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.
© Copyright 2024