American Heart Journal, Volume 161, Issue 2, Feb 2011

MHJ161_2_toc 2/2/11 2:24 PM Page 1 American Heart Journal www.ahjonline.com Table of Contents February 2011, Volume 1...

Author: Robert M. Califf | Daniel B. Mark | Patricia K. Hodgson | Rebecca L. Hines

354 downloads 1422 Views 7MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

MHJ161_2_toc 2/2/11 2:24 PM Page 1

American Heart Journal www.ahjonline.com

Table of Contents

February 2011, Volume 161, Number 2

Editorial

Curriculum in Cardiology

221 Challenge of rehospitalizations for heart failure:

241 Atrial fibrillation, anticoagulation, fall risk, and outcomes

in elderly patients

Potential of natriuretic doses of mineralocorticoid receptor antagonists Robert W. Schrier, MD and Mihai Gheorghiade, MD, Denver, CO; and Chicago, IL

Matthew B. Sellers, MD and L. Kristin Newby, MD, MHS, Durham, NC 247 Primary percutaneous coronary intervention for acute

myocardial infarction: Is it worth the wait?: The risktime relationship and the need to quantify the impact of delay

Special Articles 224 Clinical development of pharmacologic agents for acute

Giuseppe Tarantini, MD, PhD, Frans Van de Werf, MD, PhD, Claudio Bilato, MD, PhD, and Bernard Gersh, MB, ChB, DPhil, FRCP, Padua, Italy; Leuven, Belgium; and Rochester, NY

heart failure syndromes: A proposal for a mechanistic translational phase Mihai Gheorghiade, MD, Peter S. Pang, MD, Christopher M. O’Connor, MD, Krishna Prasad, MD, John McMurray, MD, John R. Teerlink, MD, Mona Fiuzat, PharmD, Hani Sabbah, PhD, and Michel Komajda, MD, Chicago, IL; Durham, NC; London, England; Glasgow, United Kingdom; San Francisco, CA; Detroit, MI; and Paris, France 233 Lessons learned from a pediatric clinical trial: The

Pediatric Heart Network Angiotensin-Converting Enzyme Inhibition in Mitral Regurgitation Study Jennifer S. Li, MD, MHS, Steven D. Colan, MD, Lynn A. Sleeper, ScD, Jane W. Newburger, MD, MPH, Victoria L. Pemberton, RNC, MS, Andrew M. Atz, MD, Meryl S. Cohen, MD, Fraser Golding, MD, Gloria L. Klein, Ronald V. Lacro, MD, Elizabeth Radojewski, RN, Marc E. Richmond, MD, and L. LuAnn Minich, MD, Durham, NC; Watertown and Boston, MA; Bethesda, MD; Charleston, SC; Philadelphia, PA; Ontario, Canada; New York, NY; and Salt Lake City, UT

Trial Design 254

Design and rationale of the RadIal Vs. femorAL access for coronary intervention (RIVAL) trial: A randomized comparison of radial versus femoral access for coronary angiography or intervention in patients with acute coronary syndromes Sanjit S. Jolly, MD, MSc, Kari Niemelä, MD, PhD, Denis Xavier, MD, Petr Widimsky, MD, Andrzej Budaj, MD, PhD, Vicent Valentin, MD, Basil S. Lewis, MD, Alvaro Avezum, MD, PhD, Philippe Gabriel Steg, MD, Sunil V. Rao, MD, John Cairns, MD, Susan Chrolavicius, BScN, Salim Yusuf, MBBS, D.Phil, and Shamir R. Mehta, MD, MSc, Ontario and Vancouver, Canada; Tampere, Finland; Bangalore, India; Prague, Czech Republic; Warsaw, Poland; Valencia, Spain; Haifa, Israel; Sao Paulo, Brazil; Paris, France; and Durham, NC

American Heart Journal (ISSN 0002-8703) is published monthly by Mosby, 360 Park Avenue South, New York, NY 10010-1710. Periodicals postage paid at New York, NY and additional mailing offices. POSTMASTER: Send address changes to American Heart Journal, Elsevier Customer Service Department, 3251 Riverport Lane, Maryland Heights, MO 63043, USA.

American Heart Journal

1A

Contents 261

continued

A randomized, partially blinded, multicenter, active-controlled, dose-ranging study assessing the safety, efficacy, and pharmacodynamics of the REG1 anticoagulation system in patients with acute coronary syndromes: Design and rationale of the RADAR Phase IIb trial Thomas J. Povsic, MD, PhD, Mauricio G. Cohen, MD, Roxana Mehran, MD, Christopher E. Buller, MD, Christoph Bode, MD, Jan H. Cornel, MD, Jaroslaw D. Kasprzak, MD, Gilles Montalescot, MD, Diane Joseph, William A. Wargin, PhD, Christopher P. Rusconi, PhD, Steven L. Zelenkofske, DO, Richard C. Becker, MD, and John H. Alexander, MD, MHS, Durham, and Chapel Hill, NC; Miami, FL; New York, NY; Ontario, Canada; Freiberg, Germany; Alkmaar, Netherlands; Lódz´, Poland; Paris, France; and Basking Ridge, NJ

269

Associations between cardiovascular parameters and uteroplacental Doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the Zwangerschap bij Aangeboren Hartafwijking (ZAHARA) II study Ali Balci, MD, MSc, Krystyna M. Sollie, MD, Barbara J. M. Mulder, MD, PhD, Monique W. M. de Laat, MD, PhD, Jolien W. Roos-Hesselink, MD, PhD, Arie P. J. van Dijk, MD, PhD, Elly M. C. J. Wajon, MD, Hubert W. Vliegen, MD, PhD, Willem Drenthen, MD, PhD, Hans L. Hillege, MD, PhD, Jan G. Aarnoudse, MD, PhD, Dirk J. van Veldhuisen, MD, PhD, and Petronella G. Pieper, MD, PhD, Groningen, Amsterdam, Rotterdam, Nijmegen, Enschede, Leiden, and Utrecht, The Netherlands Clinical Investigations

283 The influence of time from symptom onset and reper-

fusion strategy on 1-year survival in ST-elevation myocardial infarction: A pooled analysis of an early fibrinolytic strategy versus primary percutaneous coronary intervention from CAPTIM and WEST Cynthia M. Westerhout, PhD, Eric Bonnefoy, MD, Robert C. Welsh, MD, Philippe Gabriel Steg, MD, Florent Boutitie, PhD, and Paul W. Armstrong, MD, Edmonton, Canada; and Lyon and Paris, France 291 Has the ClOpidogrel and Metoprolol in Myocardial

Infarction Trial (COMMIT) of early -blocker use in acute coronary syndromes impacted on clinical practice in Canada? Insights from the Global Registry of Acute Coronary Events (GRACE) Jeremy Edwards, MD, Shaun G. Goodman, MD, MSc, Raymond T. Yan, MD, Robert C. Welsh, MD, Jan M. Kornder, MD, J. Paul DeYoung, MD, Denis Chauret, MD, Jean-Pierre Picard, MD, Kim A. Eagle, MD, and Andrew T. Yan, MD, Ontario, Alberta, British Columbia, and Quebec, Canada; and Ann Arbor, MI 298

Incidence and clinical consequences of acquired thrombocytopenia after antithrombotic therapies in patients with acute coronary syndromes: Results from the Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) trial Adriano Caixeta, MD, PhD, George D. Dangas, MD, PhD, Roxana Mehran, MD, Frederick Feit, MD, Eugenia Nikolsky, MD, PhD, Alexandra J. Lansky, MD, Jiro Aoki, MD, PhD, Jeffrey W. Moses, MD, Steven R. Steinhubl, MD, Harvey D. White, DSc, E. Magnus Ohman, MD, Steven V. Manoukian, MD, Martin Fahy, MSc, and Gregg W. Stone, MD, New York, NY; New Haven, CT; Lexington, KY; Auckland, New Zealand; Durham, NC; and Nashville, TN

Acute Ischemic Heart Disease 276 Prehospital triage in the ambulance reduces infarct

size and improves clinical outcome Sonja Postma, MSc, Jan-Henk E. Dambrink, MD, PhD, Menko-Jan de Boer, MD, PhD, A. T. Marcel Gosselink, MD, PhD, Gerrit J. Eggink, MD, Henri van de Wetering, MANP, Frans Hollak, RN, Jan Paul Ottervanger, MD, PhD, Jan C. A. Hoorntje, MD, PhD, Evelien Kolkman, MSc, Harry Suryapranata, MD, PhD, and Arnoud W. J. van ‘t Hof, MD, PhD, Zwolle, The Netherlands

2A


Valvular and Congenital Heart Disease 307 Pregnancy in women with corrected tetralogy of Fallot: Occurrence and predictors of adverse events Ali Balci, MD, MSc, Willem Drenthen, MD, PhD, Barbara J. M. Mulder, MD, PhD, Jolien W. RoosHesselink, MD, PhD, Adriaan A. Voors, MD, PhD, Hubert W. Vliegen, MD, PhD, Philip Moons, RN, PhD, Krystyna M. Sollie, MD, Arie P. J. van Dijk, MD, PhD, Dirk J. van Veldhuisen, MD, PhD, and Petronella G. Pieper, MD, PhD, Groningen, Utrecht, Amsterdam, Rotterdam, Leiden, and Nijmegen, The Netherlands; and Leuven, Belgium

Contents

continued

Imaging and Diagnostic Testing 314 Left atrial reverse remodeling and functional improve-

ment after mitral valve repair in degenerative mitral regurgitation: A real-time 3-dimensional echocardiography study Nina Ajmone Marsan, MD, Francesco Maffessanti, MS, Gloria Tamborini, MD, Paola Gripari, MD, Enrico Caiani, PhD, Laura Fusini, MD, Manuela Muratori, MD, Marco Zanobini, PhD, Francesco Alamanni, MD, and Mauro Pepi, MD, Milan, Italy Congestive Heart Failure 322 Certoparin versus unfractionated heparin to prevent

venous thromboembolic events in patients hospitalized because of heart failure: A subgroup analysis of the randomized, controlled CERTIFY study Ulrich Tebbe, MD, Sebastian M. Schellong, MD, Sylvia Haas, MD, Horst Eberhard Gerlach, MD, Claudia Abletshauser, PhD, Christian Sieder, MSc, Peter Bramlage, MD, and Hanno Riess, MD, Detmold, Dresden, München, Mannheim, Nürnberg, Mahlow, and Berlin, Germany 329

A randomized controlled trial evaluating the safety and efficacy of cardiac contractility modulation in advanced heart failure Alan Kadish, MD, Koonlawee Nademanee, MD, Kent Volosin, MD, Steven Krueger, MD, Suresh Neelagaru, MD, Nirav Raval, MD, Owen Obel, MD, Stanislav Weiner, MD, Marc Wish, MD, Peter Carson, MD, Kenneth Ellenbogen, MD, Robert Bourge, MD, Michael Parides, PhD, Richard P. Chiacchierini, PhD, Rochelle Goldsmith, PhD, Sidney Goldstein, MD, Yuval Mika, PhD, Daniel Burkhoff, MD, PhD, and William T. Abraham, MD, Chicago, IL; Inglewood, CA; Philadelphia, PA; Lincoln, NE; Amarillo, Dallas, and Tyler, TX; Atlanta, GA; Fairfax, and Richmond, VA; Birmingham, AL; New York, and Orangeburg, NY; Detroit, MI; and Columbus, OH

338

Effects of n-3 polyunsaturated fatty acids on malignant ventricular arrhythmias in patients with chronic heart failure and implantable cardioverter-defibrillators: A substudy of the Gruppo Italiano per lo Studio della Sopravvivenza nell’Insufficienza Cardiaca (GISSI-HF) trial Andrea A. Finzi, MD, Roberto Latini, MD, Simona Barlera, MSc, Maria G. Rossi, MD, Albarosa Ruggeri, MD, Alessandro Mezzani, MD, Chiara Favero, BSc, Maria G. Franzosi, BiolD, Domenico Serra, MD, Donata Lucci, MSc, Francesca Bianchini, BSc, Roberto Bernasconi, Aldo P. Maggioni, MD, Gianluigi Nicolosi, MD, Maurizio Porcu, MD, Gianni Tognoni, MD, Luigi Tavazzi, MD, and Roberto Marchioli, MD, Milano, Reggio Calabria, Veruno, Florence, Pordenone, Cagliari, S Maria Imbaro, and Cotignola, Italy; and Lugano, Switzerland

Coronary Artery Disease 344 Common oral mucosal diseases, systemic inflammation, and cardiovascular diseases in a large cross-sectional US survey Stefano Fedele, DDS, PhD, Wael Sabbah, BDS, MSc, Nikos Donos, DDS, MS, PhD, Stephen Porter, BSc, MD, PhD, and Francesco D’Aiuto, London, United Kingdom Prevention and Rehabilitation 351 Reducing cardiovascular disease risk in medically

underserved urban and rural communities Alfred A. Bove, MD, PhD, FACC, William P. Santamore, PhD, Carol Homko, RN, PhD, Abul Kashem, MD, PhD, Robert Cross, MD, Timothy R. McConnell, PhD, Gail Shirk, RN, and Francis Menapace, MD, Philadelphia, Bloomsburg, and Danville, PA Interventional Cardiology 360 Incidence and clinical outcome of minor surgery in the

year after drug-eluting stent implantation: Results from the Evaluation of Drug-Eluting Stents and Ischemic Events Registry Emmanouil S. Brilakis, MD, PhD, David J. Cohen, MD, MSc, Neal S. Kleiman, MD, Michael Pencina, PhD, Deborah Nassif, PhD, Jorge Saucedo, MD, Robert N. Piana, MD, Subhash Banerjee, MD, Michelle J. Keyes, PhD, Chen-Hsing Yen, MS, and Peter B. Berger, MD, Dallas, and Houston, TX; Kansas City, MO; Boston, MA; Oklahoma City, OK; Nashville, TN; and Danville, PA Continued on page 4A


3A

Contents

continued

367 Qualitative assessment of neointimal tissue after drug-

391 Impact of baseline thrombocytopenia on the early and

eluting stent implantation: Comparison between follow-up optical coherence tomography and intravascular ultrasound

late outcomes after ST-elevation myocardial infarction treated with primary angioplasty: Analysis from the Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction (HORIZONSAMI) trial

Sung Woo Kwon, MD, Byeong-Keuk Kim, MD, Tae-Hoon Kim, MD, Jung-Sun Kim, MD, Young-Guk Ko, MD, Donghoon Choi, MD, Yangsoo Jang, MD, and Myeong-Ki Hong, MD, Seoul, Korea 373

Standard versus high loading doses of clopidogrel in Asian ST-segment elevation myocardial infarction patients undergoing percutaneous coronary intervention: Insights from the Korea Acute Myocardial Infarction Registry Cheol Ung Choi, MD, Seung-Woon Rha, MD, Dong Joo Oh, MD, Kanhaiya L. Poddar, MBBS, Jin Oh Na, MD, Jin Won Kim, MD, Hong Euy Lim, MD, Eung Ju Kim, MD, Chang Gyu Park, MD, Hong Seog Seo, MD, Taek Jong Hong, MD, Jong-Seon Park, MD, Young Jo Kim, MD, Seung Ho Hur, MD, In Whan Seong, MD, Jei Keon Chae, MD, Myeong Chan Cho, MD, Jang Ho Bae, MD, Dong Hoon Choi, MD, Yang Soo Jang, MD, In Ho Chae, MD, Hyo Soo Kim, MD, Chong Jin Kim, MD, Jung Han Yoon, MD, Tae Hoon Ahn, MD, SeungJea Tahk, MD, Wook Sung Chung, MD, Ki Bae Seung, MD, Shung Chall Chae, MD, Seung Jung Park, MD, Young Keun Ahn, MD, and Myung Ho Jeong, MD, Seoul, Pusan, Daegu, Daejeon, Jeonju, Chongju, Bundang, Wonju, and Gwangju, South Korea

383 Comparison of 2 point-of-care platelet function tests,

VerifyNow Assay and Multiple Electrode Platelet Aggregometry, for predicting early clinical outcomes in patients undergoing percutaneous coronary intervention Young-Guk Ko, MD, Jung-Won Suh, MD, PhD, Bo Hyun Kim, BA, Chan Joo Lee, MD, Jung-Sun Kim, MD, PhD, Donghoon Choi, MD, PhD, Myeong-Ki Hong, MD, PhD, Myung-Ki Seo, MD, Tae-Jin Youn, MD, PhD, In-Ho Chae, MD, PhD, Dong Joo Choi, MD, PhD, and Yangsoo Jang, MD, PhD, Seoul, and Gyeonggi-do, Korea

4A

Diaa A. Hakim, MD, PhD, George D. Dangas, MD, PhD, Adriano Caixeta, MD, PhD, Eugenia Nikolsky, MD, PhD, Alexandra J. Lansky, MD, Jeffrey W. Moses, MD, Bimmer Claessen, MD, Elias Sanidas, MD, Harvey D. White, DSc, E. Magnus Ohman, MD, Steven V. Manoukian, MD, Martin Fahy, MSc, Roxana Mehran, MD, and Gregg W. Stone, MD, New York, NY; New Haven, CT; Auckland, New Zealand; Durham, NC; and Atlanta, GA


397

Temporal changes in the outcomes of patients with diabetes mellitus undergoing percutaneous coronary intervention in the National Heart, Lung, and Blood Institute dynamic registry Elizabeth M. Holper, MD, MPH, J. Dawn Abbott, MD, Suresh Mulukutla, MD, Helen Vlachos, MSc, Faith Selzer, PhD, Darren McGuire, MD, MHSc, David P. Faxon, MD, Warren Laskey, MD, Vankeepuram S. Srinivas, MD, Oscar C. Marroquin, MD, and Alice K. Jacobs, MD, Dallas, TX; Providence, RI; Pittsburgh, PA; Boston, MA; Albuquerque, NM; and New York, NY

Surgery 404 Clopidogrel loading dose and bleeding outcomes in patients undergoing urgent coronary artery bypass grafting Nicholas L. M. Cruden, PhD, MBChB, MRCP, Kristin Morch, Daniel R. Wong, MD, MPH, FRCSC, W. Peter Klinke, MD, FRCPC, John Ofiesh, MD, FRCSC, and J. David Hilton, MD, FRCPC, British Columbia, Canada; and Edinburgh, United Kingdom

Contents

continued

Pediatrics 411 Factors associated with the physical activity level of

children who have the Fontan procedure Patricia E. Longmuir, PhD, Jennifer L. Russell, MD, FRCP(C), Mary Corey, PhD, Guy Faulkner, PhD, and Brian W. McCrindle, MD, FRCP(C), Toronto, Canada 418

Corrections

Letters to the Editor e5 An important indirect drug interaction between

dronedarone and warfarin that may be extrapolated to other drugs that can alter gastrointestinal function James A. Reiffel, MD, New York, NY e7 Shirolkar’s reply to Reiffel’s letter to the editor

Shailesh C. Shirolkar, MD, Durham, NC


5A

Editor in Chief: Robert M. Califf, MD, Durham, NC Editor: Daniel B. Mark, MD, MPH, Durham, NC Executive Editor: Patricia K. Hodgson, Durham, NC Managing Editor: Rebecca L. Hines, Durham, NC Editorial Assistant: Brenda McCoy, Durham, NC Issue Manager: Todd C. Reiss, New York, NY

AHJ


Founded in 1925

Associate Editors Sana M. Al-Khatib, MD, Durham Thomas M. Bashore, MD, Durham Richard C. Becker, MD, Durham Jeffrey S. Berger, MD, MS New York Salvador Borges - Neto, MD, Durham Mark P. Donahue, MD, Durham Mark H. Drazner, MD, MSc, Dallas

G. Mike Felker, MD, Durham Christopher B. Granger, MD, Durham Robert A. Harrington, MD, Durham John K. Harrison, MD, Durham Paul A. Heidenreich, MD, MS, Palo Alto Adrian F. Hernandez, MD, Durham Jennifer S. Li, MD, MHS, Durham

Renato D. Lopes, MD, PhD, MHS, Durham Kenneth W. Mahaffey, MD, Durham Darren K. McGuire, MD, MHSc, Dallas Rajendra H. Mehta, MD, Durham L. Kristin Newby, MD, Durham E. Magnus Ohman, MD, Durham Manesh R. Patel, MD, Durham

Thomas J. Povsic, MD, Durham Sunil V. Rao, MD, Durham Matthew T. Roe, MD, MHS, Durham Joseph G. Rogers, MD, Durham Lynda A. Szczech, MD, Durham David J. Whellan, MD, MHS Philadelphia

Editorial Board Frank V. Aguirre, MD John H. Alexander, MD Karen P. Alexander, MD Andrew S. Allen, PhD Larry A. Allen, MD Kevin J. Anstrom, PhD Elliott M. Antman, MD Paul W. Armstrong, MD Alvaro Avezum, MD, PhD Steven R. Bailey, MD Gust H. Bardy, MD Gregory W. Barsness, MD Eric R. Bates, MD Jeroen J. Bax, MD, PhD George A. Beller, MD Peter B. Berger, MD Deepak L. Bhatt, MD Philip F. Binkley, MD John A. Bittl, MD Vera Bittner, MD, MSPH Roger S. Blumenthal, MD Eric Boersma, MSc, PhD Robert O. Bonow, MD Sorin J. Brener, MD Ralph G. Brindis, MD, MPH Christopher E. Buller, MD Javed Butler, MD, MPH Alfred E. Buxton, MD Christopher P. Cannon, MD Warren J. Cantor, MD Blase A. Carabello, MD Melvin D. Cheitlin, MD W. Randolph Chitwood Jr, MD David J. Cohen, MD, MSc Marc Cohen, MD Mauricio G. Cohen, MD Lawrence H. Cohn, MD Heidi M. Connolly, MD Stuart J. Connolly, MD Anne B. Curtis, MD Ralph J. Damiano, MD James P. Daubert, MD 6A American Heart Journal

Charles J. Davidson, MD James A. De Lemos, MD Elizabeth R. DeLong, PhD David L. DeMets, PhD J. Michael DiMaio, MD John P. DiMarco, MD, PhD Pamela S. Douglas, MD Vladimir Dzavik, MD Kim A. Eagle, MD Kenneth A. Ellenbogen, MD Stephen G. Ellis, MD Nathan R. Every, MD, MPH David P. Faxon, MD Gregg C. Fonarow, MD Gary S. Francis, MD John K. French, MB, PhD Thomas C. Gerber, MD, PhD Bernard J. Gersh, MB, ChB, DPhil Myron C. Gerson, MD Jalal K. Ghali, MD Mihai Gheorghiade, MD Raymond J. Gibbons, MD C. Michael Gibson, MS, MD Robert P. Giugliano, MD Donald D. Glower, MD Lee Goldman, MD Pascal J. Goldschmidt-Clermont, MD John R. Guyton, MD Frank E. Harrell Jr, PhD David Hasdai, MD Edward P. Havranek, MD Robert C. Hendel, MD Timothy D. Henry, MD Charles A. Herzog, MD James A. Hill, MD Mark A. Hlatky, MD Judith S. Hochman, MD David R. Holmes Jr, MD Michael P. Hudson, MD, MHSc Alice K. Jacobs, MD James L. Januzzi, MD Marc E. Jolicoeur, MD

David E. Kandzari, MD Padma Kaul, PhD Dean J. Kereiakes, MD Morton J. Kern, MD Neal S. Kleiman, MD George J. Klein, MD Richard A. Krasuski, MD William E. Kraus, MD Joel Kupersmith, MD Gervasio A. Lamas, MD Kerry L. Lee, PhD Thomas H. Lee Jr, MD A. Michael Lincoff, MD Gregory Y.H. Lip, MD James E. Lock, MD Thomas H. Marwick, MB, PhD Frederick A. Masoudi, MD, MSPH Barry M. Massie, MD Charles Maynard, PhD George A. Mensah, MD D. Douglas Miller, MD Julie M. Miller, MD Todd D. Miller, MD David J. Moliterno, MD Fred Morady, MD David A. Morrow, MD, MPH Joseph Brent Muhlestein, MD Debabrata Mukherjee, MD Sean M. O’Brien, PhD Christoper M. O’Connor, MD Gerald T. O’Connor, MD Elizabeth O. Ofili, MD Peter M. Okin, MD Brian Olshansky, MD Catherine M. Otto, MD Douglas L. Packer, MD Richard L. Page, MD Patricia A. Pellikka, MD Carl J. Pepine, MD Eric D. Peterson, MD Marc A. Pfeffer, MD, PhD Matthias E. Pfisterer, MD

Karen S. Pieper, MS Louise Pilote, MD, MPH, PhD Bertram Pitt, MD Jeanne E. Poole, MD Jeffrey J. Popma, MD Eric N. Prystowsky, MD Paolo Raggi, MD Rita F. Redberg, MD, MSc Harmony Reynolds, MD Vera H. Rigolin, MD Veronique L. Roger, MD, MPH Melvyn Rubenfire, MD John S. Rumsfeld, MD, PhD Thomas Ryan, MD Gillian D. Sanders, PhD Hartzell V. Schaff, MD Kevin A. Schulman, MD Svati H. Shah, MD Leslee Shaw, PhD Thomas N. Skelton, MD Steven R. Steinhubl, MD Lynne W. Stevenson, MD William G. Stevenson, MD Carla A. Sueta, MD, PhD W.H. Wilson Tang, MD Allen J. Taylor, MD James E. Tcheng, MD Alan N. Tenaglia, MD Pierre Theroux, MD Eric J. Topol, MD Jack V. Tu, MD, PhD Hector O. Ventura, MD David A. Vorchheimer, MD Lars Wallentin, MD, PhD Andrew Wang, MD Tracy Y. Wang, MD, MHS W. Douglas Weaver, MD William S. Weintraub, MD J. Marcus Wharton, MD Harvey D. White, DSc Clyde W. Yancy, MD James B. Young, MD

AHJ


Founded in 1925

Editorial Communications If you have questions or comments regarding original articles and editorial management, please contact Robert M. Califf, MD, Editor in Chief, American Heart Journal, Duke Clinical Research Institute Communications, 300 West Morgan Street, Suite 800, Durham, NC 27701, Telephone 919-668-5948.

Author Inquiries For inquiries relating to the submission of articles (including electronic submission where available) please visit this journal’s homepage at http://www.ahjonline. You can track accepted articles at http://www.elsevier. com/trackarticle and set up e-mail alerts to inform you of when an article’s status has changed. Also accessible from here is information on copyright, frequently asked questions and more. Contact details for questions arising after acceptance of an article, especially those relating to proofs, will be provided by the publisher.

American Heart Journal (ISSN 0002-8703) is published monthly by Mosby, 360 Park Avenue South, New York, NY 10010-1710. Periodicals postage paid at New York, NY, and additional mailing offices. POSTMASTER: Send address changes to American Heart Journal, Elsevier Customer Service Department, 3251 Riverport Lane, Maryland Heights, MO 63043, USA.

Funding Body Agreements and Policies Elsevier has established agreements and developed policies to allow authors whose articles appear in journals published by Elsevier, to comply with potential manuscript archiving requirements as specified as conditions of their grant awards. To learn more about existing agreements and policies please visit http://www.elsevier.com/fundingbodies.

Annual Subscription Rates United States and possessions: Individual $298; Institution $699; Student/ Resident: $142. All other countries: Individual $395; Institution $803; Student/Resident $188. Single Issues $67. Prices include postage and are subject to change without notice. For additional information contact Elsevier Inc. Customer Support Department, P.O. Box 945, New York, NY 10010. Tel: (212) 633-3730. Toll-Free (for customers in North America): 1-888-4ES-INFO. Fax: (212) 633-3680. E-mail: [email protected]. For author support contact: [email protected].

Copyright ©2011 Mosby Inc. All rights reserved. This journal and the individual contributions contained in it are protected under copyright by Mosby Inc., and the following terms and conditions apply to their use:

Information for Readers

Orders, Claims, and Journal Inquiries Please contact the Elsevier Customer Service Department nearest you: St. Louis: Elsevier Customer Service Department, 3251 Riverport Lane, Maryland Heights, MO 63043, USA; phone: (800) 654-2452 [toll free within the USA]; (+1) (314) 447-8871 [outside the USA]; fax: (+1) (314) 447-8029; e-mail: [email protected]. Oxford: Elsevier Customer Service Department, The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK; phone: (+44) (1865) 843434; fax: (+44) (1865) 843970; e-mail: JournalsCustomerServiceEMEA@ elsevier.com. Tokyo: Elsevier Customer Service Department, 4F Higashi-Azabu, 1-Chome Bldg, 1-9-15 Higashi-Azabu, Minato-ku, Tokyo 106-0044, Japan; phone: (+81) (3) 5561 5037; fax: (+81) (3) 5561 5047; e-mail: [email protected]. Singapore: Elsevier Customer Service Department, 3 Killiney Road, #08-01 Winsland House I, Singapore 239519; phone: (+65) 63490222; fax: (+65) 67331510; e-mail: [email protected]. Advertising Information Advertising orders and inquiries can be sent to: USA, Canada, and South America: Cunningham Associates, 180 Old Tappan Road, Old Tappan, NJ 07675, USA; phone (201) 767-4170; fax (201) 767-8065. Classified advertising orders and inquiries can be sent to Aileen Rivera, Marketing Solutions, Elsevier, 360 Park Avenue South, New York, NY 10010-1710, USA; phone (212) 633-3721; fax (212) 633-3820. Europe and the rest of the world: Julie Toop; phone +44(0) 1865 843016; fax +44 (0) 1865 843976; e-mail: [email protected]. Reprints To order 100 or more reprints for educational, commercial, or promotional use, contact the Commercial Reprints Department, Elsevier Inc., 360 Park Avenue South, New York, NY 10010-1710; E-mail: [email protected]. Author Information For full and complete Author Information, please go to: http://www.ahjonline. com/authorinfo.

