selected for not having had any previous CAD; in additiona, the effect of medical therapy can change the prognosis.
1. Kuettner A, Kopp AF, Schroeder S, Rieger T, Brunn J, Meisner C, Heuschmid M, Trabold T, Burgstahler C, Martensen J, Schoebel W, Selbmann HK, Claussen CD. Diagnostic accuracy of multidetector computed tomography coronary angiography in patients with angiographically proven coronary artery disease. J Am Coll Cardiol 2004; 43:831 839. 2. Achenbach S, Moselewski F, Ropers D, Ferencik M, Hoffmann U, MacNeill B, Pohle K, Baum U, Anders K, Jang IK, Daniel WG, Brady TJ. Detection of calcied and non-calcied coronary atherosclerotic plaque by contrast-enhanced, submillimeter multidetector spiral computed tomography: a segment-based comparison with intravascular ultrasound. Circulation 2004;109:14 17. 3. Leber AW, Becker A, Knez A, von Ziegler F, Sirol M, Nikolaou K, Ohnesorge B, Fayad ZA, Becker CR, Reiser M, Steinbeck G, Boekstegers P. Accuracy of 64-slice computed tomography to classify and quantify plaque volumes in the proximal coronary system: a comparative study using intravascular ultrasound. J Am Coll Cardiol 2006; 47:672 677. 4. Leber AW, Knez A, Becker A, Becker C, von Ziegler F, Nikolaou K, Rist C, Reiser M, White C, Steinbeck G, Boekstegers P. Accuracy of multidetector spiral computed tomography in identifying and differentiating the composition of coronary atherosclerotic plaques: a comparative study with intracoronary ultrasound. J Am Coll Cardiol 2004; 43:12411247.

5. Diamond GA, Forrester JS. Analysis of probability as an aid in the clinical diagnosis of coronary artery disease. N Engl J Med 1979;300: 1350 1358. 6. Morise AP, Haddad WJ, Beckner D. Development and validation of a clinical score to estimate the probability of coronary artery disease in men and women presenting with suspected coronary disease. Am J Med 1997;102:350 356. 7. Morise AP, Jalisi F. Evaluation of pretest and exercise test scores to assess all-cause mortality in unselected patients presenting for exercise testing with symptoms of suspected coronary artery disease. J Am Coll Cardiol 2003;42:842 850. 8. Austen WG, Edwards JE, Frye RL, Gensini GG, Gott VL, Grifth LS, McGoon DC, Murphy ML, Roe BB. A reporting system on patients evaluated for coronary artery disease. Report of the Ad Hoc Committee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association. Circulation 1975;51:5 40. 9. Min JK, Shaw LJ, Devereux RB, Okin PM, Weinsaft JW, Russo DJ, Lippolis NJ, Berman DS, Callister TQ. Prognostic value of multidetector coronary computed tomographic angiography for prediction of all-cause mortality. J Am Coll Cardiol 2007;50:11611170. 10. Myocardial infarction redeneda consensus document of the Joint European Society of Cardiology/American College of Cardiology committee for the redenition of myocardial infarction. Eur Heart J 2000;21:15021513. 11. Braunwald F. Unstable angina. A classication. Circulation 1989;80: 410 414. 12. Mann JM, Davies MJ. Vulnerable plaque. Relation of characteristics to degree of stenosis in human coronary arteries. Circulation 1996;94: 928 931. 13. Ambrose JA, Tannenbaum MA, Alexopoulos D, Hjemdahl-Monsen CE, Leavy J, Weiss M, Borrico S, Gorlin R, Fuster V. Angiographic progression of coronary artery disease and the development of myocardial infarction. J Am Coll Cardiol 1988;12:56 62. 14. Little WC, Constantinescu M, Applegate RJ, Kutcher MA, Burrows MT, Kahl FR, Santamore WP. Can coronary angiography predict the site of a subsequent myocardial infarction in patients with mild-tomoderate coronary artery disease? Circulation 1988;78:11571566. 15. Kolodgie FD, Burke AP, Farb A, Gold HK, Yuan J, Narula J, Finn AV, Virmani R. The thin-cap broatheroma: a type of vulnerable plaque: the major precursor lesion to acute coronary syndromes. Curr Opin Cardiol 2001;16:285292. 16. Pundziute G, Schuijf JD, Jukema JW, Boersma E, de Roos A, van der Wall EE, Bax JJ. Prognostic value of multislice computed tomography coronary angiography in patients with known or suspected coronary artery disease. J Am Coll Cardiol 2007;49:6270. 17. Carrigan TP, Nair D, Schoenhagen P, Curtin RJ, Popovic ZB, Halliburton S, Kuzmiak S, White RD, Flamm SD, Desai MY. Prognostic utility of 64-slice computed tomography in patients with suspected but no documented coronary artery disease. Eur Heart J 2009;30:362371. 18. Hadamitzky M, Freissmuth B, Meyer T, Hein F, Kastrati A, Martinoff S, Schmig A, Hausleiter J. Prognostic value of coronary computed tomographic angiography for prediction of cardiac events in patients with suspected coronary artery disease. JACC Cardiovasc Imaging 2009;2:404 411. 19. van Werkhoven JM, Gaemperli O, Schuijf JD, Jukema JW, Kroft LJ, Leschka S, Alkadhi H, Valenta I, Pundziute G, de Roos A, van der Wall EE, Kaufmann PA, Bax JJ. Multislice computed tomography coronary angiography for risk stratication in patients with an intermediate pretest likelihood. Heart 2009;95:16071611. 20. Meijboom WB, van Mieghem CA, Mollet NR, Pugliese F, Weustink AC, van Pelt N, Cademartiri F, Nieman K, Boersma E, de Jaegere P, Krestin GP, de Feyter PJ. 64-slice computed tomography coronary angiography in patients with high, intermediate, or low pretest probability of signicant coronary artery disease. J Am Coll Cardiol 2007; 50:1469 1475.

Usefulness of Cooling and Coronary Catheterization to Improve Survival in Out-of-Hospital Cardiac Arrest
Dion Stub, MBBSa,b,*, Christopher Hengel, MBBSa, William Chan, MBBSa,b, Damon Jackson, MBBSa, Karen Sanders, RN, GradDipEda, Anthony M. Dart, BA, BM, BCh, DPhila,b, Andrew Hilton, MBBSc, Vincent Pellegrino, MBBSc, James A. Shaw, MBBS, PhDa,b, Stephen J. Duffy, MBBS, PhDa, Stephen Bernard, MBBS, MDc, and David M. Kaye, MBBS, PhDa,b
Survival rates after out-of-hospital cardiac arrest (OHCA) continue to be poor. Recent evidence suggests that a more aggressive approach to postresuscitation care, in particular combining therapeutic hypothermia with early coronary intervention, can improve prognosis. We performed a single-center review of 125 patients who were resuscitated from OHCA in 2 distinct treatment periods, from 2002 to 2003 (control group) and from 2007 to 2009 (contemporary group). Patients in the contemporary group had a higher prevalence of cardiovascular risk factors but similar cardiac arrest duration and prehospital treatment (adrenaline administration and direct cardioversion). Rates of cardiogenic shock (48% vs 41%, p 0.2) and decreased conscious state on arrival (77% vs 86%, p 0.2) were similar in the 2 cohorts, as was the incidence of ST-elevation myocardial infarction (33% vs 43%, p 0.1). The contemporary cohort was more likely to receive therapeutic hypothermia (75% vs 0%, p <0.01), coronary angiography (77% vs 45%, p <0.01), and percutaneous coronary intervention (38% vs 23%, p 0.03). This contemporary therapeutic strategy was associated with better survival to discharge (64% vs 39%, p <0.01) and improved neurologic recovery (57% vs 29%, p <0.01) and was the only independent predictor of survival (odds ratio 5.5, 95% condence interval 1.2 to 26.2, p 0.03). Longer resuscitation time, presence of cardiogenic shock, and decreased conscious state were independent predictors of poor outcomes. In conclusion, modern management of OHCA, including therapeutic hypothermia and early coronary angiography is associated with signicant improvement in survival to hospital discharge and neurologic recovery. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:522527) In a previous study conducted in Australia before the widespread adoption of postresuscitation strategies such as cooling, survival after admission to hospital for out-of-hospital cardiac arrest (OHCA) was 25%, comparable to most registry data.1 Given the rapid uptake of such approaches since that time, we hypothesized that advances in basic life support and postresuscitation hospital care have improved outcomes. Accordingly, we performed a single-center retrospective review of all patients with OHCA admitted to our hospital from 2002 to 2003 and from 2007 to 2009. Methods Melbourne has approximately 3.9 million inhabitants, which is served by a comprehensive centrally co-ordinated
a Alfred Hospital Heart Centre, Melbourne, Victoria, Australia; bBaker IDI Heart Diabetes Institute, Melbourne, Victoria, Australia; cAlfred Hospital Intensive Care Unit, Melbourne, Victoria, Australia. Manuscript received August 25, 2010; revised manuscript received and accepted October 5, 2010. Dr. Stub is supported by a scholarship from the Cardiac Society of Australia & New Zealand, Sydney, NSW and an award from Baker IDI Heart and Diabetes Institute, Melbourne, Victoria, Australia. *Corresponding author: Tel: 613-9076-2000; fax: 613-9076-2461. E-mail address: d.stub@alfred.org.au (D. Stub).

ambulance system, which is described elsewhere.1 The Alfred Hospital (Melbourne, Victoria, Australia) is a large tertiary-/quaternary-care referral center that provides 24hour emergency coronary and cardiac surgical interventions for patients with acute coronary syndromes. In this retrospective analysis we evaluated clinical characteristics and outcomes of all patients who had an out-ofhospital ventricular brillation arrest with sustained return of spontaneous circulation (ROSC), dened as 20 minutes, and who were subsequently hospitalized at the Alfred Hospital. Analysis was performed for 2 treatment periods: a modern treatment paradigm, 2007 to 2009 (contemporary group), and a historical control group, 2002 to 2003. Data were obtained from Ambulance Victoria and hospital records. The study was performed in accordance with the Alfred Hospital ethics committee guidelines. Interrogation of the hospital database identied 326 patients with presumed OHCA. Seventy-seven patients were excluded secondary to noncardiac causes such as trauma, stroke, and drug overdose. Excluded were 4 patients who did not have ROSC on arrival, 11 patients transferred from other institutions, and 109 patients because of asystole or pulseless electrical activity as their initial rhythm. Our study population, therefore, consisted of 125 patients with OHCA secondary to ventricular arrhythmia.
www.ajconline.org

0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.011

Coronary Artery Disease/Cooling and Catheterization in Cardiac Arrest Table 1 Baseline characteristics Characteristic Age (years) Men Current/ex-smoker Diabetes mellitus Hypertension Hyperlipidemia History of coronary disease Initial rhythm Ventricular brillation Ventricular tachycardia Basic life support Witnessed arrest Bystander resuscitation Ambulance response Call to arrival (minutes) Total time until return of circulation (minutes) Number of shocks Adrenaline administration Cause Acute coronary syndrome ST-segment elevation myocardial infarct Condition on arrival to hospital Unconscious Cardiogenic shock Interventions Therapeutic hypothermia Coronary angiography Emergent angiography Percutaneous coronary intervention Coronary bypass graft surgery Control (n 44) 64 17 34 (77%) 10 (23%) 3 (7%) 13 (30%) 5 (11%) 14 (32%) 40 (91%) 4 (9%) 41 (93%) 33 (75%) 6 (59) 26 (1535) 3.9 3.5 35 (80%) 30 (68%) 19 (43%) Contemporary (n 81) 61 16 68 (84%) 45 (56%) 9 (11%) 47 (58%) 37 (46%) 28 (35%) 77 (95%) 4 (5%) 75 (93%) 57 (70%) 7 (610) 23 (1430) 3.8 4.3 58 (72%) 50 (62%) 27 (33%) p Value NS NS 0.01 NS 0.01 0.01 NS NS NS NS NS NS NS NS NS NS NS Table 2 Outcome of coronary angiography in study population Coronary Angiographic Variable Normal Single-vessel disease Multivessel disease Coronary arteries 50% stenosis, mean SD Infarct-related artery Left anterior descending coronary artery Left circumex coronary artery Right coronary artery Grafts Left main coronary artery Multivessel with no clear culprit Preintervention Thrombolysis In Myocardial Infarction grade 02 ow Thrombus-containing lesion Control (n 20) 6 (32%) 12 (63%) 2 (11%) 1.5 0.8 Contemporary (n 62) 13 (21%) 17 (27%) 31 (50%) 1.9 0.8

523

p Value 0.36 0.02 0.01 NS

10 (53%) 0 (0%) 1 (5%) 0 (0%) 1 (5%) 2 (10%) 8 (42%)

17 (27%) 6 (12%) 14 (23%) 1 (2%) 3 (2%) 8 (16%) 23 (47%)

0.02 NS NS NS NS NS NS

4 (21%)

11 (22%)

NS

38 (86%) 18 (41%) 0 (0%) 20 (45%) 11 (25%) 10 (23%) 0 (0%)

62 (77%) 39 (48%) 61 (75%) 62 (77%) 49 (61%) 31 (38%) 4 (5%)

NS NS 0.01 0.01 0.01 0.03 NS

Values are presented as mean SD, number of patients (percentage), or median (interquartile range).

Ambulance Victoria uses a 2-tier system of ambulance paramedics, most of whom have advanced life support skills, and intensive care paramedics who are authorized to perform endotracheal intubation and administer a range of cardiac drugs. Melbourne also uses a medical emergency response program in which ambulance and re brigade services respond to cardiac arrests.2 Cardiac arrest protocols follow the recommendations of the Australian Resuscitation Council.3 After hemodynamic stabilization, patients are transported urgently to the nearest hospital. In the 2 treatment periods, hospital care for patients with OHCA was modeled on relevant International Liaison Committee on Resuscitation guidelines at the time.4,5 Decision regarding need for cardiac catheterization was made by the treating cardiologist. Intensive care treatment including target hemodynamic and metabolic parameters and choice of inotropic agents were decided by the treating physician according to general critical care guidelines. Therapeutic hypothermia was induced and maintained through a combination of ice-cold intravenous uids, simple ice packs, and surface cooling blankets. The 2 signicant changes to postresuscitative care in patients with OHCA during the study period were use of mild therapeutic hypothermia for

unconscious patients (to preserve neurologic function) and increasing use of emergency coronary angiography to assess and treat underlying coronary artery disease as the cause for OHCA. The primary outcome was survival to hospital discharge. Secondary outcome was good neurologic recovery, dened as cerebral performance categories (CPCs) 1 and 2. The CPC is a simple-to-use widely used cerebral performance measurement.6 Statistical analyses were performed with SPSS 16 (SPSS, Inc., Chicago, Illinois). Numerical normally distributed data were analyzed using Students t test (presented as mean SD) and non-normal data were compared by Mann-Whitney test (presented as median with interquartile range). Proportions were analyzed with Fishers exact test. A p value 0.05 was regarded as statistically signicant. Prognostic factors that were found to be signicant (p 0.10) in preliminary univariate analyses were entered into a multivariate logistic regression analysis. All variables were entered into the equation simultaneously to control for effects of confounding (a subsequent stepwise analysis provided similar results). Results Baseline characteristics of the study population are presented in Table 1. Important prehospital factors including rates of witnessed cardiac arrest, bystander cardiopulmonary resuscitation, time until ROSC, and adrenaline administration by paramedics were similar in the control and contemporary cohorts. On arrival to the emergency department the incidence of cardiogenic shock, dened as systolic blood pressure 90 mm Hg or requiring inotropic support, did not differ signicantly between treatment periods (41% vs 48%, p NS) and rates of decreased conscious state requiring intubation did not differ signicantly (86% vs 77%, p NS).

524

The American Journal of Cardiology (www.ajconline.org) Table 4 Signicant univariate predictors of survival Characteristic Odds Ratio 0.24 3.3 0.08 0.83 0.29 0.11 2.7 7.6 2.1 2.9 95% Condence Interval 0.110.5 1.57.5 0.030.21 0.730.94 0.100.84 0.050.26 1.16.4 3.217.5 0.954.4 1.36.1 p Value 0.01 0.01 0.01 0.01 0.02 0.01 0.02 0.01 0.07 0.01

Table 3 Outcome of percutaneous coronary intervention in study population Variable Historical Control PCI (n 10) Modern PCI (n 31) p Value NS NS

Procedural success 9 (90%) 29 (94%) Final Thrombolysis In 8 (80%) 28 (90%) Myocardial Infarction grade 3 ow Door-to-balloon time 145 (112 to 345) 120 (105 to 167) (minutes) Stents per patient 1.2 0.4 1.3 1.1 Mean stent length (mm) 17.9 5 19.8 10 Mean stent diameter (mm) 3.1 0.8 3.7 3 Drug-eluting stents 0 (0%) 7 (23%) Multivessel intervention 0 (0%) 6 (19%) Peak troponin mean (range) 92 (0186) 53 (0179) Intra-aortic balloon pump 4 (40%) 8 (30%) Glycoprotein IIb/IIIa 4 (40%) 23 (56%) inhibitor Aspiration catheter 0 5 (16%)

NS 0.03 NS NS NS NS NS NS NS NS

Age Basic life support Bystander resuscitation Ambulance response Return of circulation 20 minutes Number of shocks Condition on arrival Unconscious Cardiogenic shock Interventions Cooling* Coronary angiography Successful coronary intervention Contemporary management * For unconscious patients only.

Values are presented as mean SD, number of patients (percentage), or median (interquartile range).

Figure 1. Outcomes based on contemporary (black bars) and historical control (gray bars) treatment periods (*p 0.01).

In the contemporary treatment group 75% of all patients received therapeutic hypothermia, representing 98% of comatose patients after ventricular brillation (Table 1). No patients in the control group received therapeutic hypothermia. Rates of coronary angiography and percutaneous coronary intervention (PCI) were signicantly increased in the contemporary treatment group (77% vs 45%, p 0.01; 38% vs 23%, p 0.03, respectively; Table 1). Of patients undergoing coronary angiography, 50% of patients did not go on to PCI for various clinical reasons (Table 2). PCI was successful in 90% of patients in the 2 treatment groups (see Table 3). Incidence of the left anterior descending coronary artery (LAD) as the culprit infarct-related artery was higher in the control group (53% vs 27%, p 0.02). Multivessel disease (dened as multiple coronary lesions

with 50% stenosis) was more prevalent in the contemporary treatment group (50% vs 11%, p 0.01). Survival to hospital discharge in the contemporary treatment group was 64% compared to 39% in the historical control (p 0.01). Discharge with favorable neurologic outcome (CPC 1 or 2) was also signicantly improved (57% vs 30%, p 0.01; Figure 1). Of survivors in the contemporary treatment group, 89% made a good neurologic recovery. Cause of death was similar in the 2 periods with 70% of patients dying due to poor neurologic outcome, 25% due to persistent cardiac dysfunction, and 5% due to multiorgan failure. Unadjusted predictors associated with survival are presented in Table 4. Survivors were signicantly more likely to be managed by the contemporary treatment paradigm and undergo coronary angiography and successful PCI. In unconscious patients, there was a significant increase in survival (61% vs 37%, p 0.03) and good neurologic outcome (54% vs 27%, p 0.01) in those patients receiving therapeutic hypothermia. When adjusting for key prehospital and postresuscitative factors (Figure 2, Table 5), negative predictors of survival included cardiogenic shock, resuscitation times 20 minutes, and decreased conscious state. The contemporary treatment regimen was a signicant independent predictor of survival (odds ratio 5.5, 95% condence interval 1.2 to 26.2, p 0.03). Discussion This study has demonstrated that a contemporary treatment paradigm with focused co-ordinated postresuscitative care combining therapeutic hypothermia with coronary angiography is associated with signicant improvements in short-term clinical outcomes. The 64% survival to discharge rate is signicantly better than other registry data of patients with OHCA and favorably compares to other institutions with similar treatment protocols.7,8 In the contemporary treatment group, 98% of comatose patients received therapeutic hypothermia. Of these patients, 60% survived, and of the survivors, 89% made a

Coronary Artery Disease/Cooling and Catheterization in Cardiac Arrest

525

Figure 2. Independent predictors of survival. CI condence interval; OR odds ratio.

Table 5 Multivariate predictors of survival Characteristic Odd Ratio 95% Condence Interval 1.226.2 0.9719 0.020.54 0.030.55 0.010.9 p Value

Contemporary management Coronary angiography Cardiogenic shock Return of circulation 20 minutes Unconscious on arrival

5.5 4.3 0.12 0.12 0.13

0.03 0.06 0.01 0.01 0.05

favorable neurologic recovery. This observation is important because uncertainty about neurologic recovery of patients with OHCA has previously cast doubt about the merits (and futility) of early invasive strategies such as coronary intervention. With the introduction of therapeutic hypothermia and the recognized difculty in predicting neurologic outcomes when patients rst arrive, there is a small role for early neurologic prognostication as a basis for further treatment decisions.9,10 Therapeutic hypothermia was a signicant unadjusted predictor of survival despite a large proportion of patients with cardiogenic shock who have previously been excluded from randomized studies. Use of mild therapeutic hypothermia is supported by 2 large randomized controlled trials11,12 and is a recommended part of a standardized treatment strategy for comatose survivors of cardiac arrest.13 ST-segment elevation on initial electrocardiogram was similar in the 2 groups in approximately 1/3 of patients. There have been several observational studies and systematic reviews highlighting the importance of emergency PCI in patients with ST-segment elevation myocardial infarction and OHCA.14 16 Recent reports have illustrated the further benets of combining therapeutic hypothermia with early coronary intervention.1719 This has led to the recommen-

dation by the American Heart Association that all patients with ST-segment elevation myocardial infarction and OHCA be managed at centers capable of 24-hour coronary intervention.20 Absence of ST-segment elevation in the setting of cardiac arrest has been shown to occur in up to 40% of OHCAs caused by unstable coronary plaques and coronary thrombosis.21,22 Likewise, in our study 42% of patients undergoing emergency PCI did not have ST-segment elevation on electrocardiogram. This has led to the increasing adoption of emergency coronary angiography for all patients with OHCA of suspected cardiac origin and to developing appropriate systems of care to cater to such treatment protocols.2326 The caveat to this approach is the signicant number of patients with OHCA who undergo coronary angiography and do not go on to emergency revascularization. In our study 50% of patients undergoing angiography did not go on to PCI and 23% had angiographically normal coronary arteries. As part of an early cardiac catheterization protocol they received antiplatelet and antithrombotic agents, which have been hypothesized to have their own positive effects in the setting of cardiac arrest associated with coagulation disruption.27,28 It was interesting to note that signicantly more patients from the control group had the LAD as the infarct-related artery compared to the contemporary cohort (53% vs 27%, p 0.02). This is possibly explained by the relatively small numbers of patients in the control group undergoing coronary angiography (n 20). Most LAD infarcts are generally larger and associated with greater hemodynamic disturbance than non-LAD infarcts and more than likely inuenced the decision to proceed to coronary angiography in the control group. There was a trend to shorter resuscitation times in the contemporary treatment group, which may have contributed

526

The American Journal of Cardiology (www.ajconline.org) protocol for post resuscitation care after out-of-hospital cardiac arrest. Resuscitation 2007;73:29 39. Werling M, Thoren AB, Axelsson C, Herlitz J. Treatment and outcome in post-resuscitation care after out-of-hospital cardiac arrest when a modern therapeutic approach was introduced. Resuscitation 2007;73: 40 45. Young GB. Clinical practice. Neurologic prognosis after cardiac arrest. N Engl J Med 2009;361:605 611. Wijdicks EF, Hijdra A, Young GB, Bassetti CL, Wiebe S. Practice parameter: prediction of outcome in comatose survivors after cardiopulmonary resuscitation (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2006;67:203210. Bernard SA, Gray TW, Buist MD, Jones BM, Silvester W, Gutteridge G, Smith K. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med 2002;346:557 563. The Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002;346:549 556. Neumar RW, Nolan JP, Adrie C, Aibiki M, Berg RA, Bottiger BW, Callaway C, Clark RS, Geocadin RG, Jauch EC, Kern KB, Laurent I, Longstreth WT Jr, Merchant RM, Morley P, Morrison LJ, Nadkarni V, Peberdy MA, Rivers EP, Rodriguez-Nunez A, Sellke FW, Spaulding C, Sunde K, Vanden Hoek T. Post-cardiac arrest syndrome: epidemiology, pathophysiology, treatment, and prognostication. A consensus statement from the International Liaison Committee on Resuscitation (American Heart Association, Australian and New Zealand Council on Resuscitation, European Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Asia, and the Resuscitation Council of Southern Africa); the American Heart Association Emergency Cardiovascular Care Committee; the Council on Cardiovascular Surgery and Anesthesia; the Council on Cardiopulmonary, Perioperative, and Critical Care; the Council on Clinical Cardiology; and the Stroke Council. Circulation 2008;118:24522483. Noc M, Radsel P. Urgent invasive coronary strategy in patients with sudden cardiac arrest. Curr Opin Crit Care 2008;14:287291. Garot P, Lefevre T, Eltchaninoff H, Morice MC, Tamion F, Abry B, Lesault PF, Le Tarnec JY, Pouges C, Margenet A, Monchi M, Laurent I, Dumas P, Garot J, Louvard Y. Six-month outcome of emergency percutaneous coronary intervention in resuscitated patients after cardiac arrest complicating ST-elevation myocardial infarction. Circulation 2007;115:1354 1362. Hosmane VR, Mustafa NG, Reddy VK, Reese CL, DiSabatino A, Kolm P, Hopkins JT, Weintraub WS, Rahman E. Survival and neurologic recovery in patients with ST-segment elevation myocardial infarction resuscitated from cardiac arrest. J Am Coll Cardiol 2009;53: 409 415. Hovdenes J, Laake JH, Aaberge L, Haugaa H, Bugge JF. Therapeutic hypothermia after out-of-hospital cardiac arrest: experiences with patients treated with percutaneous coronary intervention and cardiogenic shock. Acta Anaesthesiol Scand 2007;51:137142. Knafelj R, Radsel P, Ploj T, Noc M. Primary percutaneous coronary intervention and mild induced hypothermia in comatose survivors of ventricular brillation with ST-elevation acute myocardial infarction. Resuscitation 2007;74:227234. Kern KB, Rahman O. Emergent percutaneous coronary intervention for resuscitated victims of out-of-hospital cardiac arrest. Catheter Cardiovasc Interv 2010;75:616 624. Nichol G, Aufderheide TP, Eigel B, Neumar RW, Lurie KG, Bufalino VJ, Callaway CW, Menon V, Bass RR, Abella BS, Sayre M, Dougherty CM, Racht EM, Kleinman ME, OConnor RE, Reilly JP, Ossmann EW, Peterson E. Regional systems of care for out-of-hospital cardiac arrest: A policy statement from the American Heart Association. Circulation 2010;121:709 729. Spaulding CM, Joly LM, Rosenberg A, Monchi M, Weber SN, Dhainaut JF, Carli P. Immediate coronary angiography in survivors of out-of-hospital cardiac arrest. N Engl J Med 1997;336:1629 1633. Dumas F, Cariou A, Manzo-Silberman S, Grimaldi D, Vivien B, Rosencher J, Empana JP, Carli P, Mira JP, Jouven X, Spaulding C. Immediate percutaneous coronary intervention is associated with better survival after out-of-hospital cardiac arrest: insights from the

to improved outcomes. However, this did not reach statistical signicance (p 0.09). During the study period several key changes to prehospital care included adopting a chest compression-to-ventilation ratio of 30:2, in line with International Liaison Committee on Resuscitation guidelines, and focusing on uninterrupted chest compressions.4 As in other recent studies on optimizing prehospital care, it is reasonable to assume that these measures also contributed to improved patient survival.29,30 There are several limitations to our study. This is a single-center retrospective review and thus subject to potential confounders and selection bias contributing to results. The patients comprise a selected group who achieved ROSC before transport to hospital. Hospital and intensive care paramedics have extensive experience in managing patients with OHCA, and therefore the results applicability to other health care networks is uncertain. In the assessment of neurologic recovery, the CPC score was chosen because of its ease of use and wide reporting in the literature. Although simple to use, this scoring system has not been well validated and was retrospectively assigned based on patient follow-up and clinical notes. With regard to therapeutic hypothermia, time spent at target temperature range was not recorded, making the quality of hypothermia difcult to ascertain. The study, however, does indicate that in a contemporary treatment era, an aggressive approach to patients with OHCA of suspected cardiac origin is associated with signicantly improved survival to hospital discharge and neurologic recovery. Further study and randomized trials with particular focus on establishing systems of prehospital care and role of early coronary intervention are required.
1. Jennings PA, Cameron P, Walker T, Bernard S, Smith K. Out-ofhospital cardiac arrest in Victoria: rural and urban outcomes. Med J Aust 2006;185:135139. 2. Smith KL, McNeil JJ. Cardiac arrests treated by ambulance paramedics and re ghters. Med J Aust 2002;177:305309. 3. Morley PT, Walker T. Australian Resuscitation Council: adult advanced life support (ALS) guidelines 2006. Crit Care Resuscitation 2006;8:129 131. 4. Timerman S, Gonzalez MM, Mesquita ET, Marques FR, Ramires JA, Quilici AP, Timerman A. The International Liaison Committee on Resuscitation (ILCOR). Roll in guidelines 20052010 for cardiopulmonary resuscitation and emergency cardiovascular care. Arq Bras Cardiol 2006;87(suppl):e201 e208. 5. American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2005; 112(suppl):IV1IV203. 6. Jacobs I, Nadkarni V, Bahr J, Berg RA, Billi JE, Bossaert L, Cassan P, Coovadia A, DEste K, Finn J, Halperin H, Handley A, Herlitz J, Hickey R, Idris A, Kloeck W, Larkin GL, Mancini ME, Mason P, Mears G, Monsieurs K, Montgomery W, Morley P, Nichol G, Nolan J, Okada K, Perlman J, Shuster M, Steen PA, Sterz F, Tibballs J, Timerman S, Truitt T, Zideman D. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplication of the Utstein templates for resuscitation registries. A statement for healthcare professionals from a task force of the international liaison committee on resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa). Resuscitation 2004;63:233249. 7. Sunde K, Pytte M, Jacobsen D, Mangschau A, Jensen LP, Smedsrud C, Draegni T, Steen PA. Implementation of a standardised treatment

8.

9. 10.

11.

12.

13.

14. 15.

16.

17.

18.

19.

20.

21.

22.

Coronary Artery Disease/Cooling and Catheterization in Cardiac Arrest PROCAT (Parisian Region Out of Hospital Cardiac Arrest) Registry. Circ Cardiovasc Interv 2010;3:200 207. Peberdy MA, Ornato JP. Post-resuscitation care: is it the missing link in the chain of survival? Resuscitation 2005;64:135137. Ewy GA, Kern KB. Recent advances in cardiopulmonary resuscitation: cardiocerebral resuscitation. J Am Coll Cardiol 2009;53:149 157. Merchant RM, Abella BS, Khan M, Huang KN, Beiser DG, Neumar RW, Carr BG, Becker LB, Vanden Hoek TL. Cardiac catheterization is underutilized after in-hospital cardiac arrest. Resuscitation 2008;79: 398 403. Callaway CW, Schmicker R, Kampmeyer M, Powell J, Rea TD, Daya MR, Aufderheide TP, Davis DP, Rittenberger JC, Idris AH, Nichol G. Receiving hospital characteristics associated with survival after outof-hospital cardiac arrest. Resuscitation 2010;81:524 529.

527

23. 24. 25.

26.

27. Bottiger BW, Motsch J, Bohrer H, Boker T, Aulmann M, Nawroth PP, Martin E. Activation of blood coagulation after cardiac arrest is not balanced adequately by activation of endogenous brinolysis. Circulation 1995;92:25722578. 28. Schneider A, Bottiger BW, Popp E. Cerebral resuscitation after cardiocirculatory arrest. Anesth Analg 2009;108:971979. 29. Garza AG, Gratton MC, Salomone JA, Lindholm D, McElroy J, Archer R. Improved patient survival using a modied resuscitation protocol for out-of-hospital cardiac arrest. Circulation 2009;119: 25972605. 30. Lund-Kordahl I, Olasveengen TM, Lorem T, Samdal M, Wik L, Sunde K. Improving outcome after out-of-hospital cardiac arrest by strengthening weak links of the local chain of survival; quality of advanced life support and post-resuscitation care. Resuscitation 2010;81:422 426.

Two-Year Safety and Effectiveness of Sirolimus-Eluting Stents (from a Prospective Registry)

Bimmer E. Claessen, MDa,b, Roxana Mehran, MDa,c, Martin B. Leon, MDa,d, Eric A. Heller, MDa,d, Giora Weisz, MDa,d, George Syros, MDa,d, Gary S. Mintz, MDa,d, Theresa Franklin-Bond, MSa,d, Irene Apostolidou, MDa,d, Jose P.S. Henriques, MD, PhDb, Gregg W. Stone, MDa,d, Jeffrey W. Moses, MDa,d, and George D. Dangas, MD, PhDa,c,*
Uncertainty exists about the long-term safety and efcacy outcomes of sirolimus-eluting stents (SESs) in unselected patients. The present study was performed to evaluate the safety and efcacy of the SES in treatment of patients with coronary artery disease in an unselected population. Over a 2-year period, 1,504 consecutive patients undergoing percutaneous coronary intervention with >1 SES were enrolled. The primary end point was the occurrence of target vessel failure (TVF; a composite of cardiac death, myocardial infarction, or clinically driven target vessel revascularization). An independent clinical event committee adjudicated all adverse events up to 2-year follow-up. Dual antiplatelet therapy was recommended for >1 year throughout the study period. Mean age was 65 11 years; 75% were men, and 34% were diabetics. SESs were implanted for off-label indications in 86% of cases. TVF rates were 3.3%, 6.9%, 11.5%, and 15.5% at 30-day, 6-month, 1-year, and 2-year follow-ups, respectively. The 2-year cumulative rate of denite/probable stent thrombosis was 0.9%; 0.2% was very late thrombosis, occurring from 1 year to 2 years. Patients off dual antiplatelet therapy at 6 months had a signicantly increased rate of subsequent death from noncardiac causes. Patients off dual antiplatelet therapy at 1 year had a signicantly decreased rate of subsequent clinically driven target lesion revascularization. In conclusion, use of SESs in unselected patients with coronary artery disease was associated with a low TVF rate at 2 years with an acceptable incidence of stent thrombosis. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107: 528 534) The United States Food and Drug Administration (FDA) approved the use of drug-eluting stents (DESs) for treatment of coronary artery disease based on results of randomized controlled trials showing a signicant decrease in restenosis and need for repeat revascularization with DESs compared to bare metal stents.13 Typically, randomized pivotal trials have excluded patients with complex coronary artery disease with high risk for cardiac events. In clinical practice, DESs have also been used for off-label indications.4,5 However, expanded use of DES in everyday clinical practice is less well studied, and the possibility of unrecognized complications may exist.6 8 The Comprehensive Assessment of Sirolimus-Eluting Stents in Complex Lesions (MATRIX) registry was designed to evaluate the safety and efcacy of the sirolimus-eluting stent (SES) in an unselected population of patients with obstructive coronary artery disease.
Cardiovascular Research Foundation, New York, New York; bAcademic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; cMount Sinai Medical Center, New York, New York; dColumbia University Medical Center, New York, New York. Manuscript received September 22, 2010; revised manuscript received and accepted October 5, 2010. The MATRIX registry was funded by a research grant from Cordis/ Johnson and Johnson, Warren, New Jersey, to the Cardiovascular Research Foundation, New York, New York. *Corresponding author: Tel: 212-241-7014; fax 212-241-0273 E-mail address: gdangas@crf.org (G.D. Dangas). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.010
a

The SES was the only DES approved by the FDA at the time of study initiation. This report focuses on clinical outcomes in the MATRIX registry up to 2-year follow-up. Methods The MATRIX registry was conducted under an FDAapproved investigative device exemption and was a prospective, open-label, nonrandomized registry of 1,504 consecutive patients 18 years undergoing percutaneous coronary intervention (PCI) requiring the placement of 1 DES (Cypher, Cordis, Johnson and Johnson, Warren, New Jersey) for single- or multivessel coronary artery disease. Inclusion criteria were (1) 1 lesion with 50% diameter stenosis in a native coronary artery or a bypass graft requiring PCI with stenting not to exceed 108 mm of stent length, (2) de novo and restenotic lesions including in stent restenosis and radiation failure, (3) reference diameter from 2.5 to 3.5 mm, and (4) ability to understand and grant written informed consent. Exclusion criteria were (1) conrmed pregnancy at time of index PCI, (2) known allergies to aspirin, clopidogrel, or ticlopidine, (3) known allergies to heparin and bivalirudin, (4) known allergy to any component of a SES, and (5) a signicant medical condition that in the investigators opinion might interfere with a patients optimal participation in this study. From March 2004 to August 2006, consecutive patients (n 1,504) who underwent PCI with the approved SES at the 2 participating sites
www.ajconline.org

Coronary Artery Disease/Two-Year Outcomes of Sirolimus-Eluting Stents Table 1 Baseline characteristics (n 1,504) Age (years) Body mass index (kg/m2) Man Diabetes mellitus Insulin-dependent diabetes mellitus Hypertension* Hyperlipidemia Smoker Previous myocardial infarction Previous percutaneous coronary intervention Previous coronary artery bypass grafting Indication for index procedure Stable angina with abnormal stress test Unstable angina pectoris Acute myocardial infarction Number of coronary arteries narrowed 1 2 3 Moderate/severe left ventricular dysfunction 64.8 11.4 29 5.6 74.5% 33.7% 7.2% 82.5% 84.7% 10.9% 33.2% 44.1% 20.9% 63.7% 33.2% 3.4% 2.0 0.9 35.8% 33.2% 31.0% 8.2%

529

Figure 1. Proportions of patients treated with a thienopyridine (medium blue), aspirin (light blue), thienopyridine and aspirin (dark blue), none (black), or unknown (blue-gray), at 30 days, 180 days, 1 year, and 2 years.

Data are presented as mean SD or percentage. * Dened as a documented history of hypertension diagnosed and/or treated by a physician. Denes as a documented history of hyperlipidemia diagnosed and/or treated by a physician. Table 2 Procedural characteristics (n 1,504) Number of lesions treated Number of lesions treated per patient Number of vessels treated per patient Treated vessel Right coronary artery Left anterior descending coronary artery Left circumex coronary artery Ramus intermedius Left main coronary artery Saphenous vein graft Arterial bypass graft Lesion length (mm) Reference vessel diameter (mm) Preprocedural minimum lumen diameter (mm) Postprocedural minimum lumen diameter (mm) Acute gain (mm) American Heart Association/American College of Cardiology lesion type B2/C Number of stents per patient Number of stents per lesion Stent length per lesion (mm) Stent diameter (mm) Predilatation performed Postdilatatation performed Maximum ination pressure (atm) Data are presented as mean SD or percentage. 2,879 1.9 1.0 1.3 0.5 32.0% 44.5% 35.4% 4.4% 3.3% 4.5% 0.6% 18.0 9.8 3.00 0.46 0.68 0.39 2.34 0.43 1.67 0.48 66.6% 2.1 1.2 1.1 0.5 24.1 12.7 3.03 0.42 64.7% 37.6% 15.5 2.6

Figure 2. Kaplan-Meier event rates of clinical end points cardiac death (light blue), myocardial infarction (teal), target vessel revascularization (dark blue), and target vessel failure (deep blue) at 30 days, 180 days, 1 year, and 2 years. Kaplan-Meier event rates of Q-wave and nonQ-wave myocardial infarction were 0.2% and 2.3% at 30 days, 0.3% and 2.5% at 180 days, 0.4% and 2.8% at 1 year, and 0.4% and 3.5% at 2 years.

(Lenox Hill Hospital, New York, New York, and Columbia University Medical Center, New York, New York) were considered for enrollment in this study. The respective institutional review boards approved the protocol and all patients granted written informed consent. PCI and stent implantation were performed in the stan-

dard manner. Heparin was administered to maintain an activated clotting time 250 seconds, and bivalirudin was used as an alternative anticoagulant in most cases (85%) according to standard clinical practice in the 2 clinical sites. After intracoronary injection of nitroglycerin, pre- and postprocedural angiographies of the involved vessel(s) were performed in 2 near orthogonal views showing the target lesion free of foreshortening or vessel overlap to allow for accurate quantitative coronary angiographic measurements.9 Pre- and postdilatation were performed at the operators discretion. In the event of an additional stent requirement, an SES was used. The following SES sizes were used in the MATRIX registry: 8, 18, 23, and 33 mm in length, with diameters of 2.5, 3.0, and 3.5 mm. Use of glycoprotein IIb/IIIa inhibitors was left to the discretion of the operator. Successful stent implantation was dened as the achievement of a nal diameter stenosis 50% by quantitative coronary angiography after

530

The American Journal of Cardiology (www.ajconline.org)

Figure 3. Target vessel failure at 2 years in high-risk subgroups. Table 3 Two-year Kaplan-Meier event rates of landmark analysis of patients on and off dual antiplatelet therapy On Landmark set at 6 months Subjects Death From cardiac causes From noncardiac causes Myocardial infarction Target vessel revascularization Target vessel failure Landmark set at 1 yr Subjects Death From cardiac causes From noncardiac causes Myocardial infarction Target vessel revascularization Target vessel failure 1,199 2.5% 1.0% 1.6% 1.1% 8.7% 9.9% 1,106 1.5% 0.6% 0.9% 0.7% 4.7% 5.5% Off 45 8.0% 0.0% 8.0% 0.0% 0.0% 0.0% 180 2.9% 1.2% 1.8% 1.2% 0.6% 3.0% p Value

0.047 0.539 0.001 0.519 0.056 0.041

Figure 4. Incidence and timing of denite/probable stent thrombosis (ST) (crosses) in patients using thienopyridines (gray bars) and those with unknown thienopyridine status (white bars); patients off thienopyridines could not be conrmed. No stent thrombosis event occurred when patients were conrmed to be off thienopyridines.

0.166 0.383 0.152 0.490 0.0279 0.179

stent implantation with normal ow. According to protocol, physicians prescribed aspirin 325 mg/day for 1 month, 81 mg/day thereafter plus clopidogrel 75 mg/day for 1 year after the procedure. Ticlopidine was an option for possible clopidogrel allergy. A patient was provided with the prescription and a written instruction sheet. General practitioners and general cardiologists were informed by written transmission of the preliminary and nal procedure reports that included information on dose and duration of dual antiplatelet therapy. Follow-up was planned at 30 days, 6 months, 1 year, and 2 years after the index procedure; at these respective time points follow-up was available for 99.1%, 97.8%, 95.5%, and 85.3% of patients. At each follow-up time point, information was collected on antiplatelet medication adherence and occurrence of clinical events. All clinical end points (see below) were adjudicated by an independent clinical events committee. The primary end point was the occurrence of target vessel failure (TVF), a composite of cardiac death, Q-wave-

and nonQ-wave myocardial infarctions, or clinically driven target vessel revascularization. Q-wave myocardial infarction was dened as the development of new pathologic Q waves 0.04 second in duration in 2 contiguous leads as assessed by the electrocardiographic core laboratory with creatine kinase or creatine kinase-MB levels increased above normal. NonQ-wave myocardial infarction was dened as an increase of creatine kinase levels to 2 times the upper normal limit with increased creatine kinase-MB in the absence of new pathologic Q waves. Target vessel revascularization was considered clinically driven in patients with a positive functional study result, ischemic changes on electrocardiogram consistent with the target vessel, an in-lesion diameter stenosis 50% by quantitative coronary angiography if the patient has ischemic symptoms, or an in-lesion diameter stenosis 70% by quantitative coronary angiography in the absence of ischemic symptoms. Secondary clinical endpoints included rates of individual clinical events and stent thrombosis. Stent thrombo-

Coronary Artery Disease/Two-Year Outcomes of Sirolimus-Eluting Stents Table 4 Two-year clinical event rates of patients treated with sirolimus-eluting stents in randomized trials and registries Year of Publication Randomized trials SIRIUS, E-SIRIUS, and C-SIRIUS pooled analysis RAVEL ENDEAVOR III SIRTAX ISAR-TEST-3 LEADERS SORT OUT III* Observational studies MATRIX Roy et al. Ong et al. Kaltoft et al. Kimura et al. Number of SES-Treated Patients 758 120 112 503 202 850 1,170 1,504 2,099 508 2,202 10,778 TVF Death/MI Death MI TVR TLR

531

ST

2006 2003 2007 2008 2009 2009 2010 present 2008 2006 2009 2009

10.9% NA 11.6% 12.3% NA 15.4% 4.5% 15.4% 22.6% 15.4% NA NA

NA NA NA NA 6.4% 9.1% NA 7.4% NA 9.7% NA NA

2.1% 5.0% 4.5% 5.0% 5.0% 5.1% 2.7% 3.9% NA 5.8% 5.9% 7.2%

4.7% 2.5% 3.6% 3.6% 2.0% 5.8% 0.9% 3.9% NA NA 4.1% 1.5%

NA NA NA 9.3% NA 8.8% 3.3% 12.1% 13.2% 8.2% NA NA

5.7% 2.5% 4.5% 7.8% 10.4% 7.3% 1.7% 10.0% NA NA NA 10.2%

0.9% 0.0% NA 2.4% 1.0% 2.5% 0.5% 0.9% 1.8% 0.4% 1.7% 0.9%

Direct cross-trial comparisons are discouraged. *Eighteen-month event rates. This trial excluded all postprocedural events for a 5-day period. C-SIRIUS Canadian study of sirolimus-coated stent in treatment of patients with de novo coronary artery lesions; ENDEAVOR III randomized controlled trial of the Medtronic endeavor drug [ABT-578] eluting coronary stent system versus the cypher sirolimus-eluting coronary stent system in de novo native coronary artery lesions; E-SIRIUS European study of sirolimus-coated stent in treatment of patients with de novo coronary artery lesions; ISAR-TEST-3 rapamycineluting stents with different polymer coating to reduce restenosis; LEADERS limus eluted from a durable versus erodable stent coating; MI myocardial infarction; NA not available; RAVEL randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization; SIRIUS sirolimus-coated stent in treatment of patients with de novo coronary artery lesions; SIRTAX sirolimus-eluting versus paclitaxel-eluting stents for coronary revascularization; SORT OUT III Danish organization on randomized trials with clinical outcome III; ST stent thrombosis; TLR target lesion revascularization; TVR target vessel revascularization.

sis was categorized according to denitions proposed by the Academic Research Consortium as denite or probable stent thrombosis.9 Timing of stent thrombosis was classied as acute (24 hours), subacute (24 hours to 30 days), late (1 month to 1 year), and very late (1 year). Continuous variables were summarized using mean SD. Categorical variables were summarized using frequencies. Survival curves using all available follow-up data were constructed for time-to-event variables using the Kaplan-Meier method and compared by log-rank test. Data on patients who were lost to follow-up were censored at the time of the last contact. To investigate the impact of cessation of dual antiplatelet therapy on subsequent clinical events, we performed landmark analyses comparing event rates (death, myocardial infarction, and clinically driven target vessel revascularization) between patients on and off dual antiplatelet therapy. Two landmark time points were considered, 6 months and 1 year. Statistical analyses were performed using SAS 9.1 (SAS Institute, Cary, North Carolina). Results In total 1,504 patients were enrolled in the MATRIX registry; mean age was 65 11 years, 75% were men, and 34% had diabetes mellitus. Additional baseline characteristics are listed in Table 1. SESs were successfully implanted in 98.6% of lesions and a mean of 2.1 1.2 SESs (per patient) was implanted during the index procedure. Table 2 lists procedural characteristics for the study cohort. Most patients (86%) underwent stenting for off-label indications,

including multivessel stenting (n 462, 30.7%), bifurcation lesions (n 295, 19.6%), saphenous vein grafts (n 67, 4.5%), long-term total occlusions (n 58, 3.9%), and acute myocardial infarction (n 49, 3.3%). At time of hospital discharge, 99.6% of patients were being treated with aspirin and 99.6% of patients were treated with clopidogrel or ticlopidine. Figure 1 shows the proportion of patients on aspirin and clopidogrel or ticlopidine during follow-up; high levels of adherence were documented. Rates of cardiac death, myocardial infarction, and ischemia-driven target vessel revascularization and the composite of TVF at 30-day, 6-month, 1-year, and 2-year follow-up are shown in Figure 2. Figure 3 shows TVF rates at 2 years for selected high-risk subgroups (multivessel stenting, diabetes mellitus, bifurcation lesions, saphenous vein grafts, longterm total occlusions, acute myocardial infarction, and previous brachytherapy). The highest 2-year TVF rate (39.0%) was observed in the brachytherapy group due to high rates of clinically driven target vessel revascularization (35.8%) and stent thrombosis (6.9%). In the diabetic subgroup we observed a trend toward a higher incidence of TVF at 2 years in insulin-treated diabetics compared to noninsulin-treated diabetics (24.9% vs 18.7%, p 0.11). Figure 4 shows the timing of denite/probable stent thrombosis events and whether a patient was using dual antiplatelet therapy at time of stent thrombosis. Rates of acute, subacute, late, and very late denite/probable stent thrombosis were 0%, 0.3%, 0.3%, and 0.2%, respectively.

532

The American Journal of Cardiology (www.ajconline.org)

Of the 13 patents with stent thrombosis, 10 patients (77%) had the event while on conrmed dual antiplatelet therapy and 3 patients (23%) while at an unknown antiplatelet therapy status. Antiplatelet therapy follow-up information was available for 88.8% and 89.5% of patients at 6 months and 1 year, respectively. Table 3 presents results of landmark analyses comparing event rates between patients on and off dual antiplatelet therapy. Patients off dual antiplatelet therapy at 6 months had a signicantly increased rate of subsequent death from noncardiac causes. Patients off dual antiplatelet therapy at 1 year had a signicantly decreased rate of subsequent clinically driven target lesion revascularization. Discussion Results from this rst registry under an FDA investigative device exemption showed that use of SESs in treatment of coronary artery disease in a high-risk unselected population (86% of patients received SESs for off-label indications) was associated with low TVF (15.5%) and acceptable stent thrombosis (0.9%) rates up to 2 years. The 1-year TVF rate in the present report was 11.5%, which is slightly higher than those reported in 2 recently published all-comer trials comparing next-generation DESs (6% to 9% TVF at 1 year).10,11 However, the present trial enrolled a larger proportion of patients with diabetes mellitus (34%) compared to patients in COMPARE (18%) and RESOLUTE All Comers (23%). Moreover, stent design improvements in second-generation DESs might also explain the favorable results in these recent trials, although prospective randomized data to conrm this hypothesis is still pending. One-year denite/probable stent thrombosis rates in COMPARE and RESOLUTE All Comers were 0.7% to 3.0%, which is comparable to the rate in the present trial, which was 0.7%. Table 4 presents how 2-year clinical outcomes from the present study compare to those from randomized studies and nonrandomized registries evaluating the SES for which 2-year results have been published.6,1221 Of note, among registries 2-year TVF rate was the lowest in the present study, on par with the TVF rate reported by Ong et al,16 although diabetes mellitus was considerably more frequent in our cohort (34% vs 18%). Therefore, our data add to the growing body of evidence suggesting use of SESs for offlabel indications is safe and effective. Two-year safety and efcacy data have been reported for the 3 other FDA-approved DES types. The most published data are available for the paclitaxel-eluting stents; 2 year TVF rates range from 13.1% to 21.1%, and stent thrombosis rates range from 1.3% to 2.4%.17,2224 Two-year TVF and stent thrombosis rates for the zotarolimus-eluting stent were 11.1% and 1.9%, respectively, in a randomized trial comparing paclitaxel-eluting stents to zotarolimus-eluting stents.14 Two-year TVF and stent thrombosis rates for the everolimus-eluting stent were 10.4% and 1.2%, respectively, in a pooled analysis of 2 randomized trials comparing everolimus-eluting stents to paclitaxel-eluting stents.24 Analysis of various high-risk subgroups showed a high 2-year TVF rate (30.8%) in patients treated for lesions in a saphenous vein graft. The increased rate of adverse clinical

events after treatment of saphenous vein graft lesions is well established and is due to the nature of diffusely diseased degenerated vein grafts with an increased risk of periprocedural complications due to potential distal embolization and an increased need for repeat intervention.2527 We observed a very high 2-year TVF rate (39.0%) in the subgroup of patients with a history of intracoronary brachytherapy; high stent thrombosis rates after intracoronary brachytherapy are well documented and have led to a decrease of its use over time.28,29 In the subgroup of 58 patients treated for long-term total occlusions, TVF rate was 19.1% at 2 years, which is relatively similar to the 15.5% overall TVF rate in MATRIX. This is consistent with observations from a randomized controlled trial of SESs in a lower-risk patient population with (sub)total coronary occlusions, which reported a favorable event rate of 10.0% at 3 years.30 Our results conrm the use of SESs as an effective treatment strategy for these complex lesions, typically treated with multiple (and mostly long) stents. The 2-year TVF rate of diabetic patients in MATRIX (20.1%) compares well to those from diabetic patients in the Registro Regionale Angioplastiche Emilia-Romagna (REAL) registry (23.3%) and the Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital (RESEARCH) registry (18.2%).22,31 TVF rates were relatively lower in the remaining highrisk in subgroups of acute myocardial infarction (15.0%), multiple vessels (19.6%), and bifurcation lesions (16.9%). In 2 small randomized clinical trials evaluating the use of SESs in acute myocardial infarction, 2-year TVF rates were 24.2% and 21%.32,33 To our knowledge, no previous studies have reported 2-year clinical event rates for treatment of multiple vessels or bifurcation lesions with SESs. However, 3-year death, myocardial infarction, and revascularization rates in the Arterial Revascularization Strategies-II (ARTSII) trial, which evaluated SESs in patients with multivessel disease, were 3.0%, 2.8%, and 11.0%, respectively, compared to 4.1%, 4.6%, and 13.2% at 2 years in the present study.34 Patients off dual antiplatelet therapy at 6 months had a signicantly increased rate of subsequent death from noncardiac causes, possibly due to noncardiac diagnoses leading to antiplatelet discontinuation. Although information on medical co-morbidities (e.g., cancer, lung disease, etc.) was not routinely collected in the MATRIX registry, these results suggest that patients who discontinue dual antiplatelet therapy are typically patients with severe co-morbidity who are more likely to have a fatality from noncardiac causes and less likely to undergo repeat intervention. A second analysis showed no increased risk of developing a clinical event if dual antiplatelet therapy was discontinued after 12 months. Interestingly, none of patients who sustained a stent thrombosis were conrmed to be off dual antiplatelet therapy before the event. Previous studies have shown an association between clopidogrel discontinuation and subsequent stent thrombosis; this association was not found in the MATRIX registry.6 8 Possible explanations could be the very low overall incidence of stent thrombosis in MATRIX and the high percent long-term adherence to dual antiplatelet therapy. These 2 factors may have limited the power to show an association between clopidogrel discontinuation

Coronary Artery Disease/Two-Year Outcomes of Sirolimus-Eluting Stents

533

and stent thrombosis in this 1,504-patient registry. However, our present observation is in accordance with a recent investigation in an Asian population suggesting no clinical benet of continuing dual antiplatelet therapy 1 year after stenting.35 All clinical end points up to 2 years were adjudicated by an independent clinical event committee. However, longerterm follow-up is needed to conrm if the low TVF and stent thrombosis rates are maintained. Furthermore, although MATRIX is a large 1,504-patient registry, this study was underpowered to evaluate a denitive association between premature discontinuation of dual antiplatelet therapy and incidence of stent thrombosis. Acknowledgment: We thank the clinical event adjudication committee: Allen Jeremias, MD, Steven O. Marx, MD, and Stanley Schneller, MD.
1. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban HE, 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: 17731780. 2. Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, OShaughnessy 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;349:13151323. 3. Caixeta A, Leon MB, Lansky AJ, Nikolsky E, Aoki J, Moses JW, Schofer J, Morice MC, Schampaert E, Kirtane AJ, Popma JJ, Parise H, Fahy M, Mehran R. 5-Year clinical outcomes after sirolimus-eluting stent implantation insights from a patient-level pooled analysis of 4 randomized trials comparing sirolimus-eluting stents with bare-metal stents. J Am Coll Cardiol 2009;54:894 902. 4. Marroquin OC, Selzer F, Mulukutla SR, Williams DO, Vlachos HA, Wilensky RL, Tanguay JF, Holper EM, Abbott JD, Lee JS, Smith C, Anderson WD, Kelsey SF, Kip KE. A comparison of bare-metal and drug-eluting stents for off-label indications. N Engl J Med 2008;358: 342352. 5. Win HK, Caldera AE, Maresh K, Lopez J, Rihal CS, Parikh MA, Granada JF, Marulkar S, Nassif D, Cohen DJ, Kleiman NS. Clinical outcomes and stent thrombosis following off-label use of drug-eluting stents. JAMA 2007;297:20012009. 6. Kimura T, Morimoto T, Nakagawa Y, Tamura T, Kadota K, Yasumoto H, Nishikawa H, Hiasa Y, Muramatsu T, Meguro T, Inoue N, Honda H, Hayashi Y, Miyazaki S, Oshima S, Honda T, Shiode N, Namura M, Sone T, Nobuyoshi M, Kita T, Mitsudo K. Antiplatelet therapy and stent thrombosis after sirolimus-eluting stent implantation. Circulation 2009;119:987995. 7. Lasala JM, Cox DA, Dobies D, Baran K, Bachinsky WB, Rogers EW, Breall JA, Lewis DH, Song A, Starzyk RM, Mascioli SR, Dawkins KD, Baim DS. Drug-eluting stent thrombosis in routine clinical practice: two-year outcomes and predictors from the TAXUS ARRIVE registries. Circ Cardiovasc Interv 2009;2:285293. 8. van Werkum JW, Heestermans AA, Zomer AC, Kelder JC, Suttorp MJ, Rensing BJ, Koolen JJ, Brueren BR, Dambrink JH, Hautvast RW, Verheugt FW, ten Berg JM. Predictors of coronary stent thrombosis: the Dutch Stent Thrombosis Registry. J Am Coll Cardiol 2009;53: 1399 1409. 9. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A, Hamon M, Krucoff MW, Serruys PW. Clinical end points in coronary stent trials: a case for standardized denitions. Circulation 2007;115: 2344 2351. 10. Serruys PW, Silber S, Garg S, van Geuns RJ, Richardt G, Buszman PE, Kelbaek H, van Boven AJ, Hofma SH, Linke A, Klauss V, Wijns W, Macaya C, Garot P, Dimario C, Manoharan G, Kornowski R, Ischinger T, Bartorelli A, Ronden J, Bressers M, Gobbens P, Negoita

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

M, Van LF, Windecker S. Comparison of zotarolimus-eluting and everolimus-eluting coronary stents. N Engl J Med 2010;363:136 146. Kedhi E, Joesoef KS, McFadden E, Wassing J, Van MC, Goedhart D, Smits PC. Second-generation everolimus-eluting and paclitaxel-eluting stents in real-life practice (COMPARE): a randomised trial. Lancet 2010;375:201209. Kaltoft A, Jensen LO, Maeng M, Tilsted HH, Thayssen P, Bottcher M, Lassen JF, Krusell LR, Rasmussen K, Hansen KN, Pedersen L, Johnsen SP, Sorensen HT, Thuesen L. 2-Year clinical outcomes after implantation of sirolimus-eluting, paclitaxel-eluting, and bare-metal coronary stents: results from the WDHR (Western Denmark Heart Registry). J Am Coll Cardiol 2009;53:658 664. Klauss V, Serruys PW, van Es GA, Buszman P, Ischinger T, Eberli F, Corti R, Wijns W, Morice MC, Di Mario C, Schuler G, Linke A, Juni P, Windecker S. Biolimus-eluting stent with biodegradable polymer versus sirolimus-eluting stent with durable polymer for coronary revascularisation (leaders): a randomised non-inferiority trial: 24 months follow-up. Am J Cardiol 2009;104:XVI. Leon MB, Kandzari DE. Two-year outcomes from the ENDEAVOR III trial: a randomized trial of the ENDEAVOR zotarolimus-eluting stent compared with the cypher sirolimus-eluting stent. J Am Coll Cardiol 2007;49(suppl):45B. Morice MC, Serruys P, Costantini C, Wuelfert E, Wijns W, Fajadet J, Columbo A, Guagliumi G, Molnar F, Hayashi EB, Sousa JEM, Perin M. Two-year follow-up of the RAVEL study: A randomized study with the sirolimus-eluting Bx VELOCITY stent in the treatment of patients with de-novo native coronary artery lesions. J Am Coll Cardiol 2003;41(suppl):32A. Ong AT, van Domburg RT, Aoki J, Sonnenschein K, Lemos PA, Serruys PW. Sirolimus-eluting stents remain superior to bare-metal stents at two years: medium-term results from the Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital (RESEARCH) registry. J Am Coll Cardiol 2006;47:1356 1360. Roy P, de la Bonello L, De LA, Okabe T, Pinto Slottow TL, Steinberg DH, Torguson R, Smith K, Xue Z, Satler LF, Kent KM, Suddath WO, Pichard AD, Waksman R. Two-year outcome of patients treated with sirolimus- versus paclitaxel-eluting stents in an unselected population with coronary artery disease (from the REWARDS Registry). Am J Cardiol 2008;102:292297. Schampaert E, Moses JW, Schofer J, Schluter M, Gershlick AH, Cohen EA, Palisaitis DA, Breithardt G, Donohoe DJ, Wang H, Popma JJ, Kuntz RE, Leon MB. Sirolimus-eluting stents at two years: a pooled analysis of SIRIUS, E-SIRIUS, and CSIRIUS with emphasis on late revascularizations and stent thromboses. Am J Cardiol 2006; 98:36 41. Rasmussen K, Maeng M, Kaltoft A, Thayssen P, Kelbaek H, Tilsted HH, Abildgaard U, Christiansen EH, Engstrom T, Krusell LR, Ravkilde J, Hansen PR, Hansen KN, Abildstrom SZ, Aaroe J, Jensen JS, Kristensen SD, Botker HE, Madsen M, Johnsen SP, Jensen LO, Sorensen HT, Thuesen L, Lassen JF. Efcacy and safety of zotarolimus-eluting and sirolimus-eluting coronary stents in routine clinical care (SORT OUT III): a randomised controlled superiority trial. Lancet 2010;375:1090 1099. Billinger M, Beutler J, Taghetchian KR, Remondino A, Wenaweser P, Cook S, Togni M, Seiler C, Stettler C, Eberli FR, Luscher TF, Wandel S, Juni P, Meier B, Windecker S. Two-year clinical outcome after implantation of sirolimus-eluting and paclitaxel-eluting stents in diabetic patients. Eur Heart J 2008;29:718 725. Byrne RA, Kufner S, Tiroch K, Massberg S, Laugwitz KL, Birkmeier A, Schulz S, Mehilli J. Randomised trial of three rapamycin-eluting stents with different coating strategies for the reduction of coronary restenosis: 2-year follow-up results. Heart 2009;95:1489 1494. Balducelli M, Ortolani P, Marzaroli P, Piovaccari G, Menozzi A, Manari A, Sangiorgio P, Tarantino F, Rossi R, Maresta A, Tondi S, Passerini F, Guastaroba P, Grilli R, Marzocchi A. Comparison of 2-year clinical outcomes with sirolimus and paclitaxel-eluting stents for patients with diabetes: results of the Registro Regionale AngiopLastiche Emilia-Romagna Registry. Catheter Cardiovasc Interv 2009;75:327334. Leon MB, Kandzari DE, Eisenstein EL, Anstrom KJ, Mauri L, Cutlip DE, Nikolsky E, OShaughnessy C, Overlie PA, Kirtane AJ, McLaurin BT, Solomon SL, Douglas JS Jr, Popma JJ. Late safety, efcacy, and cost-effectiveness of a zotarolimus-eluting stent compared with a paclitaxel-eluting stent in patients with de novo coronary lesions: 2-year

534

The American Journal of Cardiology (www.ajconline.org) follow-up from the ENDEAVOR IV trial (Randomized, Controlled Trial of the Medtronic Endeavor Drug [ABT-578] Eluting Coronary Stent System Versus the Taxus Paclitaxel-Eluting Coronary Stent System in De Novo Native Coronary Artery Lesions). JACC Cardiovasc Interv 2009;2:1208 1218. Onuma Y, Serruys PW, Kukreja N, Veldhof S, Doostzadeh J, Cao S, Stone GW. Randomized comparison of everolimus- and paclitaxeleluting stents: pooled analysis of the 2-year clinical follow-up from the SPIRIT II and III trials. Eur Heart J 2010;31:10711078. Hoffmann R, Hamm C, Nienaber CA, Levenson B, Bonzel T, Sabin G, Senges J, Zahn R, Tebbe U, Pfannebecker T, Richardt HG, Schneider S, Kelm M. Implantation of sirolimus-eluting stents in saphenous vein grafts is associated with high clinical follow-up event rates compared with treatment of native vessels. Coron Artery Dis 2007;18:559 564. Okabe T, Lindsay J, Buch AN, Steinberg DH, Roy P, Slottow TL, Smith K, Torguson R, Xue Z, Satler LF, Kent KM, Pichard AD, Weissman NJ, Waksman R. Drug-eluting stents versus bare metal stents for narrowing in saphenous vein grafts. Am J Cardiol 2008;102: 530 534. van Twisk PH, Daemen J, Kukreja N, van Domburg RT, Serruys PW. Four-year safety and efcacy of the unrestricted use of sirolimus- and paclitaxel-eluting stents in coronary artery bypass grafts. EuroIntervention 2008;4:311317. Costa MA, Sabate M, van der Giessen WJ, Kay IP, Cervinka P, Ligthart JM, Serrano P, Coen VL, Levendag PC, Serruys PW. Late coronary occlusion after intracoronary brachytherapy. Circulation 1999;100:789 792. Waksman R. Late thrombosis after radiation. Sitting on a time bomb. Circulation 1999;100:780 782. Rahel BM, Laarman GJ, Kelder JC, ten Berg JM, Suttorp MJ. Threeyear clinical outcome after primary stenting of totally occluded native coronary arteries: a randomized comparison of bare-metal stent implantation with sirolimus-eluting stent implantation for the treatment of total coronary occlusions (Primary Stenting of Totally Occluded Native Coronary Arteries [PRISON] II study). Am Heart J 2009;157:149 155. Daemen J, Garcia-Garcia HM, Kukreja N, Imani F, de Jaegere PP, Sianos G, van Domburg RT, Serruys PW. The long-term value of sirolimus- and paclitaxel-eluting stents over bare metal stents in patients with diabetes mellitus. Eur Heart J 2007;28:26 32. Menichelli M, Parma A, Giulivi A, Violini R. Randomized trial of Sirolimus-Eluting Stent Versus Bare-Metal Stent in Acute Myocardial Infarction (SESAMI 2 years results) (abstract). Am J Cardiol 2007;100 (suppl):210L. Valgimigli M, Campo G, Arcozzi C, Malagutti P, Carletti R, Ferrari F, Barbieri D, Parrinello G, Percoco G, Ferrari R. Two-year clinical follow-up after sirolimus-eluting versus bare-metal stent implantation assisted by systematic glycoprotein IIb/IIIa inhibitor infusion in patients with myocardial infarction: results from the STRATEGY study. J Am Coll Cardiol 2007;50:138 145. Serruys PW, Daemen J, Morice MC, De BB, Colombo A, Macaya C, Richardt G, Fajadet J, Hamm C, Dawkins KD, Vranckx P, Bressers M, Van DR, Schuijer M, Wittebols K, Pieters M, Stoll HP. Three-year follow-up of the ARTS-II#sirolimus-eluting stents for the treatment of patients with multivessel coronary artery disease. EuroIntervention 2008;3:450 459. Park SJ, Park DW, Kim YH, Kang SJ, Lee SW, Lee CW, Han KH, Park SW, Yun SC, Lee SG, Rha SW, Seong IW, Jeong MH, Hur SH, Lee NH, Yoon J, Yang JY, Lee BK, Choi YJ, Chung WS, Lim DS, Cheong SS, Kim KS, Chae JK, Nah DY, Jeon DS, Seung KB, Jang JS, Park HS, Lee K. Duration of dual antiplatelet therapy after implantation of drug-eluting stents. N Engl J Med 2010;362:1374 1382.

24.

31.

25.

32.

33.

26.

27.

34.

28.

35.

29. 30.

