The American Journal of Cardiology

The American Journal of
Cardiology
Volume 100, Issue 6, Pages.913-1046 (15 September 2007)
1.
Editorial Board
Page A2
2.
Contents
Pages A4-A6
Coronary Artery Disease
3.
Comparison of Utilization of Statin Therapy at Hospital Discharge and Six-Month
Outcomes in Patients With an Acute Coronary Syndrome and Serum Low-Density
Lipoprotein ≥100 mg/dl Versus <100 mg/dl
Pages 913-918
Frederick A. Spencer, Robert J. Goldberg, Joel M. Gore, Keith A.A. Fox, Alvaro
Avezum, Giancarlo Agnelli, Leonard Kritharides, Frederick A. Anderson, Shaun G.
Goodman, Gordon FitzGerald, et al.
4.
Meta-Analysis of the Role of Statin Therapy in Reducing Myocardial Infarction
Following Elective Percutaneous Coronary Intervention
Pages 919-923
Girish R. Mood, Anthony A. Bavry, Henri Roukoz and Deepak L. Bhatt
5.
Racial Disparity in the Utilization of Implantable-Cardioverter Defibrillators
Among Patients With Prior Myocardial Infarction and an Ejection Fraction of ≤35%
Pages 924-929
Kevin L. Thomas, Sana M. Al-Khatib, Richard C. Kelsey II, Heather Bush, Lynne
Brosius, Eric J. Velazquez, Eric D. Peterson and F. Roosevelt Gilliam
6.
Comparison of Myocardial Infarct Size Assessed With Contrast-Enhanced
Magnetic Resonance Imaging and Left Ventricular Function and Volumes to
Predict Mortality in Patients With Healed Myocardial Infarction
Pages 930-936
Stijntje D. Roes, Sebastian Kelle, Theodorus A.M. Kaandorp, Thomas Kokocinski,
Don Poldermans, Hildo J. Lamb, Eric Boersma, Ernst E. van der Wall, Eckart
Fleck, Albert de Roos, et al.
7.
Prevalence and Prognostic Implications of ST-Segment Deviations from
Ambulatory Holter Monitoring After ST-Segment Elevation Myocardial Infarction
Treated With Either Fibrinolysis or Primary Percutaneous Coronary Intervention (a
Danish Trial in Acute Myocardial Infarction-2 Substudy)
Pages 937-943
Lars Idorn, Dan Eik Høfsten, Kristian Wachtell, Henning Mølgaard and Kenneth
Egstrup
8.
Rapid Triage and Transport of Patients With ST-Elevation Myocardial Infarction
for Percutaneous Coronary Intervention in a Rural Health System
Pages 944-948
James C. Blankenship, Thomas A. Haldis, G. Craig Wood, Kimberly A. Skelding,
Thomas Scott and Francis J. Menapace
9.
Safety of Drug-Eluting Stents in the Coronary Artery in ST-Elevation Myocardial
Infarction at a Single High-Volume Medical Center
Pages 949-952
Rahul Bose, Gaurav Gupta, Paul A. Grayburn, Emily A. Laible, Mi Jung Kang and
James W. Choi
10.
Comparison of Virtual Histology to Intravascular Ultrasound of Culprit Coronary
Lesions in Acute Coronary Syndrome and Target Coronary Lesions in Stable
Angina Pectoris
Pages 953-959
Myeong-Ki Hong, Gary S. Mintz, Cheol Whan Lee, Jon Suh, Jeong-Hoon Kim,
Duk-Woo Park, Seung-Whan Lee, Young-Hak Kim, Sang-Sig Cheong, Jae-Joong
Kim, et al.
11.
Prevalence of Obstructive Coronary Artery Disease in Patients With and Without
Prior Stroke Undergoing Coronary Angiography for Suspected Coronary Artery
Disease
Pages 960-961
Rasham Sandhu, Wilbert S. Aronow, Rishi Sukhija and Archana Rajdev
12.
Examination of the Microcirculation Damage in Smokers Versus Nonsmokers
With Vasospastic Angina Pectoris
Pages 962-964
Takashi Ashikaga, Mitsuhiro Nishizaki, Hiroyuki Fujii, Saori Niki, Shingo Maeda,
Noriyoshi Yamawake, Yukio Kishi and Mitsuaki Isobe
13.
Correlates of Clinical Restenosis Following Intracoronary Implantation of DrugEluting Stents
Pages 965-969
Probal Roy, Teruo Okabe, Tina L. Pinto Slottow, Daniel H. Steinberg, Kimberly
Smith, Rebecca Torguson, Zhenyi Xue, Natalie Gevorkian, Lowell F. Satler,
Kenneth M. Kent, et al.
14.
Comparison of Drug-Eluting Stents Versus Surgery for Unprotected Left Main
Coronary Artery Disease
Pages 970-973
Marcelo Sanmartín, José Antonio Baz, Ramon Claro, Vanesa Asorey, Darío
Durán, Gonzalo Pradas and Andrés Iñiguez
Roundtable Discussion
15.
The Editor's Roundtable: Arterial Thrombosis and Acute Coronary Syndromes
Pages 974-980
Vincent E. Friedewald, Eric R. Bates, Christopher B. Granger, Salim Yusuf and
William C. Roberts
Preventive Cardiology
16.
Ethnic Differences in Coronary Artery Calcium in a Healthy Cohort Aged 60 to 69
Years
Pages 981-985
Joan M. Fair, Alexandre Kiazand, Ann Varady, Mohammed Mahbouba, Linda
Norton, Geoffrey D. Rubin, Carlos Iribarren, Alan S. Go, Mark A. Hlatky and
Stephen P. Fortmann
Arrhythmias and Conduction Disturbances
17.
Endotoxemia, Inflammation, and Atrial Fibrillation
Pages 986-988
Christopher J. Boos, Gregory Y.H. Lip and Bernd Jilma
18.
Levels of Circulating Procoagulant Microparticles in Nonvalvular Atrial
Fibrillation
Pages 989-994
Stéphane Ederhy, Emanuele Di Angelantonio, Ziad Mallat, Bénédicte Hugel,
Sandra Janower, Catherine Meuleman, Franck Boccara, Jean-Marie Freyssinet,
Alain Tedgui and Ariel Cohen
19.
Prevalence of Interatrial Block in Young Healthy Men <35 Years of Age
Pages 995-997
Elias Gialafos, Theodora Psaltopoulou, Theodore G. Papaioannou, Andreas
Synetos, Polychronis Dilaveris, George Andrikopoulos, Konstantinos Vlasis, John
Gialafos and Christodoulos Stefanadis
Heart Failure
20.
Reliability of Echocardiography for Hemodynamic Assessment of End-Stage
Heart Failure
Pages 998-1001
Nicolas Mansencal, Laure Revault d’Allonnes, Alain Beauchet, Séverine Fabre,
Franck Digne, Jean-Christian Farcot, Thierry Joseph and Olivier Dubourg
21.
Optimizing the Programation of Cardiac Resynchronization Therapy Devices in
Patients With Heart Failure and Left Bundle Branch Block
Pages 1002-1006
Bàrbara Vidal, Marta Sitges, Alba Marigliano, Victoria Delgado, Ernesto DíazInfante, Manel Azqueta, David Tamborero, José María Tolosana, Antonio
Berruezo, Félix Pérez-Villa, et al.
22.
Comparison of the Effects of Cardiac Resynchronization Therapy in Patients With
Class II Versus Class III and IV Heart Failure (from the InSync/InSync ICD Italian
Registry),
Pages 1007-1012
Maurizio Landolina, Maurizio Lunati, Maurizio Gasparini, Massimo Santini, Luigi
Padeletti, Augusto Achilli, Stefano Bianchi, Francesco Laurenzi, Antonio Curnis,
Antonio Vincenti, et al
Miscellaneous
23.
Frequency, Determinants, and Clinical Relevance of Acute Coronary SyndromeLike Electrocardiographic Findings in Patients With Acute Aortic Syndrome
Pages 1013-1019
Elena Biagini, Carla Lofiego, Marinella Ferlito, Rossella Fattori, Guido Rocchi,
Maddalena Graziosi, Luigi Lovato, Lara di Diodoro, Robin M.T. Cooke, Elisabetta
Petracci, et al.
24.
Predictors of Survival in Patients With End-Stage Renal Disease Evaluated for
Kidney Transplantation
Pages 1020-1025
Fadi G. Hage, Stuart Smalheiser, Gilbert J. Zoghbi, Gilbert J. Perry, Mark
Deierhoi, David Warnock, Ami E. Iskandrian, Angelo M. de Mattos and Raed A.
Aqel
25.
Prognosis of Idiopathic Recurrent Pericarditis as Determined from Previously
Published Reports
Pages 1026-1028
Massimo Imazio, Antonio Brucato, Yehuda Adler, Giovanni Brambilla, Galit
Artom, Enrico Cecchi, Giancarlo Palmieri and Rita Trinchero
26.
Comparison of Frequency of Complex Ventricular Arrhythmias in Patients With
Positive Versus Negative Anti-Ro/SSA and Connective Tissue Disease
Pages 1029-1034
Pietro Enea Lazzerini, Pier Leopoldo Capecchi, Francesca Guideri, Francesca
Bellisai, Enrico Selvi, Maurizio Acampa, Agnese Costa, Roberta Maggio, Estrella
Garcia-Gonzalez, Stefania Bisogno, et al.
27.
Effect of Growth Hormone on Cardiac Contractility in Patients With Adult Onset
Growth Hormone Deficiency
Pages 1035-1039
Goo-Yeong Cho, In-Kyung Jeong, Seong Hwan Kim, Min-Kyu Kim, Woo-Jung
Park, Dong-Jin Oh and Hyung-Joon Yoo
Review
28.
Crack Whips the Heart: A Review of the Cardiovascular Toxicity of Cocaine
Pages 1040-1043
Luis Afonso, Tamam Mohammad and Deepak Thatai
Editorial
29.
The “Clopidogrel Resistance” Trap
Pages 1044-1046
Victor L. Serebruany
EDITOR IN CHIEF
William C. Roberts,
ASSOCIATE EDITORS
Paul A. Grayburn
Clyde W. Yancy
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
ASSISTANT EDITORS
Vincent E. Friedewald
Robert C. Kowal
Jeffrey M. Schussler
Carlos E. Velasco
EDITORIAL BOARD
CARDIOVASCULAR
MEDICINE
In Adults
Antonio Abbate
J. Dawn Abbott
George S. Abela
Jonathan Abrams
Joseph S. Alpert
Martin A. Alpert
Ezra A. Amsterdam
Jeffrey L. Anderson
Richard W. Asinger
Pablo Avanzas
Gary John Balady
Thomas M. Bashore
Eric Bates
Jeroen J. Bax
George A. Beller
William E. Boden
Monty M. Bodenheimer
Robert O. Bonow
Jeffrey S. Borer
Harisios Boudoulas
Martial G. Bourassa
Eugene Braunwald
Jeffrey A. Brinker
David L. Brown
Alfred E. Buxton
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
Gregory J. Dehmer
Efthymios N. Deliargyris
James A. de Lemos
Anthony N. DeMaria
Pablo Denes
George A. Diamond
John P. DiMarco
Michael J. Domanski
Gerald Dorros
Uri Elkayam
Kenneth A. Ellenbogen
Myrvin H. Ellestad
Stephen G. Ellis
Toby R. Engel
Andrew E. Epstein
A2
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
James S. Forrester
Joseph A. Franciosa
Gary S. Francis
W. Bruce Fye
William H. Gaasch
William Ganz
Julius M. Gardin
Bernard J. Gersh
Mihai Gheorghiade
Raymond Gibbons
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
Richard H. Helfant
Philip D. Henry
L. David Hillis
David R. Holmes, Jr.
Mun K. Hong
Yuling Hong
William G. Hundley
Ami S. Iskandrian
Allan S. Jaffe
Joel S. Karliner
John A. Kastor
Sanjiv Kaul
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 Kligfield
Robert A. Kloner
John B. Kostis
Charles Landau
Richard L. Lange
Carl J. Lavie
Carl V. Leier
Joseph Lindsay, Jr.
Gregory Y.H. Lip
Joseph Loscalzo
G.B. John Mancini
Francis E. Marchlinski
Frank I. Marcus
Barry J. Maron
Randolph P. Martin
Attilo Maseri
Dean T. Mason
Charles Maynard
Michael D. McGoon
Darren K. McGuire
Raymond G. McKay
Jawahar L. Mehta
Richard S. Meltzer
Franz H. Messerli
Eric L. Michelson
Richard V. Milani
Alan B. Miller
Wayne L. Miller
Gary S. Mintz
Fred Morady
Arthur J. Moss
James E. Muller
Robert J. Myerburg
Gerald B. Naccarelli
Navin C. Nanda
Christopher O’Connor
Robert A. O’Rourke
Erik Magnus Ohman
Antonio Pacifico
Richard L. Page
Sebastian T. Palmeri
Eugene R. Passamani
Alan S. Pearlman
Carl J. Pepine
Joseph K. Perloff
Bertram Pitt
Don Poldermans
Philip J. Podrid
Arshed A. Quyyumi
Charles E. Rackley
C. Venkata Ram
Nathaniel Reichek
Robert Roberts
William J. Rogers
Maurice E. Sarano
Melvin M. Scheinman
David J. Schneider
John S. Schroeder
Pravin M. Shah
Prediman K. Shah
Jamshid Shirani
Robert J. Siegel
Marc A. Silver
Mark E. Silverman
Ross J. Simpson, Jr.
Steven N. Singh
Sidney C. Smith, Jr.
Burton E. Sobel
John C. Somberg
David H. Spodick
Lynne W. Stevenson
John R. Stratton
Jonathan M. Tobis
Eric J. Topol
Teresa S. M. Tsang
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
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
Thomas L. Spray
RELATED SPECIALISTS
L. Maximilian Buja
Michael Emmett
Barry A. Franklin
Charles B. Higgins
Jeffrey E. Saffitz
Renu Virmani
THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 100, NO. 6 SEPTEMBER 15, 2007
CONTENTS
Coronary Artery Disease
Comparison of Utilization of Statin Therapy at
Hospital Discharge and Six-Month Outcomes in
Patients With an Acute Coronary Syndrome and
Serum Low-Density Lipoprotein >100 mg/dl Versus
<100 mg/dl ..................................................913
Frederick A. Spencer, Robert J. Goldberg, Joel M. Gore,
Keith A.A. Fox, Alvaro Avezum, Giancarlo Agnelli,
Leonard Kritharides, Frederick A. Anderson,
Shaun G. Goodman, Gordon FitzGerald,
Jeanna Allegrone, and David Brieger, for the GRACE
Investigators
Meta-Analysis of the Role of Statin Therapy in
Reducing Myocardial Infarction Following Elective
Percutaneous Coronary Intervention .................919
Girish R. Mood, Anthony A. Bavry, Henri Roukoz, and
Deepak L. Bhatt
Racial Disparity in the Utilization of ImplantableCardioverter Defibrillators Among Patients With
Prior Myocardial Infarction and an Ejection Fraction
of <35% ........................................................924
Kevin L. Thomas, Sana M. Al-Khatib,
Richard C. Kelsey II, Heather Bush, Lynne Brosius,
Eric J. Velazquez, Eric D. Peterson, and
F. Roosevelt Gilliam
James C. Blankenship, Thomas A. Haldis, Craig Wood,
Kimberly A. Skelding, Thomas Scott, and Francis J. Menapace
Safety of Drug-Eluting Stents in the Coronary Artery
in ST-Elevation Myocardial Infarction at a Single
High-Volume Medical Center ............................949
Rahul Bose, Gaurav Gupta, Paul A. Grayburn,
Emily A. Laible, Mi Jung Kang, and James W. Choi
Comparison of Virtual Histology to Intravascular
Ultrasound of Culprit Coronary Lesions in Acute
Coronary Syndrome and Target Coronary Lesions in
Stable Angina Pectoris ....................................953
Myeong-Ki Hong, Gary S. Mintz, Cheol Whan Lee,
Jon Suh, Jeong-Hoon Kim, Duk-Woo Park,
Seung-Whan Lee, Young-Hak Kim, Sang-Sig Cheong,
Jae-Joong Kim, Seong-Wook Park, and
Seung-Jung Park
Prevalence of Obstructive Coronary Artery Disease
in Patients With and Without Prior Stroke
Undergoing Coronary Angiography for Suspected
Coronary Artery Disease .................................960
Rasham Sandhu, Wilbert S. Aronow, Rishi Sukhija, and
Archana Rajdev
Comparison of Myocardial Infarct Size Assessed
With Contrast-Enhanced Magnetic Resonance
Imaging and Left Ventricular Function and Volumes
to Predict Mortality in Patients With Healed
Myocardial Infarction ......................................930
Stijntje D. Roes, Sebastian Kelle,
Theodorus A.M. Kaandorp, Thomas Kokocinski,
Don Poldermans, Hildo J. Lamb, Eric Boersma,
Ernst E. van der Wall, Eckart Fleck, Albert de Roos,
Eike Nagel, and Jeroen J. Bax
Prevalence and Prognostic Implications of
ST-Segment Deviations from Ambulatory Holter
Monitoring After ST-Segment Elevation Myocardial
Infarction Treated With Either Fibrinolysis or
Primary Percutaneous Coronary Intervention (a
Danish Trial in Acute Myocardial Infarction-2
Substudy) .......................................................937
Lars Idorn, Dan Eik Høfsten, Kristian Wachtell,
Henning Mølgaard, and Kenneth Egstrup, for the
DANAMI-2 Investigators
A4 THE AMERICAN JOURNAL OF CARDIOLOGY姞
Rapid Triage and Transport of Patients With STElevation Myocardial Infarction for Percutaneous
Coronary Intervention in a Rural Health System .....944
Examination of the Microcirculation Damage in
Smokers Versus Nonsmokers With Vasospastic
Angina Pectoris ..............................................962
Takashi Ashikaga, Mitsuhiro Nishizaki, Hiroyuki Fujii,
Saori Niki, Shingo Maeda, Noriyoshi Yamawake,
Yukio Kishi, and Mitsuaki Isobe
Correlates of Clinical Restenosis Following
Intracoronary Implantation of Drug-Eluting
Stents .............................................................965
Probal Roy, Teruo Okabe, Tina L. Pinto Slottow,
Daniel H. Steinberg, Kimberly Smith, Rebecca Torguson,
Zhenyi Xue, Natalie Gevorkian, Lowell F. Satler,
Kenneth M. Kent, William O. Suddath,
Augusto D. Pichard, and Ron Waksman
Comparison of Drug-Eluting Stents Versus Surgery
for Unprotected Left Main Coronary Artery
Disease ..........................................................970
Marcelo Sanmartı́n, José Antonio Baz, Ramon Claro,
Vanesa Asorey, Darı́o Durán, Gonzalo Pradas, and
Andrés Iñiguez
VOL. 100
SEPTEMBER 15, 2007
Roundtable Discussion (CME)
The Editor’s Roundtable: Arterial Thrombosis and
Acute Coronary Syndromes .............................974
Vincent E. Friedewald, Eric R. Bates,
Christopher B. Granger, Salim Yusuf, and
William C. Roberts
Preventive Cardiology
Ethnic Differences in Coronary Artery Calcium in a
Healthy Cohort Aged 60 to 69 Years ................981
Joan M. Fair, Alexandre Kiazand, Ann Varady,
Mohammed Mahbouba, Linda Norton,
Geoffrey D. Rubin, Carlos Iribarren, Alan S. Go,
Mark A. Hlatky, and Stephen P. Fortmann
Arrhythmias and Conduction
Disturbances
Endotoxemia, Inflammation, and Atrial
Fibrillation ......................................................986
Christopher J. Boos, Gregory Y.H. Lip, and
Bernd Jilma
Levels of Circulating Procoagulant Microparticles in
Nonvalvular Atrial Fibrillation ..........................989
Comparison of the Effects of Cardiac
Resynchronization Therapy in Patients With Class II
Versus Class III and IV Heart Failure (from the
InSync/InSync ICD Italian Registry) .................1007
Maurizio Landolina, Maurizio Lunati,
Maurizio Gasparini, Massimo Santini, Luigi Padeletti,
Augusto Achilli, Stefano Bianchi, Francesco Laurenzi,
Antonio Curnis, Antonio Vincenti, Sergio Valsecchi, and
Alessandra Denaro, on behalf of the InSync/InSync ICD
Italian Registry Investigators
Miscellaneous
Frequency, Determinants, and Clinical Relevance of
Acute Coronary Syndrome-Like Electrocardiographic
Findings in Patients With Acute Aortic
Syndrome .....................................................1013
Elena Biagini, Carla Lofiego, Marinella Ferlito,
Rossella Fattori, Guido Rocchi, Maddalena Graziosi,
Luigi Lovato, Lara di Diodoro, Robin M.T. Cooke,
Elisabetta Petracci, Letizia Bacchi-Reggiani,
Romano Zannoli, Angelo Branzi, and Claudio Rapezzi
Predictors of Survival in Patients With End-Stage
Renal Disease Evaluated for Kidney
Transplantation .............................................1020
Stéphane Ederhy, Emanuele Di Angelantonio,
Ziad Mallat, Bénédicte Hugel, Sandra Janower,
Catherine Meuleman, Franck Boccara,
Jean-Marie Freyssinet, Alain Tedgui, and Ariel Cohen
Fadi G. Hage, Stuart Smalheiser, Gilbert J. Zoghbi,
Gilbert J. Perry, Mark Deierhoi, David Warnock,
Ami E. Iskandrian, Angelo M. de Mattos, and
Raed A. Aqel
Prevalence of Interatrial Block in Young Healthy
Men <35 Years of Age ...................................995
Prognosis of Idiopathic Recurrent Pericarditis as
Determined from Previously Published
Reports ........................................................1026
Elias Gialafos, Theodora Psaltopoulou,
Theodore G. Papaioannou, Andreas Synetos,
Polychronis Dilaveris, George Andrikopoulos,
Konstantinos Vlasis, John Gialafos, and
Christodoulos Stefanadis
Heart Failure
Reliability of Echocardiography for Hemodynamic
Assessment of End-Stage Heart Failure .............998
Nicolas Mansencal, Laure Revault d’Allonnes,
Alain Beauchet, Séverine Fabre, Franck Digne,
Jean-Christian Farcot, Thierry Joseph, and
Olivier Dubourg
Optimizing the Programation of Cardiac
Resynchronization Therapy Devices in Patients With
Heart Failure and Left Bundle Branch Block .....1002
Bàrbara Vidal, Marta Sitges, Alba Marigliano,
Victoria Delgado, Ernesto Dı́az-Infante, Manel Azqueta,
David Tamborero, José Marı́a Tolosana,
Antonio Berruezo, Félix Pérez-Villa, Carles Paré,
Lluı́s Mont, and Josep Brugada
Massimo Imazio, Antonio Brucato, Yehuda Adler,
Giovanni Brambilla, Galit Artom, Enrico Cecchi,
Giancarlo Palmieri, and Rita Trinchero
Comparison of Frequency of Complex Ventricular
Arrhythmias in Patients With Positive Versus
Negative Anti-Ro/SSA and Connective Tissue
Disease ........................................................1029
Pietro Enea Lazzerini, Pier Leopoldo Capecchi,
Francesca Guideri, Francesca Bellisai, Enrico Selvi,
Maurizio Acampa, Agnese Costa, Roberta Maggio,
Estrella Garcia-Gonzalez, Stefania Bisogno,
Gabriella Morozzi, Mauro Galeazzi, and
Franco Laghi-Pasini
Effect of Growth Hormone on Cardiac Contractility
in Patients With Adult Onset Growth Hormone
Deficiency .....................................................1035
Goo-Yeong Cho, In-Kyung Jeong, Seong Hwan Kim,
Min-Kyu Kim, Woo-Jung Park, Dong-Jin Oh, and
Hyung-Joon Yoo
CONTENTS
A5
Review
Crack Whips the Heart: A Review of the
Cardiovascular Toxicity of Cocaine .................1040
Luis Afonso, Tamam Mohammad, and Deepak Thatai
Instructions to Authors can be found at the AJC
website: www.AJConline.org
Full Text: www.ajconline.org
Editorial
The “Clopidogrel Resistance” Trap ..................1044
Victor L. Serebruany
Visit our INTERNET Home Page:
http://www.AJConline.org
Classifieds on pages A15–A17
A6 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 100
SEPTEMBER 15, 2007
Comparison of Utilization of Statin Therapy at Hospital
Discharge and Six-Month Outcomes in Patients With an Acute
Coronary Syndrome and Serum Low-Density
Lipoprotein >100 mg/dl Versus <100 mg/dl
Frederick A. Spencer, MDa,*, Robert J. Goldberg, PhDb, Joel M. Gore, MDb,
Keith A.A. Fox, MB, ChBd, Alvaro Avezum, MDe, Giancarlo Agnelli, MDf,
Leonard Kritharides, MBBS, PhDg, Frederick A. Anderson, PhDc, Shaun G. Goodman, MD, MSch,
Gordon FitzGerald, PhDc, Jeanna Allegrone, BAc, and David Brieger, MBBS, PhDg, for the
GRACE Investigators
The use of, factors associated with, and long-term outcomes related to statin therapy in
patients with acute coronary syndromes and low-density lipoprotein (LDL) levels <100
mg/dl at the time of hospital presentation are unclear. This report describes the use of
statins at hospital discharge in 8,492 patients with acute coronary syndromes enrolled in
the Global Registry of Acute Coronary Events (GRACE; 1999 to 2005) according to
baseline LDL levels (<100 vs >100 mg/dl) and compares 6-month outcomes in each group
stratified by the use or nonuse of statin therapy. Seventy-two percent of patients with LDL
levels >100 mg/dl, compared with 55% of patients with LDL levels <100 mg/dl, were
discharged on statin therapy. Sociodemographic, clinical, and treatment variables varied
between patients discharged on statins and those who were not. Patients receiving statins
at discharge were twofold (LDL >100 mg/dl) to threefold (<100 mg/dl) more likely to be
receiving statin therapy at 6 months compared with those not receiving statins at discharge.
Statin use at discharge was associated with a significantly lower rate of 6-month cardiac
complications in patients with LDL levels <100 mg/dl (adjusted odds ratio for the composite end point of myocardial infarction, stroke, and death 0.64, 95% confidence interval
0.47 to 0.88). In conclusion, data from this large observational study suggest that patients
with acute coronary syndromes and LDL levels <100 mg/dl are much less likely to be
prescribed statin therapy at hospital discharge or to be receiving statin therapy at 6 months
but benefit from the prescription of statins at hospital discharge as much as patients with
levels >100 mg/dl. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:
913–918)
The appropriateness and safety of early statin therapy in
patients with acute coronary syndromes (ACS) and lower
low-density lipoprotein (LDL) cholesterol levels has previa
Department of Medicine, McMaster University, Faculty of Health
Sciences, Hamilton, Ontario, Canada; bDepartment of Medicine, Division
of Cardiovascular Medicine, and cCenter for Outcomes Research, University of Massachusetts Medical School, Worcester, Massachusetts; dDepartment of Cardiology, The University and the Royal Infirmary of Edinburgh,
Edinburgh, United Kingdom; eDante Pazzanese Institute of Cardiology,
São Paulo, Brazil; fDivision of Internal and Cardiovascular Medicine,
University of Perugia, Perugia, Italy; gCoronary Care Unit, Concord Hospital, Sydney, Australia; hCanadian Heart Research Centre and Terrence
Donnelly Heart Centre, Division of Cardiology, St. Michael’s Hospital,
University of Toronto, Toronto, Ontario, Canada. Manuscript received
January 14, 2007; revised manuscript received and accepted April 24,
2007.
The Global Registry of Acute Coronary Events (GRACE) is supported
by an unrestricted educational grant from Sanofi-Aventis, Paris, France.
*Corresponding author: Tel: 905-521-2100 ext. 76973; fax: 905-5212336.
E-mail address: [email protected] (F.A. Spencer).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.023
ously been questioned.1 Although subsequent studies have
suggested that patients with acute myocardial infarctions
(AMIs) with lower LDL levels also benefit from statins,2,3
we hypothesized that the use of statin therapy in this group
might be suboptimal. Using data from a large multinational
registry, we describe the use of statin therapy at hospital
discharge in patients with ACS, describe factors associated
with the prescription of statins at discharge, and evaluate the
potential impact of statin therapy at discharge on 6-month
use and outcomes according to baseline LDL level (⬍100
vs ⱖ100 mg/dl).
Methods
Full details of the Global Registry of Acute Coronary
Events (GRACE) study design and the approaches used for
data collection have been published and are briefly summarized.4 – 6 GRACE is designed to reflect an unselected population of patients with ACS, irrespective of geographic
region. Currently, 94 hospitals located in 13 countries are
participating in this observational study.
Patients entered in the registry had to be ⱖ18 years old
www.AJConline.org
914
The American Journal of Cardiology (www.AJConline.org)
and alive at the time of hospital presentation, be admitted
for ACS as a presumptive diagnosis (i.e., have symptoms
consistent with acute ischemia), and have ⱖ1 of the following: electrocardiographic changes consistent with ACS, serial increases in serum biochemical markers of cardiac necrosis, and documentation of coronary artery disease.4 – 6
The qualifying ACS must not have been precipitated by
significant noncardiovascular co-morbidities (e.g., trauma
or surgery). The study aimed to enroll an unselected population of patients with ACS; sites were encouraged to recruit
the first 10 to 20 consecutive eligible patients each month.
Regular audits are performed at all participating hospitals.
Where required, study investigators received approval from
their local hospital ethics or institutional review boards for
the conduct of this study.
Data were collected at each study site by a trained study
coordinator using a standardized case report form. Demographic characteristics, medical histories, presenting symptoms, durations of prehospital delay, biochemical and electrocardiographic findings, treatment practices, and hospital
outcome data were collected. Standardized definitions of all
patient-related variables, clinical diagnoses, and outcomes
were used. All cases were assigned to 1 of the following
categories: ST-segment elevation MI, non–ST-segment elevation MI, unstable angina, and other cardiac or noncardiac diagnoses that have been previously described.4 – 6 Statin use at discharge was based on data in medical records
and recorded as a yes–no variable. At a targeted interval of
6 months after hospital discharge, standardized follow-up
forms were completed during telephone calls to patients or
their next of kin to ascertain statin use and the occurrence of
selected long-term outcomes, including recurrent MI,
stroke, and death.
The study population consisted of all patients enrolled in
GRACE from August 1999 to December 2005 who survived until hospital discharge (n ⫽ 52,688) and met none of
the following exclusion criteria: lipid profile not drawn
during hospitalization (n ⫽ 30,381), transferred into or out
of a participating GRACE medical center (n ⫽ 11,858),
previously taking a statin or other lipid-lowering medication
(n ⫽ 4,904), or subsequently found to have a noncardiac or
other cardiac (non-ACS) primary diagnosis (n ⫽ 1,514). Of
11,397 patients meeting these criteria, 6-month follow-up
data were available for 8,492 patients from 13 countries.
Differences in the demographic and clinical characteristics as well as additional treatment practices of patients in
relation to statin therapy at discharge, further stratified according to LDL levels (⬍100 vs ⱖ100 mg/dl), were analyzed using chi-square tests for discrete variables. Wilcoxon’s rank-sum test was used to analyze differences between
respective comparison groups for continuous variables. Differences in statin use at 6 months after hospital discharge
and in 6-month clinical outcomes (death, stroke, AMI, and
the composite end point of death, stroke, and/or AMI) in
relation to statin use at discharge, further stratified by LDL
level (⬍100 vs ⱖ100 mg/dl, as well as ⬍70, 70 to 99, 100
to 129, 130 to 160, and ⬎160 mg/dl for 6-month death),
were analyzed with similar statistical tests. Logistic regression analyses were used to evaluate the association of statin
therapy at discharge with total mortality and the primary
composite end point of recurrent MI, stroke, and/or death at
6 months after hospital discharge. Candidate variables in
our regression models included demographic characteristics, medical history, previous medications, type of ACS,
clinical presentation characteristics, in-hospital therapies
and procedures, in-hospital complications, and medications
at discharge (specific variables are listed in Table 1). This
analysis was performed in the total patient subset as well as
in patients further stratified according to baseline LDL level.
Candidate variables possibly associated with the outcomes
of interest (p ⬍0.25 after univariate analysis) were included
in the multivariate models. Variables with p ⬎0.05 were
eliminated in a stepwise fashion so that only those with
statistically significant associations with the outcome of
interest were included in the final regression models. We
also tested for any possible interactions between statin use
at hospital discharge, LDL level (as a continuous variable or
stratified into cut points: ⬍70, 70 to 99, 100 to 130, or ⬎130
mg/dl), and the occurrence of 6-month death or our composite end point.
Results
Among the 2,326 patients with hospital LDL levels ⬍100
mg/dl (median LDL 81 mg/dl), 1,281 (55%) were receiving
statin therapy at the time of hospital discharge; of the 6,166
patients with hospital LDL levels ⱖ100 mg/dl (median LDL
136 mg/dl), 4,429 (72%) were receiving statin therapy at
hospital discharge. Irrespective of baseline LDL level, patients taking statins at hospital discharge were younger,
more likely to be men, and more likely to have histories of
hyperlipidemia or smoking, but less likely to have histories
of congestive heart failure or vascular disease, than those
not prescribed statin therapy (Table 1). They were also less
likely to have previously been taking aspirin, ␤ blockers, or
angiotensin-converting enzyme inhibitors. Regardless of
LDL level, patients receiving statins at the time of hospital
discharge were more likely to have been treated with effective cardiac medications or to have undergone revascularization procedures than those not receiving statin therapy.
Patients receiving statin therapy at discharge were more
likely to have final diagnoses of ST-segment-elevation MI
and to be discharged on aspirin, angiotensin-converting
enzyme inhibitors, and ␤ blockers than patients not taking
statins at hospital discharge.
In patients with LDL levels ⬍100 mg/dl, those taking
statins at discharge were less likely to have had congestive
heart failure, pulmonary edema, or renal failure during hospitalization than those not taking statins at hospital discharge. Patients with LDL levels ⱖ100 mg/dl who were
taking statins at hospital discharge were less likely to have
developed congestive heart failure, pulmonary edema, major bleeding, or renal failure during hospitalization but were
more likely to have had cardiac arrest than patients with the
same LDL levels who were not taking statins at discharge.
Patients with LDL levels ⬍100 and ⱖ100 mg/dl at
baseline who were taking statins at hospital discharge were,
respectively, 3.4 and 2.5 times more likely to be receiving
statin therapy at 6 months than those who were not prescribed statin therapy at discharge (Table 2). Only 25% of
patients surviving ACS with baseline LDL levels ⬍100
mg/dl who were not prescribed statins at discharge were
Coronary Artery Disease/LDL Levels, Statin Use, and ACS
915
Table 1
Characteristics of patients (n ⫽ 8,492)* according to baseline low-density lipoprotein level and statin therapy at hospital discharge
LDL ⬍100 mg/dl
Variable
Age (yrs), median
Women
Heart failure
Diabetes mellitus
Hypertension
Hyperlipidemia
Smoker
Vascular disease
Aspirin use
ACE inhibitor use
␤-blocker use
Presenting clinical characteristics
Heart rate (beats/min), median
Systolic blood pressure (mm Hg), median
Positive cardiac markers
ST deviation on index electrocardiogram
Killip class
I (no heart failure)
II (rales)
III (pulmonary edema)
IV (cardiogenic shock)
Initial creatinine (mg/dl)
In-hospital management
Aspirin
ACE inhibitor
␤ blocker
Glycoprotein IIb/IIIa inhibitor
Low-molecular-weight heparin
Unfractionated heparin
Thienopyridine
Thrombolytic drug
Cardiac catheterization
Coronary angioplasty
Coronary bypass
In-hospital complications
Recurrent ischemic symptoms
Heart failure/pulmonary edema
Cardiogenic shock
Cardiac arrest
Major bleeding
Stroke
Renal failure
Final diagnosis
ST-segment elevation MI
Non–ST-segment elevation MI
Unstable angina
Discharge therapies
Aspirin
ACE inhibitor
␤ blocker
LDL ⱖ100 mg/dl
Statin (⫹)
(n ⫽ 1,281)
Statin (0)
(n ⫽ 1,045)
p Value
Statin (⫹)
(n ⫽ 4,429)
Statin (0)
(n ⫽ 1,737)
p Value
65.5
26.7%
7.1%
25.7%
60.3%
31.4%
61.8%
38.8%
29.2%
21.8%
23.1%
69.4
33.7%
14.2%
26.5%
59.8%
19.1%
53.7%
46.2%
32.9%
25.4%
26.7%
⬍0.001
⬍0.001
⬍0.001
0.68
0.81
⬍0.001
⬍0.001
⬍0.001
0.06
0.04
0.04
61.2
29.5%
3.7%
17.2%
50.0%
38.1%
64.6%
26.1%
21.3%
15.8%
16.5%
67.7
36.2%
8.4%
19.0%
57.8%
29.0%
55.1%
39.1%
30.8%
21.6%
24.4%
⬍0.001
⬍0.001
⬍0.001
0.10
⬍0.001
⬍0.001
⬍0.001
⬍0.001
0.06
⬍0.001
⬍0.001
76
138
51.4%
65.8%
80
139
48.4%
57.4%
⬍0.001
0.80
0.16
⬍0.001
⬍0.001
76
140
50.1%
67.6%
78
140
45.2%
59.0%
0.002
0.91
⬍0.001
⬍0.001
⬍0.001
86.8%
10.5%
2.3%
0.5%
1.00
78.6%
16.0%
4.7%
0.7%
1.08
⬍0.001
88.5%
9.1%
2.0%
0.5%
1.00
83.6%
11.7%
4.1%
0.6%
1.04
⬍0.001
97.1%
75.9%
89.4%
43.4%
62.3%
50.9%
69.2%
15.9%
78.1%
56.9%
3.6%
92.6%
66.5%
82.0%
23.3%
56.9%
52.0%
41.1%
14.4%
57.1%
28.5%
4.4%
⬍0.001
⬍0.001
⬍0.001
⬍0.001
0.009
0.6
⬍0.001
0.33
⬍0.001
⬍0.001
0.32
96.8%
72.3%
90.2%
38.8%
66.7%
48.2%
67.5%
19.8%
76.6%
55.7%
4.0%
93.4%
65.6%
82.0%
22.9%
64.3%
47.6%
47.6%
16.8%
60.9%
34.8%
6.6%
⬍0.001
⬍0.001
⬍0.001
⬍0.001
0.08
0.63
⬍0.001
0.008
⬍0.001
⬍0.001
⬍0.001
17.5%
11.9%
1.9%
1.7%
3.8%
0.7%
3.2%
16.8%
16.4%
2.3%
1.3%
4.7%
0.5%
4.7%
0.67
0.002
0.48
0.36
0.25
0.48
0.06
⬍0.001
18.8%
9.3%
1.5%
2.0%
2.0%
0.4%
1.6%
22.2%
11.9%
1.6%
1.2%
3.0%
0.6%
2.5%
0.003
0.003
0.77
0.05
0.03
0.31
0.02
⬍0.001
50.2%
29.9%
19.9%
38.9%
35.7%
25.5%
50.6%
30.3%
19.1%
38.2%
32.9%
28.9%
94.4%
70.9%
84.3%
85.2%
57.2%
70.5%
95.1%
68.9%
84.4%
86.9%
57.0%
70.2%
⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001
* Patient enrollment by country: Argentina 370 (4.4%), Australia and New Zealand 209 (2.5%), Austria 121 (1.4%), Belgium 966 (11.4%), Brazil 804
(9.5%), Canada 530 (6.2%), France 1,253 (14.8%), Germany 444 (5.2%), Italy 206 (2.4%), Poland 1,054 (12.4%), Spain 739 (8.7%), United Kingdom 150
(1.8%), United States 1,646 (19.4%).
ACE ⫽ angiotensin-converting enzyme.
taking statins at 6 months, compared with 35% of patients
with baseline LDL levels ⱖ100 mg/dl who were not prescribed statins at discharge.
Patients receiving statin therapy at hospital discharge
were significantly less likely to have died or to have
developed our composite study end point during the 6
months after hospital discharge, irrespective of LDL
level (Table 2 and Figure 1). Patients with LDL levels
916
The American Journal of Cardiology (www.AJConline.org)
Table 2
Statin use and patient outcomes at six months stratified by initial low-density lipoprotein level and statin therapy at hospital discharge
LDL ⬍100 mg/dl
Variable
Taking statins at 6 months
6-mo outcomes
AMI
Stroke
Death
AMI/stroke/death
LDL ⱖ100 mg/dl
All Patients
Statin (⫹)
(n ⫽ 1,281)
Statin (0)
(n ⫽ 1,045)
p Value
Statin (⫹)
(n ⫽ 4,429)
Statin (0)
(n ⫽ 1,737)
p Value
Statin (⫹)
(n ⫽ 5,710)
Statin (0)
(n ⫽ 2,782)
p Value
88.0%
25.8%
⬍0.001
88.2%
34.8%
⬍0.001
88.2%
31.6%
⬍0.001
3.2%
0.7%
4.0%
7.0%
4.9%
0.9%
8.6%
13.5%
0.05
0.65
⬍0.001
⬍0.001
2.5%
0.5%
2.4%
5.1%
2.8%
0.7%
5.2%
7.8%
0.55
0.51
⬍0.001
⬍0.001
2.7%
0.6%
2.7%
5.5%
3.6%
0.7%
6.5%
9.9%
⬍.05
0.36
⬍0.001
⬍0.001
Figure 1. Unadjusted 6-month death rates in patients discharged on statin therapy compared with those not discharged on statin therapy according to baseline
LDL cholesterol levels. *Referent category: patients not discharged on statin therapy.
⬍100 mg/dl who were prescribed statins at discharge
were less likely to experience subsequent AMIs than
those not prescribed statins.
After multivariate analysis, statin prescription at the time
of hospital discharge in the total study population was
associated with a significant reduction in 6-month all-cause
death rates (odds ratio [OR] 0.66, 95% confidence interval
[CI] 0.51 to 0.85) and in the composite end point (OR 0.76,
95% CI 0.63 to 0.93). Statin prescription at discharge was
associated with a similar reduction in 6-month mortality in
patients with LDL levels ⬍100 mg/dl (OR 0.69, 95% CI
0.46 to 1.03) or ⱖ100 mg/dl (OR 0.74, 95% CI 0.53 to
1.01), but these associations were no longer statistically
significant. Statin prescription at hospital discharge remained significantly associated with a decreased occurrence
of the composite end point (adjusted OR 0.64, 95% CI 0.47
to 0.88) in patients with LDL levels ⬍100 mg/dl, whereas in
those with LDL ⱖ100 mg/dl, there was no significant reduction in the composite end point in patients treated with
statin therapy at hospital discharge versus those not treated
(OR 0.92, 95% CI 0.71 to 1.19). C-statistics for the 6
multiple logistic regression models ranged from 0.73 to
0.81.
No significant interaction was noted in our regression
model between statin use, LDL level when expressed as a
continuous variable, and rates of death or the composite end
point at 6 months (p ⫽ 0.70 and p ⫽ 0.09, respectively).
Similarly, no significant interactions were detected among
statin use, LDL further stratified using clinically relevant cut
points (⬍70, 70 to 99, 100 to 130, and ⬎130 mg/dl), and the
occurrence of either death or the composite end point (p ⫽
0.89 and p ⫽ 0.44, respectively).
Discussion
In this analysis of ⬎8,000 patients surviving hospitalization
for ACS, only half of patients with baseline LDL levels
⬍100 mg/dl were prescribed statin therapy at hospital discharge.
Of hospital survivors with baseline LDL levels ⬍100
mg/dl who were not prescribed statins at discharge, only 1/4
were receiving statin therapy at 6 months. Patients who had
been prescribed statins at discharge were ⬎3 times as likely
to be receiving statin therapy at 6 months as those who had
not been prescribed statins.
As seen in patients with LDL levels ⱖ100 mg/dl, patients with LDL levels ⬍100 mg/dl who were prescribed
statins at hospital discharge were younger, more likely to be
men, and less likely to have a histories of vascular disease
or to experience congestive heart failure or pulmonary
edema during hospitalization than those not prescribed statins. We have demonstrated similar patterns of prescription
Coronary Artery Disease/LDL Levels, Statin Use, and ACS
in other studies: lower risk patients are more likely to be
treated with effective therapies, including statins, than
higher risk patients.7,8 Recognition of the continued underuse of simple preventive therapies in patients at highest risk
is critical if we are to realize further improvement in ACS
survival.
In the past, concern had been raised about the potential
risks associated with treating patients with ACS and low
LDL levels at hospital presentation. In a subgroup analysis
of 12,365 patients enrolled in 2 large randomized trials of
oral glycoprotein IIb/IIIa inhibitor therapy for ACS, the
association between the initiation of early statin treatment
after ACS and 90-day and 1-year outcomes was evaluated.1
Among the 2,711 patients for whom baseline lipid levels
were recorded, a significant interaction among total cholesterol and LDL levels, early statin treatment, and postdischarge mortality was noted. For patients with LDL levels
⬍112 mg/dl, the use of early statins was associated with a
significant increase in 90-day postdischarge mortality.
These data suggested that early treatment with statins might
be harmful as baseline cholesterol levels decreased. Accordingly, the investigators of this study appropriately suggested
that caution should be exercised in starting statin therapy in
patients with ACS who do not meet current treatment guidelines until randomized controlled trials could further clarify
this observation.
Trials indirectly addressing these issues have subsequently been conducted. In the Myocardial Ischemia Reduction With Aggressive Cholesterol Lowering (MIRACL)
study, the initiation of high-dose atorvastatin ⬍1 to 4 days
after hospitalization for ACS was associated with a borderline statistical reduction in the primary end point of death,
MI, or recurrent ischemia at 16 weeks compared with placebo.2 No significant interaction between baseline LDL
level and outcome was noted. The Pravastatin or Atorvastatin Evaluation and Infection Therapy–Thrombolysis In
Myocardial Infarction 22 (PROVE-IT–TIMI 22) trial evaluated the use of intensive versus moderate lipid-lowering
treatment with statins in 4,135 patients with ACS and baseline total cholesterol levels ⬍240 mg/dl.6 Patients receiving
high-dose atorvastatin therapy achieved a median LDL level
of 62 mg/dl, whereas patients receiving pravastatin
achieved a median LDL level of 95 mg/dl. Patients randomized to atorvastatin had a 16% reduction in the combined
end point of death, MI, unstable angina, revascularization,
or stroke compared with patients receiving pravastatin.
These data suggest that target LDL levels ⬍70 mg/dl may
be preferable in patients with ACS. However, the benefits of
intensive lipid-lowering treatment appeared confined to patients with baseline LDL ⬎125 mg/dl. Patients with baseline LDL ⬍125 mg/dl who received atorvastatin realized no
significant benefit compared with patients treated with pravastatin.
In summary, in previous randomized trials (or studies
deriving data from randomized trials) the benefits of statin
therapy were confined to patients with higher LDL levels or
were independent of baseline LDL. Given the relative homogeneity of clinical trial populations, it is likely that clinical characteristics and outcomes in patients with lower
compared with higher LDL levels did not differ as greatly as
in our community-based analysis. As such, the findings
917
from these clinical trials are not necessarily generalizable to
patients with ACS who are seen in the community setting.
In our large multinational registry, patients with low LDL
levels at presentation were older and more likely to have
histories of other important co-morbidities than patients
with higher LDL levels. They also experienced markedly
worse long-term outcomes: overall 6-month mortality for
patients with LDL levels ⬍100 mg/dl was 6.5%, compared
with 2.7% for patients with LDL levels ⱖ100 mg/dl. As
such, it is not necessarily surprising that these high-risk
patients would benefit as much or more from the receipt of
early statin therapy as patients with higher LDL levels. This
hypothesis would also be consistent with the increasingly
accepted concept that the beneficial effects of statin treatment are not isolated to LDL lowering but stem from pleiotropic activities that result in plaque stabilization in highrisk patients.9
Although current Adult Treatment Panel III guidelines
do not presently recommend treatment with LDL-lowering
therapy in high-risk patients when serum LDL levels are
⬍100 mg/dl,10 a recent update endorsed by the American
Heart Association, the American College of Cardiology,
and the National Heart, Lung, and Blood Institute suggests
that this would be a reasonable therapeutic decision in stable
high-risk patients, given available data supporting a relation
between LDL lowering and decreased cardiac risk irrespective of baseline LDL level.11 The most recent American
College of Cardiology and American Heart Association
guidelines for the management of AMI now recommend the
prescription of statins at hospital discharge to all patients,
even those with baseline LDL levels ⬍100 mg/dl (class I
recommendation).12
Our data suggest that many physicians are reluctant to
initiate statin therapy during hospitalization in patients with
ACS with low LDL levels. In addition, it appears that when
this therapy is initiated, it is preferentially used in younger,
lower risk patients. Most physicians are aware that LDL
levels early after AMI may be falsely lowered. Thus, some
practitioners may be delaying the prescription of statins
until more accurate determinations of lipid levels may be
obtained. However, the low rate of statin use at 6 months in
these patients suggests that this is not routinely occurring.
Although one must always be cautious in interpreting the
effect of nonrandomized treatments on outcomes, our data
do not suggest any hazard, and possibly a benefit, associated
with early statin use in survivors of ACS with LDL levels
⬍100 mg/dl. In our study, patients with LDL levels ⬍100
mg/dl discharged on statin therapy were 3.4 times more
likely to be receiving statin therapy at 6 months, had a 31%
reduction in the adjusted odds of 6-month mortality, and had
a 36% reduction in the adjusted odds of the 6-month composite end point of death, MI, and/or stroke compared with
those not treated with statins. Long-term benefits associated
with statin use in patients with ACS and low LDL at
presentation have been described in 1 other report. In an
analysis of 155 patients admitted to the University of Michigan Medical Center with ACS and LDL ⬍80 mg/dl, statin
therapy at discharge was associated with increased survival
after discharge and a decreased incidence of death, reinfarction, or stroke at 6 months.13
As with the interpretation of findings from any observa-
918
The American Journal of Cardiology (www.AJConline.org)
tional study, considerable caution must be exercised when
evaluating a potential association between a nonrandomly
assigned treatment and short or longer term outcomes. By
virtue of our extensive data collection efforts, we were able
to control for a variety of potentially confounding variables
in examining the association among LDL levels, the receipt
of statin therapy, and long-term outcomes, but we cannot
claim to have identified or adequately measured all potential
confounders of this association. Indeed, the observed reduction in 6-month mortality with statin use at discharge was
greater than the reductions in AMI or stroke (by which
statins would be expected to exert their beneficial effect).
This suggests that the highest risk patients in our study were
most likely to be left untreated with statins and that we were
unable to completely control for this bias.
It must also be acknowledged that information about the
use of statins at 6 months was obtained by patient self-report
via telephone but was not confirmed by pharmaceutical
records or other means. We were also unable to comment on
the use of specific statin types or dosages or other cardiac
medications at 6 months.
4.
5.
6.
7.
8.
9.
Acknowledgments: We thank the physicians and nurses
who are participating in GRACE. Sophie Rushton-Smith,
PhD, provided editorial assistance and was funded by
Sanofi-Aventis, Paris France. To find out more about
GRACE, visit the Web site at http://www.outcomes.org/
grace. Dr. Spencer had full access to all of the data in the
study and takes responsibility for the integrity of the data
and the accuracy of the data analysis.
1. Newby LK, Kristinsson A, Bhapkar MV, Aylward PE, Dimas AP,
Klein WW, McGuire DK, Moliterno DJ, Verheugt FW, Weaver WD,
Califf RM, SYMPHONY and 2nd SYMPHONY Investigators. Sibrafiban vs. Aspirin to Yield Maximum Protection From Ischemic Heart
Events Post-acute Coronary Syndromes. Early statin initiation and
outcomes in patients with acute coronary syndromes. JAMA 2002;287:
3087–3095.
2. Schwartz GG, Olsson AG, Ezekowitz MD, Ganz P, Oliver MF, Waters
D, Zeiher A, Chaitman BR, Leslie S, Stern T, Myocardial Ischemia
Reduction with Aggressive Cholesterol Lowering (MIRACL) Study
Investigators. Effects of atorvastatin on early recurrent ischemic events
in acute coronary syndromes: the MIRACL study: a randomized controlled trial. JAMA 2001;285:1711–1718.
3. Cannon CP, Braunwald E, McCabe CH, Rader DJ, Rouleau JL, Belder
R, Joyal SV, Hill KA, Pfeffer MA, Skene AM, Pravastatin or Atorvastatin Evaluation and Infection Therapy-Thrombolysis in Myocar-
10.
11.
12.
13.
dial Infarction 22 Investigators. Intensive versus moderate lipid
lowering with statins after acute coronary syndromes. N Engl J Med
2004;350:1495–1504.
The GRACE Investigators. Rationale and design of the GRACE
(Global Registry of Acute Coronary Events) Project. A multinational
registry of patients hospitalized with acute coronary syndromes. Am
Heart J 2001;141:190 –199.
Eagle KA, Goodman SG, Avezum A, Budaj A, Sullivan CM, LopezSendon J. Practice variations and missed opportunities for reperfusion
in ST-segment elevation myocardial infarction: findings from the
Global Registry of Acute Coronary Events (GRACE). Lancet 2002;
359:373–377.
Steg PG, Goldberg RJ, Gore JM, Fox KA, Eagle KA, Flather MD,
Sadiq I, Kasper R, Rushton-Mellor SK, Anderson FA, GRACE Investigators. Baseline characteristics, management practices, and in-hospital outcomes of patients hospitalized with acute coronary syndromes
in the Global Registry of Acute Coronary Events (GRACE). Am J
Cardiol 2002;90:358 –363.
Spencer FA, Lessard D, Yarzebski J, Gore JM, Goldberg RJ. Decadelong changes in the use of combination evidence-based medical therapy at discharge for patients surviving acute myocardial infarction. Am
Heart J 2005;150:838 – 844.
Spencer FA, Allegrone J, Goldberg RJ, Gore JM, Fox KA, Granger
CB, Mehta RH, Brieger D, GRACE Investigators. Association of statin
therapy with outcomes of acute coronary syndromes: the GRACE
study. Ann Intern Med 2004;140:857– 866.
Rosenson RS, Tangney CC. Antiatherothrombotic properties of statins: implications for cardiovascular event reduction. JAMA 1998;
279:1643–1650.
Expert Panel on Detection, Evaluation, and Treatment of High Blood
Cholesterol in Adults. Executive summary of the third report of the
National Cholesterol Education Program (NCEP) Expert Panel on
Detection, Evaluation and Treatment of High Blood Cholesterol in
Adults (Adult Treatment Panel III). JAMA 2001;285:2486 –2497.
Grundy SM, Cleeman JI, Merz CN, Brewer HB, Clark LT, Hunninghake DB, Pasternak RC, Smith SC, Stone NJ; National Heart, Lung,
and Blood Institute; American College of Cardiology Foundation;
American Heart Association. Implications of recent clinical trials for
the National Cholesterol Education Program Adult Treatment Panel III
guidelines. Circulation 2004;110:227–239.
Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand
M, Hochman JS, Krumholz HM, Kushner FG, Lamas GA, et al;
American College of Cardiology; American Heart Association Task
Force on Practice Guidelines; Canadian Cardiovascular Society. ACC/
AHA guidelines for the management of patients with ST-elevation
myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines
(Committee to Revise the 1999 Guidelines for the Management of
Patients With Acute Myocardial Infarction). Circulation 2004;110:
e82– e292.
Tsai TT, Nallamothu BK, Mukherjee D, Rubenfire M, Fang J, Chan P,
Kline-Rogers E, Patel A, Armstrong DF, Eagle KA, Goldberg AD.
Effect of statin use in patients with acute coronary syndromes and a
serum low-density lipoprotein ⱕ80 mg. Am J Cardiol 2005;96:1491–
1493.
Meta-Analysis of the Role of Statin Therapy in Reducing Myocardial Infarction
Following Elective Percutaneous Coronary Intervention
Girish R. Mood, MD, Anthony A. Bavry, MD, MPH, Henri Roukoz, MD,
and Deepak L. Bhatt, MD*
Statin medications initiated during percutaneous coronary intervention have been evaluated in clinical trials mainly to assess if this therapy reduces subsequent restenosis. The
benefit of statin therapy on individual cardiovascular outcomes other than restenosis is
largely unknown. Hence, a meta-analysis of the available randomized trials was conducted
to evaluate individual cardiovascular outcomes with statin therapy compared with placebo
after elective percutaneous coronary intervention. In all, there were 6 studies available for
analysis (Prevention of Restenosis by Elisor After Transluminal Coronary Angioplasty
[PREDICT], Fluvastatin Angioplasty Restenosis [FLARE], the Lescol Intervention Prevention Study [LIPS], German Atorvastatin Intravascular Ultrasound [GAIN], Atorvastatin for Reduction of Myocardial Damage During Angioplasty [ARMYDA], and a study by
Briguori et al) that randomized 3,941 patients (1,967 to statins and 1,974 to placebos).
Clinical follow-up ranged from 1 day to 45 months. The incidence of myocardial infarction
was 3.0% in the statin group and 5.2% in the placebo group (odds ratio [OR] 0.57, 95%
confidence interval [CI] 0.42 to 0.78, p <0.0001). The incidence of all-cause mortality was
2.3% versus 3.0% (OR 0.74, 95% CI 0.50 to 1.1, p ⴝ 0.14), that of cardiovascular mortality
was 0.71% versus 1.2% (OR 0.58, 95% CI 0.30 to 1.11, p ⴝ 0.10), and that of repeat surgical
or percutaneous revascularization was 19.6% versus 21.9% (OR 0.89, 95% CI 0.78 to 1.02,
p ⴝ 0.098) in the statin arm versus the placebo arm, respectively. The incidence of stroke
was 0.4% in the statin arm and 0.08% in the placebo arm (OR 3.00, 95% CI 0.60 to 14.77,
p ⴝ 0.18). In conclusion, statin therapy initiated at the time of elective percutaneous
coronary intervention significantly reduces myocardial infarction. © 2007 Elsevier Inc.
All rights reserved. (Am J Cardiol 2007;100:919 –923)
Percutaneous coronary intervention (PCI) can result in myocardial injury that is reflected by an increase in creatinine
kinase-MB and troponin I or T isoenzymes.1 This periprocedural myocardial necrosis may be a significant predictor
of future adverse cardiac events,2,3 so research efforts have
focused on the prevention of these events with statin medications.1 Observational studies have shown that when this
therapy is initiated during elective PCI, myocardial damage
can be reduced.4,5 Clinical trials have also shown that statin
medications improve survival in patients for secondary prevention,6 as well as acute coronary syndromes.7 The effect
of this therapy after PCI is less well established. Accordingly, we sought to determine if statin therapy initiated at
the time of PCI reduces myocardial infarction (MI) and
other individual cardiovascular outcomes through metaanalysis.
Methods and Results
A computerized search of research published in the English
language from 1996 to 2006 using the Medline (National
Library of Medicine, Bethesda, Maryland) and Google
Scholar (Google Inc., Mountain View, California) databases
Department of Cardiovascular Medicine, Cleveland Clinic, Cleveland,
Ohio. Manuscript received February 21, 2007; revised manuscript received
and accepted April 24, 2007.
*Corresponding author: Tel: 216-445-4042; fax: 216-445-8531.
E-mail address: [email protected] (D.L. Bhatt).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.022
was conducted for randomized clinical trials using the
search terms “percutaneous coronary transluminal angioplasty,” “coronary intervention,” and “3-hydroxy-3-methylglutaryl-CoA reductase inhibitors.” We selected studies that
randomized patients who underwent elective PCI to statin
therapy versus placebo or usual care. Studies that were
performed entirely in the setting of unstable angina or acute
MI were excluded. We required that the statin was initiated
around the time of coronary intervention and that individual
outcome data were available. The primary end point was
MI. Secondary end points were all-cause mortality, cardiovascular mortality, surgical or percutaneous revascularization, and stroke. For each outcome, 2 independent reviewers
(GRM, HR) tabulated the number of events that occurred in
each arm of a trial. Discrepancies were resolved through a
third reviewer (AAB). Baseline information was also tabulated, such as patient demographics, time of statin initiation,
lipid profile, and the extent of clinical follow-up.
We used the intention-to-treat principle to calculate risks.
The incidence of an outcome was defined as the number of
cardiac events that occurred in subjects randomized to a
certain type of therapy during the extent of clinical followup. Odds ratios (ORs) were defined as the odds of an
outcome in those who received statin therapy compared
with the odds in those who received placebo or usual care.
A Mantel-Haenszel model was used to calculate each summary statistic. We assessed for heterogeneity between studies by computing the Q statistic and for publication bias by
www.AJConline.org
920
The American Journal of Cardiology (www.AJConline.org)
Table 1
Baseline characteristics of the study participants
Variable
Baseline demographics
Patients*
Mean age (yrs)
Men
Previous MI
Diabetes mellitus
Stroke
Hypertension
Smoking
Baseline medication
Aspirin
␤ blocker
Clopidogrel or ticlopidine
Statin therapy
Placebo arm
Time of statin initiation
relative to revascularization
(d)
Duration of follow-up (mo)
Completeness of follow-up
PREDICT8
FLARE9
GAIN10
LIPS11
ARMYDA12
347/348
58.3
83.7%
37.1%
7.2%
1.9%
30.6%
33.7%
409/425
60
82.5%
33.0%
4.1%
3.5%
33.1%
28.9%#
65/66
60.2
84.7%
35.1%
14.5%
—
53.4%
55.7%#
844/833
60.0
83.8%
44.4%
12.0%
2.6%
38.6%
71.5%†
76/77
64.5
85.6%
22.2%
23.5%
—
74.5%
24.8%#
100%‡
—
—
Pravastatin
40 mg/d
Placebo
1 d after
6
90%
100%
—
—
Fluvastatin 40
mg twice
daily
Placebo
15–30 d
before
10
93%
98%
56%
50%¶
Atorvastatin 20–40
mg/d
⬎1 agent**
1 d after
12
76%
97.6%
70.1%
—
Fluvastatin 40
mg twice
daily
Placebo
0–22 d after
45储
93%
Briguori et al13
226/225
62.5
86.7%
31%
22.2%
—
64.3%
32.1%#
96%‡
34%
97%
Atorvastatin
40 mg/d
100%
55.4%
100%
Per primary care
physician§
Placebo
7 d before
Placebo
3–31 d before
1
100%
⬍24 h
100%
* Statin arm/placebo arm.
†
Current and previous smokers.
‡
Aspirin 100 mg/day.
§
Atorvastatin 22 ⫾ 9 mg/day (29%), pravastatin 32 ⫾ 10 mg/day (29%), simvastatin 24 ⫾ 9 mg/day (39%), fluvastatin 80mg/day (3%).
储
Median period.
¶
Ticlopidine.
#
Current smokers.
** Statins other than atorvastatin 48.5%, fibrates 43.9%, cholestyramine 36.4%.
constructing Begg’s funnel plot. A paired Student’s t test
was used to compare means. All p values were 2 tailed, with
statistical significance set at 0.05, and confidence intervals
(CIs) were calculated at the 95% level. All analyses were
performed using Stata version 9.0 (StataCorp LP, College
Station, Texas).
A total of 58 studies were initially identified through the
review of published research. Among these, 6 studies involving 3,941 patients met our selection criteria.8 –13 The
Lescol Intervention Prevention Study (LIPS) was the largest
trial, with 1,677 patients. Most of the patient population
(⬎80%) underwent PCI for stable coronary disease,
whereas in LIPS, half the study participants had stable
angina. Clinical follow-up ranged from 1 day to 45 months
from the index PCI. The completeness of patient follow-up
ranged from 76% to 100% (Tables 1 and 2).
The cumulative incidences of MI in the statin group and
the placebo group were 3.0% and 5.2%, respectively (OR
0.57, 95% CI 0.42 to 0.78, p ⬍0.0001) with a weighted
mean duration of follow-up of 22.7 months and an absolute
difference between the groups of 2.2% (p ⬍0.0001) (Table
3). The chi-square value for heterogeneity (5 degrees of
freedom) among the studies was 5.35 (p ⫽ 0.38), with no
evidence of publication bias (p ⫽ 0.44). The OR of an early
MI, defined as an event occurring up to 1 month after PCI,
was 0.45 (95% CI 0.28 to 0.72, p ⫽ 0.001), whereas the OR
of a late MI, occurring 1 month after PCI, was 0.68 (95% CI
0.45 to 1.03, p ⫽ 0.071). The OR of a MI after the initiation
of statin therapy before PCI was 0.42 (95% CI 0.27 to 0.66,
p ⬍0.0001), whereas the OR of an MI subsequent to the
initiation of statin therapy after PCI was 0.77 (95% CI 0.50
to 1.19, p ⫽ 0.23) (Figure 1).
Among the patients randomized to statin therapy versus
placebo, the incidence of all-cause mortality was 2.3% versus 3.0%, respectively (OR 0.74, 95% CI 0.50 to 1.1, p ⫽
0.14). The weighted mean duration of follow-up was 22.7
months with an absolute difference between the groups of
0.7% (p ⫽ 0.15). The chi-square value for heterogeneity (2
degrees of freedom) among the studies was 2.86 (p ⫽ 0.24).
The cumulative incidence of cardiovascular mortality was
0.71% in the statin arm and 1.2% in placebo arm (OR 0.58,
95% CI 0.30 to 1.11, p ⫽ 0.10), with a weighted mean
duration of follow-up of 20.6 months and an absolute difference between the groups of 0.8% (p ⫽ 0.10). The chisquare value for heterogeneity (1 degree of freedom) among
the studies was 1.08 (p ⫽ 0.30). The cumulative incidences
of surgical or percutaneous revascularization in the statin
and placebo arms were 19.6% and 21.9%, respectively (OR
0.89, 95% CI 0.78 to 1.02, p ⫽ 0.098), with a weighted
mean duration of follow-up of 22.7 months. The chi-square
value for heterogeneity (3 degrees of freedom) was 5.11 (p
⫽ 0.16). The absolute difference between the groups was
2.3% (p ⫽ 0.098). The cumulative incidences of stroke in
the statin arm and the placebo arm were 0.4% and 0.08%,
respectively (OR 3.00, 95% CI 0.60 to 14.77, p ⫽ 0.18),
with a weighted mean duration of follow-up of 20.6 months
Coronary Artery Disease/Statin Therapy and Coronary Intervention
921
Table 2
Cholesterol data of study participants (statin arm/placebo arm)
Variable
Baseline
Total cholesterol (mg/dl)
LDL cholesterol (mg/dl)
Follow-up
Total cholesterol (mg/dl)
LDL cholesterol (mg/dl)
PREDICT8
FLARE9
GAIN10
LIPS11
ARMYDA12
Briguori et al13
228/231
155/157
222/223
153/153
228/242
155/166
200/199
131/132
—
—
197/196
121/122
195/239
119/159
—
102/149
156/215
86/140
—
95/147
—
—
168/193*
93/121*
* Level at index procedure.
LDL ⫽ low-density lipoprotein.
Table 3
Cardiac events among study participants (statin arm/placebo arm)
Adverse Outcome
All-cause mortality
Cardiovascular mortality
Nonfatal MI
Stroke
Revascularization§
PREDICT8
FLARE9
GAIN10
LIPS11
ARMYDA12,‡
Brigouri et al13,*
4/1
1/0
4/4
1/0†
66/75
3/7
—
3/10
—
86/82
0/0
0/0
0/2
2/0
7/16
36/49
13/24
30/38
2/1†
167/193
—
—
4/14
—
—
0/0
—
18/35
—
—
* Large non–Q-wave MI defined as creatinine kinase-MB ⬎5 times the upper limit of normal.
†
Fatal strokes.
‡
MI defined as creatinine kinase-MB ⬎2 times the upper limit of normal.
§
Includes percutaneous and surgical revascularization.
and an absolute difference between the groups of 0.3% (p ⫽
0.16). The chi-square value for heterogeneity (2 degrees of
freedom) was 0.23 (p ⫽ 0.89).
Discussion
In our analysis of 6 randomized studies with 3,941 patients,
we found that periprocedural statin therapy reduces subsequent MI by 43% in patients who undergo elective PCI
compared with placebo or usual care, with a weighted mean
follow-up period of 22.7 months. The number of patients
needed to treat to prevent 1 MI was 45. The reduction in MI
appeared to occur early and was sustained late after PCI,
and it is possible that the initiation of statin therapy before
PCI may be preferential to initiation after the procedure.
There was no significant reduction in the risk for all-cause
mortality, cardiovascular mortality, and revascularization
from statin therapy compared with placebo, although these
outcomes all trended toward reductions. Interestingly, the
incidence of stroke was nonsignificantly increased in the
statin group compared with the placebo group. A previous
large meta-analysis of statin therapy for secondary prevention documented a reduction in stroke as well as other
adverse cardiovascular outcomes,6 so it is likely that with
sufficient power and a longer duration of follow-up we
would have also shown a reduction in stroke from the use of
statin therapy.
Statins appear to be an important adjunct to routine
pharmacotherapy during elective PCI for stable coronary
artery disease. There was a significant improvement in cholesterol levels in the statin group, which was seen as early as
2 weeks. The safety of statin therapy was reported in 3
studies (Fluvastatin Angiographic Restenosis [FLARE],
LIPS and Atorvastatin for Reduction of Myocardial Damage During Angioplasty [ARMYDA]), with elevated liver
enzymes in 0.68% of patients receiving statin therapy compared with 0.23% receiving placebo. The incidence of discontinuation of study medication because of intolerance or
adverse effects was seen in 6.05% of patients in the statin
group compared with 6.32% in the placebo group. Because
there were no significant adverse effects noted between the
statin and placebo groups, the addition of statin therapy
should be considered to benefit the patients who undergo
elective PCI.
It is unclear whether the benefit of statin therapy is due
to acute anti-inflammatory effects, long-term lipid-lowering
effects, or both. In the FLARE study, statins were administered before PCI, and patients were followed for 10
months, which may have reduced MI because of a combination of these 2 mechanisms. Moreover, Briguori et al’s13
study, in which statins were administered approximately 2
weeks before PCI, showed a significant reduction in total
and low-density lipoprotein cholesterol despite short-term
follow-up of ⬍1 day after the procedure. This point illustrates the difficulty in delineating the mechanism of the
beneficial effect of periprocedural statin therapy.
Large cohort studies have reported the role of statins in
improving survival and morbidity after PCI.4,14 Apart from
lowering low-density lipoprotein cholesterol, statins are
postulated to be beneficial because of antiproliferative and
anti-inflammatory effects and improved endothelial function.15–18 An observational study showed that in-hospital
complications after coronary intervention were independent
922
The American Journal of Cardiology (www.AJConline.org)
Figure 1. Odds of MI after PCI. PREDICT ⫽ Prevention of Restenosis by Elisor After Transluminal Coronary Angioplasty.
of hyperlipidemia, and there was a trend toward increasing
benefits in those who were receiving statin therapy.19 Randomized trials with significant power are required to support
this hypothesis.20 Our analysis has further limitations. In the
German Atorvastatin Intravascular Ultrasound (GAIN)
study, the control arm included statin medications other
than atorvastatin, which was used in the statin arm. This
might have favored the control arm’s masking the actual
incidence of major adverse cardiovascular events. Also, a
wide range of follow-up periods, from 1 day to 45 months,
resulted in an inability to assess the long-term benefits of
statin therapy.
1. Bhatt DL, Topol EJ. Does creatinine kinase-MB elevation after percutaneous coronary intervention predict outcomes in 2005? Periprocedural cardiac enzyme elevation predicts adverse outcomes. Circulation 2005;112:906 –915.
2. Blankenship JC, Haldis T, Feit F, Hu T, Kleiman NS, Topol EJ,
Lincoff AM. Angiographic adverse events, creatine kinase-MB elevation, and ischemic end points complicating percutaneous coronary
intervention (a REPLACE-2 substudy). Am J Cardiol 2006;97:1591–
1596.
3. Ricciardi MJ, Davidson CJ, Gubernikoff G, Beohar N, Eckman LJ,
Parker MA, Bonow RO. Troponin I elevation and cardiac events after
percutaneous coronary intervention. Am Heart J 2003;145:522–528.
4. Chan AW, Bhatt DL, Chew DP, Quinn MJ, Moliterno DJ, Topol EJ,
Ellis SG. Early and sustained survival benefit associated with statin
therapy at the time of percutaneous coronary intervention. Circulation
2002;105:691– 696.
5. Mulukutla SR, Marroquin OC, Smith C, Varghese R, Anderson WD,
Lee JS, Cohen HA, Counihan PJ, Lee AB, Gulati V, McNamara D.
Effect of statin therapy prior to elective percutaneous coronary intervention on frequency of periprocedural myocardial injury. Am J Cardiol 2004;94:1363–1366.
6. Baigent C, Keech A, Kearney PM, Blackwell L, Buck G, Pollicino C,
Kirby A, Sourjina T, Peto R, Collins R, Simes R. Efficacy and safety
of cholesterol-lowering treatment: prospective meta-analysis of data
from 90,056 participants in 14 randomised trials of statins. Lancet
2005;366:1267–1278.
7. Bavry AA, Mood GR, Kumbhani DJ, Borek PP, Askari AT, Bhatt DL.
Long term benefit of statin therapy initiated during the hospitalization
for an acute coronary syndrome: a systematic review of randomized
trials. Am J Cardiovasc Drugs 2007;7:135–141.
8. Bertrand ME, McFadden EP, Fruchart JC, Van Belle E, Commeau P,
Grollier G, Bassand JP, Machecourt J, Cassagnes J, Mossard JM, et al.
Effect of pravastatin on angiographic restenosis after coronary balloon
angioplasty. The PREDICT Trial Investigators. Prevention of Restenosis by Elisor After Transluminal Coronary Angioplasty. J Am Coll
Cardiol 1997;30:863– 869.
9. Serruys PW, Foley DP, Jackson G, Bonnier H, Macaya C, Vrolix M,
Branzi A, Shepherd J, Suryapranata H, de Feyter PJ, et al. A randomized placebo-controlled trial of fluvastatin for prevention of restenosis
after successful coronary balloon angioplasty; final results of the Fluvastatin Angiographic Restenosis (FLARE) trial. Eur Heart J 1999;
20:58 – 69.
10. Schartl M, Bocksch W, Koschyk DH, Voelker W, Karsch KR, Kreuzer
J, Hausmann D, Beckmann S, Gross M. Use of intravascular ultrasound to compare effects of different strategies of lipid-lowering
therapy on plaque volume and composition in patients with coronary
artery disease (GAIN). Circulation 2001;104:387–392.
11. Serruys PW, de Feyter P, Macaya C, Kokott N, Puel J, Vrolix M,
Branzi A, Bertolami MC, Jackson G, Strauss B, Meier B. Fluvastatin
for prevention of cardiac events following successful first percutaneous coronary intervention: a randomized controlled trial. JAMA 2002;
287:3215–3222.
12. Pasceri V, Patti G, Nusca A, Pristipino C, Richichi G, Di Sciascio G.
Randomized trial of atorvastatin for reduction of myocardial damage
during coronary intervention: results from the ARMYDA (Atorvastatin for Reduction of Myocardial Damage During Angioplasty) study.
Circulation 2004;110:674 – 678.
13. Briguori C, Colombo A, Airoldi F, Violante A, Focaccio A, Balestrieri
P, Paolo Elia P, Golia B, Lepore S, et al. Statin administration before
percutaneous coronary intervention: impact on periprocedural myocardial infarction. Eur Heart J 2004;25:1822–1828.
14. Chan AW, Bhatt DL, Chew DP, Reginelli J, Schneider JP, Topol
EJ, Ellis SG. Relation of inflammation and benefit of statins after
percutaneous coronary interventions. Circulation 2003;107:1750 –
1756.
15. Patel TN, Shishehbor MH, Bhatt DL. A review of high-dose statin
Coronary Artery Disease/Statin Therapy and Coronary Intervention
therapy: targeting cholesterol and inflammation in atherosclerosis. Eur
Heart J 2007;28:664 – 672.
16. Shishehbor MH, Patel T, Bhatt DL. Using statins to treat inflammation
in acute coronary syndromes: are we there yet? Cleve Clin J Med
2006;73:760 –766.
17. Shishehbor MH, Bhatt DL. Inflammation and atherosclerosis. Curr
Atheroscler Rep 2004;6:131–139.
18. Mason RP, Walter MF, Jacob RF. Effects of HMG-CoA reductase
923
inhibitors on endothelial function: role of microdomains and oxidative
stress. Circulation 2004;109:II34 –II41.
19. Singh M, Lennon RJ, Roger VL, Rihal CS, Halligan S, Lerman A, Yang
E, Holmes DR Jr. Relation of preprocedural statin therapy to in-hospital
procedural complications following percutaneous coronary interventions
in patients with hyperlipidemia. Am J Cardiol 2006;98:325–330.
20. Bhatt DL, Topol EJ. Need to test the arterial inflammation hypothesis.
Circulation 2002;106:136 –140.
Racial Disparity in the Utilization of Implantable-Cardioverter
Defibrillators Among Patients With Prior Myocardial Infarction
and an Ejection Fraction of <35%
Kevin L. Thomas, MDa,*, Sana M. Al-Khatib, MDa, Richard C. Kelsey II, BSb, Heather Bush, MSc,
Lynne Brosius, MSc, Eric J. Velazquez, MDa, Eric D. Peterson, MDa, and F. Roosevelt Gilliam, MDa
Age-adjusted sudden cardiac death rates are highest for black patients compared with
other racial groups. The prophylactic implantation of an implantable cardioverter-defibrillator (ICD) provides a significant reduction in sudden cardiac death and overall mortality
in patients after myocardial infarctions with significant left ventricular systolic dysfunction. The purpose of this study was to determine whether black patients with left ventricular systolic dysfunction were less likely than white patients to receive ICDs for the
primary prevention of sudden cardiac death. Data from the National Registry to Advance
Heart Health (ADVANCENT) were analyzed to determine which patients with histories of
myocardial infarctions and ejection fractions <35% received ICDs for the primary prevention of sudden cardiac death. Multivariate logistic regression was used to evaluate the
association of patients’ race with ICD implantation. Overall, 7,830 patients were identified
as eligible candidates for ICDs. Black patients (n ⴝ 660) were younger, more often women,
had less education, had more co-morbidities, and had a lower mean ejection fraction
compared with white patients (n ⴝ 7,170). More than 90% of the study population were
insured, and approximately 88% of participants in the registry were enrolled by cardiologists. Blacks were significantly less likely than whites to receive ICDs (30% vs 41%, p
<0.001). This difference in ICD use persisted after adjusting for demographics, clinical
characteristics, and socioeconomic factors (odds ratio 0.62, 95% confidence interval 0.50 to
0.75, p <0.001). In conclusion, among patients at an increased risk for sudden cardiac
death, blacks were significantly less likely to receive ICDs than whites. © 2007 Elsevier
Inc. All rights reserved. (Am J Cardiol 2007;100:924 –929)
Our objective was to determine whether a racial disparity
exists in the use of implantable cardioverter-defibrillators
(ICDs) for the primary prevention of sudden cardiac death.
Using a national heart failure registry, broadly inclusive
across adult ages, race, and payer mix, we studied this
question in patients with ejection fractions (EFs) ⱕ35%
after myocardial infarctions (MIs).
Methods
We identified our study population using the National Registry
to Advance Heart Health (ADVANCENT), a multicenter observational registry of 26,302 patients.1 This registry was
designed to collect and report clinical characteristics, diagnostic test results, therapies, and interventions in a wide-ranging
sample of patients with left ventricular systolic dysfunction and
EFs ⱕ40% and was funded by Guidant Corporation (Minneapolis, Minnesota).
a
Duke University Medical Center, Durham, North Carolina; bGuidant
Corporation, Minneapolis, Minnesota; and cREGISTRAT Inc., Lexington,
Kentucky. Manuscript received February 21, 2007; revised manuscript
received and accepted April 13, 2007.
This study was funded by an educational grant from Guidant Corporation, Minneapolis, Minnesota.
*Corresponding author: Tel: 919-225-4341; fax: 919-681-6448.
E-mail address: [email protected] (K.L. Thomas).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.024
Sites were selected to represent academic and nonacademic medical centers in the United States. A total of 106
centers enrolled patients from June 30, 2003, until study
enrollment was closed on December 22, 2004. Sites were
not selected on specific requirements concerning heart failure admissions or the use of specific ICD devices. Sites
were reimbursed a nominal fee for the enrollment of patients and the completion of case report forms.
Clinicians who actively managed left ventricular systolic
dysfunction patients in various practice settings nationwide
served as investigators. Patients were enrolled in ADVANCENT by internists (0.5%), family practitioners (0.1%),
general cardiologists (29%), heart failure subspecialists
(18%), interventional cardiologists (27%), and electrophysiologists (13%). The enrolling physicians were unspecified
for 12.4% of patients.
Health care providers were encouraged to enroll consecutive patients during routine follow-up visits. Eligibility
was not contingent on the use of any particular therapeutic
agent or device. ADVANCENT neither mandated nor restricted treatments or devices provided to patients. The goal
of the registry was to enroll a representative patient sample
receiving care for their heart failure. The study population
consisted of patients aged ⱖ18 years with measured EFs
ⱕ40% by echocardiography, contrast ventriculography, nuclear methods (multigated angiography, gated single-photon
emission computed topography), or magnetic resonance imwww.AJConline.org
Coronary Artery Disease/Racial Disparity in ICD Utilization
aging. Patients within 30 days of MIs and with previous
episodes of ventricular tachycardia or ventricular fibrillation
were excluded to replicate the sudden cardiac death primary
prevention population studied in the Multicenter Automatic
Defibrillator Implantation Trial (MADIT) I and II.2,3 Patients with nonischemic cardiomyopathy were not included
in this analysis, because at the time the registry was closed
for enrollment (December 2004), these patients would not
have been eligible for ICD implantation according to the
criteria established by the Centers for Medicare and Medicaid Services. Race and ethnicity were separate categorical
variables on the data collection form. The racial categories
were American Indian or Alaskan Native, Asian, Native
Hawaiian or other Pacific Islander, and black or AfricanAmerican and were assessed by study coordinators by direct
questions to patients, observation, or review of medical
records.
ADVANCENT was a standardized, paperless electronic
registry. Designated center personnel (e.g., investigators,
study coordinators, nurses) completed and submitted the
enrollment form and all subsequent information for each
patient via the ADVANCENT electronic patient management interface. At enrollment, information on co-morbidities, clinical characteristics, and demographics were compiled from patient interviews, medical records, and data
collection forms. Patient information was entered and remained on the centers’ resident computer systems until the
investigators or their designees verified patient eligibility
and the accuracy of the data. Once verified, completed data
fields were transmitted via a secure Internet connection
from the registry center to REGISTRAT Inc. (Lexington,
Kentucky), an independent company responsible for clinical
data management, statistical analyses, and data reporting.
Centers provided follow-up information at approximately 6, 12, and 18 months for enrolled patients who
remained active in the registry.
The primary outcome of interest, the receipt of an ICD
before enrollment or during the registry follow-up, was
determined through contact with patients during follow-up
visits and review of the medical record.
We determined for the total population, and by race, the
number of patients eligible for ICD implantation relative to
those who received ICDs either at enrollment or during
follow-up in the registry. Additionally, certain categorical
and continuous variables were further categorized for ease
in describing trends. Education was dichotomized according
to those with less than high school education and those with
high school education or greater. Payer status was also
dichotomized as private or public. Finally, to account for
site differences, a variable was created to categorize patients
according to the size of the enrolling site. Each site was
categorized according to the number of patients enrolled in
the ADVANCENT registry. Sites were divided into 3
groups: those with ⬍100 patients, 100 to 500 patients,
⬎500 patients.
Characteristics of black and white patients were analyzed
using chi-square tests for categorical variables and Student’s t tests for continuous variables to compare potential
confounding variables and race. To adjust for possible confounding, a logistic regression was performed in which the
probability of having an ICD was modeled. Several ap-
925
proaches were considered to determine which potential confounders to include in the model. For each model considered, there were no real differences in model fit using R2,
measures of association, and the Hosmer-Lemeshow goodness-of-fit statistic. Additionally, the inclusion or exclusion
of most potential confounders did not dramatically change
the estimate of the odds ratio for race. Two main approaches
were used in determining the potential confounders to include in the model. To start, all potential confounders were
included, and forward selection was used to determine a
minimum set of variables to include in the model. These
variables were age, gender, EF, syncope, New York Heart
Association class, spironolactone use, digitalis use, QRS
width, geographical region of enrollment, antiarrhythmic
use, and the size of the enrolling site. From this model,
additional potential confounders were added on the basis of
a significance level of ␣ ⫽ 0.15 when considering bivariate
tests with race. The additional variables included were indicators for high school education, diabetes mellitus, previous coronary artery bypass grafting, previous percutaneous
intervention, enrollment by an electrophysiologist, atrial
fibrillation, hypertension, angiotensin receptor blocker use,
diuretic use, and renal disease. Interactions among race,
gender, and enrollment site size were also included. This
model demonstrated slightly higher measures of fit (c statistic ⫽ 0.71, R2 ⫽ 0.18, measures of association and
Hosmer-Lemeshow goodness-of-fit statistic p ⫽ 0.92). All
significance tests were 2 sided, and we assumed that a p
value ⬍0.05 indicated statistical significance. All analyses
were performed using SAS version 9.1 (SAS Institute Inc.,
Cary, North Carolina).
No patients had missing values for race. With the exception of education, the percentage of patients with missing
data was ⬍1%. However, ⬎2,000 patients had unspecified
values for education. Education was not significant in any of
the models considered, and there was little difference in the
fit or the adjusted odds ratio for race when education was
excluded or included. Additionally, missing educational status was independent of race and ICD receipt.
Results
We identified 10,943 patients with previous MIs and EFs
ⱕ35%. However, after excluding 3,113 patients, 7,830 (660
blacks and 7,170 whites) were included in the final study
population (Figure 1). There was no statistically significant
difference between the proportions of whites and blacks
excluded. Compared with whites, blacks were younger and
more often women. Black patients were as likely to have
earned high school diplomas or associate’s degrees as white
patients but less likely to have obtained bachelor’s or master’s degrees (Table 1). Approximately 93% of patients in
the ADVANCENT registry had Medicare or Medicaid or
private insurance, and there were no racial differences in
payer status among patients who were insured. Clinically,
blacks had a lower mean EF and higher prevalences of
hypertension, diabetes mellitus, and renal insufficiency.
White patients had more atrial fibrillation and were more
likely to be taking antiarrhythmic medications and to be
enrolled in the registry by electrophysiologists than black
patients. There were no racial differences in the use of
926
The American Journal of Cardiology (www.AJConline.org)
tomatic heart failure relative to whites. Additionally, there
were significant racial differences in the type of revascularization, with blacks having undergone more percutaneous
coronary intervention and whites more coronary artery bypass grafting (Table 1).
The mean follow-up period for patients in this analysis
was 234.9 ⫾ 129 days. Follow-up was complete for 65% of
the population and was similar between races (p ⫽ 0.11).
Defibrillators were implanted in 3,124 of the 7,830 patients
(40%) with previous MIs and EFs ⱕ35%. Among all patients who underwent ICD placement, 200 black patients
(30%) and 2,924 white patients (41%) had ICDs implanted
(p ⬍0.001). The median time from MI to ICD implantation
was 1,179 days (25th to 75th interquartile range 215 to
3,102) among black patients, compared with 2,732 days
(25th to 75th interquartile range 488 to 5,295) for white
patients. Among patients who had ICDs implanted, 94%
received their ICDs before enrollment in the ADVANCENT
registry. There was no significant racial difference in patients who received ICDs during the registry follow-up
period (p ⫽ 0.37).
After adjusting for differences in clinical characteristics
and socioeconomic factors, black patients were significantly
less likely to receive ICDs than white patients (odds ratio
[OR] 0.62, 95% confidence interval [CI] 0.50 to 0.75, p
⬍0.001). Men were more likely to receive ICDs than
women (OR 1.70, 95% CI 1.49 to 1.95, p ⬍0.001). In the
multivariate model, no significant interaction was observed
between race and gender (p ⫽ 0.41) or race and center size
(p ⫽ 0.76). Additional variables associated with greater
odds of ICD receipt included younger age; history of syncope; previous coronary artery bypass graft; lower EF;
wider QRS width; the use of antiarrhythmic therapy, spironolactone, or digoxin; enrollment in the registry by an
electrophysiologist; and larger center size (Table 2). Payer
status and education were not independent predictors of
ICD implantation.
Discussion
Figure 1. Exclusion criteria applied to patients enrolled in the ADVANCENT registry. The arrows denote the level in which various factors led to
the exclusion of patients.
post-MI evidence-based medical therapies, including aspirin, ␤ blockers, and angiotensin-converting enzyme inhibitors. Blacks were treated more often with diuretics and
spironolactone, consistent with the finding of more symp-
Among a broad multipayer population of socioeconomically
diverse patients with previous MIs and low EFs, blacks
received ICDs for the primary prevention of sudden cardiac
death at a significantly lower rate relative to whites. This
disparity does not appear to be explained by potential confounders, including co-morbidities, access to cardiologists,
or payer status. Given that ICD implantation can be lifesaving in this patient population, the disparities in care found in
this study are concerning.