Photocopying Single photocopies of single articles may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. For information on how to seek permission visit www.elsevier.com/permissions or call: (+44) 1865 843830 (UK) / (+1) 215 239 3804 (USA). Derivative Works Subscribers may reproduce tables of contents or prepare lists of articles including abstracts for internal circulation within their institutions. Permission of the Publisher is required for resale or distribution outside the institution. Permission of the Publisher is required for all other derivative works, including compilations and translations (please consult www.elsevier.com/permissions). Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this journal, including any article or part of an article (please consult www.elsevier.com/permissions). Except as outlined above, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. Although all advertising material is expected to conform to ethical (medical) standards, inclusion in this publication does not constitute a guarantee or endorsement of the quality or value of such product or of the claims made of it by its manufacturer. The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper) American Heart Journal 7A

Editorial

Challenge of rehospitalizations for heart failure: Potential of natriuretic doses of mineralocorticoid receptor antagonists Robert W. Schrier, MD, a and Mihai Gheorghiade, MD b Denver, CO; and Chicago, IL

There are N1 million hospitalizations for heart failure (HF) each year in the United States. During the 60 to 90 days postdischarge for HF, readmission rates are approximately 30%. Taken together, recurrent hospitalizations account for N75% of the US $46 billion in annual HF expenditures. Three fourths of hospitalizations are in patients with exacerbations of previously diagnosed HF. Prospective randomized studies in patients with severe HF have shown improved survival by blocking the mineralocorticoid receptor, primarily with nonnatriuretic doses of spironolactone (RALES).1 Nevertheless, the results of treating patients hospitalized for HF are still disappointing. According to the results of the Acute Decompensated Heart Failure Registry in hospitalized patients, at least 50% of these patients are discharged with continued symptoms. Despite treatment with intravenous loop diuretics in 90% of these patients, 33% are discharged with ≤5 lb of weight loss; and 16% are actually discharged with an increase in body weight.2 Thirty percent of these patients were considered to be resistant to diuretic therapy. Moreover, loop diuretics block sodium chloride entry into the macula densa, which causes further stimulation of the renin-angiotensin-aldosterone system (RAAS). Because angiotensin and aldosterone have both been shown to contribute to cardiac fibrosis and remodeling, this may be a negative component to loop diuretics therapy in HF patients. However, 84% of the patients are admitted with dyspnea, 67% with rales, and 66% with peripheral edema.2 Moreover, these physical findings predict 1-year cardiovascular rehospitalization and mortality; and currently, loop diuretics are the therapy of choice for these symptoms.3 There is, therefore, a need for additional approaches to treat congestive HF, a problem that will no doubt increase with the aging of the population. Gheorghiade et al4 have suggested that early intervention should be undertaken in these patients. Hemodynamic congestion, defined as a high left ventricular filling pressure, generally precedes clinical congestion that then leads to hospitalization for HF. Whether earlier diuretic intervention in patients with

From the aUniversity of Colorado Denver School of Medicine, Denver, CO, and b Northwestern University Feinberg School of Medicine, Chicago, IL. Submitted September 23, 2010; accepted October 29, 2010. 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.10.039

hemodynamic congestion could have a major impact on hospitalizations and readmissions for clinical congestion remains to be proven. The rate of fluid removal must, however, be judicious because mobilization of interstitial fluid in HF patients may be limited to 12 to 14 mL/min.5 Thus, theoretically, a diuresis in excess of this rate with a loop diuretic may worsen arterial underfilling with diminished cardiac and renal function. There is a more modest diuretic approach that is not often used to treat HF patients with either hemodynamic or clinical congestion, namely, natriuretic doses of mineralocorticoid receptor antagonists. This approach has been used successfully in cirrhotic patients for many years.6 The pathophysiology of sodium and water retention in HF and cirrhosis is very similar.7 The arterial baroreceptors sense arterial underfilling in HF secondary to a decrease in cardiac output, whereas, in cirrhosis, unloading of these arterial baroreceptors occurs secondary to primary systemic arterial vasodilation. Approximately 50% of HF patients have preserved left ventricular function. This is, however, at the expense of increased filling pressure that is associated with increased ventricular wall stress, endomyocardial ischemia, decreased cardiac venous drainage, and secondary mitral and even tricuspid insufficiency.4,8 In cirrhosis, the arterial vasodilation occurs primarily in the splanchnic circulation and is associated with a secondary increase in cardiac output. The pathophysiology of high-output cardiac failure, for example, beriberi and thyrotoxicosis, has a similar pathogenesis as cirrhosis. With this arterial underfilling in both HF and cirrhosis, the neurohumoral axis is stimulated to maintain circulatory integrity; but tradeoffs occur that can be deleterious. Perhaps most important are the increases in the RAAS, sympathetic activity, and the nonosmotic release of arginine vasopressin. Because of the effects of angiotensin and α-adrenergic stimulation to enhance tubular epithelial sodium reabsorption in the more proximal sites in the nephron, the normal escape from the sodium-retaining effects of aldosterone in the renal collecting duct does not occur in patients with severe heart or liver failure. Thus, secondary hyperaldosteronism is pivotal in both congestion in HF and ascites formation in cirrhosis. On this background, natriuretic doses of mineralocorticoid receptor antagonists, not loop diuretics, are the initial diuretics of choice in cirrhotic patients with

222 Schrier and Gheorghiade

American Heart Journal February 2011

Figure 1

Diuretics, digoxin, and angiotensin-converting enzyme inhibitors were withdrawn 4 days before admission to the General Clinical Research Center. The subjects were placed on a constant daily diet of 100 mEq sodium and 60 mEq potassium. A (white bars), Positive cumulative sodium balance in the 6 patients (4 ischemic heart disease, 1 idiopathic cardiomyopathy, and 1 aortic valvular disease). B (black bars), In the same patients, significant negative cumulative sodium balance during 200 mg BID spironolactone (P b .01). C, Increase in urine Na−:K+ concentration ratio during spironolactone in all 6 patients (P b .05), a finding compatible with aldosterone antagonism. Mean plasma potassium increased from 3.86 ± 0.2 to 4.1 ± 0.2 mEq/L during spironolactone treatment (P b .05). Mean systolic blood pressure (112 ± 7 vs 110 ± 5 mm Hg, P = not significant) and creatinine clearance (87 ± 7 vs 87.2 ± 8 mL/min, not significant) did not change with spironolactone treatment. Plasma human atrial natriuretic peptide decreased significantly with spironolactone (147 ± 58 vs 83 ± 30 mg/L, P b .05). Fluid intake was not restricted, and a mean of 2-kg weight loss occurred.

ascites.6 The International Ascites Club has designated diuretic resistance in cirrhosis as a failure to respond to 400 mg of spironolactone and 160 mg of furosemide. In

practice, however, the spironolactone doses used to treat cirrhotic patients with ascites are rarely N100 to 200 mg/d. Because these patients are not receiving

American Heart Journal Volume 161, Number 2

RAAS inhibitors and are generally receiving a potassiumlosing loop diuretic, hyperkalemia is generally not a problem. In contrast, the standard mean daily dose of spironolactone used in HF patients, based on the RALES study, is 25 mg. The dose-finding RALES article demonstrated no increase in urinary sodium excretion with 25 mg of spironolactone. Therefore, the interpretation of the RALES results to improve survival in HF was an effect of spironolactone to block the nongenomic, nonnatriuretic effects of aldosterone on fibrosis, oxidant injury, and remodeling of the heart.1 There is a paucity of results using natriuretic doses in HF patients.9 A publication in 1961 demonstrated a natriuretic response to 100 mg of spironolactone in 3 patients with HF.10 There was also a small study in diuretic-resistant patients on modest dose of angiotensinconverting enzyme inhibitor that demonstrated a natriuretic response to 100 mg/d of spironolactone.11 Another prospective study was performed on 6 patients with severe HF.12 All other treatments were discontinued, and the HF patients were shown to be avidly retaining sodium. Spironolactone therapy 200 mg BID then was shown to reverse totally the renal sodium retention over a 4-day period (Figure 1). In these small, short-term studies, hyperkalemia was not observed. These studies did not have a clinically important rise in serum potassium concentration. A recent study from Scotland using a record linkage database examined all patients receiving one or more prescriptions for spironolactone in patients with and without HF between 1994 and 2007. Despite a large increase in spironolactone prescriptions, there was no increase in hospitalizations for hyperkalemia.13 Thus, the role of secondary hyperaldosteronism and failure of aldosterone escape in HF patients with hemodynamic or clinical congestion is of pivotal importance. Careful titration of mineralocorticoid receptor antagonists to natriuretic doses (eg, 50-100 mg/d), while monitoring serum potassium concentration and urinary sodium-potassium ratio, is necessary. In this setting, the HF patients should be on a potassium-restricted diet and a potassium-losing loop diuretic without potassium supplementation before starting natriuretic doses of mineralocorticoid antagonists. There should be particular caution for hyperkalemia (N5.5 mEq/L)-related arrhythmias when instituting natriuretic doses of spironolactone in HF patients, particularly those with an estimated glomerular filtration rate b45 ml/min and baseline plasma potassium concentrations N4.5 mEq/L.14 A prospective randomized study needs to be undertaken to test the hypothesis that early treatment of hemodynamic congestion or clinical congestion with natriuretic doses of mineralocorticoid receptor antagonists can decrease symptomology, decrease hospitalizations and readmissions, and be safe and cost-effective in HF patients. A dose-range study of spironolactone in HF was performed by the RALES investigators, and a

Schrier and Gheorghiade 223

significant decrease was observed in sodium retention with the 50 mg and 75 mg/day doses at 9 days.15 The HF population in the United States is approximately 6 million, and effectively blocking the sodium retention secondary to hyperaldosteronism could have an important impact both medically and economically.

References 1. Pitt B, Zannad F, Remme WJ, et al. for the Randomized Aldactone Evaluation Study Investigators. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med 1999;341:709-17. 2. Adams Jr KF, Fonarow GC, Emerman CL, et al. For the ADHERE Scientific Advisory Committee and Investigators. Characteristics and outcomes of patients hospitalized for heart failure in the United States: rationale, design, and preliminary observations form the first 100,000 cases in the Acute Decompensated Failure National Registry (ADHERE). Am Heart J 2005;149:209-16. 3. Dunlay SM, Gheorghiade M, Reid K, et al. Critical elements of clinical follow-up after hospital discharge for heart failure: insights from the EVEREST trial. Eur J Heart Fail 2010;12:367-74. 4. Gheorghiade M, Filippatos G, De Luca L, et al. Congestion in acute heart failure syndromes: an essential target of evaluation and treatment. Am J Med 2006;119:S3-S10. 5. Fauchauld P. Effects of ultrafiltration of body fluid and transcapillary colloid osmotic gradient in hemodialysis patients, improvements in dialysis therapy. Contrib Nephrol 1989;74:170-5. 6. Ginès P, Schrier RW. Renal failure in cirrhosis. N Engl J Med 2009; 361:1279-90. 7. Bansal S, Lindenfeld JA, Schrier RW. Sodium retention in heart failure and cirrhosis: potential role of natriuretic doses of mineralocorticoid antagonist? Circ Heart Fail 2009;2:370-6. 8. Schrier RW. Role of diminished renal function in cardiovascular mortality: marker or pathogenetic factor? J Am Coll Cardiol 2006;47: 1-8. 9. Schrier RW, Masoumi A, Elhassan E. Aldosterone: role in edematous disorders, hypertension, chronic renal failure, and metabolic syndrome. Clin J Am Soc Nephrol 2010;5:1132-40. 10. Braunwald E, Plauth Jr WH, Morrow AGA. Method for the detection and quantification of impaired sodium excretion. Results of an oral sodium tolerance test in normal subjects and in patients with heart disease. Circulation 1965;32:223-31. 11. Van Vliet AA, Donker AJ, Nauta JJ, et al. Spironolactone in congestive heart failure refractory to high-dose loop diuretic and low-dose angiotensin-converting enzyme inhibitor. Am J Cardiol 1993;71:21A-8A. 12. Hensen J, Abraham WT, Durr JA, et al. Aldosterone in congestive heart failure: analysis of determinants and role in sodium retention. Am J Nephrol 1991;11:441-6. 13. Li W, Struthers AD, Fahey T, et al. Spironolactone use and renal toxicity: population based longitudinal analysis. BMJ 2010: 340-c1768. 14. Weir MR, Rolfe M. Potassium homeostasis and renin-angiotensinaldosterone system inhibitors. Clin J Am Soc Nephrol 2010;5: 531-48. 15. The RALES Investigators. Effectiveness of spironolactone added to an angiotensin-converting enzyme inhibitor and a loop diuretic for severe chronic congestive heart failure (The Randomized Aldactone Evaluation Study [RALES]. Am J Cardiol 1996;78:902-7.

Special Articles

Clinical development of pharmacologic agents for acute heart failure syndromes: A proposal for a mechanistic translational phase Mihai Gheorghiade, MD, a Peter S. Pang, MD, a,b Christopher M. O'Connor, MD, c Krishna Prasad, MD, d John McMurray, MD, e John R. Teerlink, MD, f Mona Fiuzat, PharmD, c Hani Sabbah, PhD, g and Michel Komajda, MD h Chicago, IL; Durham, NC; London, England; Glasgow, United Kingdom; San Francisco, CA; Detroit, MI; and Paris, France

Hospitalization for acute heart failure syndromes (AHFS) predicts a poor prognosis, with postdischarge mortality and rehospitalization rates reaching 45% within 60 to 90 days. Despite the use of evidence-based therapies and adherence to national process measures, these event rates have largely remained the same over the past decade. Given the current and growing burden of AHFS, there exists a substantial unmet need for novel therapies that improve outcomes. However, attempts to improve symptoms and/or reduce postdischarge events have failed to produce positive results, either because of safety and/or efficacy. These negative results may be related to the drug itself, the protocol in terms of patient selection and/or end points, and/or the trial execution. Although experts may not agree on the exact reasons to explain the lack of success to date of phase III trials in AHFS, there is agreement that clinical benefits observed in phase II trials were not reproduced in phase III trials. A different approach may be needed. In November of 2009, a meeting was held at the Food and Drug Administration with the primary purpose of identifying the reasons why benefits observed during phase II did not translate into benefits in phase III to improve future trial design. Although multiple domains of trial design were discussed, the participants identified a lack of in-depth understanding of novel molecules before pivotal trials in AHFS as a possible contributor to the disappointing results of recent large trials. In this brief report, we outline the T1 or translational phase of research for AHFS clinical development as an important first step toward greater success in AHFS clinical trials. (Am Heart J 2011;161:224-32.)

Acute heart failure syndromes (AHFS) have been defined as the gradual or rapid onset of signs or symptoms of heart failure (HF) requiring urgent therapy.1 The cost of AHFS is high; N1 million admissions occur every year in the United States, consuming the majority of the US $39 billion spent on HF care, with similar numbers in Europe.2-5 Heart failure is the most common reason for rehospitalization and the most expensive From the aExperimental Therapeutics, Center for Cardiovascular Innovation, Northwestern University Feinberg School of Medicine, Chicago, IL, bDepartment of Emergency Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL, cDuke University Medical Center, Durham, NC, dMHRA/St Thomas' Hospital, London, England, eWestern Infirmary and the British Heart Foundation Glasgow Cardiovascular Research Centre, University of Glasgow, Glasgow, United Kingdom, fSection of Cardiology, San Francisco Veterans Affairs Medical Center and School of Medicine, University of California-San Francisco, San Francisco, CA, gDivision of Cardiology, Wayne State University, Henry Ford Hospital, Detroit, MI, and hDepartment of Cardiology, Hôpital Pitié-Salpêtrière, Paris, France. Submitted August 26, 2010; accepted October 15, 2010. Reprint requests: Mihai Gheorghiade, MD, Division of Cardiology, Center for Cardiovascular Innovation, Experimental Therapeutics, Northwestern University Feinberg School of Medicine, 645 N. Michigan Avenue, Suite 1006 Chicago, IL 60611. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.10.023

hospital diagnosis for Medicare beneficiaries.6 Most importantly, the postdischarge mortality and rehospitalization rates of patients admitted with AHFS can be as high as 15% and 30%, respectively, within 60 to 90 days.7-9 Despite the use of evidence-based therapies and adherence to national process measures, these event rates have largely remained the same over the past decade.7-9 Given the current and growing burden of AHFS, there exists a substantial unmet need for novel therapies. Unfortunately, every large trial conducted to date in AHFS has failed to produce positive results in terms of efficacy and/or safety.10-18 These neutral or negative results may be related to the drug itself, the trial design in terms of patient selection and/or end points, and/or the trial execution.

Issues related to clinical trials conducted to date Novel therapies in AHFS are developed to improve signs and symptoms during hospitalization, and/or prevent death or rehospitalization postdischarge. There is no simple, all-encompassing answer to explain why


Gheorghiade et al 225

Table I. Contributors to lack of success in phase III AHFS trials Pathophysiology and epidemiology

The therapy (experimental drug or device)

The protocol

Study execution

1. 2. 3. 4. 5. 6.

Poor understanding of the pathophysiology in AHFS Heterogeneous patient population in terms of pathophysiology, etiology, and clinical presentation Uncertain relationship between hemodynamics and neurohormonal changes and clinical outcomes Clinical course (particularly soon after discharge) of AHFS patients has not been well studied. Cardiac and noncardiac comorbid conditions influence the outcome and interaction with therapy. The transition from animal studies to clinical studies has occurred without comprehensive understanding of the mechanistic properties of the drug in specific patient subgroups. Not “knowing” the drug 7. Not having the correct dose 8. Possible variation of efficacy and safety with time (given significant fluctuations in symptoms, hemodynamics, neurohormones, renal function, and myocardial injury during the course of hospitalization and postdischarge, the efficacy and/or safety of the drug may be dependent on the time of the invention) 9. The majority of drugs tested thus far reduce systemic BP, which may potentially decrease coronary perfusion, thereby contributing to myocardial and/or kidney injury 10. Patient selection 11. Surrogate end points and clinical outcomes in phase II do not predict the results of a phase III trial. 12. Choice of end points 13. Because most patients' signs and symptoms improve with standard therapy, it is difficult to prove that novel therapies are producing further and substantial improvements. 14. Selection of incorrect patients and/or less than ideal follow-up

Stage A trials include those short-term therapies (hours, days) targeted at treatment during the initial presentation, Stage B trials are conducted during hospitalization in those patients who continue to have signs and symptoms despite initial therapies (therapy administered for a few days), Stage C trials are therapies initiated shortly before or after discharge (therapy administered over longer duration; weeks or months). BP, Blood pressure.

trials to date have not safely achieved either of these goals (Table I). Although experts may not agree on the exact reasons to explain the lack of success to date of phase III trials in AHFS, there is agreement that clinical benefits observed in phase II trials were not reproduced in phase III trials. Several examples are listed in Table II. After a decade of efforts and hundreds of millions of research dollars invested, it still is not a certainty that effective therapies can be developed for AHFS. It is clear however that a different approach is needed. In November of 2009, a meeting was held at the Food and Drug Administration (FDA) with the goal of identifying the reasons why benefits observed during phase II did not translate into benefits in phase III to improve future trial design. This meeting was not sponsored by any company, organization, or governmental entity; and no extramural funding was used to support this work. The authors are solely responsible for the drafting and editing of the paper and its contents. Principal investigators from 4 major drug development programs (tolvaptan, tezosentan, levosimendan, and rolofylline) took part, as well as representatives from the European Society of Cardiology, the European Medicines Agency, and the FDA. A critical review of the clinical development of these drugs from phase II onward was presented in detail. Although multiple domains of trial design were discussed, the participants identified the lack of in-depth understanding of molecules before pivotal trials in AHFS as a possible contributor to the disappointing results seen to date. The Clinical Research Roundtable at the Institute of Medicine outlined 2 overarching obstacles to the realization of tangible health benefits, which might be

derived from the success of basic science research. They named these obstacles translational blocks42 (Figure 1). The first block, described as T1 or the first translational block,43 is “the transfer of new understandings of disease mechanisms gained in the laboratory into the development of new methods for diagnosis, therapy, and prevention and their first testing in humans.”42 The second translational block, or T2, is “the translation of results from clinical studies into everyday clinical practice and health decision making.”42 From the perspective of clinical trial design in AHFS, overcoming the T1 block was the primary focus of discussion at the FDA meeting; specifically, whether early T1 mechanistic studies—translating the beneficial effects of experimental agents observed in animal models of HF into patients with HF—might increase the likelihood of successful development.

Purpose of studies before pivotal testing Once successful animal model studies have been completed, traditional drug development follows a paradigm of sequential phases of development (Figure 2), culminating in a pivotal phase III trial. Variations on this traditional paradigm, or combination or adaptive designs have also been used in the past. After initial phase I trials, primarily conducted for safety in healthy subjects, phase II studies are designed for initial hypothesis testing in the desired target population to understand the mechanistic properties of the drug, selection of dose, and broad safety. Detailed knowledge of these broad domains, although not a guarantee, maximizes the likelihood of success while minimizing risk to patients. Knowing who to target and when is important


226 Gheorghiade et al

Table II. Clinical Trials in AHFS - From Phase II to Phase III Development program/ phase II studies Tolvaptan ACTIV in CHF19 (n = 319)

ECLIPSE20 (n = 181)

METEOR21 (n = 240)

Tezosentan RITZ-1 22 –symptom study (n = 669) RITZ-2 23 –hemodynamic study (n = 292) RITZ-4 24 –AHFS and ACS (n = 193) RITZ-5 25 –pulmonary edema study (n = 84) Levosimendan Dose ranging versus placebo/dobutamine26 (n = 151)

Dose escalation27 (n = 146) Hemodynamic study29 (n = 85) RUSSLAN 30 –LV failure post-MI (n = 504) LIDO 32 –low-output HF versus dobutamine (n = 203) CASINO33 (n = 227) REVIVE-134,35 (n = 100) Rolofylline CKI-20136 (n = 159) (146 received tx) CKI-202 36 (n = 35) (descriptive study in diuretic-resistant patients) CKI-20337 (n = 32) PROTECT–pilot38 (n = 301)

Nesiritide Nesiritide Efficacy Trial40 (n = 127) Nesiritide Comparative Trial (n = 305) VMAC18 (n = 489)

Phase II primary end point reached?

Primary end point(s)

1. In-hospital outcome: change in body weight at 24 h 2. Outpatient outcome: worsening HF (death, hospitalization, or unscheduled visit for HF) at 60 d

PCWP peak change from baseline within 3 to 8 h after treatment administration (80% power to detect 3–mm Hg difference) Change from baseline in LV end-diastolic volume index at 54-wk visit

Yes—for body weight reduction. No—for worsening HF or postdischarge mortality (in retrospect, patients with severe congestion, hyponatremia, and abnormal renal function had decreased mortality in response to oral tolvaptan) Yes (is borderline)

No. In retrospect, patients assigned to tolvaptan had a lower rate of mortality and hospitalizations.

Dyspnea at 24 h

No

Mean change in CI at 6 h

Yes

Composite of death, worsening HF, recurrent ischemia, and recurrent or new myocardial infarction within 72 h Change in oxygen saturation from baseline to 1 h

No

Proportion achieving in each treatment group at least one of the following: (1) N15% increase in SV at 23 to 24 h; (2) N25% decrease in PCWP (and N4 mm Hg) at 23 to 24 h; (3) N40% increase in CO (with change in HR b20%); (4) N50% decrease in PCWP during 2 consecutive measurements Proportion of patients with N25% increase in SV or decrease in PCWP at 6 h Changes in hemodynamics between 6 and 24 h Change in hemodynamics between 24 and 48 h Hypotension or myocardial ischemia of clinical significance Hemodynamic improvement (increase of N30% in CO and a decrease of N 25% in PCWP) at 24 h Combined death or rehospitalization for worsening HF Global outcomes at 5 d

Yes—positive dose-response relationship for all 5 doses

No

Yes Yes Yes (noninferior) Yes Yes Yes—positive trends

Total urine output at 6 h after first dose

Yes

Change in hourly urine output over 24 h and change in CrCl

Yes—positive trends

Improvement in renal function (GFR) and RBF Trichotomous classification of patients: Success: improvement in dyspnea (Likert scale— moderately or markedly better) at 24 and 48 h or day of discharge, and not meeting criteria for treatment failure. Failure: death, HF readmission within 7 d, worsening HF (by physician assessment by day 7), or persistent renal impairment. Unchanged: neither criteria for success or failure.

Yes Yes–positive trends

Change in PCWP at 6 h Global clinical status and clinical symptoms

Yes—Efficacy Trial Neutral for Comparative Trial

1. Change in PCWP at 3 h 2. Change in dyspnea (Likert) at 3 h

Yes However, retrospective analysis suggests worsening renal function and increased mortality.


Phase III studies Tolvaptan EVEREST12 (n = 4133)

EVEREST Short Term Trials11


Phase III primary end point reached?

Primary end point(s)

Contributors to lack of success per Table I⁎ 4, 6, 7. 8, 11

Long term 1. All-cause mortality (superiority and noninferiority) 2. Cardiovascular death or hospitalization for HF (superiority only)

No

Short term Composite of global clinical status and body weight reduction

Yes (driven by body weight reduction)

2, 3, 4, 5, 7, 8, 9, 10,13

Tezosentan

VERITAS13 (n = 1448)

1. Change in dyspnea (visual analog scale) over 24 h (in the individual trials) 2. Death or worsening HF at 7 d (in both trials combined)

No

3, 5, 6, 7, 9, 11

Levosimendan

REVIVE-2 28 (n = 600)

Composite of clinical signs/symptoms of HF and: 1. Patient reported moderately or markedly improved at 6 h, 24 h, and 5 d (and no criteria for worsening) 2. Worsening (death, patient reported moderate or severe deterioration at any time point) or worsening symptoms at any time, or persistent severe symptoms after 24 h requiring rescue therapy (ie, intravenous diuretic, vasodilator or inotropic agents) 3. Unchanged

Yes—excess hypotension, atrial and ventricular arrhythmias, and trends toward early mortality in levosimendan

SURVIVE31 (n = 1327)

All-cause mortality at 180 d

No

PROTECT39 (n = 2033)

No Trichotomous classification of patients: Success: improvement in dyspnea (Likert scale–moderately or markedly better) at 24 and 48 h or day of discharge, and not meeting criteria for treatment failure. Failure: death, HF readmission within 7 d, worsening HF (by physician assessment by day 7), or persistent renal impairment. Unchanged: neither criteria for success or failure.

3, 4, 7, 8, 9, 12, 13

Nesiritide

ASCEND-HF41

1, 2, 4, 5, 6, 7, 8, 10, 11, 12, 13

1. Composite all-cause mortality + HF rehospitalization at 30 d 2. Dyspnea at 6 and 24 h

No



Figure 1

Figure 2

T1 Block: Translation from Basic Sciences to Human Studies

Basic Biomedical Research

T2 Block: Translation of New Knowledge into Clinical Practice & Health Decision Making

Clinical Science & Knowledge

Improved Health

Proposal for AHFS Clinical Development: The T1 Phase

Translational blocks (adapted and reproduced with permission from Sung et al42).

both for efficacy and safety. Study design is drug/class specific. “Hard” clinical outcomes such as mortality and/or rehospitalization are not the purpose of phase II studies, which are, by design, underpowered to answer such questions. Outcomes analyses will continue to be commonly conducted, however, as pressures to continue late phase development often require such analysis. Overall, well-designed phase II studies will continue to be an important benchmark in clinical development. As previously mentioned, success in phase II has not led to success in phase III. After careful review of past major AHFS development programs, a new hypothesis was proposed: improved phase II trial design preceded by indepth mechanistic testing of the drug after animal studies but before phase II would increase the probability of success in phase III. Broadly, the purpose of such earlyphase studies would be to safely replicate and explore in detail the mechanistic efficacy seen in animal models through translational research in humans. Pharmacokinetic and pharmacodynamic studies would also be conducted because, unlike animal models to date, patients with HF are commonly on chronic background therapy for HF and have significant cardiac and noncardiac comorbidities, which might influence safety, efficacy, and outcomes. Although large animal testing remains a fundamental step in novel drug development, this phase does not fully replicate the conditions of a patient with HF. Human HF has many etiologies and comorbidities; in addition, a patient with HF typically receives extensive background therapy, both cardiac and noncardiac, before being exposed to an experimental drug. Heterogeneity of the HF population combined with long-term exposure to background therapy can potentially attenuate both the safety and efficacy signal of an experimental drug, particularly when compared with highly controlled

Current paradigm for drug development.

preclinical studies. Some of these difficulties can be overcome by exposing animals to background therapy and assessing the safety and efficacy of new drugs as “additive” to background therapy; however, this does not fully resolve the issue. Patients with HF are often on years of background therapy, a state difficult to reproduce in preclinical models.

A proposal: the T1 or mechanistic translational phase (Figure 3B) The transition from preclinical to clinical development of a drug for HF requires the utmost of care. There is an increased awareness that, absent attention to detail, specifically regarding mechanisms of drug action, patient selection and background therapy during this transition can derail an otherwise promising drug development program. The preclinical paradigm, performed under tightly controlled conditions, has not been fully explored in human AHFS clinical development, despite careful


Figure 3


Table III. T1 concept 1. A more thorough understanding of all of a molecule's effects on the heart (effects on viable but noncontractile myocardium, coronary perfusion, diastolic function, etc) is important. 2. Reproduce the results obtained in large animal HF models in homogeneous group of patients taking into account systolic and diastolic dysfunction, extent and severity of CAD, viable but dysfunctional myocardium, etc. 3. These in-depth evaluations should take advantage of recent progress made in noninvasive methods of assessment of cardiac function and structure (echocardiography, MRI spectroscopy, etc). 4. These studies would also expand our understanding of the pharmacokinetic and pharmacodynamic properties of novel molecules because, unlike animal models to date, patients with HF are commonly on background therapy for HF and have substantial comorbid conditions that might influence safety, efficacy, and outcomes. 5. These studies should be conducted in dedicated centers that have the patient population, technology, and expertise to conduct such technically challenging studies.

The current schema (A) and the proposed paradigm (B) for drug development.

inclusion/exclusion criteria. Examples from recent clinical trials are listed below. Selective vasopressin antagonists, such as tolvaptan, an oral V2 receptor antagonist, were developed without fully knowing the effects of unopposed V1 activity on the heart and the vasculature as a result of increasing vasopressin levels (Table II). Although the hypotensive effects of levosimendan were well known, the potential deleterious effects of decreased coronary perfusion were not well studied, especially in patients with coronary artery disease (CAD) (Table II). Although promising renal-protective effects of adenosine blocking agents were seen in small, early-phase trials, a well-powered study to assess the effects of these agents on renal function were not performed before pivotal trials assessing whether renal protection would lead to improved outcomes (Table II). The hemodynamic effects of tezosentan at relatively low doses were not tested in-depth before its hemodynamic properties were studied in a large trial.44 Patient selection to maximize homogeneity during early phases of T1 research forms the basis of our proposal below. Front-loading development programs to determine which patients are more likely to respond as well as

which patients to avoid will establish a foundation upon which to build the later phases of clinical development. In essence, homogeneity strengthens the signal to noise ratio. Although positive signals for efficacy and safety are ideal, clarity of the signal is the critical need during early phases of development. Positive signals may be particularly alluring, yet may be falsely positive if the negative signal was minimal or absent in a small, but heterogeneous sample size. A robust negative signal might halt development, or allow for a careful pause, before continued utilization of limited resources. Conversely, promising novel drugs may have been prematurely halted when negative signals in a small, but heterogeneous sample size drown out the positive effects. For example, patients with myocardial hibernation due to chronic ischemia may respond differently to an inotrope compared with patients with a primary cardiomyopathy. Similarly, patients with viable but dysfunctional myocardium due to causes other than ischemia may respond differently to a drug compared with those who have myocardial scarring. A dedicated mechanistic phase of development, what we have termed the T1 phase to distinguish from phase II, is proposed to ensure a seamless transition from animal studies to phase II (Table III). This phase would occur after initial human studies demonstrate safety to proceed with further development (phase 0, phase I). T1 studies will be designed in an analogous manner to animal studies, with relatively small numbers of homogenous patients for initial hypothesis testing. For example, patients entered into a T1 study would have the same HF etiology, a narrow range of age, ejection fraction, background therapy, kidney function, etc—in other words, a “homogenous” population in whom one would expect similar outcomes of similar magnitude. Such a strong emphasis on homogeneity should minimize background noise. Importantly, multiple homogenous groups will be



studied to best capture which subgroup within the overall heterogeneous HF population would be the initial target population for larger studies. However, emphasizing homogeneity will be a significant hurdle to recruitment, which should be mitigated by the relatively small sample size required for T1 studies. As these studies are mechanistic in nature, clearly defined end points (eg, change in pulmonary capillary wedge pressure, renal function) would be described. The small sample size would also allow for the introduction of uniform study standards across centers to assess cardiac function and performance and minimize geographic variation.

Technical expertise Given the mechanistic focus of T1 and the small numbers of patients, T1 studies would be very technical in nature and would use the latest advances in cardiac imaging and measurement. These advances, from magnetic resonance imaging (MRI) to new biomarkers, have not been well represented in clinical trial design to date. In terms of experience and expertise, studies requiring invasive hemodynamics are a good example, as right heart catheterization becomes less and less part of routine clinical practice. Other examples include echocardiographic expertise (eg, tissue Doppler imaging, 3-dimensional echocardiography) or other modalities (eg, cardiac magnetic resonance imaging [MRI], nuclear imaging/ testing, novel use of biomarkers, coronary perfusion, positron emission tomography, myocardial metabolism, myocardial spectroscopy) as technology and cardiovascular science continue to advance. In addition to cardiac studies, renal function, renal and coronary perfusion, and/or renal and myocardial injury should be studied in detail. With a focus on technical quality, highly skilled examiners would provide bedside quality control of all collected measures. Experts other than clinical trialists are essential to successful T1 studies. These include, but are not limited to, experts in acute kidney injury, biomarkers, imaging modalities, electrophysiology, and angiography. Finally, a small but robust consortium of experienced centers with ready access to a large HF population will be needed and is currently being created. Standardization Traditionally, both the active and placebo arms in clinical trials have used “standard” therapy as background therapy, which has generally been left to the discretion of individual investigators and study sites. However, standard therapy is not uniform, reflecting geographical norms and patient types.45 To minimize variability, standardizing therapy to the extent possible across T1 centers is recommended. This has already been attempted in a large phase IV study, ASCEND-HF.46 As a

general rule, studies involving novel agents should be placebo controlled.