Comparison of Morbidity and Mortality in Diabetics Versus Nondiabetics Having Isolated Coronary Bypass Versus Coronary Bypass plus Valve Operations Versus Isolated Valve Operations
Serenella Castelvecchio, MDa,b,*, Lorenzo Menicanti, MDa, Ekaterina Baryshnikova, PhDb, Carlo de Vincentiis, MDa, Alessandro Frigiola, MDa, and Marco Ranucci, MDb, for the Surgical and Clinical Outcome Research (SCORE) Group
The impact of diabetes mellitus (DM) on the outcome of patients requiring cardiac surgery has been investigated in previous decades. However, the prole of cardiac surgical practice is changing in addition to changes in patients risk prole, making the results inconclusive. In this study we sought to investigate the impact of DM on operative mortality and morbidity in patients undergoing cardiac surgery and adjust for patient and disease characteristics. In total 10,709 patients (9,229 nondiabetics and 1,480 diabetics) were admitted to the study; 5,557 patients (1,012 diabetics) underwent an isolated coronary operation, 1,775 patients (278 diabetics) underwent coronary plus valve operations, and 3,337 patients (209 diabetics) underwent valve operations. To control for differences in patient and disease characteristics, a propensity score (for DM) was performed. DM increased crude morbidity and this difference was maintained after risk adjustment for propensity score; conversely, the crude operative mortality risk was higher in diabetics but not signicantly after adjustment for propensity score. Thereafter, DM remained independently associated to operative mortality risk in the valve population only (odds ratio 2.53, 95% condence interval 1.45 to 4.4, p 0.001). In conclusion, DM has a signicant impact on operative mortality of patients undergoing heart valve surgery. Although diabetic patients undergoing coronary operations are not at increased risk of operative mortality, morbidity is signicantly affected in the overall population. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:535539) The global burden of diabetes mellitus (DM) is rapidly increasing and approximately 8% of adults in developed countries have DM.1 Because of the proportion, it is expected that an increasing number of patients with DM will undergo cardiac surgery in the future, making diabetesrelated operative risk assessment an important tool. The impact of DM on the outcome of patients requiring coronary operations has been systematically investigated in previous decades. However, results have been changing over time from series where DM was found to be associated with increased early and 30-day mortalities2,3 to more recent studies where this nding was not conrmed.4,5 More recently, because the prole of cardiac surgical practice is changing (coronary surgery is decreasing, whereas the number of patients requiring valve surgery is increasing)6 in addition to changes in patients risk prole, more attention has been placed on risk assessment of patients undergoing heart valve surgery. However, currently few risk models addressing this important issue have been proposed710 and, most important, these still have strict limitations that can
Departments of aCardiac Surgery and bCardiothoracic and Vascular Anesthesia and ICU, IRCCS Policlinico San Donato, Milan, Italy. Manuscript received July 22, 2010; revised manuscript received and accepted October 5, 2010. *Corresponding author: Tel: 39-02-5277-4842; fax: 39-02-5277-4615. E-mail address: castelvecchio.serenella@gmail.com (S. Castelvecchio). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.009

lead to an underestimation of the weight of DM in risk stratication. In this study we sought to investigate the impact of DM on operative mortality and morbidity in a single large institutional series of patients undergoing cardiac surgery and consider patient and disease characteristics with propensity-score adjustment. Methods This is a retrospective study based on our prospective institutional database of cardiac surgical patients. The local ethics committee approved the study design and waived the need for an informed consent of patients. All patients provided written consent to the scientic treatment of their data in an anonymous form at time of hospitalization. All patients operated from April 2000 to April 2009 were admitted to this study. Patients were assigned to the diabetic or nondiabetic group according to their condition at hospital admission. Exclusion criteria were age 18 years and congenital heart operations. The study population included 10,709 patients; 5,557 patients underwent an isolated coronary operation, 1,775 patients underwent coronary plus valve operations, and 3,337 underwent valve operations (single or multiple, including valve plus ascending aorta operations). Patients with DM were identied as those receiving oral antidiabetic treatment and/or insulin at time of surgery. Patients receiving nutritional modications as the sole treatwww.ajconline.org

536

The American Journal of Cardiology (www.ajconline.org) Table 2 Factors determining propensity for being a diabetic patient at multivariable stepwise forward logistic regression analysis Factor With Diabetes Age (years) Weight (kg) Hematocrit (%) Serum creatinine (mg/dl) Left ventricular ejection fraction Recent myocardial infarction Congestive heart failure Previous vascular surgery Chronic obstructive pulmonary disease Previous cerebrovascular accident Combined operation Isolated coronary artery bypass grafting Constant Regression Coefcient 0.009 0.013 0.053 0.090 0.012 0.262 0.518 0.350 0.282 0.372 0.606 1.140 1.836 p Value for Association 0.002 0.001 0.001 0.002 0.001 0.001 0.001 0.004 0.005 0.002 0.001 0.001

Table 1 Demographics and preoperative and operative details in diabetic and nondiabetic patients Variable Age (years) Men Weight (kg) Hematocrit (%) Serum creatinine (mg/dl) Long-term dialytic treatment Left ventricular ejection fraction Recent myocardial infarction Unstable angina pectoris Congestive heart failure Preoperative intra-aortic balloon pump Active endocarditis Previous vascular surgery Previous cardiac surgery Chronic obstructive pulmonary disease Previous cerebrovascular accident Isolated coronary artery bypass grafting Coronary artery bypass grafting plus valve operation Valve operation Urgent operation Cardiopulmonary bypass duration (minutes) Cross-clamp time (minutes) Nondiabetics (n 9,229) 65.5 11.7 6,406 (69%) 73.1 14.7 39.4 4.5 1.19 0.8 61 (0.7%) 0.52 0.11 1,352 (14.6%) 560 (6.1%) 414 (4.5%) 50 (0.5%) 60 (0.7%) 364 (7.9%) 490 (5.3%) 617 (6.7%) 402 (4.4%) 4,545 (49.2%) 1,497 (16.2%) Diabetics (n 1,480) 67.6 8.5 1,011 (68%) 75.5 14.4 38.1 4.7 1.35 0.9 26 (1.8%) 0.49 0.12 367 (24.8%) 110 (7.4%) 127 (8.6%) 15 (1.0%) 8 (0.5%) 107 (7.2%) 57 (3.9%) 152 (10.3%) 110 (7.4%) 1,012 (68.4%) 278 (18.7%) p Value 0.001 0.394 0.001 0.001 0.001 0.001 0.001 0.001 0.044 0.001 0.030 0.622 0.001 0.018 0.001 0.001 0.001 0.015

3,128 (33.9%) 369 (4.0%) 75.6 37 50 26

209 (14.1%) 72 (4.9%) 72.2 34 45.7 23

0.001 0.119 0.001 0.001

Data expressed as mean SD or number of patients (percentage).

ment for hyperglycemia were not considered diabetic; patients in whom DM was discovered and treated during hospitalization were considered diabetics. Insulin treatment of transient postoperative hyperglycemia was not considered a criterion for being included in the diabetic group. During and after the operation, blood glucose levels were controlled by insulin infusion to maintain a level 180 mg/dl. At the time of surgery, all patients were on state-ofthe-art optimized medical therapy. Mean duration of DM was not available, and glucose tolerance, insulin resistance, and glycosylated hemoglobin levels were not evaluated. Demographic (age, gender, and weight) data were collected. Preoperative laboratory assays included serum creatinine value (milligrams per deciliter) and hematocrit (percentage). Cardiac function was assessed by left ventricular ejection fraction measured before the operation with echocardiographic assessment. For repeated different measurements, the lowest value was used; other cardiac-related factors collected were recent (within 30 days) myocardial infarction, unstable angina, congestive heart failure (HF), preoperative use of intra-aortic balloon pump, and presence of active endocarditis. The following co-morbid conditions

were recorded: chronic obstructive pulmonary disease treated with medication at time of surgery, previous cerebrovascular accident, chronic renal failure on dialytic treatment, previous vascular surgery, and previous cardiac surgery. Operative data recorded were isolated coronary artery bypass graft (CABG) operation, CABG plus valve operation, valve operation, urgent operation, and cardiopulmonary bypass (CPB) duration (minutes). Postoperative outcome data included mechanical ventilation time (hours), intensive care unit stay (days), postoperative hospital stay (days), acute renal failure (peak postoperative serum creatinine level 2.0 mg/dl and 2 times the preoperative value), sepsis, stroke, and surgical re-exploration. Major morbidity was dened according to the STS National Database Risk Stratication Subcommittee as 1 of the following: mechanical ventilation time 48 hours, sepsis or mediastinitis, acute renal failure, stroke, or surgical re-exploration. Myocardial infarction and acute respiratory insufciency were separately considered additional outcome variables. Operative mortality was dened as in-hospital mortality or mortality within 30 days from the operation. Data in text and tables are expressed as mean SD of the mean for continuous variables and as number and percentage for categorical variables. Differences between means were assessed using unpaired Students t test and Pearson chi-square test for differences between percentages. A propensity-score analysis determining the propensity of belonging to the diabetic group was performed with the following steps: (1) factors signicantly different between diabetics and nondiabetics were established as previously stated, (2) factors associated to diabetic conditions (p 0.1) were entered into a multivariable stepwise forward logistic regression analysis, (3) the logistic equation provided a propensity score for likelihood of being diabetic, and (4) this score was used as an adjustment factor for all subsequent analyses. Accuracy of the propensity score was checked using receiver operating curve analysis. Calibration of the propensity score was checked with Hosmer-Lemeshow statistics. Morbidity and mortality rates for each factor in the diabetic and nondiabetic groups were assessed using univariate and multivariable logistic regression analyses, re-

Coronary Artery Disease/Diabetes and Cardiac Surgery Table 3 Outcome data in diabetic and nondiabetic patients Variable Nondiabetics (n 9,229) 26 86 3.1 5.0 6.5 7.1 Diabetics (n 1,480) 34 108 3.8 6.7 7.6 9.9 p Value Crude 0.002

537

Adjusted 0.232

Mechanical ventilation (hours) Intensive care unit stay (days) Hospital stay (days)

0.001 0.001

0.036 0.001

Data expressed as mean SD.

Figure 1. Crude and adjusted values for propensity-score morbidity.

Figure 2. Crude and adjusted values for propensity-score mortality in overall population and in subgroups.

spectively, obtaining crude and adjusted odds ratios (ORs) with 95% condence intervals (CIs). Sensitivity analysis for subgroups of isolated coronary, coronary plus valve, and valve operations was performed. Statistical signicance was settled at a p value 0.05; all tests were 2-sided; statistical analyses were performed with SPSS 13.0 (SPSS, Inc., Chicago, Illinois). Results The number of diabetic patients in our population was 1,480 (13.8%), with 9,229 nondiabetic patients. The 2 groups differed signicantly in demographics, co-morbidities, and operative details (Table 1). The diabetic group had a more severe risk prole due to older age, lower baseline hematocrit and ejection fraction, higher serum creatinine level, higher rate of many co-morbidities (long-term dialytic treatment, recent myocardial infarction, unstable angina, congestive HF, previous vascular and cardiac operations,

chronic obstructive pulmonary disease, previous cerebrovascular accident), and a longer CPB duration. Patients in the nondiabetic group had a slightly lower rate of CABG plus valve operations, a lower rate of isolated CABG operations, and a higher rate of valve operations. Twelve factors (Table 2) were independently associated with diabetes and were used for developing the propensity score. According to this score, the patient population was stratied for the propensity of belonging to the diabetic group, with a likelihood of 1.6% to 85.4%. Accuracy of the propensity score was conrmed by receiver operating curve analysis, which revealed an area under the curve of 0.72 (95% CI 0.706 to 0.734, p 0.001) and the calibration was conrmed by Hosmer-Lemeshow statistics (chi-square 11.5, p 0.175). At univariate analysis, diabetic patients demonstrated a worse outcome, with longer mechanical ventilation time, longer intensive care unit and hospital stay, and higher rate of major morbidity, acute renal failure, stroke, and mortality (Figures 1 and 2, Table 3). Rates of myocardial infarction were 2% in nondiabetic patients and 1.6% in diabetic patients (p 0.356). Conversely, diabetic patients showed a higher rate of acute respiratory insufciency (3.7% vs 2.5% in nondiabetic patients, p 0.005). When adjusted for propensity score, diabetic patients still demonstrated a longer hospital stay and a higher risk of major morbidity and acute renal failure but not a different mortality risk. The OR for major morbidity was 1.19 (95% CI 1.02 to 1.39, p 0.024) and the OR for acute renal failure was 1.36 (95% CI 1.08 to 1.32, p 0.01). Sensitivity analysis on mortality was performed for isolated coronary operations, coronary plus valve operations, and valve operations. There were 5,557 patients (1,012 diabetics, 18.2%) who underwent an isolated coronary operation, 1,775 (278 diabetics, 15.6%) who underwent a combined operation, and 3,337 (209 diabetics, 6.3%) who underwent a valve operation. Within groups of patients undergoing coronary operations, DM was a signicant risk factor for operative mortality at univariate analysis. After adjustment for propensity score, DM remained independently associated to operative mortality risk in the valve population only (OR 2.53, 95% CI 1.45 to 4.4, p 0.001). Operative mortality rate in diabetic patients undergoing a valve operation was 9.4% (18 patients). Thirteen patients had a cardiac cause for mortality, with low cardiac output and multiorgan failure, 3 patients had refractory liver fail-

538

The American Journal of Cardiology (www.ajconline.org)

ure, 1 had severe sepsis, and 1 died intraoperatively due to inability to wean from CPB. Discussion There are major ndings of this study. (1) DM increases crude morbidity and this difference is maintained after risk adjustment for propensity score. (2) Conversely, the crude operative mortality risk is higher in diabetics but not signicantly after adjustment for propensity score. (3) However, when cardiac operations were split into coronary, coronary combined with valve, and valve operations, operative mortality risk was signicantly higher in diabetic patients undergoing valve operations even after adjustment for the propensity score. Results of the present study partly agree with those of other investigators. This may be related to different denitions of morbidity or different subsets of patients. Moreover, operative conduct may have been different according to different institutional protocols. Thus far, we found an association between DM and postoperative stroke in agreement with Szab et al11 and Herlitz et al12 who found the incidence of stroke to be signicantly increased in diabetic patients undergoing CABG. Conversely, Kubal et al4 found no association between DM and stroke; they speculated that the results could be explained with the higher percentage of diabetic patients undergoing off-pump CABG in agreement with previous reports showing a signicant decrease in stroke when CPB was avoided.13 However, more recently, the Randomized On/Off Bypass (ROOBY) trial showed no difference between on-pump and off-pump procedures in occurrence of the primary short-term end point including stroke.14 Previous studies have reported an association between DM and postoperative renal insufciency in coronary11 or valve15 surgery. More specically, Grayson et al15 found that insulin-dependent DM was an independent risk factor for developing acute renal failure in all cardiac surgery with an adjusted OR of 3.31 (95% CI 1.75 to 6.26, p 0.001). Our ndings partly concur with that; however, lack of information regarding type of DM in our dataset makes the comparison inappropriate. Previous studies have reported controversial results on the impact of DM on early mortality of patients undergoing CABG surgery. Thourani et al3 in the late 1990s reported an observed mortality of 3.9% in diabetic patients after CABG surgery and found DM to be an independent predictor of in-hospital mortality. Other studies from previous series had similar ndings.2,16 More recent studies have reported no increased risk of in-hospital mortality for diabetic patients undergoing CABG surgery,4,5 suggesting different reasons in interpreting the results as advances in surgical, perfusion, and anesthetic techniques and/or strict perioperative glucose management. Our results obtained in a large, more recent series conrm that operative mortality risk, adjusted for propensity score, is not affected by DM in the overall population or in patients undergoing isolated coronary operations or coronary plus valve operations. However, although coronary artery disease is the most common cardiac manifestation in diabetic patients, DM appears to be strongly linked to HF and it has a greater

impact on the prognosis of HF than of coronary artery disease.17 DM and congestive HF commonly coexist; each condition increases the likelihood of developing the other, and when they occur together in the same patient risk of mortality increases markedly.18 20 Our data conrm the relation between DM and HF with the percentage of patients with congestive HF being higher in diabetic patients compared to nondiabetics. However, such differences did not similarly affect operative mortality risk when cardiac operations were split into coronary, coronary combined with valve, and valve operations. It is well known that valve surgery produces higher operative mortality, from 3% to 7% in symptomatic patients.21 This nding partly supports our data showing an even higher operative mortality rate in diabetic patients undergoing a valve operation (9.4%); moreover, operative mortality risk remained signicantly higher in diabetic patients undergoing valve operations even after adjustment for propensity score. At least 2 explanations may account for this nding. Patients affected by valve disease without coronary artery involvement are usually referred to surgery when symptoms of congestive HF occur, further increasing operative mortality risk in patients with a higher percentage of congestive HF due to DM. Alternatively, the existence of a primary myocardial disease in diabetic patients22 leading to left ventricular dysfunction, which in turn could increase operative risk, cannot be excluded. Therefore, it could be hypothesized that an early postoperative prognosis is more affected by the underlying pathophysiologic disease rather than by the surgery itself. This study has some limitations. First, DM was not systematically assessed using standardized diagnostic criteria. Mean duration of DM was not available, and glucose tolerance, insulin resistance, and glycosylated hemoglobin levels were not evaluated, limiting the accuracy of our analysis. However, all patients were followed by the referring clinician before being scheduled for cardiac surgery, and the diagnosis of DM was known before hospital admittance. Second, this is a retrospective observational study and cannot account for all variables affecting outcome or variables included in this analysis might be overestimated. However, retrospective comparisons with propensity-score adjustment are recognized as highly robust and may in some cases be acceptable as randomized control trials.
1. Harris MI, Flegal KM, Cowie CC, Eberhardt MS, Goldstein DE, Little RR, Wiedmeyer HM, Byrd-Holt DD. Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in U.S. adults. The Third National Health and Nutrition Examination Survey, 1988 1994. Diabetes Care 1998;21:518 524. 2. Cohen Y, Raz I, Merin G, Mozes B. Comparison of factors associated with 30-day mortality after coronary artery bypass grafting in patients with versus without diabetes mellitus. Israeli Coronary Artery Bypass (ISCAB) Study Consortium. Am J Cardiol 1998;81:711. 3. Thourani VH, Weintraub WS, Stein B, Gebhart SS, Craver JM, Jones EL, Guyton RA. Inuence of diabetes mellitus on early and late outcome after coronary artery bypass grafting. Ann Thorac Surg 1999; 67:10451052. 4. Kubal C, Srinivasan AK, Grayson AD, Fabri BM, Chalmers JA. Effect of risk-adjusted diabetes on mortality and morbidity after coronary artery bypass surgery. Ann Thorac Surg 2005;79:1570 1576. 5. Rajakaruna C, Rogers CA, Suranimala C, Angelini GD, Ascione R. The effect of diabetes mellitus on patients undergoing coronary surgery: a risk-adjusted analysis. J Thorac Cardiovasc Surg 2006;132: 802 810.

Coronary Artery Disease/Diabetes and Cardiac Surgery 6. Rabkin E, Schoen FJ. Cardiovascular tissue engineering. Cardiovasc Pathol 2002;11:305317. 7. Edwards FH, Peterson ED, Coombs LP, DeLong ER, Jamieson WR, Shroyer ALW, Grover FL. Prediction of operative mortality after valve replacement surgery. J Am Coll Cardiol 2001;37:885 892. 8. Nowicki ER, Birkmeyer NJ, Weintraub RW, Leavitt BJ, Sanders JH, Dacey LJ, Clough RA, Quinn RD, Charlesworth DC, Sisto DA, Uhlig PN, Olmstead EM, OConnor GT; Northern New England Cardiovascular Disease Study Group and the Center for Evaluative Clinical Sciences, Dartmouth Medical School. Multivariable prediction of inhospital mortality associated with aortic and mitral valve surgery in Northern New England. Ann Thorac Surg 2004;77:1966 1977. 9. Florath I, Rosendahl UP, Mortasawi A, Bauer SF, Dalladaku F, Ennker IC, Ennker JC. Current determinants of operative mortality in 1400 patients requiring aortic valve replacement. Ann Thorac Surg 2003; 76:75 83. 10. Ambler G, Omar RZ, Royston P, Kinsman R, Keogh BE, Taylor KM. Generic, simple risk stratication model for heart valve surgery. Circulation 2005;112:224 231. 11. Szab Z, Hkanson E, Svedjeholm R. Early postoperative outcome and medium-term survival in 540 diabetic and 2239 nondiabetic patients undergoing coronary artery bypass grafting. Ann Thorac Surg 2002;74:712719. 12. Herlitz J, Wognsen GB, Emanuelsson H, Haglid M, Karlson BW, Karlsson T, Albertsson P, Westberg S. Mortality and morbidity in diabetic and nondiabetic patients during a 2-year period after coronary artery bypass grafting. Diabetes Care 1996;19:698 703. 13. Patel NC, Deodhar AP, Grayson AD, Pullan DM, Keenan DJ, Hasan R, Fabri BM. Neurological outcomes in coronary surgery: independent effect of avoiding cardiopulmonary bypass. Ann Thorac Surg 2002; 74:400 406. 14. Shroyer AL, Grover FL, Hattler B, Collins JF, McDonald GO, Kozora E, Lucke JC, Baltz JH, Novitzky D. Veterans Affairs Randomized

539

15. 16.

17.

18.

19.

20.

21.

22.

On/Off Bypass (ROOBY) Study Group. On-pump versus off-pump coronary-artery bypass surgery. N Engl J Med 2009;361:18271837. Grayson AD, Khater M, Jackson M, Fox MA. Valvular heart operation is an independent risk factor for acute renal failure. Ann Thorac Surg 2003;75:1829 1835. Morris JJ, Smith LR, Jones RH, Glower DD, Morris PB, Muhlbaier LH, Reves JG, Rankin JS. Inuence of diabetes and mammary artery grafting on survival after coronary bypass. Circulation 1991; 84(suppl):III275III284. De Groote P, Lamblin N, Mouquet F, Plichon D, McFadden E, Van Belle E, Bauters C. Impact of diabetes mellitus on long-term survival in patients with congestive heart failure. Eur Heart J 2004;25:656 662. MacDonald MR, Petrie MC, Hawkins NM, Petrie JR, Fisher M, McKelvie R, Aguilar D, Krum H, McMurray JJ. Diabetes, left ventricular systolic dysfunction, and chronic heart failure. Eur Heart J 2008;29:1224 1240. Gustafsson I, Brendorp B, Seibaek M, Burchardt H, Hildebrandt P, Kber L, Torp-Pedersen C. Danish Investigators of Arrhythmia and Mortality on Dofetilide Study Group. Inuence of diabetes and diabetes-gender interaction on the risk of death in patients hospitalized with congestive heart failure. J Am Coll Cardiol 2004;43:771777. Pocock SJ, Wang D, Pfeffer MA, Yusuf S, McMurray JJ, Swedberg KB, Ostergren J, Michelson EL, Pieper KS, Granger CB. Predictors of mortality and morbidity in patients with chronic heart failure. Eur Heart J 2006;27:6575. Vahanian A, Baumgartner H, Bax J, Butchart E, Dion R, Filippatos G, Flachskampf F, Hall R, Iung B, Kasprzak J, Nataf P, Tornos P, Torracca L, Wenink A. Guidelines on the management of valvular heart disease: the Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology. Eur Heart J 2007;28: 230 268. Boudina S, Abel ED. Diabetic cardiomyopathy revisited. Circulation 2007;115:32133223.

Relation of Bundle Branch Block to Long-Term (Four-Year) Mortality in Hospitalized Patients With Systolic Heart Failure
Alon Barsheshet, MDa,*, Ilan Goldenberg, MDa, Moshe Garty, MD, MScb, Shmuel Gottlieb, MDc, Amir Sandach, PhDd, Avishag Laish-Farkash, MDc, Michael Eldar, MDc, and Michael Glikson, MDc
There is controversy regarding type of bundle branch block (BBB) that is associated with increased mortality risk in patients with heart failure (HF). The present study was designed to explore the association between BBB pattern and long-term mortality in hospitalized patients with systolic HF. Risk of 4-year all-cause mortality was assessed in 1,888 hospitalized patients with systolic HF (left ventricular ejection function <50%) without a pacemaker in a prospective national survey. Cox proportional hazards regression modeling was used to compare mortality risk in patients with right BBB (RBBB; 10%), left BBB (LBBB; 14%), and no BBB (76%) on admission electrocardiogram. At 4 years of follow up, mortality rates were highest in patients with RBBB (69%), intermediate in those with LBBB (63%), and lowest in those without BBB (50%, p <0.001). Multivariate analysis demonstrated a signicant 36% increased mortality risk in patients with RBBB versus no BBB (p 0.002) but no signicant difference in mortality risk for patients with LBBB versus no BBB (hazard ratio 1.04, p 0.66). RBBB versus LBBB was associated with a 29% (p 0.035) increased risk for 4-year mortality in the total population and with a 58% (p 0.015) increased risk in patients with ejection fraction <30%. In conclusion, RBBB but not LBBB on admission electrocardiogram is associated with a signicant increased long-term mortality risk in hospitalized patients with systolic HF. Deleterious effects of RBBB compared to LBBB appear to be more pronounced in patients with more advanced left ventricular dysfunction. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:540 544) Prolongation of QRS interval (120 ms) in patients with heart failure (HF) is common (14% to 47%)1 and is associated with higher all-cause mortality, cardiovascular death, or hospitalization for HF compared to patients with HF and normal QRS interval.2,3 There is controversy regarding type of bundle branch block (BBB) that is associated with poorer outcome in patients with HF,4 8 with most studies showing that left BBB (LBBB) is an independent prognostic marker, whereas right BBB (RBBB) is a weaker marker or not associated with worse prognosis. Conversely, we previously showed in hospitalized patients with HF that RBBB, but not LBBB, is associated with increased 1-year mortality risk, an association that was stronger for patients with systolic HF, particularly for patients with severe left ventricular (LV) dysfunction.9 However, currently there are limited data regarding the effect of BBB pattern on long-term mortality in
Cardiology Division, University of Rochester Medical Center, Rochester, New York; bRecanati Center for Internal Medicine and Research, Rabin Medical Center, Petah Tiqva, Israel; cHeart Institute, Sheba Medical Center and Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel; dDepartment of Mathematics, Bar-Ilan University, Ramat Gan, Israel. Manuscript received August 14, 2010; revised manuscript received and accepted October 1, 2010. The Heart Failure Survey in Israel 2003 was supported by the Israel Center for Disease Control, Ramat Gan, Israel; The Israel Medical Association, Ramat Gan, Israel; Teva, Petah Tiqva, Israel; Levant, Herzelia, Israel; Neopharm, Petah Tiqva, Israel; Pzer, Herzelia, Israel; Aventis, Netania, Israel; Dexxon, Or Akiva, Israel; Medisson, Petah Tiqva, Israel; Novartis, Petah Tiqva, Israel; and Schering-Plough, Petah Tiqva, Israel. *Corresponding author: Tel: 585-276-5228; fax: 585-273-5283. E-mail address: alon.barsheshet@heart.rochester.edu (A. Barsheshet). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.007
a

patients with LV dysfunction. Accordingly, the present study aimed to investigate the association between QRS morphology and long-term mortality in 1,888 patients hospitalized with systolic HF who were prospectively followed-up over an extended 4-year period. Methods Baseline and admission characteristics of patients were extracted from the Heart Failure Survey in Israel (HFSIS; 2003) database. Design and methods of the HFSIS registry have been described previously.10 Briey, the survey, conducted in March and April 2003, included 4,102 patients admitted with a diagnosis of HF. Criteria used for diagnosis of HF were symptoms of HF (at rest or during exercise) and objective evidence of cardiac dysfunction at rest.11 There were 3 subgroups of diagnoses for hospitalized patients: (1) acute de novo HF, (2) worsening of chronic HF, and (3) chronic stable HF with hospitalization unrelated to HF exacerbation. There were 2,090 patients with HF and LV ejection fraction (LVEF) 50% as demonstrated by echocardiography. We excluded from the study 188 patients who had a permanent pacemaker including a biventricular pacemaker and 14 patients who lacked electrocardiographic data. Thus, the nal analysis included 1,888 patients. The end point of the study was all-cause mortality, which was assessed for all patients by matching their identication numbers with the Israeli National Population Registry. Mortality data were obtained for all study patients at a 4-year period from hospitalization, providing an extended follow-up to the previously reported 1-year outcome study.9
www.ajconline.org

Heart Failure/BBB and Mortality in Heart Failure Patients Table 1 Clinical characteristics of study patients by bundle branch block pattern Variable Total (n 1,888) Bundle Branch Block Left (n 306, 14%) 76 (6881) 33% 65% 55% 31% 82% 51% 55% 13% 17% 29% 60% Right (n 193, 10%) 74 (6781) 25% 69% 55% 38% 82% 45% 67% 11% 17% 28% 57% None (n 1,389, 76%) 71 (6279) 35% 64% 53% 36% 82% 35% 62% 14% 18% 24% 61%

541

p Value*

History Age (years) Women Hypertension Diabetes mellitus Smoker Coronary heart disease New York Heart Association functional class III to IV Previous myocardial infarction Previous stroke Chronic obstructive pulmonary disease Atrial brillation Acute heart failure Left ventricular ejection fraction estimated by echocardiography (%) 4049 3039 30 Admission systolic blood pressure (mm Hg) Admission heart rate (beats/min) Admission laboratory values Creatinine (mg/dl) Sodium (mmol/L) Hemoglobin (g/dl) Long-term medications Blockers Angiotensin-converting enzyme inhibitors or angiotensin receptor blockers Furosemide Spironolactone Digoxin Statins

73 (6380) 33% 65% 53% 35% 82% 40% 62% 13% 18% 25% 61%

0.001 0.026 0.392 0.554 0.144 0.995 0.001 0.021 0.563 0.855 0.120 0.588

30% 36% 34% 135 (118157) 81 (7098) 1.2 (0.91.6) 138 (136141) 12.5 (11.013.8) 72% 81% 75% 25% 17% 53%

15% 34% 51% 135 (118153) 80 (6995) 1.3 (1.01.8) 138 (135141) 12.6 (11.213.7) 69% 83% 89% 34% 28% 52%

20% 40% 40% 131 (116156) 80 (6892) 1.4 (1.11.8) 138 (135141) 12.5 (11.313.8) 71% 82% 81% 32% 23% 43%

35% 36% 29% 136 (119158) 82 (70100) 1.2 (0.91.6) 139 (136141) 12.4 (10.913.8) 73% 80% 71% 22% 14% 55%

0.001

0.362 0.061 0.160 0.903 0.563 0.332 0.510 0.001 0.001 0.001 0.005

Data are presented as median (interquartile range) or percentage of patients. * For overall difference among the 3 subgroups. To convert creatinine to micromoles per liter, multiply by 88.4.

LBBB was dened as QRS duration 120 ms, upright complexes with notched R waves in leads I, V5, and V6, and QS or rS pattern in lead V1. RBBB was dened as QRS duration 120 ms, a monophasic R wave in lead V1 or rSR in leads V1 and V2, and deep slurred S waves in leads I, V5, and V6. LVEF classes determined by echocardiography with visual assessment were classied as normal (50%), mildly impaired (40% to 49%), moderately impaired (30% to 39%), and severely impaired (30%). Median and interquartile range timing of echocardiography were 0 month and 0 month to 6 months. Characteristics of patients categorized by BBB type were compared by nonparametric Kruskal-Wallis test or chisquare test. Cumulative probability of survival by BBB type was graphically displayed according to the Kaplan-Meier method with comparison by log-rank test. To examine the relation between RBBB, LBBB, and no BBB and mortality, several models were applied. First, potential variables (identied in previous published studies as risk factors for mortality or clinical variables that were associated with mortal-

ity) were evaluated by univariate analysis and selected based on clinical and statistical signicance. Second, multivariate analysis was carried out using Cox proportional hazards regression modeling adjusted for age (continuous), gender, New York Association (NYHA) functional classes III to IV versus I to II, previous myocardial infarction, atrial brillation, previous stroke, diabetes, chronic obstructive pulmonary disease, cirrhosis, malignant tumor, LVEF class, admission creatinine levels (continuous), systolic blood pressure 115 versus 115 mm Hg, sodium 136 versus 136 mEq/L, hemoglobin 10 versus 10 g/dl and long-term use of statins, blockers, and angiotensin-converting enzyme inhibitors or angiotensin receptor blockers. Analyses were conducted with SAS 9.2 (SAS Institute, Cary, North Carolina). Results Of the 1,888 patients with systolic HF, 306 (14%) had LBBB on admission electrocardiogram and 193 (10%) had RBBB. Table 1 presents baseline clinical characteristics of

542

The American Journal of Cardiology (www.ajconline.org)

Figure 1. Kaplan-Meier survival curves for patients with heart failure and left ventricular ejection fraction 50% analyzed according to bundle branch block patterns. Table 2 Multivariate analysis: bundle branch block pattern as a predictor of four-year mortality Variable Left ventricular ejection fraction 50% All patients Right versus no bundle branch block Left versus no bundle branch block Right versus left bundle branch block Left ventricular ejection fraction 30% All patients Right versus no bundle branch block Left versus no bundle branch block Right versus left bundle branch block Number of Patients Crude Mortality Hazard Ratio (95% condence interval) 1.36 (1.121.65) 1.04 (0.881.23) 1.29 (1.021.64) 1.55 (1.142.11) 1.03 (0.811.31) 1.58 (1.092.28) p Value

1,800 182 vs 1,322 296 vs 1,322 182 vs 296 609 70 vs 389 150 vs 389 70 vs 150

976 (54%) 124 (68%) vs 665 (50%) 187 (63%) vs 665 (50%) 124 (68%) vs 187 (63%) 396 (65%) 54 (77%) vs 242 (62%) 100 (67%) vs 242 (62%) 54 (77%) vs 100 (67%)

0.002 0.655 0.035 0.005 0.823 0.015

Adjusted for age, gender, New York Heart Association functional classes III and IV, admission creatinine levels, sodium level 136 mmol/L, hemoglobin 10 g/dl, systolic blood pressure 115 mm Hg, left ventricular ejection fraction class, previous myocardial infarction, history of atrial brillation, history of stroke, diabetes mellitus, chronic obstructive pulmonary disease, malignancy, hepatic cirrhosis, and long-term use of blockers, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, and statins. Thirty-seven patients were missing New York Heart Association functional class data, 23 were missing sodium levels, 10 were missing creatinine levels, and 10 were missing hemoglobin levels.

patients indicating that patients with LBBB were older than those with RBBB or no BBB. Prevalence of NYHA functional classes III to IV and low EF was highest in LBBB, lower in RBBB, and lowest in no BBB. Long-term therapy with furosemide, spironolactone, and digoxin was more frequent in patients with LBBB compared to those with RBBB or no BBB. In the RBBB subgroup the proportion of women was smaller than in the LBBB or no-BBB subgroups. Systolic pulmonary arterial pressure data were available for 841 patients (45%). There was a trend toward a higher systolic pulmonary arterial pressure in patients with RBBB compared to those with LBBB or no BBB (mean SD 46 14, 43 15, and 43 16 mm Hg, respectively, p 0.107 for overall comparison among the 3 subgroups). For the total study population, Kaplan-Meier survival curves (Figure 1) demonstrated that during the rst 8

months of follow-up mortality rates were higher in patients with RBBB and LBBB compared to those who had no BBB. After 8 months and until the end of the extended 4-year follow-up period, curves of RBBB and LBBB separated, showing graded decrements in survival in patients with RBBB, LBBB, and no BBB, respectively. Mortality rates at 4 years were 69%, 63%, and 50% in the RBBB, LBBB, and no-BBB groups, respectively (p 0.001; Figure 1). Consistent with these ndings, multivariate analysis (Table 2) showed a signicant increase in mortality risk in patients with RBBB compared to those with no BBB, whereas no statistically signicant difference in mortality risk was observed between patients with LBBB and those with no BBB. Comparison of the outcome between patients with RBBB and those with LBBB showed a signicant 29% (p 0.035) increase in long-term mortality risk in the former subgroup

Heart Failure/BBB and Mortality in Heart Failure Patients Table 3 Independent predictors of four-year mortality for patients with heart failure and left ventricular ejection fraction 50% and 30% Variable 50% Hazard Ratio (95% condence interval) Right versus no bundle branch block Left versus no bundle branch block Age per decade New York Heart Association class III to IV Left ventricular ejection fraction 30% Diabetes mellitus Chronic obstructive pulmonary disease Creatinine per 1 mg/dl* Sodium 136 mmol/L Systolic blood pressure 115 mm Hg Hepatic cirrhosis Malignancy 1.36 (1.121.65) 1.04 (0.881.23) 1.42 (1.331.52) 1.62 (1.421.85) 1.47 (1.241.74) 1.47 (1.291.68) 1.40 (1.191.64) 1.22 (1.161.29) 1.20 (1.041.38) 1.44 (1.241.68) 1.63 (1.162.27) 1.63 (1.302.04) p Value 0.002 0.655 0.001 0.001 0.001 0.001 0.001 0.001 0.012 0.001 0.004 0.001 Left Ventricular Ejection Fraction 30% Hazard Ratio (95% condence interval) 1.55 (1.142.11) 1.03 (0.811.31) 1.31 (1.181.44) 1.45 (1.181.78) 1.30 (1.051.61) 1.59 (1.242.03) 1.15 (1.051.26) 1.26 (1.011.57) 1.38 (1.121.72) 2.07 (1.233.46) 1.62 (1.082.41)

543

p Value 0.005 0.823 0.001 0.001 0.016 0.001 0.004 0.043 0.003 0.006 0.019

* To convert creatinine to micromoles per liter, multiply by 88.4.

Figure 2. Kaplan-Meier survival curves for patients with heart failure and left ventricular ejection fraction 30% analyzed according to bundle branch block patterns.

compared to the latter subgroup. Similar results were found for the 2 subgroups of acute and chronic HF. Notably, RBBB was associated with a similar magnitude of risk increase as demonstrated for other known predictors of mortality risk in this population including age, diabetes mellitus, advanced LV dysfunction, or renal dysfunction (Table 3). Long-term mortality risk associated with RBBB compared to no BBB or LBBB on admission electrocardiogram was even more pronounced in patients with more advanced LV dysfunction (LVEF 30%). Kaplan-Meier survival curves showed a signicant increase in 4-year mortality in patients with RBBB compared to those with LBBB or no BBB throughout follow-up (77%, 66%, and 62%, respec-

tively, p 0.008; Figure 2). Multivariate analysis (Table 2) comparing long-term outcome of patients with BBB showed a 58% increase in mortality risk (p 0.015) in patients with RBBB compared to those who had LBBB. Discussion The present study is the rst to assess the long-term effect of BBB pattern on mortality risk in hospitalized patients with systolic HF. The ndings of the present study extend our previous observation regarding the short-term (1-year) outcome of patients with HF and RBBB.9 We have shown that in patients with systolic HF RBBB is associated with a signicant increased long-term (4-years) mortality

544

The American Journal of Cardiology (www.ajconline.org) 1. Kashani A, Barold SS. Signicance of QRS complex duration in patients with heart failure. J Am Coll Cardiol 2005;46:21832192. 2. Wang NC, Maggioni AP, Konstam MA, Zannad F, Krasa HB, Burnett JC Jr, Grinfeld L, Swedberg K, Udelson JE, Cook T, Traver B, Zimmer C, Orlandi C, Gheorghiade M. Clinical implications of QRS duration in patients hospitalized with worsening heart failure and reduced left ventricular ejection fraction. JAMA 2008;299:2656 2566. 3. Iuliano S, Fisher SG, Karasik PE, Fletcher RD, Singh SN. QRS duration and mortality in patients with congestive heart failure. Am Heart J 2002;143:10851091. 4. McCullough PA, Hassan SA, Pallekonda V, Sandberg KR, Nori DB, Soman SS, Bhatt S, Hudson MP, Weaver WD. Bundle branch block patterns, age, renal dysfunction, and heart failure mortality. Int J Cardiol 2005;102:303308. 5. Baldasseroni S, Gentile A, Gorini M, Marchionni N, Marini M, Masotti G, Porcu M, Maggioni AP. Intraventricular conduction defects in patients with congestive heart failure: left but not right bundle branch block is an independent predictor of prognosis. A report from the Italian Network on Congestive Heart Failure (IN-CHF database). Ital Heart J 2003;4:607 613. 6. Baldasseroni S, Opasich C, Gorini M, Lucci D, Marchionni N, Marini M, Campana C, Perini G, Deorsola A, Masotti G, Tavazzi L, Maggioni AP. Left bundle-branch block is associated with increased 1-year sudden and total mortality rate in 5517 outpatients with congestive heart failure: a report from the Italian network on congestive heart failure. Am Heart J 2002;143:398 405. 7. Mueller C, Laule-Kilian K, Klima T, Breidthardt T, Hochholzer W, Perruchoud AP, Christ M. Right bundle branch block and long-term mortality in patients with acute congestive heart failure. J Intern Med 2006;260:421 428. 8. Zimetbaum PJ, Buxton AE, Batsford W, Fisher JD, Haey GE, Lee KL, OToole MF, Page RL, Reynolds M, Josephson ME. Electrocardiographic predictors of arrhythmic death and total mortality in the multicenter unsustained tachycardia trial. Circulation 2004;110:766 769. 9. Barsheshet A, Leor J, Goldbourt U, Garty M, Schwartz R, Behar S, Luria D, Eldar M, Glikson M. Effect of bundle branch block patterns on mortality in hospitalized patients with heart failure. Am J Cardiol 2008;101:13031308. 10. Barsheshet A, Garty M, Grossman E, Sandach A, Lewis BS, Gottlieb S, Shotan A, Behar S, Caspi A, Schwartz R, Tenenbaum A, Leor J. Admission blood glucose level and mortality among hospitalized nondiabetic patients with heart failure. Arch Intern Med 2006;166:1613 1619. 11. Remme WJ, Swedberg K. Comprehensive guidelines for the diagnosis and treatment of chronic heart failure. Task force for the diagnosis and treatment of chronic heart failure of the European Society of Cardiology. Eur J Heart Fail 2002;4:1122. 12. Abdel-Qadir HM, Tu JV, Austin PC, Wang JT, Lee DS. Bundle branch block patterns and long-term outcomes in heart failure. Int J Cardiol 2010. [Epub ahead of print]. DOI:10.1016/j.ijcard.2010.01.012. 13. Petrov DB. Appearance of right bundle branch block in electrocardiograms of patients with pulmonary embolism as a marker for obstruction of the main pulmonary trunk. J Electrocardiol 2001;34:185188. 14. Ocal A, Yildirim N, Ozbakir C, Saricam E, Ozdogan OU, Arslan S, Tufekcioglu O, Sabah I. Right bundle branch block: a new parameter revealing the progression rate of mitral stenosis. Cardiology 2006;105: 219 222. 15. Abd El Rahman MY, Abdul-Khaliq H, Vogel M, Alexi-Meskishvili V, Gutberlet M, Lange PE. Relation between right ventricular enlargement, QRS duration, and right ventricular function in patients with tetralogy of Fallot and pulmonary regurgitation after surgical repair. Heart 2000;84:416 420. 16. Robalino BD, Whitlow PL, Underwood DA, Salcedo EE. Electrocardiographic manifestations of right ventricular infarction. Am Heart J 1989;118:138 144. 17. Grines CL, Bashore TM, Boudoulas H, Olson S, Shafer P, Wooley CF. Functional abnormalities in isolated left bundle branch block. The effect of interventricular asynchrony. Circulation 1989;79:845 853.

risk compared to a no-BBB or LBBB pattern on baseline electrocardiogram. Furthermore, our data suggest that the association between RBBB and increased mortality risk is more pronounced in patients with a lower LVEF. Prolonged QRS interval was shown to be an independent predictor of high postdischarge morbidity and mortality in patients hospitalized due to HF.2,3 Several studies investigating the predictive value of QRS morphology in patients with HF yielded conicting results regarding mortality risk associated with BBB pattern.4 8 Some studies have shown that in patients with HF the prevalence of LBBB is higher than in those with RBBB (13% to 25% vs 6% to 14%, respectively),4,5,7 LBBB is associated with more severe HF characterized by advanced NYHA functional class and decreased LVEF, whereas RBBB is more prevalent in men and is not associated with advanced HF symptoms or ventricular dysfunction.5,6,12 In contrast to previous studies, we have shown that RBBB but not LBBB is an independent predictor of 4-year mortality. Notably, RBBB compared to LBBB was associated with a 29% increase (p 0.035) in long-term mortality in patients with LVEF 50% and a 58% increase (p 0.015) in long-term mortality in patients with LVEF 30%. Contrasting results among the different studies may be related to important differences in characteristics among the populations studied. Notably, the present study consisted mostly of elderly men with coronary heart disease, which is a relatively unique population. Acquired RBBB is often associated with pulmonary hypertension13,14 and right-sided HF,15,16 whereas presence of acquired LBBB in patients with HF is more closely correlated with LV structure and function.17 Consistently in the present study we observed a trend toward higher systolic pulmonary arterial pressure in patients with RBBB compared to those with LBBB or no BBB. We did not collect information regarding clinical or echocardiographic presence of right-sided HF. However, patients with left systolic HF and RBBB may have higher systolic pulmonary artery pressure and worse right ventricular function, leading to increased mortality. Furthermore, in the present study the risk associated with RBBB was more pronounced in patients with more advanced LV dysfunction. Thus, in this population coexistence of RBBB with advanced LV dysfunction may represent biventricular failure. In contrast, presence of LBBB may not provide incremental prognostic information in patients who already have more advanced LV dysfunction.7 The present study has several limitations; we could not determine the type of death in the study population and accordingly did not determine the mechanism related to the observed increase in long-term mortality risk. Furthermore, because we did not collect QRS duration data, except for criteria used for RBBB and LBBB (which included QRS duration 120 ms), we were unable to determine whether QRS duration or type of BBB is a better predictor of mortality.

Characteristics of Depression Remission and Its Relation With Cardiovascular Outcome Among Patients With Chronic Heart Failure (from the SADHART-CHF Study)
Wei Jiang, MDa,b,*, Ranga Krishnan, MDb, Maragatha Kuchibhatla, PhDc, Michael S. Cuffe, MDa, Carolyn Martsberger, PhDb, Rebekka M. Arias, BSd, and Christopher M. OConnor, MDa, for the SADHART-CHF Investigators
Depression is prevalent in patients with heart failure and is associated with a signicant increase in hospitalizations and death. Primary results of the Sertraline Against Depression and Heart Disease in Chronic Heart Failure (SADHART-CHF) trial revealed that sertraline and placebo had comparable effects on depression and cardiovascular outcomes. In this study, we explored whether remission from depression was associated with better survival and aimed to characterize participants who remitted during the trial. Based on depression response during the 12-week treatment phase, SADHART-CHF participants were divided into 2 groups: (1) remission, dened as participants whose last measured Hamilton Depression Rating Scale (HDRS) score was <8, and (2) nonremission, dened as participants whose last measured HDRS score was >8. Patients who dropped out before having any repeat HDRS were not included. Baseline characteristics and survival differences up to 5 years were evaluated between the remission and nonremission groups. Of the 469 SADHART-CHF participants, 208 (44.3%) achieved remission, 194 (41.4%) remained depressed, and 67 (14.3%) dropped out or died without any repeat HDRS assessment. Patients in the remission group had signicantly fewer cardiovascular events than those in the nonremission group (1.34 1.86 vs 1.93 2.71, adjusted p 0.01). Men patients were more likely to remit than women patients (56.5 vs 44.8%, p 0.02). The remission group had milder depressive symptoms at baseline compared to the nonremission group (HDRS 17.0 5.4 vs 19.6 5.5, Beck Depression Inventory scale 17.9 6.5 vs 20.3 7.2, p <0.001 for the 2 comparisons). In conclusion, this study indicates that remission from depression may improve the cardiovascular outcome of patients with heart failure. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:545551) Depression is a common and well-documented co-morbidity in patients with heart failure (HF).1,2 It is associated with substantial morbidity and mortality and with lower quality of life and functional status.29 The adverse relation of depression to HF is independent of HF cause and other conventional risk factors. Sertraline appeared to be safe in depressed patients with HF10; however, it did not demonstrate any superiority over placebo for depression and HF survival in the Sertraline Against Depression and Heart Disease in Chronic Heart Failure (SADHART-CHF) trial.10 Recently, Carney et al11 and Glassman et al12 reported that patients with ischemic heart disease whose depression signicantly decreased during the study periods that was not necessarily related to the trial intervention had better survival than those patients whose depression persisted. Whether this phenomenon applies to patients with HF is unknown. This study therefore aimed to explore if remission is associated with better cardiovascular outcome of patients with HF and major depressive disorder and to examine characteristics that may differentiate patients whose depression remitted from ones whose depression remained during the trial. Methods The SADHART-CHF database was used for this analysis. A detailed method of the SADHART-CHF trial has been previously published.10,13 SADHART-CHF was a randomized double-blind study of sertraline versus matching placebo in patients with HF and co-morbid major depression. In addition, all participants received nurse-facilitated support (NFS). Primary end points of the SADHART-CHF trial were change across time in severity of depression as measured by the Hamilton Depression Rating Scale (HDRS) total score and change in composite cardiovascular status.10,13 The protocol was reviewed and approved by the local institutional review board at each participating center, and all participants provided written, voluntary, informed consent before enrollment.
www.ajconline.org

Departments of aMedicine and bPsychiatry and Behavioral Sciences and cCenter for Aging, Duke University Medical Center, Durham, North Carolina; dDuke Clinical Research Institute, Durham, North Carolina. Manuscript received August 10, 2010; revised manuscript received and accepted October 1, 2010. The SADHART-CHF study was funded by the National Institute of Mental Health, Bethesda, Maryland. *Corresponding author: Tel: 919-668-0762; fax: 919-668-5271. E-mail address: jiang001@mc.duke.edu (W. Jiang). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.013

546 Table 1 Baseline characteristics Characteristics

The American Journal of Cardiology (www.ajconline.org)

Remission Group Sertraline (n 106) Placebo (n 102) 61.2 9.8 66.7% 55.9% 35.3% 8.9% 29.7 10.2 27.5% 51.1% 21.6% 67.7% 62.4% 44.1% 83.3% 73.5% 45.1% 24.5% 31.4% 3.9% 13.7% 79.4% 76.5% 9.8% 84.3% 80.4% 18.6% 65.7% 59.8% 4.9% 17.1 5.2 (730)

Nonremission Group Sertraline (n 90) 61.9 10.5 50.0% 56.7% 40.0% 3.3% 32.2 9.7 32.2% 43.3% 24.4% 68.9% 52.2% 51.1% 87.8% 71.1% 51.1% 24.4% 28.9% 8.9% 22.2% 72.2% 72.2% 10.0% 91.1% 81.1% 15.6% 71.1% 61.1% 13.3% 19.6 5.8 (535) Placebo (n 104) 62.0 11.3 55.8% 60.6% 34.6% 4.8% 29.8 10.1 27.9% 45.2% 26.9% 69.2% 72.1% 43.3% 83.7% 82.7% 47.1% 28.9% 32.7% 5.8% 22.1% 79.8% 68.3% 5.7% 83.7% 84.6% 18.3% 69.2% 61.5% 10.6% 19.5 5.2 (733)

p Value*

Age (years), mean SD Men Race White Black Other Left ventricular ejection fraction, mean SD New York Heart Association class II III IV Ischemic cause 1 heart failure hospitalization in previous 12 months Previous myocardial infarction History of hypertension Hyperlipidemia Diabetes mellitus Current smoking Previous coronary bypass Pacemaker Implantable cardioverterdebrillator Inpatient at enrollment Baseline medications Angiotensin-converting enzyme inhibitor Angiotensin receptor blocker Blocker Aspirin Digoxin Statin Loop diuretic Calcium channel blocker Baseline Hamilton Depression Rating Scale, mean SD (range) Depression severity (Hamilton Depression Rating Scale score) 17 1822 23 Baseline Beck Depression Inventory, mean SD (range) Depressive symptom dimension (Beck Depression Inventory) Cognitive/affective Somatic/affective Appetitive Antidepressant use before enrollment History of major depressive disorder

63.0 9.9 62.3% 52.8% 40.6% 6.6% 31.5 8.9 32.8% 51.9% 16.0% 68.9% 57.6% 52.8% 93.4% 83.0% 55.7% 32.1% 36.8% 7.6% 14.2% 70.8% 68.9% 9.4% 84.9% 86.8% 14.2% 72.6% 60.4% 8.5% 16.9 5.6 (630)

0.52 0.09 0.71

0.20 0.56

0.99 0.03 0.41 0.10 0.09 0.43 0.82 0.69 0.52 0.20 0.29 0.54 0.67 0.43 0.57 0.79 0.73 0.99 0.22 0.0002 0.0001/0.96

31.6% 24.8% 18.5% 18.3 6.9 (940)

30.5% 24.1% 17.4% 17.6 6.2 (834)

16.4% 25.6% 29.4% 21.3 7.2 (941)

21.5% 25.6% 34.8% 19.5 5.3 (740)

0.0009

4.4 3.7 (019) 11.8 3.9 (526) 2.1 1.7 (07) 6.6% 15.1%

4.0 3.4 (021) 11.5 3.7 (321) 2.0 1.8 (06) 4.9% 8.8%

5.5 4.0 (016) 13.7 4.5 (428) 2.1 1.6 (06) 8.9% 22.2%

4.6 4.1 (019) 12.4 4.0 (424) 2.4 1.9 (07) 8.0% 16.4%

0.07 0.003 0.28 0.73 0.08

* The p values apply to differences among the 4 groups. These p values demonstrate that depression severity was signicantly associated with remission status (p 0.0001) but not with treatment assignment (p 0.96).

SADHART-CHF trial patients were randomized 1:1 to sertraline 50 mg/day or matched placebo for a 12-week acute treatment phase. Study drug dose was titrated up in 50-mg/day increments every 2 weeks with a maximum dose of 200 mg/day, depending on severity of depressive symptoms and tolerability of participants to the study drug. Participants who were unable to tolerate the 50-mg/day dose

were allowed to remain in the study provided a minimum 25-mg/day dose was tolerated. The study medication was tapered off after the 12-week acute treatment phase. All participants, regardless of acute-phase completion, entered the long-term follow-up phase and were contacted at 6 months 12 months, and annually thereafter to evaluate clinical events and vital status. The long-term follow-up phase

Heart Failure/Depression Remission and Heart Failure Prognosis Table 2 Performance of each group during 12-week acute treatment phase Characteristics Remission Group (n 208) Sertraline Treatment assignment Days in 12-week intervention Dosing Last Hamilton Depression Rating Scale score, mean SD (range) Last Beck Depression Inventory score, mean SD (range) 26.4% 77.4 19.5 (19119) 64.1 30.7 (50150) 3.8 2.1 (07) 4.8 3.2 (016) Placebo 25.4% 78.8 18.9 (1110) 65.8 29.1 (50200) 3.5 2.1 (07) 4.9 3.6 (019) Sertraline 22.4% 73.5 21.8 (13104) 112.2 57.2 (50200) 13.9 5.4 (831) 13.6 7.8 (341) Nonremission Group (n 194) Placebo 25.9% 71.8 24.1 (11108) 108.7 55.5 (50200) 13.5 4.7 (827) 12.5 7.2 (236)

547

p Value

0.061 0.0001 0.0002 0.0009

Table 3 Reasons for termination during 12-week treatment phase Reasons for Termination During Acute Phase Patient withdrew Patient lost to follow-up Side effect Withdraw by study physician Noncompliance Death Other Remission Group (n 208, 22.5%) 1 (0.5%) 4 (1.9%) 13 (6.3%) 5 (2.4%) 19 (9.1%) 3 (1.44%) 2 (0.9%) Nonremission p Group Value (n 194, 33.5%) 4 (2.1%) 7 (3.6%) 16 (8.3%) 14 (7.2%) 13 (6.7%) 5 (2.58%) 6 (3.1%) 0.15 0.30 0.44 0.02 0.37 0.42 0.13

continued until the last enrolled participant completed a 6-month follow-up. All participants received NFS as a mechanism to ascertain safety and study compliance. Primary goals for the research personnel applying supportive measures were to increase recruitment and assessment of participants, to ascertain safety of participants, and to increase participant compliance and retention. Such support was provided by nurses and other study personnel with experience or training in clinical psychiatry and supervised by the study psychiatrist. Supportive measures included 10 hours of active interaction between research personnel and study participants during the 12-week acute phase. The interaction consisted of 3 face-to-face visits (1 during recruitment/baseline assessment followed by 2 visits conducted primarily in participants homes) and 4 follow-up telephone contacts. Research personnel aimed to provide psychological support and conduct medical and psychiatric health evaluations. Personnel were instructed to provide active and empathetic listening and validation skills and soothing and other emotional support strategies. Research personnel were asked not to push for dialog with participants who were more reserved and aimed to establish an individual interpersonal relationship with each participant. For the present study, SADHART-CHF participants were divided into 2 groups, irrespective of treatment assignment: (1) remission, dened as participants whose last measured HDRS score was 8, and (2) nonremission, dened as participants whose last measured HDRS score was 8. Patients who dropped out before having any repeat HDRS were not included in this study.

Figure 1. Survival by patients in the depression remission group assigned to placebo (blue line), patients with remission assigned to sertraline (red line), patients in the nonremission group who were assigned to placebo (green line), and patients in the nonremission group assigned to sertraline (black line). Unadjusted hazard ratio for the nonremission versus remission groups was 1.23 (95% condence interval 0.95 to 1.59).

Primary end points of this analysis were survival or time to death and rate of recurrent cardiovascular events and/or death until last follow-up. Cardiovascular events were adjudicated by a blinded clinical events committee as a component of the primary trial.10 Clinical characteristics examined among these groups included effects of randomization, dosing of study medication, and depressive symptomatology measured by the HDRS and the Beck Depression Inventory (BDI) scale. Baseline depressive symptomatology was examined in 3 ways (1) severity of depression based on HDRS scores (mild 17, moderate 18 to 22, severe 23), (2) overall HDRS and BDI scores, and (3) depressive symptom dimension, i.e., cognitive/affective (items 2, 3, 5, 6, 7, 8, 9, 12, and 14), somatic/affective (items 1, 4, 10, 11, 13, 15, 16, 17, 21, and 22), and appetitive (items 18 and 19/20) based on the BDI scale.14 Comparative analysis of baseline characteristics was performed on remission status with respect to treatment assignment. Change in HDRS by remission status over the 12-week treatment period was analyzed by random coefcient models. The nal model included treatment, remission status, time in weeks, and the square of time and the interaction of each of these 2 time variables with remission status and study site. The model also in-

548

The American Journal of Cardiology (www.ajconline.org)

Table 4 Number of cardiovascular events per participant among groups Cardiovascular Events Remission Group (n 208) Sertraline Any nonfatal cardiovascular event* Acute myocardial infarction Exacerbation of heart failure Unstable angina pectoris Arrhythmia Syncope Stroke Transient ischemic attack Other Any cardiovascular event and/or death * Adjusted p 0.02;

Nonremission Group (n 194) Placebo Sertraline 1.82 3.08 0.07 0.29 0.88 2.02 0.20 0.62 0.20 0.71 0.01 0.11 0.06 0.23 0.02 0.15 0.40 0.96 2.1 3.13 Placebo 1.52 2.28 0.02 0.14 0.70 1.34 0.18 0.63 0.14 0.53 0.05 0.21 0.04 0.21 0 0.42 0.83 1.79 2.3

1.04 1.73 0.04 0.19 0.42 1.05 0.11 0.48 0.05 0.25 0.04 0.19 0.05 0.25 0 0.36 0.81 1.27 1.81

1.18 1.86 0.07 0.29 0.57 1.40 0.07 0.39 0.11 0.37 0 0.03 0.22 0 0.32 0.64 1.40 1.90

adjusted p 0.01 for comparisons between the remission and nonremission groups.

cluded the random effects of patient, patient-by-time interaction, and square of patient-by-time interaction. All participants (remission and nonremission groups) were included in the analysis. Cox proportional regression modeling was used to evaluate survival differences between participants classied in the remission and nonremission groups. Data were censored at time to death. Logistic regression analysis was used to test differences of cardiovascular events between groups. Baseline HDRS scores, treatment assignment (sertraline or placebo), age, gender, baseline left ventricular ejection fraction, New York Heart Association (NYHA) class, ischemic cause, and study site were included in the regression model. Assumptions of the model were assessed using standard techniques. Logistic regression was also used to determine which subsets of patients were likely to remit over the 12-week treatment phase. Investigated characteristics were gender, baseline HDRS scores (17 vs 17), BDI scores (16 vs 16), and somatic/affective scores (lower vs higher than or equal to mean). Statistical analyses using SAS 9.1 (SAS Institute, Cary, North Carolina) were performed by statistical personnel within Duke University Medical Center (Durham, North Carolina). Results In total 469 participants were enrolled in SADHARTCHF at Duke University Medical Center and 3 Duke Health System afliates from August 13, 2003 to March 3, 2008. The primary analysis of SADHART-CHF has been previously reported.13 Of the 469 SADHART-CHF participants, 208 patients (44.3%) had HDRS scores that improved to 8, whereas 194 (41.4%) patients had a HDRS 8 at the end of 12-week intervention. There were 67 subjects (14.3%) who did not have any repeat HDRS during the 12-week acute phase intervention and were excluded from this study. These participants had higher NYHA class HF than the other 2 groups (NYHA class II 17.9 vs 29.9, p 0.05). Ten patients (14.9%) died before any repeat HDRS assessment. Most baseline characteristics were similar between the remission and nonremission groups with respect to sertraline or placebo assignment (Table 1). More men remitted

than women (64.4 vs 53.1%, p 0.02), although differences between groups were not statistically signicant. Hospitalization due to HF exacerbation within 1 year before enrollment was higher in the nonremission group (p 0.03; Table 1). The remission group had longer treatment duration (78 19 days) during the 12-week acute treatment phase than the nonremission group (73 23, p 0.001; Table 2). The interaction of remission by time was signicant (p 0.001) but not remission by square of time (p 0.19). Change in HDRS total score over the course of the 12-week acute phase (i.e., time) intervention and square of time were signicant (time p 0.001, square of time p 0.001). The difference in the change of HDRS scores between the remission and nonremission groups was statistically signicant (mean SE 4.8 0.49, p 0.001). Although they remained depressed, symptoms of patients in the nonremission group improved notably from baseline (mean SE 4.1 0.33, p 0.001). In the nonremission group, 40 remained signicantly depressed (HDRS total score 17). There were no differences in these measurements between treatment assignments. Doses of sertraline and placebo were signicantly different between the remission and nonremission groups, but there was no difference within treatment assignment (Table 2). One hundred twelve of the 402 participants (27.9%) dropped out during the 12-week treatment phase after providing 1 repeat measurement of depressive symptoms. Reasons for dropping out during the acute treatment phase are listed in Table 3. Average length of follow-up was 798 493 days (range 1 to 1,832) for the entire study population. Patients whose depression remitted had longer survival than those whose depression remained (866 479 vs 793 483 days); however, the result of the Cox proportional regression analysis revealed that the difference of survival between the 2 groups was not statistically signicant (hazard ratio 1.23 for nonremission vs remission group, 95% condence interval 0.95 to 1.59). KaplanMeier survival curves for the remission and nonremission groups with respect to treatment assignment are shown in Figure 1.

Heart Failure/Depression Remission and Heart Failure Prognosis Table 5 Logistic analysis assessing baseline characteristics that best predict remission Characteristics Male gender Male gender Hamilton Depression Rating Scale score 17 Male gender Beck Depression Inventory score 16 Male gender somatic/affective score no higher than mean Male gender Hamilton Depression Rating Scale score 17 somatic/affective score no higher than mean Odds Ratio 1.57 2.48 2.27 2.01 2.68 95% Condence Interval 1.0552.345 1.5443.973 1.4413.554 1.3243.042 1.5834.524

549

There were 606 cardiovascular events in the entire study population including deaths, with 230 in the remission group and 323 in the nonremission group. Table 4 present a summary of differences of various cardiovascular events among the 4 groups. Patients whose depression remitted had a smaller number of overall nonfatal cardiovascular events per participant and fatal and nonfatal combined cardiovascular events (mean SD 1.11 1.79 and 1.34 1.86) compared to those whose depression remained (mean SD 1.66 2.68 and 1.92 2.71). Differences were statistically signicant and remained so after covariate analysis with age, gender, baseline ejection fraction, NYHA classes, ischemic cause of HF, baseline HDRS scores, and treatment assignment (p 0.05 for all comparisons). There was no association of baseline depressive symptoms and cardiovascular events (p 0.82 for overall nonfatal cardiovascular events, p 0.88 for fatal and nonfatal combined cardiovascular events). Baseline HDRS and self-rated BDI scores were lower in the remission group (HDRS 17.0 5.4, BDI 17.9 6.5) than in the nonremission group (HDRS 19.6 5.5, BDI 20.3 7.2, p 0.001 for the 2 comparisons). Further analysis of depressive symptom dimension revealed that it was the dimension of the somatic/affective symptoms measured by BDI at baseline (Table 1) that separated patients whose depression remitted from patients whose depression remained (p 0.003). Although the nonremission group had higher baseline cognitive/affective symptoms than the remission group, the difference was not statistically significant (p 0.07; Table 1). Odds ratios for the 4 characteristics of remission are presented in Table 5. Discussion This analysis of the SADHART-CHF trial demonstrated an association of depression remission during the 12-week active intervention and signicantly decreased fatal and nonfatal cardiovascular events. Such ndings support the hypothesis that successful depression treatment may be associated with a cardiac benet in patients with HF. Carney et al11 performed a post hoc analysis in the Enhancing Recovery in Coronary Heart Disease (ENRICHD) study to examine whether depression remission at 6 months from

baseline was associated with survival. They found that patients who were randomized into the intervention arm and remained depressed at the 6-month follow-up had signicantly worse survival than those patients whose depression remitted, although there was no survival difference between the intervention and control groups. Furthermore, Glassman et al12 examined a 7-year survival difference between the SADHART trial participants whose depression signicantly remitted and those whose depression remained based on the Clinical Global Impression-Improvement (CGI) subscale score during the 24-week sertraline or placebo trial intervention. The investigators demonstrated that depressed patients with ischemic heart disease whose CGI score was decreased to 1 or 2 had signicant higher survival than those patients whose CGI score was 2 irrespective of treatment assignment. The SADHART-CHF study design followed the traditional phase II to III clinical trial model of randomization and placebo control, and it focused primarily on examining differences between active treatment assignments, i.e., drug or psychotherapy versus placebo or usual care. Sertraline compared to placebo did not result in a statistically significant higher rate of remission in the SADHART-CHF study (54.1% vs 49.5%, p 0.36).10 Several factors may have contributed to this nding including the placebo effect; the therapeutic impact of NFS, or the uniqueness of co-morbid depression in patients with HF. Therapeutic response to placebo has been recognized in research and in clinical practice.15,16 Several meta-analyses have demonstrated a placebo response 50% in depression trials, a nding that was particularly evident among nonpublished trials.1719 In recent years, an increase in scientic attention to the placebo effect has yielded evidence that the effect may have a neurobiological foundation.20,21 However, the potential impact of NFS on treatment response cannot be overlooked. One major predisposing factor for depression is a weak social support network. Persistent and negative life stressors coupled with limited supportive structures are believed to result in a decrease of mental function in patients with chronic illnesses. The NFS-fostered relationship may have replaced ineffective or insufcient social networks and established an alliance that contributed to functional improvement.2224 Specic evidence to support the therapeutic impact of NFS includes maintenance of remission in these participants after 4 to 6 weeks when placebo effects tend to decrease.25 Most of the study population was naive to psychological and psychiatric interventions; therefore, these participants may have been innately more responsive to psychosocial supportive measures. In addition, the fact that a large proportion of study participants had a mild baseline depression in severity (i.e., HDRS total score 8 to 17) at study entry may have contributed to the high remission rate for those participants. Other studies of this type have failed to demonstrate survival or prognostic benet of active treatment compared to controls.11,26,27 Trials that failed to show treatment benet of pharmacologic agents hypothesized that the lack of a statistically signicant difference in outcomes was due to insufcient power.2730 SADHART-CHF had a similar design to other studies in patients with heart disease and co-morbid depression. In SADHART-CHF, the sample size was believed to be ade-

550

The American Journal of Cardiology (www.ajconline.org)

quate to evaluate differences in depression and clinical outcome; however, there was no benet observed with antidepressant treatment over placebo.10 In contrast, in the ENRICHD trial, open-label treatment of selective serotonin reuptake inhibitors showed a statistically signicant 40% decrease in death or nonfatal myocardial infarction, with a crude hazard of 0.61 (95% condence interval 0.41 to 0.90) and an adjusted hazard ratio of 0.53 (95% condence interval 0.38 to 0.84).28 In this study, because antidepressant prescription was at the discretion of the study physicians and not randomized or controlled, the impact of the nding is controversial. Whether or not a survival benet was associated with depression remission was not reported in the primary analysis. However, in the subgroup analysis for ENRICHD, successful treatment of depression appeared to decrease the risk of cardiovascular events.11 The present ndings of the SADHART-CHF studies raise several important questions including the need to determine whose depression is easier to remit versus those who are more resistant to interventions, to examine whether subsets of depressed patients with HF may be at higher risk for cardiovascular events, and to identify effective antidepressant(s) for this particular population. Our research indicated that men with HF and mild depression (HDRS 17) have nearly a 2.5-fold chance for depression remission compared to the rest of the study population. In contrast, patients who had greater somatic affective symptoms were less likely to have remission. These data suggest that future trials testing therapeutics may need to focus on patients whose depression is more severe and do not respond to general supportive measures. This information also may be useful for clinical practice. Nevertheless, future studies are needed to conrm the observed association between depression remission and cardiovascular outcomes. NFS appears to be a powerful therapeutic technique that results in depression remission in certain patients with HF. A more sophisticated study design and renement of support measures will be needed to fully evaluate the effects of an NFS in patients with HF or other medical conditions. Cause of depression in patients with cardiac disease or other long-term medical conditions may be more heterogenous than in populations studied in traditional trials that test antidepressant efcacy. Therefore, outcome studies testing the association between depression treatment and prognosis may need to be aligned with real-world clinical practice techniques such as employing a naturalistic approach in the acclimatization of individual differences and identifying characteristics of populations who respond differently to various antidepressive treatments. Results of this analysis should be interpreted in the context of limitations. This was a secondary analysis of a trial, which was originally designed to assess treatment response to sertraline versus placebo, examining subgroups that are dened by changes in measurements over the course of a trial. Therefore, any changes in depression and cardiovascular outcomes between the remission and nonremission groups may be considered due to pre-existing baseline characteristic differences between those who do and those who do not respond to sertraline or other interventional measurements. Whether these participants received depression treatment after the 12-week acute phase of trial was not further

evaluated, similar to the studies of Carney et al11 and Glassman et al.12 Although the analysis was adjusted for differences in baseline variables and baseline depression severity was not associated with the prognosis, the fact that participants with remission remitted had a baseline lower depressive score, especially by the self-administered test, might have reected a subpopulation who had perceived their illness milder and/or were more receptive to psycho-supportive intervention. In addition, this study had limited statistical power to evaluate survival. Acknowledgment: We thank Pzer, Inc., New York, New York, for supplying the study drug.
1. Pelle AJM, Gidron YY, Szabo BM, Denollet J. Psychological predictors of prognosis in chronic heart failure. J Card Fail 2008;14:341 350. 2. Rutledge T, Reis VA, Linke SE, Greenberg BH, Mills PJ. Depression in heart failure- A meta-analytic review of prevalence, intervention effects, and associations with clinical outcomes. J Am Coll Cardiol 2006;48:15271537. 3. Jiang W, Kuchibhatla M, Clary GL, Cuffe MS, Christopher EJ, Alexander JD, Califf RM, Krishnan RR, OConnor CM. Congestive heart failurerelationship between depressive symptoms and long-term mortality in patients with heart failure. Am Heart J 2007;154:102108. 4. Johansson P, Dahlstrom U, Alehagen U. Depressive symptoms and six-year cardiovascular mortality in elderly patients with and without heart failure. Scand Cardiovasc J 2007;41:299 307. 5. Muller-Tasch T, Peters-Klimm F, Schellberg D, Holzapfel N, Barth A, Junger J, Szecsenyi J, Herzog W. Depression is a major determinant of quality of life in patients with chronic systolic heart failure in general practice. J Card Fail 2007;13:818 824. 6. Penninx B, Beekman ATF, Honig A, Deeg DJH, Schoevers RA, van Eijk JTM, van Tilburg W. Depression and cardiac mortalityresults from a community-based longitudinal study. Arch Gen Psychiatry 2001;58:221227. 7. Rozzini R, Sabatini T, Frisoni GB, Trabucchi M. Depression and major outcomes in older patients with heart failure. Arch Intern Med 2002;162:362363. 8. Rumsfeld JS, Havranek E, Masoudi FA, Peterson ED, Jones P, Tooley JF, Krumholz HM, Spertus JA, Cardiovascular Outcomes Res C. Depressive symptoms are the strongest predictors of short-term declines in health status in patients with heart failure. J Am Coll Cardiol 2003;42:18111817. 9. Vaccarino V, Kasl SV, Abramson J, Krumholz HR. Depressive symptoms and risk of functional decline and death in patients with heart failure. J Am Coll Cardiol 2001;38:199 205. 10. OConnor CM, Jiang W, Kuchibhatla M, Silva S, Cuffe M, Callwood D, Zakhary B, Stough W, Arias R, Rivelli S, Krishnan R, for the SADHART-CHF Investigators. Safety and efcacy of sertraline for depression in patients with heart failure: results of the SADHARTCHF trial. J Am Coll Cardiol 2010;56:692 699. 11. Carney RM, Blumenthal JA, Freedland KE, Youngblood M, Veith RC, Burg MM, Cornell C, Saab PG, Kaufmann PG, Czajkowski SM, Jaffe AS, for the ENRICHD Investigators. Depression and late mortality after myocardial infarction in the Enhancing Recovery in Coronary Heart Disease (ENRICHD) study. Psychosom Med 2004;66:466 474. 12. Glassman AH, Bigger JT, Gaffney M. Psychiatric characteristics associated with Long-term mortality among 361 patients having an acute coronary syndrome and Major depression seven-year follow-up of SADHART participants. Arch Gen Psychiatry 2009;66:10221029. 13. Jiang W, OConnor C, Silva SG, Kuchibhatla M, Cuffe MS, Callwood DD, Zakhary B, Henke E, Arias RM, Krishnan R, for the SADHARTCHF Investigators. Safety and efcacy of sertraline for depression in patients with CHF (SADHART-CHF): a randomized, double-blind, placebo-controlled trial of sertraline for major depression with congestive heart failure. Am Heart J 2008;156:437 444. 14. Linke SE, Rutledge T, Johnson BD, Vaccarino V, Bittner V, Cornell CE, Eteiba W, Sheps DS, Krantz DS, Parashar S, Merz CNB. Depressive symptom dimensions and cardiovascular prognosis

Heart Failure/Depression Remission and Heart Failure Prognosis among women with suspected myocardial ischemia. A report from the National Heart, Lung, and Blood Institute-sponsored womens ischemia syndrome evaluation. Arch Gen Psychiatry 2009;66: 499 507. Katz J. The Silent World of Doctor and Patient. New York: Free Press, 1984. Shapiro A, Shapiro E. The placebo: is it much ado about nothing? In: Harrington A, ed. The Placebo Effect: an Interdisciplinary Exploration, 1st Ed. Cambridge, MA: Harvard University Press, 1997:1236. Kirsch I. Listening to Prozac but hearing placebo: a meta-analysis of antidepressant medication. Prev Treat 1998;1:0002a. Kirsch I, Scoboria A, Nicholls S. The emperors new drugs: an analysis of antidepressant medication data submitted to the U.S. Food and Drug Administration. Prev Treat 2002;5:23. Khan A, Detke M, Khan SRF, Mallinckrodt C. Placebo response and antidepressant clinical trial outcome. J Nerv Ment Dis 2003;191:211 218. Leuchter AF, Cook IA, Witte EA, Morgan M, Abrams M. Changes in brain function of depressed subjects during treatment with placebo. Am J Psychiatry 2002;159:122129. Mayberg HS, Silva JA, Brannan SK, Tekell JL, Mahurin RK, McGinnis S, Jerabek PA. The functional neuroanatomy of the placebo effect. Am J Psychiatry 2002;159:728 737. Ong LML, Dehaes J, Hoos AM, Lammes FB. Doctor-patient communicationa review of the literature. Soc Sci Med Soc Sci Med 1995; 40:903918. Stewart MA. Effective physician-patient communication and health outcomesa review. CMAJ 1995;152:14231433.