This analysis is among the first to report rates of ICD use
by race among patients eligible for ICDs for the primary
prevention of sudden cardiac death. Previous reports of ICD
use that revealed a racial disparity were limited to Medicare
beneficiaries who were older and were survivors of cardiac
arrest, a distinctly different population. Moreover, our results are drawn from a registry, which, unlike large administrative data sets, generally provide more detailed information on patient characteristics, and treatments.4,5
Racial disparities in the use of cardiac procedures have
been described for ⬎2 decades. An Institute of Medicine
report6 and a Kaiser Family Foundation and American Col-
Coronary Artery Disease/Racial Disparity in ICD Utilization
927
Table 1
Characteristics of patients by race
Variable
Blacks
(n ⫽ 660)
Whites
(n ⫽ 7,170)
All Patients
(n ⫽ 7,830)
p Value
Age (yrs), mean ⫾ SE
Women
Education
Master’s degree or greater
Associate’s or bachelor’s degree
High school diploma
Less than high school
Unspecified
Payer status
Private
Public
Unspecified
Geographic location
East
West
Diabetes mellitus
Hypertension
Renal disease
Atrial fibrillation
Syncope
Anemia
Cancer (active)
Cancer (remission)
EF, mean ⫾ SE
New York Heart Association class
I
II
III
IV
QRS width (ms)
⬍120
120–150
⬎150
Percutaneous coronary intervention
Coronary artery bypass grafting
Medications
␤ blockers
ACE inhibitors
Angiotensin receptor blockers
Aspirin
Spironolactone
Diuretics
Digoxin
Antiarrhythmics (not amiodarone)
Registry enrollment by an electrophysiologist
63 ⫾ 0.5
151 (32%)
69 ⫾ 0.11
1,022 (17%)
68 ⫾ 0.11
1,173 (18%)
⬍0.001
⬍0.001
⬍0.001
1%
13%
36%
19%
32%
5%
17%
35%
13%
30%
4%
16%
35%
14%
31%
33%
66%
1%
32%
65%
3%
32%
65%
3%
54%
46%
44%
87%
12%
18%
5%
10%
2%
6%
25 ⫾ 0.3%
61%
39%
34%
73%
6%
28%
8%
9%
2%
9%
27 ⫾ 0.1%
60%
40%
35%
74%
7%
27%
8%
9%
2%
9%
26 ⫾ 0.1%
15%
46%
37%
3%
17%
52%
29%
3%
16%
52%
29%
3%
60%
19%
21%
70%
48%
53%
23%
24%
60%
69%
52%
24%
24%
60%
67%
⬍0.001
⬍0.001
83%
64%
23%
73%
29%
76%
39%
14%
12%
83%
65%
19%
75%
21%
66%
37%
19%
15%
83%
65%
19%
75%
21%
67%
37%
19%
15%
NS
NS
0.028
NS
⬍0.001
⬍0.001
NS
0.004
0.012
NS
0.008
⬍0.001
⬍0.001
⬍0.001
⬍0.001
0.014
NS
NS
⬍0.001
0.002
0.009
p Values were calculated using either the chi-square test or Student’s t test. p values refer only to the difference between blacks and whites.
ACE ⫽ angiotensin-converting enzyme.
lege of Cardiology Foundation report that summarized the
weight of the evidence supporting racial and ethnic differences in cardiac care brought this issue to the forefront of
medical, political, and social agendas in the United States.
Despite increased awareness of the problem, the inequalities
persist, and there remains doubt among health care professionals that racial disparities in cardiovascular care exist. In
fact, as recently as 2 years ago, only 1/3 of cardiologists
who responded to a survey agreed with the statement that
“clinically similar patients receive different cardiovascular
care based on their race and ethnic background.”7
Potential explanations for the racial disparity in ICD use
observed in this study are important to address. The basis
for disparities is multifactorial and varies across health
indicators and health care settings. Disparities in the delivery of cardiovascular care may emanate from a complex
interaction of physiologic (differential burden of risk factors), cultural (patient preferences, mistrust, and patientprovider communication), and socioeconomic (provider racial or ethnic bias, poverty, and education) factors.6,8 The
Institute of Medicine report6 identified the following factors
as major contributors to disparities in cardiovascular disease
928
The American Journal of Cardiology (www.AJConline.org)
Table 2
Characteristics associated with the receipt of an implantable cardioverter-defibrillator
Characteristic
QRS width
EF (by 10% decrease)
Antiarrhythmic therapy (not amiodarone)
Male gender
Age (by 10-yr decrease)
Race: black
Center size (larger)
Previous coronary artery bypass grafting
Digitalis use
Spironolactone use
Syncope
Registry enrollment by an electrophysiologist
Chi-square
OR
95% CI
p Value
189
133
160
78
26
15
12
10
17
22
148
280
1.59
1.38
2.14
1.70
1.18
0.62
1.13
1.18
1.23
1.32
2.16
5.86
1.49–1.70
1.33–1.43
1.86–2.45
1.49–1.95
1.14–1.22
0.50–0.75
1.04–1.23
1.06–1.23
1.10–1.37
1.16–1.50
1.78–2.61
5.02–6.83
⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001
0.002
0.002
0.002
⬍0.001
⬍0.001
⬍0.001
ORs presented are also adjusted for New York Heart Association class, history of coronary artery bypass grafting, history of percutaneous coronary
intervention, diabetes mellitus, atrial fibrillation, history of hypertension, renal disease, region of enrollment, and payer status.
management and outcomes: clinical appropriateness, patient
choice, the health care delivery system, and bias. Current
research has focused on the health care delivery system as a
modifiable contributor to racial disparities in cardiovascular
disease management. Recent studies have found that black
patients are more likely to be treated by physicians with
lower qualifications and have less access to subspecialty
care.9,10 This inequality appears to contribute significantly
to worse cardiovascular outcomes for black patients.11,12
Few studies have analyzed racial differences in ICD
use.13–16 Groeneveld et al,14 using Medicare claims data,
conducted 2 studies examining racial differences in ICD
use. The first analysis examined cardiac arrest survivors
aged 66 to 74 years and found lower survival among blacks
relative to whites (hazard ratio 1.30, 95% CI 1.09 to 1.55).
Black and white patients in the study had a mortality benefit
from ICD therapy, but blacks were less likely to receive
ICDs (OR 0.58, 95% CI 0.36 to 0.94). Groeneveld et al’s15
second analysis similarly found that the rate of ICD implantation was lower for blacks relative to whites in patients
hospitalized with admitting, primary, or secondary diagnoses of ventricular fibrillation, ventricular tachycardia, or
cardiac arrest. Additionally, this study demonstrated some
reduction in the racial disparity during the period from 1990
to 2000 that was partially attributed to geographical differences in the diffusion of ICD technology. Despite the decrease in the racial disparity in ICD use, black patients
continued to have significantly lower odds of receiving
ICDs compared with white patients (OR 0.69, 95% CI 0.61
to 0.78).
We were unable to control for physician variables (physician race or board certification status) and health system–
related variables (ICD implantation capability and hospital
quality-of-care performance) that contribute to racial disparities in cardiovascular care.11,17 Clinical trials evaluating
the efficacy of ICDs in post-MI patients with left ventricular
systolic dysfunction typically have required left ventricular
EF assessments ⬍3 months after device implantation. We
were unable to reliably report the relation between ICD
implantation and most recent EF assessment. It is unclear
the effect this may have had on our results, but there are no
data to suggest that there are racial differences in EF
changes after an MI, and therefore it is unlikely to fully
explain the racial inequality in ICD use in this study. Our
analysis likely involved referral or selection bias, because
⬎90% of participants in the ADVANCENT registry were
enrolled by cardiologists; therefore, disease management in
the study sample may not represent that of the heart failure
population cared for by noncardiologists. This fact likely
explains the high rate of ICD implantation in patients who
met primary prevention criteria for ICD therapy. In this
study, 40% of all eligible patients received ICDs. Current
estimates indicate that in the United States, only 20% to
25% of patients who are currently eligible for ICD benefits
under Medicare receive these devices.18
We were unable to verify the error rate in the ascertainment of ICD implantation status, because there was no
coordinating center in charge of data recording during follow-up. It is possible that individuals received devices during the follow-up period that were not captured and thus
could have affected our findings. An additional limitation of
this analysis is the lack of longitudinal follow-up regarding
patients’ clinical outcomes, which made it difficult to determine if differences in ICD implantation led to survival
differences. Although we did not have mortality data, given
the magnitude of risk reduction provided by ICD therapy in
clinical trials,2,3,19,20 coupled with the mortality differences
seen in previous studies examining racial differences in ICD
use, it is plausible that the racial disparity found would have
resulted in a survival advantage for white patients with
previous MIs and EFs ⱕ35%.
Acknowledgment: We would like to acknowledge Diane
King-Hageman, BS, and Beverly Raleigh, MS, for their role
in the acquisition of data for this analysis.
1. Hanna IR, Heeke B, Bush H, Brosius L, King-Hageman D, Beshai JF,
Langberg JJ. The relationship between stature and the prevalence of
atrial fibrillation in patients with left ventricular dysfunction. J Am
Coll Cardiol 2006;47:1683–1688.
2. Moss AJ, Hall WJ, Cannom DS, Daubert JP, Higgins SL, Klein H,
Levine JH, Saksena S, Waldo AL, Wilber D, et al. Improved
survival with an implanted defibrillator in patients with coronary
disease at high risk for ventricular arrhythmia. Multicenter Automatic Defibrillator Implantation Trial Investigators. N Engl J Med
1996;335:1933–1940.
Coronary Artery Disease/Racial Disparity in ICD Utilization
3. Moss AJ, Zareba W, Hall WJ, Klein H, Wilber DJ, Cannom DS,
Daubert JP, Higgins SL, Brown MW, Andrews ML. Prophylactic
implantation of a defibrillator in patients with myocardial infarction
and reduced ejection fraction. N Engl J Med 2002;346:877– 883.
4. Havranek EP, Masoudi FA, Westfall KA, Wolfe P, Ordin DL, Krumholz HM. Spectrum of heart failure in older patients: results from the
National Heart Failure project. Am Heart J 2002;143:412– 417.
5. Joshi AV, D’Souza AO, Madhavan SS. Differences in hospital lengthof-stay, charges, and mortality in congestive heart failure patients.
Congest Heart Fail 2004;10:76 – 84.
6. Smedley BD, Nelson AR, Unequal Treatment: Confronting Racial and
Ethnic Disparities in Health Care. Washington, District of Columbia:
National Academy Press, 2002.
7. Lurie N, Fremont A, Jain AK, Taylor SL, McLaughlin R, Peterson E,
Kong BW, Ferguson TB Jr. Racial and ethnic disparities in care: the
perspectives of cardiologists. Circulation 2005;111:1264 –1269.
8. Yancy CW, Benjamin EJ, Fabunmi RP, Bonow RO. Discovering the
full spectrum of cardiovascular disease: Minority Health Summit
2003: executive summary. Circulation 2005;111:1339 –1349.
9. Bach PB, Pham HH, Schrag D, Tate RC, Hargraves JL. Primary care
physicians who treat blacks and whites. N Engl J Med 2004;351:575–
584.
10. LaVeist TA, Arthur M, Morgan A, Rubinstein M, Kinder J, Kinney
LM, Plantholt S. The cardiac access longitudinal study. A study of
access to invasive cardiology among African American and white
patients. J Am Coll Cardiol 2003;41:1159 –1166.
11. Skinner J, Chandra A, Staiger D, Lee J, McClellan M. Mortality after
acute myocardial infarction in hospitals that disproportionately treat
black patients. Circulation 2005;112:2634 –2641.
929
12. Tonne C, Schwartz J, Mittleman M, Melly S, Suh H, Goldberg R.
Long-term survival after acute myocardial infarction is lower in more
deprived neighborhoods. Circulation 2005;111:3063–3070.
13. Alexander M, Baker L, Clark C, McDonald KM, Rowell R, Saynina O,
Hlatky MA. Management of ventricular arrhythmias in diverse populations in California. Am Heart J 2002;144:431– 439.
14. Groeneveld PW, Heidenreich PA, Garber AM. Racial disparity in
cardiac procedures and mortality among long-term survivors of cardiac
arrest. Circulation 2003;108:286 –291.
15. Groeneveld PW, Heidenreich PA, Garber AM. Trends in implantable
cardioverter-defibrillator racial disparity: the importance of geography.
J Am Coll Cardiol 2005;45:72–78.
16. Gauri AJ, Davis A, Hong T, Burke MC, Knight BP. Disparities in the
use of primary prevention and defibrillator therapy among blacks and
women. Am J Med 2006;119:167.e17–167.e21.
17. LaVeist TA, Arthur M, Morgan A, Plantholt S, Rubinstein M. Explaining racial differences in receipt of coronary angiography: the role
of physician referral and physician specialty. Med Care Res Rev
2003;60:453– 467.
18. McClellan MB, Tunis SR. Medicare coverage of ICDs. N Engl J Med
2005;352:222–224.
19. Bardy GH, Lee KL, Mark DB, Poole JE, Packer DL, Boineau R,
Domanski M, Troutman C, Anderson J, Johnson G, et al. Amiodarone
or an implantable cardioverter-defibrillator for congestive heart failure.
N Engl J Med 2005;352:225–237.
20. Buxton AE, Lee KL, Fisher JD, Josephson ME, Prystowsky EN,
Hafley G. A randomized study of the prevention of sudden death in
patients with coronary artery disease. Multicenter Unsustained Tachycardia Trial Investigators. N Engl J Med 1999;341:1882–1890.
Comparison of Myocardial Infarct Size Assessed With ContrastEnhanced Magnetic Resonance Imaging and Left Ventricular
Function and Volumes to Predict Mortality in Patients With
Healed Myocardial Infarction
Stijntje D. Roes, MDa,*, Sebastian Kelle, MDb, Theodorus A.M. Kaandorp, MDa,
Thomas Kokocinski, MDb, Don Poldermans, MDc, Hildo J. Lamb, MDa, Eric Boersma, PhDd,
Ernst E. van der Wall, MDe, Eckart Fleck, MDb, Albert de Roos, MDa, Eike Nagel, MDb,
and Jeroen J. Bax, MDe
Currently, left ventricular (LV) ejection fraction (EF) and/or LV volumes are the established predictors of mortality in patients with coronary artery disease (CAD) and severe LV
dysfunction. With contrast-enhanced magnetic resonance imaging (MRI), precise delineation of infarct size is now possible. The relative merits of LVEF/LV volumes and infarct size
to predict long-term outcome are unknown. The purpose of this study was to determine the
predictive value of infarct size assessed with contrast-enhanced MRI relative to LVEF and
LV volumes for long-term survival in patients with healed myocardial infarction. Cine MRI
and contrast-enhanced MRI were performed in 231 patients with healed myocardial
infarction. LVEF and LV volumes were measured and infarct size was derived from
contrast-enhanced MRI. Nineteen patients (8.2%) died during a median follow-up of 1.7
years (interquartile range 1.1 to 2.9). Cox proportional hazards analysis revealed that
infarct size defined as spatial extent (hazard ratio [HR] 1.3, 95% confidence interval [CI]
1.1 to 1.6, chi-square 6.7, p ⴝ 0.010), transmurality (HR 1.5, 95% CI 1.1 to 1.9, chi-square
8.9, p ⴝ 0.003), or total scar score (HR 6.2, 95% CI 1.7 to 23, chi-square 7.4, p ⴝ 0.006) were
stronger predictors of all-cause mortality than LVEF and LV volumes. In conclusion,
infarct size on contrast-enhanced MRI may be superior to LVEF and LV volumes for
predicting long-term mortality in patients with healed myocardial infarction. © 2007
Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:930 –936)
The main cause of death in patients with coronary artery
disease (CAD) and severely depressed left ventricular (LV)
function is end-stage heart failure, whereas sudden cardiac
death is more common in patients with CAD and preserved
or moderately depressed LV function.1–5 Risk stratification
of patients with CAD is necessary for optimization of treatment. Previous studies showed that LV function and LV
end-systolic volume (ESV) were the strongest predictors of
cardiac death.6,7 However, other variables to optimize risk
stratification are needed to identify patients at high-risk for
mortality among patients with preserved and moderately
depressed LV function. Preliminary findings in patients
with acute myocardial infarction and moderate LV dysfunction showed that infarct size assessed with contrast-enhanced magnetic resonance imaging (MRI) was a better
predictor of adverse clinical outcome than LV function.8
Departments of aRadiology and eCardiology, Leiden University Medical Center, Leiden; and cDepartment of Cardiology, Thorax Center, Rotterdam and dDepartment of Cardiology, Erasmus University, Rotterdam,
The Netherlands; and bDepartment of Cardiology/Internal Medicine, German Heart Institute, Berlin, Germany. Manuscript received March 19,
2007; revised manuscript received and accepted April 6, 2007.
*Corresponding author: Tel: 31-71-5262993; fax: 31-71-5248256.
E-mail address: [email protected] (S.D. Roes).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.029
However, the prognostic value of infarct size determined
with contrast-enhanced MRI in patients with healed myocardial infarction is unknown. Accordingly, this study examines the predictive value of infarct size assessed with
contrast-enhanced MRI relative to LV function and volumes
for long-term survival of patients with healed myocardial
infarction.
Methods
Study population: This was a prospective, follow-up
study that involved 2 hospitals. Consecutive patients (n ⫽
231), referred for MRI to evaluate cardiac function and
extent of scar tissue for clinical reasons, with a history of
CAD and evidence of scar tissue on contrast-enhanced MRI,
were enrolled. Patients with myocardial infarction ⬍3
months before cardiac MRI were excluded. Other exclusion
criteria were (supra-) ventricular arrhythmias, pacemakers,
intracranial clips, and claustrophobia. Patient characteristics
are listed in Table 1. The study was approved by the local
ethics committees of both institutions and informed consent
was obtained.
Magnetic resonance imaging: DATA ACQUISITION: A
1.5-T Gyroscan ACS-NT/Intera MRI scanner (Philips Medical Systems, Best, The Netherlands) equipped with powertrack 6000 gradients and 5-element cardiac synergy coil
www.AJConline.org
Coronary Artery Disease/Infarct Size Best Predictor of Mortality
931
Table 1
Baseline clinical variables
Variable
Total Population
(n ⫽ 231)
Survivors
(n ⫽ 212)
Nonsurvivors
(n ⫽ 19)
Age (yrs)
Men
Diabetes mellitus
Hypertension
Hypercholesterolemia*
Smoker
Previous myocardial infarction
Q wave
Infarct location
Anterior
Inferior
No. of coronary arteries narrowed on angiogram
1
2
3
Medications
␤ blocker
Calcium channel blocker
ACE inhibitor
Oral anticoagulant
Statin
Nitrate
Diuretic
64 (58, 69)
201 (87%)
41/222 (18%)
125/221 (57%)
175/225 (78%)
103/219 (47%)
193 (84%)
109 (47%)
64 (57, 69)
184 (87%)
34/204 (17%)
120/203 (59%)
163/206 (79%)
95/202 (47%)
176 (83%)
98 (46%)
67 (61, 75)
17 (89%)
7/18 (39%)
5/18 (28%)
12/19 (63%)
8/17 (47%)
17 (89%)
11 (58%)
0.1
1.0
0.029
0.013
0.1
1.0
0.7
0.3
34/109 (31%)
75/109 (69%)
31/98 (32%)
67/98 (68%)
3/11 (27%)
8/11 (73%)
1.0
24 (10%)
67 (29%)
140 (61%)
23 (11%)
62 (29%)
127 (60%)
1 (5%)
5 (26%)
13 (68%)
0.7
176/225 (78%)
47/224 (21%)
177/225 (79%)
222/226 (98%)
191/225 (85%)
69/224 (31%)
106/224 (47%)
164/206 (80%)
42/205 (20%)
163/206 (79%)
203/207 (98%)
180/206 (87%)
61/205 (30%)
90/205 (44%)
12/19 (63%)
5/19 (26%)
14/19 (74%)
19/19 (100%)
11/19 (58%)
8/19 (42%)
16/19 (84%)
p Value
0.1
0.6
0.6
1.0
0.003
0.3
0.001
Continuous data are expressed as median (interquartile range); categorical data are expressed as number/total patients with complete data (percent).
* Total cholesterol ⬎200 mg/dl.
ACE ⫽ angiotensin-converting enzyme.
was used. Patients were positioned in the supine position.
Images were acquired during breath-holds of approximately
15 seconds using vector electrocardiographic gating.
The heart was imaged from apex to base,9 with 10 to 12
imaging levels (dependent on the heart size) in the shortaxis view using a balanced, fast-field echo sequence with
parallel imaging (SENSE, acceleration factor 2). Typical
parameters were a field of view of 400 ⫻ 400 mm2, matrix
of 256 ⫻ 256 pixels, slice thickness of 10.00 or 8.00 mm, no
slice gap, flip angle of 50°, time to echo of 1.82 ms, and
time to repeat of 3.65 ms. Temporal resolution was 25 to 39
ms. Geometric settings of baseline scans were stored and
repeated for contrast-enhanced images to ensure matching
of the same slices (and hence, myocardial segments).
Contrast-enhanced images were acquired approximately
15 minutes after bolus injection of gadolinium diethylenetriamine penta-acetic acid (Magnevist, Schering/Berlin,
Germany; 0.15 mmol/kg or 0.20 mmol/kg) with an inversion-recovery 3-dimensional spoiled gradient echo sequence; inversion time was determined with real-time plan
scan. Typical parameters were a field of view of 400 ⫻ 400
mm2, matrix of 256 ⫻ 256 pixels, slice thickness of 5.00
mm, overlapping slices (50%), flip angle of 15°, time to
echo of 1.36 ms, and time to repeat of 4.53 ms.
DATA ANALYSIS: To determine global function, endocardial borders were outlined manually on short-axis cine images with previously validated software (MASS, Medis,
The Netherlands/ViewForum, Philips, The Netherlands).10
Papillary muscles were regarded as part of the ventricular
cavity, and epicardial fat was excluded. LVESV and LV
end-diastolic volume (EDV) were calculated. Subsequently,
ESV was subtracted from EDV and LV ejection fraction
(EF) was calculated.
End-diastolic wall thickness (EDWT) was measured
quantitatively at the center of the infarct region.
Contrast-enhanced images were scored visually by 2
experienced observers (blinded to other MRI and clinical
data) using the 17-segment model as recently proposed.11
Each segment was graded on a 5-point scale (segmental scar
score), with 0, absence of hyperenhancement; 1, hyperenhancement of 1% to 25% of LV wall thickness; 2, hyperenhancement extending to 26% to 50%; 3, hyperenhancement extending to 51% to 75%; and 4, hyperenhancement
extending to 76% to 100%.12
To quantify and define the extent/transmurality of scar
tissue, the following definitions were used13: (1) spatial
(circumferential) extent, the number of affected segments;
(2) transmurality, the number of segments with a segmental
scar score of 3 or 4; and (3) total scar score, summed
segmental scar scores per patient divided by 17 (which
reflects the damage per patient).
Follow-up: The long-term follow-up was performed by
chart review and telephone contact. No patients were lost to
follow-up. The primary end point was all-cause mortality,
which was defined as death caused by end-stage heart failure or acute myocardial infarction, sudden cardiac death,
and noncardiac death. Myocardial infarction was defined by
clinical presentation, elevated cardiac enzyme levels, and/or
typical changes on electrocardiography. Sudden cardiac
death was defined as unexpected natural death from a car-
932
The American Journal of Cardiology (www.AJConline.org)
Table 2
Baseline magnetic resonance imaging variables
MRI Variables
LVEF (%)
LVESV (ml)
LVEDV (ml)
End-diastolic wall thickness (mm)
Spatial extent
Transmurality
Total scar score
Total Population
(n ⫽ 231)
Survivors
(n ⫽ 212)
Nonsurvivors
(n ⫽ 19)
p
Value
43 (30, 55)
109 (69, 189)
197 (158, 258)
4.3 (3.1, 5.5)
6 (4, 8)
3 (1, 5)
0.8 (0.5, 1.3)
43 (32, 56)
103 (69, 179)
191 (154, 250)
4.3 (3.1, 5.5)
6 (4, 8)
2 (1, 4)
0.8 (0.5, 1.2)
27 (17, 41)
212 (122, 343)
283 (223, 414)
3.9 (3.2, 5.0)
8 (6, 13)
5 (3, 8)
1.3 (0.8, 1.8)
⬍0.001
⬍0.001
⬍0.001
0.4
⬍0.001
0.008
0.004
Data are expressed as median (interquartile range).
diac cause within 1 hour from the onset of symptoms, in a
person without any prior condition that would appear
fatal.14
Statistical analysis: Most continuous variables had nonnormal distribution (as evaluated by Kolmogorov-Smirnov
tests). For reasons of uniformity, summary statistics for all
continuous variables are therefore presented as medians
together with the 25th and 75th percentiles. Categorical data
are summarized as frequencies and percentages. Differences
in baseline characteristics between patients who reached the
primary end point and those who did not were analyzed
using Wilcoxon-Mann-Whitney tests or Fisher’s exact tests,
as appropriate.
We aimed to study to what extent MRI results were
associated with all-cause mortality. For this purpose, 3 Cox
proportional hazards regression models were constructed,
with spatial extent (first model), transmurality (second
model), and the total scar score (third model) as main
exposure, and age, LV function, LV dimensions, diabetes
mellitus, hypertension, diuretic usage, and statin usage as
confounding factors. The latter variables appeared to be
associated with all-cause mortality at the p ⬍0.10 level in
univariable analysis (and we had to limit the number of
covariables because of the relatively small number of end
point events). Unadjusted and adjusted hazard ratios with
their corresponding 95% confidence intervals are reported.
All the continuous variables were assessed for linearity
by entering a transformed variable in addition to the variable of interest. The natural logarithm and square transformations were used. A significant change in the ⫺2 loglikelihood was considered as a sign of nonlinearity,
otherwise the linearity assumption was accepted. All variables met the linearity assumption. To check the proportional hazard assumption (i.e., that the hazard ratio for 2
subjects with fixed predictors is constant over time)
log(⫺log[survival probability]) for different categories was
plotted against time to ensure that the curves were reasonably parallel. In general, all proportionality assumptions
were appropriate.
After adjustment for multiple confounders (discussed
previously) spatial extent as determined by MRI appeared
significantly related with the all-cause mortality. Therefore,
in a post-hoc analysis, the study population was divided into
2 groups, based on the observed median value of the spatial
extent, and the survival of both cohorts was further analyzed
by the method of Kaplan-Meier. Difference in survival over
time was evaluated by a log-rank test.
For all tests, p ⬍0.05 was considered statistically significant. All tests were 2-sided.
Results
Study population: Clinical data are presented in Table 1;
231 patients with scar tissue on contrast-enhanced MRI were
included (106 at the Leiden University Medical Center, The
Netherlands, and 125 at the German Heart Institute, Germany).
All patients had evidence of CAD on angiography and 84%
had a previous myocardial infarction; 16% had a clinically
unrecognized myocardial infarction. MRI was performed ⬎2
years after infarction in 52% of the patients.
MRI variables: MRI findings are listed in Table 2. Median LVEF in the total study population was 43% (25th to
75th percentile 30 to 55). LVEF was significantly higher in
survivors than in nonsurvivors. Median LVESV and
LVEDV were significantly lower in survivors than in nonsurvivors. No difference in EDWT values between survivors and nonsurvivors was detected.
By definition, all patients had evidence of scar tissue on
contrast-enhanced MRI. The median spatial extent, transmurality, and total scar score were significantly higher in
nonsurvivors than in survivors.
Clinical outcome of patients during follow-up: The
median duration of follow-up was 1.7 years (range 1.1 to
2.9); 19 patients (8.2%) died during follow-up. Fourteen
patients (6.1%) died of end-stage heart failure, 2 patients
(0.9%) died of sudden cardiac death, 1 patient (0.4%) died
after acute myocardial infarction, and noncardiac death was
reported in 2 patients (0.9%). Seventy-one patients (31%)
underwent revascularization after MRI. None of these revascularized patients died during follow-up.
Predictors of mortality: As demonstrated in Table 3,
LVEF, LVESV, LVEDV, spatial extent, transmurality of scar
tissue, total scar score, diabetes mellitus, hypertension, diuretic
usage, and statin usage were significantly associated with allcause mortality. Diuretic usage, but not statin usage, was a true
confounder for the relation between infarct size and all-cause
mortality (Table 4). After adjustment for multiple (true or
potential) confounders (see Methods section), infarct size defined as spatial extent, transmurality, and total scar score remained important outcome determinants (Table 5). Infarct size
(as indicated by different MRI measurements) appears to be a
stronger predictor (based on the observed chi-square value) of
all-cause mortality than LVEF and LV volumes in all 3 models. Even when the LVEF and LV volumes were entered
Coronary Artery Disease/Infarct Size Best Predictor of Mortality
933
Table 3
Univariable analysis for prediction of all-cause mortality
MRI Variables
LVEF
LVESV
LVEDV
Spatial extent
Transmurality
Total scar score
EDWT
Clinical variables
Age /10 ys
Men
Diabetes mellitus
Hypertension
Hypercholesterolemia*
Smoker
Previous myocardial infarction
Q wave
No. of coronary arteries narrowed on angiogram†
2
3
Medications
␤ blocker
Calcium channel blocker
ACE inhibitor
Oral anticoagulant‡
Statin
Nitrate
Diuretic
Hazard Ratio
Confidence Interval
Chi-Square
p Value
0.95/%
1.07/10 ml
1.08/10 ml
1.3/unit
1.3/unit
4.0/unit
0.82/mm
0.92–0.98
1.04–1.1
1.04–1.1
1.2–1.6
1.1–1.5
1.7–9.4
0.61–1.1
9.4
20
19
16
9.1
10
1.5
0.002
⬍0.001
⬍0.001
⬍0.001
0.003
0.002
0.2
1.55
0.99
3.1
0.32
0.48
1.0
1.5
1.3
0.93–2.6
0.23–4.3
1.2–8.0
0.11–0.89
0.19–1.2
0.40–2.7
0.35–6.7
0.54–3.4
2.8
0.0
5.4
4.8
2.3
0.0
0.3
0.4
0.09
1.0
0.020
0.029
0.1
0.9
0.6
0.5
1.8
2.0
(0.21–15)
(0.27–16)
0.3
0.5
0.6
0.5
0.49
1.5
0.75
0.19–1.2
0.54–4.2
0.27–2.1
2.3
0.6
0.3
0.1
0.4
0.6
0.27
1.5
7.2
0.11–0.68
0.62–3.8
2.1–25
7.8
0.9
9.7
0.005
0.4
0.002
* Total cholesterol ⬎200 mg/dl.
†
Increased risk of mortality as compared with 1-vessel disease.
‡
No events were reported in the patients who did not use oral anticoagulants.
Abbreviation as in Table 1.
Table 4
Relation between diuretic usage, statin usage, and infarct size on contrast-enhanced magnetic resonance imaging
Spatial extent
Transmurality
Total scar score
Spatial extent
Transmurality
Total scar score
Patients Using Diuretics
Patients Not Using
Diuretics
p Value
7 (5, 9)
4 (2, 5)
1.1 (0.7, 1.4)
5 (4, 7)
2 (0, 4)
0.7 (0.4, 1.1)
⬍0.001
⬍0.001
⬍0.001
Patients Using Statins
Patients not Using Statins
p Value
6 (4, 8)
3 (1, 5)
0.8 (0.5, 1.3)
7 (5, 9)
2 (1, 4)
0.8 (0.6, 1.1)
0.2
0.8
0.8
Data are expressed as median (interquartile range).
separately in the models, the spatial extent of scar tissue on
MRI remained the strongest predictor.
The spatial extent of scar tissue on contrast-enhanced
MRI was used to separate patients at high-risk (spatial
extent larger than or equal to the median 6, n ⫽ 116) from
those at low-risk (spatial extent ⬍6, n ⫽ 115) for mortality.
Three-year mortality in high-risk patients was 20.0% versus
2.4% in their low-risk counterparts (p ⫽ 0.005; Figure 1).
Discussion
The main finding in this study is that myocardial infarct size
on contrast-enhanced MRI, expressed as either spatial ex-
tent, transmurality of scar tissue, or total scar score is a
stronger predictor of long-term mortality than LV function
and/or LV volumes in patients with healed myocardial infarction.
MRI has emerged as a reliable noninvasive technique for
assessment of scar tissue in patients with CAD.12,15 Kim et al16
validated the value of contrast-enhanced MRI to detect scar
tissue in an animal model. In addition, previous studies demonstrated good correlation between infarct size on contrastenhanced MRI and peak release of creatinine kinase-MB.12,17
Assessment of infarct size using contrast-enhanced MRI
can also predict functional recovery after acute myocardial
934
The American Journal of Cardiology (www.AJConline.org)
Table 5
Multivariable Cox proportional hazards model for prediction of all-cause mortality
Hazard Ratio
95% Confidence Interval
Chi-square
p Value
A. Model 1, using the spatial extent as variable of
scar tissue on contrast-enhanced MRI
Spatial Extent
LVEF
LVESV
LVEDV
Age
Diuretic usage
Statin usage
Diabetes mellitus
Hypertension
1.3
1.04/%
0.90/10 ml
1.2/10 ml
1.5/10 yrs
5.0
0.45
1.6
0.40
1.1–1.6
0.96–1.1
0.67–1.2
0.89–1.5
0.77–2.9
1.3–19
0.16–1.3
0.55–4.7
0.13–1.2
6.7
0.8
0.4
1.2
1.4
5.7
2.3
0.9
2.8
0.010
0.4
0.5
0.3
0.2
0.017
0.1
0.4
0.1
B. Model 2, using the transmurality as variable of
scar tissue on contrast-enhanced MRI
Transmurality
LVEF
LVESV
LEVDV
Age
Diuretic usage
Statin usage
Diabetes mellitus
Hypertension
1.5
1.06/%
0.85/10 ml
1.3/10 ml
1.5/10 yrs
4.6
0.24
1.04
0.26
1.1–1.9
0.97–1.2
0.62–1.2
0.95–1.7
0.81–2.8
1.2–18
0.08–0.79
0.33–3.2
0.09–0.79
8.9
1.5
1.0
2.5
1.7
5.0
5.5
0.0
5.7
0.003
0.2
0.3
0.1
0.2
0.025
0.019
1.0
0.017
C. Model 3, using the total scar score as variable
of scar tissue on contrast-enhanced MRI
Total scar score
LVEF
LVESV
LVEDV
Age
Diuretic usage
Statin usage
Diabetes mellitus
Hypertension
6.2
1.06/%
0.86/10 ml
1.2/10 ml
1.4/10 yrs
4.8
0.27
1.1
0.30
1.7–23
0.97–1.2
0.63–1.2
0.93–1.6
0.76–2.7
1.2–19
0.08–0.88
0.38–3.6
0.10–0.91
7.4
1.5
0.9
2.2
1.2
5.2
4.8
0.0
4.6
0.006
0.2
0.3
0.1
0.3
0.023
0.029
0.8
0.033
The different models use the different variables for scar tissue on contrast-enhanced MRI.
Figure 1. Kaplan-Meier curve analysis showing the difference in mortality
when patients are stratified according to a large extent of scar tissue (spatial
extent ⱖ6) or a small extent of scar tissue (spatial extent ⬍6) on contrastenhanced MRI.
infarction. Gerber et al18 evaluated 20 patients after acute
infarction with contrast-enhanced MRI and myocardial tagging and noted that improvement in circumferential shortening was inversely related to the regional extent of hyperenhancement on contrast-enhanced images.
Assessment of scar tissue using contrast-enhanced MRI
also plays an important role in chronic CAD. Kim et al15
evaluated patients with chronic ischemic LV dysfunction
and reported that an increasing transmurality of scar tissue
was significantly related with absence of functional recovery after revascularization.
Survivors of acute myocardial infarction are at increased
risk for subsequent fatal and nonfatal cardiovascular
events.19 Early studies demonstrated that total cardiac enzyme release, as an indicator of the extent of myocardial
necrosis, is related with short- and long-term prognosis after
myocardial infarction.20,21 Subsequent studies demonstrated
that the degree of LV dysfunction correlates well with
mortality and is useful in risk stratification of patients after
acute myocardial infarction.22,23 White et al7 evaluated 605
patients after acute infarction, with a mean follow-up of 78
months (range 15 to 165), showing the powerful prognostic
Coronary Artery Disease/Infarct Size Best Predictor of Mortality
value of LVEF and LVESV. More recently, Sharir et al6
demonstrated in a large population (n ⫽ 2,686 patients) that
poststress LVEF on gated single-photon emission computed
tomographic imaging was the best predictor of cardiac
death. Accordingly, LVEF and/or LV volumes have become
the established predictors for mortality in patients with
CAD. Patients with severe LV dilatation and remodeling are
at high risk for development of heart failure. In patients with
severe heart failure (New York Heart Association class III
to IV) annual mortality is high1,3 and the main cause of
death is progressive pump failure.1– 4 In mild heart failure
(New York Heart Association class II), the overall annual
mortality ranges from 5% to 15%, with a relatively high
percentage of sudden cardiac deaths.2,5 Although the precise
mechanism underlying lethal ventricular arrhythmias is not
clear, it has been demonstrated that scar tissue may serve as
a substrate for these arrhythmias.24,25
Recent data have identified scar tissue and severely depressed LVEF (derived from gated single-photon emission
computed tomography) as important predictors of death or
ventricular arrhythmias in patients with CAD.26 Wu et al27
studied 44 patients with cine and contrast-enhanced MRI
and found that infarct size was directly related with cardiovascular complications after myocardial infarction, whereas
LV volumes and LVEF had no significant predictive value
for clinical outcome. A more recent study of Bello et al28
demonstrated that infarct size determined with contrastenhanced MRI was superior to LVEF for identification of
patients with a substrate for inducible ventricular tachycardia. In addition, Yan et al29 demonstrated that the extent of
the peri-infarct zone characterized by contrast-enhanced
MRI provides incremental prognostic value beyond LVEF
and LVESV. Based on the studies discussed,26 –29 it is conceivable that scar tissue may be superior to LVEF and LV
volumes for prediction of all-cause mortality because of its
additional value to predict death due to ventricular arrhythmias in patients with preserved or moderately depressed LV
function, who are not likely to die of heart failure. Preliminary data revealed that infarct size on contrast-enhanced
MRI was a better predictor for survival compared with
LVEF in patients with recent myocardial infarction.8 This
observation agrees with the current results in 231 patients
with healed myocardial infarction, identifying the extent of
scar tissue on contrast-enhanced MRI as a better predictor
for all-cause mortality than LV function and/or dimensions.
A further explanation for the superior prognostic value of
scar tissue to LVEF and/or LV volumes could be that the
extent of scar tissue is a direct marker of infarct size,
whereas LVEF only indirectly reflects myocardial damage.28
Several limitations of the present study need to be addressed. First, the small number of mortality end points
precludes exclusion of all potential confounding factors and
the present conclusion requires confirmation in substantially
larger patient groups. Diuretic usage was associated with
relatively poor MRI results (i.e., extensive scar tissue) and
with increased mortality, and hence acted as true confounder. Most likely, diuretics are a substitute for severe
heart failure, and thus reflects a poor clinical condition
caused by LV dysfunction, rather than that the usage itself
should be considered the cause of mortality. Simulta-
935
neously, patients who have developed heart failure have a
larger extent of myocardial damage. It is important to note,
however, that the prognostic value of MRI variables was
maintained after adjustment for diuretic usage. Statin usage
was also associated with poor prognosis. However, statin
users and nonusers had similar MRI results (Table 4). Thus,
statin usage did not act as confounder, and did not influence
the relation between MRI results and the primary end point.
Second, the power of the study is limited because of the
small number of events. Furthermore, the small number of
events does not permit distinction between heart failure
death and sudden cardiac death. Larger studies with subsequently higher event rates are needed to confirm that infarct
size on contrast-enhanced MRI is superior to LV function
and volumes in predicting mortality and to further assess the
value of infarct size as a predictor of mode of death. Also,
the duration of follow-up was limited, and studies with
longer follow-up are needed.
Third, not only viability and scar tissue are important for
prognosis, but also stress-inducible ischemia is a relevant
factor. To provide the full picture on jeopardized myocardium (viability and ischemia), contrast-enhanced MRI
should be combined with stress–rest perfusion MRI. Future
studies are needed to verify this hypothesis.
1. The CONSENSUS Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North
Scandinavian Enalapril Survival Study (CONSENSUS). N Engl J Med
1987;316:1429 –1435.
2. Effect of metoprolol CR/XL in chronic heart failure: Metoprolol
CR/XL Randomised Intervention Trial in Congestive Heart Failure
(MERIT-HF). Lancet 1999;353:2001–2007.
3. Califf RM, Adams KF, McKenna WJ, Gheorghiade M, Uretsky BF,
McNulty SE, Darius H, Schulman K, Zannad F, Handberg-Thurmond
E, et al. A randomized controlled trial of epoprostenol therapy for
severe congestive heart failure: The Flolan International Randomized
Survival Trial (FIRST). Am Heart J 1997;134:44 –54.
4. Carson P, Anand I, O’Connor C, Jaski B, Steinberg J, Lwin A,
Lindenfeld J, Ghali J, Barnet JH, Feldman AM, Bristow MR. Mode of
death in advanced heart failure: the Comparison of Medical, Pacing,
and Defibrillation Therapies in Heart Failure (COMPANION) trial.
J Am Coll Cardiol 2005;46:2329 –2334.
5. Gradman A, Deedwania P, Cody R, Massie B, Packer M, Pitt B,
Goldstein S. Predictors of total mortality and sudden death in mild to
moderate heart failure. Captopril-Digoxin Study Group. J Am Coll
Cardiol 1989;14:564 –570.
6. Sharir T, Germano G, Kang X, Lewin HC, Miranda R, Cohen I,
Agafitei RD, Friedman JD, Berman DS. Prediction of myocardial
infarction versus cardiac death by gated myocardial perfusion SPECT:
risk stratification by the amount of stress-induced ischemia and the
poststress ejection fraction. J Nucl Med 2001;42:831– 837.
7. White HD, Norris RM, Brown MA, Brandt PW, Whitlock RM, Wild
CJ. Left ventricular end-systolic volume as the major determinant of
survival after recovery from myocardial infarction. Circulation 1987;
76:44 –51.
8. Kansal P, Ortiz J, Bucciarelli-Ducci C, Lee DC, Holly TA, Klocke FJ,
Bonow RO, Wu E. Infarct size by contrast-enhanced magnetic resonance imaging predicts cardiovascular outcome after acute myocardial
infarction (abstr). J Cardiovasc Magnetic Res 2006; 8:109.
9. Lamb HJ, Doornbos J, van der Velde EA, Kruit MC, Reiber JH, de
Roos A. Echo planar MRI of the heart on a standard system: validation
of measurements of left ventricular function and mass. J Comput Assist
Tomogr 1996;20:942–949.
10. van der Geest RJ, Buller VG, Jansen E, Lamb HJ, Baur LH, van der
Wall EE, de Roos A, Reiber JH. Comparison between manual and
semiautomated analysis of left ventricular volume parameters from
short-axis MR images. J Comput Assist Tomogr 1997;21:756 –765.
936
The American Journal of Cardiology (www.AJConline.org)
11. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey
WK, Pennell DJ, Rumberger JA, Ryan T, Verani MS. Standardized
myocardial segmentation and nomenclature for tomographic imaging
of the heart: a statement for healthcare professionals from the Cardiac
Imaging Committee of the Council on Clinical Cardiology of the
American Heart Association. Circulation 2002;105:539 –542.
12. Wu E, Judd RM, Vargas JD, Klocke FJ, Bonow RO, Kim RJ. Visualisation of presence, location, and transmural extent of healed Q-wave
and non-Q-wave myocardial infarction. Lancet 2001;357:21–28.
13. Kaandorp TA, Bax JJ, Lamb HJ, Viergever EP, Boersma E, Poldermans D, van der Wall EE, de Roos A. Which parameters on magnetic
resonance imaging determine Q waves on the electrocardiogram? Am J
Cardiol 2005;95:925–929.
14. Engelstein ED, Zipes DP. Sudden cardiac death. In: Wayne Alexander,
ed. The Heart, Arteries and Veins. 9th ed. New York: McGraw-Hill,
1998:1081–1112.
15. Kim RJ, Wu E, Rafael A, Chen EL, Parker MA, Simonetti O, Klocke
FJ, Bonow RO, Judd RM. The use of contrast-enhanced magnetic
resonance imaging to identify reversible myocardial dysfunction.
N Engl J Med 2000;343:1445–1453.
16. Kim RJ, Fieno DS, Parrish TB, Harris K, Chen EL, Simonetti O,
Bundy J, Finn JP, Klocke FJ, Judd RM. Relationship of MRI delayed
contrast enhancement to irreversible injury, infarct age, and contractile
function. Circulation 1999;100:1992–2002.
17. Choi KM, Kim RJ, Gubernikoff G, Vargas JD, Parker M, Judd RM.
Transmural extent of acute myocardial infarction predicts long-term
improvement in contractile function. Circulation 2001;104:1101–
1107.
18. Gerber BL, Garot J, Bluemke DA, Wu KC, Lima JA. Accuracy of
contrast-enhanced magnetic resonance imaging in predicting improvement of regional myocardial function in patients after acute myocardial
infarction. Circulation 2002;106:1083–1089.
19. Kannel WB, Sorlie P, McNamara PM. Prognosis after initial myocardial infarction: the Framingham study. Am J Cardiol 1979;44:53–59.
20. Sobel BE, Bresnahan GF, Shell W, Yoder RD. Estimation of infarct size
in man and its relation to prognosis. Circulation 1972;46:640 – 648.
21. Thompson PL, Fletcher EE, Katavatis V. Enzymatic indices of myocardial necrosis: influence on short- and long-term prognosis after
myocardial infarction. Circulation 1979;59:113–119.
22. Risk stratification and survival after myocardial infarction. N Engl
J Med 1983;309:331–336.
23. Stadius ML, Davis K, Maynard C, Ritchie JL, Kennedy JW. Risk
stratification for 1 year survival based on characteristics identified
in the early hours of acute myocardial infarction. The Western
Washington Intracoronary Streptokinase Trial. Circulation 1986;
74:703–711.
24. Geltman EM, Ehsani AA, Campbell MK, Schechtman K, Roberts R,
Sobel BE. The influence of location and extent of myocardial infarction on long-term ventricular dysrhythmia and mortality. Circulation
1979;60:805– 814.
25. Bolick DR, Hackel DB, Reimer KA, Ideker RE. Quantitative analysis
of myocardial infarct structure in patients with ventricular tachycardia.
Circulation 1986;74:1266 –1279.
26. van der Burg AE, Bax JJ, Boersma E, Pauwels EK, van der Wall EE,
Schalij MJ. Impact of viability, ischemia, scar tissue, and revascularization on outcome after aborted sudden death. Circulation 2003;108:
1954 –1959.
27. Wu KC, Zerhouni EA, Judd RM, Lugo-Olivieri CH, Barouch LA,
Schulman SP, Blumenthal RS, Lima JA. Prognostic significance of
microvascular obstruction by magnetic resonance imaging in patients with acute myocardial infarction. Circulation 1998;97:765–
772.
28. Bello D, Fieno DS, Kim RJ, Pereles FS, Passman R, Song G, Kadish
AH, Goldberger JJ. Infarct morphology identifies patients with substrate for sustained ventricular tachycardia. J Am Coll Cardiol 2005;
45:1104 –1108.
29. Yan AT, Shayne AJ, Brown KA, Gupta SN, Chan CW, Luu TM, Di
Carli MF, Reynolds HG, Stevenson WG, Kwong RY. Characterization
of the peri-infarct zone by contrast-enhanced cardiac magnetic resonance imaging is a powerful predictor of post-myocardial infarction
mortality. Circulation 2006;114:32–39.
Prevalence and Prognostic Implications of ST-Segment Deviations
from Ambulatory Holter Monitoring After ST-Segment Elevation
Myocardial Infarction Treated With Either Fibrinolysis or Primary
Percutaneous Coronary Intervention (a Danish Trial in Acute
Myocardial Infarction-2 Substudy)
Lars Idorn, MDa,*, Dan Eik Høfsten, MDa, Kristian Wachtell, MD, PhDb,
Henning Mølgaard, MD, DMScc, and Kenneth Egstrup, MD, DMSca,
for the DANAMI-2 Investigators
Ambulatory Holter monitoring has been shown to be useful in stratifying cardiovascular
risk after acute myocardial infarction. However, it remains unclear whether ST-segment
deviations might predict clinical outcomes in a population treated with primary percutaneous coronary intervention (PCI) compared with thrombolysis. Holter monitoring was
initiated at discharge from ST-segment elevation myocardial infarction in 958 patients
followed for 2,773 patient-years, randomized to immediate revascularization with either
fibrinolysis (n ⴝ 474) or PCI (n ⴝ 484). The primary end point was all-cause mortality, and
the secondary end point was a composite of death, reinfarction, and disabling stroke. The
prevalences of ST-segment depression (STd) and ST-segment elevation (STe) were similar
in patients treated with fibrinolysis or PCI (both p ⴝ NS). During follow-up, 58 patients
died (primary PCI vs fibrinolysis hazard ratio 0.74, p ⴝ 0.25). The secondary end point was
reached in 113 patients (primary PCI vs fibrinolysis hazard ratio 0.66, p ⴝ 0.03). In
fibrinolysis-treated patients, mortality and the secondary end point were significantly
higher in patients with STe (both end points p <0.001), an association that remained
statistically significant after adjustment for age, gender, anterior infarction, ␤-blocker
treatment, left ventricular systolic function, and STd (p ⴝ 0.03 and p ⴝ 0.005, respectively).
Significant associations were not observed for STd. In PCI-treated patients, there was no
association between either STe or STd and outcome. In conclusion, immediate revascularization with PCI during STe myocardial infarction does not affect the subsequent prevalence of ST-segment deviation compared with fibrinolysis. However, although STe is an
independent predictor of mortality and nonfatal major cardiovascular events in patients
treated with fibrinolysis, it does not predict outcome after PCI, perhaps because of more
complete revascularization. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;
100:937–943)
The primary objective of this study was to determine the
prevalence of ST-segment depression (STd) and ST-segment elevation (STe) detected by Holter monitoring in survivors of acute STe myocardial infarction and the prognos-
tic value of this information in comparison with other
established risk stratification variables, such as exercise
capacity, left ventricular (LV) systolic function, and other
clinical information.
Methods
a
Department of Medical Research, Funen Hospital, Svendborg; bDepartment of Cardiology, Copenhagen University Hospital, Rigshospitalet,
Copenhagen; and cDepartment of Cardiology, Aarhus University Hospital,
Skejby Sygehus, Aarhus, Denmark. Manuscript received February 7, 2007;
revised manuscript received and accepted April 13, 2007.
The Danish Trial in Acute Myocardial Infarction–2 (DANAMI-2) was
supported by grants from the Danish Heart Foundation, Copenhagen,
Denmark; the Danish Medical Research Council, Copenhagen, Denmark;
AstraZeneca, London, United Kingdom; Bristol-Myers Squibb, Cincinnati,
Ohio; Cordis, Miami, Florida; Pfizer, Inc., New York, New York; Pharmacia-Upjohn, London, United Kingdom; Boehringer Ingelheim, Ingelheim, Germany; and Guerbet Villepinte, France.
*Corresponding author: Tel: 45-6320-2404; fax: 45-6320-2407.
E-mail address: [email protected] (L. Idorn).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.032
The design and results of the Danish Trial in Myocardial
Infarction–2 (DANAMI-2) main study have previously
been reported.1,2 DANAMI-2 was a Danish multicenter
study with 29 participating centers (24 referral and 5 invasive). The main inclusion criteria were age ⱖ18 years, the
presence of symptoms characteristic of acute myocardial
infarction (AMI) for ⱖ30 minutes but ⬍12 hours, and
cumulative STe of ⱖ4 mm in ⱖ2 contiguous leads. Patients
with contraindications to either fibrinolysis or percutaneous
coronary intervention (PCI) and patients with left bundle
branch block on electrocardiography were excluded. Patients judged to be at high risk during transportation because
www.AJConline.org
938
The American Journal of Cardiology (www.AJConline.org)
of cardiogenic shock or severe heart failure (sustained systolic blood pressure ⱕ65 mm Hg), persistent life-threatening arrhythmias, or a need for mechanical ventilation were
also excluded. Patients randomized to PCI at noninvasive
referral centers were immediately transferred to the regional
invasive center, whereas those randomized to fibrinolysis
were treated at the admitting hospitals.
The study was conducted in accordance with the Declaration of Helsinki and was approved by the National Ethics
Committee of Denmark. All eligible patients provided written informed consent. The present substudy and end points
were prespecified in the DANAMI-2 protocol.
Twenty-four-hour Holter monitoring was performed by
the randomizing center and initiated on the day of discharge.
Detailed written information was supplied, and the teaching
of nurses and technicians was available on request.
Routinely, 2 bipolar leads were recorded (CM5 and CM3),
but the use of alternative leads was allowed, if necessary, to
avoid leads with pathologic Q waves or to ensure technically
adequate recordings. Calibrated high-quality tapes (TDK C-90;
TDK U.S.A. Corporation, Garden City, New York) were used.
Recordings were submitted unanalyzed to a core laboratory
(Department of Medical Research, Funen Hospital, Svendborg, Denmark) and analyzed using a semiautomated program
(Reynolds Pathfinder 700; Del Mar Reynolds Medical Ltd.,
Hertford, United Kingdom). The complete recordings were
subsequently reviewed manually by an experienced observer,
blinded to all patient information. ST-segment deviations detected by the program were all analyzed beat by beat by the
observer and corrected if necessary. Recordings containing
pace rhythm or atrial fibrillation or with QRS durations ⬎120
ms were excluded. We prospectively defined a significant
episode of ST-segment deviation to be a shift of ⱖ1 mm from
baseline at the ST-segment position measured 80 ms from the
J point and lasting ⱖ1 minute. Furthermore, only horizontal or
down-sloping STd was considered significant. Patients with
persistent ST-segment deviations were considered to have STdeviations only if a further dynamic shift ⱖ1 mm from baseline was observed.
Echocardiography was performed at the time of discharge from the randomizing center using video recording
and submitted to a core laboratory (Department of Medicine, Glostrup University Hospital, Glostrup, Denmark),
where LV systolic function was assessed by a blinded observer using a previously validated wall motion index
(WMI) scoring system.3 The WMI was estimated semiquantitatively using a 16-segment score and assessed using the
parasternal long- and short-axis views and apical views. The
WMI score was visually assessed, accounting for the contribution of each LV segment to the systolic reduction of LV
volume; a score of 2 was assigned to each segment with
normal thickening; scores of 1.5, 1.0, and 0.5 were assigned
to mildly, moderately, and severely hypokinetic segments,
respectively; and scores of 0 and ⫺1 were assigned to
akinetic and dyskinetic segments, respectively. The WMI
was calculated by adding the score of each segment and
then dividing by the number of segments. From the resulting
WMI score, the LV ejection fraction was estimated using a
previously validated method.4 All echocardiograms were
blindly read by 2 established echocardiographic readers.
Coronary angiograms obtained during the primary revas-
cularization procedures in patients randomized to PCI were
evaluated by an independent core laboratory (Cardialysis,
Rotterdam, The Netherlands), where the number of narrowed coronary arteries (i.e., vessels with ⱖ50% diameter
stenosis) and post-PCI coronary flow grade were assessed
according to the Thrombolysis In Myocardial Infarction
(TIMI) classification.5
Physicians responsible for the follow-up and treatment of
the patients were kept unaware of the results of Holter
monitoring throughout the study. The survival status of all
patients was obtained from the Danish Civil Registration
System, in which all deaths in the country are recorded
within 2 weeks. The primary end point of this study was
all-cause mortality. The secondary end point was a composite of death, reinfarction, or disabling stroke. Patients who
reached the nonfatal secondary end point after randomization but before the initiation of Holter monitoring were
excluded from the secondary analyses (myocardial reinfarction n ⫽ 18, stroke n ⫽ 3). Patient follow-up was 3 years.
Patients experiencing nonfatal events continued follow-up
for the primary mortality end point. Smoking cessation
within the past 6 months was considered current smoking.
Statistical analyses were conducted with a commercially
available software package (Stata/SE version 9.2; StataCorp
LP, College Station, Texas). Results were analyzed according to the intention-to-treat principle. Continuous variables
are reported as mean ⫾ SD and prevalence as percentage.
Comparison of patient characteristics and the prevalence of
predefined Holter measurements was performed with Student’s t test (unpaired, 2 tailed) for continuous variables,
and categorical variables were compared using a chi-square
test or Fisher’s exact test as appropriate. Interaction between
randomized treatment and differences in clinical characteristics associated with the prespecified Holter variables was
tested using logistic regression. Survival curves were plotted using the Kaplan-Meier method and compared using
log-rank tests. Using multivariate Cox proportional-hazards
regression analysis, the prespecified Holter variables were
adjusted for the following variables, chosen before the inspection of data: age, gender, diabetes, anterior index AMI,
and the LV ejection fraction. Proportionality of hazards was
tested graphically on the basis of visual inspection of loglog survival curves and by performing a formal test of
proportionality on the basis of Schoenfeld residuals for all
variables in the model. Furthermore, for age and the LV
ejection fraction, linearity of hazards was tested by adding
squared values of the variables to the model. Assumptions
of proportionality and linearity of hazards were met for all
variables. A 2-sided p value ⬍0.05 was considered statistically significant in all analyses.
Results
Of the total of 1,462 patients discharged after an index AMI,
Holter monitoring was performed in 1,159 patients for a
median of 6 days (interquartile range 3) after randomization.
Patients in whom Holter monitoring was not performed (n
⫽ 303) were significantly older (64.7 ⫾ 13.8 vs 61.7 ⫾ 11.8
years, p ⬍0.001) and experienced a significantly higher
mortality rate during follow-up (hazard ratio [HR] 2.1, 95%
confidence interval [CI] 1.4 to 3.0). In addition, 201 record-
Coronary Artery Disease/ST-Segment Deviation After Coronary Revascularization
939
Table 1
Baseline characteristics of patients treated with fibrinolysis or percutaneous coronary intervention
Variable
Age (yrs), mean ⫾ SD
Men
Current smoker
Body mass index (kg/m2), mean ⫾ SD
Treated hypertension
Previous myocardial infarction
Previous stroke or transient ischemic attack
Diabetes mellitus
Anterior index myocardial infarction
LV ejection fraction, mean ⫾ SD
LV ejection fraction ⱕ35%*
Angiographic features
Nonstenotic coronary vessels
Single-vessel coronary disease
Double-vessel coronary disease
Triple-vessel coronary disease
Medication at discharge
Aspirin
␤ blockers
Angiotensin-converting enzyme inhibitors
Diuretics
Statins
Fibrinolysis
(n ⫽ 474)
Primary PCI
(n ⫽ 484)
p Value
61.2 ⫾ 11.9
356 (75.1%)
290 (61.8%)
26.6 ⫾ 4.2
98 (20.7%)
49 (10.3%)
24 (5.1%)
28 (5.9%)
236 (49.8%)
46.4 ⫾ 11.1%
80 (17.7%)
61.2 ⫾ 11.4
365 (75.4%)
299 (62.0%)
26.2 ⫾ 3.8
92 (19.0%)
49 (10.1%)
11 (2.3%)
30 (6.2%)
245 (50.6%)
47.0 ⫾ 11.0%
78 (16.6%)
0.98
0.91
0.95
0.10
0.52
0.91
0.02
0.85
0.80
0.47
0.65
19 (4.1%)
148 (31.7%)
148 (31.7%)
152 (32.6%)
463 (97.7%)
417 (88.0%)
175 (36.9%)
126 (26.6%)
250 (52.7%)
468 (96.7%)
431 (89.0%)
156 (32.2%)
101 (20.9%)
242 (50.0%)
0.36
0.60
0.13
0.04
0.40
* Percentages are relative to the number of patients for whom LV ejection fractions were available (n ⫽ 923).
ings were excluded because of pace rhythm (n ⫽ 5), atrial
fibrillation (n ⫽ 26), QRS duration ⬎120 ms (n ⫽ 21), or
technically inadequate quality of the recordings (n ⫽ 149),
leaving a total of 958 recordings available for the study.
Patients excluded because of the prespecified electrocardiographic abnormalities experienced a higher mortality rate
during follow-up (HR 3.0, 95% CI 1.5 to 6.1), which was
not the case for patients excluded for technical reasons (HR
0.6, 95% CI 0.2 to 1.4). Five patients were lost to follow-up.
The remaining patients were followed for 3 years or until
the primary end point was met. Patient characteristics are
listed in Table 1.
ST-segment deviations were found in 263 patients
(27.5%). In 100 patients (10.4%), only STd was found; in
139 patients (14.5%), only STe was found; and in 24 patients (2.5%), STd and STe were found. In the fibrinolysis
group, ST-segment deviations were found in 141 patients
(29.7%), compared with 122 (25.2%) in the PCI group (p ⫽
0.12). STd was found in 66 patients (13.9%) in the fibrinolysis group, compared with 58 (12.0%) in the PCI group
(p ⫽ 0.37). STe was found in 90 patients (19.0%) in the
fibrinolysis group, compared with 73 (15.1%) in the PCI
group (p ⫽ 0.11).
The clinical characteristics of patients with STe and STd
in the 2 treatment groups and in the total study population
are listed in Tables 2 and 3. Estimated by echocardiography,
LV ejection fractions were significantly lower in patients
with STe, whereas STd was significantly more frequent in
patients with positive exercise test results. Additionally,
male gender and anterior index AMI were significantly
more frequent in patients with STe. Patients with STe and
STd were significantly older than patients without. The
presence of pathologic Q waves on electrocardiography at
rest was not significantly associated with the presence of
STd or STe. In the PCI group, the presence of TIMI grade
3 flow after revascularization was not significantly associated with STe or STd.
Patients were asked to keep diaries of symptoms during
the Holter monitoring. Of the total number of STe events,
symptoms were reported in only 7.5% (14.7% for STd). In
none of these cases, patients were clinically evaluated as
having had new AMIs.
During 2,773 patient-years of follow-up, 58 patients died
(primary PCI vs fibrinolysis HR 0.74, p ⫽ 0.25). The
composite end point of death, reinfarction, or disabling
stroke was reached in 113 patients (primary PCI vs fibrinolysis HR 0.66, p ⫽ 0.03). Of these patients, 62 had reinfarctions and 17 had disabling strokes. In the fibrinolysis
group, STe was associated with increased mortality (HR
4.3, 95% CI 2.2 to 8.4, p ⬍0.001; Figure 1). A similar
association was found between the secondary end point and
STe in the fibrinolysis group (HR 2.6, 95% CI 1.6 to 4.3, p
⬍0.001). In patients randomized to primary PCI, however,
no association between STe and outcome was observed for
mortality (HR 0.23, 95% CI 0.03 to 1.7, p ⫽ 0.15) or the
composite secondary end point (HR 0.7, 95% CI 0.27 to 1.7,
p ⫽ 0.41). For STd, a similar tendency was observed for
all-cause mortality, mainly in patients treated with fibrinolysis, although this difference did not reach statistical significance (Figure 2).
Multivariate Cox proportional-hazards analysis was performed in the total study population using the following
variables: age, gender, anterior wall infarction, diabetes,
␤-blocker treatment at discharge, the LV ejection fraction,
STd, STe, and randomized treatment. Separate analyses
were also performed for each of the 2 randomized treatment
940
The American Journal of Cardiology (www.AJConline.org)
Table 2
Clinical characteristics by ST-segment elevation
Fibrinolysis (n ⫽ 474)
Variable
Age (yrs), mean ⫾ SD
Q wave on electrocardiogram
Positive exercise test results*
LV ejection fraction ⱕ35%*
Men
Anterior myocardial infarction
Narrowed coronary arteries*
1
2
3
TIMI grade 3 flow†
PCI (n ⫽ 484)
STe
(n ⫽ 90)
No STe
(n ⫽ 384)
p Value
STe
(n ⫽ 73)
No STe
(n ⫽ 411)
p Value
63.6 ⫾ 13.0
61 (67.8%)
21 (28.0%)
31 (35.2%)
75 (83.3%)
62 (68.9%)
60.7 ⫾ 11.5
261 (68.0%)
110 (32.4%)
83 (22.9%)
281 (73.2%)
174 (45.3%)
0.03
0.97
0.46
0.02
0.05
⬍0.001
59.2 ⫾ 11.5
56 (76.7%)
16 (24.6%)
26 (36.6%)
67 (91.8%)
55 (75.3%)
61.6 ⫾ 11.4
262 (63.7%)
63 (17.7%)
86 (21.6%)
298 (72.5%)
190 (46.2%)
28 (40.6%)
20 (29.0%)
21 (30.4%)
56 (86.2%)
120 (31.7%)
128 (33.8%)
131 (34.6%)
307 (83.9%)
0.10
0.03
0.19
0.01
⬍0.001
⬍0.001
0.35
—
—
—
0.64
p Value for
Interaction
0.008
0.24
0.14
0.80
0.12
0.45
—
—
—
—
—
* Percentage is relative to the number of patients who performed exercise tests (n ⫽ 835), echocardiography performed (n ⫽ 920), and significant stenosis
in the primary PCI (n ⫽ 465).
†
TIMI grade 3 flow refers to post-angioplasty flow in the culprit lesion.
Table 3
Clinical characteristics by ST-segment depression
Fibrinolysis (n ⫽ 474)
Variable
Age (yrs), mean ⫾ SD
Q wave on electrocardiogram
Positive exercise test results*
LV ejection fraction ⱕ35%*
Men
Anterior myocardial infarction
Narrowed coronary arteries*
1
2
3
TIMI grade 3 flow†
PCI (n ⫽ 484)
STd
(n ⫽ 66)
No STd
(n ⫽ 408)
p Value
STd
(n ⫽ 58)
No STd
(n ⫽ 426)
p Value
64.6 ⫾ 12.4
46 (69.7%)
25 (43.1%)
20 (32.8%)
47 (71.2%)
37 (56.1%)
60.8 ⫾ 11.7
276 (67.6%)
106 (29.7%)
94 (24.2%)
309 (75.7%)
199 (48.8%)
0.04
0.74
0.04
0.15
0.43
0.27
62.9 ⫾ 12.1
36 (62.1%)
17 (34.7%)
11 (19.6%)
47 (81.0%)
30 (51.7%)
61.0 ⫾ 12.1
282 (66.2%)
62 (16.7%)
101 (24.4%)
318 (74.6%)
215 (50.5%)
17 (33.2%)
18 (33.0%)
19 (33.8%)
44 (86.3%)
131 (33.2%)
130 (33.0%)
133 (33.8%)
319 (83.9%)
0.23
0.53
0.002
0.43
0.29
0.86
0.96
—
—
—
0.67
p Value for
Interaction
0.60
0.34
0.38
0.16
0.19
0.53
—
—
—
—
—
* Percentage is relative to the number of patients who performed exercise tests (n ⫽ 835), echocardiography performed (n ⫽ 920), and significant stenosis
in the primary PCI (n ⫽ 465).
†
TIMI grade 3 flow refers to post-angioplasty flow in the culprit lesion.
groups, and interactions with randomized treatment and
each of the included variables were tested (Tables 4 and 5).
In the fibrinolysis group, STe was independently associated
with increased mortality and with the composite secondary
end point, whereas in the PCI group, this association was
not found (Tables 4 and 5). Accordingly, for the 2 end
points, a significant interaction between randomized treatment and the prognostic value of STe was demonstrated
(Tables 4 and 5). No significant association between STd
and either of the 2 end points was found in either of the 2
treatment groups or in the total study population after multivariate adjustment.
Discussion
This is the first randomized study to compare the prevalence
and prognostic significance of ST-segment deviations after
PCI and fibrinolysis. Our study demonstrates that PCI treatment does not affect the subsequent prevalence of STsegment deviations compared with fibrinolysis treatment.
Conversely, although STe is associated with increased mortality and major cardiovascular events in patients treated
with fibrinolysis, it does not have similar prognostic implications after PCI. Furthermore, STd did not predict outcome
in any of the 2 treatment groups.
Within the past decade, a large amount of data involving
Holter ST-segment monitoring early after myocardial infarction have become available. The main interest has concentrated on transient episodes of STd, and several reports
have consistently demonstrated that most such episodes
reflect myocardial ischemia.6,7 It has also been demonstrated that ambulatory STd is associated with adverse outcomes, although there is considerable disagreement about
how the prognostic information is expressed in terms of
cardiac events.8 –11 In contrast, the clinical importance of
transient STe has been only sporadically evaluated. In some
studies, STe has been interpreted as a marker of myocardial
ischemia,10,12–14 whereas others analyzed only episodes of
STd.15–17 From the limited data available, it appears that
Coronary Artery Disease/ST-Segment Deviation After Coronary Revascularization
Figure 1. Kaplan-Meier survival estimates according to the presence
(dashed line) or absence (solid line) of STe on Holter recording.
episodes of STe occurring in leads with abnormal Q waves
are associated with established markers of large infarcts,
severe LV systolic dysfunction, and poor clinical outcomes in terms of a composite of death, reinfarction, and
disabling stroke.18 These observations are supported by
our findings: STe was particularly prevalent in patients
with large anterior infarcts but not associated with the
presence of exercise-inducible ischemia. In most previous studies, however, it has not been reported whether
941
Figure 2. Kaplan-Meier survival estimate according to the presence
(dashed line) or absence (solid line) of STd on Holter recording.
episodes of STe occurred in leads with or without pathologic Q waves.10 –14,17 Furthermore, disagreement about
the validity of ST-segment deviations in leads with
pathologic Q waves further complicates the standard of
reference among studies.
Because PCI in most cases leads to fast and complete
revascularization of the infarct-related area, it could be
speculated that subsequent ST-segment deviations on Holter
monitoring would be less common than observed among
942
The American Journal of Cardiology (www.AJConline.org)
Table 4
Multivariate analysis, predictors of all-cause mortality
Variable
HR (95% CI)
Univariate
Multivariate
All Patients
STe
STd
Age
LV ejection fraction§
Men
Anterior myocardial infarction
Diabetes mellitus
␤-blocker treatment
Randomized to primary PCI
p Value for
Interaction
†
2.06 (1.17–3.62)
1.82 (0.96–3.44)
1.10 (1.08–1.13)‡
0.94 (0.92–0.96)‡
1.03 (0.57–1.88)
2.65 (1.49–4.72)†
1.85 (0.79–4.31)
0.43 (0.23–0.80)†
0.74 (0.44–1.24)
All Patients
Fibrinolysis
Primary PCI
1.32 (0.72–2.41)
1.36 (0.70–2.65)
1.10 (1.07–1.13)‡
0.96 (0.93–0.98)‡
1.33 (0.71–2.51)
1.37 (0.73–2.58)
1.80 (0.76–4.23)
0.47 (0.25–0.88)*
077 (0.44–1.34)
2.25 (1.10–4.62)*
1.67 (0.74–3.76)
1.10 (1.05–1.14)‡
0.96 (0.92–0.99)*
1.43 (0.62–3.34)
1.45 (0.61–3.48)
1.43 (0.43–4.81)
0.45 (0.20–1.04)
—
0.20 (0.03–1.57)
0.95 (0.27–2.43)
1.10 (1.05–1.15)‡
0.95 (0.91–0.99)†
1.31 (0.49–3.51)
1.24 (0.49–3.17)
2.00 (0.56–7.09)
0.44 (0.16–1.23)
—
0.03
0.44
0.97
0.80
0.51
0.65
0.65
0.99
—
* p ⬍0.05; † p ⬍0.01; ‡ p ⬍0.001.
Absolute percentage of LV ejection fraction.
§
Table 5
Multivariate analysis, predictors of death, reinfarction, or major stroke
Variable
HR (95% CI)
Univariate
All Patients
STe
STd
Age
Men
Diabetes mellitus
LV ejection fraction§
Anterior myocardial infarction
␤-blocker treatment
Randomized to primary PCI
1.76 (1.15–2.68)*
1.47 (0.91–2.39)
1.05 (1.03–1.07)‡
1.15 (0.74–1.80)
2.36 (1.35–4.13)†
0.97 (0.95–0.98)‡
1.61 (1.11–2.35)*
0.59 (0.36–0.96)
0.66 (0.45–0.96)†
p Value for
Interaction
Multivariate
All Patients
Fibrinolysis
1.47 (0.94–2.31)
1.10 (0.66–1.84)
1.05 (1.03–1.07)‡
1.35 (0.84–2.15)
2.70 (1.52–4.78)‡
0.98 (0.96–1.00)*
1.13 (0.74–1.72)
0.66 (0.40–1.07)
0.66 (0.45–0.97)*
Primary PCI
†
2.14 (1.25–3.64)
1.34 (0.72–2.48)
1.04 (1.02–1.06)‡
1.36 (0.74–2.50)
1.64 (0.65–4.12)
0.98 (0.96–1.01)
0.90 (0.52–1.55)
0.65 (0.34–1.21)
—
0.69 (0.27–1.81)
0.67 (0.25–1.77)
1.07 (1.04–1.10)‡
1.50 (0.72–3.11)
4.37 (2.00–9.56)‡
0.98 (0.96–1.01)
1.65 (0.84–3.26)
0.67 (0.29–1.54)
—
0.05
0.33
0.18
0.71
0.08
0.50
0.23
0.90
—
* p ⬍0.05; † p ⬍0.01; ‡ p ⬍0.001.
Absolute percentage of LV ejection fraction.
§
patients treated with fibrinolysis. However, this could not be
confirmed in our study. It has been shown in a communitybased patient sample that the added absolute benefit of PCI
increases with higher overall risk,19 and this was also evident in the DANAMI-2 trial.20 The trend toward better
survival of patients with AMI made possible by PCI could
potentially increase the number of patients with ST-segment
deviations in this group and thereby offer an explanation for
the similar prevalence compared with patients treated with
fibrinolysis. However, this was not evident in our study,
because other risk factors, such as age, anterior wall AMI,
and impaired LV systolic function, were equally distributed
between the 2 treatment groups. In accordance with this
finding, long-term mortality rates were also similar in the 2
treatment groups, although there was a trend toward lower
mortality in PCI-treated patients. This is most likely attributable to the fact that the selection of either fibrinolysis or
PCI was determined by randomization, whereas in everyday
clinical practice, the risk profile of patients selected for PCI
may differ from that of patients selected for fibrinolysis, and
accordingly, so may the subsequent prevalence of several
risk factors, including ST-segment deviations.
The prevalence of ST-segment deviations in patients
recovering from recent AMIs in previous studies ranges
from 11% to 46%.6,10,11,15,17,21 These inconsistent findings
are probably caused by a variety of reasons, including differences in study design, differences in treatment, heterogenous study groups, and differences in the applied definitions of ST-segment deviation and the timing of Holter
monitoring. Bjerregaard et al22 showed that the outcome of
Holter analysis for ischemic events varied from 4.5% to
24% in a sample of patients with acute ischemic syndrome,
depending on how a significant episode of STd was defined.
Furthermore, in previous studies, the timing of Holter recording has varied from 1 day to 3 weeks after index
infarction.10,12,13,17 Studies assessing the ideal time for
Holter monitoring after myocardial infarction are therefore
warranted.
In the present study, there was a significant association
between STd on Holter monitoring and positive exercise
test results in the 2 treatment groups. ST-segment deviation
during exercise testing is generally accepted for prognostic
assessment of patients after AMI, indicating residual myocardial ischemia. We therefore suggest that most episodes of
STd observed in the present study were ischemic episodes.
In a previously reported DANAMI-2 substudy, Valeur et
al23 showed that patients randomized to fibrinolysis developed STd to a greater extent during exercise testing com-
Coronary Artery Disease/ST-Segment Deviation After Coronary Revascularization
pared with patients randomized to PCI. In contrast, we
found the same prevalence of STd in Holter monitoring in
the 2 treatment groups. It is possible that fibrinolysis-treated
patients had less complete reperfusion of the infarct-related
area compared with those treated with PCI, explaining the
greater extent of STd during maximal exercise testing. We
also found a significant association between STe and reduced LV systolic function in the 2 treatment groups,
whereas no association was found between STe and positive
exercise test results. This suggests that STe, rather than
reflecting transient myocardial ischemia, is a marker of
significant LV remodeling, a hypothesis supported by the
fact that STe was also significantly more frequent in patients
with anterior wall myocardial infarction.
The timing of Holter recording may in some part have
affected our results. In the present study, recordings were
performed on the day of discharge. It is possible that the
degree of reperfusion and ventricular myocardial remodeling became less frequent during the months after STe myocardial infarction, affecting incident transient ST-segment
deviations. It is therefore possible that the prevalence of
ST-segment deviations would differ in the 2 treatment
groups if Holter monitoring had been performed at a later
stage. However, Mickley et al6 found no differences in the
prevalence of STd in similar patients examined at discharge
and 6 and 12 months after infarction. Another major limitation of the study relates to the general risk profile of the
investigated cohort. Because of the exclusion criteria of the
DANAMI-2 trial, and the fact that Holter monitoring was
performed at discharge, excluding all patients dying before
the Holter recording was possible, our cohort was at a lower
risk. It is therefore possible that the prevalence and prognostic implications of the investigated variables differ from
what would be found in an unselected STe myocardial
infarction population.
Acknowledgment: We gratefully acknowledge all the
DANAMI-2 investigators and study nurses for their efforts
throughout the DANAMI-2 study. We would also like to
thank the members of the DANAMI-2 steering committee
for their support.
1. Andersen HR, Nielsen TT, Rasmussen K, Thuesen L, Kelbaek H,
Thayssen P, Abildgaard U, Pedersen F, Madsen JK, Grande P, et al. A
comparison of coronary angioplasty with fibrinolytic therapy in acute
myocardial infarction. N Engl J Med 2003;349:733–742.
2. Andersen HR, Nielsen TT, Vesterlund T, Grande P, Abildgaard U,
Thayssen P, Pedersen F, Mortensen LS. Danish multicenter randomized study on fibrinolytic therapy versus acute coronary angioplasty in
acute myocardial infarction: rationale and design of the Danish Trial in
Acute Myocardial Infarction-2 (DANAMI-2). Am Heart J 2003;146:
234 –241.
3. Gislason GH, Gadsboll N, Quinones MA, Kober L, Seibaek M, Burchardt H, Torp-Pedersen C. The reliability of echocardiographic left
ventricular wall motion index to identify high-risk patients for multicenter studies. Echocardiography 2006;23:1– 6.
4. Berning J, Rokkedal NJ, Launbjerg J, Fogh J, Mickley H, Andersen
PE. Rapid estimation of left ventricular ejection fraction in acute
myocardial infarction by echocardiographic wall motion analysis. Cardiology 1992;80:257–266.
943
5. TIMI Study Group. The Thrombolysis In Myocardial Infarction
(TIMI) trial. Phase I findings. N Engl J Med 1985;312:932–936.
6. Mickley H, Pless P, Nielsen JR, Moller M. Changing circadian variation of transient myocardial ischemia during the first year after a first
acute myocardial infarction. Am J Cardiol 1992;70:1117–1122.
7. Egstrup K. Transient myocardial ischaemia during ambulatory monitoring out of hospital in patients with chronic stable angina pectoris.
Acta Med Scand 1988;224:311–318.
8. Currie P, Ashby D, Saltissi S. Prognostic significance of transient
myocardial ischemia on ambulatory monitoring after acute myocardial
infarction. Am J Cardiol 1993;71:773–777.
9. Gill JB, Cairns JA, Roberts RS, Costantini L, Sealey BJ, Fallen EF,
Tomlinson CW, Gent M. Prognostic importance of myocardial ischemia detected by ambulatory monitoring early after acute myocardial
infarction. N Engl J Med 1996;334:65–70.
10. Gottlieb SO, Gottlieb SH, Achuff SC, Baumgardner R, Mellits ED,
Weisfeldt ML, Gerstenblith G. Silent ischemia on Holter monitoring
predicts mortality in high-risk postinfarction patients. JAMA 1988;259:
1030 –1035.
11. Tzivoni D, Gavish A, Zin D, Gottlieb S, Moriel M, Keren A, Banai S,
Stern S. Prognostic significance of ischemic episodes in patients with
previous myocardial infarction. Am J Cardiol 1988;62:661– 664.
12. Jereczek M, Andresen D, Schroder J, Voller H, Bruggemann T, Deutschmann C, Schroder R. Prognostic value of ischemia during Holter
monitoring and exercise testing after acute myocardial infarction. Am J
Cardiol 1993;72:8 –13.
13. Ouyang P, Chandra NC, Gottlieb SO. Frequency and importance of
silent myocardial ischemia identified with ambulatory electrocardiographic monitoring in the early in-hospital period after acute myocardial infarction. Am J Cardiol 1990;65:267–270.
14. Petretta M, Bonaduce D, Bianchi V, Vitagliano G, Conforti G, Rotondi
F, Themistoclakis S, Morgano G. Characterization and prognostic
significance of silent myocardial ischemia on predischarge electrocardiographic monitoring in unselected patients with myocardial infarction. Am J Cardiol 1992;69:579 –583.
15. Mickley H, Nielsen JR, Berning J, Junker A, Moller M. Prognostic
significance of transient myocardial ischaemia after first acute myocardial infarction: five year follow up study. Br Heart J 1995;73:320 –
326.
16. Moss AJ, Goldstein RE, Hall WJ, Bigger JT Jr, Fleiss JL, Greenberg
H, Bodenheimer M, Krone RJ, Marcus FI, Wackers FJ, et al. Detection
and significance of myocardial ischemia in stable patients after recovery from an acute coronary event. Multicenter Myocardial Ischemia
Research Group. JAMA 1993;269:2379 –2385.
17. Langer A, Minkowitz J, Dorian P, Casella L, Harris L, Morgan CD,
Armstrong PW. Pathophysiology and prognostic significance of
Holter-detected ST segment depression after myocardial infarction.
The Tissue Plasminogen Activator: Toronto (TPAT) Study Group.
J Am Coll Cardiol 1992;20:1313–1317.
18. Mickley H, Nielsen JR, Berning J, Junker A, Moller M. Characteristics
and prognostic importance of ST-segment elevation on Holter monitoring early after acute myocardial infarction. Am J Cardiol 1995;76:
537–542.
19. Kent DM, Schmid CH, Lau J, Selker HP. Is primary angioplasty for
some as good as primary angioplasty for all? J Gen Intern Med
2002;17:887– 894.
20. Thune JJ, Hoefsten DE, Lindholm MG, Mortensen LS, Andersen HR,
Nielsen TT, Kober L, Kelbaek H. Simple risk stratification at admission to identify patients with reduced mortality from primary angioplasty. Circulation 2005;112:2017–2021.
21. Genovesi EA, Paperini L, Baldini U, Raugi M, Digiorgio A, Magini G.
Holter-detected myocardial ischemia. Impact for prognosis and decision making after acute myocardial infarction. Minerva Cardioangiol
2002;50:117–123.
22. Bjerregaard P, El Shafei A, Kotar SL, Labovitz AJ. ST segment
analysis by Holter Monitoring: methodological considerations. Ann
Noninvasive Electrocardiol 2003;8:200 –207.
23. Valeur N, Clemmensen P, Saunamaki K, Grande P. The prognostic
value of pre-discharge exercise testing after myocardial infarction
treated with either primary PCI or fibrinolysis: a DANAMI-2 substudy. Eur Heart J 2005;26:119 –127.
Rapid Triage and Transport of Patients With ST-Elevation
Myocardial Infarction for Percutaneous Coronary Intervention
in a Rural Health System
James C. Blankenship, MDa,*, Thomas A. Haldis, DOc, G. Craig Wood, MSb,
Kimberly A. Skelding, MDa, Thomas Scott, MDa, and Francis J. Menapace, MDa
This study was conducted to evaluate door-to-treatment times before and after the implementation of a rapid triage and transfer system for patients with ST-elevation myocardial infarction
transferred from community hospitals to a rural angioplasty center for primary percutaneous
coronary intervention (PCI). The system was developed in late 2004 and implemented at a rural
percutaneous coronary intervention center in early 2005. Helicopter transport was available for
97% of requests for transfer from community hospitals. All patients with ST-elevation myocardial infarction transferred during 2004 and 2005 (n ⴝ 226) were evaluated with respect to
presentation and treatment times. Time from community hospital presentation to wire crossing
decreased during the study from 205 to 105 minutes (p ⴝ 0.0001). One fourth of patients were
treated <90 minutes after presentation, and 2/3 were treated in <120 minutes. In conclusion,
the implementation of a rapid triage, transfer, and treatment protocol can achieve a significant
shortening of presentation-to-treatment times. Efficient community hospitals working with an
efficient angioplasty center can achieve presentation–to–wire crossing times of <90 minutes for
some patients. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:944 –948)
Several trials have documented the benefits of transfer for
percutaneous coronary intervention (PCI) compared with
on-site lytic therapy.1–3 However, in these trials, mostly
outside the United States, transfer for PCI was very efficient. Data from the National Registry of Myocardial Infarction suggest that transfer for PCI in the United States is
often much slower.4 Only 15% of these patients had doorto-balloon times of ⬍120 minutes, and rural teaching hospital door-to-balloon times for transferred patients were 73
minutes slower than those of their urban counterparts.5
Although efficient triage and treatment of patients with
ST-elevation myocardial infarction (STEMI) presenting to
facilities with PCI capability is possible,6 – 8 it has not been
demonstrated that patients with STEMI can be effectively
triaged and transferred for STEMI PCI in the United States,
particularly in rural areas.9 Because the outcome of STEMI
treated with PCI depends in part on how fast PCI is delivered,10,11 developing systems for the rapid transfer of patients with STEMI presenting to community hospitals is
important. Our facility instituted a rapid transfer protocol on
January 1, 2005. We report here the characteristics and
results of patients with STEMI transferred to our medical
center before and after the initiation of this protocol.
a
Department of Cardiology and bCenter for Health Research, Geisinger
Medical Center, Danville, Pennsylvania; and cMeritCare Medical Center,
Fargo, North Dakota. Manuscript received February 16, 2007; revised
manuscript received and accepted April 17, 2007.
This study was supported by the Geisinger Interventional Cardiology
Research Fund, Geisinger Medical Center, Danville, Pennsylvania.
*Corresponding author: Tel: 570-271-8067; fax: 570-271-8056.
E-mail address: [email protected] (J.C. Blankenship).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.031
Methods
The PCI center in this study is a 437-bed tertiary care
hospital in rural central Pennsylvania. It is a teaching affiliate of a Philadelphia medical college and has an independent cardiology fellowship program. It serves 37 counties
with a population of 2.4 million. It has provided primary
PCI services since 1985 and has treated patients with
STEMI exclusively with PCI since 1995. The PCI center is
part of a health system that also owns and operates 4
helicopters based at 4 sites in central Pennsylvania. The
helicopter communications center is located in the emergency department of the PCI center. During the study period, the helicopter was able to respond to approximately
97% of all requests for transports.
All patients transferred to our PCI center from January 1,
2004, to December 31, 2005, for PCI within 12 hours of the
onset of STEMI were considered for inclusion in this study,
including those with cardiogenic shock, hemodynamic instability, or cardiac arrest. We excluded patients transferred
for failed thrombolytic therapy and patients with contraindications to PCI, including patient refusal, allergy to contrast dye, major co-morbidities precluding PCI, and warfarin anticoagulation. Patients whose initial presentation was
to the PCI center were also excluded. The study was approved by the institutional investigational review board.
During 2004, patients were treated according to standard
practice. No door-to-balloon time quality improvement projects had been previously attempted. Prehospital electrocardiography was not available to emergency medical services
in rural central Pennsylvania. Patients with STEMI presenting to community hospitals were evaluated there by emergency physicians, primary care physicians, or cardiologists.
Most community hospitals treated patients with STEMI
www.AJConline.org
Coronary Artery Disease/Rapid Transport for Direct PCI
with aspirin 325 mg chewed, intravenous nitroglycerin, and
heparin bolus and infusion. Most community hospitals preferentially transferred patients with STEMI for PCI; a minority received on-site thrombolytic therapy. Commonly,
the transfer of patients with STEMI was arranged by calling
the PCI center emergency physician or the general cardiologist on call. The PCI center emergency physician screened
the patient for appropriateness for PCI, checked cardiac
intensive care unit bed availability, and then requested helicopter transfer. When bad weather prevented helicopter
transport, transfer was undertaken by local ambulance services. On arrival at the PCI center, the patient was taken to
the emergency department, where the emergency physician
and the general cardiologist on call assessed the patient,
checked laboratory results, and consulted the interventional
cardiologist. If appropriate, the interventional cardiologist
directed the paging operator to group-page the catheterization team. All members of the team were expected to be in
the laboratory ⬍30 minutes after first contact.
In mid-2004, the PCI center emergency department instituted a streamlined helicopter dispatch protocol. It was
designed so that 1 call from a community hospital emergency department to the PCI center emergency physicians
on a dedicated phone line prompted immediate dispatch of
the helicopter. Independent of this, over the last half of
2004, a rapid triage, transfer, and treatment protocol for
patients with STEMI was developed. Some of its innovations were informally incorporated into practice (through
rapid cycle tests of change) by individual physicians at the
angioplasty center over that time. The new protocol, implemented on January 1, 2005, encouraged referring physicians
to directly call the PCI center emergency department. After
reviewing a faxed copy of the electrocardiogram and
screening the patient with a 9-question checklist, the emergency department physician dispatched a helicopter and
alerted the interventional cardiologist on call, who then
called the operator to group-page the catheterization laboratory team. All catheterization team members were expected in the laboratory ⬍30 minutes after first notification.
In most cases, the interventional cardiologist and catheterization team met the patient and helicopter crew at the
helipad elevator doors and escorted them directly to the
catheterization laboratory. After obtaining a brief history,
examination, and informed consent, PCI was performed.