Dose finding We propose, at minimum, 3 distinct doses plus placebo during T1 studies, with a clear rationale for both how the dosing was chosen and, if not, why such studies were not performed. It is necessary to ensure that doses studied and chosen have sufficient pharmacologic rationale with demonstrable dose response. The “correct” dose, by itself, may not lead to improved outcomes. However, the “wrong” dose may lead to a negative or neutral study—a good drug, good target, but wrong dose scenario. Further dose studies may be needed in phase II.

Other considerations Should we go straight into phase III? Depending on the drug, one may proceed more quickly to phase III, by either skipping phase II altogether or doing adaptive studies, where confirmatory phase II studies are designed to segue right into phase III, which has certain appeal. In such a design, pilot study results may count toward the overall pivotal trial. Well-designed, large clinical trials might better identify the “real-world” effect of novel therapies. However, at the present time, we would advocate against proceeding straight to phase III, given the lack of success to date. A more methodological approach may be necessary until a better pathophysiologic understanding of AHFS is achieved. Such due diligence needs to be carefully considered in light of key business development milestones, such as patent expiration. However, if T1 phase testing is considered early during strategic planning, appropriate timelines and milestones can be created. Patent pressures, regulatory landscape, and industry collaboration As each of the authors will attest, multiple factors are brought into play when outlining the overall strategy and then subsequent tactical plan for clinical development. The time limits of patent protection, regulatory considerations, as well as justifiable concerns by for-profit small biotech and larger pharmaceutical corporations all play a role and are just a fraction of the overall contributors and influences to drug development. Although it would be easiest to “blame” industry pressures circumventing “quality” science for the disappointing results seen to date, this not only would be an oversimplification, but would in fact be wrong. The lack of success to date in AHFS clinical trials, combined with the myriad of opinions and advice from key opinion leaders, has been cited as one reason why some trials never go forward at all. Overall, it is clear that every clinical development program ultimately remains a


hypothesis to be tested and that clarity, teamwork, consensus, and a relentless focus on execution, guided by the best science, are all requirements for successful development, thus, our proposal for a T1 phase of development. Although this may add a step to the overall timeline, this additional time would be offset by the substantial benefits such knowledge would bring, at a fraction of the overall cost of a large phase III trial.

Next steps An upcoming consensus meeting is currently being planned to discuss T1 phase trial design in more detail, which will include respective experts in various domains, such as biomarkers, renal injury, and statistical analysis, given the technical nature of T1 studies. In addition, consensus regarding end points for later stages of development is critical, that is, whether the focus should be on efficacy during the acute phase (inpatient phase), where it appears that standard therapy improves signs and symptoms, or the recovery/subacute phase (early postdischarge period), where there appears to be a vulnerable phase in terms of morbidity/mortality.47

Conclusion A significant unmet need exists for novel therapies for AHFS, given the high postdischarge mortality and rehospitalization rates despite available therapy. Although clinical trials to date have largely disappointed, much has been learned. In hindsight, phase II trials have been inconsistent in methodological rigor. This is likely a major contributor to the lack of success in phase III. A comprehensive, mechanistic understanding of a molecule before pivotal trials is critical. Such knowledge will facilitate identification of the right patient population as well as end points. We propose the T1 or translational phase as a means to achieve this end. This will require a small, but highly motivated, skilled, and experienced network of investigators to perform early translational AHFS research. The T1 phase should be followed by a carefully crafted and well-executed phase II and III trial.

Acknowledgements We would like to thank Norman Stockbridge, MD, PhD, for hosting the meeting held at the FDA in November 2009 and for his thoughtful review of this manuscript.

References 1. Gheorghiade M, Zannad F, Sopko G, et al. Acute heart failure syndromes: current state and framework for future research. Circulation 2005;112:3958-68.


2. Dickstein K, Cohen-Solal A, Filippatos G, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur Heart J 2008;29: 2388-442. 3. Fang J, Mensah GA, Croft JB, et al. Heart failure-related hospitalization in the U.S., 1979 to 2004. J Am Coll Cardiol 2008; 52:428-34. 4. Hunt SA, Abraham WT, Chin MH, et al. Focused update incorporated into the ACC/AHA 2005 guidelines for the diagnosis and management of heart failure in adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in collaboration with the International Society for Heart and Lung Transplantation. J Am Coll Cardiol 2009;53:e1-e90. 5. Lloyd-Jones D, Adams RJ, Brown TM, et al. Heart disease and stroke statistics—2010 update: a report from the American Heart Association. Circulation 2010;121:e46-e215. 6. Jencks SF, Williams MV, Coleman EA. Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med 2009; 360:1418-28. 7. Bueno H, Ross JS, Wang Y, et al. Trends in length of stay and short-term outcomes among Medicare patients hospitalized for heart failure, 1993-2006. JAMA 2010;303:2141-7. 8. Gheorghiade M, Abraham WT, Albert NM, et al. Systolic blood pressure at admission, clinical characteristics, and outcomes in patients hospitalized with acute heart failure. JAMA 2006;296: 2217-26. 9. Fonarow GC, Peterson ED. Heart failure performance measures and outcomes: real or illusory gains. JAMA 2009;302:792-4. 10. Cuffe MS, Califf RM, Adams Jr KF, et al. Short-term intravenous milrinone for acute exacerbation of chronic heart failure: a randomized controlled trial. Journal of the American Medical Association 2002;287:1541-7. 11. Gheorghiade M, Konstam MA, Burnett Jr JC, et al. Short-term clinical effects of tolvaptan, an oral vasopressin antagonist, in patients hospitalized for heart failure: the EVEREST clinical status trials. JAMA 2007;297:1332-43. 12. Konstam MA, Gheorghiade M, Burnett Jr JC, et al. Effects of oral tolvaptan in patients hospitalized for worsening heart failure: the EVEREST outcome trial. JAMA 2007;297:1319-31. 13. McMurray JJV, Teerlink JR, Cotter G, et al. Effects of tezosentan on symptoms and clinical outcomes in patients with acute heart failure: the VERITAS randomized controlled trials. JAMA 2007;298: 2009-19. 14. Mebazaa A, Nieminen MS, Packer M, et al. Levosimendan vs dobutamine for patients with acute decompensated heart failure: the SURVIVE randomized trial. JAMA 2007;297: 1883-1891. 15. Sackner-Bernstein J, Kowalski M, Fox M, et al. Short-term risk of death after treatment with nesiritide for decompensated heart failure. JAMA 2005;293:1900-5. 16. Sackner-Bernstein JD, Kowalski M, Fox M, et al. Short-term risk of death after treatment with nesiritide for decompensated heart failure: a pooled analysis of randomized controlled trials. JAMA 2005;293: 1900-5. 17. Sackner-Bernstein JD, Skopicki HA, Aaronson KD. Risk of worsening renal function with nesiritide in patients with acutely decompensated heart failure. Circulation 2005;111:1487-91.



18. VMAC Investigators. Intravenous nesiritide vs nitroglycerin for treatment of decompensated congestive heart failure: a randomized controlled trial. JAMA 2002;287:1531-40. 19. Gheorghiade M, Gattis WA, O'Connor CM, et al. Effects of tolvaptan, a vasopressin antagonist, in patients hospitalized with worsening heart failure: a randomized controlled trial. JAMA 2004; 291:1963-71. 20. Udelson JE, Orlandi C, Ouyang J, et al. Acute hemodynamic effects of tolvaptan, a vasopressin V2 receptor blocker, in patients with symptomatic heart failure and systolic dysfunction: an international, multicenter, randomized, placebo-controlled trial. J Am Coll Cardiol 2008;52:1540-5. 21. Udelson JE, McGrew FA, Flores E, et al. Multicenter, randomized, double-blind, placebo-controlled study on the effect of oral tolvaptan on left ventricular dilation and function in patients with heart failure and systolic dysfunction. J Am Coll Cardiol 2007;49:2151-9. 22. Coletta AP, Cleland JG. Clinical trials update: highlights of the scientific sessions of the XXIII Congress of the European Society of Cardiology—WARIS II, ESCAMI, PAFAC, RITZ-1 and TIME. Eur J Heart Fail 2001;3:747-50. 23. Torre-Amione G, Young JB, Colucci WS, et al. Hemodynamic and clinical effects of tezosentan, an intravenous dual endothelin receptor antagonist, in patients hospitalized for acute decompensated heart failure. J Am Coll Cardiol 2003;42:140-7. 24. O'Connor CM, Gattis WA, Adams Jr KF, et al. Tezosentan in patients with acute heart failure and acute coronary syndromes: results of the Randomized Intravenous TeZosentan Study (RITZ-4). J Am Coll Cardiol 2003;41:1452-7. 25. Kaluski E, Kobrin I, Zimlichman R, et al. RITZ-5: randomized intravenous TeZosentan (an endothelin-A/B antagonist) for the treatment of pulmonary edema: a prospective, multicenter, doubleblind, placebo-controlled study. J Am Coll Cardiol 2003;41: 204-10. 26. Nieminen MS, Akkila J, Hasenfuss G, et al. Hemodynamic and neurohumoral effects of continuous infusion of levosimendan in patients with congestive heart failure. J Am Coll Cardiol 2000;36: 1903-12. 27. Slawsky MT, Colucci WS, Gottlieb SS, et al. Acute hemodynamic and clinical effects of levosimendan in patients with severe heart failure. Study Investigators. Circulation 2000;102:2222-7. 28. Cleland JG, Freemantle N, Coletta AP, et al. Clinical trials update from the American Heart Association: REPAIR-AMI, ASTAMI, JELIS, MEGA, REVIVE-II, SURVIVE, and PROACTIVE. Eur J Heart Fail 2006; 8:105-10. 29. Kivikko M, Lehtonen L, Colucci WS. Sustained hemodynamic effects of intravenous levosimendan. Circulation 2003;107:81-6. 30. Moiseyev VS, Poder P, Andrejevs N, et al. Safety and efficacy of a novel calcium sensitizer, levosimendan, in patients with left ventricular failure due to an acute myocardial infarction. A randomized, placebo-controlled, double-blind study (RUSSLAN). Eur Heart J 2002;23:1422-32. 31. Mebazaa A, Nieminen MS, Packer M, et al. Levosimendan vs dobutamine for patients with acute decompensated heart failure: the SURVIVE Randomized Trial. JAMA 2007;297:1883-91. 32. Follath F, Cleland JG, Just H, et al. Efficacy and safety of intravenous levosimendan compared with dobutamine in severe low-output heart failure (the LIDO study): a randomised double-blind trial. Lancet 2002;360:196-202. 33. Cleland JG, Ghosh J, Freemantle N, et al. Clinical trials update and cumulative meta-analyses from the American College of Cardiology:

34.

35.

36.

37.

38.

39.

40.

41.

42.

43. 44.

45.

46.

47.

WATCH, SCD-HeFT, DINAMIT, CASINO, INSPIRE, STRATUS-US, RIO-Lipids and cardiac resynchronisation therapy in heart failure. Eur J Heart Fail 2004;6:501-8. Garratt CPM, Colucci W, Fisher L, et al. Development of a comprehensive new endpoint for the evaluation of new treatments for acute decompensated heart failure: results with levosimendan in the REVIVE I study (abstract). Presented at: 24th International Symposium on Intensive Care and Emergency Medicine, Brussels, Belgium; 2004. http://ccforum.com/content/8/S1/P89. Earl GL, Fitzpatrick JT. Levosimendan: a novel inotropic agent for treatment of acute, decompensated heart failure. Ann Pharmacother 2005;39:1888-96. Givertz MM, Massie BM, Fields TK, et al. The effects of KW-3902, an adenosine A1-receptor antagonist, on diuresis and renal function in patients with acute decompensated heart failure and renal impairment or diuretic resistance. J Am Coll Cardiol 2007;50: 1551-60. Dittrich HC, Gupta DK, Hack TC, et al. The effect of KW-3902, an adenosine A1 receptor antagonist, on renal function and renal plasma flow in ambulatory patients with heart failure and renal impairment. J Card Fail 2007;13:609-17. Cotter G, Dittrich HC, Weatherley BD, et al. The PROTECT pilot study: a randomized, placebo-controlled, dose-finding study of the adenosine A1 receptor antagonist rolofylline in patients with acute heart failure and renal impairment. J Card Fail 2008;14: 631-40. Massie BM, O'Connor CM, Metra M, et al. Rolofylline, an adenosine A1-receptor antagonist, in acute heart failure. N Engl J Med 2010; 363:1419-28. Colucci WS, Elkayam U, Horton DP, et al. Intravenous nesiritide, a natriuretic peptide, in the treatment of decompensated congestive heart failure. Nesiritide Study Group. N Engl J Med 2000;343: 246-53. Hernandez AF, O'Connor CM, Starling RC, et al. Rationale and design of the Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure Trial (ASCEND-HF). Am Heart J 2009; 157:271-7. Sung NS, Crowley Jr WF, Genel M, et al. Central challenges facing the national clinical research enterprise. JAMA 2003;289: 1278-87. Woolf SH. The meaning of translational research and why it matters. JAMA 2008;299:211-3. Cotter G, Kaluski E, Stangl K, et al. The hemodynamic and neurohormonal effects of low doses of tezosentan (an endothelin A/B receptor antagonist) in patients with acute heart failure. Eur J Heart Fail 2004;6:601-9. Blair JE, Zannad F, Konstam MA, et al. Continental differences in clinical characteristics, management, and outcomes in patients hospitalized with worsening heart failure results from the EVEREST (Efficacy of Vasopressin Antagonism in Heart Failure: Outcome Study with Tolvaptan) program. J Am Coll Cardiol 2008;52: 1640-8. Ezekowitz JA, Hernandez AF, Starling RC, et al. Standardizing care for acute decompensated heart failure in a large megatrial: the approach for the Acute Studies of Clinical Effectiveness of Nesiritide in Subjects with Decompensated Heart Failure (ASCEND-HF). Am Heart J 2009;157:219-28. Gheorghiade M, Bonow RO. Heart failure: early follow-up after hospitalization for heart failure. Nat Rev Cardiol 2010;7: 422-4.

Lessons learned from a pediatric clinical trial: The Pediatric Heart Network Angiotensin-Converting Enzyme Inhibition in Mitral Regurgitation Study Jennifer S. Li, MD, MHS, a ,j Steven D. Colan, MD, b,j Lynn A. Sleeper, ScD, b,j Jane W. Newburger, MD, MPH, c,j Victoria L. Pemberton, RNC, MS, d,j Andrew M. Atz, MD, e,j Meryl S. Cohen, MD, f,j Fraser Golding, MD, g,j Gloria L. Klein, b,j Ronald V. Lacro, MD, c,j Elizabeth Radojewski, RN, g,j Marc E. Richmond, MD, h,j and L. LuAnn Minich, MD i,j Durham, NC; Watertown and Boston, MA; Bethesda, MD; Charleston, SC; Philadelphia, PA; Ontario, Canada; New York, NY; and Salt Lake City, UT

Background Mitral regurgitation is the most common indication for reoperation in children following repair of atrioventricular septal defect (AVSD). We hypothesized that angiotensin-converting enzyme inhibitor therapy would decrease the severity of mitral regurgitation and limit left ventricular volume overload in children following AVSD repair. Methods The Pediatric Heart Network designed a placebo-controlled randomized trial of enalapril in this population. The primary aim was to test the effect of enalapril on the change in left ventricular end-diastolic dimension body surface area– adjusted z score. Before the launch of the trial, a feasibility study was performed to estimate the number of patients with at least moderate mitral regurgitation following AVSD repair. Trial experience Seventeen months after the start of the study, 349 patients were screened, 8 were trial eligible, and only 5 were enrolled. The study was subsequently terminated because of low patient accrual. Several factors led to the problems with patient accrual, including (1) the use of criteria to assess disease severity in the feasibility study that were not identical to those used in the trial, (2) failure to achieve equipoise for the study among clinicians and referring physicians, (3) reliance on methodology developed in adult populations with different disease mechanisms, and (4) absence of adequate data to define the natural history of the disease process under study. Progress in the treatment of children with cardiovascular disease will depend on the future of multicenter collaborative clinical trials. The lessons learned from this study may contribute to improvements in this research. (Am Heart J 2011;161:233-40.) Randomized clinical trials have resulted in remarkable advances in cardiovascular care. During the 1990s, results from more cardiovascular clinical trials were published than in the previous 3 decades combined, ushering in the current era of “evidence-based cardiovascular medicine.”1 Cardiovascular controlled trials have established novel treatments resulting in major improvements in patient outcomes and have also enhanced our understanding of From the aDuke University Medical Center, Durham, NC, bNew England Research Institute, Watertown, MA, cChildren's Hospital, Boston, MA, dNational Heart, Lung, and Blood Institute, NIH, Bethesda, MD, eMedical University of South Carolina, Charleston, SC, f Children's Hospital of Philadelphia, Philadelphia, PA, gHospital for Sick Children, Toronto, Ontario, Canada, hColumbia University, New York, NY, and iUniversity of Utah, Salt Lake City, UT. j For the Pediatric Heart Network Investigators. ClinicalTrials.gov Identifier: NCT00113698. J. Michael DiMaio, MD served as guest editor for this article. Submitted July 21, 2010; accepted October 19, 2010. Reprint requests: Jennifer S. Li, MD, MHS, Division of Cardiology, Department of Pediatrics, Duke University Medical Center, Duke Clinical Research Institute, PO Box 17969, Durham, NC 27715. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.10.030

heart disease and the impact of risk factors and adverse events. Certain populations, however, have been underrepresented in cardiovascular clinical trials, including women, the elderly, minority populations, and children.1-3 Many barriers impede the design and conduct of randomized clinical trials in children, including the relative rarity of specific diseases, disease heterogeneity, incompletely defined natural history, lack of research infrastructure, ethical issues in pediatric research, and difficulty in identifying valid clinical end points. Systematic controlled studies of medications in children with congenital heart disease have thus been limited, and most medications are not labeled for pediatric use.4 Therefore, treatment decisions in this population are often based on clinical experience, small observational studies, and extrapolation from adult data, rather than clinical trial evidence. In an attempt to narrow the scientific gap in the pediatric cardiovascular population, the National Heart, Lung, and Blood Institute (NHLBI) established the Pediatric Heart Network (PHN) in 2001.5 In 2004, the PHN launched a placebo-controlled randomized trial of the angiotensin-converting enzyme (ACE) inhibitor enalapril for use in infants and children

234 Li et al

with mitral regurgitation following atrioventricular septal defect (AVSD) repair. Our failure to complete this trial has led us to reflect on crucial issues that arose in its design and execution. These issues should be of interest to investigators, clinical practitioners, research networks, participating institutions, sponsors, data and safety monitoring boards (DSMBs), and families of the children whom we serve. Our purpose is to encourage discussion so that standards can be developed to better inform future trials in pediatric medicine.

The PHN ACE Inhibition in Mitral Regurgitation experience Rationale Mitral regurgitation remains the most common indication for reoperation (about 10% of patients) in children following repair of AVSD. Postoperatively, moderate mitral regurgitation occurs in 15% to 30% and severe regurgitation occurs in 5% to 19% of patients.6-10 Residual mitral regurgitation causes volume overload and places a hemodynamic burden on the left ventricle that induces a series of compensatory adjustments, including up-regulation of the renin-angiotensin-aldosterone system.11,12 Although these adjustments may initially restore cardiac output, the disease process is often progressive; and compensatory mechanisms fail. Reports are conflicting from trials of ACE inhibitor therapy for the treatment of mitral regurgitation in both the adult and pediatric populations, and use of ACE inhibitors is not recommended for treatment of mitral regurgitation in the current consensus paper from the American College of Cardiology and the American Heart Association.13-17 Despite this, we have observed an escalating trend toward adoption of this therapy in children, based primarily on data indicating a favorable response to acute ACE inhibitor therapy in children.14,15 We hypothesized that ACE inhibition therapy would decrease the severity of mitral regurgitation and thus limit left ventricular volume overload in children following AVSD repair. We designed the ACE Inhibition in Mitral Regurgitation (ACEi in MR) trial to evaluate the efficacy and safety of the ACE inhibitor enalapril for the treatment of significant mitral regurgitation in children (ClinicalTrials.gov ID: NCT00113698) (Figure 1). Children who had residual mitral (left atrioventricular valve) regurgitation after repair of an AVSD were selected as the study population for 2 reasons: (1) the regurgitant orifice in AVSD responds dynamically to left ventricular size and therefore might respond to medical therapy, and (2) this is a reasonably homogeneous group with a relatively high incidence of at least moderate mitral regurgitation postoperatively. Study design Before the trial, we performed a feasibility study. Each network center undertook a chart review of the


assessment of mitral regurgitation in patients following AVSD repair according to local echocardiography reports from 2003 to 2004. The degree of mitral regurgitation was collected at each center from the reports determined by subjective assessment of the color Doppler width or area (eg, none, trivial, mild, moderate, or severe). The image data were not re-reviewed, and data on left ventricular size were not collected for the purposes of the feasibility study. A total of 1,005 reports covering a 5-year period from July 1, 1997, to June 30, 2002, were reviewed; and 195 patients were identified as having at least moderate mitral regurgitation based on subjective assessment following AVSD repair. Based on these reports, we estimated the incidence of at least moderate mitral regurgitation to be 20% in postoperative AVSD patients followed at PHN centers. The primary aim of the trial (drug phase) was to test the effect of enalapril therapy on the 6-month change in left ventricular end-diastolic dimension (LVEDD) body surface area (BSA)–adjusted z score. Secondary aims included assessment of the effects of enalapril on changes in echocardiographic measures of left ventricular geometry and hemodynamics (ejection fraction, regurgitant fraction, fiber stress, sphericity index, and BSA-adjusted z scores for mass and end-diastolic and end-systolic dimensions and volumes), the change in level of neurohormonal activation evaluated by measurement of B-type natriuretic peptide, and the incidence of adverse effects. Two groups of patients who had undergone repair of an AVSD were targeted for the study. Inclusion and exclusion criteria are shown in Table I. Group 1 patients were to be evaluated at the time of AVSD repair. Eligible patients ≤5 years of age were to be enrolled and observed for 6 months in an observational phase to allow the heart to adapt to the surgical intervention and thus to allow the effects of ACE inhibition to be assessed independently of the acute results of the surgery. Group 2 patients were those who had had their surgery performed either before the study launch or at non-PHN institutions but were referred to a PHN center, so that they could not be enrolled into the observational phase. At ≥6 months after surgery, subjects in both groups with at least moderate mitral regurgitation were to be randomized to receive enalapril or placebo for 6 months. We defined moderate mitral regurgitation on the study echocardiogram used to assess trial eligibility as (1) either a proximal regurgitant jet area ≥6 mm2/m2 or a regurgitant fraction ≥30% (extrapolating from data on adults with mitral regurgitation) and (2) left ventricular volume overload as shown by the presence of an LVEDD BSA-adjusted z score of ≥2.0. 18 An echocardiography core laboratory centrally assessed all study echocardiograms. We assumed an SD of 0.75 for the change in LVEDD BSA-adjusted z score in conjunction with a 2-sided 0.05


Li et al 235

Figure 1

Angiotensin-Converting Enzyme Inhibition in Mitral Regurgitation trial.

type I error rate, 85% power, a minimum clinically significant group difference in mean z score change of 0.5, and one interim look at the data. These assumptions resulted in a total required sample size of 82 subjects. A loss to follow-up rate of 10% was assumed to occur over the 6 months. Thus, the total required sample size was 92 subjects (46 per group) for the drug phase. All participating centers received Institutional Review Board approval. A DSMB appointed by the NHLBI monitored the conduct of the study. Written informed consent was obtained from the parent or guardian of each patient participating in either phase of the study. Assent from the subject was obtained when age appropriate.

Enrollment The observational phase of the ACEi in MR trial began on June 1, 2004; and the start date for randomization in the drug phase was 6 months later, on December 1,

2004. Recruitment in the drug phase was estimated to require 2 years. Two groups of patients were screened for eligibility—those who completed the observational phase (group 1) and those who were already ≥6 months postoperative from their repair (group 2). By April 2005, 257 patients from both groups were screened; but only one patient had been randomized in the drug phase. With the goal of increasing study enrollment, the protocol was amended to broaden the drug phase eligibility criteria by (1) extending the upper age limit for eligibility from 5 to 18 years and (2) lowering the LVEDD BSA z score eligibility criterion from 2.0 to 1.5 to be more inclusive of borderline left ventricular dilation. In addition, community outreach and advertising to referring cardiologists and advocacy groups were implemented. An additional 92 patients were subsequently screened for a total of 349 patients screened by November 2005 (Figure 2).


236 Li et al

Table I. Inclusion and exclusion criteria Observational phase

Inclusion criteria

Exclusion criteria

Drug phase

Inclusion criteria

Exclusion criteria

No longer requires intensive care ≤28 d after repair of an AVSD (including primum atrial septal defect, transitional AVSD, and complete AVSD) Age ≤5 y Complete transthoracic echocardiogram ≤28 d after repair Informed consent of legal guardian Tetralogy of Fallot and total or partial anomalous venous connection Associated mitral stenosis with a mean diastolic gradient of N10 mm Hg Other sources of LV volume overload, (eg, Nmild aortic regurgitation, defined as a regurgitant jet/annulus ratio N20%, or significant left-to right shunt, defined as an echocardiographically determined Qp/Qs ≥1.5 Associated obstructive LV outflow lesions that are more than trivial, defined as an LV outflow peak instantaneous gradient of N20 mm Hg Significant residual coarctation, defined as an arm/leg systolic pressure difference of N20 mm Hg Associated hypertrophic obstructive cardiomyopathy ≥6 m postoperation from repair of an AVSD (primum atrial septal defect, transitional AVSD, and complete AVSD) Age ≤5 y At least moderate mitral regurgitation, defined as both LV end-diastolic dimension BSA-adjusted z score ≥2.0 AND Either a proximal regurgitant jet area ≥6 mm2/m2 or a regurgitant fraction ≥30% Asymptomatic or minimally symptomatic, defined by Ross or NYHA heart failure class I or II Atrioventricular synchrony (paced or intrinsic) Girls of child-bearing potential must either abstain from sexual activity or provide assurance of birth control. Informed consent of legal guardian and assent when appropriate Tetralogy of Fallot, total or partial anomalous venous connection, and obstructed pulmonary venous return Reoperation for mitral regurgitation after the initial repair of an AVSD LV dysfunction, defined as an ejection fraction b55% Treatment with an ACE inhibitor within 6 m of randomization Associated mitral stenosis with a mean diastolic gradient of N10 mm Hg Other sources of LV volume overload, including Nmild aortic regurgitation, defined as a regurgitant jet/annulus ratio N20%, or significant left-to right shunt, defined as an echocardiographically determined Qp/Qs ≥1.5 Systemic hypertension, defined as sustained systolic or diastolic blood pressure ≥95th percentile for age Associated obstructive LV outflow with an LV outflow peak instantaneous gradient of N20 mm Hg Significant residual coarctation with an arm/leg systolic pressure difference of N20 mm Hg Associated hypertrophic obstructive cardiomyopathy History of chronic renal or hepatic dysfunction or persistent neutropenia Severe mitral regurgitation that is expected to need surgery ≤6 m Residual postoperative lesions that are expected to require additional surgery ≤6 m Need for ACE inhibitor therapy expected within 6 m in the opinion of the attending cardiologist History of intolerance to ACE inhibitors Pregnancy or intent to become pregnant

LV, Left ventricular; NYHA, New York Heart Association.

In total, only 8 patients were ultimately trial eligible and 5 were enrolled, all from group 2 (ie, subjects who were ≥6 months postoperative). The PHN ACEi in MR Subcommittee and the PHN Steering Committee reviewed these patient accrual numbers and recommended that the drug phase of this study be halted because the recruitment rate was much lower than expected. The DSMB of the PHN met on November 4, 2005, and concurred with the recommendation that was accepted by NHLBI.

Cost Pediatric Heart Network clinical sites receive an infrastructure budget that supports personnel and administrative costs for all studies. For each study, the sites also receive reimbursement for expenses related to screening, enrollment, and follow-up of subjects. For this study, the sites received $150 per patient for screening; $2,150 per patient for the observational phase; and $10,120 per patient for the drug phase. Based on 196 patients screened (but not enrolled in the observational phase),


Li et al 237

Figure 2 Group 1: ≤28 days postop (N = 84)

Group 2: ≥ 6 mo postop (N = 265)

Screened for Trial (N = 349)

At least mild to moderate MR * (N = 139)

Potentially trial eligible (N = 54)

At least moderate MR** (N = 9)

Residual shunt (N = 1)

ACE inhibitor use (N = 47)

(BY CORE LAB ECHO)

(BY LOCAL ECHO)

Associated complex defect (N = 8)

Less than moderate MR (N = 42)

Trial eligible & no exclusion criteria (N = 8)

Other clinical exclusion (N = 17)

<Mild MR or unknown MR (N = 210)

Not approached (N = 13)

Unknown MR (N = 3)

*Qualitative assessment **Quantitative assessment MR, mitral regurgitation

Randomized (N = 5)

Patient flow diagram.

153 patients in the observational phase, and 5 patients in the drug phase (as of November 4, 2005), a total of $408,950 was paid to the PHN centers for patient-related costs for the entire study; but only $50,600 was spent on patients enrolled in the trial. This represents just 2% of the total patient care budget for the PHN for the study period. The personnel, administrative, and data-related costs for the clinical sites, the Data Coordinating Center, and core laboratory are difficult to estimate because they are spread across multiple studies; but it is unlikely that these represented more than another $100,000 for the drug phase. The total expenditure was able to be limited because of careful tracking and early recognition of inadequate enrollment despite multiple adjustments to the protocol and the agreement of the research team to terminate the drug arm of the study.