551

15. 16.

17. 18.

19.

20.

21.

22.

23.

24. Wolfaardt UB, Reddon JR, Joyce AS. Assessing the efcacy of antidepressants: the transactional paradigm. Med Hypotheses 2005;64: 1229 1236. 25. Walsh BT, Seidman SN, Sysko R, Gould M. Placebo response in studies of major depressionvariable, substantial, and growing. JAMA 2002;287:1840 1847. 26. Frasure-Smith N. The Montreal Heart Attack Readjustment Trial. J Cardiopulm Rehabil 1995;15:103106. 27. Glassman AH, OConnor CM, Califf RM, Swedberg K, Schwartz P, Bigger JT, Krishnan KRR, van Zyl LT, Swenson JR, Finkel MS, Landau C, Shapiro PA, Pepine CJ, Mardekian J, Harrison WM. Grp S. Sertraline treatment of major depression in patients with acute MI or unstable angina. JAMA 2002;288:701709. 28. Berkman LF, Blumenthal J, Burg M, Carney RM, Catellier D, Cowan MJ, Czajkowski SM, DeBusk R, Hosking J, Jaffe A, Kaufmann PG, Mitchell P, Norman J, Powell LH, Raczynski JM, Schneiderman N, ENRICHD Investigators. Effects of treating depression and lowperceived social support on clinical events after myocardial infarctionthe Enhancing Recovery in Coronary Heart Disease patients (ENRICHD) randomized trial. JAMA 2003;289:3106 3116. 29. Taylor CB, Youngblood ME, Catellier D, Veith RC, Carney RM, Burg MM, Kaufmann PG, Shuster J, Mellman T, Blumenthal JA, Krishnan R, Jaffe AS. Investigators E. Effects of antidepressant medication of morbidity and mortality in depressed patients after myocardial infarction. Arch Gen Psychiatry 2005;62:792798. 30. Van Melle JP, De Jonge P, Honig A, Schene AH, Kuyper AMG, Crijns H, Schins A, Tulner D, Van den Berg MP, Ormel J, Investigators M-I. Effects of antidepressant treatment following myocardial infarction. Br J Psychiatry 2007;190:460 466.

Warfarin Use and Outcomes in Patients With Advanced Chronic Systolic Heart Failure Without Atrial Fibrillation, Prior Thromboembolic Events, or Prosthetic Valves
Marjan Mujib, MD, MPHa, Abu-Ahmed Z. Rahman, MDb, Ravi V. Desai, MDc, Mustafa I. Ahmed, MDa, Margaret A. Feller, MPHa, Inmaculada Aban, PhDa, Thomas E. Love, PhDd, Michel White, MDe, Prakash Deedwania, MDf, Wilbert S. Aronow, MDg, Gregg Fonarow, MDh, and Ali Ahmed, MD, MPHa,i,*
Warfarin is often used in patients with systolic heart failure (HF) to prevent adverse outcomes. However, its long-term effect remains controversial. The objective of this study was to determine the association of warfarin use and outcomes in patients with advanced chronic systolic HF without atrial brillation (AF), previous thromboembolic events, or prosthetic valves. Of the 2,708 BEST patients, 1,642 were free of AF without a history of thromboembolic events and without prosthetic valves at baseline. Of these, 471 patients (29%) were receiving warfarin. Propensity scores for warfarin use were estimated for each patient and were used to assemble a matched cohort of 354 pairs of patients with and without warfarin use who were balanced on 62 baseline characteristics. Kaplan-Meier and Cox regression analyses were used to estimate the association between warfarin use and outcomes during 4.5 years of follow-up. Matched participants had a mean age SD of 57 13 years with 24% women and 24% African-Americans. All-cause mortality occurred in 30% of matched patients in the 2 groups receiving and not receiving warfarin (hazard ratio 0.86, 95% condence interval 0.62 to 1.19, p 0.361). Warfarin use was not associated with cardiovascular mortality (hazard ratio 0.97, 95% condence interval 0.68 to 1.38, p 0.855), or HF hospitalization (hazard ratio 1.09, 95% condence interval 0.82 to 1.44, p 0.568). In conclusion, in patients with chronic advanced systolic HF without AF or other recommended indications for anticoagulation, prevalence of warfarin use was high. However, despite a therapeutic international normalized ratio in those receiving warfarin, its use had no signicant intrinsic association with mortality and hospitalization. Published by Elsevier Inc. (Am J Cardiol 2011;107:552557)

Heart failure (HF) is a hypercoagulable state, and patients with HF and low left ventricular ejection fraction (LVEF) may be at increased risk of LV thrombus formation and thromboembolic events.13 Although use of anticoagulants is recommended in patients with HF and atrial brillation (AF) and/or a previous thromboembolic event,4 there is conicting evidence of the benet of anticoagulation use
University of Alabama at Birmingham, Birmingham, Alabama; bCape Fear Valley Hospital, Fayetteville, North Carolina; cLehigh Valley Hospital, Allentown, Pennsylvania; dCase Western Reserve University, Cleveland, Ohio; eMontreal Heart Institute, Montreal, Quebec, Canada; fUniversity of California at San Francisco, Fresno, California; gNew York Medical College, Valhalla, New York; hUniversity of California at Los Angeles, Los Angeles, California; iVA Medical Center, Birmingham, Alabama. Manuscript received September 21, 2010; revised manuscript received and accepted October 5, 2010. Dr. Ahmed is supported by Grants R01-HL085561 and R01-HL097047 from the National Heart, Lung, and Blood Institute/National Institutes of Health, Bethesda, Maryland, and a generous gift from Ms. Jean B. Morris of Birmingham, Alabama. The Beta-Blocker Evaluation of Survival Trial (BEST) is conducted and supported by the National Heart, Lung, and Blood Institute in collaboration with the BEST study investigators. *Corresponding author: Tel: 205-934-9632; fax: 205-975-7099. E-mail address: aahmed@uab.edu (A. Ahmed). 0002-9149/11/$ see front matter Published by Elsevier Inc. doi:10.1016/j.amjcard.2010.10.012
a

in patients with HF without AF and/or previous thromboembolic events.59 However, because risk of LV thrombus formation increases with decreasing LVEF, clinicians are often concerned about the risk of LV thrombus formation in their patients with HF and markedly low LVEF. The objective of the present study was to determine the association of warfarin use and outcomes in patients with advanced chronic systolic HF without AF and/or previous thromboembolic events. Methods We conducted a post hoc analysis of the public-use copy of the Beta-Blocker Evaluation of Survival Trial (BEST) data for the present study. The BEST was a multicenter randomized placebo-controlled clinical trial of bucindolol, a blocker, in HF, the methods and results of which have been previously published.10 Briey, 2,708 patients with advanced chronic systolic HF were enrolled from 90 different sites across the United States and Canada from May 1995 to December 1998. All but 1 patient consented to be part of the public-use copy of the data. At baseline, patients had a mean duration of 49 months of HF and had a mean LVEF of 23%. All patients had New York Heart Association (NYHA) class III to IV symptoms and 90% of all
www.ajconline.org

Warfarin and Outcomes in Chronic Systolic Heart Failure Table 1 Baseline patient characteristics by use of warfarin before and after propensity matching Variable Before Propensity Matching No warfarin (n 1,171) Age (years) Women African-American Current smoking Body mass index (kg/m2) New York Heart Association class III Medical history Heart failure duration (months) Coronary artery disease Angina pectoris Hypertension Diabetes mellitus Hyperlipidemia Ventricular brillation Peripheral vascular disease Medications Bucindolol Angiotensin-converting enzyme inhibitors/ angiotensin receptor blocker Digitalis Diuretics Vasodilators Aspirin Statins Physical examination Pulse (beats/min) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Jugular venous distention S3 gallop Pulmonary rales Hepatomegaly Edema Laboratory data Hemoglobin (g/dl) Serum creatinine (mg/dl) Serum potassium (mEq/L) Plasma norepinephrine (pg/ml) Partial thromboplastin time (seconds) International normalized ratio Left bundle branch block by electrocardiogram Cardiothoracic ratio by chest x-ray Pulmonary edema by chest x-ray Left ventricular ejection fraction by nuclear scan (%) Right ventricular ejection fraction by nuclear scan (%) 60 12 283 (24%) 302 (26%) 213 (18%) 37 9 1,086 (93%) 46 48 670 (57%) 624 (53%) 715 (61%) 452 (39%) 527 (45%) 79 (7%) 164 (14%) 600 (51%) 1,137 (97%) 1,056 (90%) 1,086 (93%) 504 (43%) 738 (63%) 273 (23%) 82 13 119 19 72 11 399 (42%) 477 (41%) 162 (14%) 115 (10%) 288 (25%) 13.9 1.6 1.2 0.4 4.34 0.46 484 272 28 8 1.1 0.3 303 (26%) 54.8 7.2 114 (10%) 23.5 7.2 35.5 11.7 Warfarin (n 471) 56 12 107 (23%) 108 (23%) 94 (20%) 37 8 429 (91%) 44 44 265 (56%) 238 (51%) 239 (51%) 156 (33%) 208 (44%) 54 (12%) 78 (17%) 229 (49%) 458 (97%) 449 (95%) 433 (92%) 204 (43%) 98 (21%) 117 (25%) 83 13 114 16 71 11 836 (48%) 240 (51%) 38 (8%) 46 (10%) 96 (20%) 14.2 1.6 1.2 0.4 4.29 0.47 524 321 34 8 2.2 1.0 124 (26%) 55.5 6.9 51 (11%) 21.4 7.3 33.6 12.5 p Value 0.001 0.532 0.226 0.406 0.643 0.255 0.505 0.724 0.312 0.001 0.038 0.756 0.002 0.186 0.337 0.875 0.001 0.573 0.920 0.001 0.511 0.100 0.001 0.114 0.003 0.001 0.001 0.973 0.068 0.003 0.722 0.046 0.011 0.001 0.001 0.850 0.101 0.505 0.001 0.003 After Propensity Matching No warfarin (n 354) 57 14 87 (25%) 84 (24%) 73 (21%) 37 9 322 (91%) 45 47 190 (54%) 172 (49%) 191 (54%) 124 (35%) 145 (41%) 31 (9%) 53 (15%) 168 (48%) 340 (96%) 333 (94%) 326 (92%) 146 (41%) 91 (26%) 87 (25%) 83 13 115 16 72 11 134 (38%) 165 (47%) 36 (10%) 36 (10%) 76 (22%) 14.0 1.6 1.2 0.4 4.31 0.47 511 325 29 13 1.1 0.5 98 (28%) 55.1 7.3 37 (11%) 22.4 7.5 33.7 11.9 Warfarin (n 354) 57 12 83 (23%) 87 (25%) 68 (19%) 37 8 327 (92%) 45 45 192 (54%) 178 (50%) 191 (54%) 118 (33%) 149 (42%) 33 (9%) 52 (15%) 172 (49%) 345 (98%) 334 (94%) 327 (92%) 150 (42%) 96 (27%) 83 (23%) 83 13 115 16 71 11 147 (42%) 173 (49%) 35 (10%) 39 (11%) 74 (21%) 14.1 1.6 1.2 0.4 4.31 0.46 509 302 34 7 2.2 0.9 94 (27%) 55.4 6.9 36 (10%) 22.1 7.2 34.1 12.4

553

p Value 0.353 0.781 0.857 0.709 0.446 0.583 0.939 0.937 0.708 1.000 0.693 0.825 0.896 1.000 0.821 0.425 1.000 1.000 0.818 0.640 0.794 0.979 0.708 0.738 0.356 0.582 1.000 0.807 0.924 0.522 0.808 0.807 0.917 0.001 0.001 0.796 0.590 1.000 0.520 0.660

Values presented as number of patients (percentages) or mean SD.

patients were receiving angiotensin-converting enzyme inhibitors, diuretics, and digitalis. Data on use of warfarin at baseline were available in all 2,707 participants. For the present analysis, we excluded 692 patients with AF, 343 patients with a history of thromboembolic diseases, and 30 patients with prosthetic valves at baseline. Thus, our nal sample was 1,642, of which 471 patients (29%) were receiving warfarin at baseline. Consid-

ering the signicant imbalances in baseline characteristics between the 2 groups (Table 1), we used propensity scores to assemble a matched cohort of 354 pairs of patients who were well balanced on 62 baseline characteristics.1117 Propensity scores for warfarin use were estimated for each of the 1,642 patients using a nonparsimonious multivariable logistic regression model.18,19 Absolute standardized differences were estimated to evaluate the prematch imbalance

554

The American Journal of Cardiology (www.ajconline.org)

Figure 1. Absolute standardized differences comparing covariate values for patients with and without warfarin use before and after propensity score matching.

and postmatch balance and presented as a Love plot. An absolute standardized difference of 0% indicates no residual bias and differences 10% are considered inconsequential. BEST participants were followed for a minimum of 18 months and a maximum of 4.5 years.10 Primary outcome for the present analysis was all-cause mortality during 4.1 years of follow-up (mean 2, range 10 days to 4.14 years). Secondary outcomes were cardiovascular and HF mortalities and all-cause and HF hospitalizations. Kaplan-Meier and Cox regression analyses were used to determine associations between warfarin use and outcomes during 4.1 years of follow-up. Log-minus-log scale survival plots were used to check proportional hazards assumptions. Subgroup analyses were conducted to determine the homogeneity of association between use of warfarin and all-cause mortality. All statistical tests were 2-tailed with a p value 0.05 considered statistically signicant. All data analyses were performed using SPSS 18 for Windows (SPSS, Inc., Chicago, Illinois). Results Matched patients had a mean age SD of 57 13 years, 170 24% were women, and 171 24% were African-Americans. Before matching, patients receiving warfarin were younger, had a lower mean of LVEF and right ventricular EF, a lower prevalence of hypertension and diabetes mellitus, but had a greater symptom burden such as

Figure 2. Kaplan-Meier plots for all-cause mortality by use of warfarin. CI condence interval; HR hazard ratio.

jugular venous distention and S3 gallop. These and other imbalances in baseline characteristics were well balanced after matching (Figure 1, Table 1). After matching, absolute standardized differences for all measured covariates were 10% (most were 5%), suggesting substantial covariate balance across groups (Figure 1). Median international normalized ratios (INRs; interquartile range) were 2.0 (1.1) and 1.0

Warfarin and Outcomes in Chronic Systolic Heart Failure Table 2 Effects of warfarin on all-cause mortality in BEST trial Outcomes Events (%) No Warfarin Before matching Patients Unadjusted Multivariable-adjusted Propensity-adjusted After matching Patients All-cause mortality 1,171 321 (27%) 354 106 (30%) Warfarin 471 159 (34%) 354 106 (30%) Absolute Risk Increase* Hazard Ratio (95% condence interval)

555

p Value

7%

1.20 (0.991.45) 1.14 (0.911.43) 1.08 (0.871.35)

0.062 0.253 0.477

0%

0.86 (0.621.19)

0.361

* Absolute rate increase was calculated by subtracting rates of events in the group receiving from those not receiving warfarin (before values were rounded). Multivariable model includes all covariates displayed in Figure 1.

Figure 3. Association of use of warfarin and all-cause mortality in subgroups of matched patients.

(0.1) for matched patients receiving and not receiving warfarin, respectively. Overall, 212 matched patients (30%) died from all causes during a 2.1-year median follow-up. All-cause mortality occurred in 30% and 30% of matched patients receiving and not receiving warfarin, respectively (hazard ratio 0.86, 95% condence interval 0.62 to 1.19, p 0.361; Figure 2, Table 2). The association between warfarin use and all-cause mortality was homogenous across a wide spectrum of participants, including those with LVEFs 20% and 20% (Figure 3). When we used LVEF as a continuous variable, we still did not observe any statistically signicant interaction between use of warfarin and LVEF (p for interaction

0.815). Prematch-unadjusted, multivariable-adjusted, and propensity scoreadjusted hazard ratios between warfarin and no-warfarin use in the 1,642 patients before matching are listed in Table 2. Warfarin had no association with cardiovascular and HF mortalities and all-cause and HF hospitalizations after matching (Table 3). Discussion Findings from the present study demonstrate that in patients with advanced chronic systolic HF without AF and/or previous thromboembolic events, prevalence of warfarin use was relatively high. Nevertheless, our data suggest that

556

The American Journal of Cardiology (www.ajconline.org)

Table 3 Effects of warfarin on other outcomes in BEST trial Outcomes Events (%) No Warfarin Before matching Patients Cardiovascular mortality Sudden cardiac death Heart failure mortality All-cause hospitalization Heart failure hospitalization After matching Patients Cardiovascular mortality Heart failure mortality All-cause hospitalization Heart failure hospitalization 1,171 268 (23%) 151 (13%) 91 (8%) 702 (60%) 412 (35%) 354 84 (24%) 27 (8%) 217 (61%) 131 (37%) Warfarin 471 143 (30%) 81 (17%) 45 (10%) 291 (62%) 186 (40%) 354 95 (27%) 31 (9%) 225 (64%) 141 (40%) Absolute Risk Increase* Hazard Ratio (95% condence interval) p Value

7% 4% 2% 2% 5% 3% 1% 3% 3%

1.29 (1.061.58) 1.31 (1.001.71) 1.19 (0.831.70) 0.99 (0.861.14) 1.13 (0.951.34)

0.013 0.052 0.336 0.883 0.174

0.97 (0.681.38) 0.73 (0.381.39) 0.96 (0.761.22) 1.09 (0.821.44)

0.855 0.332 0.763 0.568

* Absolute rate increase was calculated by subtracting rates of events in the group receiving from those not receiving warfarin (before values were rounded).

despite achieving a mean therapeutic INR, warfarin use did not provide any intrinsic survival benet in patients with advanced chronic systolic HF who had no other established indications for anticoagulation. These ndings are important because many practicing physicians perceive advanced chronic systolic HF as a prethrombotic stage and prescribe warfarin, although current American College of Cardiology/ American Heart Association guidelines have no clear recommendation on this.4 Thus, warfarin might be prescribed without any proved benet and with a potential increased risk of bleeding and other adverse effects.20 The unadjusted association of warfarin use with increased risk of cardiovascular mortality is rather surprising because pre-match patients receiving warfarin were younger and had lower or similar baseline prevalences of cardiovascular co-morbidities (Table 1). These suggest strong confounding by a history of ventricular brillation, greater symptom burden, and lower mean LVEF and right ventricular EF, which are known to increase risk of cardiovascular death. The near signicant unadjusted association between warfarin use and sudden cardiac death is likely due to increased prevalence of ventricular brillation in warfarin users. Despite a greater burden of HF symptoms in warfarin users, lack of signicant unadjusted associations of warfarin use with HF mortality and HF hospitalization is intriguing but may suggest that LVEF and HF symptoms were rather weak confounders. Lack of signicant associations of warfarin use with mortality and hospitalization in matched patients suggests lack of an intrinsic effect of warfarin on outcomes in patients with advanced systolic HF without AF and/or thromboembolic disorders. Although a low LVEF is often considered an indication for warfarin use in these patients, ndings from our subgroup analysis suggest that the association between warfarin use and all-cause mortality was similar regardless of LVEF categories. Lack of an intrinsic effect of warfarin in patients with advanced systolic HF without AF and/or thromboembolic events, despite a therapeutic INR, suggests that thromboembolic events may not underlie mechanisms of death or hospitalization in these

patients. Findings from our study provide further evidence supporting current guideline recommendations that use of warfarin in patients with HF should be restricted to those with AF and/or previous thromboembolic events.4 There is conicting evidence in the literature regarding the role of warfarin in patients with advanced systolic HF without AF or other indications for anticoagulation.5,21 In 1 study warfarin use was associated with lower mortality and morbidity in patients with mild to moderate (2/3 had NYHA class I to II symptoms) chronic systolic HF.6 Our study is distinguished by patients with more advanced HF (all with NYHA class III to IV symptoms) and use of a propensitymatched design that allowed the assembly of a balanced cohort. Findings from our study are consistent with those from the largest randomized clinical trial of anticoagulation in patients with HF and normal sinus rhythm to date in which there was no difference in outcomes between patients receiving warfarin (open label), aspirin, or clopidogrel.8 However, this study was not considered denitive because it was terminated prematurely because of slow enrollment resulting in an estimated power of only 40% to detect a 20% difference. In the absence of another large clinical trial with adequate power, observational studies such as ours add further evidence of lack of benet for therapeutic anticoagulation in these patients. As in all observational studies, a key limitation of our study is potential confounding by an unmeasured covariate. Sensitivity analysis would normally help quantify the degree of a hidden bias that would need to be present to invalidate conclusions based on signicant associations in an observational study. However, sensitivity analyses can be performed only if the observed association is statistically signicant.22 Another limitation is lack of data on other cardiovascular events including stroke and adverse events such as bleeding. In conclusion, in patients with advanced chronic systolic HF without AF and/or other indications for anticoagulation, despite a mean INR that was therapeutic, use of warfarin was not associated with clinical outcomes.

Warfarin and Outcomes in Chronic Systolic Heart Failure 1. Gibbs CR, Blann AD, Watson RD, Lip GY. Abnormalities of hemorheological, endothelial, and platelet function in patients with chronic heart failure in sinus rhythm: effects of angiotensin-converting enzyme inhibitor and beta-blocker therapy. Circulation 2001;103:1746 1751. 2. Jafri SM, Ozawa T, Mammen E, Levine TB, Johnson C, Goldstein S. Platelet function, thrombin and brinolytic activity in patients with heart failure. Eur Heart J 1993;14:205212. 3. Lip GY, Gibbs CR. Does heart failure confer a hypercoagulable state? Virchows triad revisited. J Am Coll Cardiol 1999;33:1424 1426. 4. Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG, Jessup M, Konstam MA, Mancini DM, Michl K, Oates JA, Rahko PS, Silver MA, Stevenson LW, Yancy CW, Antman EM, Smith SC Jr, Adams CD, Anderson JL, Faxon DP, Fuster V, Halperin JL, Hiratzka LF, Jacobs AK, Nishimura R, Ornato JP, Page RL, Riegel B. ACC/ AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure): developed in collaboration with the American College of Chest Physicians and the International Society for Heart and Lung Transplantation: endorsed by the Heart Rhythm Society. Circulation 2005;112(suppl)e154 e235. 5. Lip GY, Gibbs CR. Anticoagulation for heart failure in sinus rhythm: a Cochrane systematic review. QJM 2002;95:451 459. 6. Al-Khadra AS, Salem DN, Rand WM, Udelson JE, Smith JJ, Konstam MA. Warfarin anticoagulation and survival: a cohort analysis from the Studies of Left Ventricular Dysfunction. J Am Coll Cardiol 1998;31: 749 753. 7. Cleland JG, Findlay I, Jafri S, Sutton G, Falk R, Bulpitt C, Prentice C, Ford I, Trainer A, Poole-Wilson PA. The Warfarin/Aspirin Study in Heart failure (WASH): a randomized trial comparing antithrombotic strategies for patients with heart failure. Am Heart J 2004;148:157 164. 8. Massie BM, Collins JF, Ammon SE, Armstrong PW, Cleland JG, Ezekowitz M, Jafri SM, Krol WF, OConnor CM, Schulman KA, Teo K, Warren SR. Randomized trial of warfarin, aspirin, and clopidogrel in patients with chronic heart failure: the Warfarin and Antiplatelet Therapy in Chronic Heart Failure (WATCH) trial. Circulation 2009; 119:1616 1624. 9. Ripley TL, Nutescu E. Anticoagulation in patients with heart failure and normal sinus rhythm. Am J Health Syst Pharm 2009;66:134 141. 10. Beta-Blocker Evaluation of Survival Trial Investigators. The BEST investigators. A trial of the beta-blocker bucindolol in patients with advanced chronic heart failure. N Engl J Med 2001;344:1659 1667.

557

11. Wahle C, Adamopoulos C, Ekundayo OJ, Mujib M, Aronow WS, Ahmed A. A propensity-matched study of outcomes of chronic heart failure (HF) in younger and older adults. Arch Gerontol Geriatr 2009;49:165171. 12. Meyer P, Ekundayo OJ, Adamopoulos C, Mujib M, Aban I, White M, Aronow WS, Ahmed A. A propensity-matched study of elevated jugular venous pressure and outcomes in chronic heart failure. Am J Cardiol 2009;103:839 844. 13. Ekundayo OJ, DellItalia LJ, Sanders PW, Arnett D, Aban I, Love TE, Filippatos G, Anker SD, Lloyd-Jones DM, Bakris G, Mujib M, Ahmed A. Association between hyperuricemia and incident heart failure among older adults: a propensity-matched study. Int J Cardiol 2010; 142:279 287. 14. Desai RV, Banach M, Ahmed MI, Mujib M, Aban I, Love TE, White M, Fonarow G, Deedwania P, Aronow WS, Ahmed A. Impact of baseline systolic blood pressure on long-term outcomes in patients with advanced chronic systolic heart failure (insights from the BEST trial). Am J Cardiol 2010;106:221227. 15. Bowling CB, Pitt B, Ahmed MI, Aban IB, Sanders PW, Mujib M, Campbell RC, Love TE, Aronow WS, Allman RM, Bakris GL, Ahmed A. Hypokalemia and outcomes in patients with chronic heart failure and chronic kidney disease: ndings from propensity-matched studies. Circ Heart Fail 2010;3:253260. 16. Alper AB, Campbell RC, Anker SD, Bakris G, Wahle C, Love TE, Hamm LL, Mujib M, Ahmed A. A propensity-matched study of low serum potassium and mortality in older adults with chronic heart failure. Int J Cardiol 2009;137:1 8. 17. Ahmed MI, Ekundayo OJ, Mujib M, Campbell RC, Sanders PW, Pitt B, Perry GJ, Bakris G, Aban I, Love TE, Aronow WS, Ahmed A. Mild hyperkalemia and outcomes in chronic heart failure: A propensity matched study. Int J Cardiol 2009;144:383388. 18. Rubin DB. Using propensity score to help design observational studies: Application to the tobacco litigation. Health Serv Outcomes Res Methodol 2001;2:169 188. 19. Rosenbaum PR, Rubin DB. The central role of propensity score in observational studies for causal effects. Biometrika 1983;70:4155. 20. Fihn SD, Callahan CM, Martin DC, McDonell MB, Henikoff JG, White RH. The risk for and severity of bleeding complications in elderly patients treated with warfarin. The National Consortium of Anticoagulation Clinics. Ann Intern Med 1996;124:970 979. 21. Konstam MA. Antithrombotic therapy in heart failure: WATCHful wondering. Circulation 2009;119:1559 1561. 22. Rosenbaum PR. Sensitivity to hidden bias. In: Rosenbaum PR, ed. Observational Studies, Vol 1. New York: Springer-Verlag; 2002:105 170.

The Risk of Thromboembolism in Heart Failure: Does It Merit Anticoagulation Therapy?

Eduard Shantsila, MD, and Gregory Y.H. Lip, MD*
The development of thromboembolism in patients with systolic heart failure (HF) is well described. Thus, one may naturally presume that anticoagulant therapy could be benecial in patients with HF. However, although the benets of adequate anticoagulation are of no doubt in those with concomitant atrial brillation (AF), is there enough evidence to advocate routine anticoagulation therapy for patients with HF in sinus rhythm?1,2 Prothrombotic Factors in Heart Failure: Virchows Triad Revisited Various abnormalities seen in patients with severe left ventricular (LV) systolic dysfunction, including endothelial damage and dysfunction, abnormal blood stasis, and a hypercoagulable state, provide a milieu contributing to thrombogenesis and thromboembolism. These 3 components are representative of Virchows triad of thrombogenesis, originally proposed 150 years ago. Indeed, endothelial damage and dysfunction are a hallmark of HF, and their presence adversely affects prognosis in this condition. Endothelial dysfunction is accompanied by the imbalance of pro- and anticoagulant systems, with clear shift toward a prothrombotic direction. Plasma markers of endothelial damage and dysfunction, such as von Willebrand factor, soluble thrombomodulin, and soluble E-selectin, are signicantly increased in patients with acute and chronic HF.3,4 A dysfunctional endothelium also actively produces inammatory cytokines (e.g., tumor necrosis factor and interleukin-1), which are signicantly upregulated in HF and further promote thrombogenesis.5 In addition, the presence of dilated and/or dysfunctional cardiac chambers creates areas of blood stasis, particularly in patients with dilated cardiomyopathy, large anterior myocardial infarctions, and LV aneurysms. Stasis accelerates the activation of the coagulation system and brin formation. Also, abnormal blood constituents leading to a prothrombotic or hypercoagulable state have been reported in HF patients, as reected by high circulating biomarker levels, including brinogen, antithrombin III, brinopeptide A, and brin D-dimer.6,7 Activation of the neuroendocrine system, especially upregulation of angiotensin and endothelin production, further enhances the prothrombotic state in HF.8 Thrombosis in Chronic Heart Failure Evidence of an increased risk for thromboembolic events (stroke, pulmonary and peripheral thromboembolism) in
University of Birmingham Centre for Cardiovascular Sciences, City Hospital, Birmingham, United Kingdom. Manuscript received October 18, 2010; revised manuscript received and accepted October 19, 2010. *Corresponding author: Tel: 44-121-507-5080; fax: 44-121-554-4083. E-mail address: g.y.h.lip@bham.ac.uk (G.Y.H. Lip). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.029

patients with chronic HF free of AF in large prospective cohorts is relatively limited, although retrospective analyses of HF treatment trials and data from some epidemiologic cohort studies are available. Many older cohort studies have included proportions of patients with AF, which may well be asymptomatic and/or paroxysmal in nature, which is itself a strong independent risk factor for thromboembolism. Without associated AF, the risk for thromboembolism may be small; for example, 1 analysis of patients with HF in New York Heart Association class II and III without AF found only a 1% annual risk for thromboembolism.9 However, almost half of sudden cardiac deaths in HF have been shown to be due to acute myocardial infarction or coronary thrombosis, and 27% of deaths in HF originally classied as progressive congestive HF are actually caused by coronary thrombosis.10 Given that sudden cardiac death is a major contributor to mortality in HF populations, it is probable that the impact of thromboembolism in HF might be underestimated. The risk for thromboembolism may be particularly high in patients with severely depressed cardiac contractility.11,12 In the Survival and Ventricular Enlargement (SAVE) trial, for example, the risk for stroke was twofold higher in patients with ejection fractions (EFs) 28% compared to those with EFs 28%, and every 5% reduction in the EF was associated with an 18% increase in stroke risk.12 In the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT), from which patients with AF were excluded, the 4-year rate of thromboembolic events was 3.5% in those with EFs of 30% to 35%, 3.6% in those with EFs of 20% to 30%, and 4.6% in those with EFs 20%, equivalent to annual rates of 0.9%, 0.9%, and 1.2%, respectively.13 Oral Anticoagulation in Heart Failure Retrospective analyses of large HF trials have produced controversial results on the role of oral anticoagulation. For example, the analyses from the Studies of Left Ventricular Dysfunction (SOLVD) and SAVE trials found that warfarin seemed to be benecial in patients with HF, being associated with a signicant 24% relative risk reduction in allcause mortality as well as a lower risk for HF hospitalization, but not with a reduction of thromboembolic events. However, no benets from warfarin for thromboembolism prevention were evident in the Veterans Affairs VasodilatorHeart Failure Trial (V-HeFT)14 and SCD-HeFT.13 Recent controlled clinical trials of oral anticoagulation for HF have had poor recruitment and small numbers and were underpowered.1517 The largest published study was the Warfarin and Antiplatelet Therapy in Chronic Heart Failure (WATCH) trial, which randomized patients with HF with EFs 30% to receive warfarin or aspirin or clopidogrel, but the study was terminated early because of poor
www.ajconline.org

Editorial/Anticoagulation Therapy in Heart Failure

559

recruitment. Despite being underpowered, the WATCH study did show a strong trend favoring warfarin over aspirin for the reduction of nonfatal stroke (0.7% vs 2.1%), as well as fewer hospitalizations in the warfarin group (16.1%) compared to aspirin (22.2%) and clopidogrel (18.3%). However, the grim side of warfarin therapy was a signicant increase in the bleeding rate (5.5%) compared to aspirin (3.6%) and clopidogrel (2.5%).16 In this issue of The American Journal of Cardiology, Mujib et al18 provide further evidence that the universal administration of oral anticoagulation in patients with HF may not be benecial. The study has the advantage of studying 1,600 patients with severe LV dysfunction, and the investigators conclude that despite therapeutic international normalized ratios in those receiving warfarin, its use had no signicant effect on mortality and hospitalization. However, the study represents a post hoc analysis of a previously published randomized clinical trial, and administration of anticoagulants was at the discretion of the treating physician and not randomized. Second, only 471 patients received warfarin, with a median international normalized ratio of 2.0, which perhaps suggests that about half the patients treated with warfarin received suboptimal anticoagulation. Indeed, good anticoagulation control (expressed as time in therapeutic range) is crucial for the best outcomes when warfarin is used. Third, detailed data on stroke and thromboembolism incidence were not recorded. Indeed, given the very high mortality of patients with HF (30% in this study), relatively small changes in thromboembolism-related mortality could pass unnoticed. Given that the secondary end point of stroke was significantly reduced in WATCH, attention to this outcome in future analyses would be needed in any future clinical trials, rather than the focus on overall mortality, which may not be signicantly affected. Although Mujib et al18 tried to exclude patients with AF from their study, many patients with HF could develop AF in paroxysms and asymptomatically; thus, it is essentially how hard one looks to exclude underlying AF from being a major confounder. Current Approach and Future Directions Current guidelines from the American Heart Association and American College of Cardiology, the American College of Chest Physicians, and the European Society of Cardiology do not support the routine use of warfarin in cardiomyopathy in sinus rhythm.19 21 In many patients with HF, who are often elderly with multiple co-morbidities and polypharmacy, assessment of bleeding risk would also be useful; recent guidelines for AF have recommended the HASBLED bleeding risk score (hypertension, abnormal renal or liver function, stroke, bleeding history or predisposition, labile international normalized ratio, elderly [aged 65 years], and drugs or alcohol) as a simple, practical score to assess bleeding risk,22 but this has yet to be validated in a large cohort of patients with HF. The ongoing double-blind, multicenter Warfarin Aspirin Reduced Cardiac Ejection Fraction (WARCEF) trial aims to adequately address the utility of oral anticoagulation compared to aspirin in patients with HF with EFs 35%, and the results are awaited. Finally, the development of novel oral

anticoagulants that overcome the limitations and disutility of warfarin may also change the place of oral anticoagulant therapy in the management of HF. Time will tell.
1. Rietbrock S, Plumb JM, Gallagher AM, van Staa TP. How effective are dose-adjusted warfarin and aspirin for the prevention of stroke in patients with chronic atrial brillation? An analysis of the UK General Practice Research Database. Thromb Haemost 2009;101:527534. 2. Hughes M, Lip GY; Guideline Development Group, National Clinical Guideline for Management of Atrial Fibrillation in Primary and Secondary Care, National Institute for Health and Clinical Excellence. Stroke and thromboembolism in atrial brillation: a systematic review of stroke risk factors, risk stratication schema and cost effectiveness data. Thromb Haemost 2008;99:295304. 3. Shantsila E, Lip GY. The endothelium and thrombotic risk in heart failure. Thromb Haemost 2009;102:185187. 4. Chong AY, Lip GY, Freestone B, Blann AD. Increased circulating endothelial cells in acute heart failure: comparison with von Willebrand factor and soluble E-selectin. Eur J Heart Fail 2006;8:167172. 5. Lip GY, Gibbs CR. Does heart failure confer a hypercoagulable state? Virchows triad revisited. J Am Coll Cardiol 1999;33:1424 1426. 6. Jug B, Vene N, Salobir BG, Sebestjen M, Sabovic M, Keber I. Prognostic impact of haemostatic derangements in chronic heart failure. Thromb Haemost 2009;102:314 320. 7. Gombos T, Mak V, Cervenak L, Papassotiriou J, Kunde J, Hrsfalvi J, Frhcz Z, Pozsonyi Z, Borgulya G, Jnoskuti L, Prohszka Z. Levels of von Willebrand factor antigen and von Willebrand factor cleaving protease (ADAMTS13) activity predict clinical events in chronic heart failure. Thromb Haemost 2009;102:573580. 8. Sbarouni E, Bradshaw A, Andreotti F, Tuddenham E, Oakley CM, Cleland JG. Relationship between hemostatic abnormalities and neuroendocrine activity in heart failure. Am Heart J 1994;127:607 612. 9. Freudenberger RS, Wilson AC, Kostis JB; AFFIRM Investigators and Committees. Comparison of rate versus rhythm control for atrial brillation in patients with left ventricular dysfunction (from the AFFIRM study). Am J Cardiol 2007;100:247252. 10. Uretsky BF, Thygesen K, Armstrong PW, Cleland JG, Horowitz JD, Massie BM, Packer M, Poole-Wilson PA, Ryden L. Acute coronary ndings at autopsy in heart failure patients with sudden death: results from the assessment of treatment with lisinopril and survival (ATLAS) trial. Circulation 2000;102:611 616. 11. Dries DL, Rosenberg YD, Waclawiw MA, Domanski MJ. Ejection fraction and risk of thromboembolic events in patients with systolic dysfunction and sinus rhythm: evidence for gender differences in the studies of left ventricular dysfunction trials. J Am Coll Cardiol 1997; 29:1074 1080. 12. Loh E, Sutton MS, Wun CC, Rouleau JL, Flaker GC, Gottlieb SS, Lamas GA, Moy LA, Goldhaber SZ, Pfeffer MA. Ventricular dysfunction and the risk of stroke after myocardial infarction. N Engl J Med 1997;336:251257. 13. Freudenberger RS, Hellkamp AS, Halperin JL, Poole J, Anderson J, Johnson G, Mark DB, Lee KL, Bardy GH; SCD-HeFT Investigators. Risk of thromboembolism in heart failure: an analysis from the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT). Circulation 2007; 115:26372641. 14. Dunkman WB, Johnson GR, Carson PE, Bhat G, Farrell L, Cohn JN; The V-HeFT VA Cooperative Studies Group. Incidence of thromboembolic events in congestive heart failure. Circulation 1993;87:VI94 VI101. 15. Cleland JG, Findlay I, Jafri S, Sutton G, Falk R, Bulpitt C, Prentice C, Ford I, Trainer A, Poole-Wilson PA. The Warfarin/Aspirin Study in Heart Failure (WASH): a randomized trial comparing antithrombotic strategies for patients with heart failure. Am Heart J 2004;148:157 164. 16. Massie BM, Collins JF, Ammon SE, Armstrong PW, Cleland JG, Ezekowitz M, Jafri SM, Krol WF, OConnor CM, Schulman KA, Teo K, Warren SR; WATCH Trial Investigators. Randomized trial of warfarin, aspirin, and clopidogrel in patients with chronic heart failure: the Warfarin and Antiplatelet Therapy in Chronic Heart Failure (WATCH) trial. Circulation 2009;119:1616 1624. 17. Cokkinos DV, Haralabopoulos GC, Kostis JB, Toutouzas PK; HELAS investigators. Efcacy of antithrombotic therapy in chronic heart failure: the HELAS study. Eur J Heart Fail 2006;8:428 432.

560

The American Journal of Cardiology (www.ajconline.org) 21. Swedberg K, Cleland J, Dargie H, Drexler H, Follath F, Komajda M, Tavazzi L,Smiseth OA, Gavazzi A, Haverich A, Hoes A, Jaarsma T, Korewicki J, Lvy S, Linde C, Lopez-Sendon JL, Nieminen MS, Pirard L, Remme WJ; Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005): the Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Eur Heart J 2005;26:11151140. 22. Pisters R, Lane DA, Nieuwlaat R, de Vos CB, Crijns HJ, Lip GY. A novel user-friendly score (HAS-BLED) to assess one-year risk of major bleeding in atrial brillation patients: the Euro Heart Survey. Chest. 2010;138:10931100.

18. Mujib M, Rahman AAZ, Desai RV, Ahmed MI, Feller MA, Aban I, Love TE, White M, Deedwania P, Aronow WS, Fonarow G, Ahmed A. Warfarin use and outcomes in patients with advanced chronic systolic heart failure without atrial brillation, prior thromboembolic events, or prosthetic valves. Am J Cardiol 2010;107:552557. 19. Hirsh J, Fuster V, Ansell J, Halperin JL; American Heart Association/ American College of Cardiology Foundation. American Heart Association/American College of Cardiology Foundation guide to warfarin therapy. J Am Coll Cardiol 2003;41:16331652. 20. Albers GW, Amarenco P, Easton JD, Sacco RL, Teal P. Antithrombotic and thrombolytic therapy for ischemic stroke: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126:483S512S.

Trials on the Effect of Cardiac Resynchronization on Arterial Blood Pressure in Patients With Heart Failure
Sameer Ather, MDa, Sripal Bangalore, MD, MHAb, Srinath Vemuri, MDa, Long B. Cao, MDa, Biykem Bozkurt, MD, PhDa, and Franz H. Messerli, MDc,*
Cardiac resynchronization therapy (CRT) increases cardiac performance in patients with heart failure, but its effect on arterial pressure is not well established. To determine the effect of CRT on systolic blood pressure (SBP), diastolic blood pressure (DBP), and pulse pressure (PP) a systematic review using standard nomenclatures for CRT was done in Scopus (MEDLINE and Embase), Cochrane Controlled Trials Register, National Institutes of Health http://www.ClinicalTrials.gov database, and bibliography of select meta-analyses for studies evaluating CRT in patients with dilated cardiomyopathy. Two independent investigators extracted the articles based on predened criteria. The primary outcome was difference in arterial pressure parameters from baseline to after CRT in nonrandomized cohort trials. This was then validated by comparing the change in arterial pressure between CRT and medical therapy groups in randomized controlled trials. A random-effects model was used for analyses. Analyses of 15 nonrandomized studies showed that CRT resulted in an increase (from baseline) in SBP by 4.4 mm Hg (95% condence interval [CI] 0.8 to 8.0, p 0.02), no change in DBP (p 0.21), and an increase in PP by 2.8 mm Hg (95% CI 1.0 to 4.6, p 0.003). Results from the 3 randomized controlled trials were concordant with an increase in SBP by 3.9 mm Hg (95% CI 1.1 to 6.8, p 0.007), no effect on DBP (p 0.40), and an increase in PP by 4.3 mm Hg (95% CI 4.1 to 4.5, p <0.001) compared to medical therapy. In conclusion, CRT is associated with a modest increase in SBP and PP in patients with heart failure. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107: 561568) More than 90% of patients with heart failure (HF) have a history of hypertension.1 In contrast, in severe HF decreasing left ventricular function is unable to sustain the high blood pressure (BP) despite compensatory mechanisms (salt and water retention, vasoconstriction, sympathetic stimulation and desensitization, cardiac hypertrophy, and cellular changes including appearance of slow myosin, prolongation of action potential, post-translational modications in calcium handling proteins, and increase in collagen).2,3 Thus, cardiac output decreases in parallel with systolic BP (SBP) and pulse pressure (PP).4 In patients with HF, cardiac resynchronization therapy (CRT) improves left ventricular systolic function,5 HF symptoms,6 quality of life,7 exercise tolerance,8 maladaptive remodeling,9 morbidity (HF admissions), and mortality.10 American College of Cardiology/American Heart Association guidelines11 recommend (class I) CRT in patients with ejection fraction 35%, QRS duration 120 ms, sinus rhythm, and New York Heart Association class III/ambulatory class IV HF symptoms on optimal medical therapy. However, it is not known if this improvement in systolic function translates into an increase in BP. Limited data seem to suggest an increase in SBP but this has not been consistently reported or studied.8,10
Baylor College of Medicine, Houston, Texas; bBrigham and Womens Hospital, Boston, Massachusetts; cSt. LukesRoosevelt Hospital Center and Columbia University, College of Physicians and Surgeons, New York, New York. Manuscript received September 22, 2010; revised manuscript received and accepted October 7, 2010. *Corresponding author: Tel: 212-523-7373; fax: 212-523-7765. E-mail address: messerli.f@gmail.com (F.H. Messerli). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.014
a

We hypothesized that CRT with its associated improvement in cardiac function would result in an increase in BP prole. Thus, our primary objective was to evaluate the effect of CRT on SBP, diastolic BP (DBP), and PP. Methods Eligible studies were prospective nonrandomized cohort studies that reported BP prole at baseline and after CRT with rst follow-up within 6 months of CRT or randomized controlled trials of CRT (with or without implantable cardioverter debrillator) that reported BP prole in the CRT and medical therapy groups. To avoid major medication changes over follow-up period and additional cardiovascular insults, a limit of 6 months of follow-up was used for nonrandomized cohort studies. Studies were identied by searching electronic databases, including Scopus (MEDLINE 1966 to October 2009, Embase 1980 to October 2009), Cochrane Trials Register, and National Institutes of Health http://www.ClinicalTrials.gov database (closed studies only) using the terms cardiac resynchronization or biventricular pacing or biventricular pacemaker or multisite pacing or multisite pacemaker or dual-site pacing or dual-site pacemaker or left ventricular pacing or left ventricular pacemaker. In addition, reference lists of select meta-analyses were searched for reports of relevant studies.1217 Studies in which the intervention included revascularization (coronary artery bypass grafting or percutaneous coronary intervention) at the time of CRT were excluded as revascularization could have potentially confounded the effect of CRT on BP outcomes (Figure 1).
www.ajconline.org

562

The American Journal of Cardiology (www.ajconline.org)

Figure 1. Schematic representation of data search and acquisition. DCMP dilated cardiomyopathy; F/U follow-up.

Eligible studies had to fulll the following inclusion criteria: (1) randomized or nonrandomized cohort studies of CRT in patients with dilated cardiomyopathy, (2) studies reporting outcomes of interest (SBP, DBP, or PP; before and after CRT in nonrandomized cohort studies and CRT vs medical therapy in randomized controlled trials), and (3) follow-up 6 months for nonrandomized cohort studies. There were no restrictions based on language or year of publication. Studies were restricted to published data. Studies that had duplicated data, including same group of patients or for whom there were updated results available, were excluded. We included only studies that did not exclude nonresponders from their analyses to prevent bias towards a positive result. Further, studies including patients with ischemic and/or nonischemic cardiomyopathy were included in this meta-analysis, whereas studies evaluating the effect of CRT, specically in pediatric patients, congenital heart disease, hypertrophic cardiomyopathy, restrictive cardiomyopathy, chemotherapy-induced cardiomyopathy, and infectious cardiomyopathy, e.g., Chagas disease, were excluded. Primary analyses were changes in BP parameters, i.e., SBP, DBP, and PP from baseline (before CRT) to that at follow-up (after CRT) in nonrandomized cohort studies. This was validated by comparing changes in BP parameters between a CRT group (with/without implantable cardioverter debrillator) and a medical therapy group (with/ without implantable cardioverter debrillator) in randomized controlled trials. The 2 analyses were done separately without pooling the data. Eligibility assessment and data abstraction were performed independently by 2 authors (S.V. and L.B.C) and supervised by S.A. We extracted inclusion criteria, exclusion criteria, baseline data, outcomes, and report quality. Disagreements were resolved by consensus. To assess risk of bias in nonrandomized cohort studies,

presence of single or double blinding and documentation of withdrawal were ascertained. For nonrandomized cohort studies, intermediate risk of bias was dened as a low possibility of bias in the 2 domains. As the studies were nonrandomized, none were considered at low risk. For randomized controlled trials, methodologic quality was assessed by reported allocation generation, allocation concealment, blinding, documentation of withdrawal, selective reporting, and intention-to-treat analysis in line with the recommendation by the Cochrane Collaboration.18 For randomized controlled trials, high risk of bias was dened as a possibility of bias in 4 domains, moderate as a possibility of bias in 2 to 3 domains, and low risk as a possibility of bias in 1 domain. Statistical analyses were done using Comprehensive MetaAnalysis 2.2.046 in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement (for randomized controlled trials) and Meta-analysis Of Observational Studies in Epidemiology (MOOSE) statement for others. Mean difference was chosen as the principal measurement of effect as the unit of measurement was same across all studies. Studies included reported mean or difference in means and standard deviation (SD) or p value for the variables. If the SD was not available for 1 of the 3 variables (SBP, DBP, PP), then the SD was calculated using a beforeversus-after correlation of 0.7 from the other 2 SDs available. Data were analyzed for heterogeneity by I2 statistic proposed by Higgins and Thompson19 separately for nonrandomized cohort studies and randomized controlled trials. Values 30% indicated mild heterogeneity and those 50% substantial heterogeneity.19 In the presence of heterogeneity, a random effects model (DerSimonianLaird approach)20 was used to pool the data; otherwise, a xedeffects model (inverse variance) was used. Publication bias was assessed and quantied using the regression intercept of Egger et al21 and corrected by the trim-and-ll method of Duval and Tweedie.22 Results Fifteen nonrandomized cohort studies and 3 randomized controlled trials met our inclusion criteria for analyses (Figure 1). Of the 18 studies included in the meta-analyses, 15 (nonrandomized controlled trials) compared variables before and after CRT2337 in 492 patients, whereas 3 studies (randomized controlled trials) compared CRT (n 1,637) to optimal medical management (n 727).8,10,38 Baseline characteristics and inclusion and exclusion criteria are presented in Tables 1 and 2. There were 2 studies on the Pacing Therapies in Congestive Heart Failure (PATH-CHF) trial but separate variables were extracted from each study.23,24 Two trials with 3 published studies included patients with epicardial pacing,23,24,34 whereas the other studies used transvenous biventricular pacing. All nonrandomized cohort studies reported withdrawals or crossovers or had no withdrawal. In 1 study, patients were unaware of their treatment 27 and 1 study documented that it was single blinded but did not specify who was blinded.37 Thus, these 2 studies were considered at intermediate risk of bias, whereas the rest of the nonrandomized cohort studies were at high risk of bias. Among the randomized controlled trials, based on the 6 parameters sug-

Heart Failure/Cardiac Resynchronization and Blood Pressure Table 1 Baseline characteristics of included trials Study Randomized controlled trials* Piccirillo et al,38 2006 Bristow et al,8 2004 Cleland et al,10 2005 Nonrandomized cohort studies Auricchio et al,24 1999 Auricchio et al,23 2003 Bakker et al,25 2000 Braunschweig et al,26 2000 Flevari et al,27 2006 Fung et al,28 2008 Inage et al,29 2008 Knaapen et al,30 2004 Kubanek et al,31 2006 Madaric et al,32 2007 Mullens et al,33 2008 Nelson et al,34 2000 Sogaard et al,35 2001 Vanderheyden et al,36 2008 Waggoner et al,38 2006 Subjects Age (years) 65/65 68/67 66/67 62 60 64 64 66 64 63 58 62 67 66 59 61 64 61 NYHA Class 3.7/3.7 3.2/3.1 3.1/3.1 3.2 NA 3.7 3.1 3.2 3.2 3.2 3.1 3.2 3 3.4 NA 3.4 NA 3.2 Baseline EF (%) 0.22/0.23 0.22/0.21 0.25/0.25 0.21 0.23 0.15 0.22 0.27 0.26 0.25 0.25 0.22 0.25 0.21 0.2 0.23 0.19 0.25 QRS Width (ms) 159/160 158/160 160/160 168 155 194 181 195 NA 161 173 195 171 177 175 184 179 180 Sinus Rhythm (%) 1/1 1/1 1/1 1 1 1 0.56 1 1 NA 1 0.79 1 NA 1 1 1 1 Ischemic Cause (%) 1/1 0.59/0.54 0.36/0.40 0.33 0.38 0.33 0.69 0.44 0.42 0.12 0.43 0.42 0.5 0.55 0.23 0.56 0.4 0.25 Men (%)

563

EF After CRT 0.28/0.22 NA NA NA NA 0.21 NA 0.34 0.34 0.33 0.37 0.24 0.28 0.34 NA 0.23 0.34 0.34

15/16 308/1,212 404/409 25 85 12 16 25 97 17 14 43 28 19 22 22 10 57

0.80/0.81 0.69/0.67 0.73/0.74 0.56 0.66 0.42 0.94 0.76 0.74 0.65 0.57 0.86 0.82 0.8 NA 0.88 0.8 0.76

* For randomized controlled trials data are presented as medical therapy/cardiac resynchronization arms. NA absence of data; NYHA New York Heart Association; EF ejection fraction.

gested by the Cochrane Review, the Cardiac ResynchronizationHeart Failure (CARE-HF)10 and Comparison of Medical Therapy, Pacing, and Debrillation in Heart Failure (COMPANION)8 trials were at low risk of bias with low risk in 4 of 6 categories, whereas the third study by Piccirillo et al38 was at high risk of bias across all categories. In nonrandomized cohort studies, compared to baseline there was a signicant increase in SBP (difference 4.4 mm Hg, 95% condence interval [CI] 0.8 to 8.0, p 0.02) after CRT (Figure 2). This was concordant in the randomized controlled trials, where there was a signicantly higher SBP (difference 3.9 mm Hg, 95% CI 1.1 to 6.8, p 0.007) in the CRT group compared to the medical therapy alone group (Figure 3). In nonrandomized cohort studies, compared to baseline there was no change in DBP (difference 1.0 mm Hg, 95% CI 0.6 to 2.6, p 0.20) after CRT (Figure 4). Similarly, in the randomized controlled trials, there was no change in DBP (difference 0.5 mm Hg, 95% CI 0.7 to 1.7, p 0.40) in the CRT group compared to the medical therapy alone group (Figure 5). In nonrandomized cohort studies, compared to baseline there was a signicant increase in PP (difference 2.8 mm Hg, 95% CI 1.0 to 4.6, p 0.003) after CRT (Figure 6). Similarly, in randomized controlled trials, there was a signicant increase in PP (difference 4.3 mm Hg, 95% CI 4.1 to 4.5, p 0.001) in the CRT group compared to the medical therapy alone group (Figure 7). There was signicant heterogeneity for analyses of SBP, DBP, and PP. Sensitivity analyses based on follow-up period, year of publication, size of study, New York Heart Association class, mean ejection fraction, mean QRS duration, presence of ischemic cardiomyopathy, and industry funding were performed but did not

change the assessed heterogeneity. Thus, a random-effects model was used to minimize the effect of heterogeneity. There was no signicant publication bias based on the Egger regression intercept in any of the comparisons (funnel plots not included). There were 2 studies each with low and intermediate risk of bias. On redoing the analysis with only low-/ intermediate-risk studies, there was a trend toward an increase in SBP compared to baseline in 2 nonrandomized cohort studies after CRT (difference 4.8 mm Hg, 95% CI 0.6 to 10.1, p 0.08), whereas SBP was signicantly higher in the CRT group compared to the optimal medical therapy group in the 2 randomized controlled trials (difference 3.9 mm Hg, 95% CI 0.6 to 7.3, p 0.02).8,10,27,37 Similarly, on analysis of low-/intermediate-risk studies, the 2 nonrandomized cohort trials showed a trend towards an increase in PP (difference 2.5 mm Hg, 95% CI 0.5 to 5.6, p 0.099) compared to baseline, whereas the CARE-HF trial showed a signicant increase in PP in the CRT group compared to the optimal medical therapy group (difference 4.3 mm Hg, 95% CI 4.1 to 4.5, p 0.001).10,27,37 There was no effect of CRT on DBP in low-/intermediate-risk studies in randomized controlled trials or nonrandomized cohort trials. In addition, to rule out the effect of change in medication on BP prole, we analyzed 11 nonrandomized cohort studies in which none of the BP medications were changed during follow-up. In this subanalysis there was no change in the results with an increase in SBP (difference 5.6 mm Hg, 95% CI 1.0 to 10.2, p 0.02),24 26,29,30,3236 an increase in PP (difference 2.8 mm Hg, 95% CI 0.5 to 5.1, p 0.02),23,25,26,30,32,34 and no effect on DBP (difference 0.4 mm Hg, 95% CI 1.7 to 2.6, p 0.4).24 26,30,32,34

564

The American Journal of Cardiology (www.ajconline.org)

Table 2 Inclusion criteria and baseline blood pressure prole in included trials Study Randomized controlled trials* Piccirillo et al,38 2006 Bristow et al,8 2004 Cleland et al,10 2005 Nonrandomized cohort studies Auricchio et al,24 1999 Auricchio et al,23 2003 Bakker et al,25 2000 Braunschweig et al,26 2000 Flevari et al,27 2006 Fung et al,28 2008 Inage et al,29 2008 Knaapen et al,30 2004 Kubanek et al,31 2006 Madaric et al,32 2007 Mullens et al,33 2008 Inclusion Criteria Patients with HF with NYHA class III, EF 0.35, QRS duration 120 ms, sinus rhythm, and ischemic cause Patients with HF with NYHA class III, EF 0.35, QRS duration 120 ms, sinus rhythm, PR 150 ms, HF hospitalization in previous year Patients with HF 18 years old with NYHA class III, EF 0.35, QRS duration 120 ms, sinus rhythm, LVEDD 30 mm Patients with HF with NYHA class III, QRS duration 120 ms, sinus rhythm, PR 150 ms Patients with HF 1875 years old with NYHA class II, EF 0.30, QRS duration 120 ms, sinus rhythm, peak VO2 18 ml/min/kg on maximal exercise Patients with HF 1875 years old with NYHA class III, LBBB with QRS duration 120 ms, sinus rhythm Patients with HF with NYHA class III, EF 0.40, QRS duration 150 ms Patients with HF with NYHA class III, EF 0.35, QRS duration 120 ms, sinus rhythm Standard CRT indication: NYHA class III, EF 0.35, QRS duration 120 ms Patients with HF with NYHA class III, QRS duration 120 ms Patients with HF with NYHA class III, EF 0.35, QRS duration 120 ms, sinus rhythm, LVEDD 55 mm Patients with HF with NYHA class III, EF 0.35, QRS duration 140 ms, LVEDD 60 mm Standard indication for CRT: NYHA class III, EF 0.35, QRS duration 120 ms, sinus rhythm Patients with HF with NYHA class III, EF 0.35, LBBB with QRS duration 120 ms, HR 70 beats/min, sum of dyssynchrony 102 ms, preserved AV conduction Patients with HF with NYHA class III, EF 0.35, QRS duration 140 ms, sinus rhythm Patients with HF with NYHA class III, QRS duration 120 ms, sinus rhythm Patients with HF with NYHA class III, EF 0.25, LBBB with QRS duration 140 ms, sinus rhythm, total sum of dyssynchrony 102 ms, HR 70 beats/ min Patients with HF with NYHA class III, EF 0.35, QRS duration 150 ms, sinus rhythm, LVEDD 60 mm SBP (mm Hg) 109/112 112/111 110/110 DBP (mm Hg) 69/68 64/68 70/70

90 113 NA 109 110 105 98 114 120 112 111

57 NA NA 70 70 NA NA 71 76 71 NA

Nelson et al,34 2000 Sogaard et al,35 2001 Vanderheyden et al,36 2008

110 103 113

72 NA NA

Waggoner et al,37 2006

113

67

* Randomized controlled trials data are presented as medical therapy/cardiac resynchronization arms. AV atrioventricular; HR heart rate; LBBB left bundle branch block; LVEDD left ventricular end-diastolic dimension; VO2 oxygen consumption. Other abbreviations as in Table 1.

Figure 2. Effect of cardiac resynchronization therapy on systolic blood pressure shown by paired comparison of baseline data to data after cardiac resynchronization therapy from nonrandomized cohort studies. *p 0.05; **p 0.01; ***p 0.001. difference.

Heart Failure/Cardiac Resynchronization and Blood Pressure

565

Figure 3. Effect of cardiac resynchronization therapy on systolic blood pressure shown by comparison of cardiac resynchronization therapy to optimal medical therapy from randomized controlled trials. *p 0.05; **p 0.01; ***p 0.001. Abbreviation as in Figure 2.

Figure 4. Effect of cardiac resynchronization therapy on diastolic blood pressure shown by paired comparison of baseline data to data after cardiac resynchronization therapy from nonrandomized cohort studies. *p 0.05; **p 0.01; ***p 0.001. Abbreviation as in Figure 2.

Figure 5. Effect of cardiac resynchronization therapy on diastolic blood pressure by comparison of cardiac resynchronization therapy to optimal medical therapy from randomized controlled trials. *p 0.05; **p 0.01; ***p 0.001. Abbreviation as in Figure 2.

Figure 6. Effect of cardiac resynchronization therapy on pulse pressure shown by paired comparison of baseline data to data after cardiac resynchronization therapy from nonrandomized cohort studies. *p 0.05; **p 0.01; ***p 0.001. Abbreviation as in Figure 2.

Discussion Our results indicate that in patients with HF with standard indications for CRT, there was a moderate increase in SBP and PP compared to baseline. This was

conrmed in analyses of patients from randomized controlled trials, where there was a modest increase in BP parameters compared to medical therapy alone. However, we did not nd any difference in DBP between groups.

566

The American Journal of Cardiology (www.ajconline.org)

Figure 7. Effect of cardiac resynchronization therapy on pulse pressure shown by comparison of cardiac resynchronization therapy to optimal medical therapy from randomized controlled trials. *p 0.05; **p 0.01; ***p 0.001. Abbreviation as in Figure 2.

There is a close correlation between HF and hypertension. Hypertension ultimately leads to HF and patients with HF have a high prevalence of hypertension. However, in patients with advanced HF, a low SBP is usually seen even in patients who were previously hypertensive. This is termed decapitated hypertension in which patients who are hypertensive to begin with develop normal/low BP as HF progresses. This results from decreasing pump function and cardiac output despite the presence of compensatory mechanisms such as peripheral vasoconstriction. This inability to generate higher SBP is accepted as an indicator of poor pump function. Patients with decapitated hypertension are difcult to manage because of their inability to tolerate HF medications that can potentially lower BP such as angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, diuretics, and blockers. Although CRT increases ejection fraction in patients with HF, increase in BP has not been conclusively shown. Our results show that CRT is associated with a modest increase in BP in this patient population, which can potentially lead to reversal of decapitated hypertension. This potential increase in SBP using CRT can provide a window to the treating cardiologist to introduce/continue HF medications. This interplay among high BP, hypertensive HF, and dilated cardiomyopathy was lucidly discussed by Oakley39 about 3 decades ago: The development of left ventricular failure because of hypertension determines a decrease of the previously raised BP to normal levels and, since the failure usually persists the BP remains normal. If the patient recovers from HF, then the BP rises and the diagnosis is likely to be hypertension. In other words, the diagnosis varies between dilated cardiomyopathy and hypertension according to left ventricular function and only if a patient with dilated cardiomyopathy, HF and normal BP actually recovers and develops high BP can the causal or conditioning role of high BP be proved. Although hypertension is associated with development of incident HF,40 higher SBP has a protective survival effect in patients with established HF.41,42 In our analysis, CRT increased SBP in patients with advanced HF who had normal to low-normal SBP. Whether this improvement in BP improves survival is not clearly dened. In a recent study by Tanaka et al,43 an increase in SBP after CRT was associated with a decrease in the combined end point of HF hospitalization and all-cause mortality. However, further studies are needed to prove if improvement in arterial pressure improves survival independent of ejection fraction.

In the setting of decreased cardiac output, there is sympathetic activation and parasympathetic withdrawal.44 CRT improves cardiac function by reverting asynchronous mechanical events in patients with HF, especially in patients with a wide QRS complex or with echocardiographic dyssynchrony. Moreover, this improvement in myocardial function is associated with restitution of the defunct autonomic function in patients with HF.45 Thus, improvement in myocardial function without a concomitant increase in adrenergic activity could be a possible mechanism behind improvement in mortality using CRT. Given the paucity of data, we used both randomized controlled trials and nonrandomized cohort studies for our analysis. The effect of CRT on BP prole was heterogeneous. Because extensive sensitivity analyses could not explain this heterogeneity, a random-effects model was used to mitigate the effect of this heterogeneity. As a result, although we could pool the studies, we were not able to explain the possible cause of this heterogeneity.
1. Levy D, Larson MG, Vasan RS, Kannel WB, Ho KK. The progression from hypertension to congestive heart failure. JAMA 1996;275:1557 1562. 2. Kushnir A, Shan J, Betzenhauser MJ, Reiken S, Marks AR. Role of CaMKIIdelta phosphorylation of the cardiac ryanodine receptor in the force frequency relationship and heart failure. Proc Natl Acad Sci U S A 2010;107:10274 10279. 3. Katz AM. Cardiomyopathy of overload. A major determinant of prognosis in congestive heart failure. N Engl J Med 1990;322:100 110. 4. Nohria A, Lewis E, Stevenson LW. Medical management of advanced heart failure. JAMA 2002;287:628 640. 5. Kawaguchi M, Murabayashi T, Fetics BJ, Nelson GS, Samejima H, Nevo E, Kass DA. Quantitation of basal dyssynchrony and acute resynchronization from left or biventricular pacing by novel echocontrast variability imaging. J Am Coll Cardiol 2002;39:2052 2058. 6. Auricchio A, Stellbrink C, Sack S, Block M, Vogt J, Bakker P, Huth C, Schondube F, Wolfhard U, Bocker D, Krahnefeld O, Kirkels H. Pacing Therapies in Congestive Heart Failure (PATHCHF) Study Group. Long-term clinical effect of hemodynamically optimized cardiac resynchronization therapy in patients with heart failure and ventricular conduction delay. J Am Coll Cardiol 2002; 39:2026 2033. 7. Werling C, Weisse U, Siemon G, Kiessling AH, Rameken M, Schwacke H, Saggau W, Senges J, Seidl K. Biventricular pacing in patients with ICD: how many patients are possible candidates? J Thorac Cardiovasc Surg 2002;50:6770. 8. Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T, Carson P, DiCarlo L, DeMets D, White BG, DeVries DW, Feldman AM. Comparison of Medical Therapy, Pacing, and Debrillation in Heart Failure (COMPANION) Investigators. Cardiacresynchronization therapy with or without an implantable debrillator in advanced chronic heart failure. N Engl J Med 2004;350: 2140 2150.

Heart Failure/Cardiac Resynchronization and Blood Pressure 9. St John Sutton MG, Plappert T, Abraham WT, Smith AL, DeLurgio DB, Leon AR, Loh E, Kocovic DZ, Fisher WG, Ellestad M, Messenger J, Kruger K, Hilpisch KE, Hill MR, Multicenter InSync Randomized Clinical Evaluation (MIRACLE) Study Group. Effect of cardiac resynchronization therapy on left ventricular size and function in chronic heart failure. Circulation 2003;107:19851990. 10. Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, Tavazzi L. Cardiac ResynchronizationHeart Failure (CARE-HF) Study Investigators. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005; 352:1539 1549. 11. Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA III, Freedman RA, Gettes LS, Gillinov AM, Gregoratos G, Hammill SC, Hayes DL, Hlatky MA, Newby LK, Page RL, Schoenfeld MH, Silka MJ, Stevenson LW, Sweeney MO, Smith SC Jr, Jacobs AK, Adams CD, Anderson JL, Buller CE, Creager MA, Ettinger SM, Faxon DP, Halperin JL, Hiratzka LF, Hunt SA, Krumholz HM, Kushner FG, Lytle BW, Nishimura RA, Ornato JP, Page RL, Riegel B, Tarkington LG, Yancy CW, American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/ AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices), American Association for Thoracic Surgery, Society of Thoracic Surgeons. ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices): developed in collaboration with the American Association for Thoracic Surgery and Society of Thoracic Surgeons. Circulation 2008;117(suppl):e350 e408. 12. Bradley DJ, Bradley EA, Baughman KL, Berger RD, Calkins H, Goodman SN, Kass DA, Powe NR. Cardiac resynchronization and death from progressive heart failure: a meta-analysis of randomized controlled trials. JAMA 2003;289:730 740. 13. Rivero-Ayerza M, Theuns DA, Garcia-Garcia HM, Boersma E, Simoons M, Jordaens LJ. Effects of cardiac resynchronization therapy on overall mortality and mode of death: a meta-analysis of randomized controlled trials. Eur Heart J 2006;27:26822688. 14. Lam SK, Owen A. Combined resynchronisation and implantable debrillator therapy in left ventricular dysfunction: Bayesian network meta-analysis of randomised controlled trials. BMJ 2007;335:925. 15. McAlister FA, Ezekowitz JA, Wiebe N, Rowe B, Spooner C, Crumley E, Hartling L, Klassen T, Abraham W. Systematic review: cardiac resynchronization in patients with symptomatic heart failure. Ann Intern Med 2004;141:381390. 16. McAlister FA, Ezekowitz J, Hooton N, Vandermeer B, Spooner C, Dryden DM, Page RL, Hlatky MA, Rowe BH. Cardiac resynchronization therapy for patients with left ventricular systolic dysfunction: a systematic review. JAMA 2007;297:25022514. 17. Abdulla J, Haarbo J, Kober L, Torp-Pedersen C. Impact of implantable debrillators and resynchronization therapy on outcome in patients with left ventricular dysfunctiona meta-analysis. Cardiology 2006; 106:249 255. 18. Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.2 (updated September 2009). The Cochrane Collaboration, 2009. Available at: www.cochrane-handbook.org. 19. Higgins JP, Thompson SG. Quantifying heterogeneity in a metaanalysis. Stat Med 2002;21:1539 1558. 20. DerSimonian R, Laird N. Meta-analysis in clinical trials. Controlled Clin Trials 1986;7:177188. 21. Egger M, Davey Smith G, Schneider M, Minder C. Bias in metaanalysis detected by a simple, graphical test. BMJ 1997;315:629 634. 22. Duval S, Tweedie R. Trim and ll: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 2000;56:455 463. 23. Auricchio A, Stellbrink C, Butter C, Sack S, Vogt J, Misier AR, Bocker D, Block M, Kirkels JH, Kramer A, Huvelle E. Pacing therapies in Congestive Heart Failure II Study Group, Guidant Heart Failure Research Group. Clinical efcacy of cardiac resynchronization therapy using left ventricular pacing in heart failure patients stratied by severity of ventricular conduction delay. J Am Coll Cardiol 2003;42:2109 2116. 24. Auricchio A, Stellbrink C, Block M, Sack S, Vogt J, Bakker P, Klein H, Kramer A, Ding J, Salo R, Tockman B, Pochet T, Spinelli J. The

567

25.

26. 27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39. 40.