Letters describing the program were mailed to community hospital emergency department directors in the first
week of 2005. In the first half of 2005, a PCI center interventional cardiologist (JCB) visited the 8 hospitals that
provided most referrals, presenting the program through a
didactic conference and written materials. Strategies for
rapid referral diagnosis, triage, and referral were discussed,
and emergency department staff members were encouraged
to streamline the diagnosis and triage of patients with
STEMI with strategies including the following: perform
electrocardiography ⬍10 minutes after presentation for all
patients with chest pain, call the PCI center emergency
physician immediately on a dedicated hotline before performing chest x-rays, and make treatment and triage decisions before contacting primary care physicians or local
cardiologists. A pharmacologic regimen was recommended
that included aspirin 325 mg chewed, metoprolol 5 mg
945
intravenously times 3 when not contraindicated, heparin 70
U/kg bolus without infusion, sublingual nitroglycerin or
optional topical nitropaste without routine intravenous infusion, and optional clopidogrel 600 mg orally. Eliminating
intravenous infusions of heparin and nitroglycerin were
suggested as time-saving measures. Community hospitals
adopted these recommendations at their own discretion.
PCI was performed at the PCI center by 3 experienced
(⬎200 PCIs/year) staff interventionists. PCI was typically
performed with 6Fr guide catheters, heparin anticoagulation, predilatation before stenting, and vascular closure with
an intra-arterial collagen plug device. Noninfarct lesions
were not treated during the initial PCI. Details of diagnostic
catheterization and PCI (e.g., imaging of the noninfarct
vessel before vs after PCI, postdilatation, glycoprotein IIb/
IIIa inhibitor treatment) varied among the interventionists.
After PCI, patients were admitted to the cardiac intensive
care unit under the care of the general cardiology service.
Data on all patients who underwent PCI were collected
and reported to the American College of Cardiology National Cardiovascular Data Registry. Registry forms were
completed immediately after PCI by the operator. Hospital
discharge information was obtained at discharge by an interventional physician’s assistant.
Transfer times for the first 6 months of 2004 were collected retrospectively using logs from the helicopter service,
cardiac catheterization laboratories, and patient charts. Precise times were available for all key data points for all
patients. Starting July 1, 2004, transfer times were collected
immediately after PCI on standardized forms. All data for
the study period were entered into data fields that were
added, for study patients, to the National Cardiovascular
Data Registry.
The time of onset of symptoms was that reported by the
patient, as recorded in the chart or obtained directly from the
patient. The time of initial presentation was taken as the first
time found on any documentation from the initial hospital.
The time of first electrocardiography was taken directly
from the first electrocardiogram performed at the referring
hospital. Dispatch time was the time the helicopter service
was called with a request for transfer. Helicopter touchdown
and liftoff times from the referral hospital and PCI center
were taken from helicopter flight logs as reported by the
pilot to the helicopter communications center. The catheterization laboratory arrival time was defined as the time
that the patient entered the catheterization laboratory. Catheterization laboratory arrival and wire crossing times were
taken from the catheterization laboratory procedure logs as
recorded in real time. The rationale for using wire crossing
time is discussed further later but centered on the fact that
the angioplasty balloon inflation was often not the first
reperfusion modality used.
Demographic and procedural data are reported as percentage and number of patients with each characteristic.
Times are reported as medians because all time distributions
were skewed and were compared across time intervals using
the nonparametric Kruskal-Wallis test for independent samples. Helicopter transfer patients were used for the calculation of times from electrocardiography to helicopter dispatch; all patients were used for all other time analyses.
When comparing categorical and continuous data, Pear-
946
The American Journal of Cardiology (www.AJConline.org)
Table 1
Baseline demographic characteristics (n ⫽ 226)
Variable
Men
Age (mean ⫾ SD) (yrs)
Weight (mean ⫾ SD) (kg)
Hypertension
Diabetes mellitus
Smoker
Hypercholesterolemia*
Previous heart failure
Previous myocardial infarction
Previous percutaneous coronary intervention
Previous coronary bypass grafting
Value
80%
58 ⫾ 12
89 ⫾ 21
50%
17%
64%
54%
7%
6%
7%
3%
* Defined as total cholesterol ⬎200 mg/dl, low-density lipoprotein ⬎130
mg/dl, or high-density lipoprotein ⬍30 mg/dl.
Table 2
Procedural characteristics (n ⫽ 226)
Variable
Preintervention
Culprit coronary artery
Left anterior descending
Right
Left circumflex/ramus
Initial stenosis (mean ⫾ SD) (%)
Postintervention
Ejection fraction on left ventriculography (mean ⫾ SD) (%)
Final Thrombolysis In Myocardial Infarction flow
grade 2 or 3
Final stenosis (mean ⫾ SD) (%)
Value
34%
50%
16%
95 ⫾ 10
45 ⫾ 13
100%
3 ⫾ 14
son’s chi-square statistic or Fisher’s exact test and 2-sample
Student’s t tests, were used, respectively. All tests were 2
sided, and p values ⬍0.05 were considered significant. SAS
version 9.1 (SAS Institute Inc., Cary, North Carolina) was
used for all statistical testing.
Results
During the 2-year study period, 226 patients with STEMI
were transferred from 14 community hospitals and underwent PCI. Helicopter transport was used for 197 patients
(87%). Transfer by local ambulance services was used for
29 patients, in 6 cases (3%) because of bad weather and in
23 cases (10%) because community hospital emergency
physicians judged that ambulance transport would be
quicker. During this same period, 65 patients with STEMI
presenting directly to the PCI center emergency department
underwent direct PCI, and 1,462 others underwent nonSTEMI PCI. There were no deaths or significant complications during helicopter transport. Demographic and angiographic characteristics (Tables 1 and 2) did not change over
time during the study (data not shown). There was a trend
toward more cardiogenic shock in patients presenting in
2005 (15%) compared with 2004 (6%) (p ⫽ 0.051).
Time periods for various milestones in the triage, transfer, and treatment process are listed in Table 3. Times from
community hospital arrival to electrocardiography varied
from 6 to 8 minutes and did not change significantly during
the study period. Times from electrocardiography to heli-
copter dispatch decreased from 34.5 to 16 minutes during
the study period (p ⫽ 0.0004). Times from helicopter dispatch to arrival at the PCI center decreased from 56 to 45
minutes (p ⫽ 0.002). Times from arrival at the PCI center to
wire crossing decreased from 91 to 29 minutes (p ⫽
0.0001).
Overall time from community hospital arrival to wire
crossing decreased over the study period from 205 to 105
minutes (p ⫽ 0.0001). Of the decrease in time, 37% (36
minutes) was due to more efficient diagnosis and retrieval of
patients with STEMI, and 63% (62 minutes) was due to
more efficient treatment at the PCI center. The percentage of
patients with door-to-wire times ⬍90 minutes increased
from 0% to 24% (p ⫽ 0.0001), and the percentage with
door-to-wire times ⬍120 minutes increased from 2% to
67% (p ⫽ 0.0001).
Table 4 includes data from the 8 individual community
hospitals that referred ⱖ14 patients to the PCI center during
the study period. (The other 6 referring hospitals together
transferred only 22 patients [range 1 to 7] during the 2-year
study period.) Time from presentation to wire crossing
decreased from 2004 to 2005 for all 8 hospitals. In 2005, 2
hospitals achieved median door-to-wire times of 91 and 92
minutes, and 6 of the 8 hospitals achieved median times
⬍120 minutes. Table 4 demonstrates a wide variation in
median door-to-wire times, from 91 to 145 minutes. Outcomes during initial hospitalization were favorable, with a
mortality rate of 3.5% (Table 5).
Discussion
The most important finding of this study is that the rapid
diagnosis, triage, transfer, and treatment of patients with
STEMI is possible in a real-world rural American setting.
The median door-to-wire time at the end of the study
period for patients transferred for PCI (105 minutes)
approached the median door-to-balloon time (100 minutes) for patients who underwent on-site PCI at 365
hospitals during a similar time period, as reported by
Bradley et al.12 One fourth of patients were treated within
the guideline goal of 90 minutes recommended by the
American College of Cardiology and the American Heart
Association, and 2/3 were treated in ⬍120 minutes.
Current guidelines suggest that patients with STEMI
treated with PCI, including those requiring transfer, should
receive this therapy ⬍90 minutes after first medical contact.13 This may be more feasible in Europe1,3 than the
United States. Consequently, the appropriateness of primary
PCI for patients presenting to non-PCI facilities in the
United States has been questioned.14 –17 Our protocol resulted in door-to-wire times ⬎1 hour shorter than the 171minute door-to-balloon times reported by Nallamothu et al4
from the National Registry of Myocardial Infarction, with 2
hospitals only 1 to 2 minutes away from meeting the 90minute guideline goal for half of their patients. We conclude
that the 90-minute goal is achievable for some transferred
patients, even in a rural setting.
A second important finding of this study is that dramatic
improvements in time from initial presentation to PCI for
transferred patients can be achieved simply, using minimal
resources. The protocols we implemented included many
but certainly not all policies and practices advocated by
Coronary Artery Disease/Rapid Transport for Direct PCI
947
Table 3
Treatment times
Variable
Onset of chest pain to arrival at community hospital
Arrival at community hospital to arrival at angioplasty center
Arrival at community hospital to electrocardiography
Electrocardiography to helicopter dispatch
Helicopter dispatch to arrival at angioplasty center
Arrive at angioplasty center to wire crossing
Arrival at catheterization laboratory to wire crossing
Total time (arrival at community hospital to wire crossing)
Total time ⬍90 min
Total time ⬍120 min
January to June
2004
(n ⫽ 59)
July to December
2004
(n ⫽ 49)
85 (52, 142)
112 (88, 140)
8.5 (3, 16)
34.5 (17, 56)
56.5 (47, 69)
91 (74, 120)
40 (31, 50)
205 (179, 292)
0%
2%
67 (45, 126)
80 (66, 101)
6 (3, 12)
16 (10, 22)
49.5 (41, 61)
70 (50, 91)
29 (23, 42)
152 (124, 194)
2%
22%
January to June
2005
(n ⫽ 69)
99 (60, 165)
86 (66, 108)
6 (3, 10)
21 (32, 58)
55.5(43, 67)
35 (26, 48)
24 (16, 31)
120 (98, 165)
12%
51%
July to December
2005
(n ⫽ 49)
p Value
75 (57, 158)
76 (55, 104)
8 (5, 14)
16 (9, 25)
45 (35, 58)
29 (24, 42)
15 (11, 21)
105 (92, 129)
24%
67%
0.32*
0.0001*
0.45*
0.0004*
0.002*
0.0001*
0.0001*
0.0001*
0.0001†
0.0001†
All times are medians with 25th and 75th percentiles.
* Kruskall-Wallis test.
†
Chi-square test.
Table 4
Median treatment times for individual community hospitals
Hospital (flight time)
A (10 min)
B (4 min)
C (7 min)
D (17 min)
E (32 min)
F (14 min)
G (6 min)
H (6 min)
Time from Arrival at Community Hospital to
Helicopter Dispatch
Time from Arrival at Community Hospital to Wire
Crossing
2004
2005
p Value
2004
2005
p Value
21 (17, 40)
42 (35, 46)
34 (23, 49)
33 (25, 138)
23 (20, 54)
37 (17, 77)
26 (10, 63)
80 (39, 93)
19.5 (14, 26)
29 (19.5, 55)
18 (13, 24)
25 (12, 26)
40 (23, 64)
32 (28, 45)
23 (15, 39)
26 (16, 31)
0.72*
0.34*
0.0015*
0.18*
0.80*
0.87*
0.81*
0.041*
183 (146, 214)
187 (166, 258)
173 (123, 201)
219 (197, 305)
195 (130, 214)
194 (185, 310)
159 (126, 180)
215 (187, 244)
102 (97, 149)
105 (74, 119)
91 (76, 110)
118 (104, 230)
141 (111, 177)
145 (128, 154)
112 (103, 204)
92 (89, 98)
0.0087*
0.009*
0.0001*
0.24*
0.0096*
0.027*
0.097*
0.027*
All times are medians with 25th and 75th percentiles.
* Kruskall-Wallis test.
Table 5
Outcomes
Variable
At hospital discharge
Death, n (%)
Urgent revascularization, n (%)
Death or urgent revascularization, n (%)
Length of stay (d), median
(25th percentile, 75th percentile)
2004
2005
p
(n ⫽ 108) (n ⫽ 118) Value
2 (2%)
1 (1%)
3 (3%)
3 (2,4)
6 (5%)
3 (3%)
8 (8%)
3 (2,4)
0.28*
0.62*
0.14*
0.46†
* Fisher’s exact test.
†
Kruskall-Wallis test.
Henry et al18 and Bradley et al,12 which were published after
our study was completed.
Our study documented heterogeneity in door-to-wire times
among the frequently referring community hospitals. This was
attributable mostly to practices at the community hospitals;
flight time was not significantly associated with overall doorto-wire time. Formal surveys of community hospital practices
were not undertaken, but informal communications revealed
dramatic differences in the extent to which community hospital
emergency departments embraced the new triage, treatment,
and transfer protocols. Improvements in door-to-treatment
times among hospitals varied widely, with 1 hospital decreasing its time by 123 minutes to go from second slowest to
second fastest among referring hospitals. In contrast, the fastest
hospital at the outset reduced its time by only 47 minutes,
ending up fifth fastest among referring hospitals. This demonstrates that the most efficient hospitals in our region are significantly faster than others and that they can improve doorto-treatment times for their patients to a greater extent than
others. Raising all hospitals to the levels of efficiency of the
fastest community hospitals would have resulted, in our study,
in almost half of transferred patients meeting the 90-minute
guideline goal.
We used wire crossing time instead of balloon inflation
time for several reasons. First, balloon inflation is an arbitrary end point that often does not reflect time of reperfusion. Many arteries reperfuse spontaneously or with wire
crossing before balloon inflation, and often the first balloon
inflation does not provide reperfusion. Second, many procedures began with rheolytic or aspiration thrombectomy,
significantly delaying first balloon inflation. Third, direct
stenting was performed when possible, often after dottering
a lesion without balloon inflation, and delays in choosing
948
The American Journal of Cardiology (www.AJConline.org)
and preparing the stent would also introduce bias. Fourth,
when a balloon was used as the initial reperfusion device,
wire-to-balloon times were consistently only 2 to 3 minutes,
representing a small increase compared with door-to-wire
times. Fifth, using “balloon times” as a guideline metric
may encourage gaming strategies, such as inflating the balloon in the aorta before it reaches the target lesion.19
There were several limitations to this study. The number of
patients was small. The generalizability of the results may be
limited by 2 unique features of the PCI center: the catheterization laboratory is staffed by a single group of experienced,
employed interventionists, and the helicopter service is owned
by the same health system that operates the PCI center. Prehospital electrocardiography was not available to local emergency medical center units, and its use may further improve
transfer efficiency.20 –23 Standardized community hospital order sets and pretransfer treatment strategies were sporadically
implemented. If universally implemented, these might have
provided further improvements in efficiency. Patients in this
study were not given facilitation therapy; some type of adjuvant medical therapy may in the future prove beneficial for
patients with STEMI.24 Many patients did not receive reperfusion ⬍90 minutes after presentation, and for some it was
delayed beyond 120 minutes. When such delays are foreseeable, a strategy of immediate medical thrombolysis may be
preferable, as recommended by the American College of Cardiology guidelines.13, 17
8.
9.
10.
11.
12.
13.
14.
15.
Acknowledgment: The past and current success of our
program for rapid triage and transfer for STEMI is due
entirely to the dedicated efforts of emergency physicians
and nurses at our center and at community hospitals, Lifeflight personnel, and the staff members and physicians of
our catheterization laboratories.
1. Widimsky P, Budesinsky T, Vorac D, Groch L, Zelizko M, Aschermann M, Branny M, St’asek J, Formanek P. Long distance transport
for primary angioplasty versus immediate thrombolysis in acute myocardial infarction. Eur Heart J 2003;24:94 –104.
2. Grines CL, Westerhausen DR, Grines LL, Hanlon JT, Logemann TL,
Niemela M, Weaver WD, Balestrini C. A randomized trial of transfer for
primary angioplasty versus on-site thrombolysis in patients with high-risk
myocardial infarction. J Am Coll Cardiol 2002;39:1713–1722.
3. Andersen HR, Nielsen TT, Rasmussen K, Thuesen L, Kelbaek H,
Thayssen P, Abildgaard U, Pedersen F, Madsen JK, Grande P, et al. A
comparison of coronary angioplasty with fibrinolytic therapy in acute
myocardial infarction. N Engl J Med 2003;349:733–742.
4. Nallamothu BK, Bates ER, Herrin ER, Wang Y, Bradley EH, Krumholz HM. Times to treatment in transfer patients undergoing primary
percutaneous coronary intervention in the United States. Circulation
2005;111:761–767.
5. Shavelle DM, Rasouli ML, Frederick P, Gibson CM, French WJ.
Outcome in patients transferred for percutaneous coronary intervention
(a National Registry of Myocardial Infarction 2/3/4 analysis). Am J
Cardiol 2005;96:1227–1232.
6. Waters RE, Singh KP, Roe MT, Lotfi M, Sketch MH, Mahaffey KW,
Newby LK, Alexander JH, Harrington RA, Califf RM, Granger CB.
Rationale and strategies for implementing community-based transfer
protocols for primary percutaneous coronary intervention for acute
ST-segment elevation myocardial infarction. J Am Coll Cardiol 2004;
43:2153–2159.
7. Caputo RP, Kalon KL, Stoler RC, Sukin CA, Lopez JJ, Cohen DJ,
Kuntz RE, Berman A, Carrozza JP, Baim DS. Effect of continuous
quality improvement analysis on the delivery of primary percutaneous
16.
17.
18.
19.
20.
21.
22.
23.
24.
transluminal coronary angioplasty for acute myocardial infarction.
Am J Cardiol 1997;79:1159 –1164.
Caputo RP, Kosinski R, Walford G, Giambartolomei A, Grant W,
Reger MJ, Simons A, Esente P. Effect of continuous quality improvement analysis on the delivery of primary percutaneous revascularization for acute myocardial infarction. Catheter Cardiovasc Interv 2005;
64:428 – 433.
Stukel TA, Lucas FL, Wennberg DE. Long-term outcomes of regional variations in intensity of invasive vs medical management of
Medicare patients with acute myocardial infarction. JAMA 2005;
293:1329 –1337.
McNamara RL, Wang Y, Herrin J, Curtis JP, Bradley EH, Magid DJ,
Peterson ED, Blaney M, Frederick PD, Krumholz HM. Effect of
door-to-balloon time on mortality in patients with ST-segment elevation myocardial infarction. J Am Coll Cardiol 2006;47:2180 –2186.
Brodie BR, Hansen C, Stuckey TD, Richter S, VerSteeg DS, Gupta N,
Downey WE, Pulsipher M. Door-to-balloon time with primary percutaneous coronary intervention for acute myocardial infarction impacts
late cardiac mortality in high-risk patients and patients presenting early
after the onset of symptoms. J Am Coll Cardiol 2006;47:289 –295.
Bradley EH, Herrin J, Wang Y, Barton BA, Webster TR, Mattera JA,
Roumanis SA, Curtis JP, Nallamothu BK, Magid DJ, et al. Strategies
for reducing the door-to-balloon time in acute myocardial infarction.
N Engl J Med 2006;355:2308 –2320.
Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand
M, Hochman JS, Krumholz HM, Kushner FG, Lamas GA, et al.
ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: executive summary: a report of the ACC/
AHA Task Force on Practice Guidelines (Committee to Revise the
1999 Guidelines on the Management of Patients with Acute Myocardial Infarction). J Am Coll Cardiol 2004;44:671–719.
Williams DO. Treatment delayed is treatment denied. Circulation 2004;
109:1806 –1808.
Hermann HC. Transfer for primary angioplasty: the importance of
time. Circulation 2005;111:718 –720.
Wharton TP. Should patients with acute myocardial infarction be transferred to a tertiary center for primary angioplasty or receive it at qualified
hospitals in the community? Circulation 2005;112:3509 –3534.
Larson DM, Sharkey SW, Unger BT, Henry TD. Implementation of
acute myocardial infarction guidelines in community hospitals. Acad
Emerg Med 2005;12:522–528.
Henry TD, Atkins JM, Cunningham MS, Francis GS, Groh WJ, Hong
RA, Kern KB, Larson DM, Ohman EM, Ornato JP, et al. ST-segment
elevation myocardial infarction: recommendations on triage of patients
to heart attack centers: is it time for a national policy for the treatment
of ST-segment elevation myocardial infarction? J Am Coll Cardiol
2006;47:1339 –1345.
Dehmer GJ, Powell W. Pay for quality—what every interventional
cardiologist needs to know: part II. Catheter Cardiovasc Interv 2006;
68:338 –341.
Bonnefoy E, Lapostolle F, Leizorovicz A, Steg G, McFadden EP,
Dubien PY, Cattan S, Boullenger E, Machecourt J, Lacroute JM, et al;
Comparison of Angioplasty and Prehospital Thromboysis in Acute
Myocardial Infarction Study Group. Primary angioplasty versus prehospital fibrinolysis in acute myocardial infarction: a randomized
study. Lancet 2002;360:825– 829.
van de Loo A, Saurbier B, Kalbhenn J, Koberne F, Zehender M. Primary
percutaneous coronary intervention in acute myocardial infarction: direct
transportation to catheterization laboratory by emergency teams reduces
door-to-balloon time. Clin Cardiol 2005;29:112–116.
Garvey JL, MacLeod BA, Sopko G, Hand MM. Pre-hospital 12-lead
electrocardiography programs. A call for implementation by emergency medical services systems providing advanced life support. J Am
Coll Cardiol 2006;47:485– 491.
Sekulic M, Hassunizadeh B, McGraw S, David S. Feasibility of early
emergency room notification to improve door-to-balloon times for
patients with acute ST elevation myocardial infarction. Catheter Cardiovasc Interv 2005;66:316 –319.
Gersh BJ, Stone GW, White HD, Holmes DR Jr. Pharmacological
facilitation of primary percutaneous coronary intervention for acute
myocardial infarction: is the slope of the curve the shape of the future?
JAMA 2005;293:973–986.
Safety of Drug-Eluting Stents in the Coronary Artery in
ST-Elevation Myocardial Infarction at a Single High-Volume
Medical Center
Rahul Bose, MD, Gaurav Gupta, MD, Paul A. Grayburn, MD, Emily A. Laible, BSN,
Mi Jung Kang, MS, and James W. Choi, MD*
Several trials have shown the effectiveness of drug-eluting stents (DES) in reducing
restenosis. Acute ST-elevation myocardial infarction (STEMI) has been an exclusion criterion in most trials evaluating the safety and efficacy of DES. There is recent randomized
trial data evaluating the use and safety of DES for acute myocardial infarction. However,
there is a need for “real world” data on the efficacy and safety of DES in STEMI.
A single-center retrospective analysis was performed on 188 consecutive patients with
STEMI treated with primary or rescue coronary angioplasty between March 2004 and July
2005. The study consisted of 3 groups: 115 patients treated with paclitaxel-eluting stents,
55 with sirolimus-eluting stents, and 18 with bare metal stents. Outcomes were assessed
from 12 to 28 months (mean 20, median 19) for major adverse cardiac events (MACEs)
including myocardial infarction, in-stent thrombosis, clinical restenosis, and death. There
were 4 in-stent thromboses in the paclitaxel group (3.4%) and 2 in-stent thromboses in the
sirolimus group (3.6%). The thromboses ranged from acute (within 24 hours) to as late as
8 months. Clinical restenosis occurred in 4 patients (3.4%) in the paclitaxel group and in
2 patients (3.6%) in the sirolimus. None of the 18 patients with bare metal stents had
thrombosis or clinical restenosis. There were 7 total deaths, all related to complications
from the index STEMI: 1 in the bare metal group, 1 in the sirolimus group, and 5 in the
paclitaxel group. The postdischarge MACE rate was 7% with no deaths. In conclusion, the
use of DES in acute STEMI is associated with a low postdischarge MACE rate and a
3.5% in-stent thrombosis rate, which is similar to reported rates in earlier randomized
trials. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:949 –952)
There are only a few studies examining the safety of drugeluting stents (DES) used in the setting of an acute STelevation myocardial infarction (STEMI), and some suggestion that the use of DES in STEMI may be associated with
a higher risk of stent thrombosis.1–3 Therefore, the aim of
this study is to report on the “real world” experience regarding the safety of DES used in the setting of an acute
STEMI.
Methods
A single center, retrospective analysis was performed on
188 consecutive patients admitted from March 2004 to July
2005 to Baylor University Medical Center with STEMI and
treated with primary or rescue angioplasty within 24 hours
of onset of symptoms. Baylor University Medical Center
consists of 3 private practice cardiology groups with a total
of 13 interventional cardiologists. Patients eligible for the
study had to be ⱖ18 years old, have electrocardiogram
documentation with ⱖ1 mm ST elevation in ⱖ2 consecutive leads, and primary or rescue percutaneous coronary
Baylor University Medical Center, Dallas, Texas. Manuscript received
February 27, 2007; revised manuscript received and accepted April 10,
2007.
*Corresponding author: Tel: 214-824-8721; fax: 214-824-4943.
E-mail address: [email protected] (J.W. Choi).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.030
intervention performed within 24 hours of onset of symptoms. Three groups consisted of those patients who were
treated with paclitaxel-eluting stents, sirolimus-eluting
stents, and bare metal stents. The type of stent used and
recommended aspirin dose was at the interventional cardiologist’s discretion. All patients were prescribed aspirin (81
or 325 mg) daily. In addition, clopidogrel (75 mg/day) was
prescribed for a minimum duration of 1, 3, and 6 months,
respectively, for bare metal, sirolimus, and paclitaxel stents.
Using telephone interview, in-hospital and office follow-up chart review, the patients were analyzed for major
adverse cardiac events (MACEs) occurring during the index
hospitalization, and at up to 2 years of follow-up. MACE
was defined as cardiac death, repeat myocardial infarction,
in-stent thrombosis, clinical restenosis, and target lesion
revascularization (clinical restenosis and in-stent thrombosis). Non-STEMI was defined as creatine kinase-MB ⬎3
times the upper limit of normal. Stent thrombosis was defined as positive angiographic documentation or unexplained cardiac death. Myocardial infarctions due to in-stent
thrombosis were defined as the following: acute thrombosis
is an event within 24 hours of stent implantation, subacute
thrombosis is a thrombosis occurring between 24 hours and
30 days, and late thrombosis is a thrombosis occurring ⬎1
month after stent implantation. Demographic data along
with history of cardiac risk factors were also collected.
www.AJConline.org
950
The American Journal of Cardiology (www.AJConline.org)
Table 1
Demographic data
Variable
All Patients
(n ⫽ 188)
Paclitaxel
(n ⫽ 115)
Sirolimus
(n ⫽ 55)
Bare Metal
(n ⫽ 18)
Age (yrs)
Men
Diabetes mellitus
History of hypertension
History of hyperlipidemia
Smoker
59.8 ⫾ 11.8
134 (71%)
39 (21%)
91 (48%)
68 (36%)
87 (46%)
60.1 ⫾ 12.3
82 (71%)
19 (17%)
43 (37%)
37 (32%)
48 (42%)
59.8 ⫾ 11.2
36 (65%)
15 (27%)
36 (65%)
25 (45%)
27 (49%)
58.1 ⫾ 10.9
16 (89%)
5 (28%)
12 (67%)
6 (33%)
12 (67%)
Variable
All Patients
(n ⫽ 188)
Paclitaxel
(n ⫽ 115)
Sirolimus
(n ⫽ 55)
Bare Metal
(n ⫽ 18)
Left anterior descending
Left circumflex
Right coronary artery
Saphenous vein grafts
Stent diameter (mm)
Stent length (mm)
73 (39%)
31 (16%)
81 (43%)
3 (2%)
3.0 ⫾ 0.4
21.8 ⫾ 7.0
47 (41%)
24 (21%)
44 (38%)
—
2.9 ⫾ 0.4
21.3 ⫾ 6.8
21 (38%)
6 (11%)
27 (49%)
1 (2%)
3.1 ⫾ 0.4
22.3 ⫾ 7.4
5 (28%)
1 (5%)
10 (56%)
2 (11%)
3.6 ⫾ 0.5
24.2 ⫾ 7.4
p ⫽ not significant.
Table 2
Procedural data
p ⫽ not significant.
Table 3
Clinical outcome measures
Variable
Acute stent thrombosis
Subacute stent thrombosis
Late stent thrombosis
Death
Myocardial infarction
Clinical restenosis
Target lesion revascularization
Table 4
Drug-eluting stent (DES) clinical outcomes after index hospitalization
Paclitaxel
(n ⫽ 115)
Sirolimus
(n ⫽ 55)
Bare Metal
(n ⫽ 18)
0
3
1
5*
0
4
8
1
1
0
1*
1
2
4
0
0
0
1*
0
0
0
* Related to complications from index myocardial infarction: anoxic
brain injury, refractory cardiogenic shock, and sepsis.
p ⫽ not significant.
Procedural data consisted of stent diameter and length, and
vessel stented.
Results
A total of 188 patients were analyzed. The 3 groups consisted of 115 patients treated with paclitaxel-eluting stents,
55 patients treated with sirolimus-eluting stents, and 18
patients treated with bare metal stents. Table 1 lists the
patients’ demographics. Rescue intervention was performed
in 55 (29%) patients and primary intervention in 133 (71%).
Ninety-three (49%) of the index procedures were performed
by 3 of the 13 interventionalists.
Table 2 lists the procedural data. Only 3 cases involved
saphenous vein grafts (2 to the left anterior descending
artery, 1 to the right coronary artery). All patients received
aspirin and clopidogrel after coronary stenting.
Table 3 lists the clinical outcomes. The follow-up time
ranged from 12 to 28 months (average 19.6, median 19).
All 188 patients completed 1-year clinical follow-up.
Variable
Subacute stent thrombosis
Late stent thrombosis
Death
Myocardial infarction
Clinical restenosis
MACEs
Paclitaxel
(n ⫽ 110)
Sirolimus
(n ⫽ 54)
Total
(n ⫽ 164)
2 (2%)
1 (1%)
0
0
4 (4%)
7 (6%)
1 (2%)
0
0
1 (2%)
2 (4%)
4 (7%)
3 (2%)
1 (1%)
0
1 (1%)
6 (4%)
11 (7%)
p ⫽ not significant.
Eighteen-month follow-up was completed in 139 patients
(74%) and 24 patients (26%) had 2-year follow-up. Of
the 188 patients included in the study, 7 died during their
index hospitalization (5 in the paclitaxel group, 1 in the
sirolimus group, and 1 in the bare metal group). Six of the
7 deaths occurred due to complications from the index
infarction (refractory cardiogenic shock in 3 and anoxic
brain injury from prolonged cardiac arrest in 3). One
death occurred due to sepsis. No follow-up deaths were
observed in the surviving discharged patients. A summary of MACEs among the discharged surviving patients
in the 2 DES groups is outlined in Table 4.
One patient in the sirolimus group also experienced a
non-STEMI within 2 weeks of her index percutaneous intervention. There were no other non-STEMIs in any of the
other groups.
In-stent thrombosis occurred in 6 patients (3.5% of the
total DES group) during the follow-up period, 4 in the
paclitaxel group, and 2 in the sirolimus group. No deaths
occurred as a result of stent thrombosis, and all patients
underwent successful percutaneous target lesion revascularization. Two of the 4 events in the paclitaxel group occurred
Coronary Artery Disease/DES and STEMI
off dual antiplatelet therapy (1 at 2 weeks due to noncompliance and 1 occurred at 8 months when the patient stopped
taking aspirin secondary to a noncardiac illness). Two other
patients in the paclitaxel group experienced in-stent thromboses while on dual antiplatelet therapy (1 on day 2 of
hospitalization and 1 at 10 days). Both patients in the sirolimus group who experienced thromboses were taking their
dual antiplatelet medicines (1 within 24 hours and 1 at 2
weeks). No patients in the bare metal group experienced
thromboses. Three (5%: 1 paclitaxel, 2 sirolimus) of the 55
patients who initially presented for rescue intervention had
stent thrombosis. Three (2%: all paclitaxel) of the 133
patients who presented for primary intervention had stent
thrombosis.
Data were also collected on clinical restenoses. Four
patients in the paclitaxel group underwent target vessel
revascularization due to clinical restenoses. These procedures occurred as early as 3 months and as late as 12
months. Two patients in the sirolimus group underwent
target vessel revascularization secondary to clinical restenoses (1 at 2 months and 1 at 20 months). No patients in the
bare metal group experienced clinical restenosis and may in
part be due to the larger stent diameters used (Table 2).
Three (5%: 2 paclitaxel, 1 sirolimus) of the 55 patients who
initially presented for rescue intervention had clinical restenosis. Three (2%: 2 paclitaxel, 1 sirolimus) of the 133
patients who presented for primary intervention had clinical
restenosis.
Discussion
A Korean prospective study looking at all patients who were
implanted with DES (elective procedures, non-STEMI and
STEMI) showed that primary stenting in acute myocardial
infarction was a predictor for stent thrombosis.4 Another
study attempting to evaluate the incidence and risk factors
for subacute stent thrombosis in patients implanted with
DES excluded patients who had suffered an acute myocardial infarction but did report a higher incidence of stent
thrombosis than reported in previous trials.5 Hofma et al6
compared the efficacy of paclitaxel-eluting stents versus
sirolimus-eluting stents at 30 days and 1 year in patients
with an acute STEMI. They concluded that there was no
difference between the DES with regard to the risk of stent
thrombosis.
Recently, randomized prospective studies have assessed
the safety and efficacy of DES when used in the setting of
an acute myocardial infarction.1–3 Recent data suggest that
the DES may be associated with a higher rate of in-stent
thrombosis. Two studies reported an in-stent thrombosis
rate of 3.1% to 3.7% in paclitaxel, sirolimus, and bare metal
stents, virtually identical rates to that in our study. Our data
were minimal for the bare metal stents and therefore limits
any comparisons between DES and bare metal stents.
The rates of clinically driven restenoses were also similar
between the 2 DES groups (4 of 116 patients, 3.4% for
paclitaxel and 2 of 55 patients, 3.6% for sirolimus). The
efficacy of DES with regard to restenosis has been well
established by previous trials.7–9 Our observed clinical restenosis rates are in line with those observed in other studies.
951
Although a total of 6 cardiac deaths occurred, all were
due to complications of the index myocardial infarction
(prolonged hypoxia, refractory cardiogenic shock, and sepsis) and none of the deaths were a consequence of complications arising from the intervention, stent thrombosis, or
type of stent used.
In-stent restenosis and subacute thrombosis are more
common in longer length stents.10 The present study had
patients with an average stent length of 21.8 mm. The
average length of the bare metal stents was longer than the
average length of both the paclitaxel-eluting stent and the
sirolimus-eluting stent groups. Pasceri et al11 studied the use
of sirolimus-eluting stents in acute myocardial infarctions.
Two of the patients in their study group had subacute
thrombosis when they stopped taking ticlidopine. The average stent length in their study was 30 ⫾ 13 mm, which is
longer than the stent lengths in the present study.
Previous studies have shown that a potential problem
with the DES is late in-stent thrombosis.12 Multiple studies
have shown that the use of antiplatelet agents decreases the
risk and the interruption of the antiplatelet therapy is associated with in-stent thrombosis.4,5,13 All the patients in this
study were placed on aspirin and clopidogrel after stent
implantation, as recommended. Two patients in the paclitaxel group were prematurely taken off their antiplatelet
therapy, and this likely played a role in the observed MACE
events. One of these patients also had completed the recommended regimen of clopidegrel but had a late event off
of aspirin.
This study is relatively small in size and was not randomized. Follow-up angiograms were not routinely performed (performed only on patients who had clinical indications for angiography). Angiographic evaluations of
index procedures were not performed to assess for dissection, bifurcation location, stent underdeployment, and residual thrombus, which all have been associated with in-stent
thrombosis. This type of analysis may have helped explain
some of the early in-stent thrombosis seen in the present
study.
1. Laarman GJ, Suttorp MJ, Dirksen MT, van Heerebeek L, Kiemeneij F,
Slagboom T, van der Wieken LR, Tijssen JGP, Rensing BJ, Patterson
M. Paclitaxel-eluting versus uncoated stents in primary percutaneous
coronary intervention. N Engl J Med 2006;355:1105–1113.
2. Spaulding C, Henry P, Teiger E, Beatt K, Bramucci E, Carrie D, Slama
MS, Merkely B, Erglis A, Margheri M, et al. Sirolimus-eluting versus
uncoated stents in acute myocardial infarction. N Engl J Med 2006;
355:1093–1104.
3. Menichelli M, Parma A, Pucci E, Fiorilli R, Felice FD, Nazzaro M,
Giulivi A, Alborino D, Azzelino A, Violini R. Randomized trial of
sirolimus-eluting stent versus bare-metal stent in acute myocardial
infarction (SESAMI). J Am Coll Cardiol 2007;49:1924 –1930.
4. Park DW, Park SW, Park KH, Lee BK, Kim YH, Lee CW, Hong MK,
Kim JJ, Park SJ. Frequency of and risk factors for stent thrombosis
after drug-eluting stent implantation during long-term follow-up. Am J
Cardiol 2006;98:352–356.
5. Iakovou I, Schmidt T, Bonizzoni E, Ge L, Sangiorgi G, Stankovic G,
Airoldi F, Chieffo A, Montorfano M, Carlino M, et al. Incidence,
predictors and outcome of thrombosis after successful implantation of
drug-eluting stents. JAMA 2005;293:2126 –2130.
6. Hofma S, Ong A, Aoki J, Van Mieghem C, Granillo G, Vaglimigli
M, Regar E, de Jaegere P, McFadden EP, Sianos G, et al. One year
clinical follow up of paclitaxel eluting stents for acute myocardial
infarction compared with sirolimus eluting stents. Heart 2005;91:
1176 –1180.
952
The American Journal of Cardiology (www.AJConline.org)
7. Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann
J, Turco M, Caputo R, Bergin P, Greenberg J, Popma J, Russell M, for
the TAXUS IV Investigators. A polymer-based, paclitaxel-eluting
stent in patients with coronary artery disease. N Engl J Med 2004;350:
221–231.
8. Moses J, Leon M, Popma JJ, Fitzgerald PJ, Holmes D, O’Shaughnessy
C, Caputo R, Kereiakes D, Williams D, Teirstein P, Jaeger J, Kuntz,
Richard El, for the SIRIUS Investigators. Sirolumus-eluting stents
versus standard stents in patients with stenosis in native coronary
artery. N Engl J Med 2003;349:1315–1323.
9. Park SJ, Shim WH, Ho DS, Raizner AE, Park SW, Hong M, Lee CW,
Choi D, Jang Y, Lam R, Weissman NJ, Mintz GS. A paclictaxel stent
for the prevention of coronary restenosis. N Engl J Med 2003;348:
1537–1545.
10. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin
M, Colombo A, Schuler G, Barragan P, Guagliumi G, Molnar F,
Falotico R, for the RAVEL Study Group. A randomized comparison
of a sirolimus-eluting stent with a standard stent for coronary
revascularization. N Engl J Med 2002;346:1773–1780.
11. Pasceri V, Granatelli A, Pristipino C, Pelliccia F, Speciale G, Pironi B,
Roncella A, Richichi G. A randomized trial of rapamycin-eluting stent
in acute myocardial infarction: preliminary results (abstr). Am J Cardiol 2003;92(suppl 1):1.
12. Regar E, Lemos P, Saia F, Degerteckia M, Tanabe K, Lee C, Arampatzis C, Hoye A, Sianos G, Feyter P, et al. Incidence of thrombotic
stent occlusion during the first three months after sirolimus-eluting
stent implantation in 500 consecutive patients. Am J Cardiol 2004;93:
1271–1275.
13. Ong A, Hoye A, Aoki J, Van Mieghem C, Granillo G, Sonnenschein
K, Regar E, McFadden EP, Sianos G, van der Giessen WJ, et al.
Thirty-day incidence and six-month clinical outcome of thrombotic
stent occlusion after bare-metal, sirolimus, or paclitaxel stent implantation. J Am Coll Cardiol 2005;45:947–953.
Comparison of Virtual Histology to Intravascular Ultrasound of
Culprit Coronary Lesions in Acute Coronary Syndrome and Target
Coronary Lesions in Stable Angina Pectoris
Myeong-Ki Hong, MD, PhDa, Gary S. Mintz, MDc, Cheol Whan Lee, MD, PhDa, Jon Suh, MDa,
Jeong-Hoon Kim, MDa, Duk-Woo Park, MD, PhDa, Seung-Whan Lee, MD, PhDa,
Young-Hak Kim, MD, PhDa, Sang-Sig Cheong, MD, PhDb, Jae-Joong Kim, MD, PhDa,
Seong-Wook Park, MD, PhDa, and Seung-Jung Park, MD, PhDa,*
Coronary plaque composition cannot be assessed accurately using gray-scale intravascular
ultrasound (IVUS). Using virtual histology IVUS (VH-IVUS), a comparison of coronary
plaque composition between acute coronary syndromes (ACS) and stable angina pectoris
(SAP) was performed. Preintervention IVUS of de novo culprit and target lesions was
performed in 318 patients (123 with ACS and 195 with SAP). Using VH-IVUS, plaque was
characterized as fibrotic, fibrofatty, dense calcium, and necrotic core. VH-IVUS-derived
thin-cap fibroatheroma (VH-TCFA) was defined as necrotic core >10% of plaque area
without overlying fibrous tissue in a plaque burden >40%. Lesions were classified into 3
groups: ruptured, VH-TCFA, and non–VH-TCFA plaque. Unstable lesions were defined as
either VH-TCFA or ruptured plaque. Compared with patients with SAP, those with ACS
had significantly more unstable lesions (89% vs 62%, p <0.001). Planar VH-IVUS analysis
at the minimum luminal site and at the largest necrotic core site and volumetric analysis
over a 10-mm-long segment centered at the minimum luminal site showed that the
percentage of necrotic core was significantly greater and that the percentage of fibrofatty
plaque was significantly smaller in patients with ACS. The percentages of fibrotic and
fibrofatty plaque areas and volumes were smaller, and the percentages of necrotic core
areas and volumes were larger in VH-TCFAs compared with non-TCFAs. Ruptured
plaques in VH-IVUS analyses showed intermediate findings between VH-TCFAs and
non–VH-TCFAs. In conclusion, culprit lesions in patients with ACS were more unstable
and had greater amounts of necrotic core and smaller amounts of fibrofatty plaque
compared with target lesions in patients with SAP. © 2007 Elsevier Inc. All rights
reserved. (Am J Cardiol 2007;100:953–959)
Clinical symptoms and presentations, not lesion morphology, define acute coronary syndromes (ACS) and differentiate ACS from stable angina pectoris (SAP). However,
unstable clinical symptoms in patients with ACS are associated with unstable plaque characteristics.1– 4 Conventional
gray-scale intravascular ultrasound (IVUS) has significant
limitations in accurately assessing atheromatous plaque
composition. These limitations have been partially addressed by virtual histology IVUS (VH-IVUS), which characterizes plaque as calcified, fibrotic, fibrofatty, or necrotic
core.5,6 The purpose of this study was to use VH-IVUS to
a
Department of Medicine, University of Ulsan College of Medicine,
Asan Medical Center, Seoul; bDepartment of Medicine, University of
Ulsan College of Medicine, Asan Medical Center, Gangneung, Korea; and
c
Cardiovascular Research Foundation, New York, New York. Manuscript
received March 5, 2007; revised manuscript received and accepted April
13, 2007.
This study was supported in part by the Cardiovascular Research
Foundation, Seoul, Korea, and Grant 0412-CR02-0704-0001 from the
Korea Health 21 R&D Project, Ministry of Health & Welfare, Seoul,
Korea.
*Corresponding author: Tel: 82-2-3010-3152; fax: 82-2-475-6898.
E-mail address: [email protected] (S.-J. Park).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.034
compare coronary plaque composition between patients
with ACS and patients with SAP.
Methods
Study population: From June 2005 to July 2006, 318
patients (123 with ACS and 195 with SAP) with de novo
culprit and target lesions underwent preintervention VHIVUS at Asan Medical Center. The ACS group included 41
patients with unstable angina, 28 patients with non–STsegment elevation myocardial infarction, and 54 patients
with ST-segment elevation myocardial infarction. Acute
myocardial infarction was defined as continuous chest pain
at rest with abnormal levels of cardiac enzymes (creatinine
kinase-MB or troponin T). SAP was defined as no change in
the frequency, duration, or intensity of symptoms within 6
weeks before the intervention.7 The culprit lesion in ACS or
the target lesion in SAP was identified by the combination
of left ventricular wall motion abnormalities, electrocardiographic findings, angiographic lesion morphology, and scintigraphic defects. In patients with SAP who underwent
multivessel intervention, the lesion with the worst diameter
stenosis and more complex morphology in the territory of
scintigraphic reversible defects was selected as the target
www.AJConline.org
954
The American Journal of Cardiology (www.AJConline.org)
Table 1
Baseline clinical characteristics
Variable
Age (yrs)
Men
Hypertension
Diabetes mellitus
Cigarette smoker
Lipid profiles at baseline (mg/dl)
Total cholesterol
HDL cholesterol
LDL cholesterol
Triglycerides
C-reactive protein (mg/dl)
Coronary intervention
No. of stents used
No. of narrowed coronary arteries
1
2
3
ACS
(n ⫽ 123)
SAP
(n ⫽ 195)
p Value
59 ⫾ 11
104 (85%)
47 (38%)
22 (18%)
65 (53%)
60 ⫾ 9
126 (65%)
97 (50%)
49 (25%)
38 (20%)
0.7
⬍0.001
0.044
0.131
⬍0.001
185 ⫾ 42
39 ⫾ 11
117 ⫾ 38
176 ⫾ 147
0.6 ⫾ 0.9
121 (98%)
1.2 ⫾ 0.4
168 ⫾ 35
44 ⫾ 13
94 ⫾ 27
158 ⫾ 93
0.3 ⫾ 0.6
191 (98%)
1.2 ⫾ 0.4
⬍0.001
0.004
⬍0.001
0.25
⬍0.001
0.6
0.21
0.018
71 (58%)
35 (28%)
17 (14%)
139 (71%)
44 (23%)
12 (6%)
Variables
HDL ⫽ high-density lipoprotein; LDL ⫽ low-density lipoprotein.
Table 2
Gray-scale intravascular ultrasound findings between acute coronary
syndromes and stable angina pectoris
Variable
Minimum luminal area
EEM area (mm2)
Luminal area (mm2)
P&M area (mm2)
Remodeling index
Largest necrotic core
EEM area (mm2)
Luminal area (mm2)
P&M area (mm2)
Volumetric analysis
EEM volume (mm3)
Luminal volume (mm3)
P&M volume (mm3)
Table 3
Virtual histology intravascular ultrasound findings between acute
coronary syndromes and stable angina pectoris
ACS
(n ⫽ 123)
SAP
(n ⫽ 195)
p Value
17.1 ⫾ 4.5
3.7 ⫾ 1.0
13.4 ⫾ 4.4
1.07 ⫾ 0.18
15.0 ⫾ 4.5
3.8 ⫾ 0.9
11.2 ⫾ 4.4
1.02 ⫾ 0.19
⬍0.001
0.3
⬍0.001
0.038
17.4 ⫾ 4.4
4.8 ⫾ 1.7
12.6 ⫾ 4.2
15.7 ⫾ 5.4
5.0 ⫾ 2.1
10.7 ⫾ 4.4
0.003
0.3
⬍0.001
167.7 ⫾ 43.8
59.5 ⫾ 15.6
108.3 ⫾ 36.7
149.2 ⫾ 40.5
60.1 ⫾ 14.1
89.1 ⫾ 34.4
⬍0.001
0.7
⬍0.001
Minimum luminal area site
Absolute areas (mm2)
Fibrotic (green)
Fibrofatty (yellow-green)
Dense calcium (white)
Necrotic core (red)
Percentages
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Largest necrotic core site
Absolute areas (mm2)
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Percentages
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Volumetric analysis
Absolute volumes (mm3)
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Percentages
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
ACS
(n ⫽ 123)
SAP
(n ⫽ 195)
p Value
5.3 ⫾ 2.7
0.5 ⫾ 0.6
0.8 ⫾ 0.7
3.1 ⫾ 1.9
4.6 ⫾ 3.0
0.5 ⫾ 0.6
0.6 ⫾ 0.6
2.1 ⫾ 1.3
0.030
0.6
0.001
⬍0.001
53 ⫾ 15
5⫾5
9⫾7
33 ⫾ 14
56 ⫾ 15
7⫾6
8⫾8
29 ⫾ 14
0.073
0.020
0.4
0.015
5.0 ⫾ 4.3
0.4 ⫾ 0.4
0.9 ⫾ 0.7
3.4 ⫾ 2.0
4.0 ⫾ 2.8
0.4 ⫾ 0.5
0.7 ⫾ 0.7
2.3 ⫾ 1.6
0.015
0.6
0.003
⬍0.001
50 ⫾ 15
4⫾4
10 ⫾ 7
36 ⫾ 13
53 ⫾ 15
5⫾5
9⫾8
33 ⫾ 14
0.105
0.024
0.5
0.034
41.9 ⫾ 22.4
4.7 ⫾ 4.5
6.4 ⫾ 5.1
20.3 ⫾ 12.6
32.3 ⫾ 20.8
4.5 ⫾ 4.7
4.4 ⫾ 4.6
14.3 ⫾ 9.5
56 ⫾ 13
6⫾5
9⫾7
29 ⫾ 12
57 ⫾ 13
8⫾5
9⫾8
27 ⫾ 11
⬍0.001
0.7
0.001
⬍0.001
0.3
0.045
0.5
0.081
lesion for VH-IVUS analysis.8 Total occlusions, bifurcation
lesions, lesions with severe angulations, and heavily calcified lesions (all because of technical difficulties in performing smooth, motorized catheter pullback) were excluded
from this study.
IVUS imaging and analysis: VH-IVUS examination
was performed before any intervention and after the intracoronary administration of nitroglycerin 0.2 mg using a
motorized transducer pullback system (0.5 mm/s). A 2.9Fr
IVUS imaging catheter (Eagle Eye; Volcano Corporation,
Rancho Cordova, California) incorporated a 20-MHz
phased-array transducer.
Conventional gray-scale quantitative IVUS analyses
were performed according to criteria of the clinical expert
consensus document on IVUS to include external elastic
membrane (EEM), luminal, and plaque and media (P&M;
defined as EEM minus luminal) areas.9 Plaque burden was
defined as P&M area divided by EEM area. A remodeling
Figure 1. Incidence of unstable and stable plaques in patients with ACS
compared with those with SAP. Unstable and stable lesions were defined
as either VH-TCFA or ruptured plaque and non–VH-TCFA plaque, respectively.
index was calculated as the lesion EEM divided by the mean
reference EEM area. IVUS signs of plaque rupture were a
cavity that communicated with the lumen with an overlying
residual fibrous cap fragment.7,8
Coronary Artery Disease/Plaque Compositions by VH-IVUS
955
Table 4
Gray-scale intravascular ultrasound findings comparing ruptured plaques with virtual histology intravascular ultrasound– derived thin-cap fibroatheromas
and non–virtual histology intravascular ultrasound– derived thin-cap fibroatheromas
Variable
Minimum luminal area site
EEM area (mm2)
Luminal area (mm2)
P&M area (mm2)
Remodeling index
Largest necrotic core site
EEM area (mm2)
Luminal area (mm2)
P&M area (mm2)
Volumetric analysis
EEM volume (mm3)
Luminal volume (mm3)
P&M volume (mm3)
Plaque Rupture
(n ⫽ 74)
p Value†
Non–Plaque Rupture
VH-TCFA
(n ⫽ 156)
Non-VH-TCFA
(n ⫽ 88)
18.7 ⫾ 4.4
3.8 ⫾ 0.8
14.9 ⫾ 4.4
1.10 ⫾ 0.20
15.0 ⫾ 4.4*
3.7 ⫾ 0.9
11.3 ⫾ 4.2*
1.03 ⫾ 0.19*
14.8 ⫾ 4.3
3.7 ⫾ 0.9
11.1 ⫾ 4.2
1.00 ⫾ 0.15
⬍0.001
1.0
⬍0.001
0.003
19.2 ⫾ 5.8
5.1 ⫾ 2.5
14.1 ⫾ 4.6
15.7 ⫾ 4.7*
4.9 ⫾ 1.9
10.8 ⫾ 4.0*
15.0 ⫾ 4.3
4.6 ⫾ 1.7
10.4 ⫾ 4.0
⬍0.001
0.3
⬍0.001
183.6 ⫾ 40.9
60.5 ⫾ 16.4
122.9 ⫾ 38.4
150.0 ⫾ 42.4*
59.1 ⫾ 14.4
91.2 ⫾ 33.8*
144.5 ⫾ 34.5
60.8 ⫾ 13.6
83.8 ⫾ 27.8
⬍0.001
0.6
⬍0.001
* p ⬍0.05 comparing VH-TCFAs and ruptured plaques. There were no differences when VH-TCFAs were compared with non–VH-TCFAs.
Analysis of variance.
†
Planar VH-IVUS analysis was performed at the site of
the minimal luminal area and the site of the largest
necrotic core. Volumetric VH-IVUS analysis was performed along a 10-mm segment centered on the minimal
luminal area; calculations were made using Simpson’s
rule. VH-IVUS analysis classified and color-coded tissue
as green (fibrotic), yellow-green (fibrofatty), white
(dense calcium), and red (necrotic core).5,6 VH-IVUS
analyses are reported in absolute amounts and as percentages (relative amounts) of plaque area and volume. VHIVUS-derived thin-cap fibroatheroma (VH-TCFA) was
defined as a necrotic core ⱖ10% of plaque area at either
the minimal luminal area site or the largest necrotic core
site in ⱖ3 consecutive frames without evident overlying fibrous tissue in the presence of ⱖ40% plaque
burden.6
Because plaque rupture is 1 of the final fates of TCFA,
and the identification of TCFA before the rupture of plaque
occurs is clinically significant, plaque rupture has different
characteristics from TCFA. Differentiation between TCFA
and non-TCFA is also clinically important because plaque
rupture occurs mainly in TCFA lesions rather than nonTCFA lesions. Therefore, according to gray-scale and VHIVUS findings, culprit and target lesions were classified into
3 groups: ruptured plaque, VH-TCFA, and non–VH-TCFA
plaque. Unstable lesions contained either VH-TCFAs or
ruptured plaques.
Statistical analysis: Statistical analysis was performed
with SPSS (SPSS, Inc., Chicago, Illinois). Data are presented as frequencies or mean ⫾ SD. Comparisons were
performed with chi-square statistics or Fisher’s exact test
and the unpaired Student’s t test, the Mann-Whitney U test,
or analysis of variance. Multiple stepwise logistic regression analysis was performed to assess independent predictors
for VH-TCFA. A p value ⬍0.05 was considered statistically
significant.
Results
Baseline clinical characteristics are listed in Table 1. Table 2
lists gray-scale IVUS findings of culprit and target lesions,
comparing patients with ACS and those with SAP. Patients
with ACS had significantly larger EEM and P&M areas and
larger remodeling indexes than those with SAP.
Table 3 lists VH-IVUS analysis of culprit and target
lesions, comparing patients with ACS and those with SAP.
Planar VH-IVUS analysis at the minimum luminal site and
at the largest necrotic core site showed that the percentage
of necrotic core area was significantly greater (33% vs 29%,
p ⫽ 0.015, and 36% vs 33%, p ⫽ 0.034, respectively) and
that the percentage of fibrofatty plaque (5% vs 7%, p ⫽
0.020, and 4% vs 5%, p ⫽ 0.024, respectively) was significantly smaller in patients with ACS. There was a tendency
toward smaller percentages of fibrotic plaque (53% vs 56%,
p ⫽ 0.073, and 50% vs 53%, p ⫽ 0.105, respectively) in
patients with ACS. Volumetric VH-IVUS analysis over a
10-mm-long segment centered at the minimum luminal site
supported these observations.
VH-TCFAs: Culprit lesions in the 123 patients with
ACS contained 45 ruptured plaques (37%), 64 VH-TCFAs
(52%), and 14 non–VH-TCFA (11%); conversely, target
lesions in 195 patients with SAP contained 29 ruptured
plaques (15%), 92 VH-TCFAs (47%), and 74 non–VHTCFAs (38%) (p ⬍0.001, patients with ACS vs patients
with SAP). Therefore, compared with patients with SAP,
those with ACS had significantly more unstable lesions
(89% vs 62%, p ⬍0.001; Figure 1).
There were no significant differences in gray-scale IVUS
findings between nonruptured lesions with VH-TCFA and
nonruptured lesions without VH-TCFA (Table 4). However,
VH-TCFAs had smaller EEM and P&M areas and volumes
compared with the ruptured plaques.
In the VH-IVUS analysis, there were smaller percentages
of fibrotic and fibrofatty plaque areas and volumes and a
956
The American Journal of Cardiology (www.AJConline.org)
Table 5
Virtual histology intravascular ultrasound findings comparing ruptured plaques with virtual histology intravascular ultrasound– derived thin-cap
fibroatheromas and non–virtual histology intravascular ultrasound– derived thin-cap fibroatheromas
Variable
Minimum luminal area site
Absolute areas (mm2)
Fibrotic (green)
Fibrofatty (yellow-green)
Dense calcium (white)
Necrotic core (red)
Percentages
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Largest necrotic core site
Absolute areas (mm2)
Fibrotic (green)
Fibrofatty (yellow-green)
Dense calcium (white)
Necrotic core (red)
Percentages
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Volumetric analysis
Absolute volumes (mm3)
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Percentages
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Plaque Rupture
(n ⫽ 74)
p Value‡
Non–Plaque Rupture
VH-TCFA
(n ⫽ 156)
Non–VH-TCFA
(n ⫽ 88)
6.5 ⫾ 2.9
0.7 ⫾ 0.7
0.8 ⫾ 0.7
3.2 ⫾ 1.9
4.0 ⫾ 2.4ⴱ,†
0.4 ⫾ 0.5ⴱ,†
0.7 ⫾ 0.6
2.8 ⫾ 1.6*
5.0 ⫾ 3.1
0.7 ⫾ 0.7
0.6 ⫾ 0.8
1.4 ⫾ 0.9
⬍0.001
⬍0.001
0.14
⬍0.001
58 ⫾ 15
7⫾6
7⫾7
28 ⫾ 14
50 ⫾ 13ⴱ,†
4 ⫾ 4ⴱ,†
9⫾8
37 ⫾ 12ⴱ,†
64 ⫾ 15
9⫾7
8⫾9
20 ⫾ 10
⬍0.001
⬍0.001
0.3
⬍0.001
6.2 ⫾ 2.2
0.6 ⫾ 0.6
0.9 ⫾ 0.8
3.5 ⫾ 2.0
3.6 ⫾ 2.2†
0.3 ⫾ 0.3ⴱ,†
0.7 ⫾ 0.6
3.0 ⫾ 1.7*
4.4 ⫾ 2.9
0.6 ⫾ 0.6
0.7 ⫾ 0.8
1.6 ⫾ 1.2
⬍0.001
⬍0.001
0.09
⬍0.001
53 ⫾ 15
5⫾5
9⫾7
33 ⫾ 14
46 ⫾ 12ⴱ,†
3 ⫾ 3ⴱ,†
10 ⫾ 7
41 ⫾ 10ⴱ,†
60 ⫾ 15
7⫾6
10 ⫾ 10
23 ⫾ 10
⬍0.001
⬍0.001
0.5
⬍0.001
52.2 ⫾ 25.4
6.6 ⫾ 5.7
6.5 ⫾ 5.4
21.8 ⫾ 11.9
30.5 ⫾ 18.2†
3.3 ⫾ 3.8ⴱ,†
5.5 ⫾ 4.7*
17.8 ⫾ 11.4ⴱ,†
32.0 ⫾ 18.4
5.2 ⫾ 4.2
3.4 ⫾ 4.2
10.1 ⫾ 6.1
⬍0.001
⬍0.001
⬍0.001
⬍0.001
59 ⫾ 12
7⫾5
8⫾6
26 ⫾ 11
53 ⫾ 12ⴱ,†
5 ⫾ 4ⴱ,†
10 ⫾ 7ⴱ,†
32 ⫾ 10ⴱ,†
62 ⫾ 13
10 ⫾ 5
8⫾9
21 ⫾ 9
⬍0.001
⬍0.001
0.009
⬍0.001
* p ⬍0.05 comparing TCFAs with non–VH-TCFAs.
p ⬍0.05 comparing TCFAs with ruptured plaques.
‡
Analysis of variance.
†
larger percentage of necrotic core areas and volumes in
VH-TCFAs compared with non-TCFAs (Table 5). Ruptured
plaques in VH-IVUS analyses showed intermediate findings
between VH-TCFA and non-VH-TCFA plaques; the percentages of fibrotic and fibrofatty plaque areas and volumes
were greater, and the percentage of necrotic core area and
volume was smaller in ruptured plaques compared with
VH-TCFAs (Table 5).
Clinical characteristics of subgroup patients without
plaque rupture are listed in Table 6. Multiple stepwise
logistic regression analysis including all clinical variables
with p values ⬍0.2 in univariate analysis (male gender,
low-density lipoprotein and high-density lipoprotein cholesterol levels, and ACS) indicated that ACS was the only
independent predictor of VH-TCFA (odds ratio 2.739, 95%
confidence interval 1.252 to 5.993, p ⫽ 0.012) in patients
without ruptured plaques.
Table 7 lists VH-IVUS analysis of culprit lesions, comparing patients with non–ST-segment-elevation myocardial
infarction or unstable angina and those with ST-segment
elevation myocardial infarction.
Discussion
The present preintervention gray-scale and VH-IVUS analysis of 318 patients showed distinctly different VH-IVUS
morphologies between culprit lesions in patients with ACS
and target lesions in patients with SAP. Necrotic cores were
larger and fibrofatty plaque was less, and there were more
unstable lesions (either plaque rupture or VH-TCFA) in
patients with ACS compared with those with SAP.
The angiographic assessment of coronary luminal stenosis has been considered a surrogate marker of the severity of
atherosclerosis. However, coronary angiography, a luminogram, has low predictive value to assess atherosclerotic
plaque burden or to predict ACS events.10 –12 Although
IVUS provides cross-sectional morphometric detail (e.g.,
identification of ruptured plaques, assessment of remodel-
Coronary Artery Disease/Plaque Compositions by VH-IVUS
957
Table 6
Baseline clinical characteristics of patients without plaque rupture comparing patients with virtual histology intravascular ultrasound– derived thin-cap
fibroatheromas with patients without virtual histology intravascular ultrasound– derived thin-cap fibroatheromas
Variable
Age (yrs)
Men
Hypertension
Diabetes mellitus
Cigarette smoker
Lipid profiles at baseline (mg/dl)
Total cholesterol
HDL cholesterol
LDL cholesterol
Triglycerides
C-reactive protein (mg/dl)
ACS
No. of narrowed coronary arteries
1
2
3
Lesions With VH-TCFA
(n ⫽ 156)
Lesions Without VH-TCFA
(n ⫽ 88)
p Value
61 ⫾ 10
111 (73%)
67 (43%)
35 (22%)
45 (29%)
59 ⫾ 10
53 (60%)
44 (50%)
21 (24%)
21 (24%)
0.2
0.08
0.3
0.8
0.4
172 ⫾ 35
42 ⫾ 11
104 ⫾ 33
163 ⫾ 130
0.4 ⫾ 0.7
64 (41%)
170 ⫾ 39
45 ⫾ 15
94 ⫾ 29
156 ⫾ 88
0.3 ⫾ 0.5
14 (16%)
0.7
0.08
0.07
0.6
0.5
⬍0.001
0.5
105 (67%)
36 (23%)
15 (10%)
64 (73%)
19 (22%)
5 (5%)
Abbreviations as in Table 1.
Table 7
Virtual histology intravascular ultrasound findings between non–ST-segment elevation myocardial infarction or unstable angina pectoris and ST-segmentelevation myocardial infarction
Variable
Minimum luminal area site
Absolute areas (mm2)
Fibrotic (green)
Fibrofatty (yellow-green)
Dense calcium (white)
Necrotic core (red)
Percentages
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Largest necrotic core site
Absolute areas (mm2)
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Percentages
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Volumetric analysis
Absolute volumes (mm3)
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Percentages
Fibrotic
Fibrofatty
Dense calcium
Necrotic core
Non–ST-Segment Elevation Myocardial
Infarction/Unstable Angina Pectoris
(n ⫽ 69)
ST-Segment Elevation
Myocardial Infarction
(n ⫽ 54)
p Value
5.1 ⫾ 2.7
0.5 ⫾ 0.5
0.9 ⫾ 0.8
3.1 ⫾ 2.1
5.6 ⫾ 2.7
0.6 ⫾ 0.6
0.8 ⫾ 0.6
3.2 ⫾ 1.8
0.3
0.3
0.5
0.8
52 ⫾ 17
5⫾5
10 ⫾ 9
34 ⫾ 15
55 ⫾ 13
6⫾6
8⫾5
32 ⫾ 13
0.3
0.2
0.102
0.5
5.1 ⫾ 5.3
0.4 ⫾ 0.4
1.0 ⫾ 0.7
3.4 ⫾ 2.2
4.9 ⫾ 2.3
0.4 ⫾ 0.4
0.8 ⫾ 0.7
3.3 ⫾ 1.7
0.8
0.2
0.192
0.7
49 ⫾ 16
4⫾4
11 ⫾ 8
37 ⫾ 14
51 ⫾ 13
5⫾5
8⫾6
35 ⫾ 12
0.3
0.186
0.06
0.6
38.0 ⫾ 20.2
3.9 ⫾ 3.7
6.6 ⫾ 5.2
20.5 ⫾ 13.5
46.8 ⫾ 24.3
5.8 ⫾ 5.2
6.0 ⫾ 5.0
20.1 ⫾ 11.6
0.03
0.025
0.5
0.9
54 ⫾ 14
6⫾5
10 ⫾ 7
31 ⫾ 12
58 ⫾ 12
7⫾6
8⫾6
26 ⫾ 11
0.072
0.076
0.136
0.054
958
The American Journal of Cardiology (www.AJConline.org)
ing) and quantifies atherosclerotic plaque area and volume
and plaque burden, more recent studies have suggested that
gray-scale IVUS has limited value for the identification of
specific plaque components.4 – 6,13,14
Spectral analysis (i.e., VH-IVUS) has the potential to
provide detailed qualitative and quantitative information;
the identification of 4 specific plaque components has been
validated in explanted human coronary segments as well as
in retrieved directional coronary atherectomy specimens.5,15
We applied these histologically validated VH-IVUS findings to real clinical practice and compared coronary plaque
composition between patients with ACS and those with
SAP. The amounts of necrotic core were larger and the
amounts of fibrotic and fibrofatty plaque were less in patients with ACS. The results of the present study are similar
to those of a previous pathologic study showing larger
necrotic cores in TCFAs compared with stable plaques
(24% vs 12%, p ⫽ 0.01).3
The rupture of a vulnerable plaque and subsequent
thrombus formation are the most important mechanisms
leading to ACS.1–3 Retrospective pathologic studies of patients with coronary artery disease who died suddenly
showed culprit lesion plaque rupture in about 70% of patients.1,2,16 IVUS studies have reported a 16% to 66% incidence of plaque rupture in the culprit lesions of patients
with ACS.8,17,18 Wide variation in the incidence of ruptured
plaque might be partly related to different clinical presentations, different study populations, different lesion segments (culprit vs nonculprit lesions), and different resolution powers of IVUS catheters (a 20-MHz phased-array
transducer vs a rotating 30- or 40-MHz transducer). In the
present study, ruptured plaques detected with gray-scale
IVUS were more common in patients with ACS than in
those with SAP, and most patients with ACS had unstable
lesion morphology. Several pathologic studies have suggested that TCFAs are particularly prone to rupture and
result in acute coronary artery occlusions.1–3 Using a VHIVUS definition of TCFA similar to the one used in the
present study, Rodriguez-Granillo et al6 reported that 23
patients with ACS had a significantly higher prevalence of
VH-TCFAs in secondary, nonobstructive lesions (diameter
stenosis ⬍50%) compared with 32 patients with SAP. In the
present study, primary lesions (lesions targeted for intervention) were also more often unstable in patients with ACS
than in those with SAP.
A previous pathologic study showed that the percentage
of necrotic core was significantly larger in ruptured plaques
than in TCFAs.3 There are 2 explanations why this was not
seen in the present study. VH-IVUS cannot identify the
thrombus that commonly follows plaque rupture. It most
often appears “green” and is classified as fibrotic plaque,
reducing the calculated relative size of the necrotic core. In
addition, we studied lesions after rupture, after the necrotic
core may have embolized.
The present study also detected ruptured plaques in 15%
of target lesions in patients with SAP (similar to previous
studies7,8) and VH-TCFAs in 47% of target lesions in patients with SAP. VH-TCFAs may stabilize without plaque
rupture or clinical presentation. Unstable clinical symptoms
may depend on the severity of the original and/or coexisting
stenosis or on thrombus formation, not just on plaque rup-
ture.19 The relatively higher incidence of VH-TCFAs in
patients with SAP in the present study compared with histologic reports may be partly explained by the different
definitions of TCFA used and the different patient populations studied. However, it is interesting to speculate that the
presence of unstable lesion morphology (ruptured plaque or
VH-TCFA) may have contributed to the clinical progression
to intervention-requiring symptoms in many of the patients
with SAP in the present study.
This study was a single-center, retrospective study. VHIVUS cannot determine the presence of thrombus. Total
occlusions, bifurcation lesions, lesions with severe angulations, and heavily calcified lesions were excluded from this
study. Therefore, this study might not represent the whole
spectrum of patients with ACS and patients with SAP.
1. Naghavi M, Libby P, Falk E, Casscells SW, Litovsky S, Rumberger J,
Badimon JJ, Stefanadis C, Moreno P, Pasterkamp G, et al. From
vulnerable plaque to vulnerable patient: a call for new definitions and
risk assessment strategies: part 1. Circulation 2003;108:1664 –1672.
2. Virmani R, Kolodgie FD, Burke AP, Farb A, Schwartz SM. Lessons
from sudden coronary death: a comprehensive morphological classification scheme for atherosclerotic lesions. Arterioscler Thromb Vasc
Biol 2000;20:1262–1275.
3. Virmani R, Burke AP, Farb A, Kolodgie FD. Pathology of the vulnerable plaque. J Am Coll Cardiol 2006;47:C13–C18.
4. Schoenhagen P, Ziada KM, Kapadia SR, Crowe TD, Nissen SE, Tuzcu
EM. Extent and direction of arterial remodeling in stable versus unstable coronary syndromes. an intravascular ultrasound study. Circulation 2000;101:598 – 603.
5. Nair A, Kuban BD, Tuzcu EM, Schoenhagen P, Nissen SE, Vince G.
Coronary plaque classification with intravascular ultrasound radiofrequency data analysis. Circulation 2002;106:2200 –2206.
6. Rodriguez-Granillo GA, García-García HM, Mc Fadden EP, Valgimigli M, Aoki J, de Feyter P, Serruys PW. In vivo intravascular
ultrasound-derived thin-cap fibroatheroma detection using ultrasound
radiofrequency data analysis. J Am Coll Cardiol 2005;46:2038 –2042.
7. Mintz GS, Maehara A, Bui AB, Weissman NJ. Multiple versus single
coronary plaque ruptures detected by intravascular ultrasound in stable
and unstable angina pectoris and in acute myocardial infarction. Am J
Cardiol 2003;91:1333–1335.
8. Hong MK, Mintz GS, Lee CW, Kim YH, Lee SW, Song JM, Han KH,
Kang DH, Song JK, Kim JJ, et al. Comparison of coronary plaque
rupture between stable angina and acute myocardial infarction: a
three-vessel intravascular ultrasound study in 235 patients. Circulation
2004;110:928 –933.
9. Mintz GS, Nissen SE, Anderson WD, Bailey SR, Erbel R, Fitzgerald
PJ, Pinto FJ, Rosenfield K, Siegel RJ, Tuzcu EM, Yock PG. American
College of Cardiology clinical expert consensus document on standards for acquisition, measurement and reporting of intravascular
ultrasound studies (IVUS): a report of the American College of Cardiology Task Force on Clinical Expert Consensus Documents. J Am
Coll Cardiol 2001;37:1478 –1492.
10. Gerard Pasterkamp G, Falk E, Woutman H, Borst C. Techniques
characterizing the coronary atherosclerotic plaque: influence on clinical decision making? J Am Coll Cardiol 2000;36:13–21.
11. 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.
12. 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;78:1157–1166.
13. Gussenhoven EJ, Essed CE, Lancee CT, Mastik F, Frietman P, van
Egmond FC, Reiber J, Bosch H, van Urk H, Roelandt J. Arterial wall
characteristics determined by intravascular ultrasound imaging: an in
vitro study. J Am Coll Cardiol 1989;14:947–952.
Coronary Artery Disease/Plaque Compositions by VH-IVUS
14. Hodgson JM, Reddy KG, Suneja R, Nair RN, Lesnefsky EJ, Sheehan
HM. Intracoronary ultrasound imaging: correlation of plaque morphology
with angiography, clinical syndrome and procedural results in patients
undergoing coronary angioplasty. J Am Coll Cardiol 1993;21:35– 44.
15. Nasu K, Tsuchikane E, Katoh O, Vince DG, Virmani R, Surmely JF,
Murata A, Takeda Y, Ito T, Ehara M, et al. Accuracy of in vivo
coronary plaque morphology assessment. A validation study of in vivo
virtual histology compared with in vitro histopathology. J Am Coll
Cardiol 2006;47:2405–2412.
16. Falk E, Shah PK, Fuster V. Coronary plaque disruption. Circulation
1995;92:657– 671.
17. Rioufol G, Finet G, Ginon I, André-Fouët X, Rossi R, Vialle E, Desjoyaux E, Convert G, Huret JF, Tabib A. Multiple atherosclerotic plaque
959
rupture in acute coronary syndrome: a three-vessel intravascular
ultrasound study. Circulation 2002;106:804 – 808.
18. Kotani JI, Mintz GS, Castagna MT, Pinnow E, Berzingi CO, Bui AB,
Pichard AD, Satler LF, Suddath WO, Waksman R, et al. Intravascular
ultrasound analysis of infarct-related and non-infarct-related arteries in
patients who presented with an acute myocardial infarction. Circulation 2003;107:2889 –2893.
19. Fujii K, Kobayashi Y, Mintz GS, Takebayashi H, Dangas G, Moussa
I, Mehran R, Lansky AJ, Kreps E, Collins M, et al. Intravascular
ultrasound assessment of ulcerated ruptured plaques: a comparison of
culprit and nonculprit lesions of patients with acute coronary syndromes and lesions in patients without acute coronary syndromes.
Circulation 2003;108:2473–2478.
Prevalence of Obstructive Coronary Artery Disease in Patients With
and Without Prior Stroke Undergoing Coronary Angiography
for Suspected Coronary Artery Disease
Rasham Sandhu, MDa, Wilbert S. Aronow, MDa,*, Rishi Sukhija, MDb, and Archana Rajdev, MDa
This study was conducted to investigate the prevalence and severity of obstructive coronary
artery disease (CAD) in 64 men and 38 women (mean age 71 ⴞ 9 years) with previous stroke
and in 102 age- and gender-matched patients with similar coronary risk factors without
previous stroke who underwent coronary angiography for chest pain. Obstructive CAD was
present in 100 of 102 patients (98%) with previous stroke and in 84 of 102 (82%) patients
without previous stroke (p <0.001). Obstructive 3-vessel CAD was present in 56 of 102 patients
(55%) with previous stroke and in 35 of 102 patients (34%) without previous stroke (p <0.005).
The prevalence of 2-vessel CAD and of 1-vessel CAD was not significantly different between
patients with and without previous stroke. In conclusion, patients with previous stroke have a
significantly higher prevalence of obstructive CAD and of obstructive 3-vessel CAD than ageand gender-matched patients with similar coronary risk factors without previous stroke who
undergo coronary angiography for chest pain. © 2007 Elsevier Inc. All rights reserved. (Am
J Cardiol 2007;100:960 –961)
Patients with previous stroke have a higher prevalence of
coronary artery disease (CAD) than patients without previous stroke.1–3 However, to the best of our knowledge, the
prevalence and severity of obstructive CAD documented by
coronary angiography have not previously been reported in
age- and gender-matched patients with similar coronary risk
factors with and without previous stroke. We report data
from 102 patients with previous stroke and from a control
group of 102 patients without previous stroke who underwent coronary angiography because of chest pain.
Methods and Results
The patients included 64 men and 38 women (mean age
71 ⫾ 9 years) with previous stroke and 102 age- and
gender-matched patients without previous stroke who underwent coronary angiography because of chest pain. The
patients with and without previous stroke had no significant
difference in the prevalence of current cigarette smoking
(36% vs 38%, respectively), the prevalence of systemic
hypertension (88% vs 86%, respectively), the prevalence of
diabetes mellitus (45% vs 42%, respectively), and the prevalence of hypercholesterolemia (75% vs 74%, respectively).
All 102 patients with previous stroke had cerebral infarction
documented by computed tomographic brain scans. Obstructive CAD was diagnosed if a patient had ⬎50% narrowing of ⱖ1 coronary artery.
Student’s t tests were used to analyze continuous variables. Chi-square tests were used to analyze dichotomous
variables.
a
Department of Medicine, Cardiology Division, New York Medical
College, Valhalla, New York; and bDepartment of Medicine, University of
Arkansas School of Medicine, Little Rock, Arkansas. Manuscript received
March 1, 2007; revised manuscript received and accepted April 24, 2007.
*Corresponding author: Tel: 914-493-5311; fax: 914-235-6274.
E-mail address: [email protected] (W.S. Aronow).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.033
Table 1
Prevalence and severity of obstructive coronary artery disease in patients
with and without previous stroke who underwent coronary angiography
for chest pain
Prior Stroke
No. of Coronary
Arteries Narrowed ⬎50%
Yes
(n ⫽ 102)
No
(n ⫽ 102)
p Value
1–3
3
2
1
100 (98%)
56 (55%)
23 (23%)
21 (21%)
84 (82%)
35 (34%)
21 (21%)
28 (28%)
⬍0.001
⬍0.005
NS
NS
Table 1 lists the prevalence and severity of obstructive
CAD in patients with and without previous stroke who
underwent coronary angiography because of chest pain.
Table 1 also lists levels of statistical significance.
Discussion
Patients with previous stroke have a higher prevalence of
CAD than patients without previous stroke.1–3 Patients with
cerebrovascular disease are also at increased risk for developing new coronary events.1,4 –7
To the best of our knowledge, the prevalence and severity of obstructive CAD documented by coronary angiography have not previously been reported in age- and gendermatched patients with similar coronary risk factors with and
without previous stroke who undergo coronary angiography
because of chest pain. The present study showed that the
prevalence of obstructive CAD was 98% in patients with
previous stroke and 82% in patients without previous stroke
(p ⬍0.001). The prevalence of obstructive 3-vessel CAD
was 55% in patients with previous stroke and 34% in patients without previous stroke (p ⬍0.005). The prevalence
of obstructive 2-vessel CAD and of obstructive 1-vessel
CAD was not significantly different between patients with
and without previous stroke.
www.AJConline.org
Coronary Artery Disease/Stroke and Coronary Artery Disease
Because patients with previous stroke have a significantly
higher prevalence of obstructive CAD and of obstructive
3-vessel CAD than patients without previous stroke, patients
with previous stroke should have more intensive treatment of
modifiable coronary risk factors and should be considered
more aggressively for coronary revascularization than patients
without previous stroke. Patients with previous stroke should
also be screened for myocardial ischemia.
1. Chimowitz MI, Mancini GBJ. Asymptomatic coronary artery disease in
patients with stroke. Prevalence, prognosis, diagnosis, and treatment.
Stroke 1992;23:433– 436.
2. Aronow WS, Ahn C. Prevalence of coexistence of coronary artery disease,
peripheral arterial disease, and atherothrombotic brain infarction in men
and women ⱖ62 years of age. Am J Cardiol 1994;74:64 – 65.
961
3. Ness J, Aronow WS. Prevalence of coexistence of coronary artery
disease, ischemic stroke, and peripheral arterial disease in older persons,
mean age 80 years, in an academic hospital-based geriatrics practice
J Am Geriatr Soc 1999;47:1255–1256.
4. Hertzer NR, Lees CD. Fatal myocardial infarction following carotid
endarterectomy: three hundred and thirty-five cases followed 6-11 years
after operation. Ann Surg 1981;194:212–218.
5. Chambers BR, Norris JW. Outcome in patients with asymptomatic neck
bruits. N Eng J Med 1986;315:860 – 865.
6. Aronow WS, Schoenfeld MR. Forty-five month follow-up of extracranial carotid arterial disease for new coronary events in elderly patients.
Coronary Artery Dis 1992;3:249 –251.
7. Aronow WS, Ahn C, Schoenfeld MR, Mercando AD, Epstein S. Prognostic significance of silent myocardial ischemia in patients ⬎61 years
of age with extracranial internal or common carotid arterial disease with
and without previous myocardial infarction. Am J Cardiol 1993;71:
115–117.
Examination of the Microcirculation Damage in Smokers Versus
Nonsmokers With Vasospastic Angina Pectoris
Takashi Ashikaga, MDa,*, Mitsuhiro Nishizaki, MDa, Hiroyuki Fujii, MDa, Saori Niki, MDa,
Shingo Maeda, MDa, Noriyoshi Yamawake, MDa, Yukio Kishi, MDb, and Mitsuaki Isobe, MDc
Endothelial dysfunction is considered one of the mechanisms underlying vasospastic
angina pectoris (VSA). It is also known that smokers have abnormalities in endothelial
dysfunction. Although smoking is a major risk factor for coronary artery disease, microvascular abnormalities have not been well shown. We investigated clinical characteristics
and coronary reactivity with adenosine triphosphate in smokers with VSA. Twenty-two
consecutive patients whose coronary spasm was documented in the left anterior descending
(LAD) coronary artery with acetylcholine were enrolled. Coronary blood flow responses
were also evaluated by intracoronary Doppler flow velocity recordings in the LAD coronary
artery. Average peak velocities (APVs) were measured at baseline and intracoronary
administration of adenosine triphosphate (50 ␮g) in 11 smokers (age 60 ⴞ 9 years; 8 men)
and 11 nonsmokers (age 61 ⴞ 10 years, 5 men). Coronary flow reserve (CFR) was
calculated by the ratio of baseline to hyperemic APV. Multivessel spasm was demonstrated
in 6 smokers and only 2 nonsmokers (p <0.05). APV at rest in smokers (13.4 ⴞ 3.0 cm/s)
was similar to that in nonsmokers (13.5 ⴞ 2.9 cm/s). However, CFR in smokers (2.6 ⴞ 0.7)
was significantly lower than in nonsmokers (3.4 ⴞ 0.8; p <0.05). In conclusion, multivessel
spasm was demonstrated in smokers in clinical settings, and microcirculation damage is
prominent in smokers with VSA. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol
2007;100:962–964)
Endothelial dysfunction is considered one of the mechanisms underlying vasospastic angina pectoris (VSA).1,2 Cigarette smoking has been known as an independent risk
factor for coronary spasm.3 In humans, cigarette smoke
exposure impaired endothelium-dependent vasodilation in
macrovascular beds such as coronary and brachial arteries
and in microvascular beds.4 – 6 However, the exact components of long-term cigarette smoking and the mechanisms
responsible for this association have not been clearly elucidated. Although acute cigarette smoke exposure may also
increase coronary artery vascular resistance, reducing coronary blood flow, less is known about the influence of
long-term smoking on coronary blood flow.7 Therefore, we
examined the effect of long-term cigarette smoking on macrovascular and microvascular coronary arteries.
Methods and Results
A consecutive series of 22 patients whose coronary spasm
was induced in the left anterior descending (LAD) coronary
artery were enrolled in this study. The study population
consisted of 11 current smokers (age 60 ⫾ 9 years; 8 men)
with ⬎20 pack-years history of smoking (1 pack-year ⫽ 20
cigarettes per day for 1 year) and 11 nonsmokers (age 61 ⫾
10 years; 5 men). Smokers refrained from cigarette smoking
a
Department of Cardiology, Yokohama Minami Kyosai Hospital,
Yokohama; bDepartment of Preventive Medicine, Tokyo Kyosai Hospital,
Tokyo; and cDepartment of Cardiology, Tokyo Medical and Dental University, Tokyo, Japan.
*Corresponding author: Tel: 81-45-782-2101; fax: 81-45-701-9159.
E-mail address:[email protected] (T. Ashikaga).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.035
ⱖ2 days before the study. None of the patients revealed any
significant coronary artery stenosis. Intracoronary injection
of acetylcholine into the right and left coronary arteries (20,
50, and 70 ␮g in the right coronary artery and 20, 50, and
100 ␮g in the left coronary artery) were performed in all
patients to induce coronary spasm. All patients with VSA
had episodes of spontaneous typical chest pain with or
without documentation of ischemic ST-T segment changes
on an electrocardiogram. Coronary spasm was defined as
total or subtotal coronary spasm with associated chest pain
and/or ischemic ST-T segment changes on an electrocardiogram.8,9 All patients had normal physical examinations and
no history of diabetes mellitus and renal disease. No patients
had abnormal findings on 12-lead electrocardiograms at
rest, 2-dimensional echocardiography, and Doppler echocardiography. The ethics committee at our institution approved the study protocol and all patients gave informed
consent before participation in this study.
All medications were withheld ⱖ72 hours except sublingual nitroglycerin, which was withheld ⱖ2 hours before
catheterization. The Judkins technique was used to perform
coronary angiography in the morning. After documentation
of coronary spasm, intracoronary Doppler flow measurements
were performed with a 0.014-mm Doppler guidewire
(FloWire; Cardiometrics, Mountain View, California). The
Doppler wire was positioned in the middle segment of the
LAD coronary artery. After coronary flow velocity at rest
was measured, maximal hyperemic flow velocity was induced by intracoronary injection of 50 ␮g of adenosine
triphosphate to determine the coronary flow reserve (CFR)
(10). Average peak velocities (APVs) were measured at
www.AJConline.org
Coronary Artery Disease/Vasospastic Angina in Smokers vs Nonsmokers
963
Table 1
Clinical characteristics of study patients
Variables
Age (yrs)
Women
Body mass index (kg/m2)
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Total cholesterol (mg/dl)
HDL cholesterol (mg/dl)
LDL cholesterol (mg/dl)
Triglyceride (mg/dl)
Smokers
(n ⫽ 11)
Nonsmokers
(n ⫽ 11)
60 ⫾ 8
3
23.8 ⫾ 2.9
130 ⫾ 16
70 ⫾ 10
185 ⫾ 52
44 ⫾ 14
137 ⫾ 39
143 ⫾ 70
61 ⫾ 10
6
23.5 ⫾ 3.5
129 ⫾ 19
72 ⫾ 13
205 ⫾ 26
62 ⫾ 23*
123 ⫾ 35
142 ⫾ 81
Figure 2. Baseline APV in patients with VSA.
Data are presented as means ⫾ SDs where applicable.
* Significant at p ⬍0.05.
LDL ⫽ low-density lipoprotein.
Figure 3. CFR in patients with VSA. ⴱp ⬍0.05 compared with smokers.
Discussion
Figure 1. Angiographic characteristics in patients with VSA. White bars,
single-vessel spasm; gray bars, double-vessel spasm; black bars, triplevessel spasm.
baseline and at intracoronary administration of adenosine
triphosphate. All patients were in sinus rhythm at the time of
study. Measurements were performed in the LAD coronary
artery. Mean arterial blood pressure and heart rate were
monitored during the study. All data are expressed means ⫾
1 SD. Statistical analysis was performed with a 2-tailed
Student’s t test for paired or unpaired observations. Fisher’s
exact test was used for discrete data. These analyses were
performed using the StatView 4.51J program (SAS, Cary,
North Carolina). A p value ⬍0.05 was considered significant.
Baseline clinical characteristics in smokers and nonsmokers are listed in Table 1. There were no significant
differences between the 2 groups in age, body mass index,
systolic blood pressure, diastolic blood pressure, total cholesterol, low-density lipoprotein cholesterol, and triglyceride level. Only high-density lipoprotein (HDL) cholesterol
was higher in nonsmokers compared with smokers. Multiple-vessel spasm was demonstrated in 6 smokers (2 with
double-vessel spasm and 4 with triple-vessel spasm) compared with only 2 nonsmokers (1 with double-vessel spasm
and 1 triple-vessel spasm; p ⬍0.05; Figure 1).
APV at rest in smokers (13.0 ⫾ 3.0 cm/s) was similar to
that in nonsmokers (13.5 ⫾ 2.9 cm/s; Figure 2). However,
CFR in smokers (2.6 ⫾ 0.7) was significantly lower than in
nonsmokers (3.4 ⫾ 0.8; p ⬍0.05; Figure 3).
To our knowledge this is the first study to demonstrate that
microcirculation abnormality is prominent in smokers with
VSA. From the responses to acetylcholine, which is an
endothelium-dependent vasodilator, endothelial dysfunction
was demonstrated in the LAD coronary artery in this study.
Abnormal CFR can be a result of narrowing of the epicardial arteries, as well microcirculation abnormality.11 Because angiography showed no detectable coronary artery
stenosis, coronary microcirculation abnormality existed in
the LAD coronary artery. The mechanism of adenosine
triphosphate was believed to be endothelium-dependent and
endothelium-independent vasodilation.12 Endothelium-independent factors could not be ruled out in this study.