Factors leading to underaccrual In this NHLBI-funded randomized trial of ACE inhibition to treat mitral regurgitation, we enrolled only 5% of the intended study sample and spent substantial effort before enrollment was suspended. It was disheartening that, despite screening 349 subjects, only 8 were eligible; and only 5 of the planned 92 participants were randomized. This experience, however, is not atypical: N80% of all clinical trials fall short of their accrual goals.19,20 As a result, trials are stopped, timelines are extended, or the studies accrue inadequate

sample sizes, resulting in CIs that are so wide or power so low that they are uninterpretable. As randomized trials involving patients with congenital heart disease are more commonly undertaken, it is worthwhile to analyze why this trial failed. We believe that several factors likely led to this outcome. 1. The available sample size was estimated from a retrospective chart review at participating centers using criteria that differed from trial entry criteria. Our feasibility study involved chart review of the subjective assessment of mitral regurgitation according to echocardiogram reports from the year before the design phase; the incidence of at least moderate mitral regurgitation was estimated to be 20%. In contrast, our trial entry criteria required quantitative measurements to operationalize the definition of moderate to severe mitral regurgitation instead of the qualitative grading assessment used in the chart review. Some echocardiographic measurements specified in the entry criteria, for example, regurgitant fraction and proximal regurgitant jet area, were not routinely available in the retrospective feasibility study. Other echocardiographic measurements used in the trial entry criteria, such as the degree of left ventricular dilation secondary to mitral regurgitation detectable within 6 months of surgery, were not

238 Li et al

collected in the feasibility study. If we had conducted the feasibility study using the defined trial eligibility criteria, we might have been alerted to the inadequate sample size available for this trial. 2. Clinical equipoise could not be established at the centers. The ethics of clinical research require equipoise—a state of uncertainty on the part of the clinical investigator regarding the comparative therapeutic merits of each arm in a trial.21 We, as PHN investigators, had clinical equipoise with each arm in the drug trial based on conflicting data reported from trials of ACE inhibitor therapy for the treatment of mitral regurgitation in both the adult and pediatric populations.12-16 Equipoise, however, could not be uniformly established among our colleagues at the participating centers. A total of 47 patients screened for the trial with at least mild to moderate mitral regurgitation noted in the chart were already on ACE inhibitors or had ACE inhibitor use planned by their primary cardiologists. These physicians may have felt compelled to use medical therapy because they believed that such therapy might delay the need for a second mitral valve surgery despite the lack of evidence showing benefit of ACE inhibitors in this setting. Others may have used ACE inhibitors because they were not fully informed of the study and used the medicine routinely in their patients after AVSD surgery. This culture of off-label use of drugs in pediatrics is historically based because most drugs before recent years have not been labeled for use in children. Off-label use of cardiovascular medications in children hospitalized with congenital and acquired heart disease remains common. A recent study analyzed data from 31,432 children in the Pediatric Health Information System database hospitalized during 2005 with cardiovascular disease and showed that 78% received ≥1 cardiovascular medication off-label and 31% received ≥3 cardiovascular medications off-label.4 The practice of off-label use without adequate information on drug efficacy and safety places children at risk for adverse events and ineffective therapy. In 9 reports evaluating pediatric drug-related adverse events, off-label prescriptions were involved in 23% to 60%.3 These data highlight the need for further study to determine which treatments are effective and should be used more frequently and which are unsafe or ineffective in children hospitalized with cardiovascular disease. In launching this trial, we did not fully appreciate how daunting it could be to change the culture among pediatric cardiologists from experiential practice to evidence-based medicine. 3. Quantitative eligibility criteria for the trial had not been verified in children. Of 349 patients screened, only 8 met the study definition of at least moderate mitral regurgitation and


left ventricular dilation, and had no exclusion criteria. Many quantitative methods are used to estimate the severity of mitral regurgitation including area of the regurgitant jet, size of the regurgitant orifice, regurgitant fraction, and effective regurgitant orifice area by proximal isovelocity surface area.18,22,23 For inclusion in this trial, we defined at least moderate mitral regurgitation to require either a proximal regurgitant jet area ≥6 mm2/m2 or a regurgitant fraction ≥30% with evidence of left ventricular dilation demonstrated by a LVEDD BSAadjusted z score ≥2 (later 1.5). Our definition was of necessity adapted from published guidelines in adults.18 A number of the techniques recommended in adult guidelines have not been evaluated in children with AVSD. We relied on the presence of left ventricular dilation in conjunction with either a regurgitant jet or a regurgitant fraction consistent with at least moderate regurgitation. Regurgitant jet area rather than jet width was selected because the regurgitant orifice in AVSD is nearly always noncircular. The selection of the specific criteria for the proximal regurgitant jet area was based on adjustment of the published adult criteria for BSA. The selection of these 2 criteria combined a quantitative measure of the severity of mitral regurgitation with evidence of left ventricular enlargement. We included left ventricular dilation as an entry criterion based on the data that left ventricular volume overload resulting from the mitral regurgitation is the primary cause of the clinically important outcomes that therapy is intended to avert.10,11 Of the 349 screened patients with initial local echocardiograms available for analysis (Figure 2), 40% (139/349) were judged to have greater than mild mitral regurgitation and 18% were judged to have moderate to severe mitral regurgitation by the site investigators using qualitative assessment on the local echocardiogram. Of these, however, only 54 patients had no other exclusion criteria (including ACE inhibitor use) based on the local echocardiogram and received a study echocardiogram to assess quantitative criteria (Figure 2). Subsequently, only 8 met the study definition of at least moderate mitral regurgitation (regurgitant fraction ≥30% or proximal regurgitant jet area/BSA ≥6 mm2/m2) with left ventricular dilation (defined as LVEDD z score ≥2 [later 1.5]) in the absence of any newly identified clinical exclusion criteria. For 3 patients, the degree of mitral regurgitation was indeterminate. Analysis of the 42 remaining patients showed that none had a regurgitant fraction N30% or an LVEDD z score ≥2 (later 1.5) and that 81% had a proximal regurgitant jet area/BSA N6 mm2/m2. Thus, most patients were excluded from trial entry because of the absence of left ventricular dilation. We therefore attempted to broaden this requirement by decreasing the LVEDD z score requirement to ≥1.5; but ultimately, this decrease had little effect on enrollment. The lack of left ventricular dilation may be related to the possibility that a 6-month period was not long enough to


observe left ventricular remodeling. Another possibility is that young children following AVSD surgery may have restrictive left ventricular physiology that inhibits dilation with short-term volume overload. Alternatively, we cannot exclude the possibility that the proximal regurgitant jet area value of 6 mm2/m2 overestimated the actual degree of mitral regurgitation because the left ventricle did not dilate. Neither this method of quantitative assessment of mitral regurgitation nor echocardiographic estimation of regurgitant fraction has been validated in children after repair of AVSD. In this study, we presumed that left ventricular dilation would be a mechanistic consequence of significant mitral regurgitation; but in fact, we failed to detect ventricular dilation in a significant number of subjects who met adult quantitative criteria for at least moderate mitral regurgitation. Because the pathophysiologic consequences of mitral regurgitation are mediated through left ventricular volume overload, it would be difficult to justify a clinical trial of medical therapy for mitral regurgitation in children who have no evidence of left ventricular volume overload. Before initiating this trial, there were insufficient natural history data to define the frequency and timing of onset of ventricular dilation as a consequence of mitral regurgitation after repair of AVSD; and this lack of natural history data continues to be a barrier to the design and execution of trials of this kind.

Conclusion Despite our detailed planning and execution of a feasibility analysis, the slow enrollment rate and ultimate failure of this trial were due to several identifiable factors: 1. Use of criteria to assess disease severity in the feasibility study that were not identical to those used in the trial. 2. Failure to achieve equipoise for the study among clinicians and referring physicians. 3. Reliance on methodology developed in adult populations with different disease mechanisms. 4. Absence of adequate data to define the natural history of the disease process under study. The PHN's sustainable infrastructure, which supports multiple simultaneous studies, meant that we did not have to dismantle an entire clinical trial structure and thus were able to minimize the financial losses associated with closing this trial. Given evidence indicating different toxicities and benefits for drugs in children compared with adults,24-26 pediatric clinical trials are necessary to appropriately assess therapeutic agents. When there is limited market value in conducting such trials, the National Institutes of Health becomes one of the few organizations willing to sponsor such randomized trials. Progress in the treatment

Li et al 239

of children with cardiovascular disease will depend on the future of these multicenter collaborative clinical trials. Studies that better define the natural history of pediatric cardiovascular conditions will facilitate the conduct of randomized clinical trials involving these conditions.27-29 We have used the lessons learned from this study to better plan subsequent studies within the PHN. This includes the ongoing randomized clinical trial of atenolol versus losartan in individuals with Marfan syndrome for which the criteria used in the feasibility phase were identical to those being used in the actual trial.30 We hope that the issues discussed and the lessons learned from this study will facilitate improved research for children.

Disclosures Supported by U01 grants from the National Heart, Lung, and Blood Institute (HL068269, HL068270, HL068279, HL068281, HL068285, HL068292, HL068290, HL068288, HL085057).

References 1. Lee PY, Alexander KP, Hammill BG, et al. Representation of elderly persons and women in published randomized trials of acute coronary syndromes. JAMA 2001;286:708-13. 2. Heiat A, Gross CP, Krumholz HM. Representation of the elderly, women, and minorities in heart failure clinical trials. Arch Intern Med 2002;162:1682-8. 3. Shah SS, Hall M, Goodman D, et al. Off-label drug use in hospitalized children. Arch Pediatr Adolesc Med 2007;161:282-90. 4. Pasquali SK, Hall M, Slonim AD, et al. Off-label use of cardiovascular medications in children hospitalized with congenital and acquired heart disease. Circ Cardiovasc Qual Outcomes 2008;1:74-83. 5. Mahony L, Sleeper LA, Anderson PA, et al. The Pediatric Heart Network: a primer for the conduct of multicenter studies in children with congenital and acquired heart disease. Pediatr Cardiol 2006; 27:191-8. 6. Hanley FL, Fenton KN, Jonas RA, et al. Surgical repair of complete atrioventricular canal defects in infancy. Twenty-year trends. J Thorac Cardiovasc Surg 1993;106:387-94 [discussion 394-7]. 7. Moran AM, Daebritz S, Keane JF, et al. Surgical management of mitral regurgitation after repair of endocardial cushion defects: early and midterm results. Circulation 2000;102(19 Suppl 3):III 160-5. 8. Crawford Jr FA, Stroud MR. Surgical repair of complete atrioventricular septal defect. Ann Thorac Surg 2001;72:1621-8 [discussion 1628-9]. 9. Michielon G, Stellin G, Rizzoli G, et al. Repair of complete common atrioventricular canal defects in patients younger than four months of age. Circulation 1997;96(9 Suppl):II-316-22. 10. Ten Harkel AD, Cromme-Dijkhuis AH, Heinerman BC, et al. Development of left atrioventricular valve regurgitation after correction of atrioventricular septal defect. Ann Thorac Surg 2005;79:607-12. 11. Pisacane C, Pacileo G, Santoro G, et al. New insights in the pathophysiology of mitral and aortic regurgitation in pediatric age: role of angiotensin-converting enzyme inhibitor therapy. Ital Heart J 2001;2:100-6. 12. Gaasch WH, Aurigemma GP. Inhibition of the renin-angiotensin system and the left ventricular adaptation to mitral regurgitation. J Am Coll Cardiol 2002;39:1380-3.

240 Li et al

13. Wisenbaugh T, Sinovich V, Dullabh A, et al. Six month pilot study of captopril for mildly symptomatic, severe isolated mitral and isolated aortic regurgitation. J of Heart Valve Dis 1994;3:197-204. 14. Calabro R, Pisacane C, Pacileo G, et al. Hemodynamic effects of a single oral dose of enalapril among children with asymptomatic chronic mitral regurgitation. Am Heart J 1999;138(5 Pt 1): 955-61. 15. Gupta DK, Kapoor A, Garg N, et al. Beneficial effects of nicorandil versus enalapril in chronic rheumatic severe mitral regurgitation: six months follow up echocardiographic study. J Heart Valve Dis 2001; 10:158-65. 16. Seguchi M, Nakazawa M, Momma K. Effect of enalapril on infants and children with congestive heart failure. Cardiol Young 1992;2: 14-9. 17. Bonow RO, Carabello B, de Leon AC, et al. ACC/AHA guidelines for the management of patients with valvular heart disease. Executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients with Valvular Heart Disease). J Heart Valve Dis 1998;7:672-707. 18. Zoghbi WA, Enriquez-Sarano M, Foster E, et al. Recommendations for the evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiog 2003;16:777-802. 19. Rendell JM, Licht RW. Under-recruitment of patients for clinical trials: an illustrative example of a failed study. Acta Psychiatr Scand 2007; 115:337-9. 20. Gajewski BJ, Simon SD, Carlson SE. Predicting accrual in clinical trials with Bayesian posterior predictive distributions. Stat Med 2008; 17:2328-40.


21. Freedman B. Equipoise and the ethics of clinical research. N Engl J Med 1987;317:141-5. 22. Alharthi MS, Mookadam F, Tajik AJ. Echocardiographic quantitation of mitral regurgitation. Expert Rev Cardiovasc Ther 2008;6:1151-60. 23. Patel AR, Mochizuki Y, Yao J, et al. Mitral regurgitation: comprehensive assessment by echocardiography. Echocardiography 2000; 17:275-83. 24. Cuzzolin L, Atzei A, Fanos V. Off-label and unlicensed prescribing for newborns and children in different settings: a review of the literature and a consideration about drug safety. Expert Opin Drug Safety 2006;5:703-18. 25. Roberts R, Rodriguez W, Murphy D, et al. Pediatric drug labeling: improving the safety and efficacy of pediatric therapies. JAMA 2003; 290:905-11. 26. Benjamin Jr DK, Smith PB, Murphy MD, et al. Peer-reviewed publication of clinical trials completed for pediatric exclusivity. JAMA 2006;296:1266-73. 27. McCrindle BW, Zak V, Sleeper LA, et al. Laboratory measures of exercise capacity and ventricular characteristics and function are weakly associated with functional health status after Fontan procedure. Circulation 2010;121:34-42. 28. Williams IA, Sleeper LA, Colan SD, et al. Functional state following the Fontan procedure. Cardiol Young 2009;15:1-11. 29. Anderson PA, Sleeper LA, Mahony L, et al. Contemporary outcomes after the Fontan procedure: a Pediatric Heart Network multicenter study. J Am Coll Cardiol 2008;52:85-98. 30. Lacro RV, Dietz HC, Wruck LM, et al. Rationale and design of a randomized clinical trial of beta blocker therapy (atenolol) vs. angiotensin II receptor blocker therapy (losartan) in individuals with Marfan syndrome. Am Heart J 2007;154:624-31.

Curriculum in Cardiology

Atrial fibrillation, anticoagulation, fall risk, and outcomes in elderly patients Matthew B. Sellers, MD, a and L. Kristin Newby, MD, MHS a,b,c Durham, NC

Atrial fibrillation (AF) affects 2.5 million patients in the United States. The incidence of this condition increases with age, such that approximately 5% of people N65 years of age have AF. Because of the lack of organized atrial contraction and thrombus formation in the left atrium, patients with AF are at increased risk of stroke. The estimated risk of stroke among all AF patients is 5% per year. Among patients without mitral stenosis, there is a graded relationship of stroke risk with the number of CHADS2 risk factors. Warfarin is the recommended treatment for embolic stroke prophylaxis in AF in intermediate- to high-risk patients. However, elderly patients who are deemed to be at risk of falls are often not started on warfarin therapy secondary to a perceived higher risk of bleeding complications. These risks have been evaluated, but conclusive data regarding the riskbenefit trade-off are elusive. This review summarizes available data on the use of warfarin in elderly patients with AF, focusing on the risk of bleeding, and will specifically address the utility of falls risk assessment in the decision to initiate warfarin therapy for AF. (Am Heart J 2011;161:241-6.)

Atrial fibrillation (AF) is the most common cardiac arrhythmia and currently affects nearly 2.5 million people in the United States. Approximately 5% of people N65 years of age carry this diagnosis1; and with a growing geriatric population in the United States, this proportion is expected to increase by 2.5-fold over the next 50 years.2 The risk of stroke in the setting of AF increases with age and may be as high as 23.5% in patients 80 to 90 years of age.3 Furthermore, not only is AF associated with an increased risk of stroke; but the Framingham Study also demonstrated that mortality from AF-related strokes is almost double that of strokes unrelated to AF, and functional deficits after AF-related strokes were more likely to be severe.4 Thus, anticoagulation therapy to prevent stroke is of paramount importance in patients at high risk for thromboembolic stroke. However, the bleeding risk associated with warfarin therapy has led physicians to be cautious in using (and, arguably, to underuse) warfarin in older patients, especially those perceived as being at risk for falls.5,6 Some of these fears may stem from studies showing that antithrombotic therapy can double the risk of intracranial hemorrhage (ICH), especially fatal hemorrhagic events.7-9 However, several other studies have called into question From the aDepartment of Medicine, Duke University Medical Center, Durham, NC, b Division of Cardiology, Duke University Medical Center, Durham, NC, and cThe Duke Clinical Research Institute, Durham, NC. George J. Klein, MD served as guest editor for this article. Submitted April 2, 2010; accepted November 2, 2010. Reprint request: L. Kristin Newby, MD, MHS, Duke Clinical Research Institute, DUMC Box 17969, Durham, NC 27715-7969. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.11.002

the use of “falls risk” as a contraindication to antithrombotic therapy in the setting of AF for high-risk patients10,11; indeed, there is little agreement on how to define or assess “falls risk.” Consequently, elderly patients who have the highest risk of stroke and of worse outcomes with stroke in the setting of AF3,4 may be frequently undertreated despite a lack of evidence to support withholding therapy. This review summarizes some of the robust data behind the use of anticoagulants for stroke prevention in the setting of AF; the data behind adverse events, specifically hemorrhagic events, associated with the use of warfarin and aspirin; and what is presently known about the relationship of falls with bleeding in the context of the general risks and benefits of warfarin therapy in AF. No external sources of funding were used to support this work. The authors are solely responsible for the development of the concept for this review paper, as well as the drafting and editing of the paper and its final contents.

Stroke prevention Multiple clinical trials and subsequent meta-analyses have demonstrated the benefit of aspirin compared with placebo, as well as warfarin compared with placebo, in reducing stroke risk among AF patients (Table I). In a pooled analysis of 3 randomized controlled trials (RCTs), the Atrial Fibrillation Investigators found that the relative risk reduction for aspirin versus placebo was 21% (95% CI 0%-38%, P = .05).12 In addition, another meta-analysis of 6 RCTs of aspirin versus placebo for stroke prevention found that aspirin reduced stroke by 22% (95% CI 2%38%), with absolute risk reductions of 1.5% per year for


242 Sellers and Newby

Table I. Relative risk reduction in meta-analysis of anticoagulation treatment Relative risk reduction No. of trials 3 6 6 5

ASA versus placebo

Warfarin versus placebo

ASA versus warfarin

Reference

21% (0%-38%) 22% (2%-38%) – –

– – 62% (48%-72%) –

– – – 36% (14%-52%)

AF Investigators12 Hart et al13 Hart et al13 Hart et al13

ASA, Acetylsalicylic acid.

primary prevention and 2.5% per year for secondary prevention.13 Even larger reductions have been demonstrated for warfarin relative to placebo. In 6 trials of warfarin versus placebo, adjusted-dose warfarin resulted in a risk reduction of 62% (95% CI 48%-72%). Finally, 5 RCTs showed that adjusted-dose warfarin compared with aspirin yielded a relative risk reduction of 36% (95% CI 14%-52%).13 Patient risk in these warfarin versus aspirin trials was further stratified using the now-validated CHADS2 criteria to determine which patients should be treated with aspirin and which should be treated more aggressively with vitamin K antagonists. Following from these clinical trials, the Seventh American College of Chest Physicians guidelines and the American College of Cardiology/American Heart Association/European Society of Cardiology guidelines for AF recommend either vitamin K antagonist or aspirin therapy for stroke reduction based on the CHADS2 risk stratification scheme.14,15

Hemorrhagic complications The relative benefit of warfarin compared with aspirin in preventing embolic stroke is known12,13; however, warfarin therapy is not without risk. The association of hemorrhagic complications with warfarin use is well established, and elderly patients appear to be at higher risk.9,16 In one analysis, when compared with patients b50 years of age, the unadjusted relative risk of patients ≥80 years of age having a life-threatening or fatal bleed on warfarin was 4.5 (95% CI 1.3-15.6); after adjusting for intensity of anticoagulation, the increased risk remained (relative risk [RR] 4.6, 95% CI 1.2-18.1).16 Among patients treated with warfarin for deep venous thrombosis, age N65 years was an independent risk factor for bleeding (hazard ratio [HR] 1.3, 95% CI 1.0-1.7)9; and among patients ≥85 years of age compared with patients aged 70 to 74 years, the risk of ICH is substantially increased (adjusted odds ratio [OR] 2.5, 95% CI 1.3-4.7).17 Furthermore, among elderly patients who do have a severe hemorrhagic complication (such as an intracerebral hemorrhage), warfarin use appears to be associated with significantly higher mortality. In one prospective cohort, 3-month mortality after ICH among patients

receiving warfarin at the time of an ICH was 52%, compared with 25.8% in patients not taking warfarin. Warfarin use was an independent predictor of death, with an OR of 2.2 (95% CI 1.3-3.8)—roughly a doubling in mortality in a dose-dependent manner.8 In addition to fear of hemorrhagic complications, there may be other reasons physicians choose aspirin over warfarin therapy, such as warfarin's narrow therapeutic window, the atherosclerotic benefits of aspirin therapy, patient preference, and the lack of monitoring and ease of administration associated with aspirin. But as with all interventions, physicians must weigh the risks and benefits of treatment.

Relationship of international normalized ratio with hemorrhage and stroke Warfarin monotherapy Recommendations for anticoagulation therapy in AF patients should consider the balance of stroke risk, bleeding risk, and other complications of warfarin therapy, all of which appear to be at least in part associated with the intensity of anticoagulation. Prothrombin time ratio is a strong predictor of bleeding risk in all age groups.18 In addition, international normalized ratio (INR) may be associated with worse stroke outcomes. For example, in one study of AF patients taking warfarin who presented with stroke, patients with INR N2.0 had an increased risk of a severe stroke and an increased risk of death within 30 days relative to patients with an INR b2.0 (HR 3.4, 95% CI 1.1-10.1).19 It appears that an INR of 2.0 to 3.0 provides the best balance between bleeding risk and stroke prevention benefit. Fang et al17 showed that, compared with an INR b2.0, an INR of 3.5 to 3.9 was associated with an increased risk of ICH (adjusted OR 4.6, 95% CI 2.3-9.4), but an INR of 2.0 to 3.0 was not (adjusted OR 1.3, 95% CI 0.8-2.2). In analyses from the SPORTIF III and SPORTIF V trials, rates of bleeding were higher among patients who had poor INR control compared with those with good control (goal INR 2.0-3.0)20 (Table II). In summary, there is a critical need for appropriate monitoring of INR in patients (particularly elderly patients) taking warfarin to limit hemorrhagic complica-


Sellers and Newby 243

Table II. Hemorrhagic complications in anticoagulation therapy Authors

Anticoagulant

Fihn et al16 White et al9 Fang et al17 Hylek et al19 Fang et al17 White et al20 Mant et al21

Warfarin Warfarin Warfarin Warfarin Warfarin Warfarin ASA + warfarin

Analysis Age (>79 vs 65) Age (>84 vs 70-74) INR (>2 vs 75 versus 65 >84 versus 70-74 50-98 (mean 78.3) >84 versus 70-74 70.9 (mean) >75 (mean 81.5)

RR (hemorrhage) 4.6%⁎ HR (hemorrhage) 1.3% OR (ICH) 2.5% HR (30-d mortality) 3.4% OR (ICH) 4.6% 3.85% versus 1.58% (hemorrhage) RR (hemorrhage) 0.96%

95% CI/P value 1.2%-18.1% 1.0%-1.7% 1.3%-4.7% 1.1%-10.1% 2.3%-9.4% P < .01 0.53%-1.75%

⁎ Adjusted for intensity of anticoagulation. † Poor control (INR 2-3 75% of time).

tions. Anticoagulation clinics may provide one means of decreasing the rate of hemorrhagic complications from warfarin therapy. In one study, patients who were treated in an anticoagulation clinic were 59% less likely to experience a bleeding complication than patients receiving usual care (HR 0.41, 95% CI 0.24-0.70).22

Aspirin and warfarin treatment The balance of risk and benefit with combined use of aspirin and warfarin in patients with AF is also an important question, as many patients also have an indication for aspirin therapy. In the SPORTIF trials, there was no significant reduction in stroke, systemic embolism, or myocardial infarction with the use of warfarin plus aspirin; but major bleeding occurred significantly more often with the combined use of warfarin and aspirin (3.9% per year) compared with monotherapy with warfarin (2.3% per year, P b .01).23 Aspirin monotherapy Because ICH is a major concern in elderly AF patients treated with warfarin therapy (as well as in others considered to be at high risk for ICH or other bleeding complications), one approach for physicians is to use aspirin monotherapy. As discussed previously, aspirin therapy does indeed decrease the risk of stroke in AF; however, aspirin is not as effective as warfarin.12,13 Thus, electing to use aspirin instead of warfarin assumes that aspirin therapy offers a lower risk of ICH than warfarin therapy, which balances the lower efficacy for stroke prevention. However, among patients N75 years of age in the Birmingham Atrial Fibrillation Treatment of the Aged trial, there was no difference in the rates of ICH between aspirin- and warfarin-treated groups (RR 0.96, 95% CI 0.531.75) with a goal INR of 2.0 to 3.023 (Table II). However, the SPINAF II trial24 (goal INR of 4.5) and the Japanese Nonvalvular Atrial Fibrillation–Embolism Secondary Prevention trial25 both found significantly higher rates of ICH among warfarin-treated patients than among those treated with aspirin. These mixed results do not necessarily support the decision to favor aspirin therapy over warfarin therapy when treating patients with AF who are at high risk for falls or hemorrhagic complications.

Anticoagulation and risk of falls Although increasing age is consistently associated with increased bleeding risk in warfarin therapy, an evaluation that specifically focused on fall-related hemorrhagic events showed that warfarin treatment was not associated with an increased risk of bleeding complications. In this study, the cohort treated with warfarin (379 falls patients) exhibited a hemorrhagic event rate of 6%, compared with 11% among patients (2,256 falls) not treated with warfarin (P = .01).26 However, these results were likely subject to selection bias because patients who are selected for warfarin therapy are less likely to be at risk for falls5,6 and have fewer comorbid conditions, decreasing their risk of complications. In a large retrospective study of 1,245 Medicare patients, approximately 50% of whom were prescribed warfarin, patients at high risk of falls suffered ICH more than twice as often as other subjects.10 The status of high risk for falls was based on documentation in the medical record; therefore, the definition of high risk was not standardized or defined in this retrospective study. Few studies have addressed the relationship of falls or predicted fall risk with bleeding in the setting of anticoagulation for AF. A meta-analysis of antithrombotic therapy in elderly patients at risk for falls concluded that the propensity for falling in elderly patients should not be an important factor when deciding whether or not a patient is a good candidate for anticoagulation for AF.11 In this analysis, the quality-adjusted life expectancy was greatest for warfarin, followed by aspirin, followed by no therapy. This remained true unless the annual stroke risk was b2%. Considering these numbers, an elderly patient taking warfarin would have to fall approximately 300 times per year for the risk of bleeding complications from falling to outweigh the benefits for prevention of embolic stroke. However, the authors were not able to estimate similar rates for subdural and intracerebral hemorrhages because there were too few events. This of itself suggests that the risk of ICH among elderly patients at risk for falls is low. Finally, the stroke rate may have been overestimated, and complications underestimated, in the RCTs in the meta-analysis compared with clinical practice; and patients in the trials may have been monitored more intensely than is usual in clinical practice.11



Figure 1

CHADS2 Score 0-1

CHADS2 Score 2-6

0.5

1

2

Hazard Ratio (95% CI)

Hazard ratio for out-of-hospital death, hospitalization for stroke, myocardial infarction or bleeding for warfarin treatment versus no warfarin treatment in atrial fibrillation patients according to CHADS2 risk score. Adapted from Gage et al.10

In another assessment of falls risk and anticoagulation therapy for AF in 19,506 patients, after accounting for baseline factors associated with risk of ICH, neither warfarin nor aspirin treatment was associated with risk of ICH (HR 1.0, 95% CI 0.8-1.4 for warfarin and HR 1.1, 95% CI 0.8-1.4 for aspirin).10 Importantly, in this study, the increased risk of stroke appeared to outweigh the risk of ICH. Among patients at high risk for falls, the HR for stroke was 1.3 (95% CI 1.1-1.6, P = .002) compared with patients who were not at high risk for falls. Furthermore, among patients at high risk for falls, the HR for stroke for each 1point increase in CHADS2 score was 1.42 (95% CI 1.371.47, P b .0001). The HR for the primary composite outcome of out-of-hospital death, hospitalization for stroke, myocardial infarction, and hemorrhage on warfarin compared with no warfarin was 0.98 (95% CI 0.56-1.72, P = .94) for a CHADS2 score of 0 to 1 and 0.75 (95% CI 0.610.91, P = .004) for a CHADS2 score of 2 to 6 (Figure 1). These results support the contention that patients at risk for falls but with concomitant increased stroke risk as manifested by a CHADS2 score of ≥2 would benefit overall from anticoagulation, specifically, warfarin therapy, even in the setting of an increased risk of hemorrhage.10 One underlying central limitation of the body of literature on falls risk in elderly patients is that there is no unifying definition of which patients are at risk of falls. Many of the documented trials rely on physician reporting of falls risk, reports that may be multifactorial and not necessarily based on actual risk of falling. This potentially introduces multiple confounders that may contribute to risk of ICH but are not related to the fall itself.

Future directions The ACTIVE trials explored the role of clopidogrel (an irreversible inhibitor of the platelet P2Y12 receptor) plus aspirin versus aspirin alone in warfarin-intolerant patients

(ACTIVE A) and clopidogrel plus aspirin versus warfarin (ACTIVE W) in AF patients who were able to take warfarin.27 These trials included patients with AF at enrollment or ≥2 episodes of AF in the previous 6 months, in addition to ≥1 of the following risk factors for stroke: age N74 years, hypertension, previous stroke or transient ischemic attack, non–central nervous system embolism, ejection fraction b45%, peripheral vascular disease, or age 55 to 74 years with diabetes or coronary artery disease. Warfarin was found to be superior to clopidogrel + aspirin in ACTIVE W28; but clopidogrel + aspirin reduced risk of stroke and systemic embolism in patients intolerant of warfarin compared with aspirin alone, albeit at the expense of increased major bleeding that was most prominent in patients N65 years of age.29 Thus, warfarin remains the cornerstone of treatment of most moderate- to high-risk patients with AF. However, several novel oral anticoagulants are expected to shift the balance of benefit and risk in anticoagulation for AF. In the RELY trial, the oral direct thrombin inhibitor dabigatran was compared at 2 doses with warfarin to a target INR of 2 to 3 in 18,113 patients with a mean age of 71.5 years.30 In this trial, low-dose dabigatran (110 mg twice daily) was noninferior to warfarin in preventing stroke or systemic embolism and exhibited a better bleeding profile than warfarin, with 2.7% of patients assigned to dabigatran experiencing a major hemorrhage (RR 0.80, 95% CI 0.69-0.93). Highdose dabigatran (150 mg twice daily) was superior to warfarin in preventing stroke or systemic embolism (RR 0.66, 95% CI 0.53-0.82), with similar rates of major bleeding (RR 0.93, 95% CI 0.81-1.07). Importantly, both the high- and low-dose dabigatran groups had significantly lower rates of ICH compared with the warfarin group: 27 (0.23% per year) intracranial bleeds in the lowdose group, 36 (0.30% per year) in the high-dose group, and 87 (0.74% per year) in the warfarin group (RR 0.31, 95% CI 0.20-0.47, low-dose vs warfarin; RR 0.40, 95% CI 0.27-0.60, high-dose vs warfarin). Dabigatran was recently approved by the US Food and Drug Administration for the prevention of stroke and embolism in AF and will be available in 2 doses: a 75-mg, twice-daily dose intended for patients with severe renal dysfunction and a 150-mg, twice-daily dose.31,32 In addition, at the 2010 European Society of Cardiology scientific sessions, results were presented from an RCT examining apixaban (an oral competitive factor Xa antagonist) compared with aspirin in warfarin-intolerant patients.33 The study identified a N50% reduction in thromboembolic complications in apixaban-treated patients with an acceptable bleeding risk, which resulted in early termination of the trial by the data monitoring committee. Full publication of the results is pending. The ongoing ROCKET AF trial is comparing the efficacy and safety of rivaroxaban (another oral factor Xa


inhibitor) versus warfarin in a superiority trial in patients with moderate to high CHADS2 scores34 ; and the ARISTOTLE trial is assessing the efficacy and safety of apixaban compared with warfarin across the spectrum of CHADS2 risk scores in an ongoing noninferiority trial.35 The results of both trials are expected within the year and may provide additional alternatives to warfarin anticoagulation with lower bleeding risk that may be particularly beneficial in elderly patients. Finally, pharmacogenetic guidance of warfarin dosing may improve the safety of warfarin use. Trials completed to date have been small and have not shown benefit on intermediate end points or bleeding,36 but larger RCTs are ongoing.