Pacing Therapies for Congestive Heart Failure Study Group. The Guidant Congestive Heart Failure Research Group. Effect of pacing chamber and atrioventricular delay on acute systolic function of paced patients with congestive heart failure. Circulation 1999;99: 29933001. Bakker PF, Meijburg HW, de Vries JW, Mower MM, Thomas AC, Hull ML, Robles De Medina EO, Bredee JJ. Biventricular pacing in end-stage heart failure improves functional capacity and left ventricular function. J Interv Card Electrophysiol 2000;4:395 404. Braunschweig F, Linde C, Gadler F, Ryden L. Reduction of hospital days by biventricular pacing. Eur J Heart Fail 2000;2:399 406. Flevari P, Theodorakis G, Paraskevaidis I, Kolokathis F, Kostopoulou A, Leftheriotis D, Kroupis C, Livanis E, Kremastinos DT. Coronary and peripheral blood ow changes following biventricular pacing and their relation to heart failure improvement. Europace 2006;8:44 50. Fung JW, Yip GW, Zhang Q, Fang F, Chan JY, Li CM, Wu LW, Chan GC, Chan HC, Yu CM. Improvement of left atrial function is associated with lower incidence of atrial brillation and mortality after cardiac resynchronization therapy. Heart Rhythm 2008;5:780 786. Inage T, Yoshida T, Hiraki T, Ohe M, Takeuchi T, Nagamoto Y, Fukuda Y, Gondo T, Imaizumi T. Chronic cardiac resynchronization therapy reverses cardiac remodelling and improves invasive haemodynamics of patients with severe heart failure on optimal medical treatment. Europace 2008;10:379 383. Knaapen P, van Campen LM, de Cock CC, Gotte MJ, Visser CA, Lammertsma AA, Visser FC. Effects of cardiac resynchronization therapy on myocardial perfusion reserve. Circulation 2004;110:646 651. Kubanek M, Malek I, Bytesnik J, Fridl P, Riedlbauchova L, Karasova L, Lanska V, Kautzner J. Decrease in plasma B-type natriuretic peptide early after initiation of cardiac resynchronization therapy predicts clinical improvement at 12 months. Eur J Heart Fail 2006;8:832 840. Madaric J, Vanderheyden M, Van Laethem C, Verhamme K, Feys A, Goethals M, Verstreken S, Geelen P, Penicka M, De Bruyne B, Bartunek J. Early and late effects of cardiac resynchronization therapy on exercise-induced mitral regurgitation: relationship with left ventricular dyssynchrony, remodelling and cardiopulmonary performance. Eur Heart J 2007;28:2134 2141. Mullens W, Bartunek J, Wilson Tang WH, Delrue L, Herbots L, Willems R, De Bruyne B, Goethals M, Verstreken S, Vanderheyden M. Early and late effects of cardiac resynchronization therapy on force-frequency relation and contractility regulating gene expression in heart failure patients. Heart Rhythm 2008;5:5259. Nelson GS, Curry CW, Wyman BT, Kramer A, Declerck J, Talbot M, Douglas MR, Berger RD, McVeigh ER, Kass DA. Predictors of systolic augmentation from left ventricular preexcitation in patients with dilated cardiomyopathy and intraventricular conduction delay. Circulation 2000;101:27032709. Sogaard P, Kim WY, Jensen HK, Mortensen P, Pedersen AK, Kristensen BO, Egeblad H. Impact of acute biventricular pacing on left ventricular performance and volumes in patients with severe heart failure. A tissue Doppler and three-dimensional echocardiographic study. Cardiology 2001;95:173182. Vanderheyden M, Mullens W, Delrue L, Goethals M, Verstreken S, Wijns W, de Bruyne B, Bartunek J. Endomyocardial upregulation of beta1 adrenoreceptor gene expression and myocardial contractile reserve following cardiac resynchronization therapy. J Card Fail 2008; 14:172178. Waggoner AD, Rovner A, de las Fuentes L, Faddis MN, Gleva MJ, Sawhney N, Davila-Roman VG. Clinical outcomes after cardiac resynchronization therapy: importance of left ventricular diastolic function and origin of heart failure. J Am Soc Echocardiogr 2006;19:307 313. Piccirillo G, Magri D, di Carlo S, De Laurentis T, Torrini A, Matera S, Magnanti M, Bernardi L, Barilla F, Quaglione R, Ettorre E, Marigliano V. Inuence of cardiac-resynchronization therapy on heart rate and blood pressure variability: 1-year follow-up. Eur J Heart Fail 2006;8:716 722. Oakley C. Diagnosis and natural history of congested (dilated) cardiomyopathies. Postgrad Med J 1978;54:440 450. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, Jones DW, Materson BJ, Oparil S, Wright JT Jr, Roccella EJ, National Heart, Lung, and Blood Institute, Joint National Committee on Prevention. Detection, evaluation, and treat-

568

The American Journal of Cardiology (www.ajconline.org) 43. Tanaka Y, Tada H, Yamashita E, Sato C, Irie T, Hori Y, Goto K, Iwamoto J, Manni H, Yokokawa M, Naito S, Oshima S, Taniguchi K. Change in blood pressure just after initiation of cardiac resynchronization therapy predicts long-term clinical outcome in patients with advanced heart failure. Circ J 2009;73:288 294. 44. Cohn JN. The management of chronic heart failure. N Engl J Med 1996;335:490 498. 45. Marijon E, Boveda S, Chevalier P, Bulava A, Winter JB, Lambiez M, Defaye P, on behalf of the Mona Lisa Study Group. Monitoring of heart rate variability in heart failure patients with cardiac resynchronisation therapy: interest of continuous and didactic algorithm. Int J Cardiol 2010;144:166169.

ment of high blood pressure. National High Blood Pressure Education Program Coordinating Committee. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;289:2560 2572. 41. Cheng RK, Horwich TB, Fonarow GC. Relation of systolic blood pressure to survival in both ischemic and nonischemic systolic heart failure. Am J Cardiol 2008;102:1698 1705. 42. Lee TT, Chen J, Cohen DJ, Tsao L. The association between blood pressure and mortality in patients with heart failure. Am Heart J 2006;151:76 83.

Patient Perception Versus Medical Record Entry of Health-Related Conditions Among Patients With Heart Failure
Adnan S. Malik, MDa, Grigorios Giamouzis, MD, PhDb, Vasiliki V. Georgiopoulou, MDa, Lucy V. Fike, MPHa, Andreas P. Kalogeropoulos, MDa, Catherine R. Norton, MDa, Dan Sorescu, MDa, Sidra Azim, MDa, Sonjoy R. Laskar, MDa, Andrew L. Smith, MDa, Sandra B. Dunbar, RN, DSNc, and Javed Butler, MD, MPHa,*
A shared understanding of medical conditions between patients and their health care providers may improve self-care and outcomes. In this study, the concordance between responses to a medical history self-report (MHSR) form and the corresponding provider documentation in electronic health records (EHRs) of 19 select co-morbidities and habits in 230 patients with heart failure were evaluated. Overall concordance was assessed using the statistic, and crude, positive, and negative agreement were determined for each condition. Concordance between MHSR and EHR varied widely for cardiovascular conditions ( 0.37 to 0.96), noncardiovascular conditions ( 0.06 to 1.00), and habits ( 0.26 to 0.69). Less than 80% crude agreement was seen for history of arrhythmias (72%), dyslipidemia (74%), and hypertension (79%) among cardiovascular conditions and lung disease (70%) and peripheral arterial disease (78%) for noncardiovascular conditions. Perfect agreement was observed for only 1 of the 19 conditions (human immunodeciency virus status). Negative agreement >80% was more frequent than >80% positive agreement for a condition (15 of 19 [79%] vs 8 of 19 [42%], respectively, p 0.02). Only 20% of patients had concordant MSHRs and EHRs for all 7 cardiovascular conditions; in 40% of patients, concordance was observed for <5 conditions. For noncardiovascular conditions, only 28% of MSHR-EHR pairs agreed for all 9 conditions; 37% agreed for <7 conditions. Cumulatively, 39% of the pairs matched for <15 of 19 conditions. In conclusion, there is signicant variation in the perceptions of patients with heart failure compared to providers records of co-morbidities and habits. The root causes of this variation and its impact on outcomes need further study. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107: 569 572) Heart failure (HF) prevalence is growing and primarily affects the elderly.1 The complex array of physiologic, psychological, social, and health care delivery issues that accompany HF make it a difcult chronic disease to manage.2 Optimal self-care behavior is important for achieving the best outcomes for chronic diseases such as HF. For patients to actively participate in their care, however, it is important for them to have a clear understanding of their healthrelated problems.3 This is particularly critical for patients with HF, as they tend to be older, have a higher co-morbidity burden, and often require complex treatment plans.4 From a provider perspective, medical record documentation
Cardiology Division, Emory University School of Medicine, Atlanta, Georgia; bDepartment of Cardiology, Larissa University Hospital, Larissa, Greece; and cEmory University School of Nursing, Atlanta, Georgia. Manuscript received August 13, 2010; revised manuscript received and accepted October 11, 2010. Funding: This project was funded by the Emory University Heart and Vascular Board grant titled The Atlanta Cardiomyopathy Consortium, and supported in part by PHS grant (UL1 RR025008, KL2 RR025009 or TL1 RR025010) from the Clinical and Translational Science Award program, National Institutes of Health, National Center for Research Resources. *Corresponding author: Tel: 404-778-5273; fax: 404-778-5285. E-mail address: javed.butler@emory.edu (J. Butler). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.017
a

of disease states is an essential part of care provision.5 This is especially true in the current era of increasing use of electronic health records (EHRs), as many providers communicate information exclusively through this medium.6 It may be assumed that what is documented in EHRs is the same as patients understanding and reporting. However, if this is not true, this discordance may lend itself to poor patient self-care behavior (related to not understanding or not reporting their conditions) or to insufcient medical care (due to misunderstanding by providers). In the current era, whether EHR entries are congruent with patients reporting of health-related conditions, and to what extent, is not known. In this study, we sought to assess and compare patient self-report versus EHR documentation of cardiovascular and noncardiovascular conditions and behavioral habits in patients with HF. Methods The data for this study were derived from patients enrolled in the Atlanta Cardiomyopathy Consortium. This prospective cohort study is enrolling patients from the Emory University Hospital, Emory University Hospital Midtown, and the Grady Memorial Hospital in Atlanta, Georgia. All patients undergo detailed medical history surveys, electrocardiography, 6-minute walk tests, standardwww.ajconline.org

570

The American Journal of Cardiology (www.ajconline.org) Table 1 Baseline characteristics (n 230) Characteristic Age (years) Male White Education (years) Living alone Insured Married Ischemic cause of HF Left ventricular ejection fraction (%) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Heart rate (beats/min) Creatinine (mg/dl) Sodium (mEq/L) Hemoglobin (g/dl) Brain natriuretic peptide (ng/L) -blocker use Angiotensin-converting enzyme inhibitor or angiotensin receptor blocker use Debrillator/pacemaker Value 56.6 11.9 149 (64.8%) 127 (55.2%) 14.1 3.1 41 (17.9%) 212 (92.2%) 143 (62.2%) 69 (31.3%) 39.3 14.6 112 18 71 11 72 11 1.4 1.1 138 3 13.3 1.8 202 (73664) 219 (94.8%) 197 (85.6%) 145 (64.5%)

ized questionnaires, and collection of blood and urine samples at baseline. Every 6 months, patients are contacted to assess outcomes, including interim medication changes, procedures, new disease diagnoses, and hospitalizations. Mortality data are collected through medical record review, information obtained from family members, and Social Security Death Index query. The institutional review board has approved the study. At the time of this analysis, a total of 238 patients were enrolled; we included 230 of these patients (96.6%), excluding 8 patients who did not complete medical history surveys. Research nurses abstracted data from the EHRs independently without discussion with the patients or their survey documentation. The main source of EHR data (n 222 [96.5%]) was Emory Healthcares electronic medical record system, which is based on the Cerner Millennium platform (Cerner Corporation, Kansas City, Missouri). The system provides a comprehensive view of clinical data collected across hospitals and clinics. Data on 8 patients (3.5%) were collected from the EHR system at the Grady Memorial Hospital, which is based on the Siemens Medical Solutions (Malvern, Pennsylvania) Health Services platform. All patients completed a medical history self-report (MHSR) form, which included questions regarding cardiovascular conditions (history of heart attack or myocardial infarction, high blood pressure or hypertension, high cholesterol, heart rhythm problems or arrhythmias, coronary artery bypass graft surgery, coronary stent placement, and implantable cardioverter debrillator or pacemaker implantation), and noncardiovascular conditions (diabetes mellitus, peripheral arterial disease, pulmonary disease, liver disease, peptic ulcer disease, thyroid disease, cancer, osteoarthritis, and human immunodeciency virus (HIV) infection). The pulmonary disease question was open ended, allowing patients to manually enter specic diagnoses. Data on history of tobacco, alcohol, and cocaine use were also obtained. To assess the reliability of EHR data abstraction, data on 10% of the total charts, selected using a random number generator (http://www.random.org), were independently abstracted. Cumulative agreement between the 2 independent EHR data abstractions for all study variables was 93.1%. EHR data for each condition (yes or no) were compared with the data from MHSR forms (yes or no), and concordance was assessed using the statistic. Crude, positive, and negative agreement were calculated to facilitate interpretation of values.7 Crude agreement is equal to the number of pairs that agree divided by the number of pairs available for analysis. The number of pairs available differed for each condition because of missing values in patient responses. Positive and negative agreement measures were calculated. The positive agreement measure is the ratio of total concordant positive responses over the average positive responses of patients and EHRs. The negative agreement is the ratio of total concordant negative responses over the average negative responses of patients and EHRs. Kappa statistics were interpreted as follows8: values of 0.93 to 1.00 denote almost perfect agreement, 0.81 to 0.92 very good agreement, 0.61 to 0.80 substantial agreement, 0.41 to 0.60 moderate agreement, 0.21 to 0.40 fair agreement, 0.01 to 0.20 slight agreement, and 0 no agreement. Finally, to

Data are expressed as mean SD, number (percentage), or median (interquartile range).

summarize agreement by patient, the sum of the number of concordant conditions per participant was calculated. There were 7 cardiovascular and 9 noncardiovascular conditions and 3 habits included in the summary measure. McNemars statistic was calculated for paired comparisons. Finally, patients responses as dont know to select conditions were captured and compared with EHR data. All analyses were performed using SAS version 9.2 (SAS Institute Inc., Cary, North Carolina). Results The baseline patient characteristics and treatment pattern are listed in Table 1. The mean age of patients was 56.6 11.9 years; 64.5% were men, and 55.2% were white. The mean left ventricular ejection fraction was 39.3 14.6%. Table 2 lists the agreement data. There was fair agreement for arrhythmia history and moderate agreement for dyslipidemia and hypertension. The strongest agreement was noted for procedural care, including coronary artery bypass grafting and implantable cardioverter-debrillator and/or pacemaker implantation. For noncardiovascular conditions, there was only fair agreement for pulmonary disease; of the 47 of 82 patients (57%) who entered specic diagnoses, there was poor agreement for chronic obstructive pulmonary disease, asthma, and sleep apnea (not listed in Table 2). There was very good agreement for cancer and diabetes mellitus and perfect agreement for HIV. For alcohol use, there was fair agreement. In 80% of the patients (12 of 15) in whom there was disagreement, the patients did not report alcohol use when the EHRs suggested histories. There was moderate agreement for cocaine and tobacco use. Dont know responses were uncommon, including 13 for myocardial infarction (12 had no EHR documentation), 6 for stents and 1 for coronary bypass surgery (all with no

Heart Failure/Patient Perception Versus Medical Record Entry Table 2 Agreement for cardiovascular and noncardiovascular conditions Co-Morbidity Cardiovascular Arrhythmia Dyslipidemia* Hypertension Stent Myocardial infarction Coronary bypass surgery Debrillator/pacemaker Noncardiovascular Peptic ulcer disease Peripheral arterial disease Lung disease Osteoarthritis Liver disease Thyroid Cancer Diabetes mellitus HIV infection Behavioral Excess alcohol use Cocaine abuse Tobacco Number 191 223 224 221 196 225 221 218 230 203 215 225 214 229 228 182 218 224 203 Yes/Yes 103 94 127 15 50 39 140 1 4 29 10 3 26 34 69 1 3 11 64 No/No 35 71 49 185 123 183 77 198 176 113 174 212 174 185 144 181 200 198 108 Yes/No 36 34 26 21 16 3 3 4 0 39 23 2 7 8 6 0 3 14 31 No/Yes 17 24 22 0 7 0 1 15 50 22 8 8 7 2 9 0 12 1 0 Crude Agreement 72% 74% 79% 90% 88% 99% 98% 91% 78% 70% 86% 96% 93% 96% 93% 100% 93% 93% 85% Positive Agreement 80% 76% 84% 59% 81% 97% 99% 10% 14% 49% 39% 38% 79% 87% 90% 100% 29% 59% 81% Negative Agreement 57% 71% 67% 95% 91% 99% 98% 95% 88% 79% 92% 98% 96% 97% 95% 100% 96% 96% 87%

571

0.37 0.48 0.51 0.54 0.73 0.95 0.96 0.06 0.11 0.28 0.32 0.36 0.75 0.85 0.85 1.00 0.26 0.56 0.69

Kappa p values 0.001 for all conditions except peripheral arterial disease (p 0.006) and peptic ulcer disease (p 0.27). All yes and no citations are by patient report rst, followed by EHR documentation. * Use of lipid-lowering medications or fullling the National Cholesterol Education Program Adult Treatment Panel III criteria.

ditions. For noncardiovascular conditions, 28% agreed for all 9 conditions, and 37% agreed for 7 conditions. Cumulatively, 39% of the pairs matched for 15 of 19 co-morbidities. For cardiovascular conditions, positive agreement ranged from 59% (stent placement) to 99% (implantable cardioverter-debrillator and/or pacemaker implantation). Negative agreement ranged from 57% (arrhythmias) to 99% (coronary bypass surgery). For noncardiovascular conditions, positive agreement ranged from 9.5% (peptic ulcer disease) to 100% (HIV infection), whereas negative agreement ranged from 79% (pulmonary disease) to 100% (HIV infection). Positive agreement for habits ranged from 29% for alcohol use to 81% for tobacco use, and negative agreement ranged from 87% for tobacco to 96% for alcohol and cocaine use. Negative agreement of 80% was more frequent than positive agreement (15 of 19 [79%] vs 8 of 19 [42%] conditions, respectively, p 0.02). Approximately 40% of pairs were discordant for 4 conditions. Discussion
Figure 1. Summary measures of patient versus EHR concordance for co-morbidities. Suboptimal proportional concordance was noted for cardiovascular and noncardiovascular co-morbidities and for the behavioral habits assessed.

EHR entries), and 2 for diabetes mellitus (1 had EHR documentation). A summary measure of patient versus EHR agreement is shown in Figure 1. Of the 7 cardiovascular conditions, only 20% of patients had concordant MHSR responses and EHR entries for all conditions, and 40% agreed for 5 con-

In this study, we observed considerable variability in patient report versus EHR entry of a range of medical conditions and habits, including many conditions for which optimization of care and outcomes requires participation on behalf of patients. Agreement was expectedly better for conditions involving interventions (e.g., debrillator implantation, coronary bypass surgery). We found better negative agreement between MHSR and EHRs (i.e., when the condition was absent) than positive (i.e., when the condition was present). These results provide insights into an understudied area of health care delivery that may inuence outcomes. The accurate capture of patient information by

572

The American Journal of Cardiology (www.ajconline.org) 1. Lloyd-Jones D, Adams R, Brown T, Carnethon M, Dai S, De Simone G, Ferguson T, Ford E, Furie K, Gillespie C, Go A, Greenlund K, Haase N, Hailpern S, Ho P, Howard V, Kissela B, Kittner S, Lackland D, Lisabeth L, Marelli A, McDermott M, Meigs J, Mozaffarian D, Mussolino M, Nichol G, Roger V, Rosamond W, Sacco R, Sorlie P, Stafford R, Thom T, Wasserthiel-Smoller S, Wong N, Wylie-Rosett J. Heart disease and stroke statistics2010 update: a report from the American Heart Association. Circulation 2010;121:e46 e215. 2. Liu L. Changes in cardiovascular hospitalization and comorbidity of heart failure in the United States: ndings from the National Hospital Discharge Surveys 1980 2006. Int J Cardiol In press. 3. Epstein R, Alper B, Quill T. Communicating evidence for participatory decision making. JAMA 2004;291:2359 2366. 4. Baker D, Asch S, Keesey J, Brown J, Chan K, Joyce G, Keeler E. Differences in education, knowledge, self-management activities, and health outcomes for patients with heart failure cared for under the chronic disease model: the improving chronic illness care evaluation. J Card Fail 2005;11:405 413. 5. Bayliss EA, Ellis JL, Steiner JF. Subjective assessments of comorbidity correlate with quality of life health outcomes: initial validation of a comorbidity assessment instrument. Health Qual Life Outcomes 2005;3:51. 6. Smith P, Araya-Guerra R, Bublitz C, Parnes B, Dickinson L, Van Vorst R, Westfall J, Pace W. Missing clinical information during primary care visits. JAMA 2005;293:565571. 7. Cicchetti DV, Feinstein AR. High agreement but low kappa: II. Resolving the paradoxes. J Clin Epidemiol 1990;43:551558. 8. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159 174. 9. OMalley AS, Grossman JM, Cohen GR, Kemper NM, Pham HH. Are electronic medical records helpful for care coordination? Experiences of physician practices. J Gen Intern Med 2010;25:177185. 10. Braunstein JB, Anderson GF, Gerstenblith G, Weller W, Niefeld M, Herbert R, Wu AW. Noncardiac comorbidity increases preventable hospitalizations and mortality among Medicare beneciaries with chronic heart failure. J Am Coll Cardiol 2003;42:1226 1233. 11. Riegel B, Moser DK, Anker SD, Appel LJ, Dunbar SB, Grady KL, Gurvitz MZ, Havranek EP, Lee CS, Lindenfeld J, Peterson PN, Pressler SJ, Schocken DD, Whellan DJ. State of the science: promoting self-care in persons with heart failure: a scientic statement from the American Heart Association. Circulation 2009;120:1141 1163. 12. Velagaleti RS, Massaro J, Vasan RS, Robins SJ, Kannel WB, Levy D. Relations of lipid concentrations to heart failure incidence: the Framingham Heart Study. Circulation 2009;120:23452351. 13. Kostis JB, Davis BR, Cutler J, Grimm RH Jr, Berge KG, Cohen JD, Lacy CR, Perry HM Jr, Blaufox MD, Wassertheil-Smoller S, Black HR, Schron E, Berkson DM, Curb JD, Smith WM, McDonald R, Applegate WB; SHEP Cooperative Research Group. Prevention of heart failure by antihypertensive drug treatment in older persons with isolated systolic hypertension. JAMA 1997;278:212216. 14. Iversen KK, Kjaergaard J, Akkan D, Kober L, Torp-Pedersen C, Hassager C, Vestbo J, Kjoller E. The prognostic importance of lung function in patients admitted with heart failure. Eur J Heart Fail 2010;12:685 691.

EHRs depends on patient awareness and the documentation practices of providers; however, the effectiveness with which patient information is captured is unknown. OMalley et al9 surveyed physicians with EHR experience, chief medical ofcers of EHR vendors, and thought leaders and showed a signicant gap between policy makers expectations and clinicians assessments of EHR as a tool to improve care coordination. A large portion of health care costs and hospitalizations in HF are related not only to worsening HF but also to the high burden of co-morbidities seen in these patients. Braunstein et al10 showed that 39% of patients with HF had 5 noncardiac co-morbidities, and only 4% had none. Importantly, patients with HF with 5 co-morbidities accounted for 81% of total inpatient days. Given the burgeoning cost and poor outcomes for patients with HF, increasing emphasis is being placed on nonpharmacologic care, including self-care, as a mechanism for improving outcomes.11 Patients recognition of their medical conditions is critical to effective self-care. Our study showed that 37% of MHSREHR pairs exhibited less than moderate agreement, underscoring a problem as well as an opportunity for improving care. Many conditions had agreement of 80%. There could be multiple explanations for these results (e.g., patients may not fully understand the terminology or the signicance of a disease or meaning of their symptoms). Alternatively, providers may not be documenting or asking patients pertinent questions. We also observed a high discordance for history of lipid abnormalities, peripheral arterial disease, and hypertension; these co-morbidities commonly accompany HF, and patient participation is important for optimal treatment.12,13 Similarly, coexisting pulmonary disease may exacerbate or be confused with HF symptoms and affects HF prognosis and treatment options.14 Many instances of cocaine and tobacco use were not mentioned in the medical records. The inability to identify these behaviors naturally leads to inadequate patient counseling on the importance of cessation. Also, a signicant segment of patients denied alcohol use that was nevertheless documented in their EHRs. Interestingly, we noted better negative (absence of disease) as opposed to positive (presence of disease) agreement. Whether this is related to perceptual challenges on behalf of patients, lack of documentation by providers, incomplete co-morbidity classication, or the variable prevalence of different disease states needs further study. This study was limited by its size and by the fact that data were collected at a single academic medical center.

Effectiveness of Serial Increases in Amino-Terminal ProB-Type Natriuretic Peptide Levels to Indicate the Need for Mechanical Circulatory Support in Children With Acute Decompensated Heart Failure
Derek T.H. Wong, MDa, Kristen George, MScNb, Judith Wilson, MScNb, Cedric Manlhiot, BScb, Brian W. McCrindle, MDb, Khosrow Adeli, MD, PhDb, and Paul F. Kantor, MBBChb,*
We sought to determine prospectively whether serial assessment of the natriuretic peptide prohormone, amino-terminal proB-type natriuretic peptide (NTpro-BNP), correlated with clinical severity and outcomes in children hospitalized for acute decompensated heart failure (ADHF). Patients (>1 month of age) admitted from 2005 to 2007 with ADHF requiring intravenous vasoactive/diuretic therapy for ADHF were eligible. Serum NTpro-BNP levels were obtained within 24 hours of admission and at prespecied intervals, and clinical caregivers were blinded to these levels. End points included hospital discharge, death or cardiac transplantation, and care escalation including the need for mechanical circulatory support (MCS) was noted. Twenty-four patients were enrolled: 22 survived to hospital discharge and 2 died. Ten required MCS (of which 6 underwent cardiac transplantation). Two patients underwent transplantation without MCS. For the entire cohort, NTpro-BNP levels peaked at days 2 to 3 after admission, with a subsequent gradual decrease until discharge. However, for those who did require MCS, NTpro-BNP failed to decrease consistently until after MCS initiation. At discharge, NTpro-BNP levels were signicantly decreased from admission levels but remained well above normal for all patients. Single-point NTpro-BNP levels on admission did not correlate with independently assessed clinical scores of heart failure severity or predict the need for MCS in this cohort. In conclusion, serial NTpro-BNP levels demonstrated an incremental trend after 48 hours in patients who went on to require MCS but decreased in all other patients and may therefore assist the decision to initiate or avoid MCS after admission for pediatric ADHF. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:573578) In children, acute decompensated heart failure (ADHF) due to cardiomyopathy or failed congenital heart disease repair constitutes a common indication for cardiac transplantation.1 Admission with severe HF frequently raises concerns that a patient has declared the need for transplantation assessment.2 In this regard, recent advances have made use of mechanical circulatory support (MCS) in pediatrics a more popular and feasible method of bridging to transplantation or to recovery.3 However, the appropriate timing of MCS device placement is sometimes difcult to determine. We investigated whether initial assessment of amino-terminal proB-type natriuretic peptide (NTproBNP) in children admitted for management of ADHF correlated with formal assessment of clinical status, and whether serial assessment of NTpro-BNP might be a useful tool to predict the need for MCS in children with ADHF of diverse causes. Methods This was a prospective observational study conducted with the approval of the Hospital for Sick Children (Toronto, Ontario, Canada) research ethics board. Patients admitted to this institution from May 2005 to July 2007 with a diagnosis of ADHF requiring escalation of HF management involving a need for intravenous diuretics, inotropic medication, or MCS were eligible for enrollment. Diagnosis of HF was made by the clinician responsible for admission and not the research team. Two of the investigators (D.W. and P.K.) adjudicated each case for eligibility based on specic criteria before enrollment. Inclusion criteria were (1) a clinical diagnosis of ADHF, (2) age 1 month to 18 years at admission, and (3) escalation of HF therapy with intravenous diuretic therapy, inotropic medication, or MCS. Patients with congenital or acquired heart disease were equally eligible. Exclusion criteria included (1) planned corrective cardiovascular surgery or catheter-based intervention on the same admission (to avoid the effect of surgical or catheter intervention as a confounder of patient outcome) and (2) patients 1 month of age because their
www.ajconline.org

Department of Pediatrics, Division of Pediatric Cardiology, Childrens Hospital of Eastern Ontario, Ottawa, Ontario, Canada; bDepartment of Pediatrics, Labatt Family Heart Centre, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada. Manuscript received January 25, 2010; revised manuscript received and accepted October 11, 2010. *Corresponding author: Tel: 416-813-7239; fax: 416-813-7547. E-mail address: paul.kantor@sickkids.ca (P.F. Kantor). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.015

574

The American Journal of Cardiology (www.ajconline.org)

Figure 1. Flow diagram depicting outcomes for entire patient cohort.

NTpro-BNP levels are known to be increased as part of the normal postnatal circulatory adaptation.4 6 All patients who met the criteria for enrollment were approached and informed consent was obtained. On admission, demographic information, underlying cardiac diagnosis, laboratory investigations (i.e., biochemistry, radiologic studies, electrocardiograms), and echocardiographic data were recorded by the research team. A brief symptomatic history and physical examination were performed at time of admission and at time of discharge by members of the research team (D.W., J.W., and K.G.). A patients clinical course in hospitallength of stay, discharge status and medicationsand need for inotropic support, diuretics, vasoactive medications, antiarrhythmics, and anticoagulation were recorded. We noted whether the outcome of each admission was 1 of the predened end points of discharge, death, or heart transplantation. Other clinically relevant events were also dened and noted prospectively, including CCCU admission, mechanical ventilation, and inotropic and vasoactive medication usage. At time of admission and at time of attaining a dened end point, the previously validated New York University Pediatric Heart Failure Index (NYU-PHFI) score, Ross classication, and where age appropriate the New York Heart Association (NYHA) score were determined.7,8 To isolate the effect of the NTpro-BNP level from clinical severity assessment and the decision to escalate hemodynamic support, clinicians and the research team were blinded to the NTpro-BNP data until the study was completed. NTpro-BNP levels were obtained within 24 hours of admission, at intervals of 2 to 3, 6 to 8, and 13 to 15 days after admission, and weekly thereafter. Patients who were admitted to the critical care unit had NTpro-BNP levels drawn daily. If patients were on stable long-term MCS for 1 week, NTpro-BNP levels were drawn weekly while in the critical care unit. A nal sample was drawn within 48 hours of planned discharge or immediately before heart transplantation. Samples were stored in clotted blood specimen containers at 80C and analyzed after completion of the entire study by our clinical laboratory as a research protocol. Assays were performed on an Elecsys electro-

chemiluminescent immunoassay system using Elecsys 1010/2010 immunoassay analyzers (Roche Diagnostics, Laval, Quebec, Quebec, Canada) with appropriate calibration and control methods.9 Because neither NTpro-BNP nor BNP levels were routinely measured in our institution at the time of this study, no patient received any clinically indicated BNP or NTpro-BNP assay during admission. Investigators were also blinded to results of the NTpro-BNP assays until after the clinical component of the study was completed. Data are presented as mean SD, medians with minimum and maximum values, and frequencies as appropriate. NTpro-BNP levels over time are represented as box-plots to express the severely skewed level distribution. To account for the skewed distribution of NTpro-BNP, a natural logarithmic transformation was applied to NTpro-BNP in all analyses. Basic comparisons between end point groups (MCS vs no MCS) were obtained through Fishers exact test, Students t test with Satterthwaite correction, and Kruskal-Wallis analysis of variance for continuous variables with skewed distribution. Factors associated with need for MCS were sought in univariable logistic regression models using need for MCS before discharge as a binary variable. No multivariable modeling was attempted due to the limited number of patients enrolled in the study. Changes in NTpro-BNP over time and associated factors were assessed in linear regression models adjusted for repeated measurements over time to an autoregressive (rst order with model-based estimation of covariance structure). Regression parameters were estimated using generalized estimating equations. Mean values were used for imputation of missing variables when necessary. Time since admission in days was treated as a continuous variable. An interaction criterion was created between study groups (MCS vs not) and time to estimate differences in rate of change in NT pro-BNP for each group. Because of the small number of patients who died or required heart transplantation we were not able to search for predictors of negative outcomes (even as a combined end point). All statistical analyses were performed using SAS 9.1 (SAS Institute, Cary, North Carolina).

Heart Failure/Serial NTpro-BNP in Pediatric Heart Failure Table 1 Clinical and laboratory data at presentation and serial median B-type natriuretic peptide levels in patients who did or did not require mechanical circulatory support at prespecied intervals Patient Characteristics New York University Pediatric Heart Failure Index Ross/New York Heart Association classication Lactate (mmol/L) pH Creatinine (mol/L) Left ventricular ejection fraction (%) Amino-terminal proB-type natriuretic peptide (pg/ml), median (range) On admission 23 days 68 days 1315 days At discharge/exit Maximum MCS (n 10) 13 (124) 3.5 (24) 7.8 (3.312.6) 7.25 (7.037.41) 89 (31150) 17 (941) No MCS (n 14) 13 (921) 3 (14) 2.0 (1.25.8) 7.41 (7.287.46) 46 (483) 30 (964)

575

p Value NS 0.06 0.02 0.01 0.02 0.05

30,406 (11,40063,862) 39,653 (20,79387,899) 26,510 (3,52939,252) 16,245 (3,20443,740) 3,984 (76648,397) 40,713 (25,20487,899)

10,743 (3,76062,221) 9,262 (3,09470,580) 8,058 (56863,847) 8,594 (1,07032,775) 8,933 (3,61238,270) 15,680 (7,11170,580)

0.03 0.01 0.03 0.12 0.56 0.001

Note that patients requiring mechanical circulatory support are not segregated into before or after mechanical circulatory support in this tabulation.

Results The clinical pathway overview for the cohort is diagrammed in Figure 1. In all, 24 consecutive patients consented to enrollment. Median age (range) was 8.7 years (0 to 17.7) and median weight was 29 kg (3 to 80). Cardiomyopathy was the underlying cause of HF in 17 cases, including idiopathic (8), myocarditis (2), ischemic (1), arrhythmiainduced (2), and unclassied (4) phenotypes. Structural congenital heart disease was present in 4 patients, with 3 other patients including 1 with post-transplant rejection and hemodynamic failure. Eighteen patients were in Ross/NYHA class III or IV at admission, with 5 in class II. One patient (an infant with a mitochondrial cardiomyopathy) appeared clinically asymptomatic to the research team but later deteriorated precipitously, requiring MCS. Most patients showed a change in symptomatic status between admission and the end point, with all those dying or requiring transplantation reaching stage NHYA/Ross class IV before that end point. Median ejection and shortening fractions, as measured by echocardiography on admission for the entire cohort, were 23% (range 6 to 64) and 14% (range 3 to 50), respectively. Only 7 patients were treated with diuretic therapy alone, followed by initiation of oral angiotensin-converting enzyme inhibiter and/or blocker. Most patients in this cohort (17 of 24) required inotropic or inodilator medications17 received milrinone, 9 received epinephrine, 6 received vasopressin, 4 received norepinephrine, and 3 received dobutamine. Eleven patients required a combination of multiple inotropic and/or vasoactive medications. No patients received nesiritide, which was unavailable in Canada at the time of this study. Three patients received digoxin during their hospitalization. Twelve patients (50%) were admitted to the critical care unit, and all of these required mechanical ventilation. Ten patients required MCS, of which 7 were managed with extracorporeal membranous oxygenation and 3 patients required the Berlin Heart EXCOR (Berlin Heart, Berlin, Ger-

Figure 2. Trend of amino-terminal proB-type natriuretic peptide (picograms per milliliter) over time for entire cohort. ADMIT admission; D23 days 2 to 3 of admission; D710 days 7 to 10 of admission; D1315 days 13 to 15 of admission; DC time of discharge from study.

many) ventricular assist device. Differences in clinical presentation characteristics between those patients who went on to require MCS and those who did not are presented in Table 1. Patients who eventually required MCS had a signicantly lower pH and a higher lactate and creatinine on admission than those patients who did not require MCS. In the group requiring MCS median time between admission and initiation of support was 13 days (range 0 to 36). Mean duration on MCS was 26 33 days (median 17, range 2 to 107). Eight patients underwent heart transplantation, 6 of whom underwent bridging by MCS before their heart transplantation. Two patients died1 patient immediately after heart transplantation (after needing MCS for failed single-ventricle palliation) and the other patient having care withdrawn after a large intracranial hemorrhage while on MCS for HF due to dilated cardiomyopathy. Overall survival to hospital discharge was 92% for this cohort, with a transplant-free survival to discharge of 63%.

576

The American Journal of Cardiology (www.ajconline.org)

Figure 3. Plots of amino-terminal proB-type natriuretic peptide (picograms per milliliter) versus clinical scores of heart failure severity.

Figure 4. Trend of amino-terminal proB-type natriuretic peptide (picograms per milliliter) over time for the rst 14 days of admission for patients segregated according to nal outcome of requiring mechanical circulatory support. On admission those who were later to require mechanical circulatory support form a single cohort that then diverges into 2 curves over time as patients are placed onto mechanical circulatory support. Over time, as patients require mechanical circulatory support, their amino-terminal proB-type natriuretic peptide levels no longer contribute to the Need MCSPrior to Support curve and contribute to the Needed MCSOn Support curve as patients move from 1 curve to the other (p 0.01 for the 2 groups before mechanical circulatory support and for those who required mechanical circulatory support comparing levels before and after initiation of mechanical circulatory support).

Serum NTpro-BNP levels were available for all patients. Duration of hospitalization varied signicantly depending on the clinical course of each patient (range 2 to 125 days), and as a result the number of NTpro-BNP measurements per patient also varied. Sixteen patients had measurements that spanned all 5 specied intervals. The trend of NTpro-BNP over time for the entire cohort is displayed in Figure 2. It can be appreciated that the level of NTpro-BNP was 100 to 1,000 times the upper limit of normal (normal range 200 ng/ml). Despite a carefully conducted clinical severity assessment, we found that the

correlation coefcient between HF symptom scores at admission and NTpro-BNP levels was weak (Figure 3). This applied to the NYHA/Ross class status (r 0.15, p 0.52) and the NYU-PHFI (r 0.17, p 0.46). After admission we observed that NTpro-BNP levels continued to increase for the overall cohort initially, peaking at days 2 to 3 of hospitalization, with an overall gradual decrease over time thereafter. Peak level of NTpro-BNP recorded in patients who required MCS during admission was 40,713 pg/ml (range 25,204 to 87,899) compared to 15,680 pg/ml (range 7,111 to 70,580) for those who did not require MCS (p 0.001). Although this difference is signicant, absolute peak level was not found to be predictive of the outcome of MCS, in part because the time at which the peak occurred was not consistent. In regression models adjusted for repeated measurements, serial NTpro-BNP levels decreased in those patients who did not require MCS (953 pg/ml/day), but not for those who were eventually placed on MCS, before the initiation of MCS (601 pg/ml/ day, p 0.04, vs no MCS; Figure 4). For those patients who required MCS, NTpro-BNP decreased only after being placed on MCS, with a signicantly different rate of change over time compared to previously being on MCS (1,354 pg/ml/day, p 0.001, vs before MCS). At time of discharge, levels of NTpro-BNP were signicantly lower in all surviving patients compared to admission. However, these levels were still well above the normal reference range, even in patients with a normalization of ejection fraction. At time of discharge NTpro-BNP levels did not correlate with Ross classication score, NYHA class, or NYU-PHFI score (Figure 3). Due to the small numbers of deaths/cardiac transplantations, it was not possible to determine in this cohort whether NTpro-BNP levels were predictive of death or need for transplantation as a composite end point. Discussion In this study, we investigated whether initial assessment of NTpro-BNP levels correlated well with HF severity at admission and whether serial assessment of NTpro-BNP

Heart Failure/Serial NTpro-BNP in Pediatric Heart Failure

577

levels indicated the important clinical outcome of MCS requirement. Although we were able to demonstrate a marginal difference in the independently administered NYHA/ Ross score (p 0.06) on admission between those who did and those who did not go on to require MCS, the difference in these scores (3.0 vs 3.5) was not very marked and would likely not be clinically discernable. The more detailed NYU-PHFI, which incorporates medical therapies and physical signs and symptoms, was also not discriminatory in this setting for the primary outcome of MCS when administered independently of the clinical team. This nding emphasizes the limitation of subjective and patient-/parentreported functional status (for the Ross/NYHA score) but also indicates that a single-point assessment incorporating several objective parameters may not be very useful in predicting the course of ADHF in children. Serial assessment of NTpro-BNP levels, however, revealed important differences between patients who required MCS and those who did not. We also noted signicant differences in single-point (at admission) assessments of arterial pH, lactate, creatinine and to a lesser extent left ventricular ejection fraction in children who later required MCS. These traditional measurements may have played a signicant role in the decision-making process for MCS by the clinical caregivers, and we cannot comment on their independent predictive value for this outcome given the small sample. Although the levels of NTpro-BNP we encountered were 100 to 1,000 times the established normal ranges and were extremely increased compared to published values of adults with congestive HF,10,11 other pediatric studies (using the same assay method) have shown similar extreme increases in NTpro-BNP.1214 Single time-point natriuretic peptide levels in the outpatient context have shown good prognostic value for later hospitalization, transplantation, or death.15 In contrast to Ratnasamy et al16 and other investigators,6,17 we found that single-point assessment of NT pro-BNP on admission did not correlate closely with independently administered clinical scores of congestive HF severity. This loss of discriminatory power may reect the skewed increase in acuity and severity of congestive HF in our cohort of patients: all required hospitalization for their congestive HF and 50% required mechanical ventilation. Of note, no patients in our study previously had NTpro-BNP levels measured and 75% of patients had not been followed at all in an HF clinic setting. Our data suggest that interpretation of NTpro-BNP levels becomes more complex when admission for ADHF is required. Although maximal attained levels of NTpro-BNP were higher in patients who required MCS during admission than those who did not, absolute levels were not predictive of this outcome in a regression model. We found a statistically signicant difference in rate of change of NTproBNP between those who did and those who did not require MCS. NTpro-BNP levels decreased in those patients who did not require MCS, but not for those who were eventually placed on MCS, before initiating MCS. After being placed on MCS the rate of decrease was faster after being placed on MCS than that noted before being on MCS (p 0.001). The lack of decrease of NTpro-BNP before MCS may be an indicator of failure of medical management to successfully treat the HF. We speculate that the rate of decrease of

NTpro-BNP may be related to the degree of volume unloading of the left ventricle, as has been described by Milting et al18 in adult subjects. We found that for those patients on MCS a more rapid decrease in NTpro-BNP was observed after MCS than before, possibly representing more efcient ventricular decompression. As a result, we conclude that after admission for ADHF, serial assessments of NTpro-BNP, rather than its absolute level, may be a useful indicator of patient risk in pediatric ADHF. This is supported by recent data of Heise et al14 who showed a dramatic and sustained decrease in NTproBNP levels within 1 week of EXCOR implantation. Moreover, Sodian et al19 demonstrated that the rate of decrease of BNP was more rapid in those who were weaned off MCS compared to those who died or required heart transplantation. Due to our small numbers, we were not able to determine whether use of extracorporeal membranous oxygenation versus the EXCOR correlated with a greater or lesser decrease after MCS in NT pro-BNP. Limitations: Severe ADHF in children remains an uncommon scenario beyond the newborn period, and this cohort does not reect the variety of symptom severities seen in a more stable outpatient population already on therapy. Therefore, the ability of NTpro-BNP levels to discriminate between symptom severity groups in a broader setting is not refuted by our data. Also, this study was not powered to assess the predictive value of NTpro-BNP for the outcome of death or transplantation because this end point was too infrequent. Larger patient numbers or the creation of a pediatric ADHF registry may allow for greater statistical power that will allow clinicians to determine the utility of NTpro-BNP in the various phases of pediatric HF management. Our population was predominantly made up of children with cardiomyopathies. Application of these data to children with complex congenital heart lesions and in particular Fontan circulation is uncertain. Acknowledgment: The researchers thank Roche Laboratories for donating the NTpro-BNP assays.
1. Kirk R, Edwards LB, Aurora P, Taylor DO, Christie JD, Dobbels F, Kucheryavaya AY, Rahmel AO, Stehlik J, Hertz MI. Registry of the international society for heart and lung transplantation: twelfth ofcial pediatric heart transplantation report-2009. J Heart Lung Transplant 2009;28:9931006. 2. Kantor PF, Mertens LL. Clinical practice: heart failure in children. Part I: clinical evaluation, diagnostic testing, and initial medical management. Eur J Pediatr 2009;169:269 279. 3. Kirklin JK. Mechanical circulatory support as a bridge to pediatric cardiac transplantation. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2008;80 85. 4. Mir TS, Marohn S, Laer S, Eiselt M, Grollmus O, Weil J. Plasma concentrations of N-terminal pro-brain natriuretic peptide in control children from the neonatal to adolescent period and in children with congestive heart failure. Pediatrics 2002;110(suppl):e76. 5. Koch A, Singer H. Normal values of B type natriuretic peptide in infants, children, and adolescents. Heart 2003;89:875 878. 6. Mir TS, Laux R, Hellwege HH, Liedke B, Heinze C, von Buelow H, Laer S, Weil J. Plasma concentrations of aminoterminal pro atrial natriuretic peptide and aminoterminal pro brain natriuretic peptide in healthy neonates: marked and rapid increase after birth. Pediatrics 2003;112:896 899.

578

The American Journal of Cardiology (www.ajconline.org) 14. Heise G, Lemmer J, Weng Y, Hubler M, Alexi-Meskishvili V, Bottcher W, Hetzer R, Berger F, Stiller B. Biomarker responses during mid-term mechanical cardiac support in children. J Heart Lung Transplant 2008;27:150 157. 15. Price JF, Thomas AK, Grenier M, Eidem BW, OBrian Smith E, Deneld SW, Towbin JA, Dreyer WJ. B-type natriuretic peptide predicts adverse cardiovascular events in pediatric outpatients with chronic left ventricular systolic dysfunction. Circulation 2006;114: 10631069. 16. Ratnasamy C, Kinnamon DD, Lipshultz SE, Rusconi P. Associations between neurohormonal and inammatory activation and heart failure in children. Am Heart J 2008;155:527533. 17. Ohuchi H, Takasugi H, Ohashi H, Okada Y, Yamada O, Ono Y, Yagihara T, Echigo S. Stratication of pediatric heart failure on the basis of neurohormonal and cardiac autonomic nervous activities in patients with congenital heart disease. Circulation 2003;108:2368 2376. 18. Milting H, Ellinghaus P, Seewald M, Cakar H, Bohms B, Kassner A, Korfer R, Klein M, Krahn T, Kruska L, El Banayosy A, Kramer F. Plasma biomarkers of myocardial brosis and remodeling in terminal heart failure patients supported by mechanical circulatory support devices. J Heart Lung Transplant 2008;27:589 596. 19. Sodian R, Loebe M, Schmitt C, Potapov EV, Siniawski H, Muller J, Hausmann H, Zurbruegg HR, Weng Y, Hetzer R. Decreased plasma concentration of brain natriuretic peptide as a potential indicator of cardiac recovery in patients supported by mechanical circulatory assist systems. J Am Coll Cardiol 2001;38:19421949.

7. Ross RD, Daniels SR, Schwartz DC, Hannon DW, Shukla R, Kaplan S. Plasma norepinephrine levels in infants and children with congestive heart failure. Am J Cardiol 1987;59:911914. 8. Connolly D, Rutkowski M, Auslender M, Artman M. The New York University Pediatric Heart Failure Index: a new method of quantifying chronic heart failure severity in children. J Pediatr 2001;138: 644 648. 9. Yeo KT, Dumont KE, Brough T. Elecsys NT-ProBNP and BNP assays: are there analytically and clinically relevant differences? J Card Fail 2005;11(suppl):S84 S88. 10. James SK, Lindahl B, Siegbahn A, Stridsberg M, Venge P, Armstrong P, Barnathan ES, Califf R, Topol EJ, Simoons ML, Wallentin L. N-terminal pro-brain natriuretic peptide and other risk markers for the separate prediction of mortality and subsequent myocardial infarction in patients with unstable coronary artery disease: a Global Utilization of Strategies To Open occluded arteries (GUSTO)-IV substudy. Circulation 2003;108:275281. 11. Westerhout CM, Fu Y, Lauer MS, James S, Armstrong PW, Al-Hattab E, Califf RM, Simoons ML, Wallentin L, Boersma E. Short- and long-term risk stratication in acute coronary syndromes: the added value of quantitative ST-segment depression and multiple biomarkers. J Am Coll Cardiol 2006;48:939 947. 12. Cohen S, Springer C, Avital A, Perles Z, Rein AJ, Argaman Z, Nir A. Amino-terminal pro-brain-type natriuretic peptide: heart or lung disease in pediatric respiratory distress? Pediatrics 2005;115:13471350. 13. Fried I, Bar-Oz B, Perles Z, Rein AJ, Zonis Z, Nir A. N-terminal proB-type natriuretic peptide levels in acute versus chronic left ventricular dysfunction. J Pediatr 2006;149:28 31.

Relation of Obesity to Recurrence Rate and Burden of Atrial Fibrillation

Maya Guglin, MD*, Kuldeep Maradia, MD, Ren Chen, MD, MPH, and Anne B. Curtis, MD
Obesity is associated with new-onset atrial brillation (AF). However, the effect of obesity on AF recurrence or burden has not been studied. The aim of this study was to investigate the relation between AF recurrence, AF burden, and body mass index (BMI). A limitedaccess data set from the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) trial provided by the National Heart, Lung, and Blood Institute was used. Statistical analysis was done with a generalized linear mixed model. In 2,518 patients who had BMIs recorded, higher BMI was associated with a higher number of cardioversions (odds ratio [OR] 1.017, 95% condence interval [CI] 1.005 to 1.029 for a BMI increase of 1 kg/m2; OR 1.088, 95% CI 1.024 to 1.155 for a BMI increase of 5 kg/m2; OR 1.183, 95% CI 1.049 to 1.334 for a BMI increase of 10 kg/m2; p 0.006 for each). Increased BMI was also associated with a higher likelihood of being in AF on follow-up (OR 1.020, 95% CI 1.002 to 1.038 per 1 kg/m2 increased BMI, p 0.0283; OR 1.104, 95% CI 1.011 to 1.205 per 5 kg/m2 increased BMI, p 0.0283; OR 1.218, 95% CI 1.021 to 1.452 per 10 kg/m2 increased BMI, p 0.0283). In a multivariate analysis, left atrial size but not BMI was an independent predictor of AF recurrence and AF burden. Because left atrial size was correlated with BMI, the effect of BMI on AF can be likely explained by greater left atrial size in subjects with higher BMIs. In conclusion, obesity is associated with a higher incidence of recurrence of AF and greater AF burden. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:579 582) Obesity is a risk factor for the development of newonset atrial brillation (AF). Multiple studies have documented a strong and independent association between body mass index (BMI) and the incidence of AF.1 4 In the Framingham Heart Study, obese participants had a 45% to 50% increased risk for incident AF compared to participants with normal BMI, independent of other cardiovascular risk factors.1 In a Danish study, overweight subjects were also at increased risk for incident AF.2 In addition to increasing the susceptibility of developing AF, a recent longitudinal cohort study over 21 years5 suggested that obesity was an independent predictor of progression from paroxysmal to permanent AF. However, the association between obesity and total AF burden or recurrence rate has not been studied. Methods To evaluate the relation of obesity with recurrence of AF or burden of AF, we used a limited access data set from the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) trial, provided by the National Heart, Lung, and Blood Institute (Bethesda, Maryland). Detailed selection criteria for the study population, their baseline characteristics, and randomization into rate-control versus rhythm-control arms was previously explained.6 Our main independent variable was BMI. We used BMI (calculated as weight in kilograms divided by height in meters squared) entered in the data set by the AFFIRM investigators as a surrogate measure of obesity. It was analyzed as a continuous and a categorical variable. Two outcome measures were AF recurrence and AF burden. We used the number of cardioversions done throughout the follow-up period (electrical as well as pharmacological) as a surrogate marker of AF recurrence, and the number of follow-up visits when patients were in AF as a surrogate marker of AF burden. All the data were analyzed with a generalized linear mixed model using SAS (SAS Institute Inc., Cary, North Carolina). A univariate analysis was done rst to identify variables linked to the recurrence rate and total burden of AF. The connection between BMI as our main variable of interest and the 2 outcomes was then examined in detail, for the whole AFFIRM population and for the rate- and rhythm-control arms separately. The following variables were also checked for association with AF recurrence rate or AF burden: age, use of angiotensinogen-converting enzyme inhibitors, use of blockers, systolic blood pressure, history of hypertension, history of coronary artery disease, history of coronary artery bypass surgery, history of congestive heart failure, history of diabetes, history of cardiomyopathy, history of myocardial infarction, New York Heart Association class at baseline, the left ventricular ejection fraction, and left atrial size. Variables found to be signicantly associated with the outcomes were then put in a multivariate model. Because left ventricular ejection fractions were missing in 50% of the cases, we calculated fractional shortening on the basis of left ventricular systolic and diastolic dimensions (fractional
www.ajconline.org

Department of Medicine, University of South Florida, Tampa, Florida. Manuscript received August 17, 2010; revised manuscript received and accepted October 7, 2010. *Corresponding author: Tel: 813-259-0992; fax: 813-259-0665. E-mail address: mguglin@health.usf.edu (M. Guglin). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.018

580

The American Journal of Cardiology (www.ajconline.org)

Table 1 Variables linked to number of cardioversions and burden of atrial brillation in a univariate analysis Variable OR BMI Age Hypertension* Left atrial size * By history. Table 2 Effect of body mass index as a continuous variable on number of cardioversions BMI Increase (kg/m2) 1 5 10 Study Arm OR Rate control Rhythm control Rate control Rhythm control Rate control Rhythm control 1.023 1.015 1.121 1.079 1.256 1.164 Cardioversions 95% CI 0.991.058 1.0031.028 0.9491.323 1.0131.148 0.9011.751 1.0271.319 p Value 0.1788 0.0178 0.1788 0.0178 0.1788 0.0178 1.016 0.99 1.236 1.345 Number of Cardioversions 95% CI 1.0041.029 0.9810.999 1.0641.50 1.1581.561 p Value 0.0086 0.0244 0.0075 0.0001 OR 1.01 1.02 0.99 1.32 Number of Follow-Up Visits in AF 95% CI 0.991.02 0.821.26 0.981.00 1.141.54 p Value 0.4233 0.857 0.1403 0.0003

shortening [left ventricular diastolic dimension left ventricular systolic dimension]/left ventricular diastolic dimension) and used it for the nal analysis. A p value of 0.05 was considered statistically signicant. Results In the AFFIRM study, 4,060 patients were enrolled at baseline. We excluded 1,542 patients who did not have baseline BMI information. Of the remaining 2,518 patients, 1,255 were assigned to the rate-control arm and 1,263 to the rhythm-control arm; the mean BMIs were 29.0 and 28.8 kg/m2, respectively. These 2,518 patients had 22,753 follow-up visits and a total of 1,094 cardioversions, either pharmacologic or electrical: 888 in the rhythm-control arm and 206 in the rate-control arm. In a univariate analysis, BMI, left atrial size, age, and history of hypertension were independently associated with a higher AF recurrence rate (Table 1). In the study population as a whole (n 2,518), higher BMI was associated with a greater number of cardioversions. The odds ratios (OR) of receiving cardioversion were 1.017 (95% condence interval [CI] 1.005 to 1.029, p 0.006) for a BMI increase of 1 kg/m2, 1.088 (95% CI 1.024 to 1.155, p 0.006) for a BMI increases of 5 kg/m2, and 1.183 (95% CI 1.049 to 1.334, p 0.006) for a BMI increase of 10 kg/m2. In the rhythm-control arm (n 1,263), in which a higher rate of cardioversions was expected, the ORs for cardioversion were 1.015 (95% CI 1.003 to 1.028), 1.079 (95% CI 1.013 to 1.148), and 1.164 (95% CI 1.027 to 1.319) for BMI increases of 1, 5, and 10 kg/m2, respectively (p 0.0178 for each). In the rate-control arm, in which there was a lower rate of cardioversions, the association between BMI and number of cardioversions was not signicant (Table 2).

When patients were classied into underweight (BMI 18.5 kg/m2), normal weight (BMI 18.5 to 24.9 kg/m2), overweight (BMI 25 to 29.9 kg/m2), and obese (BMI 30 kg/m2), obese patients were more likely to undergo cardioversion (OR 1.268, p 0.0194; Table 3), with normalweight patients used as a reference. We did not nd a signicant association between obesity and the number of cardioversions in the rate-control arm. However, in the rhythm-control arm, the OR of requiring cardioversion in obese subjects was 1.291 (p 0.0173), with normal weight used as a reference. During each follow-up visit, the current rhythm was recorded as AF versus no AF (presumed sinus rhythm). Of 22,374 follow-up visits, patients were found to be in AF or atrial utter on 8,686 visits: 6,289 in the rate-control group and 2,397 in the rhythm-control arm. Using a linear mixed model, the ORs of a patient being in AF or atrial utter were 1.020 (95% CI 1.002 to 1.038) per 1 kg/m2 BMI increase, 1.104 (95% CI 1.011 to 1.205) per 5 kg/m2 BMI increase, and 1.218 (95% CI 1.021 to 1.452) per 10 kg/m2 BMI increase (p 0.0283 for each). In the rate-control arm, obese (BMI 30 kg/m2) patients had an OR of 1.55 (p 0.0484) of being in AF on a follow-up visit, when normal weight (BMI 18.5 to 24.9 kg/m2) was used as a reference (Table 4). No signicant association between BMI and the likelihood of being in AF was found in the rhythm-control arm. Of the possible confounding variables that could inuence AF recurrence rate and AF burden, a history of hypertension, left ventricular fractional shortening, and left atrial size were found to be signicantly associated with AF recurrence rate (p 0.025, p 0.004, and p 0.001, respectively). After adjusting for age, history of hypertension, fractional shortening, and left atrial size, the latter appeared to be the only determinant of both outcomes in a multivariate analysis. At the same time, BMI was correlated signicantly with left atrial size (Spearmans correlation coefcient 0.22, p 0.0001). Discussion In this analysis of a limited-access data set from the AFFIRM trial, we have demonstrated for the rst time that obesity is associated with a higher recurrence rate and greater burden of AF compared to nonobese patients. Because in the rate-control arm, the strategy was not to restore sinus rhythm, patients spent more time in AF than patients in the rhythm-control arm. The difference between AF burden in obese versus nonobese patients was signicant in the

Arrhythmias and Conduction Disturbances/Obesity and Atrial Fibrillation Table 3 Effect of body mass index as a categorical variable on the number of cardioversions Study Arm Total BMI (kg/m2) 18.5 18.524.9 2529.9 30 18.5 18.524.9 2529.9 30 18.5 18.524.9 2529.9 30 Number of Cardioversions 10 (6.2%) 231 (4.2%) 401 (4.5%) 452 (5.4%) 1 (2.1%) 39 (1.5%) 87 (1.9%) 79 (2.0%) 9 (7.8%) 192 (6.8%) 314 (7.5%) 373 (8.7%) OR 1.518 1.056 1.268 1.511 1.164 1.236 1.176 1.107 1.291 95% CI 0.683.391 Reference 0.8631.291 1.0391.548 0.13816.513 Reference 0.7331.849 0.7731.977 0.5332.594 Reference 0.8921.372 1.0461.593

581

p Value 0.3086 0.5987 0.0194 0.7353 0.5194 0.3755 0.6875 0.3562 0.0173

Rate control

Rhythm control

Table 4 Body mass index effect on atrial brillation burden Study Arm Total BMI (kg/m2) 18.5 18.524.9 2529.9 30 18.5 18.524.9 2529.9 30 18.5 18.524.9 2529.9 30 Number of Visits in AF 30 (19.7%) 1,952 (36.3%) 3,451 (39.8%) 3,253 (39.8%) 12 (31.6%) 1,366 (52.5%) 2,594 (57.0%) 2,317 (58.9%) 18 (15.8%) 586 (21.1%) 857 (20.8%) 936 (22.1%) OR 0.426 1.268 1.333 0.39 1.37 1.55 0.895 1.001 1.156 95% CI 0.1231.473 Reference 0.9441.704 0.9891.797 0.0344.54 Reference 0.8962.096 1.0032.395 0.2673.004 Reference 0.7051.421 0.8161.637 p Value 0.1776 0.115 0.0587 0.4521 0.1461 0.0484* 0.858 0.9965 0.4155

Rate control

Rhythm control

* p 0.05.

rate-control arm as well as in the whole data set, but not in the rhythm-control arm. In contrast, more cardioversions, pharmacologic or electrical, were performed in the rhythmcontrol arm. More cardioversions in obese versus nonobese patients were demonstrated in this arm and in the whole data set, but not in the rate-control arm. Obesity was rst reported as an important, potentially modiable risk factor for new-onset AF by the Framingham investigators. A 4% increase in AF risk per 1 kg/m2 increase in BMI was observed, with adjusted hazard ratios for AF associated with obesity of 1.52 (95% CI 1.09 to 2.13, p 0.02) and 1.46 (95% CI 1.03 to 2.07, p 0.03) for men and women, respectively, compared to subjects with normal BMIs.1 Subsequently, it was shown that the association of obesity with sustained AF is stronger than for transitory or intermittent AF. On average, AF risk is 3% higher per unit increase in BMI. The risk is higher by 7% per BMI unit increase for sustained AF, by 4% for intermittent AF, and by 1% for transitory AF. The obesity-AF association appears to be partially mediated by diabetes mellitus but minimally through other cardiovascular risk factors.3 In the longitudinal cohort study from Olmsted County, Minnesota, BMI independently predicted progression to permanent AF. Compared to normal BMI, obesity (BMI 30 to 34.9 kg/m2) and severe obesity (BMI 35 kg/m2) were associated with increased risk for progression to permanent

AF. This relation was not weakened by left atrial volume, which was independent of and incremental to BMI for the prediction of progression to permanent AF.5 Similarly, in the Swedish Primary Prevention Study, body surface area at age 20 years (calculated from recalled weight and measured height) was strongly related to subsequent AF (p 0.0001), as were midlife BMI and weight gain from age 20 years to midlife (p 0.0001).7 A meta-analysis of 16 studies enrolling a total of 123,249 subjects found that obese subjects have an associated 49% increased risk for developing AF compared to nonobese subjects. In postoperative AF, however, BMI did not appear to play an important role8 and was even associated with a lower incidence of AF.9 In a recently published study by Tedrow et al,4 it was demonstrated for the rst time that the risk for incident AF is especially high in subjects who gained weight rapidly. Even more important, they proved that this risk decreases after normalization of BMI. Obesity therefore appears to be a reversible risk factor for AF. The association between obesity, left atrial size, and AF is well established.10 Obesity is identied as the most important determinant of left atrial enlargement. In our study, left atrial size was independently correlated with BMI. Therefore, obesity may increase the rate of new-onset AF, the recurrence rate, the transition from paroxysmal to permanent AF, and total AF burden, not directly but through

582

The American Journal of Cardiology (www.ajconline.org) 2. Frost L, Hune LJ, Vestergaard P. Overweight and obesity as risk factors for atrial brillation or utter: the Danish Diet, Cancer, and Health Study. Am J Med 2005;118:489 495. 3. Dublin S, French B, Glazer NL, Wiggins KL, Lumley T, Psaty BM, Smith NL, Heckbert SR. Risk of new-onset atrial brillation in relation to body mass index. Arch Intern Med 2006;166:23222328. 4. Tedrow UB, Conen D, Ridker PM, Cook NR, Koplan BA, Manson JE, Buring JE, Albert CM. The long- and short-term impact of elevated body mass index on the risk of new atrial brillation: the WHS (Womens Health Study). J Am Coll Cardiol 2010;55:2319 2327. 5. Tsang TS, Barnes ME, Miyasaka Y, Cha SS, Bailey KR, Verzosa GC, Seward JB, Gersh BJ. Obesity as a risk factor for the progression of paroxysmal to permanent atrial brillation: a longitudinal cohort study of 21 years. Eur Heart J 2008;29:22272233. 6. Wyse DG, Waldo AL, DiMarco JP, Domanski MJ, Rosenberg Y, Schron EB, Kellen JC, Greene HL, Mickel MC, Dalquist JE, Corley SD. A comparison of rate control and rhythm control in patients with atrial brillation. N Engl J Med 2002;347:18251833. 7. Rosengren A, Hauptman PJ, Lappas G, Olsson L, Wilhelmsen L, Swedberg K. Big men and atrial brillation: effects of body size and weight gain on risk of atrial brillation in men. Eur Heart J 2009;30: 11131120. 8. Wanahita N, Messerli FH, Bangalore S, Gami AS, Somers VK, Steinberg JS. Atrial brillation and obesityresults of a meta-analysis. Am Heart J 2008;155:310 315. 9. Banach M, Goch A, Misztal M, Rysz J, Jaszewski R, Goch JH. Predictors of paroxysmal atrial brillation in patients undergoing aortic valve replacement. J Thorac Cardiovasc Surg 2007;134:1569 1576. 10. Stritzke J, Markus MR, Duderstadt S, Lieb W, Luchner A, Doring A, Keil U, Hense HW, Schunkert H, Investigators MK. The aging process of the heart: obesity is the main risk factor for left atrial enlargement during aging the MONICA/KORA (Monitoring of Trends and Determinations in Cardiovascular Disease/Cooperative Research in the Region of Augsburg) study. J Am Coll Cardiol 2009;54:19821989. 11. Ardestani A, Hoffman HJ, Cooper HA. Obesity and outcomes among patients with established atrial brillation. Am J Cardiol 2010;106: 369 373. 12. Badheka AO, Rathod A, Kizilbash MA, Garg N, Mohamad T, Afonso L, Jacob S. Inuence of obesity on outcomes in atrial brillation: yet another obesity paradox. Am J Med 2010;123:646 651.

increased left atrial size. It is noteworthy that in the study from the Framingham cohort establishing the link between obesity and new-onset AF, after adjustment for left atrial diameter, BMI was no longer associated with AF risk.1 The investigators concluded that effect of obesity was mediated by left atrial dilatation. In contrast, left atrial remodeling and enlargement is a well-known effect of AF itself, and obesity contributing to AF recurrence and burden and therefore promoting left atrial dilatation is another possible course of the events. Two other studies used the same data set to analyze effects of obesity in AF, and both concluded that extra weight is associated with lower cardiovascular mortality.11,12 Neither study, however, addressed the issues of recurrence and burden of AF in obese versus nonobese patients. Our ndings are in concurrence with other studies investigating the connection between BMI and AF. Although the mechanism of obesity-related increased risk for AF is unclear, a consistent pattern of increased AF incidence, prevalence, recurrence, and overall burden suggests that lifestyle modications directed toward a healthier weight may reduce AF and all the risks and complications associated with it. We analyzed only the data available from the limitedaccess data set of the AFFIRM trial and did not have access to the complete study data. In addition, this was a retrospective analysis of the main trial; therefore, the results should be interpreted with caution. The number of clinic visits at which patients appeared to be in AF was used as a surrogate of AF burden, and the number of cardioversions as a surrogate of AF recurrence rate. Data for left atrial size were missing in 24% of patients with known BMIs.
1. Wang TJ, Parise H, Levy D, DAgostino RB Sr, Wolf PA, Vasan RS, Benjamin EJ. Obesity and the risk of new-onset atrial brillation. JAMA 2004;292:24712477.

The Editors Roundtable: Implantable Cardioverter-Debrillators in Primary Prevention of Sudden Cardiac Death and Disparity-Related Barriers to Implementation
Vincent E. Friedewald, MDa,*, Gregg C. Fonarow, MDb, Brian Olshansky, MDc, Clyde W. Yancy, MDd, and William C. Roberts, MDe
Acknowledgment This CME activity is supported by an educational grant from Medtronic, Inc., Minneapolis, Minnesota. Disclosures Dr. Friedewald has received honoraria for speaking from Novartis, East Hanover, New Jersey. Dr. Fonarow has received honoraria for speaking and consulting and research grants from Medtronic; and GlaxoSmithKline, Research Triangle Park, North Carolina. Dr. Olshansky has received honoraria for speaking and consulting from Medtronic; and Boston Scientic Corporation, Natick, Massachusetts. Dr. Olshansky has received honoraria for consulting and is a member of the advisory board for Novartis. Dr. Yancy has no relevant nancial relationships to disclose. Dr. Roberts has received honoraria for speaking from Merck, Whitehouse Station, New Jersey; AstraZeneca, Wilmington, Delaware; and Novartis. Objectives Upon completion of the activity, the physician should be able to: 1. Diagnose patients with congestive heart failure who are candidates for implantable cardioverter-debrillator (ICD) therapy. 2. Explain the risks and benets of ICD therapy to patients. 3. Decrease gender and ethnic disparities in treatment with ICD therapy. Target Audience: This activity is designed for cardiologists and all other health care specialists caring for patients with acute and chronic coronary heart disease. CME Credit: The A. Webb Roberts Center for Continuing Medical Education of Baylor Health Care System, Dallas, Texas, designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit. Physicians should only claim credit commensurate with the extent of their participation in the activity. The A. Webb Roberts Center for Continuing Medical Education of Baylor Health Care System, Dallas, Texas, is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. CME Provider Privacy Policy and Contact Information: The A. Webb Roberts Center for Continuing Medical Education of Baylor Health Care System (214-820-2317) observes the privacy and condentiality of CME information and the personal information of CME participants. Third parties receive only aggregated data about CME activities that are relevant to their interests and/or the activities they support. CME Instructions: After reading this article, go online at www.AJConline.org to register, complete a post-test with a minimum score of 80%, complete an evaluation, and print a certicate. Combination of Media: Print and Internet Computer Requirements: Windows 2000, Pentium 3 or greater, 512 ram, 80 gigabytes storage Estimated Time to Complete: 1 hour Release Date: February 2010 Termination Date: February 2011 Introduction The ICD was rst placed into human subjects in 1980 by Mirowski, after several years of nonhuman animal testing.1,2 In the 30 years since the introduction of ICDs, ICD therapy in the United States has become commonplace, with 2 broad categories of use for preventing sudden cardiac death (SCD): primary prevention involves the prevention of SCD in patients without histories of cardiac arrest or sustained ventricular tachycardia, and secondary prevention involves the prevention of SCD in patients who have survived prior cardiac arrest, sustained ventricular tachycardia, or other major cardiac events. This Editors Roundtable focuses on ICD therapy for primary prevention, which mainly involves patients with ischemic and nonischemic
www.ajconline.org

a Associate Editor, The American Journal of Cardiology, Clinical Professor, Department of Internal Medicine, The University of Texas Medical School at Houston, Houston, Texas, and Research Professor, University of Notre Dame, Notre Dame, Indiana; bThe Eliot Corday Professor in Cardiovascular Medicine and Science, UCLA Division of Cardiology, University of California, Los Angeles, Los Angeles, California; cProfessor of Medicine, University of Iowa College of Medicine, Iowa City, Iowa; d Medical Director, Baylor Heart and Vascular Institute of Baylor University Medical Centerand Chief, Cardiothoracic Transplantation, Baylor University Medical Center, Dallas, Texas; and eEditor-in-Chief, The American Journal of Cardiology and Baylor University Medical Center Proceedings, Executive Director, Baylor Heart and Vascular Institute of Baylor University Medical Centerand Dean, A. Webb Roberts Center for Continuing Medical Education of Baylor Health Care System, Dallas, Texas. This discussion took place at Baylor University Medical Center, Dallas, Texas, on October 9, 2008. *Corresponding author: Tel: 574-631-6675; fax: 574-631-4505. E-mail address: vfriedew@nd.edu (V.E. Friedewald).