Although previous reports have shown that smoking is
the only major risk factor for coronary artery diseases, the
association between long-term smoking and microcirculation abnormality is not as well defined.13 Vita et al5 showed
that coronary response to intracoronary acetylcholine did
not correlate with smoking. Czernin et al7 showed that
myocardial blood flow and flow reserve were similar in
terms of long-term smoking in young healthy volunteers.
However, Kauffman et al14 demonstrated that smoking affects the regulation via epicardial arteries and coronary
microcirculation in healthy volunteers. However, they could
not exclude epicardial coronary artery stenosis because the
study was performed with coronary angiography rather than
positron emission tomography. In this study, long-term
smoking was demonstrated to regulate epicardial and microcirculation abnormality in patients with VSA.
Cigarette smoke contains a large number of oxidants.6
There is a general consensus that cigarette smoke targets the
vascular endothelial cells. It is believed that nitric oxide and
964
The American Journal of Cardiology (www.AJConline.org)
its vasodilatory function are altered by cigarette smoking.1
In addition, these oxidants may cause endothelial and microcirculation dysfunction through nitric oxide synthesis.15
However, the exact oxidant in our study is unknown, and
further study will be required.
Although total and low-density lipoprotein cholesterol
concentrations were similar in smokers and nonsmokers,
HDL concentrations were slightly lower in smokers. This is
not unexpected, because cigarette smoking is known to be
associated with a selective reduction in HDL cholesterol
concentration.16 This lower HDL cholesterol level was believed to be a consequence of smoking rather than an independent coronary risk factor. Because HDL cholesterol is
documented to increase the endothelial production of prostacyclin or reduce its catabolism, resulting in coronary vasodilation and its antiplatelet effects, this potential role of
HDL cholesterol may partially explain our results.17
Multivessel spasm was demonstrated to be a high risk
among subjects with variant angina.18 Our data showed that
multivessel spasm was frequent in smokers compared with
nonsmokers. Smoking may deteriorate not only coronary
atherosclerosis via endothelial dysfunction, but may also
confer a poor prognosis in patients with VSA.10
1. Kugiyama K, Yasue H, Ohgushi M, Motoyama T, Kawano H, Inobe
Y, Hirashima O, Sugiyama S. Deficiency in nitric oxide bioactivity in
epicardial coronary arteries of cigarette smokers. J Am Coll Cardiol
1996;28:1161–1167.
2. Sumida H, Watanabe H, Kugiyama K, Ohgushi M, Matsumura T,
Yasue H. Does passive smoking impair endothelium-dependent coronary artery dilation in women? J Am Coll Cardiol 1998;31:811– 815.
3. Sugiuchi M, Takatsu F. Cigarrete smoking is a major risk factor for
coronary spasm. Circulation 1993;87:76 –79.
4. Ijzerman RG, Serne EH, van Weissenbruch MM, van Weissenbruch
MM, de Jongh RT, Stehouwer CD. Cigarette smoking is associated
with acute impairment of microvascular function in humans. Cli Sci
(Lond) 2003;104:247–252.
5. Vita JA, Treasure CB, Nabel EG, McLenachan JM, Fish D, Yueng
AC, Vekshtein VI, Selwyn AP, Ganz P. Coronary vasomotor response
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
to acetylcholine relates to risk factors for coronary artery disease.
Circulation 1990;81:491– 497.
Church DF, Pryor WA. Free radical chemistry of cigarette some and
its toxicological implications. Environ Health Perspect 1985;64:
111–126.
Czernin J, Sun K, Brunken R, Böttcher M, Phelps M, Schelbert H.
Effect of acute and long-term smoking on myocardial blood flow and
flow reserve. Circulation 1995;91:2891–2897.
Ashikaga T, Nishizaki M, Arita M, Yamawake N, Suzuki M, Hashimoto Y, Kishi Y, Numano F, Hiraoka M. Effect of dipyridamole on
QT dispersion in vasospastic angina. Am J Cardiol 1999;84:807– 810.
Suzuki M, Nishizaki M, Arita M, Ashikaga T, Yamawake N, Kakuta
T, Numano F, Hiraoka M. Increased QT dispersion in patients with
vasospastic angina. Circulation 1998;98:435– 440.
Sonoda S, Takeuchi M, Nakashima Y, Kuroiwa A. Safety and optimal
dose of intracoronary adenosine 5’-triphosphate for the measurement
of coronary flow reserve. Am Heart J 1998;35:621– 627.
Uren NG, Melin JA, De Bruyne B, Wijins W, Baudhuin T, Camici PG.
Relation between myocardial blood flow and the severity of coronary
artery stenosis. N Engl J Med 1994;330:1782–1788.
Reis SE, Holubkov R, Lee JS, Sharaf B, Reichek N, Rogers WJ, Walsh
EG, Fuisz AR, Kerensky R, Detre KM, Sopko G, Pepine CJ. Coronary
flow velocity response to adenosine characterizes coronary microvascular function in women with chest pain and no obstructive coronary
disease. J Am Coll Cardiol 1999;33:1469 –1475.
Ramsdale DR, Faragher EB, Bray CL, Benett DH, Ward C, Beton DC.
Smoking and coronary artery disease assessed by routine coronary
arteriography. Br Med J 1985;290:197–200.
Kauffmann PA, Gnecchi-Ruscone T, di Terlizzi M, Schäfers KP,
Lüscher TF, Garnici PG. Coronary heart disease in smokers: vitamin
C restores coronary microcirculatory function. Circulation 2000;102:
1233–1238.
Hein WH, Kuo L. LDLs impair vasomotor function of the coronary
microcirculation: role of superoxide anions. Circ Res 1998;83:404 –
414.
Shennann NM, Seed M, Wynn V. Variation in serum lipid and lipoprotein levels associated with changes in smoking behavior in nonobese caucasian males. Atherosclerosis 1985;58:17–25.
Mineo C, Deguchi H, Griffin JH, Shaul PW. Endothelial and antithrombotic avtions of HDL. Circ Res 2006;98:1352–1364.
Onaka H, Hirota Y, Shimada S, Suzuki S, Kono T, Suzuki J, Sakai Y,
Kawamura K. Prognostic significance of the pattern of multivessel
spasm with variant angina. Jpn Circ J 1999;63:509 –513.
Correlates of Clinical Restenosis Following Intracoronary
Implantation of Drug-Eluting Stents
Probal Roy, MD, Teruo Okabe, MD, Tina L. Pinto Slottow, MD, Daniel H. Steinberg, MD,
Kimberly Smith, BS, Rebecca Torguson, BS, Zhenyi Xue, MS, Natalie Gevorkian, MD,
Lowell F. Satler, MD, Kenneth M. Kent, MD, William O. Suddath, MD,
Augusto D. Pichard, MD, and Ron Waksman, MD*
Despite significant decreases in restenosis and repeated intervention achieved using drugeluting stents (DESs), the benefit has not been homogenous across all patient and lesion
subsets. Identification of correlates of DES restenosis may allow a differing management
approach and lead to improved patient outcomes. The study population consisted of 3,535
consecutive patients (5,046 lesions) who underwent successful sirolimus- or paclitaxeleluting stent implantation for >1 native coronary artery or bypass graft lesion from April
2003 to September 2006. From this cohort, 197 patients (237 lesions) were identified to have
in-stent restenosis (ISR) requiring revascularization within 12 months of stent implantation. This group was compared with the remainder of the patient population. Logistic
regression analysis was performed to identify independent predictors of DES ISR. Independent correlates of DES ISR using multivariate analysis included both clinical and
procedural factors. Clinical predictors were age, hypertension, and unstable angina. Procedural predictors were left anterior descending artery intervention, number of stents
implanted, stented length/lesion, and lack of intravascular ultrasound guidance. Implantation of >3 stents was associated with a significantly higher restenosis risk (9.7% vs 5.1%;
p ⴝ 0.0003). A 10-mm increase in stented length was associated with an adjusted odds ratio
of 1.18 (95% confidence interval 1.03 to 1.35). Diabetes, stent diameter, and stent type were
found not to be predictive of DES ISR. In conclusion, correlates of DES ISR included both
clinical and procedural factors. Limiting the number of stents and stented length, in
addition to intravascular ultrasound guidance, may minimize DES ISR. © 2007 Elsevier
Inc. All rights reserved. (Am J Cardiol 2007;100:965–969)
Percutaneous treatment of coronary artery disease with
bare-metal stent implantation remains limited by in-stent
restenosis (ISR).1 Bare-metal stent ISR is nonuniform, with
certain patient and lesion subsets more prone than others
(patients with diabetes, small vessels, and long lesions).2,3
By allowing the local delivery of antiproliferative drugs that
suppress neointimal growth,4,5 drug-eluting stents (DESs)
have realized dramatic decreases in ISR and repeated interventions.6 – 8 Despite this major advance, the efficacy of
DESs in decreasing restenosis has not been uniform across
all patient and lesions subsets. The identification of predictors of DES ISR assumes greater importance in an environment in which nondiscriminatory use was cautioned by
reports of increased mortality and nonfatal myocardial infarction.9,10 To date, the most consistently reported predictors of DES ISR include vessel diameter, lesion length, and
stent type.11,12 Other correlates reported include female gender,13 diabetes, and ostial, ISR, and left anterior descending
artery lesions.14 Identification of correlates of DES ISR may
allow a differing management approach and improved paDivision of Cardiology, Washington Hospital Center, Washington, DC.
Manuscript received March 1, 2007; revised manuscript received and
accepted April 13, 2007.
*Corresponding author: Tel.: 202-877-2812; fax: 202-877-2715.
E-mail address: [email protected] (R. Waksman).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.036
tient outcomes. Studies performed to date included angiographic follow-up, which has the potential to bias revascularization rates. We report correlates of DES ISR requiring
clinically driven repeated intervention from a large series of
consecutive patients undergoing DES implantation in a single medical center.
Methods
The study population consisted of 3,535 consecutive patients (5,046 lesions) who underwent successful DES implantation for ⱖ1 coronary artery or bypass graft lesion
from April 2003 to September 2006. From this cohort, 197
patients (237 lesions) were identified to have DES ISR
requiring revascularization within 12 months of the procedure. This group was compared with the remaining 3,338
patients (4,809 lesions) to identify correlates of DES ISR.
For purposes of the study, all patients presenting with stent
thrombosis were excluded and only the first procedure for
each patient was considered. Other stented segments that
remained patent in patients with DES ISR were also excluded from analysis.
All patients underwent their procedures at Washington
Hospital Center, Washington, DC, a tertiary referral hospital with 11 catheterization laboratories serviced by 31 independent interventional cardiologists. All patients signed a
www.AJConline.org
966
The American Journal of Cardiology (www.AJConline.org)
Table 2
Procedural characteristics
Table 1
Baseline Clinical Characteristics
DES ISR
Variable
Men
Age (yrs)
Diabetes mellitus
Insulin-requiring diabetes
mellitus
Smoker
Systemic hypertension
Dyslipidemia*
Family history
Dialysis requiring renal
failure
Previous myocardial
infarction
Previous coronary artery
bypass surgery
Previous percutaneous
coronary intervention
Peripheral vascular disease
Unstable angina pectoris
Acute myocardial infarction
Cardiogenic shock
Left ventricular ejection
fraction
DES ISR
Yes
(n ⫽ 197)
No
(n ⫽ 3,338)
p Value
127 (64.5%)
62.9 ⫾ 11.0
65 (33.2%)
21 (10.7%)
2,178 (65.2%)
65.5 ⫾ 11.6
1,190 (36.0%)
390 (11.8%)
NS
0.002
NS
NS
32 (16.2%)
172 (87.3%)
172 (88.7%)
93 (49.5%)
4 (2.1%)
609 (18.2%)
2,653 (79.9%)
2,803 (85.1%)
1,788 (56.0%)
97 (2.9%)
NS
0.01
NS
NS
NS
65 (35.5%)
1,049 (33.2%)
NS
35 (17.9%)
584 (17.6%)
NS
58 (30.4%)
846 (26.4%)
NS
38 (19.6%)
100 (51.0%)
25 (12.8%)
7 (3.6%)
50 ⫾ 12
527 (16.0%)
1,463 (43.9%)
370 (11.1%)
91 (2.8%)
48 ⫾ 14
NS
NS
NS
NS
0.03
* Includes patients with a previously documented diagnosis of dyslipidemia. The patient may be treated with diet or medication. A new diagnosis
can be made during this hospitalization with increased cholesterol ⬎250
mg/dl. Does not include increased triglycerides.
consent document for the procedure, and the study was
conducted under local institutional review board approval.
Percutaneous coronary intervention was performed using a standard technique through the femoral approach in
most patients. Patients received either sirolimus-eluting
stents (Cypher, Johnson & Johnson Cordis Corp., Warren, New Jersey) with diameters of 2.5 to 3.5 mm and
lengths of 8 to 33 mm or paclitaxel-eluting stents (Taxus,
Boston Scientific Corp., Natick, Massachusetts) with diameters of 2.5 to 3.5 mm and lengths of 8 to 32 mm, or
both. Use of adjunctive devices (intravascular ultrasound
[IVUS], atherectomy, distal protection, and so on) was at
the operators’ discretion. All patients were treated with
aspirin 325 mg before percutaneous coronary intervention and loaded with clopidogrel 300 to 600 mg orally if
not on a maintenance dose. Dual antiplatelet therapy
(aspirin 81 mg and clopidogrel 75 mg/day) was strongly
recommended to all patients for ⱖ12 months. Aspirin therapy was to continue indefinitely thereafter. During percutaneous coronary intervention, patients underwent systemic
anticoagulation with either bivalirudin (a bolus of 0.75
mg/kg, followed by an intravenous infusion of 1.75 mg/kg/
hour) or unfractionated heparin (a bolus of 40 U/kg and
additional heparin to achieve an activated clotting time of
250 to 300 seconds). Platelet glycoprotein IIb/IIIa inhibitors
were administrated at the operators’ discretion.
ISR was defined as recurrent stenosis in the stented
segment requiring revascularization for clinical indications
Variable
Target vessel
Left main
Left anterior descending
artery
Left circumflex
Right
Saphenous vein graft
Lesion location
Ostial
Proximal
Mid
Distal
Lesion type (ACC/AHA
classification)
Type A
Type B
Type C
ISR
Procedural details
No. of lesions treated
Sirolimus-eluting stent only*
Paclitaxel-eluting stent only*
Stent diameters (mm)
Stented length/lesion (mm)
No. of implanted stents
IVUS*
Direct stenting
After dilatation
Glycoprotein IIb/IIIa use*
Procedural length (min)†
Angiographic success
Yes
(n ⫽ 237)
No
(n ⫽ 4,809)
p
Value
4 (1.7%)
121 (51.1%)
90 (1.9%)
1,834 (38.1%)
NS
⬍0.001
41 (17.3%)
57 (24.1%)
12 (5.1%)
1,100 (22.9%)
1,538 (32.0%)
225 (4.7%)
0.05
0.01
NS
12 (5.1%)
112 (47.7%)
82 (34.9%)
28 (11.9%)
188 (4.0%)
2,144 (45.7%)
1,758 (37.4%)
587 (12.5%)
NS
NS
NS
NS
10 (4.5%)
161 (72.5%)
51 (23.0%)
19 (8.1%)
292 (6.5%)
3,275 (72.7%)
939 (20.8%)
236 (4.9%)
NS
NS
NS
0.03
2.05 ⫾ 1.0
120 (60.9%)
57 (28.9%)
2.98 ⫾ 1.3
25.44 ⫾ 12.8
1.78 ⫾ 0.9
132 (69.1%)
73 (31.5%)
50 (26.0%)
31 (15.7%)
63.30 ⫾ 32.6
232 (97.9%)
1.79 ⫾ 2.0
1,993 (59.7%)
1,115 (33.4%)
3.03 ⫾ 0.6
23.61 ⫾ 11.6
1.46 ⫾ 0.8
2,359 (73.6%)
1,715 (36.1%)
974 (24.4%)
421 (12.7%)
59.52 ⫾ 46.2
4,675 (98.6%)
0.002
NS
NS
NS
0.02
⬍0.001
NS
NS
NS
NS
NS
NS
* Variables are patient based.
Minimum ⫾ SD.
ACC/AHA ⫽ American College of Cardiology/American Heart Association.
†
(symptoms or demonstrable ischemia on stress testing).
Target lesion revascularization was defined as revascularization, either percutaneous or surgical, for a stenosis within
the stent or in the 5-mm segments proximal or distal to the
stent. Angiographic success was defined as residual stenosis
⬍30% with Thrombolysis In Myocardial Infarction grade 3
flow.
Demographic, clinical, and procedural data were collected and entered into a prospective database. These data
were obtained using hospital chart review by independent
research personnel blinded to the study objectives. All data
management and analysis were performed by a dedicated
data coordinating center (Data Center, Cardiovascular Research Institute, Washington, DC). Clinical follow-up was
performed at 1, 6, and 12 months by trained quality assurance nurses who worked exclusively with the database to
determine clinical events after percutaneous coronary intervention. Clinical follow-up was performed using telephone
contact or office visit. A committee independently adjudicated all subsequent clinical events.
Statistical analysis was performed using SAS, version
Coronary Artery Disease/Correlates of DES Clinical Restenosis
967
Table 3
Predictors of drug-eluting stent in-stent restenosis (ISR) using univariate
analysis
Variable
OR
CI
p Value
Age
Diabetes mellitus
Systemic hypertension
Dyslipidemia
Unstable angina pectoris
Left anterior descending artery
ISR lesion
Length of procedure
No. of lesions treated
No. of stents
Stented length/lesion
Stent diameter
Sirolimus-eluting stent
IVUS guidance
Glycoprotein IIb/IIIa inhibitor use
0.98
0.88
1.73
1.37
1.33
1.37
1.48
1.0
1.03
1.55
1.03
0.99
1.05
0.80
1.28
0.97–0.99
0.65–1.20
1.13–2.65
0.87–2.15
1.0–1.78
1.03–1.83
0.89–2.45
1.0–1.004
0.99–1.08
1.33–1.80
1.01–1.04
0.75–1.31
0.78–1.41
0.58–1.1
0.86–1.91
0.003
NS
0.01
NS
NS
0.03
NS
NS
NS
⬍0.0001
⬍0.0001
NS
NS
NS
NS
Figure 1. Bar graph shows the increased risk of DES ISR with implantation
of ⱖ3 stents.
Table 4
Predictors of drug-eluting stent in-stent restenosis (ISR) using
multivariate analysis
Variable
OR
CI
p Value
Age
Diabetes mellitus
Systemic hypertension
Dyslipidemia
Unstable angina pectoris
Left anterior descending artery
ISR lesion
Length of procedure
No. of lesions treated
No. of stents
Stented length/lesion
Stent diameter
Sirolimus-eluting stent
IVUS guidance
Glycoprotein IIb/IIIa inhibitor use
0.98
0.80
2.28
1.17
1.44
1.43
1.35
1.0
1.04
1.34
1.02
1.03
1.22
0.68
1.23
0.96–0.99
0.57–1.12
1.36–3.82
0.70–1.97
1.04–2.0
1.03–1.97
0.78–2.34
1.0–1.005
0.95–1.13
1.10–1.65
1.003–1.03
0.80–1.34
0.87–1.71
0.48–0.97
0.79–1.93
0.0006
NS
0.002
NS
0.03
0.03
NS
NS
NS
0.005
0.015
NS
NS
0.035
NS
9.1 (SAS Institute, Cary, North Carolina). Continuous variables were expressed as mean ⫾ SD and compared using
Student’s t test. Categorical variables were expressed as
percentage and compared using either chi-square test or
Fisher’s exact test. Univariate and multivariate analyses
were performed using logistic regression with variables in
Tables 1 and 2 to determine independent predictors of stent
DES ISR. A p value ⬍0.05 was considered statistically
significant.
Results
Baseline clinical and procedural characteristics are listed in
Tables 1 and 2. Patients with DES ISR were more likely to
be younger and hypertensive and have a higher left ventricular ejection fraction. Other clinical parameters were similar
between the 2 groups. There was a trend toward more
unstable angina in patients with DES ISR.
From a procedural viewpoint, patients with restenosis
were more likely to have left anterior descending artery and
ISR lesions treated, more lesions treated, a larger number of
stents implanted, and longer stented lengths per lesion.
Figure 2. Bar graph shows the increased risk of DES ISR with increasing
stented length.
Predictors of DES ISR using univariate analysis are
listed in Table 3. Clinical predictors identified were age and
hypertension. Procedural predictors were left anterior descending artery lesion, number of stents implanted, and
stented length/lesion.
Multivariate analysis identified the clinical predictors of
DES ISR as age, hypertension, and unstable angina. Procedural predictors were left anterior descending artery lesion,
number of stents implanted (odds ratio [OR] 1.34, confidence interval [CI] 1.10 to 1.65, p ⫽ 0.005), stented length/
lesion (OR 1.02, CI 1.003 to 1.03, p ⫽ 0.015), and IVUS
guidance (OR 0.68, CI 0.48 to 0.97, p ⫽ 0.035). In particular, diabetes, stent type, and stent diameter were found not
to be correlates of restenosis using multivariate analysis
(Table 4).
The number of stents implanted was assessed comparing
patients who received ⱕ2 stents (per Food and Drug Administration approval) versus patients who received ⱖ3
stents. Those receiving ⱖ3 stents were at a significantly
higher restenosis risk (9.7% vs 5.1%; p ⫽ 0.0003; Figure 1).
Restenosis risk increased with incremental increases in
stented length. A 10-mm increase in stented length was
associated with an adjusted OR of 1.18 (95% CI 1.03 to
1.35; Figure 2).
968
The American Journal of Cardiology (www.AJConline.org)
Discussion
The present study identified a number of correlates of restenosis after intracoronary implantation of DESs in an
unselected population. Unlike previous studies, this study
was performed without angiographic follow-up and focused
solely on clinical restenosis. Hence, the findings were free
from bias induced by angiographic follow-up and were
reflective of real-world practice. Correlates of DES ISR
included both clinical and procedural factors. Clinical correlates of DES ISR were age, hypertension, and unstable
angina. Procedural correlates were left anterior descending
artery lesion, number of stents implanted, stented length/
lesion, and lack of IVUS guidance. These findings suggested that minimization of number of stents and stented
length, in addition to IVUS-guided percutaneous coronary
intervention, may decrease DES ISR. The discussion focuses on these modifiable procedural correlates of DES ISR.
Substantial decreases in both clinical and angiographic
restenosis with DESs in pivotal randomized studies6 – 8 and
post-DES approval registries15 led to renewed enthusiasm
for the percutaneous treatment of patients with coronary
artery disease. This resulted in physicians adopting a more
liberal approach to complex disease, in particular, longer
lesions. Although DESs were proved superior to bare-metal
stents in the treatment of patients with diffuse disease in
small single-center studies,16 –18 lesion or stented length
remained a consistent predictor of DES ISR.12–14 Although
complete lesion coverage remained important to avoid edge
restenosis,19 our findings suggested that the generous stent/
lesion ratios13 adopted in the DES era should be curtailed to
limit DES ISR. Alternatively, surgical revascularization bypassing lengthy lesions may provide better long-term patient outcomes, particularly in light of the increased risk of
stent thrombosis with longer stented lengths.20 The number
of stents implanted was not previously identified as a correlate of DES ISR. The larger number of stents in patients
with DES ISR may correlate more with use of overlapping
stents for the treatment of diffuse disease rather than stenting of multiple discrete lesions. A meta-analysis of 5 clinical trials studying sirolimus-eluting stents showed that
overlapping sirolimus-eluting stents were safe and efficacious in the reduction of restenosis compared with overlapping bare-metal stents. Nevertheless, at 1 year in the same
meta-analysis, overlapping sirolimus-eluting stents were associated with a target lesion revascularization rate of 4.7%,
target vessel revascularization rate of 9.5%, and target vessel failure rate of 11.6%.21 The present study suggested that
limiting the number of stents will aid in minimizing DES
ISR.
The role of IVUS guidance in percutaneous coronary
intervention remains uncertain, particularly in the DES era.
Studies from the bare-metal stent era suggested a benefit of
IVUS guidance in percutaneous coronary intervention. In
the Can Routine Ultrasound Influence Stent Expansion
(CRUISE)22 and Thrombocyte activity evaluation and effects of Ultrasound guidance in Long Intracoronary stent
Placement (TULIP)23 studies, ultrasound-guided percutaneous coronary intervention compared with angiographic
guidance alone was associated with significantly better angiographic and clinical outcomes. No randomized trial
studying IVUS guidance in percutaneous coronary intervention with DES exists. Our findings suggested a benefit of
IVUS guidance in the prevention of DES ISR. Because DES
ISR was predominantly focal, mechanical causes were believed to be contributory in most cases. IVUS remained the
best modality to assess stent architecture after the procedure
and hence ensure an optimal result. Stent underexpansion
was the most important mechanical cause leading to DES
ISR. Fujii et al24 showed a correlation between minimal
stent area after the procedure on IVUS examination and
DES ISR in patients undergoing sirolimus-eluting stent implantation for the treatment of ISR. Lower maximal inflation pressures correlated with DES underexpansion and the
need for target lesion revascularization.25 Conversely, overexpansion of the stent can result in loss of stent integrity,
fracture, and inhomogeneous drug delivery,26 which can
result in DES ISR.27 IVUS-guided percutaneous coronary
intervention in the DES era can potentially decrease restenosis by eliminating stent underexpansion, allowing appropriate stent diameters to decrease the need for aggressive
postdilation, and assessing lesion length, thus ensuring complete lesion coverage.
This study failed to identify diabetes, stent type, and
stent diameter as correlates of DES ISR. Diabetes was not
consistently reported as a predictor of DES ISR.11–13 Our
findings agreed with this observation and suggested that
DES potentially ameliorated the propensity for ISR in this
subgroup of patients. Studies identifying correlates of DES
ISR with both sirolimus- and paclitaxel-eluting stents
showed paclitaxel-eluting stents to be an independent predictor of ISR.11,12 These studies included angiographic follow-up, which has the potential to bias revascularization
rates against paclitaxel-eluting stents. In the present study,
in which revascularization procedures were clinically
driven, stent type was not predictive of ISR. In this study,
the smallest stent diameter was 2.5 mm. Hence, assessment
of smaller caliber vessels was unable to be performed. This
may explain the absence of stent diameter as a predictor.
Alternatively, stent diameter did not prove predictive of ISR
in the study, which may reflect the efficacy of DESs in
smaller vessels.
This is an observational single-center study with the
inherent limitations of this kind of analysis. As mentioned,
assessment of smaller caliber vessels (⬍2.5 mm diameter)
was not made in the study.
1. Al Suwaidi J, Berger PB, Holmes DR Jr. Coronary artery stents. JAMA
2000;284:1828 –1836.
2. Kastrati A, Schomig A, Elezi S, Schuhlen H, Dirschinger J, Hadamitzky M, Wehinger A, Hausleiter J, Walter H, Neumann FJ. Predictive
factors of restenosis after coronary stent placement. J Am Coll Cardiol
1997;30:1428 –1436.
3. Cutlip DE, Chauhan MS, Baim DS, Ho KK, Popma JJ, Carrozza JP,
Cohen DJ, Kuntz RE. Clinical restenosis after coronary stenting:
perspectives from multicenter clinical trials. J Am Coll Cardiol 2002;
40:2082–2089.
4. Marx SO, Marks AR. Bench to bedside: the development of rapamycin
and its application to stent restenosis. Circulation 2001;104:852– 855.
5. Rowinsky EK, Donehower RC. Paclitaxel (taxol). N Engl J Med
1995;332:1004 –1014.
6. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin
M, Colombo A, Schuler G, Barragan P, Guagliumi G, Molnar F,
Falotico R; RAVEL Study Group. Randomized study with the sirolimus-coated Bx Velocity balloon-expandable stent in the treatment of
Coronary Artery Disease/Correlates of DES Clinical Restenosis
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
patients with de novo native coronary artery lesions. A randomized
comparison of a sirolimus-eluting stent with a standard stent for
coronary revascularization. N Engl J Med 2002;346:1773–1780.
Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR,
O’Shaughnessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein
PS, Jaeger JL, Kuntz RE; SIRIUS Investigators. Sirolimus-eluting
stents versus standard stents in patients with stenosis in a native
coronary artery. N Engl J Med 2003;349:1315–1323.
Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann
JT, Turco M, Caputo R, Bergin P, Greenberg J, Popma JJ, Russell ME;
TAXUS-IV Investigators. A polymer-based, paclitaxel-eluting stent in
patients with coronary artery disease. N Engl J Med 2004;350:221–
231.
Camenzind E, Steg PG, Wijns W. Stent thrombosis late after implantation of first-generation drug-eluting stents: a cause for concern.
Circulation 2007;115:1440 –1445.
Pfisterer M, Brunner-La Rocca H, Buser P, Rickenbacher P, Hunziker
P, Mueller C, Jeger R, Bader F, Osswald S, Kaiser C; BASKET-LATE
Investigators. Late clinical events after clopidogrel discontinuation
may limit the benefit of drug-eluting stents; an observational study of
drug-eluting versus bare-metal stents. J Am Coll Cardiol 2006;48:
2584 –2591.
Kastrati A, Dibra A, Mehilli J, Mayer S, Pinieck S, Pache J, Dirschinger J, Schomig A. Predictive factors of restenosis after coronary
implantation of sirolimus- or paclitaxel-eluting stents. Circulation
2006;113:2293–2300.
Lee CW, Park DW, Lee BK, Kim YH, Hong MK, Kim JJ, Park SW,
Park SJ. Predictors of restenosis after placement of drug-eluting stents
in one or more coronary arteries. Am J Cardiol 2006;97:506 –511.
Berenguer A, Mainar V, Bordes P, Valencia J, Gomez S, Lozano T.
Incidence and predictors of restenosis after sirolimus-eluting stent
implantation in high-risk patients. Am Heart J 2005;150:536 –542.
Lemos PA, Hoye A, Goedhart D, Arampatzis CA, Saia F, van der
Giessen WJ, McFadden E, Sianos G, Smits PC, Hofma SH, et al.
Clinical, angiographic, and procedural predictors of angiographic restenosis after sirolimus-eluting stent implantation in complex patients:
an evaluation from the Rapamycin-Eluting Stent Evaluated At Rotterdam Cardiology Hospital (RESEARCH) study. Circulation 2004;109:
1366 –1370.
Ong AT, Serruys PW, Aoki J, Hoye A, van Mieghem CA, RodriguezGranillo GA, Valgimigli M, Sonnenschein K, Regar E, van der Ent M,
et al. The unrestricted use of paclitaxel- versus sirolimus-eluting stents
for coronary artery disease in an unselected population: one-year
results of the Taxus-Stent Evaluated at Rotterdam Cardiology Hospital
(T-SEARCH) registry. J Am Coll Cardiol 2005;45:1135–1141.
Kim YH, Park SW, Lee CW, Hong MK, Gwon HC, Jang Y, Lee MM,
Koo BK, Oh DJ, Seung KB, et al. Comparison of sirolimus-eluting
stent, paclitaxel-eluting stent, and bare metal stent in the treatment of
long coronary lesions. Catheter Cardiovasc Interv 2006;67:181–187.
969
17. Tsagalou E, Chieffo A, Iakovou I, Ge L, Sangiorgi GM, Corvaja N,
Airoldi F, Montorfano M, Michev I, Colombo A. Multiple overlapping
drug-eluting stents to treat diffuse disease of the left descending
coronary artery. J Am Coll Cardiol 2005;45:1570 –1573.
18. Aoki J, Ong AT, Rodriguez Granillo GA, McFadden EP, van
Mieghem CA, Valgimigli M, Tsuchida K, Sianos G, Regar E, de
Jaegere PP, et al. “Full metal jacket” (stented length ⬎ or ⫽ 64 mm)
using drug-eluting stents for de novo coronary artery lesions. Am
Heart J 2005;150:994 –999.
19. Sakurai R, Ako J, Morino Y, Sonoda S, Kaneda H, Terashima M,
Hassan AH, Leon MB, Moses JW, Popma JJ, et al; SIRIUS Trial
Investigators. Predictors of edge stenosis following sirolimus-eluting
stent deployment (a quantitative intravascular ultrasound analysis from
the SIRIUS trial). Am J Cardiol 2005;96:1251–1253.
20. Park DW, Park SW, Park KH, Lee BK, Kim YH, Lee CW, Hong MK,
Kim JJ, Park SJ. Frequency of and risk factors for stent thrombosis
after drug-eluting stent implantation during long-term follow-up. Am J
Cardiol 2006;98:352–356.
21. Kereiakes DJ, Wang H, Popma JJ, Kuntz RE, Donohoe DJ, Schofer J,
Schampaert E, Meier B, Leon MB, Moses JW. Periprocedural and late
consequences of overlapping Cypher sirolimus-eluting stents: pooled
analysis of five clinical trials. J Am Coll Cardiol 2006;48:21–31.
22. Fitzgerald PJ, Oshima A, Hayase M, Metz JA, Bailey SR, Baim DS,
Cleman MW, Deutsch E, Diver DJ, Leon MB, et al. Final results of the
Can Routine Ultrasound Influence Stent Expansion (CRUISE) study.
Circulation 2000;102:523–530.
23. Oemrawsingh PV, Mintz GS, Schalij MJ, Zwinderman AH, Jukema
JW, van der Wall EE; TULIP Study. Thrombocyte activity evaluation
and effects of Ultrasound guidance in Long Intracoronary stent Placement: intravascular ultrasound guidance improves angiographic and
clinical outcome of stent implantation for long coronary artery stenoses: final results of a randomized comparison with angiographic guidance (TULIP Study). Circulation 2003;107:62– 67.
24. Fujii K, Mintz GS, Kobayashi Y, Carlier SG, Takebayashi H, Yasuda
T, Moussa I, Dangas G, Mehran R, Lansky AJ, et al. Contribution of
stent underexpansion to recurrence after sirolimus-eluting stent implantation for restenosis. Circulation 2004;109:1085–1088.
25. Ako J, Morino Y, Terashima M, Honda Y, Sonoda S, Leon MB, Moses
JW, Yock PG, The SIRIUS Investigators. Optimal geometry is still
important with sirolimus eluting stents: incomplete stent expansion as
a risk for target lesion revascularization. J Am Coll Cardiol 2004;
43(suppl A):84A.
26. Schofer J, Schluter M. Coronary restenosis after implantation of drugeluting stents. Minerva Cardioangiol 2005;53:43– 48.
27. Iakovou I, Stankovic G, Orlic D, Vitrella G, Sangiorgi G, Corvaja N,
Chieffo A, Michev I, Airoldi F, Spanos V, Colombo A. Overdilation
of Cypher 3.0mm 6 cells stent: clinical consequence. J Am Coll
Cardiol 2004;43(suppl A):45A.
Comparison of Drug-Eluting Stents Versus Surgery for
Unprotected Left Main Coronary Artery Disease
Marcelo Sanmartín, PhDa,*, José Antonio Baz, MDa, Ramon Claro, MDa, Vanesa Asorey, MDb,
Darío Durán, PhDb, Gonzalo Pradas, MDb, and Andrés Iñiguez, PhDa
This study was conducted to compare the clinical outcomes of drug-eluting stents (DESs) with
those of standard bypass surgery for the treatment of patients with left main lesions in a
single-center experience. From January 2000 to October 2005, a total of 96 patients with
significant unprotected left main disease were treated with DES implantation, and 245 with
bypass surgery. Baseline features, such as Euroscore, were similar between groups, except for
diabetes and hypertension, which were more frequent in the surgical group. The combination
of death, Q-wave myocardial infarction, stroke, and repeated revascularization (major adverse
cardiac and cerebrovascular events [MACCEs]) at 30 days occurred in 2.1% after DES implant
and 9.0% after surgery (p ⴝ 0.03). At 1 year, DES-treated patients more frequently needed
repeat revascularization (5.2% vs 0.8%; p ⴝ 0.02), although combined MACCE rates were
similar (10.4% for DES, 11.4% for surgery; p ⴝ 0.50). Moreover, after a mean follow-up of 1.3
ⴞ 0.8 and 3.2 ⴞ 1.6 years for the DES and surgical groups, there were no significant differences
in MACCEs, respectively. In conclusion, in our experience, percutaneous treatment of patients
with unprotected left main disease with DESs provided similar clinical results compared with
surgical revascularization at a midterm follow-up. © 2007 Elsevier Inc. All rights reserved.
(Am J Cardiol 2007;100:970 –973)
Surgical revascularization is the current recommended treatment for patients with unprotected left main disease.1 Although 6-month and 1-year results with drug-eluting stents
(DESs) appear promising in single-center studies,2–11 percutaneous intervention (PCI) is still discouraged in current
guidelines.12,13 However, percutaneous treatment of unprotected left main lesions with DESs is increasingly performed in many institutions, although data comparing DES
implantation with coronary artery bypass grafting (CABG)
are still scarce. Moreover, recent doubts about the long-term
safety of DESs14 pose additional uncertainties about the
validity of this strategy for treating patients with unprotected left main lesions. Accordingly, the main goal of this
study is to compare the midterm safety and efficacy of
percutaneous revascularization of unprotected left main stenosis using DESs and surgical revascularization.
Methods
The present study includes patients with significant unprotected left main disease treated using either CABG or DES
implantation from January 2000 through October 2005 in a
single center. Patients needing concomitant valvular surgery or
with previous CABG were excluded. The PCI group included
patients treated from June 2003 through October 2005 in
whom ⱖ1 paclitaxel- (Taxus, Boston Scientific, Natick, Massachusetts) or sirolimus-eluting stent (Cypher, Cordis, Johnson
a
Unidad de Cardiología Intervencionista and bUnidad de Cirugía
Cardíaca, Complexo Hospitalario Universitario de Vigo, Medtec, Vigo,
Spain. Manuscript received February 14, 2007; revised manuscript received and accepted April 24, 2007.
*Corresponding author: Tel.: 34-986-413-144; fax: 34-986-421-439.
E-mail address: [email protected] (M. Sanmartín).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.037
& Johnson Corp., Miami, Florida) was implanted for the treatment of a significant left main stenosis (n ⫽ 96). The CABG
group included 245 patients with significant left main disease
treated using surgical revascularization before this period. The
revascularization strategy was selected using careful individual
case analysis by the interventional cardiology team and after
detailed discussion of treatment strategies with patients and
attending physicians. The study complied with the Declaration
of Helsinki. All patients signed an informed consent before
invasive procedures.
Surgical revascularization was performed according to
standard techniques. PCI procedures were performed from
either a transradial or transfemoral approach and generally
used 6Fr guiding catheters. Lesions affecting the distal left
main artery were managed in most cases using a “provisional
T-stenting” strategy. Type and size of stents used, predilatation
strategy, and antithrombotic regimen were left to the operators’
decision. Recommendations about antiplatelet treatment for
patients who underwent percutaneous revascularization included a combination of aspirin and clopidogrel for ⱖ6
months.
Cardiac enzyme analysis and 12-lead electrocardiography were performed in all patients ⱖ8 hours after revascularization. Periprocedural myocardial infarction (MI) was
considered if there was the appearance of new persistent Q
waves after revascularization with abnormal creatine kinase
or troponin increases. Cerebrovascular events were defined
as any disabling ischemic or hemorrhagic stroke.
The primary end point was the combined rate of death,
Q-wave MI, cerebrovascular event, or target vessel revascularization during the entire follow-up. Secondary end points
included in-hospital and 1-year rates of major adverse cardiac
and cerebrovascular events (MACCEs). Continuous variables
www.AJConline.org
Coronary Artery Disease/Stent Versus Surgery for Left Main Disease
Table 1
Baseline clinical characteristics
Age (yrs)
Women
Hypertension
Diabetes
Insulin-treated
Smoker
Dyslipemia
Previous percutaneous
revascularization
Euroscore ⱖ6
Ejection fraction ⬍50%
Significant right coronary artery
disease
Serum creatinine ⬎1.5 mg/dl
Indication
Stable angina pectoris
Acute coronary syndrome
Recent Q-wave MI
971
Table 3
Surgical revascularization procedural data (n ⫽ 245)
PCI
(n ⫽ 96)
CABG
(n ⫽ 245)
p Value
66.0 ⫾ 12.5
18 (19%)
42 (44.2%)
18 (19%)
1 (6.3%)
37 (38.5%)
40 (42%)
14 (14.6%)
66.0 ⫾ 10.2
33 (13%)
148 (60.2%)
78 (32%)
20 (27%)
112 (45.5%)
112 (46%)
21 (8.5%)
0.97
0.23
0.01
0.01
0.11
0.27
0.54
0.11
26 (27.0%)
25 (32.5%)
44 (45.8%)
50 (25.3%)
53 (24.4%)
145 (58.9%)
0.77
0.18
0.02
2 (2.1%)
8 (3.3%)
0.73
37 (38.5%)
49 (51%)
12 (12.5%)
74 (30.2%)
153 (62.4%)
40 (16.9%)
0.16
0.07
0.41
Table 2
Main data from percutaneous coronary interventions (n ⫽ 96)
Left main location
Ostium/midshaft
Distal left main
Involvement of both distal branches
Technique
Single stent
2-Stent technique (T stent, crush)
Final kissing-balloon
Glycoprotein IIb/IIIa inhibitor
Circulatory support
Taxus stent
Cypher stent
Postprocedural length of stay (d)
High-risk variables
Significant calcification
Occluded right coronary artery
Left dominant artery
Multivessel intervention
35 (38.5%)
59 (61.5%)
16 (16.7%)
85 (90.6%)
9 (9.4%)
33 (34.4%)
16 (16.7%)
4 (4.2%)
69 (71.9%)
27 (28.1%)
1.5 ⫾ 1.4
14 (7.4%)
16 (16.7%)
5 (5.2%)
49 (51.0%)
were presented as mean ⫾ SD. Comparisons between means
were performed using Student’s t test, whereas differences in
categorical variables were assessed using Fisher’s exact test.
Survival curves were generated using the Kaplan-Meier
method, and differences were analyzed using the Breslow test.
Variables associated with MACCEs in univariate analysis
were entered in a stepwise logistic regression analysis. To
account for possible baseline differences between the DES and
CABG groups, a propensity score analysis was performed and
used as a covariate in multivariate regression analysis. The
propensity score was generated using a logistic regression
model with clinical and angiographic baseline characteristics
(Tables 1 and 2). Long-term predictors of death were evaluated
using Cox regression analysis. All tests were 2 tailed, and
differences were considered significant at p ⬍0.05. SPSS version 12.0 (SPSS Inc., Chicago, Illinois), was used for statistical
processing.
30-d clinical events
Q-Wave MI
Cerebrovascular accident
Urgent revascularization
Death
Combined
1-yr cumulative clinical events
Q-Wave MI
Cerebrovascular accident
Repeat revascularization
Death
Combined
PCI
(n ⫽ 96)
CABG
(n ⫽ 245)
p Value
0 (0%)
0 (0%)
0 (0%)
2 (2.1%)
2 (2.1%)
3 (1.3%)
2 (0.8%)
2 (0.8%)
15 (6.1%)
22 (9.0%)
0.44
1.0
1.0
0.17
0.03
0 (0%)
0 (0%)
5 (5.2%)
5 (5.2%)
10 (10.4%)
3 (1.3%)
2 (0.8%)
2 (0.8%)
20 (8.4%)
27 (11.4%)
0.44
1.0
0.02
0.37
0.50
Table 4
Cumulative clinical outcomes at 30 days and 1 year
Distal anastomoses per patient
Intra-aortic balloon pump
Priority
Emergent
Urgent
Left internal thoracic artery grafting
Repeated surgery
Packed blood cell transfusion
Postoperative length of stay (d)
3.2 ⫾ 0.9
12 (4.9%)
7 (2.9%)
61 (25%)
234 (98%)
4 (1.6%)
97 (39%)
9.5 ⫾ 7.6
Results
Baseline clinical data are listed in Table 1. Despite higher
prevalences of diabetes and hypertension in the CABG group,
study groups had similar Euroscore values (4.0 ⫾ 2.5 and 3.9
⫾ 3.0 for the DES and CABG groups, respectively; p ⫽ 0.80).
Procedural characteristics are listed in Tables 2 and 3.
All CABG procedures included conventional cardiopulmonary bypass circulation.
In the DES group (Table 2), only 9 patients were treated
using a 2-stent technique, which included 3 patients with the
crush technique and 6 cases with T stenting. Moderate to
severe depression of left ventricular systolic function was
present in 10 patients (10.4%). An intra-aortic balloon pump
was used in 2 patients, and a percutaneous left ventricular
assist device in another 2 patients (Impella Recover LP 2.5,
Impella CardioSystems GmbH, Aachen, Germany).
In-hospital stay after the procedure was significantly
higher in the CABG group (9.5 ⫾ 7.6 vs 1.5 ⫾ 1.4 days; p
⬍0.01).
Clinical events of DES- and CABG-treated patients are
listed in Table 4. Combined MACCEs at 30 days were
significantly lower in the DES group (2.1% vs 9.0%; p ⫽
0.03). In the DES group, 2 patients had significant creatine
kinase-MB increases (⬎5 times the upper limit of normal),
but no patient experienced a Q-wave periprocedural MI.
Only 2 patients (2.1%) in the DES group died early after the
procedure. Both deaths were attributable to cardiovascular
causes. One patient had free-wall left ventricular rupture
with cardiac tamponade 5 days after an inferior MI. Another
patient experienced out-of-hospital sudden death 3 days
after implantation of a 3.5 ⫻ 18-mm Cypher stent in the left
972
The American Journal of Cardiology (www.AJConline.org)
Figure 1. Kaplan-Meier plots show event-free survival analysis in the surgical (CABG) and percutaneous revascularization (PCI) groups. CVA ⫽
cerebrovascular accident.
main artery. Close relatives affirmed proper compliance
with prescribed antiplatelet agents (aspirin and clopidogrel).
At 1 year, DES-treated patients more frequently needed
repeat revascularization (5.2% vs 0.8%; p ⫽ 0.02), which
included 3 patients with surgical revascularization because
of significant left main restenosis and 2 patients with repeated PCI for non–left main–related lesions. All 3 patients
requiring repeated revascularization because of left main
restenosis initially had lesions involving the distal bifurcation, including 1 patient treated using the crush technique,
and the other 2 patients, using a single stent crossing the
circumflex ostium and final kissing-balloon inflation.
Scheduled or ischemia-driven angiographic follow-up
was performed in 55 patients in the DES group (57%). Of
these, 4 patients had angiographic restenosis (ⱖ50%) of the
left main or 1 of its major branches (7%).
In the DES group, there were an additional 3 deaths.
These included an 82-year-old man with poor left ventricular function at baseline assessment treated using a paclitaxel-eluting stent who died suddenly at day 183, 2 days
after withdrawal of clopidogrel treatment; another patient
with left ventricular dysfunction who died at day 235 from
heart failure; and 1 patient who died after 271 days from
complications related to colon cancer.
Mean follow-up was achieved in 98% of patients at
1.3 ⫾ 0.8 and 3.2 ⫾ 1.6 years in the DES and CABG groups,
respectively. Of DES-treated patients, 49 had a clinical follow-up ⬎1 year after the revascularization procedure. Figure 1
shows the Kaplan-Meier event-free survival analysis for the
entire period. Remarkably, there were no significant differences in total mortality (p ⫽ 0.34) or the combined MACCE
end point (p ⫽ 0.88). Repeated revascularization occurred
more frequently in DES-treated patients (p ⫽ 0.004). When
combined rates of Q-wave MI, cerebrovascular accident, and
death were analyzed separately, there was a nonsignificant
trend to lower event rates in the PCI group (p ⫽ 0.07).
Of all baseline clinical variables analyzed, only CABG
treatment and presentation with recent Q-wave MI showed a
significant correlation with long-term mortality. However, after Cox regression analysis, only recent Q-wave MI represented a significant predictor of total mortality (p ⫽ 0.014).
Discussion
The main finding of this study was that in a single-center
experience and considering careful patient selection, PCI
using a DES provided similar results compared with surgical revascularization in patients with significant unprotected
Coronary Artery Disease/Stent Versus Surgery for Left Main Disease
left main coronary artery disease. As expected, long-term
repeat revascularization increased with PCI (5.2% vs 0.8%
at 1 year), but the present results suggested that hard end
points, such as death, stroke, and Q-wave MI did not increase with DES use and may even be less frequent.
Other previous single-center studies suggested that treatment of patients with sirolimus- or paclitaxel-eluting stents
provided good immediate results, with rates of repeated revascularization that ranged from 0% to 38%.2– 8 Reasons for this
wide variation in target vessel failure rates are probably related
mostly to case selection. Importantly, distal left main involvement was shown to be an important predictor of adverse events
after percutaneous revascularization using DESs.15,16
The present results compare well with other 3 recent
single-center retrospective studies evaluating DESs versus
CABG for patients with left main coronary artery disease.9 –11 However, 2 important differences should be noted
between these studies and the present one. First, we avoided
inclusion of CABG-treated patients after DESs became
widely available in our institution. This design was intended
to mitigate selection bias because patients not treated with
DESs during this period probably had more diffusely diseased coronary vessels. Second, final PCI treatment of the
left main bifurcation included implantation of a single stent
in most patients in the present study. The optimal percutaneous approach for bifurcated left main lesions was not
established, but recent data have suggested that a simple
strategy involving implantation of a single stent across 1
side branch (usually the circumflex artery) provided better
results than systematic stenting of both branches.17 Perhaps
selection of patients with a higher probability of obtaining
an optimal angiographic result with only 1 stent helped
improve clinical outcomes, especially by decreasing adverse events related to repeated target vessel revascularization.
The main limitation of this study was related to its nonrandomized character. Despite selection of a historic surgical
group from a pre-DES era, there were some baseline clinical
differences between study groups, highlighting the inevitable
selection bias related to patient selection. In the present study,
surgically treated patients probably had more extensive coronary disease, especially related to the higher prevalences of
diabetes mellitus and hypertension. However, important surgical risk estimates, such as age, left ventricular ejection fraction,
and Euroscore, were similarly distributed. In addition, none of
the baseline clinical features analyzed had a significant correlation with long-term adverse events, and imbalances in study
groups were attenuated by propensity score analysis. A second
limitation involved the relatively short follow-up for DEStreated patients (mean follow-up 1.3 years).
1. Eagle KA, Guyton RA, Davidoff R, Edwards FH, Ewy GA, Gardner TJ,
Hart JC, Herrmann HC. ACC/AHA 2004 guidelines update for coronary
artery bypass graft surgery. Circulation 2004;110:1168–1176.
2. de Lezo JS, Medina A, Pan M, Delgado A, Segura J, Pavlovic D, Melian
F, Romero M, Burgos L, Hernandez E, Urena I, Herrador J. Rapamycineluting stents for the treatment of unprotected left main coronary disease.
Am Heart J 2004;148:481–485.
3. Arampatzis CA, Lemos PA, Hoye A, Saia F, Tanabe K, van der
Giessen WJ, Smits PC, McFadden E, de Feyter P, Serruys PW.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
973
Elective sirolimus eluting stent implantation for left main coronary
artery disease: six-month angiographic follow-up and 1-year clinical
outcome. Catheter Cardiovasc Interv 2004;62:292–296.
Chieffo A, Stankovic G, Bonizzoni E, Tsagalou E, Iakovou I, Montorfano M, Airoldi F, Michev I, Sangiorgi MG, Carlino M, Vitrella G,
Colombo A. Early and mid-term results of drug-eluting stent implantation in unprotected left main. Circulation 2005;111:791–795.
Park SJ, Kim YH, Lee BK, Lee SW, Lee CW, Hong MK, Kim JJ,
Mintz GS, Park SW. Sirolimus-eluting stent implantation for unprotected left main coronary artery stenosis: comparison with bare metal
stent implantation. J Am Coll Cardiol 2005;45:351–356.
Valgimigli M, van Mieghem CA, Ong AT, Aoki J, Granillo GA,
McFadden EP, Kappetein AP, de Feyter PJ, Smits PC, Regar E, et al.
Short- and long-term clinical outcome after drug-eluting stent implantation for the percutaneous treatment of left main coronary artery
disease: insights from the Rapamycin-Eluting and Taxus Stent Evaluated At Rotterdam Cardiology Hospital registries (RESEARCH and
TSEARCH). Circulation 2005;111:1383–1389.
Sadeghi HM, O’Neill WW, Grines CL. Percutaneous intervention of
unprotected left main coronary artery. J Interv Cardiol 2003;16:281–
288.
Price MJ, Cristea E, Sawhney N, Kao JA, Moses JW, Leon MB, Costa
RA, Lansky AJ, Teirstein PS. Serial angiographic follow-up of sirolimus-eluting stents for unprotected left main coronary artery revascularization. J Am Coll Cardiol 2006;47:871– 877.
Lee MS, Kapoor N, Jamal F, Czer L, Aragon J, Forrester J, Kar S,
Dohad S, Kass R, Eigler N, et al. Comparison of coronary artery
bypass surgery with percutaneous coronary intervention with drugeluting stents for unprotected left main coronary artery disease. J Am
Coll Cardiol 2006;47:864 – 870.
Chieffo A, Morici N, Maisano F, Bonizzoni E, Cosgrave J, Montorfano M, Airoldi F, Carlino M, Michev I, Melzi G, et al. Percutaneous
treatment with drug-eluting stent implantation versus bypass surgery
for unprotected left main stenosis: a single-center experience. Circulation 2006;113:2542–2547.
Palmerini T, Marzocchi A, Marrozzini C, Ortolani P, Saia F, Savini C,
Bacchi-Reggiani L, Gianstefani S, Virzì S, Manara F, et al. Comparison between coronary angioplasty and coronary artery bypass surgery
for the treatment of unprotected left main coronary artery stenosis (the
Bologna Registry). Am J Cardiol 2006;98:54 –59.
Silber S, Albertsson P, Aviles FF, Camici PG, Colombo A, Hamm C,
Jorgensen E, Marco J, Nordrehaug JE, Ruzyllo W, et al. Task Force for
Percutaneous Coronary Interventions of the European Society of Cardiology. Guidelines for percutaneous coronary interventions. The Task
Force for Percutaneous Coronary Interventions of the European Society of Cardiology. Eur Heart J 2005;26:804 – 847.
Smith SC Jr, Feldman TE, Hirshfeld JW Jr, Jacobs AK, Kern MJ, King
SB III, Morrison DA, O’Neill WW, Schaff HV, Whitlow PL,
et al. ACC/AHA/SCAI 2005 guideline update for percutaneous coronary
intervention—summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines
(ACC/AHA/SCAI Writing Committee to Update the 2001 Guidelines for
Percutaneous Coronary Intervention). Circulation 2006;113:156 –175.
Maisel WH. Unanswered questions. Drug-eluting stents and the risk of
late thrombosis. N Engl J Med 2007;356:981–984.
Hoye A, Iakovou I, Ge L, van Mieghem CA, Ong AT, Cosgrave J,
Sangiorgi GM, Airoldi F, Montorfano M, Michev I, et al. Long-term
outcomes after stenting of bifurcation lesions with the “crush” technique:
predictors of an adverse outcome. J Am Coll Cardiol 2006;47:1949 –58.
Valgimigli M, Malagutti P, Rodriguez-Granillo GA, Garcia-Carcia HM,
Polad J, Tsuchida K, Regar E, van der Giessen WJ, Jaegere P, De Feyter
P, Serruys PW. Distal left main coronary disease is a major predictor of
outcome in patients undergoing percutaneous intervention in the drugeluting stent era: an integrated clinical and angiographic analysis based on
the rapamycin-eluting stent evaluated at Rotterdam Cardiology Hospital
(RESEARCH) and Taxus-stent evaluated at Rotterdam Cardiology Hospital (T-SEARCH) Registries. J Am Coll Cardiol 2006;47:1530 –1537.
Kim YH, Park SW, Hong MK, Park DW, Park KM, Lee BK, Song JM,
Han KH, Lee CW, Kang DH, et al. Comparison of simple and complex
stenting techniques in the treatment of unprotected left main coronary
artery bifurcation stenosis. Am J Cardiol 2006;97:1597–1601.
The Editor’s Roundtable: Arterial Thrombosis and Acute
Coronary Syndromes
Vincent E. Friedewald, MDa,*, Eric R. Bates, MDb, Christopher B. Granger, MDc,
Salim Yusuf, DPhild, and William C. Roberts, MDe
Acknowledgment
This CME activity is supported by an educational grant
from GlaxoSmithKline, Research Triangle Park, North
Carolina.
2. Decide among the anticoagulant drugs which agent(s)
are most appropriate in various ACS settings.
3. Combine different anticoagulant drugs for maximum
efficacy.
4. Use anticoagulant drugs with lowest possible risk for
bleeding complications.
Disclosure
Dr. Friedewald has no relevant financial relationships to disclose. Dr. Bates has received honoraria for speaking and advisory boards for GlaxoSmithKline, sanofi-aventis, Bridgewater, New Jersey, and The Medicines Company, Parsippany,
New Jersey. Dr. Granger has received research grants as an
investigator for AstraZeneca, Wilmington, DE, Procter &
Gamble, Mason, Ohio, Alexion, Cheshire, Connecticut,
sanofi-aventis, Novartis, East Hanover, New Jersey, Genentech, South San Francisco, California, Bristol Myers
Squibb, New York, New York, The Medicines Company,
and honoraria for speaking from Boehringer Ingelheim,
Ridgefield, Connecticut, Bayer Healthcare, West Haven,
Connecticut, GlaxoSmithKline, and consulting fees from
Bristol-Myers Squibb, The Medicines Company, and
AstraZeneca. Dr. Roberts has received honoraria for
speaking from Merck, Whitehouse Station, New Jersey,
Schering Plough, Kenilworth, New Jersey, Pfizer, New
York, New York, AstraZeneca, and Novartis. Dr. Yusuf
has received honoraria for speaking, consulting fees, and
research grants from GlaxoSmithKline, sanofi-aventis,
and Bristol-Myers Squibb.
Objectives
Upon completion of the activity, the physician should be
able to:
1. Recognize the indications for anticoagulant therapy in
patients with acute coronary syndrome (ACS).
a
Assistant Editor, American Journal of Cardiology; Clinical Professor,
Department of Internal Medicine, The University of Texas Medical School
at Houston, Houston, Texas; Visiting Professor, University of Notre Dame,
Notre Dame, Indiana; bProfessor, Internal Medicine, University of Michigan, Ann Arbor, Michigan; cAssociate Professor of Medicine, Department
of Medicine, Duke University Medical Center, Durham, North Carolina;
d
Professor of Medicine, McMaster University, Hamilton, Ontario, Canada;
and eEditor-in-Chief, American Journal of Cardiology; Executive Director,
Baylor Heart and Vascular Institute; and Dean, A. Webb Roberts Center
for Continuing Medical Education, Baylor Health Care System, Dallas,
Texas. Manuscript received and accepted May 10, 2007.
This manuscript is based upon a meeting of the authors that took place
on November 21, 2006.
*Corresponding author: Tel: 512-264-1611; fax: 512-264-7034.
E-mail address: [email protected] (V. Friedewald).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.05.025
Introduction
Occlusive thrombus superimposed on coronary atherosclerotic plaque is usually present in patients with ST-segment
elevation acute myocardial infarction (AMI)1–3 and is a
primary target of therapy, such as fibrinolysis, percutaneous
coronary intervention (PCI), and anticoagulant drugs. Occlusive thrombus is infrequently associated with other types
of ACS—non–Q-wave AMI, unstable angina pectoris, and
sudden coronary death. This Editor’s Roundtable discusses
the pathophysiology, diagnosis, and anticoagulant treatment
of patients with ACS.
Dr. Friedewald: What is the relation between coronary
arterial thrombosis and ACS?
Dr. Roberts: The necropsy findings in the coronary
arteries of patients with fatal coronary disease are variable.
Thrombus is present in 10% to 20% of sudden coronary
deaths, and is termed a “mural thrombus” because it occupies only a small area of the residual coronary arterial
lumen. I do not believe that these thrombi are the cause of
sudden coronary death. In patients with the first episode of
angina pectoris, the role of thrombi is unknown. In unstable
angina pectoris, thrombus is uncommon. In patients with
AMI who die, 50% have thrombus at autopsy and the other
50% may have an initial thrombus with subsequent lysis.
These thrombi are superimposed on plaque.
Dr, Friedewald: Does thrombus play a role in long-term
plaque formation?
Dr. Roberts: Yes, I believe it does.
Dr. Bates: When you say “thrombus,” are you referring
to actual blood clot?
Dr. Roberts: Coronary artery thrombus is a fibrin platelet aggregate adherent to the intimal surface of the plaque.
Dr. Granger: ACS with coronary arterial thrombosis is
best detected by elevated blood troponin. Such patients are
at high risk for acute myocardial infarction (AMI) and
death, and they might benefit from glycoprotein IIb/IIIa
inhibition, aggressive antithrombin therapy, and coronary
intervention.
Dr. Roberts: In my experience at necropsy, ruptured
plaque occurs in about 70% of patients with AMI. In the
other 30%, thrombus forms over a non-ruptured plaque.
Dr. Yusuf: I recall a past debate whether clot was an
initial or secondary event in AMI. The debate was settled by
the work of DeWood2 and others who found that clot is the
www.AJConline.org
Roundtable Discussion/Arterial Thrombosis and ACS
initiator of AMI. In 1 study, Fulton4 put radioactive material
into patients when they were first admitted to the coronary
care unit, and found at necropsy that the radioactive material
was located in the periphery of the clot, suggesting the clot
occurred early in the course of AMI. In a study of persons
with sudden cardiac death by Michael Davies,5 gross clot
rupture and occlusive thrombus were absent in ⬍30% of
patients. He did find clot erosion and platelet thrombus in
almost all AMI patients. Emboli from nonocclusive thrombi
causing arrhythmic events and micronecrosis may be an
important mechanism in sudden cardiac death.
Dr. Roberts: “Plaque erosion,” which means rubbing off
the surface layer, is a difficult histologic diagnosis that I
never make.
Dr. Bates: Angioscopic data suggesting the presence of
thrombus weeks and months after clinical events are often
reported.
Dr. Roberts: The number of cases following AMI studied with angioscopy is relatively small, and the studies are
uncontrolled.
Dr. Granger: Patients who survive cardiac arrest do not
appear to benefit from fibrinolytic therapy. This is consistent with the premise that acute complete thrombosis is not
an important mechanism of sudden cardiac death.
Dr. Roberts: I agree. Persons who have sudden cardiac
death would not inevitably have had an AMI if they had
survived. In 1 study of patients who had cardiac arrest outside
the hospital, only one-half subsequently had an AMI.
Dr. Yusuf: I wonder whether friable platelet clots form
emboli and cause arrhythmias and acute cardiac events, but
not necessarily AMI.
Dr. Bates: There has been much discussion about inflammation and peripheral clot embolization related to endothelial and microcirculatory function. Can some of those
processes be stimulated in the absence of plaque rupture or
thrombus formation, leading to an acute cardiac event?
Dr. Roberts: Those are difficult questions for histological study and answers would require electron microscopic
studies of endothelial cells.
Dr. Bates: What acute changes in the coronary artery
cause fatal arrhythmias?
Dr. Roberts: I do not believe there has to be any acute
change within the coronary artery for a fatal arrhythmia.
Rather than plaque rupture, my view is that the events
related to sudden cardiac death due to coronary artery disease are due to excess plaque quantity, or “plaque burden.”
Patients who die suddenly, patients who have an AMI, and
patients with unstable angina pectoris usually have huge
plaque quantity. Even patients with stable angina pectoris
who die of cancer or some other disease unrelated to atherosclerosis have large amounts of plaque in the coronary
arteries. The plaque quantity is similar in all of these groups.
I believe that when the plaque quantity is large, only small
added factors are needed to transition the myocardium from
adequate to inadequate oxygenation.
Dr. Yusuf: Something acute must happen, however, for
a person to die suddenly. What is it?
Dr. Roberts: That is the question that makes the argument for plaque rupture so appealing, because it is a morphological explanation. There are numerous, documented
episodes of sudden cardiac death, such as a father running to
975
greet his daughter whom he has not seen in 20 years, and he
becomes so excited that he has a cardiac arrest. Those
events happen to persons only when they have a large
amount of plaque, and the emotion of the moment trips them
into a terminal arrhythmia.
Dr. Yusuf: There are 2 sets of data I find persuasive
suggesting that clot plays a pathophysiological role in acute
cardiac events. One is DeWood’s classic data2 describing a
high proportion of clots that occurred early in patients with
ST-segment elevation AMI, and subsequently underwent
spontaneous lysis. In patients with non–Q-wave AMI, the
clots appeared later in the course. These observations, however, do not tell us whether clot is a primary or secondary
event in AMI. The second line of evidence is found in
people with unstable angina pectoris who are threatening to
have an AMI, and giving them an antiplatelet agent prevents
clot formation.
Dr. Roberts: I believe clots play some role in all these
patients, but without necessarily causing total coronary arterial occlusion.
Dr. Bates: Quite often with angiography we see edges of
plaque that appear to have ruptured. In acute cases, we see
a crater, resembling a volcano, with heaped-up edges. When
you refer to plaque burden, are you talking only about
atherosclerosis, or do you observe coronary ulceration apart
from thrombosis? Perhaps ulceration is a substrate, complementary to thrombosis, and such ulcers heal.
Dr. Roberts: I am not certain, but stomach ulcers heal,
so I believe plaque ulcers can also heal.
Dr. Bates: I believe they do heal. Ulcerations in ACS are
part of the causative lesion. Such craters often correlate with
electrocardiographic changes and positive biomarkers, and
we conclude they caused the ACS.
Dr. Roberts: When you look at 5-mm sections of all of
the major coronary arteries of patients with anterior wall
AMI, the location of the infarct tells where the thrombus
is located. It does not tell the quantity of plaque in the
coronary arteries. I have found that the quantity of plaque
in the right, left anterior descending, and left circumflex
coronary arteries is similar regardless of whether patients
have had unstable angina pectoris, AMI, or healed myocardial infarction.
Dr. Yusuf: In most patients with unstable angina pectoris, you cannot localize the “culprit” artery by electrocardiogram, although you can make a reasonable guess. For
example, if there are deep, symmetrical T waves in leads V1
and V2, the left anterior descending coronary artery is involved. When there is only ST-segment depression, however, you cannot determine which artery is involved.
Dr. Bates: In 1 report of a patient having a percutaneous
coronary intervention (PCI) followed by an AMI, magnetic
resonance imaging showed a tight, small defect correlating
with the coronary artery where the balloon caused a distal
coronary artery embolus.
Dr. Friedewald: Is there anything inherently different
about coronary artery thrombus from thrombi in other blood
vessels?
Dr. Roberts: Thrombus consists of fibrin, platelets,
white cells, and red cells and, in my opinion, must be
adherent to the vessel wall to be called a “thrombus.”
Patients with abdominal aortic aneurysms have thrombi
976
The American Journal of Cardiology (www.AJConline.org)
within the aneurysm. Those thrombi, which are often
present for years, consist entirely of fibrin and are not
organized. In contrast, thrombi in coronary arteries organize
and disappear in patients who survive the AMI, eventually
retracting and becoming small scars in the arterial wall.
Other types of thrombi may even totally lyse, which is well
known to occur with pulmonary emboli.
Dr. Friedewald: What is the role of anticoagulants
in AMI?
Dr. Yusuf: The first anticoagulant was unfractionated
heparin for treating patients with unstable angina pectoris
and non–ST-segment elevation AMI. Although initial outcome studies were inconclusive, subsequent trials showed a
composite reduction in fatal outcomes in AMI in patients
treated with anticoagulation. The Fragmin During Instability in Coronary Artery Disease (FRISC I) study6 showed a
60% to 70% reduction in AMI in patients with unstable
angina pectoris receiving unfractionated heparin. We began
using unfractionated heparin in the early 1980s. Unfractionated heparin inhibits factor IIa and factor X to a lesser
extent. It is a sort of “dirty” molecule in that it also binds
platelets, its bioavailability is unpredictable, and it has other
negative attributes. Low-molecular-weight heparin, however, causes less platelet activation and a slightly higher
ratio of factor X inhibition than unfractionated heparin.
Both of these forms of heparin cause excessive bleeding
complications.
The next set of anticoagulants was direct thrombin inhibitors, including hirudin, which is a pure factor IIa inhibitor. These agents were created by recombinant molecular
technology. They were more effective than heparin in preventing AMI, but had a fine line between efficacy and
bleeding. Next, bivalirudin was developed, and we still do
not have convincing data that it is superior to unfractionated
heparin. Factor IIa inhibition is, however, important in the
clinical setting of an induced injury such as PCI, so bivalirudin may have value with this procedure.
The factor Xa inhibitor fondaparinux came next, and is
now used extensively in clinical practice. In 4 of 5 clinical
trials7–11 of prevention of deep venous thrombosis comparing fondaparinux with the low molecular weight heparin,
enoxaparin, fondaparinux had superior efficacy. In the fifth
trial there was a trend favoring fondaparinux. The collective
data from these 5 trials showed that fondaparinux was about
2 times as effective as enoxaparin in preventing deep venous thrombosis.
In the OASIS (Organization for the Assessment of Strategies for Ischemic Syndrome)-5 Trial,7 fondaparinux was
compared to enoxaparin in patients with non–ST-segment
elevation myocardial infarction, and this large trial of 20,000
individuals demonstrated similar efficacy at the end of the
9-day treatment, but substantial reduction in the rates of major
bleeding, including life-threatening and fatal bleeds with
fondaparinux compared to enoxaparin. This translated into a
statistically significant (20%) reduction in the risk of death,
myocardial infarction and strokes with fondaparinux compared
to enoxaparin at 30 days.
The OASIS-6 study12 compared fondaparinux versus
standard care in patients with ST-segment elevation myocardial infarction, and demonstrated a significant reduction
in death alone and in the composite of death and myocardial
infarction, without an increase in bleeding or an increase in
strokes. Thus, fondaparinux is the only antithrombotic agent
that has been demonstrated to reduce mortality in patients
with acute coronary syndromes or ST-segment-elevation
myocardial infarction, without increasing hemorrhagic
complications or stroke risk.
Dr. Roberts: Were IIb/IIIa inhibitors used in the
OASIS trials?
Dr. Yusuf: About 25% of the 20,000 patients studied
received a glycoprotein IIb/IIIa inhibitor in both the enoxaparin and fondaparinux groups. In both groups, the results
showed similar efficacy but superiority for fondaparinux in
reducing bleeding complications, which was associated
with decreased mortality. The initial analyses, however, did
not distinguish whether patients who bled were more likely
to die because of the bleeds or that they were only sicker.
OASIS-5 showed at 1 month and 6 months lower mortality
with fondaparinux compared to enoxaparin, and this was
almost entirely related to decreased rates of bleeding. There
was also a significant reduction in the incidence of strokes
in patients taking fondaparinux. We tested these agents in
patients with ST-segment elevation AMI and in those receiving primary PCI, and there was no difference between
fondaparinux and unfractionated heparin. The increased
bleeding in the OASIS-6 study differed from OASIS-5, in
that the bleeding was due to cardiac rupture, which we do
not regard as an actual bleeding complication. Recently, we
had another trial, the ExTRACT-TIMI 25 (Enoxaparin and
Thrombolysis Reperfusion for Acute Myocardial Infarction
Treatment–Thrombolysis in Myocardial Infarction 25) study,
looking at low-molecular-weight heparin in patients with
ST-segment elevation AMI treated with thrombolytic
agents. Study patients showed a reduction in re-infarction
and a nonsignificant trend toward reduced mortality, but
they had a substantial increase in bleeding complications,
including intracranial hemorrhage.
Dr. Roberts: Were all forms of bleeding decreased with
fondaparinux compared to the older agents?
Dr. Granger: All types of bleeding were reduced, and to
a similar extent.
Dr. Yusuf: Minor bleeding was reduced by about 40%,
major bleeding by 50%, and fatal bleeding by ⬎65%.
Dr. Granger: Over the past 15 years, starting with
aspirin therapy, our focus has been on the thrombus and
re-opening the coronary artery in AMI. We previously
regarded bleeding as a nuisance that should be avoided,
but not an important problem. In numerous recent studies, however, we have observed a 2- to 6-fold increased
risk of death associated with a bleeding event, after
adjusting for all other variables. This is supported by data
from the fondaparinux trials showing that an agent that
reduces bleeding also reduces subsequent mortality.
Likewise, patients who receive blood transfusions appear
to have worse outcomes. It is possible that transfusion
may also be causing harm that had not been recognized,
although this remains controversial.
Dr. Yusuf: The increased mortality associated with minor bleeds when patients do not receive blood transfusions
tells us that bleeding per se has an adverse effect.
Dr. Granger: The reasons for the increased mortality are
not defined, but may relate to hemodynamic compromise
Roundtable Discussion/Arterial Thrombosis and ACS
due to the bleeding event, hyperadrenergic state, increased
inflammation, or perhaps most important, changes in treatment away from the use of antithrombotic drugs following
bleeding episodes.
Dr. Roberts: Was the favorable difference in bleeding
complications related to the age of patients given
fondaparinux?
Dr. Yusuf: No. Bleeding complications were lower with
fondaparinux at all ages. The rate of bleeding with or without any agent, however, rises with increasing age. The risk
factors for bleeding are age, female gender, low body
weight, and decreased creatinine clearance.
Dr. Granger: The absolute advantage of fondaparinux
in causing less bleeding, however, was greater in people
who had higher risks for bleeding. For that reason, its use
was particularly favorable in older persons.
Dr. Yusuf: With increasing age, the increased risk for
bleeding is associated with decreased creatinine clearance.
The benefit of reduced bleeding with fondaparinux is consistent across every subgroup.
Dr. Friedewald: What is the role of bivalirudin?
Dr. Granger: Based in part on the ACUITY (Acute
Catheterization and Urgent Intervention Triage Strategy)
trial,13 bivalirudin appears to be an effective drug. It has a
very short half-life, and therefore, may have some advantages, including being more reversible. Some of these advantages might relate to lower rates of bleeding, especially
if one could use it as a sole agent rather than using the
combination of unfractionated heparin and a glycoprotein
IIb/IIIa inhibitor, which is standard for high-risk patients.
Compared to unfractionated heparin and a glycoprotein IIb/
IIIa inhibitor, either in the setting of high-risk coronary
angioplasty or ACS—as in the ACUITY trial— bivalirudin
has a substantially lower risk of bleeding.
Dr. Yusuf: Compared to unfractionated heparin alone,
bivalirudin does not cause a lower rate of bleeding complications.
Dr. Granger: We do not know that for certain. We do
know from the ACUITY trial that there is no difference in
bleeding complication frequency between a glycoprotein
IIb/IIIa inhibitor with bivalirudin versus a glycoprotein IIb/
IIIa inhibitor with unfractionated heparin or low-molecularweight heparin.
Dr. Yusuf: I agree.
Dr. Bates: This raises the important point about our
difficulty in comparing drugs as stand-alone agents without
considering drug dose, duration of therapy, and adjunctive
events like the use of antithrombotic medications and invasive procedures that put patients at increased risk for bleeding. The OASIS-5 trial has the benefit of a direct comparison, but in all other trials there are different baseline
characteristics, different doses, different durations, and even
differences in the definition of “bleeding.”
Dr. Yusuf: No single drug “does it all,” so it is important
to use the right combination of drugs for each particular
clinical setting.
Dr. Roberts: When you say “combination,” are you
referring to glycoprotein IIb/IIIa agents and if so, are they
even going to be used much longer?
Dr. Bates: Only a few years ago we were focused on
paraprocedural AMI, which was one of the reasons for using
977
glycoprotein IIb/IIIa inhibitors with coronary intervention.
Since clopidogrel has reduced the utilization of those drugs,
and we now have 4 other anticoagulant drugs to choose
from, we have shifted our focus to bleeding, which is how
these drugs are being contrasted by their advocates.
Dr. Bates: The enoxaparin dose in OASIS-5 might have
been too high, based on what has been subsequently learned
about the need to down-titrate the dose when correcting for
increased age or decreased renal function. The lesson from
ACUITY is that using 2 or 3 anticoagulant medications in
combination carries a greater risk for bleeding than does the
use of 1 drug alone or 2 drugs in combination.
Dr. Yusuf: The dose of enoxaparin in OASIS-5 was
what is currently approved by the Food and Drug Administration and supported by the American College of Cardiology and American Heart Association treatment guidelines. There is speculation that lower doses of enoxaparin
should be prescribed for older patients than is stated in the
current recommendations.
Dr. Roberts: Does that apply to both men and women?
Dr. Yusuf: Yes, because the increased risk of bleeding is
determined by creatinine clearance, which is lower in
women. The one trial that tested a lower dose in older
patients, the ExTRACT-TIMI 25 study,14 showed excess
bleeding in those patients with unfractionated heparin. It
also appeared to show less efficacy at lower doses, but we
have not seen the full data.
Dr. Bates: ExTRACT underestimated the importance of
adjusting the dose for lower creatinine clearances. As we
learn more about the overlap between renal function and
drug dosing, it is apparent that we need to know the creatinine clearance before we prescribe many drugs, both in the
catheterization laboratory and outside the hospital.
Dr. Friedewald: Are there any new thoughts about
aspirin?
Dr. Yusuf: Aspirin is very effective at the lower dose of
81 mg/day. North American guidelines recommend 325
mg/day, and European guidelines recommend under 100
mg/day—not a trivial difference.
Dr. Bates: The limitation of guidelines is that they adopt
dose recommendations based on the doses used in clinical
trial protocols, which form the evidence base for drawing
the guideline. It is possible that updated American guidelines will be more consistent with the European guidelines,
based on Dr Yusuf’s analysis showing that if one uses dual
antiplatelet therapy and is concerned about bleeding, 81
mg/day of aspirin is the safest dose when combined with
clopidogrel.
Dr. Yusuf: This raises the important issue of the limitations of our data, which were post hoc analyses. These
should be addressed prospectively by doing a proper randomized trial. If recommendations are going to be changed,
they should acknowledge the need for such a future trial.
Dr. Bates: A range of acceptable treatments would be a
better way to present these guidelines, since it varies by
concomitant therapy and region of the world.
Dr. Roberts: Is there evidence whether 81 mg or 325 mg
of aspirin is better?
Dr. Yusuf: For efficacy in long-term treatment with
aspirin, there are no data showing that 81 mg and 325
mg/day are distinguishable. For safety, there are good data
978
The American Journal of Cardiology (www.AJConline.org)
that lower doses are safer at all dose levels, from 1000
mg/day down to ⬍100 mg/day.
Dr. Bates: One compromise is to use 325/per day in the
hospital, when the patient is acutely ill. Almost everyone
agrees that 81 mg/day is the safest dose for the long-term.
Dr. Yusuf: There are not enough trials using direct
comparisons, so I would like to see more trials, but there are
enough data to make recommendations.
Dr. Granger: We have enough data about aspirin for
bleeding complications but not for efficacy. We are trying to
do a National Institutes of Health-sponsored trial of aspirin
doses for prevention to address the issue of 81 mg versus
325 mg/day.
Dr. Yusuf: There are also concerns with using the higher
dose in the acute care setting where bleeding complication
rates are more frequent because of protocols employing
other antithrombotic agents and invasive procedures.
Dr. Roberts: How often do you use glyoprotein IIb/IIIa
agents?
Dr. Bates: I use glyoprotein IIb/IIIa agents infrequently.
There are data strongly suggesting that they do not combine
well with fibrinolytic therapy. Although some experts recommend them for rescue coronary angioplasty, I avoid them
because patients who come to rescue angioplasty have already survived an acute event, and we do not want to
increase the bleeding risk in a patient who is currently
stable. There are some data supporting their use in PCI. The
1 elective indication for IIb/IIIa glycoprotein inhibitors
might be for patients with non–ST-segment elevation
AMI or biomarker-positive ACS. This indication is based
on data from ISAR-REACT 2 (Intracoronary Stenting
and Antithrombotic Regimen: Rapid Early Action for
Coronary Treatment 2),15 in which they were shown to be
beneficial when used with aspirin and clopidogrel.
Dr. Yusuf: In our center, glycoprotein IIb/IIIa inhibitors
are used almost entirely with PCI.
Dr. Bates: In primary PCI or PCI for patients with
non–ST-segment elevation AMI?
Dr. Yusuf: Mostly in patients with non–ST-segment
elevation AMI.
Dr. Bates: There is little reason to use them in patients
who are biomarker-negative coming into the catheterization
laboratory for an elective procedure.
Dr. Granger: At Duke, patients coming to the catheterization laboratory, the emergency department, and the coronary care unit with dynamic ST depression or who are
troponin-positive usually receive a glycoprotein IIb/IIIa inhibitor. In addition, all high-risk patients receive a glycoprotein IIb/IIIa inhibitor at the time of coronary angioplasty.
Dr. Bates: Clopidogrel does not work for several hours
after ingestion. Thus, the patient undergoing PCI who has
not received clopidogrel before the procedure will not get
the benefit of its antiplatelet effect. A glycoprotein IIb/IIIa
inhibitor would have some benefit for high-risk patients in
that situation. Patients who can be pretreated with clopidogrel for at least 1 day before the PCI have less need for a
glycoprotein Iib/IIIa inhibitor.
Dr. Granger: In CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes
with Early Implementation of the American College of Cardiology/American Heart Association Guidelines),16 which is a
480 hospital registry, ⬎40% of patients receiving heparin,
low-molecular-weight heparin, or glycoprotein IIb/IIIa inhibitors were overdosed, mainly due to failure to adjust
doses for renal insufficiency. This overdosing was associated with a much higher risk of bleeding complications.
Dr. Bates: Creatinine clearance should be elevated to the
level of a vital sign. Patients having cardiac catheterization
should have the creatinine clearance determined before they
are prescribed these drugs, which are excreted by the kidneys. Such increased attention to the creatinine clearance
would greatly improve drug safety.
Dr. Roberts: When is clopidogrel not given for ACS?
Dr. Bates: Everyone with an ST-segment elevation AMI
and a biomarker-positive ACS, and who receives an intracoronary stent should receive clopidogrel—probably for at
least 1 year after the procedure. It is much less clear,
however, whether all patients with ACS who are biomarkernegative should be treated with clopidogrel because many
of these patients have noncardiac causes of chest pain.
Dr. Yusuf: Patients with a high probability of an ACS—
even when biomarker-negative— should receive clopidogrel. Clinical trials show that clopidogrel does prevent
AMI in such patients.
Dr. Bates: Clinical trials increasingly focus on the highest-risk patients for ACS because they are more likely to
show differences among various treatment strategies.
Among the 10 million people a year who come to the
emergency department with chest pain, few have an ACS.
My fear of stating in practice guidelines that every patient
with possible ACS should be given clopidogrel in the emergency department is that this would result in hundreds of
thousands of patients receiving a drug they do not need.
Dr. Yusuf: It takes only 24 to 48 hours to determine
whether a patient has ACS and the bleeding risk of taking 2
doses of clopidogrel is lower than the risk of aspirin. Thus,
if you give aspirin to a patient while you are making the
diagnosis, you should also give clopidogrel.
Dr. Roberts: When the diagnosis is ACS you prescribe
clopidogrel?
Dr. Yusuf: Yes.
Dr. Bates: Do you start clopidogrel in the emergency
department or later? The consensus is moving toward giving
it in the emergency department because it carries no risk and
there is value for early platelet inactivation prior to implanting an intracoronary stent, which many of these patients
receive.
Dr. Yusuf: Clopidogrel is expensive. If it were as inexpensive as aspirin, there would be no debate.
Dr. Granger: At our institution, patients who can afford
clopidogrel continue it indefinitely after receiving a drugeluting intracoronary stent.
Dr. Yusuf: That policy makes drug-eluting intracoronary stents even more expensive.
Dr. Granger: Following ACS, most of the benefit of
clopidogrel seems to occur within 3 months. Although this
is being challenged, I currently believe that patients should
be continued for 9 to 12 months, provided cost and safety
are not important issues.
Dr. Bates: It is difficult to decide what to recommend
as a general guideline for clinical practice, and it is even
harder to decide what to do for specific patients with
Roundtable Discussion/Arterial Thrombosis and ACS
varying risks and benefits, financial abilities, and comorbid diseases.
Dr. Yusuf: Perhaps, “How long to give clopidogrel after
implanting a drug-eluting stent?” is the wrong question.
Instead, maybe we need to ask whether there is a compelling
reason for using drug-eluting stents in the first place, and if
so, in whom? Coronary arterial restenosis after PCI is usually benign, and it can be “corrected” when it is clinically
important. Acute stent thrombosis, however, is life-threatening, and to prevent additional angioplasties, is it really
worth using these expensive drug-eluting stents with their
potential hazards of acute, fatal events, the potential hazards
of long-term combination antithrombotic therapy, and the
greatly increased cost?
Dr. Bates: We need better information about the risks and
costs of treating coronary narrowings with bare metal stents.
Dr. Friedewald: How do you use antithrombotic therapy
in patients with ACS?
Dr. Bates: I give 2 doses of aspirin, the first dose 325 mg
and the second, 81 mg. I also give 600 mg of clopidogrel.
Some cardiologists might give 300 mg of clopidogrel, but
600 mg carries no additional risk, and its onset of action is
a few hours earlier. I also give unfractionated heparin or
enoxaparin. Bivalirudin has not been tested in ACS. One
could also use fondaparinux as a background anticoagulant.
I also might give a glycoprotein IIb/IIIa inhibitor for a large
AMI. During a PCI, I give only heparin, and no anticoagulant therapy afterwards. I have patients ambulate early and
discharge them in 2 or 3 days on aspirin and clopidogrel.
Dr. Yusuf: What drugs do you prescribe for the
long-term?
Dr. Bates: I would also prescribe an angiotensin-converting enzyme inhibitor and a statin.
Dr. Roberts: What dose of statin?
Dr. Bates: I would start with 40 mg of atorvastatin or its
equivalvent.
Dr. Yusuf: I give unfractionated heparin, aspirin, and
clopidogrel, and quickly take them to the catheterization
laboratory for coronary angioplasty and implantation of a
bare metal stent. I then start long-term aspirin, a statin, an
angiotensin-converting enzyme inhibitor, and a ␤ blocker.
Dr. Roberts: Do you prefer an angiotensin-converting
inhibitor to an angiotensin receptor blocker?
Dr. Yusuf: I prefer the angiotensin-converting inhibitor,
ramipril.
Dr. Granger: The data are sufficiently strong that I also
give ezetimibe following a primary PCI.
Dr. Yusuf: What if coronary angioplasty is not immediately available?
Dr. Bates: In the first 3 hours, I would give a bolus of a
fibrinolytic agent, aspirin, clopidogrel, probably enoxaparin,
and a ␤ blocker if the patient is hemodynamically stable.
Dr. Yusuf: I would give aspirin, clopidogrel, fondaparinux, and thrombolytic therapy—possibly streptokinase.
Dr. Granger: A major concern with fondaparinux is its
association with increased risk for thrombosis with PCI,
based on the findings of OASIS-5 and OASIS-6. This complication is overcome by adding unfractionated heparin in a
dose of about 50 U/kg, which seems to prevent or at least
substantially reduce the risk of catheter thrombus without
causing a major increase in bleeding.
979
Dr. Bates: There are enough data now available to justify transitioning patients on enoxaparin through the catheterization laboratory without having to switch anticoagulants. There is a theory that patients already receiving
fondaparinux can receive whatever anticoagulant one ordinarily uses, in the same dose.
Dr. Yusuf: We have found an increased incidence of
PCI-induced catheter thrombosis with both enoxaparin and
fondaparinux in a study of several hundred patients, and I
agree with using unfractionated heparin in them. The important point is that there is no single agent that suffices in
all patients with ACS.
Dr. Bates: Another question is whether, in the USA—
where hospitals stays are typically only 2 to 3 days with
ACS—patients should continue to receive subcutaneous anticoagulant therapy for at least a few days after hospital
discharge.
Dr. Yusuf: The answer is possibly “yes,” because in
OASIS-5 it was clear that, with fondaparinux, there was a
striking reduction in bleeding complications regardless of
the length of hospital stay.
Dr. Granger: The FRISC II trial found that patients
should continue to receive antithrombin therapy until the
time of revascularization. In this setting, fondaparinux is
highly effective and has a very low risk of bleeding complications.
Dr. Friedewald: What are the unanswered questions
about anticoagulant therapy for ACS?
Dr. Bates: An important question is whether we should
be giving subcutaneous anticoagulation therapy longer than
the 48 hours of intravenous anticoagulation therapy recommended in current guidelines.
Dr. Granger: We need more data on the best way to use
fondaparinux, including its use in patients having coronary
angioplasty. Until we learn more, we should use it in combination with unfractionated heparin. Another question is
whether we should use glycoprotein IIb/IIIa inhibitors as
aggressively as some advocate and if so, when and for how
long? And what are the optimal doses and timing of drugs—
including aspirin—when using an antithrombotic protocol?
Dr. Yusuf: The most important question is, “How can
we institute earlier treatment of ACS patients with thrombolytic drugs, primary PCI, and antithrombotic agents?”
Earlier treatment will have the greatest impact. A second
issue is determining what can be achieved using other ACS
treatment models, including glucose-lowering and treatment
with antioxidants and free radicals.
Dr. Roberts: The best antithrombotic and antilytic therapy might be a statin taken before the onset of ACS.
Dr. Friedewald: Thank you.