Conclusions The population of elderly patients with AF presents challenges with regard to the decision to provide anticoagulation treatment as well as which therapy, aspirin or warfarin, to choose. A higher likelihood of drug-drug interactions with warfarin, more adverse effects, and more comorbidities are at play in making these decisions. However, the available data suggest that physicians' decisions are guided more by their concerns over bleeding than an evaluation of the patient's risk for stroke; in many cases, their concerns regarding bleeding appear to be overemphasized in the equation. Overall, warfarin appears to be generally underused in the treatment of elderly AF patients despite fairly clear evidence that it reduces embolic and ischemic events, benefits that outweigh bleeding risk. We conclude that the risk of falling in the elderly population should not be an absolute or relative contraindication to the initiation of warfarin therapy, but that physicians should use their clinical judgment, weighing the evidence for risk and benefit with each case they are presented, including consideration of newer anticoagulants as they become clinically available.

References 1. Feinberg WM, Blackshear JL, Laupacis A, et al. Prevalence, age distribution, and gender of patients with atrial fibrillation. Analysis and implications. Arch Intern Med 1995;155:469-73. 2. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA 2001;285:2370-5. 3. Garwood CL, Corbett TL. Use of anticoagulation in elderly patients with atrial fibrillation who are at risk for falls. Ann Pharmacother 2008;42:523-32. 4. Lin HJ, Wolf PA, Kelly-Hayes M, et al. Stroke severity in atrial fibrillation. The Framingham Study. Stroke 1996;27:1760-4. 5. Dharmarajan TS, Varma S, Akkaladevi S, et al. To anticoagulate or not to anticoagulate? A common dilemma for the provider: physicians' opinion poll based on a case study of an older long-term care facility resident with dementia and atrial fibrillation. J Am Med Dir Assoc 2006;7:23-8.

Sellers and Newby 245

6. Man-Son-Hing M, Laupacis A. Anticoagulant-related bleeding in older persons with atrial fibrillation: physicians' fears often unfounded. Arch Intern Med 2003;163:1580-6. 7. He J, Whelton PK, Vu B, Klag MJ. Aspirin and risk of hemorrhagic stroke: a meta-analysis of randomized controlled trials. JAMA 1998; 280:1930-5. 8. Rosand J, Eckman MH, Knudsen KA, et al. The effect of warfarin and intensity of anticoagulation on outcome of intracerebral hemorrhage. Arch Intern Med 2004;164:880-4. 9. White RH, Beyth RJ, Zhou H, et al. Major bleeding after hospitalization for deep-venous thrombosis. Am J Med 1999;107: 414-24. 10. Gage BF, Birman-Deych E, Kerzner R, et al. Incidence of intracranial hemorrhage in patients with atrial fibrillation who are prone to fall. Am J Med 2005;118:612-7. 11. Man-Son-Hing M, Nichol G, Lau A, et al. Choosing antithrombotic therapy for elderly patients with atrial fibrillation who are at risk for falls. Arch Intern Med 1999;159:677-85. 12. The efficacy of aspirin in patients with atrial fibrillation. Analysis of pooled data from 3 randomized trials. The Atrial Fibrillation Investigators. Arch Intern Med 1997;157:1237-40. 13. Hart RG, Benavente O, McBride R, et al. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med 1999;131:492-501. 14. Fuster V, Ryden LE, Cannom DS, et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation 2006;114:e257-354. 15. Singer DE, Albers GW, Dalen JE, et al. Antithrombotic therapy in atrial fibrillation: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126:429S-56S. 16. Fihn SD, Callahan CM, Martin DC, et al. The risk for and severity of bleeding complications in elderly patients treated with warfarin. The National Consortium of Anticoagulation Clinics. Ann Intern Med 1996;124:970-9. 17. Fang MC, Chang Y, Hylek EM, et al. Advanced age, anticoagulation intensity, and risk for intracranial hemorrhage among patients taking warfarin for atrial fibrillation. Ann Intern Med 2004;141:745-52. 18. Hart RG, Tonarelli SB, Pearce LA. Avoiding central nervous system bleeding during antithrombotic therapy: recent data and ideas. Stroke 2005;36:1588-93. 19. Hylek EM, Go AS, Chang Y, et al. Effect of intensity of oral anticoagulation on stroke severity and mortality in atrial fibrillation. N Engl J Med 2003;349:1019-26. 20. White HD, Gruber M, Feyzi J, et al. Comparison of outcomes among patients randomized to warfarin therapy according to anticoagulant control: results from SPORTIF III and V. Arch Intern Med 2007;167: 239-45. 21. Mant J, Hobbs FD, Fletcher K, et al. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet 2007;370: 493-503. 22. Nichol MB, Knight TK, Dow T, et al. Quality of anticoagulation monitoring in nonvalvular atrial fibrillation patients: comparison of anticoagulation clinic versus usual care. Ann Pharmacother 2008;42: 62-70.


23. Flaker GC, Gruber M, Connolly SJ, et al. Risks and benefits of combining aspirin with anticoagulant therapy in patients with atrial fibrillation: an exploratory analysis of stroke prevention using an oral thrombin inhibitor in atrial fibrillation (SPORTIF) trials. Am Heart J 2006;152:967-73. 24. SPAF II Investigators. Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study. Lancet 1994;343:687-91. 25. Yamaguchi T. Optimal intensity of warfarin therapy for secondary prevention of stroke in patients with nonvalvular atrial fibrillation : a multicenter, prospective, randomized trial. Japanese Nonvalvular Atrial Fibrillation-Embolism Secondary Prevention Cooperative Study Group. Stroke 2000;31:817-21. 26. Bond AJ, Molnar FJ, Li M, et al. The risk of hemorrhagic complications in hospital in-patients who fall while receiving antithrombotic therapy. Thromb J 2005;3:1. 27. Connolly SJ, Yusuf S, Budaj A, et al. Rationale and design of ACTIVE: the Atrial Fibrillation Clopidogrel Trial with Irbesartan for Prevention of Vascular Events. Am Heart J 2006;151:1187-93. 28. ACTIVE Writing Group of the ACTIVE Investigators. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet 2006;367:1903-12. 29. Connolly SJ, Pogue J, Hart RG, et al. Effect of clopidogrel added to aspirin in patients with atrial fibrillation. N Engl J Med 2009;360: 2066-78.


30. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009;361: 1139-51. 31. U.S. Food and Drug Administration Web site. FDA news release. FDA approves Pradaxa to prevent stroke in people with atrial fibrillation (October 19, 2010). Available at: http://www.fda.gov/News Events/Newsroom/PressAnnouncements/ucm230241.htm (accessed October 22, 2010). 32. Wood S, O'Riordan M. FDA approves dabigatran for stroke prevention, embolism in AF patients. October 20, 2010. Available at: http://www.theheart.org/article/1138703.do (accessed October 22, 2010). 33. Eikelboom JW, O'Donnell M, Yusuf S, et al. Rationale and design of AVERROES: apixaban versus acetylsalicylic acid to prevent stroke in atrial fibrillation patients who have failed or are unsuitable for vitamin K antagonist treatment. Am Heart J 2010;159:348-53 e1. 34. Rivaroxaban-once daily, oral, direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and Embolism Trial in Atrial Fibrillation: rationale and design of the ROCKET AF study. Am Heart J 2010;159:340-347 e1. 35. Lopes RD, Alexander JH, Al-Khatib SM, et al. Apixaban for reduction in stroke and other ThromboemboLic events in atrial fibrillation (ARISTOTLE) trial: design and rationale. Am Heart J 2010;159: 331-9. 36. Kangelaris KN, Bent S, Nussbaum RL, Garcia DA, Tice JA. Genetic testing before anticoagulation? A systematic review of pharmacogenetic dosing of warfarin. J Gen Intern Med 2009;24:656-64.

Primary percutaneous coronary intervention for acute myocardial infarction: Is it worth the wait? The risk-time relationship and the need to quantify the impact of delay Giuseppe Tarantini, MD, PhD, a Frans Van de Werf, MD, PhD, b Claudio Bilato, MD, PhD, a and Bernard Gersh, MB, ChB, DPhil, FRCP c Padua, Italy; Leuven, Belgium; and Rochester, NY

The efficacy of reperfusion therapy is dependent not only by the duration of symptoms before therapy but also by the baseline risk of the individual and the circumstances (time and context) of the occurrence. All these variables play a crucial role in determining the choice of best therapy (fibrinolysis or primary angioplasty [primary percutaneous coronary intervention, PPCI]), thereby confirming the admonition that one size does not fit all. It is generally accepted that patients are best served by PPCI when times to therapy are equal between PPCI and fibrinolysis, whereas pivotal issues that are less well supported by evidence include whether a single time interval is appropriate with regard to the “acceptable” PPCI-related delay and what degree of transfer-related delay is acceptable in patients presenting “early” to a non–percutaneous coronary intervention (PCI)-capable facility. The aim of this perspective is to use available data to individualize the approach to reperfusion therapy, taking into account temporal delays and the overall mortality risk on a case-by-case basis. (Am Heart J 2011;161:247-53.)

“It is not enough that we do our best; sometimes we have to do what is required” Sir Winston Churchill In the 20 years since the GISSI-1 and ISIS-2 were published, therapy for ST-segment elevation myocardial infarction (STEMI) has continuously evolved.1,2 Several meta-analyses, including randomized trials, have demonstrated that primary percutaneous coronary intervention (PPCI) is superior to in-hospital fibrinolysis for the treatment of patients with STEMI, even among patients admitted to hospitals without interventional facilities, in which interhospital transfer is necessary.3-6 It is generally accepted that PPCI is the preferred reperfusion strategy for all, but particularly for patients with STEMI with large From the aDivision of Cardiology, Department of Cardiac, Thoracic and Vascular Sciences, University of Padua Medical School, Padua, Italy, bUniversity Hospital Gasthuisberg, Leuven, Belgium, and cDivision of Cardiovascular Diseases, Mayo Clinic College of Medicine, Rochester, NY. Submitted September 30, 2010; accepted November 7, 2010. Reprint requests: Giuseppe Tarantini, MD, PhD, Division of Cardiology, Department of Cardiac, Thoracic and Vascular Sciences, Policlinico Universitario, Via Giustiniani, 2, 35128 Padova, Italy. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.11.003

infarct and low bleeding risk, provided it is performed within 90 minutes of first medical contact (particularly in patients presenting early) by an experienced team.7,8 However, the availability of catheterization laboratories among countries and within countries strongly affects the ability to deliver mechanical reperfusion therapy.9 Thus, according to the guidelines by the American College of Cardiology/American Heart Association7 and the European Society of Cardiology,8 patients with STEMI who cannot undergo PPCI in a timely manner, such as those presenting to a hospital without PCI capability and cannot be transferred to a PCI center, should be treated with fibrinolytic therapy within 30 minutes of hospital presentation as a system goal unless fibrinolytic therapy is contraindicated (class I, level of evidence B). Notwithstanding, controversies still exist in the care of patients with STEMI, especially in the everyday clinical practice. Should a 50-year-old male diabetic patient with a 3-hour anterior STEMI who is hemodynamically stable, is admitted to a hospital without PCI capability, and is 70 minutes away from the tertiary center with PPCI facilities be treated in the same manner as a 74-yearold male patient with a 3-hour anterior STEMI who is hemodynamically unstable (ie, heart rate N100 beat/min) and is admitted to the same hospital? These 2 cases raise a number of important questions and unresolved issues. What is the maximum acceptable PPCI-related


248 Tarantini et al

Figure 1

Table I. Risk scores

TIMI score for STEMI (0-14)

TIMI risk index CADILLAC (0-18)

Independent clinical predictors of 30-day mortality.

delay compared with fibrinolysis that can justify immediate transfer for PPCI as opposed to initial treatment with a fibrinolytic agent? Is interhospital transfer necessary for both patients? Does a specific PPCI-related delay fit both cases? Is there any role for a pharmacoinvasive strategy? The aim of this perspective is to use available data to individualize the approach to reperfusion therapy, taking into account temporal delays and the overall mortality risk.

Risk-dependent benefit of the reperfusion therapy: key modulator of the reperfusion choice? Choices among alternative therapies or decisions regarding the allocation of clinical resources are based on a baseline assessment of the patient's risk. It is well documented,10 and N80% of the prognostic information is based on 3 characteristics, namely, age, hemodynamic status (eg, heart rate, blood pressure, and Killip class), and the location of the infarction (Figure 1), whereas other variables reflecting the circumstances under which the infarction was treated (eg, time to treatment and type of reperfusion therapy) are of lesser but nonetheless modifiable importance.10,11 One of the first validated and clinically useful risk scores for STEMI was the Thrombolysis In Myocardial Infarction (TIMI) score, which was derived from fibrinolytic therapy trials.11 The TIMI score incorporates clinical and electrocardiographic (ECG) characteristics (Table I) and has a robust prognostic performance

PAMI (0-15)

GRACE (0-372)

Risk factor

Score

Age, 65-74/≥75 y Systolic blood pressure, b100 mm Hg Heart rate, N100 beat/min Killip classification II-IV Anterior STEMI or left branch bundle block Diabetes mellitus, hypertension, or angina pectoris Weight, b67 kg Time to treatment, N4 h Heart rate × (age/10)2/systolic blood pressure Baseline left ventricle ejection fraction, b40% Renal insufficiency Killip classification II-IV Final TIMI flow 0-2 Age, N65 y Anemia⁎

2/3 3 2 2 1 1

3-Vessel disease Age, N75 y Age, 65-75 y Killip classification N1 Heart rate, N100 beat/min Diabetes mellitus Anterior STEMI or left branch bundle block Age (y) b30 30-39 40-49 50-59 60-69 70-79 80-89 ≥90 Heart rate (beat/min) b50 50-69 70-89 90-109 110-149 150-199 N200 Systolic blood pressure (mm Hg) b80 80-99 100-119 120-139 140-159 160-199 N200 Creatinine (mg/dL) 0-0.39 0.4-0.79 0.8-1.19 1.2-1.59 1.6-1.99 2-3.99 N4 Killip classification I II III IV

1 1

4 3 3 2 2 2 2 7 3 2 2 2 2 0 8 25 41 58 75 91 100 0 3 9 15 24 38 46 58 53 43 34 24 10 0 1 4 7 10 13 21 28 0 20 39 59


Tarantini et al 249

Table I (continued ) Risk factor Cardiac arrest at admission Increased cardiac markers ST-segment deviation

Score 39 14 28

CADILLAC, Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications; PAMI, Primary Angioplasty in Myocardial Infarction; GRACE, Global Registry for Acute Coronary Events. ⁎ Anemia is defined as baseline hematocrit b39% for men and 36% for women.

across the heterogeneous spectrum of patients with STEMI.12 With the increasing adoption of PPCI as the preferred reperfusion strategy, risk scores based on PPCI trials were subsequently developed (Table I) and validated, although with a notable difference in predictive accuracy.13 The Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications risk score is based on clinical and angiographic parameters, whereas the Primary Angioplasty in Myocardial Infarction score relies on clinical and ECG characteristics (Table I). In addition, the Global Registry for Acute Coronary Events score, based on a large registry of patients across the entire spectrum of acute coronary syndromes, incorporates clinical and ECG characteristics. This risk score was validated and showed to be predictive for all acute coronary syndromes but had lower predictive accuracy for early mortality as compared with other scores.13 None of the models, however, have been tested prospectively by randomizing patients to a reperfusion strategy based on the estimated mortality at presentation. The quantitative analysis by Keeley et al3 that compared PPCI with fibrinolysis showed an absolute decrease of approximately 2% of risk of death by PPCI over in-hospital thrombolysis. However, mortality benefit is related to the risk stratification at admission, being more evident in high-risk patients (low proportion of patients), as demonstrated by a subgroup analysis of the DANAMI-2 trial.14 Indeed, individual trials and risk-benefit metaregression analyses confirm that the absolute difference in mortality at 30 days between PPCI and fibrinolysis increases in favor of PPCI as the estimated risk of mortality with fibrinolysis grows.14-16 Conversely, if the estimated mortality benefit with fibrinolysis declines, the absolute mortality benefit of PPCI decreases, although definitive evidence on the exact equipoise is still lacking. On the other hand, when the estimated mortality with fibrinolysis is extremely high (eg, patients with cardiogenic shock), compelling evidence exists favoring a PPCI strategy, as demonstrated by the SHOCK trial17 and the National Registry of Myocardial Infarction II.18 If PPCI is unavailable, the potential benefits of fibrinolytic therapy need to take into consideration the risk of a lifethreatening bleed.19

Time-dependent benefit of reperfusion therapy: overemphasized or overlooked? Time from the symptom onset The early open-artery theory suggests that benefits of reperfusion in patients with STEMI are directly related to the speed and completeness of the restoration of patency of the infarct-related coronary artery. Several clinical studies strongly confirm the relationship between achieving prompt antegrade coronary flow and improvement of myocardial salvage and clinical outcomes, both for PPCI and fibrinolysis.20,21 As shown in experimental and clinical models, the extent of transmurality and the presence of microvascular injury are strongly dependent on the duration of ischemia before reperfusion, with a close relationship among myocardial and microvascular injury, myocardial viability, left ventricular function, and clinical outcomes.22-25 A clear relationship between the extent of delay and mortality has been recently described26: for each additional 30 minutes of treatment delay, there is a 7.5% increase in mortality. Similarly, it has been recently reported that the delivery of reperfusion therapy for patients with STEMI outside guidelinerecommended delays is associated with adverse outcomes.27 According to the hypothetical construct of the association between mortality and treatment delay described by Gersh et al,28 this demonstrates a striking benefit within the first 2 to 3 hours, emphasizing the narrow “golden window of opportunity,” as reported also by Boersma et al.21 At a later stage, on the “flat part of the curve,” there is a continued but decreasing magnitude of mortality benefit of PPCI over time. The pathophysiologic factors that potentially could modulate the relationship between time and clinical outcomes are the presence of functioning coronary collaterals, ischemic preconditioning, myocardial oxygen demand, myocardial territory at risk, endothelial and microvascular function, and spontaneous coronary reperfusion. In clinical practice, symptom-onset-to-needle time or symptom-onset-to-balloon time is often and inappropriately considered as surrogate of the total ischemic time. However, it should be emphasized that the precise time from the symptom onset may be troublesome to determine by patient's history (presentation delay) because of inaccurate recollection, nonspecific symptoms, silent ischemia, intermittent spontaneous reperfusion, collateral circulation, or prodromal angina. Moreover, the time from the symptom onset cannot be a surrogate of the true ischemic time if the infarctrelated artery is already open at the time of PPCI or at the onset of fibrinolytic therapy.22 Thus, although time from the symptom onset has become central in the diagnosis, triage, and management of patients with STEMI, its prognostic role may be overemphasized because of the substantial differences between the

250 Tarantini et al

pathophysiology of STEMI in a patient and experimental models.29 Accordingly, several efforts have been made to evaluate characteristics of the ECG at the time of presentation to provide additional insight into the stages of the STEMI process and to discriminate between “pseudo early comers” (patients with perceived short time from the symptom onset but with new Q waves on the admission ECG) and “pseudo latecomers” (patients with perceived long time from the symptom onset but without Q waves). Indeed, the presence of Q waves on the admission ECG provides incremental value to the presentation delay in predicting 30-day mortality after either fibrinolysis or PPCI30 and might help to evaluate patients with STEMI for triage and potential transfer to tertiary centers for planned PCI especially if the duration of the symptoms is unclear.

Time from the symptom onset–risk interaction A significant interaction between ischemic time and baseline risk in patients with STEMI treated with PPCI has been reported with an adverse impact on 1-year mortality for any delay in reperfusion, particularly if patients are at high risk and/or preprocedural TIMI 2-3 flow is absent.31 Time-to-treatment analyses, however, may have been confused by other variables. First, it has been recently observed that early presenters have the highest risk score and the largest magnitude of cumulative ST elevation, reflecting a larger area at risk.29,32 Late presenters, on the other hand, may be considered, at least in part, survivors (eg, they did not die out of hospital); therefore, they may be at lower risk compared with early presenters, but this may be confounded by comorbities, for example, renal failure, diabetes, advanced age, etc. Second, with regard to the differential benefit of PPCI over fibrinolysis as a function of symptom duration, the current American College of Cardiology/American Heart Association guidelines state that fibrinolysis is an appropriate alternative in patients presenting with symptom onset duration ≤3 hours depending on the extent of transfer delay. This statement is based on a subgroup analysis of the CAPTIM trial,33 which demonstrated a higher mortality rate in PPCI-treated than in fibrinolysis-treated patients presenting within 2 hours from the symptom onset. On the other hand, Boersma et al,4 using individual data from randomized trials (not including the CAPTIM trial), provided evidence that lower mortality is seen in PPCItreated than in fibrinolysis-treated patients for each specific time-to-treatment period. Therefore, the key question, for which data are unavailable, is whether fibrinolysis at 60 to 70 minutes is better than PPCI at 120 to 140 minutes and in which patients. Similarly, the recent article by Lambert et al27 emphasized clearly the importance of reperfusion within the guideline-recom-


mended delays, but it is not conclusive to answer this question.

Primary percutaneous coronary intervention–related delay The mortality benefit of PPCI over fibrinolysis is also dependent on the additional PPCI-related delay. Current guidelines suggested an acceptable PPCI-related delay of b60/120 minutes based on the currently published trials.7,8 However, other authors found different times to equipoise according to different modeling of the data,34 for example, stratifying the DANAMI-2 study patients according to whether they were initially admitted to an interventional or noninterventional center or the inclusion of trials using non–fibrin-specific fibrinolytic therapy. To this regard, in a meta-analysis, Nallamothu et al35 showed that the survival advantage of PPCI was lost for PPCI delay N1 hour only when fibrin-specific agents were considered. Irrespective of whatever categorization, precise cutoff value for PPCI-related delay is, to some extent, oversimplification from a clinical standpoint. Primary percutaneous coronary intervention–related delay-risk interaction By collecting patient-specific data for most randomized trials and using a PPCI-related delay defined at hospital level, Boersma et al4 showed that PPCI remained superior to fibrinolysis when PPCI-related delay was either b35 minutes or N79 minutes. Because the mortality rates in patients with longer delays were higher, it is likely that an interaction between acceptable PPCI-related delay and severity of baseline mortality risk was present. Other large registry studies have also demonstrated that longer PPCI-related delay does not negate the benefits of PPCI,36 but all these data are subject to the confounding factors and selection bias inherent of registry studies. Recently, Pinto et al,37 using data from the National Registry of Myocardial Infarction registry, reported a time to equivalent benefit in mortality between PPCI and fibrinolysis at 114 minutes. The authors assessed also the impact of PCI-related reperfusion delay across cohorts of patients with various risk factors (anterior vs nonanterior infarction and age b65 or ≥65 years) and further classified them into those presenting before versus ≥2 hours after symptom onset. This analysis demonstrated that in young patients with STEMI, either anterior or nonanterior, within 2 hours after symptom onset with low risk of bleeding, the PPCI-related reperfusion delay leading to equivalent mortality between PPCI and fibrinolysis was b60 minutes. In older patients, this balance swayed toward a longer acceptable delay. This, although nonrandomized, registry further supports the risk in modulating the “acceptable” PCI-related reperfusion delay. The Vienna registry documented a comparable mortality in PPCI- and fibrinolysis-treated patients


Tarantini et al 251

Figure 2

The estimated PPCI delay to equipoise for patient 2 is 200 minutes, according to its time from symptom onset (180 minutes) and risk (TIMI risk score 5 [12.4% mortality rate]) (see Tarantini et al16 for details). Within this time interval, however, loosing time still means loosing benefit. For example, 90 minutes of extra delay from “A” to “B” determines a significant reduction of the PPCI benefit as shown by the increase of the number needed to treat. Finally$2, an expected system delay longer than acceptable should imply the choice of fibrinolysis (shift from “B” to “D”) to avoid harming the patient. NNH, Number needed to harm; NNT, number needed to treat. Modified from Tarantini et al16 with permission from Elsevier.

even for PPCI-related delay of 138 minutes, with a significant advantage in cardiogenic shock or if age was ≥75 years.38 Finally, a recent meta-regression analysis confirmed that the variability in baseline mortality risk significantly modified the acceptable time delay in choosing the appropriate reperfusion strategy and that the time delay that nullifies the survival benefit of PPCI over fibrinolysis can be calculated from the baseline mortality risk and the length of presentation delay.16 Despite the fact that increasing treatment delay, particularly the modifiable health system–related delay, results in higher mortality in high-risk patients,27,29,31,39,40 PPCI may still be the preferred strategy in high-risk subgroups presenting to hospitals without interventional facilities unless the patient presents extremely early (eg, b60 minutes).16,37,40 In this respect, a recent comprehensive registry from the province of Quebec, Canada, identified that being transferred for PPCI was a major predictor of lower 30-day mortality.27 Based on the described multifaceted risk-time relationship, it comes out that there is a clinical need of a customized quantification of the impact of treatment delay (“waiting score”) in order to choose the most appropriate reperfusion therapy on an individual basis. While waiting for more robust data on the field, how do we apply the bulk of the available evidences to the 2 patients described in the introduction? The preferred approach to the first patient would be the prompt initiation of fibrinolytic therapy, whereas in the

case of the second patient, it would be the immediate transfer for PPCI, based on the quantification of the impact of delay, that is, b1 hour and N2 hours, respectively. Figure 2 shows schematically how the survival advantage of PPCI over fibrinolysis may vary as function of the related PPCI delay in our case vignette 2. Primary percutaneous coronary intervention remains the treatment of choice even when longer delays are unavoidable, although the concept that, particularly in high-risk patients, the longer the delay the lower the absolute survival advantage of PPCI, as shown by others,31,40 remains true. In summary, although accepting that the preferred approach for most patients is PPCI, within the confines of guideline-recommended PCI-related delays, the data suggest that, in certain circumstances based on baseline levels of mortality risk, longer delays to choose PPCI instead of fibrinolysis might be acceptable.4,37 It should be noteworthy, however, that there are major shortcomings in the available data to take into account. First, the prognostic impact of the presentation delay may be underestimated because few patients are admitted within the “golden time window of opportunity.”28 Thus, a shorter time to equipoise may be expected in patients presenting very early (when the thrombus is more vulnerable to fibrinolysis) and to a drug-invasive strategy. Second, other confounders may play a role. For example, (1) longer presentation delays are associated with longer door-to-balloon or door-to-needle times,39,41 and (2) high-

252 Tarantini et al

quality and large-volume PCI centers may be associated with shorter delays because of more efficient systems of care.41,42 Third, the calculation of a definite waiting score is difficult and needs to be validated prospectively in the context of large clinical trials. Finally, patients treated with the pharmacoinvasive approach were not included in the reported analyses. The pharmacoinvasive strategy implies primary reperfusion therapy by thrombolysis with routine angiography 3 to 24 hours after apparently successful thrombolysis or as a “rescue” procedure. There is mounting evidence that a routine but nonemergent invasive strategy after fibrinolytic therapy is not only safe and effective but also the preferred approach,43 as recognized by European and US guidelines.7,8 However, few of the currently published studies compare a pharmacoinvasive strategy (with fibrinolysis) to transfer for “plain” PPCI. Therefore, if the maximum acceptable PPCI-related delay depends on the baseline risk of the patient, we need data comparing PPCI with pharmacoinvasive strategy, tailored on the baseline mortality risk. This is currently being studied prospectively in the STREAM trial in which high-risk patients presenting early (b3 hours) to an ambulance crew to a regional hospital with PPCI facilities are randomized to either transfer for PPCI or a pharmacoinvasive approach with tenecteplase, clopidogrel, and enoxaparin.44

Conclusions In patients who present to hospitals with requisite facilities and documented expertise, PPCI is the preferred strategy. Fibrinolysis without delay may provide maximal advantage in younger patients at low risk of hemorrhage and mortality presenting earlier (only if the delay to PPCI is beyond the acceptable time frame for that patient), whereas high-risk patients, except those who present very early without Q waves on the admission ECG, might benefit from PPCI, even when longer delays are unavoidable. Developing systems geared to rapid transfer may be more cost-effective overall than the development of a profusion of low-volume, PPCI-capable facilities.45

References 1. GISSI trial: early results and late follow-up. Gruppo Italiano per la Sperimentazione della streptokinasi nell'Infarto Miocardico. J Am Coll Cardiol 1987;10(5 Suppl B):33B-9B. 2. ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancet 1988;2:349-60. 3. Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction a quantitative review of 23 randomized trials. Lancet 2003;361: 13-20.