Am J Cardiol 2011;107:583590 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.003

584

The American Journal of Cardiology (www.ajconline.org)

heart failure who are receiving optimal medical therapy and have a reasonable expectation of survival with good functional status for 1 year.3 Discussion Dr. Friedewald: When did ICD therapy appear? Dr. Olshansky: ICD therapy was developed and rst tested in nonhuman animal models by Michel Mirowski in the 1970s.2 Although the concept of an ICD was not initially well received, Mirowski paved the way for clinical acceptance when he performed the rst human implantation in 1980.1 The rst ICD was large, weighing about 9 oz, with a large battery and a capacitor to shock the heart. It had few settings, its sensing ability was rudimentary, and it lasted only about 18 months. It required major surgery because the device was implanted in the abdomen and the chest had to be opened, requiring at least 1 week of hospitalization. The rst ICD devices were handmade, so the supply was limited, and patients sometimes had to wait for weeks before 1 could be obtained. The technology progressed dramatically over the next 30 years as the devices became multiprogrammable, used smaller batteries with longer battery life, had better capabilities to debrillate with biphasic shocks, were made programmable for cardiac pacing, had better leads that could be placed intravenously, and were implanted into the upper chest, an easy procedure to perform. Today, ICD implantation is a low-risk procedure carried out worldwide. Dr. Yancy: What are the current risks of ICD use? Dr. Olshansky: There are several. Although the focus in the lay press has been placed on device recalls and lead problems, improper working devices are rare, maybe 1 in 10,000 implants. Other complications including myocardial lead perforation, infection, pneumothorax, lead dislodgement, and inappropriate shocks (a shock delivered for a reason other than a life-threatening ventricular tachyarrhythmia). The ICD is designed, however, to protect life at the expense of an occasional inappropriate shock. The risk for inappropriate shock is about 25%. Dr. Yancy: What is the frequency of serious problems with ICDs? Dr. Olshansky: Serious problems such as infection and device failure occur in about 1% of ICD implants. Dr. Yancy: What is the current role of the ICD in preventing SCD? Dr. Olshansky: There has been a signicant movement to ICD use for primary prevention of SCD. At 1 time, before receiving an ICD, patients had to experience 2 separate out-of-hospital cardiac arrests, so few patients used to qualify for an ICD. We no longer require a prior event, only that a patient is likely to have cardiac arrest (Appendix3). Although there is some controversy about the criteria, most ICD implantations are for primary prevention, not secondary prevention. Dr. Fonarow: Much of the growth of ICD utilization is due to the recognition that most antiarrhythmic drugs are ineffective for both primary and secondary prevention and are sometimes proarrhythmic, thereby increasing the risk for SCD. Two decades ago, ecainide and encainide were 2 of the top 10 cardiac medications prescribed, but they subsequently were found to increase all-cause mortality and are

proarrhythmic.4 Other antiarrhythmic drugs, such as amiodarone, also fail to protect against SCD. Dr. Friedewald: What is the relation between left ventricular (LV) dysfunction and SCD? Dr. Fonarow: Patients with signicant LV dysfunction even in the absence of a prior cardiac event, ventricular ectopic beats on ambulatory monitoring, or inducible arrhythmia on electrophysiologic studyare at increased risk for SCD. Because up to 1/2 of deaths in patients with LV dysfunction are sudden, prophylactic ICD placement in this patient population is often indicated. Prospective randomized clinical trials in patients receiving optimal heart failure (HF) treatment with subsequent placement of the ICD demonstrated that they aborted SCD when compared to drug treatment alone, in patients with both ischemic and nonischemic forms of cardiomyopathy. Dr. Friedewald: Do drugs that are not directly antiarrhythmic, but proven benecial in treating patients with HF (i.e., blockers and renin-angiotensin aldosterone inhibitors) reduce the risk of SCD in patients with HF? Dr. Fonarow: Beta blockers reduce death from progressive HF as well as SCD in patients with LV dysfunction. Patients on blockers, however, have a greater relative reduction in death from progressive HF, resulting in increased incidence of SCD in this population. The predominant effect of angiotensin-converting enzyme inhibitors is on death from progressive HF with possibly a slight reduction in the frequency of SCD. Aldosterone antagonists also decrease the risk of death from progressive HF, with possibly a slight reduction in the frequency of SCD. Thus, patients on optimal medical therapy for LV dysfunction and HF have enough residual risk for SCD that usually justies primary ICD placement. Dr. Yancy: Is there a role for antiarrhythmic drugs in patients with an ICD? Dr. Olshansky: There may be a role for the use of antiarrhythmic drugs in addition to ICD therapy in patients who receive multiple ICD shocks for ventricular and atrial tachyarrhythmias. As primary therapy to reduce total mortality or arrhythmic death, however, antiarrhythmic drugs have no role. The important point is that ICD therapy reduces the incidence of both SCD and overall mortality. Dr. Fonarow: It is important to separate absolute risk and proportional risk. The proportion of SCD relative to death from progressive left ventricular dysfunction is higher in patients with less severe HF symptomsNew York Heart Association class I or II compared to patients in class III or IV HF, in which a greater proportion of deaths are due to progressive HF. Thus, although SCD occurs in patients in class III and IV HF, the absolute risk for deaths both from progressive HF death as well as SCD rises with increasing severity of HF. There is little benet in preventing SCD in a patient who shortly thereafter dies from progressive HF. Thus, identifying patients who derive the greatest absolute benet from the therapy and in whom the benet outweighs the potential ICD risks is essential. In patients with class I to III HF treated with an ICD, the benet outweighs the risk and prolongs survival. In class IV patients, however, because of the ICD impairment on quality of life and functional status, HF not amenable to optimal medical therapy precludes ICD therapy. Class II and III HF

Roundtable Discussion/ICDs in Primary Prevention of SCD

585

patients frequently die suddenly. Many of them also die from progressive HF, so they need protection from both major modes of death. Dr. Yancy: Thus, patients with class IV HF may not be helped by ICD implantment. Dr. Fonarow: That is true, except in class IV HF patients, who improve to a better functional class with other forms of treatment. Thus, class IV patients who are not candidates for cardiac resynchronization therapy, cardiac transplantation, or ventricular assist devices, and who remain persistently in class IV are not candidates for ICD under current guidelines because there is insufcient evidence that they benet. Ambulatory class IV HF patients who receive cardiac resynchronization therapy combined with ICD, however, may have improved survival compared to patients on optimal medical treatment alone. Dr. Yancy: Is SCD risk related to functional capacity in patients with HF? Dr. Olshansky: Patients in functional class I and II HF are less likely to die, and when they die it is more often due to an arrhythmia. They are the patients that derive the greatest potential benet from an ICD. Functional class III patients also have a high risk of arrhythmic death, but they also have greater risk of mortality from progressive HF than functional class II patients. Thus, the functional class III patient can benet from an ICD. Functional class IV patients may not benet from ICD therapy. Dr. Roberts: Are you saying that the greater the LV dilatation, the greater the chance that death will be caused by pump failure, and the less the LV dilatation, the greater the chance that death will be due to an arrhythmia? Dr. Olshansky: It is probably true that the larger the LV cavity, the greater chance of death from LV failure. Many LV parameters have been studied, however, and only 2 are clinically useful: (1) New York Heart Association functional class and (2) LV ejection fraction. Although they are not perfect, they are the best we have now. Dr. Roberts: The LV ejection fraction is proportional to LV cavity size? Dr. Olshansky: The possibility that LV end-diastolic or LV end-systolic volume is an independent predictor has not been studied. Thus, we use LV ejection fraction. Dr. Roberts: The same is true for angina pectoris. With increasing LV cavity dilatation, there is less frequency of angina, and patients with grade 4 angina pectoris generally have normal LV cavity size. Dr. Olshansky: One of the most challenging issues is risk assessment. Although our discussion is focused on primary prevention in patients with HF, there are many different patient types at increased risk for SCD whose LV function is well preserved. Patients with LV diastolic dysfunction alone, which is difcult to dene, also are at increased risk of SCD. There also are patients at increased risk of SCD with certain genetic abnormalities, such as the Brugada syndrome, the longQT interval syndrome, and hypertrophic cardiomyopathy, who do not have HF and have normal LV ejection fraction. For example, consider a 17-year-old girl with syncope and a QT interval of 560 ms on her electrocardiogram. Compare that patient to an 85-year-old man with an LV ejection fraction of 10% and functional class II HF on optimal medical therapy, who has a much greater risk of SCD, and a much greater chance of an

ICD stimulus than the 17-year-old girl. Thus, clinical characteristics other than LV ejection fraction and functional class must be considered. The younger individual might use the debrillator less, but her dying from cardiac disease almost surely would be from a cardiac arrhythmia. In the older individual, however, progressive HF and co-morbidities associated with age (renal dysfunction, atrial arrhythmias, and diabetes mellitus) are additional independent risks for mortality. Another group to consider for ICD therapy is comprised of patients with ischemic cardiomyopathy and nonsustained ventricular tachycardia (VT) who have relatively well-preserved LV ejection fraction and are in New York Heart Association functional class I. They also have sufcient risk of VT and ventricular brillation that may warrant ICD placement. Dr. Roberts: Let us assume that a 60-year-old man develops substernal chest pain, goes to the hospital emergency department, and 3 hours after the onset of chest pain has cardiac arrest. Is that a form of SCD? Dr. Olshansky: By some denitions, yes. Dr. Roberts: By your denition? Dr. Olshansky: Probably not. Dr. Roberts: The World Health Organization uses death within 24 hours after a change in health status in their criteria for SCD, which is ridiculous! If a 24-hour denition from sudden change of previous health to an event is used, a lot of patients with acute myocardial infarction are included, and I believe they should not be included in the category of SCD. Dr. Fonarow: The denition of SCD is important. There are, however, many denitions of SCD, including a stringent 15- or 30-minute denition or when it is completely unexpected and not explained by something else such as acute myocardial infarction. At one time, it was thought that many episodes of SCD in HF patients were not caused by tachyarrhythmias, rather by other mechanisms such as bradyarrhythmias, which would not be prevented by an ICD. We now know that a substantial proportion of out-of-hospital SCD events are caused by tachyarrhythmias and can be effectively detected and aborted with the ICD. The number of deaths that are bradyarrhythmic or due to other mechanisms such as myocardial infarction or are hyperkalemia mediated and cannot be aborted by the ICD constitute only a small proportion of SCD in patients with signicant LV dysfunction after myocardial infarction or ambulatory HF. Dr. Roberts: I use a 6-hour denition for SCD because within 6 hours after onset of chest pain, there is no histological evidence of acute myocardial infarction. Thus, patients with occluding thrombus are not included. Dr. Olshansky: That is reasonable, but there is a lot we do not understand about SCD. Many cases of SCD, for example, are due to asystole rather than a ventricular arrhythmia. Dr. Friedewald: According to current guidelines, who should receive an ICD? Dr. Fonarow: The newest guidelines from the American College of Cardiology, American Heart Association, and Heart Rhythm Society are based on randomized clinical trials That establish ICD efcacy.3 For primary prevention, placement of an ICD is indicated for (1) patients with LV ejection fraction 35% who are at least 40 days after onset of the infarction with New York Heart Association class II or III HF; (2) patients with nonischemic dilated cardiomy-

586

The American Journal of Cardiology (www.ajconline.org)

opathy and LV ejection fraction 35% who are functional class II or III, do not have contraindications for an ICD, and have a reasonable expectation for good functional status for at least 1 year; and (3) patients at least 40 days post myocardial infarction who do not have class II or III HF but have class I failure with LV ejection fraction 30%. These categories comprise a large number of eligible patients who do not have any prior arrhythmias but are candidates for ICD by virtue of their clinical state, symptom class, and degree of LV dysfunction determined by LV ejection fraction. Dr. Friedewald: What is the basis for waiting 40 days after an acute myocardial infarction? Dr. Fonarow: Placement of an ICD soon after acute myocardial infarction has not shown a net benet because some postmyocardial infarction patients have substantial improvement in LV function with reperfusion therapy and optimal medical treatment.4 Dr. Yancy: The guidelines strongly advise that patients be on reasonable medical therapy. What is reasonable, and for how long should medical treatment be employed before ICD placement is considered? Dr. Fonarow: Many patients with nonischemic cardiomyopathy have spontaneous improvement in LV function, even when untreated. Thus, they need a chance for recovery before receiving an ICD. Prior guidelines, which have been controversial, left this time frame up to the clinician. In most patients we can start blockers and angiotensinconverting enzyme inhibitors at the same time in patients with HF and up-titrate them within 2 to 3 months to establish optimal medical therapy and then reassess LV function. Dr. Yancy: It would seem difcult to withhold an ICD from a patient with signicant LV dysfunction at risk for SCD for such a relatively long period of time. Dr. Olshansky: Yes, especially in the common scenario of patients who are not receiving optimal medical therapy for nonischemic and ischemic cardiomyopathy, as they are at risk for SCD. Other patients of special concern are those with a recent myocardial infarction who have poor LV function, are being discharged from the hospital, and face a period of time before they are seen again. That time period encompasses an increased risk of SCD. Because patients with poor LV ejection fractions or impaired LV function have a high risk of nonarrhythmic death early, it is not clear that the ICD offsets that risk. The concern whether we are doing enough for those patients, however, remains. Some clinicians use bridging approaches, such as life vests and the automatic external debrillator, but the efcacy data for such devices are poor. Dr. Roberts: What percent of patients with HF die in the hospital? Dr. Yancy: With acute cardiac decompensation, the average mortality is about 4%, but the range is 2% to 20%. Dr. Roberts: When a patient with severe HF dies suddenly, do you attribute the death to sudden ventricular brillation or asystole, rather than to pump failure? Dr. Yancy: This is a dilemma of clinical trials. When we try to adjudicate such deaths, we use arbitrary denitions. How do you set the time clock, and what does that mean? SCD is an unexpected or unanticipated event at some point after a patient is deemed stable and comfortable. The denition of sudden is a problem for many clinical trials.

Dr. Roberts: In-hospital death should be separated from out-of-hospital death. Dr. Olshansky: I agree. There are some in-hospital sudden deaths that occur while patients are on the ICD, and these are reported as SCD. The mechanism can be either a ventricular arrhythmia or pulseless electrical activity. Dr. Roberts: But patients receiving ICD therapy also die of HF. Dr. Friedewald: Is patient proximity to medical care considered among the indications for ICD implantation? Dr. Fonarow: For individuals having out-of-hospital SCD, even those who next door to the hospital, the chance of survival with intact neurological function is incredibly low, about 7%. Thus, geography is not a factor for ICD placement. Every second without adequate debrillation makes survival after cardiac arrest less likely. Dr. Friedewald: Does a low likelihood of patient adherence to cardiac medications affect ICD usage? Dr. Fonarow: Patient adherence to HF medications is not a factor in the decision whether to implant an ICD. Patients with ICDs need to take medications. The device is not a replacement for medications. Patients with ICDs also require optimal medical treatment. The device does nothing to prevent progression of underlying LV failure. It is critically important to continue optimal medical treatment along with the ICD. Dr. Yancy: What is the experience with the ICD in clinical settings in the context of guideline indications? Dr. Fonarow: Among all patients hospitalized for HF in the American Heart Association Get With the Guidelines Heart Failure program, only about 25% of eligible patients receive an ICD. Some eligible patients, however, refuse the ICD, and some do not meet additional criteria such as HF time of onset and the presence of co-morbid conditions. When these patients are excluded, about 33% of eligible patients are treated with an ICD. In the outpatient setting utilizing a registry like Improve the Use of Evidence-Based Heart Failure Therapies in the Outpatient Setting5 (IMPROVE-HF), the utilization rate is higher. About 50% of HF patients who are candidates under the guidelines receive ICDs. This means, however, that 50% of HF eligible outpatients in cardiology practices do not receive ICDs. In primary care settings the rate of ICD placement is even lower. Thus, despite excellent clinical trial evidence of their efcacy, a substantial proportion of patients who are eligible for ICD treatment do not receive it. Dr Friedewald: Why is ICD therapy so underutilized? Dr. Fonarow: It is very difcult to understand why evidence-based therapies, using either medications or devices, are not better adhered to, especially in view of all the information disseminated about guidelines. One factor with many other analogous therapies is a substantial lag from the time the therapy becomes evidence-based and recommended in the guidelines to when it is widely adopted. ICD treatment is relatively new, so the customary time lag is a factor in its underutilization. Another factor is that this treatment requires much more than a simple written prescription. ICD placement involves referral to an electrophysiologist, an invasive procedure, and close follow-up. Thus, some patients, even when fully informed about the benets of ICD, decline implantation.

Roundtable Discussion/ICDs in Primary Prevention of SCD

587

Dr. Olshansky: I am surprised that the ICD utilization is as high as 50% in outpatients. It has been my experience not supported by any published datathat many physicians either do not know the data supporting ICD use, do not believe the data, or do not want to believe the data, so they simply do not even think about ICD therapy. There also have been negative stories published in the public media. Here is a quote from a major newspaper, which is representative of some of the publics perception: The number of patients receiving ICDs has actually declined as more doctors and patients decide the risks and uncertainties the device has posed may outweigh the potential benets. Industry estimates and medical studies show that ICDs have saved the lives of only 10% of the 600,000 people who will receive them at most. Nine of 10 people who receive ICDs receive no medical benet. This is a misunderstanding of the science. When the number of persons needed to treat to get a benet of saving 1 life is only 10 people, that is a tremendously effective therapy. We do not expect that every patient who receives an ICD will have his or her life saved by the ICD. We only expect risk reduction. Dr. Yancy: We have excellent data demonstrating utility of the ICD in the appropriate patient with reduction in total mortality and sudden death. Some patients who receive an ICD never have a ring and when they do, it is an inappropriate shock. Others are concerned about costs and related risks. What do you think is required: more data? More time? Different perspectives? How do we optimize use of the ICD? Dr. Olshansky: The issue comes down to the risk/benet ratio and what is expected in The extension of reasonably functional life, which is difcult to dene. If the life of an 85-year-old person is extended by a few months, is that of value to society or only to the individual? Is it worth the cost? Dr. Yancy: Is this an ethical argument or a scientic argument? Dr. Roberts: It is both. Are we going to have rationing of medical care? We are talking about expensive devices in a country that is broke. Dr. Friedewald: What is the average increase in longevity in the group with the ICD? Dr. Olshansky: An exact number has not been calculated, at least in the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT).6 In that trial, which is probably our best database of patients who have functional class II and III HF on the best medical therapy, there is a 23% relative risk reduction with a mean follow-up of about 45 months. Dr. Roberts: What is the absolute risk reduction? Dr. Fonarow: The absolute risk reduction over the 4 years of the study is 7%, so 70 lives would have been lost for each 1,000 patients not receiving an ICD. The mean age was 65 years, and they were primarily functional class II to III. Dr. Roberts: The study patients had an LV ejection fraction of about 35%. What do we know about patients who are 40 days post myocardial infarction with greater LV ejection fractions? Dr. Fonarow: An interesting paradox is that patients with better-preserved LV ejection fractions derive less benet. It is the group of patients with LV ejection fraction 35% and especially 30% that derive much greater benet. Thus, a cutoff of the LV ejection fraction of about 35% is reasonable, because among patients with an LV ejection

fraction 35%, the number needed to treat to show a benet is substantially higher. Dr. Roberts: What is the range of error for the measurement of LV ejection fraction? Dr. Fonarow: For most studies, the range is 5%. Dr. Yancy: Do you believe there is a greater physician bias against ICD placement than patient bias? Dr. Olshansky: There is a signicant physician bias, but I cannot quantitate it. Maybe we overemphasize the potential benets because some trials do not show as much benet as others. There are other post hoc analytic data about specic patient populations suggesting that certain types of patients do not achieve the same benet as others. Patients with atrial brillation, patients with functional class III HF, and patients with renal dysfunction may derive less benet. Older patients also may not have as much improvement of functional quality of life with the ICD as younger persons. Dr. Fonarow: Such data can be misleading because trials are powered to look at the overall population. We can be misguided when we retrospectively study underpowered subgroups that appear to have less benet with the ICD. Dr. Yancy: There are data showing no statistical difference in outcomes with ICD therapy as a function of gender or race. There are, however, gender and racial differences among patients who receive ICD therapy, given the same indications. How does this apply to the American Heart Associations Get With the GuidelinesHeart Failure? Dr. Fonarow: Get With the GuidelinesHeart Failure is a hospital-based registry and performance improvement program from the American Heart Association. There are currently 424 US hospitals from all regions of the USA participating. The registry is comprised of 200,000 patients and includes data on demographics, treatment, LV function, previous device placement, hospitalization, and primary discharge diagnosis of HF or HF as the predominant reason for hospitalization. A study by Hernandez and colleagues7 looked at the use of the ICD in appropriate patients to see if there were disparities on the basis of patient age, gender, or race. One third of eligible patients received an ICD or planned ICD placement post discharge. There were substantial disparities by race and gender: 44% of eligible white men and 33% of eligible black men received an ICD; 30% of white women and 28% of eligible black women received an ICD. Adjustment for co-morbidities, insurance coverage, and other factors did not change these ndings. Dr. Yancy: The same disparities have been found in the use of cardiac resynchronization therapy. Dr. Fonarow: Many published reports have focused on disparities of other cardiovascular procedures, surgery, and device implants. Among newer or more expensive therapies, a certain proportion of patients are less likely to be treated, and race and gender seem to be signicant inuences. Dr. Olshansky: Do such disparities also occur with medical treatment? Dr. Yancy: There is variance in the use of blockers after acute myocardial infarction. For other evidence-based treatments, there is no demonstrable evidence of disparity in the medical care of inpatients. For activities such as counseling for smoking cessation or appropriate discharge or-

588

The American Journal of Cardiology (www.ajconline.org)

ders, disparities emerge. In mammography for breast cancer screening, revascularization compared to amputation for peripheral vascular disease, hemodialysis compared to renal transplantation, and thrombolytic therapy compared to acute percutaneous coronary intervention, there is clear evidence of disparate health care. Dr. Olshansky: Some disparities are patient related and others are due to physician ignorance or physician bias. I wonder if they depend on the physician-patient relationship. There are differences, for example, in the care of white women who see white female physicians, white male physicians, African American male physicians, and African American female physicians. Dr. Yancy: Physicians who are culturally sensitive provide care in a more equitable manner. Dr. Roberts: Is cigarette smoking a factor in ICD use? Dr. Fonarow: Cigarette smoking is not a relative or absolute contraindication for ICD therapy. Dr. Roberts: How do you advise patients who decline an ICD? Dr. Fonarow: All of the other appropriate aspects of HF therapy are important, including medications, lifestyle modications, and appropriate exercise in a cardiac rehabilitation program. Dr. Olshansky: Some health care providers have a strong bias against persons who smoke cigarettes. Such an attitude toward cigarette smokers is a different type of bias from that of gender and race. Dr. Yancy: Bias occurs in many different modalities and representations, which is why I believe we should consider an across the board performance improvement strategy for ICD therapy. Dr. Olshansky: I am not an advocate of electronic medical records, but perhaps electronic medical records can be used to enforce proper medical management. For example, these guidelines could be built into electronic medical records in a way that a red ag would appear when a patient satised the indications for an ICD. If the physician did not prescribe it, or the patient refused it, the reasons for not using it would be recorded. This approach also might encourage device implantation in a manner that is blind to race and gender. Dr. Fonarow: Whether systems such as electronic medical records can be successful in improving treatment with modalities as complex as placement of ICDs and cardiac resynchronization devices requires further study. With large gaps in treatment and the disparities that have been identied, we need to do something different from the present conventional hit or miss treatment approach. A more reliable, safe, effective, and unbiased delivery system consistently delivering appropriate therapies in all settings should be the goal. Dr. Friedewald: Are there signs of disparity reductions for ICDs in the Get With the GuidelinesHeart Failure program? Dr. Fonarow: Preliminary data suggest that there has been some improvement, but it is too early to be certain. In other areas, such as use of renin-angiotensin system inhibitors and blockers, there is profound evidence that these types of programs improve the quality of care and treatment rates.

Dr. Friedewald: Perhaps guidelines themselves should place greater emphasis on race and gender issues. Dr. Fonarow: I agree. The 2005 HF guidelines on indications for ICD implantation included a section on special populations indicating that the recommendations were applicable to both genders, all races and ethnicities, and to older patients. There is also tremendous variation in ICD use in the outpatient setting compared to the hospital setting. There are some hospitals that have very high rates of qualied patients being treated appropriately with an ICD, and others where the use is nearly zero. Sharing best practice information can decrease these variations, indicating that it is possible to achieve high treatment rates of appropriate guideline-recommended therapies. Dr. Roberts: It seems to be much easier to set up standards and diagnostic and therapeutic criteria in institutions compared to the private practice setting. Dr. Fonarow: I agree. In many hospital settings, where there are standardized practice protocols and multidisciplinary teams, quality improvement initiatives are more likely to succeed. This is true regardless of the hospital type: teaching or nonteaching, rural or urban, large or small. The outpatient setting, however, is a far greater challenge, but there have been successes in this area as well. Dr. Friedewald: What is the role of electrophysiologic testing in assessing people for possible ICD implantation? Dr. Olshansky: The electrophysiology test has a potential role in patients who have nonsustained VT, are not post myocardial infarction, have had no recent acute intervention, are clinically relatively stable, and have a LV ejection fraction 40% with New York Heart Association functional class I HF. This subset may benet from electrophysiology testing to assess the risk for primary prevention with the ICD. T-wave alternans has been mentioned as a possible noninvasive predictor, but it will never be helpful in determining who should receive an ICD. T-wave alternans will ultimately fall the way of other noninvasive tests. We are left with only a hope for a good noninvasive predictor, so we must rely mainly on New York Heart Association functional class and LV ejection fraction. Dr. Friedewald: Does persistent atrial brillation affect the indications for ICD placement? Dr. Yancy: The patient with atrial brillation is at greater risk for HF and SCD. Dr. Olshansky: For ICD indications, atrial brillation is a 2-edged sword. A post hoc analysis of atrial brillation patients or syncope patients in the SCD-HeFT trial found that those patients do not benet from an ICD but that they are at higher risk of both total mortality and fatal arrhythmia.6 Because those ndings are post hoc analyses, it is hard to use those predictors to determine who would benet the most from ICD placement. We do not have good prospective data to determine the highest risk patients. Some of the highest risk patients are also the oldest patients and the most likely to have other co-morbidities so that they are going to die anyway and therefore may not benet much by receiving an ICD. Our challenge is to nd a middle ground. The patient with an LV ejection fraction that is not too abnormal is not going to benet much from an ICD, but when the LV ejection fraction is too low the patient also is not going to benet from an ICD. Thus, there is an in-between popu-

Roundtable Discussion/ICDs in Primary Prevention of SCD

589

lationwith a modestly poor LV ejection fraction and modestly reduced functional classwho seem to derive the most benet from an ICD. Dr. Roberts: How much does the ICD cost? Dr. Yancy: The usual metric for HF is $50,000 per quality-adjusted life-year. For an ICD, the cost is about $37,500. Thus, by accepted denitions, it seems reasonably cost effective. Dr. Olshansky: It is at least as cost effective as many of our other standard medical therapies. Dr. Fonarow: I agree. Dr. Roberts: The most common mode of death in patients with angina pectoris is cardiac arrest. Would you personally have an ICD inserted if you had chronic stable angina? Dr. Yancy: No, I would want surgical revascularization. Dr. Fonarow: There is no evidence that ICD placement offers substantial value in chronic angina pectoris, but we do have great evidence in favor of antiplatelet drugs, blockers, renin-angiotensin system inhibitors, and aggressive statin therapy. Dr. Roberts: If you had an acute myocardial infarction 50 days ago, an ejection fraction of 40%, and class I HF, would you like to receive an ICD if money were not a factor? Dr. Fonarow: No, because I would believe that with appropriate medical therapy, my risk would be low enough. Dr. Yancy: What if your LV ejection fraction was 35%? Dr. Fonarow: Then sign me up for an ICD! Dr. Roberts: But the ejection fraction margin of error, at best, is 5%. Dr. Fonarow: Yes, so there is wiggle room for LV ejection fraction criteria. If I am measured at 35%, there is a possibility that I will be at 30%, and that places me in the group that derives greatest benet from ICD placement. There is a little built-in error ratio around guideline recommendations. Dr. Friedewald: What is the future of ICD therapy? Dr. Olshansky: The future of ICD therapy depends in part on what the health care system can afford. While the price of debrillators might be driven down by improved technology, the ICD will remain expensive. I assume, however, that technology will reduce device size and risk. For example, leadless debrillators are being developed. They are big, but it is possible for these devices to be put in without leads, so there would be no leads to break. There also is progress in extending battery life. Dr. Roberts: What is the current average battery life? Dr. Olshansky: About 5 to 6 years, but if a patient only needs 1 device and if the device could be put in without a lead, the implantation procedure would be much easier and perhaps less expensive. Another issue is to better determine which patient populations to target, because we want to treat patients who are at the highest risk and are going to get the most benet from the device. We have to dene what is meant by absolute risk that is required for the ICD and which patients truly benet from implantation. We also need better markers. While there should not be any bias with regard to who gets an ICD, there may be genetic differences among populations who benet. This area needs research because genetic testing may help identify patients who will benet the most.

Dr. Friedewald: What more should we do about ethnic disparities in treatment? Dr. Yancy: I favor efforts that improve across-the-board adherence to evidence-based medical and device therapies. My observations over the last decade has demonstrated that there are few substantive differences between groups of individuals, especially on the basis of race, and that most of the differences in outcomes are due to the extent to which people receive evidence-based medical or device therapies. Performance improvement strategies that are deployed in a race-blind, gender-blind, age-blind manner are best. One mistake we made several years ago when we launched evidence-based therapy on the basis of race was that we forced the medical community to emphasize race, which was polarizing. The uptake in therapy that should have occurred did not occur, and many people suffered unnecessarily. In an era in which we have limited resources the best way to utilize them is to make certain that every patient affected with disease has the best possible chance for the best possible outcome, the lowest utilization of hospital services, and the greatest improvement in survival. The right approach is not a focus targeted toward a single group, because that does not work. Rather, the right approach is to raise the bar for all individuals with disease, recognizing that some will have a greater gain than others. Thus, wide deployment of quality-driven performance improvement strategies makes the most sense and is doable. Dr. Friedewald: Thank you. Appendix: Recommendations for Implantable CardioDebrillators3 Class I 1. ICD therapy is indicated in patients who are survivors of cardiac arrest due to VF or hemodynamically unstable sustained VT after evaluation to dene the cause of the event and to exclude any completely reversible causes. (Level of Evidence: A) 2. ICD therapy is indicated in patients with structural heart disease and spontaneous sustained VT, whether hemodynamically stable or unstable. (Level of Evidence: B) 3. ICD therapy is indicated in patients with syncope of undetermined origin with clinically relevant, hemodynamically signicant sustained VT or VF induced at electrophysiological study. (Level of Evidence: B) 4. ICD therapy is indicated in patients with LV ejection fraction 35% due to prior myocardial infarction who are at least 40 days post-myocardial infarction and are in NYHA functional Class II or III. (Level of Evidence: A) 5. ICD therapy is indicated in patients with nonischemic dilated cardiomyopathy who have an LV ejection fraction 35% and who are in NYHA functional Class II or III. (Level of Evidence: B) 6. ICD therapy is indicated in patients with LV dysfunction due to prior myocardial infarction who are at least 40 days post-myocardial infarction, have an LV ejection fraction 30%, and are in NYHA functional Class I. (Level of Evidence: A) 7. ICD therapy is indicated in patients with nonsustained VT due to prior myocardial infarction, LV ejection

590

The American Journal of Cardiology (www.ajconline.org)

fraction 40%, and inducible VF or sustained VT at electrophysiological study. (Level of Evidence: B) Class IIa 1. ICD implantation is reasonable for patients with unexplained syncope, signicant LV dysfunction, and nonischemic dilated cardiomyopathy. (Level of Evidence: C) 2. ICD implantation is reasonable for patients with sustained VT and normal or near-normal ventricular function. (Level of Evidence: C) 3. ICD implantation is reasonable for patients with hypertrophic cardiomyopathy who have 1 major risk factor for SCD. (Level of Evidence: C) 4. ICD implantation is reasonable for the prevention of SCD in patients with arrhythmogenic right ventricular dysplasia/cardiomyopathy who have 1 risk factor for SCD. (Level of Evidence: C) 5. ICD implantation is reasonable to reduce SCD in patients with long-QT syndrome who are experiencing syncope and/or VT while receiving beta-blockers. (Level of Evidence: B) 6. ICD implantation is reasonable for nonhospitalized patients awaiting transplantation. (Level of Evidence: C) 7. ICD implantation is reasonable for patients with the Brugada syndrome who have had syncope. (Level of Evidence: C) 8. ICD implantation is reasonable for patients with the Brugada syndrome who have documented VT that has not resulted in cardiac arrest. (Level of Evidence: C) 9. ICD implantation is reasonable for patients with catecholaminergic polymorphic VT who have syncope and/or documented sustained VT while receiving beta blockers. (Level of Evidence: C) 10. ICD implantation is reasonable for patients with cardiac sarcoidosis, giant cell myocarditis, or Chagas disease. (Level of Evidence: C) Class IIb 1. ICD therapy may be considered in patients with nonischemic heart disease who have an LV ejection fraction 35% and who are in NYHA functional Class I. (Level of Evidence: C) 2. ICD therapy may be considered for patients with the long-QT syndrome and risk factors for SCD. (Level of Evidence: B) 3. ICD therapy may be considered in patients with syncope and advanced structural heart disease in whom thorough invasive and noninvasive investigations have failed to dene a cause. (Level of Evidence: C) 4. ICD therapy may be considered in patients with a familial cardiomyopathy associated with sudden death. (Level of Evidence: C) 5. ICD therapy may be considered in patients with LV noncompaction. (Level of Evidence: C) Class III 1. ICD therapy is not indicated for patients who do not have a reasonable expectation of survival with an acceptable functional status for at least 1 year, even if 2. 3.

4.

5.

6.

7.

they meet ICD implantation criteria specied in the Class I, IIa, and IIb recommendations above. (Level of Evidence: C) ICD therapy is not indicated for patients with incessant VT or VF. (Level of Evidence: C) ICD therapy is not indicated in patients with signicant psychiatric illnesses that may be aggravated by device implantation or that may preclude systematic follow-up. (Level of Evidence: C) ICD therapy is not indicated for NYHA Class IV patients with drug-refractory HF who are not candidates for cardiac transplantation or implantation of a cardiac resynchronization therapy device that incorporates both pacing and debrillation capabilities. (Level of Evidence: C) ICD therapy is not indicated for syncope of undetermined cause in a patient without inducible ventricular tachyarrhythmias and without structural heart disease. (Level of Evidence: C) ICD therapy is not indicated when VF or VT is amenable to surgical or catheter ablation (e.g., atrial arrhythmias associated with the Wolff-ParkinsonWhite syndrome, right ventricular or LV outow tract VT, idiopathic VT, or fascicular VT in the absence of structural heart disease). (Level of Evidence: C) ICD therapy is not indicated for patients with ventricular tachyarrhythmias due to a completely reversible disorder in the absence of structural heart disease (e.g., electrolyte imbalance, drugs, or trauma). (Level of Evidence: B)

1. Mirowski M, Reid PR, Mower MM, Watkins L, Gott VL, Schauble JF, Langer A, Heilman MS, Kolenik SA, Frischell RD, Weisfeldt ML. Termination of malignant ventricular arrhythmias with an implantable automatic debrillator in human beings. N Engl J Med 1980;303:322324. 2. Mirowski M, Mower MM, Langer A, Heilman MS, Schreibman L. A chronically implanted system for automatic debrillation in active conscious dogs: experimental model for treatment of sudden death from ventricular brillation. Circulation 1978;58:90 94. 3. Epstein AE, DiMarco JP, Ellenbogen KA, Estes NAM, Freedman RA, Gettes LS, Gillinov AM, Gregoratos G, Hammill SC, Hayes DL, Hlatky MA, Newby LK, Page RL, Schonfeld MH, Sitka MJ, Stevenson LW, Sweeny MO. ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: executive summary. Circulation 2008;117:2820 2840. 4. Hohnloser SH, Kuck KH, Dorian P, Roberts RS, Hampton JR, Hatala R, Fain E, Gent M, Connolly SJ, for the DINAMIT Investigators. Prophylactic use of an implantable cardioverter-debrillator after acute myocardial infarction. N Engl J Med 2004;351:24812488. 5. Fonarow GC, Yancy CW, Albert NM, Curtis AB, Gattis Stough W, Gheorghiade M, Heywood JT, Mehra M, OConnor CM, Reynolds D, Walsh NM. Improving the use of evidence-based heart failure therapies in the outpatient setting: the IMPROVE HF performance improvement registry. Am Heart J 2007;154:1238. 6. Bardy GH, Lee KL, Mark DB, Poole JE, Packer DL, Boineau R, Domanski M, Troutman C, Anderson J, Johnson G, McNulty SE, Clapp-Channing N, Davidson LD, Fraulo ES, Fishbein DP, Luceri RM, Ip JH, for the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) Investigators. Amiodarone or an implantable cardioverter-debrillator for congestive heart failure. N Engl J Med 2005;352:225237. 7. Hernandez AF, Fonarow GC, Liang L, AI-Khatib SM, Curtis LH, LaBresh KA, Yancy CW, Albert NM, Peterson ED. Sex and racial differences in the use of implantable cardioverter-debrillators among patients hospitalized with heart failure. JAMA 2007;298: 15251532.

Comparison of the Effectiveness and Safety of Low-Molecular Weight Heparin Versus Unfractionated Heparin Anticoagulation After Heart Valve Surgery
Claudia Bucci, PharmDa,b,*, William H. Geerts, MDc, Andrew Sinclair BScPhmb, and Stephen E. Fremes, MDd
Although unfractionated heparin (UFH) is used routinely after heart valve surgery at many institutions, cardiovascular surgery patients have a particularly high risk for developing heparin-induced thrombocytopenia (HIT). The aim of this study was to compare the efcacy and safety of low-molecular-weight heparin (LMWH) or UFH after heart valve surgery by conducting a retrospective evaluation of consecutive cardiovascular surgery patients in whom the LMWH dalteparin (n 100) was used as the postoperative anticoagulant. This group was compared to an earlier group of patients who received UFH (n 103). The main outcomes included the efcacy of the anticoagulant regimens (determined by the incidence of valve thrombosis, arterial thromboembolic events, and venous thromboembolic events) and the safety (determined by major bleeding, HIT, thrombotic events in HIT-positive cases, and death). Overall, there were for fewer thrombotic events in the LMWH-treated group (4% vs 11%, p 0.11). There was a higher rate of bleeding events in the UFH-treated group (10% vs 3%, p 0.08). Six patients in the UFH-treated group developed HIT, 4 of whom had thrombotic events (HIT with thrombosis). In the LMWHtreated group, 3 patients developed HIT, 1 of whom had HIT with thrombosis. In conclusion, in this study, an LMWH regimen after heart valve surgery was effective and safe, with fewer thrombotic, bleeding, HIT, and HIT with thrombosis events. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:591594) In the past, intravenous unfractionated heparin (UFH) was used routinely at our institution after heart valve replacement surgery to prevent thrombotic complications (Appendix A). LMWH is associated with a substantially lower rate of heparin-induced thrombocytopenia (HIT) and HIT with thrombosis than UFH and may be a safer alternative after heart valve surgery.1 We replaced intravenous and subcutaneous UFH with subcutaneous low-molecularweight heparin (LMWH) in prophylactic or therapeutic doses for early anticoagulation after heart valve replacement surgery (Appendix B). The objective of the study was to assess the efcacy and safety of anticoagulation with LMWH after heart valve surgery compared to UFH. Methods In March 2006, we implemented an avoid-heparin policy after cardiovascular surgery (Appendixes A and B). Intraoperative UFH was used in all cases. We conducted a retrospective evaluation of consecutive patients in whom the LMWH dalteparin was used, and we compared this group to an earlier group of patients who received UFH. This study was approved by the ethics review board of Sunnybrook Health Sciences Centre. The main outcome measures included the efcacy (determined by the incidence of valve thrombosis, arterial thromboembolic events, and venous thromboembolic events) and the safety (determined by major bleeding, HIT, thrombotic events in HIT-positive cases, and death) of the 2 anticoagulant regimens. All outcomes collected occurred during the operative hospital admission. Conrmed HIT was dened by 1 of the following: positive serotonin release assay, positive HIT enzyme-linked immunosorbent assay plus high clinical probability for HIT
www.ajconline.org

a Department of Pharmacy and bFaculty of Pharmacy, University of Toronto; and cThromboembolism Program, Department of Medicine, and d Division of Cardiovascular Surgery, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada. Manuscript received July 15, 2010; revised manuscript received and accepted October 11, 2010. Dr. Bucci has received research grant support from AstraZeneca, Wilmington, Delaware. Dr. Bucci is a consultant for Sano-Aventis, Paris, France; Bristol-Myers Squibb, New York, New York; Bayer Healthcare, Munich, Germany; Boehringer Ingelheim, Ingelheim, Germany; and Eli Lilly & Company, Indianapolis, Indiana. Dr. Geerts has received research grant support from Bayer Healthcare; Pzer, Inc., New York, New York; and Sano-Aventis. Dr. Geerts is a consultant for Bayer Healthcare; Boehringer Ingelheim; GlaxoSmithKline, London, United Kingdom; LEO Pharma A/S, Ballerup, Denmark; Pzer, Inc.; and Sano-Aventis. Dr. Geerts has received honoraria for presentations from Bayer Healthcare, Boehringer Ingelheim, Pzer, Inc., and Sano-Aventis. Dr. Fremes has received honoraria from Sano-Aventis; Bayer Healthcare; Astellas Pharma US, Inc., Deereld, Illinois; Novo Nordisk A/S, Bagsvrd, Denmark; Novadaq, Bonita Springs, Florida; Medtronic, Inc., Minneapolis, Minnesota; Edwards Lifesciences, Irvine, California; Sorin Group USA, Inc., Arvada, Colorado; and Merck & Company, Whitehouse Station, New Jersey. Dr. Fremes has received research support from St. Jude Medical, Inc., St. Paul, Minnesota; Aventis; Proctor & Gamble, Cincinnati, Ohio; Medicure, Winnipeg, Manitoba, Canada; and Merck & Company. *Corresponding author: Tel: 416-480-6755; fax: 416-480-5887. E-mail address: claudia.bucci@sunnybrook.ca (C. Bucci).

0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.020

592 Table 1 Baseline characteristics Characteristic Age (years) Age 75 years Men/women Left ventricular function 1 2 3 4 Atrial brillation* Renal dysfunction Valve replacement Single Mechanical aortic Tissue aortic Mechanical mitral Tissue mitral Mitral annuloplasty Tricuspid repair Double Coronary bypass Length of surgery (hours) Length of stay after surgery (days)

The American Journal of Cardiology (www.ajconline.org)

Heparin (n 103) 65.1 12.6 (3187) 23 (22%) 72/31 57 (55%) 22 (21%) 16 (16%) 8 (8%) 66 (64%) 13 (13%) 95 (92%) 27 21 17 14 15 1 8 (8%) 43 (42%) 4.9 1.7 (2.310.2) 15.9 9.9 (561)

Dalteparin (n 100) 66.1 12.1 (4384) 24 (24%) 67/33 52 (52%) 28 (28%) 15 (15%) 5 (5%) 68 (68%) 16 (16%) 90 (90%) 23 19 21 8 18 1 10 (10%) 32 (32%) 5.3 1.9 (2.712.9) 16.1 9.5 (664)

p Value 0.57 0.87 0.76 0.65

0.67 0.55 0.63

0.63 0.19 0.09 0.87

Data are expressed as mean SD (range) or as number (percentage). * Includes transient and chronic episodes. Transient atrial brillation occurred in 49 and 43 patients in the heparin and LMWH groups, respectively (p 0.57). Chronic atrial brillation was present preoperatively and/or was persistent after surgery and occurred in 17 and 25 patients in the heparin and LMWH groups, respectively (p 0.17). Creatinine clearance 30 ml/min.

or strongly positive HIT enzyme-linked immunosorbent assay (optical density 1.0). HIT was ruled out in patients with negative results on HIT enzyme-linked immunosorbent assay or serotonin release assay. Major bleeding was dened as any overt bleeding meeting 1 of the following criteria: proved fatal bleeding, intracranial hemorrhage (computed tomography or magnetic resonance imaging required), retroperitoneal bleeding (ultrasound, computed tomography, or magnetic resonance imaging required), bleeding requiring an intervention (pericardial bleeding requiring reoperation or catheter drainage of blood, pleural bleeding requiring thoracotomy or chest tube, gastrointestinal bleeding requiring surgery or endoscopic treatment, wound bleeding requiring reoperation), other life-threatening bleeding at a critical site, bleeding requiring transfusion of 2 U of red blood cells, or bleeding that resulted in chronic sequelae or prolongation of the hospital stay. Bleeding requiring pericardiocentesis, thoracentesis, or diagnostic endoscopy alone was not considered major. Nonmajor bleeding was dened as any of the following: epistaxis requiring nasal packing, airway bleeding, hematuria, hematemesis (but not just coffee grounds), or gastrointestinal bleeding (frank blood or melena stools) not requiring an intervention. All analyses were done using InStat version 3 (GraphPad Software, San Diego, California). All statistical tests were 2 sided and used a p value of 0.05 as the threshold for statistical signicance. Baseline discrete variables are presented as frequencies or percentages, while continuous variables are presented as mean SD or as median (interquartile range).

The frequencies of the clinical end points were compared using a chi-square or Fishers exact tests. Results The control group consisted of 103 consecutive patients treated with UFH after heart valve surgery from April 2004 to May 2006. These patients received only UFH in therapeutic (83%) or prophylactic (17%) doses. The control group patients were compared to 100 heart valve patients given therapeutic (73%) or prophylactic (27%) dalteparin postoperatively from March 2006 to August 2007. The 2 groups were similar for a large number of demographic and clinical characteristics (Table 1). The mean age was approximately 65 years, and 68% of patients were men. Thrombotic and bleeding risk factors in the 2 groups were similar apart from greater postoperative aspirin use in the UFH patients (Table 2). Approximately 60% of the study population had 1 risk factor for thrombosis, and 80% of patients had 1 risk factor for bleeding. Overall, there were fewer thrombotic events in the LMWH-treated group, although the difference was not statistically signicant (4% vs 11%, p 0.11; Table 3). In the UFH group, there were 11 thrombotic events (5 strokes, 1 valve thrombosis, 1 ischemic bowel, 2 transient global amnesia, 1 foot embolus, 1 kidney infarction). In the dalteparin group, there were 4 thrombotic events (3 strokes, 1 ischemic bowel). The thromboembolic events are listed in Table 4 . In the UFH-treated group, 4 thrombotic events occurred in

Valvular Heart Disease/ Table 2 Risk factors for thrombosis and bleeding Risk Factor Thrombotic Atrial brillation* Grade 4 left ventricle Previous thromboembolism Left atrial enlargement Previous myocardial infarction Bleeding Postoperative aspirin Renal dysfunction Clopidogrel Nonsteroidal antiinammatory drug Intraoperative stroke International normalized ratio 5 (postoperatively) Coagulation disorder Therapeutic anticoagulation (postoperative) Warfarin (postoperative) Heparin (n 103) 17 (17%) 8 (8%) 6 (6%) 17 (17%) 13 (13%) 93 (90%) 13 (13%) 3 (3%) 12 (12%) 1 (1%) 4 (4%) 86 (83%) 100 (97%) Dalteparin (n 100) 26 (26%) 5 (5%) 12 (12%) 12 (12%) 10 (10%) 32 (32%) 16 (16%) 1 (1%) 27 (27%) 2 (2%) 3 (3%) 73 (73%) 97 (97%)

593

bleed. Most of the bleeds occurred in patients receiving therapeutic anticoagulation with UFH or LMWH or with warfarin. Concomitant aspirin increased the risk for bleeding events (odds ratio 7.43, 95% condence interval 1.85 to 126). Six patients in the UFH-treated group developed HIT, 4 of whom had thrombotic events (HIT with thrombosis). In the LMWH-treated group, 3 patients developed HIT, 1 of whom had HIT with thrombosis. There was 1 death in each group during hospitalization, both related to HIT (Table 3). Discussion Patients who have undergone cardiac surgery routinely receive UFH during and after surgery. The product monographs for LMWHs warn against the use of LMWH for prevention of thromboembolism in patients with prosthetic heart valves, including those who are pregnant.2 This is based on 2 cases of valve thrombosis in pregnant women receiving enoxaparin.3 The Anticoagulation in Prosthetic Valves and Pregnancy Consensus Report Panel concluded that the level of anticoagulation with enoxaparin may not have been optimal in these cases.4 LMWH may be a safer alternative to UFH in cardiac surgery patients because of the lower risk for HIT. However, to date, the safety and efcacy of LMWH after mechanical heart valve surgery has been poorly evaluated. In a nonrandomized case series of 208 patients who underwent heart valve replacement, therapeutic anticoagulation was more rapidly and predictably achieved with LMWH than with UFH.5 In a larger study with no control group, the use of the LMWH enoxaparin as bridging to therapeutic anticoagulation with warfarin after mechanical valve replacement appeared to be safe and effective.6 In a small study, LMWH patients were matched to patients who received UFH after mechanical heart valve implantation.7 Although bridging with LMWH was as safe and effective as bridging with UFH, LMWH was associated with reduced length of hospital stay and costs. In this study, we found that an LMWH regimen after heart valve surgery was effective and safe. In addition, there was a lower risk for thrombosis and bleeding in the LMWHtreated group. No cases of valve thrombosis occurred in the LMWH group of this study. There were also fewer HIT and HIT with thrombosis events in patients receiving LMWH. Our study provides data supporting the use of LMWH in patients with newly implanted heart valves. We believe that this is the rst report assessing the development of HIT in this patient population. Our study was retrospective in nature and consisted of a relatively small sample. Although the baseline characteristics were similar, confounding may be present. Confounding from unmeasured factors such as the motivation of the clinical staff and increased attention to postoperative anticoagulation may have been present. We believe that this is the rst intervention study to systematically attempt to reduce HIT in cardiac surgery. The results of this preliminary study are important as part of ongoing quality assurance of the protocol implemented at our institution.

* Represents chronic atrial brillation (i.e., present preoperatively or at discharge). Includes factor VII deciency (n 1) and sickle-cell disease (n 2). Table 3 Comparison of efcacy and safety outcomes Outcome Thrombotic events Stroke Valve thrombosis Ischemic bowel Transient global Amnesia Foot embolus Infarction kidney Bleeding events Pericardial Retroperitoneal Hemothorax Patients with HIT Patients with HIT with thrombosis Death due to HIT Heparin (n 103) 11 (11%) 5 1 1 2 1 1 10 (10%) 5 4 1 6 (6%) 4 (4%) 1 (1%) Dalteparin (n 100) 4 (4%) 3 1 p Value .11

3 (3%) 2 1 3 (3%) 1 (1%) 1 (1%)

0.08

0.50 0.37 1.00

HIT-positive patients, while in the LMWH-treated group, 1 thrombotic event occurred in an HIT-positive patient. There was a higher rate of bleeding events in the UFHtreated group (10% vs 3%, p 0.08; Table 3). There were 10 major bleeding events in the UFH-treated group (5 pericardial bleeds, 4 retroperitoneal bleeds, and 1 hemothorax; all these patients received concomitant aspirin and/or clopidogrel) and 3 major bleeding events in the LMWH-treated group (2 pericardial bleeds, 1 hemothorax; 2 of these patients received concomitant aspirin and/or clopidogrel). One patient in each group had a nonmajor lower gastrointestinal

594

The American Journal of Cardiology (www.ajconline.org)

Appendix A: Unfractionated Heparin Anticoagulation Protocol After Heart Valve Replacement

Valve Position Mechanical valves Aortic Aortic Mitral Additional Risk Factors* No Yes No Heparin Therapy Aspirin Use Target INR

Mitral

Yes

5,000 U SC q12h until INR 2. If INR 2 on POD 4, start IV heparin. 5,000 U SC q12h until INR 2. If INR 2 on POD 4, start IV heparin. 500 U/hour IV starting 12 hours postoperatively for 96 hours. If INR 2 on POD 4, start IV heparin. 500 U/hour IV starting 12 hours postoperatively for 96 hours. If INR 2 on POD 4, start IV heparin. None (except for DVT prophylaxis). 5,000 U SC q12h until INR 2. If INR 2 on POD 4, start IV heparin. 5,000 U SC q12h until INR 2. If INR 2 on POD 4, start IV heparin. 5,000 U SC q12h until INR 2. If INR 2 on POD 4, start IV heparin.

None ECASA 81 mg/day None

2.03.0 2.03.0 2.53.5

ECASA 81 mg/day

2.53.5

Tissue valves Aortic Aortic Mitral and mitral annuloplasty Mitral and mitral annuloplasty

No Yes No Yes

ECASA 325 mg/day None ECASA 325 mg/day; start after 3 months of warfarin None

No warfarin 2.03.0 2.03.0; warfarin only for 3 months 2.03.0

* Atrial brillation, large left atrium, left atrial thrombus, previous thromboembolism. DVT deep venous thrombosis; ECASA enteric-coated aspirin; INR international normalized ratio; IV intravenous; POD postoperative day; q12h every 12 hours; SC subcutaneous.

Appendix B: Low-Molecular-Weight Heparin Anticoagulation Protocol After Heart Valve Replacement

Valve Position Mechanical valves Aortic LMWH Therapy Dalteparin 2,500 U SC qhs. If INR 2 at 96 hours postoperatively, increase dalteparin dose.* Dalteparin 5,000 U SC qhs. If INR 2 at 96 hours postoperatively, increase dalteparin dose.* None Dalteparin 2,500 U SC qhs. If INR 2 at 96 hours postoperatively, increase dalteparin dose.* Dalteparin 2,500 U SC qhs. If INR 2 at 96 hours postoperatively, increase dalteparin dose.* None Aspirin Target INR 2.03.0

Mitral

None

2.53.5

Tissue valves Aortic without risk factors Aortic with risk factors*

ECASA 325 mg/day None

None 2.03.0

Mitral and mitral annuloplasty

None

2.03.0 Warfarin for 3 months (unless risk factors are present), then ECASA 325 mg/day is added

* Sustained or intermittent atrial brillation lasting 48 hours, maze procedure. ECASA enteric-coated aspirin; INR international normalized ratio; qhs at bedtime; SC subcutaneous.

1. Martel N, Lee J, Wells PS. Risk for heparin-induced thrombocytopenia with unfractionated heparin and low-molecular-weight heparin thromboprophylaxis: a meta-analysis. Blood 2005;106:2710 2715. 2. Canadian Pharmacists Association. Product monograph heparins: low molecular weight. In: e-CPS: Compendium of Pharmaceuticals and Specialties. Available at: http://www.pharmacists.ca/content/products/ ecps_english.cfm. Accessed February 4, 2010. 3. Shapira Y, Sagie A, Battler A. Low-molecular-weight heparin for the treatment of patients with mechanical heart valves. Clin Cardiol 2002;25:323327. 4. Anticoagulation and enoxaparin use in patients with prosthetic heart vavles and/or pregnancy. Clin Cardiol Consensus Rep 2002;3:120.

5. Montalescot G, Polle V, Collet JP, Leprince P, Bellanger A, Gandjbakhch I. Low molecular weight heparin after mechanical heart valve replacement. Circulation 2000;101:10831086. 6. Meurin P, Tabet JY, Weber H, Renaud N, Ben Driss A. Low-molecularweight heparin as a bridging anticoagulant early after mechanical heart valve replacement. Circulation 2006;113:564 569. 7. Fanikos J, Tsilimingras K, Kucher N, Rosen AB, Hieblinger MD, Goldhaber SZ. Comparison of efcacy, safety and cost of low-molecular weight heparin with continuous-infusion unfractionated heparin for initiation of anticoagulation after mechanical prosthetic valve implantation. Am J Cardiol 2004;93:247250.

Seeking Optimal Relation Between Oxygen Saturation and Hemoglobin Concentration in Adults With Cyanosis from Congenital Heart Disease
Craig S. Broberg, MDa,b,*, Ananda R. Jayaweera, PhDa, Gerhard P. Diller, MDb, Sanjay K. Prasad, MDb, Swee Lay Thein, MDc, Bridget E. Bax, PhDd, John Burman, MDb, and Michael A. Gatzoulis, MD, PhDb
In patients with cyanosis from congenital heart disease, erythropoiesis is governed by many factors that can alter the expected relation between the oxygen saturation (O2sat) and hemoglobin concentration. We sought to dene the relation between the O2sat and hemoglobin in such patients and to predict an ideal hemoglobin concentration for a given O2sat. Adults with congenital heart defects and cyanosis were studied prospectively using blood tests and exercise testing. Nonoptimal hemoglobin was dened as any evidence of inadequate erythropoiesis (i.e., iron, folate, or vitamin B12 deciency, increased erythropoietin, reticulocytosis, or a right-shifted oxygen-hemoglobin curve). For patients without these factors, a linear regression equation of hemoglobin versus O2sat was used to predict the optimal hemoglobin for all patients. Of the 65 patients studied, 21 met all the prestudy criteria for an optimal hemoglobin. For all patients, no correlation was found between O2sat and hemoglobin (r 0.22). However, a strong linear correlation was found for those meeting the criteria for optimal hemoglobin (r 0.865, p <0.001). The optimal hemoglobin regression equation was as follows: predicted hemoglobin 57.5 (0.444 O2sat). A negative correlation was found between the hemoglobin difference (predicted minus measured) and exercise duration on cardiopulmonary exercise testing (r 0.396, p 0.005) and the 6-minute walk distance (r 0.468, p <0.001). In conclusion, a strong relation between O2sat and hemoglobin concentration can be shown in stable cyanotic patients and used to predict an optimal hemoglobin. This relation might be useful in dening functional anemia in this group. 2011 Published by Elsevier Inc. (Am J Cardiol 2011;107:595599)

In the clinical care of cyanotic patients with congenital heart disease, it is necessary to assess the appropriateness of a measured hemoglobin level for a given oxygen saturation (O2sat). Although multiple factors can inuence both hemoglobin and O2sat, a tool to predict the optimal relation between these variables would be valuable, particularly for situations in which the hemoglobin might be signicantly

a Adult Congenital Heart Disease Program, Oregon Health and Sciences University, Portland, Oregon; bRoyal Brompton and Hareeld National Health Service Trust, National Heart and Lung Institute, Imperial College London, London, United Kingdom; cDepartment of Haematological Medicine, Kings College London School of Medicine and Kings College Hospital, London, United Kingdom; dChild Health, St. Georges, University of London, London, United Kingdom. Manuscript received July 30, 2010; manuscript received and accepted October 1, 2010. The study was funded by the Clinical Research Committee, Royal Brompton Hospital. Dr. Broberg has received support from the Waring Trust (London, United Kingdom) through the Royal Brompton Hospital and the Tartar Trust (Portland, Oregon) through the Oregon Health and Sciences University. Dr. Gatzoulis and the Royal Brompton Adult Congenital Heart Centre have received support from the British Heart Foundation, London, United Kingdom and unrestricted research funds from Actelion UK (London, United Kingdom). *Corresponding author: Tel: (503) 494-8750; fax: (503) 494-8550. E-mail address: brobergc@ohsu.edu (C. Broberg).

less than expected, such as postoperatively or after severe hemoptysis. We hypothesized that by controlling for factors that might alter this relation, particularly those that might limit erythropoiesis, such as iron deciency, an ideal linear relation could be found that would dene an optimal hemoglobin level for a given O2sat. We also hypothesized that patients with an optimal hemoglobin might have a clinical advantage, as measured by the exertional capacity. We therefore prospectively measured the variables that could potentially alter the hemoglobinO2sat association to determine their optimal relation. Methods We prospectively enrolled consecutive adults with congenital heart disease in a descriptive cross-sectional study. Patients gave consent, and institutional ethics review approved the protocol. Patients were included if they had a known congenital defect with a right-to-left shunt. We included patients with a wide range of O2sat, including some patients who had undergone previous repair and had a normal O2sat at the study. All tests were obtained within a 24-hour period. Other data from the present study have been previously reported.1,2 The patients were recruited and seen at the Royal Brompton Hospital. Additional blood testing was done at
www.ajconline.org

0002-9149/11/$ see front matter 2011 Published by Elsevier Inc. doi:10.1016/j.amjcard.2010.10.019

596

The American Journal of Cardiology (www.ajconline.org)

Table 1 Prestudy criteria for determination of adequate erythropoiesis Variable Transferrin saturation Red blood cell folate (g/L) Vitamin B12 (ng/L) Serum erythropoietin (IU/L) Reticulocyte count (%) Hypochromic cell count (%) P50 of oxygenhemoglobin dissociation curve (mm Hg) Cutoff 20% 200 180 25 2 6 29 Normal Range 2045 164900 180900 variable 2% 6% 2529 Excluded Patients (n) 26 0 2 9 8 7 10

dence intervals, around this line. Using the regression equation, the values for the predicted hemoglobin were made, and the difference between the predicted and measured hemoglobin was obtained (hemoglobin difference) for each patient. The clinical variables between the patients with and without an optimal hemoglobin level were compared using Students t test and correlations using Pearsons coefcient. The data are expressed as the mean SD, and p 0.05 was considered statistically signicant. No adjustment was made for multiple comparisons. Results A total of 65 patients were studied (mean age 36 12 years, 67% women). For the whole group, the O2sat at rest was 81 8%, hemoglobin was 19.6 2.9 g/dl, and hematocrit was 60 8%. The anatomic diagnoses are listed in Table 2, as well as their group designation. All but 1 patient had pulmonary vascular disease. Of the 65 patients, 20 had likely been cyanotic at birth, with 45 (largely with simple shunts) having developed right to left shunting over time. No patient was found to have signicant renal, liver, or thyroid dysfunction. Of the 65 patients studied, 44 met 1 criteria for exclusion, and most of them met at least 2 criteria (Table 1). The exclusions on clinical grounds included 16 using supplemental oxygen, 8 with differential cyanosis, 10 with recent phlebotomy, and 2 with recent hemoptysis. The most common exclusion criterion was iron deciency, and many of those patients met additional criteria associated with iron deciency (e.g., phlebotomy, hemoptysis, P50 shift, or erythropoietin elevation). After the exclusions, 21 patients had met all the criteria for adequate erythropoiesis. For the entire cohort, no signicant relation was found between O2sat and hemoglobin (Figure 1). In contrast, when patients with evidence of inadequate erythropoiesis were excluded, a strong linear relation was found. The slope and intercept for the regression line dened a predicted optimal hemoglobin as follows: predicted hemoglobin 0.444(O2sat) 57.5. From this, the predicted hemoglobin for a given O2sat, including the upper and lower condence intervals, were calculated (Table 3 and Figure 2). To establish the clinical relevance of our predicted line, we sought correlations with the functional parameters. Those with an optimal hemoglobin level had a better 6-minute walk test distance (415 119 vs 333 112 m, p 0.011) and treadmill exercise duration (7.48 2.70 vs 5.13 2.01 minutes, p 0.001). The correlation coefcient values between the hemoglobin difference and the outcome variables are listed in Table 4. A signicant inverse correlation was found with the 6-minute walk distance (Figure 3) and exercise duration, such that a greater difference (i.e., measured hemoglobin less than predicted) was associated with poorer function. A similar relation was seen with the ventilatory efciency slope and heart rate reserve. These same variables did not correlate with the measured hemoglobin (Figure 3 and Table 4). The correlation with the peak oxygen consumption and percentage of the maximum predicted oxygen consumption did not reach statistical signicance. Blood

Patients who met all these criteria were considered to have optimal hemoglobin. Additional clinical exclusions listed in text.

Kings College Hospital and St. Georges Hospital (London, United Kingdom). The analyses were performed at the Oregon Health and Sciences University (Portland, Oregon). O2sat was measured using transcutaneous spectrometry in the nger after 5 minutes of rest in the sitting position. The blood was drawn in the morning with the patient in a nonfasting state using a venous cannula in the antecubital region. The hemoglobin concentration, packed cell volume, platelet count, basic serum chemistry panels, liver function tests, iron, ferritin, transferrin saturation, red blood cell vitamin B12, folate, thyroid stimulating hormone, and serum erythropoietin3 were measured. The percentage of hypochromic cells and reticulocyte count were measured using an automated Coulter counter (Advia 120, Bayer, United Kingdom). The partial oxygen pressure at half saturation (P50) of the O2 hemoglobin dissociation curve was also measured (Hem-O-Scan, American Instrument Company, Silver Springs, Maryland). The whole blood viscosity over a range of shear was measured using a rotational viscometer. The viscosity was then remeasured after the hematocrit had been diluted to 45% using autologous serum.1 The patients also performed a 6-minute walk test and treadmill exercise, with the measured oxygen consumption and ventilatory efciency recorded, as previously described.2 After the collection of all data, we identied those patients with any evidence of potentially inadequate or excessive erythropoiesis, according to the presence of 1 of the following a priori criteria: evidence of iron deciency, vitamin B12 or folate deciency, elevated serum erythropoietin, reticulocytosis, hypochromia, or a signicant rightward shift of the O2 hemoglobin dissociation curve (Table 1). We also excluded patients using various clinical criteria, including acute hospitalization, therapeutic phlebotomy within the previous 6 months, and recent signicant hemoptysis (requiring hospitalization). Patients with a patent ductus arteriosus and differential cyanosis were also excluded from the optimal category because of the uncertainty of what the mean O2sat would be. Patients using supplemental oxygen regularly were excluded because their O2sat at room air might not have accurately reected their average daily saturation. After exclusion of any patient who had met these criteria, a plot of the O2sat and measured hemoglobin was made. A linear regression equation was dened, together with con-

Congenital Heart Disease/Optimal Hemoglobin in Cyanosis Table 2 Study population by anatomic lesion Diagnosis Atrial septal defect Ventricular septal defect Atrioventricular septal defect Patent ductus arteriosus Truncus arteriosus Transposition of great arteries Pulmonary atresia with ventricular septal defect Other complex Total Patients (n) 4 27 8 8 6 5 2 5 65 Cyanotic Since Birth 0 2 0 0 6 5 2 5 20 Cyanosis Developed* 4 25 8 8 0 0 0 0 45 Excluded (Nonoptimal) 2 11 7 8 6 4 2 4 44

597

Included (Optimal) 2 16 1 0 0 1 0 1 21

All transposition patients also had ventricular septal defect. * Developed right-to-left shunt in setting of elevated pulmonary vascular resistance (Eisenmenger reaction).

Figure 1. Scatter plot for hemoglobin concentration versus at rest O2sat. For the entire population, no signicant relation was found between hemoglobin and O2sat (dotted line). For patients meeting the criteria for adequate erythropoiesis, a strong linear relation was found (solid line). Regression equations for optimal patients and for all patients shown. Table 3 Prediction of optimal hemoglobin for a given oxygen saturation Oxygen Saturation (%) 93 90 87 85 83 80 77 75 73 CI, condence interval. Predicted Hemoglobin (g/dl) 16.1 17.5 18.8 19.7 20.6 21.9 23.2 24.1 25.0 95% CI (g/dl) 14.417.9 16.019.0 17.520.1 18.421.0 19.221.9 20.423.4 21.425.0 22.126.1 22.827.3

Discussion The concept of determining an ideal set point for erythropoiesis in congenital heart disease with cyanosis is not new. An optimal hematocrit between oxygen delivery and hyperviscosity was studied decades ago,4 although limited by the use of ex vivo models.5 Few clinical studies have addressed this relation and largely only in children or adolescents.5,6 A linear relation has been shown, although less steep than ours.5,7,8 A right shift of the oxyhemoglobin dissociation curve has been seen in iron-decient children.9 We also previously reported a less steep relation in ironreplete adults with patients with Eisenmenger syndrome.6 Our present study, in contrast, was more fastidious with the exclusions, accounting for the steeper slope and narrower condence interval found. Hence, the results of the present study have shown a strong linear association between the hemoglobin level and O2sat that could clinically distinguish patients according to their exercise capacity. For the expected range of O2sat, we estimated the relation could be

viscosity was not different between the optimal and nonoptimal groups, even after adjustment for hematocrit (50 10 vs 49 10 mPa.s at high shear, p 0.82, and 4.29 0.5 vs 4.45 1.1 mPa.s at low shear, p 0.63, respectively).

598

The American Journal of Cardiology (www.ajconline.org)

Figure 2. Predicted O2sat hemoglobin relation. Relation based on regression slope obtained for optimal patients. Upper and lower condence intervals shown. Raw values are listed in Table 3. Table 4 Correlations between exercise parameters and hemoglobin Functional Variable Measured Hemoglobin Hemoglobin Difference (Predicted Measured) r 0.468 0.396 0.261 0.328* 0.388* p Value 0.001* 0.005* 0.080 0.026* 0.007* Figure 3. Correlations between hemoglobin and measured walk distance. (A) Relation between 6-minute walk distance and predictedmeasured hemoglobin was signicant. Negative hemoglobin difference indicated hemoglobin was greater than predicted. (B) In contrast, relation between 6-minute walk distance and measured hemoglobin was poor.

r 6-Minute walk distance (m) Exercise duration (minutes) Peak oxygen consumption (ml/kg/min) Ventilatory efciency slope Heart rate reserve (beats/min) 0.036 0.130 0.157 0.107 0.054

p Value 0.787 0.377 0.298 0.477 0.717

Correlations with measured hemoglobin were not signicant; whereas correlations with hemoglobin difference was signicant. Negative relation indicates patients with greater difference (i.e., hemoglobin less than expected) had poorer exercise capacity. * Statistically signicant.

simplied more conveniently as follows: predicted hemoglobin 61 (O2sat/2). Because all patients with an O2sat 75% met the criteria for having a nonoptimal hemoglobin, it was impossible to predict the optimal hemoglobin for such patients. Extrapolation of our data would predict a very high hemoglobin, which might not be achievable without serious hyperviscosity.10 However, any patient with an O2sat 70% is arguably not in a state of balanced erythropoiesis, because this must reect increased tissue oxygen extraction. The greatest hemoglobin in our optimal group was 25 g/dl (packed cell volume 73%) in a stable patient. We previously reported no adverse effects of viscosity on exercise capacity in this population.1 Hyperviscosity and its related symptoms are likely far more complex than ex vivo methods can measure and very different at the capillary level, in particular.11 Daily activity will vary from patient to patient and will also affect the drive to erythropoiesis. The presence of an optimal hemoglobin does not mean

the patient is asymptomatic, because many factors will contribute to symptoms in this group. We did not address whether the manipulation of hemoglobin levels to an optimal level as we have dened would have any effect on symptoms, exercise capacity, or prognosis, although we, and others, have shown improvement after treatment of iron deciency in patients with chronic cyanosis.12,13 Exercise capacity also has multiple determinants. The purpose of comparing the exercise data in the present study was solely to establish whether the predictive formula had any functional relevance. Because at least a component of exertional capacity correlated with hemoglobin difference but not hemoglobin concentration itself (Figure 3), the relation we dened seems to have clinical signicance. We know of no other method of validating our results. Other limitations deserve comment. As an initial exploration of this association, we had no guidance on which factors would be most important, and not all were important. Gender differences were not considered, because our study did not have a large enough sample size to justify a separate analysis of men versus women. We did not study patients with Fontan physiology, although often such patients are cyanotic. We have no reason to suspect this

Congenital Heart Disease/Optimal Hemoglobin in Cyanosis

599

relation would not be relevant to this group also, and this deserves additional investigation. However, we do not think our prediction formula should be applied to other cyanotic conditions, such as lung disease, nor to children with congenital heart defects.
1. Broberg CS, Bax BE, Okonko DO, Rampling MW, Bayne S, Harries C, Davidson SJ, Uebing A, Khan AA, Thein S, Gibbs JS, Burman J, Gatzoulis MA. Blood viscosity and its relation to iron deciency, symptoms, and exercise capacity in adults with cyanotic congenital heart disease. J Am Coll Cardiol 2006;48:356 365. 2. Broberg CS, Ujita M, Prasad S, Li W, Rubens M, Bax BE, Davidson SJ, Bouzas B, Gibbs JS, Burman J, Gatzoulis MA. Pulmonary arterial thrombosis in Eisenmenger syndrome is associated with biventricular dysfunction and decreased pulmonary ow velocity. J Am Coll Cardiol 2007;50:634 642. 3. Tyndall MR, Teitel DF, Lutin WA, Clemons GK, Dallman PR. Serum erythropoietin levels in patients with congenital heart disease. J Pediatr 1987;110:538 544. 4. Crowell JW, Smith EE. Determinant of the optimal hematocrit. J Appl Physiol 1967;22:501504. 5. Berman W Jr, Wood SC, Yabek SM, Dillon T, Fripp RR, Burstein R. Systemic oxygen transport in patients with congenital heart disease. Circulation 1987;75:360 368. 6. Diller GP, Dimopoulos K, Broberg CS, Kaya MG, Naghotra US, Uebing A, Harries C, Goktekin O, Gibbs JS, Gatzoulis MA. Presen-

7. 8.