Needs Assessment: The need for this activity for cardiologists and other healthcare specialists in cardiovascular
medicine is based on the following:
1. In 2007, ⬎1.2 million persons in the USA will have a
new or recurrent AMI.17
2. Acute occlusive thrombus is present in most persons
having a transmural AMI.
980
The American Journal of Cardiology (www.AJConline.org)
3. There are a variety of interventional and medical
strategies for treating acute coronary thrombosis.
4. No drug used alone is considered optimal therapy for
acute coronary thrombosis.
5. The risk of hemorrhagic complications must be considered when using antithrombotic therapy.
Target Audience: This activity is designed for cardiologists and all other healthcare 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, designates this educational activity for a maximum of 1
AMA PRA Category 1 Credit(s).™ Physicians should only
claim credit commensurate with the extent of their participation in the activity.
The A. Webb Roberts Center for Continuing Medical
Education for Baylor Health Care System, Dallas, is accredited by the Accreditation Council for Continuing Medical
Education to provide continuing medical education for
physicians.
CME Instructions: After reading this article, go online at www.AJConline.org to register, complete a posttest with a minimum score of 80%, complete an evaluation
and print a certificate.
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: September 2007
Termination Date: September 2008
1. Davies MJ, Woolf N, Robertson WB. Pathology of acute myocardial
infarction with particular reference to occlusive coronary thrombi. Brit
Heart J 1976;38:659 – 664.
2. De Wood MA, Spores J, Notske R, Mouser LT, Burroughs R, Golden MS,
Lang HT. Prevalence of total coronary occlusion during the early hours of
transmural myocardial infarction. N Engl J Med 1980;303:897–902.
3. Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of
coronary artery disease and the acute coronary syndromes. N Engl
J Med 1992;326:242–250;310 –318.
4. Fulton RM, Duckett K. Plasma-fibrinogen and thromboemboli after
myocardial infarction. Lancet 1976;2:1161–1164.
5. Thomas AC, Knapman A, Kirkler DM, Davies MJ. Community
study of the causes of “natural” sudden death. BMJ 1988;
297(6661):1453–1456.
6. FRISC Study Group. Low-molecular-weight heparin during instability
in coronary artery disease. Lancet 1996;347:561–568.
7. The Fifth Organization to Assess Strategies in Acute Ischemic
Syndromes Investigators. Comparison of fondaparinux and
enoxaparin in acute coronary syndromes. N Engl J Med 2006;354:
1464 –1476.
8. The Fifth Organization to Assess Strategies in Acute Ischemic
Syndromes Investigators. Effects of fondaparinux on mortality
and reinfarction in patients with acute ST-segment elevation myocardial infarction: the OASIS-6 randomized trial. JAMA 2006;295:
1519 –1530.
9. Turpie AGG, Bauer KA, Eriksson BI, Lassen MR, Fondaparinux vs
enoxaparin for the prevention of venous thromboembolism in major
orthopaedic surgery. A meta-analysis of 4 randomized double blind
studies. Arch Intern Med 2003;62:1833–1840.
10. Buller HR, Davidson BL, Decousus H, Gallus A, Gent M, Piovella F,
Prins MH, Raskob G, Segers AEM, Cariou R, et al, for the Matisse
Investigators. Fondaparinux or enoxaparin for the initial treatment of
symptomatic deep venous thrombosis: a randomized trial: Ann Intern
Med 2004;140:867– 873.
11. The Matisse Investigators. Subcutaneous fondaparinux versus intravenous unfractionated heparin in the initial treatment of pulmonary
embolism. N Engl J Med 2003;349:1695–1702; erratum, N Engl J Med
2004;350:423.
12. The Oasis-6 Trial Group. Effects of fondaparinux on mortality and
reinfarction in patients with acute ST-segment elevation myocardial
infarction. JAMA 2006;295:1519 –1530.
13. Stone GW, McLaurin BT, Cox DA, Bertrand ME, Lincoff AM, Moses
JW, White HD, Pocock SJ, Ware JH, Feit F, et al, for the ACUITY
Investigators. Bivalirudin for patients with acute coronary syndromes.
N Engl J Med 2006;355:2203–2216.
14. Antman EM, Morrow DA, McCabe CH, Murphy SA, Ruda M, Sadowski Z, Budaj A, López-Sendón JL, Guneri S, Jiang F, et al, for the
ExTRACT-TIMI 25 Investigators. Enoxaparin versus unfractionated
heparin with fibrinolysis for ST-elevation myocardial infarction.
N Engl J Med 2006;354:1477–1488.
15. Schühlen H, Hadamitzky M, Walter H, Ulm K, Schömig A. Major
benefit from antiplatelet therapy for patients at high risk for adverse
cardiac events after coronary Palmaz-Schatz stent placement: analysis
of a prospective risk stratification protocol in the Intracoronary Stenting and Antithrombotic Regimen (ISAR) trial. Circulation 1997;95(8):
2015–2021.
16. Singh KP, Roe MT, Peterson ED, Chen AY, Mahaffey KW, Goodman
SG, Harrington RA, Smith SC Jr, Gibler WB, Ohman EM, Pollack CV
Jr; for the CRUSADE Investigators. Low-molecular-weight heparin
compared with unfractionated heparin for patients with non-ST-segment elevation acute coronary syndromes treated with glycoprotein
IIb/IIIa inhibitors: results from the CRUSADE initiative. J Thromb
Thrombolysis 2006;21:211–220.
17. American Heart Association Statistics Committee and Stroke Statistics
Committee. Heart disease and stroke statistics—2007 update. Circulation 2007;115:e69 – e171.
Ethnic Differences in Coronary Artery Calcium in a Healthy
Cohort Aged 60 to 69 Years
Joan M. Fair, ANP, PhDa,*, Alexandre Kiazand, MDa, Ann Varady, MSa,
Mohammed Mahbouba, MDa, Linda Norton, MSNa, Geoffrey D. Rubin, MDb,
Carlos Iribarren, MD, PhDd, Alan S. Go, MDd,e,f,g, Mark A. Hlatky, MDc,
and Stephen P. Fortmann, MDa
Measurement of coronary artery calcium (CAC) has been proposed as a screening tool, but
CAC levels may differ according to race and gender. Racial/ethnic and gender distributions
of CAC were examined in a randomly selected cohort of 60- to 69-year-old healthy subjects.
Demographic, race/ethnicity (R/E), and clinical characteristics and assessment of CAC
were collected. There were 723 white/European, 105 African-American, 73 Hispanic, and
67 East Asian subjects (597 men, 369 women) included in this analysis. Men had a
significantly higher prevalence of any CAC (score >10) than women (76% vs 41%;
p <0.0001). For men, the unadjusted odds of having any CAC was 2.2 (95% confidence
interval [CI] 1.3 to 3.8) for whites compared with African-Americans. For women, CAC
scores were not significantly different across ethnic groups. After adjustment for coronary
risk factors, African-American and East Asian R/E remained associated with a lower
prevalence of CAC in men (adjusted odds ratios [ORs] 0.33 and 0.47, respectively), as well
as older age (OR 1.2, 95% CI 1.1 to 1.3), known hyperlipidemia (OR 1.7, 95% CI 1.1 to 2.7),
and history of hypertension (OR 2.2, 95% CI 1.4 to 3.3). In women, Asian R/E (OR 2.5, 95%
CI 1.1 to 5.7), history of smoking (adjusted OR 2.8, 95% CI 1.3 to 6.1), and known
hyperlipidemia (adjusted OR 2.0, 95% CI 1.3 to 3.1) were associated with a higher
prevalence of CAC independent of other risk factors. In conclusion, our data indicate that
the presence of CAC varied significantly across selected race/ethnic groups independent of
traditional cardiovascular risk factors. © 2007 Elsevier Inc. All rights reserved. (Am J
Cardiol 2007;100:981–985)
Measurement of coronary artery calcium (CAC) has been
proposed as a screening tool for coronary artery disease, but
CAC levels may differ according to race and gender.1– 6 In this
report, we examined race/ethnicity (R/E) and gender distributions of CAC in a randomly selected cohort of subjects aged 60
to 69 years who were free of known coronary disease and
participated in the Atherosclerotic Disease, VAscular FunctioN and GenetiC Epidemiology (ADVANCE) Study.
Methods
This cross-sectional analysis was performed in a presumably healthy older control cohort recruited for a study investigating genetic and environmental determinants of atherosclerosis. Details of the full study recruitment were
published elsewhere.7–9 Briefly, the older control cohort
a
Stanford Prevention Research Center, Stanford University, Stanford,
California; Departments of bRadiology and cHealth Research and Policy,
Stanford University School of Medicine, Stanford; dKaiser Permanente of
Northern California Division of Research, Oakland; and Departments of
e
Epidemiology, fBiostatistics, and gMedicine, University of California, San
Francisco, San Francisco, California. Manuscript received January 20,
2007; revised manuscript received and accepted April 13, 2007.
This work was supported in part by a grant from the Donald W.
Reynolds Foundation, Las Vegas, Nevada.
*Corresponding author: Tel.: 650-723-0774; fax: 650 725-6247.
E-mail address: [email protected] (J.M. Fair).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.038
included subjects aged 60 to 69 years identified from the
automated databases of Kaiser Permanente of Northern California, a large integrated health care plan providing care for
⬎3 million members in the San Francisco Bay area. An
initial pool of 84,590 potentially eligible subjects were
identified after exclusion of those with diagnostic codes for
cardiovascular, cerebrovascular, or peripheral vascular disease; congestive heart failure; malignancies; significant
liver or renal disease; or outpatient dementia or who lived
⬎50 miles from the research facility. Recruitment letters
were sent to subjects randomly selected from the eligible
pool in lots of 200 until a recruitment goal of ⱖ1,000
subjects was achieved. Recruitment was adjusted in the final
year from a random sampling strategy to oversample men
and persons of color, particularly Hispanic and AfricanAmerican men, to better frequency match the gender and
R/E distribution of the case cohorts in the ADVANCE
Study.
The final older control cohort consisted of 639 men and
385 women free of diagnosed cardiovascular disease. At
study baseline, all subjects completed a comprehensive
clinic visit, including a health survey, blood samples for
biomarker and genetic testing, blood pressure, anthropometric measurements, ankle-brachial index, electrocardiogram
at rest, brachial artery reactivity, heart rate variability, and
coronary computed tomography. Risk factors were measured using standardized protocols. Standardized and valiwww.AJConline.org
982
The American Journal of Cardiology (www.AJConline.org)
Table 1
Demographic, clinical history, and risk factor characteristics of 597 men without known cardiovascular disease who underwent screening for coronary
artery calcium from December 2001 to January 2004
Variable
White
(n ⫽ 443)
African-American
(n ⫽ 71)
Hispanic
(n ⫽ 44)
East Asian
(n ⫽ 41)
Age (yrs)
Education ⬍12th grade
Family history of coronary artery disease
Hyperlipidemia
Hypertension
Diabetes mellitus
Metabolic syndrome
On statin drug
On hypertension drug
Ever smoked
Total cholesterol (mg/dl)
Low-density lipoprotein (mg/dl)
High-density lipoprotein (mg/dl)
Triglycerides (mg/dl)
Cholesterol/high-density lipoprotein ratio
Body mass index (kg/m2)
Waist circumference (cm)
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Framingham risk score
Fasting glucose (mg/dl)
Fasting insulin (␮U/ml)
C-Reactive protein (mg/L)
Kilocalories from saturated fat (%)
Physical activity (kcal/kg/d)
65.9 ⫾ 3
11 (2.5%)
205 (46%)
173 (39%)
175 (40%)
80 (18%)
113 (26%)
102 (23%)
183 (41%)
288 (65%)
201 ⫾ 35
124 ⫾ 30
50 ⫾ 13
146 ⫾ 96
4.3
28 ⫾ 5
100 ⫾ 13
131 ⫾ 16
75 ⫾ 8
17.6 ⫾ 0.1
106 ⫾ 25
11.8 ⫾ 10
2.6 ⫾ 6
10.4 ⫾ 3
35 ⫾ 4
65.7 ⫾ 3
4 (5.6%)
18 (25%)*
35 (50%)
45 (63%)*
23 (32%)*
17 (25%)
18 (25%)
44 (62%)*
48 (68%)
197 ⫾ 43
123 ⫾ 37
51 ⫾ 15
116 ⫾ 66*
4.0*
30 ⫾ 6*
102 ⫾ 15
136 ⫾ 22*
78 ⫾ 10*
17.7 ⫾ 0.1
108 ⫾ 32
22.5 ⫾ 51*
3.2 ⫾ 5
10.6 ⫾ 3
35 ⫾ 4
66.0 ⫾ 3
6 (13.6%)*
22 (50%)
21 (48%)
24 (55%)
15 (34%)*
18 (42%)*
12 (27%)
25 (57%)
28 (64%)
199 ⫾ 33
122 ⫾ 31
45 ⫾ 9
164 ⫾ 85
4.5
29 ⫾ 3
100 ⫾ 9
135 ⫾ 16
74 ⫾ 11
20.8 ⫾ 0.1*
110 ⫾ 23
14.0 ⫾ 10
2.2 ⫾ 2
9.9 ⫾ 3
35 ⫾ 3
65.8 ⫾ 2
0 (0%)
11 (27%)
16 (39%)
26 (63%)*
11 (27%)
9 (23%)
12 (29%)
26 (63%)*
21 (51%)
200 ⫾ 39
123 ⫾ 35
49 ⫾ 12
152 ⫾ 97
4.3
26 ⫾ 3*
90 ⫾ 10*
135 ⫾ 16
74 ⫾ 11
17.5 ⫾ 0.1
114 ⫾ 26
13.9 ⫾ 14
2.5 ⫾ 8
8.9 ⫾ 2*
35 ⫾ 3
Values expressed as mean ⫾ SD or proportions. White R/E is always the referent group.
* Significant difference at p ⱕ0.05.
dated questionnaires were used to collect physical activity
and food-frequency data (questionnaires available upon request).
In genetic studies, self-reported R/E status was generally
viewed as a surrogate measure of ancestral origin (which
will affect genetic variation), whereas in epidemiologic
studies, this variable was generally viewed as an indicator of
social and cultural influences. In this study, it was deemed
important to classify to balance both objectives. Self-reported R/E was collected during the eligibility screening
survey, on the baseline health survey, and from health plan
databases. In addition, information for self-reported birthplace, grandparents’ R/E, and grandparents’ country of origin was also collected. A uniform algorithm was used to
assign R/E. If self-reported R/E and all grandparents’ R/E
were concordant, subjects were coded to that R/E category
(⬎80% of subjects met these criteria). Discordant cases
were reviewed and the hierarchy of grandparents’ R/E,
grandparents’ country of origin, subjects’ self-reported ethnicity, and screening data self-reported R/E was applied to
assign proportional R/E ancestry. Subjects of such mixed
R/E ancestry with majority indication of Hispanic heritage
were assigned as Hispanic. Other admixtures were assigned
to a mixed non-Hispanic category.
The primary outcome for this study was the presence of
CAC. Detection and quantification of coronary calcium was
performed using 4- or 16-row detector computed tomography (Siemens Medical Solutions, Erlangen, Germany).
Electrocardiographic leads were attached to the subject’s
chest. After a scout image was acquired, an electrocardiographically-triggered breath-held scan through the heart was
performed using a 1.25- to 1.5-mm detector gated to 80% of
the RR interval. Two identical scans were acquired sequentially. Each scan was performed with x-ray tube potential of
120 kV, current of 100 mA, and gantry rotation of 360 to
400 ms. Data were reconstructed using a half-scan reconstruction algorithm to give an effective temporal resolution
of 200 ms per section and 3-mm section thickness. Both
traditional Agatston and volume scoring methods were used
to quantify coronary calcium.10 An attenuation threshold of
130 HU was applied for differentiating calcification from
soft tissues. A calcific lesion was defined as ⱖ2 contiguous
pixels with attenuation of ⱖ130 HU. Measurements were
made on the 2 scans, and Agatston total scores reported are
the average of scores from each of the 2 scans.
Demographic and clinical characteristics are presented as
mean ⫾ SD for continuous variables and count and proportion for categorical variables. Generalized linear models
were used to compare CAC scores across race/ethnic
groups, with white/European as the reference group. Multivariable logistic regression models were performed to
evaluate the association of R/E with the presence of any
CAC after accounting for the effect of traditional coronary
risk factors. CAC scores (log [CAC ⫹ 1]) and triglyceride
levels (log [tg ⫹ 1]) were log transformed for all analyses.
CAC score ⬎10 was used as the cut-off value for the
presence of coronary calcium.11 Because of marked differ-
Preventive Cardiology/Ethnic Differences in Coronary Artery Calcium
983
Table 2
Demographic, clinical history, and risk factor characteristics for 369 women without known cardiovascular disease who underwent screening for
coronary artery calcium from December 2001 to January 2004
Variable
White
(n ⫽ 280)
African-American
(n ⫽ 34)
Hispanic
(n ⫽ 29)
East Asian
(n ⫽ 26)
Age (yrs)
Education ⬍12th grade
Family history of coronary artery disease
Hyperlipidemia
Hypertension
Diabetes mellitus
Metabolic syndrome
On statin drug
On hypertension drug
Ever smoked
Total cholesterol (mg/dl)
Low-density lipoprotein (mg/dl)
High-density lipoprotein (mg/dl)
Triglycerides (mg/dl)
Cholesterol/high-density lipoprotein ratio
Body mass index (kg/m2)
Waist circumference (cm)
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Framingham risk score
Fasting glucose (mg/dl)
Fasting insulin (␮U/ml)
C-Reactive protein (mg/L)
Kilocalories from saturated fat (%)
Physical activity (kcal/kg/d)
66 ⫾ 3
7 (2.5%)
150 (54%)
96 (34%)
111 (40%)
25 (9%)
60 (21%)
33 (12%)
105 (38%)
149 (53%)
213 ⫾ 35
126 ⫾ 31
61 ⫾ 17
135 ⫾ 80
3.7 ⫾ 1
28 ⫾ 6
85 ⫾ 14
130 ⫾ 18
72 ⫾ 9
8.4 ⫾ 0.0
99 ⫾ 18
10.3 ⫾ 9
4.3 ⫾ 7
10.1 ⫾ 3
34 ⫾ 2
65.8 ⫾ 3
0 (0%)
11 (32%)*
13 (38%)
23 (68%)*
7 (21%)*
10 (30%)
8 (24%)
21 (62%)*
22 (65%)
207 ⫾ 38
122 ⫾ 36
64 ⫾ 19
107 ⫾ 52*
3.5 ⫾ 1
30 ⫾ 7*
94 ⫾ 17*
130 ⫾ 17
74 ⫾ 9
8.3 ⫾ 0.1
103 ⫾ 22
17.2 ⫾ 15*
6.1 ⫾ 8
10.1 ⫾ 3
33 ⫾ 2
66.1 ⫾ 3
5 (17.2%)*
14 (48%)
10 (34%)
10 (34%)
6 (21%)
9 (35%)
4 (14%)
6 (21%)
10 (34%)
200 ⫾ 43
117 ⫾ 35
57 ⫾ 17
135 ⫾ 64
3.7 ⫾ 1
29 ⫾ 5
89 ⫾ 12
132 ⫾ 16
73 ⫾ 8
9.2 ⫾ 0.0
105 ⫾ 20
15.2 ⫾ 21*
4.9 ⫾ 6
9.9 ⫾ 3
34 ⫾ 3
65.7 ⫾ 3
0 (0%)
8 (31%)*
8 (31%)
11 (42%)
4 (15%)
5 (20%)
3 (12%)
10 (38%)
3 (12%)*
208 ⫾ 38
126 ⫾ 35
57 ⫾ 11
126 ⫾ 58
3.8 ⫾ 1
25 ⫾ 4*
79 ⫾ 10*
138 ⫾ 27*
73 ⫾ 10
9.1 ⫾ 0.0
106 ⫾ 26
11.2 ⫾ 8
2.5 ⫾ 4
8.5 ⫾ 2*
34 ⫾ 2
Values expressed as mean ⫾ SD or proportions. White R/E is always the referent group.
* Significant difference at p ⱕ0.05.
ences in CAC distributions for women and men, analyses
were stratified by gender.
Results
A total of 1,013 participants aged 60 to 69 years had CAC
measured, including 723 white/European, 105 AfricanAmerican, 73 Hispanic, and 67 East Asian subjects (597
men, 369 women). South Asians (n ⫽ 11) and mixed nonHispanics (n ⫽ 34) were excluded because of small sample
sizes, leaving 968 subjects.
Demographic, clinical history, and risk factor characteristics differed among R/E groups, primarily between white
and African-American men (Table 1). Compared with white
men, African-American men had significantly higher mean
body mass indexes, systolic and diastolic blood pressures,
fasting insulin levels, and prevalences of self-reported history of hypertension and diabetes. However, African-American
men had significantly lower mean triglyceride levels and
total cholesterol/high-density cholesterol ratios and a lower
prevalence of family history of coronary artery disease.
Compared with white men, Hispanic men had significantly
higher Framingham risk scores, history of diabetes mellitus,
and metabolic syndrome and lower educational attainment.
Compared with white men, Asian men reported a significantly higher history of hypertension and hypertensive drug
use, but lower family history of coronary artery disease,
body mass index, waist circumference, and dietary saturated
fat intake. For women, the pattern of differences in clinical
characteristics across race/ethnic groups was similar to that
of men, with the exception that compared with white
women, Asian women reported a significantly lower rate of
ever smoking (Table 2).
Men had a significantly higher prevalence of any CAC
(defined as CAC score ⬎10) than women (75.5% vs 40.9%;
p ⬍0.0001). The distribution of CAC scores in men differed
significantly across all ethnic groups (Table 3), primarily
because of differences between white and African-American men (median CAC score 145 vs 43; p ⫽ 0.001). Unadjusted odds of having any measurable CAC were 2.2-fold
higher in white men compared with African-American men.
In women, median CAC scores were not significantly different across ethnic groups. In contrast to findings in men,
median and 90th percentile CAC scores were higher in
African-American women compared with white women.
However, differences did not reach statistical significance.
We next conducted multivariable logistic regression
models to examine the association between R/E and any
CAC separately in men and women after adjustment for
traditional coronary risk factors (Table 4). In men, AfricanAmerican and East Asian R/E were significantly associated
with a lower prevalence of CAC (adjusted odds ratios [ORs]
0.33 and 0.47, respectively), whereas older age (OR 1.2,
95% confidence interval [CI] 1.1 to 1.3), history of hyperlipidemia (OR 1.7, 95% CI 1.1 to 2.7), and history of
hypertension (OR 2.2, 95% CI 1.4 to 3.3) were associated
with a higher prevalence of CAC. In women, Asian R/E
984
The American Journal of Cardiology (www.AJConline.org)
Table 3
Racial/ethnic differences in the prevalence and distribution of coronary artery calcium in a healthy cohort of men and women aged 60 to 69 years
R/E
Men
CAC median
25th percentile
90th percentile
Prevalence of CAC ⬎10 (%)
Women
CAC median
25th percentile
90th percentile
Prevalence of CAC ⬎10 (%)
White
African-American
Hispanic
East Asian
443 (70%)
145
15.3
1,297
78
280 (73%)
1.7
0
175
39
71 (11%)
43*
0.8
696
62
34 (9%)
2.4
0
593
44
44 (7%)
106
8.2
1,042
75
29 (8%)
0.8
0
412
38
41 (7%)
54
6.9
856
68
26 (7%)
12.9
0
283
58
* Kruskal Wallis p ⬍0.008 compared with white.
Table 4
Adjusted odds of any coronary artery calcium among race/ethnic groups
in a healthy older cohort aged 60 to 69 years
R/E
White
African-American
Hispanic
East Asian
Adjusted OR* (95% CI)
Men
Women
Referent group
0.33 (0.2–0.6)
0.69 (0.3–1.4)
0.47 (0.2–0.98)
Referent group
1.2 (0.6–2.5)
0.9 (0.4–2.1)
2.5 (1.1–5.7)
* Adjusted for the composite variables included in the Framingham risk
score of age, current cigarette smoking, high-density lipoprotein, history of
hyperlipidemia, history of hypertension, and diabetes mellitus.
(adjusted OR 2.5), ever smoking (OR 2.8, 95% CI 1.3 to
6.1), and history of hyperlipidemia (OR 1.96, 95% CI 1.3 to
3.1) were associated with a higher prevalence of CAC
independent of other risk factors. CAC prevalence did not
differ significantly between African-American and white
women after adjustment for coronary risk factors.
Discussion
In this sample of healthy 60- to 69-year-old men and
women, men had a higher prevalence of detectable CAC
than women in all race/ethnic groups. We provide new
information on the prevalence of CAC across race/ethnic
groups stratified by gender. For men, CAC score was highest in whites compared with other ethnic groups. After
adjustment for coronary risk factors, the odds of any CAC
was ⬎65% lower in African-American men and ⬎50%
lower in East Asians compared with white men. Previous
studies of CAC levels in different ethnic groups yielded
mixed results. In the Multi-Ethnic Study of Atherosclerosis
(MESA), a large study of 4 ethnic groups with a mean age
of 63 years, the reported prevalence of any CAC was highest in white men (70.4%), lower in Chinese (59%) and
Hispanics (56%), and lowest in African-American men
(52%).1 Findings comparing whites and African-Americans
were consistent with our results in a slightly older cohort
(mean age 66 years) with CAC prevalences of 78% in white
men and 62% in African-American men. In contrast, 2
population-based studies involving younger subjects found
no significant differences between these 2 racial groups.11,12
However, gender-specific results were not reported. MESA
also found that Chinese men ⬎65 years had the lowest CAC
scores compared with other racial groups.1 In our study of
60- to 69-year-olds, CAC scores were significantly lower in
East Asian compared with white men after adjustment for
risk factors, but not lower than for African-American men.
Of note, our study included several East Asian groups,
whereas MESA included only Chinese.
For the women in our study, CAC was highest in the East
Asian and lowest in the Hispanic subgroups, but differences
across the 4 ethnic groups were not statistically significant.
However, after adjustment for coronary risk factors, the
likelihood of CAC was 2.5 times higher for East Asian
women compared with white women, which was significant. Only a few studies examined CAC prevalence by
race/ethnic subgroups in asymptomatic women.1,5,13 Studies
comparing CAC prevalence in women reported a lower
prevalence in African-Americans compared with whites,
although the significance was attenuated after adjustment of
coronary risk factors. In MESA, reported prevalences of
CAC (⬎0) in women were 44.6% in whites, 36.5% in
African-Americans, and 41.9% in Chinese.1 In our slightly
older cohort of women, prevalences of CAC (⬎10) were
39.3% in whites, 44.1% in African-Americans, and 57.7%
in East Asians. These findings should be treated cautiously
given the small number of East Asian women in our study.
MESA results were based on a much larger sample drawn in
multiple sites. A recent study comparing CAC in white and
Filipino women (aged 55 to 78 years) found no significant
difference in prevalence rates for CAC between these 2
groups (52% vs 44%, respectively), although Filipino
women had higher incidences of type 2 diabetes and metabolic syndrome than white women.13 In MESA, diabetes
was significantly lower in white women compared with all
other ethnic groups.1 The magnitude of difference in diabetic status for the women in our study was similar to that
observed in MESA. However, the difference between white
and Asian women was not statistically significant, likely
because of a smaller sample size. Although Asian women
generally had higher rates of diabetes than whites, the prevalence of CAC was at least as high as that in whites,
although coronary heart disease and coronary heart disease
mortality rates were lower.14
The evidence from this study extended the findings of
other population studies that CAC prevalence differed significantly by gender and ethnicity. CAC may not fully
Preventive Cardiology/Ethnic Differences in Coronary Artery Calcium
represent the true atherosclerosis burden in older populations, especially African-Americans.15 The finding of higher
CAC levels in East Asian women with generally lower
risk-factor profiles requires replication in larger samples.
Overall, these findings suggest a need for rigorous attention
to risk-factor reduction for both men and women of all
racial/ethnic groups, even at levels of CAC not currently
considered clinically significant.
1. Bild DE, Detrano R, Peterson D, Guerci A, Liu K, Shahar E, Ouyang
P, Jackson S, Saad MF. Ethnic differences in coronary calcification:
the Multi-Ethnic Study of Atherosclerosis (MESA). Circulation 2005;
111:1313–1320.
2. Greenland P, LaBree L, Azen SP, Doherty TM, Detrano RC. Coronary
artery calcium score combined with Framingham score for risk prediction in asymptomatic individuals. JAMA 2004;291:210 –215.
3. Khurana C, Rosenbaum CG, Howard BV, Adams-Campbell LL, Detrano RC, Klouj A, Hsia J. Coronary artery calcification in black
women and white women. Am Heart J 2003;145:724 –729.
4. Lee TC, O’Malley PG, Feuerstein I, Taylor AJ. The prevalence and
severity of coronary artery calcification on coronary artery computed
tomography in black and white subjects. J Am Coll Cardiol 2003;41:39–44.
5. Newman AB, Naydeck BL, Whittle J, Sutton-Tyrrell K, Edmundowicz
D, Kuller LH. Racial differences in coronary artery calcification in
older adults. Arterioscler Thromb Vasc Biol 2002;22:424 – 430.
6. Reaven PD, Thurmond D, Domb A, Gerkin R, Budoff MJ, Goldman
S. Comparison of frequency of coronary artery calcium in healthy
Hispanic versus non-Hispanic white men by electron beam computed
tomography. Am J Cardiol 2003;92:1198 –1200.
7. Go AS, Iribarren C, Chandra M, Lathon PV, Fortmann SP, Quertermous T, Hlatky MA. Statin and beta-blocker therapy and the initial
8.
9.
10.
11.
12.
13.
14.
15.
985
presentation of coronary heart disease. Ann Intern Med 2006;144:
229 –238.
Iribarren C, Go AS, Husson G, Sidney S, Fair JM, Quertermous T,
Hlatky MA, Fortmann SP. Metabolic syndrome and early-onset coronary artery disease: is the whole greater than its parts? J Am Coll
Cardiol 2006;48:1800 –1807.
Taylor-Piliae RE, Norton LC, Haskell WL, Mahbouda MH, Fair JM,
Iribarren C, Hlatky MA, Go AS, Fortmann SP. Validation of a new
brief physical activity survey among men and women aged 60 – 69
years. Am J Epidemiol 2006;164:598 – 606.
Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M Jr.,
Detrano R. Quantification of coronary artery calcium using ultrafast
computed tomography. J Am Coll Cardiol 1990;15:827– 832.
Jain T, Peshock R, McGuire DK, Willett D, Yu Z, Vega GL, Guerra
R, Hobbs HH, Grundy SM. African Americans and Caucasians have a
similar prevalence of coronary calcium in the Dallas Heart Study. J Am
Coll Cardiol 2004;44:1011–1017.
Bild DE, Folsom AR, Lowe LP, Sidney S, Kiefe C, Westfall AO,
Zheng ZJ, Rumberger J. Prevalence and correlates of coronary calcification in black and white young adults: the Coronary Artery Risk
Development in Young Adults (CARDIA) Study. Arterioscler Thromb
Vasc Biol 2001;21:852– 857.
Araneta MR, Barrett-Connor E. Subclinical coronary atherosclerosis in
asymptomatic Filipino and white women. Circulation 2004;110:2817–
2823.
Thom T, Haase N, Rosamond W, Howard VJ, Rumsfeld J, Manolio T,
Zheng ZJ, Flegal K, O’Donnell C, Kittner S, et al. Heart disease and
stroke statistics—2006 update: a report from the American Heart
Association Statistics Committee and Stroke Statistics Subcommittee.
Circulation 2006;113:e85–151.
Manolio TA, Bild DE. Coronary calcium, race, and genes. Arterioscler
Thromb Vasc Biol 2002;22:359 –360.
Endotoxemia, Inflammation, and Atrial Fibrillation
Christopher J. Boos, MDa,b,*, Gregory Y.H. Lip, MDa, and Bernd Jilma, MDc
There is emerging evidence to support a link between inflammation and atrial fibrillation
(AF). Lipopolysaccharide (LPS) infusion is a well established experimental model used to
investigate host systemic inflammatory responses. It was hypothesized that LPS challenge,
by virtue of its provoked host inflammatory response, might increase the propensity to the
development of new-onset AF. A post hoc analysis was performed of prospective data
collected for 652 healthy men (mean age 27 ⴞ 5 years, all without history of AF) who were
challenged with LPS according to a standard experimental protocol. All subjects underwent
a detailed health screening before inclusion. After LPS challenge, all subjects underwent
continuous cardiac monitoring for a minimum of 8 hours and were reviewed again using
rhythm assessments at 24 hours and after 7 ⴞ 3 days. Effects of LPS on high-sensitivity
C-reactive protein, interleukin-6, tumor necrosis factor-␣, and neutrophil counts as indexes
of an inflammatory response were also assessed. LPS led to overall marked increases in
high-sensitivity C-reactive protein, interleukin-6, tumor necrosis factor-␣, and neutrophil
counts (all p <0.0001) and an average temperature increase of 1.1°C. There was no
evidence of new-onset AF in the subjects challenged. In conclusion, experimental LPS
challenge led to a significant increase in acute inflammatory indexes, but did not increase
the propensity to new-onset AF in a young low-risk population. © 2007 Elsevier Inc. All
rights reserved. (Am J Cardiol 2007;100:986 –988)
There is increasing evidence to support an association between inflammation and atrial fibrillation (AF). An association between AF and a variety of inflammatory markers
(e.g., high-sensitivity C-reactive protein [hs-CRP], interleukin-6 [IL-6], tumor necrosis factor-␣ [TNF-␣], and white
blood cell count) has been shown by a number of clinical
studies.1–7 Furthermore, it would seem that hs-CRP may
relate to the clinical burden of AF.7 Atrial biopsy specimens
from subjects with lone AF have shown the presence of
inflammatory infiltrates.1,8 Additionally, studies of anti-inflammatory– based therapies (e.g., steroids) supported their
potential to decrease the burden of AF with a simultaneous
decrease in hs-CRP.1,6 Lipopolysaccharide (LPS) infusion is
a well-established and extensively published experimental
model used to study host systemic inflammatory and coagulation responses to infection.9 –13 LPS infusion was consistently shown to cause marked increases in the host systemic
inflammatory response.9 –13 If acute inflammation per se
leads to the development of AF, we tested the hypothesis
that LPS infusion might be linked to an excess of new-onset
AF, even in healthy subjects.
Methods
We performed a post hoc analysis of prospective data collected from 652 healthy men challenged with LPS endoa
Haemostasis, Thrombosis and Vascular Biology Unit, University Department of Medicine, City Hospital, Birmingham; and bArmy Medical
Directorate, Camberley, United Kingdom; and cDepartment of Clinical
Pharmacology, Division of Hematology and Immunology, Medical University of Vienna, Vienna, Austria. Manuscript received February 28,
2007; revised manuscript received and accepted April 13, 2007.
*Corresponding author: Tel.: 44-7973-840-309; fax 121-554-4083.
E-mail address: [email protected] (C.J. Boos).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.039
toxin according to a standard protocol. This encompassed a
series of studies assessing the effects of LPS challenge on
aspects of inflammation, coagulation, and endothelial activation, as well as incorporating several randomized controlled trials of drug therapy aimed at controlling host systemic inflammatory response to sepsis (for detailed review
see Mayr and Jilma11).9 –13 Health status was determined
using a detailed clinical history and physical examination.
All included subjects were also required to have a normal
12-lead electrocardiogram, plasma glucose, full blood count,
renal and liver function, virologic status (including hepatitis
B and C and human immunodeficiency virus), thrombophilia, and drug screen results.
Subjects were admitted to the study ward at 8 A.M. after
an overnight fast. LPS was administered as an intravenous
bolus of 2 ng/kg (national reference endotoxin, Escherichia
coli; lot g1; US Pharmacopoeia Convention, Inc., Rockville,
Maryland) over 2 minutes. This was followed in all subjects
by an infusion of 0.9% normal saline solution at 200 ml/
hour for the first 6 hours to maintain adequate hydration.
Flu-like symptoms were induced by LPS infusion that lasted
for 4 to 6 hours. No serious or severe adverse effects were
observed to date by us or reported by others, to our knowledge.
All subjects were carefully monitored (including continuous cardiac monitoring) in bed for the first 8 hours and
allowed home after 8 to 12 hours, to return the following
morning at 24 hours after the challenge, when they underwent a repeated analysis of cardiac rhythm. Furthermore, a
repeat electrocardiogram was obtained for all patients at 3 to
7 days after the challenge. Peripheral blood sampling for
full blood count, as well as established inflammatory markers (discussed in the following), were collected in ethylenediaminetetraacetic acid bottles by Vacutainer (BectonDickinson, New Jersey) using separate venipunctures (using
www.AJConline.org
Arrhythmias and Conduction Disturbances/Endotoxemia and Atrial Fibrillation
987
Table 1
Baseline demographics of included subjects and peak changes after
endotoxin challenge
Table 2
Correlations between peak values of inflammatory parameters
Variable
TNF-␣
IL-6
hs-CRP
Neutrophil count
Caucasians
Age (yrs)
Men
Body mass index (kg/m2)
Mean arterial blood pressure (mm Hg)
Baseline
6 Hours
Systolic blood pressure (mm Hg)
Baseline
6 Hours
Diastolic blood pressure (mm Hg)
Baseline
6 Hours
Heart rate (beats/min)
Baseline
5 Hours
Temperature (°C)
Baseline
4 Hours
TNF-␣ (pg/ml)
Baseline
2 Hours
CRP
Baseline
24 Hours
Neutrophil count (⫻ 109/L)
Baseline
6 Hours
IL-6 (pg/ml)
Baseline
2 Hours
Result
p Value
652 (97%)
27.0 ⫾ 5.0
100%
23.0 ⫾ 2.0
83.0 ⫾ 9.0
77 ⫾ 8.0
125.0 ⫾ 11.0
120.0 ⫾ 10.0
TNF-␣
IL-6
hs-CRP
Neutrophil Count
—
0.66*
0.46*
0.40*
0.66*
—
0.27†
0.42*
0.46*
0.27†
—
0.19†
0.40*
0.42*
0.19†
—
* p ⬍0.001;† p ⬍0.05, Spearman ranks correlation test.
⬍0.001
0.001
64.0 ⫾ 9.0
57.0 ⫾ 8.0
⬍0.001
67.0 ⫾ 10.0
87.0 ⫾ 10.0
⬍0.001
36.2 ⫾ 0.4
37.3 ⫾ 0.6
⬍0.001
0.8 (0.5–1.2)
172 (95.6–280.0)
⬍0.001
0.07 (0.04–0.12)
2.5 (2.0–3.1)
⬍0.001
2.8 ⫾ 1.3
8.2 ⫾ 2.2
⬍0.001
1.0 (0.7–1.6)
212.0 (97.0–442.9)
⬍0.001
Values shown as mean ⫾ SD, number (percent), or median (interquartile
range).
the same arm each time) immediately before LPS and at a
minimum of 2, 4, 6, and 24 hours after LPS. The studies had
full ethical approval from the Ethics Committee of the
Medical University of Vienna, and written informed consent was obtained from all subjects.
The development of AF was defined as ⬎6 beats of an
irregularly irregular rhythm, with baseline fibrillation (f)
waves.14 We also examined the effects of LPS challenge on
inflammatory cytokines (hs-CRP, IL-6, and TNF-␣), as well
as differential blood counts, including neutrophil counts,
that were performed using an automated cell counter (Sysmex, Kobe, Japan). TNF-␣ and IL-6 were measured using
enzyme immunoassay (R&D Systems, Oxford, United
Kingdom). hs-CRP was quantified using a nephelometric
assay.15
All statistical analyses were performed using GraphPad
InStat (www.graphpad.com), version 3.00, for Windows 95
(Microsoft Corp., Redmond, Washington). Determination
of normality was assessed using data inspection and coupled
with the Kolmogorov-Smirnov test. All continuous data are
presented as mean ⫾ SD when normally distributed and
median and interquartile range for nonparametric data. For
paired data, we used paired t test and Wilcoxon’s matched-
pairs test for parametric and nonparametric data, respectively.
Results
LPS infusion led to marked inflammation, indicated by significantly increased hs-CRP (p ⬍0.0001), IL-6 (p ⬍0.0001), and
TNF-␣ (p ⬍0.0001), as well as neutrophil counts (p ⬍0.0001;
Table 1), and was associated with significant increases in both
temperature and heart rate. There were significant correlations
between all inflammatory markers studied (p ⬍0.05; Table 2).
There were no recorded or symptomatic episodes of AF
(or other arrhythmia) in any subject in response to LPS
infusion.
Discussion
In this study we showed for the first time that experimental
endotoxemia after LPS infusion did not lead to the development of acute new-onset AF in a large cohort of healthy
LPS-challenged subjects despite clear evidence of a marked
inflammatory response during the study period.
Whether inflammation genuinely has an important role in
the initiation and maintenance of AF is highly contentious
despite increasing support of the concept.1,7,16 Ellinor et al16
recently showed no evidence of increased hs-CRP in subjects with lone AF compared with healthy controls or a
relation between hs-CRP and AF burden. However, they
showed that hypertension, a condition commonly associated
with AF, was linked to increased CRP levels, suggesting
that inflammation was linked to the presence of underlying
cardiovascular disease rather than AF itself. This is in contrast to a previous report by Chung et al,7 in which not only
was hs-CRP higher in patients with lone AF compared with
controls, but there was a stepwise increase in hs-CRP with
increasing clinical AF burden (permanent greater than persistent greater than paroxysmal). However, in this later
study, 1/3 of patients with so-called lone AF had co-existing
hypertension despite no overt structural heart disease.7
These data might suggest that perhaps the inflammatory
stimulus needs to be sustained and/or exist in the presence
of a structurally abnormal heart or in patients with additional cardiovascular risk factors known to potentiate AF
development (e.g., the elderly or patients with diabetes
mellitus or hypertension).
If acute inflammation per se is linked to the development
and persistence of AF,1 we would expect to have noted an
excess of new-onset AF after endotoxin challenge. The
negative results in the present study are important in view of
the intense interest (and speculation) into inflammation/AF
coupled with increasing data suggesting that inflammation
988
The American Journal of Cardiology (www.AJConline.org)
may have an important pathophysiologic role in AF-related
thrombogenicity.17–19 It may be that for inflammation to
trigger or perpetuate AF, the inflammatory trigger needs to
be linked to corresponding localized inflammatory processes
within the atrial myocardium.1,8 This might explain, for
example, the increased propensity to AF (25% to 40%) after
cardiac surgery (and its associated myopericarditis).1,8,20,21
After coronary bypass surgery, the peak onset of AF tended
to occur on postoperative days 2 and 3 and coincided with
peak increases in white blood cell counts and hs-CRP.3,22
This study had several limitations. Although echocardiography was not undertaken as part of subject inclusion, it
can be largely assumed that because all subjects were
healthy with normal electrocardiograms and clinical history/
examination results, they would have had structurally normal hearts. However, we cannot assume that experimental
endotoxemia using LPS infusion is synonymous with clinical genuine infection/inflammation. Although there were
no observed episodes of AF, we cannot comment on the
frequency of isolated premature atrial or ventricular beats,
which are recognized triggers of AF.23,24 Moreover, we
cannot exclude that some LPS-challenged subjects may
have developed silent asymptomatic paroxysmal AF episodes subsequent to the study observation period. Finally,
we included only healthy persons, for whom the background tendency to AF would have been ⬍1%.14
1. Boos CJ, Anderson RA, Lip GY. Is atrial fibrillation an inflammatory
disorder? Eur Heart J 2006;27:136 –149.
2. Roldan V, Marin F, Martinez JG, Garcia-Herola A, Sogorb F, Lip GY.
Relation of interleukin-6 levels and prothrombin fragment 1⫹2 to a
point-based score for stroke risk in atrial fibrillation. Am J Cardiol
2005;95:881– 882.
3. Abdelhadi RH, Gurm HS, Van Wagoner DR, Chung MK. Relation of
an exaggerated rise in white blood cells after coronary bypass or
cardiac valve surgery to development of atrial fibrillation postoperatively. Am J Cardiol 2004;93:1176 –1178.
4. Conway DS, Buggins P, Hughes E, Lip GY. Predictive value of
indexes of inflammation and hypercoagulability on success of cardioversion of persistent atrial fibrillation. Am J Cardiol 2004;94:508 –510.
5. Sata N, Hamada N, Horinouchi T, Amitani S, Yamashita T, Moriyama
Y, Miyahara K. C-Reactive protein and atrial fibrillation. Is inflammation a consequence or a cause of atrial fibrillation? Jpn Heart J
2004;45:441– 445.
6. Dernellis J, Panaretou M. Relationship between C-reactive protein
concentrations during glucocorticoid therapy and recurrent atrial fibrillation. Eur Heart J 2004;25:1100 –1107.
7. Chung MK, Martin DO, Sprecher D, Wazni O, Kanderian A, Carnes
CA, Bauer JA, Tchou PJ, Niebauer MJ, Natale A, Van Wagoner DR.
C-Reactive protein elevation in subjects with atrial arrhythmias: inflammatory mechanisms and persistence of atrial fibrillation. Circulation 2001;104:2886 –2891.
8. Frustaci A, Chimenti C, Bellocci F, Morgante E, Russo MA, Maseri A.
Histological substrate of atrial biopsies in subjects with lone atrial
fibrillation. Circulation 1997;96:1180 –1184.
9. Boos CJ, Goon PK, Lip GY. The endothelium, inflammation, and
coagulation in sepsis. Clin Pharmacol Ther 2006;79:20 –22.
10. Derhaschnig U, Reiter R, Knobl P, Baumgartner M, Keen P, Jilma B.
Recombinant human activated protein C (rhAPC; drotrecogin alfa
[activated]) has minimal effect on markers of coagulation, fibrinolysis,
and inflammation in acute human endotoxemia. Blood 2003;102:
2093–2098.
11. Mayr FB, Jilma B. Coagulation interventions in experimental human
endotoxemia. Translational Research 2006;148:263–271.
12. Fiuza C, Suffredini AF. Human models of innate immunity: local and
systemic inflammatory responses. J Endotoxin Res 2001;7:385–388.
13. Jilma B. Safety of endotoxin challenge in healthy volunteers: bradycardia (lett). Intensive Care Med 2005;31:496.
14. Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen
KA, Halperin JL, Le Heuzey JY, Kay GN, Lowe JE, et al; American
College of Cardiology; American Heart Association Task Force; European Society of Cardiology Committee for Practice Guidelines;
European Heart Rhythm Association; Heart Rhythm Society. ACC/
AHA/ESC 2006 guidelines for the management of patients with atrial
fibrillation. Europace 2006;8:651–745
15. Derhaschnig U, Bergmair D, Marsik C, Schlifke I, Wijdenes J, Jilma
B. Effect of interleukin-6 blockade on tissue factor-induced coagulation in human endotoxemia. Crit Care Med 2004;32:1136 –1140.
16. Ellinor PT, Low A, Patton KK, Shea MA, Macrae CA. C-Reactive
protein in lone atrial fibrillation. Am J Cardiol 2006;97:1346 –1350.
17. Nakamura Y, Nakamura K, Fukushima-Kusano K, Ohta K, Matsubara
H, Hamuro T, Yutani C, Ohe T. Tissue factor expression in atrial
endothelia associated with nonvalvular atrial fibrillation: possible involvement in intracardiac thrombogenesis. Thromb Res 2003;111:
137–142.
18. Lip GY, Patel JV, Hughes E, Hart RG. High-sensitivity C-reactive
protein and soluble CD40 ligand as indices of inflammation and
platelet activation in 880 patients with nonvalvular atrial fibrillation:
relationship to stroke risk factors, stroke risk stratification schema, and
prognosis. Stroke 2007;38:1229 –1237.
19. Conway DS, Lip GY. Inflammation, arrhythmia burden and the thrombotic consequences of atrial fibrillation (lett). Eur Heart J 2004;25:
1761.
20. Crystal E, Connolly SJ, Sleik K, Ginger TJ, Yusuf S. Interventions on
prevention of postoperative atrial fibrillation in patients undergoing
heart surgery: a meta-analysis. Circulation 2002;106:75– 80.
21. Zimmer J, Pezzullo J, Choucair W, Southard J, Kokkinos P, Karasik P,
Greenberg MD, Singh SN. .Meta-analysis of antiarrhythmic therapy in
the prevention of postoperative atrial fibrillation and the effect on
hospital length of stay, costs, cerebrovascular accidents, and mortality
in patients undergoing cardiac surgery. Am J Cardiol 2003;91:1137–
1140.
22. Bruins P, te Velthuis H, Yazdanbakhsh AP, Jansen PG, van Hardevelt
FW, de Beaumont EM, Wildevuur CR, Eijsman L, Trouwborst A,
Hack CE. Activation of the complement system during and after
cardiopulmonary bypass surgery: postsurgery activation involves Creactive protein and is associated with postoperative arrhythmia. Circulation 1997;96:3542–3548.
23. Haissaguerre M, Jais P, Shah DC, Takahashi A, Hocini M, Quiniou G,
Garrigue S, Le Mouroux A, Le Metayer P, Clementy J. Spontaneous
initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998;339:659 – 666.
24. Watanabe H, Tanabe N, Makiyama Y, Chopra SS, Okura Y, Suzuki H,
Matsui K, Watanabe T, Kurashina Y, Aizawa Y. ST-segment abnormalities and premature complexes are predictors of new-onset atrial
fibrillation: the Niigata preventive medicine study. Am Heart J 2006;
152:731–735.
Levels of Circulating Procoagulant Microparticles in Nonvalvular
Atrial Fibrillation
Stéphane Ederhy, MDa, Emanuele Di Angelantonio, MD, MSca, Ziad Mallat, MD, PhDb,
Bénédicte Hugel, PhDc, Sandra Janower, MDa, Catherine Meuleman, MDa, Franck Boccara, MDa,
Jean-Marie Freyssinet, PhDc, Alain Tedgui, PhDb, and Ariel Cohen MD, PhDa,*
Circulating procoagulant microparticles (MPs) arising from cell activation or fragmentation
during apoptosis retain procoagulant properties and are increased in severe thrombotic states.
We investigated whether circulating procoagulant MP levels would be increased in nonvalvular
atrial fibrillation (AF). Using a hospital case-control study design, circulating procoagulant MP
levels were measured in 45 patients with permanent and/or persistent AF who were not
receiving anticoagulant therapy and 90 age-matched control subjects (45 with cardiovascular
risk factors and 45 without). Annexin V-positive MP levels (expressed as nanomoles per liter of
phosphatidylserine equivalent) were higher in patients with AF (median 9.3, interquartile range
6.8 to 17.3 nmol/L) than in control subjects with cardiovascular risk factors (median 4.9,
interquartile range 3.7 to 8.4 nmol/L) and control subjects without cardiovascular risk factors
(median 3.2, interquantile range 2.3 to 4.6 nmol/L; p <0.001). Platelet-derived MPs (captured
with antiglycoprotein Ib) and endothelial-derived MPs (captured with anti-CD31) were similar
in patients with AF and control subjects with cardiovascular risk factors but were significantly
higher than in control subjects without cardiovascular risk factors. On multiple regression
analysis, the presence of AF was a strong predictor of annexin V-positive MP level (p <.001).
In conclusion, circulating procoagulant MPs are increased in persistent and/or permanent AF
and might reflect a hypercoagulable state that could contribute to atrial thrombosis and
thromboembolism. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:989 –994)
Nonvalvular atrial fibrillation (AF) is the most common
sustained cardiac arrhythmia and is associated with an approximately fivefold increase in the risk of stroke.1 Besides
clinical2 and echocardiographic features3 that have been
shown to influence the risk of stroke in AF, a prothrombotic
state4 could predict thrombus formation in the left atrial
appendage5 and thromboembolism.6 Circulating procoagulant microparticles (MPs) are small membrane vesicles that
are shed mainly from cells in response to activation, injury,
and/or apoptosis.7,8 They mainly derive from platelets and
endothelial cells but also from leukocytes, lymphocytes, and
erythrocytes.7,8 They are also present in the circulation of
healthy subjects.9 Circulating procoagulant MPs, mainly of
platelet and endothelial origins, are rich in anionic phospholipids, particularly procoagulant phosphatidylserine, and
have been shown to bear or transfer tissue factor activity,
the initiator of coagulation in vivo. These properties suggest
an important procoagulant potential of circulating procoagulant MPs, which could contribute to initiation or perpetuation of thrombotic clinical conditions such as diabetes,10
acute coronary syndromes,11–13 and stroke.14 It has been
suggested that AF is associated with a prothrombotic state
a
Cardiology Department, Saint-Antoine University and Medical
School, Université Pierre et Marie Curie, Paris; bINSERM U689, Hôpital
Lariboisière, Paris; and cHematology and Immunology Institute, Universite
Louis Pasteur, Strasbourg; and INSERM U770, Hôpital de Bicêtre, Le
Kremlin Bicêtre, France.
*Corresponding author: Tel: 33-1-4928-2886; fax: 33-1-4928-2884.
E-mail address: [email protected] (A. Cohen).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.040
characterized by abnormalities of endothelial function and
platelet activation, with an increased risk of left atrial
thrombosis and thromboembolism. We sought to investigate, in a hospital-based case-control study, the hypothesis
that circulating procoagulant MP levels would be increased
in AF.
Methods
Using a hospital-based case-control study design, circulating procoagulant MP levels (annexin V-positive MPs and
platelet and endothelial circulating procoagulant MPs) were
compared in patients with AF and in 2 age-matched control
groups of patients in sinus rhythm with and without cardiovascular risk factors. The population initially screened included 57 patients with AF who had undergone transesophageal echocardiography as an evaluation before cardioversion in
patients with atrial arrhythmias15 before the initiation of any
oral anticoagulant treatment or heparin treatment. Lone atrial
AF was defined as AF occurring in the absence of structural
heart disease and hypertension.15,16 We excluded patients
with paroxysmal AF, as well as those in situations in which
circulating procoagulant MPs could be increased,10,13,14,17,18
such as acute ischemic stroke, stroke, acute pulmonary
embolism, or venous thromboembolism ⬍3 months earlier.
We also excluded patients with a history of infection or
inflammatory state (n ⫽ 4), acute coronary syndromes (n ⫽
2), or acute congestive heart failure (n ⫽ 4). In addition,
patients with hyperthyroidism (n ⫽ 2) and those receiving
antiinflammatory therapies were excluded, leaving a population that comprised 45 patients with permanent or persiswww.AJConline.org
990
The American Journal of Cardiology (www.AJConline.org)
Table 1
Baseline characteristics in patients with AF and control subjects with and without cardiovascular risk factors
p Value
Variable
Age (mean ⫾ SD) (yrs)
Men
Hypertension
Diabetes mellitus
Hypercholesterolemia
Current smoker
Coronary artery disease
History of heart failure
History of TIA or ischemic
stroke
Systolic blood pressure
(mean ⫾ SD) (mm Hg)
Diastolic blood pressure
(mean ⫾ SD) (mm Hg)
Concomitant treatment
Aspirin
␤ blockers
Calcium channel blockers
ACE inhibitors/ARBs
Diuretics
Statins
Fibrate
Digoxin
Amiodarone
Nitrates
Insulin
Oral antidiabetic therapy
Patients With
AF (n ⫽ 45)
Controls With
Risk Factors
(n ⫽ 45)
Controls Without
Risk Factors (n
⫽ 45)
AF vs
Controls With
Risk Factors
AF vs
Controls
Without
Risk Factors
Controls With
vs Without
Risk Factors
68.2 ⫾ 10.1
28 (62.2%)
27 (60.0%)
14 (31.1%)
17 (37.8%)
19 (42.2%)
6 (13.3%)
2 (4.4%)
2 (4.4%)
67.1 ⫾ 9.9
30 (66.7%)
25 (55.6%)
9 (20.0%)
14 (31.1%)
20 (44.4%)
9 (20.0%)
0
1 (2.2%)
67.0 ⫾ 9.5
26 (57.8% )
0
0
0
0
0
0
0
–
0.660
0.670
0.227
0.506
0.832
0.396
0.494*
1.000*
–
0.667
⬍0.001*
⬍0.001*
⬍0.001*
⬍0.001*
0.026*
0.494*
0.494*
–
0.384
⬍0.001*
0.001*
⬍0.001*
⬍0.001*
0.003*
–
1.000*
137.8 ⫾ 22.7
135.2 ⫾ 12.2
121.7 ⫾ 10.3
0.497
⬍0.001
⬍0.001
77.3 ⫾ 12.4
68.4 ⫾ 11.8
66.6 ⫾ 7.8
0.003
⬍0.001
0.182
13 (28.9%)
16 (35.6%)
9 (20.0%)
13 (28.9%)
3 (6.7%)
6 (13.3%)
2 (4.4%)
2 (4.4%)
2 (4.4%)
1 (2.2%)
2 (4.4%)
8 (17.8%)
6 (13.3%)
10 (22.2%)
4 (8.9%)
11 (14.4%)
0
6 (13.3%)
0
0
0
1 (2.2%)
2 (4.4%)
6 (13.3%)
–
–
–
–
–
–
–
–
–
–
–
–
0.071
0.163
0.134
0.634
0.242*
1.000
0.494*
0.494*
0.494
1.000*
1.000*
0.561
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
* Fisher’s exact test.
ACE ⫽ angiotensin-converting enzyme; ARB ⫽ angiotensin receptor blocker; TIA ⫽ transient ischemic attack.
tent AF. Risk factors for stroke in AF included age, gender,
hypertension, ischemic heart disease, congestive heart failure, history of stroke or transient ischemic attack, and diabetes mellitus. Because most of these risk factors for thromboembolism in AF are also cardiovascular risk factors and
could then constitute potential confounders, we compared
patients with AF versus 2 groups of age-matched control
subjects: those with cardiovascular risk factors and those
without cardiovascular risk factors. Control subjects with
cardiovascular risk factors consisted of patients with no
history of atrial arrhythmias who were undergoing routine
screening physical examinations at our outpatient cardiology clinic for cardiac symptoms, with an electrocardiogram
documenting sinus rhythm. Control subjects without cardiovascular risk factors included patients undergoing screening
examination before orthopedic surgery with no known cardiovascular risk factors, no history of atrial arrhythmias, and
no clinical evidence of disease or current cardiovascular
treatment, and who had an electrocardiogram documenting
sinus rhythm. These subjects were assessed by careful history examination and blood tests. Baseline demographic and
clinical data were available for control subjects and patients.
The study was approved by the institutional review board
and performed in accordance with institutional guidelines.
All patients gave written informed consent.
Enumeration of circulating procoagulant MPs was performed as previously described,19 with minor modifications.
Briefly, citrated blood was centrifuged at 1,500g for 15
minutes at room temperature and the supernatant was centrifuged again at 13,000g for 2 minutes to avoid platelet
contamination. Thrombin and factor Xa inhibitors—respectively, D-phenylalanyl-prolyl-arginyl chloromethyl ketone
and 1,5-dansyl-glutamyl-glycyl-arginyl chloromethyl ketone—were added to plasma samples at a final concentration of 50 ␮mol/L each and CaCl2 at a final concentration of
50 mmol/L. After capture of MPs on annexin V-coated
wells (for 30 minutes at 37°C), to take advantage of the
strong affinity of annexin V for aminophospholipids present
in MPs at the calcium concentration used, 4 washing steps
were performed with Tris buffer containing 1 mmol/L
CaCl2 and 0.05% Tween 20 for 5 minutes each at 20°C,
with the last step performed without Tween 20. The phosphatidylserine content of MPs, which is directly responsible
for their procoagulant activity, was then measured in a
prothrombinase assay. MPs were incubated with factor Xa
(50 pmol/L), factor Va (360 pmol/L), prothrombin (1.3
␮mol/L), and 2.3 mmol/L CaCl2 for 15 minutes at 37°C,
and linear absorbance changes were recorded at 405 nm
after addition of chromozym TH (380 ␮mol/L). After specific capture of platelet-derived MPs on antiglycoprotein Ib
Arrhythmias and Conduction Disturbances/Microparticles in Atrial Fibrillation
991
Table 2
Baseline correlates of annexin V-positive MPs and platelet and endothelial MP level in control subjects
Variable
Age
Men
Hypertension
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Diabetes mellitus
Hypercholesterolemia
Current smoker
Coronary artery disease
Annexin V-Positive MPs, Tertiles
ⱕ3.2
3.4–4.9
ⱖ5.0
64.2 (8.8%)
19 (61.3%)
6 (19.4%)
125.3 (14.7%)
67.4 (9.2%)
1 (3.2%)
2 (⬎6.5%)
3 (9.7%)
1 (3.2%)
68.6 (9.4%)
20 (69.0%)
7 (24.1%)
130.4 (11.0%)
67.2 (9.8%)
2 (6.9%)
7 (24.1%)
8 (27.6%)
5 (17.2%)
68.5 (10.3%)
17 (56.7%)
12 (40.0%)
129.9 (13.0%)
69.4 (11.2%)
6 (20.0%)
5 (16.7%)
9 (30.0%)
3 (10.0%)
p
Value*
Adjusted
p Value*
0.073
0.398
0.014
0.280
0.263
0.102
0.273
0.138
0.223
0.055
0.269
0.012
0.374
0.107
0.153
0.281
0.284
0.489
0.634
0.914
⬍0.001
0.021
0.130
0.026
0.003
0.017
0.008
0.622
0.864
⬍0.001
0.025
0.089
0.031
0.002
0.021
0.009
0.435
0.566
⬍0.001
0.006
0.079
0.028
0.004
0.019
0.021
0.479
0.636
⬍0.001
0.005
0.124
0.024
0.005
0.011
0.015
Platelet MPs, Tertiles
Age
Men
Hypertension
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Diabetes mellitus
Hypercholesterolemia
Current smoker
Coronary artery disease
ⱕ1.9
2.0-4.4
ⱖ4.5
64.9 (10.0%)
18 (60.0%)
3 (10.0%)
125.3 (14.5%)
67.1 (9.6%)
0
1 (3.3%)
2 (6.7%)
0
69.0 (9.3%)
21 (70.0%)
7 (23.3%)
128.1 (11.4%)
66.9 (10.1%)
4 (13.3%)
7 (23.3%)
8 (26.7%)
4 (13.3%)
67.4 (9.6%)
16 (55.2%)
14 (48.3%)
131.8 (12.9%)
69.9 (10.7%)
5 (17.2%)
6 (20.7%)
10 (34.5%)
5 (17.2%)
Endothelial MPs, Tertiles
Age
Men
Hypertension
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Diabetes mellitus
Hypercholesterolemia
Current smoker
Coronary artery disease
ⱕ0.99
0.1-0.19
ⱖ0.20
67.9 (8.5%)
16 (59.3%)
2 (7.4%)
122.5 (14.9%)
65.7 (9.0%)
0
0
2 (7.4%)
0
66.6 (9.9%)
20 (60.6%)
8 (24.2%)
129.5 (11.3%)
66.5 (9.8%)
3 (9.1%)
6 (18.2%)
9 (27.3%)
3 (9.1%)
66.8 (10.4%)
20 (62.2%)
15 (50.0%)
132.7 (11.6%)
71.7 (10.6%)
6 (20.0%)
8 (26.7%)
9 (30.0%)
6(20.0%)
* Derived from regression of log value (annexin V-positive MP, platelet MP, or endothelial MP) level on each baseline characteristic separately;
adjustments made for age and gender where noted.
antibody-coated wells, quantification was achieved after
several washing steps using a prothrombinase assay as described earlier. After specific capture of endothelium-derived MPs on anti-CD31 antibody-coated wells, quantification was achieved after several washing steps using a
prothrombinase assay as described earlier. MP levels were
expressed as nanomoles per liter of phosphatidylserine
equivalent. Intra- and interassay coefficients of variation
were approximately 10% and 20%, respectively, over a
rather wide concentration range, e.g., between 1 and approximately 10 times the normal concentration.20
Based on previous studies,10,12 we hypothesized that patients with AF would have MP levels increased by approximately 2 SDs compared with healthy control subjects and
by 1 SD compared with subjects without AF but with
cardiovascular risk factors. To achieve this with 90% power
and p ⬍0.05 among the 3 groups, 35 subjects per group
were required. To minimize the risk of a type II error and to
account for possible confounders, we recruited many more
than this number of patients with AF and control subjects
with and without cardiovascular risk factors. Categorical
variables, expressed as percentages, were compared using
chi-square test or Fisher’s exact test when appropriate. After
a test of normality, continuous data were expressed as the
mean and SD or median value with interquartile range
(IQR) as appropriate. Differences between patients and control subjects were evaluated using the 2-sample t test and
Mann-Whitney U test as appropriate. Correlations among
annexin V-positive MPs and endothelial and platelet MP
levels were evaluated using Spearman’s rank correlation
coefficients. Linear regression analyses were performed to
assess whether annexin V-positive MPs and endothelial and
platelet MP levels varied according to baseline characteristics among control subjects and according to AF status in
the overall population.
Because the values of annexin V-positive MPs and endothelial and platelet MPs were not normally distributed,
they were examined in the linear regression analysis after
logarithmic transformation. Multivariable linear regression
models were also used to describe the association between
992
The American Journal of Cardiology (www.AJConline.org)
Table 3
Effect of AF on annexin V-positive MPs and platelet and endothelial MP levels on multivariable linear regression analyses
Parameter
Expected
Change in Log
Annexin
V-positive MPs
p Value
Expected Change
in Log Platelet
MPs
p Value
Expected
Change in Log
Endothelial MPs
p Value
AF
Adjusted AF*
1.06 (0.81–1.31)
0.89 (0.63–1.15)
⬍0.001
⬍0.001
0.98 (0.58–1.39)
0.82 (0.38–1.25)
⬍0.001
⬍0.001
0.98 (0.49–1.47)
0.68 (0.16–1.21)
⬍0.001
0.011
* Adjusted for age, diabetes, hypertension, and hypercholesterolemia for log annexin V-positive MPs and for age, diabetes, hypertension, hypercholesterolemia, current smoking, and previous myocardial infarction for expected change in log platelet and endothelial MPs.
each marker (annexin V-positive MPs and endothelial and
platelet MPs) and the cardiovascular risk factors and to
identify their independent predictors after adjustment for
possible confounders. Variables were defined as potential
confounders if they were associated on univariate analysis
with each marker or if they changed the unadjusted regression coefficient of AF by ⬎5% on bivariate analysis.21
Furthermore, variables that were strongly suspected to be a
priori potential confounders were also included in the analysis (e.g., age, diabetes, and hypertension).10,23,24 The relation between circulating procoagulant MP levels and potential predictors was also presented using the estimated
regression coefficient, expressed as the percentage increase
in MP levels for the presence of each risk factor or for a unit
increase in the continuous variables such as age. All analyses were performed using STATA 9 statistical software
(STATA, College Station, Texas). A value of ⬍0.05 was
considered statistically significant.
Results
Circulating procoagulant MP levels were assayed in 45
patients with AF and in 90 control subjects: 45 with cardiovascular risk factors and 45 without. Demographic characteristics and cardiovascular risk factors of control subjects
and patients with AF are listed in Table 1. There were no
significant differences in terms of clinical cardiovascular
risk factors between patients with AF and control subjects
with cardiovascular risk factors. By design, control subjects
without cardiovascular risk factors were significantly different with regard to clinical cardiovascular risk factors from
patients with AF and from control subjects with cardiovascular risk factors. However, there was no significant difference regarding age and gender between controls with and
without cardiovascular risk factors, even if there was a
slight predominance of men in the former group. Among
baseline patient characteristics, only diastolic blood pressure was significantly different between patients with AF
and control subjects with cardiovascular risk factors.
Relations between baseline characteristics and circulating procoagulant MPs (annexin V-positive MPs and platelet
and endothelial MPs) were investigated among control subjects after adjustment for age and gender (Table 2). Annexin
V-positive MPs were positively associated with only age
and hypertension, whereas there was positive significant
association of platelet and endothelial MPs with hypertension, systolic blood pressure, diabetes, hypercholesterolemia, smoking status, and history of coronary artery disease.
A good correlation was found between annexin V-positive
MP levels and platelet MPs (␳ ⫽ 0.642, p ⬍0.001) or
endothelial MP levels (␳ ⫽ 0.415, p ⬍0.001) and between
platelet and endothelial MP levels (␳ ⫽ 0.565, p ⬍0.001).
Annexin V-positive MP levels were significantly higher
in patients with AF (median 9.3, IQR 6.8 to 17.3 nmol/L)
compared with control subjects with cardiovascular risk
factors (median 4.9, IQR 3.7 to 8.4 nmol/L of phosphatidylserine equivalent; p ⬍0.001) and controls without cardiovascular risk factors (median 3.2, IQR 2.3 to 4.6 nmol/L
of phosphatidylserine equivalent; p ⬍0.001). Furthermore,
annexin V-positive MP levels were significantly higher in
control subjects with cardiovascular risk factors than in
control subjects without cardiovascular risk factors (p
⬍0.001) (Figure 1).
Using regression analysis, after adjusting for potential
confounders, patients with AF had annexin V-positive MP
levels 2.4 times as high as subjects without AF (Table 3).
Platelet MP levels were not significantly different between patients with AF (median 7.0, IQR 3.0 to 17.5 nmol/L
of phosphatidylserine equivalent) and control subjects with
cardiovascular risk factors (median 5.5, IQR 2.6 to 11.2
nmol/L; p ⫽ 0.202). However platelet MP levels were
significantly higher in patients with AF compared with
control subjects without cardiovascular risk factors (median
1.9, IQR 1.4 to 2.4 nmol/L of phosphatidylserine equivalent; p ⬍0.001) and in control subjects with cardiovascular
risk factors compared with control subjects without cardiovascular risk factors (p ⬍0.001). Multivariable regression
analysis showed that patients with AF had platelet MP
levels 2.3 times as high as patients without AF after adjustment for potential confounders (Table 3).
Endothelial MP levels were not significantly different
between patients with AF (median 0.2, IQR 0.1 to 0.4
nmol/L of phosphatidylserine equivalent) and control subjects with cardiovascular risk factors (median 0.2, IQR 0.1
to 0.2 nmol/L of phosphatidylserine equivalent; p ⫽ 0.195).
Endothelial MP levels were significantly higher in patients
with AF compared with control subjects without cardiovascular risk factors (median 0.1, IQR 0.01 to 0.1; p ⬍0.001)
and in control subjects with cardiovascular risk factors compared with control subjects without cardiovascular risk factors (p ⬍0.001). However, after adjustment for potential
confounders, patients with AF had endothelial MP levels
twice as high as patients without AF (Table 3). No significant differences in annexin V-positive MPs (p ⫽ 0.704) or
platelet (p ⫽ 0.815) and endothelial MPs (p ⫽ 0.955) were
observed between patients with AF who were taking aspirin
and those with AF who were not taking aspirin.
Arrhythmias and Conduction Disturbances/Microparticles in Atrial Fibrillation
993
Figure 1. Annexin V-positive MP and platelet and endothelial MP levels (expressed as nanomoles per liter of phosphatidylserine equivalent) in the 2 control
groups and in patients with nonvalvular AF are presented as scatterplots and box plots. For box plots, the middle line indicates the median; bottom of box
indicates 25th percentile; and top of box indicates 75th percentile.
Discussion
The present study reports that AF was associated with
increased levels of annexin V-positive MPs. Compared with
control subjects with or without cardiovascular risk factors,
patients with AF exhibited, respectively, approximately a
twofold and threefold greater amount of annexin V-positive
MPs. These results suggest an increased prothrombotic state
supported by annexin V-positive MPs.
It has been suggested that the pathophysiology of thromboembolism in AF is multifactorial and fulfills the Virchow
triad.22 In fact, abnormal blood flow and vessel wall and
hemostatic abnormalities have been described in AF. Moreover, it has been postulated that AF is associated with a
prothrombotic or hypercoagulable biologic state.22,25 However, the role of confounding factors such as vascular risk
factors has to be taken into account. In a case-control study
from the Framingham cohort26 that compared matched subjects with and without AF, hemostatic factors such as fibrinogen, von Willebrand factor antigen, tissue plasminogen activator antigen, and plasminogen activator inhibitor–1
antigen were found to be no longer associated with AF in
multivariable analysis adjusted for cardiovascular risk factors. In addition, advancing age, previous cerebral ischemia,
recent heart failure, and diabetes were found to be independently associated with increased plasma von Willebrand
factor in a Stroke Prevention in Atrial Fibrillation substudy,25 highlighting the need for adjustment for vascular
risk factors and cardiovascular conditions.
The design of our study provides a unique opportunity to
analyze the relative contribution of AF and cardiovascular
risk factors in patients with AF. Our study confirms the
evidence of increased circulating procoagulant activity,
with more annexin V-positive MPs and endothelial and
platelet MPs in patients with AF and control subjects with
cardiovascular risk factors compared with healthy control
subjects. On multivariable analysis, AF was a strong predictor of higher levels of annexin V-positive MPs and endothelial and platelet MPs independent of cardiovascular
risk factors. Endothelial MPs have been suspected to be a
marker of endothelial damage or dysfunction,7 activation,27
and/or cell injury,28 whereas platelet MPs have been reported to be increased in severe prothrombotic states such as
acute myocardial infarction10,12,13 and stroke.14 Tissue factor, an essential component of the coagulation pathway that
triggers the coagulation cascade, has been shown to be
increased in chronic AF29 and overexpressed in the endothelium of left atrial appendages obtained from patients with
AF.30 It has also been demonstrated that a circulating storage pool of tissue factor was associated with MPs and
constituted the main reservoir of blood-borne tissue factor
activity.7 Thus, circulating procoagulant MPs could play a
role in the hypercoagulable state observed in AF.
Our study could support this hypothesis in several aspects. First, we observed a stepwise increase in annexin
V-positive MPs from control subjects in sinus rhythm without cardiovascular risk factors to control subjects in sinus
rhythm with cardiovascular risk factors and finally to patients with AF. Second, multivariable analysis isolated AF
as a strong predictor of increased annexin V-positive MPs
beyond cardiovascular risk factors, suggesting an influence
of the arrhythmia by itself on circulating procoagulant MP
levels. Third, in the blood flow, circulating procoagulant
MPs provide an additional procoagulant phospholipid surface for the assembly of the clotting enzyme complexes
leading to thrombin generation. Phosphatidylserine normally sequestered in the inner leaflet of the plasma mem-
994
The American Journal of Cardiology (www.AJConline.org)
branes of cells at rest becomes accessible during remodeling
and shedding processes that occur after stimulation or apoptosis. After it is translocated, phosphatidylserine can enhance tissue factor procoagulant activity.
In conclusion, circulating procoagulant MPs are increased in persistent and/or permanent AF and might reflect
a hypercoagulable state that could contribute to atrial thrombosis and thromboembolism.
1. Wolf PA, Dawber TR, Thomas HE Jr, Kannel WB. Epidemiologic
assessment of chronic atrial fibrillation and risk of stroke: the Framingham study. Neurology 1978;28:973–977.
2. Predictors of thromboembolism in atrial fibrillation: I. Clinical features
of patients at risk. The Stroke Prevention in Atrial Fibrillation Investigators. Ann Intern Med 1992;116:1–5.
3. Predictors of thromboembolism in atrial fibrillation: II. Echocardiographic features of patients at risk. The Stroke Prevention in Atrial
Fibrillation Investigators. Ann Intern Med 1992;116:6 –12.
4. Li-Saw-Hee FL, Blann AD, Lip GY. A cross-sectional and diurnal
study of thrombogenesis among patients with chronic atrial fibrillation.
J Am Coll Cardiol 2000;35:1926 –1931.
5. Pongratz G, Brandt-Pohlmann M, Henneke KH, Pohle C, Zink D,
Gehling G, Bachmann K. Platelet activation in embolic and preembolic status of patients with nonrheumatic atrial fibrillation. Chest
1997;111:929 –933.
6. Transesophageal echocardiographic correlates of thromboembolism in
high-risk patients with nonvalvular atrial fibrillation. The Stroke Prevention in Atrial Fibrillation Investigators Committee on Echocardiography. Ann Intern Med 1998;128:639 – 647.
7. Morel O, Toti F, Hugel B, Bakouboula B, Camoin-Jau L, DignatGeorge F, Freyssinet JM. Procoagulant microparticles: disrupting the
vascular homeostasis equation? Arterioscler Thromb Vasc Biol 2006;
26: 2594 –2604.
8. Diamant M, Tushuizen ME, Sturk A, Nieuwland R. Cellular microparticles: new players in the field of vascular disease? Eur J Clin Invest
2004;34:392– 401.
9. Berckmans RJ, Neiuwland R, Boing AN, Romijn FP, Hack CE, Sturk
A. Cell-derived microparticles circulate in healthy humans and support
low grade thrombin generation. Thromb Haemost 2001;85:639 – 646.
10. Morel O, Hugel B, Jesel L, Lanza F, Douchet MP, Zupan M, Chauvin
M, Cazenave JP, Freyssinet JM, Toti F. Sustained elevated amounts of
circulating procoagulant membrane microparticles and soluble GPV
after acute myocardial infarction in diabetes mellitus. Thromb Haemost 2004;91:345–353.
11. Mallat Z, Hugel B, Ohan J, Lesèche G, Freyssinet JM, Tedgui A. Shed
membrane microparticles with procoagulant potential in human atherosclerotic plaques: a role for apoptosis in plaque thrombogenicity.
Circulation 1999;99:348 –353.
12. Mallat Z, Benamer H, Hugel B, Benessiano J, Steg PG, Freyssinet JM,
Tedgui A. Elevated levels of shed membrane microparticles with
procoagulant potential in the peripheral circulating blood of patients
with acute coronary syndromes. Circulation 2000;101:841– 43.
13. Boulanger CM, Scoazec A, Ebrahimian T, Henry P, Mathieu E,
Tedgui A, Mallat Z. Circulating microparticles from patients with
myocardial infarction cause endothelial dysfunction. Circulation 2001;
104:2649 –2652.
14. Lee YJ, Jy W, Horstman LL, Janania J, Reyes Y, Kelley RE, Ahn YS.
Elevated platelet microparticles in transient ischemic attacks, lacunar
infarcts, and multiinfarct dementias. Thromb Res 1993;72:295–304.
15. Di Angelantonio E, Ederhy S, Benyounes N, Janower S, Boccara F,
Cohen A comparison of transesophageal echocardiographic identification of embolic risk markers in patients with lone versus non-lone
atrial fibrillation. Am J Cardiol 2005;95:592–596.
16. Scardi S, Mazzone C, Pandullo C, Goldstein D, Poletti A, Humar F.
Lone atrial fibrillation: prognostic differences between paroxysmal and
chronic forms after 10 years of follow-up. Am Heart J 1999;137:686 –
691.
17. Chirinos JA, Heresi GA, Velasquez H, Jy W, Jimenez JJ, Ahn E,
Horstman LL, Soriano AO, Zambrano JP, Ahn YS.. Elevation of
endothelial microparticles, platelets, and leukocyte activation in patients with venous thromboembolism. J Am Coll Cardiol 2005;45:
1467–1471.
18. Ogura H, Tanaka H, Koh T, Fujita K, Fujimi S, Nakamori Y, Hosotsubo H, Kuwagata Y, Shimazu T, Sugimoto H. Enhanced production
of endothelial microparticles with increased binding to leukocytes in
patients with severe systemic inflammatory response syndrome.
J Trauma 2004;56:823– 830.
19. Aupeix K, Hugel B, Martin T, Bischoff P, Lill H, Pasquali JL, Freyssinet JM. The significance of shed membrane particles during programmed cell death in vitro, and in vivo, in HIV-1 infection. J Clin
Invest 1997;99:1546 –1554.
20. Weinstabl H, Vormittag R, Sailer T, Koder S, Freyssinet JM, Pabinger
I. Adaption of a kinetic assay for the determination of blood circulating
microparticles. Hämostaseologie 2006;26:288.
21. Mickey R, Greenland S. The impact of confounder selection criteria on
effect estimation. Am J Epidemiol 1989;129:125–137.
22. Choudhury A, Lip GY. Atrial fibrillation and the hypercoagulable
state: from basic science to clinical practice. Pathophysiol Haemost
Thromb 2004;33:282–289
23. Sabatier F, Darmon P, Hugel B, Combes V, Sanmarco M, Velut JG,
Arnoux D, Charpiot P, Freyssinet JM, Oliver C, Sampol J, DignatGeorge F. Type 1 and type 2 diabetic patients display different patterns
of cellular microparticles. Diabetes 2002;51:2840 –2845.
24. Preston RA, Jy W, Jimenez JJ, Mauro LM, Horstman LL, Valle M,
Aime G, Ahn YS. Effects of severe hypertension on endothelial and
platelet microparticles. Hypertension 2003;41:211–217.
25. Conway DS, Pearce LA, Chin BS, Hart RG, Lip GY. Plasma von
Willebrand factor and soluble p-selectin as indices of endothelial
damage and platelet activation in 1321 patients with nonvalvular atrial
fibrillation: relationship to stroke risk factors. Circulation 2002;106:
1962–1967.
26. Feng D, D’Agostino RB, Silbershatz H, Lipinska I, Massaro J, Levy D,
Benjamin EJ, Wolf PA, Tofler GH. Hemostatic state and atrial fibrillation (the Framingham Offspring Study). Am J Cardiol 2001;87:168 –
171.
27. Jimenez JJ, Jy W, Mauro LM, Soderland C, Horstman LL, Ahn YS.
Endothelial cells release phenotypically and quantitatively distinct
microparticles in activation and apoptosis. Thromb Res 2003;109:175–
180.
28. Combes V, Simon AC, Grau GE, Arnoux D, Camoin L, Sabatier F,
Mutin M, Sanmarco M, Sampol J, Dignat-George F. In vitro generation of endothelial microparticles and possible prothrombotic activity
in patients with lupus anticoagulant. J Clin Invest 1999;104:93–102.
29. Conway DS, Buggins P, Hughes E, Lip GY. Relationship of interleukin-6 and C-reactive protein to the prothrombotic state in chronic atrial
fibrillation. J Am Coll Cardiol 2004;43:2075–2082.
30. Nakamura Y, Fukushima-Kusano K, Ohta K, Matsubara H, Hamuro T,
Yutani C, Ohe T. Tissue factor expression in atrial endothelia associated with nonvalvular atrial fibrillation: possible involvement in intracardiac thrombogenesis. Thromb Res 2003;111:137–142.
Prevalence of Interatrial Block in Young Healthy Men <35 Years of Age
Elias Gialafos, MDa, Theodora Psaltopoulou, MDb, Theodore G. Papaioannou, PhDa,*,
Andreas Synetos, MDa, Polychronis Dilaveris, MDa, George Andrikopoulos, MDa,
Konstantinos Vlasis, MDa, John Gialafos, MDa, and Christodoulos Stefanadis, MDa
Interatrial block (IAB; P-wave duration >110 ms) is highly prevalent and is strongly
associated with atrial tachyarrhythmias and left atrial dysfunction. Very few studies
have examined IAB in young healthy subjects. The aim of the present study was to
demonstrate the prevalence of IAB and its possible relation with clinical variables in
1,353 young healthy men. It was found that 9.1% of healthy men aged <35 years and
5.4% of those aged <20 years had P-wave durations >110 ms. The frequent presence
of IAB in leads II, V3, and V5 was also observed. Age and heart rate were independent
significant determinants of IAB. In conclusion, IAB is a frequent phenomenon, even at
young ages. Thus, the early recognition of IAB might be important, possibly contributing to the prevention of future cardiovascular complications. © 2007 Elsevier Inc.
All rights reserved. (Am J Cardiol 2007;100:995–997)
Interatrial block (IAB; P-wave duration [Pdur] ⱖ110 ms) is
associated with the pathogenesis of atrial tachyarrhythmias,
especially atrial fibrillation,1,2 left atrial electromechanical
dysfunction,3,4 and possibly embolism.5 A 40% prevalence
of IAB in a general hospital population has been reported.6,7
Although studies of IAB have been published, very few
studies examined IAB in young healthy subjects. The aim of
the present study was to demonstrate the prevalence of IAB
and its possible relation with clinical variables in a population of young healthy men.
Methods and Results
One thousand three hundred fifty-three servicemen at the Hellenic Airforce Base of Tripoli in Greece (mean age 23.7 ⫾ 3.4
years) were enrolled in this study. Subjects with known structural heart disease, histories of arrhythmias, conduction abnormalities, or bundle branch block (wide QRS interval)
were excluded. Blood pressure (systolic, diastolic, and
mean) was measured with a cuff sphygmomanometer, and
each participant in the study completed a questionnaire
concerning smoking and coffee habits and feelings of palpitation.
A digital 12-lead electrocardiogram was recorded at rest
in the supine position using a computer-based electrocardiographic system (Cardioperfect version 1.1; Cardio Control NV, Delft, The Netherlands). During the recordings, the
subjects were breathing freely and were not allowed to
speak. The onset and offset of the P wave were defined as
the junction between the P-wave pattern and the isoelectric line and marked with the cursor. Two independent
investigators measured the P waves of all electrocardiograms, without knowledge of subject assignment. The
a
First Department of Cardiology, Hippokration Hospital, and bDepartment of Hygiene and Epidemiology, Medical School, National and Kapodistrian University of Athens, Athens, Greece. Manuscript received February 19, 2007; revised manuscript received and accepted April 13, 2007.
*Corresponding author: Tel: 302108254765; fax: 302108254791.
E-mail address: [email protected] (T.G. Papaioannou).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.041
averages of the measurements of the 2 observers were
used for comparisons. Patients with measurable P waves
in ⱕ9 electrocardiographic leads were excluded. The
study was approved by the institutional committee on
human research, and all subjects gave informed consent
before entering the study.
Intra- and interobserver relative errors (the absolute difference between 2 observations divided by the mean and
expressed as a percentage) for Pdur were determined in 100
randomly selected study participants. To define intraobserver errors of measurement, 1 of the 2 investigators
measured the durations of the P waves of all 100 electrocardiograms twice. The intraobserver relative error for
Pdur was 7.7 ⫾ 7.5%, and the interobserver relative error
was 9.2 ⫾ 5.7%.
Continuous variables are presented as mean ⫾ SD, and
categorical variables are expressed as absolute numbers and
percentages. Dichotomous variables (i.e., coffee consumption, smoking) were compared among IAB and age subgroups using a 2-sided chi-square test. Two-tailed independent-sample Student’s t tests were performed to assess
differences between subjects with and without IAB. Multivariate logistic regression analysis was performed to estimate the independent relation of heart rate and age with IAB
prevalence after adjustment for blood pressure levels, coffee
consumption, and smoking. A p value ⬍0.05 was considered statistically significant. Statistical analysis was performed using SPSS version 13 for Windows (SPSS, Inc.,
Chicago, Illinois).
Of the 1,353 subjects, 49.3% were regular smokers,
66.9% systematically consumed coffee, and 36% reported
feeling of palpitations. The mean systolic and diastolic
blood pressures were 121.4 ⫾ 8.9 and 78.3 ⫾ 7.4 mm Hg,
respectively, and the mean heart rate was 76.1 ⫾ 13.5
beats/min. In the total population, the prevalence of IAB
was 9.1% when defined as Pdur ⱖ110 ms, in accordance
with a previous definition of IAB.4,8 –10 The prevalence of
IAB was 1.6% when defined as Pdur ⱖ120 ms, as recently
suggested.9,10
Subjects with IAB (n ⫽ 123) were significantly older
www.AJConline.org
996
The American Journal of Cardiology (www.AJConline.org)
Table 1
Independent determinants of interatrial block prevalence in a healthy
adult population (multiple logistic regression analysis)
Variable
Heart rate (beats/min)
Age (yrs)
Smoking
Coffee consumption
Mean blood pressure (mm Hg)
Odds Ratio
(95% Confidence Interval)
p Value
1.027 (1.013–1.040)
1.055 (1.000–1.113)
0.771 (0.520–1.144)
0.953 (0.632–1.439)
1.000 (0.973–1.027)
⬍0.001
0.049
0.196
0.820
0.972
Figure 1. Prevalence of IAB in different age groups.
than subjects without IAB (24.3 ⫾ 3.2 vs 23.6 ⫾ 3.4 years,
p ⫽ 0.03). Also, subjects with IAB had significantly higher
heart rates than those without IAB (80.7 ⫾ 15.1 vs 75.7 ⫾
13.3 beats/min, p ⬍0.001), and a greater percentage of
subjects with IAB felt palpitations (46.7% vs 34.9%, p ⫽
0.01). Coffee consumption (p ⫽ 0.80) and smoking (p ⫽
0.38) were not significantly different between subjects with
and without IAB. Multiple logistic regression analysis indicated that heart rate and age (as continuous variables)
were independent significant determinants of IAB (Table 1).
The study population was divided into 3 age subgroups:
⬍20 years (n ⫽ 312), 20 to 25 years (n ⫽ 658), and ⬎25
years (n ⫽ 383). The prevalence of IAB was significantly
different among the age groups, as shown in Figure 1. More
specifically, IAB prevalence was significantly higher in
those aged 20 to 25 years (10%) and ⬎25 years (10.4%)
than in those aged ⬍20 years (5.4%).
Discussion
The present study demonstrated that 9.1% of healthy young
men aged ⬍35 years had Pdur ⱖ110 ms, and 1.6% of the
studied population had Pdur ⱖ120 ms, and a significantly
greater percentage of subjects with IAB than those without
IAB reported feeling palpitations. Another interesting finding was the frequent presence of IAB in leads II, V3, and V5.
Age and heart rate were independent determinants of IAB.
In addition, a higher maximum Pdur in participants with IAB
was associated with a larger number of abnormal leads in
the same participants. For example, if the maximum Pdur
was ⱖ110 to 115 ms, a mean of 1.16 abnormal leads were
found, whereas if the maximum Pdur was ⱖ130 to 145 ms,
the mean number of abnormal leads was 5 (p for trend
⬍0.001; data not shown).