4. Boersma E and PCAT-2 Trialists Collaborative Group. Does time matter? A pooled analysis of randomized clinical trials comparing primary percutaneous coronary intervention and in-hospital fibrinolysis in acute myocardial infarction patients. Eur Heart J 2006;27: 779-88. 5. Dalby M, Bouzamondo A, Lechat P, et al. Transfer for primary angioplasty versus immediate thrombolysis in acute myocardial infarction: a meta-analysis. Circulation 2003;108:1809-14. 6. Andersen HR, Nielsen TT, Rasmussen K, et al, for the DANAMI-2 Investigators. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med 2003;349: 733-42. 7. Kushner FG, Hand M, Smith Jr SC, et al. 2009 Focused updates: ACC/AHA guidelines for the management of patients with STelevation myocardial infarction. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation 2009;120:2271-306. 8. Van de Werf F, Bax J, Betriu A, et al. Management of acute myocardial infarction in patients presenting with persistent STsegment elevation: the Task Force on the Management of ST-Segment Elevation Acute Myocardial Infarction of the European Society of Cardiology. Eur Heart J 2008;29:2909-45. 9. Nallamothu BK, Krumholz HM, Ko DT, et al. Development of systems of care for ST-elevation myocardial infarction patients: gaps, barriers, and implications. Circulation 2007;116:e68-e72. 10. Lee KL, Woodlief LH, Topol EJ, et al. Predictors of 30-day mortality in the era of reperfusion for acute myocardial infarction: results from an international trial of 41 021 patients. Circulation 1995;91:1659-68. 11. Morrow DA, Antman EM, Charlesworth A, et al. TIMI risk score for ST-elevation myocardial infarction: a convenient, bedside, clinical score for risk assessment at presentation: an Intravenous nPA for Treatment of Infarcting Myocardium Early II trial substudy. Circulation 2000;102:2031-7. 12. Morrow DA, Antman EM, Parsons L, et al. Application of the TIMI risk score for ST-elevation MI in the National Registry of Myocardial Infarction 3. JAMA 2001;286:1356-9. 13. Lev EI, Kornowski R, Vaknin-Assa H, et al. Comparison of the predictive value of four different risk scores for outcomes of patients with ST-elevation acute myocardial infarction undergoing primary percutaneous coronary intervention. Am J Cardiol 2008; 102:6-11. 14. Thune JJ, Hoefsten DE, Lindholm MG, et al, for the Danish Multicenter Randomized Study on Fibrinolytic Therapy Versus Acute Coronary Angioplasty in Acute Myocardial Infarction (DANAMI)-2 Investigators. Simple risk stratification at the admission to identify patients with reduced mortality from primary angioplasty. Circulation 2005;112: 2017-21. 15. Tarantini G, Razzolini R, Ramondo A, et al. Explanation for the survival benefit of primary angioplasty over thrombolytic therapy in patients with ST-elevation acute myocardial infarction. Am J Cardiol 2005;96:1503-5. 16. Tarantini G, Razzolini R, Napodano M, et al. Acceptable reperfusion delay to prefer primary angioplasty over fibrin-specific thrombolytic therapy is affected (mainly) by the patient's mortality risk: 1 h does not fit all. Eur Heart J 2010;31:676-83. 17. Hochman JS, Sleeper LA, White HD, et al, for the Should We Emergently Revascularization Occluded Coronaries for Cardiogenic Shock (SHOCK) Investigators. One-year survival following early revascularization for cardiogenic shock. JAMA 2001;285:190-2. 18. Wu AH, Parsons L, Every NR, et al, for the Second National Registry of Myocardial Infarction. Hospital outcomes in patients presenting with congestive heart failure complicating acute


19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

myocardial infarction: a report from the Second National Registry of Myocardial Infarction (NRMI-2). J Am Coll Cardiol 2002;40: 1389-94. Krumholz HM, Pasternak RC, Weinstein MC, et al. Cost effectiveness of thrombolytic therapy with streptokinase in elderly patients with suspected acute myocardial infarction. N Engl J Med 1992;327: 7-13. Cannon CP, Gibson CM, Lambrew CT, et al. Relationship of symptom-onset-to-balloon time and door-to-balloon time with mortality in patients undergoing angioplasty for acute myocardial infarction. JAMA 2000;283:2941-7. Boersma E, Maas AC, Deckers JW, et al. Early thrombolytic treatment in acute myocardial infarction: reappraisal of the golden hour. Lancet 1996;348:771-5. Tarantini G, Cacciavillani L, Corbetti F, et al. Duration of ischemia is a major determinant of transmurality and severe microvascular obstruction after primary angioplasty: a study performed with contrastenhanced magnetic resonance. J Am Coll Cardiol 2005;46:1229-35. Tarantini G, Razzolini R, Cacciavillani L, et al. Influence of transmurality, infarct size, and severe microvascular obstruction on left ventricular remodeling and function after primary coronary angioplasty. Am J Cardiol 2006;98:1033-40. Basso C, Corbetti F, Silva C, et al. Morphologic validation of reperfused hemorrhagic myocardial infarction by cardiovascular magnetic resonance. Am J Cardiol 2007;100:1322-7. Francone M, Bucciarelli-Ducci C, Carbone I, et al. Impact of primary coronary angioplasty delay on myocardial salvage, infarct size, and microvascular damage in patients with ST-segment elevation myocardial infarction: insight from cardiovascular magnetic resonance. J Am Coll Cardiol 2009;54:2145-53. De Luca G, Suryapranata H, Ottervanger JP, et al. Time delay to treatment and mortality in primary angioplasty for acute myocardial infarction. Every Minute of Delay Counts Circulation 2004;109: 1223-5. Lambert L, Brown K, Segal E, et al. Association between timeliness of reperfusion therapy and clinical outcomes in ST-elevation myocardial infarction. JAMA 2010;303:2148-55. Gersh BJ, Stone GW, White HD, et al. Pharmacological facilitation or primary percutaneous coronary intervention for acute myocardial infarction. Is the slope of the curve the shape of the future? JAMA 2005;293:979-86. Terkelsen CJ, Sørensen JT, Maeng M, et al. System delay and mortality among patients with STEMI treated with primary percutaneous coronary intervention. JAMA 2010;304:763-71. Armstrong PW, Fu Y, Westerhout CM, et al. Baseline Q-wave surpasses time from symptom onset as a prognostic marker in STsegment elevation myocardial infarction patients treated with primary percutaneous coronary intervention. J Am Coll Cardiol 2009;53: 1503-9. De Luca G, Suryapranata H, Zijlstra F, et al. Symptom onset to balloon time and mortality in patients with acute myocardial infarction treated by primary angioplasty. J Am Coll Cardiol 2003; 42:991-7. Zijlstra F, Patel A, Jones M, et al. Clinical characteristics and outing come of patients with early (less than 2 h), intermediate (2-4 h) and

Tarantini et al 253

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

late (greater than 4 h) presentation treated by primary coronary angioplasty or thrombolytic therapy for acute myocardial infarction. Eur Heart J 2002;23:550-7. Bonnefoy E, Lapostolle F, Leizorovicz A, et al. Primary angioplasty versus prehospital fibrinolysis in acute myocardial infarction: a randomised study. Lancet 2002;360:825-9. Betriu A, Masotti M. Comparison of mortality rates in acute myocardial infarction treated by percutaneous coronary intervention versus fibrinolysis. Am J Cardiol 2005;95:100-1. Nallamothu BK, Antman EM, Bates ER. Primary percutaneous coronary intervention versus fibrinolytic in acute myocardial infarction does the choice of fibrinolytic agent impact on the importance of time to treatment? Am J Cardiol 2004;94:772-4. Magid DJ, Wang Y, Herrin Y, et al. Relationship between time of day, day of week, timeliness of reperfusion, and in-hospital mortality for patients with acute ST-segment elevation myocardial infarction. JAMA 2005;294:803-12. Pinto DS, Kirtane AJ, Nallamothu BK, et al. Hospital delays in reperfusion for ST-elevation myocardial infarction. Implications when selecting a reperfusion strategy. Circulation 2006;114:2019-25. Kalla K, Christ G, Karnik R, et al, for the Vienna STEMI Registry Group. Implementation of guidelines improves the standard of care: the Viennese registry on reperfusion strategies in ST-elevation myocardial infarction (Vienna STEMI Registry). Circulation 2006; 113:2398-405. Brodie BR, Hansen C, Stuckey TD, et al. 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;47:289-95. Brodie BR, Gersh BJ, Stuckey T, et al. When is door-to-balloon time critical? Analysis from the HORIZONS-AMI (Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction) and CADILLAC (Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications) trials. J Am Coll Cardiol 2010;56:407-13. Ting HH, Bradley EH, Wang Y, et al. Delay in presentation and reperfusion therapy in ST-elevation myocardial infarction. Am J Med 2008;121:316-23. Magid DJ, Calonge BN, Rumsfeld JS, et al, for the National Registry of Myocardial Infarction 2 and 3 Investigators. Relation between hospital primary angioplasty volume and mortality for patients with acute MI treated with primary angioplasty vs thrombolytic therapy. JAMA 2000;284:3131-8. Wijeysundera HC, You JJ, Nallamothu BK, et al. An early invasive strategy versus ischemia-guided management after fibrinolytic therapy for ST-segment elevation myocardial infarction: a metaanalysis of contemporary randomized controlled trials. Am Heart J 2008;156:564-72. Armstrong PW, Gershlick A, Goldstein P, et al. The Strategic Reperfusion Early After Myocardial Infarction (STREAM) study. Am Heart J 2010;160:30-5. Concannon T, Kent D, Normand SL, et al. Comparative effectiveness of STEMI regionalization strategies. Circ Cardiovasc Qual Outcomes 2010;3:506-13.

Trial Design

Design and rationale of the RadIal Vs. femorAL access for coronary intervention (RIVAL) trial: A randomized comparison of radial versus femoral access for coronary angiography or intervention in patients with acute coronary syndromes Sanjit S. Jolly, MD, MSc, a,l Kari Niemelä, MD, PhD, b,l Denis Xavier, MD, c,l Petr Widimsky, MD, d,l Andrzej Budaj, MD, PhD, e,l Vicent Valentin, MD, f,l Basil S. Lewis, MD, g,l Alvaro Avezum, MD, PhD, h,l Philippe Gabriel Steg, MD, i,l Sunil V. Rao, MD, j,l John Cairns, MD, k,l Susan Chrolavicius, BScN, a,l Salim Yusuf, MBBS, D.Phil, a,l and Shamir R. Mehta, MD, MSc a,l Ontario and Vancouver, Canada; Tampere, Finland; Bangalore, India; Prague, Czech Republic; Warsaw, Poland; Valencia, Spain; Haifa, Israel; Sao Paulo, Brazil; Paris, France; and Durham, NC

Background Major bleeding in acute coronary syndromes (ACS) is associated with an increased risk of subsequent mortality and recurrent ischemic events. Observational data and small randomized trials suggest that radial instead of femoral access for coronary angiography/intervention results in fewer bleeding complications, with preserved and possibly improved efficacy. Radial access versus femoral access has yet to be formally evaluated in a randomized trial adequately powered for the comparison of clinically important outcomes. Objectives The aim of this study is to evaluate the efficacy and safety of radial versus femoral access for coronary angiography/intervention in patients with ACS managed with an invasive strategy. Design This was a multicenter international randomized trial with blinded assessment of outcomes. 7021 patients with ACS (with or without ST elevation) have been randomized to either radial or femoral access for coronary angiography/ intervention. The primary outcome is the composite of death, myocardial infarction, stroke, or non–coronary artery bypass graft-related major bleeding up to day 30. The key secondary outcomes are (1) death, myocardial infarction, or stroke up to day 30 and (2) non–coronary artery bypass graft-related major bleeding up to day 30. Percutaneous coronary intervention (PCI) success rates will also be compared between the two access sites. Conclusions The RIVAL trial will help define the optimal access site for coronary angiography/intervention in patients with ACS. (Am Heart J 2011;161:254-260.e4.)

From the aMcMaster University and the Population Health Research Institute, Hamilton Health Sciences, Hamilton, Ontario, Canada, bTampere University Hospital, Tampere, Finland, cSt John's Medical College and Research Institute, Bangalore, India, dCharles University, Hospital Kralovske Vinohrady, Prague, Czech Republic, ePostgraduate Medical School, Department of Cardiology, Grochowski Hospital, Warsaw, Poland, fHospital Universitari Dr Peset, Valencia, Spain, gLady Davis Carmel Medical Center, Haifa, Israel, h Dante Pazzanese Institute of Cardiology, Sao Paulo, Brazil, iINSERM U-698. “Recherche Clinique en Athérothrombose.” Université Paris 7 and Assistance Publique—Hôpitaux de Paris, Paris, France, jDuke Clinical Research Institute, Duke University, Durham, North Carolina, and kUniversity of British Columbia, Vancouver, Canada. l On behalf of the RIVAL steering committee. See online Appendix B for complete listing. Reg. number NCT01014273. Marc Cohen, MD, served as guest editor for this article. Submitted October 20, 2010; accepted November 23, 2010. Reprint requests: Sanjit S. Jolly, MD, MSc, Rm C3-118, DBCVSRI Building, Hamilton General Hospital, 237 Barton St. East, Hamilton, Ontario, Canada L8L 2X2. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.11.021

Among patients with acute coronary syndromes (ACS) (ie, ST elevation myocardial infarction [MI] [STEMI], non– ST-segment elevation MI [NSTEMI], or unstable angina), 2% to 5% experience major bleeding,1-3 a substantial proportion of which originates from the vascular access site.4 In multiple observational studies (in both ACS and percutaneous coronary intervention [PCI]), major bleeding has been independently associated with a 2- to 10-fold increased risk of death.4-7 Vascular access for coronary angiography/intervention via the radial artery, a superficial and readily compressible site, could result in a lower risk of bleeding than that associated with access via the femoral artery site. Observational studies have suggested that radial access may be independently associated with a 50% to 60% reduction in the odds of major bleeding compared to


femoral access.8,9 A meta-analysis of 18 small randomized trials (n = 4,458 patients, 61 major bleeding events) showed that radial access was associated with a 73% reduction in major bleeding compared to femoral access (0.5% vs 2.3%, respectively, odds ratio [OR] 0.27, 95% CI 0.16-0.45, P b .001).10 Currently, radial access accounts for only 6% to 12% of procedures worldwide.9,11-14 Due to technical challenges related to radial artery diameter, subclavian tortuosity, and reduced guide catheter support, there are perceptions that the radial approach may be associated with a greater PCI procedural failure rate. In addition, the inability to use large bore hemodynamic support devices such as intra-aortic balloon pumps via the radial artery may lead to concerns among operators that they may not be able to respond adequately to emergencies that arise during the procedure. On the other hand, femoral access has a long history of use, allows for larger diameter catheters for complex procedures, and, during emergency situations, has the advantage of ease of access to the femoral vein for transvenous pacing and of the availability of the femoral artery route for the insertion of an intraaortic balloon pump. Femoral access allows superior guide support and therefore may be associated with greater rates of PCI success. Although femoral artery vascular closure devices allow earlier sheath removal, they have not been shown to reduce major bleeding in randomized trials.15 There is an emerging hypothesis that major bleeding may cause recurrent ischemic events because of (1) activation of the coagulation cascade, (2) adverse effects of blood transfusion, (3) cessation of antithrombotic and antiplatelet therapies or reversal of their effects, and (4) decreased ischemic threshold due to anemia or hypovolemia. Observational studies have demonstrated an association between major bleeding and subsequent ischemic events, but causality remains uncertain.4-6,16 Several randomized trials (OASIS 5, HORIZONS) have demonstrated that a reduction in early bleeding events using safer antithrombotic therapies is strongly associated with a reduction in longer-term mortality, MI, and stroke; this relationship is likely to be causal.2,17 Whether procedural methods to reduce bleeding such as radial access could reduce ischemic events and death is an unanswered question; such a relationship would help support the causal relationship between major bleeding and mortality. Although observational studies have suggested that radial access is associated with a lower risk of mortality than femoral access, potential selection bias necessitates a cautious interpretation of the findings.8,11 The meta-analysis of small randomized trials performed by our group found that radial access was associated with a trend toward less death, MI, or stroke compared to femoral access (OR 0.71, 95% CI 0.49-1.01, P = .058).10

Jolly et al 255

The results from observational studies and meta-analyses of small randomized trials require confirmation in a large adequately powered randomized trial. Accordingly, we designed RIVAL, a multicenter randomized trial to compare the 2 procedures among 7,000 patients with ACS.

The RIVAL study Primary objective Among patients with ACS (with or without ST-segment elevation), to determine whether radial access is superior to femoral access for the composite of death, MI, stroke or non–coronary artery bypass graft (CABG)-related major bleeding up to 30 days.

Design RIVAL is a multinational, multicenter, randomized, parallel group study comparing radial versus femoral access for coronary angiography/intervention among patients with non–ST-segment elevation ACS (NSTEACS) or STEMI (Figure 1). The trial began as an investigator initiated randomized substudy of the CURRENT-OASIS 718 trial, which compared 2 regimens of clopidogrel (double dose vs standard dose) and 2 regimens of aspirin (high vs low dose) among patients with NSTE-ACS or STEMI.19,20 The main CURRENT-OASIS 7 study was completed in July 2009,19 and the RIVAL study has continued as a stand-alone study.

Eligibility criteria Patients with NSTE-ACS or STEMI who are to be managed with an invasive approach are eligible i) if the interventional cardiologist is willing to proceed with either a radial or femoral approach; ii) if an operator is available, with the requisite expertise for both the radial (≥50 such procedures in the past year) and the femoral approach; and iii) if the patient has a normal Allen test (ie, confirmation of collateral flow to the hand) (Table I).

Expertise We mandated that each operator in the RIVAL study had performed ≥50 radial procedures within the previous year. An alternative approach would have been to use “expertise-based” randomization. The advantage of this approach is that the procedure is performed by experts who are completely familiar with the technique. However, there are also disadvantages to expertise-based randomization. First, linking the randomly allocated treatment to specific operators may introduce confounding, and this could threaten the very essence of randomization (which endeavors to balance known and unknown variables). For example, radial experts may have intrinsically better catheterization skills and use lower heparin doses, which may lead to a


256 Jolly et al

Figure 1

RIVAL study design: a randomized trial of radial versus femoral access in patient with NSTE-ACS and STEMI.

difference in outcomes not because of the randomized treatment. Second, expertise-based randomization may limit the external validity of the trial because only a select few, highly specialized individuals would be eligible to participate. Finally, in our meta-analysis,10 the benefits of radial compared to femoral access were seen in both trials of radial experts (operators' preferred route was radial) and nonradial experts. The cut point of a requirement of 50 procedures in the previous year is based on published data on the learning curve of previously exclusive femoral operators to proficiently use radial access with high success rates and within an acceptable procedure time.21,22 Specifically, in a study of 4 experienced femoral operators during their first 415 radial procedures, the learning curve plateaued at 50 radial procedures per operator in procedural success, fluoroscopy, and procedural time, and these results have been replicated in other studies of the radial learning curve in experienced femoral operators.21,22 In a recent randomized trial of 1,024 patients, experienced femoral operators who had performed ≥50 radial procedures demonstrated high radial procedural success rates of 96.5% (3.5% crossover to femoral).23 Annual operator volume for both radial and femoral diagnostic and PCI procedures is being recorded on the

case report forms. Based on interim data from the first 6,925 patients randomized, the procedural volumes of the operators in the RIVAL trial were high (median 300 PCI/y, first quartile (Q1) 186 and third quartile (Q3) 400 PCI/y), and they had experience with both procedures (median proportion annual PCI radial 40%, Q1 25% and Q3 70%).

Randomization Patients are randomized in equal proportions to the 2 groups using a 24-hour computer central automated voice response system located at the coordinating center in Hamilton, Canada.

Interventions 1. Radial access to perform coronary angiography and PCI (if clinically indicated); or 2. Femoral access to perform coronary angiography and PCI (if clinically indicated). The use of an arterial vascular closure device is allowed at the discretion of the treating physician.


Jolly et al 257

Table I. Eligibility criteria Inclusion criteria ACS patients Patients with UA or NSTEMI

Ischemic symptoms suspected to represent a non–ST-segment elevation ACS (UA or NSTEMI) were defined as clinical history consistent with new onset or a worsening pattern of characteristic ischemic chest pain occurring at rest or with minimal exertion (lasting longer than 10 min) and at least one of the following: (1) ECG changes compatible with new ischemia (ST depression of at least 1 mm or transient ST elevation or ST elevation of ≤1 mm or T wave inversion N3 mm in at least 2 contiguous leads); (2) patients N60 y with normal ECG are eligible provided that there is a high degree of certainty that presenting symptoms are because of myocardial ischemia. Such patients must have documented evidence of previous CAD with at least one of the following: (a) prior MI requiring hospitalization, (b) prior revascularization procedure (N3 m ago), (c) cardiac catheterization showing significant CAD, (d) positive exercise test, and (e) other objective evidence of atherosclerotic vascular disease; or (3) already elevated cardiac biomarkers (CK-MB or troponin T or I) above the upper limit of normal. Patients with STEMI (1) Presenting with signs or symptoms of acute MI lasting ≤20 min and (2) definite ECG changes compatible with STEMIpersistent ST elevation (≥2 mm in 2 contiguous precordial leads or N1 mm in ≤2 limb leads) or new left bundle-branch block or Q wave in 2 contiguous leads. Intent to perform same-sitting coronary angiography and PCI during index hospitalization Written informed consent Suitable candidate for either radial or femoral artery PCI, including (a) palpable radial artery with documented normal Allen test, (b) previous experience of the operator with ≥50 cases within the past year of radial artery access for coronary angiography/intervention, and (c) acceptance by operator to use whichever route is assigned by randomization Exclusion Criteria Age b18 y Active bleeding or significant increased risk of bleeding (severe hepatic insufficiency, current peptic ulceration, proliferative diabetic retinopathy) Uncontrolled hypertension Cardiogenic shock Prior CABG surgery with use of N1 internal mammary artery Documented severe peripheral vascular disease precluding a femoral approach Previously entered in the study Investigational treatment (drug or device) within the previous 30 d Medical, geographic, or social factors making study participation impractical or inability to provide written informed consent and to understand the full meaning of the informed consent UA, Unstable angina; CAD, coronary artery disease.

Procedures Patients are screened before undergoing coronary angiography with the permission of the treating interventional cardiologist and then randomized after informed consent. In patients undergoing PCI, troponin, creatine kinase (CK), and CK-MB must be drawn immediately pre PCI and at 2, 6, and 12 hours post PCI. For patients requiring CABG, blood must be drawn for CK and CK-MB immediately pre CABG and at 6 and 12 hours post CABG. All patients must have electrocardiogram (ECG) pre CABG, immediately post CABG, and at the time of discharge.

Outcomes

Other outcomes 1. Death within 30 days; 2. Components of primary outcome at 48 hours and at 30 days; 3. PCI procedural success; 4. Major vascular access site complications at 48 hours and 30 days after the procedure (major vascular access site complications include pseudoaneurysms requiring ultrasound compression, thrombin injection, or surgical repair; and large hematomas requiring prolonged hospitalization, arteriovenous fistulae, limb ischemia, or damage to adjacent nerve).

Primary outcome The primary efficacy outcome is the occurrence of death, MI, stroke or non–CABG-related major bleeding within 30 days.

Study outcome definitions are listed in Table II. Mortality is the first and most important other outcome and has the potential to confirm the link between mortality and major bleeding.

Key secondary outcomes The 2 key secondary outcomes are:

Central events adjudication

1. Death, MI, or stroke within 30 days; and 2. Non–CABG-related major bleeding within 30 days.

A committee of clinicians blinded to treatment allocation will adjudicate all primary efficacy outcomes and bleeding events.


258 Jolly et al

Table II. Outcome definitions First occurrence until day 30 of any component of death, MI, stroke, non–CABG-related major bleeding Term Non–CABG-related major bleeding

Death

MI

Definition Defined as bleeding that is (a) fatal, (b) results in transfusion of ≥2 U of red blood cells or equivalent whole blood, (c) causes significant hypotension with the need for inotropes or surgical intervention (a requirement for surgical access-site repair will constitute major bleeding only if there has been significant hypotension or transfusion of ≥2 U), (d) causes significantly disabling sequellae, or (e) is intracranial and symptomatic or intraocular and leads to significant visual loss. The primary outcome uses all-cause mortality. All deaths will be subclassified as cardiovascular and noncardiovascular. All deaths with a clear cardiovascular cause or unknown will be classified as cardiovascular (including complications of procedures and bleeding). Only deaths because of documented noncardiovascular cause (eg, cancer) will be classified as noncardiovascular. The diagnosis of new MI will depend on the timing of the event after randomization (ie, within the first 24 h of randomization, between 24 h and 7 d after randomization, and N7 d after randomization), the presence or absence of an associated MI at baseline, and whether the suspected event occurred after a revascularization procedure. No associated MI at baseline In patients with no associated baseline MI, either one of the 2 following criteria satisfies the diagnosis for an acute MI: (1) typical rise and fall of biochemical markers of myocardial necrosis including troponin, CK-MB, CK to N2× ULN (or if markers are already elevated, N50% of the lowest recovery enzyme level from the index infarction) with at least one of the following: (a) ischemic symptoms, (b) development of pathologic Q waves on the ECG, and (c) ECG changes indicative of ischemia (ST-segment elevation or depression); or (2) pathologic findings of an acute MI. MI within 24 h of randomization In UA/NSTEMI patients with an associated MI at baseline or in STEMI patients, a new MI within 24 h of randomization is defined as (1) new ischemic symptoms N20 min and (2) new or recurrent ST-segment elevation or depression N1 mm in ≥2 contiguous leads, not due to changes from evolution of the index MI or STEMI patients. MI between 24 h and 7 d In UA/NSTEMI patients with an associated MI at baseline or in STEMI patients, of randomization a new MI between 24 h and 7 d is defined as (a) new ischemic symptoms N20 min and (b) elevation or reelevation of CK-MB (or total CK if CK-MB is not available) ≥2× the upper limit of normal or N50% above the previous valley level and N2× the upper limit of normal in patients with already elevated enzymes or new or recurrent ST-segment elevation or depression N1 mm or new significant Q waves in ≥2 contiguous leads discrete from the baseline MI. MI occurring after 7 d or hospital In all patients, the definition of new MI occurring after hospital discharge or discharges, whichever comes first after 7 d, whichever comes first, will be either one of the 2 following criteria satisfies the diagnosis for an acute, evolving, or recent MI: typical rise and fall of biochemical markers of myocardial necrosis (including troponin, CK-MB, and CK) to N2× ULN (or if markers are already elevated, N50% of the lowest recovery enzyme level from the index infarction and N2× ULN) with at least one of the following: (a) ischemic symptoms, (b) development of pathologic Q waves on the ECG, and (c) ECG changes indicative of ischemia (ST-segment elevation or depression); or pathologic findings of an acute MI. MI post PCI For patients with MI within 24 h after PCI, a new MI is defined by (1) CK-MB⁎ (or total CK if CK-MB is unavailable) ≥3× the upper limit of normal or increased by 50% from the preprocedural valley level and ≥3× ULN in patients with already elevated enzymes or (2) new ST-segment elevation or development of new or significant Q waves in ≥2 contiguous leads (discrete from the baseline MI in STEMI patients). MI post CABG For patients with MI within 24 h after CABG, a new MI is defined by CK-MB (or total CK, if CK-MB is unavailable) (1) ≥5× the upper limit of normal or increased by 50% from the preprocedural valley level and ≥5× ULN in patients with already elevated enzymes and development of new pathologic Q waves in ≥2 contiguous leads or (2) CK-MB value ≥10× ULN without new pathologic Q waves. In all cases of new MI, troponin T or I may be used for the diagnosis of new MI in the absence of CK-MB at the discretion of the event adjudication committee, taking into consideration all available clinical and laboratory evidence. In addition, the event adjudication committee may request further details of the revascularization procedure such as the PCI or CABG written report or supplementary narratives to assist in ascertainment of new post PCI or CABG MI.


Jolly et al 259

Table II (continued) First occurrence until day 30 of any component of death, MI, stroke, non–CABG-related major bleeding Term

Definition

Stroke

Any stroke is defined as the presence of a new focal neurologic deficit thought to be vascular in origin, with signs or symptoms lasting N24 h. It is strongly recommended (but not required) that an imaging procedure such as a CT or MRI be performed. Failure: no success at dilating attempted lesion(s) and/or failure to cross/dilate/not attempted. Partial success: one of ≥2 attempted lesions was successfully dilated and procedure performed but N50% residual or TIMI flow b3 or failure. Full success: lesion(s) attempted was successfully dilated with b50% residual or TIMI 3 flow.

PCI success

Steering committee: S. Jolly (coprincipal investigator), S.R. Mehta (coprincipal investigator), A. Avezum, A. Budaj, J. Cairns, S. Chrolavicius (project manager), R. Diaz, V. Dzavik, M.G. Franzosi, C. Granger, C. Joyner (events adjudication committee chair), M. Keltai, F. Lanas, B. Lewis, K. Niemela, S. Rao, P.G. Steg, V. Valentin, P. Widimsky, D. Xavier, and S. Yusuf⁎ (steering committee chair). Data monitoring committee: P. Sleight (chair), J.L. Anderson, D.L. DeMets, J. Hirsh, D.R. Holmes, Jr, and D.E. Johnstone. Project office staff: S. Chrolavicius (project manager), R. Afzal (statistician), L. Blake, W. Chen, S. Di Diodato, C. Cramp (research coordinator), C. Horsman (research coordinator), B. Jedrzejowski (research coordinator), M. Lawrence (event adjudication coordinator), A. Lehmann, A. Robinson (research coordinator), R. Manojlovic, L. Mastrangelo, E. Pasadyn, T. Sovereign, L. Wasala, L. Xu (statistician). ⁎ If available, troponin may be used instead of CK-MB post PCI if no associated MI at baseline. CT, computed tomography; MRI, magnetic resonance imaging. TIMI, thrombolysis in myocardial infarction; ULN, upper limit of normal.

Statistical considerations

Limitations

The originally estimated sample size was 4,000 patients, which was chosen to provide 80% power for the detection of a relative risk reduction of 25% for a 10% rate of the primary outcome with a 2-sided α ≤ .05. However, it was evident by July 2009 that the aggregate event rate was much lower than originally estimated, and a revised estimate of 6% for the primary outcome at 30 days in the femoral access group led to the calculation of a revised sample size of approximately 7,000 patients to provide 80% power to detect a relative risk reduction of 25%, with 2-sided α ≤ .05. In September 2010, before the end of enrollment and before blinding, the steering committee clarified that the major bleeding component of the primary outcome refers to non-CABG major bleeding. This was because CABG-related major bleeding is unlikely to be modified by vascular access site. The protocol clarification was issued to all sites. Coronary artery bypass graft–related major bleeding will be reported as a tertiary outcome. For the primary analysis, the relative efficacy of radial access versus femoral access will be assessed on the primary outcome by a comparison of the survival curves (estimated using the Kaplan-Meier method) for the 2 treatments using the log-rank statistic (primary test of treatment effect). Treatment effect, expressed as the hazard ratio (radial access vs femoral access) and corresponding 95% CIs, will be estimated from a Cox proportional hazards model. Statistical significance will be assessed using a 2-sided α ≤ .05. There will be no adjustment for multiple comparisons of key secondary outcomes because there are only 2, and each contains components of the primary outcome. An analysis of subgroups by tertiles of both center and operators' radial procedural volume will be performed as well.

The first potential limitation of the RIVAL trial is the exclusion of higher risk patients. For example, operators may have been unwilling to randomize morbidly obese patients because of a lack of clinical equipoise in this population. A second potential limitation is that very high-proportion radial operators (N90%) may not have commonly participated in the trial potentially because of a lack clinical equipoise between the 2 procedures among these operators. However, on the other hand, very high-proportion radial operators could have higher femoral complications rates; accordingly, we designed a pragmatic trial engaging operators with expertise in both procedures. Finally, the optimal minimum annual number of radial procedures to define expertise is unknown; accordingly, there is a potential limitation in the chosen cut point of 50 radial procedures annually. However, in practice, the RIVAL trial was successful in engaging highvolume operators who had significantly exceeded these minimum requirements for radial volume and who had extensive experience with both procedures.

Study status Enrollment was completed on November 3, 2010, with 7,021 patients randomized with the expectation that 30-day data are likely to be available by the spring or summer of 2011. The recruitment by country and site is shown in an online Appendix.

Summary The RIVAL trial will be the first large randomized trial comparing radial versus femoral access for coronary angiography/intervention among patients with ACS that

260 Jolly et al

is adequately powered to determine effects on major efficacy and safety outcomes.

Disclosures Funding for the RIVAL trial was provided by SanofiAventis, Population Health Research Institute, and the Canadian Network and Center for Trials Internationally (CANNeCTIN), which is funded by the Canadian Institutes of Health Research.

References 1. Stone GW, McLaurin BT, Cox DA, et al. Bivalirudin for patients with acute coronary syndromes. N Engl J Med 2006;355:2203-16. 2. Yusuf S, Mehta SR, Chrolavicius S, et al. Comparison of fondaparinux and enoxaparin in acute coronary syndromes. N Engl J Med 2006; 354:1464-76. 3. Fox KA, Goodman SG, Klein W, et al. Management of acute coronary syndromes. Variations in practice and outcome; findings from the Global Registry of Acute Coronary Events (GRACE). Eur Heart J 2002;23:1177-89. 4. Budaj A, Eikelboom JW, Mehta SR, et al. Improving clinical outcomes by reducing bleeding in patients with non–ST-elevation acute coronary syndromes. Eur Heart J 2008. 5. Rao SV, O'Grady K, Pieper KS, et al. Impact of bleeding severity on clinical outcomes among patients with acute coronary syndromes. Am J Cardiol 2005;96:1200-6. 6. Eikelboom JW, Mehta SR, Anand SS, et al. Adverse impact of bleeding on prognosis in patients with acute coronary syndromes. Circulation 2006;114:774-82. 7. Ndrepepa G, Berger PB, Mehilli J, et al. Periprocedural bleeding and 1-year outcome after percutaneous coronary interventions: appropriateness of including bleeding as a component of a quadruple end point. J Am Coll Cardiol 2008;51:690-7. 8. Chase AJ, Fretz EB, Warburton WP, et al. The association of arterial access site at angioplasty with transfusion and mortality: the M.O.R.T. A.L study: (Mortality benefit of Reduced Transfusion After PCI via the Arm or Leg). Heart 2008. 9. Rao SV, Ou FS, Wang TY, et al. Trends in the prevalence and outcomes of radial and femoral approaches to percutaneous coronary intervention: a report from the National Cardiovascular Data Registry. JACC Cardiovasc Interv 2008;1:379-86. 10. Jolly SS, Amlani S, Hamon M, et al. Radial versus femoral access for coronary angiography or intervention and the impact on major bleeding and ischemic events: a systematic review and meta-analysis of randomized trials. Am Heart J 2009;157:132-40.