9.

10.

11. 12.

13.

tation, survival prospects, and predictors of death in Eisenmenger syndrome: a combined retrospective and case-control study. Eur Heart J 2006;27:17371742. Gidding SS, Stockman JA III. Erythropoietin in cyanotic heart disease. Am Heart J 1988;116:128 132. Gidding SS, Bessel M, Liao YL. Determinants of hemoglobin concentration in cyanotic heart disease. Pediatr Cardiol 1990;11:121 125. Gidding SS, Stockman JA III. Effect of iron deciency on tissue oxygen delivery in cyanotic congenital heart disease. Am J Cardiol 1988;61:605 607. Rosove MH, Perloff JK, Hocking WG, Child JS, Canobbio MM, Skorton DJ. Chronic hypoxaemia and decompensated erythrocytosis in cyanotic congenital heart disease. Lancet 1986;2:313315. Kontras SB, Bodenbender JG, Craenen J, Hosier DM. Hyperviscosity in congenital heart disease. J Pediatr 1970;76:214 220. Gaiha M, Sethi HP, Sudha R, Arora R, Acharya NR. A clinicohematological study of iron deciency anemia and its correlation with hyperviscosity symptoms in cyanotic congenital heart disease. Indian Heart J 1993;45:5355. Tay EL, Peset A, Papaphylactou M, Inuzuka R, Alonso-Gonzalez R, Giannakoulas G, Tzifa A, Goletto S, Broberg C, Dimopoulos K, Gatzoulis MA. Replacement therapy for iron deciency improves exercise capacity and quality of life in patients with cyanotic congenital heart disease and/or the Eisenmenger syndrome. Int J Cardiol 2010 [Epub ahead of print].

Relation of Pulse Pressure to Blood Pressure Response to Exercise in Patients With Hypertrophic Cardiomyopathy
Kevin S. Heffernan, PhD*, Martin S. Maron, MD, Eshan A. Patvardhan, MBBS, Richard H. Karas, MD, PhD, and Jeffrey T. Kuvin, MD; the Vascular Function Study Group
Almost one third of patients with hypertrophic cardiomyopathy (HC) will have an abnormal blood pressure response (ABPR) to exercise, and this has been associated with a greater risk of sudden cardiac death. In the present study, we examined the association between the steady (mean arterial pressure) and pulsatile (pulse pressure) blood pressure components as they relate to ABPR in patients with HC (n 70). All patients completed a standard Bruce protocol during symptom-limited stress testing with concurrent hemodynamic measurements. Pulse pressure (PP) was signicantly greater in patients with HC with an ABPR (n 19) than in the patients with HC without an ABPR to exercise (p <0.05). According to binary logistic regression analysis, PP at rest was a signicant predictor of ABPR in patients with HC (p <0.05). Mean arterial pressure was not signicantly different between the 2 groups, nor was it a predictor of an ABPR in the presence of HC. Those within the greatest tertile of PP at rest were 4.8 times more likely to have an ABPR than those within the lowest PP tertile (95% condence interval 1.24 to 18.2, p <0.05). In conclusion, elevations in PP at rest might identify patients with HC at a greater risk of having an ABPR during exercise. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107: 600 603) Approximately 30% of patients with hypertrophic cardiomyopathy (HC) will have an abnormal blood pressure response (ABPR) to exercise, categorized as a failure to increase (or a potential decrease) in systolic blood pressure with an increase in exercise intensity. This ABPR has been associated with sudden cardiac death in patients with HC.1 4 Blood pressure (BP) has both pulsatile and steady components. The pulsatile component of BP, estimated by pulse pressure (PP), reects the integration of left ventricular (LV) systolic function, large artery stiffness, forward pulse wave genesis, and pulse wave reection. Arterial stiffness is an important determinant of the net cardiovascular performance and cardiac energetics at rest and during exercise.5 As such, increased PP, a manifestation of altered ventricularvascular coupling and increased pulsatile afterload, might be related to the ABPR in patients with HC, but this has yet to be examined. The purpose of the present investigation was to test the hypothesis that PP at rest would be associated with an ABPR during exercise in patients with HC. Methods A total of 70 patients with HC were recruited from the HC Center at Tufts Medical Center. The diagnosis of HC was determined using the criteria put forth by the American College of Cardiology/European Society of Cardiology clinical expert consensus document on HC. All patients had LV hypertrophy (wall thickness 15 mm according to echocardiographic demonstration) associated with a nondilated cavity in the absence of another cardiac or systemic disease that could produce this magnitude of hypertrophy. The exclusion criteria included severe valvular disease, recent myocardial infarction or unstable cardiac symptoms, peripheral vascular disease, heart failure, end-stage disease with systolic dysfunction or LV ejection fraction 40%, severe arrhythmia, chronic obstructive pulmonary disease, recent exertional syncope, previous septal myectomy, alcohol septal ablation, and coexistent aortic stenosis. Coronary artery disease was dened as the presence of ischemia or infarction on single-photon emission computed tomographic nuclear myocardial perfusion imaging or 50% stenosis of an epicardial coronary artery by angiography. The presence or absence of hypertension (systolic BP/ diastolic BP 140/90 mm Hg or taking antihypertensive medication) and clinical symptoms were obtained for each patient from a questionnaire or the medical records. All subjects gave written informed consent, and the institutional review board at Tufts Medical Center approved the present study. The patients underwent standard 2-dimensional transthoracic echocardiography for assessment of the cardiac dimensions, followed by a symptom-limited exercise test using a standard Bruce protocol with a concurrent hemodynamic assessment. BP was measured using auscultation and sphygmomanometry. The measures at rest were made with the patients with HC in the supine position after a brief quiet rest period. BP was measured thereafter at the end of the exercise stage. An ABPR was dened as a reduction in systolic BP during exercise relative to systolic BP at rest or an inability to increase systolic BP 20 mm Hg during exercise. PP was calculated as
www.ajconline.org

Hypertrophic Cardiomyopathy Center, Division of Cardiology, and Molecular Cardiology Research Institute, Tufts Medical Center, Boston, Massachusetts. Manuscript received August 13, 2010; manuscript received and accepted October 7, 2010. *Corresponding author: Tel: (217) 621-8900; fax: (617) 636-0223. E-mail address: KHeffernan@TuftsMedicalCenter.org (K.S. Heffernan). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.023

Cardiomyopathy/Blood Pressure and Exercise in HC Table 1 Patient characteristics according to blood pressure (BP) response Variable All Yes Age (years) Women Maximum left ventricular thickness (mm) Left ventricular enddiastolic diameter (mm) Left ventricular endsystolic diameter (mm) Left atrial size (mm) Systolic anterior motion (scale 04) Mitral regurgitation (scale 04) Left ventricular outow tract obstruction New York Heart Association class I II III Medications Blocker Calcium channel blocker Diuretic Angiotensin-converting enzyme inhibitor Antiarrhythmic Family history hypertrophic cardiomyopathy History of chest pain History of syncope Coronary artery disease Heart rate at rest (beats/ min) Systolic blood pressure at rest (mm Hg) Diastolic blood pressure at rest (mm Hg) Mean arterial pressure at rest (mm Hg) Pulse pressure at rest (mm Hg) 45 2 23 (33%) 19.6 0.6 43.2 0.8 25.0 0.7 39.8 0.9 1 1 33 (47%) 54 5 7 (37%) 20.0 1.1 41.3 1.3 25.2 1.4 40.5 1.5 1 1 10 (52%) ABPR No 41 2* 16 (31%) 19.5 0.6 43.9 1.0 25.0 0.9 39.6 1.0 1 1 23 (45%)

601

bution was assessed using the Kolmogorov-Smirnof and Shapiro-Wilk tests. Chi-square tests were used to compare the categorical variables. Patients with and without an ABPR were compared using analysis of variance for normally distributed variables and the Mann-Whitney U test for non-normally distributed variables. If the demographic variables differed between the 2 groups, analysis of covariance was used to adjust for the group differences. The patients were then separated into tertiles according to PP, and binary logistic regression analysis was used to examine the predictors of ABPR (entered as a discrete variable). Results The baseline demographics are listed in Table 1. Of the 70 patients, 19 had an ABPR (27%). Signicant group differences were found in age (p 0.004) and diuretic use (p 0.014) between those with and without an ABPR to exercise (Table 1). PP at rest was signicantly greater in the patients with HC and an ABPR than in those patients with a normal BP response to exercise (p 0.007; Table 1). The differences in PP remained after adjusting for age and diuretic use with analysis of covariance (adjusted mean 63 mm Hg vs 52 mm Hg, p 0.028). The differences in PP also remained after adjusting for medication use (adjusted for blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, Norpace, and other antiarrhythmic agents 64 mm Hg vs 52 mm Hg, p 0.008). Mean arterial pressure was not signicantly different between the 2 groups (p 0.738; Table 1). The prevalence of ABPR was not different in the patients with HC with versus without LV outow tract obstruction (30% vs 24%, p 0.601). PP was not different in the patients with HC who did and did not have LV outow tract obstruction (56 vs 54 mm Hg, p 0.612). No gender differences were found in the prevalence of ABPR (men, 26% vs women, 30%, p 0.776). No gender differences were found in PP (men, 56 3 mm Hg vs women, 51 3, p 0.264). When separating the patients into tertiles according to PP, the prevalence of ABPR was signicantly greater for the patients with HC with the greatest PP compared to those in the rst (reference group) and second tertile (p 0.008). Of the patients in the greatest tertile (60 mm Hg), 50% had an ABPR compared to 13% and 17% in the second (range 45 to 60 mm Hg) and third (45 mm Hg) tertiles, respectively. According to binary logistic regression analysis, after adjusting for potential confounders (age, gender, LV wall thickness, left atrial size, anterior basal septal wall thickness, history of chest pain, history of syncope, family history of HC, LV outow tract obstruction, coronary artery disease), PP at rest was a signicant predictor of the ABPR in patients with HC (p 0.016). The patients with HC and the greatest PP at rest were 4.8 times more likely to have an ABPR than those with the lowest PP ( 1.6, Wald 5.2, 95% condence interval 1.24 to 18.2, p 0.023). Systolic BP, diastolic BP, and mean arterial pressure were not signicant predictors of ABPR.

41 (59%) 16 (23%) 13 (19%) 41 (59%) 26 (37%) 7 (10%) 12 (17%) 7 (10%) 26 (37%)

10 (52%) 3 (16%) 6 (32%) 14 (74%) 7 (37%) 5 (26%) 3 (16%) 2 (11%) 9 (47%)

31 (61%) 13 (25%) 7 (14%) 27 (53%) 19 (37%) 2 (4%)* 9 (18%) 5 (10%) 17 (33%)

22 (31%) 12 (17%) 17 (24%) 74 2 126 2 70 2 88 2 55 2

6 (32%) 4 (21%) 6 (32%) 72 4 131 5 67 2 89.0 3 64 5

16 (31%) 8 (16%) 11 (22%) 75 2 125 2 71 2 88 2 52 2*

Data are presented as mean SEM or n (%). * Signicant group difference (p 0.05).

systolic BP minus diastolic BP. The patients were instructed to withhold all cardiovascular medications for 24 to 72 hours before exercise testing. The presence of LV outow tract obstruction was assessed as previously described at rest, with Valsalva maneuver, and during exercise.6 LV outow tract obstruction was dened as a peak instantaneous outow gradient of 30 mm Hg using continuous-wave Doppler echocardiography.6 Systolic anterior motion and mitral regurgitation were assessed semiquantitatively (scale 0 to 4), as previously described.6 All data are reported as the mean SEM. A priori signicance was set at p 0.05. The normality of distri-

602

The American Journal of Cardiology (www.ajconline.org)

Discussion The novel ndings of the present study were as follows. First, patients with HC and an ABPR to exercise had a signicantly greater PP at rest compared to those with a normal BP response to exercise. Second, PP at rest was a signicant predictor of ABPR during exercise in patients with HC. Our ndings suggest that PP at rest, but not the traditionally assessed BP components of cardiovascular risk assessment (i.e., systolic BP and diastolic BP) or the steady component of BP (i.e., mean arterial pressure), is a predictor of the hemodynamic response to exercise in patients with HC. Exercise testing is an integral part of the algorithm in risk stratication and the delivery of prophylactic therapy for HC.7 An ABPR to exercise is predictive of sudden cardiac death in HC, and sudden cardiac death is the most common cause of death for patients with HC.8 In the present study, the patients with HC with a high PP at rest were almost 5 times more likely to have an ABPR to exercise. In other populations, an elevated PP has been associated with LV hypertrophy,9 impaired ventricular relaxation,10 ischemia during exercise,11 heart failure,12 increased left atrial size,13 and atrial brillation,14 all clinically relevant facets of HC pathologic features. Therefore, patients with HC and an elevated PP at rest might be a particularly high-risk cohort. Given the correlation between ABPR and the risk of sudden cardiac death in patients with HC, an elevated PP could help identify high-risk patients within the HC cohort. The underlying pathophysiology that causes an ABPR in patients with HC appears to be multifaceted. The proposed mechanisms include LV systolic dysfunction15 with subsequent subendocardial myocardial ischemia,16,17 altered baroreex-mediated modulation of autonomic outow to the heart and vasculature (possibly related to the aforementioned subendocardial myocardial ischemia),18 and subsequent altered vascular response to exercise (i.e., excessive decrease in peripheral/systemic vascular resistance).19 It has also been suggested that a lower exercise capacity and an ABPR in patients with HC might be related to diastolic dysfunction20 and a blunted augmentation of stroke volume during exercise.21 Although seemingly paradoxical, these mechanisms might have a single underlying and unifying etiology related to arterial stiffness. Patients with HC will have greater arterial stiffness than controls,22 and arterial stiffness has been shown to be related to reduced exercise capacity in patients with HC.23 Our ndings have expanded on previous work by noting that PP, a manifestation of the stiffening of the conduit vessels coupled with augmented pressure from wave reections, is associated with an ABPR in those with HC. LV ejection of the stroke volume into a stiff aorta, coupled with an early return of reected pressure waves, increases PP and cardiac energetic demand, reduces myocardial oxygen supply/consumption, reduces subendocardial perfusion,24 impairs cardiac systolic and diastolic function, and blunts stroke volume genesis.25 Thus, increased arterial stiffness and pulsatile afterload might offer insight into the ndings of LV systolic/diastolic dysfunction, reduced stroke volume/cardiac output, reduced myocardial oxygen consumption, and subsequent subendocardial ischemia, contributing to an ABPR in those with HC.

Finally, a strong relation exists between arterial stiffness and integrated neural control of BP within the baroreex arc. Stiffening of the vessels housing the barosensory regions might depress mechanotransduction, resulting in a reduction of baroreceptor afferent ring per given unit of arterial pressure change, less inhibition of sympathetic outow (altering peripheral vascular tone), and lessened amplication of cardiac vagal tone (altering LV contractility).26 The patients with HC with hemodynamic instability during lower body negative pressure, a HC cohort with a greater prevalence of an ABPR, have a lower resting baroreex sensitivity and exaggerated changes in baroreex sensitivity during lower body negative pressure.19 Thus, increased arterial stiffness in HC could also alter baroreex modulation of the cardiac and vascular autonomic control, contributing to an ABPR. Similar to previous observations, the prevalence of ABPR was not inuenced by LV outow tract obstruction.1,2 Previous studies have noted that LV outow tract obstruction does not affect peripheral vascular endothelial function or vascular stiffness in those with HC.22,27 As such, LV outow tract obstruction did not inuence PP in the present study. We noted no gender difference in the ABPR in those with HC, and this too was consistent with previous reports.1 The lack of effect of gender on the ABPR might have been because men and women within the age range studied tend to have a similar brachial PP.28 Thus, LV outow tract obstruction and gender did not modulate PP in those with HC and, as such, might not be signicant determinants of an ABPR during exercise in those with HC. Few therapeutic options are available to attenuate the hypotensive BP response to exercise in patients with HC. Thaman et al19 previously reported that neither propranolol nor clonidine signicantly modulated the hypotensive BP response to exercise in those with HC. However, the selective serotonin reuptake inhibitor paroxetine was able to normalize the BP response to exercise.19 Although the mechanism for this interaction remains unknown, paroxetine might have a favorable effect on vascular function.29,30 This raises the intriguing possibility that therapies that improve vascular function in the presence of HC might favorably affect the BP response to exercise. The limitations to the present study included the lack of information on clinical end points and examination of a single surrogate marker (i.e., ABPR). Whether patients with HC and an elevated PP are truly a high-risk cohort remains to be elucidated empirically. We used PP as an indirect measure of arterial stiffness. Additional research is needed to corroborate the present ndings using more valid bioassays of arterial stiffness, such as pulse wave velocity.
1. Frenneaux MP, Counihan PJ, Caforio AL, Chikamori T, McKenna WJ. Abnormal blood pressure response during exercise in hypertrophic cardiomyopathy. Circulation 1990;82:19952002. 2. Olivotto I, Maron BJ, Montereggi A, Mazzuoli F, Dolara A, Cecchi F. Prognostic value of systemic blood pressure response during exercise in a community-based patient population with hypertrophic cardiomyopathy. J Am Coll Cardiol 1999;33:2044 2051. 3. Sadoul N, Prasad K, Elliott PM, Bannerjee S, Frenneaux MP, McKenna WJ. Prospective prognostic assessment of blood pressure response during exercise in patients with hypertrophic cardiomyopathy. Circulation 1997;96:29872991.

Cardiomyopathy/Blood Pressure and Exercise in HC 4. Ciampi Q, Betocchi S, Lombardi R, Manganelli F, Storto G, Losi MA, Pezzella E, Finizio F, Cuocolo A, Chiariello M. Hemodynamic determinants of exercise-induced abnormal blood pressure response in hypertrophic cardiomyopathy. J Am Coll Cardiol 2002;40:278 284. 5. Kelly RP, Tunin R, Kass DA. Effect of reduced aortic compliance on cardiac efciency and contractile function of in situ canine left ventricle. Circ Res 1992;71:490 502. 6. Maron MS, Olivotto I, Zenovich AG, Link MS, Pandian NG, Kuvin JT, Nistri S, Cecchi F, Udelson JE, Maron BJ. Hypertrophic cardiomyopathy is predominantly a disease of left ventricular outow tract obstruction. Circulation 2006;114:22322239. 7. Sharma S, Firoozi S, McKenna WJ. Value of exercise testing in assessing clinical state and prognosis in hypertrophic cardiomyopathy. Cardiol Rev 2001;9:70 76. 8. Maron BJ. Hypertrophic cardiomyopathy: a systematic review. JAMA 2002;287:1308 1320. 9. Gardin JM, Arnold A, Gottdiener JS, Wong ND, Fried LP, Klopfenstein HS, OLeary DH, Tracy R, Kronmal R. Left ventricular mass in the elderly: the Cardiovascular Health Study. Hypertension 1997;29: 10951103. 10. Leite-Moreira AF, Correia-Pinto J, Gillebert TC. Afterload induced changes in myocardial relaxation: a mechanism for diastolic dysfunction. Cardiovasc Res 1999;43:344 353. 11. Kingwell BA, Waddell TK, Medley TL, Cameron JD, Dart AM. Large artery stiffness predicts ischemic threshold in patients with coronary artery disease. J Am Coll Cardiol 2002;40:773779. 12. Haider AW, Larson MG, Franklin SS, Levy D. Systolic blood pressure, diastolic blood pressure, and pulse pressure as predictors of risk for congestive heart failure in the Framingham Heart Study. Ann Intern Med 2003;138:10 16. 13. Vaziri SM, Larson MG, Lauer MS, Benjamin EJ, Levy D. Inuence of blood pressure on left atrial size: the Framingham Heart Study. Hypertension 1995;25:11551160. 14. Mitchell GF, Vasan RS, Keyes MJ, Parise H, Wang TJ, Larson MG, DAgostino RB Sr, Kannel WB, Levy D, Benjamin EJ. Pulse pressure and risk of new-onset atrial brillation. JAMA 2007;297:709 715. 15. Okeie K, Shimizu M, Yoshio H, Ino H, Yamaguchi M, Matsuyama T, Yasuda T, Taki J, Mabuchi H. Left ventricular systolic dysfunction during exercise and dobutamine stress in patients with hypertrophic cardiomyopathy. J Am Coll Cardiol 2000;36:856 863. 16. Ciampi Q, Betocchi S, Losi MA, Ferro A, Cuocolo A, Lombardi R, Villari B, Chiariello M. Abnormal blood-pressure response to exercise and oxygen consumption in patients with hypertrophic cardiomyopathy. J Nucl Cardiol 2007;14:869 875. 17. Yoshida N, Ikeda H, Wada T, Matsumoto A, Maki S, Muro A, Shibata A, Imaizumi T. Exercise-induced abnormal blood pressure responses are related to subendocardial ischemia in hypertrophic cardiomyopathy. J Am Coll Cardiol 1998;32:1938 1942. 18. Kawasaki T, Azuma A, Kuribayashi T, Akakabe Y, Yamano M, Miki S, Sawada T, Kamitani T, Matsubara H, Sugihara H. Vagal enhancement

603

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

due to subendocardial ischemia as a cause of abnormal blood pressure response in hypertrophic cardiomyopathy. Int J Cardiol 2008;129:59 64. Thaman R, Elliott PM, Shah JS, Mist B, Williams L, Murphy RT, McKenna WJ, Frenneaux MP. Reversal of inappropriate peripheral vascular responses in hypertrophic cardiomyopathy. J Am Coll Cardiol 2005;46:883 892. Matsumura Y, Elliott PM, Virdee MS, Sorajja P, Doi Y, McKenna WJ. Left ventricular diastolic function assessed using Doppler tissue imaging in patients with hypertrophic cardiomyopathy: relation to symptoms and exercise capacity. Heart 2002;87:247251. Nagata M, Shimizu M, Ino H, Yamaguchi M, Hayashi K, Taki J, Mabuchi H. Hemodynamic changes and prognosis in patients with hypertrophic cardiomyopathy and abnormal blood pressure responses during exercise. Clin Cardiol 2003;26:7176. Boonyasirinant T, Rajiah P, Setser RM, Lieber ML, Lever HM, Desai MY, Flamm SD. Aortic stiffness is increased in hypertrophic cardiomyopathy with myocardial brosis: novel insights in vascular function from magnetic resonance imaging. J Am Coll Cardiol 2009;54:255 262. Austin BA, Popovic ZB, Kwon DH, Thamilarasan M, Boonyasirinant T, Flamm SD, Lever HM, Desai MY. Aortic stiffness independently predicts exercise capacity in hypertrophic cardiomyopathy: a multimodality imaging study. Heart 2010;96:13031310. Saeki A, Recchia F, Kass DA. Systolic ow augmentation in hearts ejecting into a model of stiff aging vasculature: inuence on myocardial perfusion-demand balance. Circ Res 1995;76:132141. Borlaug BA, Melenovsky V, Redeld MM, Kessler K, Chang HJ, Abraham TP, Kass DA. Impact of arterial load and loading sequence on left ventricular tissue velocities in humans. J Am Coll Cardiol 2007;50:1570 1577. Casadei B, Meyer TE, Coats AJ, Conway J, Sleight P. Baroreex control of stroke volume in man: an effect mediated by the vagus. J Physiol 1992;448:539 550. Heffernan KS, Napolitano CA, Maron MS, Patvardhan EA, Patel AR, Pandian NG, Karas RH, Kuvin JT. Peripheral vascular endothelial function in patients with hypertrophic cardiomyopathy. Am J Cardiol 2010 1;105:112115. Mitchell GF, Parise H, Benjamin EJ, Larson MG, Keyes MJ, Vita JA, Vasan RS, Levy D. Changes in arterial stiffness and wave reection with advancing age in healthy men and women: the Framingham Heart Study. Hypertension 2004;43:1239 1245. Lara N, Archer SL, Baker GB, Le Melledo JM. Paroxetine-induced increase in metabolic end products of nitric oxide. J Clin Psychopharmacol 2003;23:641 645. Chrapko W, Jurasz P, Radomski MW, Archer SL, Newman SC, Baker G, Lara N, Le Melledo JM. Alteration of decreased plasma NO metabolites and platelet NO synthase activity by paroxetine in depressed patients. Neuropsychopharmacology 2006;31:1286 1293.

Clinical Challenges of Genotype Positive ()Phenotype Negative () Family Members in Hypertrophic Cardiomyopathy
Barry J. Maron, MDa, Laura Yeates, BScb, and Christopher Semsarian, MB, BS, PhDb,c,d,*
Hypertrophic cardiomyopathy (HC), a common genetic heart disorder associated with substantial clinical and genetic heterogeneity, is the most frequent cause of sudden death in the young (including competitive athletes).13 HC is generally characterized by unexplained left ventricular (LV) hypertrophy,4 6 although with the aid of molecular diagnosis, it has become evident that disease-causing mutations can be associated with virtually any LV wall thickness.79 Indeed, many genetically affected children (and even some adults) in HC families do not demonstrate LV hypertrophy at some point in their clinical courses, with incomplete penetrance of the phenotype.8 12 Laboratory investigations over the past 2 decades have dened HC as a primary myocardial disease caused by 1,000 mutations in 13 genes encoding proteins within and associated with the sarcomere.1,2,8,9,13 This has led to an increasing recognition of a novel patient subset within the vast and ever expanding HC disease spectrum: genetically affected family members without clinical or morphologic evidence of the disease.10,14 16 Such patients are usually referred to as preclinical or genotype-positive ()phenotype-negative () (G P), and they present the paradox of a rapidly evolving new patient subgroup that requires a long period of follow-up to develop clear guidelines with regard to management. Denition of Genotype-Positive ()PhenotypeNegative () G P patients carry mutations in genes encoding proteins of the cardiac sarcomere, judged (or known) to be disease causing for HC.5,10,14,15 Such patients are usually asymptomatic, often with 12-lead electrocardiographic abnormalities but no evidence of the HC phenotype (i.e., LV hypertrophy) on 2-dimensional echocardiography and cardiovascular magnetic resonance (CMR) imaging.5,10,14,15 CMR is emerging as a highly relevant imaging modality for the identication of the HC phenotype, because of its tomographic highspatial resolution characteristics. CMR is not encumbered by certain well-recognized limitations of
Hypertrophic Cardiomyopathy Center, Minneapolis Heart Institute Foundation, Minneapolis, Minnesota; bAgnes Ginges Centre for Molecular Cardiology, Centenary Institute, Newtown, Australia; cSydney Medical School, University of Sydney, Sydney, Australia; and dDepartment of Cardiology, Royal Prince Alfred Hospital, Camperdown, Australia. Manuscript received August 23, 2010; revised manuscript received and accepted October 1, 2010. Dr. Semsarian is the recipient of a Practitioner Fellowship from the National Health and Medical Research Council, Canberra, Australia. This study was also supported in part by the Hearst Foundations, San Francisco, California (Dr. Maron). *Corresponding author: Tel: 61-2-9565-6195; fax: 61-2-9565-6101. E-mail address: c.semsarian@centenary.org.au (C. Semsarian). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.022
a

echocardiography with respect to measurements of LV wall thickness, justifying its inclusion in the assessment of G P patients.4,17 For example, CMR can provide LV wall thickness measurements with greater precision, particularly relevant to hypertrophy in the borderline zone of 12 to 15 mm. Also, in selected patients, CMR may identify segmental regions of LV hypertrophy in the anterolateral LV free wall (or apex), not reliably detected or often underestimated in magnitude by echocardiography.4,16,17 It has not been our practice to dene the HC phenotype solely by abnormalities on 12-lead electrocardiography.4,6,8 10 This consideration is due to certain predictable limitations of electrocardiography as a screening test for the clinical HC spectrum18,19: (1) difculty in establishing absolute and strict partitions for normality at all ages and body sizes (particularly in children), (2) potential confusion created by nonspecic alterations unrelated to cardiovascular disease, (3) unpredictable variability in electrocardiographic patterns over time, (4) the documented weak relation between electrocardiographic voltages and LV wall thickness, and (5) the occurrence of normal results on electrocardiography in up to 25% of phenotypically expressed HC and in about 50% of G P family members.12,16 18 In contrast, 2-dimensional echocardiography and CMR provide reproducible, quantitative measures of LV wall thickness for comparison to established normal values.4,6,9,17 On the basis of these considerations, it appears most appropriate to dene the HC phenotype with respect to LV hypertrophy (as identied directly by contemporary imaging), while an indirect measure with 12-lead electrocardiography would likely be associated with a large number of false-positive test results. Family Studies In Figure 1, we present 4 families demonstrating significant challenges that arise in G P family members, which are cornerstones of HC clinical decision making: eligibility versus disqualication from intense competitive sports20 23 and the prevention of sudden death with prophylactic implantable cardioverter-debrillators (ICDs).24 27 Family A highlights the issue of sports eligibility in G P children. Two young female children aged 10 and 12 years (III:3 and III:4) are both elite gymnasts. They showed no evidence of HC on electrocardiography and echocardiography, but both carry the causative gene mutation, Arg810His in the MYBPC3 gene, also identied in other clinically affected individuals in the family (II:1 and II:3). Should these children (III:3 and III:4) be removed from competitive sports? Family B targets the issue of whether decisions regarding sports eligibility in G P patients should be further inuenced by a family history of a sudden death event. This
www.ajconline.org

Editorial/Genotype-PositivePhenotype-Negative HC Patients

605

Figure 1. Pedigrees of 4 HC families. In family D, IV:1 had an appropriate ICD shock at 16 years of age. Numbers in parentheses indicate age in years. Shaded gray symbols with arrows G P subjects; squares men; circles women; solid black symbols clinically affected subjects with HC phenotype (i.e., LV hypertrophy); clear symbols without cardiac evaluation; symbols transected by lines HC-related cardiac arrest and death. ES end-stage HC-related heart failure; N normal on clinical screening with electrocardiography and imaging studies; / heterozygote for mutation (genotype positive); RCA resuscitated cardiac arrest; SCD sudden cardiac death; Tx heart transplantation.

family presented for evaluation after a resuscitated cardiac arrest occurred in an 11-year-old female patient (III:2). With subsequent clinical screening and genetic testing, the causative gene mutation, Arg495Gln in the MYBPC3 gene, was identied in each clinically affected family member, as well as the 9-year-old sibling (III:3), with no evidence of the disease phenotype. Should this 9-year-old G P brother be excluded from future involvement in competitive sports, and furthermore, should an ICD for primary prevention even be considered in a patient of this age? Family C explores and extends the potential role for ICDs in young G P patients with family histories of sudden death. In the rst generation, a 30-year-old man (I:1) died suddenly while jogging, but without a conrmed HC diagnosis. All 3 adult children are clinically affected, and 2 have elected prophylactic ICDs (II:1 and II:4). Genetic screening identied the Gly733Glu mutation in the MYH7 gene in each of the 3 siblings, as well as the 14-year-old grandson of the proband, with no evidence of LV hypertrophy but extensive involvement in competitive athletics (III: 2). Should this 14-year-old be withdrawn from sports and/or considered for an ICD on the basis of the sudden death event occurring decades earlier in his grandfather?

Family D illustrates the dilemma for a G P relative (III:2), who carries the Lys97Asn mutation in the TNNT2 gene. At 38 years of age, she is the sole survivor affected by HC in a family with malignant outcomes, including 4 sudden death events (III:3, III:4, III:5, and IV:1) and 2 deaths due to end-stage progression (II:1 and II:3). Should this adult relative elect a prophylactic ICD, despite absence of the HC phenotype? Commentary These HC families underscore emerging dilemmas in clinical HC practice, that is, whether the management of sudden death risk in this new subset of genetically affected relatives without clinical evidence of disease (i.e., G P) should be similar to more typical patients with HC with LV hypertrophy. As seen in families B and C, the issue of disqualication from competitive sports participation and the advisability of prophylactic ICDs for G P members of HC families are often interwoven, making many of these clinical decisions particularly complex and challenging. Sports Eligibility: Family A raises the question of whether all G P patients with HC should be disqualied

606

The American Journal of Cardiology (www.ajconline.org)

from intense competitive sports, an obvious consideration given that HC is the most common cause of sudden death in young athletes.3,20 Also, intense competitive sports increase the likelihood of these catastrophic events,20 23 and withdrawal from this vigorous lifestyle may well reduce this risk.22,23,27 Nevertheless, this issue remains unsettled. Notably, the 2 available consensus expert panel documents present diametrically opposed recommendations, with both unavoidably based on clinical inferences with little or no hard evidence.2123 For example, the European Society of Cardiology guidelines23 are particularly conservative, recommending disqualication from competitive sports for all gene carriers.2123 In contrast, the United States based Bethesda Conference 3622 permits G P patients to participate in all competitive sports until LV hypertrophy appears.2123 These divergent recommendations regarding the same clinical scenario cannot be resolved without longitudinal follow-up studies in this select subset. Families B and C raise the consideration of whether sports disqualication for G P athletes is even more relevant when there is a family history of HC-related sudden death in a close relative. The question of sports eligibility or disqualication in such subjects is increasing in frequency as more of these families pursue genetic testing in response to catastrophic events occurring in relatives. However, this particular clinical issue is not specically addressed in the aforementioned recommendations of either the United States22 or European23 consensus panel. A family history of sudden death in a close relative is an acknowledged risk factor in HC,8,9,24 26 although available data relate only to phenotype-positive patients with LV hypertrophy and clinically dened disease. While it is unresolved as to whether this risk marker can (or should) be extrapolated to genetically affected subjects without LV hypertrophy, it nevertheless seems most prudent to discourage young G P relatives with family histories of HC sudden death from engaging in intense competitive sports at an early age. ICDs For Primary Prevention: The question of whether G P patients should be considered for primary prevention ICDs because of a family history of sudden death arises most frequently in adults who are part of malignant families with multiple sudden deaths25,26 (such as family D). As demonstrated by families B and C, this treatment consideration may also arise with respect to G P children and adolescents. However, the considerable frequency of device-related complications in young patients over long follow-up periods of decades is often a mitigating factor for prophylactic implants in such G P relatives.24,25,27 Nevertheless, ICDs have proved effective in terminating life-threatening ventricular tachyarrhythmias in high-risk patients with HC with overt disease expression.24 27 However, it is largely unresolved as to whether non-hypertrophied LV muscle in patients with HC-causing mutations can constitute an electrically unstable substrate capable of potentially lethal sustained ventricular tachyarrhythmias. Therefore, prophylactic ICDs are most likely to be considered after LV hypertrophy has appeared, thereby justifying close clinical surveillance with echocardiography (and

CMR imaging, if available), probably at 12-month intervals, to identify changes in LV wall thickness. The overwhelming difculty surrounding this (and other) key questions related to management of G P HC family members is the paucity of available outcome data. The prevailing perception has been that sudden death risk in HC is virtually always linked to the presence of LV hypertrophy.8,9,24,25 Notably, however, 2 cases have been reported recently in 37-year-old and 43-year-old (nonathlete) patients with MHY7 mutations (but without clinical or phenotypic evidence of HC), who survived ventricular brillation.28 In addition, a few family members reported in the pre-genotyping era may represent similar sudden deaths in the absence of LV hypertrophy.11 Although rare, such isolated cases suggest the possibility that susceptible G P relatives can harbor arrhythmogenic substrates at a cellular and molecular level capable of triggering life-threatening ventricular tachyarrhythmias. This observation is consistent with other clinical ndings that support the notion that the nonhypertrophied LV myocardium in some G P relatives may be electrically or functionally abnormal, that is, with evidence of diastolic dysfunction,10,15 or abnormal 12-lead electrocardiographic patterns,29 as well as risk markers such as nonsustained ventricular tachycardia on ambulatory (Holter) electrocardiography, abnormal blood pressure response to exercise,30 delayed gadolinium enhancement.31 In contrast, we found no evidence of important arrhythmias on ambulatory (Holter) electrocardiographic monitoring in our G P patients, including in the 4 families reported here. Whether or not the occurrence of arrhythmic events would be enhanced by intense physical activity (such as competitive sports) is unresolved. Supportive evidence that lethal events are probably exceedingly rare in G P patients can be derived from those HC patients with only mild phenotypic expression who have generally favorable prognosis and low sudden death risk.8,9 However, it is certainly possible that the true prevalence of lethal HCrelated events in G P patients has been underestimated, given that at autopsy, these patients would have structurally normal hearts and probably not be assigned postmortem cardiac diagnoses. Conclusions and Clinical Decision Making Genetic diagnosis in HC has provided many answers but has also raised a number of important questions, including the emergence of G P family members, adding to the complexity of management in this highly heterogenous disease. Specically, given the paucity of clinical information, the relatively small numbers of G P family members identied, and the expected low HC event rate, it is likely that many years of follow-up will be necessary to acquire and formulate evidence-based insights for this HC clinical subset. For example, it is unknown what proportion of G P relatives will develop LV hypertrophy (or when), nor whether some gene carriers may achieve normal longevity without ever expressing the disease phenotype. Although morphologic conversion of the HC phenotype is most common during adolescence, it has also been documented occasionally in

Editorial/Genotype-PositivePhenotype-Negative HC Patients

607

midlife and beyond.12,32 The explanation for genetically affected patients remaining phenotype negative remains unclear but could be related to a number of factors, which (in addition to age) likely include secondary genetic factors that may inuence expression of the mutant gene (e.g., modier genes and epigenetic changes).1,2,12,32 Most important in this regard are the management considerations raised with respect to the G P subgroup. Is disqualication from intense competitive sports justied? When (or ever) should prophylactic ICDs be considered? These decisions cannot be deferred indenitely, because they are of immediate concern to patients and therefore must be resolved with the currently available information. Unavoidably, this process will often reect cultural differences and societal priorities, and reliance on the experiences and perceptions of the individual physicians and patients involved. At present, however, it is not possible to reach explicit and denitive judgments addressing these clinical questions with absolute authority in each G P patient. It is our practice that until clarity is achieved in this area, the most prudent strategy is to engage young G P patients and their families in a fundamental process predicated on the principles of full transparency, informed consent, and ultimately patient autonomy. Patients are provided with all relevant and pertinent information (or lack thereof). Specic options regarding competitive sports or prophylactic debrillator implantation are discussed in a detailed and balanced fashion, taking into account patient autonomy considerations i.e., the comfort level and desires of the fully informed patient and family in making decisions that involve the substantial ambiguity implicit when there are insufcient evidence-based data. We followed this approach in managing each of the G P family members shown in Figure 1, and this approach is also consistent with the process used with phenotypically expressed HC patients under consideration for primary prevention ICDs in the ambiguous gray zone in which sudden death risk level cannot be assessed with precision using the conventional markers, and individual clinical judgments on a case-by-case basis are necessary. In our 4 families, the G P subjects in family A chose to reduce their participation in high-level competitive sports, while those in families B to D have continued their sports participation. In addition, the G P individual in family D elected to have an ICD implanted for primary prevention. Finally, it is advisable to provide G P patients with imaging surveillance at regular intervals to determine when or if the HC phenotype develops and with ambulatory electrocardiographic monitoring for the potential detection of ventricular tachyarrhythmias. Acknowledgment: We acknowledge the contributions of Edwin Kirk and Anne Ronan from the state regional genetic services in New South Wales, Australia.
1. Alcalai R, Seidman JG, Seidman CE. Genetic basis of hypertrophic cardiomyopathy: from bench to the clinics. J Cardiovasc Electrophysiol 2008;19:104 110.

2. Seidman JG, Seidman C. The genetic basis for cardiomyopathy: from mutation identication to mechanistic paradigms. Cell 2001;104:557 567. 3. Maron BJ, Doerer JJ, Haas TS, Tierney DM, Mueller FO. Sudden deaths in young competitive athletes: analysis of 1866 deaths in the United States, 1980-2006. Circulation 2009;119:10851092. 4. Maron MS, Maron BJ, Harrigan C, Buros J, Gibson CM, Olivotto I, Biller L, Lesser JR, Udelson JE, Manning WJ, Appelbaum E. Hypertrophic cardiomyopathy phenotype revisited after 50 years with cardiovascular magnetic resonance. J Am Coll Cardiol 2009;54:220 228. 5. Maron BJ, Seidman CE, Ackerman MJ, Towbin JA, Maron MS, Ommen SR, Nishimura RA, Gersh BJ. Whats in a name? Dilemmas in nomenclature characterizing hypertrophic cardiomyopathy and left ventricular hypertrophy. Circ Cardiovasc Genet 2009;2:81 86. 6. Klues HG, Schiffers A, Maron BJ. Phenotypic spectrum and patterns of left ventricular hypertrophy in hypertrophic cardiomyopathy: morphologic observations and signicance as assessed by two-dimensional echocardiography in 600 patients. J Am Coll Cardiol 1995;26: 1699 1708. 7. Ingles J, Doolan A, Chiu C, Seidman J, Seidman C, Semsarian C. Compound and double mutations in patients with hypertrophic cardiomyopathy: implications for genetic testing and counselling. J Med Genet 2005;42:e59. 8. Maron BJ. Hypertrophic cardiomyopathy: a systematic review. JAMA 2002;287:1308 1320. 9. Maron BJ, McKenna WJ, Danielson GK, Kappenberger LJ, Kuhn HJ, Seidman CE, Shah PM, Spencer WH III, Spirito P, Ten Cate FJ, Wigle ED. American College of Cardiology/European Society of Cardiology clinical expert consensus document on hypertrophic cardiomyopathy. A report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents and the European Society of Cardiology Committee for Practice Guidelines. Eur Heart J 2003;24:19651991. 10. Maron BJ, Ho CY. Hypertrophic cardiomyopathy without hypertrophy: an emerging pre-clinical subgroup composed of genetically affected family members. JACC Cardiovasc Imaging 2009;2:65 68. 11. McKenna WJ, Stewart JT, Nihoyannopoulos P, McGinty F, Davies MJ. Hypertrophic cardiomyopathy without hypertrophy: two families with myocardial disarray in the absence of increased myocardial mass. Br Heart J 1990;63:287290. 12. Maron BJ, Niimura H, Casey SA, Soper MK, Wright GB, Seidman JG, Seidman CE. Development of left ventricular hypertrophy in adults with hypertrophic cardiomyopathy caused by cardiac myosin-binding protein C mutations. J Am Coll Cardiol 2001;38:315321. 13. Lind JM, Chiu C, Semsarian C. Genetic basis of hypertrophic cardiomyopathy. Expert Rev Cardiovasc Ther 2006;4:927934. 14. Kelly M, Semsarian C. Multiple mutations in genetic cardiovascular disease: a marker of disease severity? Circ Cardiovasc Genet 2009;2: 182190. 15. Ho CY, Sweitzer NK, McDonough B, Maron BJ, Casey SA, Seidman JG, Seidman CE, Solomon SD. Assessment of diastolic function with Doppler tissue imaging to predict genotype in preclinical hypertrophic cardiomyopathy. Circulation 2002;105:29922997. 16. Germans T, Wilde AAM, Dijkmans PA, Chai W, Kamp O, Pinto YM, van Rossum AC. Structural abnormalities of the inferoseptal left ventricular wall detected by cardiac magnetic resonance imaging in carriers of hypertrophic cardiomyopathy mutations. J Am Coll Cardiol 2006;48:2518 2523. 17. Maron MS, Lesser RJ, Maron BJ. Massive left ventricular hypertrophy in hypertrophic cardiomyopathy signicantly underestimated by echocardiography but identied by cardiovascular magnetic resonance: implications for management strategies. Am J Cardiol. In press. 18. Montgomery JV, Harris KM, Casey SA, Zenovich AG, Maron BJ. Relation of electrocardiographic patterns to phenotypic expression and clinical outcome in hypertrophic cardiomyopathy. Am J Cardiol 2005; 96:270 275. 19. McLeod CJ, Ackerman MJ, Nishimura RA, Tajik AJ, Gersh BJ, Ommen SR. Outcome of patients with hypertrophic cardiomyopathy and a normal electrocardiogram. J Am Coll Cardiol 2009;54:229 233. 20. Maron BJ. Sudden death in young athletes. N Engl J Med 2003;349: 1064 1075. 21. Pelliccia A, Zipes DP, Maron BJ. Bethesda Conference #36 and the European Society of Cardiology Consensus Recommendations revisited a comparison of U.S. and European criteria for eligibility and

608

The American Journal of Cardiology (www.ajconline.org) disqualication of competitive athletes with cardiovascular abnormalities. J Am Coll Cardiol 2008;52:1990 1996. Maron BJ, Zipes DP. 36th Bethesda Conference: eligibility recommendations for competitive athletes with cardiovascular abnormalities. J Am Coll Cardiol 2005;45:13121375. Pelliccia A, Fagard R, Bjornstad HH, Anastassakis A, Arbustini E, Assanelli D, Bif A, Borjesson M, Carr F, Corrado D, Delise P, Dorwarth U, Hirth A, Heidbuchel H, Hoffmann E, Mellwig KP, Panhuyzen-Goedkoop N, Pisani A, Solberg EE, van-Buuren F, Vanhees L, Blomstrom-Lundqvist C, Deligiannis A, Dugmore D, Glikson M, Hoff PI, Hoffmann A, Hoffmann E, Horstkotte D, Nordrehaug JE, Oudhof J, McKenna WJ, Penco M, Priori S, Reybrouck T, Senden J, Spataro A, Thiene G. Recommendations for competitive sports participation in athletes with cardiovascular disease: a consensus document from the Study Group of Sports Cardiology of the Working Group of Cardiac Rehabilitation and Exercise Physiology and the Working Group of Myocardial and Pericardial Diseases of the European Society of Cardiology. Eur Heart J 2005;26:14221445. Maron BJ, Spirito P, Shen W-K, Haas TS, Formisano F, Link MS, Epstein AE, Almquist AK, Daubert JP, Lawrenz T, Boriani G, Estes NA III, Favale S, Piccininno M, Winters SL, Santini M, Betocchi S, Arribas F, Sherrid MV, Buja G, Semsarian C, Bruzzi P. Implantable cardioverter-debrillators and prevention of sudden cardiac death in hypertrophic cardiomyopathy. JAMA 2007;298:405 412. Maron BJ. Contemporary insights and strategies for risk stratication and prevention of sudden death in hypertrophic cardiomyopathy. Circulation 2010;121:445 456. 26. Bos JM, Maron BJ, Ackerman MJ, Haas TS, Sorajja P. Nishimura RA, Gersh BJ, Ommen SR. Role of family history of sudden death in risk stratication and prevention of sudden death with implantable debrillators in hypertrophic cardiomyopathy. Am J Cardiol 2010;106: 14811486. 27. Maron BJ, Spirito P. Implantable debrillators and prevention of sudden death in hypertrophic cardiomyopathy. J Cardiovasc Electrophysiol 2008;19:1118 1126. 28. Christiaans I, Lekanne dit Deprez RH, van Langen IM, Wilde AA. Ventricular brillation in MYH7-related hypertrophic cardiomyopathy before onset of ventricular hypertrophy. Heart Rhythm 2009;6:1366 1369. 29. Panza JA, Maron BJ. Relation of electrocardiographic abnormalities to evolving left ventricular hypertrophy in hypertrophic cardiomyopathy during childhood. Am J Cardiol 1989;63:1258 1265. 30. Michels M, Soliman OI, Phefferkorn J, Hoedemaekers YM, Kofard MJ, Dooijes D, Majoor-Krakauer D, Ten Cate FJ. Disease penetrance and risk stratication for sudden cardiac death in asymptomatic hypertrophic cardiomyopathy mutation carriers. Eur Heart J 2009;30: 25932598. 31. Strijack B, Ariyarajah V, Soni R, Jassal DS, Greenberg CR, McGregor R, Morris M. Late gadolinium enhancement cardiovascular magnetic resonance in genotyped hypertrophic cardiomyopathy with normal phenotype. J Cardiovasc Magn Reson 2008;10:58. 32. Maron BJ, Seidman JG, Seidman CE. Proposal for contemporary screening strategies in families with hypertrophic cardiomyopathy. J Am Coll Cardiol 2004;44:21252132.

22. 23.

24.

25.

Usefulness of Repeated N-Terminal Pro-B-Type Natriuretic Peptide Measurements as Incremental Predictor for Long-Term Cardiovascular Outcome After Vascular Surgery
Dustin Goei, MDa, Jan-Peter van Kuijk, MDa, Willem-Jan Flu, MDa, Sanne E. Hoeks, PhDb, Michel Chonchol, MDc, Hence J.M. Verhagen, MDa, Jeroen J. Bax, MDd, and Don Poldermans, MDa,*
Plasma N-terminal proB-type natriuretic peptide (NTpro-BNP) levels improve preoperative cardiac risk stratication in vascular surgery patients. However, single preoperative measurements of NTpro-BNP cannot take into account the hemodynamic stress caused by anesthesia and surgery. Therefore, the aim of the present study was to assess the incremental predictive value of changes in NTpro-BNP during the perioperative period for long-term cardiac mortality. Detailed cardiac histories, rest left ventricular echocardiography, and NTpro-BNP levels were obtained in 144 patients before vascular surgery and before discharge. The study end point was the occurrence of cardiovascular death during a median follow-up period of 13 months (interquartile range 5 to 20). Preoperatively, the median NTpro-BNP level in the study population was 314 pg/ml (interquartile range 136 to 1,351), which increased to a median level of 1,505 pg/ml (interquartile range 404 to 6,453) before discharge. During the follow-up period, 29 patients (20%) died, 27 (93%) from cardiovascular causes. The median difference in NTpro-BNP in the survivors was 665 pg/ml, compared to 5,336 pg/ml in the patients who died (p 0.01). Multivariate Cox regression analyses, adjusted for cardiac history and cardiovascular risk factors (age, angina pectoris, myocardial infarction, stroke, diabetes mellitus, renal dysfunction, body mass index, type of surgery and the left ventricular ejection fraction), demonstrated that the difference in NTpro-BNP level between pre- and postoperative measurement was the strongest independent predictor of cardiac outcome (hazard ratio 3.06, 95% condence interval 1.36 to 6.91). In conclusion, the change in NTpro-BNP, indicated by repeated measurements before surgery and before discharge is the strongest predictor of cardiac outcomes in patients who undergo vascular surgery. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:609 614) Patients who undergo vascular surgery are at high risk for peri- and postoperative cardiac events due to underlying coronary artery disease.1 N-terminal proB-type natriuretic peptide (NTpro-BNP) levels improve preoperative cardiac risk stratication for surgical patients.2 4 NTpro-BNP is a cardiac neurohormone that is synthesized in the ventricular myocardium and is released in response to ventricular wall stretching and myocardial ischemia.5,6 However, a single preoperative measurement of NTpro-BNP cannot reect the hemodynamic changes caused by anesthesia and surgical stress. However, they might in fact be the consequence hemodynamic instabilDepartments of aVascular Surgery and bAnaesthesiology, Erasmus Medical Center, Rotterdam, The Netherlands; cDivision of Renal Diseases and Hypertension, University of Colorado Denver Health Sciences Center, Aurora, Colorado; and dDepartment of Cardiology, Leiden University Medical Center, Leiden, The Netherlands. Manuscript received August 9, 2010; revised manuscript received and accepted October 5, 2010. Dustin Goei, Jan-Peter van Kuijk, Willem-Jan Flu and Sanne E. Hoeks were supported by an unrestricted research grant from Lijf en Leven Rotterdam, the Netherlands. *Corresponding author: Tel: 31-10-7034613; fax: 31-10-7034957. E-mail address: d.poldermans@erasmusmc.nl (D. Poldermans). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.021

ities during the perioperative period, with subsequent episodes of prolonged subclinical myocardial ischemia, associated with adverse long-term cardiac outcomes.7 Previous studies have demonstrated that changes in NT pro-BNP over brief periods are related to adverse outcomes in acute coronary syndromes and acute decompensated heart failure.8,9 However, data on the use of repeated perioperative NTpro-BNP measurements in vascular surgery patients are lacking. In the present study, we evaluated the incremental predictive value of changes in NTpro-BNP during the perioperative period for long-term cardiac mortality in patients who underwent vascular surgery. Methods The study population consisted of patients who underwent elective vascular surgery at the Erasmus Medical Center (Rotterdam, The Netherlands) from 2007 to 2010. Patients were identied in a prospectively maintained database including all patients who underwent vascular surgery at this institution. The medical ethics committee of the hospital was informed about the study, and all procedures of this
www.ajconline.org

610

The American Journal of Cardiology (www.ajconline.org)

Table 1 Baseline characteristics of the source population (n 144) Variable Demographics Age (years) Men Angina pectoris Myocardial infarction Coronary revascularization Heart failure Stroke Smokers Hypertension* Diabetes mellitus Hypercholesterolemia Renal dysfunction Chronic obstructive pulmonary disease Site and type of surgery Abdominal aortic Lower extremity Carotid Open surgery Measurements Body mass index (kg/m2) LVEF (%) 30 3040 4050 50 Serum creatinine (mol/L) NTpro-BNP (pg/ml) Preoperative Postoperative Value 68 10 99 (69%) 32 (22%) 59 (41%) 41 (29%) 24 (17%) 38 (26%) 52 (26%) 100 (69%) 40 (28%) 75 (52%) 35 (24%) 26 (18%) 74 (51%) 56 (39%) 14 (10%) 106 (74%) 26 3.8 10 (7%) 30 (21%) 59 (41%) 45 (31%) 90 (72113) 314 (1361,351) 1,505 (4046,453)

Data are expressed as mean SD, number (percentage), or median (IQR). * Blood pressure 140/90 mm Hg or medical therapy to control hypertension. Plasma cholesterol 5.5 mmol/L or treatment with lipid-lowering drugs Endovascular or open vascular procedures.

Table 2 Medications at screening and discharge (n 144) Medication Aspirin Statins -blocking agents Diuretics Angiotensin-converting enzyme inhibitors Calcium antagonists Angiotensin receptor blockers At Screening 98 (68%) 110 (77%) 137 (95%) 43 (30%) 47 (33%) 29 (20%) 17 (12%) At Discharge 95 (66%) 115 (80%) 141 (98%) 43 (30%) 48 (33%) 27 (19%) 35 (24%) p Value NS NS NS NS NS NS 0.002

retrospective study met the approval of the medical ethics committee of the Erasmus Medical Center. Before surgery, a detailed cardiac history was obtained from each patient, including angina pectoris, myocardial infarction, percutaneous coronary intervention or coronary artery bypass grafting, heart failure (dened as the presence of heart failure symptoms according the New York Heart Association classication or previous hospital admission for

decompensated heart failure), and stroke or transient ischemic attack. Furthermore, cardiovascular risk factors were recorded, and included age, smoking history, hypertension (blood pressure 140/90 mm Hg or medical therapy to control hypertension), diabetes mellitus (fasting glucose level 7.0 mmol/L or medication to control diabetes), hypercholesterolemia (plasma cholesterol level 5.5 mmol/L or treatment with lipid-lowering drugs), and renal dysfunction (dened as serum creatinine 2 mg/dl). Other data collected included history of chronic obstructive pulmonary disease (dened as a forced expiratory volume in 1 second 70% of age- and gender-predicted value), site of surgery (abdominal aortic, lower extremity, or carotid), and type of procedure (endovascular or open). Finally, the use of the following medications was recorded at baseline and at the time of discharge: aspirin, statins, -blocking agents, calcium antagonists, diuretics, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers. Treatment goals were dened according to current guidelines for patients with peripheral arterial disease and included low-dose aspirin (80 mg/day), statins, -blocking agents (titrated to a perioperative heart rate of 50 to 70 beats/min) in patients with or at risk for ischemic heart disease, and angiotensin-converting enzyme inhibitors in patients with left ventricular (LV) ejection fractions (LVEFs) 40%.10 Peripheral venous blood samples were obtained for measurement of NTpro-BNP levels in all patients during the preoperative outpatient clinic or at hospital admission and before discharge. NTpro-BNP concentration was determined using an electrochemiluminescence assay on an Elecsys (Hoffman-La Roche, Basel, Switzerland). The method is a sandwich-type quantitative immunoassay, based on polyclonal antibodies against epitopes in the N-terminal part of pro-BNP. The lower detection limit was 5 pg/ml. Intraassay coefcients of variance at 271 and 6,436 pg/ml were 1.9% and 0.9%, respectively. Assays were performed by a laboratory technician blinded to the patients clinical data. Importantly, pre- and postoperative NTpro-BNP levels were unknown for the treating physician and were not used for clinical management. Preoperatively, patients underwent 2-dimensional transthoracic echocardiography during rest. Cardiac evaluation was performed using a portable Acuson Cypress ultrasound system (Siemens Medical Solutions USA, Inc., Mountain View, California) with a 3V2C transducer (3.0, 3.5, 2.5, and 2.0 MHz). The LVEF was assessed in the apical, 4-chamber, or 2-chamber view with the patient in left lateral decubitus position. Quantication of LV volumes was performed using the modied Simpsons rule, with inter- and intraobserver variability of 9% to 12% and 6%, respectively.11 The LVEF was calculated as (LV end-systolic volume LV end-diastolic volume) 100/LV end-diastolic volume. LV dysfunction was dened as impaired LV systolic function with a LVEF 40%. Of note, all echocardiographic studies performed were for research purposes and were not used for clinical management. The end point of this study was the occurrence of cardiovascular death, dened as any death with a cardiovascular cause, including those deaths after cardiac procedures, cardiac arrest, myocardial infarction, pulmonary embolus,

Miscellaneous/Repeated NTpro-BNP Measurements and Long-Term Prognosis

611

Figure 1. Receiver-operating characteristic curve of NTpro-BNP levels to predict long-term cardiovascular mortality. Sensitivity and 1 specicity are plotted for various levels. The ideal cut-off value is indicated by the arrow.

erative days were calculated as median differences with their interquartile range. We applied multivariate Cox regression analyses to evaluate the relation between plasma levels of NTpro-BNP and the subsequent changes in relation to the study end point. In the regression analyses, NTpro-BNP was entered as a continuous dependent variable and was log-transformed to obtain normality. Multivariate regression analyses were adjusted for cardiac risk factors12 and factors recognized to inuence NTpro-BNP levels1315: age, angina pectoris, myocardial infarction, stroke, diabetes mellitus, renal dysfunction, body mass index, type of surgery, and the LVEF. Interactions between renal dysfunction, intraoperative uids administered, medication at discharge (diuretics, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers) and the change in NTproBNP levels with the study end point were evaluated by forcing these interaction terms in the multivariate regression model. Interaction terms were included in the multivariate regression model only when signicant. We report crude and adjusted hazard ratios and their 95% condence intervals. For all tests, p values 0.05 (2 sided) were considered signicant. All analyses were performed using SPSS version 15.0 (SPSS, Inc., Chicago, Illinois). Results The study population consisted of 144 patients with peripheral arterial disease referred for elective noncardiac vascular surgery. Most of the patients underwent abdominal aortic surgery (n 74 [51%]). Baseline characteristics of the total study population are listed in Table 1. The mean age of the patients was 68 10 years, and 69% were men. Almost half of the patients (41%) had histories of myocardial infarction. Diabetes mellitus and renal dysfunction were present in 28% and 24% of patients, respectively. Hypertension and hypercholesterolemia were the most frequent cardiovascular risk factors and were observed in 69% and 52% of the patients, respectively. At baseline, 99 patients (69%) had LVEFs 40%. Symptomatic heart failure was based on the presence of signs and symptoms according to New York Heart Association classication and was diagnosed in 24 patients (17%) of the total study population. Of these patients, 17 (71%) had systolic or combined heart failure, while 7 (29%) were diagnosed with heart failure with preserved ejection fractions. In the remaining 75 patients (52%) with LVEFs 40%, but without clinical symptoms of heart failure, systolic or combined LV dysfunction was observed in 40 patients (53%), and 35 patients (47%) had isolated diastolic dysfunction. The high prevalence of LV dysfunction was also reected by the use of diuretics and angiotensin-converting enzyme inhibitors, which were prescribed in 1/3 of these patients. Furthermore, cardioprotective medications in the population, such as aspirin, statins, and blockers, were prescribed in 2 of 3 patients. With the exception of angiotensin receptor blockers (p 0.02), no signicant differences were found between medications at screening and at discharge (Table 2).

Figure 2. The changes in NTpro-BNP levels between the pre- and postoperative periods were compared between survivors and patients who died.

stroke, or sudden deaths not ascribed to other causes. Mortality was considered cardiovascular unless explicit proof of a noncardiac cause could be delivered. Long-term mortality was assessed by approaching the municipal civil registries. Dichotomous data are described as numbers and percentages. The continuous variables age and body mass index are described as mean SD. Continuous data with a signicant skewed distribution were compared using the Mann-Whitney U test and are expressed as median (interquartile range [IQR]). Receiver-operating characteristic curve analysis was used to assess the optimal cut-off value of NTpro-BNP for the prediction of longterm cardiac mortality. The optimal value of preoperative NTpro-BNP for predicting long-term cardiac mortality was dened as the concentration with the largest sum of sensitivity plus specicity. Changes in plasma NTproBNP levels from preoperatively until the rst 30 postop-

612 Table 3 Long-term mortality Variable HR Preoperative NTproBNP NTpro-BNP change 2.83 3.31 Univariate 95% CI 1.634.93 1.766.23

The American Journal of Cardiology (www.ajconline.org)

Multivariate* HR 2.57 3.06 95% CI 1.165.70 1.366.91

* Adjusted for age, angina pectoris, myocardial infarction, stroke, diabetes mellitus, renal dysfunction, body mass index, type of surgery, and the LVEF. NTpro-BNP was entered as a log-transformed variable.

The median preoperative NTpro-BNP level in our highrisk vascular surgery population was 314 pg/ml (IQR 136 to 1,351). During the rst 30 postoperative days, repeated measurements of NTpro-BNP were performed, and the median postoperative NTpro-BNP level increased to 1,505 pg/ml (IQR 404 to 6,453). The median difference for the total population between pre- and postoperative NTproBNP levels was 969 pg/ml (IQR 139 to 4,337). In all patients, the rst postoperative NTpro-BNP measurement was performed before hospital discharge. The mean length of hospital stay was 6.8 3.1 days. After a median follow-up period of 13 months (IQR 5 to 20), the mortality end point was reached in 29 patients (20%), 27 (93%) of whom died secondary to cardiovascular causes. The association between preoperative NTpro-BNP level and long-term cardiovascular mortality was assessed using a receiver-operating characteristic curve (Figure 1). For preoperative NT pro-BNP, the area under the curve was 0.668 (95% condence interval 0.619 to 0.716), and the optimum discriminate threshold was 350 pg/ml. The changes in NT pro-BNP levels between the pre- and postoperative periods were compared between survivors and patients who died (Figure 2). The median preoperative NTpro-BNP level in patients who died during the follow-up period was signicantly higher compared to the survivors (795 vs 269 pg/ml, p 0.002). In addition, the median difference between preand postoperative NTpro-BNP level was signicantly higher in patients who died compared to the survivors (5,336 vs 665 pg/ml, p 0.010). We found no signicant interaction between renal dysfunction, intraoperative uids administered, medications at discharge, and the change in NTpro-BNP levels with respect to the study end point. Using multivariate Cox regression analyses with adjustment for demographics and cardiac risk factors, preoperative NT pro-BNP level as a log-transformed variable was an independent predictor of long-term cardiac mortality (hazard ratio 2.57 95% condence interval 1.16 to 5.70). Importantly, the change in NTpro-BNP level between pre- and postoperative measurements was the strongest independent predictor of cardiac outcome (hazard ratio 3.06 95% condence interval 1.36 to 6.91; Table 3). Discussion The present study demonstrates that a change in NTproBNP, indicated by repeated measurements before surgery and before discharge, is an incremental and independent

predictor of an increased long-term cardiovascular mortality risk on top of clinical risk factors. Importantly, the change in NTpro-BNP level was the strongest predictor of cardiac outcome and yielded a threefold increased risk for the occurrence of long-term cardiac events. Postoperative cardiac events in patients who undergo vascular surgery are more common in patients with preoperative myocardial ischemia, LV dysfunction, and valve abnormalities compared to patients without these conditions.12,16 There has been considerable evidence demonstrating that a single determination of NTpro-BNP is a promising marker in the setting of preoperative cardiac risk stratication.2,4,17 In addition to cardiac risk factors only, NTpro-BNP above the threshold of 350 pg/ml was an excellent tool for further risk stratication (C-statistic 0.86) in patients who undergo elective noncardiac vascular surgery.17 Rodseth et al18 demonstrated that NTpro-BNP above the optimal discriminatory threshold of 280 pg/ml, determined by receiver-operating characteristic curve analysis, was associated with 30-day and intermediate-term cardiac outcomes. We observed an optimal discriminatory threshold of 350 pg/ml for long-term outcomes, which was in fact close to the median concentration at baseline (314 pg/ml). However, it seems unlikely that there is a dichotomous threshold that denes a normal or abnormal NTproBNP value. In 2 large meta-analyses of mixed cohorts of non cardiac surgery patients, the decision threshold for NTpro-BNP varied widely from 201 to 791 pg/ml.19,20 On the basis of these data, it is more likely that the perioperative cardiovascular risk increases as NTpro-BNP concentrations increase. Furthermore, noncardiac factors such as renal dysfunction, pulmonary hypertension, chronic obstructive pulmonary disease, and body mass index might inuence NTpro-BNP levels.21,22 The role of postoperative NTpro-BNP determination is less clear. Mahla et al23 hypothesized that the differentiation between preoperative and postoperative NTpro-BNP levels is important, because restriction to a single preoperative value does not reect the variable hemodynamic consequences of anesthesia and risks associated with type of surgery. In the study by Mahla et al,23 218 patients scheduled for vascular surgery were enrolled, and the optimal discriminate threshold for postoperative NTpro-BNP was calculated at 860 pg/ml. They concluded that a single postoperative NTpro-BNP determination provides important additional prognostic information to preoperative levels. However, several important limitations of that study should be acknowledged. Pre- and postoperative NTpro-BNP levels were analyzed as 2 separate predictors of long-term outcomes, but no attention was given to the change in NTpro-BNP levels during the perioperative period. Furthermore, using multivariate regression analyses, no adjustments were performed for conventional cardiovascular risk factors. Most important, however, no adjustments were performed for preoperative NTpro-BNP level, which is the most evident confounder of increased postoperative NT pro-BNP level. The present study is the rst to demonstrate that the change in NTpro-BNP level between pre- and postoperative measurement is the strongest independent predictor of cardiac outcomes, after adjustment for conventional risk

Miscellaneous/Repeated NTpro-BNP Measurements and Long-Term Prognosis

613

factors and preoperative NTpro-BNP levels. In our population the median difference between pre- and postoperative NTpro-BNP levels was 969 pg/ml. In line with other studies, the present ndings demonstrated high NTpro-BNP levels after major surgery. Importantly, the reasons for these elevations remain largely unknown but could be explained by several pathophysiologic pathways. Natriuretic peptide release is an index of activation of neurohumoral axis in the setting of LV overload and myocardial ischemia24 to reduce ventricular wall stress. It could be speculated that a postoperative increase of NTpro-BNP reects impaired cardiac function with prognostic power beyond that of a single preoperative determination because it incorporates the physiologic consequences of anesthesia, uid shifts, surgical stress, duration of the procedure, and intraoperative blood loss. Notably, a correlation between pre- and poststress levels of NTpro-BNP and the risk and extent of inducible ischemia has been demonstrated previously, in the nonsurgical setting25 and in noninvasive risk assessment before noncardiac surgery.4,26 As such, increases in plasma levels of NTpro-BNP may have a role in identifying high-risk vascular patients who may require more extensive postoperative cardiovascular follow-up. Furthermore, it offers prospect of several applications, including the selection of higher risk subjects for recruitment to therapeutic trials and potentially providing a reliable surrogate index of efcacy of new treatments. In contrast, in patients with chronic heart failure, previous clinical trials have examined the value of adding measurements of NTpro-BNP to standard heart failure treatment with the effort to improve outcomes and have returned mixed results.2730 Potential limitations of these data merit consideration. First, the study population consisted of patients referred to a tertiary referral center and may not fully represent a general population scheduled for elective vascular surgery. Second, unknown factors that inuence NTproBNP levels and their interaction with identifying risk for adverse events need to be accounted. Although currently, no consensus exists regarding the reference range of NTpro-BNP values, we used the change in NTproBNP level between pre- and postoperative measurement. Our ndings suggest the incorporation of NTpro-BNP determinations in the diagnostic procedure before and after surgery in vascular surgery patients. Changes between pre- and postoperative NTpro-BNP levels provide important additional prognostic information and should provoke clinicians to nd the cause responsible for the elevation of plasma NTpro-BNP levels. Investigators undertaking clinical trials or cohort studies should be encouraged to incorporate serial neurohormonal measurements in their study designs. Furthermore, studies in larger patient populations are required to clarify the optimal predictive cut-off value for the prediction of longterm cardiac events in addition to known cardiac risk factors.
1. Mangano DT. Perioperative cardiac morbidity. Anesthesiology 1990; 72:153184. 2. Yeh HM, Lau HP, Lin JM, Sun WZ, Wang MJ, Lai LP. Preoperative plasma N-terminal pro-brain natriuretic peptide as a marker of cardiac

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

risk in patients undergoing elective non-cardiac surgery. Br J Surg 2005;92:10411045. Dernellis J, Panaretou M. Assessment of cardiac risk before noncardiac surgery: brain natriuretic peptide in 1590 patients. Heart 2006; 92:16451650. Feringa HH, Bax JJ, Elhendy A, de Jonge R, Lindemans J, Schouten O, van den Meiracker AH, Boersma E, Schinkel AF, Kertai MD, van Sambeek MR, Poldermans D. Association of plasma N-terminal proB-type natriuretic peptide with postoperative cardiac events in patients undergoing surgery for abdominal aortic aneurysm or leg bypass. Am J Cardiol 2006;98:111115. Schnabel R, Rupprecht HJ, Lackner KJ, Lubos E, Bickel C, Meyer J, Munzel T, Cambien F, Tiret L, Blankenberg S. Analysis of N-terminal-pro-brain natriuretic peptide and C-reactive protein for risk stratication in stable and unstable coronary artery disease: results from the AtheroGene study. Eur Heart J 2005;26:241249. Kragelund C, Gronning B, Kober L, Hildebrandt P, Steffensen R. N-terminal pro-B-type natriuretic peptide and long-term mortality in stable coronary heart disease. N Engl J Med 2005;352:666 675. Landesberg G, Shatz V, Akopnik I, Wolf YG, Mayer M, Berlatzky Y, Weissman C, Mosseri M. Association of cardiac troponin, CK-MB, and postoperative myocardial ischemia with long-term survival after major vascular surgery. J Am Coll Cardiol 2003;42:15471554. Heeschen C, Hamm CW, Mitrovic V, Lantelme NH, White HD. N-terminal pro-B-type natriuretic peptide levels for dynamic risk stratication of patients with acute coronary syndromes. Circulation 2004; 110:3206 3212. Bettencourt P, Azevedo A, Pimenta J, Frioes F, Ferreira S, Ferreira A. N-terminal-pro-brain natriuretic peptide predicts outcome after hospital discharge in heart failure patients. Circulation 2004;110:2168 2174. Poldermans D, Bax JJ, Boersma E, De Hert S, Eeckhout E, Fowkes G, Gorenek B, Hennerici MG, Iung B, Kelm M, Kjeldsen KP, Kristensen SD, Lopez-Sendon J, Pelosi P, Philippe F, Pierard L, Ponikowski P, Schmid JP, Sellevold OF, Sicari R, Van den Berghe G, Vermassen F, Hoeks SE, Vanhorebeek I, Vahanian A, Auricchio A, Ceconi C, Dean V, Filippatos G, Funck-Brentano C, Hobbs R, Kearney P, McDonagh T, McGregor K, Popescu BA, Reiner Z, Sechtem U, Sirnes PA, Tendera M, Vardas P, Widimsky P, De Caterina R, Agewall S, Al Attar N, Andreotti F, Anker SD, Baron-Esquivias G, Berkenboom G, Chapouotot L, Cifkova R, Faggiano P, Gibbs S, Hansen HS, Iserin L, Israel CW, Kornowski R, Eizagaechevarria NM, Pepi M, Piepoli M, Priebe HJ, Scherer M, Stepinska J, Taggart D, Tubaro M, Guidelines for pre-operative cardiac risk assessment and perioperative cardiac management in non-cardiac surgery: the Task Force for Preoperative Cardiac Risk Assessment and Perioperative Cardiac Management in Non-Cardiac Surgery of the European Society of Cardiology (ESC) and endorsed by the European Society of Anaesthesiology (ESA). Eur Heart J 2009;30:2769 2812. McGowan JH Cleland JG. Reliability of reporting left ventricular systolic function by echocardiography: a systematic review of 3 methods. Am Heart J 2003;146:388 397. Boersma E, Poldermans D, Bax JJ, Steyerberg EW, Thomson IR, Banga JD, van De Ven LL, van Urk H, Roelandt JR. Predictors of cardiac events after major vascular surgery: Role of clinical characteristics, dobutamine echocardiography, and beta-blocker therapy. JAMA 2001;285:18651873. Goei D, Schouten O, Boersma E, Welten GMJM, Dunkelgrun M, Lindemans J, van Gestel YRBM, Hoeks SE, Bax JJ, Poldermans D. Inuence of renal function on the usefulness of N-terminal pro-B-type natriuretic peptide as a prognostic cardiac risk marker in patients undergoing noncardiac vascular surgery. Am J Cardiol 2008;101:122 126. Redeld MM, Rodeheffer RJ, Jacobsen SJ, Mahoney DW, Bailey KR, Burnett JC Jr. Plasma brain natriuretic peptide concentration: impact of age and gender. J Am Coll Cardiol 2002;40:976 982. Daniels LB, Clopton P, Bhalla V, Krishnaswamy P, Nowak RM, McCord J, Hollander JE, Duc P, Omland T, Storrow AB, Abraham WT, Wu AH, Steg PG, Westheim A, Knudsen CW, Perez A, Kazanegra R, Herrmann HC, McCullough PA, Maisel AS. How obesity affects the cut-points for B-type natriuretic peptide in the diagnosis of acute heart failure. Results from the Breathing Not Properly Multinational Study. Am Heart J 2006;151:999 1005.