IAB, defined as Pdur ⱖ110 ms in any lead, is a rather new
electrocardiographic index that is used as a possible surrogate
of atrial tachyarrhythmias and left atrial electromechanical
dysfunction LA thrombosis and systemic embolism.1,11–13
However, IAB was recently redefined as Pdur ⱖ120 ms in any
lead for optimal diagnosis.3,9,10 Its high prevalence was
demonstrated in 2 well-separated general hospital populations.6,7 In these studies, a 47% prevalence of IAB was
reported in patients who were in sinus rhythm among patients of all ages, and 59% in patients aged ⬎60 years,
which is higher than that described in the classic textbooks
and studies. This discrepancy among textbooks and studies
might be due to different methods for the estimation of Pdur,
because the textbooks restrict the measurement to a single
lead, or due to a failure to diagnose IAB, even in a tertiary
care teaching hospital, with hazardous consequences.2,14
In contrast to textbooks, recent studies have used ⬎9 leads
to estimate maximum Pdur because IAB can appear in any
lead.9,10,12,13,15 In our study, we found more frequent IAB in
leads II, V3, and V5. This finding agrees with that of Ariyarajah
et al,15 who suggested that IAB can be found in any lead but is
more frequent in leads II and V3 to V6. The more focused
examination in these leads might alert clinicians to the need for
increased awareness, which is the key to the timely detection
and recognition of IAB, as well as anticipation and even
prevention of sequelae. However, toward this direction is also
the redefinition of IAB, which might make its manual detection
easier, whereas software able to automatically generate and
calculate Pdur in the 12 leads of the electrocardiogram would be
even more useful.
Goyal and Spodick3 demonstrated that patients with IAB
had sluggish, poorly contractile left atria and that the extent
of left atrial dysfunction was related to the degree of electric
delay from IAB. Indeed, in most published research, IAB is
interpreted as left atrial enlargement, but delay in interatrial
conduction can in fact occur independent of increase in
atrial size.3,4
In conclusion, the early recognition of IAB by measuring
Pdur in all leads rather than only 1 lead might be of importance, because IAB precedes left atrial dilatation7 and might
alert clinicians to anticipate and possibly prevent complications such as stroke.5,9,13
1. Ariyarajah V̇, Asad N, Tandar A, Spodick DH. Interatrial block:
pandemic prevalence, significance, and diagnosis. Chest 2005;128:
970 –975.
2. Agarwal YK, Aronow WS, Levy JA, Spodick DH. Association of
interatrial block with development of atrial fibrillation. Am J Cardiol
2003;91:882.
3. Goyal SB, Spodick DH. Electromechanical dysfunction of the left atrium
associated with interatrial block. Am Heart J 2001;142:823– 827.
4. Bayes de Luna A. Electrocardiographic alterations due to atrial pathology. In: Bayes de Luna A, ed. Clinical Electrocardiography: A
Textbook. New York: Futura, 1998:169 –171.
5. Lorbar M, Levrault R, Phadke JG, Spodick DH. Interatrial block as a
predictor of embolic stroke. Am J Cardiol 2005;95:667– 668.
6. Jairath UC, Spodick DH. Exceptional prevalence of interatrial block in
a general hospital population. Clin Cardiol 2001;24:548 –550.
7. Asad N, Spodick DH. Prevalence of interatrial block in a general
hospital population. Am J Cardiol 2003;91:609 – 610.
Conduction Arrhythmias/Interatrial Block in Young Healthy Men
8. Willems JL, Robles de Medina EO, Bernard R, Coumel P, Fisch C,
Krikler D, Mazur NA, Meijler FL, Mogensen L, Moret P. Criteria for
intraventricular conduction disturbances and pre-excitation. World
Health Organizational/International Society and Federation for Cardiology Task Force Ad Hoc. J Am Coll Cardiol 1985;5:1261–1275.
9. Ariyarajah V, Frisella M, Spodick D. Reevaluation of the criterion for
interatrial Block. Am J Cardiol 2006;98:936 –937.
10. Ariyarajah V, Apiyasawat S, Spodick DH. Optimal P-wave duration
for bedside diagnosis of interatrial block. Ann Noninvas Electrocardiol
2006;11:259 –262.
11. Ariyarajah V, Mercado K, Apiyasawat S, Puri P, Spodick DH. Correlation of left atrial size with P-wave duration in interatrial block.
Chest 2005;128:2615–2618.
997
12. Ariyarajah V, Apiyasawat S, Najjar H, Mercado K, Puri P, Spodick
DH. Frequency of interatrial block in patients with sinus rhythm
hospitalized for stroke and comparison to those without interatrial
block. Am J Cardiol 2007;99:49 –52.
13. Ariyarajah V, Puri P, Apiyasawat S, Spodick DH. Interatrial block: a
novel risk factor for embolic stroke? Ann Noninvas Electrocardiol
2007;12:15–20.
14. Ariyarajah V, Puri P, Spodick DH. Clinical underappreciation of
interatrial block in a general hospital population. Cardiology 2005;
104:193–195.
15. Ariyarajah V, Apiyasawat S, Puri P, Spodick DH. Specific electrocardiographic markers of P-wave morphology in interatrial block. J Electrocardiol 2006;39:380 –384.
Reliability of Echocardiography for Hemodynamic Assessment of
End-Stage Heart Failure
Nicolas Mansencal, MDa,*, Laure Revault d’Allonnes, MDa, Alain Beauchet, MDb,
Séverine Fabre, MDa, Franck Digne, MDa, Jean-Christian Farcot, MDa, Thierry Joseph, MD, PhDa,
and Olivier Dubourg, MDa
The management of patients with end-stage heart failure is difficult and may require the
monitoring of intracardiac pressures. The aim of this prospective study was to assess the
reliability of echocardiography in patients with end-stage HF. Twenty consecutive patients
presenting with severe left ventricular dysfunction in end-stage heart failure were prospectively studied. All patients underwent right-sided cardiac catheterization and transthoracic
echocardiography. Right atrial pressure, estimated using a new echocardiographic parameter, was significantly improved. There was good agreement between systemic and pulmonary vascular resistance, determined by catheterization and echocardiography. All patients
with echocardiographic pulmonary vascular resistance <6 Wood units also had invasive
pulmonary vascular resistance <6 Wood units. Only echocardiographic mean right atrial
pressure was related to the use of saline infusion or bolus infusion of furosemide. All
patients requiring intravenous furosemide had pulmonary capillary wedge pressures persistently >15 mm Hg despite adequate medication. In conclusion, this study indicates that
echocardiography may be a reliable tool for the management of patients with end-stage
heart failure. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:998 –1001)
Heart failure (HF) is a major public health problem in the
industrialized world. Every year, ⬎550,000 new cases occur in the United States, where it ranks first among the
causes of cardiovascular mortality, accounting for 53,000
deaths/year.1 HF and left ventricular (LV) systolic dysfunction are associated with a poor quality of life and annual
mortality of about 60% for severe HF.2 The management of
patients with end-stage HF is difficult and may require the
monitoring of intracardiac pressures, usually using rightsided cardiac catheterization. The noninvasive echocardiographic assessment of right- and left-sided cardiac pressures
is possible and usually performed in most patients with HF
but has not been specifically validated in patients with
end-stage HF.3,4 The aim of this prospective study was to
assess the reliability of echocardiography in this specific
population of patients with end-stage HF.
Methods and Results
We prospectively studied 20 consecutive patients presenting
with severe LV dysfunction in end-stage HF. Entry criteria
included age ⬎18 years, severe systolic LV dysfunction
(LV ejection fraction ⬍30%), end-stage HF, and an indication for right-sided cardiac catheterization (performed with
informed consent).
All patients underwent right-sided cardiac catheterization and transthoracic echocardiography in the same hour.
Departments of aCardiology and bBiostatistics, University Hospital
Ambroise Paré, Assistance Publique-Hôpitaux de Paris, UFR de Médecine
Paris-Ile de France-Ouest, Faculté de Versailles-Saint Quentin en Yvelines,
Boulogne, France. Manuscript received February 25, 2007; revised manuscript received and accepted April 24, 2007.
*Corresponding author: Tel: 33-0-149095620; fax: 33-0-149095344.
E-mail address: [email protected] (N. Mansencal).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.042
Treatment was left to the discretion of the physician according to the results of invasive catheterization (with a bolus
infusion of intravenous furosemide or saline infusion, if
necessary). Patients requiring bolus infusions of intravenous
furosemide or saline infusion (n ⫽ 13) underwent continuous invasive monitoring for ⱖ24 hours, and repeat echocardiography was performed 24 hours after the initiation of
treatment using the same echocardiographic protocol.
Complete right-sided cardiac catheterization was performed to obtain the following parameters: mean right atrial
pressure, systolic pulmonary arterial pressure, mean pulmonary arterial pressure, cardiac output using the thermodilution method, pulmonary capillary wedge pressure (PCWP),
and systemic and pulmonary vascular resistance. All measurements were repeated 3 to 5 times (within a 15% range)
and were interpreted by observers blinded to the echocardiographic results. Right-sided cardiac catheterization was
considered the gold standard.
All transthoracic echocardiographic studies were performed
with a Toshiba Powervision 8000 system (Toshiba Medical
Systems Corporation, Otawara-Shi, Japan) equipped with a
multifrequency transducer by a single investigator (NM)
blinded to the hemodynamic data. The LV ejection fraction
was calculated using Simpson’s rule. The following echocardiographic parameters were systematically assessed: (1)
mean right atrial pressure using 2-dimensional percentage
collapse of the inferior vena cava; (2) systolic pulmonary
arterial pressure using tricuspid regurgitation5; (3) mean
pulmonary arterial pressure using pulmonary regurgitation6;
(4) aortic output7; (5) PCWP using the ratio of mitral inflow
to annulus (E/E=) (PCWP ⫽ 1.24 [E/E=] ⫹ 1.9)8,9; and (6)
PCWP using the ratio of mitral inflow to flow propagation
velocity (E/Vp), as previously reported by Garcia et al10
(PCWP ⫽ 5.8 [E/Vp] ⫹ 4.5). A cut-off value ⱖ15 for E/E=
www.AJConline.org
Heart Failure/Echocardiography and Heart Failure
999
Table 1
Invasive and echocardiographic results in 20 patients with end-stage heart failure
Linear Regression
Variable
Mean right atrial pressure (mm Hg)
Invasive vs conventional echocardiographic
parameter
Invasive vs new echocardiograhpic
parameter
Systolic pulmonary arterial pressure (mm Hg)
Mean pulmonary arterial pressure (mm Hg)
PCWP (mm Hg)
Invasive versus E/E= ratio
Invasive versus E/Vp ratio
Cardiac output (L/min)
Systemic vascular resistance (Wood units)
Pulmonary vascular resistance (Wood units)
Catheterization
Echocardiography
13.9 ⫾ 6.5
13.3 ⫾ 3.8
13.9 ⫾ 6.5
r Value
p Value
95% Confidence Interval
0.96
0.61
0.008
⫺8.93 to ⫹7.72
12.7 ⫾ 5.2
0.13
0.91
⬍0.0001
⫺2.88 to ⫹5.68
46.4 ⫾ 12.8
32.5 ⫾ 9.9
46.7 ⫾ 10.9
31 ⫾ 7.7
0.74
0.27
0.94
0.93
⬍0.0001
⬍0.0001
⫺10.67 to ⫹11.56
⫺5.66 to ⫹7.44
21.5 ⫾ 8.6
21.5 ⫾ 8.6
3.3 ⫾ 1
18.3 ⫾ 5.9
3.9 ⫾ 2.1
15.5 ⫾ 4.4
19.2 ⫾ 3.7
3.22 ⫾ 1.1
19 ⫾ 4.8
4.9 ⫾ 2.3
0.0001
0.08
0.91
0.44
0.0002
0.78
0.36
0.89
0.79
0.65
⬍0.0001
0.12
⬍0.0001
⬍0.0001
0.005
⫺3.6 to ⫹15.6
⫺12.2 to ⫹16.9
⫺0.57 to ⫹0.75
⫺5 to ⫹2.2
⫺6.1 to ⫹2
was also used to predict LV filling pressure ⱖ15 mm Hg,
whereas a cut-off value ⬍8 determined LV filling pressure
⬍15 mm Hg.11 A cut-off value ⱖ2.5 for E/Vp was used to
predict LV filling pressure ⱖ15 mm Hg, whereas a cut-off
value ⬍1.5 determined LV filling pressure ⬍15 mm Hg.
For the approximate estimation of mean right atrial pressure using the 2-dimensional percentage collapse of the
inferior vena cava by the subcostal view, 2 different methods were used: the conventional published method12 and a
new estimation of collapse for this population of patients, in
which collapse of the inferior vena cava ⱖ50% indicated
atrial pressure of 5 mm Hg, collapse of 10% to 50% indicated atrial pressure of 10 mm Hg, no collapse (⬍10%)
associated with an inferior vena cava diameter ⬍25 mm
indicated atrial pressure of 15 mm Hg, and no collapse
associated with an inferior vena cava diameter ⱖ25 mm
indicated atrial pressure of 20 mm Hg.13 These different
echocardiographic measurements were used to calculate
systemic and pulmonary vascular resistance.
Quantitative data are expressed as mean ⫾ SD and ranges
and qualitative data as frequencies and percentages. Comparisons of means were performed using Student’s t test or the
Mann-Whitney nonparametric test as necessary. Comparisons
of frequencies were performed using the chi-square test or
Fisher’s exact test as necessary. We performed correlations
between different echocardiographic parameters and hemodynamic data using linear regression analysis. Agreement between invasive and echocardiographic data was evaluated by
plotting the difference against the mean value of the measurements (using the Bland-Altman method). We used the most
accurate echocardiographic method for the estimation of
PCWP, according to the results of correlations and BlandAltman analysis, for the calculation of pulmonary vascular
resistance. A p value ⬍0.05 was considered statistically significant. Statistical analysis was performed using SAS version
8.2 (SAS Institute Inc., Cary, North Carolina).
Fifteen men and 5 women were included (mean age 47 ⫾
12 years). The diagnosis of systolic LV dysfunction was
dilated cardiomyopathy in 17 patients and severe coronary
artery disease in 3 patients. The mean LV ejection fraction
was 19 ⫾ 7% (range 10% to 29%). The mean LV enddiastolic diameter was 67.8 ⫾ 5.5 mm (range 61 to 79). The
p Value
Bland-Altman Analysis
indications for right-sided cardiac catheterization were refractory HF (n ⫽ 15) and cardiac transplantation (n ⫽ 5).
Invasive and echocardiographic results are listed in
Table 1. Hemodynamic results did not differ significantly
between catheterization and echocardiography, except
for PCWP using E/E= and pulmonary vascular resistance
(p ⫽ 0.0001 and p ⫽ 0.0002, respectively). For right
atrial pressure, no significant difference was noted between the use of conventional and new echocardiographic
parameters (p ⫽ 0.14), but Bland-Altman plots revealed
a higher dispersion of plots using the conventional echocardiographic method.
Using cut-off values of E/E= for the assessment of PCWP,
all patients were well classified: 3 patients had PCWPs ⬍15
mm Hg and E/E= ⬍8, and 17 patients had PCWPs ⱖ15 mm
Hg and E/E= ⬎15. Using cut-off values of E/Vp, 11 patients
(55%) had ratios of 1.5 to 2.5, which did not allow the
estimation of PCWP.
E/E= was used for the calculation of pulmonary vascular
resistance. There was good agreement between systemic
and pulmonary vascular resistance determined by catheterization and echocardiography (Figure 1). Four patients had
pulmonary vascular resistance ⱖ6 Wood units using catheterization compared with 7 patients using echocardiography (p ⫽ 0.48). All patients with invasive pulmonary vascular resistance ⱖ6 Wood units were well classified using
echocardiography. All patients with echocardiographic pulmonary vascular resistance ⬍6 Wood units (n ⫽ 13) also
had invasive pulmonary vascular resistance ⬍6 Wood units.
Continuous invasive monitoring was performed in 13
patients requiring bolus infusions of intravenous furosemide
(n ⫽ 10) or saline infusion (n ⫽ 3). Among all echocardiographic parameters studied, only mean right atrial pressure
was related to the use of saline infusion or the infusion of
intravenous furosemide: all patients with ⱖ50% collapse of
the inferior vena cava (n ⫽ 3) required saline infusion,
whereas all patients without any inferior vena cava collapse
(n ⫽ 7), corresponding to echocardiographic right atrial
pressure ⱖ15 mm Hg, required infusions of intravenous
furosemide. All patients with PCWPs ⬍15 mm Hg required
saline infusions, but PCWP ⱖ15 mm Hg was not associated
with the use of furosemide.
1000
The American Journal of Cardiology (www.AJConline.org)
Figure 1. Linear regressions and Bland-Altman plots of (A) systemic vascular resistance (SVR) and (B) pulmonary vascular resistance (PVR) for
echocardiography and right-sided cardiac catheterization. All vascular resistances are expressed in Wood units. Dashed lines represent 1.96 SDs from the
mean.
After 24 hours of treatment, echocardiographic follow-up demonstrated that in patients requiring saline infusions (n ⫽ 3), initial inferior vena cava collapse ⱖ50% was
modified, with collapse of 10% to 50%, whereas a tendency
toward increased PCWP was observed (9.3 ⫾ 4 mm Hg
initially vs 16.3 ⫾ 1.5 mm Hg after treatment, p ⫽ 0.09). In
patients requiring intravenous furosemide (n ⫽ 10), systolic
and mean pulmonary arterial pressures and PCWPs significantly decreased (p ⫽ 0.04), and inferior vena cava collapse in all cases ranged from 10% to 50%, but all these
patients presented with persistent elevated PCWPs (ⱖ15
mm Hg).
Discussion
In the 1980s and 1990s, right-sided cardiac catheterization was
routinely used to assess cardiac pressures in intensive care
units. However, iatrogenic complications, although uncommon, are possible. Furthermore, the indications for echocardiography have recently been extended, with the use of second
harmonic imaging and tissue Doppler imaging.8,11,14 We compared echocardiography with right-sided cardiac catheterization to determine the accuracy of echocardiography in this
high-risk population. Echocardiography did not give very
precise values of PCWP, but using previously published
cut-off values, E/E= correctly selected patients with low or
elevated LV filling pressures, whereas using E/Vp, echocardiographic indecision was found in most patients (55%).
The E/E= ratio should not be used for an exact value of
PCWP, because a significant difference between echocardiographic and invasive data was observed in our study
(p ⫽ 0.0001). In a normal population, PCWP ⬍15 mm Hg
is considered normal. In patients with end-stage heart disease, PCWP is usually higher, as demonstrated in our study.
In fact, PCWP ⬍15 mm Hg reflected hypovolemia, and
these patients should receive saline infusions, whereas a
value ⬎15 mm Hg could not be used to select patients
requiring intravenous furosemide. In contrast, study of the
inferior vena cava seems to be helpful in estimating right
atrial pressure and filling pressure, guides treatment, and
can be used to monitor the response to medication in patients with end-stage HF. Echocardiography should therefore be interpreted in light of the severity of HF.
One of the last cardiac catheterization indications remains
cardiac transplantation,15,16 for the measurement of all intracardiac pressures and the estimation of pulmonary vascular
resistance. This is of particular importance for the choice of
cardiac transplantation.16 Orthotopic cardiac transplantation
requires low pulmonary vascular resistance. It has been
established that a value ⱖ6 Wood units is a serious contraindication to orthotopic heart transplantation.16 In our study,
all patients with invasive values ⱖ6 Wood units were well
classified using echocardiography. In contrast, an echocardiographic value ⬍6 Wood units predicts low pulmonary
vascular resistance. So echocardiography may be used as a
reliable tool to predict pulmonary vascular resistance ⬍6
Wood units, whereas echocardiographic pulmonary vascular resistance ⱖ6 Wood units needs to be confirmed using
right-sided cardiac catheterization.
The limitation of our study is the small number of patients. The indications for right-sided cardiac catheterization
are limited, and the population of our study is rare. However, we systematically performed echocardiography and
cardiac catheterization, considered the gold standard in all
patients, and demonstrated the value of echocardiography in
this high-risk population for the assessment of hemodynamic profile and the choice of medication.
Heart Failure/Echocardiography and Heart Failure
1. American Heart Association. Heart Disease and Stroke Statistics: 2005
Update. Dallas, Texas: American Heart Association, 2005.
2. The CONSENSUS Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North
Scandinavian Enalapril Survival Study (CONSENSUS). N Engl J Med
1987;316:1429 –1435.
3. Capomolla S, Ceresa M, Pinna G, Maestri R, La Rovere MT, Febo O,
Rossi A, Paganini V, Caporotondi A, Guazzotti G, et al. Echo-Doppler
and clinical evaluations to define hemodynamic profile in patients with
chronic heart failure: accuracy and influence on therapeutic management. Eur J Heart Fail 2005;7:624 – 630.
4. Nieminen MS, Brutsaert D, Dickstein K, Drexler H, Follath F, Harjola
VP, Hochadel M, Komajda M, Lassus J, Lopez-Sendon JL, et al.
EuroHeart Failure Survey II (EHFS II): a survey on hospitalized acute
heart failure patients: description of population. Eur Heart J 2006;27:
2725–2736.
5. Yock PG, Popp RL. Noninvasive estimation of right ventricular systolic pressure by Doppler ultrasound in patients with tricuspid regurgitation. Circulation 1984;70:657– 662.
6. Raffoul H, Guéret P, Diebold B, Cohen A, Abergel E, Zelinsky R,
Ourbak P, Peronneau P, Guerin F. Intérêt de l’enregistrement du flux
pulmonaire en Doppler continu pour l’estimation de la pression artérielle pulmonaire systolique. Arch Mal Coeur 1990;83:1703–1709.
7. Coats AJ. Doppler ultrasonic measurement of cardiac output: reproducibility and validation. Eur Heart J 1990;11(suppl):49 – 61.
8. Nagueh SF, Middleton KJ, Kopelen HA, Zoghbi WA, Quinones MA.
Doppler tissue imaging: a noninvasive technique for evaluation of left
ventricular relaxation and estimation of filling pressures. J Am Coll
Cardiol 1997;30:1527–1533.
9. Mansencal N, Bouvier E, Joseph T, Farcot JC, Pilliere R, Redheuil A,
Lacombe P, Jondeau G, Dubourg O. Value of tissue Doppler imaging
10.
11.
12.
13.
14.
15.
16.
1001
to predict left ventricular filling pressure in patients with coronary
artery disease. Echocardiography 2004;21:133–138.
Garcia MJ, Ares MA, Asher C, Rodriguez L, Vandervoort P, Thomas
JD. An index of early left ventricular filling that combined with pulsed
Doppler peak E velocity may estimate capillary wedge pressure. J Am
Coll Cardiol 1997;29:448 – 454.
Ommen SR, Nishimura RA, Appleton CP, Miller FA, Oh JK, Redfield
MM, Tajik AJ. Clinical utility of Doppler echocardiography and tissue
Doppler imaging in the estimation of left ventricular filling pressures:
a comparative simultaneous Doppler-catheterization study. Circulation 2000;102:1788 –1794.
Kircher BJ, Himelman RB, Schiller NB. Noninvasive estimation of
right atrial pressure from the inspiratory collapse of the inferior vena
cava. Am J Cardiol 1990;66:493– 496.
Wong SP, Otto CM. Echocardiographic findings in acute and chronic
pulmonary disease. In: Otto CM, ed. The Practice of Clinical Echocardiography. 2nd ed. Philadelphia, Pennsylvania: W.B. Saunders,
2002:739 –760.
Mansencal N, Bordachar P, Chatellier G, Redheuil A, Diebold B,
Abergel E. Comparison of accuracy of left ventricular echocardiographic measurements by fundamental imaging versus second harmonic imaging. Am J Cardiol 2003;91:1037–1039.
Annas GJ. Regulating the introduction of heart and liver transplantation. Am J Public Health 1985;75:93–95.
Costanzo MR, Augustine S, Bourge R, Bristow M, O’Connell JB,
Driscoll D, Rose E. Selection and treatment of candidates for heart
transplantation. A statement for health professionals from the Committee on Heart Failure and Cardiac Transplantation of the Council on
Clinical Cardiology, American Heart Association. Circulation 1995;
92:3593–3612.
Optimizing the Programation of Cardiac Resynchronization
Therapy Devices in Patients With Heart Failure and Left Bundle
Branch Block
Bàrbara Vidal, MD, Marta Sitges, MD, PhD*, Alba Marigliano, MD, Victoria Delgado, MD,
Ernesto Díaz-Infante, MD, Manel Azqueta, MD, David Tamborero, MSE, José María Tolosana, MD,
Antonio Berruezo, MD, Félix Pérez-Villa, MD, PhD, Carles Paré, MD, PhD, Lluís Mont, MD, PhD,
and Josep Brugada, MD, PhD
This study was conducted to investigate the clinical impact of cardiac resynchronization
device optimization. A series of 100 consecutive patients received cardiac resynchronization
therapy. In the first 49 patients, an empirical atrioventricular delay of 120 ms was set, with
simultaneous biventricular stimulation (interventricular [VV] interval ⴝ 0 ms). In the next
51 patients, systematic atrioventricular optimization was performed. VV optimization was
also performed, selecting 1 VV delay: right or left ventricular preactivation (ⴙ30 or
ⴚ30 ms) or simultaneous (VV interval ⴝ 0 ms), according to the best synchrony obtained
by tissue Doppler– derived wall displacement. At follow-up, patients who were alive without cardiac transplantation and showed improvement of >10% in the distance walked in
the 6-minute walking test were considered responders. There were no differences between
the 2 groups at baseline. Left ventricular ejection fraction improved in the 2 groups, but left
ventricular cardiac output improved only in the optimized group. At 6 months, patients
with optimized devices walked slightly further in the 6-minute walking test (497 ⴞ 167 vs
393 ⴞ 123 m, p <0.01), with no differences in New York Heart Association functional class
or quality of life compared with nonoptimized patients. Overall, the number of nonresponders were similar in the 2 groups (27% vs 23%, p ⴝ NS). In conclusion, the echocardiographic optimization of cardiac resynchronization devices provided a slight incremental
clinical benefit at midterm follow-up. Simple and rapid methods to routinely optimize the
devices are warranted. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:
1002–1006)
Although the clinical benefit of cardiac resynchronization
therapy (CRT) has been proved in patients with advanced
heart failure and left bundle branch block (LBBB),1,2 about
30% of patients do not respond to this treatment.3,4 To
improve these results, CRT devices have been equipped to
stimulate the 2 ventricles either simultaneously or sequentially, with a specific delay (the interventricular [VV] interval). Given that some ventricular segments have delayed
conduction and contraction in patients with left ventricular
(LV) dysfunction and LBBB, it is reasonable to think that
LV preactivation may achieve better resynchronization.5,6
Furthermore, LV stimulation is currently performed from
the epicardium, whereas the right ventricle is stimulated
from the endocardium. It has been shown that pacing from
the epicardium results in a transmural delay.7,8 Lead positioning may also affect the transmission of the stimulus and
the ventricular activation sequence, causing VV and intraventricular delays, which can in turn be modified with VV
programming. Sequential biventricular stimulation has been
shown to induce acute hemodynamic improvement compared with simultaneous biventricular pacing.5,7,9 –11 However, little is known about the clinical benefit of the optimization of the device.12 The main objective of our study
was to evaluate whether the proportion of responders in the
optimized group was higher than in the nonoptimized
group. The secondary objective was to compare the extent
of LV reverse remodeling in the 2 groups.
Thorax Clinic Institute, Hospital Clínic, Institut d’Investigacions
Biomèdiques August Pi i Sunyer, University of Barcelona, Barcelona,
Spain. Manuscript received March 14, 2007; revised manuscript received
and accepted April 24, 2007.
Drs. Vidal, Delgado, and Tolosana were supported by a postresidency
award from Fundació Clínic, Barcelona, Spain. This study was supported
in part by a grant from Fundación Española del Corazón 2006, Madrid, and
by Grant FIS PI04/90069 from Fondo de Investigaciones Sanitarias, Madrid, Spain.
*Corresponding author: Tel: 34-93-227-9305; fax: 34-93-451-41-48.
E-mail address: [email protected] (M. Sitges).
Methods
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.046
In this observational cohort study, patients treated with
simultaneous biventricular pacing devices were compared
with patients who received CRT devices with sequential
stimulation capability (and were therefore optimized). The
nonoptimized group comprised 49 consecutive patients who
received CRT devices without sequential activation capability. An empirical atrioventricular (AV) interval of 120 ms
and a VV interval of 0 ms were set. The optimized group
www.AJConline.org
Heart Failure/CRT Device Optimization
included 51 consecutive patients who underwent AV and
VV interval optimization.
The inclusion criteria for this study were (1) New York
Heart Association functional class III or IV despite receiving optimal medical treatment for ⱖ2 weeks before device
implantation, (2) LV ejection fraction ⱕ35%, (3) LV enddiastolic diameter ⬎55 mm, and (4) QRS width ⬎130 ms.
Patients were excluded if (1) they had treatable cardiopathies, (2) heart transplantation was considered in ⬍6
months, or (3) they had co-morbidities that shortened life
expectancy. The investigation conformed with the principles outlined in the Declaration of Helsinki, and the study
protocol was accepted by our hospital’s ethics committee.
Written informed consent was obtained from all patients.
The study protocol included a baseline patient evaluation
with the performance of transthoracic echocardiography
(echocardiography off) to study LV anatomy, function, and
synchrony and a clinical evaluation to determine New York
Heart Association functional class, quality of life (using the
Minnesota Living With Heart Failure Questionnaire), and
6-minute walking distance. The same echocardiographic
protocol (echocardiography on) was repeated 24 to 72 hours
after device implantation, and AV and VV intervals were
optimized when the devices had the capability to stimulate
sequentially. All patients were followed for 6 months, when
new clinical and echocardiographic evaluations were performed. Patients were considered CRT clinical responders if
at 6-month follow-up they were alive, had not required heart
transplantation, and had improved the distance covered in
the 6-minute walking test by ⱖ10%.
Patients received pacemakers or defibrillators, according
to clinical indications. Patients who received pacemakers
without sequential stimulation capability were implanted
with either a ContakHF or a Contak-Renewal (Guidant
Corporation, Indianapolis, Indiana), and those assigned to
devices with VV interval programming capability received
a Contak TR or a Contak-Renewal II (Guidant Corporation). If a defibrillator was indicated, a Contak-Renewal II
(Guidant Corporation) was implanted. One electrode was
placed in the right atrium (if the patient was in sinus
rhythm), another at the apex of the right ventricle, and the
Easy Track (Guidant Corporation) electrode was implanted
through the coronary sinus and advanced to a posterolateral
branch. All LV leads were implanted transvenously.
Standard Doppler echocardiography using a commercially available system (Sonos 5500; Philips Medical Systems, Andover, Massachusetts; or Vivid 7; GE-Vingmed
Ultrasound AS, Horten, Norway) was performed just before
initiating CRT, 24 to 72 hours after the implantation of the
device and at 6-month follow-up. The same parameters
were evaluated in each echocardiographic scan: LV dimensions were measured from M-mode echocardiography in the
parasternal long-axis view, LV volumes and ejection fractions were quantified by Simpson’s method, and LV stroke
volume was calculated using quantitative Doppler echocardiography, following the recommendations of the American
Society of Echocardiography.13
Interventricular delay was calculated as the time difference between the preejective period at the pulmonary and
aortic valves (from QRS to the onset of flow).14 Intraventricular LV asynchrony was evaluated with the time differ-
1003
ence in peak contraction of the septal and posterior walls of
the left ventricle with M-mode scans from the parasternal
view15; in the group of optimized devices, intraventricular
LV synchrony was also assessed using tissue Doppler imaging (Tissue Tracking, EchoPac; GE Healthcare, Milwaukee, Wisconsin) by evaluating the synchronicity in the displacement of the lateral and septal walls and of the anterior
and inferior walls in the 4- and 2-chamber apical views,
respectively. All studies were digitally stored and analyzed
offline by 2 experienced cardiologists who were not involved in the clinical follow-up and were therefore unaware
of each patient’s clinical outcome.
The AV and VV intervals were optimized by programming the devices following a standardized protocol. The
AV interval was analyzed with pulsed Doppler, studying the
LV filling flow pattern and trying to separate the A wave
from the E wave to obtain the longest LV diastolic filling
time. LV diastolic filling time was evaluated at 3 AV intervals: 160, 140, and 120 ms, choosing the interval that
yielded the longest LV filling time without interrupting the
A wave. Once the optimal AV interval was established, the
optimal VV interval was chosen empirically: either right
ventricular preactivation (VV ⫽ 30 ms), simultaneous
biventricular pacing (VV ⫽ 0 ms), or LV preactivation
(VV ⫽ ⫺30 ms). The optimal VV interval was the interval
that yielded the best intraventricular synchrony as demonstrated by tissue Doppler imaging, with a greater superposition of the curves of displacement of 2 opposite LV walls
(Figure 1). To check that assessment of the best synchrony
correlated with the best hemodynamic response,11,16 the
velocity–time integral (VTI) at the aortic valve was evaluated with pulsed-wave Doppler in 30 random patients at
each of the VV delays tested, and the agreement for determining the best intervals was evaluated.
Quantitative variables are expressed as mean ⫾ SD,
whereas qualitative variables are expressed as number and
percentage. Student’s t test for paired data was used to
compare echocardiographic measurements. Discrete variables were compared using the chi-square test. Functional
class before and after CRT was compared using Wilcoxon’s
sign test. Concordance between aortic VTI and tissue tracking curves for assessing the optimal VV interval was analyzed with the ␬ coefficient. Statistical significance was
defined at p ⬍0.05. All data were analyzed using SPSS
version 11.0 (SPSS, Inc., Chicago, Illinois).
Results
One hundred patients who had received CRT devices with
or without implantable cardioverter defibrillators were studied (27 [55%] in the nonoptimized group and 36 [70%] in
the optimized group received defibrillators). Eighty-one patients (81%) were men, and the mean age was 70 ⫾ 8 years.
Among them, 49 patients (49%) had simultaneous biventricular pacing systems, and the remaining 51 (51%) received sequential biventricular pacing devices that were
optimized after implantation. At baseline, there were no
differences in clinical characteristics, functional status, or
LV function between the 2 groups (Table 1).
Devices without the capability for VV programming
were set at a standard AV interval of 120 ms (sensed AV
1004
The American Journal of Cardiology (www.AJConline.org)
Figure 1. VV interval optimization by assessing intraventricular synchrony with the evaluation of tissue displacement curves of the lateral and septal walls
in the 4-chamber (4Ch) apical view and of the anterior and inferior walls in the 2-chamber (2Ch) apical view. The highest level of superposition of the tracings
was observed with LV preactivation at ⫺30 ms, which was chosen as the optimal programming.
Table 1
Baseline clinical and echocardiographic characteristics (p ⫽ NS)
Variable
Age (yrs)
6-min walking test (m)
Minnesota Living With Heart Failure
Questionnaire score
NYHA functional class
Ischemic cause
QRS (ms)
LV end-diastolic volume (ml)
LV end-systolic volume (ml)
LV end-diastolic diameter (mm)
LV end-systolic diameter (mm)
LV ejection fraction (%)
LV cardiac output (L/min)
Nonoptimized
Group
(n ⫽ 49)
Optimized
Group
(n ⫽ 51)
71 ⫾ 7
307 ⫾ 88
41 ⫾ 20
69 ⫾ 8
336 ⫾ 160
43 ⫾ 21
3.0 ⫾ 0.5
23 (47%)
177 ⫾ 29
226 ⫾ 89
175 ⫾ 78
74 ⫾ 7
61 ⫾ 9
23 ⫾ 6
3.7 ⫾ 0.8
3.0 ⫾ 0.6
24 (47%)
173 ⫾ 23
215 ⫾ 80
161 ⫾ 75
73 ⫾ 10
58 ⫾ 10
25 ⫾ 8
3.8 ⫾ 1.5
NYHA ⫽ New York Heart Association.
delay ⫽ 100 ms) and a VV interval of 0 ms in all 49
patients, with a basic heart rate of 60 beats/min. Device
optimization, including optimal AV and VV interval assessment, was performed in all 51 patients who had received
devices with sequential biventricular stimulation capability.
Among these, AV optimization was not performed in 13
patients (25%) because of the presence of permanent atrial
fibrillation. In most patients (n ⫽ 20 [39%]), the optimal
AV delay was set at 140 ms; in 12 patients (23%), an AV
interval of 120 ms was selected, and in only 6 patients
(12%), an AV delay of 160 ms was set because it yielded the
best diastolic filling time. LV preactivation at ⫺30 ms was
chosen in most patients (n ⫽ 37 [72%]) as the setting that
provided the greatest degree of LV synchrony. Simultaneous biventricular pacing (VV delay ⫽ 0 ms) was considered optimal in 11 patients (21%), whereas only 3 patients
(6%) benefited more from right ventricular preactivation
(VV ⫽ 30 ms).
In 30 randomly selected patients, concordance between
optimal VV delay by tissue displacement curves and by
aortic VTI was good (␬ coefficient ⫽ 0.66, p ⬍0.01).
Ninety-eight patients (98%) completed 6-month followup. Two patients were lost to follow-up because they lived
Table 2
Events according to cardiac resynchronization therapy device
optimization at 6 month follow-up
Variable
Combined end point
(nonresponders)*
Cardiac death or heart
transplantation
Improvement ⬍10% in
6-min walking test
Distance covered in
6-min walking test (m)
⌬ distance covered in
6-min walking test (m)
NYHA functional class
Quality-of-life score
Nonoptimized
CRT
(n ⫽ 47)
Optimized
CRT
(n ⫽ 51)
p Value
11 (27%)
10 (23%)
NS
3 (6%)
4 (7%)
NS
9 (19%)
5 (15%)
NS
393 ⫾ 123
497 ⫾ 167
⬍0.01
84 ⫾ 128
136 ⫾ 160
NS
2.0 ⫾ 0.7
28 ⫾ 17
2.0 ⫾ 0.6
22 ⫾ 18
NS
NS
* Combined end point at 6-month follow-up: cardiovascular death, heart
transplantation, or ⬍10% improvement in the 6-minute walking test.
Abbreviation as in Table 1.
far away from our referring area. By this time, 1 patient
(1%) had been transplanted, 6 (6%) had died, and 14 (14%)
had not improved the distance covered in the 6-minute
walking test by ⬎10%. Therefore, according to the main
combined end point of the study, a total of 21 patients (21%)
were nonresponders at 6-month follow-up.
Although the optimized group tended to have fewer
nonresponders than the nonoptimized group, there was no
significant clinical difference in the main clinical end
points. Nor were there differences in New York Heart Association functional class or the score obtained on the quality-of-life test at 6-month follow-up. However, patients with
optimized devices performed slightly better on the 6-minute
walking test. In the optimized patients, the distance walked
in the 6-minute test was slightly longer (497 ⫾ 167 m in the
optimized patients vs 393 ⫾ 123 m in the nonoptimized
patients, p ⬍0.01) and the quality-of-life score lower (22 ⫾
18 in the optimized patients vs 28 ⫾ 17 in the nonoptimized
patients, p ⫽ NS) at 6-month follow-up (Table 2).
When we excluded patients with atrial fibrillation as the
baseline rhythm (n ⫽ 20, 7 in the nonoptimized group and
13 in the optimized group), we also found a similar impact
Heart Failure/CRT Device Optimization
Table 3
Echocardiographic characteristics at six-month follow-up
Variable
LV end-diastolic volume (ml)
LV end-systolic volume (ml)
LV end-diastolic diameter (mm)
LV end-systolic diameter (mm)
LV ejection fraction (%)
⌬ LV ejection fraction (%)
LV cardiac output (L/min)
Nonoptimized
CRT
(n ⫽ 47)
Optimized
CRT
(n ⫽ 51)
p
Value
204 ⫾ 82
151 ⫾ 71
73 ⫾ 10
54 ⫾ 13
28 ⫾ 9
20 ⫾ 23
3.6 ⫾ 0.5
197 ⫾ 82
140 ⫾ 71
71 ⫾ 10
53 ⫾ 15
30 ⫾ 9
23 ⫾ 45
4.3 ⫾ 1.4
NS
NS
NS
NS
NS
NS
⬍0.05
of device optimization on clinical outcomes, with only a
nonsignificant trend toward fewer events in the optimized
group (incidence of the combined end point: 11 [28%] in the
nonoptimized group vs 7 [19%] in the optimized group,
p ⫽ 0.4).
Table 3 lists LV dimensions and function 6 months after
device implantation in patients with optimized and nonoptimized devices. The 2 groups showed improved LV systolic function and reduced LV diameters and volumes at
follow-up. There was a nonsignificant tendency toward
greater remodeling in optimized patients, although LV cardiac output was significantly higher in the optimized group.
Discussion
Our results show that the optimization of AV and VV
intervals in CRT devices slightly increases cardiac output
and improves functional performance, as reflected in the
distance covered in the 6-minute walking test at 6-month
follow-up examination of patients with heart failure and
LBBB. However, optimization did not increase the percentage of responders or the extent of LV reverse remodeling.
Although CRT provided a significant benefit in the 2
groups, the events observed and the LV reverse remodeling
in patients with optimized devices were not significantly
better than in the nonoptimized group. With a larger population or a longer follow-up period, differences in LV reverse remodeling or hemodynamics might have become
more evident, but the small clinical differences found in our
study (on the 6-minute walking test alone or in the quality
of life when excluding patients with atrial fibrillation) suggest that the added clinical benefit of CRT optimization is
slight and should be counterbalanced by the cost and the
time required for optimization in patients with LV dysfunction and LBBB.
Our results are in concordance with those of a study by
Mortensen et al,12 who compared the response to CRT at
3-month follow-up of optimized patients and patients
treated with simultaneous biventricular pacing. They observed that although there was a significant improvement in
functional class and 6-minute walking distance with CRT in
the 2 groups, there was no significant difference in the
benefit obtained. Bordachar et al17 reported an incremental
benefit in cardiac performance and LV synchrony in patients who underwent VV optimization. However, that study
did not report any clinical follow-up. More recently, Leon et
al11 described similar findings at 6-month follow-up in a
1005
large series of patients from the Multicenter InSync Randomized Clinical Evaluation (MIRACLE) and InSync III
studies. Boriani et al16 recently reported no changes in
quality of life or functional class after VV optimization.
However, none of these studies included hard events (cardiac death or transplantation) at follow-up. As far as we
know, there is no more evidence that the optimization of AV
and VV intervals provide an incremental clinical benefit at
longer term follow-up. Accordingly, we believe that our
findings in a consecutive series of patients from a singlecenter experience, despite the limitations inherent in a nonrandomized study, provide more data regarding the clinical
utility of optimizing CRT devices. It may be also argued
that a 15% to 20% increment in distance walked is a tremendous benefit in the context of patients with heart failure,
as reported in the MIRACLE trial.18 Nonetheless, this must
be counterbalanced with the personal and time-consuming
cost of optimizing the devices.
In ⬎70% of our patients, LV preactivation at ⫺30 ms
with a relatively short AV interval was chosen as optimal.
Previous studies from our group have demonstrated that LV
transmural activation time is about 30 ms8; consequently,
considering that the LV lead lies on the epicardium, LV
preactivation at ⫺30 ms may well be the optimal sequence
in this group of patients with LBBB. Other investigators
have obtained similar results studying invasively the effect
of VV optimization on LV dP/dt.5,7 Conversely, other investigators who have used echocardiographic methods to
optimize devices, such as the effect on Doppler-derived LV
dP/dt or cardiac output, have reported that up to 25% of
patients can benefit more from simultaneous biventricular or
right ventricular preactivation.9,19 Differences in the positioning of the right ventricular lead (outflow tract or apex)
or in baseline conduction abnormalities of the studied population may account for these discrepancies in optimal VV
programming. The fact that several investigators have reported different optimal programming schedules5,7,9,19 is a
reflection of the wide variability existing in the underlying
conduction abnormalities and LV mechanical asynchrony in
patients who undergo CRT. Therefore, individualized programming of the devices is theoretically important to obtain
an incremental benefit.20 This may be of great interest in
patients with narrow QRS widths, with right bundle branch
block, or upgraded to CRT from conventional pacing in
whom the conduction abnormalities may be less homogenous than in a population with LBBB such as that included
in our study. However, in view of the relative small incremental benefit observed, simple and rapid methods are warranted to select the best programming.
In this study, we assessed only a few AV and VV delays.
Some patients might have improved more with longer delays, especially those with ischemic cardiomyopathy, who
may require longer VV intervals because of the presence of
scar tissue resulting in a slower conduction velocity.21 However, the limited data on this issue suggest that the effect of
different VV delays on hemodynamics varies within a small
range (VV ⫽ ⫹20 to ⫺20 ms). We did not check the
programming of the devices at follow-up, which could have
improved the outcomes in the optimized group. Additionally, there is no consensus on whether it is better to optimize
first the AV or the VV interval; in our patients, we checked
1006
The American Journal of Cardiology (www.AJConline.org)
the AV delay after optimizing the VV interval, and we made
no changes in any patient. Finally, the optimal AV and VV
intervals were selected on the basis of the greatest intraventricular LV synchrony obtained, because this has been
shown to have prognostic implications.22 Although there is
no established consensus on what is the most adequate or
accurate method to evaluate intraventricular LV synchrony,
tissue Doppler imaging– based techniques seem to be the
most useful at present.23 In our patients, we used the degree
of superposition between the displacement curves of 2 opposite myocardial segments (Figure 1), and we chose the
VV interval. We are aware that this method, like others used
by different investigators, is not standardized and may
therefore have resulted in inadequate optimizations of the
devices; as a result, no added clinical benefit would be
expected in comparison with simultaneous biventricular
pacing. In contrast, the measurement of time to peak velocities or peak strain may be cumbersome in patients with
suboptimal quality and noisy traces, and indeed, variability
in measurement has been pointed out as a limitation in the
use of this method to assess LV asynchrony.23,24 Finally,
in our experience, when the VV interval is evaluated with
other methodologies, such as surface electrocardiography25
or, as demonstrated in the present study, the aortic VTI, we
obtain a good correlation with echocardiography, supporting the validity of this echocardiographic optimization
method.
1. Cleland JGF, Daubert J-C, Erdmann E, Freemantle N, Gras D, Kappenberger L, Tavazzi L; Cardiac Resynchronization–Heart 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.
2. Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco
T, Carson P, Di Carlo L, De Mets D, White BG, et al. Cardiacresynchronization therapy with or without an implantable defibrillator
in advanced chronic heart failure. N Engl J Med 2004;350:2140 –2150.
3. 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.
4. Diaz-Infante E, Mont L, Leal J, Garcia-Bolao I, Fernandez-Lozano I,
Hernandez-Madrid A, Perez-Castellano N, Sitges M, Pavon-Jimenez
R, Barba J, et al. Predictors of lack of response to resynchronization
therapy. Am J Cardiol 2005;95:1436 –1440.
5. van Gelder BM, Bracke FA, Meijer A, Lakerveld LJ, Pijls NH. Effect
of optimizing the VV interval on left ventricular contractility in cardiac
resynchronization therapy. Am J Cardiol 2004;93:1500 –1503.
6. Porciani MC, Dondina C, Macioce R, Demarchi G, Pieragnoli P,
Musilli N, Colella A, Ricciardi G, Michelucci A, Padeletti L. Echocardiographic examination of atrioventricular and interventricular delay optimization in cardiac resynchronization therapy. Am J Cardiol
2005;95:1108 –1110.
7. Perego GB, Chianca R, Facchini M, Frattola A, Balla E, Zucchi S,
Cavaglia S, Vicini I, Negretto M, Osculati G. Simultaneous vs. sequential biventricular pacing in dilated cardiomyopathy: an acute hemodynamic study. Eur J Heart Fail 2003;5:305–313.
8. Berruezo A, Mont L, Nava S, Chueca E, Bartholomay E, Brugada J.
Electrocardiographic recognition of the epicardial origin of ventricular
tachycardias. Circulation 2004;109:1842–1847.
9. Sogaard P, Egeblad H, Pedersen AK, Kim WY, Kristensen BO, Hansen PS, Mortensen PT. Sequential versus simultaneous biventricular
resynchronization for severe heart failure: evaluation by tissue Doppler
imaging. Circulation 2002;106:2078 –2084.
10. Kass DA, Chen CH, Curry C, Talbot M, Berger R, Fetics B, Nevo E.
Improved left ventricular mechanics from acute VDD pacing in patients with dilated cardiomyopathy and ventricular conduction delay.
Circulation 1999;99:1567–1573.
11. Leon AR, Abraham WT, Brozena S, Daubert JP, Fisher WG, Gurley
JC, Liang CS, Wong G. Cardiac resynchronization with sequential
biventricular pacing for the treatment of moderate-to-severe heart
failure. J Am Coll Cardiol 2005;46:2298 –2304.
12. Mortensen PT, Sogaard P, Mansour H, Ponsonaille J, Gras D, Lazarus
A, Reiser W, Alonso C, Linde CM, Lunati M, et al. Sequential
biventricular pacing: evaluation of safety and efficacy. Pacing Clin
Electrophysiol 2004;27:339 –345.
13. Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gutgesell H, Reichek N, Sahn D, Schnittger I, et al. Recommendations for quantitation of the left ventricle by two-dimensional
echocardiography. American Society of Echocardiography Committee
on Standards, Subcommittee on Quantitation of Two-Dimensional
Echocardiograms. J Am Soc Echocardiogr 1989;2:358 –367.
14. St. John Sutton MG, Plappert T, Abraham WT, Smith AL, DeLurgio
DB, Leon AR, Loh E, Kocovic DZ, Fisher WG, Ellestad M, et al.
Effect of cardiac resynchronization therapy on left ventricular size and
function in chronic heart failure. Circulation 2003;107:1985–1990.
15. Pitzalis MV, Iacoviello M, Romito R, Massari F, Rizzon B, Luzzi G,
Guida P, Andriani A, Mastropasqua F, Rizzon P. Cardiac resynchronization therapy tailored by echocardiographic evaluation of ventricular asynchrony. J Am Coll Cardiol 2002;40:1615–1622.
16. Boriani G, Muller CP, Seidl KH, Grove R, Vogt J, Danschel W,
Schuchert A, Djiane P, Biffi M, Becker T, et al. Randomized comparison of simultaneous biventricular stimulation versus optimized interventricular delay in cardiac resynchronization therapy. The Resynchronization for the Hemodynamic Treatment for Heart Failure
Management II Implantable Cardioverter Defibrillator (RHYTHM II
ICD) study. Am Heart J 2006;151:1050 –1058.
17. Bordachar P, Lafitte S, Reuter S, Sanders P, Jais P, Haissaguerre M,
Roudaut R, Garrigue S, Clementy J. Echocardiographic parameters of
ventricular dyssynchrony validation in patients with heart failure using
sequential biventricular pacing. J Am Coll Cardiol 2004;44:2157–
2165.
18. Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E,
Kocovic DZ, Packer M, Clavell AL, Hayes DL, et al. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002;346:1845–
1853.
19. Leon AR, Liang CS, Abraham WT, Chinchoy E, Hill MRS; US InSync
III Investigators and Coordinators. Interventricular delay increases
stroke volume in cardiac resynchronization patients. Eur Heart J
2002;23(suppl):529.
20. Fung JW, Yu CM, Yip G, Zhang Y, Chan H, Kum CC, Sanderson JE.
Variable left ventricular activation pattern in patients with heart failure
and left bundle branch block. Heart 2004;90:17–19.
21. Rodriguez LM, Timmermans C, Nabar A, Beatty G, Wellens HJ.
Variable patterns of septal activation in patients with left bundle
branch block and heart failure. J Cardiovasc Electrophysiol 2003;14:
135–141.
22. Bader H, Garrigue S, Lafitte S, Reuter S, Jais P, Haissaguerre M,
Bonnet J, Clementy J, Roudaut R. Intra-left ventricular electromechanical asynchrony. A new independent predictor of severe cardiac events
in heart failure patients. J Am Coll Cardiol 2004;43:248 –256.
23. Bax JJ, Ansalone G, Breithardt OA, Derumeaux G, Leclercq C, Schalij
MJ, Sogaard P, St. John Sutton M, Nihoyannopoulos P. Echocardiographic evaluation of cardiac resynchronization therapy: ready for
routine clinical use? A critical appraisal. J Am Coll Cardiol 2004;
44:1–9.
24. Yu CM, Fung JW, Zhang Q, Chan CK, Chan YS, Lin H, Kum LC,
Kong SL, Zhang Y, Sanderson JE. Tissue Doppler imaging is superior
to strain rate imaging and postsystolic shortening on the prediction of
reverse remodeling in both ischemic and nonischemic heart failure
after cardiac resynchronization therapy. Circulation 2004;110:66 –73.
25. Vidal B, Tamborero D, Mont L, Sitges M, Delgado V, Berruezo A,
Diaz Infante E, Tolosana JM, Pare C, Brugada J. Electrocardiographic
optimization of interventricular delay in cardiac resynchronization
therapy: correlation with echocardiography. J Cardiovasc Electrophysiol, in press.
Comparison of the Effects of Cardiac Resynchronization Therapy
in Patients With Class II Versus Class III and IV Heart Failure
(from the InSync/InSync ICD Italian Registry)†,‡
Maurizio Landolina, MDa,*, Maurizio Lunati, MDb, Maurizio Gasparini, MDc,
Massimo Santini, MDd, Luigi Padeletti, MDe, Augusto Achilli, MDf, Stefano Bianchi, MDg,
Francesco Laurenzi, MDh, Antonio Curnis, MDi, Antonio Vincenti, MDj, Sergio Valsecchi, PhDk,
and Alessandra Denaro, MSk, on behalf of the InSync/InSync ICD Italian Registry Investigators
Cardiac resynchronization therapy (CRT) is recommended for patients with New York Heart
Association (NYHA) class III or IV heart failure and wide QRS complexes. The aim of this
study was to compare the effects of CRT in patients in NYHA class II with those in NYHA
class III or IV. Nine hundred fifty-two patients (188 in NYHA class II) consecutively implanted
with biventricular devices and enrolled in a national observational registry were studied.
Clinical outcomes were estimated after 12 months of CRT, and long-term survival was
assessed. At a median follow-up of 16 months, significantly fewer major cardiovascular events
were reported in patients in NYHA class II compared with NYHA class III or IV (rate 13 vs
23 per 100 patient-years of follow-up, p <0.001). The percentage of patients who improved in
NYHA class status after 12 months of CRT was lower in those in class II than in those in class
III or IV (34% vs 69%, p <0.001), whereas the absolute increase in the ejection fraction was
similar (8 ⴞ 9% vs 9 ⴞ 11%, p ⴝ NS), as well as the reductions in end-diastolic diameter (ⴚ3
ⴞ 8 vs ⴚ3 ⴞ 8 mm, p ⴝ NS) and end-systolic diameter (ⴚ4 ⴞ 10 vs ⴚ6 ⴞ 10 mm, p ⴝ NS).
The NYHA class II group experienced lower all-cause mortality (log-rank test p ⴝ 0.018). In
the 2 groups, patients with major cardiovascular events during follow-up exhibited less or no
reverse remodeling compared with those with better long-term clinical outcomes. In conclusion,
the results of this study indicate that CRT induced similar improvements in ventricular
function in the 2 groups, whereas the improvement in functional status was significantly lower
for patients in NYHA class II than for those in class III or IV. A positive effect of CRT on
cardiac dimensions was associated with a long-term beneficial effect on disease progression in
patients in NYHA class II. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:
1007–1012)
Cardiac resynchronization therapy (CRT) is recommended
for patients with moderate or severe symptomatic heart
failure (HF) and evidence of ventricular dyssynchrony,1,2
because it has been demonstrated to improve quality of life,
New York Heart Association (NYHA) class status, and
exercise performance and to reduce HF hospitalizations and
mortality.3– 6 The effects of CRT in less symptomatic patients with HF (NYHA class II) with systolic left ventricular
(LV) dysfunction are still controversial, because conflicting
a
Fondazione Policlinico S. Matteo IRCCS, Pavia; bNiguarda Hospital,
Milan; cIRCCS Istituto Clinico Humanitas, Rozzano; dS. Filippo Neri
Hospital, Rome; eCareggi Hospital, Florence; fBelcolle Hospital, Viterbo;
g
Fatebenefratelli Hospital, Isola Tiberina; hS. Camillo de Lellis, Rome;
i
Spedali Civili, Brescia; jS. Gerardo dei Tintori, Monza; and kMedtronic
Italia, Rome, Italy. Manuscript received January 18, 2007; revised manuscript received and accepted April 13, 2007.
*Corresponding author: Tel: 39-0382-501276; fax: 39-0382-503161.
E-mail address: [email protected] (M. Landolina).
†
Conflicts of interest: Sergio Valsecchi and Alessandra Denaro are
employees of Medtronic Italia, Rome, Italy.
‡
A list of centers and investigators participating in the InSync/InSync
ICD Italian Registry is provided in the Appendix.
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.043
data have been reported.7–10 CRT in patients with mild HF
is currently under investigation in ongoing randomized controlled trials.11,12 We report the data of the InSync/InSync
ICD Italian Registry, a prospective observational study that
includes patients with HF, LV dysfunction, and wide QRS
complexes who received a device capable of delivering
biventricular pacing. The objective of the present analysis
was to compare the effect of CRT between patients with
mild symptoms of HF with its effect in those with moderate
to severe HF (NYHA class III or IV).
Methods
Since 1999, patients successfully implanted with biventricular pacing devices for CRT delivery (CRT models 8040 and
8042 and CRT defibrillator models 7272, 7277, and 7279;
Medtronic, Inc., Minneapolis, Minnesota) have been prospectively included in the InSync/InSync ICD Italian Registry. The registry enrolls patients with mild or severe symptomatic chronic HF (NYHA classes II to IV), ejection
fractions (EFs) ⱕ35%, and wide QRS complexes (⬎130
ms). Patients with recent myocardial infarctions (⬍3
months) or with decompensated HF were excluded. All
www.AJConline.org
1008
The American Journal of Cardiology (www.AJConline.org)
Table 1
Demographics, baseline clinical parameters, and pharmacologic treatment
Parameter
Median (IQR) follow-up, mo
Men
Age (yrs)
Ischemic cause of HF
Hospitalizations for HF (n/yr)
Hospitalizations for HF (d/yr)
QRS duration (ms)
Left bundle branch block
Chronic atrial fibrillation
Previous pacemaker
EF (%)
End-diastolic diameter (mm)
End-systolic diameter (mm)
Mitral regurgitation (grade)
CRT defibrillator
Diuretics
Angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers
Nitrates
␤ blockers
Class III antiarrhythmics
Anticoagulants
NYHA Class
Total
(n ⫽ 952)
II
(n ⫽ 188)
III or IV
(n ⫽ 764)
16 (8–33)
778 (82%)
67 ⫾ 10
427 (45%)
1.7 ⫾ 1.5
21 ⫾ 13
167 ⫾ 32
650 (68%)
157 (17%)
130 (14%)
27 ⫾ 8
69 ⫾ 9
58 ⫾ 11
2.2 ⫾ 1.0
350 (37%)
830 (87%)
707 (74%)
229 (24%)
460 (48%)
306 (32%)
202 (21%)
17 (8–33)
158 (84%)
64 ⫾ 11
71 (38%)
1.1 ⫾ 1.3
11 ⫾ 7
163 ⫾ 29
129 (69%)
18 (10%)
18 (10%)
29 ⫾ 7
69 ⫾ 10
57 ⫾ 12
1.9 ⫾ 1.0
84 (45%)
163 (87%)
145 (77%)
38 (20%)
105 (56%)
52 (28%)
29 (15%)
16 (7–34)
620 (81%)
67 ⫾ 10
356 (47%)
1.8 ⫾ 1.4
22 ⫾ 13
168 ⫾ 32
521 (68%)
139 (18%)
112 (15%)
27 ⫾ 8
69 ⫾ 9
59 ⫾ 10
2.3 ⫾ 1.0
266 (35%)
667 (87%)
562 (74%)
191 (25%)
355 (47%)
254 (33%)
173 (23%)
p Value
0.782†
0.358‡
0.0001*
0.029‡
0.0001†
0.007*
0.070*
0.911‡
0.004‡
0.069‡
0.001*
0.851*
0.310*
0.027*
0.012‡
0.825‡
0.316‡
0.169‡
0.021‡
0.142‡
0.030‡
* Unpaired Student’s t test.
†
Mann-Whitney nonparametric test.
‡
Chi-square analysis.
IRQ ⫽ interquartile range.
patients provided written informed consent approved by
each hospital’s ethics committee.
The devices and the pacing leads were implanted using
standard techniques,13 with the transvenous LV lead positioned in a lateral or posterolateral cardiac vein through the
coronary sinus. When a conventional indication for an implantable cardioverter defibrillator existed, a combined device was implanted.
The baseline evaluation included demographics and
medical history, clinical examination, 12-lead electrocardiography, estimation of NYHA functional class, and 2-dimentional, M-mode, and Doppler echocardiography. The
modified biplane Simpson’s technique was used to calculate
the EF,14 and LV end-systolic and end-diastolic diameters
were determined following the guidelines of the American
Society of Echocardiography.15 The severity of mitral regurgitation was assessed from color-flow Doppler images
in the apical 4-chamber view. Mitral regurgitation was
classified as mild or grade 1 (jet area/left atrium area
ⱕ20%), moderate or grade 2 (jet area/left atrium area
20% to 40%), or severe or grade 3 (jet area/left atrium
area ⬎40%).16
After implantation, patients returned for regular clinic
visits at 1, 3, and 6 months and every 6 months thereafter.
At each follow-up visit, 12-lead electrocardiography,
NYHA classification, and echocardiographic examinations were repeated. Echocardiographically directed adjustment of the atrioventricular pacing interval was done
before patients were discharged and at follow-up to optimize hemodynamic function. Pharmacologic treatments
were based on clinical evaluations by the attending physicians.
The effects of CRT on patients’ clinical and functional
status were evaluated, comparing the baseline clinical and
echocardiographic parameters with those at 12-month follow-up for surviving patients (otherwise, the last observation was carried forward).
Major cardiac events were considered death from any
cause, urgent heart transplantation, and hospitalization for
worsening HF. In patients implanted with combined devices, appropriate defibrillator therapy for episodes of ventricular fibrillation was also evaluated. Spontaneous arrhythmic episodes detected by the devices were validated by
2 blinded expert electrophysiologists; ventricular fibrillation
was diagnosed if the recorded RR interval was ⱕ240 ms
and accompanied by changes in the morphology of stored
electrocardiograms.
Continuous data are expressed as mean ⫾ SD or medians
and interquartile ranges. Categorical data are expressed as
percentages. Differences between mean data were compared
using Student’s t test for Gaussian variables and by the
Mann-Whitney test or Wilcoxon’s nonparametric test for
non-Gaussian variables for independent or paired samples,
respectively. Differences in proportions were compared using chi-square analysis. The mortality rate was summarized
by the construction of Kaplan-Meier curves, and the distributions of the groups were compared using a log-rank test.
A p value ⬍0.05 was considered significant for all tests. All
statistical analyses were performed using SPSS for Windows version 12.0 (SPSS, Inc., Chicago, Illinois).
Heart Failure/CRT in NYHA II Patients
Figure 1. Kaplan-Meier estimates of time to death from any cause and heart
transplantation (A) and time to death from any cause, heart transplantation,
and appropriate implantable defibrillator therapy for episodes of ventricular
fibrillation (B).
Results
We included data from 952 patients in our analysis. Patients’ baseline symptom classes were distributed as follows: 188 (20%) in NYHA class II, 626 (66%) in NYHA
class III, and 138 (14%) in NYHA class IV. Table 1 lists
demographics, baseline clinical and echocardiographic parameters, and pharmacologic therapy for the entire study
population and for between-group comparisons (NYHA
class II vs NYHA class III or IV).
Patients in NYHA class II were significantly younger,
had a lower prevalence of ischemic causes of HF and
chronic atrial fibrillation, and in general had better clinical
conditions as expressed by a smaller number and shorter
duration of hospitalizations for worsening HF in the 12
months before implantation. EFs were significantly higher,
and the grade of mitral regurgitation was lower in patients in
NYHA class II. Moreover, ␤ blockers were more frequently
prescribed in patients in NYHA class II, whereas the use of
oral anticoagulants was more frequent in patients in class III
or IV. A combined device was more frequently implanted in
patients in NYHA class II than in those in class III or IV.
During a median follow-up period of 16 months (interquartile range 8 to 33), 136 patients died, and 17 underwent
urgent heart transplantation (153 of 952, rate 9.3 per 100
patient-years of follow-up). Of these, 16 of 188 patients
were in NYHA class II at baseline and 137 of 764 in NYHA
class III or IV (rate 5.1 vs 10.3, p ⫽ 0.004). Ninety-three
events were classified as cardiac death; 11 occurred in the
patients in NYHA class II and 82 in those in class III or IV
(rate 3.5 vs 6.2, p ⫽ 0.048). Death from worsening HF
occurred in 60 patients, 4 in the NYHA class II group and
1009
56 in the NYHA class III or IV group (rate 1.3 vs 4.2, p ⫽
0.011). A total of 222 patients had ⱖ1 hospitalization for
worsening HF; 28 were in the NYHA class II group and 194
in the NYHA class III or IV group (rate 8.9 vs 14.6, p ⫽
0.007). In the group of 350 patients implanted with combined devices, 20 patients had episodes of ventricular fibrillation appropriately treated by the defibrillators, 4 of 84 in
the NYHA class II group and 16 of 266 in the NYHA class
III or IV group (5% vs 6% respectively, p ⫽ NS). A major
cardiovascular event occurred in 41 of 188 patients in
NYHA class II, compared with 306 of 764 patients in class
III or IV (rate 13 vs 23, p ⬍0.001). Kaplan-Meier event-free
survival analysis showed that patients in NYHA class II had
significantly lower rates of all-cause mortality and heart
transplantation (log-rank test p ⫽ 0.018; Figure 1) and of
the combined end point of all-cause death, heart transplantation, and appropriate defibrillator therapy (log-rank test
p ⫽ 0.017; Figure 1).
At 12-month follow-up, clinical and echocardiographic
parameters had significantly improved, and hospitalizations
for worsening HF had decreased in patients in NYHA class
II and those in class III or IV at baseline (Table 2). The
percentage of patients who improved in NYHA class status
after 12 months of CRT appeared lower in those in class II
than in those in class III or IV (34% vs 69%, p ⬍0.001;
Figure 2). The proportion of patients in NYHA class II with
symptom progression to NYHA class III was low (9%). A
significant increase in the EF and significant reductions in
LV end-diastolic and end-systolic diameters occurred in
patients in NYHA class II and those in class III or IV at
baseline, whereas mitral regurgitation significantly decreased only in those in class III or IV (Table 2). The
absolute increase in the EF was similar in the patients in
NYHA class II and those in class III or IV (8 ⫾ 9% vs 9 ⫾
11%, p ⫽ NS), as well as the reductions in LV end-diastolic
diameter (⫺3 ⫾ 8 vs ⫺3 ⫾ 8 mm, p ⫽ NS) and end-systolic
diameter (⫺4 ⫾ 10 vs ⫺6 ⫾ 10 mm, p ⫽ NS). The patients
who did not experience major cardiovascular events, those
in NYHA class II (n ⫽ 147) and class III or IV (n ⫽ 458)
at baseline, had significant improvements in LV diameters
after CRT. In contrast, those with major cardiovascular
events during follow-up did not demonstrate improvement
in LV dimensions in the 2 NYHA class groups (Figure 3),
except for a slight reduction in end-systolic diameter in the
NYHA class III or IV group.
During follow-up, LV leads dislodged in 40 patients in
NYHA class III or IV (5%) and in 9 (5%) in class II. There
were 18 pocket erosions (2%), with 10 requiring device
extraction, in the NYHA class III or IV group and 2 (1%),
with 1 requiring device extraction, in the NYHA class II
group.
Discussion
This report, based on a national registry enrolling unselected
patients implanted with the same types of devices, compares
the long-term effects of CRT between patients with systolic
LV dysfunction and NYHA class II HF and those in NYHA
class III or IV. Only a few small uncontrolled studies9,10 and
randomized trials with limited numbers of patients7,8 have
assessed the short-term effects of CRT in patients with
1010
The American Journal of Cardiology (www.AJConline.org)
Table 2
Clinical and echocardiographic parameters at baseline and at 12-month follow-up
Parameter
NYHA Class
II
(n ⫽ 188)
Hospitalizations for HF (n/yr)
QRS duration (ms)
EF (%)
End-diastolic diameter (mm)
End-systolic diameter (mm)
Mitral regurgitation (grade)
III or IV
(n ⫽ 764)
Baseline
Follow-Up
Baseline
Follow-Up
1.1 ⫾ 1.3
163 ⫾ 29
29 ⫾ 7
69 ⫾ 10
57 ⫾ 12
1.9 ⫾ 1.0
0.2 ⫾ 0.5
148 ⫾ 27*
37 ⫾ 11*
65 ⫾ 11*
53 ⫾ 12*
1.8 ⫾ 0.9
1.8 ⫾ 1.4
168 ⫾ 32
27 ⫾ 8
69 ⫾ 9
59 ⫾ 10
2.3 ⫾ 1.0
0.4 ⫾ 0.9†
144 ⫾ 29*
36 ⫾ 11*
65 ⫾ 10*
53 ⫾ 12*
1.9 ⫾ 0.9*
†
* p ⬍0.01 versus baseline (paired Student’s t test).
†
p ⬍0.01 versus baseline (Wilcoxon’s nonparametric test).
Figure 2. Change in NYHA functional class after 12 months of CRT in
patients in NYHA class II and NYHA class III or IV.
Figure 3. Magnitude of LV reverse remodeling after 12 months of CRT in
patients in NYHA class III or IV and those in class II with and without
cardiovascular (CV) events during follow-up. Black bars, baseline measurements; white bars, follow-up measurements. LVEDV ⫽ LV enddiastolic diameter; LVESV ⫽ LV end-systolic diameter. *p ⬍0.05.
mildly symptomatic HF and systolic LV dysfunction. The
actual role of CRT in patients in NYHA class II has yet to
be established by randomized controlled trials such as the
ongoing Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction (REVERSE) study11 and
the Multicenter Automatic Defibrillator Implantation–Car-
diac Resynchronization Therapy (MADIT-CRT) trial.12
However, observational studies, such as the present study,
may provide additional information because the results of
carefully controlled clinical trials are not fully representative of “real-world” practices, and the clinical profile of
patients with HF enrolled in the trials is different from the
characteristics of patients receiving treatment in clinical
practice, in whom a series of factors or co-morbidities that
are under-represented in prospective trials may be present.17
As expected, patients with NYHA class II HF had better
baseline conditions, as evidenced by younger age, a low
prevalence of chronic atrial fibrillation, and a smaller number of hospitalizations for worsening HF in the 12 months
before implantation. However, despite the relatively mild
symptoms, patients in NYHA class II enrolled in this study
already showed severe LV dilatation and impaired LV function. Although it is well known that patients with mild HF
have a better prognosis,18 there are no data on the long-term
outcome of patients in NYHA class II after CRT. In our
study, during a median follow-up period of 16 months of
CRT, a significantly lower percentage of patients in NYHA
class II either died or underwent urgent heart transplantation
or had ⱖ1 hospitalization for worsening HF compared with
those in NYHA class III or IV, whereas the proportion of
patients experiencing ventricular arrhythmias treated by the
implantable defibrillator was similar in the 2 groups. The
Kaplan-Meier analysis showed that patients with NYHA
class II HF had a lower long-term rate of all-cause mortality
and heart transplantation compared with those in NYHA
class III or IV. The survival curves for the combined end
point of all-cause mortality, heart transplantation, and appropriate defibrillator therapy confirmed the lower event
rate, excluding the potential bias due to a reduction in
sudden deaths in the NYHA class II group implanted with a
significantly larger number of combined devices. It is noteworthy that the annual rate of all-cause mortality (5.1%) of
patients in NYHA class II treated with CRT was lower than
that reported for the placebo arm of the NYHA class II
subgroup enrolled in the Sudden Cardiac Death in Heart
Failure Trial (SCD-HeFT),19 although the SCD-HeFT patients were treated with the optimal medical therapy. Indeed, SCD-HeFT reported a 5-year mortality rate of 32%
for the placebo arm of patients in NYHA class II (i.e., 6.4%
per year, assuming a constant event rate during follow-up).
Heart Failure/CRT in NYHA II Patients
Furthermore, in our study, the number of hospitalizations
for worsening HF in patients in NYHA class II was significantly reduced after CRT compared with the 12 months
before implantation. These findings are in favor of a beneficial effect of CRT on disease progression in patients with
mild HF, although a control group of patients in NYHA
class II without CRT is needed to firmly draw this conclusion.
The effects of CRT on LV remodeling and functional
capacity in mildly symptomatic patients with HF are still
controversial. Kuhlkamp et al10 did not find significant
changes in clinical and echocardiographic parameters after
3 months of CRT. However, the follow-up period might
have been too short to detect an improvement in patients
with mild HF. Actually, other studies7–9 demonstrated substantial improvements in LV function after 6 months of
CRT in patients in NYHA class II, in contrast to the minimal improvement in exercise capacity and clinical symptoms. In the present study, CRT induced significant improvements in functional class, EFs, and LV dimensions in
patients in NYHA class II and those in class III or IV after
12 months, indicating that CRT promotes long-lasting reverse remodeling, even in patients with less symptomatic
HF. The magnitudes of the improvements in EFs and LV
dimensions were similar in patients in NYHA class II and
those in class III or IV, whereas fewer patients in NYHA
class II than in class III or IV improved in functional class.
Not surprisingly, given the minimal symptoms at baseline,
NYHA class status remained unchanged in 57% of patients
in NYHA class II. In contrast, the percentage of patients
exhibiting symptom progression to NYHA class III was low
(9%). Bleeker et al9 reported similar results in a few patients
followed for 6 months after CRT. In addition, they demonstrated a decrease in LV dyssynchrony with color-coded
tissue Doppler imaging shortly after CRT in patients with
mild HF, indicating resynchronization of LV contraction.
This acute resynchronization was associated with an improvement in clinical and echocardiographic parameters at 6
months. It is noteworthy that LV reverse remodeling was
not restricted to patients who improved in NYHA class but
also occurred in those without improvement in functional
class. LV function did not improve in the patients with
symptom progression to NYHA class III. These observations are in line with those of our study and point out the
minimal benefit to clinical symptoms in contrast to the
considerable LV reverse remodeling of CRT in patients
with mild HF.
We did not observe major device-related complications;
infections, most commonly restricted to the device pockets,
were relatively rare in the 2 groups. Thus, the risk/benefit
ratio of CRT appears acceptable also for patients with
mildly symptomatic HF.
This study has the limitations of all multicenter observational studies, such as potential bias in patient selection and
the lack of a control group. Patients in NYHA class II were
enrolled on the basis of preliminary evidence in favor of
beneficial effect of CRT in subjects with mild HF symptoms,7,8 before the availability of specific guidelines; however, all these patients had impaired LV function and experienced frequent hospitalization for worsening HF. Finally,
the pharmacologic therapy at enrollment was not optimal,
1011
especially concerning ␤ blockers and angiotensin-converting enzyme inhibitors. This observation indicates that the
adoption of guidelines for the management of HF in normal
clinical practice is slow.20,21 Nevertheless, pharmacologic
treatment was stable throughout the follow-up and did not
present differences between the NYHA class II group and
the class III or IV group that could have affected our results.
Acknowledgment: We would like to thank Tiziana De
Santo and Paola Di Stefano (Clinical Service Team,
Medtronic Italia) for their careful statistical analysis of the
data, and Jane Moore for her assistance in editing this
report.
Appendix
The centers and investigators participating in the InSync/
InSync ICD Italian Registry are as follows: M. Gasparini, P.
Galimberti, F. Regoli, E. Gronda, Istituto Clinico Humanitas IRCCS, Rozzano, Italy; M. Lunati, G. Cattafi, G. Magenta, M. Paolucci, R. Vecchi, Niguarda Cà Granda Hospital, Milan, Italy; M. Santini, R. Ricci, San Filippo Neri
Hospital, Rome, Italy; F. Gaita, M. Bocchiardo, P. DiDonna, D. Caponi, Civile Hospital, Asti, Italy; L. Tavazzi,
M. Landolina, F. Frattini, R. Rordorf, C. Belvito, S. Savastano, Policlinico S. Matteo IRCCS, Pavia, Italy; L.
Padeletti, P. Pieragnoli, Careggi Hospital, Florence, Italy;
A. Vincenti, S. DeCeglia, A. Cirò, S. Gerardo Dei Tintori
Hospital, Monza, Italy; A. Curnis, G. Mascioli, Spedali
Civili Hospital, Brescia, Italy; A. Puglisi, S. Bianchi, C.
Peraldo, Fatebenefratelli Hospital, Rome, Italy; M. Sassara,
A. Achilli, F. Turreni, P. Rossi, Belcolle Hospital, Viterbo,
Italy; G.B. Perego, S. Luca Auxologico Hospital, Milan,
Italy; P.A. Ravazzi, P. Diotallevi, S.S. Antonio e Biagio e
Cesare Arrigo Hospital, Alessandria, Italy; M. Tritto, Mater
Domini Hospital, Castellanza, Italy; A. Carboni, D. Ardissino, G. Gonzi, V. Serra, Civile Hospital, Parma, Italy; G.
Vergara, S. Maria Del Carmine Hospital, Rovereto, Italy; G.
Boriani, M. Biffi, C. Martignani, L. Frabetti, S. Orsola, M.
Malpighi Hospital, Bologna, Italy; G. Luzzi, Policlinico
Consorziale Hospital, Bari, Italy; F. Laurenzi, S. Camillo
Hospital, Rome, Italy; G. Pistis, Mauriziano Hospital,
Turin, Italy; A. Cesario, G.B. Grassi Hospital, Ostia, Italy;
G. Zanotto, Civile Hospital, Verona, Italy; S. Orazi, S.
Camillo Hospital, Rieti, Italy; R. Ometto, C. Bonanno, S.
Bortolo Hospital, Vicenza, Italy; G. Molon, E. Barbieri, S.
Cuore Don Calabria Hospital, Negrar, Italy; A. Raviele, G.
Gasparini, Umberto I Hospital, Mestre, Italy; G. Botto, M.
Luzi, A. Sagone, S. Anna Hospital, Como, Italy; A. Vado,
S. Croce e Carle Hospital, Cuneo, Italy; A. Montenero,
Multimedica Hospital, Sesto S. Giovanni, Italy; G. Inama,
Maggiore Hospital, Crema, Italy; B. Sassone, Civile Hospital, Bentivoglio, Italy; M. Briedda, F. Zardo, S. Maria
degli Angeli Hospital, Pordenone, Italy; E. Bertaglia,
ULSS13 Hospital, Mirano, Italy; A. Proclemer, S. Maria
della Misericordia Hospital, Udine, Italy; F. Zanon, Civile
Hospital Rovigo, Rovigo, Italy; M. Disertori, L. Gramegna,
M. DelGreco, D. Dallafior, S. Chiara Hospital, Trento, Italy;
C. Tomasi, A. Maresta, M. Piancastelli, S. Maria delle Croci
Hospital, Ravenna, Italy; A. Bridda, S. Martino Hospital,
1012
The American Journal of Cardiology (www.AJConline.org)
Belluno, Italy; R. Mantovan, Cà Foncello Hospital, Treviso,
Italy; A. Fusco, A. Vicentini, Polispecialistica Pederzoli
Hospital, Peschiera del Garda, Italy; P. Baraldi, S. Agostino
Hospital, Modena, Italy; G. Lonardi, Civile Hospital, Legnago, Italy; W. Rahue, S. Maurizio Hospital, Bolzano, Italy;
P. Delise, Civile Hospital Conegliano, Conegliano, Italy; C.
Menozzi, S. Maria Nuova, Reggio Emilia, Italy; P. Babudri,
Borgo Roma Hospital, Verona, Italy; R. Marconi, C. eG.
Mazzoni Hospital, Ascoli Piceno, Italy; G. De Fabrizio, F.
Alfano, G. Moscati, Avellino, Italy; G. Barbato, Maggiore
Hospital, Bologna, Italy; P. Gelmini, Civile Hospital
Desenzano del Garda, Garda, Italy; M. DiSabato, S.
Leopoldo Mandic Hospital, Merate, Italy; S. Ricci,
Ramazzini Hospital, Carpi, Italy; M.D. Aulerio, S. Biagio
Hospital, Domodossola, Italy; G.L. Morgagni, R. Latini,
Civile Hospital Macerata, Macerata, Italy; G. Bardelli, Fornaroli Hospital, Magenta, Italy; R. Paulichl, F. Tappeiner
Hospital, Merano, Italy; M. Bernasconi, M. Marzegalli, S.
Carlo Borromeo, Milan, Italy; G. Neri, Montebelluna Hospital, Treviso, Italy; E. Occhetta, Hospital Maggiore Della
Carità, Novara, Italy; P. Bocconcelli, S. Salvatore Hospital,
Pesaro, Italy; A. Capucci, Civile Hospital, Piacenza, Italy;
A. Campana, S. Giovanni di Dio e Ruggi d’Aragona Hospital, Salerno, Italy; N. Dibelardino, Civile Hospital, Velletri, Italy; and A. Vaglio, Giovanni e Paolo Hospital,
Venice, Italy.
1. Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats
TG, Jessup M, Konstam MA, Mancini DM, Michl K, et al; American
College of Cardiology; American Heart Association Task Force on
Practice Guidelines; American College of Chest Physicians; International Society for Heart and Lung Transplantation; Heart Rhythm
Society. ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: summary article: 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).
Circulation 2005;112:1116 –1143.
2. Swedberg K, Cleland J, Dargie H, Drexler H, Follath F, Komajda M,
Tavazzi L, Smiseth OA, Gavazzi A, Haverich A, et al; 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;26:1115–1140.
3. Cazeau S, Leclercq C, Lavergne T, Walker S, Varma C, Linde C,
Garrigue S, Kappenberger L, Haywood GA, Santini M, et al; Multisite
Stimulation in Cardiomyopathies (MUSTIC) Study Investigators. Effects of multisite biventricular pacing in patients with heart failure and
intraventricular conduction delay. N Engl J Med 2001;344:873– 880.
4. Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E,
Kocovic DZ, Packer M, Clavell AL, Hayes DL, et al; MIRACLE
Study Group. Multicenter InSync Randomized Clinical Evaluation.
Cardiac resynchronization in chronic heart failure. N Engl J Med
2002;346:1845–1853.
5. Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De
Marco T, Carson P, DiCarlo L, DeMets D, White BG, et al;
Comparison of Medical Therapy, Pacing, and Defibrillation in
Heart Failure (COMPANION) Investigators. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140 –2150.
6. Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, Tavazzi L; Cardiac Resynchronization-Heart 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.
7. Abraham WT, Young JB, Leon AR, Adler S, Bank AJ, Hall SA,
Lieberman R, Liem LB, O’Connell JB, Schroeder JS, Wheelan KR;
Multicenter InSync ICD II Study Group. Effects of cardiac resynchronization on disease progression in patients with left ventricular systolic
dysfunction, an indication for an implantable cardioverter defibrillator,
and mildly symptomatic chronic heart failure. Circulation 2004;110:
2864 –2868.
8. Higgins SL, Hummel JD, Niazi IK, Giudici MC, Worley SJ, Saxon
LA, Boehmer JP, Higginbotham MB, De Marco T, Foster E, Yong PG.
Cardiac resynchronization therapy for the treatment of heart failure in
patients with intraventricular conduction delay and malignant ventricular tachyarrhythmias. J Am Coll Cardiol 2003;42:1454 –1459.
9. Bleeker GB, Schalij MJ, Holman ER, Steendijk P, van der Wall EE,
Bax JJ. Cardiac resynchronization therapy in patients with systolic left
ventricular dysfunction and symptoms of mild heart failure secondary
to ischemic or nonischemic cardiomyopathy. Am J Cardiol 2006;98:
230 –235.
10. Kuhlkamp V; InSync 7272 ICD World Wide Investigators. Initial experience with an implantable cardioverter-defibrillator incorporating cardiac
resynchronization therapy. J Am Coll Cardiol 2002;39:790 –797.
11. Linde C, Gold M, Abraham WT, Daubert JC; REVERSE Study Group.
Rationale and design of a randomized controlled trial to assess the
safety and efficacy of cardiac resynchronization therapy in patients
with asymptomatic left ventricular dysfunction with previous symptoms or mild heart failure—the Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction (REVERSE) study. Am
Heart J 2006;151:288 –294.
12. Moss AJ, Brown MW, Cannom DS, Daubert JP, Estes M, Foster E,
Greenberg HM, Hall WJ, Higgins SL, Klein H, et al. Multicenter
Automatic Defibrillator Implantation Trial-Cardiac Resynchronization
Therapy (MADIT-CRT): design and clinical protocol. Ann Noninvas
Electrocardiol 2005;10(suppl):34 – 43.
13. Bax JJ, Abraham T, Barold SS, Breithardt OA, Fung JW, Garrigue S,
Gorcsan J III, Hayes DL, Kass DA, Knuuti J, et al. Cardiac resynchronization therapy: part 2—issues during and after device implantation and unresolved questions. J Am Coll Cardiol 2005;46:
2168 –2182.
14. Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gutgesell H, Reichek N, Sahn D, Schnittger I. Recommendations for quantification of the left ventricle by two-dimensional
echocardiography. J Am Soc Echocardiogr 1989;2:358 –367.
15. Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of
echocardiographic measurements. Circulation 1978;58:1072–1083.
16. Helmcke F, Nanda NC, Hsiung MC, Soto B, Adey CK, Goyal RG,
Gatewood RP Jr. Color Doppler assessment of mitral regurgitation
with orthogonal planes. Circulation 1987;75:175–183.
17. Tavazzi L. Ventricular pacing: a promising new therapeutic strategy in
heart failure. For whom? Eur Heart J 2000;21:1211–1214.
18. Baldasseroni S, Opasich C, Gorini M, Lucci D, Marchionni N, Marini
M, Campana C, Perini G, Deorsola A, Masotti G, et al; Italian Network
on Congestive Heart Failure Investigators. 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.
19. Bardy GH, Lee KL, Mark DB, Poole JE, Packer DL, Boineau R,
Domanski M, Troutman C, Anderson J, Johnson G, et al; Sudden
Cardiac Death in Heart Failure Trial (SCD-HeFT) Investigators. Amiodarone or an implantable cardioverter-defibrillator for congestive
heart failure. N Engl J Med 2005;352:225–237.
20. Di Lenarda A, Scherillo M, Maggioni AP, Acquarone N, Ambrosio
GB, Annicchiarico M, Bellis P, Bellotti P, De Maria R, Lavecchia R,
et al; TEMISTOCLE Investigators. Current presentation and management of heart failure in cardiology and internal medicine hospital units:
a tale of two worlds—the TEMISTOCLE study. Am Heart J 2003;
146:E12.
21. Maggioni AP, Sinagra G, Opasich C, Geraci E, Gorini M, Gronda E,
Lucci D, Tognoni G, Balli E, avazzi L; Beta Blockers in Patients With
Congestive Heart Failure: Guided Use in Clinical Practice Investigators. Treatment of chronic heart failure with beta adrenergic blockade
beyond controlled clinical trials: the BRING-UP experience. Heart
2003;89:299 –305.
Frequency, Determinants, and Clinical Relevance of Acute
Coronary Syndrome-Like Electrocardiographic Findings in Patients
With Acute Aortic Syndrome
Elena Biagini, MDa, Carla Lofiego, MDa, Marinella Ferlito, MDa, Rossella Fattori, MDb,
Guido Rocchi, MDa, Maddalena Graziosi, MDa, Luigi Lovato, MDb, Lara di Diodoro, MDa,
Robin M.T. Cooke, MAa, Elisabetta Petracci, MStata, Letizia Bacchi-Reggiani, MStata,
Romano Zannoli, MSca, Angelo Branzi, MDa, and Claudio Rapezzi, MDa,*
We investigated frequency/characteristics of acute coronary syndrome-like (ACS-like)
electrocardiographic (ECG) profiles among patients with a final diagnosis of acute aortic
syndrome (AAS), and explored pathophysiologic determinants and prognostic relevance
within each Stanford subtype. We blindly reviewed presentation electrocardiograms of 233
consecutive patients with final diagnosis of AAS (164 Stanford type A) at a regional
treatment center. Prevalence of ACS-like ECG findings was 27% (type A, 26%, type B,
29%); most were non–ST-elevation myocardial infarction-like. Patients with ACS-like
ECG findings more often had coronary ostia involvement (p ⴝ 0.002), pleural effusion (p ⴝ
0.02), significant aortic regurgitation (p ⴝ 0.01), and troponin positivity (p ⴝ 0.001).