11. Montalescot G, Ongen Z, Guindy R, et al. Predictors of outcome in patients undergoing PCI. Results of the RIVIERA study. Int J Cardiol 2007. 12. Cantor WJ, Mahaffey KW, Huang Z, et al. Bleeding complications in patients with acute coronary syndrome undergoing early invasive management can be reduced with radial access, smaller sheath sizes, and timely sheath removal. Catheter Cardiovasc Interv 2007;69: 73-83. 13. Hamon M, Steg G, Faxon D, et al. Major bleeding in patients with acute coronary syndrome undergoing early invasive management can be reduced by fondaparinux, even in the context of trans-radial coronary intervention: insights from OASIS-5 trial. Circulation 2006; 114(Suppl II):552. 14. Hamon M, Rasmussen LH, Manoukian SV, et al. Choice of arterial access site and outcomes in patients with acute coronary syndromes managed with an early invasive strategy: the ACUITY trial. EuroIntervention 2009;5:115-20. 15. Biancari F, D'Andrea V, Di Marco C, et al. Meta-analysis of randomized trials on the efficacy of vascular closure devices after diagnostic angiography and angioplasty. Am Heart J 2010;159: 518-31. 16. Manoukian SV, Feit F, Mehran R, et al. Impact of major bleeding on 30-day mortality and clinical outcomes in patients with acute coronary syndromes: an analysis from the ACUITY trial. J Am Coll Cardiol 2007;49:1362-8. 17. Stone GW, Witzenbichler B, Guagliumi G, et al. Bivalirudin during primary PCI in acute myocardial infarction. N Engl J Med 2008;358: 2218-30. 18. Mehta SR, Bassand JP, Chrolavicius S, et al. Design and rationale of CURRENT-OASIS 7: a randomized, 2 × 2 factorial trial evaluating optimal dosing strategies for clopidogrel and aspirin in patients with ST and non–ST-elevation acute coronary syndromes managed with an early invasive strategy. Am Heart J 2008;156:1080-8.e1. 19. Mehta SR, Bassand JP, Chrolavicius S, et al. Dose comparisons of clopidogrel and aspirin in acute coronary syndromes. N Engl J Med 2010;363:930-42. 20. Mehta SR, Tanguay JF, Eikelboom JW, et al. Double-dose versus standard-dose clopidogrel and high-dose versus low-dose aspirin in individuals undergoing percutaneous coronary intervention for acute coronary syndromes (CURRENT-OASIS 7): a randomised factorial trial. Lancet 2010;376:1233-43. 21. Spaulding C, Lefevre T, Funck F, et al. Left radial approach for coronary angiography: results of a prospective study. Cathet Cardiovasc Diagn 1996;39:365-70. 22. Hildick-Smith DJ, Lowe MD, Walsh JT, et al. Coronary angiography from the radial artery—experience, complications and limitations. Int J Cardiol 1998;64:231-9. 23. Brueck M, Bandorski D, Kramer W, et al. A randomized comparison of transradial versus transfemoral approach for coronary angiography and angioplasty. JACC Cardiovasc Interv 2009;2:1047-54.


Jolly et al 260.e1

Appendix A

BELGIUM (43 patients)

Table I. Recruitment by country Country

Argentina Australia Belgium Brazil Bulgaria Canada Chile China Croatia Czech Republic Finland France Germany Hungary India Ireland

No. of patients randomized

Country

No. of patients randomized

94 55 43 183 551 1475 146 208 1 187 1124 187 94 31 902 9

Israel Italy Latvia Lithuania Malaysia Mexico New Zealand Poland Romania Russia Singapore Slovakia Spain Sweden United Kingdom United States

239 129 11 80 1 17 9 583 12 1 5 20 432 50 20 122

Appendix B: RIVAL program list of principal investigators and recruitment by site ARGENTINA (94 patients) 252 Dr Eduardo Conrado Conci, Instituto Modelo De Cardiologia Priv. S.R.L. (3) 260 Dr L. L. Lobo Marquez, Instituto de Cardiologia de Tucuman—SRL (1) 265 Dr Guillermo Covelli, Clinica Privada del Prado (16) 266 Dr Miguel Angel Hominal, Sanatorio Medico de Diagnostico y Tratamiento (13) 271 Dr Gerardo Zapata, Instituto Cardiovascular de Rosario (5) 272 Dr J. C. Pomposiello, Hospital Privadomde Comunidad (12) 277 Dr Simon Salzberg, Hospital Juan A. Fernandez (15) 278 Dr Alejandro Sanchez, Policlinico Modelo Cipolletti (14) 280 Dr Daniel Santos, Instituto Cardiologico Especializado SRL (1) 282 Dr Daniel Piskorz, Sanatorio Britanico de Rosario (2) 284 Dr Mario Berli, Hospital Provinicial Dr Jose Maria Cull (5) 286 Dr Claudo Rodolfo Majul, Hospital Britanico de Buenos Aires (7) AUSTRALIA (55 patients) 233 Dr Matthew Worthley, Royal Adelaide Hospital (54) 240 Prof Peter Thompson, Sir Charles Gairdner Hospital (1)

936 Dr P-E Massart, Clinique et Maternite SainteElisabeth de Namur (21) 937 Prof Vincent Dangoisse, Cliniques Universitaires de Mont-Godinne (7) 938 Dr Marc Vincent, Clinique générale Saint-Jean (15) BRAZIL (183 patients) 315 Dr José Armando Mangione, Hospital Beneficiência Portuguesa de São Paulo (8) 320 Dr Sandra Andrade, Mendonça Hilgemberg Incor-hemocardio (9) 326 Dr Maria Sanali Moura de Oliveira Paiva, Natal Hospital Center (10) 328 Dr Gilmar Reis, Hospital São Francisco de Assis (17) 331 Dr José Francisco Kerr Saraiva, HMCP PucCampinas Hospital e Maternidade Celso Pierro (5) 332 Dr Ari Timerman, Instituto Dante Pazzanese de Cardiologia (48) 333 Dr Rogério Tadeu Tumelero, Hospital São Vicente de Paulo (38) 334 Dr Wladimir Faustino Saporito, Hospital Estadual Mario Covas (8) 336 Dr Roberto Vieira Botelho, Instituto do Coracao do Triangulo Mineir (40) BULGARIA (551 patients) 431 Dr Alexander Doganov, National Heart Hospital (23) 432 Prof Julia Jorgova-Makedonska, “St Ekaterina” University Hospital (21) 435 Dr Atanas Penev, UMHAT-Sveta Marina (3) 433 Dr Georgi Mazhdrakov, UMHAT “St Anna” (52) 436 Dr Ivan Manoukov, Clinic of Invasive Cardiology, University Hospital “Sveti Georgi” (452) CANADA (1,475 patients) 001 Dr Sanjit Jolly/Dr Shamir Mehta, Hamilton General Hospital (696) 002 Dr Michael Rokoss, Henderson Hospital (31) 003 Dr Omid Salehian, McMaster University Medical Centre (29) 004 Dr Gilbert Gosselin, Montreal Heart Institute (1) 005 Dr Jeffrey Pang/Dr Cam Joyner, Sunnybrook Health Sciences Centre (30) 006 Dr Sven Pallie, Niagara Health System (NHS)— St Catharines General Hospital Site (58) 007 Dr Anthony Fung, Vancouver General Hospital (150) 009 Dr Yun Kai Chan, NHS—Greater Niagara General Site (12) 013 Dr Farrukh Hussain, St Boniface General Hospital (10) 014 Dr C. Van Kieu, CSSSRY C.H. Honore Mercier (1) 018 Dr Asim Cheema, St Michael's Hospital (138)


260.e2 Jolly et al

019 Dr G. Calvin, MacCallum Memorial University Eastern Health Sciences Center (7) 020 Dr Frank Nigro, Intermountain Research Consultants (2) 026 Dr Patrick Beliveau, CHUQ—Hotel Dieu de Quebec (49) 030 Dr Denis-Carl Phaneuf, Hospital Pierre-Le Gardeur (3) 031 Dr Jean-Pierre Dery, Hôpital Laval (7) 033 Dr Vlad Dzavik, University Health Network (22) 035 Dr Salima Shariff, Surrey Memorial Hospital (27) 039 Dr Joseph Berlingieri, Joseph Berlingieri and William Nisker (JBN) Medical Diagnostic Services (19) 062 Dr Warren Cantor, York PCI Group Inc (53) 063 Dr Robert Boone, Providence Health Care, St Paul's (42) 064 Dr Francois Charbonneau, University of Calgary (88) CHILE (146 patients) 353 Dr R. Lamich Betancourt, Hospital Barros Luco (70) 354 Dr C.P.P. Jofre, Hospital Dr Herman Henriquez Aravena (52) 355 Dr E.E.G. Flores, Hospital Clinico Regional Valdivia (23) 360 Dr Misael Lopetegui, Hospital Clínico San Borja Arriarán (1) CHINA (208 patients) 906 Dr Shuyang Zhang, Peking Union Medical College Hospital (10) 912 Dr Yaling Han, The General Hospital of Shenyang Military C (11) 914 Dr Meng Wei, Shanghai No. 6 People Hospital (22) 915 Dr Daowen Wang, Tongji Hospital of Huazhong University (9) 917 Dr Jianan Wang, Second Affiliated Hospital Zhejiang University (22) 919 Dr Jiyan Chen, GuangDong Provincial People's Hospital (60) 924 Prof Tianchang Li, Beijing Tongren Hospital (1) 927 Dr Biao Xu, The Affiliated Drum Tower Hospital (31) 928 Dr Genshan Ma, Zhong Da Hospital (42) CROATIA (1 patient) 457 Dr Mijo Bergovec, Klinicka bolnica Dubrava (1) CZECH REPUBLIC (187 patients) 404 Assoc Prof Dr Pavel Cervinka, Masaryk Hospital (54) 405 MUDr Zdenek Coufal, Regional Hospital T. Bati a.s. Zlín (14) 410 Dr Petr Kala, Fakultní nemocnice Brno ( 10) 412 Prof Ales Linhart, General Faculty Hospital (3)

413 Dr Roman Ondrejcak, KKN a.s., Nemocnice Karlovy Vary, kardiologické odd (29) 421 Prof Jan Vojacek, Fakultni nemocnice Hradec Kralove (6) 422 Dr Richard Rokyta, I. Interní klinika, Fakultní (27) 430 Dr Michal Rezek, University Hospital St Anna (44) FINLAND (1,124 patients) 636 Oyl Saila Vikman, Heart Center, Tampere University Hospital (232) 637 Prof Juhani Airaksinen, Turku University Hospital (30) 640 Oyl Antti Ylitalo, Satakunta Central Hospital (5) 641 Oyl Matti Niemela, Oulu University Hospital (857) FRANCE (187 patients) 583 Dr Emile Ferrari, CHU de Nice—Hôpital Pasteur (160) 585 Dr Gilles Grollier, CHU de caen—Hôpital Côte de Nacre (6) 595 Prof Gabriel Steg, Hôpital Bichat Claude Bernard (21) GERMANY (94 patients) 508 Dr Dietrich Andresen, Vivantes Klinikum am Urban (8) 510 Dr Florin Laubenthal, Elisabeth-Krankenhaus Essen (1) 522 Dr Juergen vom Dahl, Kliniken Maria Hilf GmbH (62) 525 Dr Christoph Nienaber, Universitaet Rostock (7) 532 Dr Stefan Hoffmann, Vivantes Klinikum im Friedrichshain (16) HUNGARY (31 patients) 485 Dr Peter Polgár Josa Andras Teaching Hospital (1) 487 Dr Béla Nagybaczoni Bajcsy Zsilinszky Hospital (30) INDIA (902 patients) 851 Dr Kumar Rajendra Premchand, Krishna Institute of Medical Sciences (240) 852 Dr M. Bhaskar Rao, CARE Hospital (71) 854 Dr Keyur Parikh, S.A.L. Hospital & Medical Institute (161) 862 Dr Nakul Sinha, Sanjay Gandhi PGIMS (299) 863 Dr Hemang Baxi, The Heart Care Clinic (38) 872 Dr Brian Pinto, Holy Family Hospital (76) 874 Dr Sudhir R. Naik, Apollo Hospitals (2) 881 Dr Pradeep Kumar Shetty, Narayana Hrudayalaya (15) IRELAND (9 patients) 391 Dr Peter Crean, St James's Hospital (9) ISRAEL (239 patients) 971 Dr Yoav Turgeman, Haemek Medical Centre (17)


972 Prof Basil S. Lewis, Lady Davis Carmel Medical Centre (114) 980 Dr Joel Arbel, Meir Medical Center (3) 981 Dr Alon Marmor, Heart Institute (62) 983 Dr Majdi Halabi, Division of Invasive Cardiology (43) ITALY (129) 756 Dr Marco Rossi, Istituto Clinico Humanitas (1) 766 Prof Giancarlo Piovaccari, Ospedale Infermi U.O. Cardiologia (56) 767 Dott Salvatore Pirelli, Div. Di Cardiologia, Azienda Istituti Ospitalieri Cremona (43) 771 Dr Giuseppe Steffenino, Azienda Ospeddliera Santa Croce e Carle (3) 773 Dr Roberto Zanini, Azienda Ospedaliera “Carlo Poma” (12) 775 Dr Ugo Limbruno, Ospedale della Misericordia (14) LATVIA (11 patients) 396 Dr Andrejs Erglis, Pauls Stradins Clinical University Hospital (11) LITHUANIA (80 patients) 992 Dr Ramunas Unikas, Kaunas Medical University Hospital (80) MALAYSIA (1 patient) 801 Dr Chong Wei Peng, University Malaya Medical Centre (1) MEXICO (17 patients) 363 Dr Jose Luis Arenas Leon, Hospital Angeles Centro Medico Del Potos (5) 370 Dr M.S.L.Velasco, Star Medica Morelia (12) NEW ZEALAND (9 patients) 067 Dr Gerard Devlin, Waikato Hospital (6) 068 Dr Scott Harding, Wellington Hospital (3) POLAND (583 patients) 617 Prof Andrzej Budaj, Szpital Grochowski (134) 618 Dr Piotr Achremczyk, Radomski Szpital Specjalistyczny (7) 619 Dr Pawel Miekus, Szpital Miejski im. J. Brudzinskiego (6) 621 Dr Barbara Kusnierz, Wojewódzki Szpital Specjalistyczny nr 4 (19) 623 Dr Bozena Wrzosek, Wojewódzki Szpital Specjalistyczny (181) 624 Dr Jerzy Kopaczewski, Szpital Wojewodzki (16) 628 Dr Jan Wodniecki, Szpital Specjalistyczny w Zabrzu (3) 628 Dr Damian Kawecki, Szpital Specjalistyczny w Zabrzu (26)

Jolly et al 260.e3

631 Dr Maciej Dalkowski, Regionalny Osrodek Kardiologii, “Miedziowe Centrum Zdrowia” S.A. (171) 647 Prof Lech Polonski, III, Katedra i Oddzial Kliniczny Kardiologii SUM (14) 648 Prof Zbigniew Kalarus, Slaskie Center for Heart Disease (6) ROMANIA (12 patients) 442 Prof Radu Capalneanu, Institutul Inimii N. Stancioiu (12) RUSSIA (1 patient) 813 Prof Svetlana Berns, Municipal Healthcare Institution (1) SINGAPORE (5 patients) 791 Prof Tian Hai Koh, National Heart Centre (5) SLOVAKIA (20 patients) 452 Dr Roman Margozcy, Stredoslovensky ustav srdcovych a cievnych chorob (10) 454 Dr Peter Kurray, Kardiocentrum Nitra sro (10) SPAIN (432 patients) 701 Dr Vicent Valentin, Hospital Universitario Dr Peset (128) 702 Dr Nicolas Vazquez, Hospital Universitario Juan Canalejo (13) 706 Dr Juan Angel Ferrer, Hospital General Vall D'Hebron (25) 707 Dr Manel Sabate, Hospital De La Santa Creu i De Sant Pau (4) 713 Dr Inaki Lekuona, Hospital De Galdakao-Usansolo (51) 721 Dr Francisco Bosa, Hospital Universitario de Canarias (116) 722 Dr Jose Moreu, Hospital Virgen de la Salud (1) 724 Dr Andrés Iñiguez, Hospital Meixoeiro (24) 725 Dr Francisco Macaya, Hospital Clínico San Carlos (29) 730 Dr Juan Sanchis, Hospital Clinico de Valencia (4) 732 Dr Ramón López Palop, Hospital Universitario San Juan de Alicante (35) 738 Dr Ramiro Trillo, Hospital Clinico de Santiago (2) SWEDEN (50 patients) 652 Dr Aida Hot-Bjelak, Capio St Goran's Hospital (15) 653 Dr Loghman Henareh, Karolinska University Hospital (5) 654 Prof Goran Olivecrona, Lund University Hospital (30) UNITED KINGDOM (20 patients) 941 Dr Andreas Baumbach, Bristol Royal Infirmary (15)

260.e4 Jolly et al

944 Prof Adam de Belder, Royal Sussex County Hospital (2) 948 Dr Iqbal Malik, St Mary's Hospital (1) 950 Dr Mike Pitt, Heart of England NHS Foundation Trust (2) USA (122 patients) 75 Dr Mehrdad Saririan, Maricopa Integrated Health System (1)


104 Dr Ramesh Mazhari, GWU Medical Faculty Associates (7) 105 Dr Steven L. Goldberg, University of Washington (1) 172 Dr Narendra Singh, Northside Cardiology (32) 178 Dr Joseph Chambers, Endovascular Research (1) 183 Dr Stephen Thew, Heart Clinics Northwest (62) 186 Dr M. Kevin Ariani, Northridge Hospital Medical Center (1) 192 Dr Joana Magno, The Queen's Medical Center (4)

A randomized, partially blinded, multicenter, active-controlled, dose-ranging study assessing the safety, efficacy, and pharmacodynamics of the REG1 anticoagulation system in patients with acute coronary syndromes: Design and rationale of the RADAR Phase IIb trial Thomas J. Povsic, MD, PhD, a,l Mauricio G. Cohen, MD, b,l Roxana Mehran, MD, c,l Christopher E. Buller, MD, d,l Christoph Bode, MD, e,l Jan H. Cornel, MD, f,l Jarosław D. Kasprzak, MD, g,l Gilles Montalescot, MD, h,l Diane Joseph, a William A. Wargin, PhD, i Christopher P. Rusconi, PhD, j Steven L. Zelenkofske, DO, k,l Richard C. Becker, MD, a,l and John H. Alexander, MD, MHS a,l Durham, and Chapel Hill, NC; Miami, FL; New York, NY; Ontario, Canada; Freiberg, Germany; Alkmaar, Netherlands; Łódź , Poland; Paris, France; and Basking Ridge, NJ

Anticoagulants are the cornerstone of current acute coronary syndrome (ACS) therapy; however, anticoagulation regimens that aggressively reduce ischemic events are almost uniformly associated with more bleeding. REG1, an anticoagulation system, consists of RB006 (pegnivacogin), an RNA oligonucleotide factor IXa inhibitor, and RB007 (anivamersen), its complementary controlling agent. Phase I and IIa studies defined predictable relationships between doses of RB006, RB007, and degree of antifactor IX activity. The efficacy and safety of REG1 for the treatment of patients with ACS managed invasively and the safety of reversing RB006 with RB007 after cardiac catheterization are unknown. Randomized, partiallyblinded, multicenter, active-controlled, dose-ranging study assessing the safety, efficacy, and pharmacodynamics of the REG1 anticoagulation system compared to unfractionated heparin or low molecular heparin in subjects with acute coronary syndrome (RADAR) is designed to assess both the efficacy of the anticoagulant RB006 and the safety of a range of levels of RB006 reversal with RB007. The objectives of RADAR are (1) to determine the safety of a range of levels of RB006 reversal with RB007 after catheterization, (2) to confirm whether a dose of 1 mg/kg RB006 results in near-complete inhibition of factor IXa in patients with ACS, and (3) to assess the efficacy of RB006 as an anticoagulant in patients with ACS undergoing percutaneous coronary intervention. (Am Heart J 2011;161:261-268.e2.)

Anticoagulant therapies are the cornerstone of treatment of patients with acute coronary syndromes (ACS); however, they are associated with clinically significant increases

From the aDuke Clinical Research Institute, Duke University Medical Center, Durham, NC, b Miller School of Medicine, University of Miami, Miami, FL, cMount Sinai Medical Center, New York, NY, dHamilton General Hospital, Hamilton, Ontario, Canada, eUniversity of Freiberg, Freiberg, Germany, fMedisch Centrum Alkmaar, Alkmaar, Netherlands, g Medical University of Lodz, Łódź, Poland, hInstitut de Cardiologie, Pitié-Salpétrière Hospital, Paris, France, iPK-PM Associates, LLC, Chapel Hill, NC, jRegado Biosciences, Durham, NC, and kRegado Biosciences, Basking Ridge, NJ. l On behalf of the RADAR Investigators. See online Appendix B for a complete listing. RCT reg no. NCT00932100. Vladimir Dzavik, MD, served as guest editor for this article. Submitted July 1, 2010; accepted October 15, 2010.

Reprint requests: Thomas J. Povsic, MD, PhD, Alexander H. Sands Building, 303 Research Drive, Room 321, Duke University Medical Center, Durham, NC, 27710. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.10.022

in bleeding.1 As the clinical consequences of bleeding have become clearer,2-6 attention has focused on reducing bleeding without compromising antithrombotic efficacy. The ideal anticoagulant would prevent thrombus formation, cause minimal or no bleeding, be easy to administer and monitor, and be reversible should a clinical need (bleeding) arise or when anticoagulation is no longer necessary. Currently available parenteral anticoagulants have important limitations.7 Unfractionated heparin has variable and unpredictable anticoagulant effects, results in platelet activation, can cause heparininduced thrombocytopenia, and its reversal agent, protamine, has its own adverse effects.8-10 Low-molecular-weight heparins are long acting, lack effective reversal strategies, cannot be monitored, are associated with variability in achieving adequate levels of anticoagulation,11 and require a delay in sheath removal.12 Bivalirudin is rapidly cleared but requires an infusion for


262 Povsic et al

Figure 1

Table I. Components of the REG1 system

Half-life Clinical activity Clinical target Elimination

Structure of RB006-RB007 complex. From Dyke CK, Steinhubl SR, Kleiman NS, et al. Circulation 2006;114:2490-7.

longer treatment times and may be less effective at preventing ischemic events.13 Fondaparinux is associated with low bleeding rates but is less effective as an anticoagulant in the catheterization laboratory.14 One approach to more reliable inhibition of coagulation is to target upstream proteases that are activated in the early steps of coagulation and are present at lower and more predictable concentrations. Factor IX is a particularly attractive target, given its central role in the generation of factor Xa and thrombin and in the propagation of the clot on the surface of activated platelets.15-17 In addition, as tissue factor-mediated factor X activation proceeds through factor IX, effective factor IX inhibition may inhibit thrombin generation via both the intrinsic and extrinsic pathways.18 Finally, the pathophysiologic correlate of factor IX inhibition is hemophilia B. Patients with hemophilia B have prolonged activated partial thromboplastin times (aPTT), suggesting that aPTT might be used to monitor factor IXa inhibition with REG1 (Regado Biosciences, Basking Ridge, NJ).19,20

The REG1 anticoagulation system REG1 is a novel anticoagulation system made up of an RNA aptamer that binds to and selectively inactivates factor IXa (RB006, pegnivacogin) and its complementary controlling agent (RB007, anivamersen).21 Aptamers are oligonucleotides that bind to proteins based on their three-dimensional structure.22,23 Because of their nucleotide structure, they typically lack immunogenicity and toxicity, have tunable pharmacokinetics (PK), and can be formulated as either intravenous (IV) or subcutaneous injectables. Given their small size, they are frequently able to target protein interfaces that have proven difficult to pursue using larger peptides.

RB006

RB007

RB006-RB007 complex

100 h N30 h Factor IX Nucleases

b5 min Inert RB006 Nucleases

10 min Inert None Nucleases

Unique to aptamers is their inherent ability to code for their own reversal agents based on their oligonucleotide sequence and the high specificity and affinity afforded by Watson-Crick base pairing (Figure 1, Table I).20 This allows modulation of aptamer activity at any given time point depending on the clinical circumstances. RB006 (Figure 1) consists of a modified nucleotide aptamer attached to a polyethylene glycol chain to slow degradation and increase plasma half-life to approximately 100 hours with stable anticoagulant activity observed for 30 hours after single-dose administration (Table I).21 Extensive preclinical and early clinical modeling have demonstrated that RB006 results in reliable PK and pharmacodynamics (PD) based on the relationship between RB006 dosing, plasma concentrations, aPTT prolongation, and factor IX activity. These studies suggest that 0.75 to 1.0 mg/kg of RB006 results in consistent and near-complete (N99%) inhibition of factor IX in stable subjects and during elective percutaneous coronary intervention (PCI).21,24-26 RB007 has a half-life of b5 minutes and is rapidly cleared from the bloodstream by endogenous endonucleases. RB007 has no known biological effect other than rapid and irreversible binding to RB006 resulting in dissociation of RB006 from its factor IXa binding site and permanent reversal of anticoagulant effect.24,25 Administration of excess molar ratios of RB007 results in rapid (b10 minutes) sustained reversal of RB006 anticoagulant activity.25 Lower ratios of RB007-RB006 administration result in partial reversal of factor IX inhibition.

Clinical experience The REG1 program has 3 phase I studies assessing the PK and PD of the REG1 system in healthy volunteers and patients with coronary artery disease (CAD).21,24,25 A phase II pilot study demonstrated the feasibility of using RB006 as the sole anticoagulant therapy in patients undergoing elective PCI.26 The applicability of these findings to patients with ACS undergoing PCI remains unknown.

The RADAR trial RADAR (Clinicaltrials.gov identifier NCT00932100) is an international, multicenter, phase II, randomized, partially blinded, active control, clinical trial investigating


RB006 and a range of doses of RB007 in patients with ACS managed using an early invasive strategy. The purpose of this article was to describe the design and rationale for the RADAR trial. The authors are solely responsible for the design and conduct of this study, all study analyses, and the drafting and editing of the paper and its final contents.

Povsic et al 263

Figure 2

Objectives The main objectives of RADAR are (1) to determine the safety of a range of levels of RB006 reversal with RB007 after catheterization, (2) to confirm whether a dose of 1 mg/kg RB006 results in near-complete inhibition of factor IXa in patients with ACS, and (3) to assess the efficacy of RB006 as an anticoagulant in patients with ACS undergoing PCI. Secondary objectives include assessment of the safety of early post-PCI anticoagulation reversal and assessment of the feasibility of early sheath removal and patient ambulation after various degrees of RB006 reversal with RB007. Patient population A total of 800 patients will be recruited from approximately 80 sites worldwide with enrollment in Canada, France, Germany, the Netherlands, Poland, and the United States. Inclusion/exclusion criteria are listed in Figure 2. Patients are eligible if they are 18 to 80 years old and have experienced ischemic symptoms for ≥10 minutes within 72 hours of enrollment. Ischemic symptoms must be associated with (1) ST-segment depression or transient ST-segment elevation; (2) elevated troponin I, troponin T, or creatine kinase (CK)-MB; (3) documented CAD on prior angiography; or (4) a history of PCI or coronary artery bypass grafting (CABG).27 To be eligible, cardiac catheterization must be planned within 24 hours of randomization. Key exclusion criteria are acute ST-segment elevation myocardial infarction (MI), planned use of sheath sizes N7F during catheterization, and contraindications to anticoagulation including a history of intracranial bleeding or aneurysm and recent use of bivalirudin, glycoprotein (GP) IIb/IIIa inhibitors, or fibrinolytic agents. Patients receiving fondaparinux will be eligible 24 hours after their last dose of fondaparinux. Randomization and study drug Patients in RADAR will be randomized in a 2:1:1:2:2 fashion to REG1 with 25%, 50%, 75%, or 100% reversal or heparin and recommended GP IIb/IIIa inhibition (Figure 3). RADAR is designed as a partially blinded study with assignment to open-label REG1 or heparin and blinded RB007.

Inclusion and exclusion criteria.

REG1 arms RB006 dosing. Patients randomized to REG1 will receive 1 mg/kg of RB006 immediately after randomization (if no prior heparin administered) or as soon as the aPTT is b60 seconds (if prior heparin administered). Additional RB006 dosing. Given the data suggesting the consistent and high level of factor IX inhibition for up to 24 hours with 1 mg/kg of RB006, redosing is expected to be rare (Table II). Provisions have been made, however, to ensure the adequacy of anticoagulation. Local aPTTs will be drawn at 20 minutes and at 1, 4 (PK/PD substudy only), and 10 hours after RB006 administration, provided the patient has not undergone cardiac catheterization before these time points. Redosing will occur only if the patient is to undergo PCI; the most recent aPTT is b2 × upper limit of normal (ULN) of the local aPTT assay or b2 × baseline aPTT value; and no prior redosing has been performed. Redosing based on activated clotting time (ACT) will be limited to cases where aPTT data are not available and N4 hours have elapsed since RB006 administration.


264 Povsic et al

Figure 3

Study flow.

Table II A. Nomogram for RB006 redosing based on aPTT RB006 redose (mg/kg)

Relative aPTT value N2 1.81-2.0 1.61-1.8 1.41-1.6 1.21-1.4 1.0-1.2

No redose 0.2 0.4 0.6 0.6 0.8

B. Nomogram for RB006 redosing based on ACT 4-10 h post-RB006 administration (if aPTT not available) ACT (s) ≥180 b180

RB006 redose (mg/kg) No redose 0.3

N10 h post-RB006 administration (if aPTT not available) ACT (s) ≥180 b180

RB006 redose (mg/kg) No redose 0.4

Pharmacokinetics/PD substudy A key component of RADAR will be to verify the adequacy of 1 mg/kg of RB006 to achieve near-complete inhibition of factor IX activity in patients with ACS. A comprehensive PK/PD assessment will be performed in 20 patients who have not and 10 patients who have received heparin before randomization. Blood drawn

immediately before and 10 minutes after RB006 dosing and at the onset and completion of catheterization will be analyzed to ensure that RB006 concentrations are in the predicted range and that the corresponding effect on aPTT and factor IX inhibition is consistent with a high level of factor IX inhibition as was predicted from early phase studies in clinically stable patients. RB007 dosing. Patients randomized to REG1 will also be randomized to variable doses of RB007 designed to effect 25% (0.075 mg/kg, n = 200), 50% (0.2 mg/kg, n = 100), 75% (0.4 mg/kg, n = 100), or 100% (1 mg/kg, n = 200) reversal of RB006 activity (Figure 3). Upon completion of catheterization, a blinded dose of RB007 will be administered. If a patient receives RB006 but does not undergo catheterization, the blinded RB007 dose will be administered 24 hours after RB006 dosing or once the decision is made to not perform cardiac catheterization if the patient will not remain under observation for 24 hours. An aPTT will be obtained, and the sheath removed 10 minutes after administration of the randomized blinded RB007 dose. If hemostasis has not been achieved after 20 minutes or should uncontrollable bleeding occur, open-label 1 mg/kg RB007 may be administered to achieve 100% reversal of RB006.25,26

Heparin arm Patients randomized to heparin (n = 200) will be treated with enoxaparin or heparin, according to local practice, and dosed per a suggested nomogram (Table III). If a subject is to undergo PCI, unfractionated heparin will be


Povsic et al 265

Table III. Heparin-dosing nomogram

threatened closure, thrombus formation, distal embolization, or vessel dissection. To assess the effectiveness of RB006 to prevent thrombotic complications during PCI, the presence of intravascular thrombus on angiography, abrupt closure, threatened closure, side-branch closure, catheterassociate or intraprocedural thrombus formation, distal embolization, and no-reflow will be assessed in all patients.

Initial heparin dose Age

Male

Female

b60 60-85

14 U/kg 12 U/kg

12 U/kg 11 U/kg

Infusion alterations

aPTT

aPTT

b1.2 × ULN 1.2-1.5 × ULN 1.51-2.0 × ULN 2.01-2.4 × ULN 2.41-2.8 × ULN 2.81-4.0 × ULN N4.0 × ULN

b40 40-49 50-75 76-85 86-100 100-150 N150

Bolus (U)

Stop (min)

Rate (U/h) Δ 100 Δ 100

3000

30 60 60

Δ Δ Δ Δ

−100 −100 −200 −300

administered to achieve an ACT N200 seconds. Sheath removal, according to local practice, will be encouraged after documentation of an ACT b165 seconds. For patients treated with enoxaparin, a 1 mg/kg subcutaneous dose every 12 hours is recommended, with an additional 0.3 mg/kg IV dose to be administered if the most recent enoxaparin dose was ≥8 hours from the time of PCI.12 Sheath removal will occur 6 hours after the last subcutaneous dose or 4 hours after the last IV dose of enoxaparin, whichever is later. Concomitant medications. Aspirin is to be administered before or immediately after study enrollment. Thienopyridine therapy with loading doses (clopidogrel 600 mg or prasugrel 60 mg) is encouraged at or before study enrollment; however, if local practice stipulates later administration, alternative administration is permitted consistent with product labeling. Given the presentation of these patients with ACS, the use of GP IIb/IIIa inhibitors is recommended as the standard of care for patients undergoing PCI in the heparin group. Glycoprotein IIb/IIIa inhibitors may be started upon presentation or deferred until a decision is made regarding PCI according to local practice. All other concomitant medications are left to the discretion of the investigator.