614

The American Journal of Cardiology (www.ajconline.org) Hori R, Yasue H, Nakao K. Rapid ventricular induction of brain natriuretic peptide gene expression in experimental acute myocardial infarction. Circulation 1995;92:1558 1564. Sabatine MS, Morrow DA, de Lemos JA, Omland T, Desai MY, Tanasijevic M, Hall C, McCabe CH, Braunwald E. Acute changes in circulating natriuretic peptide levels in relation to myocardial ischemia. J Am Coll Cardiol 2004;44:1988 1995. Feringa HH, Schouten O, Dunkelgrun M, Bax JJ, Boersma E, Elhendy A, de Jonge R, Karagiannis SE, Vidakovic R, Poldermans D. Plasma N-terminal pro-B-type natriuretic peptide as long-term prognostic marker after major vascular surgery. Heart 2007;93:226 231. Lainchbury JG, Troughton RW, Frampton CM, Yandle TG, Hamid A, Nicholls MG, Richards AM. NTproBNP-guided drug treatment for chronic heart failure: design and methods in the BATTLESCARRED trial. Eur J Heart Fail 2006;8:532538. Psterer M, Buser P, Rickli H, Gutmann M, Erne P, Rickenbacher P, Vuillomenet A, Jeker U, Dubach P, Beer H, Yoon SI, Suter T, Osterhues HH, Schieber MM, Hilti P, Schindler R, Brunner-La Rocca HP. BNP-guided vs symptom-guided heart failure therapy: the Trial of Intensied vs Standard Medical Therapy in Elderly Patients With Congestive Heart Failure (TIME-CHF) randomized trial. JAMA 2009; 301:383392. Jourdain P, Jondeau G, Funck F, Gueffet P, Le Helloco A, Donal E, Aupetit JF, Aumont MC, Galinier M, Eicher JC, Cohen-Solal A, Juilliere Y. Plasma brain natriuretic peptide-guided therapy to improve outcome in heart failure: the STARS-BNP multicenter study. J Am Coll Cardiol 2007;49:17331739. Schou M, Gustafsson F, Videbaek L, Markenvard J, Ulriksen H, Ryde H, Jensen JC, Nielsen T, Knudsen AS, Tuxen CD, Handberg J, Sorensen PJ, Espersen G, Lind-Rasmussen S, Keller N, Egstrup K, Nielsen OW, Abdulla J, Nyvad O, Toft J, Hildebrandt PR. Design and methodology of the NorthStar Study: NT-proBNP stratied follow-up in outpatient heart failure clinicsa randomized Danish multicenter study. Am Heart J 2008;156:649 655.

16. Lee TH, Marcantonio ER, Mangione CM, Thomas EJ, Polanczyk CA, Cook EF, Sugarbaker DJ, Donaldson MC, Poss R, Ho KK, Ludwig LE, Pedan A, Goldman L. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:10431049. 17. Schouten O, Hoeks SE, Goei D, Bax JJ, Verhagen HJ, Poldermans D. Plasma N-terminal pro-B-type natriuretic peptide as a predictor of perioperative and long-term outcome after vascular surgery. J Vasc Surg 2009;49:435 441. 18. Rodseth RN, Padayachee L, Biccard BM. A meta-analysis of the utility of pre-operative brain natriuretic peptide in predicting early and intermediate-term mortality and major adverse cardiac events in vascular surgical patients. Anaesthesia 2008;63:1226 1233. 19. Karthikeyan G, Moncur RA, Levine O, Heels-Ansdell D, Chan MT, Alonso-Coello P, Yusuf S, Sessler D, Villar JC, Berwanger O, McQueen M, Mathew A, Hill S, Gibson S, Berry C, Yeh HM, Devereaux PJ. Is a pre-operative brain natriuretic peptide or N-terminal pro-Btype natriuretic peptide measurement an independent predictor of adverse cardiovascular outcomes within 30 days of noncardiac surgery? A systematic review and meta-analysis of observational studies. J Am Coll Cardiol 2009;54:1599 1606. 20. Ryding AD, Kumar S, Worthington AM, Burgess D. Prognostic value of brain natriuretic peptide in noncardiac surgery: a meta-analysis. Anesthesiology 2009;111:311319. 21. DeFilippi C, van Kimmenade RR, Pinto YM. Amino-terminal pro-Btype natriuretic peptide testing in renal disease. Am J Cardiol 2008; 101:82 88. 22. de Lemos JA, Hildebrandt P. Amino-terminal pro-B-type natriuretic peptides: testing in general populations. Am J Cardiol 2008;101:16 20. 23. Mahla E, Baumann A, Rehak P, Watzinger N, Vicenzi MN, Maier R, Tiesenhausen K, Metzler H, Toller W. N-terminal pro-brain natriuretic peptide identies patients at high risk for adverse cardiac outcome after vascular surgery. Anesthesiology 2007;106:1088 1095. 24. Hama N, Itoh H, Shirakami G, Nakagawa O, Suga S, Ogawa Y, Masuda I, Nakanishi K, Yoshimasa T, Hashimoto Y, Yamaguchi M,

25.

26.

27.

28.

29.

30.

Usefulness of At Rest and Exercise Hemodynamics to Detect Subclinical Myocardial Disease in Type 2 Diabetes Mellitus
Christine L. Jellis, MDa, Tony Stanton, MD, PhDa, Rodel Leano, BSa, Jennifer Martin, MD, PhDa, and Thomas H. Marwick, MD, PhDa,b,*
Patients with type 2 diabetes mellitus (T2DM) might have subclinical myocardial dysfunction identied at rest or unmasked during exercise. We examined the correlates of the myocardial exercise response in patients with T2DM. Myocardial dysfunction was sought during at rest and exercise echocardiography in 167 healthy patients with T2DM (97 men, 55 10 years). Myocardial ischemia was excluded using stress echocardiography. Standard echocardiography and color tissue Doppler imaging measures (early diastolic tissue velocity [Em], strain, and strain rate) were acquired at baseline and peak stress. The calibrated integrated backscatter was calculated from the at rest parasternal long-axis view. The longitudinal diastolic functional reserve index after exercise was dened as Em [1 (1/Embase)]. The clinical, anthropometric, and metabolic data were collected at rest and stress. Subclinical myocardial dysfunction at baseline (n 24) was independently associated with weight (odds ratio [OR] 1.02, p 0.04) and hemoglobin A1c (OR 1.36, p 0.03). This group displayed an impaired exercise response that was independently associated with a reduced exercise capacity (OR 0.84, p 0.034) and longitudinal diastolic functional reserve index (OR 0.69, p 0.001). Inducible myocardial dysfunction (stress Em <9.9 cm/s) was identied after exercise in 70 of the remaining 143 subjects. This nding was associated with calibrated integrated backscatter (OR 1.08, p 0.04) and lower peak heart rate (OR 0.97, p 0.002) but not metabolic control. The intensity of the metabolic derangement in patients with T2DM was associated with subclinical at rest myocardial dysfunction, but not with the myocardial exercise response. In conclusion, the association of an abnormal stress response with nonmetabolic factors, including backscatter and blunted peak heart rate, suggests potential roles for myocardial brosis and cardiac autonomic neuropathy in patients with nonischemic diabetic heart disease. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:615 621) In addition to hastening atherosclerosis, type 2 diabetes mellitus (T2DM) and the metabolic syndrome1 have been linked to myocardial disease in the absence of ischemic heart disease and hypertension.2 This is likely multifactorial, secondary to the accumulation of advanced glycated end products, myocardial brosis, microvascular disease, and autonomic neuropathy. Diabetic heart disease is initially asymptomatic; however, nonspecic symptoms of fatigue, dyspnea, or reduced exercise tolerance will gradually develop. Early detection might facilitate measures to prevent disease progression. Tissue velocity and deformation imaging can detect myocardial dysfunction when the conventional 2-dimensional echocardiographic parameters are normal.3 In early diabetic heart disease, myocardial function might be preserved at rest, with exercise unmasking a blunting of contraction and relaxation, indicative of an abnormal functional reserve.4 Longitudinal function is typically reduced initially, reective of the early involvement of the subendocardial bers.5 Impairment in the at rest and peak exercise systolic tissue velocity has been associated with common metabolic risk factors in asymptomatic patients.6 We sought to identify whether early diastolic tissue velocity (Em), deformation imaging, and tissue characterization could identify diabetic heart disease not apparent at rest and examined the correlates of myocardial dysfunction with exercise. Methods A total of 167 apparently healthy subjects with T2DM (97 men, 55 10 years) and no macro- or microvascular complications of T2DM or history of hypertension or valvular, congenital, or ischemic heart disease were recruited from the hospital clinics of the Princess Alexandra Hospital and its local community. Sinus rhythm and normal renal function were required for inclusion. Antihypertensive medications were withheld for 12 hours before testing. The human research ethics committees of Princess Alexandra Hospital and the University of Queensland (Brisbane, Australia) approved the present study.
www.ajconline.org

The University of Queensland, Brisbane, Australia; and bCleveland Clinic, Cleveland, Ohio. Manuscript received September 25, 2010; manuscript received and accepted October 5, 2010. This study was supported in part by a Centres for Clinical Research Excellence award (455832) from the National Health and Medical Research Council, Canberra, Australia. Dr. Jellis was supported by a Research Entry Scholarship from the Vincent Fairfax Family Foundation, Sydney, Australia; and the Royal Australasian College of Physicians, Sydney, Australia. *Corresponding author: Tel: (216) 445-7275; fax: (216) 445-7306. E-mail address: marwict@ccf.org (T.H. Marwick). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.024

616

The American Journal of Cardiology (www.ajconline.org)

Table 1 Characteristics of myocardial dysfunction at rest and unmasked by exercise Variable Normal (n 143) Age (years) Type 2 diabetes mellitus duration (years) Weight (kg) Body mass index (kg/m2) Fasting glucose (mmol/L) Hemoglobin A1c (%) Total cholesterol mmol/L mg/dl Low-density lipoprotein cholesterol mmol/L mg/dl Creatinine (mmol/L) Microalbuminuria (%) [albumin/creatinine (g/mol)] Statin therapy Angiotensin-converting enzyme inhibitors and/or angiotensin-receptor blockers Blockers At rest heart rate (beats/min) At rest systolic blood pressure (mm Hg) At rest diastolic blood pressure (mm Hg) Peak heart rate (beats/min) Exercise capacity (METs) At rest end-systolic volume (ml) At rest end-diastolic volume (ml) At rest ejection fraction (%) Stress end-systolic volume (ml) Stress end-diastolic volume (ml) Stress ejection fraction (%) Change in ejection fraction (%) At rest early diastolic tissue velocity (cm/s) At rest systolic tissue velocity (cm/s) Stress early diastolic tissue velocity (cm/s) Stress systolic tissue velocity (cm/s) Change in systolic tissue velocity (cm/s) Left ventricular longitudinal functional reserve index At rest strain (%) Stress strain (%) At rest strain rate (s1) Stress strain rate (s1) Calibrated integrated backscatter (dB) Height-indexed left ventricular mass (g/m2.7) 55 10 5.3 5.5 89.5 17.0 31.3 5.4 8.1 2.8 7.4 1.4 4.8 0.9 186 35 2.7 0.8 104 31 79 18 16 (11%); 0.7*; IQR 0.9 55 (38%) 58 (41%) 9 (6%) 85 13 133 17 81 10 163 20 9.2 3.2 25 10 73 19 65 7 18 7 69 19 74 5 96 5.7 1.5 4.9 1.1 9.9 2.5 8.0 2.0 3.1 1.9 3.5 2.5 20.8 3.1 21.2 3.5 1.3 0.3 1.9 0.5 17.1 5.4 51.0 16.7 At Rest Em Abnormal (n 24) 53 11 9.7 8.9 98.5 28.5 33.7 7.2 9.6 3.8 8.1 1.6 4.5 0.8 174 31 0.035 2.3 0.8 89 31 77 24 6 (25%); 1.1*; IQR 2.2 16 (67%) 18 (75%) 1 (4%) 88 16 138 17 86 8 159 19 7.5 2.5 29 12 78 26 64 8 10 8 71 23 73 6 97 3.1 1.0 2.9 1.4 6.3 4.5 5.4 2.6 2.5 1.6 1.3 2.2 19.6 3.7 21.2 3.6 1.4 0.3 2.0 0.5 16.3 5.7 45.2 16.6 2.8 0.8 108 31 78 17 7 (10%); 0.8*; IQR 0.7 24 (33%) 28 (39%) 1 (1%) 88 12 132 16 81 10 170 17 9.4 3.3 26 10 72 20 64 6 18 7 69 18 74 5 10 6 5.8 1.4 4.9 1.1 11.5 1.9 8.3 1.9 3.5 1.8 5.3 1.7 20.6 3.0 20.6 3.7 1.3 0.3 1.9 0.5 18.0 4.9 52.4 18.1 2.6 0.9 101 35 80 19 8 (11%); 0.7*; IQR 0.9 32 (46%) 30 (43%) 8 (11%) 81 13 134 17 81 9 155 20 9.0 3.0 25 10 73 19 67 7 18 7 70 21 74 6 77 5.5 1.6 4.8 1.2 8.1 1.7 7.5 2.2 2.6 1.9 1.6 1.7 20.8 3.2 22.0 3.2 1.3 0.3 1.9 0.4 15.9 5.8 49.6 15.0 p Value 0.389 0.024 0.035 0.054 0.030 0.024 0.105 Normal (n 73) 53 9 5.1 5.4 89.7 17.4 31.5 5.6 8.2 3.0 7.5 1.4 5.0 0.8 193 31 Stress Em Abnormal (n 70) 58 10 6.1 6.3 90.2 16.9 31.2 5.3 8.2 2.8 7.4 1.4 4.7 1.0 182 39 0.084 p Value 0.007 0.310 0.848 0.768 0.873 0.458 0.019

0.624 0.064 0.010 0.005 0.684 0.330 0.259 0.033 0.363 0.013 0.137 0.210 0.241 0.224 0.590 0.553 0.825 0.001 0.001 0.001 0.001 0.184 0.001 0.09 0.978 0.160 0.181 0.539 0.125

0.459 0.720 0.116 0.474 0.013 0.001 0.572 0.949 0.001 0.439 0.314 0.753 0.017 0.906 0.718 0.852 0.014 0.365 0.614 0.001 0.012 0.006 0.001 0.605 0.020 0.728 0.692 0.019 0.360

No statistically signicant difference noted between at rest or Stress Em groups for gender, smoking status, height, waist/hip ratio, triglycerides, high-density lipoprotein cholesterol, hypoglycemic therapy, or peak systolic or diastolic blood pressure. * Median value given as nonparametric distribution. IQR interquartile range.

Clinical data were collected regarding subject age, gender, weight, height, waist and hip circumference, smoking status, and duration of T2DM. Venesection was performed before exercise after the subjects had fasted for 8 hours. The tested parameters included fasting glucose, hemoglobin A1c (HbA1c), creatinine, and lipid prole. Microalbuminuria, a marker of microvascular disease, was quantied using a random urinary albumin/creatinine ratio and dened as 2.5 g/mol for men and 3.5 g/mol for women. The heart rate and blood pressure were measured at baseline,

throughout exercise, and during recovery. The peak exercise capacity was estimated in METs according to the duration of exercise using the equation: METs [speed (0.1 [grade 1.8]) 3.5]/3.5. Standard commercially available cardiac ultrasound machines (Vivid 7, General Electric Medical Systems, Milwaukee, WI) were used to perform M-mode and 2-dimensional echocardiography to assess the chamber wall thickness, valvular morphology, and chamber volumes. Baseline parasternal and apical images were acquired in

Miscellaneous/Subclinical Diabetic Myocardial Disease

617

gray scale and color tissue Doppler imaging (TDI) formats. These were used as the at rest reference images for comparison of the stress echocardiographic images in the same views and to enable off-line TDI parameter measurement. Stress echocardiography was performed to exclude inducible segmental myocardial dysfunction indicative of underlying hemodynamically signicant epicardial coronary artery disease. The subjects underwent treadmill exercise using the Bruce protocol, and the peak images in the parasternal and apical views using the same formats were acquired. The images were digitally saved for off-line analysis. The left ventricular end-systolic and end-diastolic volumes were quantied at rest and peak to calculate the ejection fraction using the modied Simpson biplane method. Conventional apical views (4-chamber, 2-chamber, and long-axis) in color TDI format at rest and peak were used to obtain the tissue velocity, strain, and strain rate curves from the 6 basal segments using standard commercial software (Echopac, GE Vingmed). The peak systolic tissue velocity (Sm) and peak Em were calculated from the tissue velocity curves by placing a sample volume at the annulus of the mitral valve. An average value from 3 consecutive tissue velocity curves was established. On some stress images, the transmitral and TDI diastolic velocity curve peaks were fused. In these instances, we waited until the heart rate had decreased and they had separated. Abnormal diastolic function at rest was dened as an at rest septal Em 2 SD of normal for age.7 Normal ageadjusted values for peak exercise stress Em have not yet been dened. Hence, abnormal exercise diastolic function was identied when the stress septal Em was 9.9 cm/s, the mean value in a nondiabetic control population.4 Longitudinal left ventricular diastolic functional reserve index, a measure of augmentation in diastolic relaxation during exercise, was calculated using the previously dened equation: longitudinal left ventricular diastolic function reserve index Em [1 (1/Embase)], where Em was the change in Em from baseline to peak exercise and Embase was the early diastolic tissue velocity at baseline measured at the septal mitral annulus.4 This normalization equation was used to express the Em with respect to the baseline Em at rest. According to previous studies, the preserved contractile reserve was dened as augmentation of the ejection fraction 4%8 and/or an increase in the stress Sm from the baseline at rest Sm (Sm) 2.4 cm/s.4 The strain and strain rate curves were derived from the apical views in color TDI format by placing sample volumes in the mid-myocardial layer of the 6 basal segments and tracking the position of the sample volume throughout the cardiac cycle. An average peak value from 3 consecutive curves was used to calculate the strain and strain rate. The angle of incidence between the transducer and the wall of interest was maintained at 20 to not underestimate the degree of deformation. The calibrated integrated backscatter (cIB) was calculated by measuring the tissue intensity of the pericardium, posterior wall, and anterior septum in the parasternal long-axis view. Automated tissue tracking enabled maintenance of the sample volume of interest within the designated myocardial segment throughout the cardiac cycle. An integrated backscatter curve was derived using commercial software (Echopac, GE Vingmed). The mean

cIB was calculated by subtracting the pericardial integrated backscatter intensity at end-diastole from the integrated backscatter intensity of the posterior wall and the anterior septum, which were then averaged. Interobserver and intraobserver variability were assessed by repeat measurement of the tissue velocity, strain, strain rate, and cIB on the at rest and peak stress images from 10 randomly selected subjects. The original observer (C.J.) was unaware of the previous measurements performed 4 weeks earlier to evaluate the reproducibility. A second observer (R.L.) was unaware of the rst observers results to assess the variability. In addition to the absolute difference between the measurements, the intraclass correlation coefcient was used to determine overall variability. This analysis of reliability was performed using a 2-way random effects model to assess absolute agreement. The results are expressed as the mean SD. The analysis between dened categorical groups (normal vs abnormal) was performed using Students independent t test for continuous variables and the chi-square test for categorical variables. Independent associations between the echocardiographic and metabolic parameters were sought with a stepwise selection method to build logistic or linear regression models of the independent variables. Candidate variables for the models were selected from the unadjusted correlates listed in Table 1 that were signicant at p 0.10 and not co-linear. Statistical analysis was performed using standard software (Statistical Package for Social Sciences, version 16, SPSS, Chicago, Illinois). p Values 0.05 were statistically signicant. Results All 167 subjects had an ejection fraction at rest of 50% and no evidence of inducible ischemia on the exercise stress echocardiogram. Sinus rhythm was maintained throughout testing in all participants. At baseline, 24 subjects had subclinical dysfunction as shown by a reduced at rest Em (septal Em 2 SD of normal for age). Differences were noted between the metabolic parameters of those with abnormal ndings and the 143 subjects with normal at rest myocardial function (Table 1). The subjects with abnormal ndings weighed more, had a greater fasting blood glucose and HbA1c level, and had had a longer duration of T2DM. Probably because of the greater use of statin therapy, the low-density lipoprotein cholesterol level was lower. No difference in serum creatinine was noted between the 2 groups, and the prevalence of microalbuminuria was low. The height-indexed left ventricular mass demonstrated no disparity between the 2 groups. In a logistic regression model, weight (odds ratio [OR] 1.02, p 0.04) and HbA1c (OR 1.36, p 0.027), but not the duration of T2DM, were associated with dysfunction at rest. On linear regression analysis, the duration of T2DM ( 0.214, p 0.006) was an independent correlate of the Em at rest and weight was of borderline signicance ( 0.148, p 0.055). The analysis of the exercise responses of the same groups showed those with abnormal myocardial dysfunction at rest displayed reduced exercise capacity. In addition to a reduced Em, this group had a low Sm at rest. The Em, Sm, and longitudinal diastolic functional reserve index were all signicantly reduced with exercise, although no differences

618

The American Journal of Cardiology (www.ajconline.org)

Figure 1. Abnormal stress Em response associated with signicantly greater cIB than normal stress Em.

were found in deformation or tissue intensity (Table 1). Logistic regression modeling of these baseline groups and the exercise response demonstrated that exercise capacity (OR 0.838, p 0.034) and longitudinal diastolic functional reserve index (OR 0.689, p 0.001) were inversely associated with myocardial dysfunction at rest. Contractile reserve, as measured by the change in ejection fraction or Sm with exercise, was similarly preserved in both groups, irrespective of their baseline Em magnitude (Table 1). Of the 143 patients with normal function at rest, 70 had inducible myocardial dysfunction after exercise (stress Em 9.9 cm/s). No difference was seen between the metabolic and physiologic markers of those with abnormal stress Em compared to those with a normal stress Em. However, patients with an abnormal stress Em were older and had lower at rest and peak heart rates. No difference in microalbuminuria or the height-indexed left ventricular mass was noted between the 2 groups. The antihypertensive regimens were similar between the 2 groups with respect to angiotensin-converting enzyme inhibitors and angiotensinreceptor blockers. -Blocker use was very low overall but marginally greater in the abnormal stress response group (Table 1). Linear regression modeling showed that the peak heart rate was an independent associate of stress Em ( 0.413, p 0.001). Patients with an abnormal stress response to exercise had a signicantly smaller longitudinal diastolic functional reserve index than those without inducible myocardial dysfunction (Table 1). When corrected for total exercise capacity, the association between a reduced longitudinal diastolic functional reserve index and abnormal Em at

Table 2 Mean difference in measurements between same and different observers Same Observer At rest early diastolic tissue velocity (cm/s) Stress early diastolic tissue velocity (cm/s) At rest strain (%) Stress strain (%) At rest strain rate (s1) Stress strain rate (s1) Calibrated integrated backscatter (dB) 0.1 0.4 0.5 1.3 1.7 1.8 0.3 1.9 0.1 0.2 0.02 0.3 0.02 4.7 Between Observers 0.6 1.3 1.7 3.0 0.7 3.5 0.5 4.2 0.1 0.3 0.1 0.4 0.2 5.7

rest and abnormal stress Em was maintained. No relation was displayed between reduction of longitudinal diastolic functional reserve index and the derangement of metabolic parameters. However, a reduced longitudinal diastolic functional reserve index was independently predicted by a reduced peak heart rate ( 0.27, p 0.001). The contractile reserve was preserved in both groups after exercise; however, this was of a signicantly greater magnitude in the normal stress Em group than in the abnormal stress Em group for both Sm and the change in ejection fraction (Table 1). Neither the strain or strain rate at rest nor the stress strain rate were associated with an abnormal at rest Em or stress Em. The independent metabolic associations of impaired strain at rest were weight ( 0.16, p 0.035), HbA1c ( 0.16, p 0.03), and triglycerides ( 0.17, p 0.03). No relation between the metabolic parameters and the strain rate at rest was identied. Peak strain was not strongly

Miscellaneous/Subclinical Diabetic Myocardial Disease

619

Figure 2. Comparison between baseline at rest and peak stress color TDI tissue velocity curves for patients with normal (patient A) and abnormal (patient B) Em responses to exercise stress. White arrows indicate Em peak; abnormal stress Em dened as 9.9 cm/s; both patients had normal Em at rest for age.

associated with the metabolic parameters, with only renal function independently related (creatinine, 0.20, p 0.01). The peak strain rate was independently associated with age ( 0.18, p 0.03), peak heart rate ( 0.21, p 0.01), and systolic blood pressure at rest ( 0.18, p 0.02). No difference was seen in the cIB between those with a normal versus an abnormal Em at rest. However, those with an abnormal stress Em demonstrated a signicantly greater cIB than those with a normal stress Em on univariate analysis (Figure 1). Differences in the cIB persisted when corrected for the use of potential antibrotic agents (angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers), although statistical signicance was lost (16 6 dB vs 18 5 dB, p 0.081), likely related to the large proportion of subjects excluded. The peak heart rate (OR 0.97, p 0.002) and cIB (OR 1.1, p 0.04), but not age or heart rate at rest, were independently predictive of myocardial dysfunction with exercise. A greater cIB was independently associated with a lower Em at rest ( 0.16, p 0.04) and greater waist/hip ratio ( 0.29). The interobserver and intraobserver measurements showed good concordance, with the results reproducible and without signicant variability from original observations. Compared to the original results, no signicant dif-

ference was seen in the at rest Em, stress Em, at rest strain, stress strain, at rest strain rate, stress strain rate, or cIB when remeasured by the original observer or second observer (Table 2). The overall reliability of the inter- and intraobserver measures was high, with an intraclass correlation of 0.986 (95% condence interval 0.981 to 0.991). Discussion The intensity of metabolic disturbances in those with T2DM has been associated with subclinical myocardial dysfunction for both diastolic and systolic parameters at rest but was unrelated to the stress response. Instead, abnormal stress responses appear to be associated with myocardial properties consistent with structural change. Diabetic heart disease is characterized by myocardial collagen deposition and myobrillar hypertrophy in the absence of valvular, congenital, hypertensive, or ischemic heart disease. Unlike a focal scar in ischemic cardiomyopathy, this brosis is a reactive and labile process governed by extrinsic, primarily metabolic, factors such as blood glucose control, accumulation of advanced glycation end products, lipid prole, blood pressure, and obesity. Many of these variables result in stimulation of inammatory pathways and activation of cytokines and the renin-angiotensin-aldosterone system,

620

The American Journal of Cardiology (www.ajconline.org)

leading to interstitial and perivascular brosis. Over time, the increasing myocardial collagen burden leads to increased ventricular stiffness and diastolic dysfunction. In addition to brosis, microvascular ischemia and cardiac autonomic neuropathy likely play important roles in the constellation of pathologic processes that combine to cause myocardial dysfunction in the setting of T2DM. The correlates of the imaging ndings at rest are supportive of previous ndings relating impaired Sm to metabolic derangement.6 Independent metabolic correlates of subclinical myocardial dysfunction have included HbA1c, weight, triglycerides, and the duration of diabetes. These ndings agree with the previously documented deleterious effects of chronic hyperglycemia, obesity,9 and hypertriglyceridemia10 on myocardial function. The closer relation between weight and Em at rest, rather than body mass index, likely reects the increasingly recognized detrimental effect of visceral adiposity in T2DM that has been recently linked to increased cardiovascular risk and mortality.11 The association between a reduction in the sensitive myocardial parameters at rest such as Em and Sm and poorer metabolic control in the present study might reect the increased stiffness and impaired myocardial relaxation associated with increased collagen deposition in the setting of myocardial brosis. Previous histologic studies have supported this nding by demonstrating exaggerated myocardial collagen deposition in the settings of hyperglycemia and arterial hypertension.12 Our ndings appear unrelated to both systolic blood pressure and left ventricular mass. Although no difference was found in the cIB between the designated at rest Em groups, a relation between the metabolic parameters, including abdominal obesity, and cIB was noted. This also supports a link between diabetic metabolic derangement and myocardial brosis. Our results lend to the speculation that increased myocardial stiffness secondary to brosis might play a more important role during exercise stress than at rest and, conversely, that the at rest metabolic disturbances are less relevant to the stress response. Because early diastole is primarily dependent on myocardial relaxation to maximize ventricular lling, it can be expected that diastolic parameters such as Em will be affected earlier in the disease process of brosis than the systolic markers. Hence, using Em as a measure of myocardial relaxation might be a more sensitive parameter of very early myocardial dysfunction, particularly in the asymptomatic population. The results of the present study have demonstrated that myocardial dysfunction unmasked by exercise (dened by an impaired stress Em) could be identied in approximately 50% of asymptomatic patients with T2DM and normal resting function (Figure 2). As expected, a signicant proportion of those with an impaired diastolic stress response also clearly demonstrated a reduced myocardial systolic response to exercise with a reduction in the stress Sm and impaired contractile reserve. In addition to the TDI parameters, myocardial reectivity measured using cIB can be employed noninvasively to characterize the myocardial tissue for evidence of collagen deposition, with a greater cIB (less negative) score indicative of increased myocardial reectivity as a measure of greater brosis.13 The correlation between backscatter and histo-

logically quantied collagen has been previously validated.14 Within our T2DM population, an inverse association was found between cIB and stress Em, with a greater cIB associated with exercise-induced myocardial dysfunction despite normal function at rest. This supports the hypothesis that brosis is responsible for the impaired relaxation seen in T2DM even early in the disease process. The cIB at rest might therefore be a useful tool in the prediction of an abnormal exercise response. Exercise capacity is well recognized as a signicant predictor of cardiovascular and all-cause mortality.15 Impaired exercise capacity has previously been shown even in patients with uncomplicated T2DM and appears to be related to reduced peak oxygen consumption, perhaps owing to factors limiting oxygen delivery, rather than glycemic control.16 This reduction in oxygen delivery might result in failure to adequately achieve the metabolic demands of many tissues during stress, including the myocardium and skeletal muscle, thereby reducing the exercise capacity. The failure of oxygen delivery during periods of increased metabolic demand might be related to impaired compensatory circulatory regulation in the setting of cardiac autonomic neuropathy, an underlying disease of the microvascular circulation impeding oxygen delivery to the muscle bed, and also failure to upregulate myocardial uptake of oxygen owing to myocyte replacement with brosis. Impaired exercise capacity has previously been associated with reduced left ventricular diastolic functional reserve in a heterogeneous population with impaired myocardial relaxation at rest.17 Our results support this relation in the T2DM population, with subjects with myocardial dysfunction at rest demonstrating both reduced longitudinal diastolic functional reserve index and reduced exercise capacity. However, a unique nding was that those with normal myocardial function at rest but an abnormal response to exercise stress (abnormal stress Em) had signicantly reduced longitudinal diastolic functional reserve index compared to those with a normal stress response (normal stress Em). These ndings highlight the important role of myocardial diastolic relaxation in maintaining normal myocardial function and exercise capacity. The strongest associations with blunted contractile response to exercise on the TDI parameters were increased patient age and a diminished peak heart rate response. A reduction of the peak heart rate was also associated with impaired longitudinal diastolic functional reserve index. Blunting of the increase in heart rate with exercise has previously been documented in patients with T2DM, with this nding most prominent in association with cardiac autonomic neuropathy involving both the parasympathetic and the sympathetic nervous systems.18 Our ndings suggest that these patients might have early cardiac autonomic neuropathy that is only revealed with maximum stimulation of the sympathetic nervous system at peak exercise. However, this could also be attributable to the concomitant use of rate-controlling medications. This reduction in the peak heart rate might result in reduced maximal cardiac output, which would further impair oxygen delivery during periods of increased metabolic demand. Although metabolic factors correlated with myocardial dysfunction evident at rest, they correlated poorly with

Miscellaneous/Subclinical Diabetic Myocardial Disease

621

myocardial dysfunction unmasked with exercise. The implication is that reliance purely on the metabolic and at rest variables might fail to detect subclinical diabetic heart disease in a proportion of patients. This could result in these patients being incorrectly classied as having normal myocardial function, resulting in them not receiving therapies that could improve their cardiac function or prevent further deterioration, such as exercise, neurohormonal antagonists, and more stringent blood pressure or blood glucose control. Using exercise to unmask patients with subclinical myocardial dysfunction in T2DM, the clinician might have a more sensitive method to detect a potentially at risk population. The present study was primarily limited by its observational nature. Thus, although associations have been found, direct causal relations cannot be attributed. The continuation of prescribed hypoglycemic, antihypertensive, and antilipid therapy during the present study was unavoidable. However, these agents were not more prevalent in the groups without myocardial dysfunction. The use of noninvasive functional stress echocardiography to exclude macrovascular ischemic heart disease means that false-negative ndings were possible. However, our center has both high sensitivity and specicity in stress echocardiogram interpretation compared to coronary angiography,19 making it unlikely that hemodynamically signicant coronary lesions were underappreciated. Invasive coronary angiography of this asymptomatic population for the purposes of the present observational study could not be justied on ethical grounds.
1. Alberti KG, Zimmet P, Shaw J. The metabolic syndromea new worldwide denition. Lancet 2005;366:1059 1062. 2. Rubler S, Dlugash J, Yuceoglu YZ, Kumral T, Branwood AW, Grishman A. New type of cardiomyopathy associated with diabetic glomerulosclerosis. Am J Cardiol 1972;30:595 602. 3. Fang ZY, Yuda S, Anderson V, Short L, Case C, Marwick TH. Echocardiographic detection of early diabetic myocardial disease. J Am Coll Cardiol 2003;41:611 617. 4. Ha JW, Lee HC, Kang ES, Ahn CM, Kim JM, Ahn JA, Lee SW, Choi EY, Rim SJ, Oh JK, Chung N. Abnormal left ventricular longitudinal functional reserve in patients with diabetes mellitus: implication for detecting subclinical myocardial dysfunction using exercise tissue Doppler echocardiography. Heart 2007;93:15711576. 5. Fang ZY, Leano R, Marwick TH. Relationship between longitudinal and radial contractility in subclinical diabetic heart disease. Clin Sci Lond 2004;106:53 60. 6. Vinereanu D, Nicolaides E, Tweddel AC, Madler CF, Holst B, Boden LE, Cinteza M, Rees AE, Fraser AG. Subclinical left ventricular dysfunction in asymptomatic patients with type II diabetes mellitus,

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

related to serum lipids and glycated hemoglobin. Clin Sci Lond 2003;105:591599. Sun JP, Popovic ZB, Greenberg NL, Xu XF, Asher CR, Stewart WJ, Thomas JD. Designation of tissue Doppler normal range. In: Marwick TH, Yu CM, Sun JP, eds. Myocardial ImagingTissue Doppler and Speckle Tracking. Massachusetts: Blackwell Publishing; 2007:36 51. Leung DY, Grifn BP, Stewart WJ, Cosgrove DM III, Thomas JD, Marwick TH. Left ventricular function after valve repair for chronic mitral regurgitation: predictive value of preoperative assessment of contractile reserve by exercise echocardiography. J Am Coll Cardiol 1996;28:1198 1205. Wong CY, OMoore-Sullivan T, Leano R, Byrne N, Beller E, Marwick TH. Alterations of left ventricular myocardial characteristics associated with obesity. Circulation 2004;110:30813087. de las Fuentes L, Waggoner AD, Brown AL, Davila-Roman VG. Plasma triglyceride level is an independent predictor of altered left ventricular relaxation. J Am Soc Echocardiogr 2005;18:12851291. Czernichow S, Kengne AP, Huxley RR, Batty GD, de Galan B, Grobbee D, Pillai A, Zoungas S, Marre M, Woodward M, Neal B, Chalmers J. Comparison of waist-to-hip ratio and other obesity indices as predictors of cardiovascular disease risk in people with type-2 diabetes: a prospective cohort study from ADVANCE. Eur J Cardiovasc Prev Rehabil Epub 2010 Jul 12. Di Bello V, Talarico L, Picano E, Di Muro C, Landini L, Paterni M, Matteucci E, Giusti C, Giampietro O. Increased echodensity of myocardial wall in the diabetic heart: an ultrasound tissue characterization study. J Am Coll Cardiol 1995;25:1408 1415. Pardo Mindan FJ, Panizo A. Alterations in the extracellular matrix of the myocardium in essential hypertension. Eur Heart J 1993; 14(Suppl J):1214. Picano E, Pelosi G, Marzilli M, Lattanzi F, Benassi A, Landini L, LAbbate A. In vivo quantitative ultrasonic evaluation of myocardial brosis in humans. Circulation 1990;81:58 64. Snader CE, Marwick TH, Pashkow FJ, Harvey SA, Thomas JD, Lauer MS. Importance of estimated functional capacity as a predictor of all-cause mortality among patients referred for exercise thallium single-photon emission computed tomography: report of 3,400 patients from a single center. J Am Coll Cardiol 1997;30:641 648. Regensteiner JG, Sippel J, McFarling ET, Wolfel EE, Hiatt WR. Effects of non-insulin-dependent diabetes on oxygen consumption during treadmill exercise. Med Sci Sports Exerc 1995;27:875 881. Ha JW, Choi D, Park S, Choi EY, Shim CY, Kim JM, Ahn JA, Lee SW, Oh JK, Chung N. Left ventricular diastolic functional reserve during exercise in patients with impaired myocardial relaxation at rest. Heart 2009;95:399 404. Bottini P, Tantucci C, Scionti L, Dottorini ML, Puxeddu E, Reboldi G, Bolli GB, Casucci G, Santeusanio F, Sorbini CA. Cardiovascular response to exercise in diabetic patients: inuence of autonomic neuropathy of different severity. Diabetologia 1995;38:244 250. Ingul CB, Stoylen A, Slordahl SA, Wiseth R, Burgess M, Marwick TH. Automated analysis of myocardial deformation at dobutamine stress echocardiography: an angiographic validation. J Am Coll Cardiol 2007;49:16511659.

Specic Characteristics of Sudden Death in a Mediterranean Spanish Population

M. Teresa Subirana, MDa,*, Josep O. Juan-Babot, MD, PhDb, Teresa Puig, MD, PhDc, Joaqun Lucena, MD, PhDd, Antonio Rico, MD, PhDd, Manuel Salguero, MD, PhDe, Juan C. Borondo, MDf, Jorge Ordez, MD, PhDg, Josep Arimany, MD, PhDh, Rafael Vzquez, MD, PhDi, Lina Badimon, MD, PhDb, Gaetano Thiene, MDj, and Antonio Bays de Luna, MD, PhDb
Most of the data reported on sudden cardiac death has been from studies of Anglo-Saxon patients. We conducted a study to ascertain the relation between sudden death (SD) and some epidemiologic, clinical, and biochemical parameters and to assess the coronary histopathologic aspects of subjects in a Spanish population who had died suddenly. A total of 204 subjects (86% men), aged 12 to 80 years (mean 54 15), who had died from out-of-hospital natural SD were evaluated. Only 15% of subjects had been previously diagnosed with heart disease. Pathologic evidence of underlying cardiovascular disease was found in 90% of cases, with coronary heart disease (CHD) the most frequent (58%). The CHD was acute coronary thrombosis in 41% and a stable plaque with luminal narrowing of >75% in 59%. An old myocardial infarction was found in 31% of the SD victims. Cardiac hypertrophy was found in 48%, with no relation between the presence of cardiac hypertrophy and CHD. Patients with stable plaques had a greater heart weight than did those with acute coronary thrombosis (p 0.02). Male gender, older age, smoking, and low-density lipoprotein cholesterol/high-density lipoprotein cholesterol ratio of >3 were associated with CHD. A greater percentage of patients with an eroded and/or ruptured plaque than patients with a stable plaque were smokers. Only smoking and a low-density lipoprotein/high-density lipoprotein cholesterol ratio of >3 were associated with an eroded and/or ruptured plaque. In conclusion, compared with the ndings from studies of AngloSaxon patients, a lower incidence of CHD and acute coronary thrombosis and a greater incidence of cardiac hypertrophy were found in SD victims of a Mediterranean Spanish population. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:622 627) Sudden death (SD) represents 12% to 13% of overall natural mortality when the temporal denition is restricted to death occurring 2 hours after the onset of symptoms,1 4 with approximately 50% of deaths in patients with cardiovascular disease.57 Sudden cardiac death (SCD) represents about 80% to 90% of all SDs.2,3,8 Therefore, SCD constitutes one of the most important challenges of modern cardiology. Our understanding of the pathophysiologic mechanisms of SCD, as well as the correlation between SCD and associated diseases and risk factors, has mainly been based on data from studies of white Anglo-Saxon patients that showed coronary heart disease (CHD) to be a fundamental cause of SCD.9,10 The Seven Countries Study demonstrated in a 25-year follow-up period that Southern European cohorts had a lower risk of fatal CHD than other European or United States cohorts.11 However, only partial data on the incidence of SCD in Spain are available,8 and it is unknown whether its clinical and pathologic characteristics differ from those reported from Anglo-Saxon countries.12 The present study was designed to investigate the relation between SCD and different epidemiologic, clinical, and biochemical parameters in a Spanish population and to characterize the coronary histopathologic features of the subjects who died suddenly. Methods A total of 204 victims of out-of-hospital natural SD aged 12 to 80 years from 2 Spanish autonomous communities, Catalonia and Andalusia, were included. Those who had died long after an aborted SD were excluded. Autopsy was performed within 18 hours after death. A
www.ajconline.org

Department of Cardiology, Hospital de la Santa Creu i Sant Pau, Barcelona, and Universitat Autnoma de Barcelona, Barcelona, Spain; b Cardiovascular Research Center, CSIC-ICCC, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain; cDepartment of Epidemiology, Hospital de la Santa Creu i Sant Pau, Barcelona, and Universitat Autnoma de Barcelona, Spain; dForensic Pathology Service, Institute of Legal Medicine, Seville, Spain; eNational Institute of Toxicology and Forensic Sciences, Seville, Spain; fNational Institute of Toxicology and Forensic Sciences, Barcelona, Spain; gDepartment of Biochemistry, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain; hInstitute of Legal Medicine of Catalonia, Barcelona, Spain; iDepartment of Cardiology, Hospital Universitario Nuestra Seora de Valme, Seville, Spain; and jDepartment of Medico-Diagnostic Sciences and Special Therapies, University of Padua Medical School, Padua, Italy. Manuscript received July 16, 2010; manuscript received and accepted October 5, 2010. This work was supported by a grant from Redes temticas de investigacin cooperativa. Instituto de Salud Carlos III (G03-078). *Corresponding author: Tel: (34) 93-556-5945; fax: (34) 93-556-5603. E-mail address: msubiranad@santpau.cat (M.T. Subirana). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.028

Miscellaneous/Sudden Death in a Mediterranean Population

623

forensic autopsy protocol13,14 was applied. Femoral blood and vesical or pelvic kidney urine samples were obtained by puncture. The coronary arteries and myocardium were specically studied. Information on the sociodemographic data, cardiovascular risk factors, and present and previous disorders suggesting cardiovascular disease in the victims and their relatives was obtained from the close family or friends by telephone questionnaire. The cardiovascular risk factors were collected as continuous (i.e., height, weight, body mass index), categorical (i.e., yes, no, exsmoker for smoking), and dichotomous (i.e., regular alcohol intake, physical activity, diabetes, hypertension, dyslipemia, and the use of psychoactive drugs) variables. In the case of previously recognized heart disease, an attempt was made to compare the data obtained from the relatives with the data retrieved from the hospital records. SD was dened as a natural, nonviolent, unexpected death occurring within 1 hour of the onset of symptoms or within 24 hours of a previously stable medical condition, if the event had not been witnessed. Death from CHD was diagnosed when cross-sectional luminal narrowing of a major coronary artery of 75% or an acute thrombosis related to rupture or erosion of a coronary plaque was found. The culprit plaque was dened as that with an acute thrombus or, in its absence, that with the greatest degree of cross-sectional luminal narrowing relative to the internal elastic lamina at the narrowest segment. An acute ruptured plaque consisted of a continuous luminal thrombus with an underlying lipid-rich core. When the thrombus was in direct contact with the intimal layer, without rupture of a lipid pool, the plaque was dened as eroded. Vulnerable plaques were dened as having a brous cap 65 m with macrophage and T-lymphocyte inltration. Stable plaques were dened as those causing luminal narrowing of 75% in the absence of luminal thrombosis and were considered vulnerable or nonvulnerable.12 A heart weight 450 g in men and 400 g in women was considered cardiac hypertrophy.15,16 We used the heart weight, instead of left ventricular thickness, because the heart rate might provide more information about the cardiac mass. In dilated hearts with an abnormal cardiac mass, the left ventricular thickness can sometimes be normal. The smoking status of each patient was classied as current daily smoker,17 nonsmoker, or exsmoker. Hypertension was dened according to the guidelines of the European Society of Hypertension/European Society of Cardiology.18 The heart tissue was xed in formaldehyde by retrograde perfusion at systemic pressure. The coronary arteries were dissected and embedded in parafn, and 5-mthick sections were stained with hematoxylin-eosin. The coronary arteries were studied by serial sectioning at 3-mm intervals after decalcication. Any segment showing cross-sectional luminal narrowing of 50% was studied histologically. Histologic images were studied in a Leica MZ-9.5 (Leica Microsistemas, Barcelona, Spain) stereomicroscope to quantify the stenotic area. Image capture and morphometric study were performed using a Sony 3CCD color video camera and processed using Visilog (Sony

Table 1 Pathologic ndings associated with sudden death (SD) (n 204) Cardiovascular disease Heart disease Coronary heart disease Hypertensive left ventricular hypertrophy Valvular heart disease Idiopathic left ventricular hypertrophy Dilated cardiomyopathy Hypertrophic cardiomyopathy Arrhythmogenic right ventricular dysplasia/cardiomyopathy Myocarditis Congenital heart disease Amyloidosis Vascular disease Pulmonary embolism Aortic dissection Cerebral hemorrhage Noncardiovascular disease 183 161 119 (58%) 20 (10%) 5 (2%) 4 (2%) 4 (2%) 3 (2%) 3 (2%) 1 (1%) 1 (1%) 1 (1%) 22 8 (4%) 9 (4%) 5 (2%) 7

ESPAC, Barcelona, Spain), version 4.1.5 (Noesis, Saint Aubin, France), software. Labeled blocks from a representative transverse slice of the anterior, lateral, and posterior free wall of the left ventricle, posterior free wall of the right ventricle, and anterior and posterior interventricular septum and 1 block from each atrium were taken for study of the myocardium. In addition, any area with a signicant macroscopic abnormality was sampled and analyzed using hematoxylin-eosin, Massons trichrome, and Van Gieson stains. Two investigators, using a double-headed light microscope, performed the analysis simultaneously. From the blood samples, the total cholesterol, triglycerides, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, very-LDL cholesterol, chylomicrons, apolipoproteins B and CIII, and lipoprotein (a) were measured. The cotinine and glucose levels were obtained from the urine samples (Roche Diagnostics, San Cugat del Valls, Barcelona, Spain and DRG Diagnostics, Marburg, Germany). Descriptive analyses were initially performed. The quantitative variables are reported using the mean and standard deviation. Relations between categorical variables were studied using the chi-square test. A comparison of the quantitative variables between the 2 groups was performed using the t test and of ordinal variables using the Mann-Whitney nonparametric U test. The variables analyzed univariately by logistic regression analysis to predict coronary artery disease and type of atherosclerotic plaques included gender, age, body mass index (24.9 vs 25 to 29.9 vs 30 kg/m2), presence or absence of smoking, hypertension, diabetes, alcohol intake, physical activity (yes vs no), diabetes (yes vs no) and LDL/HDL cholesterol ratio (3 vs 3). Variables showing statistical signicance (p 0.10) were included in the multivariate regression model to determine which were independently related to the prognosis. In all analyses, contrasts were made bilaterally with an of 5% (Statistical Package for Social Sciences, version 15.0, SPSS, Chicago, Illinois).

624

The American Journal of Cardiology (www.ajconline.org)

Table 2 Biologic risk factors for sudden death (SD) cases stratied by heart disease status and culprit plaque type Variable CHD and Non-CHD CHD (n 119) Non-CHD (n 85) Culprit Plaque Type

p Value Eroded and/or Ruptured Stable (75%) and/or Vulnerable p Value (n 49) (n 70) 0.001 0.05 0.05 0.001 0.001 0.01 0.27 0.61 0.001 0.001 0.90 0.31 6.9 2.2 2.6 1.5 1.0 0.4 4.4 1.4 4.9 2.3 1.0 0.7 1.6 1.3 0.7 0.6 1.3 0.4 0.2 0.1 30.0 80.5 71% 6.1 2.0 2.3 1.5 1.1 0.4 3.6 1.3 3.9 2.0 0.8 0.6 1.8 1.0 0.7 0.5 1.1 0.3 0.1 0.1 18.3 45.5 37% 0.05 0.23 0.40 0.01 0.008 0.08 0.29 0.88 0.01 0.33 0.14 0.001

Cholesterol (mmol/L) 6.5 2.1 5.2 1.8 Triglycerides (mmol/L) 2.5 1.5 2.0 1.2 High-density lipoprotein cholesterol (mmol/L) 1.0 0.4 1.2 0.5 Low-density lipoprotein cholesterol (mmol/L) 3.9 1.4 3.2 1.3 Low-density lipoprotein/high-density 4.3 2.2 3.1 1.8 lipoprotein cholesterol ratio Very-low-density lipoprotein cholesterol 0.9 0.6 0.7 0.6 (mmol/L) Chylomicrons (mmol/L) 1.7 1.1 1.1 0.5 Lipoprotein (a) (g/L) 0.7 0.6 0.6 0.5 Apolipoprotein B (g/L) 1.2 0.4 0.9 0.4 Apolipoprotein CIII (g/L) 0.2 0.1 0.1 0.1 Urine glucose (mmol/L) 24.1 64.9 10.7 44.5 Urine cotinine (positive) 52% 43%

All parameters were measured from blood samples, unless indicated otherwise.

Results A total of 204 subjects (86% males), who had died from out-of-hospital natural SD, were evaluated. Of the 204 subjects, 175 were males (86%) and 29 were females, with a mean age of 54 15 years (males 53 15 years; females 61 13 years; p 0.014). Only 21 subjects were 35 years (range 12 to 34). The mean body mass index was 30 8 kg/m2 and was greater for the females than for the males (34 10 vs 29 8 kg/m2; p 0.024). Of the 204 subjects, 58% (62% of males and 27% of females) were smokers, 52% had a history of regular alcohol intake, 39% had a history of hypertension, 35% a history of dyslipemia, and 18% had a history of diabetes. The smokers were younger than the nonsmokers (52 13 vs 57 18 years, respectively; p 0.001). Those with hypertension were older than those without (60 13 vs 49 16 years, respectively; p 0.0001). The mean age of those with an LDL/HDL cholesterol ratio of 3 was 54 16 years and was 55 13 years for those with an LDL/HDL cholesterol ratio 3 (p 0.59). Of the subjects with diabetes mellitus, the mean age was 61 12 years compared to 52 16 years for those without diabetes mellitus (p 0.001). The urine cotinine level was measured in 75% of the subjects. A high association (87%) was found between cotinine present in the urine and data on positive smoking status. A history of cardiovascular symptoms was found in 33% of the subjects (60% females), including dyspnea, angina, and/or syncope, but only 15% had been previously diagnosed with heart disease (15% males and 13% females), with documented myocardial infarction in 10%. Of the 204 subjects, 14% had a family history of SD and 24% had a rst-degree relative who had had myocardial infarction. Most deaths occurred while the subject was resting or doing mild exercise (71%), and, from the information provided by the family and relatives, 20% of those whose death was witnessed had complained of chest pain.

Figure 1. Culprit plaques. (A) Stable plaque. (B) Vulnerable plaque. (C) Eroded plaque. (D) Ruptured plaque. Macroscopic and microscopic images. Hematoxylin-eosin stain, original magnication 60.

Of the 204 cases, 183 (90%) could have been related to underlying cardiovascular disease. Heart disease was found in 161 subjects (79%), with CHD the most frequent (58%) and signicantly different (p 0.016) between the males (62%) and females (38%). Hypertensive cardiac hypertrophy was found in 20 subjects (9.9%), aortic dissection in 9 (4.4%), and pulmonary embolism in 8 (3.9%). In 14 subjects, the cause of death could not be ascertained (Table 1). The mean heart weight was 498 123 g (454 115 g in females and 506 123 g in males; p 0.038). Cardiac hypertrophy was diagnosed in 41% of the females and 49% of the males (48% of those with SD). In the males, a signicant relation (p 0.0001) was found between cardiac hypertrophy and a history of hypertension, with 75% of those with hypertension versus 34% of those without, having a hypertrophic heart. No relation was found between cardiac hypertrophy and CHD. On histologic study of the myocardium, a scar from an old, healed, myocardial infarction was found in 64 subjects (31% of all those with SD) and was more frequent in male patients 60 years old (47% vs 29%; p 0.015). No differences related to age were found in the female patients.

Miscellaneous/Sudden Death in a Mediterranean Population Table 3 Uni- and multivariate logistic regression analysis of signicant coronary heart disease (CHD) Variable CHD (n 119) 91% 57 12 45% 63% 22% 43% 35% 74% 20% 44% 69% Non-CHD (n 85) 79% 50 18 32% 50% 30% 28% 42% 44% 14% 40% 64% p Value Univariate 0.018 0.001 0.11 0.08 0.33

625

Multivariate 0.017 0.001 0.038

Male gender Age (years) Hypertension Cigarette smokers Body mass index (kg/m2) 24.9 2529.9 30 Low-density lipoprotein/high-density lipoprotein cholesterol ratio 3 Diabetes mellitus Physical activity Alcohol consumption Data are presented as % or mean standard deviation.

0.001 0.37 0.79 0.33

0.001

Table 4 Uni- and multivariate logistic regression analysis of culprit plaque type Variable Eroded and/or Ruptured Plaque (n 49) 94% 54 12 33% 76% 21% 45% 33% 87% 16% 37% 72% Stable (75%) and/or Vulnerable Plaque (n 70) 89% 59 12 53% 54% 23% 42% 36% 63% 23% 50% 67% p Value Univariate 0.33 0.04 0.06 0.03 0.89 Multivariate 0.04

Male gender Age (years) Hypertension Cigarette smokers Body mass index (kg/m2) 24.9 2529.9 30 Low-density lipoprotein/high-density lipoprotein cholesterol ratio 3 Diabetes mellitus Physical activity Alcohol consumption Data are presented as % or mean standard deviation.

0.004 0.38 0.23 0.37

0.005

Finally, 46% of these cases did not meet the criteria (eroded or ruptured plaque/luminal cross-sectional narrowing 75%) to consider SD related to CHD. The coronary arteries were evaluated in all SD subjects. A single vessel (left anterior descending coronary artery, circumex coronary artery, or right coronary artery) was affected in 43 subjects (36%); 2 vessels in 48 (40%), and 3 vessels in 28 (24%). No females had the main left coronary artery affected, and in only 15 cases (14% of males) was this vessel affected. In 74% of the subjects with CHD, the left anterior descending coronary artery was affected. The biologic risk factors for SCD according to a diagnosis of CHD or non-CHD and the type of culprit plaque are listed in Table 2. Regarding coronary plaque morphology, an eroded (53%) and/or ruptured plaque (47%) was observed as the culprit plaque in 49 subjects (41%). In the remaining (59%), a stable plaque with cross-sectional luminal narrowing of 75% was considered the culprit plaque, with a vulnerable anatomy in 4 (Figure 1). An old myocardial infarction was present in 45% of those with an eroded or ruptured culprit plaque and in 60% of those with a stable culprit plaque.

A relation was found between the type of culprit plaque and heart weight, with patients with stable plaques having a greater cardiac weight (528 119 g vs 479 99 g; p 0.025). The same relation was found when a healed myocardial infarction was present in either of these groups (564 122 g vs 491 107 g; p 0.017). In hearts with acute thrombosis, the mean heart weight was 511 90 g for those with a ruptured plaque and 451 100 g for those with an eroded plaque (p 0.023). An eroded or ruptured plaque was found in 33% of those with a history of hypertension versus 67% of those without hypertension (p 0.04) and in only 13% of the group with cardiac hypertrophy and a history of hypertension. The patients with CHD were signicantly older than those without (57 12 vs 50 18 years; p 0.01). The CHD prevalence was signicantly greater in the males than in the females (91% vs 9%; p 0.01). Regarding biologic cardiovascular risk factors, no signicant differences were found between SCD with CHD and SCD without CHD, except for apolipoprotein B, apolipoprotein CIII, and the lipid prole. HDL cholesterol,

626

The American Journal of Cardiology (www.ajconline.org)

LDL cholesterol, and very-LDL cholesterol were signicantly greater in those with CHD (Table 2). From the multivariate regression analysis, only male gender, older age, smoking, and an LDL/HDL cholesterol ratio of 3 were signicantly associated with CHD (Table 3). Those with an eroded and/or ruptured culprit plaque were younger (54 12 vs 59 12 years; p 0.04) and more likely to be smokers (76% vs 54%; p 0.03) than those with a stable culprit plaque. Using univariate regression analysis, younger age, smoking history, and an LDL/ HDL cholesterol ratio of 3 were associated with an eroded and/or ruptured culprit plaque (acute thrombosis). However, on multivariate regression analysis, only smoking history (odds ratio 2.5; p 0.04) and an LDL/HDL cholesterol ratio of 3 (odds ratio 4.2; p 0.005) were associated with these types of plaque (Table 4). Furthermore, using multivariate regression analysis, no signicant difference between a ruptured or an eroded plaque and the cardiovascular risk factors studied was found. Discussion In Anglo-Saxon countries, CHD has been the underlying cause of SD in 80% to 90% of cases.19,20 The incidence of SCD in Spain has been estimated to be one of the lowest in the industrialized countries.8,21 However, the prevalence of cardiovascular risk factors in the Mediterranean area22,23 is not as low as one might expect. The results of the present study have provided epidemiologic and, in particular, anatomopathologic information on SD in Spain that might explain the differences in SD between Anglo-Saxon and some Mediterranean countries; these differences are probably related to lifestyle and environment.24 As reported in other studies,25,26 a signicant observation was the high percentage of those who died from SD who had a history of SD (14%) or myocardial infarction (24%) in rst-degree relatives. Logically, this would support the idea of a genetic factor involved in SD and, in particular, CHD, as recently reported.27 A case-control study by Friedlander et al25 revealed that a family history of acute myocardial infarction or SD was more common among those who had died from SD than in control subjects, with this association mostly independent of other common risk factors with familial aggregation. According to the information provided by the family and relatives, only 20% of those who had died from SD had complained of chest pain, lower than the 37% reported in the Maastrich study.28 This is consistent with the lower incidence of underlying CHD in our study population. It might indicate that SD could be the rst manifestation of cardiovascular disease, as has been reported by other studies,9,29 making it difcult to establish methods of preventing SD in the general population. In the present study, 90% of cases were associated with cardiovascular disease, with CHD the most frequent. Compared to the ndings from studies of Anglo-Saxon patients, we found a clearly lower incidence of CHD (58% vs 80% to 90%9) and acute coronary thrombosis (41% vs 52%12). These ndings support the lower incidence of acute coronary syndrome reported in the Mediterranean area, long considered a consequence of diet30 and/or, in a broader

aspect, the Mediterranean culture. However, the involvement of factors such as genetics should also be considered. It is well known that when examining the same levels of cholesterol, the incidence of myocardial infarction has been lower in Spain than in Anglo-Saxon countries.22,31 In contrast to what has been reported in Anglo-Saxon populations, we found a greater percentage of cardiac hypertrophy without signicant disarray (48% vs 13 to 15%).32,33 In our male subjects, this was related to hypertension (p 0.001). The Massa Ventricolare Sinistra Nellipertensione Arteriosa study34 showed the strong, continuous, and independent relation between the left ventricular mass and subsequent cardiovascular morbidity, including SD. Pathologic signs of CHD were found predominantly in our male subjects (62% vs 38%); however, a low number of female subjects were included in the present study. In 59% of those with CHD, a stable plaque was considered the culprit plaque responsible for SD and a healed myocardial infarction was found in 60% of these cases. This might suggest that if acute myocardial ischemia is a cause of SD, arrhythmia in the setting of myocardial scars could also be a very important component. A relation was found between stable plaque and cardiac hypertrophy, with patients with stable plaques having a greater cardiac mass (528 19 vs 479 99 g; p 0.025). As with the ndings from Burke et al,12,35 our study showed a lower frequency of acute coronary thrombosis in patients with cardiac hypertrophy and a history of hypertension. An eroded or ruptured plaque was found in only 33% of those with a history of hypertension compared to 67% of those without hypertension. Regarding the remaining coronary risk factors, a significant relation was found between CHD and male gender, older age, smoking, and LDL/HDL cholesterol ratio of 3. When we attempted to analyze its inuence on the plaque type, patients with eroded and/or ruptured plaques were found to be younger (p 0.04), more likely to be smokers (p 0.03), and to have a greater probability of an LDL/ HDL cholesterol ratio of 3 (p 0.004). In contrast, gender, diabetes, regular alcohol intake, and hypertension could not be related to the type of plaque. The possible limitations of the present study included that, although it was performed prospectively, it was not possible to perform autopsy studies every day, rendering it impossible to obtain exact information on the incidence and prevalence of SD in this population. Nevertheless, we believe our ndings can provide important epidemiologic and anatomopathologic information and offer us the possibility of establishing comparative data with the data from AngloSaxon countries. Also, in accordance with previous reports,36 we found a lower prevalence of CHD in females with SD (Table 3). Nevertheless, we must emphasize the low number of females subjects included in our study. Finally, although we have previously justied the use of heart weight instead of the left ventricular wall thickness as a variable, this could be considered a possible limitation. Therefore, we conducted a supplementary analysis to study the relation between the thickness of the left ventricle and the presence of coronary disease, without nding a relation between these 2 variables. Moreover, a signicant linear

Miscellaneous/Sudden Death in a Mediterranean Population

627

correlation (r2 0.144) was seen between the heart weight and ventricular wall thickness. In 14 cases, no abnormal pathologic ndings were found. At least some of these cases might correspond to an undiagnosed channelopathy. Even though minor structural myocardial abnormalities had been reported in some symptomatic or asymptomatic patients with a channelopathy, such as Brugada syndrome,37 usually no structural alterations will be demonstrated by routine invasive and noninvasive examinations. We believe that 1 of our patients might have presented with Brugada syndrome because in a pre-employment medical examination, an atypical right bundle branch block on the electrocardiogram was reported. Acknowledgment: We thank A. Bartomeu, J. Guindo, and J. Medallo for their excellent collaboration in this study.
1. Engelstein ED, Zipes DP. Sudden cardiac death. In: Alexander RW, Schlant RC, Fuster V, eds. The Heart, Arteries and Veins. 9th ed. New York: McGraw-Hill; 1998:10811112. 2. Myerburg RJ, Castellanos A. Cardiac arrest and sudden death. In: Braunwald E, ed. Heart Disease: A Textbook of Cardiovascular Medicine. 5th ed. Philadelphia: WB Saunders; 1997:742779. 3. Kuller L, Lilienfeld A, Fisher R. An epidemiological study of sudden and unexpected deaths in adults. Medicine (Baltimore) 1967;46:341 361. 4. Schatzkin A, Cupples LA, Heeren T, Morelock S, Kannel WB. Sudden death in the Framingham Heart Study: differences in the incidence and risk factors by sex and coronary disease status. Am J Epidemiol 1984;120:888 899. 5. Demirovic J, Myerburg RJ. Epidemiology of sudden coronary death: an overview. Prog Cardiovasc Dis 1994;37:39 48. 6. Kannel WB, Schatzkin A. Sudden death: lessons from subsets in population studies. J Am Coll Cardiol 1985;5:141B149B. 7. Thiene G, Basso C, Corrado D. Cardiovascular causes of sudden death. In: Silver MD, Gotlieb AI, Schoen FJ, eds. Cardiovascular Pathology. 3rd ed. Philadelphia: Churchill Livingstone; 2001:326 374. 8. Marrugat J, Elosua R, Gil M. Epidemiology of sudden cardiac death in Spain. Rev Esp Cardiol 1999;52:717725. 9. Priori SG, Aliot E, Blomstrom-Lundqvist C, Bossaert L, Breithardt G, Brugada P, Camm AJ, Cappato R, Cobbe SM, Di Mario C, Maron BJ, McKenna WJ, Pedersen AK, Ravens U, Schwartz PJ, Trusz-Gluza M, Vardas P, Wellens HJ, Zipes DP. Task Force on Sudden Cardiac Death of the European Society of Cardiology. Eur Heart J 2001;22:1374 1450. 10. Huikuri HV, Castellanos A, Myerburg RJ. Sudden death due to cardiac arrhythmias. N Engl J Med 2001;345:14731482. 11. Keys A. Coronary heart disease in seven countries. Circulation 1970; 41(Suppl 1):1211. 12. Burke AP, Farb A, Malcom GT, Liang Y-H, Smialek J, Virmani R. Coronary risk factors and plaque morphology in men with coronary disease who died suddenly. N Engl J Med 1997;336:1276 1282. 13. Brinkmann B. Harmonisation of medico-legal autopsy rules. Int J Leg Med 1999;113:114. 14. Basso C, Burke M, Fornes P, Gallagher PJ, de Gouveia RH, Sheppard M, Thiene G, van der Wal A. Guidelines for autopsy investigation of sudden cardiac death. Virchows Arch 2008;452:1118. 15. Linzbach A. Heart failure from the point of view of quantitative anatomy. Am J Cardiol 1960;5:370 382.

16. Grant RP. Aspects of cardiac hypertrophy. Am Heart J 1953;46:154 158. 17. Dobson A, Kuulasmaa K, Moltchanov V, Evans A, Fortmann SP, Jamrozik K, Sans S, Tuomilehto J. Changes in cigarette smoking among adults in 35 populations in the mid 1980s: Who MONICA Project. Tob Control 1998;7:14 21. 18. Guidelines Committee. European Society of Hypertension-European Society of Cardiology guidelines for the management of arterial hypertension. J Hypertens 2003;21:10111053. 19. Kuller LH. Sudden death denition and epidemiologic considerations. Prog Cardiovasc Dis 1980;23:112. 20. Myerburg RJ, Interian A Jr, Mitrani RM, Kessler KM, Castellanos A. Frequency of sudden cardiac death and proles of risk. Am J Cardiol 1997;80:10F19F. 21. Grupo valenciano de Estudios sobre la Muerte Sbita. Muerte sbita en la ciudad de Valencia. Rev Esp Cardiol 1987;40(Suppl):8594. 22. Masi R, Pena A, Marrugat J, Sala J, Vila J, Pavesi M, Covas M, Aub C, Elosua R. High prevalence of cardiovascular risk factors in Gerona, Spain, a province with low myocardial infarction incidence: REGICOR Investigators. J Epidemiol Commun Health 1998;52:707715. 23. WHO Scientic Group. Technical Report Series 726. Sudden Cardiac Death. Geneve: World Health Organization: 1985:525. 24. Keys A, KeysM. In: How to Eat Well and Stay Well. The Mediterranean Way. Garden City, NY: Doubleday; 1975. 25. Friedlander Y, Siscovick DS, Weinmann S, Austin MA, Psaty BM, Lemaitre RN, Arbogast P, Raghunathan TE, Cobb LA. Family history as a risk factor for primary cardiac arrest. Circulation 1998;97:155 160. 26. Jouven X, Desnos M, Guerot C, Ducimetiere P. Predicting sudden death in the population: the Paris Prospective Study I. Circulation 1999;99:1978 1983. 27. Myocardial Infarction Genetics Consortium. Genome-wide association of early onset myocardial infarction with single nucleotide polymorphisms and copy number variants. Nat Genet 2009;41:334 341. 28. de Vreede-Swagemakers JJ, Gorgels AP, Dubois-Arbouw WI, van Ree JW, Daemen MJ, Houben LG, Wellens HJ. Out-of-hospital cardiac arrests in the 1990s: a population based study in the Maastricht area on incidence, characteristics and survival. J Am Coll Cardiol 1997;30: 1500 1505. 29. Lown B. Sudden cardiac death: the major challenge confronting contemporary cardiology. Am J Cardiol 1979;43:313328. 30. de Lorgeril M, Salen P, Martin JL, Monjaud I, Delaye J, Mamelle N. Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: nal report of the Lyon Diet Heart Study. Circulation 1999;99:779 785. 31. Kromhout D. On the waves of the Seven Countries Study: a public health perspective on cholesterol. Eur Heart J 1999;20:796 802. 32. Virmani R, Burke AP, Farb A. Sudden cardiac death. Cardiovasc Pathol 2001;10:211218. 33. Doolan A, Langlois N, Semsarian C. Causes of sudden cardiac death in young Australians. Med J Aust 2004;180:110 112. 34. Verdecchia P, Carini G, Circo A, Dovellini E, Giovannini E, Lombardo M, Solinas P, Gorini M, Maggioni AP. Left ventricular mass and cardiovascular morbidity in essential hypertension: the MAVI study. J Am Coll Cardiol 2001;38:1829 1835. 35. Burke AP, Farb A, Liang Y, Smialek J, Virmani R. Effect of hypertension and cardiac hypertrophy on coronary artery morphology in sudden cardiac death. Circulation 1996;94:3138 3145. 36. Albert CM, McGovern BA, Newell JB, Ruskin JN. Sex differences in cardiac arrest survivors. Circulation 1996;93:1170 1176. 37. Frustaci A, Russo MA, Climenti C. Structural myocardial abnormalities in asymptomatic family members with Brugada syndrome and SCN5A gene mutation. Eur Heart J 2009;30:11763.