ACS-like ECG profile in type A disease was independently associated with coronary ostia
involvement (odds ratio [OR] 5.27, 95% confidence interval [CI] 1.75 to 15.88). ACS-like
ECG profile predicted in-hospital mortality (OR 2.90, 95% CI 1.24 to 6.12), as did age (each
incremental 10-year: OR 1.59, 95% CI 1.14 to 2.22), and syncope at presentation (OR 2.90,
95% CI 1.16 to 7.24). In conclusion, about 25% of our AAS patients (in either Stanford
subtype) presented ACS-like ECG patterns— often with non–ST-elevation myocardial
infarction characteristics—which could cause misdiagnosis. ACS-like ECG profile was
associated with more complicated disease, and in type A disease was a strong independent
predictor of in-hospital mortality. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol
2007;100:1013–1019)
Acute aortic syndrome (AAS) is a life-threatening cardiovascular emergency requiring rapid recognition and, when indicated, prompt therapeutic intervention.1–7 Acute coronary syndrome (ACS) often masks AAS at presentation.8 –11 Clinical/
instrumental similarities between AAS and “classic” ACS can
lead to dramatic consequences. In a patient with AAS, prompt
use of antithrombotics and coronary angioplasty (in line with
present for ACS) will inevitably postpone urgent surgery, and
can lead to catastrophic intraoperative complications such as
aortic wall rupture or major bleeding.12–15 Although studies
based on the International Registry of Acute Aortic Dissection
(IRAD) provide important data on the frequency of various
electrocardiographic (ECG) abnormalities at presentation in
patients harboring AAS,1,3– 6 relatively little is known about the
frequencies, determinants, and clinical relevance of non–STelevation myocardial infarction (STEMI) and non-STEMI
a
Institute of Cardiology and bCardiovascular Radiology Unit, Cardiothoracic Department, University of Bologna and S.Orsola-Malpighi Hospital, Bologna, Italy. Manuscript received March 6, 2007; revised manuscript received and accepted April 12, 2007.
The study was partially supported by the Fanti Melloni Foundation,
Bologna, Italy. Dr Biagini was supported by a grant from Centro per la
Lotta contro l’Infarto–Fondazione Onlus, Rome, Italy.
*Corresponding author: Tel: 39-051-349858; fax: 39-051-344859.
E-mail address: [email protected] (C. Rapezzi).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.044
(NSTEMI)-like patterns.16 –18 We assessed the frequency and
characteristics of ACS-like ECG profiles at presentation, and
explored their possible determinants and prognostic relevance.
Methods
Setting and study design: We studied all patients with a
final diagnosis of spontaneous AAS admitted to our center
from 1990 to June 2005. Because this is the only regional
center providing treatment facilities for AAS, referral of
patients diagnosed with AAS from surrounding hospitals is
automatic. Study design was cross sectional, but with a
sequential clinical outcome measure (in-hospital mortality,
as recorded in the centralized hospital database) providing
relevant prognostic information. We assessed prevalence of
ACS-like ECG findings at presentation in patients with a
final diagnosis of AAS. Possible determinants of ACS-like
ECG findings were investigated by exploring clinical, anatomic, and functional associations. The main eligibility criterion was final diagnosis of AAS (as confirmed at surgery,
autopsy, or imaging) recorded in the institutional database.
AAS comprised both “classic” aortic dissection and intramural hematoma, defined and categorized according to the
Stanford classification. Data collection was based on institutional database records and detailed reviews of patients’
original presentation electrocardiograms, transesophageal
and transthoracic echocardiographic recordings, computed
www.AJConline.org
1014
The American Journal of Cardiology (www.AJConline.org)
Table 1
Patients’ diagnostic characteristics (including availability of imaging examinations) according to type of acute aortic syndrome
Variable
Overall
(n ⫽ 233)
Type A
(n ⫽ 164)
Type B
(n ⫽ 69)
p Value
(type A vs type B)
Age, mean ⫾ SD (yrs)
Men
Intramural hematoma
History of hypertension
Antihypertensive therapy
Marfan syndrome
Bicuspid aortic valve
Aortic coarctation
Pregnancy
Diabetes
Coronary artery disease
Atrial fibrillation
Previous stroke
Known thoracic aortic aneurysm
Previous ascending aorta and/or aortic valve surgery
Systolic blood pressure, mean ⫾ SD (mm Hg)
Systolic blood pressure ⬎160 mm Hg
Systolic blood pressure ⱕ90 mm Hg
Chest pain
Back pain
Abdominal pain
Migratory pain
Pain in other sites
Dyspnea
Syncope
Paraplegia
Shock within first 12 h
Cardiac tamponade
Pericardial effusion
Pleural effusion
Periaortic effusion
Moderate/severe aortic regurgitation
Imaging examinations
Transthoracic echocardiography
Transesophageal echocardiography
Computed tomography
Magnetic resonance imaging
Angiography
61 ⫾ 13
156 (67%)
28 (12%)
155 (66%)
137 (59%)
12 (5%)
4 (2%)
1 (0.4%)
1 (0.4%)
2 (1%)
12 (5%)
13 (5%)
8 (3%)
9 (4%)
14 (6%)
138 ⫾ 39
80 (34%)
38 (16%)
161 (69%)
92 (39%)
67 (29%)
48 (21%)
21 (9%)
24 (10%)
27 (11%)
5 (2%)
35 (15%)
31 (13%)
90 (39%)
46 (20%)
39 (17%)
59 (25%)
63 ⫾ 12
105 (64%)
19 (12%)
105 (70%)
93 (57%)
5 (3%)
2 (1%)
0
1 (1%)
0
6 (4%)
9 (5%)
7 (4%)
7 (4%)
8 (5%)
127 ⫾ 36
39 (24%)
33 (20%)
129 (79%)
50 (30%)
42 (26%)
28 (17%)
15 (9%)
19 (12%)
27 (16%)
2 (1%)
32 (19%)
31 (19%)
77 (47%)
18 (11%)
15 (9%)
57 (35%)
59 ⫾ 14
51 (74%)
9 (13%)
50 (72%)
44 (64%)
7 (10%)
2 (3%)
1 (1%)
0
2 (3%)
6 (9%)
4 (6%)
1 (1%)
2 (3%)
6 (9%)
167 ⫾ 34
41 (59%)
5 (7%)
32 (46%)
42 (61%)
25 (36%)
20 (29%)
6 (9%)
5 (7%)
0
3 (4%)
3 (4%)
0
13 (19%)
28 (41%)
24 (35%)
2 (3%)
0.04
0.14
0.76
0.21
0.32
0.02
0.37
0.12
0.52
0.03
0.11
0.93
0.28
0.62
0.26
⬍0.001
⬍0.001
0.02
⬍0.001
⬍0.001
0.10
0.04
0.91
0.32
⬍0.001
0.13
⬍0.01
⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001
170 (73%)
177 (76%)
160 (69%)
44 (19%)
73 (31%)
118 (72%)
134 (82%)
105 (64%)
10 (6%)
43 (26%)
52 (75%)
43 (62%)
55 (80%)
34 (49%)
30 (43%)
—
—
—
—
—
tomographic scans, and magnetic resonance imaging examinations. The presentation electrocardiogram (i.e., the first recording after onset) was retrieved for each patient and blindly
reviewed by 3 experienced cardiologists (MF, CL, and CR),
who followed standard diagnostic criteria (differences were
settled by group discussion). All available echocardiographic
recordings were blindly reviewed by a team of 3 expert echocardiographers (MF, EB, and CR), who noted the anatomic
details of the AAS, and identified cases of coronary ostia
involvement of the dissection/hematoma by assessment of the
dynamic spatial relations between the ostia and the intimal
flap (or intramural hematoma). Computed tomographic
scans and magnetic resonance imaging were reviewed by 2
cardiovascular radiologists (RF and LL) to define anatomic
details. In a subgroup of patients, data were available on
cardiac troponin I testing at hospital admission, performed
according to the standard protocol used in chest pain units
(blood samples taken at presentation, and after 8 hours or
until a correct diagnosis of aortic dissection was reached).
For patients with ACS-like electrocardiograms at presenta-
tion, subsequent in-hospital electrocardiograms were retrieved whenever possible to establish evolution of the ECG
pattern (in terms of persistence/disappearance of ACS-like
pattern and appearance of Q waves). The study was conducted in accordance with the guiding principles of the
Declaration of Helsinki and all patients provided written
informed consent for anonymous data publication. Because
application for formal approval of this observational study
from an Ethical Committee would not have been deemed
appropriate in our setting, ethical considerations were discussed at our Institute’s internal investigational review
panel. No potential issue was identified.
Definitions: Following present guidelines,19 –24 ECG
finding was considered to be ACS-like if it presented ⱖ1 of
the following characteristics in ⱖ2 contiguous leads: STsegment elevation ⱖ0.1 mV; ST-segment depression ⱖ0.1
mV; T-wave inversion ⱖ0.2 mV. (Electrocardiograms with
ST-segment deviation ⬍0.1 mV or T-wave inversion ⬍0.2
mV were classified as nonspecific minor repolarization
Miscellaneous/ACS-Like Electrocardiography in AAS
1015
Table 2
Electrocardiographic characteristics at presentation
Variable
Normal electrocardiogram
Sinus rhythm
Atrial fibrillation
Other rhythms
Left ventricular hypertrophy
Right bundle branch block
Left bundle branch block
Left anterior hemiblock
Low voltages
Old Q waves
Nonspecific ST-T abnormalities
ACS-like profile
STEMI
Non-STEMI
2 ST
2 ST ⫹ T wave abnormalities
T-waves ⱖ2 mm in ⱖ2 contiguous leads
Overall Population
(n ⫽ 233)
Type A
(n ⫽ 164)
Type B
(n ⫽ 69)
p Value
44 (19%)
214 (92%)
13 (5%)
6 (2%)
72 (31%)
13 (5%)
5 (2%)
21 (9%)
10 (4%)
12 (5%)
145 (62%)
62 (27%)
10 (4%)
52 (22%)
5 (2%)
17 (7%)
30 (13%)
31 (19%)
149 (91%)
9 (5%)
6 (4%)
47 (29%)
11 (7%)
4 (2%)
14 (8%)
9 (5%)
8 (5%)
96 (59%)
42 (26%)
7 (4%)
35 (21%)
2 (1%)
11 (7%)
22 (13%)
13 (19%)
65 (94%)
4 (6%)
0 (0%)
25 (36%)
2 (3%)
0 (0%)
7 (10%)
1 (1%)
4 (6%)
49 (71%)
20 (29%)
3 (4%)
17 (25%)
3 (4%)
6 (9%)
8 (12%)
1
0.55
1
0.25
0.32
0.40
0.45
0.89
0.30
1
0.07
0.71
0.98
0.58
0.13
0.59
0.71
changes.) ACS-like ECG findings with ST elevation were
classified as ST-elevation myocardial infarction-like, or
NSTEMI-like. Old Q waves, bundle branch blocks, hemiblocks, and low QRS voltage were defined according to
present standard criteria. Left ventricular hypertrophy was
defined using the Sokolow-Lyon voltage definition.25 Repolarization abnormalities in lateral leads were considered consistent with the presence of typical strain when there was a
downsloping convex ST segment with an inverted asymmetrical T wave with polarity opposite the main QRS deflection.26
We used standard definitions of shock, cardiac tamponade,
pleural effusion, and pericardial effusion.27–29 Severe and moderate-to-severe aortic regurgitation at transthoracic/transesophageal echocardiography was considered hemodynamically significant. Periaortic hematoma was diagnosed by transthoracic/
transesophageal echocardiography, computed tomographic
scan, or magnetic resonance imaging scan.6 Ascertainment of
new myocardial infarction (as a complication of AAS and/or
surgical treatment of AAS) was based on instrumental/laboratory criteria (typical troponin and/or creatine phosphokinase-MB change, or appearance of new pathologic Q waves in
the absence of alternative explanations).27
Statistical analysis: Descriptive statistics are reported as
mean ⫾ SD medians (interquartile range) for skewed distributions or numbers (percentages). Based on the very
different prognosis (and treatment strategies) generally associated with the 2 Stanford subtypes, type A and type B
patients were considered separately throughout. Group
comparisons were performed by Student’s t test for unpaired data, or Mann-Whitney U test, as appropriate. Chisquare analysis was used for nominal data. p Values ⬍0.05
were considered statistically significant. Multivariable logistic regression analyses were performed to identify independent predictors of (1) ACS-like ECG findings at presentation and (2) in-hospital mortality. Variables considered at
univariate analysis were tested, using a forward stepwise
inclusion procedure with backward elimination. Model discrimination was assessed with the c-statistic, and model
calibration was assessed with the Hosmer-Lemeshow statistic. To assess linearity, we categorized continuous variables as intervals and performed the score test for trend of
odds on the proportions of death at each interval, using
STATA 9.0. All other statistical analyses were performed
using SPSS for Windows, release 13.0 (SPSS Inc., Chicago,
Illinois).
Results
Patients: During the study period, 233 patients received
a final diagnosis of AAS, which was classified as Stanford
type A in 164 (70%) patients. Twenty-eight (12%) patients
(19 of 164 [12%] with Stanford type A, and 9 of 69 [13%]
with Stanford type B) had intramural hematomas. All 233
patients had an available admission electrocardiogram plus
ⱖ1 imaging examination (echocardiography, computed tomographic scan, or magnetic resonance imaging scan). Table 1 lists the patients’ diagnostic characteristics (including
availability of imaging examinations) by Stanford subtype.
Surgical interventions were performed in 126 (77%) type A
patients (in-hospital mortality, 21%; n ⫽ 26). In the remaining 38 type A patients (in-hospital mortality, 61%; n ⫽ 23),
surgery was not performed due to advanced age, co-morbidity, patient refusal, or death. Surgical or interventional
treatments were performed in 18 (26%) type B patients, due
to thoracic–abdominal malperfusion or impending rupture
of the false lumen. In-hospital mortality was 17% (3 of 18)
among type B patients who received surgical/interventional
treatments, 12% (6 of 51) among the other patients.
ACS-like ECG patterns: Table 2 lists ECG findings at
presentation by Stanford subtype. Within each Stanford
subtype, ⬎25% of the patients had ACS-like ECG patterns,
most often displaying NSTEMI-like characteristics. Remarkably, most (61%) of the NSTEMI-like patterns were
exclusively characterized by negative T waves. Interestingly, of the 10 STEMI-like patterns, 3 (30%) were recorded in patients with type B disease. Furthermore, 2
1016
The American Journal of Cardiology (www.AJConline.org)
Table 3
Comparisons of patients with and without ACS-like electrocardiography
Variable
Type A
ACS-like
(n ⫽ 42)
Age, mean ⫾ SD (yrs)
Men
Intramural hematoma, n (%)
Hours from symptoms to final diagnosis, median
(interquartile range)
History of high blood pressure
Coronary artery disease
Systolic blood pressure, mean ⫾ SD (mm Hg)
Systolic blood pressure ⬎160 mm Hg
Pericardial effusion
Pleural effusion
Periaortic effusion
Significant aortic regurgitation
Shock
Coronary ostia involvement
Troponin positivity, n/N
Type B
Not ACS-like
(n ⫽ 122)
p Value
ACS-like
(n ⫽ 20)
Not ACS-like
(n ⫽ 49)
63 ⫾ 12
29 (69%)
4 (9%)
5.6 (10.1)
63 ⫾ 13
77 (63%)
15 (12%)
5.5 (15.0)
0.926
0.488
0.628
0.933
62 ⫾ 12
14 (70%)
2 (9%)
14.0 (127.3)
57 ⫾ 15
36 (73%)
7 (14%)
5.7 (24.6)
0.21
0.77
0.63
0.52
26 (63%)
1 (2%)
131 ⫾ 33
8 (19%)
18 (43%)
5 (12%)
3 (7%)
21 (51%)
11 (27%)
9 (22%)
11/25 (44%)
79 (65%)
5 (4%)
144 ⫾ 40
30 (25%)
59 (48%)
13 (11%)
12 (10%)
36 (30%)
21 (17%)
6 (5%)
5/39 (13%)
0.740
0.609
0.270
0.463
0.538
0.823
0.602
0.016
0.205
0.001
0.001
14 (67%)
3 (14%)
184 ⫾ 32
16 (76%)
5 (25%)
14 (67%)
6 (28%)
2 (9%)
2 (9%)
0 (0%)
5/9 (55%)
36 (73%)
3 (6%)
167 ⫾ 35
26 (53%)
8 (16%)
15 (31%)
18 (37%)
0 (0%)
1 (2%)
0 (0%)
0/10 (0%)
0.77
0.24
0.06
0.04
0.40
⬍0.01
0.59
0.03
0.14
0.03
occurred in patients with intramural hematoma (1 type A, 1
type B).
Regarding evolution of ACS-like patterns, among the 30
patients with NSTEMI-like patterns due to isolated negative T
waves, 8 continued to display the alterations without developing true myocardial infarction, 5 showed resolution of abnormalities (evolution was not verifiable in 17). Among the 22
patients with NSTEMI-like patterns due to downsloping ST
with or without negative T waves, 8 had a true myocardial
infarction in hospital, 2 continued to display the alterations
without developing true myocardial infarction, and the abnormalities disappeared in 5 (evolution not verifiable in 7). Regarding the 10 patients with STEMI-like patterns at presentation, 2 had true myocardial infarction, the abnormalities
disappeared in 2 (in the remaining 6, the evolution could not be
verified either due to death or lack of troponin data).
Morphologic, clinical, and functional characteristics
accompanying ACS-like ECG patterns: Table 3 lists
characteristics of patients with or without ACS-like ECG
profiles in the overall population and by Stanford subtype. For type A, coronary ostia involvement (Figure 1)
was about 4 times more common in patients with ACSlike ECG findings. For type B, pleural effusion was about
twice as common in patients with ACS-like ECG patterns, who more often had severe hypertension at first
presentation (of note, the hemodynamically significant
aortic regurgitation recorded in 2 patients with ACS-like
ECG findings was preexistent). Troponin positivity was
more often recorded in patients with ACS-like ECG patterns (irrespective of Stanford subtype) within the subset
of patients with available test results. Based on number
considerations, multivariate analysis was performed only
for type A disease (only coronary ostia involvement,
odds ratio [OR] 5.27, 95% confidence interval [CI] 1.75
to 15.88, p ⫽ 0.003, entered the model).
Remarkably, 8 patients with ACS-like ECG findings (6
type A, 2 type B) initially received either intensive anti-
p Value
thrombotic therapy with glycoprotein IIb/IIIa inhibitors
(n ⫽ 6) or thrombolytics (n ⫽ 2). An additional 4 patients
(3 type A, 1 type B) who showed repolarization abnormalities, but did not fulfill guideline criteria for ACS-like ECG
patterns, also received thrombolytics.
ACS-like ECG findings and in-hospital mortality:
Table 4 reports the results of univariate and multivariate analysis of ACS-like ECG findings and other risk factors for
in-hospital mortality among patients with Stanford type A. In
addition to age and dyspnea at presentation, the ACS-like ECG
profile was an independent predictor of in-hospital mortality
(OR 2.76, 95% CI 1.24 to 6.12, p ⫽ 0.01). The predictive
accuracy of the model correlated with the observed events
(77.4% of the correct classification; c-statistics, 0.623; Hosmer-Lemeshow goodness-of-fit, p ⫽ 0.15). Of note, when we
performed a supplemental univariate analysis of the separate
prognostic contributions of the 2 types of ACS-like pattern,
neither type of pattern reached statistical significance
(NSTEMI-like, OR 1.90, 95% CI 0.87 to 4.17; p ⫽ 0.109;
STEMI-like, OR 2.47, 95% CI 0.59 to 10.29; p ⫽ 0.22).
Regarding therapy, the role of inappropriate intensive antithrombotics or thrombolytics did not reach significance at
univariate analysis (possibly due to number considerations). Of
note, because survival is known to be strictly associated with
surgical treatment, we decided to construct a supplemental
model that included this variable. In addition to absence of
surgical treatment (OR 5.44, 95% CI 2.40 to 12.32; p ⬍0.001),
other predictors were ACS-like ECG findings (OR 2.43, 95%
CI 1.05 to 5.59, p ⫽ 0.04), dyspnea at presentation (OR 3.08,
95% CI 1.08 to 8.76, p ⫽ 0.04), and syncope (OR 2.95, 95%
CI 1.13 to 7.70, p ⫽ 0.03).
Within the Stanford type B subset, the ACS-like ECG
profile did not reach significance at univariate analysis (OR
1.26, 95% CI 0.28 to 5.64, p ⫽ 0.76). The only variable that
reached significance at univariate analysis of this small
population was migratory pain (OR 6.57, 95% CI 1.45 to
Miscellaneous/ACS-Like Electrocardiography in AAS
1017
Figure 1. Examples of different substrates for myocardial infarction/ischemia in type A aortic dissection patients. (A) Flap extending into the right coronary
ostium (arrow) in a patient with inferior and right ventricular STEMI. (B) Flap obstructing the left coronary ostium in the diastolic phase (arrow),
accompanied by ECG signs of inferolateral subendocardial ischemia suggesting NSTEMI. FL ⫽ false lumen; LA ⫽ left atrium; LVOT ⫽ left ventricular
outflow tract; TL ⫽ true lumen.
Table 4
Univariate and multivariate analysis of predictors of in-hospital mortality among patients with Stanford type A disease
Variable
Age (each incremental 10-yr)
Men
History of systemic hypertension
Systolic blood pressure ⱖ160 mm Hg
Abdominal pain
Migratory pain
Neurologic symptoms
Dyspnea at presentation
Syncope
ACS-like profile
Cardiac tamponade
Shock
Pericardial effusion
Pleural effusion
Periaortic effusion
Aortic insufficiency
Onset of symptoms to final diagnosis (each incremental h)
Intramural hematoma
Intensive antithrombotic therapy/thrombolytics
Univariate
OR (95% CI)
p Value
Multivariate
OR (95% CI)
p Value
1.65 (1.21–2.27)
0.66 (0.33–1.31)
1.23 (0.61–2.50)
1.24 (0.57–2.67)
1.66 (0.79–3.48)
0.93 (0.38–2.27)
1.65 (0.83–3.28)
3.02 (1.14–7.99)
3.14 (1.35–7.32)
2.20 (1.06–4.59)
3.23 (1.44–7.24)
2.54 (1.15–5.64)
2.45 (1.21–4.96)
3.41 (1.21–4.96)
0.61 (0.16–2.30)
1.24 (0.62–2.50)
1.00 (0.99–1.00)
0.84 (0.33–2.15)
2.43 (0.47–12.50)
0.002
0.23
0.56
0.59
0.18
0.87
0.16
0.03
0.01
0.04
0.004
0.02
0.01
0.02
0.47
0.55
0.98
0.72
0.29
1.59 (1.14–2.22)
0.007
2.90 (1.16–7.24)
2.76 (1.24–6.12)
0.02
0.01
1018
The American Journal of Cardiology (www.AJConline.org)
29.73, p ⫽ 0.02). Of note, neither of 2 type B patients
treated with antithrombotics died in hospital.
Discussion
To our knowledge, this is the first study of AAS to focus
specifically on ACS-like ECG findings. About 25% of patients with AAS in our regional referral center had ACS-like
ECG findings (fulfilling guideline definitions), which could
lead to misdiagnosis and inappropriate treatment. Surprisingly, most of the ACS-like electrocardiograms showed
NSTEMI features. On practical grounds, ACS-like ECG
patterns tended to be associated with more complicated
features. In type A disease, an ACS-like ECG profile appears to be an incremental risk factor for in-hospital mortality.
We used ECG definitions adopted at present for diagnosis of ACS. Unlike other studies that were not specifically
dedicated to electrocardiography, we decided to identify
patterns satisfying a guideline definition of ACS-like abnormalities, as to gain information regarding the treacherous
decision-making phase. About 25% of our patients had
ACS-like patterns. The finding that inappropriate treatment
was sometimes administered not only to patients satisfying
the restrictive study definition of ACS-like changes but also
to patients with less specific repolarization abnormalities
underscores the extent of the potential for misdiagnosis.
It is difficult to compare our frequency data with other
studies that used quite different ECG definitions. Some data
can be gleaned from IRAD, where “new Q-wave or STsegment shifts” were recorded in 3% to 7% patients; “ischemia” in 15% to 17%; and “nonspecific ST-segment or
T-wave changes” in 41% to 42%.1– 6 A smaller study focusing on ECG findings within 6 hours of onset of acute aortic
dissection (not intramural hematomas) recorded acute repolarization changes in 40% of the patients studied.16
The results of the present study underscore the relevance
of NSTEMI-like presentations, which is sixfold more common than STEMI-like changes.16 Many reports concern the
pitfalls caused by the coexistence of aortic dissection and
STEMI,8 –11 but little attention has been focused on
NSTEMI-like alterations.
In each Stanford subtype, ACS-like ECG findings at
presentation seemed to be associated with more complicated
disease. In type A disease, ACS-like ECG profile at presentation was associated with coronary ostia involvement and
hemodynamically significant aortic regurgitation (type B
patients with ACS-like ECG patterns more often had pleural
effusion and severe hypertension at presentation). Interestingly, looking at the small group of patients with ACS-like
ECG findings and available troponin data, we found that
about half tested negative. Few data are available on this
relevant topic. The single available study indicated that
abnormal troponin values could (depending on their definition) be found in 10% to 23% of patients with type A acute
aortic dissection.30
Knowledge of the evolution of ECG abnormalities, cardiac enzymes measurements, and left ventricular function
would be valuable to assess to what extent ACS-like ECG
findings are associated with true myocardial infarction or
prolonged ischemia. Unfortunately, at least in type A pa-
tients, it is scarcely feasible to study the natural evolution,
as the diagnosis is promptly followed by surgery, leaving a
very narrow time window for more detailed diagnostic investigation, and precluding meaningful before and after
comparisons. Nevertheless we endeavored to collect data
and provide indications regarding possible evolutions. We
recorded occurrences of 3 main scenarios: (1) development
of true myocardial infarction; (2) persistence of the ECG
alterations despite absence of myocardial infarction; and (3)
regression of the ECG alterations without any sign of myocardial infarction (the low numbers and incompleteness of
the laboratory data collection preclude any meaningful conclusions regarding the relative frequencies of these 3 evolutions). Thus, ACS-like ECG patterns do not seem to be
invariably the consequence of an acute ischemic episode at
the time of the acute aortic event.
Conceivable alternative explanations for the presence of
an ACS-like ECG profile include preexisting repolarization
abnormalities due either to previous ischemic episodes,
chronic coronary artery disease, or hypertension-induced
left ventricular overload; ST-T abnormalities due to acute
pericardial effusion; and “early repolarization” casually associated with AAS. In cases where ACS-like ECG profile is
determined by true myocardial ischemia, a variety of possibly co-existing mechanisms can be postulated, including
intimal flap interfering with the patency of the coronary
ostia acute left ventricular pressure overload, and global
myocardial ischemia due either to cardiac tamponade or to
other causes of low cardiac output or shock. An underlying
coronary artery disease could in turn amplify the effects of
any of these plausible acute mechanisms. Although coronary ostia involvement, cardiac tamponade, and acute aortic
insufficiency should be exclusive to type A disease; the
other proposed mechanisms could occur in either Stanford
subtype.
This was a retrospective study set in a single tertiary
center (albeit characterized by referral of all diagnosed
cases of AAS from surrounding hospitals). It was designed
specifically to provide important clues for clinical management of referred patients who reach a final diagnosis of
AAS (it is scarcely feasible to study those patients who die
before reaching a final diagnosis of AAS, who may have
systematically different characteristics). Questions regarding differential diagnosis between ACS and AAS could not
be addressed in this context. Confirmation of the findings
and further investigation (including serial troponin data) are
needed in a multicenter context.
In conclusion, ACS-like ECG findings appear to be
rather common at presentation, and are more often characterized by NSTEMI features. ACS-like ECG patterns were
common in type B as well as type A disease (and in
intramural hematomas as well as classic aortic dissection).
ACS-like ECG findings may stem from very different substrates, ranging from preexisting repolarization abnormalities, pericardially mediated modifications, or global myocardial ischemia to segmental ischemia or true myocardial
infarction. The ACS-like presentation electrocardiogram
was generally associated with more complicated forms of
AAS, and in Stanford type A disease with higher in-hospital
mortality.
Miscellaneous/ACS-Like Electrocardiography in AAS
1. Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ,
Russman PL, Evangelista A, Fattori R, Suzuki T, Oh JK, et al. The
International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. JAMA 2000;283:897–903.
2. Nienaber CA, von Kodolitsch Y, Nicolas V, Siglow V, Piepho A,
Brockhoff C, Koschyk DH, Spielmann RP. The diagnosis of thoracic
aortic dissection by noninvasive imaging procedures. N Engl J Med
1993;328:1–9.
3. Collins JS, Evangelista A, Nienaber CA, Bossone E, Fang J, Cooper
JV, Smith DE, O’Gara PT, Myrmel T, Gilon D, et al. International
Registry of Acute Aortic Dissection (IRAD). Differences in clinical
presentation, management, and outcomes of acute type a aortic dissection in patients with and without previous cardiac surgery. Circulation 2004;110(suppl):II237–II242.
4. Nienaber CA, Fattori R, Mehta RH, Richartz BM, Evangelista A,
Petzsch M, Cooper JV, Januzzi JL, Ince H, Sechtem U, Bossone E, et
al. International Registry of Acute Aortic Dissection. Gender-related
differences in acute aortic dissection. Circulation 2004;109:3014 –
3021.
5. Suzuki T, Mehta RH, Ince H, Nagai R, Sakomura Y, Weber F,
Sumiyoshi T, Bossone E, Trimarchi S, Cooper JV, et al. International
Registry of Aortic Dissection. Clinical profiles and outcomes of acute
type B aortic dissection in the current era: lessons from the International Registry of Aortic Dissection (IRAD). Circulation 2003;
108(suppl):II312–II317.
6. Evangelista A, Mukherjee D, Mehta RH, O’Gara PT, Fattori R, Cooper JV, Smith DE, Oh JK, Hutchison S, Sechtem U, et al, for the
International Registry of Aortic Dissection (IRAD) Investigators.
Acute intramural hematoma of the aorta: a mystery in evolution.
Circulation 2005;111:1063–1070.
7. Klompas M. Does this patient have an acute thoracic aortic dissection?
JAMA 2002;287:2262–2272.
8. Satler LF, Levine S, Kent KM, Pearle DL, Green CE, del Negro A,
Rackley CE. Aortic dissection masquerading as acute myocardial
infarction: implication for thrombolytic therapy without cardiac catheterization. Am J Cardiol 1984;54:1134 –1135.
9. Ortega-Carnicer J. Impending complete rupture of proximal aortic
dissection manifested by ST-segment elevation. Intensive Care Med
2001;27:940 –941.
10. Neri E, Toscano T, Papalia U, Frati G, Massetti M, Capannini G, Tucci
E, Buklas D, Muzzi L, Oricchio L, Sassi C. Proximal aortic dissection
with coronary malperfusion: presentation, management, and outcome.
J Thorac Cardiovasc Surg 2001;121:552–560.
11. Chang HC, Yeh KH, Huang HL, Wang CC, Chang YS. Aortic intramural hematoma presenting as acute inferior wall MI with cardiogenic
shock. Am J Emerg Med 2004;22:433– 436.
12. Davis DP, Grossman K, Kiggins DC, Vilke GM, Chan TC. The
inadvertent administration of anticoagulants to ED patients ultimately
diagnosed with thoracic aortic dissection. Am J Emerg Med 2005;23:
439 – 442.
13. Eriksen UH, Molgaard H, Ingerslev J, Nielsen TT. Fatal haemostatic
complications due to thrombolytic therapy in patients falsely diagnosed as acute myocardial infarction. Eur Heart J 1992;13:840 – 843.
14. Weiss P, Weiss I, Zuber M, Ritz R. How many patients with acute
dissection of the thoracic aorta would erroneously receive thrombolytic therapy based on the electrocardiographic findings on admission?
Am J Cardiol 1993;72:1329 –1330.
15. Eriksen UH, Molgaard H, Ingerslev J, Nielsen TT. Fatal haemostatic
complications due to thrombolytic therapy in patients falsely diagnosed as acute myocardial infarction. Eur Heart J 1992;13:840 – 843.
16. Hirata K, Kyushima M, Asato H. Electrocardiographic abnormalities
in patients with acute aortic dissection. Am J Cardiol 1995;76:1207–
1212.
17. Barabas M, Gosselin G, Crepeau J, Petitclerc R, Cartier R, Theroux P.
Left main stenting-as a bridge to surgery for acute type A aortic
dissection and anterior myocardial infarction. Catheter Cardiovasc
Interv 2000;51:74 –77.
1019
18. Patel VB, Griffin BP. Anomalous coronary artery, aortic dissection,
and acute myocardial infarction. J Am Soc Echocardiogr 1999;12:
326 –330.
19. Khan R, Amaram S, Gomes JA, Kelen GJ, Lynfield J, El-Sherif N.
Myocardial infarction following acute aortic dissection. Cathet Cardiovasc Diagn 1980;6:181–184.
20. Braunwald E, Antman EM, Beasley JW, Califf RM, Cheitlin MD,
Hochman JS, Jones RH, Kereiakes D, Kupersmith J, Levin TN, et al.
American College of Cardiology/American Heart Association Task
Force on Practice Guidelines (Committee on the Management of
Patients With Unstable Angina). ACC/AHA guideline update for the
management of patients with unstable angina and non-ST-segment
elevation myocardial infarction—2002: summary article: a report of
the American College of Cardiology/American Heart Association Task
Force on Practice Guidelines (Committee on the Management of
Patients With Unstable Angina). Circulation 2002;106:1893–1900.
21. Braunwald E, Antman EM, Beasley JW, Califf RM, Cheitlin MD,
Hochman JS, Jones RH, Kereiakes D, Kupersmith J, Levin TN, et al.
ACC/AHA guidelines for the management of patients with unstable
angina and non-ST-segment elevation myocardial infarction. A report
of the American College of Cardiology/American Heart Association
Task Force on Practice Guidelines (Committee on the Management of
Patients With Unstable Angina). J Am Coll Cardiol 2000;36:970 –
1062.
22. Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand
M, Hochman JS, Krumholz HM, Kushner FG, Lamas GA, et al, for the
American College of Cardiology; American Heart Association; Canadian Cardiovascular Society. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction— executive
summary report of the ACC/AHA Task Force on Practice Guidelines
(Committee to Revise the 1999 Guidelines on the Management of
Patients With Acute Myocardial Infarction. J Am Coll Cardiol 2004;
44:671–719.
23. Van de Werf F, Ardissino D, Betriu A, Cokkinos DV, Falk E, Fox KA,
Julian D, Lengyel M, Neumann FJ, Ruzyllo W, et al, for the Task
Force on the Management of Acute Myocardial Infarction of the
European Society of Cardiology. Task Force on the Management of
Acute Myocardial Infarction of the European Society of Cardiology.
Management of acute myocardial infarction in patients presenting with
ST-segment elevation. The Task Force on the Management of Acute
Myocardial Infarction of the European Society of Cardiology. Eur
Heart J 2003;24:28 – 66.
24. Bertrand ME, Simoons ML, Fox KA, Wallentin LC, Hamm CW,
McFadden E, De Feyter PJ, Specchia G, Ruzyllo W, for the Task Force
on the Management of Acute Coronary Syndromes of the European
Society of Cardiology. Management of acute coronary syndromes in
patients presenting without persistent ST-segment elevation. Eur
Heart J 2002;23:1809 –1840.
25. Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarction
redefined—a consensus document of The Joint European Society of
Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36:959 –969.
26. Sokolow M, Lyon TP. The ventricular complex in left ventricular
hypertrophy as obtained by unipolar precordial and limb leads. Am
Heart J 1949;37:161–186.
27. Okin PM, Devereux RB, Nieminen MS, Jern S, Oikarinen L, Viitasalo
M, Toivonen L, Kjeldsen SE, Julius S, Snapinn S, Dahlof B, for the
LIFE Study Investigators. Electrocardiographic strain pattern and prediction of cardiovascular morbidity and mortality in hypertensive patients. Hypertension 2004;44:48 –54.
28. Hollenberg SM, Kavinsky CJ, Parrillo JE. Cardiogenic shock. Ann
Intern Med 1999;131:47–59.
29. Spodick DH. Acute cardiac tamponade. N Engl J Med 2003;349:684 –
690.
30. Bonnefoy E, Godon P, Kirkorian G, Chabaud S, Touboul P. Significance of serum troponin I elevation in patients with acute aortic
dissection of the ascending aorta. Acta Cardiol 2005;60:165–170.
Predictors of Survival in Patients With End-Stage Renal Disease
Evaluated for Kidney Transplantation
Fadi G. Hage, MDa, Stuart Smalheiser, MDc, Gilbert J. Zoghbi, MDa, Gilbert J. Perry, MDa,
Mark Deierhoi, MDb, David Warnock, MDb, Ami E. Iskandrian, MDa, Angelo M. de Mattos, MDb,
and Raed A. Aqel, MDa,*
Cardiovascular disease is the major cause of mortality in patients with end-stage renal
disease (ESRD). This study examined the all-cause mortality in 3,698 patients with ESRD
evaluated for kidney transplantation at our institution from 2001 to 2004. Mean age for the
cohort was 48 ⴞ 12 years, and 42% were women. Stress myocardial perfusion imaging was
done in 2,207 patients (60%) and coronary angiography in 260 patients (7%). There were
622 deaths (17%) during a mean follow-up period of 30 ⴞ 15 months. The presence and
severity of coronary disease on angiography was not predictive of survival. Coronary
revascularization did not impact survival (p ⴝ 0.6) except in patients with 3-vessel disease
(p ⴝ 0.05). The best predictor of death was left ventricular ejection fraction, measured by
gated myocardial perfusion imaging, with 2.7% mortality increase for each 1% ejection
fraction decrease. In conclusion, left ventricular ejection fraction is a strong predictor of
survival in patients with ESRD awaiting renal transplantation. Strategies to improve
cardiac function or earlier renal transplantation deserve further studies. © 2007 Elsevier
Inc. All rights reserved. (Am J Cardiol 2007;100:1020 –1025)
It is well known that cardiovascular disease is the major
cause of mortality in patients with end-stage renal disease
(ESRD).1 In 2005, patients with ESRD had an annual incidence of cardiovascular mortality of 7.5%, which constituted 45% of all-cause mortality in this population.2 This
increased cardiovascular risk has been extended recently
from ESRD to all patients with chronic kidney disease,3–5
thereby exponentially expanding the population at risk. Cardiac assessment with stress myocardial perfusion imaging (MPI) before kidney transplantation has been suggested in patients at increased risk for cardiovascular
events.6 The only study that showed a benefit of coronary
revascularization (CR) before renal transplantation was
conducted in only 26 patients.7 Present guidelines for CR
in the general population are all evidence-based8,9; however, the prospective randomized clinical trials have systematically excluded patients with ESRD10 and are therefore not helpful in deciding on the best therapy in this
population. This study examined the predictors of allcause mortality in a large cohort of patients with ESRD
awaiting renal transplantation.
Methods
Study population: The University of Alabama at Birmingham Renal Transplant Database is one of the largest
and most comprehensive renal transplant data sources in the
a
Divisions of Cardiovascular Disease and bNephrology, University of
Alabama at Birmingham, Birmingham, Alabama; and cDivision of Cardiovascular Disease, University of Florida at Jacksonville, Jacksonville, Florida. Manuscript received March 2, 2007; revised manuscript received and
accepted April 6, 2007.
*Corresponding author: Tel: 205-934-7898; fax: 205-558-4714.
E-mail address: [email protected] (R.A. Aqel).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.045
United States. Detailed information pertaining to ⬎8,000
individuals seeking renal transplants since 1993 has been
collected in a prospective fashion. The section of transplant
nephrology employs 3 research nurses and 1 supervising
data analyst who are responsible for continuous update of
active patient files. Baseline information is collected at time
of evaluation; follow-up data are collected and entered prospectively thereafter. Several other sources of information,
including contact with patients and patients’ family members, referring physicians and health care professionals, our
hospital and outside hospital records, and the US Social
Security Death Master File were used to correctly ascertain
outcomes.
For the purposes of this study, all adult patients with
ESRD who underwent evaluation for kidney transplantation
between January 2001 and December 2004 at our institution
were screened for inclusion. We excluded all patients with
previous heart, combined heart-lung, lung, or liver transplants. There were no other exclusion criteria. Socioeconomic status was categorized using educational level attained as a surrogate marker (i.e., ⬍12 years of schooling
defined low socioeconomic status). Left ventricular (LV)
hypertrophy was defined by 12-lead electrocardiogram.
Body mass index was calculated by the formula: weight (in
kilograms)/square of height (in meters). Age at evaluation,
gender, diabetes mellitus status, hypertension, and tobacco
smoking history were extracted from the database. Survival
time was defined by the length of time between the date of
coronary angiography or MPI to the date of death or up to
December 31, 2005. To further ascertain mortality, a US
Social Security Death Master File search was conducted
against the entire cohort as of June 30, 2006. The study
protocol was reviewed and approved by the local institution
review board.
www.AJConline.org
Miscellaneous/Survival Predictors in End-stage Renal Disease
Cardiac evaluation: Our institution follows the renal
transplant candidates evaluation guidelines put forth by the
American Society of Transplantation11 as stress single-photon emission computed tomographic MPI is performed routinely in renal transplant candidates ⬎50 years, in subjects
with diabetes mellitus for ⬎10 years, in patients with previous cardiovascular events, and in patients with at least two
of the following risk factors: family history of ischemic
cardiac disease, LV hypertrophy, hypertension, dyslipidemia, or significant smoking history. More than 90% of all
stress MPI in our database was routinely performed at our
institution. The stress MPI studies were mostly performed at
midweek to reduce the influence of volume overload and
uremia, and since 1998 they have been performed in conjunction with gating to allow measurement of the LV ejection fraction (EF). Results of the stress MPI studies were
obtained from clinical reports and were labeled as showing
abnormal perfusion if reversible or fixed perfusion defects
were present. Most stress MPI studies were performed with
adenosine, according to methods previously described from
our laboratory.12
For the purpose of this study, coronary angiograms were
performed electively if the stress MPI was abnormal or at
the discretion of the consulting cardiologist based on clinical presentation (usually in cases of previously known
coronary disease). The results of the coronary angiography
were obtained from clinical reports. Coronary artery disease
(CAD) was termed 1-, 2-, or 3-vessel disease based on
⬎50% lumen diameter narrowing in any of the three major
coronary arteries (left anterior descending, left circumflex,
or right coronary) or any of their major branches. For
purpose of this analysis ⬎50% lumen diameter narrowing
of the left main coronary artery was considered equivalent
to 2-vessel disease. The angiograms were interpreted without knowledge of the stress MPI results and vice-versa.
Statistical analysis: Distribution of patients’ characteristics at evaluation were tested by chi-square analysis or
analysis of variance, as appropriate. Event-free survival
curves were constructed using the product-limit method
(Kaplan-Meier) and differences among survival curves were
estimated by the log-rank test. Cox regression modeling was
used to estimate crude (univariate) and adjusted (multivariate) risks, using variables found to have p ⱕ0.2 in the
univariate analysis, or when considered of biological significance (i.e., age, gender, smoking status). Direct visualization of Log [⫺log (survival time)] versus log (survival
time) plots was used to validate the proportionality of hazard increments assumption. LVEF by gated single photon
emission computed tomography and age at transplantation
were tested in the models, both as continuous and as categorical variables (LVEF ⱕ30%, 31% to 40%, 41% to 50%,
51% to 60%, or ⬎60%; age ⬍50 or ⱖ50 years). The
decision to describe a variable either as continuous or categorical was based on the best-fit of the multivariable models. Estimated risks were reported as hazard ratios (HR)
with correspondent 95% confidence intervals (CI). All p
values reported were 2-sided. SPSS version 11.5 for Windows (SPSS Inc., Chicago, Illinois) statistical software was
used in all the analysis.
1021
Results
Baseline characteristics: During the 4-year period from
January 2001 to December 2004, 3,698 patients (831 in
2001, 954 in 2002, 975 in 2003, and 938 in 2004) were
evaluated for kidney transplantation. Baseline characteristics of this cohort are listed in Table 1. The follow-up was
30 ⫾ 15 months, accumulating 8,599 patient-years. There
were a total of 622 deaths (17%), with a mean time to death
of 21 ⫾ 14.1 months.
Cardiac evaluation and intervention: According to
present guidelines, 12-lead electrocardiography was performed in all patients and stress MPI on 60% in the patients
evaluated for possible renal transplantation, with coronary
angiography reserved for those with an abnormal stress MPI
or with known CAD. The results of these tests are summarized in Figure 1. The mean LVEF, which was available for
2,048 patients, was 53 ⫾ 12%. LVEF was not available for
1,650 patients, and this group of patients was younger and
had significantly less co-morbidities (Table 1). Of the 260
patients who had coronary angiography, 94 (36%) underwent CR (1/3 by coronary artery bypass grafting and 2/3 by
percutaneous coronary interventions using stents). Stress
MPI was performed in 60% and coronary angiography in
7% of the entire cohort of 3,698 patients evaluated for renal
transplantation. Three-vessel CAD was found in 2% by
angiography, and 3% underwent CR.
Survival predictors: The strongest predictor of survival
by univariate analysis was LVEF with a stepwise increase in
mortality with decreasing LVEF (p ⬍0.001) (Figure 2 and
Table 2). This translated to an increased adjusted mortality
risk of 2.7% (HR 0.973, 95% CI 0.965 to 0.981, p ⬍0.001)
for each 1% decrease in the LVEF (adjusted for age, gender,
race, abnormal stress MPI, tobacco smoking, LV hypertrophy, diabetes mellitus, obesity, and socioeconomic status).
Patients who did not have LVEF measured had better survival as expected from their lower morbidities (Figure 2).
Myocardial perfusion abnormalities were predictive of
worse survival (p ⬍0.001) with no difference between patients having reversible defects and fixed defects (p ⫽ 0.45).
Presence and severity of CAD by angiography in the subset
of patients that underwent angiography was not predictive
of survival (2-year survival of 80%, 88%, 86%, and 78% for
0-, 1-, 2-, and 3-vessel disease, respectively, p ⫽ 0.6). CR
did not impact survival (2-year survival of 77% vs 81%, for
no CR and CR, respectively, p ⫽ 0.7), except in patients
with 3-vessel disease with no prior bypass surgery (p ⫽
0.05; Figure 3).
By multivariate analysis, LVEF ⱕ40%, reversible defects, age ⬎45 years, diabetes mellitus, and LV hypertrophy
were independently associated with mortality (Table 2).
Tobacco smoking, obesity, gender, and socioeconomic status did not reach statistical significance. This observation
was confirmed by analyzing LVEF as a continuous variable
(adjusted HR 0.973, 95% CI 0.965 to 0.981, p ⬍0.001) with
a 2.7% increased risk per each percentage point decrease in
the LVEF. Analysis of LVEF as categories demonstrated a
grade-reverse association between LVEF and survival (Figure 2). Analysis of LVEF as categories demonstrated a
1022
The American Journal of Cardiology (www.AJConline.org)
Table 1
Baseline characteristics at time of evaluation for kidney transplantation
Variable
LVEFⱕ40%
(n ⫽ 305)
Age (yrs, mean ⫾ SD)
Age ⬎45 years
Women
Diabetes mellitus
Low SES*
Tobacco use (history)
Body mass index ⱖ30
Black race
LVH by ECG†
Albumin (g/dl, mean ⫾ SD)
Creatinine (mg/dl, mean ⫾ SD)
Hematocrit (%, mean ⫾ SD)
Months on dialysis (mean ⫾ SD)
Original renal disease‡
Diabetes nephropathy
Hypertension
Polycystic kidneys
Lupus nephritis
51 ⫾ 9.8
㛳
223 (73%)
§,㛳
88 (29%)
㛳
195 (64%)
§,㛳
198 (65%)
§,㛳
122 (40%)
92 (30%)
171 (56%)
§,㛳
162 (53%)
§,㛳
3.4 ⫾ 0.47
§,㛳
8.7 ⫾ 4.64
36 ⫾ 5.1
§
20 ⫾ 27.9
㛳
㛳
149 (49%)
95 (31%)
㛳
14 (5%)
㛳
7 (2%)
LVEF⬎40%
(n ⫽ 1,743)
¶
52 ⫾ 10.7
¶
1307 (75%)
784 (45%)
¶
1063 (61%)
1011 (58%)
¶
575 (33%)
575 (33%)
924 (53%)
558 (32%)
¶
3.5 ⫾ 0.48
¶
8.0 ⫾ 4.07
35 ⫾ 7.4
¶
16 ⫾ 23.8
¶
784 (45%)
488 (28%)
¶
75 (4%)
¶
24 (1%)
LVEF not Measured
(n ⫽ 1,650)
Overall
(n ⫽ 3,698)
42 ⫾ 12.3
627 (38%)
693 (42%)
264 (16%)
924 (56%)
429 (26%)
512 (31%)
940 (57%)
544 (33%)
3.6 ⫾ 0.49
9.5 ⫾ 4.86
36 ⫾ 5.4
19 ⫾ 31.7
48 ⫾ 12
2158 (58%)
1558 (42%)
1552 (42%)
2133 (58%)
1119 (30%)
1182 (32%)
2047 (55%)
1264 (34%)
3.5 ⫾ 0.49
8.7 ⫾ 4.54
36 ⫾ 6.4
18 ⫾ 27.9
165 (10%)
528 (32%)
127 (8%)
106 (6%)
1072 (29%)
1109 (30%)
214 (6%)
137 (4%)
* Low SES is defined as ⬍12 years of formal education.
†
LVH on the ECG was determined by the cardiologist reading the ECG for clinical purposes.
‡
Percentages do not add to 100% because of multiple diagnoses.
§
p ⬍0.05 for LVEF ⱕ40% versus LVEF ⬎40%.
㛳
p ⬍0.05 for LVEF ⱕ40% versus LVEF not measured.
¶
p ⬍0.05 for LVEF ⬎40% versus LVEF not measured.
ECG ⫽ electrocardiogram; LVH ⫽ left ventricular hypertrophy; SES ⫽ socioeconomic status.
days after the transplantation 9 patients had an increase of
serum troponin I from baseline. Four of these patients did
not have CR (1 patient with no CAD, 1 patient with 2-vessel
CAD, and 2 patients with 3-vessel CAD) and 5 patients had
CR (2 patients with 2-vessel CAD and 3 patients with
3-vessel CAD) before transplantation. Only 5 patients were
documented clinically to have had a myocardial infarction
(1 patient with 2-vessel disease who had CR, 2 patients with
3-vessel CAD who had no CR, and 2 patients with 3-vessel
CAD who had CR). There were 2 deaths in the perioperative period (1 from retroperitoneal bleeding and 1 from
intracerebral bleeding), both of whom had no CAD by
angiography.
Figure 1. Cardiac evaluation and revascularization. Distribution of results
from (A) stress MPI, (B) LVEF by gated-SPECT, and (C) coronary angiography. (D) Patients who underwent coronary angiography and revascularization. BMS ⫽ bare metal stent; CABG ⫽ coronary artery bypass
grafting; DES ⫽ drug-eluting stent; PTCA ⫽ percutaneous transluminal
coronary angioplasty; SPECT ⫽ single-photon emission computed tomography. *Total number of patients included in the pie chart and its percentage
of the entire cohort of 3,698 patients. †Ischemia on stress MPI denotes any
evidence of ischemia even if accompanied with scar, whereas scar denotes
only scar with no evidence of ischemia. ‡More than 50% coronary artery
diameter narrowing in the stated number of coronary arteries as determined by
the cardiologist reading the angiograms for clinical purposes.
grade-reverse association between LVEF and survival (Figures 2 and 4).
Perioperative myocardial infarction and mortality: Of
the 260 patients who had coronary angiography, 57 eventually underwent kidney transplantation. Within the first 30
Discussion
Renal transplantation is the treatment of choice in patients
with ESRD as mortality is significantly lower in transplant
recipients compared with those that continue on chronic
dialysis.13,14 Patients awaiting renal transplantation are at
high risk of cardiovascular morbidity and mortality. Our
analysis showed that the strongest predictor of survival in
this patient population is a normal LVEF. Revascularization
in patients with renal dysfunction has been noted to carry
higher rates of complications, including death, compared
with patients with normal renal function whether it involves
bypass surgery15,16 or percutaneous coronary interventions,17,18 although newer modalities including using coronary stents have decreased the risk slightly.19,20 However,
renal dysfunction is an independent marker of morbidity
and mortality in patients with cardiovascular disease regardless of revascularization. In a retrospective examination,
Miscellaneous/Survival Predictors in End-stage Renal Disease
1023
Figure 2. Kaplan-Meier curves showing survival after transplant evaluation. (A) A severely reduced left ventricular function (LVEF ⱕ40%) portends a worse
survival. (B) Decreasing LVEF from ⬎60% to ⱕ40% predicts a stepwise increase in mortality.
Table 2
Variables associated with survival in patients with end-stage renal
disease being evaluated for renal transplantation
Variable
LVEF ⱕ40%
Abnormal stress MPI
Diabetes mellitus
Age ⬎45 yrs
LVH (by ECG)
Tobacco smoking
Women
Obesity
Socioeconomic status
Black race
Crude
HR
Adjusted
HR*
95% CI†
p Value
2.2
1.7
2.0
2.1
1.6
1.2
1.0
0.9
1.1
0.8
1.9
1.4
1.6
1.4
1.4
1.1
0.9
0.9
0.9
0.8
1.49–2.34
1.10–1.68
1.26–1.90
1.14–1.82
1.17–1.70
0.9–1.33
0.78–1.15
0.71–1.07
0.76–1.11
0.67–1.09
⬍0.001
0.005
⬍0.001
0.002
⬍0.001
0.4
0.6
0.2
0.4
0.5
* Adjusted by all the variables listed Table 1.
†
CI for the adjusted HR.
Abbreviations as in Table 1.
patients with renal dysfunction who presented with acute
coronary syndrome had lower long-term mortality when
they were managed with revascularization compared with
medical therapy.21 Similarly, when a large database of pa-
tients who underwent coronary angiography in Alberta,
Canada, was examined, patients with ESRD had better longterm (median follow-up 2.1 years) survival with CR.22 It is
significant to note that aside from a small study7 looking at
the outcome of 26 patients who were randomized to CR
versus medical therapy, there is a paucity of data on the
benefit of CR in asymptomatic patients with ESRD. Our
findings reveal that revascularization did not impact survival in the whole population of patients evaluated for renal
transplantation. This might be due to small numbers of
patients with 3-vessel CAD in whom there was a marginal
benefit of CR compared with medical therapy. The number
of perioperative myocardial infarcts was too small to determine whether CR had any impact, and the 2 deaths were not
due to cardiac complications.
At present, patients with ESRD awaiting renal transplantation are screened by stress MPI if they are ⬎50 years,
have diabetes mellitus for ⬎10 years, had previous cardiovascular events, or have significant risk factors. The prevalence of abnormal perfusion was low even in this group of
patients. Although patients with abnormal perfusion are at
higher risk for death, most deaths occur in patients with
1024
The American Journal of Cardiology (www.AJConline.org)
Figure 3. Kaplan-Meier curves showing the survival of patients with end-stage renal disease in relation to coronary artery disease presence and severity by
coronary angiography as well as coronary revascularization. LHC ⫽ left heart catheterization.
normal perfusion. The cause of death is not clear, but
sudden death is a likely mechanism. Sudden death is likely
due to LV hypertrophy, electrolyte imbalance, and possibly
subendocardial ischemia. There is a high likelihood that
increased vascular stiffness contributes to LV hypertrophy,
heart failure, and sudden death. The role of parathyroid
hormone, vitamin D and calcium, and phosphorus homeostasis in cardiovascular mortality is a subject of increasing interest at present.23–25 These metabolic alterations are
present well before patients reach the ESRD state,26 highlighting the importance of early detection and the potential
for early intervention.27
Previous studies28,29 have reported that stress MPI could
be used as a prognostic tool in asymptomatic patients with
ESRD and concluded that patients with perfusion abnormalities have worse cardiac outcome. Our data not only confirm
those findings, but because all our stress MPI were gated,
we were able to determine that the LVEF is even a stronger
predictor of survival in this population. In addition to LVEF
and perfusion, older age, diabetes mellitus, and LV hypertrophy contributed to increased mortality in this patient
population. The strong relation between LVEF and outcome
suggests that more attention needs to be given to interventions that could improve the LVEF while awaiting renal
transplantation and in their absence of triaging these patients earlier to renal transplantation (Figure 4). Wali et al30
has recently shown remarkable recovery of LVEF after
renal transplantation unrelated to CR, and we surely have
seen many examples of this observation (unpublished data).
There are many limitations to this study. First it is observational, single-center, and nonrandomized with the inherent limitations of any study of this nature, but as discussed, we do not have large multicenter prospective
randomized studies. These results should be taken as proof
of concept and necessary for sample size calculations for
more definitive studies. Second, the number of patients who
had coronary angiography and CR is small and a benefit
could have been present for CR but missed (type II error).
Figure 4. The effect of decreasing LVEF on the mortality risk using the
group with LVEF ⬎60% as the reference. Mortality is adjusted by age,
gender, race, ischemia, obesity, left ventricular hypertrophy, tobacco exposure, and socioeconomic status. *p ⬍0.001; †p ⫽ 0.002; ‡p ⫽ 0.4.
Third, we used all-cause mortality as an end point because
it is least controversial. Obviously patients with ESRD
could die from other causes. We did not evaluate other end
points such as nonfatal myocardial infarction, heart failure,
serious but nonlethal arrhythmias, and admissions for acute
coronary syndromes as the data were incomplete on these
events. The results of MPI were qualitative in terms of
presence or absence of ischemia or scar. It is possible that a
more quantitative index of extent and severity would have
increased the predictive power of MPI. Another limitation is
the lack of data on pharmacologic treatment, lipid levels,
markers of inflammation such as C-reactive protein, and
echocardiographic variables at the time of cardiac evaluation, as these parameters were not routinely obtained in all
patients and thus were not captured in the database. However, observational studies have reported a reverse association between hyperlipidemia and cardiovascular outcomes
in the ESRD population.31,32 Recently published interventional trials have failed to show significant cardiovascular
Miscellaneous/Survival Predictors in End-stage Renal Disease
benefit of treatment of dyslipidemia in dialysis33 and demonstrated a limited benefit in renal transplant recipients.34
1. Foley RN, Parfrey PS, Sarnak MJ. Epidemiology of cardiovascular
disease in chronic renal disease. J Am Soc Nephrol 1998;9:S16 –S23.
2. U.S. Renal Data System, USRDS 2005 Annual Data Report: Atlas of
End-Stage Renal Disease in the United States, National Institutes of
Health, National Institute of Diabetes and Digestive and Kidney Diseases. Bethesda, MD: National Institutes of Health, 2006.
3. Tonelli M, Wiebe N, Culleton B, House A, Rabbat C, Fok M, McAlister
F, Garg AX. Chronic kidney disease and mortality risk: a systematic
review. J Am Soc Nephrol 2006;17:2034 –2047.
4. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY. Chronic kidney
disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004;351:1296 –1305.
5. Szczech LA, Best PJ, Crowley E, Brooks MM, Berger PB, Bittner V,
Gersh BJ, Jones R, Califf RM, Ting HH, et al. Outcomes of patients
with chronic renal insufficiency in the bypass angioplasty revascularization investigation. Circulation 2002;105:2253–2258.
6. Pilmore H. Cardiac assessment for renal transplantation. Am J Transplant 2006;6:659 – 665.
7. Manske CL, Wang Y, Rector T, Wilson RF, White CW. Coronary
revascularisation in insulin-dependent diabetic patients with chronic
renal failure. Lancet 1992;340:998 –1002.
8. Eagle KA, Guyton RA, Davidoff R, Edwards FH, Ewy GA, Gardner
TJ, Hart JC, Herrmann HC, Hillis LD, Hutter AMJr, et al. ACC/AHA
2004 guideline update for coronary artery bypass graft surgery:
a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the
1999 Guidelines for Coronary Artery Bypass Graft Surgery). Circulation 2004;110:e340 – e437.
9. Smith SC, Jr, Feldman TE, Hirshfeld JW, Jr, Jacobs AK, Kern MJ,
King SB3rd, Morrison DA, O’Neil WW, Schaff HV, Whitlow PL, et
al. ACC/AHA/SCAI 2005 guideline update for percutaneous coronary
intervention: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/
SCAI Writing Committee to Update 2001 Guidelines for Percutaneous
Coronary Intervention). Circulation 2006;113:e166 – e286.
10. Coca SG, Krumholz HM, Garg AX, Parikh CR. Underrepresentation
of renal disease in randomized controlled trials of cardiovascular
disease. JAMA 2006;296:1377–1384.
11. Kasiske BL, Cangro CB, Hariharan S, Hricik DE, Kerman RH, Roth
D, Rush DN, Vazquez MA, Weir MR. The evaluation of renal transplantation candidates: clinical practice guidelines. Am J Transplant
2001;1(Suppl 2):3–95.
12. Hage FG, Dubovsky EV, Heo J, Iskandrian AE. Outcome of patients
with adenosine-induced ST-segment depression but with normal perfusion on tomographic imaging. Am J Cardiol 2006;98:1009 –1011.
13. Schnuelle P, Lorenz D, Trede M, Van Der Woude FJ. Impact of renal
cadaveric transplantation on survival in end-stage renal failure: evidence for reduced mortality risk compared with hemodialysis during
long-term follow-up. J Am Soc Nephrol 1998;9:2135–2141.
14. Wolfe RA, Ashby VB, Milford EL, Ojo AO, Ettenger RE, Agodoa LY,
Held PJ, Port FK. Comparison of mortality in all patients on dialysis,
patients on dialysis awaiting transplantation, and recipients of a first
cadaveric transplant. N Engl J Med 1999;341:1725–1730.
15. Cooper WA, O’Brien SM, Thourani VH, Guyton RA, Bridges CR,
Szczech LA, Petersen R, Peterson ED. Impact of renal dysfunction on
outcomes of coronary artery bypass surgery: results from the Society
of Thoracic Surgeons National Adult Cardiac Database. Circulation
2006;113:1063–1070.
16. Anderson RJ, O’Brien M, MaWhinney S, VillaNueva CB, Moritz TE,
Sethi GK, Henderson WG, Hammermeister KE, Grover FL, Shroyer
AL. Renal failure predisposes patients to adverse outcome after coronary artery bypass surgery. VA Cooperative Study #5. Kidney Int
1999;55:1057–1062.
17. Best PJ, Lennon R, Ting HH, Bell MR, Rihal CS, Holmes DR, Berger
PB. The impact of renal insufficiency on clinical outcomes in patients
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
1025
undergoing percutaneous coronary interventions. J Am Coll Cardiol
2002;39:1113–1119.
Rubenstein MH, Harrell LC, Sheynberg BV, Schunkert H, Bazari H,
Palacios IF. Are patients with renal failure good candidates for percutaneous coronary revascularization in the new device era? Circulation 2000;102:2966 –2972.
Stigant C, Izadnegahdar M, Levin A, Buller CE, Humphries KH.
Outcomes after percutaneous coronary interventions in patients with
CKD: improved outcome in the stenting era. Am J Kidney Dis 2005;
45:1002–1009.
Hassani SE, Chu WW, Wolfram RM, Kuchulakanti PK, Xue Z,
Gevorkian N, Suddath WO, Satler LF, Kent KM, Pichard AD, Weissman NJ, Waksman R. Clinical outcomes after percutaneous coronary
intervention with drug-eluting stents in dialysis patients. J Invasive
Cardiol 2006;18:273–277.
Keeley EC, Kadakia R, Soman S, Borzak S, McCullough PA. Analysis
of long-term survival after revascularization in patients with chronic
kidney disease presenting with acute coronary syndromes. Am J Cardiol 2003;92:509 –514.
Hemmelgarn BR, Southern D, Culleton BF, Mitchell LB, Knudtson
ML, Ghali WA. Survival after coronary revascularization among patients with kidney disease. Circulation 2004;110:1890 –1895.
Rostand SG, Drueke TB. Parathyroid hormone, vitamin D, and cardiovascular disease in chronic renal failure. Kidney Int 1999;56:383–
392.
Kamycheva E, Sundsfjord J, Jorde R. Serum parathyroid hormone
levels predict coronary heart disease: the Tromso Study. Eur J Cardiovasc Prev Rehabil 2004;11:69 –74.
Sambrook PN, Chen JS, March LM, Cameron ID, Cumming RG, Lord
SR, Schwarz J, Seibel MJ. Serum parathyroid hormone is associated
with increased mortality independent of 25-hydroxy vitamin d status,
bone mass, and renal function in the frail and very old: a cohort study.
J Clin Endocrinol Metab 2004;89:5477–5481.
Shinohara K, Shoji T, Tsujimoto Y, Kimoto E, Tahara H, Koyama H,
Emoto M, Ishimura E, Miki T, Tabata T, Nishizawa Y. Arterial
stiffness in predialysis patients with uremia. Kidney Int 2004;65:936 –
943.
London GM, Marchais SJ, Guerin AP, Metivier F, Adda H. Arterial
structure and function in end-stage renal disease. Nephrol Dial Transplant 2002;17:1713–1724.
Hase H, Joki N, Ishikawa H, Fukuda H, Imamura Y, Saijyo T, Tanaka
Y, Takahashi Y, Inishi Y, Nakamura M, Moroi M. Prognostic value of
stress myocardial perfusion imaging using adenosine triphosphate at
the beginning of haemodialysis treatment in patients with end-stage
renal disease. Nephrol Dial Transplant 2004;19:1161–1167.
Rabbat CG, Treleaven DJ, Russell JD, Ludwin D, Cook DJ. Prognostic
value of myocardial perfusion studies in patients with end-stage renal
disease assessed for kidney or kidney-pancreas transplantation: a metaanalysis. J Am Soc Nephrol 2003;14:431– 439.
Wali RK, Wang GS, Gottlieb SS, Bellumkonda L, Hansalia R, Ramos
E, Drachenberg C, Papadimitriou J, Brisco MA, Blahut S, et al. Effect
of kidney transplantation on left ventricular systolic dysfunction and
congestive heart failure in patients with end-stage renal disease. J Am
Coll Cardiol 2005;45:1051–1060.
Iseki K, Yamazato M, Tozawa M, Takishita S. Hypocholesterolemia is
a significant predictor of death in a cohort of chronic hemodialysis
patients. Kidney Int 2002;61:1887–1893.
Liu Y, Coresh J, Eustace JA, Longenecker JC, Jaar B, Fink NE, Tracy
RP, Powe NR, Klag MJ. Association between cholesterol level and
mortality in dialysis patients: role of inflammation and malnutrition.
JAMA 2004;291:451– 459.
Wanner C, Krane V, Marz W, Olschewski M, Mann JF, Ruf G, Ritz
E. Atorvastatin in patients with type 2 diabetes mellitus undergoing
hemodialysis. N Engl J Med 2005;353:238 –248.
Holdaas H, Fellstrom B, Jardine AG, Holme I, Nyberg G, Fauchald P,
Gronhagen-Riska C, Madsen S, Neumayer HH, Cole E, et al. Effect of
fluvastatin on cardiac outcomes in renal transplant recipients: a multicentre, randomised, placebo-controlled trial. Lancet 2003;361:2024 –
2031.
Prognosis of Idiopathic Recurrent Pericarditis as Determined
from Previously Published Reports
Massimo Imazio, MDa,*, Antonio Brucato, MDb, Yehuda Adler, MDc, Giovanni Brambilla, MDd,
Galit Artom, MDc, Enrico Cecchi, MDa, Giancarlo Palmieri, MDe, and Rita Trinchero, MDa
After a systematic review of all publications on recurrent pericarditis from 1966 to 2006, we
identified 8 major clinical series including a total of 230 patients with idiopathic recurrent
pericarditis (mean age 46 years, men/women ratio: 0.9). After a mean follow-up of 61
months, the complication rate was 3.5% cardiac tamponade and 0% constrictive pericarditis and left ventricular dysfunction. The overall life prognosis is excellent in idiopathic
recurrent pericarditis and complications are uncommon. In conclusion constrictive pericarditis was never reported despite numerous recurrences, and the risk is lower than in
idiopathic acute pericarditis (approximately 1%). Thus, it is important to reassure patients
on their prognosis, explaining the nature of the disease, and the likely course. Therapeutic
choices should take into account of the overall good outcome of these patients, including
less toxic agents. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:
1026 –1028)
In clinical practice, idiopathic recurrent pericarditis is the
most troublesome complication of acute pericarditis, with a
strong negative impact on the quality of life of patients.1–3
Physicians and patients are often worried about the possible
long-term prognosis and especially for the risk of constrictive pericarditis. These fears often cause treating physicians
to seek alternative treatments, triggering frequent consultation with “experts.” To assess the prognosis of idiopathic
recurrent pericarditis, we reviewed all published series and
analyzed the reported complications rate. The aim of this
work is to study the frequency of complications (cardiac
tamponade, constrictive pericarditis, and left ventricular
dysfunction) of cases of idiopathic recurrent pericarditis in
major published clinical series of recurrent pericarditis.
Methods and Results
To reflect the best available evidence, we performed a
comprehensive Medline search of all publications from
1966 to 2006 with the MeSH terms “pericarditis,” “recurrent pericarditis,” “pericardial constriction,” and “cardiac
tamponade.” Additional publications were sought using the
reference lists of identified papers, the published reviews on
the topic, and a search of abstracts from the scientific sessions of the American College of Cardiology, the American
Heart Association, and the European Society of Cardiology.
In the analysis of complications, only patients with idiopathic recurrent pericarditis were included. Cases of recurrent pericarditis after post-pericardiotomy syndromes,
a
Cardiology Department, Ospedale Maria Vittoria, Torino; bDepartment of Internal Medicine, Ospedali Riuniti, Bergamo; Departments of
d
Emergency Medicine and eInternal Medicine, Ospedale Niguarda, Milan,
Italy; and cCardiac Rehabilitation Institute, Chaim Sheba Medical Center,
Tel-Hashomer and Sackler Faculty of Medicine, Tel-Aviv, Israel. Manuscript received March 4, 2007; revised manuscript received and accepted
March 23, 2007.
*Corresponding author: Tel: 39-011-4393391; fax: 39-011-4393334.
E-mail address: [email protected] (M. Imazio).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.047
acute myocardial infarction, connective tissue diseases, or
other systemic disease or known etiology were excluded.
Excluding duplicate publications and pooled data, we
identified 8 major clinical series of recurrent pericarditis,4 –11 including a total of 230 patients with idiopathic
recurrent pericarditis (mean age 46 years, men/women ratio:
0.9). To be included in the present study patients had to
have an original attack of acute pericarditis, documented by
ⱖ2 of the following features: chest pain, pericardial friction
rub, evidence of pericardial effusion, or characteristic electrocardiographic changes. There was evidence of ⱖ1 recurrent attack of pericarditis ⱖ1 month after the first attack.
Although precise definition of recurrent pericarditis varied
among different studies (Table 1), the clinical diagnosis of
recurrence was based at least on recurrent chest pain and ⱖ1
of the following signs: fever, pericardial friction rub, electrocardiographic changes, echocardiographic evidence of
new or worsening pericardial effusion, increased C-reactive
protein, or increased erythrocyte sedimentation rate. There
was no evidence of a definable chronic systemic illness
(connective tissue disease, neoplasm, or infection) or any
other specific cause. Patients with recurrent pericarditic pain
without objective evidence of disease12,13 were not considered in the present study, because they are not represented
in published clinical series of recurrent pericarditis and do
not fulfil criteria for true recurrence.
After a mean follow-up of 61 months, the complication
rate was 3.5% cardiac tamponade, generally occurring during the initial attacks (62.5%; Table 2), 0% constrictive
pericarditis, and left ventricular dysfunction. Medical therapy was similar in the different studies, and corticosteroid
therapy was prescribed to most patients (mean 63.9%).
Complications and percentages of corticosteroid therapy
prescription are listed in Table 2.
Discussion
A systematic review of all publications on recurrent pericarditis shows that the overall life prognosis is excellent in
www.AJConline.org
Miscellaneous/Good Prognosis of Idiopathic Recurrent Pericarditis
1027
Table 1
Definitions of recurrent pericarditis in previously published articles
First Author
Year of Report
Definition of Recurrent Pericarditis
Robinson4
1968
Clementy5
1979
Fowler6
1986
Marcolongo7
1995
Raatikka8
2003
Imazio9
2005
CORE10
2005
Brucato11
2006
Precise definition is lacking in the methods. Pain was the first symptom in most cases, all cases
had electrocardiographic changes and elevated erythrocyte sedimentation rate.
Precise definition is lacking in the methods. All patients had chest pain, fever, diffuse
electrocardiographic changes, 50% of cases showed pericardial friction rubs.
Recurrent attack of pericarditis ⱖ3 months after the first attack. Diagnostic criteria for
pericarditis: chest pain and evidence of pericardial friction rub, echocardiographic evidence of
pericardial effusion, or characteristic electrocardiographic changes.
Persistence or recurrence of long-lasting chest pain ⱖ3 months after the original attack,
associated with fever or pericardial friction rub, specific electrocardiographic changes, or
echocardiographic findings of pericardial effusion.
The criterion for a relapse was recurrence of characteristic symptoms and findings (investigators
reported electrocardiographic changes, pericardial effusion, elevated erythrocyte sedimentation
rate, and C-reactive protein) after a period of ⱖ1 month since the onset of the previous attack.
Recurrence was documented by recurrent pain and ⱖ1 of the following signs: fever, pericardial
friction rub, electrocardiographic changes, echocardiographic evidence of pericardial effusion,
elevations in the white blood cell count or C-reactive protein, and/or erythrocyte sedimentation
rate.
Recurrence was documented by recurrent pain and ⱖ1 of the following signs: fever, pericardial
friction rub, electrocardiographic changes, echocardiographic evidence of pericardial effusion,
and elevations in the white blood cell count or erythrocyte sedimentation rate, or C-reactive
protein.
The minimum criteria for diagnosis of recurrences were the combination of: (1) typical chest
pain, (2) increased C-reactive protein, (3) electrocardiographic and/or echocardiographic
alterations. Investigators specified that C-reactive protein had to be elevated in all cases during
an acute attack.
Idiopathic cases were reported after a documented first attack of acute pericarditis, and exclusion of known etiologies.
Table 2
Complication rate in published clinical series on recurrent pericarditis (only cases of idiopathic recurrent pericarditis were included)
First Author
Year of
Report
No. of Patients
Mean Age
(yrs)
Men/
Women
Mean
Follow-up
(mo)
Steroids
Tamponade
Constriction
Left
Ventricular
Dysfunction
Robinson4
Clemency5
Fowler6
Marcolongo7
Raatikka8
Imazio9
CORE10
Brucato11
Overall
1968
1979
1986
1995
2003
2005
2005
2006
20
20
24
9
8
36
70
43
230
37 (13–70)
40 (22–67)
38 (14–57)
36 (15–65)
12 (7–16)
53 (18–75)
54 (18–79)
48 (19–75)
46 (7–79)
13/7
13/7
13/11
5/4
5/3
13/23
24/46
25/18
111/119
—
—
87.6
41.6
96.0
74.0
20.0
97.2
60.9
14/20
13/20
20/24
9/9
7/8
17/36
25/70
42/43
147/230 (63.9%)
2/20
2/20
0/24
0/9
0/8
0/36
0/70
4/43*
8/230† (3.5%)
0/20
0/20
0/24
0/9
0/8
0/36
0/70
0/43
0/230 (0.0%)
0/20
0/20
0/24
0/9
0/8
0/36
0/70
0/43
0/230 (0.0%)
* Only in the initials attacks.
Five of 8 (62.5%) in the initial attack.
†
idiopathic recurrent pericarditis, and complications are uncommon. Despite the common concern for the risk of constriction, constrictive pericarditis was never reported in
these patients despite numerous recurrences, and the overall
risk is lower than in idiopathic acute pericarditis (approximately 1%).3 Thus, it is important to reassure patients on
their prognosis, explaining the nature of the disease, and the
likely course. Drug treatment should take into account of
this good outcome, to avoid toxic agents, such as immunosuppressive drugs, that are only supported by weak evidence-based data.7,14 –16 Although corticosteroids are often
used in patients with recurrent pericarditis, either at the
beginning or after failure of previous treatments, and such
therapy may be more likely to be used in patients with
pericarditis resistant to initial treatment, there is growing
evidence showing that treating pericarditis with corticosteroids may increase the risk of recurrences,1,3,17 either in
acute pericarditis18 or recurrent pericarditis (Table 3).9 –11,17
Evidence from serologic studies, pericardial fluid, and tissue analyses have suggested that a viral cause due to a new
infection, a chronic infection, or a reactivation of a previous
viral infection may be responsible for recurrences.3–5,19 In such
cases, corticosteroids could promote viral replication and thus
further recurrences, and this may also explain possible deleterious effects of these drugs in viral or idiopathic acute pericarditis.18,20 In addition, corticosteroids are associated with significant side effects.
On this basis, the restriction of corticosteroid drugs may
1028
The American Journal of Cardiology (www.AJConline.org)
Table 3
Main results of published logistic regression analyses for predicted recurrences after corticosteroid treatment in acute and recurrent pericarditis
Study
Artom et al17
Imazio et al9
CORE study10
COPE study18
Patients
Condition
Follow-up (month)
OR
95% CI
p Value
119
55
84
120
ⱖ2 recurrences
ⱖ1 recurrences
First recurrence
Acute pericarditis
NA (1–185)
72
20
24
6.68
10.35
2.89
4.30
1.65–27.02
4.46–23.99
1.10–8.26
1.21–15.25
0.008
⬍0.001
0.040
0.024
CI ⫽ Confidence interval; OR ⫽ odds ratio.
play a major role in reducing the subsequent recurrence rate.
In addition, colchicine might be a useful drug in acute
pericarditis for the primary prevention of recurrences.1,2,18
Colchicine is safe and effective in the management of
recurrences, not only after failure of standard treatment,1 but
also in the treatment of the first recurrence.10 Thus, colchicine might be also the first choice drug for the secondary
prevention of recurrences.
Some investigators have suggested that physical invasion of the pericardium (such as pericardiotomy) might
promote recurrences and thus, they recommended to restrict invasive procedures as much as possible to prevent
further recurrences.20
In the present study, patients with recurrent pericarditic pain
without objective evidence of disease12,13 were not reported
because they were excluded from published studies. Nevertheless, the presence of pericardial pain alone is not uncommon in
clinical practice and has been described.12,13 About 10% of
patients with a previous acute pericarditis attack might present
recurrent pain without clinical evidence of disease. Female
gender, previous use of corticosteroids, and frequent recurrences were found to be risk factors for this syndrome.13 The
reason for this phenomenon is unknown, but it seems reasonable to avoid, whenever possible, starting a new course of
corticosteroids. Instead, a determined effort should be made to
treat the pain with simple analgesics. This approach to treatment requires the understanding and cooperation of the patient.
These patients must be followed carefully because this syndrome can precede clinical evidence of a true recurrence.
We acknowledge that the use of aspirin, nonsteroidal antiinflammatory drugs, steroids, colchicine, and other drugs
might be different in the studies we analyzed, but the overall
rate of events, such as cardiac tamponade, constrictive pericarditis, and left ventricular dysfunction was low to preclude an
analysis of the possible effect of the different therapies on these
complications. Nevertheless, the overall prognosis remains
good despite possible differences in drug treatment.
Although it would be desirable to establish in every recurrence whether the episode is a manifestation of autoimmunity
or infection, it is often difficult to perform a precise diagnosis
in clinical practice. Research is needed to find reliable, possibly
noninvasive, methods to distinguish autoimmune from infectious cases to develop a tailored treatment.
In conclusion, the overall prognosis in idiopathic recurrent pericarditis is good, and feared complications are uncommon.
1. Adler Y, Finkelstein Y, Guindo J, Rodriguez de la Serna A, Shoenfeld
Y, Bayes-Genis A, Sagie A, Bayes de Luna A, Spodick DH. Colchi-
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
cine treatment for recurrent pericarditis. A decade of experience.
Circulation 1998;97:2183–2185.
Maisch B, Seferovic PM, Ristic AD, Erbel R, Rienmuller R, Adler Y,
Tomkowski WZ, Thiene G, Yacoub MH, for the Task Force on the
Diagnosis and Management of Pricardial Diseases of the European
Society of Cardiology. Guidelines on the Diagnosis and Management
of Pericardial Diseases. Eur Heart J 2004;25:587– 610.
Soler-Soler J, Sagrista-Sauleda J, Permanyer-Miralda G. Relapsing
pericarditis. Heart 2004;90:1364 –1368.
Robinson J, Bridgen W. Recurrent pericarditis. BMJ 1968;2:272–275.
Clementy J, Jambert H, Dallacchio M. Les pericardites aigues recidivantes: 20 observations. Arch Mal Coeur 1979;72:857– 861.
Fowler NO, Harbin AD. Recurrent acute pericarditis: follow-up study
of 31 patients. J Am Coll Cardiol 1986;7:300 –305.
Marcolongo R, Russo R, Laveder F, Noventa F, Agostini C. Immunosoppressive therapy prevents recurrent pericarditis. J Am Coll Cardiol 1995;26:1276 –1279.
Raatikka M, Pelkonen PM, Karjalainen J, Jokinen EV. Recurrent
pericarditis in children and adolescents. J Am Coll Cardiol 2003;42:
759 –764.
Imazio M, Demichelis B, Parrini I, Cecchi E, Demarie D, Ghisio A,
Belli R, Bobbio M, Trinchero R. Management, risk factors, and outcomes in recurrent pericarditis. Am J Cardiol 2005;96:736 –739.
Imazio M, Bobbio M, Cecchi E, Demarie D, Pomari F, Moratti M,
Ghisio A, Belli R, Trinchero R. Colchicine as first choice therapy for
recurrent pericarditis: results of the CORE (COlchicine for REcurrent
pericarditis) trial. Arch Intern Med 2005;165:1987–1991.
Brucato A, Brambilla G, Moreo A, Alberti A, Munforti C, Ghirardello
A, Doria A, Shinar Y, Livneh A, Adler Y, et al. Long-term outcomes
in difficult-to-treat patients with recurrent pericarditis Am J Cardiol
2006;98:267–271.
Shabetai R. Often neglected yet important: the pericardium and its
disease. Herz 2000;25:717–719.
Imazio M, Demichelis B, Parrini I, Cecchi E, Pomari F, Demarie D,
Gaschino G, Ghisio A, Belli R, Trinchero R. Recurrent pericarditic
pain without objective evidence of disease in patients with previous
acute pericarditis. Am J Cardiol 2004;94:973–975.
Brucato A, Brambilla G, Adler Y, Spodick DH. Recurrent pericarditis:
therapy of refractory cases. Eur Heart J 2005;26:2600 –2601.
Asplen CH, Levine HD. Azatioprine therapy of steroid-responsive
pericarditis. Am Heart J 1970;80:109 –111.
Peterlana D, Puccetti A, Simeoni S, Tinazzi E, Corrocher R, Lunardi
C. Efficacy of intravenous immunoglobulin in chronic idiopathic pericarditis: report of four cases. Clin Rheumatol 2005;1:18 –21.
Artom G, Koren-Morag N, Spodick DH, Brucato A, Guindo J, Bayesde-Luna A, Brambilla G, Finkelstein Y, Granel B, Bayes-Genis A,
Schwammenthal E, Adler Y. Pretreatment with corticosteroids attenuates
the efficacy of colchicine in preventing recurrent pericarditis: a multicentre all-case analysis. Eur Heart J 2005;26:723–727.
Imazio M, Bobbio M, Cecchi E, Demarie D, Demichelis B, Pomari F,
Moratti M, Gaschino G, Giammaria M, Ghisio A, Belli R, Trinchero
R. Colchicine in addition to conventional therapy for acute pericarditis: results of the COlchicine for acute PEricarditis (COPE) trial.
Circulation 2005;112:2012–2016.
Maisch B. Recurrent pericarditis: mysterious or not so mysterious?
Eur Heart J 2005;26:631– 633.
Lange RA, Hillis LD. Acute pericarditis. N Engl J Med 2004;351:
2195–2202.
Comparison of Frequency of Complex Ventricular Arrhythmias
in Patients With Positive Versus Negative Anti-Ro/SSA
and Connective Tissue Disease
Pietro Enea Lazzerini, MDa,*, Pier Leopoldo Capecchi, MDa, Francesca Guideri, MDa,
Francesca Bellisai, MDb, Enrico Selvi, MDb, Maurizio Acampa, MDc, Agnese Costa, MDa,
Roberta Maggio, MDb, Estrella Garcia-Gonzalez, MDb, Stefania Bisogno, MDb,
Gabriella Morozzi, BiolDb, Mauro Galeazzi, MDb, and Franco Laghi-Pasini, MDa
A previous study of electrocardiography at rest showed that anti-Ro/SSA–positive patients
with connective tissue disease (CTD) frequently had corrected QT (QTc) interval prolongation. Because QTc interval prolongation is a definite risk factor for arrhythmic sudden
death in the general population, a 24-hour electrocardiographic monitoring study was
performed to investigate the possible relation between QTc interval prolongation and
incidence of ventricular arrhythmias as a possible expression of immunomediated electric
instability of the myocardium in anti-Ro/SSA–positive patients with CTD. The study
population consisted of 46 patients with CTD; 26 anti-Ro/SSA–positive and 20 anti-Ro/SSA–
negative (control group) patients (Sjögren’s syndrome, 9 and 3 patients; systemic lupus erythematosus, 4 and 9 patients; systemic sclerosis, 2 and 4 patients; undifferentiated CTD, 8 and
1 patients; mixed CTD, 2 and 2 patients, and polymyositis/dermatomyositis, 1 and 1 patient,
respectively). All patients underwent ambulatory Holter electrocardiography to obtain 24-hour
monitoring of the QTc interval and ventricular arrhythmias. With respect to the control group,
anti-Ro/SSA–positive patients with CTD (1) commonly showed QTc interval prolongation
(46% vs 5%), and this abnormality, when present, persisted for the 24 hours (global mean
24-hour QTc interval 440.5 ⴞ 23.4 vs 418.2 ⴞ 13.2 ms); (2) had a higher incidence of complex
ventricular arrhythmias (i.e., Lown classes 2 to 5, 50% vs 10%) also in the absence of
detectable cardiac abnormalities; and (3) in patients with CTD, there is a direct relation
between global mean 24-hour QTc interval and ventricular arrhythmic load independently of age and disease duration. In conclusion, anti-Ro/SSA–positive patients with
CTD seemed to have a particularly high risk of developing ventricular arrhythmias.
The risk appeared related mainly to abnormalities in ventricular electrophysiologic
characteristics emerging in the clinical setting as QTc interval prolongation. © 2007
Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:1029 –1034)
Cardiovascular involvement is common in patients with
connective tissue diseases (CTDs),1 and rhythm disturbances
represent a frequent manifestation of CTD-associated cardiovascular damage.2 Of the underlying arrhythmogenic mechanisms, a possible direct arrhythmogenicity of anti-Ro/SSA
antibodies (targeting an extractable nuclear antigen, the
ribonucleoprotein Ro, composed of 2 subunits of 60 and 52
kd, respectively3) was hypothesized. Comparing 31 anti-Ro/
SSA–positive patients and 26 anti-Ro/SSA–negative patients with CTD, we found significant prolongation of the
corrected QT interval (QTc) and a high incidence of QTc
intervals higher than the upper limit of normal (440 ms) in
anti-Ro/SSA–positive subjects.4 In patients with coronary
artery disease, frequent or complex ventricular arrhythmias
were associated with increased cardiac event rates and mortality.5 Furthermore, QTc interval prolongation (and dispersion) predicted the presence of frequent or complex ventricular arrhythmias in hypertensive patients with left
ventricular hypertrophy (also in untreated patients newly
presenting with hypertension).6,7 Finally, Kajiyama et al8
showed that an increase in QT interval dispersion during
percutaneous coronary angioplasty was associated with ventricular tachyarrhythmias. The aim of the study was to
investigate the possible relation between QTc interval prolongation and incidence of ventricular arrhythmias during
24-hour electrocardiographic monitoring in anti-Ro/SSA–
positive patients with CTD as a possible expression of
immunomediated electric instability of the myocardium.
Methods
Department of Clinical Medicine and Immunological Sciences, Divisions of aClinical Immunology, bRheumatology, and cInternal Medicine,
University of Siena, Siena, Italy. Manuscript received February 15, 2007;
revised manuscript received and accepted April 13, 2007.
*Corresponding author: Tel.: 39-0577-585-741; fax: 39-0577-4-4114.
E-mail address: [email protected] (P.E. Lazzerini).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.048
The study population consisted of 46 patients with different
CTDs9 –14 allocated to 2 groups of 26 anti-Ro/SSA–positive
and 20 anti-Ro/SSA–negative patients (Table 1). None of
the subjects participated in the previous study of electrocardiography (ECG) at rest4 or used drugs potentially influencwww.AJConline.org
1030
The American Journal of Cardiology (www.AJConline.org)
Table 1
Demographic characteristics and ongoing treatment of study subjects
Variable
Anti-Ro/SSA–Positive
(n ⫽ 26)
Anti-Ro/SSA–Negative
(n ⫽ 20)
47.4 ⫾ 14.3
23–73
24/2
82.8 ⫾ 61.9
44.3 ⫾ 17.1
24–70
18/2
96.7 ⫾ 75.1
9
4
2
8
2
1
3
9
4
1
2
1
13 (8.1 mg/d)
4 (400 mg/d)
2 (12.5 mg/wk)
2 (20 mg/d)
8
0
1
9
10 (15.7 mg/d)
8 (333 mg/d)
2 (9.2 mg/wk)
0
6
1
1
4
Age (yrs)
Age range (yrs)
Women/men
Mean disease duration (mo)
Diagnosis
Sjögren’s syndrome
Systemic lupus erythematosus
Systemic sclerosis
Undifferentiated CTD
Mixed CTD
Polymyositis/dermatomyositis
Treatment
Corticosteroids*
Hydroxychloroquine
Methotrexate
Leflunomide
Nonsteroidal anti-inflammatory drugs
Iloprost (pulsed therapy)
Intravenous immunoglobulins (pulsed therapy)
No therapy
Data expressed as mean ⫾ SD or number of patients unless noted otherwise.
* Prednisone-equivalent dose. Data in parentheses expressed as mean dose.
Table 2
Echocardiographic findings in the population under study
Variable
Ejection fraction (%)
Left atrial size (mm)
Left ventricular internal
dimension (mm)
Aortic root (mm)
Posterior wall thickness
(mm)
Venticular septum (mm)
Estimated pulmonary
artery pressure
(mm Hg)
Anti-Ro/SSA–Positive Anti-Ro/SSA–Negative
(n ⫽ 26)
(n ⫽ 20)
62.6 ⫾ 3.4
34.8 ⫾ 4.9
46.8 ⫾ 3.1
63.1 ⫾ 3.9
33.5 ⫾ 3.4
46.1 ⫾ 4.4
32.9 ⫾ 3.9
8.2 ⫾ 1.9
29.8 ⫾ 4.2
9.0 ⫾ 1.1
9.8 ⫾ 1.2
26.2 ⫾ 4.8
9.7 ⫾ 0.9
25.5 ⫾ 2.2
No patient had evidence of pericarditis/myocarditis/endocarditis, diastolic dysfunction, or significant valve abnormalities.
ing the QTc interval, except for hydroxychloroquine (Table 1).
Patients did not have electrocardiographic and/or echocardiographic abnormalities (Table 2) and/or a history
consistent with coronary artery disease (or other organic
cardiomyopathies), diabetes, renal failure, or electrolyte
imbalance.15,16 However, all patients for whom Lown class
4B was found on 24-hour electrocardiographic recording
adjunctively underwent stress ECG to more strongly rule
out possible underlying asymptomatic coronary artery disease. In the 3 patients with Lown class 4B, the stress test did
not induce ischemia. Of 46 patients, 4 had mild arterial
hypertension (3 patients in the anti-Ro/SSA–positive group,
1 in the anti-Ro/SSA–negative group; 2-sided Fisher’s exact
test p ⫽ 0.62 [NS]), but none had an echocardiographic
pattern of hypertrophy. Patients with branch bundle block
(standard QRS duration ⬎120 ms) were excluded from the
study. Informed consent was obtained from patients before
entering the study.
Antinuclear antibody testing was performed using indirect immunofluorescence on Hep-2000 (ImmunoConcepts,
Sacramento, California), a special substrate that uses a patented Hep2 cell line genetically engineered to increase sensitivity to anti-Ro/SSA autoantibodies. Then, anti-extractable nuclear antigen analysis was performed in all sera
using Western blot analysis with a commercial kit (MarDx,
Carlsbad, California) for which antigenic preparation is
denaturized. This method allowed us to identify antibodies
against SSA of 52 and 60 kd. All selected patients underwent ambulatory 24-hour ECG to monitor QTc interval and
the occurrence of ventricular arrhythmias. The QTc interval
was monitored for 24 hours using a Holter electrocardiographic recorder with a 12-channel system (Prima-Holter, Cardioline, Remco, Vignate-Milano, Italy). The channel with
the best signal-to-noise ratio was selected for automatic QT
analysis. Analog electrocardiographic signals were digitalized using a resolution of 12 bits with a data acquisition
system installed on an Acer-Veritron-7700G computer
(Acer Italy S.R.L., Lainate, Italy). To measure QTc intervals, the onset of the QRS and end of the T wave were
detected using an algorithm including the basic steps of
electrocardiographic signal preprocessing with a low-pass
differentiator, corrected QRS detection and definition, Twave end definition, and corrected QT value selection
(Prima 3/12 CH Model SW QT, Cardioline, Remco). The
T-wave end definition is based on the first derivative of the
electrocardiographic signal. To identify T-wave end, a
search window was defined from the QRS position. Finally,
QT interval was corrected using Bazett’s formula.17 Data
were recorded in digitalized form and represented graphically as a trend. For every patient, mean QTc interval was
Miscellaneous/Arrhythmias in Anti-Ro/SSA–Positive CTD
1031
Table 3
Parameters of 24-hour corrected QT (QTc) intervals in anti-Ro/SSA–positive and –negative subjects
Variable
Global mean 24-h QTc (ms)
Patients with prolonged mean 24-h QTc (mean QTc ⬎440 ms)
Hours/24 h with mean QTc ⬎440 ms
Maximum mean 24-h QTc (ms)
Minimum mean 24-h QTc (ms)
Median (50th percentile) (ms)
Anti-Ro/SSA–Positive
(n ⫽ 26)
Anti-Ro/SSA–Negative
(n ⫽ 20)
p Value
440.5 ⫾ 23.4
12 (46%)
16 (66%)
497.0 ⫾ 12.0
414.7 ⫾ 13.3
438.6
418.2 ⫾ 13.2
1 (5%)
0 (0%)
446.5 ⫾ 8.2
397.7 ⫾ 13.4
414.8
⬍0.001*
⬍0.005†
* Two-tail unpaired t test.
Two-sided Fisher’s exact test.
†
calculated for every single hour of the entire day for each of
the 12 leads. The lead with the longest mean QTc interval
for every single hour and the lead with the longest mean
QTc interval for the entire 24-hour registration (global mean
24-hour QTc interval) were considered.
Subjects with ventricular arrhythmias were categorized
according to Lown and Wolf.18 Groups were defined as
grade 0, no ventricular ectopic beat; class 1A, ⬍720 ventricular ectopic beats/24 hours, with ⱕ1 ventricular ectopic
beat/min; class 1B, ⬍720 ventricular ectopic beats/24
hours, with ⱖ2 ventricular ectopic beats/min; class 2, ⱖ720
ventricular ectopic beats/24 hours; class 3, multiform ventricular extrasystole or bigeminal or trigeminal extrasystole;
class 4A, ventricular extrasystoles in couplets; class 4B,
ventricular tachycardias; and class 5, ventricular extrasystole of the R-on-T type. Frequent or complex ventricular
arrhythmias were defined as Lown classes 2 to 5.5 After
Lown classification,18 we also defined an arbitrary score
(ventricular arrhythmia score) aimed at identifying the ventricular arrhythmic load of every patient. Scores ranged from 0 to
7 and were defined as score 0, Lown class 0; score 1, Lown
class 1A, score 2, Lown class 1B; score 3, Lown class 2; score
4, Lown class 3; score 5, Lown class 4A; score 6, Lown class
4B; and score 7, Lown class 5.
Statistical evaluation of differences between groups was
performed using 2-tail Student’s t test for unpaired data for
normally distributed continuous variables and 2-tail MannWhitney test for continuous variables not normally distributed. The 2-sided Fisher’s exact test was performed to
evaluate categorical variables. Correlation between 2 parameters was tested using linear regression analysis. In any
case, p ⬍0.05 was considered significant (GraphPad-InStat,
version 3.00 for Windows 95, GraphPad, San Diego, CA;
Microsoft Corp., Redmond, WA).
Results
In the patient group, 15 of 26 were positive for antibodies to
52- and 60-kd Ro proteins; 17 of 26, for anti– 60-kd Ro; and
23 of 26, for anti–52-kd Ro. Eleven patients were also
positive for anti-La/SSB antibodies. None of the controls
was positive for anti-Ro/SSA and/or anti-La/SSB antibodies. Anti-Ro/SSA–positive patients showed significant prolongation of mean QTc interval during the 24-hour monitoring period with respect to anti-Ro/SSA–negative subjects
(Table 3). Moreover, this difference was also significant for
every single hour of registration (Figure 1), with mean
hourly values ranging from 436.1 ⫾ 24.5 to 444.3 ⫾ 23.3
ms in anti-Ro/SSA–positive patients vs 415.2 ⫾ 16.4 to
421.3 ⫾ 17.2 ms in anti-Ro/SSA–negative patients (Figure 1).
In the patient group, mean hourly QTc values were higher
than the upper normal limit of 440 ms for 16 of 24 hours
(66%) monitored versus 0 of 24 hours (0%) in controls (Figure
1; Table 3). The 24-hour QTc monitoring in every single
patient showed that 12 of 26 anti-Ro/SSA–positive patients
(46%) had a global mean 24-hour QTc interval ⬎440 ms
(Table 3). Conversely, only 1 of 20 controls (5%) had a
prolonged global mean 24-hour QTc interval (Table 3). In
more detail, of positive patients, 6 of 12 (23% of overall
positive group) with a prolonged global mean 24-hour
QTc interval had values ⬎450 ms, with a maximal 24hour mean value of 497.0 ⫾ 12.0 ms (Table 3). The only
anti-Ro/SSA–negative subject with a prolonged QTc interval had a global mean 24-hour value of 446.5 ⫾ 8.2 ms
(Table 3).
No differences in global mean 24-hour QTc intervals
were detected in the anti-Ro/SSA–positive group when patients were stratified on the basis of antibody subtype positivity. The presence of ventricular arrhythmias (ⱖ1 ventricular ectopic beat/24 hours) was detected in 16 of 26
anti-Ro/SSA–positive patients (61.5%) and in 6 of 20 controls (30%; Table 4). This difference became more evident
when considering frequent or complex ventricular arrhythmias (i.e., ventricular arrhythmias in Lown classes 2
to 5), with a prevalence of 50% (13 of 26) in the patient
group (8 patients Lown class 3, 3 patients Lown class 4A,
and 2 patients Lown class 4B) versus 10% (2 of 20) in
controls (1 control Lown class 4A, 1 control Lown class
4B; Table 4). In more detail, in anti-Ro/SSA–positive patients, we found multiform ventricular ectopic beats in
about 1/3 and bigeminal/trigeminal or couplet ventricular
ectopic beats in about 1/4 of patients. Finally, episodes of
unsustained ventricular tachycardia were detected in 2 of 26
anti-Ro/SSA–positive patients (8%; Table 4). Conversely,
we did not find multiform or bigeminal/trigeminal ventricular ectopic beats in anti-Ro/SSA–negative patients (except
for 1 patient), but only infrequent monomorphic and isolated ventricular ectopic beats. However, couplets and
1 episode of unsustained ventricular tachycardia were also
registered in 2 different anti-Ro/SSA–negative patients with
CTD (Table 4). Moreover, considering ventricular arrhythmias included in the higher risk Lown classes 4 to 5 only,
we found a twofold higher prevalence in anti-Ro/SSA–
1032
The American Journal of Cardiology (www.AJConline.org)
Figure 1. Plot of mean hourly QTc intervals of anti-Ro/SSA–positive and –negative patients with CTDs. Vertical bars represent SD. * p ⬍0.01; ** p ⬍0.001,
2-tail unpaired t test (anti-Ro/SSA–positive vs –negative patients for every hour).
Table 4
Ventricular arrhythmias at the 24-hour ambulatory electrocardiogram recording in anti-Ro/SSA–positive and –negative subjects
Ventricular ectopic beats
Lown class 2–5
Multiform ventricular ectopic beats
Bigeminal/trigeminal ventricular ectopic beats
Couplets
Ventricular tachycardia
Mean ventricular arrhythmia score
Anti-Ro/SSA–Positive
(n ⫽ 26)
Anti-Ro/SSA–Negative
(n ⫽ 20)
p Value
16 (61%)
13 (50%)
8 (31%)
6 (23%)
5 (19%)
2 (8%)
2.38 ⫾ 2.28
6 (30%)
2 (10%)
1 (5%)
1 (5%)
1 (5%)
1 (5%)
0.75 ⫾ 1.68
⬍0.05*
⬍0.005*
⬍0.05†
* Two-sided Fisher’s exact test.
†
Two-tail Mann-Whitney test.
positive versus –negative patients (19% vs 10%). However,
this difference did not reach statistical significance. Ventricular arrhythmia scores were significantly different in the
2 groups of patients, with a threefold higher mean value in
anti-Ro/SSA–positive patients versus controls (Table 4). No
differences in mean ventricular arrhythmia scores were detected in the anti-Ro/SSA–positive group when patients
were stratified on the basis of antibody subtype positivity. A
positive correlation was found between ventricular arrhythmia scores and global mean 24-hour QTc intervals considering the entire study population (Figure 2). However, no
correlation was present between ventricular arrhythmia
scores and age (r ⫽ 0.2492; p ⫽ NS) or disease duration
(r ⫽ 0.2334; p ⫽ NS). Furthermore, global mean 24-hour
QTc interval was significantly longer in patients with CTD
(anti-Ro/SSA–positive and –negative) in Lown classes 2 to
5 (n ⫽ 15) than in patients with CTD in Lown classes 0 to
1 (n ⫽ 31; 442.7 ⫾ 24.9 vs 425.0 ⫾ 17.3 ms; 2-tail
Mann-Whitney test, p ⫽ 0.017).
Discussion
The main results arising from the present study were that (1)
anti-Ro/SSA–positive patients with CTD commonly showed
QTc interval prolongation, and this abnormality, when present,
was persistent throughout the 24-hour observation period; (2)
in the same patients, a high incidence of complex ventricular
arrhythmias was shown, also in the absence of detectable
cardiac abnormalities of structural origin; and (3) in patients
with CTD, a direct relation existed between global mean 24hour QTc interval and ventricular arrhythmic load independent
of age and disease duration.
A possible limitation of our study was the nonequal
distribution of different CTDs in the 2 groups, as expected
from epidemiologic studies. In a previous study of ECG at
rest, we showed that anti-Ro/SSA–positive patients frequently (58% of patients) showed QTc interval prolongation
(with mean values higher than the upper normality limit of
440 ms).4 Thus, we hypothesized that anti-Ro/SSA antibodies may exert a direct arrhythmogenic effect, thereby pro-
Miscellaneous/Arrhythmias in Anti-Ro/SSA–Positive CTD
1033
Figure 2. Relation between ventricular arrhythmia score (VAS) and global mean 24-hour QTc interval in the entire population under study (n ⫽ 46).
viding a risk of developing life-threatening arrhythmias,
because QTc interval prolongation is a definite risk factor
for arrhythmic death in the general population.15 Conversely, a later study by Costedoat-Chalumeau et al19 evaluating 32 anti-Ro/SSA–positive compared with 57 anti-Ro/
SSA–negative patients with CTD found no difference in
QTc interval duration. The investigators studied a very
selected cohort of subjects consisting almost exclusively of
patients with systemic lupus erythematosus (close to 90% in
each group). However, in a study by Gordon et al,20 QTc
interval was reported to be longer in the anti-Ro/SSA–
positive than –negative group, and this difference, although
not significant, approached significance (p ⫽ 0.063).
The finding of the present study, designed to minimize
possible errors related to a single QTc interval evaluation,
more strongly confirmed the evidence that anti-Ro/SSA–
positive patients with CTD had an increased risk to present
electrocardiographic signs of cardiac repolarization prolongation. The new data are that QTc interval prolongation,
when present, persisted for a preponderant portion of the
daytime (⬎60%). The plot of mean hourly QTc intervals
clearly showed 2 populations presenting constantly different
values, notwithstanding the similarity of changes during the
24-hour period. None of the patients under study used drugs
potentially influencing the QTc interval, except for hydroxychloroquine. The number of subjects using this drug
was higher in the anti-Ro/SSA–negative than –positive
group; therefore, virtually ruling out a possible role of
hydroxychloroquine in the genesis of the detected QTc
interval prolongation in anti-Ro/SSA–positive patients. Accordingly, more recent studies showed the cardiovascular
safety of hydroxychloroquine.21 As discussed in a previous
report,4 we hypothesized that QTc interval abnormalities
were a consequence of direct interference of anti-Ro/SSA
antibodies on ventricular repolarization, even if the pos-
sible electrophysiologic mechanism was still unknown.
However, on the basis of recent data22 showing that
anti-Ro/SSA antibodies inhibited specific calcium currents (L and T types), thereby providing a putative pathogenetic mechanism of complete atrioventricular block
and sinus bradycardia in anti-Ro/SSA–positive newborns, the possibility of selective blocking activity of
anti-Ro/SSA antibodies on potassium currents cannot be
ruled out. In ventricular cardiomyocytes, repolarization
was mainly dependent on potassium channels.16
1. Riboldi P, Gerosa M, Luzzana C, Catelli L. Cardiac involvement in systemic
autoimmune diseases. Clin Rev Allergy Immunol 2002;23:247–261.
2. Lazzerini PE, Capecchi PL, Guideri F, Acampa M, Galeazzi M, Laghi
Pasini F. Connective tissue diseases and cardiac rhythm disorders: an
overview. Autoimmun Rev 2006;5:306 –313.
3. Franceschini, F, Cavazzana I. Anti-Ro/SSA and La/SSB antibodies.
Autoimmunity 2005;38:55– 63.
4. Lazzerini PE, Acampa M, Guideri F, Capecchi PL, Campanella V,
Morozzi G, Galeazzi M, Marcolongo R, Laghi Pasini F. Prolongation of
the corrected QT interval in adult patients with anti-Ro/SSA-positive
connective tissue diseases. Arthritis Rheum 2004;50:1248 –1252.
5. Engstrom G, Hedblad B, Janzon L, Juul-Moller S. Ventricular arrhythmias during 24-h ambulatory ECG recording: incidence, risk factors
and prognosis in men with and without a history of cardiovascular
disease. J Intern Med 1999;246:363–372.
6. Kulan K, Ural D, Komsuoglu B, Agacdiken A, Goldeli O, Komsuoglu
SS. Significance of QTc prolongation on ventricular arrhythmias in
patients with left ventricular hypertrophy secondary to essential hypertension. Int J Cardiol 1998;64:179 –184.
7. Saadeh A, Evans S, James M, Jones J. QTc dispersion and complex
ventricular arrhythmias in untreated newly presenting hypertensive
patients. J Hum Hypertens 1999;13:665– 669.
8. Kajiyama A, Saito D, Murakami T, Shiraki T, Oka T, Doi M, Masaka
T, Tanemoto K, Tsuji T. Relation of QT-interval variability to ventricular arrhythmias during percutaneous transluminal coronary angioplasty. Jpn Circ J 2001;65:779 –782.
9. Vitali C, Bombardieri S, Moutsopoulos HM, Balestrieri G, Bencivelli
W, Bernstein RM, Bjerrrum KB, Braga S, Coll J, de Vita S, et al;
European Study Group on Diagnostic Criteria for Sjogren’s Syndrome.
1034
10.
11.
12.
13.
14.
15.
The American Journal of Cardiology (www.AJConline.org)
Preliminary criteria for the classification of Sjogren’s syndrome:
results of a prospective concerted action supported by the European
Community. Arthritis Rheum 1993;36:340 –347.
Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF,
Schaller JG, Winchester RJ. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;23:
1271–1277.
Subcommittee for Scleroderma Criteria of the American Rheumatism
Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis (scleroderma).
Arthritis Rheum 1980;25:581–590.
Mosca M, Tavoni A, Neri R, Bencivelli W, Bombardieri S. Undifferentiated connective tissue diseases: the clinical and serological profile
of 91 patients followed for at least 1 year. Lupus 1998;7:95–100.
Alarcon-Segovia D, Cardiel MH. Comparison between 3 diagnostic
criteria for mixed connective tissue disease: study of 593 patients.
J Rheumatol 1989;16:328 –334.
Dalakas MC, Hohlfeld R. Polymyositis and dermatomyositis. Lancet
2003;362:971–982.
Helming H, Brendorp B, Kober L, Sahebzadah N, Torp-Petersen C.
QTc interval in the assessment of cardiac risk. Card Electrophysiol
Rev 2002;6:289 –294.
16. Witchel HJ, Hancox JC. Familial and acquired long QT syndrome and
the cardiac rapid delayed rectifier potassium current. Clin Exp Pharmacol Physiol 2000;27:753–766.
17. Bazett HC. An analysis of the time relations of electrocardiograms.
Heart 1920;7:353–376.
18. Lown B, Wolf M. Approaches to sudden death from coronary heart
disease. Circulation 1971;44:130 –142.
19. Costedoat-Chalumeau N, Amoura Z, Hulot JS, Ghillani P, FunckBrentano C, Piette JC. In response to “Prolongation of the corrected
QT interval in adult patients with anti-Ro/SSA-positive connective
tissue diseases” by Lazzerini et al in Arthritis Rheum 2004;50:
1248 –1252. Arthritis Rheum 2005;52:676 – 677; author reply 677–
678.
20. Gordon PA, Rosenthal E, Khamashta MA, Huges GR. Absence of
conduction defects in the electrocardiograms of mothers with children with congenital complete heart block. J Rheumatol 2001;28:
366 –369.
21. Costedoat-Chalumeau N, Amoura Z, Houng du LT, Lechat P, Piette JC.
Safety of hydroxychloroquine in pregnant patients with connective tissue
diseases. Review of the literature. Autoimmun Rev 2005;4:111–115.
22. Xiao GQ, Hu K, Boutjdir M. Direct inhibition of expressed cardiac Land T-type calcium channels by IgG from mothers whose children
have congenital heart block. Circulation 2001;103:1599 –1604.
Effect of Growth Hormone on Cardiac Contractility in Patients
With Adult Onset Growth Hormone Deficiency
Goo-Yeong Cho, MD, PhDa,*, In-Kyung Jeong, MD, PhDb, Seong Hwan Kim, MDa,
Min-Kyu Kim, MDa, Woo-Jung Park, MD, PhDa, Dong-Jin Oh, MD, PhDa,
and Hyung-Joon Yoo, MD, PhDa
This study was conducted to investigate the effect of growth hormone (GH) replacement on
cardiac function assessed by standard or tissue Doppler echocardiography in GH deficiency.
Ten patients (mean age 47 ⴞ 14 years) received GH at a dose of 1.0 IU/day (6 times/week).
After 6 months of GH replacement, GH substitution was discontinued. Echocardiography was
performed at baseline, after 6 months of therapy, and 1 year after the withdrawal of GH
replacement. All parameters were compared with those from 11 healthy controls matched for
age, gender, and left ventricular (LV) mass index. After GH replacement, LV ejection fractions
were nonsignificantly increased. However, fractional shortening, LV dimensions, and LV
volumes did not change. Compared with controls, peak strain (ⴚ18.9 ⴞ 4.8% vs ⴚ15.7 ⴞ 6.9%,
p <0.01) and strain rate (ⴚ1.3 ⴞ 0.4/s vs ⴚ1.0 ⴞ 0.5/s, p <0.01) at baseline were significantly
decreased in patients with GH deficiency. Strain and strain rate increased significantly after 6
months of replacement but returned to baseline levels after 12 months off therapy. In conclusion, GH replacement in adult-onset GH deficiency demonstrated beneficial effects on cardiac
contractility assessed by strain and strain rate, but these parameters returned to baseline levels
after the withdrawal of GH. Strain and strain rate can be used to evaluate subtle changes in
myocardium after GH replacement. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol
2007;100:1035–1039)
Growth hormone (GH) deficiency in adults may be associated with increased cardiovascular death.1 Some echocardiographic studies have demonstrated that GH deficiency
decreases left ventricular (LV) mass index and LV systolic
function compared with controls and that these parameters
improve after GH replacement.2 Furthermore, GH replacement results in improvements in hemodynamics in idiopathic dilated cardiomyopathy.3 However, not all studies
have shown positive results of GH replacement.4,5 These
discrepancies may be related to the small sample sizes of
most studies or the fact that conventional echocardiography
may be not sensitive enough to assess myocardial contractility in subtle changes of cardiac effect caused by GH.
Strain or strain rate has been proposed as a sensitive tool to
detect early systolic function abnormalities.6 –9 We investigated the effect of GH replacement and its persistence on
cardiac function assessed by 2-dimensional or tissue Doppler strain echocardiography in adult-onset GH deficiency
with normal global systolic function.
Methods
We studied patients with adult-onset panhypopituitarism
who underwent replacement therapy with hormones other
a
Department of Medicine, Hallym University Sacred Heart Hospital,
Hallym University, Anyang City; and bDepartment of Endocrinology and
Metabolism, College of Medicine, Kyung Hee University, Seoul, South
Korea. Manuscript received February 23, 2007; revised manuscript received and accepted April 13, 2007.
*Corresponding author: Tel: 82-31-380-5922; fax: 82-31-386-2269.
E-mail address: [email protected] (G.-Y. Cho).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.051
than GH as appropriate for ⬎1 year. The adequacy of
hormone replacement therapy was assessed periodically. Of
these, patients who were fully informed of the procedure
and the aim of the study and gave written informed consent
before the study were candidates for inclusion. They were
not admitted to the study if any of the following criteria
were present: (1) history of hypertension or diabetes, (2)
ischemic heart disease or significant valvular heart disease,
(3) a depressed LV ejection fraction (⬍50%), and (4) significant LV hypertrophy calculated using Devereux’s formula (⬎135 g/m2 in men, ⬎110 g/m2 in women).10
Ten patients (mean age 47.1 ⫾ 14.1 years; 6 women)
were evaluated. All patients received GH (somatotropin) at
a starting dose of 1.0 IU/day injected subcutaneously
(6 times/week), with titration according to serum insulinlike growth factor–1 (IGF-1) level (measured by radioimmunoassay; BioSource Europe, S.A., Nivelles, Belgium) to
normalize for age and gender. All patients reached the
optimal level of IGF-1 in the 4 weeks after replacement.
After 6 months of GH replacement, GH substitution was
discontinued. Standard and tissue Doppler echocardiography were performed at baseline (pre-GH), after 6 months of
therapy (post-GH), and 1 year after the discontinuation of
GH replacement (off GH). All echocardiographic parameters were compared with those from 11 healthy controls
(mean age 48.2 ⫾ 19.2 years; 6 women) matched for age,
gender, and LV mass index. The study protocol was approved by the ethics committee of Hallym University.
All images were obtained with a standard ultrasound
machine (System 5; GE Vingmed Ultrasound AS, Horten,
Norway) with a 2.5-MHz probe. Standard techniques were
www.AJConline.org
1036
The American Journal of Cardiology (www.AJConline.org)
Figure 1. Strain (A) and strain rate (B) in the septum. Peak strain and strain rate were defined as the peak negative values on the strain curve during the whole
cycle. Yellow, basal septum; blue, midseptum; red, apical septum.
Table 1
Influence of growth hormone replacement on echocardiographic parameters
Variable
LV end-systolic dimension (cm)
LV end-diastolic dimension (cm)
Fractional shortening (%)
LV mass index (g/m2)
LV end-systolic volume (ml)
LV end-diastolic volume (ml)
LV ejection fraction (%)
E/A ratio
Deceleration time (ms)
Annular velocity (cm/s)
Systole
Early diastole
Atrial contraction
E/early diastolic tissue velocity
Control
Pre-GH
Post-GH
Off GH
2.9 ⫾ 0.4
4.9 ⫾ 0.5
40 ⫾ 5
106 ⫾ 22
30 ⫾ 8
74 ⫾ 18
60 ⫾ 4
1.3 ⫾ 0.5
202 ⫾ 51
3.1 ⫾ 0.5
5.0 ⫾ 0.4
37 ⫾ 6
107 ⫾ 21
28 ⫾ 11
68 ⫾ 26
59 ⫾ 7
1.3 ⫾ 0.4
195 ⫾ 35
3.0 ⫾ 0.4
4.8 ⫾ 0.6
38 ⫾ 7
110 ⫾ 34
24 ⫾ 13
65 ⫾ 24
65 ⫾ 7
1.5 ⫾ 0.6
205 ⫾ 54
3.0 ⫾ 0.3
4.8 ⫾ 0.4
38 ⫾ 6
83 ⫾ 25*
24 ⫾ 9
70 ⫾ 26
65 ⫾ 4
1.2 ⫾ 0.3
190 ⫾ 29
7.2 ⫾ 1.8
8.2 ⫾ 2.8
8.3 ⫾ 1.8
10.6 ⫾ 3.9
6.1 ⫾ 0.6*
6.8 ⫾ 1.5
7.9 ⫾ 1.7
11.5 ⫾ 2.9
7.5 ⫾ 1.6
7.8 ⫾ 2.4
7.9 ⫾ 1.8
10.5 ⫾ 3.2
6.2 ⫾ 1.5*
6.4 ⫾ 1.9*
10.3 ⫾ 7.2
11.8 ⫾ 3.1
* p ⬍0.05 versus post-GH by Wilcoxon’s sign-rank test.
used to obtain M-mode, 2-dimensional, and Doppler measurements in accordance with the guidelines of the American Society of Echocardiography. Tissue Doppler– derived
peak systolic and early and late diastolic velocities were
derived from the septal mitral annulus. LV end-systolic and
end-diastolic volumes along with the ejection fraction were
calculated using the modified Simpson’s method from the
apical 4-chamber view. The percentage of LV fractional shortening was calculated as [(LV end-diastolic dimension ⫺ LV
end-systolic dimension)/LV end-diastolic dimension] ⫻
100.
The parasternal long-axis view was chosen for the analysis
of the integrated backscatter (IB) curve. The sample volume (5
⫻ 5 mm) was placed in the midmyocardium of the basal
anteroseptum and posterior wall and tracked manually to avoid
the inclusion of the specular reflections of the endocardium or
the pericardium during the cardiac cycle. Calibrated IB was
obtained by subtracting average pericardial IB intensity from
the average myocardial IB intensity of the posterior wall.8
Strain or strain rate was measured from tissue Doppler data on
the basis of instantaneous difference in myocardial tissue velocity. For the purposes of this study, each of 4 walls (septum,
lateral, inferior, and anterior wall) of the left ventricle was
analyzed at 3 levels (basal, mid, and apex) from apical 4- and
2-chamber views. A frame rate of 96 to 130 frames/s was used
to acquire the data. All images were stored digitally on mag-
neto-optical discs and analyzed off-line. Strain and strain rate
curves were extracted from color tissue Doppler images using
standard software (EchoPac; GE Vingmed Ultrasound AS;
Figure 1). The sample volume (12 ⫻ 6 mm) was placed in the
midmyocardium and maintained at the same position during
the cardiac cycle by manually tracking to avoid blood or
pericardial contamination. Peak strain and strain rate were
defined as the peak negative value on the strain curve during
the whole cardiac cycle and averaged across all 12 segments
for each patient. Strain analysis was performed by 1 investigator.
All values are expressed as mean ⫾ SD. Data were
analyzed using standard statistical software (SPSS version
13.0; SPSS, Inc., Chicago, Illinois). Differences between
patients and control subjects were compared using the
Mann-Whitney U test or Student’s t test for unpaired data.
Comparisons between baseline (pre-GH) and after 6 months
of GH replacement (post-GH) and after 1 year off therapy
(off GH) were made using Wilcoxon’s sign-rank test or
paired-samples t tests. A p value ⬍0.05 was considered
significant.
Results
The mean systolic and diastolic blood pressures in GH
deficiency were 117 ⫾ 6 and 77 ⫾ 5 mm Hg, respectively,
Miscellaneous/Cardiac Effect of Growth Hormone
1037
Table 2
Influence of growth hormone replacement on tissue characterization and strain analysis
Variable
Heart rate (beats/min)
Calibrated IB (dB)
Anteroseptum
Posterior
Strain (%)
Septum
Lateral
Inferior
Anterior
Average (12 segments)
Strain rate (per second)
Septum
Lateral
Inferior
Anterior
Average (12 segments)
Control
64 ⫾ 9
Pre-GH
Post-GH
Off GH
61 ⫾ 7
62 ⫾ 7
66 ⫾ 5
⫺26.8 ⫾ 4.9
⫺23.4 ⫾ 3.6
⫺25.1 ⫾ 5.5
⫺27.3 ⫾ 5.7
⫺25.4 ⫾ 6.6
⫺27.4 ⫾ 6.6
⫺26.9 ⫾ 3.9
⫺24.5 ⫾ 2.6
⫺19.2 ⫾ 5.2
⫺18.6 ⫾ 4.8
⫺19.7 ⫾ 4.6
⫺19.3 ⫾ 6.0
⫺18.9 ⫾ 4.8
⫺17.4 ⫾ 5.9*
⫺18.2 ⫾ 5.5
⫺17.2 ⫾ 5.4*
⫺17.5 ⫾ 7.9*
⫺15.7 ⫾ 6.9ⴱ,†
⫺18.8 ⫾ 5.8
⫺19.2 ⫾ 6.5
⫺18.8 ⫾ 5.3
⫺18.7 ⫾ 7.2‡
⫺18.5 ⫾ 7.3‡
⫺18.5 ⫾ 6.5
⫺14.6 ⫾ 5.2ⴱ,†
⫺16.8 ⫾ 6.0*
⫺15.3 ⫾ 6.2*
⫺16.3 ⫾ 6.1ⴱ,†
⫺1.2 ⫾ 0.5
⫺1.4 ⫾ 0.5
⫺1.1 ⫾ 0.3
⫺1.4 ⫾ 0.6
⫺1.3 ⫾ 0.4
⫺1.1 ⫾ 0.5
⫺1.3 ⫾ 0.5
⫺1.0 ⫾ 0.3*
⫺1.2 ⫾ 0.7*
⫺1.0 ⫾ 0.5*
⫺1.2 ⫾ 0.4
⫺1.4 ⫾ 0.7
⫺1.1 ⫾ 0.3‡
⫺1.3 ⫾ 0.6‡
⫺1.2 ⫾ 0.6†,‡
⫺1.1 ⫾ 0.4
⫺1.0 ⫾ 0.4ⴱ,†
⫺1.0 ⫾ 0.3*
⫺1.0 ⫾ 0.5*
⫺1.0 ⫾ 0.4ⴱ,†
* p ⬍0.05 versus controls by independent-samples Student’s t test;† p ⬍0.05 versus post-GH by paired-samples t test;‡ p ⬍0.05 versus pre-GH by
paired-samples t test.
Figure 2. Peak strain (percent) (A) and strain rate (per second) (B) before GH replacement (pre-GH), 6 months after GH replacement (post-GH), and after
12 months off therapy (off-GH).
which were not statistically different from those of controls
(121 ⫾ 17 and 72 ⫾ 17 mm Hg) and did not change after
GH replacement. After 6 months of GH replacement, IGF-1
levels were significantly increased (106 ⫾ 36 vs 417 ⫾ 186
ng/ml, p ⫽ 0.008). Among conventional echocardiographic
parameters (Table 1), LV ejection fractions were nonsignificantly increased after 6 months of GH replacement.
However, fractional shortening, LV chamber dimensions
and volumes, and LV mass indexes were not statistically
different from those in the control group and did not change
after GH replacement. Tissue velocity of the mitral annulus
during systole (Sm) was significantly increased after GH
replacement. There was a trend toward an increase in early
diastolic tissue velocity (Em) after GH replacement, but the
difference did not reach statistical significance. However,
Sm and Em returned to baseline conditions after 12 months
off therapy. The other LV diastolic parameters did not differ
after GH replacement or after 12 months off therapy. The
calibrated IB of the anteroseptum and posterior wall was not
significantly different compared with healthy controls and
did not change after 6 months of therapy or 12 months off
therapy (Table 2). Before GH replacement, although LV
ejection fractions were matched with those of healthy controls, average peak strain (⫺18.9 ⫾ 4.8% vs ⫺15.7 ⫾ 6.9%,
p ⬍0.01) and strain rate (⫺1.3 ⫾ 0.4/s vs ⫺1.0 ⫾ ⫺0.5/s,
p ⬍0.01) were significantly decreased in patients with GH
deficiency (Table 3). Peak strain and strain rate were
increased significantly after 6 months of GH replacement,
with values improved up to the levels of healthy controls,
but returned to baseline after 12 months off therapy
(Figure 2).
Discussion
We demonstrated that average peak strain and peak strain
rate were significantly decreased in patients with GH
deficiency despite normal LV systolic function measured
by conventional echocardiography, and decreased cardiac
contractility was improved after 6 months of GH replacement.
1038
The American Journal of Cardiology (www.AJConline.org)
The primary goal of the treatment of GH deficiency in
children is to allow growth to normal height. In adult-onset
GH deficiency, the goals are to relieve symptoms associated
with the deficiency rather than to increase height. GH is also
known to play an important role in heart function and
morphology. However, contrary to treatment in children,
there are significant questions about treating GH deficiency
in adults. The magnitude of improvement may not be worthwhile in terms of cost-effectiveness or long-term safety.11
Epidemiologic studies have shown that GH deficiency
may be associated with cardiovascular death.1 Vasan et al12
prospectively studied the association between serum IGF-1
levels and the incidence of congestive heart failure. These
investigators showed that there was a decrease in the risk for
heart failure with increments in IFG-1. However, cardiac
structural or functional change in GH deficiency has been
the subject of much debate,13 and studies of the cardiac
effects of GH replacement are less consistent.2,4,5,14,15 These
discrepancies may be related to the different durations and
severities of GH deficiency and to the small sample sizes of
these studies.4,13,16,17
Furthermore, although echocardiography is the most
widely used modality for the evaluation of systolic function,
the LV ejection fraction may be not sensitive enough to
detect subclinical myocardial disease. Tissue Doppler–
derived strain and strain rate have been validated18,19 and
can be used for noninvasive indexes for contractility,20
which appears to be extremely effective for the identification of subclinical LV dysfunction in diabetic or obese
patients.6,8 In a more recent study, patients with GH deficiency did not show cardiac structural or functional differences compared with healthy controls, with no significant
changes after GH treatment.4 These results would agree
with ours had we measured only conventional echocardiographic parameters. In this study, although LV systolic
function measured using conventional echocardiography in
GH deficiency was apparently normal and matched with
that of healthy controls, LV contractility assessed with
strain rate was significantly reduced. After GH replacement
and withdrawal, conventional echocardiographic parameters failed to demonstrate myocardial change of GH. In
strain analysis, however, all 4 ventricular walls uniformly
increased after GH replacement and returned to prereplacement levels after withdrawal. This means that short-term
GH replacement may affect a small degree of myocardial
function, which may not manifest clinically and may go
undetected by conventional echocardiography. From this
point of view, strain and strain rate could be used as excellent diagnostic tools for evaluating myocardial change
of GH.
The question still remains in adult-onset GH deficiency
of how long replacement is required once it is initiated. In
our study, the positive effect of short-term GH replacement
on cardiac contractility disappeared after the discontinuation of GH replacement. This finding agrees with a previous
study in that the positive effect of replacement is not persistent.16 LV mass was not changed after GH replacement
and significantly decreased after withdrawal. However, interpretation should be made with caution because of the
small sample size.
We have also evaluated myocardial structural change
after GH replacement using tissue characterization. It has
been reported that GH replacement in adults with childhood-onset GH deficiency results in improvements in cardiac structure assessed by tissue characterization.21 This
result contrasts with our study, in which calibrated IB in GH
deficiency did not change after GH replacement. This discrepancy may be related to the duration of the GH deficiency period or treatment duration, affecting the degree of
myocardial structure. Em by tissue Doppler is directly dependent on loading conditions, LV relaxation, and myocardial structure characterized by interstitial fibrosis.22 In our
study, Em was increased after GH replacement, which
means that GH may have a positive effect on myocardial
structure.
1. Tomlinson JW, Holden N, Hills RK, Wheatley K, Clayton RN, Bates
AS, Sheppard MC, Stewart PM. Association between premature mortality and hypopituitarism. West Midlands Prospective Hypopituitary
Study Group. Lancet 2001;357:425– 431.
2. Maison P, Chanson P. Cardiac effects of growth hormone in adults
with growth hormone deficiency: a meta-analysis. Circulation 2003;
108:2648 –2652.
3. Fazio S, Sabatini D, Capaldo B, Vigorito C, Giordano A, Guida R,
Pardo F, Biondi B, Sacca L. A preliminary study of growth hormone
in the treatment of dilated cardiomyopathy. N Engl J Med 1996;334:
809 – 814.
4. Climent VE, Pico A, Sogorb F, Aznar S, Lip GY, Marin F. Growth
hormone therapy and the heart. Am J Cardiol 2006;97:1097–1102.
5. Meyers DE, Cuneo RC. Controversies regarding the effects of growth
hormone on the heart. Mayo Clin Proc 2003;78:1521–1526.
6. Marwick TH. Tissue Doppler imaging for evaluation of myocardial
function in patients with diabetes mellitus. Curr Opin Cardiol 2004;
19:442– 446.
7. Weidemann F, Strotmann JM. Detection of subclinical LV dysfunction
by tissue Doppler imaging. Eur Heart J 2006;27:1771–1772.
8. Wong CY, O’Moore-Sullivan T, Leano R, Byrne N, Beller E, Marwick TH. Alterations of left ventricular myocardial characteristics
associated with obesity. Circulation 2004;110:3081–3087.
9. Wong CY, O’Moore-Sullivan T, Fang ZY, Haluska B, Leano R,
Marwick TH. Myocardial and vascular dysfunction and exercise capacity in the metabolic syndrome. Am J Cardiol 2005;96:1686 –1691.
10. Devereux RB, Reichek N. Echocardiographic determination of left
ventricular mass in man. Anatomic validation of the method. Circulation 1977;55:613– 618.
11. Orme SM, McNally RJ, Cartwright RA, Belchetz PE. Mortality and
cancer incidence in acromegaly: a retrospective cohort study. United
Kingdom Acromegaly Study Group. J Clin Endocrinol Metab 1998;
83:2730 –2734.
12. Vasan RS, Sullivan LM, D’Agostino RB, Roubenoff R, Harris T,
Sawyer DB, Levy D, Wilson PW. Serum insulin-like growth factor I
and risk for heart failure in elderly individuals without a previous
myocardial infarction: the Framingham Heart Study. Ann Intern Med
2003;139:642– 648.
13. Monson JP, Besser GM. The potential for growth hormone in the
management of heart failure. Heart 1997;77:1–2.
14. Shahi M, Beshyah SA, Hackett D, Sharp PS, Johnston DG, Foale RA.
Myocardial dysfunction in treated adult hypopituitarism: a possible
explanation for increased cardiovascular mortality. Br Heart J 1992;
67:92–96.
15. Minczykowski A, Gryczynska M, Ziemnicka K, Czepczynski R, Sowinski J, Wysocki H. The influence of growth hormone (GH) therapy on
cardiac performance in patients with childhood onset GH deficiency.
Growth Horm IGF Res 2005;15:156 –164.
16. Amato G, Carella C, Fazio S, La MG, Cittadini A, Sabatini D, Marciano-Mone C, Sacca L, Bellastella A. Body composition, bone metabolism, and heart structure and function in growth hormone (GH)deficient adults before and after GH replacement therapy at low doses.
J Clin Endocrinol Metab 1993;77:1671–1676.
Miscellaneous/Cardiac Effect of Growth Hormone
17. Cuocolo A, Nicolai E, Colao A, Longobardi S, Cardei S, Fazio S,
Merola B, Lombardi G, Sacca L, Salvatore M. Improved left ventricular function after growth hormone replacement in patients with hypopituitarism: assessment with radionuclide angiography. Eur J Nucl
Med 1996;23:390 –394.
18. Neilan TG, Jassal DS, Perez-Sanz TM, Raher MJ, Pradhan AD, Buys
ES, Ichinose F, Bayne DB, Halpern EF, Weyman AE, et al. Tissue
Doppler imaging predicts left ventricular dysfunction and mortality in
a murine model of cardiac injury. Eur Heart J 2006;27:1868 –1875.
19. Abraham TP, Laskowski C, Zhan WZ, Belohlavek M, Martin EA,
Greenleaf JF, Sieck GC. Myocardial contractility by strain echocardiography: comparison with physiological measurements in an in vitro
model. Am J Physiol Heart Circ Physiol 2003;285:H2599 –H2604.
1039
20. Greenberg NL, Firstenberg MS, Castro PL, Main M, Travaglini A,
Odabashian JA, Drinko JK, Rodriguez LL, Thomas JD, Garcia MJ.
Doppler-derived myocardial systolic strain rate is a strong index of left
ventricular contractility. Circulation 2002;105:99 –105.
21. Minczykowski A, Gryczynska M, Ziemnicka K, Sowinski J, Wysocki
H. The influence of growth hormone therapy on ultrasound myocardial
tissue characterization in patients with childhood onset GH deficiency.
Int J Cardiol 2005;101:257–263.
22. Shan K, Bick RJ, Poindexter BJ, Shimoni S, Letsou GV, Reardon
MJ, Howell JF, Zoghbi WA, Nagueh SF. Relation of tissue Doppler
derived myocardial velocities to myocardial structure and betaadrenergic receptor density in humans. J Am Coll Cardiol 2000;
36:891– 896.
Crack Whips the Heart: A Review of the Cardiovascular
Toxicity of Cocaine
Luis Afonsob,*, Tamam Mohammada, and Deepak Thataib
Cocaine is an extremely powerful reinforcing psychostimulant with highly addictive properties. Over the last few decades, cocaine addiction has attained epidemic proportions in
North America, imposing a tremendous burden on society and the health care system. The
cardiovascular complications of cocaine abuse are adrenergic mediated and range from
cocaine-associated acute coronary syndromes to aortic dissection and sudden cardiac
death. Concomitant alcohol and cigarette smoking exacerbate the cardiotoxicity of cocaine.
This contemporary review discusses the spectrum of cardiac complications arising from
cocaine use, operant pathophysiologic mechanisms and controversies surrounding the
pharmacotherapy of cocaine-associated acute coronary syndromes. © 2007 Elsevier Inc.
All rights reserved. (Am J Cardiol 2007;100:1040 –1043)
The local anesthetic, stimulatory, medicinal, aphrodisiac,
and salutary effects of the “coca leaf” have been common
knowledge for centuries, mystically revered as a “gift of the
gods” by the ancient Incan civilizations of Peru and other
pre-Columbian Andean societies. In the modern world, cocaine has been a feature of the counterculture for over a
century, afflicting celebrities and public figures alike. It has
remained the most feared illicit drug in the United States
since the 1920s. Over the last few decades, its recreational
use has attained epidemic proportions, being increasingly
intertwined with violence and crime, particularly in the
lower socioeconomic stratum. Cocaine is a highly addictive
and potent sympathomimetic drug with potentially lethal
cardiovascular effects. This review focuses on the pharmacology of this drug, its mechanism of action and addresses
the wide spectrum of cardiovascular complications associated with cocaine.
Epidemiologic Perspective and Scope of the Problem
In the United States, 5 million people report using cocaine
annually on a regular basis,1,2 and in 2003 alone, 34.9 million
Americans, ⱖ12 years reported using cocaine at least once.
Cocaine is reportedly the most commonly abused drug in
patients presenting to emergency departments and accounts for
a quarter of all nonfatal myocardial infarction (MI) in young
adults.3 Cocaine also happens to be the most frequent cause of
drug-related deaths reported by medical examiners. The economic burden on the health system, stemming from the cardiovascular toxicity of cocaine is substantial.
Pharmacologic Considerations
Cocaine (or “crack” in its impure freebase form) is a crystalline tropane alkaloid that is obtained from the leaves of
Divisions of aInternal Medicine and bCardiology, Wayne State University, Detroit, Michigan. Manuscript received April 4, 2007; revised manuscript received and accepted April 16, 2007.
*Corresponding author: Telephone: 313-745-2620; fax: 313-993-8627.
E-mail address: [email protected] (L. Afonso).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.049
the coca plant (Erythroxylon coca), which grows primarily
in South America. Cocaine is available in 2 forms: the
hydrochloride salt and the “free base.” Cocaine hydrochloride is prepared by dissolving the alkaloid in hydrochloric
acid to form a water-soluble powder or granule that decomposes when heated. Freebase cocaine is a form produced
when the user mixes cocaine hydrochloride with a liquid
base, such as baking soda or ammonia; dissolves the resultant alkaloidal cocaine in a solvent, such as ether; and finally
heats it to evaporate the liquid. The result is pure smokeable
cocaine or crack (heat stable, melting point 98°C). Cocaine
(half-life 0.5 to 1.5 hours) is metabolized by plasma and
liver cholinesterases to water-soluble metabolites, primarily
benzoylecgonine and ecgonine methyl ester; these have
half-lives of 5 to 8 and 3.5 to 6 hours, respectively,4 and are
excreted in the urine. Although cocaine can be detected in
blood or urine for a only few hours, its metabolites are
detectable in blood or urine for 24 to 36 hours after ingestion.5 Biochemical analysis of hair (radioimmunoassay) is
touted to be a very sensitive marker of cocaine use in the
preceding weeks or months.6
Pathophysiologic Mechanisms
During depolarization, cocaine inhibits membrane permeability to sodium, thereby blocking the initiation and transmission of electrical signals, a property that accounts for its
anesthetic effects. Regardless of the route of administration,
a high blood concentration of cocaine is achieved due to its
excellent absorption via mucous membranes.
Genesis of ischemia: A variety of mechanisms have been
implicated in the development of cocaine-induced myocardial
ischemia and necrosis. Cocaine blocks the presynaptic uptake
of catecholamines and dopamine, leading to post-synaptic
sympathetic activation and dopaminergic receptor stimulation.5 These sympathomimetic effects result in an augmentation of ventricular contractility, blood pressure, heart rate, and
escalating myocardial oxygen demand. The ensuing supply–
demand deficit may manifest as angina.
www.AJConline.org
Review/Cocaine and Cardiac Complications
Prothrombotic effects: Cocaine increases platelet aggregability and may incite intracoronary thrombosis in the
setting of hypoxia compounded by vasoconstriction of large
epicardial and small coronary resistance vessels.7 Intranasal
cocaine administration is associated with an increase in
plasma plasminogen activator inhibitor (PAI-1), conceivably potentiating vascular thrombosis.8
Mechanisms of vasospasm: Cocaine has been shown to
provoke coronary vasospasm in several small studies performed in the catheterization laboratory.9,10 Even minute
doses used for topical anesthesia have been shown to lead to
coronary vasoconstriction, increases in heart rate and blood
pressure, as well as reduction of coronary sinus blood
flow.10 Cocaine-induced vasoconstriction has been shown to
be ␣-adrenergic receptor mediated, more pronounced in
stenotic as opposed to nonstenotic coronary segments, and
is generally believed to be accentuated by ␤-adrenergic
blockade (unopposed ␣-adrenergic effects). Cocaine-induced vasoconstriction may be facilitated by additional
mechanisms, including the release of plasma endothelin-1
(vasoconstrictor effects) and an impairment of the peripheral production of nitric oxide (local vasodilator effects).
Interestingly, a reduction of coronary flow velocity and
increased coronary resistance has also been demonstrated in
cocaine users in the absence of significant epicardial vessel
spasm, coronary artery disease (CAD) or MI. Recurrent
spasm occurring approximately 90 minutes after initial administration is temporally related to increasing blood concentration of the main cocaine metabolite ethyl methyl
ecgonine.
Accelerated Coronary Atherosclerosis
Premature coronary atherosclerosis is common in young
cocaine abusers and appears to further elevate their risk of
recurrent ischemic events. Obstructive CAD is typically
found in 35% to 55% of patients undergoing coronary
angiography for cocaine-associated chest pain, with no particular predilection to coronary vascular territories. This has
also been demonstrated quantitatively (cross-sectional coronary plaque area) in a case-controlled necropsy study of
cocaine addicts.11 Cocaine has also been reported to cause
coronary artery aneurysms and ectasia, suggesting yet another potential mechanism for MI in these patients.
Cocaine-Associated Chest Pain
Cocaine may be imbibed by smoking, nasal insufflation, or
intravenous injection; the occurrence of an ischemic event is
not related to the amount ingested, the route, or frequency of
administration.5,12 The association between cocaine use and
myocardial ischemia and infarction is well recognized.
Temporally, these events typically occur when the blood
cocaine concentration is at its highest, but may manifest
when the concentration is low or undetectable, as attributed
to the vasoconstrictive effects of cocaine’s major metabolites.9 Chest pain characteristics and presence of traditional
risk factors for atherosclerosis do not help identify patients
who have cocaine-related MI. It is noteworthy that although
chest pain is the most common presenting symptom, the
incidence of enzymatic or biomarker evidence of acute MI
1041
or myonecrosis is relatively low and reported to be approximately 6% in patients presenting with chest pain.13 Myocardial ischemia, infarction, and occasionally rhabdomyolysis or cocaine-related barotraumas (pneumomediastinum,
pneumothorax, or pneumopericardium) explain the presence of chest pain in a small fraction of patients; however,
the etiology of chest pain remains largely obscure and
poorly understood in most patients. Finally, although approximately a third of patients with cocaine-induced MI
develop complications such as congestive heart failure or
arrhythmias, the overall mortality in hospitalized patients
remains exceedingly low.14
Therapeutic Recommendations for Cocaine-Associated
Chest Pain
Calcium channel blockers: Trouve and Nahas15 first
reported the benefit of calcium channel blockers and
showed that nitrendipine increased the survival and protected rat hearts from arrhythmias and other toxic effects of
cocaine. Negus and colleagues16 later reported alleviation of
cocaine-induced coronary vasoconstriction with intravenous
verapamil. Cocaine-induced coronary vasospasm in both
atherosclerotic and normal segments can be abolished by
sublingual nitroglycerin, calcium channel antagonists, and
morphine sulfate,17 and blocked by the adrenergic-receptor
blocker phentolamine.10,18 The American College of Cardiology/American Heart Association 2002 guidelibes recommend nitroglycerin and oral calcium antagonists for patients
with ST-segment elevation or depression that accompanies
ischemic chest discomfort (class I indication), and intravenous calcium antagonists for patients with ST-segment deviation suggestive of ischemia (class IIa indication).19
The beta-blocker controversy: There is considerable
debate as to the role of ␤ blockers in the setting of cocainerelated chest pain or MI. The use of ␤ blockers in this
setting is fraught with controversy largely stemming largely
from the paucity of clinical trials addressing this issue.20,21
Based on the results of a single, double-blind, randomized,
placebo-controlled trial, some observers believe that ␤-adrenergic blockers be avoided altogether as they accentuate
cocaine-induced coronary spasm.20 In this study, 30 patient
volunteers who underwent cardiac catheterization for the evaluation of chest pain were randomized to topically administered
intranasal cocaine or saline followed by intracoronary propranolol or placebo in blinded fashion. Intranasal cocaine-induced
coronary vasoconstriction with further increases in coronary
vascular resistance were noted after the administration of intracoronary propranolol. Labetalol, which has both ␣- and
␤-adrenergic blocking activity, in comparison, reverses the
cocaine-induced increase in arterial pressure but does not
alleviate cocaine-induced coronary vasoconstriction.22 The
American College of Cardiology/American Heart Association
guidelines19 advocate the use of ␤ blockers (class II A indication, level of evidence: C) for cocaine-associated chest pain in
the presence of hypertension or sinus tachycardia. No data on
the effects of ␤ blockers in the setting of cocaine-related MI
could be found in the literature.
Benzodiazepines reduce heart rate and systemic arterial
pressure, and in animals, attenuate cocaine’s toxic effects on
the heart; thus, representing a viable alternative to nitro-
1042
The American Journal of Cardiology (www.AJConline.org)
glycerin. Finally, it is reasonable to administer aspirin to
patients with cocaine-induced myocardial ischemia to inhibit platelet aggregation.
Role of thrombolytics: There are limited data on the
safety of thrombolytic therapy and reports of severe complications associated with its use in cocaine users.23 The
frequent presence of contraindications to thrombolysis, including severe hypertension, seizures, intracerebral hemorrhage, and aortic dissection in cocaine abusers precludes the
liberal use of thrombolytic agents. Moreover, the lack of
specificity of standard electrocardiographic criteria (ST elevation) to identify cocaine-induced MI predicates the cautious use of thrombolytics, after treatment with oxygen,
aspirin, nitrates, and benzodiazepines have failed, particularly, if immediate coronary angiography and angioplasty
are not feasible.5 The American College of Cardiology/
American Heart Association guidelines19 advocate the use
of thrombolytic therapy if ST segments remain elevated
despite nitroglycerin and calcium antagonists and coronary
arteriography is not possible (class II A indication).
Recommendations for coronary arteriography: Immediate coronary angiography is advisable in patients with
ST-segment elevation persisting after administration of nitroglycerine and calcium antagonists (Class I, level of evidence C). If available, angiography is also recommended if
de novo ST-segment depression or isolated T-wave changes
do not respond to the previously mentioned medical therapy
(class II a, level of evidence C).19
Cocaine-Related Myocardial Dysfunction
Cocaine may cause an acute or chronic deterioration of left
ventricular performance. Acute left ventricular systolic and
diastolic function has been attributed to the effects of cocaine or its metabolites in myocyte calcium handling and
has been documented after binge use.24 Focal myocyte necrosis, focal myocarditis, sarcoplasmic vacuolization, and
myofibrillar loss have been variably demonstrated in myocardial biopsy specimens obtained in the setting of acute
cocaine toxicity.
Acute myocardial depression may also manifest as left
ventricular apical ballooning syndrome or Tako-tsubo Syndrome an entity associated with high circulating levels of
catecholamines, myocyte injury, and microvascular dysfunction with close histopathologic resemblances to cocaine-mediated cardiotoxicity. Studies in animals have shown that cocaine
alters cytokine production in the endothelium and in circulating leukocytes, inducing the transcription of genes responsible
for changes in the composition of myocardial collagen and
myosin, and myocyte apoptosis. Chronic left ventricular systolic dysfunction is also a well-known occurrence in a small
percent (7%) of asymptomatic long-term abusers.24
Electrocardiographic Changes and Dysrythmias
The interpretation of electrocardiograms in patients with
cocaine-associated chest pain can be challenging. Cocaineinduced MI has been documented in patients with normal as
well as abnormal electrocardiograms.Conversely, a significant proportion of patients without MI meet the electrocar-
diographic criteria for ST-elevation MI. The sensitivity of
the electrocardiogram for detecting MI is reportedly as low
as 36%.25
Cocaine has direct effects on blocking the sodium channels and may produce or exacerbate cardiac arrhythmias,
including sinus tachycardia, sinus bradycardia, supraventricular tachycardias, asystole, accelerated idioventricular
rhythm, ventricular tachycardia, ventricular fibrillation, and
Torsade de pointes.5 In canine experiments, cocaine has
been demonstrated to reduce ventricular fibrillation thresholds. It is speculated that the increase in left ventricular
mass (associated with long-term cocaine use) and the development of contraction band necrosis serve as the underlying anatomic substrate, increasing the propensity for ischemia and arrhythmias.
The cocaine-associated long QTc interval might be related to the effects of cocaine and its metabolites on conduction in the Human Ether-a-go-go Related Gene (HERG)encoded potassium channel.26 A Brugada pattern (right
bundle branch block with ST elevation in leads V1,V2,V3)
provoked by cocaine has recently gained recognition and is
believed to occur from modulation or unmasking of the
sodium channels.27 Wide-complex dysrhythmias responding to sodium bicarbonate (pH-dependency) may also occur
after exposure to cocaine as a result of the direct effects of
cocaine on the sodium channels.
Polysubstance Abuse: Cocaine, Tobacco,
and Alcohol Interactions
Concomitant cigarette smoking has been shown to exacerbate the deleterious effects (adrenergically-mediated) of cocaine on myocardial supply– demand balance.28 This combination increases the myocardial metabolic requirements
for oxygen while simultaneously decreasing the diameter of
the diseased coronary artery.21 Likewise, the combination of
cocaine and ethanol is considered be more lethal than either
substance alone as it increases myocardial oxygen demand
and is reported to be the most frequent substance abuse
combination encountered in the emergency department. The
transesterification of cocaine and ethanol to the metabolite
cocaethylene occurs in the liver. Cocaethylene is considered
more lethal than cocaine because it blocks the re-uptake of
dopamine and appears to mediate the increase in catastrophic cardiovascular complications observed with this
combination.
Endocarditis
Cocaine, because it is administered intravenously more than
any other illicit drug, causes endocarditis, particularly involving the left-sided cardiac valves.29 The mechanism of
this propensity for endocarditis is unclear, but may relate to
a hemodynamically-mediated injury of cardiac valves, facilitating bacterial invasion,5 possibly compounded by the
purported immunosuppressive effects of cocaine.
Aortic Dissection
This is a diagnosis that should be considered in patients with
severe cocaine-associated chest pain. Most patients tend to
be young and have poorly controlled or untreated hyperten-
Review/Cocaine and Cardiac Complications
sion. The cocaine-induced surge in catecholamine levels
presumably leads to elevated shear-stress forces, increasing
the proclivity for intimal tears and aortic dissection. In a
retrospective study by Hsue and colleagues,30 cocaine was
implicated in 37% of aortic dissections studied in an inner
city population. The location of dissection seems to be
equally distributed between types A and B, as is the case
with dissections not related to cocaine. Aortic intramural
hematoma and coronary artery dissection have also been
reported to be associated with cocaine abuse.
1. Cregler LL, Mark H. Relation of acute myocardial infarction to cocaine abuse. Am J Cardiol 1985;56:794.
2. Hughes AL. The prevalence of illicit drug use in six metropolitan areas
in the United States: results from the 1991 National Household Survey
on Drug Abuse. Br J Addict 1992;87:1481–1485.
3. Qureshi AI, Suri MF, Guterman LR, Hopkins LN. Cocaine use and the
likelihood of nonfatal myocardial infarction and stroke: data from the
Third National Health and Nutrition Examination Survey. Circulation
2001;103:502–506.
4. Hollander JE, Hollander RS. In: Goldfrank LR, ed. Goldfrank’s Toxicologic Emergencies. New York: McGraw-Hill Medical Pub. Division, 2002:1004 –1019.
5. Lange RA, Hillis LD. Cardiovascular complications of cocaine use.
N Engl J Med 2001;345:351–358.
6. Ness RB, Grisso JA, Hirschinger N, Markovic N, Shaw LM, Day NL,
Kline J. Cocaine and tobacco use and the risk of spontaneous abortion.
N Engl J Med 1999;340:333–339.
7. Minor RL, Jr., Scott BD, Brown DD, Winniford MD. Cocaine-induced
myocardial infarction in patients with normal coronary arteries. Ann
Intern Med 1991;115:797– 806.
8. Moliterno DJ, Lange RA, Gerard RD, Willard JE, Lackner C, Hillis
LD. Influence of intranasal cocaine on plasma constituents associated
with endogenous thrombosis and thrombolysis. Am J Med 1994;96:
492– 496.
9. Brogan WC III, Lange RA, Glamann DB, Hillis LD. Recurrent coronary vasoconstriction caused by intranasal cocaine: possible role for
metabolites. Ann Intern Med 1992;116:556 –561.
10. Lange RA, Cigarroa RG, Yancy CW, Jr., Willard JE, Popma JJ, Sills
MN, McBride W, Kim AS, Hillis LD. Cocaine-induced coronaryartery vasoconstriction. N Engl J Med 1989;321:1557–1562.
11. Dressler FA, Malekzadeh S, Roberts WC. Quantitative analysis of
amounts of coronary arterial narrowing in cocaine addicts. Am J
Cardiol 1990;65:303–308.
12. Kloner RA, Rezkalla SH. Cocaine and the heart. N Engl J Med
2003;348:487– 488.
13. Weber JE, Chudnofsky CR, Boczar M, Boyer EW, Wilkerson MD,
Hollander JE. Cocaine-associated chest pain: how common is myocardial infarction? Acad Emerg Med 2000;7:873– 877.
14. Hollander JE, Hoffman RS, Burstein JL, Shih RD, Thode HC, Jr.
Cocaine-associated myocardial infarction. Mortality and complications. Cocaine-Associated Myocardial Infarction Study Group. Arch
Intern Med 1995;155:1081–1086.
1043
15. Trouve R, Nahas G. Nitrendipine: an antidote to cardiac and lethal
toxicity of cocaine. Proc Soc Exp Biol Med 1986;183:392–397.
16. Negus BH, Willard JE, Hillis LD, Glamann DB, Landau C, Snyder
RW, Lange RA. Alleviation of cocaine-induced coronary vasoconstriction with intravenous verapamil. Am J Cardiol 1994;73:510 –513.
17. Saland KE, Hillis LD, Lange RA, Cigarroa JE. Influence of morphine
sulfate on cocaine-induced coronary vasoconstriction. Am J Cardiol
2002;90:810 – 811.
18. Brogan WC III, Lange RA, Kim AS, Moliterno DJ, Hillis LD. Alleviation of cocaine-induced coronary vasoconstriction by nitroglycerin.
J Am Coll Cardiol 1991;18:581–586.
19. Braunwald E, Antman EM, Beasley JW, Califf RM, Cheitlin MD,
Hochman JS, Jones RH, Kereiakes D, Kupersmith J, Levin TN, et al.
ACC/AHA 2002 guideline update for the management of patients with
unstable angina and non-ST-segment elevation myocardial infarction–
summary article: a report of the American College of Cardiology/
American Heart Association task force on practice guidelines (Committee on the Management of Patients With Unstable Angina). J Am
Coll Cardiol 2002;40:1366 –1374.
20. Lange RA, Cigarroa RG, Flores ED, McBride W, Kim AS, Wells PJ,
Bedotto JB, Danziger RS, Hillis LD. Potentiation of cocaine-induced
coronary vasoconstriction by beta-adrenergic blockade. Ann Intern
Med 1990;112:897–903.
21. Moliterno DJ, Willard JE, Lange RA, Negus BH, Boehrer JD, Glamann DB, Landau C, Rossen JD, Winniford MD, Hillis LD. Coronaryartery vasoconstriction induced by cocaine, cigarette smoking, or both.
N Engl J Med 1994;330:454 – 459.
22. Boehrer JD, Moliterno DJ, Willard JE, Hillis LD, Lange RA. Influence
of labetalol on cocaine-induced coronary vasoconstriction in humans.
Am J Med 1993;94:608 – 610.
23. Hoffman RS, Hollander JE. Thrombolytic therapy and cocaine-induced myocardial infarction. Am J Emerg Med 1996;14:693– 695.
24. Bertolet BD, Freund G, Martin CA, Perchalski DL, Williams CM,
Pepine CJ. Unrecognized left ventricular dysfunction in an apparently
healthy cocaine abuse population. Clin Cardiol 1990;13:323–328.
25. Hollander JE, Hoffman RS, Gennis P, Fairweather P, DiSano MJ,
Schumb DA, Feldman JA, Fish SS, Dyer S, Wax P, et al. Prospective
multicenter evaluation of cocaine-associated chest pain. Cocaine Associated Chest Pain (COCHPA) Study Group. Acad Emerg Med 1994;
1:330 –339.
26. Ferreira S, Crumb WJ, Jr., Carlton CG, Clarkson CW. Effects of
cocaine and its major metabolites on the HERG-encoded potassium
channel. J Pharmacol Exp Ther 2001;299:220 –226.
27. Daga B, Minano A, de la Puerta I, Pelegrin J, Rodrigo G, Ferreira I.
Electrocardiographic findings typical of Brugada syndrome unmasked
by cocaine consumption. Rev Esp Cardiol 2005;58:1355–1357.
28. Winniford MD, Wheelan KR, Kremers MS, Ugolini V, van den Berg
E, Jr., Niggemann EH, Jansen DE, Hillis LD. Smoking-induced coronary vasoconstriction in patients with atherosclerotic coronary artery
disease: evidence for adrenergically mediated alterations in coronary
artery tone. Circulation 1986;73:662– 667.
29. Chambers HF, Morris DL, Tauber MG, Modin G. Cocaine use and the
risk for endocarditis in intravenous drug users. Ann Intern Med 1987;
106:833– 836.
30. Hsue PY, Salinas CL, Bolger AF, Benowitz NL, Waters DD. Acute aortic
dissection related to crack cocaine. Circulation 2002;105:1592–1595.
The “Clopidogrel Resistance” Trap
Victor L. Serebruany, MD, PhDa,ⴱ
Numerous randomized trials have indicated the benefits
of clopidogrel either as an alternative1 or as an adjunct2 to
aspirin for the secondary prevention of acute vascular
events, including absolute mortality reduction in the largest
ever acute myocardial infarction study.3 Despite proved
efficacy, antiplatelet protection with clopidogrel has several
potential limitations, such as delayed onset of platelet inhibition even after loading regimens,4,5 substantial response
variability in the acute setting,6,7 remaining risk for the
development of vascular thrombosis,8,9 and higher rates of
perioperative bleeding complications during cardiac surgery10,11 because of the irreversible nature of platelet P2Y12
receptor blockade. It is unclear to what extent clopidogrel
per se is responsible for all these shortcomings, how damaging are they in real-life clinical scenarios, and, most
important, what can be done to prevent, minimize, or compensate for such limitations. All real or perceived limitations associated with response to clopidogrel can be divided
into 2 categories: those driven by measuring multiple biomarkers in the platelet studies (the variability and durability of response, excess timing needed to exhibit full-scale
antiplatelet potency, and inefficient inhibition due to increased baseline preexisting platelet activity) and those scenarios really observed in clinical practice (recurrent vascular events, including stent thrombosis, and increased
bleeding risks). Insufficient platelet inhibition with clopidogrel has been termed “clopidogrel resistance.” However,
such “resistance” still remains a laboratory research finding
rather than a proved, clinically relevant fact, despite numerous attempts to link low response to clopidogrel with worsened vascular outcomes12,13 in general and the development
of stent thrombosis14,15 in particular. All these small studies
fall short of proving that changes in certain platelet biomarkers may predict outcomes after clopidogrel, because they
are overpowered by discrepancies with available randomized clinical data and contradicting epidemiologic evidence.
It seems that noncompliance is a major and the most
logical practical reason for nonresponse to clopidogrel.
With regard to compliance, it is critical to divide the evidence into acute (in-hospital) and maintenance (outpatient)
long-term settings. In fact, clopidogrel administration is
controlled much better in the hospital than in outpatients.
Therefore, platelet data suggesting “resistance” may have
merit when properly assessed during loading regimens because clopidogrel is indeed on board. However, these studies cannot overcome the power and validity of the Clopidogrel and Metoprolol in Myocardial Infarction Trial
(COMMIT),3 in which a combination of moderate-dose
a
HeartDrug Research Laboratories, Johns Hopkins University, Towson,
Maryland. Manuscript received January 25, 2007; revised manuscript
received and accepted April 13, 2007.
*Corresponding author: Tel: 410-847-9490; fax: 443-583-0205.
E-mail address: [email protected] (V.L. Serebruany).
0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjcard.2007.04.050
aspirin, clopidogrel, and streptokinase saved 119 lives in
patients with acute myocardial infarction over those treated
with aspirin and a fibrinolytic agent only. Critical to remember is that absolute mortality benefit has been achieved
exclusively in patients who received no-load (75 mg) clopidogrel and that most such patients will be considered
“clopidogrel resistant” if assessed by any platelet tests.16
Should “resistance” be real, COMMIT will not yield the
best outcome among all clopidogrel trials. In the outpatient
setting, many noncompliant patients will be considered
“clopidogrel resistant.” In fact, no platelet study ever controlled for compliance to clopidogrel. Obviously, pill counts
and telephone interviews are not sufficient to document
strict compliance. The determination of active (thiol) and/or
intact (carboxyl) clopidogrel metabolites with the simultaneous assessment of platelet activity in autologous samples
is mandatory to prove that a patient experiencing a second
vascular event indeed takes clopidogrel. One established
team is working in the right direction, linking changes in
platelet biomarkers with the plasma levels of clopidogrel
metabolites,17,18 although the data are not yet sufficient to
draw any definite conclusions. Compliance monitoring is
especially important considering alarming double-digit rates
of noncompliance reported for ␤ blockers,19 calcium channel antagonists,20 angiotensin receptor blockers,21 angiotensin-converting enzyme inhibitors,22 selective serotonin reuptake inhibitors,23 and even statins.24 Compliance with
antiplatelet agents is less known but also is common, mainly
because of the preventive (expected, not immediate) nature
of the possible benefit and because of annoying minor
bleeding episodes, making routine tasks as shaving or
brushing the teeth challenging. The combined analyses of
Sibrafiban Versus Aspirin to Yield Maximum Protection
from Ischemic Heart Events Post-Acute Coronary Syndromes (SYMPHONY) and 2nd SYMPHONY trials revealed that ⱖ17% of patients admitted discontinuing aspirin.25 Noncompliance with ticlopidine reached 11% in
patients after stroke when controlled by platelet aggregometry.26 There are few available reports of noncompliance
with clopidogrel. Unjustified cessation of clopidogrel therapy has been observed in ⬎15% of patients with coronary
artery disease27 and in 18.4% (at 3 months) and up to 38.4%
(at 1 year) in a cohort of patients after stroke.28 These high
rates of noncompliance are far greater than any reasonably
determined rates of “clopidogrel resistance.” Therefore, the
postulate that no response, or low response, after clopidogrel may cause worsened vascular outcomes is not valid.
Quite the opposite: the logical explanation of such an adverse association is that excess vascular events occur more
frequently not in “resistant” patients but in patients not
treated with or who discontinue antiplatelet agents. Moreover, if minor bleeding events are responsible for drug
withdrawal, such patients will most likely stop taking not
only clopidogrel but aspirin as well. This chain of events
may lead to rebound platelet activation and second acute
www.AJConline.org
Editorial/Response After Clopidogrel and Clinical Outcomes
vascular events, as documented for cyclooxygenase-2 inhibitors and nonsteroidal anti-inflammatory drugs,29 aspirin,30 and clopidogrel.31 Taken together, it is reasonable to
suspect that even minor bleeding complications are enough
of a deterrent to stop therapy for some patients, especially
when the benefits of the drug are not readily apparent. This
limitation of clopidogrel stands in contrast to drugs that
alleviate actual symptoms, rather than merely preventing
acute events. Obviously, we cannot expect platelets to be
inhibited when an antiplatelet agent is not on board. Moreover, the reported rates of noncompliance are higher than
those of “clopidogrel resistance.”
The cost of an error in judging why a patient with activated
platelets develops a vascular event is enormous but happens in
everyday clinical practice much more often than one might
imagine. Indeed, if “clopidogrel resistance” is a real, meaningful finding, then higher loading and maintenance doses of
clopidogrel, and the introduction of much more potent antiplatelet strategies with prasugrel and AZD 6140, are well
justified and will result in better outcomes. However, should
“resistance” be a laboratory artifact frequently observed in
noncompliant patients, then higher doses, and/or more aggressive antiplatelet regimens are harmful and will not only cause
more bleeding but result in higher drug discontinuation rates
and rebound platelet activation, followed by worsened vascular
outcomes. Considering modern trends to use aggressive (although unjustified by randomized outcome evidence) doubled,
or even tripled, clopidogrel loading doses, combined with the
controversy regarding the greater thrombotic risks observed
with drug-eluting stents, promoting “clopidogrel resistance”
hurts rather than helps patients, exposing them to increased
bleeding and thrombotic risks. Moreover, this hysteria is developing on top of the wide use of glycoprotein IIb/IIIa inhibitors, fibrinolytic agents,32 and bivalirudin,33 which are also
potent antiplatelet agents. Therefore, the hypothesis that “clopidogrel resistance” causes vascular thrombosis is widespread
but is far from being conclusive without a definitive outcome
study is and not supported by randomized clinical evidence.
After COMMIT,3 the strongest arguments challenging
the “clopidogrel resistance” concept can be found in the
results of Management of Atherothrombosis With Clopidogrel in High-Risk Patients (MATCH)34 and Clopidogrel
for High Atherothrombotic Risk, Ischemic Stabilization,
Management, and Avoidance (CHARISMA).35 The lack of
efficacy in patients after stroke and for primary prevention
was not caused by low response to clopidogrel. Quite the
opposite: significant increases in bleeding events were observed in the 2 trials, suggesting that platelet inhibition after
no-load clopidogrel regimens was adequate, or even excessive. Should “resistance” be indeed responsible for the lack
of efficacy in these studies, then bleeding rates will be
lower, or at least unchanged. There is also no reasonable
proof that “clopidogrel resistance” is associated with stent
thrombosis. Controlled studies are lacking, and platelet data
are not supported by levels of clopidogrel metabolites confirming compliance, but there are 2 additional arguments
that preclude linking “clopidogrel resistance” with stent
closure. First, the expansion of drug-eluting stents per se is
most likely a trade-off between restenosis and rethrombosis,
suggesting a potential mortality disadvantage compared
with bare-metal stents.36 Higher rates of late in-stent throm-
1045
Table 1
Clinical and epidemiologic evidence against “clopidogrel resistance”
Finding
Randomized evidence
Absolute mortality benefit with no-load regimen
Excess bleeding and lack of efficacy after stroke
Excess bleeding and lack of efficacy in primary
prevention
Epidemiologic evidence
Higher patient noncompliance than “resistance” rates
Lower stent thrombosis than “resistance” rates
Source
COMMIT3
MATCH31
CHARISMA32
Various19–28
Various7,12,36–42
bosis may indeed require more aggressive antiplatelet regimens, but in-stent thrombosis is probably caused by the
prothrombotic properties of slow-released eluting agents
and has nothing to do with “resistance” to clopidogrel.
Second, despite the controversy over drug-eluting stents, the
rates of stent thrombosis are still as low as 0.5%,37 0.6%,38
1.0%,39 and 1.27%,27 but no more than 3.4%.40 In contrast,
the rates of “clopidogrel resistance” are much higher: low at
4.2%,7 mild at 5% to 11%,41 and high at 25% to 30%,12,42
depending on the platelet test used, patient selection, and
compliance. Therefore, only a small portion of “resistant”
patients develop stent thrombosis, challenging the concept
that these events are directly related. Major facts challenging the importance of “clopidogrel resistance” as a clinically
relevant phenomenon are listed in Table 1.
On the basis of the available evidence, it seems that the
entire issue of “clopidogrel resistance” has recently grown
out of proportion and should be avoided, as suggested by the
International Society of Thrombosis and Haemostasis, the
European Society of Cardiology, and the American Association of Chest Physicians, unless further randomized evidence
suggests the opposite. Last but not least, there are already
plenty of controversies, ranging from real threats of second
thrombotic events to unjustified excessive bleeding risks for
patients treated with escalating doses of clopidogrel. There is
no need to add more confusion; moreover, no platelet data will
help solve or even further advance this controversy. Only a
multicenter, randomized study with a hard outcome or, ideally,
survival end point, supported by comprehensive serial platelet
assessment and strict compliance rules including the measurement of clopidogrel metabolites, will determine whether “clopidogrel resistance” is a real danger (as suggested by platelet
biomarkers) or an artificial tool (as suggested by randomized
clinical evidence) introduced to help novel antiplatelet agents
gain the vascular market share.
1. CAPRIE Steering Committee. A randomized, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events. Lancet
1996;348:1329 –1339.
2. CURE Trial Investigators. Effects of clopidogrel in addition to aspirin
in patients with acute coronary syndromes without ST-segment elevation. N Engl J Med 2001;345:494 –502.
3. COMMIT Collaborative Group. Addition of clopidogrel to aspirin in
45,852 patients with acute myocardial infarction: randomised placebocontrolled trial. Lancet 2005;366:1607–1621.
4. Nguyen TA, Diodati JG, Pharand C. Resistance to clopidogrel: a
review of the evidence. J Am Coll Cardiol 2005;45:1157–1164.
5. Lev EI, Patel RT, Maresh KJ, Guthinkonda S, Granada J, DeLao T,
Bray PF, Kleiman NS. Aspirin and clopidogrel drug response in
patients undergoing percutaneous coronary intervention: the role of
dual drug resistance. J Am Coll Cardiol 2006;47:27–33.
1046
The American Journal of Cardiology (www.AJConline.org)
6. Gachet C, Aleil B. The inter-individual variability of the response to
clopidogrel. Arch Mal Coeur Vaiss 2005;98:216 –225.
7. Serebruany VL, Steinhubl SR, Berger PB, Malinin AI, Bhatt DL,
Topol EJ. Variability in platelet responsiveness to clopidogrel among
544 individuals. J Am Coll Cardiol 2005;45:246 –251.
8. Vats HS, Hocking WG, Rezkalla SH. Suspected clopidogrel resistance
in a patient with acute stent thrombosis. Nat Clin Pract Cardiovasc
Med 2006;4:226 –230.
9. Duffy B, Bhatt DL. Antiplatelet agents in patients undergoing percutaneous coronary intervention: how many and how much? Am J Cardiovasc Drugs 2005;5:307–318.
10. Kapetanakis EI, Medlam DA, Petro KR, Haile E, Hill PC, Dullum
MK, Bafi AS, Boyce SW, Corso PJ. Effect of clopidogrel premedication in off-pump cardiac surgery: are we forfeiting the benefits of
reduced hemorrhagic sequelae? Circulation 2006;113:1667–1674.
11. von Heymann C, Redlich U, Moritz M, Sander M, Vargas Hein O,
Grubitzsch H, Konertz WF, Spies C. Aspirin and clopidogrel taken
until 2 days prior to coronary artery bypass graft surgery is associated
with increased postoperative drainage loss. Thorac Cardiovasc Surg
2005;53:341–345.
12. Matetzky S, Shenkman B, Guetta V, Shechter M, Bienart R, Goldenberg I, Novikov I, Pres H, Savion N, Varon D, Hod H. Clopidogrel
resistance is associated with increased risk of recurrent atherothrombotic events in patients with acute myocardial infarction. Circulation
2004;109:3171–3175.
13. Geisler T, Langer H, Wydymus M, Gohring K, Zurn C, Bigalke B,
Stellos K, May AE, Gawaz M. Low response to clopidogrel is associated with cardiovascular outcome after coronary stent implantation.
Eur Heart J 2006;27:2420 –2425.
14. Wenaweser P, Dorffler-Melly J, Imboden K, Windecker S, Togni M,
Meier B, Haeberli A, Hess OM. Stent thrombosis is associated with an
impaired response to antiplatelet therapy. J Am Coll Cardiol 2005;45:
1748 –1752.
15. Barragan P, Bouvier JL, Roquebert PO, Macaluso G, Commeau P,
Comet B, Lafont A, Camoin L, Walter U, Eigenthaler M. Resistance
to thienopyridines: clinical detection of coronary stent thrombosis by
monitoring of vasodilator-stimulated phosphoprotein phosphorylation.
Catheter Cardiovasc Interv 2003;59:295–302.
16. Serebruany VL. Mortality benefit of no-load clopidogrel in COMMIT:
not a surprise. J Cardiovasc Pharm Ther 2006;11:99 –100.
17. von Beckerath N, Taubert D, Pogatsa-Murray G, Wieczorek A,
Schomig E, Schomig A, Kastrati A. A patient with stent thrombosis,
clopidogrel-resistance and failure to metabolize clopidogrel to its active metabolite. Thromb Haemost 2005;93:789 –791.
18. von Beckerath N, Taubert D, Pogatsa-Murray G, Schomig E, Kastrati
A, Schomig A. Absorption, metabolization, and antiplatelet effects of
300-, 600-, and 900-mg loading doses of clopidogrel: results of the
ISAR-CHOICE (Intracoronary Stenting and Antithrombotic Regimen:
Choose Between 3 High Oral Doses for Immediate Clopidogrel Effect)
trial. Circulation 2005;112:2946 –2950.
19. Kramer JM, Hammill B, Anstrom KJ, Fetterolf D, Snyder R, Charde
JP, Hoffman BS, Allen LaPointe N, Peterson E. National evaluation of
adherence to beta-blocker therapy for 1 year after acute myocardial
infarction in patients with commercial health insurance. Am Heart J
2006;152:454.e1– 454.e8.
20. Wogen J, Frech F. Patient adherence with hypertension medication. J
Manag Care Pharm 2004;10:90 –91.
21. Burke TA, Sturkenboom MC, Lu SE, Wentworth CE, Lin Y, Rhoads
GG. Discontinuation of antihypertensive drugs among newly diagnosed hypertensive patients in UK general practice. J Hypertens 2006;
24:1193–2000.
22. Eagle KA, Kline-Rogers E, Goodman SG, Gurfinkel EP, Avezum A,
Flather MD, Granger CB, Erickson S, White K, Steg PG. Adherence to
evidence-based therapies after discharge for acute coronary syndromes: an
ongoing prospective, observational study. Am J Med 2004;117:73– 81.
23. Cantrell CR, Eaddy MT, Shah MB, Regan TS, Sokol MC. Methods for
evaluating patient adherence to antidepressant therapy: a real-world
comparison of adherence and economic outcomes. Med Care 2006;
44:300 –303.
24. Ho PM, Spertus JA, Masoudi FA, Reid KJ, Peterson ED, Magid DJ,
Krumholz HM, Rumsfeld JS. Impact of medication therapy discontinuation on mortality after myocardial infarction. Arch Intern Med 2006;
166:1842–1847.
25. Newby LK, Bhapkar MV, White HD, Moliterno DJ, LaPointe NM,
Kandzari DE, Verheugt FW, Kramer JM, Armstrong PW, Califf RM;
SYMPHONY and 2nd SYMPHONY Investigators. Aspirin use postacute coronary syndromes: intolerance, bleeding and discontinuation.
J Thromb Thrombolys 2003;16:119 –128.
26. Komiya T, Kudo M, Urabe T, Mizuno Y. Compliance with antiplatelet
therapy in patients with ischemic cerebrovascular disease. Assessment
by platelet aggregation testing. Stroke 1994;25:2337–2342.
27. Kuchulakanti PK, Chu WW, Torguson R, Ohlmann P, Rha SW, Clavijo
LC, Kim SW, Bui A, Gevorkian N, Xue Z, et al. Correlates and long-term
outcomes of angiographically proven stent thrombosis with sirolimus- and
paclitaxel-eluting stents. Circulation 2006;113:1108 –1113.
28. Hamann GF, Weimar C, Glahn J, Busse O, Diener HC; German Stroke
Data Bank. Adherence to secondary stroke prevention strategies—
results from the German Stroke Data Bank. Cerebrovasc Dis 2003;
15:282–288.
29. Serebruany VL, Malinin AE, Bhatt D. Paradoxical rebound platelet
activation after painkillers cessation: missing risk for vascular events?
Am J Med 2006;119:11–16.
30. Biondi-Zoccai GG, Lotrionte M, Agostoni P, Abbate A, Fusaro M,
Burzotta F, Testa L, Sheiban I, Sangiorgi G. A systematic review and
meta-analysis on the hazards of discontinuing or not adhering to
aspirin among 50,279 patients at risk for coronary artery disease. Eur
Heart J 2006;27:2667–2674.
31. Angiolillo DJ, Fernandez-Ortiz A, Bernardo E, Ramirez C, Sabate M,
Jimenez-Quevedo P, Hernandez R, Moreno R, Escaned J, Alfonso F,
et al. Clopidogrel withdrawal is associated with proinflammatory and
prothrombotic effects in patients with diabetes and coronary artery
disease. Diabetes 2006;55:780 –784.
32. Serebruany VL, Malinin AI, Callahan KP, Binbreck AS, van de Werf
F, Alexander JH, Granger CB, Gurbel PA. Effect of tenecteplase
versus alteplase on platelets in patients during the first three hours of
treatment of acute myocardial infarction (the ASSENT-2 Platelet Substudy). Am Heart J 2003;245:636 – 642.
33. Verstraete M. Modulating platelet function with selective thrombin
inhibitors. Haemostasis 1996;26(suppl):70 –77.
34. Diener HC, Bogousslavsky J, Brass LM, Cimminiello C, Csiba L,
Kaste M, Leys D, Matias-Guiu J, Rupprecht HJ; MATCH Investigators. Aspirin and clopidogrel compared with clopidogrel alone after
recent ischaemic stroke or transient ischaemic attack in high-risk
patients (MATCH): randomised, double-blind, placebo-controlled
trial. Lancet 2004;364:331–337.
35. CHARISMA Trial Investigators. Clopidogrel and aspirin versus aspirin alone for the prevention of atherothrombotic events. N Engl J Med
2006;354:1706 –1717.
36. Bavry AA, Kumbhani DJ, Helton TJ, Borek PP, Mood GR, Bhatt DL.
Late thrombosis of drug-eluting stents: a meta-analysis of randomized
clinical trials. Am J Med 2006;119:1056 –1061.
37. Lee CW, Park KH, Kim YH, Hong MK, Kim JJ, Park SW, Park SJ.
Clinical and angiographic outcomes after placement of multiple overlapping drug-eluting stents in diffuse coronary lesions. Am J Cardiol
2006;98:1028 –1032.
38. Abizaid A, Chan C, Lim YT, Kaul U, Sinha N, Patel T, Tan HC,
Lopez-Cuellar J, Gaxiola E, Ramesh S, et al; WISDOM Investigators.
Twelve-month outcomes with a paclitaxel-eluting stent transitioning
from controlled trials to clinical practice (the WISDOM Registry).
Am J Cardiol 2006;98:918 –922.
39. Laarman GJ, Suttorp MJ, Dirksen MT, van Heerebeek L, Kiemeneij F,
Slagboom T, van der Wieken LR, Tijssen JG, Rensing BJ, Patterson
M. Paclitaxel-eluting versus uncoated stents in primary percutaneous
coronary intervention. N Engl J Med 2006;355:1105–1113.
40. Spaulding C, Henry P, Teiger E, Beatt K, Bramucci E, Carrie D, Slama
MS, Merkely B, Erglis A, Margheri M, et al; TYPHOON Investigators. Sirolimus-eluting versus uncoated stents in acute myocardial
infarction. N Engl J Med 2006;355:1093–1104.
41. Muller I, Besta F, Schulz C, Massberg S, Schonig A, Gawaz M.
Prevalence of clopidogrel non-responders among patients with stable
angina pectoris scheduled for elective coronary stent placement.
Thromb Haemost 2003;89:783–787.
42. Cuisset T, Frere C, Quilici J, Morange PE, Saidi LN, Carvajal J,
Lambert M, Mouret JP, Alessi MC, Bonnet JL. Beneficial effect of a
loading dose of 600 mg of clopidogrel on platelet parameters in
patients admitted for acute coronary syndrome. Arch Mal Coeur Vaiss
2006;99:889 – 893.