Clinical and Prognostic Relevance of Echocardiographic Evaluation of Right Ventricular Geometry in Patients With Idiopathic Pulmonary Arterial Hypertension
Stefano Ghio, MDa,*, Anna Sara Pazzano, MDa, Catherine Klersy, MDb, Laura Scelsi, MDa, Claudia Raineri, MDa, Rita Camporotondo, MDa, Andrea DArmini, MDa, and Luigi Oltrona Visconti, MDa
The aim of the present study was to assess the clinical and prognostic signicance of right ventricular (RV) dilation and RV hypertrophy at echocardiography in patients with idiopathic pulmonary arterial hypertension. Echocardiography and right heart catheterization were performed in 72 consecutive patients with idiopathic pulmonary arterial hypertension admitted to our institution. The median follow-up period was 38 months. The patients were grouped according to the median value of RV wall thickness (6.6 mm) and the median value of the RV diameter (36.5 mm). On multivariate analysis, the mean pulmonary artery pressure (p 0.018) was the only independent predictor of RV wall thickness, and age (p 0.011) and moderate to severe tricuspid regurgitation (p 0.027) were the independent predictors of RV diameter. During follow-up, 22 patients died. The death rate was greater in the patients with a RV diameter >36.5 mm than in patients with a RV diameter <36.5 mm: 15.9 (95% condence interval 9.4 to 26.8) vs 6.6 (95% condence interval 3.3 to 13.2) events per 100-person years (p 0.0442). In contrast, the death rate was similar in patients with RV wall thickness above or below the median value. However, among the patients with a RV wall thickness >6.6 mm, a RV diameter >36 mm was not associated with a poorer prognosis (p 0.6837). In conclusion, in patients with idiopathic pulmonary arterial hypertension, a larger RV diameter is a marker of a poor prognosis but a greater RV wall thickness reduces the risk of death associated with a dilated right ventricle. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:628 632) The aim of the present study was to assess the determinants and prognostic relevance of right ventricular (RV) dilation and RV hypertrophy at echocardiography in patients with idiopathic pulmonary arterial hypertension (IPAH) and to test the hypothesis that a greater RV wall thickness is associated with better circulatory function and a better prognosis in such patients. Methods From July 1996 to March 2009, 72 patients were consecutively admitted to our institution for the evaluation of chronic pulmonary hypertension and were diagnosed with IPAH. The diagnosis was made after having ruling out the known causes of pulmonary hypertension.1 Patients with different etiologies of pulmonary hypertension were excluded. All patients underwent right heart catheterization and ultrasound examination during the hospitalization period. The patients were followed up for a median of 38 months (interquartiles range 14 to 71). During follow-up,
Cardiac, Thoracic, and Vascular Department and bBiometry and Clinical Epidemiology, Fondazione IRCCS Policlinico S. Matteo, Pavia, Italy. Manuscript received July 30, 2010; manuscript received and accepted October 5, 2010. *Corresponding author: Tel: (0039) 382-50-3713; fax: (0039) 382-503159. E-mail address: s.ghio@smatteo.pv.it (S. Ghio). 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.027
a

the patients were treated according to international guidelines. The enrollment of patients expanded over several years; thus, the treatments varied over time. However, the clinical decisions were never made on the basis of the echocardiographic parameters of RV function because this was never recommended by the guidelines. The echocardiographic examinations were performed in the same laboratory using commercially available ultrasound equipments. The complete echocardiographic protocol and intra- and interobserver agreement for the most important echocardiographic parameters have been previously reported.2 All echocardiographic data were averaged for 3 beats. The 2 parameters of interest in the present analysis (i.e., the RV end-diastolic diameter and the thickness of the RV free wall) were determined in the parasternal view (Figure 1).3 A Swan Ganz thermodilution catheter (American Edwards Laboratories, Irvine, California) was inserted transcutaneously by way of the right internal jugular vein. The thermistor was connected to a dedicated computer to display the cardiac output on-line. The following hemodynamic parameters were measured or calculated: systemic blood pressure (arm cuff sphygmomanometer), right atrial pressure, systolic, diastolic, and mean pulmonary artery pressure, pulmonary wedge pressure, cardiac output, cardiac index, systemic vascular resistance, and pulmonary vascular resistance. All the thermodilution measurements were obtained in triplicate. In most cases, the ultrasound examinawww.ajconline.org

Miscellaneous/Right Ventricular Geometry in Pulmonary Hypertension

629

Table 2 Clinical, hemodynamic, and echocardiographic parameters according to right ventricular (RV) hypertrophy Variable RV Wall Thickness (mm) 6.6 (n 37) Age (years) Gender Women Men Functional class III or IV Systolic pulmonary artery pressure (mm Hg) Mean pulmonary artery pressure (mm Hg) Pulmonary capillary wedge pressure (mm Hg) Pulmonary vascular resistance (Wood Units) Cardiac index (L/min/m2) Right atrial pressure (mm Hg) Right ventricular diameter (mm) Transtricuspid gradient (mm Hg) Right ventricular fractional area change Tricuspid annular plane systolic excursion (mm) Moderate/severe tricuspid regurgitation Inferior vena cava collapsibility Left ventricular end-diastolic volume (ml) Left ventricular ejection fraction 55.0 13.4 30 7 76% 70.8 20.8 46.0 12.7 9.7 4.3 10.5 5.8 6.6 (n 35) 48.3 19.1 22 13 76% 89.9 16.6 56.3 13.1 9.6 4.1 14.6 6.1 p Value

0.086 0.116

1.000 0.000 0.001 0.555 0.004 0.031 0.837 0.022 0.000 0.188 0.702 0.293 0.328 0.470 0.474

Figure 1. Parasternal long-axis view of patient with IPAH. Arrows indicate how RV diameter and RV wall thickness were measured. Note, focus in near gain to better visualize RV wall.

Table 1 Clinical, hemodynamic, and echocardiographic parameters according to right ventricular (RV) diameter Variable RV Diameter (mm) 36.5 (n 36) Age (years) Gender Women Men Functional class III or IV Systolic pulmonary artery pressure (mm Hg) Mean pulmonary artery pressure (mm Hg) Pulmonary capillary wedge pressure (mm Hg) Pulmonary vascular resistance (Wood Units) Cardiac index (L/min/m2) Right atrial pressure (mm Hg) Right ventricular wall thickness (mm) Transtricuspid gradient (mm Hg) Right ventricular fractional area change Tricuspid annular plane systolic excursion (mm) Moderate/severe tricuspid regurgitation Inferior vena cava collapsibility Left ventricular end-diastolic volume (ml) Left ventricular ejection fraction 50.3 16.8 31 5 69% 76.1 22.5 48.0 13.7 9.5 4.4 10.7 4.6 2.4 0.6 6.2 3.4 6.5 2.0 70.0 18.8 28.7 10.3% 16.6 3.3 49% 85% 62 26 61 7% 36.5 (n 36) 53.1 16.6 21 15 83% 84.0 19.3 54.0 13.8 10.3 4.0 14.3 7.2 2.1 0.7 9.0 5.3 6.8 1.9 75.4 19.0 20.6 8.5% 13.2 4.4 97% 33% 54 12 60 8% p Value

2.4 0.7 2.1 0.6 7.5 4.7 7.7 4.7 35.6 6.7 40.2 9.9 64.7 16.3 81.5 17.9 26.3 9.3% 22.9 11.0% 15.1 4.3 67% 66% 60 26 60 7% 14.7 4.2 79% 53% 56 15 61 9%

0.483 0.017

0.173 0.116 0.066 0.453 0.014 0.044 0.010 0.434 0.237 0.001 0.000 0.000 0.000 0.131 0.410

tion and the right heart catheterization were performed on the same day. The data are shown as the mean SD for the continuous variables and as the absolute and relative frequencies for the categorical variables. Pearsons coefcient was used to evaluate the correlations between the continuous variables. The RV wall thickness and RV diameter were dichotomized

on the basis of their median distribution and the betweengroup differences in the clinical, hemodynamic, and echocardiographic characteristics were compared. The mean group values were compared using the 2-tailed t test or the Mann-Whitney U test and proportions using the Fisher exact test. To identify the independent predictors of RV wall thickness and RV diameter, log-linear models were tted; noncollinear variables showing p 0.2 on univariate analysis were included in the models. The relative risks and 95% condence intervals (CIs) were computed. Cumulative survival was calculated using the Kaplan-Meier estimates. The relative risk of dying and its 95% CI were computed using a Cox model. All survival models included the year of diagnosis to account for the prolonged enrollment period of the study. Cardiac death was the only end point of the survival analysis; lung transplantation was considered as a censored observation, and those patients were withdrawn from the analysis at the intervention. p Values 0.05 were retained for statistical signicance. The computations were made using Stata, version 11 (StataCorp, College Station, Texas). Results Of the 72 patients, 20 were men and 52 were women. Their mean age was 52 16 years. The World Health Organization functional class at referral was II for 17 pa-

630

The American Journal of Cardiology (www.ajconline.org)

tients and III-IV for 55 patients. The main echocardiographic and hemodynamic characteristics were as follows: RV diameter 37.8 8.7 mm; RV wall thickness 6.7 1.9 mm; tricuspid annular plane systolic excursion 14.9 4.2 mm; moderate to severe tricuspid regurgitation 71.8% of patients; left ventricular ejection fraction 60 8%; mean pulmonary artery pressure 51.0 13.8 mm Hg; pulmonary vascular resistance 12.5 6.3 Wood Units; cardiac index 2.3 0.7 L/min/m2; and right atrial pressure 7.6 4.7 mm Hg. The median RV diameter was 36.5 mm (range 7 to 13). The patients were accordingly divided into 2 groups: patients with a RV diameter 36.5 mm and those with a RV diameter of 36.5 mm. Their clinical, hemodynamic, and echocardiographic characteristics are listed in Table 1. More female patients had a RV diameter of 36.5 mm. The right heart hemodynamic prole of the patients with a RV diameter greater than the median was characterized by greater pulmonary vascular resistance, greater right atrial pressure, and a lower cardiac index. Greater RV dysfunction was observed at echocardiography in the patients with a RV diameter 36.5 mm. On multivariate analysis, age (relative risk 1.01, 95% CI 1.00 to 1.02, p 0.011) and tricuspid regurgitation (relative risk 9.9, 95% CI 1.3 to 77.9, p 0.027) were the independent variables associated with the RV diameter. The median RV wall thickness was 6.6 mm (range 3 to 13), and the patients were divided into 2 groups: those with a RV wall thickness 6.6 mm and those with a RV wall thickness of 6.6 mm. Their clinical, hemodynamic, and echocardiographic characteristics are listed in Table 2. The right heart hemodynamic prole of the patients with a RV wall thickness greater than the median was characterized by greater systolic and mean pulmonary artery pressures, greater pulmonary vascular resistance, and a slightly lower cardiac index. The clinical and echocardiographic characteristics were substantially similar within the 2 groups, except for a slightly greater RV diameter in the patients with a greater RV wall thickness. The mean pulmonary artery pressure (relative risk 1.02, 95% CI 1.00 to 1.04, p 0.018) was the only variable independently associated with the RV wall thickness on multivariate analysis. During the follow-up period, 22 patients died. The death rate per 100 person-years was 6.6 (95% CI 3.3 to 13.2) for the patients with a RV diameter of 36.5 mm and 15.9 (95% CI 9.4 to 26.8) for the patients with a RV diameter 36.5 mm. The enrollment-year adjusted hazard ratio was 2.64 (95% CI 1.06 to 6.57; p 0.036). The death rate per 100 person-years was 11.2 (95% CI 6.3 to 19.7) for the patients with a RV wall thickness of 6.6 mm and 9.8 (95% CI 5.3 to 18.3) for the patients with a RV wall thickness 6.6 mm. The enrollment-year adjusted hazard ratio was 0.88 (95% CI 0.38 to 2.08; p 0.785). A clear interaction was seen between the RV wall thickness and RV diameter on the prognosis. In patients with a RV wall thickness at or less than the median, dilation had a strong negative prognostic effect. The death rate per 100 person-years was 5.4 (95% CI 2.0 to 14.5) for the patients with a RV wall thickness of 6.6 mm and RV diameter of 36.5 mm, but it was 23.5 (95% CI 11.8 to 47.1) for the patients with a RV wall thickness of 6.6 mm and RV diameter 36.5 mm.

Figure 2. (A) Kaplan-Meier survival estimates for 36 patients with IPAH and RV wall thickness of 6.6 mm. Continuous line indicates patients with RV diameter 36.5 mm; dashed line, patients with RV diameter 36.5 mm. End point of survival analysis was cardiac death. (B) Kaplan-Meier survival estimates for 36 patients with IPAH with RV wall thickness 6.6 mm. Continuous line indicates patients with RV diameter of 36.5 mm; dashed line, patients with RV diameter 36.5 mm. End point of survival analysis was cardiac death.

The adjusted hazard ratio was 4.23 (95% CI 1.21 to 14.82; p 0.024; Figure 2). In contrast, in patients with a RV wall thickness greater than the median, dilation had no signicant negative prognostic effect. The death rate per 100 person-years was 8.4 (95% CI 3.2 to 22.5) for the patients with a RV wall thickness 6.6 mm and RV diameter of 36.5 mm and was 11.1 (95% CI 5.0 to 24.6) for the patients with a RV wall thickness 6.6 mm and RV diameter 36.5 mm. Adjusted hazard ratio 1.70 (95% CI 0.46 to 6.34; p 0.428; Figure 2). Discussion The anatomic characteristics of the pressure overloaded right ventricle have been described using various imaging techniques.4 7 However, the lack of knowledge on the relation between RV structure and function and how such structural characteristics of the right ventricle affect the prognosis in patients with pulmonary arterial hypertension is still substantial. The novelty of the present study was that

Miscellaneous/Right Ventricular Geometry in Pulmonary Hypertension

631

a simple echocardiographic evaluation of the RV geometry provided relevant clinical and prognostic information for patients with IPAH. Although both a larger RV diameter and a thicker RV wall at standard echocardiography have been associated with a more advanced right heart hemodynamic prole, RV dilation also predicted a poor prognosis. A greater RV wall thickness was not, per se, related to the prognosis, but it might reduce the risk of death associated with a dilated right ventricle. In the present study, tricuspid regurgitation was the only hemodynamic determinant of RV dilation in patients with pressure overload of the right ventricle. This result can be explained by the variable degree of tricuspid regurgitation, which is a common nding among patients with pulmonary hypertension. Also, enlargement of the RV chamber, particularly in the free wall to septum minor axis, rather than in the long axis, is the main mechanism of adaptation to volume overload.8,9 The prognostic relevance of RV dilation in patients with pulmonary hypertension is an issue that has not yet received much attention in published studies, even though in the landmark epoprostenol study, continuous infusion of prostacyclin for 12 weeks resulted in improved survival and less RV dilation.10 The reason could be that the RV volumes can be accurately assessed only using cardiac magnetic resonance (CMR).4,5 A study performed at a center with high CMR experience demonstrated that the RV end-diastolic volume is an independent determinant of prognosis in patients with IPAH. In particular, the investigators observed that progressive RV dilation during treatment predicts treatment failure and a poor long-term outcome.11 Such data from patients with IPAH parallel what has been well demonstrated in patients with heart failure (i.e., changes in the left ventricular volumes more accurately predict the outcome than do baseline values).12 The evidence provided in the present study that the RV diameter at standard echocardiography is related to prognosis can therefore be considered of clinical relevance and lays the basis for an easy assessment of the changes in RV dilation during follow-up of patients with IPAH for centers without CMR available. Cardiac hypertrophy is the mechanism of adaptation to altered mechanical or biochemical stimuli in which increased pressure overload plays a key role. It has been suggested that a greater RV hypertrophic adaptation in response to the chronic increase in RV afterload might be associated with a better hemodynamic prole and a better prognosis in patients with IPAH.13,14 This hypothesis has not received further attention, although a consensus exists that a better adaptation of the right ventricle to the high afterload is one reason patients with pulmonary arterial hypertension associated with congenital heart disease have a substantially better prognosis than those with IPAH.15 In the present series of patients with IPAH, those with a thicker RV free wall had an hemodynamic prole characterized by greater pulmonary pressure and vascular resistance and by a substantially similar cardiac index compared to those with moderate hypertrophy, supporting the concept that the sustained increase in RV wall stress was the main stimulus triggering hypertrophy. These data agree with previous CMR studies demonstrating that the RV mass is directly related to pulmonary artery pressure and not as strongly

related to mortality as RV dilation.16 19 Whichever the imaging technique, it must be noted that understanding to what extent hypertrophy is adaptive or maladaptive is not possible. The protective effects of myocyte hypertrophy could coexist to a variable extent with the detrimental effects of perivascular brosis of intramyocardial coronary arteries and varying degrees of interstitial brosis, as well as contractile dysfunction.20 22 Hypertrophy might also be associated with subendocardial ischemia, although a greater thickness of the RV wall was not necessarily associated with the presence or absence of RV perfusion defects at myocardial scintigraphy.23 Little is known about the cellular and molecular mechanisms that underlie the transition from compensated hypertrophy to RV dilation and failure.24 In the present study, the patients with a RV free wall thickness greater than the median had a prognosis similar to that of the patients with RV wall thickness at or less than the median, despite having greater pulmonary pressure and greater vascular resistance. This could imply that the effects of hypertrophy on survival are not neutral but somehow benecial in patients with IPAH. However, the protective effects of hypertrophy can be inferred considering the interaction observed between the wall thickness and diameter. In patients with a RV wall thickness less than the median, a RV diameter greater than the median value increased by the risk of death 3-fold but in patients with thicker RV walls, a RV diameter greater than the median was not associated with a worse prognosis. We acknowledge that it is necessary to verify to what extent the evaluation of RV geometry provides signicant additive clinical information compared to the echocardiographic assessment of RV function, which has demonstrated prognostic relevance in patients with IPAH. However, this should be better explored in a multicenter setting and larger populations. In addition, only in a multicenter setting, would it be possible to integrate the echocardiographic, hemodynamic, and functional data of patients with IPAH. Another limitation of the present study was that the enrollment of patients occurred for several years, and the treatment of the patients changed during that period, possibly affecting RV function to a different extent. To avoid this limitation, we used a statistical analysis that took into account the time of diagnosis. Finally, we also acknowledge that echocardiography is not the ideal imaging technique to assess the RV structure. The RV mass and volumes are correctly measured using CMR, a technique that could also help in identifying brosis within the myocardial walls.25 However, the widespread availability of echocardiography represents an unsurpassed advantage that should lead us to consider the 2 techniques complementary, rather than alternative.
1. Gali N, Hoeper MM, Humbert M, Torbicki A, Vachiery JL, Barbera JA, Beghetti M, Corris P, Gaine S, Gibbs JS, Gomez-Sanchez MA, Jondeau G, Klepetko W, Opitz C, Peacock A, Rubin L, Zellweger M, Simonneau G; Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS), endorsed by the International Society of Heart and Lung Transplantation (ISHLT). Guidelines for the diagnosis and treatment of pulmonary hypertension. Eur Heart J 2009;30:24932537.

632

The American Journal of Cardiology (www.ajconline.org) cardiac remodeling. On behalf of an International Forum on Cardiac Remodeling. J Am Coll Cardiol 2000;35:569 582. Bristow MR, Zisman LS, Lowes BD, Abraham WT, Badesch DB, Groves BM, Voelkel NF, Lynch DM, Quaife RA. The pressureoverloaded right ventricle in pulmonary hypertension. Chest 1998;114: 101S106S. Abraham WT, Raynolds MV, Gottschall B, Badesch DB, Wynne KM, Groves BM, Lowes BD, Bristow MR, Perryman MB, Voelkel NF. Importance of angiotensin-converting enzyme in pulmonary hypertension. Cardiology 1995;86(Suppl):9 15. Hopkins WE, Ochoa LL, Richardson GW, Trulock EP. Comparison of the hemodynamics and survival of adults with severe primary pulmonary hypertension or Eisenmenger syndrome. J Heart Lung Transplant 1996;15:100 105. Quaife RA, Lynch D, Badesch DB, Voelkel NF, Lowes BD, Robertson AD, Bristow MR. Right ventricular phenotypic characteristics in subjects with primary pulmonary hypertension or idiopathic dilated cardiomyopathy. J Card Fail 1999;5:46 54. Saba TS, Foster J, Cockburn M, Cowan M, Peacock AJ. Ventricular mass index using magnetic resonance imaging accurately estimates pulmonary artery pressure. Eur Respir J 2002;20:1519 1524. Roeleveld RJ, Marcus JT, Boonstra A, Postmus PE, Marques KM, Bronzwaer JG, Vonk-Noordegraaf A. A comparison of noninvasive MRI based methods of estimating pulmonary artery pressure in pulmonary hypertension. J Magn Reson Imaging 2005;22:6772. Vonk-Noordegraaf A, Lankhaar J-W, Gotte MJV, Marcus JT, Postmus PE, Westerhof N. Magnetic resonance and nuclear imaging of the right ventricle in pulmonary arterial hypertension. Eur H J Suppl 2007; 9(Suppl H):H29 H34. Katz AM. Cardiomyopathy of overload. N Engl J Med 1990;322:100 110. Weber KT, Janicki JS, Shroff SG, Pick R, Chen RM, Bashey RI. Collagen remodeling of the pressure-overloaded non human primate myocardium. Circ Res 1988;62:757765. Weber KT, Pick R, Jalil JE, Janicki JS, Carroll EP. Pattern of myocardial brosis. J Mol Cell Cardiol 1989;21(Suppl 5):121131. Gomez A, Bialostozky D, Zajarias A, Santos E, Palomar A, Martnez ML, Sandoval J. Right ventricular ischemia in patients with primary pulmonary hypertension. J Am Coll Cardiol 2001;38:11371142. Bogaard HJ, Abe K, Vonk Noordegraaf A, Voelkel NF. The right ventricle under pressure: cellular and molecular mechanisms of rightheart failure in pulmonary hypertension. Chest 2009;135:794 804. McCann GP, Gan CT, Beek AM, Niessen HW, Vonk Noordegraaf A, van Rossum AC. Extent of MRI delayed enhancement of myocardial mass is related to right ventricular dysfunction in pulmonary artery hypertension. AJR Am J Roentgenol 2007;188:349 355.

2. Ghio S, Klersy C, Magrini G, DArmini AM, Scelsi L, Raineri C, Pasotti M, Serio A, Campana C, Vigan M. Prognostic relevance of the echocardiographic assessment of right ventricular function in patients with idiopathic pulmonary arterial hypertension. Int J Cardiol 2010;140:272278. 3. Baker BJ, Gammill J, Massengill J, Schubert E, Karin A, Doherty JE. Echocardiographic detection of right ventricular hypertrophy. Am Heart J 1983;105:611 614. 4. Boxt LM, Katz J, Kolb T, Czegledy FP, Barst RJ. Direct quantitation of right and left ventricular volumes with nuclear magnetic resonance imaging in patients with primary pulmonary hypertension. J Am Coll Cardiol 1992;19:1508 1515. 5. Katz J, Whang J, Boxt LM, Barst RJ. Estimation of right ventricular mass in normal subjects and in patients with primary pulmonary hypertension by nuclear magnetic resonance imaging. J Am Coll Cardiol 1993;21:14751481. 6. Hoeper MM, Tongers J, Leppert A, Baus S, Maier R, Lotz J. Evaluation of right ventricular performance with a right ventricular ejection fraction thermodilution catheter and MRI in patients with pulmonary hypertension. Chest 2001;120:502507. 7. Raman SV, Shah M, McCarthy B, Garcia A, Ferketich AK. Multidetector row cardiac computed tomography accurately quanties right and left ventricular size and function compared with cardiac magnetic resonance. Am Heart J 2006;151:736 744. 8. Reynertson SI, Kundur R, Mullen GM, Costanzo MR, McKiernan TL, Louie EK. Asymmetry of right ventricular enlargement in response to tricuspid regurgitation. Circulation 1999;100:465 467. 9. Mutlak D, Aronson D, Lessick J, Reisner SA, Dabbah S, Agmon Y. Functional tricuspid regurgitation in patients with pulmonary hypertension: is pulmonary artery pressure the only determinant of regurgitation severity? Chest 2009;135:115121. 10. Hinderliter AL, Willis PW IV, Barst RJ, Rich S, Rubin LJ, Badesch DB, Groves BM, McGoon MD, Tapson VF, Bourge RC, Brundage BH, Koerner SK, Langleben D, Keller CA, Murali S, Uretsky BF, Koch G, Li S, Clayton LM, Jbsis MM, Blackburn SD Jr, Crow JW, Long WA; Primary Pulmonary Hypertension Study Group. Effects of long-term infusion of prostacyclin (epoprostenol) on echocardiographic measures of right ventricular structure and function in primary pulmonary hypertension. Circulation 1997;95:1479 1486. 11. van Wolferen SA, Marcus JT, Boonstra A, Marques KM, Bronzwaer JG, Spreeuwenberg MD, Postmus PE, Vonk-Noordegraaf A. Prognostic value of right ventricular mass, volume, and function in idiopathic pulmonary arterial hypertension. Eur Heart J 2007;28:1250 1257. 12. Cohn JN, Ferrari R, Sharpe N. Cardiac remodeling-concepts and clinical implications: a consensus paper from an international forum on

13.

14.

15.

16.

17. 18.

19.

20. 21. 22. 23. 24. 25.

Clinically Signicant Incidental Findings Among Human Immunodeciency Virus-Infected Men During Computed Tomography for Determination of Coronary Artery Calcium
Nancy Crum-Cianone, MD, MPHa,b,*, James Stepenosky, MDc, Sheila Medina, MPHa,b, Dylan Wessman, MDd, David Krause, MDd, and Gilbert Boswell, MDc
Those infected with the human immunodeciency virus (HIV) have a greater risk of cardiovascular disease and might undergo computed tomographic (CT) scans for early detection. Incidental ndings on cardiac CT imaging are important components of the benets and costs of testing. We determined the prevalence and factors associated with incidental ndings on CT scans performed to screen for coronary artery calcium (CAC) among HIV-infected men. A clinically signicant nding was dened as requiring additional workup or a medical referral. A total of 215 HIV-infected men were evaluated. Their median age was 43 years; 17% were current tobacco users; the median CD4 count was 580 cells/mm3; and 83% were receiving antiretroviral medications. Also, 34% had a positive CAC score of >0. An incidental nding was noted among 93 participants (43%), with 36 (17%) having >1 clinically signicant nding. A total of 139 ndings were noted, most commonly pulmonary nodules, followed by granulomas, scarring, and hilar adenopathy. Most of the incidental ndings were stable on follow-up, and no malignancies were detected. The factors associated with the presence of an incidental nding in the multivariate model included increasing age (odds ratio 1.6 per 10 years, p <0.01), positive CAC score (odds ratio 2.3, p <0.01), and current tobacco use (odds ratio 2.5, p 0.02). In conclusion, incidental ndings were common among HIV-infected men undergoing screening CT imaging for CAC determination. The incidental ndings were more common among older patients and those with detectable CAC. 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:633 637) Although several studies have used coronary artery calcium (CAC) scores to measure subclinical heart disease in human immunodeciency virus (HIV)-infected persons,13 no study to date has determined the frequency and types of incidental abnormalities found during these tests. Studies evaluating the incidental ndings detected on coronary computed tomographic (CT) imaging (CAC and angiography) in the general population have been published but did not specically evaluate HIV-infected subjects.4 18 HIVinfected persons might be at a greater risk of infectious and malignant conditions,19,20 which could have implications for the type and number of incidental ndings on imaging studies. Incidental ndings are important, because they not only affect the clinical usefulness, but also the cost-effectiveness, of screening CT scans. We report the frequency and clinical signicance of incidental ndings among HIV-infected persons undergoing CT screening for CAC determination. Methods We evaluated HIV-infected men who were screened for CAC using noncontrast CT imaging of the heart from December 9, 2008 to March 1, 2010. The primary study objective was to determine the prevalence of subclinical coronary atherosclerosis using the Agatston scoring method.21 The objectives of the present substudy were to describe the prevalence and types of incidental ndings on CT scans during standard imaging for CAC scores, and to compare the characteristics of HIV-infected men with and without incidental ndings. The inclusion criteria for study participation were a positive HIV test (enzyme-linked immunosorbent assay conrmed by Western blot) and being a military beneciary, which included active duty members, retirees, and family members. All participants provided written informed consent, and the governing institutional review board approved
www.ajconline.org

a Infectious Disease Clinical Research Program, Uniformed Services University of the Health Sciences, Bethesda, Maryland; bInfectious Disease Division and Departments of cRadiology and dCardiology, Naval Medical Center San Diego, San Diego, California. Manuscript received September 8, 2010; manuscript received and accepted October 11, 2010. Support for this study (Infectious Disease Clinical Research Program018) was provided by the Infectious Disease Clinical Research Program, a Department of Defense program executed through the Uniformed Services University of the Health Sciences. This project was funded in whole, or in part, with federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, under Inter-Agency Agreement Y1-AI-5072. The content of this publication is the sole responsibility of the authors and does not necessarily reect the views or policies of the National Institutes of Health or the Department of Health and Human Services, Department of Defense, or Departments of the Army, Navy, or Air Force. The mention of trade names, commercial products, or organizations does not imply endorsement by the United States Government. *Corresponding author: Tel: (619) 532-8134; fax: (619) 532-8137. E-mail address: nancy.crum@med.navy.mil (N. Crum-Cianone).

0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.10.026

634

The American Journal of Cardiology (www.ajconline.org)

Table 1 Study population characteristics stratied by incidental ndings on cardiac computed tomographic (CT) imaging Characteristic* Total Cohort (n 215) 43 (3650) 110 (51%) 48 (22%) 57 (27%) 37 (17%) 109 (51%) 12 (520) 13 (6%) 64 (30%) 66 (31%) 26.7 (24.129.5) 94 (85100) 30 (14%) 10 (718) 47 (22%) 74 (34%) 12 (5 to 19) 580 (386729) 260 (138366) 150 (70%) 178 (83%) 16 (7%) Incidental Findings Yes (n 93) 48 (3952) 50 (54%) 21 (22%) 22 (24%) 22 (23%) 54 (58%) 17 (826) 10 (11%) 34 (37%) 27 (29%) 26.8 (23.829.6) 94 (87103) 14 (15%) 13 (923) 26 (28%) 43 (46%) 13 (7 to 21) 563 (334682) 230 (100360) 59 (63%) 74 (80%) 9 (10%) No (n 122) 41 (3447) 60 (49%) 27 (22%) 35 (29%) 15 (12%) 55 (45%) 10 (518) 3 (2%) 30 (25%) 39 (32%) 26.5 (24.429.1) 93 (8598) 16 (13%) 10 (715) 21 (17%) 31 (25%) 9 (5 to 19) 600 (457754) 278 (184367) 91 (75%) 104 (85%) 7 (6%) 0.01 0.69 p-value

Age (years) Ethnicity White Black Other Tobacco use Current Ever Years of use Diabetes mellitus Hypertension Hypercholesteremia Body mass index (kg/m2) Waist circumference (cm) C-reactive protein 0.5 mg/dl Erythrocyte sedimentation rate (mm/hour) Erythrocyte sedimentation rate 20 mm/hour Presence of coronary atherosclerosis (coronary artery calcium 0) Human immunodeciency virus duration (years) Current CD4 cell count (cells/mm3) Nadir CD4 cell count (cells/mm3) Undetectable human immunodeciency virus RNA level (50 copies/ml) Current antiretroviral therapy History of opportunistic infection

0.04 0.07 0.01 0.02 0.07 0.66 0.86 0.31 0.70 0.01 0.07 0.01 0.08 0.02 0.06 0.10 0.28 0.30

*Categorical variables expressed as n (%) and continuous variables as median (interquartile range). Among those with a history of tobacco use. Hypercholesteremia dened as total cholesterol 200 mg/dl.

the study. The study was registered at ClinicalTrials.gov (study identier NCT00889577). The clinical data collected at the time of the CT scan included demographics, tobacco use history, body mass index, waist circumference, C-reactive protein (lower limit of detection 0.5 mg/dl; particle enhanced immunoturbidimetric Assay, Roche, Indianapolis, Indiana), erythrocyte sedimentation rate (modied Westergren method), CD4 cell counts (ow cytometry), plasma HIV RNA levels (Roche Amplicor, undetectable at 50 copies/ml), and antiretroviral therapy use. The diagnosis of hypercholesterolemia (total cholesterol 200 mg/dl), hypertension, and diabetes (determined by the use of medications for these conditions) and a history of an opportunistic infection were recorded. Participants underwent noncontrast CT imaging using a multidetector CT machine. Prospectively gated, axial, 3-mm images were obtained at 120 kV on a Siemens Definition Dual Source CT scanner (Siemens Medical Solutions, Forsheim, Germany). The scanning protocol captured images with a 330-ms gantry rotation time, an individual detector width of 0.6 mm with a reconstructed section width of 3 mm and a temporal resolution of 165 ms. Contiguous 3-mm-thick sections were reconstructed during peak inspiration with a 16- to 25-cm eld of view, depending on the heart size. The images were processed on an Impax 6.3 workstation (Agfa-Gevaert Group, Mortsel, Belgium). The full eld of view reconstructions were performed of the

entire lungs at the levels from the carina to just below the cardiac apex. The CT images also included the mediastinum, hilum, and diaphragm. The abdominal organs were not visualized unless the diaphragm had been shifted cranially because of underlying pathology. CAC scoring was performed using an Aquarius workstation (TeraRecon, San Mateo, California) and calculated as the sum of all lesions in each of the coronary arteries using Agatston units, as previously described.21 A CAC score of 0 was considered positive for detectable calcium. The images were read by a board-certied radiologist (G.B.) for incidental noncoronary ndings at CAC scoring. Any nding requiring additional clarication was reread by the same radiologist. Incidental ndings were captured only if the nding was not solely age-related (e.g., calcication of the aorta), due to trauma (e.g., old rib fractures), or postoperative. The incidental ndings were classied as clinically signicant by the need for additional workup, including imaging or medical referral. The need for additional radiologic imaging was determined using available evidence-based criteria; for example, the management of pulmonary nodules was determined using the Fleischner guidelines.22 A pulmonary nodule was dened as a 3-cm lesion in the lung parenchyma and a granuloma as a calcied opacity. The patients with emphysema or bronchiectasis were recommended for referral to pulmonary medicine. The medical records of the participants with a

Miscellaneous/Incidental Cardiac Imaging Findings in HIV Table 2 Number and type of incidental ndings by location among 93 HIV-infected men* Finding Lung Parenchymal/bronchi Nodule Granuloma Scar Emphysema Bronchiectasis Parenchymal opacity/ consolidation Bronchial opacity Bullae Pleura Pleural thickening Heart Cardiac valves with signicant calcication Papillary muscle fat/calcication Mediastinum Lymph nodes Mass Hilum/subcarina Lymph node Granuloma Pericardiac Lymph node Other Gynecomastia Hiatal hernia Axillary adenopathy Aortic dilation/aneurysm Thymus mass Paralyzed hemidiaphragm Diaphragm eventration Pectus excavatum Enlarged pulmonary artery Large splenic cyst Total

635

Incidental Findings (n) 78 23 19 13 9 6 4 2 1 1 4 2 2 9 7 2 18 17 1 1 1 29 8 7 7 1 1 1 1 1 1 1 139

Findings With Clinical Signicance (n) 37 18 0 0 9 6 3 1 0 0 2 2 0 1 0 1 0 0 0 1 1 5 0 0 0 1 1 1 0 0 1 1 46

The statistical analysis included descriptive statistics of the prevalence of incidental ndings. Categorical variables are described as numbers with proportions and continuous variables as medians with interquartile ranges. The incidental ndings were divided into clinically signicant and not clinically signicant, as dened. The comparisons of participants with and without incidental ndings were performed using Fishers exact and rank-sum tests for categorical and continuous variables, respectively. A multivariate logistic regression model was performed to evaluate the factors associated with the incidental ndings on cardiac CT scanning. The variables with p 0.10 on univariate testing were placed in the full multivariate model, and a backward stepwise approach was used to derive the nal model. p-values 0.05 were considered statistically signicant. All analyses were performed using Stata, version 10 (StataCorp, College Station, Texas). Results The study population consisted of 215 HIV-infected men (Table 1). An incidental nding was noted for 93 patients (43%), with 36 (17%) having a clinically signicant nding that required follow-up or medical referral. The range of incidental ndings per person was 0 to 5. Of those with an incidental nding, 62 patients had 1 nding, 19 had 2 ndings, 10 had 3 ndings, 1 had 4 ndings, and 1 had 5 ndings. The number and type of incidental ndings are listed in Table 2. A total of 139 incidental ndings were noted, with some participants having multiple ndings. The most common ndings were pulmonary nodules, followed by granulomas and scars. A total of 15 patients (7% of the cohort) had ndings consistent with emphysema or bronchiectasis. Several extrapulmonary ndings were noted, including gynecomastia, hiatal hernia, pathologic aortic dilation, and a large (15 13 cm) asymptomatic splenic cyst. Of the incidental ndings, 46 (33%) were deemed clinically signicant. Most often, these were pulmonary nodules (Table 2). On the follow-up evaluation at a median of 15 months (range 6 to 21) after the initial CT scan for CAC scoring, 15 of 18 of the pulmonary nodules were reimaged. Of the 15 pulmonary nodules, 13 were unchanged; 1 had increased in size (from 7 to 20 mm) but the workup ndings, including bronchoscopy with cultures and cytology, have been negative; and 1 was determined not to be a nodule on reimaging. The remaining 3 participants with pulmonary nodules failed to undergo the scheduled follow-up imaging. The patients with pneumonia/opacities received antibiotics, and the cases had resolved or were improving for 4 of the 5 who underwent repeat imaging. The subjects with emphysema or bronchiectasis had no changes in management, except for 1, for whom bronchodilators were prescribed. The aortic dilation detected on imaging remained stable. The single patient with the large splenic cyst underwent splenectomy because of the high risk of rupture. Only 1 participant died during the follow-up period; the cause of death was not associated with the incidental nding (i.e., nodule) on CT imaging. The HIV-infected persons with an incidental nding were older (48 vs 41 years, p 0.01), were current tobacco

* Some patients had multiple incidental ndings. Medical referral recommended. Cyst was 15 13 cm in diameter. Table 3 Final multivariate model for factors associated with incidental ndings on computed tomographic (CT) imaging for coronary artery calcium (CAC) determination Factor Age (per 10 years) Positive coronary artery calcium score* Current Tobacco use OR (95% CI) 1.6 (1.22.2) 2.3 (1.34.3) 2.5 (1.34.3) p-value 0.01 0.01 0.02

CI, condence interval, OR, odds ratio. * Dened as CAC score of 0.

clinically signicant incidental nding were reviewed in August 2010 for clinical outcomes, including data on follow-up imaging.

636

The American Journal of Cardiology (www.ajconline.org)

users (23% vs 12%, p 0.04), were diabetic (11% vs 2%, p 0.02), had a greater erythrocyte sedimentation rate (13 vs 10 mm/hour, p 0.01), had a positive CAC score (46% vs 25%, p 0.01), and had a lower current CD4 cell count (563 vs 600 cells/mm3, p 0.02) compared to those without an incidental nding (Table 1). Of the tobacco users, a longer duration of use was also associated with an incidental nding (p 0.01). In the nal multivariate model, increasing age (odds ratio 1.6 per 10 years, p 0.01), the presence of CAC (odds ratio 2.3, p 0.01), and current tobacco use (odds ratio 2.5, p 0.02) were associated with the presence of an incidental nding (Table 3). No HIV-specic factor was signicant in the nal model. The nal model had a good t (likelihood-ratio test, 26.7, p 0.01). Discussion The present study is the rst to evaluate the prevalence of incidental ndings among HIV-infected persons undergoing CT imaging for coronary atherosclerosis screening. We found that 43% of HIV-infected persons undergoing CAC determination had 1 incidental nding. In the general population, 5 studies have evaluated the prevalence of incidental ndings on CT scans for CAC determination,711 with additional studies examining other types of CT cardiac imaging,5,6,1218 and noted 8% to 53% had incidental ndings, with a mean prevalence of 38%.7,8,10,11 Our prevalence of incidental ndings was similar to that in these studies and investigations using similar CT imaging techniques (41% to 56%).10,12 However, our study was performed among patients substantially younger (median age 43 years) than those in the general population (60 years). Because incidental ndings increase with age,5,6 our data suggest that HIV-infected persons might have a greater number of incidental ndings. Additionally, our HIV-positive cohort had a greater frequency of incidental ndings than HIV-negative military personnel (8%) of similar ages.11 The reason for the greater rate of incidental ndings among HIV-infected persons might be because of their greater risk of pulmonary disease.19 Most incidental ndings in our study were pulmonary nodules, granulomas, scarring, and bronchiectasis, all suggestive of previous respiratory infections. In addition, HIV-infected persons often have greater rates of tobacco use compared to the general population, which might have contributed to an increased number of pulmonary ndings.10,23 Finally, HIV-infected persons might have unique incidental ndings not typically seen in the general population, including gynecomastia, likely the result of altered fat distribution patterns due to the HIV and/or antiretroviral agents.24,25 In our study, 17% of patients with HIV had a clinically signicant nding that required follow-up or medical referral. This compared with a range of 1% to 52% in the general population.4 6 The wide variation in rates is a result of the varying denitions and elds of view used for imaging. We used classications within recent publications and multidetector CT imaging to allow for comparability.4,10 Similar to the general population, the most common signicant ndings were pulmonary nodules.4,8 10

Several of our participants had incidental ndings of unquestionable clinical signicance, including pneumonia treated with antibiotics and a large splenic cyst. The discovery of such ndings could result in added value of screening CT scans.4 In contrast, the detection of ndings such as pulmonary nodules, which are often benign or resolve on follow-up imaging,5,17 could result in added nancial costs, risks (including radiation exposure during repeat imaging), and increased patient anxiety.5 Most of our incidental ndings were stable over time and did not result in changes in management. Also, our study did not detect any undiagnosed cancers. As such, an assessment of the frequency and medical signicance of incidental ndings among specic populations are important. The factors associated with incidental ndings included increasing age, current tobacco use, and the presence of CAC; no HIV-specic factors were identied. Only 1 other published study has evaluated the predictors of incidental ndings.10 Age and tobacco use are known to be associated with pulmonary disease, including scarring and nodules. Of particular interest was our nding that the HIV-infected participants with a positive CAC score had a 2-fold greater frequency of incidental ndings after adjusting for age and smoking. The precise relation between CAC and incidental ndings is unknown. A possible explanation is that the increased inammation associated with previous infections (which leads to incidental pulmonary ndings) might contribute to the development of vascular disease.26 Additional studies are needed to evaluate this potential association. Other explanations include the shared factors between CAC and incidental ndings, such as exposure or concurrent conditions, which were unmeasured in our study. Overall, these data suggest that the benet of identifying calcied coronary plaques should be balanced against the greater number of incidental ndings among these patients. Our study had potential limitations. As noted in other studies, no standard denition for clinically signicant incidental ndings has been determined. Our study did use the recent classications4 and standardized criteria.22 Second, although our report has provided information on the outcomes of the initial ndings, follow-up is ongoing. Two previous studies with follow-up data showed that a single death in each study was attributed to an incidental nding on screening CT imaging; however, the duration of follow-up was limited.5,16 Third, we did not enroll an agematched HIV-negative control group to determine whether HIV-infected persons have a greater prevalence of incidental ndings but used historical information. Fourth, our HIV patients had low overall rates of tobacco and illicit drug use; thus, we might have underestimated the prevalence of incidental ndings compared to other HIV populations. Finally, because only men were evaluated in our study, future studies examining HIV-infected women are advocated.
1. Mangili A, Gerrior J, Tang AM, OLeary DH, Polak JK, Schaefer EJ, Gorbach SL, Wanke CA. Risk of cardiovascular disease in a cohort of HIV-infected adults: a study using carotid intima-media thickness and coronary artery calcium score. Clin Infect Dis 2006;43:14821489. 2. Talwani R, Falusi OM, Mendes de Leon CF, Nerad JL, Rich S, Proia LA, Sha BE, Smith KY, Kessler HA. Electron beam computed tomography for assessment of coronary artery disease in HIV-infected men

Miscellaneous/Incidental Cardiac Imaging Findings in HIV receiving antiretroviral therapy. J Acquir Immune Dec Syndr 2002; 30:191195. Mangili A, Jacobson DL, Gerrior J, Polak JF, Gorbach SL, Wanke CA. Metabolic syndrome and subclinical atherosclerosis in patients infected with HIV. Clin Infect Dis 2007;44:1368 1374. Jacobs PC, Mali WP, Grobbee DE, van der Graaf Y. Prevalence of incidental ndings in computed tomographic screening of the chest: a systematic review. J Comput Assist Tomogr 2008;32:214 221. Machaalany J, Yam Y, Ruddy TD, Abraham A, Chen L, Beanlands RS, Chow BJW. Potential clinical and economic consequences of noncardiac incidental ndings on cardiac computed tomography. J Am Coll Cardiol 2009;54:15331541. Lee CI, Tsai EB, Sigal BM, Plevritis SK, Garber AM, Rubin GD. Incidental extracardiac ndings at coronary CT: clinical and economic impact. AJR Am J Roentgenol 2010;194:15311538. Hunold P, Schmermund A, Seibel RM, Grnemeyer DH, Erbel R. Prevalence and clinical signicance of accidental ndings in electronbeam tomographic scans for coronary artery calcication. Eur Heart J 2001;22:1748 1758. Schragin JG, Weissfeld JL, Edmundowicz D, Strollo DC, Fuhrman CR. Non-cardiac ndings on coronary electron beam computed tomography scanning. J Thorac Imaging 2004;19:82 86. Horton KM, Post WS, Blumenthal RS, Fishman EK. Prevalence of signicant noncardiac ndings on electron-beam computed tomography coronary artery calcium screening examinations. Circulation 2002;106:532534. Burt JR, Iribarren C, Fair JM, Norton LC, Mahbouba M, Rubin GD, Hlatky MA, Go AS, Fortmann SP; Atherosclerotic Disease, Vascular Function and Genetic Epidemiology (ADVANCE) study. Incidental ndings on cardiac multidetector row computed tomography among healthy older adults: prevalence and clinical correlates. Arch Intern Med 2008;168:756 761. Elgin EE, OMalley PG, Feuerstein I, Taylor AJ. Frequency and severity of incidentalomas encountered during electron beam computed tomography for coronary calcium in middle-aged army personnel. Am J Cardiol 2002;90:543545. Gil BN, Ran K, Tamar G, Shmuell F, Eli A. Prevalence of signicant noncardiac ndings on coronary multidetector computed tomography angiography in asymptomatic patients. J Comput Assist Tomogr 2007; 31:1 4. Onuma Y, Tanabe K, Nakazawa G, Aoki J, Nakajima H, Ibukuro K, Hara K. Noncardiac ndings in cardiac imaging with multidetector computed tomography. J Am Coll Cardiol 2006;48:402 406. Haller S, Kaiser C, Buser P, Bongartz G, Bremerich J. Coronary artery imaging with contrast-enhanced MDCT: extracardiac ndings. AJR Am J Roentgenol 2006;187:105110.

637

3. 4. 5.

6. 7.

8. 9.

10.

11.

12.

13. 14.

15. Koonce J, Schoepf JU, Nguyen SA, Northam MC, Ravenel JG. Extracardiac ndings at cardiac CT: experience with 1,764 patients. Eur Radiol 2009;19:570 576. 16. Law YM, Huang J, Chen K, Cheah FK, Chua T. Prevalence of signicant extracoronary ndings on multislice CT coronary angiography examinations and coronary artery calcium scoring examinations. J Med Imaging Radiol Oncol 2008;52:49 56. 17. Iribarren C, Hlatky MA, Chandra M, Fair JM, Rubin GD, Go AS, Burt JR, Fortmann SP. Incidental pulmonary nodules on cardiac computed tomography: prognosis and use. Am J Med 2008;121:989 996. 18. Lehman SJ, Abbara S, Cury RC, Nagurney JT, Hsu J, Goela A, Schlett CL, Dodd JD, Brady TJ, Bamberg F, Hoffmann U. Signicance of cardiac computed tomography incidental ndings in acute chest pain. Am J Med 2009;122:543549. 19. Davis JL, Fei M, Huang L. Respiratory infection complicating HIV infection. Curr Opin Infect Dis 2008;21:184 190. 20. Patel P, Hanson DL, Sullivan PS, Novak RM, Moorman AC, Tong TC, Holmberg SD, Brooks JT. Adult and Adolescent Spectrum of Disease Project and HIV Outpatient Study Investigators. Incidence of types of cancer among HIV-infected persons compared with the general population in the United States, 19922003. Ann Intern Med 2008;148: 728 736. 21. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M Jr, Detrano R. Quantication of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990;15:827 832. 22. MacMahon H, Austin JH, Gamsu G, Herold CJ, Jett JR, Naidich DP, Patz EF Jr, Swensen SJ; Fleischner Society. Guidelines for management of small pulmonary nodules detected on CT scans: a statement from the Fleischner Society. Radiology 2005;237:395 400. 23. Currier JS. Update on cardiovascular complications in HIV infection. Top HIV Med 2009;17:98 103. 24. Mira JA, Lozano F, Santos J, Ramayo E, Terrn A, Palacios R, Len EM, Mrquez M, Macas J, Fernndez-Palacin A, Gmez-Mateos J, Pineda JA, Grupo Andaluz para el Estudio de las Enfermedades Infecciosas. Gynaecomastia in HIV-infected men on highly active antiretroviral therapy: association with efavirenz and didanosine treatment. Antivir Ther 2004;9:511517. 25. Strub C, Kaufmann GR, Flepp M, Egger M, Kahlert C, Cavassini M, Battegay M; Swiss HIV Cohort Study. Gynecomastia and potent antiretroviral therapy. AIDS 2004;18:13471349. 26. Ross AC, Rizk N, ORiordan MA, Dogra V, El-Bejjani D, Storer N, Harrill D, Tungsiripat M, Adell J, McComsey GA. Relationship between inammatory markers, endothelial activation markers, and carotid intima-media thickness in HIV-infected patients receiving antiretroviral therapy. Clin Infect Dis 2009;49:1119 1127.

Self-Terminated Ventricular Fibrillation and Recurrent Syncope

Yuval Konstantino, MD*, Angela Morello, MD, Peter J. Zimetbaum, MD, and Mark E. Josephson, MD
Ventricular brillation (VF) is a lethal arrhythmia that requires immediate cardioversion and is rarely self-terminating. Spontaneous termination is typically associated with more organized activation than sustained VF terminated by shock, but the precise mechanism is unclear. In the present case, we describe a patient with recurrent syncope and documented self-terminating VF, who ultimately underwent implantable cardioverter debrillator insertion. Assessment of the rhythm strip revealed organization of a chaotic rhythm into monomorphic ventricular tachycardia before termination, in supportive of previous reports. In conclusion, self-terminating VF is a very rare condition that can cause syncope. 2011 Published by Elsevier Inc. (Am J Cardiol 2011;107:638 640)

The cardiac causes of syncope, primarily bradyarrhythmias and tachyarrhythmias, account for 10% to 20% of syncopal episodes.1 Of the tachyarrhythmias, ventricular tachycardia (VT) and, less commonly, supraventricular tachycardia can cause syncope. In contrast, ventricular Cardiovascular Institute, Beth Israel Deaconess Medical Center, Boston, Massachusetts. Manuscript received August 22, 2010; manuscript received and accepted October 1, 2010. *Corresponding author: Tel: (617) 632-27713; fax: (617) 632-7377. E-mail address: ykonstan@gmail.com (Y. Konstantino).

brillation (VF) is an extremely rare cause of syncope. VF is a life-threatening arrhythmia that requires immediate cardioversion and is rarely self-terminating. In the present case, we describe a patient with recurrent syncope and self-terminating VF detected by implantable loop recorder (ILR). An implantable cardioverter debrillator was inserted. Case Report A 68-year-old man was admitted to our hospital with recurrent syncope. Previously, he had had syncope after sitting up in bed. At that time, he had a prodrome of nausea,

Figure 1. ILR tracing demonstrating VF that organized into monomorphic VT, followed by spontaneous termination and sinus bradycardia. 0002-9149/11/$ see front matter 2011 Published by Elsevier Inc. doi:10.1016/j.amjcard.2010.10.025 www.ajconline.org

Case Report/Ventricular Fibrillation and Recurrent Syncope

639

Figure 2. VF initiated by early coupling VPC, as recorded on telemetry during hospitalization.

ushing, and dizziness. The cardiac evaluation included an electrocardiogram with normal ndings, an echocardiogram that revealed normal biventricular function and no wall motion abnormality, and an imaging stress test that demonstrated a xed infero-posterolateral defect. Angiography revealed 100% occlusion of the right coronary artery and a 50% narrowing of the obtuse marginal artery. One year later, he lost consciousness at rest, without any preceding symptoms. Cardiopulmonary resuscitation was

initiated immediately. On medical team arrival, he was awake and free of symptoms. His vital signs were normal. The electrocardiogram revealed sinus rhythm without evidence of previous infarction, acute ischemia, or Brugada, long QT, or short QT syndromes. The electrolytes and cardiac biomarker levels were normal. The echocardiogram displayed normal biventricular function with no wall motion abnormality. An electrophysiologic study was performed, revealing a normal voltage map, without evidence of an

640

The American Journal of Cardiology (www.ajconline.org)

endocardial scar. Ventricular arrhythmia was not induced, with extra stimuli delivered from the right ventricular apex and the right ventricular outow tract, with and without isoproterenol. Sinus node function and atrioventricular conduction were normal. Given recurrent unexplained syncope, an ILR was placed. Two weeks later, a third syncopal episode occurred. The ILR tracing demonstrated VF, which organized into monomorphic VT and spontaneously terminated (Figure 1). During hospitalization, another syncopal episode occurred, with no preceding symptoms. VF was documented (Figure 2), and brief cardiopulmonary resuscitation was initiated with spontaneous VF termination. Cardioversion or medications were not required. Assessment of the initiating ventricular premature complex (VPC) revealed a Q wave in lead II, suggestive of a VPC originating from an inferior scar. An implantable cardioverter debrillator was inserted. Discussion In 3 studies that analyzed data from ILRs from a total of 365 patients with recurrent syncope, arrhythmia was documented in 20% to 25%. Bradycardia was seen in 15% to 19% and tachyarrhythmia in 4% to 6% (VT in 1 to 3% and VF in none).2 4 Yu et al5 described a patient with recurrent syncope diagnosed as recurrent VF from the ILR ndings The VF was eliminated by radiofrequency catheter ablation of the triggering VPCs. Leenhardt et al6 had previously reported a group of young patients at high risk of sudden death, with no structural heart disease and normal QT who presented with syncope related to a short-coupled variant of torsade de pointes. The initiating VPC had a very short coupling interval, and the arrhythmia deteriorated into VF in most of patients soon after its onset. This was unlikely to be the etiology in our patient, given his older age, history of coronary artery disease, and a much longer coupling interval of the initiating VPC.

The precise mechanism of self-termination of VF is unknown. Self-terminating polymorphic VT has been previously described in patients with coronary artery disease and a normal QT interval undergoing an electrophysiologic study,7 but it is extremely uncommon. Spontaneous reversion to sinus rhythm typically occurs when the localized chaotic activity is conned to small areas of the heart and gradually becomes regular.7 Mkikallio et al8 have found that spontaneously terminating VF displays more organized local activation dynamics than sustained VF terminated by shock, suggesting that the dynamic behavior of local cardiac activation might be related to the maintenance of ventricular tachyarrhythmias. The present case demonstrated the organization of the cardiac electrocardiogram into monomorphic VT, before VF termination, in support of the previous reports.
1. Calkins H, Zipes DP. Hypotension and syncope. In: Zipes DP, Braunwald E, eds. Braunwalds Heart Disease: A Textbook of Cardiovascular Medicine, 7th ed. Philadelphia: Elsevier Saunders; 2005:912. 2. Entem FR, Enriquez SG, Cobo M, Expsito V, Llano M, Ruiz M, Jose Olalla J, Otero-Fernandez M. Utility of implantable loop recorders for diagnosing unexplained syncope in clinical practice. Clin Cardiol 2009; 32:28 31. 3. Seidl K, Rameken M, Breunung S, Senges J, Jung W, Andresen D, van Toor A, Krahn AD, Klein GJ. Reveal-investigators. Diagnostic assessment of recurrent unexplained syncope with a new subcutaneously implantable loop recorder. Europace 2000;2:256 262. 4. Farwell DJ, Freemantle N, Sulke N. The clinical impact of implantable loop recorders in patients with syncope. Eur Heart J 2006;27:351356. 5. Yu CC, Tsai CT, Lai LP, Lin JL. Successful radiofrequency catheter ablation of idiopathic ventricular brillation presented as recurrent syncope and diagnosed by an implanted loop recorder. Int J Cardiol 2006;110:112113. 6. Leenhardt A, Glaser E, Burguera M, Nrnberg M, Maison-Blanche P, Coumel P. Short-coupled variant of torsade de pointes: a new electrocardiographic entity in the spectrum of idiopathic ventricular tachyarrhythmias. Circulation 1994;89:206 215. 7. Josephson ME. Clinical Cardiac Electrophysiology, 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2008:516 524. 8. Mkikallio TH, Huikuri HV, Myerburg RJ, Seppnen T, Kloosterman M, Interian A Jr, Castellanos A, Mitrani RD. Differences in the activation patterns between sustained and self-terminating episodes of human ventricular brillation. Ann Med 2002;34:130 135.

READERS COMMENTS
Comparison of 600 Versus 300-mg Clopidogrel Loading Dose in Patients With ST-Segment Elevation Myocardial Infarction Undergoing Primary Coronary Angioplasty The rationale for the use of a double dose of clopidogrel in the study of Mangiacapra et al1 for those with STsegment elevation myocardial infarction (STEMI) who underwent primary coronary angioplasty was to rapidly suppress platelet activity to avoid subsequent cardiovascular ischemic events. As is known, other adjuvant antiplatelet therapy trials for patients with STEMI who undergo primary angioplasty have revealed similar results, with reductions of cardiovascular end points and events of acute and subacute instant thrombosis. However, additional benets have been emphasized, especially for those with high risk proles. For example, in the Korean Acute Myocardial Infarction Registry, Chen et al2 pointed out that cilostazol-based triple therapy should be used for elderly, female, and diabetic patients with STEMI for primary angioplasty. De Luca et al3 performed a meta-regression analysis of randomized trials and demonstrated a mortality benet proportional to baseline risk in abciximab-adjuvant triple therapy. In the Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) trial, Dangas et al4 reported that Killip class I had the lowest hazard ratio (0.36), and a clopidogrel loading dose of 600 mg had a hazard ratio of 0.67 for the occurrence of cardiovascular events at 30 days after percutaneous coronary intervention, implying that the benets of a higher loading dose of clopidogrel will be blunted in the condition of Killip class I.5 Recently, Pocock et al6 reported that the highest death rate from ischemia occurred on days 0 and 1 (hazard ratio 15.57), but the highest death rate from major bleeding occurred during days 8 to 30 (hazard ratio 4.80) after primary angioplasty. This nding is consistent with the results of the Clopidogrel Optimal Loading Dose Usage to Reduce Recurrent Events/Optimal Antiplatelet Strategy for Interventions (CURRENTOASIS 7) trial that clopidogrel 600 mg/ day can be safely used for 1 week without increasing the rate of major bleeding.7 Accordingly, the use of a 600-mg loading dose of clopidogrel should be recommended in patients with STEMI with high risk proles for primary angioplasty. How to identify nonresponders to clopidogrel will be the next major issue for those who require a loading dose of clopidogrel 600 mg for primary angioplasty.
Gen-Min Lin, MD Yi-Hwei Li, PhD Hualien, Taiwan 8 November 2010

vidual patient risk and mortality impact of ischemic and hemorrhagic complications: assessment from the Acute Catheterization and Urgent Intervention Triage Strategy trial. Circulation 2010;121:4351. 7. Mehta SR, Tanguay JF, Eikelboom JW, Jolly SS, Joyner CD, Granger CB, Faxon DP, Rupprecht HJ, Budaj A, Avezum A, Widimsky P, Steg PG, Bassand JP, Montalescot G, Macaya C, Di Pasquale G, Niemela K, Ajani AE, White HD, Chrolavicius S, Gao P, Fox KA, Yusuf S; on behalf of the CURRENT-OASIS 7 Trial Investigators. Double-dose versus standard-dose clopidogrel and high-dose versus low-dose aspirin in individuals undergoing percutaneous coronary 1intervention for acute coronary syndromes (CURRENT-OASIS 7): a randomised factorial trial. Lancet 2010;376:12331243.
doi:10.1016/j.amjcard.2010.11.001

1. Mangiacapra F, Muller O, Ntalianis A, Trana C, Heyndrickx GR, Bartunek J, Vanderheyden M, Wijns W, De Bruyne B, Barbato E. Comparison of 600 versus 300-mg clopidogrel loading dose in patients with ST-segment elevation myocardial infarction undergoing primary coronary angioplasty. Am J Cardiol 2010;106:1208 1211. 2. Chen KY, Rha SW, Li YJ, Poddar KL, Jin Z, Minami Y, Wang L, Kim EJ, Park CG, Seo HS, Oh DJ, Jeong MH, Ahn YK, Hong TJ, Kim YJ, Hur SH, Seong IW, Chae JK, Cho MC, Bae JH, Choi DH, Jang YS, Chae IH, Kim CJ, Yoon JH, Chung WS, Seung KB, Park SJ; Korea Acute Myocardial Infarction Registry Investigators. Triple versus dual antiplatelet therapy in patients with acute ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention. Circulation 2009;119: 32073214. 3. De Luca G, Suryapranata H, Stone GW, Antoniucci D, Tcheng JE, Neumann FJ, Bonizzoni E, Topol EJ, Chiariello M. Relationship between patients risk prole and benets in mortality from adjunctive abciximab to mechanical revascularization for ST-segment elevation myocardial infarction: a meta-regression analysis of randomized trials. J Am Coll Cardiol 2006;47:685 686. 4. Dangas G, Mehran R, Guagliumi G, Caixeta A, Witzenbichler B, Aoki J, Peruga JZ, Brodie BR, Dudek D, Kornowski R, Rabbani LE, Parise H, Stone GW; HORIZONS-AMI Trial Investigators. Role of clopidogrel loading dose in patients with ST-segment elevation myocardial infarction undergoing primary angioplasty: results from the HORIZONS-AMI (Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction) trial. J Am Coll Cardiol 2009;54:1438 1446. 5. Lin GM, Han CL. The loading dose of clopidogrel in patients with ST-segment elevation myocardial infarction undergoing primary angioplasty. Am J Emerg Med 2010;28:382383. 6. Pocock SJ, Mehran R, Clayton TC, Nikolsky E, Parise H, Fahy M, Lansky AJ, Bertrand ME, Lincoff AM, Moses JW, Ohman EM, White HD, Stone GW. Prognostic modeling of indi-

Long-Term Follow Up of Atrioventricular Block in Transcatheter Aortic Valve Implantation We have read with great interest the report by Roten et al1 regarding predictors of atrioventricular (AV) conduction impairment after transcatheter aortic valve implantation with the CoreValve prosthesis (Medtronic, Inc., Minneapolis, Minnesota). We have published our early experience with this prosthesis2 and recently performed electrophysiologic studies in some pacemaker-free patients immediately before and after valve implantation. An electrode was placed on the His bundle during valve implantation, and data were continuously recorded during the procedure. We agree that the type of AV block is intra- or infrahisian; in fact, our group has published for the rst time a report of intrahisian AV block in a patient who underwent percutaneous CoreValve prosthesis implantation.3 However, we are extremely surprised by the high AV conduction recovery rate, taking in account that infrahisian blocks are usually permanent. In our series with the rst 50 patients who underwent percutaneous implantation of the CoreValve prosthesis, pacemaker implantation was needed in 22 patients (44%), 20 patients because of complete AV block and 2 patients because of rst-degree AV block with newly developed left bundle branch block. Of the 20 patients, considering a subgroup of 15 patients who were discharged alive from the hospital and afwww.ajconline.org

Am J Cardiol 2011;107:641 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved.

READERS COMMENTS
Comparison of 600 Versus 300-mg Clopidogrel Loading Dose in Patients With ST-Segment Elevation Myocardial Infarction Undergoing Primary Coronary Angioplasty The rationale for the use of a double dose of clopidogrel in the study of Mangiacapra et al1 for those with STsegment elevation myocardial infarction (STEMI) who underwent primary coronary angioplasty was to rapidly suppress platelet activity to avoid subsequent cardiovascular ischemic events. As is known, other adjuvant antiplatelet therapy trials for patients with STEMI who undergo primary angioplasty have revealed similar results, with reductions of cardiovascular end points and events of acute and subacute instant thrombosis. However, additional benets have been emphasized, especially for those with high risk proles. For example, in the Korean Acute Myocardial Infarction Registry, Chen et al2 pointed out that cilostazol-based triple therapy should be used for elderly, female, and diabetic patients with STEMI for primary angioplasty. De Luca et al3 performed a meta-regression analysis of randomized trials and demonstrated a mortality benet proportional to baseline risk in abciximab-adjuvant triple therapy. In the Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) trial, Dangas et al4 reported that Killip class I had the lowest hazard ratio (0.36), and a clopidogrel loading dose of 600 mg had a hazard ratio of 0.67 for the occurrence of cardiovascular events at 30 days after percutaneous coronary intervention, implying that the benets of a higher loading dose of clopidogrel will be blunted in the condition of Killip class I.5 Recently, Pocock et al6 reported that the highest death rate from ischemia occurred on days 0 and 1 (hazard ratio 15.57), but the highest death rate from major bleeding occurred during days 8 to 30 (hazard ratio 4.80) after primary angioplasty. This nding is consistent with the results of the Clopidogrel Optimal Loading Dose Usage to Reduce Recurrent Events/Optimal Antiplatelet Strategy for Interventions (CURRENTOASIS 7) trial that clopidogrel 600 mg/ day can be safely used for 1 week without increasing the rate of major bleeding.7 Accordingly, the use of a 600-mg loading dose of clopidogrel should be recommended in patients with STEMI with high risk proles for primary angioplasty. How to identify nonresponders to clopidogrel will be the next major issue for those who require a loading dose of clopidogrel 600 mg for primary angioplasty.
Gen-Min Lin, MD Yi-Hwei Li, PhD Hualien, Taiwan 8 November 2010

vidual patient risk and mortality impact of ischemic and hemorrhagic complications: assessment from the Acute Catheterization and Urgent Intervention Triage Strategy trial. Circulation 2010;121:4351. 7. Mehta SR, Tanguay JF, Eikelboom JW, Jolly SS, Joyner CD, Granger CB, Faxon DP, Rupprecht HJ, Budaj A, Avezum A, Widimsky P, Steg PG, Bassand JP, Montalescot G, Macaya C, Di Pasquale G, Niemela K, Ajani AE, White HD, Chrolavicius S, Gao P, Fox KA, Yusuf S; on behalf of the CURRENT-OASIS 7 Trial Investigators. Double-dose versus standard-dose clopidogrel and high-dose versus low-dose aspirin in individuals undergoing percutaneous coronary 1intervention for acute coronary syndromes (CURRENT-OASIS 7): a randomised factorial trial. Lancet 2010;376:12331243.
doi:10.1016/j.amjcard.2010.11.001

1. Mangiacapra F, Muller O, Ntalianis A, Trana C, Heyndrickx GR, Bartunek J, Vanderheyden M, Wijns W, De Bruyne B, Barbato E. Comparison of 600 versus 300-mg clopidogrel loading dose in patients with ST-segment elevation myocardial infarction undergoing primary coronary angioplasty. Am J Cardiol 2010;106:1208 1211. 2. Chen KY, Rha SW, Li YJ, Poddar KL, Jin Z, Minami Y, Wang L, Kim EJ, Park CG, Seo HS, Oh DJ, Jeong MH, Ahn YK, Hong TJ, Kim YJ, Hur SH, Seong IW, Chae JK, Cho MC, Bae JH, Choi DH, Jang YS, Chae IH, Kim CJ, Yoon JH, Chung WS, Seung KB, Park SJ; Korea Acute Myocardial Infarction Registry Investigators. Triple versus dual antiplatelet therapy in patients with acute ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention. Circulation 2009;119: 32073214. 3. De Luca G, Suryapranata H, Stone GW, Antoniucci D, Tcheng JE, Neumann FJ, Bonizzoni E, Topol EJ, Chiariello M. Relationship between patients risk prole and benets in mortality from adjunctive abciximab to mechanical revascularization for ST-segment elevation myocardial infarction: a meta-regression analysis of randomized trials. J Am Coll Cardiol 2006;47:685 686. 4. Dangas G, Mehran R, Guagliumi G, Caixeta A, Witzenbichler B, Aoki J, Peruga JZ, Brodie BR, Dudek D, Kornowski R, Rabbani LE, Parise H, Stone GW; HORIZONS-AMI Trial Investigators. Role of clopidogrel loading dose in patients with ST-segment elevation myocardial infarction undergoing primary angioplasty: results from the HORIZONS-AMI (Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction) trial. J Am Coll Cardiol 2009;54:1438 1446. 5. Lin GM, Han CL. The loading dose of clopidogrel in patients with ST-segment elevation myocardial infarction undergoing primary angioplasty. Am J Emerg Med 2010;28:382383. 6. Pocock SJ, Mehran R, Clayton TC, Nikolsky E, Parise H, Fahy M, Lansky AJ, Bertrand ME, Lincoff AM, Moses JW, Ohman EM, White HD, Stone GW. Prognostic modeling of indi-

Long-Term Follow Up of Atrioventricular Block in Transcatheter Aortic Valve Implantation We have read with great interest the report by Roten et al1 regarding predictors of atrioventricular (AV) conduction impairment after transcatheter aortic valve implantation with the CoreValve prosthesis (Medtronic, Inc., Minneapolis, Minnesota). We have published our early experience with this prosthesis2 and recently performed electrophysiologic studies in some pacemaker-free patients immediately before and after valve implantation. An electrode was placed on the His bundle during valve implantation, and data were continuously recorded during the procedure. We agree that the type of AV block is intra- or infrahisian; in fact, our group has published for the rst time a report of intrahisian AV block in a patient who underwent percutaneous CoreValve prosthesis implantation.3 However, we are extremely surprised by the high AV conduction recovery rate, taking in account that infrahisian blocks are usually permanent. In our series with the rst 50 patients who underwent percutaneous implantation of the CoreValve prosthesis, pacemaker implantation was needed in 22 patients (44%), 20 patients because of complete AV block and 2 patients because of rst-degree AV block with newly developed left bundle branch block. Of the 20 patients, considering a subgroup of 15 patients who were discharged alive from the hospital and afwww.ajconline.org

Am J Cardiol 2011;107:641 0002-9149/11/$ see front matter 2011 Elsevier Inc. All rights reserved.

642

The American Journal of Cardiology (www.ajconline.org)

ter a median follow-up period of 435 days, only 2 patients recovered cardiac rhythm, and the remaining patients had escape rhythm, with a mean heart rate of 28 beats/min. Compared with Roten et als1 study, these differences may be explained in part by technical aspects of the procedure and differences in the prostheses implanted, as in our hospital, we implant only CoreValve prostheses. Considering our data and those of Roten et al1 (median follow-up period 79 days) together, we can speculate that AV conduction impairment seems to be

slowly progressive over time, which supports early pacemaker implantation in patients who develop second or third AV block despite detecting recovery of AV block.
Jos Rubn, MD, PhD Pablo Avanzas, MD, PhD Raquel del Valle, MD Cesar Mors, MD, PhD Oviedo, Spain 10 November 2010

1. Roten L, Wenaweser P, Delacrtaz E, Hellige G, Stortecky S, Tanner H, Pilgrim T, Kadner A, Eberle B, Zwahlen M, Carrel T, Meier B,

Windecker S. Incidence and predictors of atrioventricular conduction impairment after transcatheter aortic valve implantation. Am J Cardiol 2010;106:14731480. 2. Avanzas P, Muoz-Garca AJ, Segura J, Pan M, Alonso-Briales JH, Lozano I, Mors C, Surez de Lezo J, Hernndez-Garca JM. Percutaneous implantation of the CoreValve selfexpanding aortic valve prosthesis in patients with severe aortic stenosis: early experience in Spain. Rev Esp Cardiol 2010;63:141148. 3. Rubn J, Avanzas P, Calvo D, Moris C. Intrahisian block during transcatheter aortic valve implantation with the CoreValve prosthesis. Rev Esp Cardiol. In press.
doi:10.1016/j.amjcard.2010.11.013