Cardiac catheterization Cardiac catheterization and, if appropriate, PCI are to be performed within 24 hours of RB006 administration. If patients received prior heparin, cardiac catheterization will be delayed until N3 hours after REG1 administration to ensure that any heparin-related anticoagulant effect is gone by the end of the procedure. If PCI is indicated, provisional GP IIb/IIIa inhibitor use is permitted at the discretion of the operator in the event of vessel closure or

Sheath removal and bleeding assessments. Sheath removal procedures have been formalized to ensure standard bleeding assessments and processes across groups. If manual or mechanical compression is used to achieve hemostasis, bleeding assessments will be performed every 10 minutes, and the time to hemostasis recorded. Hemostasis achieved through the use of vascular closure devices (VCDs) will follow similar procedures, with deployment of the VCD and arterial closure 10 minutes after administration of RB007. Patients in whom VCDs are used will undergo identical assessment. If complete hemostasis fails after VCD deployment, manual compression will be performed for up to 20 minutes before any further interventions. Formal bleeding assessments will be performed and recorded for all treatment arms at 10 and 20 minutes and at 1, 2, 4, 8, 16, and 24 hours, provided the patient has not been discharged.

Primary and secondary outcomes The primary outcome is a composite of major and minor bleeding through 30 days using a modified ACUITY bleeding scale (online Appendix A).28 As most bleeding events in this study are likely to be access site related, the ACUITY bleeding scale was felt to be most appropriate. Secondary outcomes include ischemic events (death, nonfatal MI, recurrent ischemia in target vessel distribution, or urgent target lesion revascularization [TLR]) (online Appendix A), level of factor IXa inhibition as measured by effect on measures of coagulation, duration of hospitalization, feasibility of early sheath removal, and clinical and laboratory markers of anticoagulant activity. A blinded clinical events committee based at the Duke Clinical Research Institute will adjudicate all suspected bleeding events, MIs, urgent TLR, and ischemic events in the target vessel distribution using standard prespecified definitions. Clinical follow-up Patient assessments, including history, physical exam, concomitant medication assessment, and safety laboratories will occur at discharge and at 30-day postrandomization. To ensure capture of any bleeding occurring after hospital discharge, patients will be contacted by telephone at 7 days.


266 Povsic et al

Data analysis and statistical assumptions As a phase II, dose-finding study, RADAR is designed to test several hypotheses. All analyses will be considered significant at a nominal α level of .05 with no adjustments for multiple comparisons. With respect to the primary end point of major or minor bleeding, with a sample size of 200 patients per arm and assuming a 20% event rate in the heparin arm, RADAR has 80% power to detect a 10% absolute reduction with RB006 with 100% reversal compared with heparin.29 Similarly, assuming a 20% event rate in the RB006 with 25% reversal arm, RADAR has 80% power to detect a 10% absolute reduction with RB006 with 100% reversal compared with 25% reversal. To assess the effect of RB006 on ischemic end points, the study will compare the 600 patients randomized to RB006 with the 200 patients randomized to heparin. Formal noninferiority testing is not feasible; however, assuming an 8% event rate in the heparin arm, RADAR will be able to exclude an absolute increase of N6.2% in the REG1 group with 80% power and an increase of 7.2% with 90% power. Adaptive design An independent Data Safety Monitoring Board (DSMB) will meet after enrollment of 100, 200, and 400 patients to assess bleeding rates. They may recommend that the Steering Committee discontinue individual reversal arms if excess bleeding is observed in the low reversal arms that is consistent with a dose-dependent effect of RB007. Major bleeding rates will be compared with the historical rate of approximately 5% observed in ACUITY. If bleeding rates in the lowest reversal arm exceed this rate with 95% confidence, the DSMB may recommend discontinuation of that arm. Under these guidelines, termination of individual arms may be considered if • ≥4 events in 25 patients (rate 16%, lower 95% CI 5.02%); • ≥7 events in 50 patients (rate 14%, lower 95% CI 6.15%); • ≥9 events in 75 patients (rate 12%, lower 95% CI 5.9%); • ≥10 events in 100 patients (rate 10%, lower 95% CI 5.12%). Should open-label 100% RB007 reversal be used at a rate of ≥25% that is consistent with a dose-dependent effect of RB007, a recommendation to discontinue a reversal arm may also be considered. If an RB007 reversal arm is dropped, additional patients will be allocated to the remaining REG1 arms.

Committees An academic steering committee (online Appendix B) composed of international representation from the

academic interventional community will provide scientific direction and assess trial progress. The steering committee is responsible for trial design and publication of major trial findings. The committee will also encourage and prioritize additional analyses for publication suggested by investigators. The DSMB (online Appendix B) will consist of 2 interventional cardiologists, an expert in coagulation, a statistician, and a nonvoting member.

Discussion RADAR is the first study to address the clinical effectiveness of a drug-controlling agent combination and offers unique challenges in design and interpretation. RADAR is designed to assess the REG1 system in patients with ACS by addressing 3 critical issues. First, to determine the safety of a range of levels of RB006 reversal with RB007 after catheterization with respect to sheath removal. Second, to confirm whether a dose of 1 mg/kg RB006 results in near-complete inhibition of factor IXa based on pharmacodynamic markers in patients with ACS. Third, to preliminarily assess the efficacy of RB006 as an anticoagulant in patients with ACS undergoing PCI. Understanding these issues will be critical to the design of subsequent phase III clinical trials of REG1 in patients with the spectrum of ACS and undergoing PCI.

Drug reversal: advantages and complexities A unique feature of REG1 is that RB006 can be actively, partially, or completely reversed using RB007. One approach to achieving a desirable balance between the prevention of ischemic events and avoiding bleeding is to use a therapy that is actively reversible, allowing modulation of the level of anticoagulation to fit the individual patient's clinical circumstances. Aptamers are unique in their ability to code for their own controlling agents, which allows for active modulation of their activity.21,24,25 Assessing the efficacy and safety of a combination product is challenging because dosing of each component must be independently evaluated. Phase I studies of REG1 established the relationship between weightbased and fixed RB006 doses and degree of factor IX inhibition, as well as the ability of RB007 to completely reverse the anticoagulant effects of RB006 in stable volunteers and patients with CAD 21,24 ; phase Ic determined the relationship between RB007 dose and the degree of RB006 reversal.25 RB006 dosing in ACS The dose of RB006 being studied in RADAR is based on preclinical and early clinical data.20,21,24,25 The dose of 1 mg/kg that was chosen is expected to achieve consistent and near-complete factor IX inhibition. Given that RB006


is a selective inhibitor of factor IX, this dose is expected to be maximally or near maximally effective. Investigation of a high dose of RB006 is facilitated by the availability of RB007-mediated reversal should clinical bleeding occur and, at least in part, mitigates the need for extensive dose ranging for RB006. The PK/PD substudy will confirm that the dose of RB006 investigated in RADAR achieves consistent and near-complete inhibition of factor IX in patients with ACS. Finally, RADAR will begin to assess the efficacy of factor IX inhibition in patients with ACS undergoing PCI by assessing rates of ischemic events both during PCI and after RB007 reversal in patients treated with RB006. Event rates with RB006 will be compared with those of patients treated with heparin with or without a GP IIb/IIIa inhibitor in RADAR and from historical data.

RB007 reversal A key objective in RADAR is to determine the range of RB006 reversal with RB007 that is clinically feasible after cardiac catheterization. Unlike other anticoagulants, REG1 offers the possibility of tailoring the anticoagulation to the clinical circumstances of the patient. Because REG1 allows precise modulation of the level of antithrombotic therapy during the course of treatment, RADAR will be the first study to explore the consequence of variable levels of anticoagulation in the postcatheterization/PCI period. RADAR has statistical power to detect a 10% absolute difference in major or minor bleeding; an assessment of bleeding rates across all reversal arms will inform the minimal RB007 dose that will allow early sheath removal without excessive bleeding. Because patients with hemophilia B exhibit clinically evident bleeding only after losing N90% of factor IX activity, it is apparent that even partial RB006 reversal may permit arterial sheath removal. As would happen in practice, if patients have residual or recurrent bleeding, an additional dose of RB007 sufficient to fully reverse RB006 can be given. Blinding of RB006 study drug and study monitoring Given the different modes of administration, the timing of aPTT monitoring, and the timing of reversal for REG1, it is logistically challenging to fully blind patients and study personnel to treatment assignment. Therefore, a partially blinded design was chosen with open-label assignment to REG1 versus heparin and blinded randomization to the degree of reversal in REG1-treated patients. Required femoral access We required femoral access in RADAR to reduce the heterogeneity in groin management and procedurerelated bleeding. Both femoral closure and compression devices are allowed during the study. Although VCDs have been shown to reduce time to ambulation, registry

Povsic et al 267

data and meta-analyses of randomized studies are inconsistent with respect to demonstrating reductions in bleeding complications with VCDs as compared with manual compression.30-36

Conclusions RADAR is a unique phase II clinical trial investigating a combined anticoagulant thrombotic and complementary active reversal agent. The data from RADAR will answer important questions about each component of the REG1 anticoagulation system and define the best strategy to support the development of adequately powered phase III clinical trials.

Disclosures Funding: RADAR is funded by Regado Biosciences (Basking Ridge, NJ).

References 1. Harrington RA, Becker RC, Cannon CP, et al. Antithrombotic therapy for non–ST-segment elevation acute coronary syndromes: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2008;133:670S-707S. 2. Stone GW, Witzenbichler B, Guagliumi G, et al. Bivalirudin during primary PCI in acute myocardial infarction. N Engl J Med 2008;358: 2218-30. 3. Rao SV, O'Grady K, Pieper KS, et al. Impact of bleeding severity on clinical outcomes among patients with acute coronary syndromes. Am J Cardiol 2005;96:1200-6. 4. Rao SV, O'Grady K, Pieper KS, et al. A comparison of the clinical impact of bleeding measured by two different classifications among patients with acute coronary syndromes. J Am Coll Cardiol 2006;47: 809-16. 5. Rao SV, Kaul PR, Liao L, et al. Association between bleeding, blood transfusion, and costs among patients with non–ST-segment elevation acute coronary syndromes. Am Heart J 2008;155:369-74. 6. Rao SV, Jollis JG, Harrington RA, et al. Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA 2004;292:1555-62. 7. Hirsh J, O'Donnell M, Eikelboom JW. Beyond unfractionated heparin and warfarin: current and future advances. Circulation 2007;116: 552-60. 8. Kimmel S, Sekeres M, Berlin J, et al. Adverse events after protamine administration in patients undergoing cardiopulmonary bypass: risks and predictors of under-reporting. J Clin Epidemiol 1998;51: 1-10. 9. Seifert H, Jobes D, Ten Have T, et al. Adverse events after protamine administration following cardiopulmonary bypass in infants and children. Anesth Analg 2003;97:383-9. 10. Weiler JM, Gellhaus MA, Carter JG, et al. A prospective study of the risk of an immediate adverse reaction to protamine sulfate during cardiopulmonary bypass surgery. J Allergy Clin Immunol 1990;85: 713-9. 11. Collett JP, Montalescot G, Lison L, et al. Percutaneous coronary intervention after subcutaneous enoxaparin pretreatment in patients with unstable angina pectoris. Circulation 2001;103:658-63. 12. Ferguson JJ, Califf RM, Antman EM, et al. Enoxaparin vs unfractionated heparin in high-risk patients with non–ST-segment

268 Povsic et al

13.

14.

15.

16. 17.

18.

19.

20. 21.

22. 23. 24.

elevation acute coronary syndromes managed with an intended early invasive strategy: primary results of the SYNERGY randomized trial. JAMA 2004;292:45-54. Stone GW, Ware JH, Bertrand ME, et al. Antithrombotic strategies in patients with acute coronary syndromes undergoing early invasive management: one-year results from the ACUITY trial. JAMA 2007; 298:2497-506. Fifth Organization to Assess Strategies in Acute Ischemic Syndromes Investigators. Comparison of fondaparinux and enoxaparin in acute coronary syndromes. N Engl J Med 2006;354:1464-76. Kjalke M, Monroe DM, Hoffman M, et al. Active site-inactivated factors VIIa, Xa, and IXa inhibit individual steps in a cell-based model of tissue factor-initiated coagulation. Thromb Haemost 1998;80: 578-84. Mann K, Brummel K, Butenas S. What is all that thrombin for? J Thromb Haemost 2003;1:1504-14. Monroe DM, Hoffman M, Roberts HR. Transmission of a procoagulant signal from tissue factor-bearing cells to platelets. Blood Coagul Fibrinolysis 1996;7:459-64. Howard EL, Becker KC, Rusconi CP, et al. Factor IXa inhibitors as novel anticoagulants. Arterioscler Thromb Vasc Biol 2007;27: 722-7. Zelenkofske SL, Rusconi CP, Damiento CM, et al. Subcutaneous RB006, a direct FIXa inhibitor, provides prolonged anticoagulation with rapid reversal: the first clinical experience with the REG2 system. Poster presented at: Arteriosclerosis, Thrombosis and Vascular Biology 2010 Scientific Session Sessions; April 9, 2010; San Francisco, CA; 2010. Rusconi CP, Scardino E, Layzer J, et al. RNA aptamers as reversible antagonists of coagulation factor IXa. Nature 2002;419:90-4. Dyke CK, Steinhubl SR, Kleiman NS, et al. First-in-human experience of an antidote-controlled anticoagulant using RNA aptamer technology: a phase 1a pharmacodynamic evaluation of a drug-antidote pair for the controlled regulation of factor IXa activity. Circulation 2006;114:2490-7. Gold L, Polisky B, Uhlenbeck O, et al. Diversity of oligonucleotide functions. Ann Rev Biochem 1995;64:763-97. Nimjee SM, Rusconi CP, Sullenger BA. APTAMERS An emerging class of therapeutics. Ann Rev Med 2005;56:555-83. Chan MY, Cohen MG, Dyke CK, et al. Phase 1b randomized study of antidote-controlled modulation of factor IXa activity in patients with stable coronary artery disease. Circulation 2008;117:2865-74.


25. Chan MY, Rusconi CP, Alexander JH, et al. A randomized, repeatdose, pharmacodynamic and safety study of an antidote-controlled factor IXa inhibitor. J Thromb Haemost 2008;6:789-96. 26. Cohen MG, Purdy DA, Rossi JS, et al. First clinical application of an actively reversible direct factor IXa inhibitor as an anticoagulation strategy in patients undergoing percutaneous coronary intervention. Circulation 2010;122:614-22. 27. Thygesen K, Alpert JS, White HD. Universal definition of myocardial infarction. J Am Coll Cardiol 2007;50:2173-95. 28. Stone GW, Bertrand M, Colombo A, et al. Acute Catheterization and Urgent Intervention Triage strategY (ACUITY) trial: study design and rationale. Am Heart J 2004;148:764-75. 29. Lincoff AM, Kleiman NS, Kereiakes DJ, et al. Long-term efficacy of bivalirudin and provisional glycoprotein IIb/IIIa blockade vs heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary revascularization: REPLACE-2 randomized trial. JAMA 2004;292:696-703. 30. Biancari F, D'Andrea V, Di Marco C, et al. Meta-analysis of randomized trials on the efficacy of vascular closure devices after diagnostic angiography and angioplasty. Am Heart J 2010;159: 518-31. 31. Vaitkus PT. A meta-analysis of percutaneous vascular closure devices after diagnostic catheterization and percutaneous coronary intervention. J Invasive Cardiol 2004;16:243-6. 32. Koreny M, Riedmüller E, Nikfardjam M, et al. Arterial puncture closing devices compared with standard manual compression after cardiac catheterization: systematic review and meta-analysis. JAMA 2004;291:350-7. 33. Sanborn TA, Ebrahimi R, Manoukian SV, et al. Impact of femoral vascular closure devices and antithrombotic therapy on access site bleeding in acute coronary syndromes: The ACUITY Trial. Circ Cardiovasc Interv 2010;3:57-62. 34. Marso JP, Amin AP, House JA, et al. Association between use of bleeding avoidance strategies and risk of periprocedural bleeding among patients undergoing percutaneous coronary intervention. JAMA 2010;303:2156-64. 35. Dauerman HL, Applegate RJ, Cohen DJ. Vascular closure devices: the second decade. J Am Coll Cardiol 2007;50:1617-26. 36. Arora N, Matheny ME, Sepke C, et al. A propensity analysis of the risk of vascular complications after cardiac catheterization procedures with the use of vascular closure devices. Am Heart J 2007;153:606-11.


Appendix A. End point definitions Bleeding Major bleeding is defined as intracranial, intraocular, or retroperitoneal hemorrhage; access site hemorrhage requiring radiologic or surgical intervention; 5-cm diameter hematoma at puncture site; clinically overt blood loss resulting in a decrease in hemoglobin concentration of N3 g/dL; any decrease in hemoglobin concentration of N4 g/ dL; reoperation for bleeding; use of blood product transfusion; and hemarthrosis. Minor bleeding is defined as clinically overt bleeding that does not meet criteria for major bleeding. Ischemic end points Ischemic events will be defined as all deaths, regardless of cause, nonfatal MI, urgent TLR, or recurrent ischemia in the target vessel distribution. These are defined below. Myocardial infarction. Myocardial infarction will be defined based on occurrence in the setting of recent revascularization (CABG or PCI) and its relation to recent cardiac markers. (A) No recent MI, no recent revascularization in the previous 24 hours: any elevation of troponin I or T above the ULN, OR any elevation of CK-MB above the ULN or total CK above 2 × ULN (if no troponin data available), or new significant (N0.04 seconds) Q waves in ≥2 contiguous leads, or clinical evidence of ischemia (N20 minutes) with new or recurrent ST-segment elevation of N0.1 mV in 2 contiguous limb leads or N0.2 mV in 2 contiguous precordial leads prompting urgent cardiac catheterization AND resulting in documented poor flow (Thrombolysis In Myocardial Infarction b3) in the infarct-related artery prompting urgent intervention (PCI or CABG). (B) Baseline MI present, but no recent revascularization (within 24 hours): clinical evidence of recurrent ischemic pain (N20 minutes) with new or recurrent ST-segment elevation N0.1 mV in 2 contiguous limb leads or N0.2 mV in 2 contiguous precordial leads prompting urgent cardiac catheterization AND resulting in documentation poor flow (TIMI b3) in the infarct-related artery prompting urgent intervention (PCI or CABG), or new significant (≥ 0.04 seconds) Q waves in ≥ 2 contiguous leads and discrete from enrollment MI, or cardiac markers of necrosis or electrocardiographic (ECG) evidence as follows: i. re-elevation of troponin I or T NULN (if prior level was normal); or ii. re-elevation of troponin I or T NULN and N50% above the nadir prior level (if prior level was above normal); or iii. re-elevation of CK-MB to ≥ULN (if prior level was normal); or iv. CK-MB ≥ ULN and N50% above the prior nadir level (if prior level was above normal); or

Povsic et al 268.e1

v. total CK ≥ 2 × ULN and increased by ≥50% over the previous nadir value (if CK-MB is unavailable). (C) No recent MI prior and recent revascularization: for patients undergoing PCI, CK-MB (or total CK, if CK-MB is unavailable) ≥ 3 × ULN. In the absence of CK-MB and total CK data, the troponin value must exceed N3 × ULN or new significant (N0.04 seconds) Q waves in 2 contiguous ECG leads. For patients undergoing CABG, CK-MB (or total CK, if CK-MB is unavailable) ≥ 5 × ULN. In the absence of CKMB and total CK data, the troponin value must exceed N5 × ULN or new significant (N0.04 seconds) Q waves in 2 contiguous ECG leads. (D) For patients in whom the cardiac markers of necrosis are elevated before PCI, an end point MI is defined as follows: CK-MB (or total CK, if CK-MB is unavailable) ≥ 3 × ULN and increased by at least 20% from the level before the procedure or new significant (N0.04 seconds) Q waves in 2 contiguous ECG leads. In the absence of CK-MB and total CK data, troponin value must exceed 3 × ULN and represent a N50% increase in the nadir preprocedural troponin on 2 consecutive assessments. For patients in whom the cardiac markers of necrosis are elevated before CABG, an end point MI is defined as follows: CK-MB (or total CK, if CK-MB is unavailable) ≥ 5 × ULN and increased by at least 20% from level before the procedure. In the absence of CK-MB and total CK data, the troponin value must exceed 5 × ULN and represent a N50% increase in the nadir preprocedural troponin on 2 consecutive assessments or new significant (N0.04 seconds) Q waves in 2 contiguous ECG leads. Urgent revascularization and urgent TLR. Any revascularization procedure (PCI or CABG) performed because of a clinical event that occurs after randomization will constitute an urgent revascularization. If urgent revascularization occurs at a previously treated site (PCI or CABG) as part of RADAR, this will constitute a TLR. Recurrent myocardial ischemia in target vessel distribution. Recurrent ischemia in target vessel distribution will include any symptoms of chest discomfort or equivalent ischemic symptoms of ≥10 minutes' duration not fulfilling criteria for an MI but prompting medical intervention (eg, thrombolytics, GP IIb/IIIa inhibitors, anticoagulation, prolonged hospitalization, or additional medical procedures, etc) and with clear documentation that the ischemia originates from the target or culprit vessel(s). The culprit lesion(s) should be identified by imaging studies (eg, cardiac catheterization, radionucleotide study, echocardiographic wall motion study, etc). This end point is designed to capture recurrent events in the target vessel requiring additional medical or procedural intervention (ie, intracoronary lytics and additional systemic GP IIb/IIIa inhibition) but not

268.e2 Povsic et al

resulting in MI or revascularization caused by ischemic or thrombotic complications.


Appendix B. Committee members

MD, Duke Clinical Research Institute; Jan Cornel, MD, Medisch Centrum Alkmaar; Gilles Montalescot, MD, PitiéSalpétrière Hospital; Jaroslav Kasprzak, MD, Medical University of Lodz; Steven Zelenkofske, MD, Regado Biosciences.

Academic Steering Committee John H. Alexander, MD (chair), Duke Clinical Research Institute; Roxana Mehran, MD, Mount Sinai Medical Center; Christoph Bode, MD, University of Freiberg; Mauricio G. Cohen, MD, University of Miami; Christopher Buller, MD, Hamilton General Hospital; Richard Becker, MD, Duke Clinical Research Institute; Thomas Povsic,

Data Safety Monitoring Board Ronald Waksman, MD, Washington Heart Center (chair); Stefan James, MD, Uppsala Clinical Research Center; Jack Ansel, MD, Lenox Hill Hospital; Vic Hasselblad, PhD, Duke Clinical Research Institute; and Rebecca Torguson, MPH, Washington Heart Center.

Associations between cardiovascular parameters and uteroplacental Doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the Zwangerschap bij Aangeboren Hartafwijking (ZAHARA) II study Ali Balci, MD, MSc, a,j,k Krystyna M. Sollie, MD, b,k Barbara J. M. Mulder, MD, PhD, d,k Monique W. M. de Laat, MD, PhD, e Jolien W. Roos-Hesselink, MD, PhD, f Arie P. J. van Dijk, MD, PhD, g Elly M. C. J. Wajon, MD, h Hubert W. Vliegen, MD, PhD, i Willem Drenthen, MD, PhD, a Hans L. Hillege, MD, PhD, a,c Jan G. Aarnoudse, MD, PhD, b Dirk J. van Veldhuisen, MD, PhD, a and Petronella G. Pieper, MD, PhD a Groningen, Amsterdam, Rotterdam, Nijmegen, Enschede, Leiden, and Utrecht, The Netherlands

Background Previous research has shown that women with congenital heart disease (CHD) are more susceptible to cardiovascular, obstetric, and offspring events. The causative pathophysiologic mechanisms are incompletely understood. Inadequate uteroplacental circulation is an important denominator in adverse obstetric events and offspring outcome. The relation between cardiac function and uteroplacental perfusion has not been investigated in women with CHD. Moreover, the effects of physiologic changes on pregnancy-related events are unknown. In addition, long-term effects of pregnancy on cardiac function and exercise capacity are scarce. Methods Zwangerschap bij Aangeboren Hartafwijking (ZAHARA) II, a prospective multicenter cohort study, investigates changes in and relations between cardiovascular parameters and uteroplacental Doppler flow patterns during pregnancy in women with CHD compared to matched healthy controls. The relation between cardiovascular parameters and uteroplacental Doppler flow patterns and the occurrence of cardiac, obstetric, and offspring events will be investigated. At 20 and 32 weeks of gestation, clinical, neurohumoral, and echocardiographic evaluation and fetal growth together with Doppler flow measurements in fetal and maternal circulation are performed. Maternal evaluation is repeated 1 year postpartum. Implications By identifying the factors responsible for pregnancy-related events in women with CHD, risk stratification can be refined, which may lead to better pre-pregnancy counseling and eventually improve treatment of these women. (Am Heart J 2011;161:269-275.e1.)

Because of improved long-term survival, most women with congenital heart disease (CHD) reach child-bearing age and many pursue pregnancy. In women with uncorrected maternal congenital heart defects or with

residual sequelae after correction, the hemodynamic changes in pregnancy can have negative effects on the health of both mother and her (unborn) child. Cardiac events are rare in healthy women (b1%), while arrhythmias occur in 4.5% and heart failure in 4.8% of women with CHD.1 In complex CHD, cardiac event

From the aDepartment of Cardiology, University Medical Centre Groningen, University of Groningen, Groningen, The Netherlands, bDepartment of Obstetrics, University Medical Centre Groningen, University of Groningen, Groningen, The Netherlands, cDepartment of Epidemiology, University Medical Centre Groningen, University of Groningen, Groningen, The Netherlands, dDepartment of Cardiology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands, eDepartment of Obstetrics, Academic Medical

k On behalf of the ZAHARA-II investigators. See the online Appendix for complete listing. This study is supported by a grant from the Netherlands Heart Foundation (NHF) (2007B75). DJvV is clinically established investigator of the NHF (D97-017). The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper and its final contents. Submitted August 26, 2010; accepted October 18, 2010.

Centre, University of Amsterdam, Amsterdam, The Netherlands, fDepartment of Cardiology, Erasmus Medical Centre, Erasmus University, Rotterdam, The Netherlands, g Department of Cardiology, Radboud University Nijmegen Medical Centre, Radboud University Nijmegen, Nijmegen, The Netherlands, hDepartment of Cardiology, Medical Spectrum Twente, Enschede, The Netherlands, iDepartment of Cardiology, Leiden University Medical Centre, University of Leiden, Leiden, the Netherlands, and jInteruniversity Cardiology Institute of the Netherlands (ICIN), Utrecht, The Netherlands.

Reprint requests: Petronella G. Pieper, MD, PhD, Department of Cardiology, University Medical Centre Groningen, University of Groningen, PO Box 30.001, 9700 RB, Groningen, The Netherlands. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.10.024

Background


270 Balci et al

rate can be even higher.1-4 Women with CHD are not only more susceptible to cardiac events, but obstetric and offspring events are also more prevalent.1-3 Important obstetric events are postpartum hemorrhage (8%, up to 29%), pregnancy-induced hypertension (5.5%, up to 13%), preeclampsia (PE; 3.2%, up to 10%), and preterm delivery (16%, up to 65%); whereas in healthy pregnant women, the prevalence of these events is much lower.1 Frequently observed events in offspring of women with CHD are intrauterine growth restriction (IUGR), prematurity, and mortality.1-3 The magnitude of these risks depends, at least in part, on the type and severity of maternal CHD. In the general population, both IUGR and PE are associated with a lower cardiac output (CO), an elevated total vascular resistance (TVR), and abnormal uterine and umbilical artery Doppler waveform patterns.5,6 These parameters can be used to identify women at risk for PE and IUGR. The higher incidence of these events in women with CHD may be caused by inadequate maternal hemodynamics, resulting in insufficient uteroplacental circulation.7,8 Nevertheless, the interaction between echocardiographic, hemodynamic, and neurohumoral parameters on one side and uteroplacental Doppler flow patterns on the other side in relation to pregnancy outcome have not been studied in women with CHD and have not been compared to those in healthy pregnant women. In addition, although some studies identified predictors of cardiac events in women with both acquired and CHD, prospective data relating echocardiographic parameters of ventricular or valvular function to maternal cardiac events are still scarce and little is known about the long-term effects of pregnancy on cardiac function or exercise capacity in women with CHD.2,3,9 The mid- and longterm effects of pregnancy on cardiovascular hemodynamics have been described in a few select subgroups. 10,11 In these studies, however, exercise capacity or cardiac biomarkers were not assessed. In this article, we will introduce the study design and describe the rationale of the ZAHARA II study.

Methods Study objectives The primary objective of the present study is to compare cardiovascular, neurohumoral, and uteroplacental Doppler flow changes during pregnancies of women with CHD with age- and parity-matched healthy controls and to relate these changes to the occurrence of cardiovascular and obstetric events and to offspring outcome. The secondary objective of this study is to evaluate the incidence of permanent postpartum cardiovascular deterioration in women with CHD.

Study design This is an observational prospective multicenter cohort study.

Table I. Inclusion and exclusion criteria of ZAHARA II Women with CHD Inclusion criteria - Age ≥18 y - Morphological CHD - Presentation at ≤20 wk of gestation - Presentation in one of the participating medical centers Exclusion criteria - Miscarriage or termination of pregnancy b20 wk of gestation - Alcohol abuse - Illicit drugs use Healthy controls Inclusion criteria - Age ≥18 y - Presentation at ≤20 wk of gestation Exclusion criteria - Miscarriage or termination of pregnancy b20 wk of gestation - Women who are on chronic medication - Women who are under specialist control - Alcohol abuse - Illicit drugs use

Study population Women with any morphological CHD with a pregnancy of b20 weeks duration, presenting in the participating centers, who meet all the inclusion and none of the exclusion criteria are eligible (Table I). During a 3-year period, a minimum of 160 women with CHD are enrolled, and simultaneously, 60 healthy, age- and parity-matched women are recruited from a low-risk midwife practice in Groningen and in Rotterdam, the Netherlands, to serve as controls. We will use a subclassification for our cohort where appropriate, using division in disease complexity, the adapted World Health Organization classification for estimating pregnancy risk, or clustering of the morphological and functional comparable diseases as in previous studies.2,9,12,13

Measurements Baseline data are recorded at the first prenatal visit using medical records and include underlying heart disease, prior interventions, cardiac sequelae, prior cardiac events, comorbidity, and obstetric history. Maternal age, parity, present cardiac status (including New York Heart Association functional class, physical examination, oxygen saturation, and echocardiographic data), use of medication, intoxications, educational status, and current employment are also recorded. Clinical evaluation at gestational weeks 20 and 32 as well as at 1 year postpartum is performed for follow-up data and for registration of events. During follow-up echocardiograms, electrocardiograms and 24hour electrocardiographic registrations as well as obstetric evaluation and blood and urinalysis are conducted.

Echocardiography Standardized echocardiograms according to disease-specific protocols are performed at 20 and 32 weeks of gestation and at 1 year postpartum. Echocardiograms are evaluated off-line in the University Medical Center Groningen, Groningen, the Netherlands. Morphological left and right ventricular size and function (if feasible, ejection fraction according to Simpson's


Balci et al 271

Table II. Adverse events as defined in ZAHARA II Primary cardiovascular events during pregnancy, up to 6 m postpartum Need for urgent invasive cardiovascular procedures Heart failure: according to the guidelines of the European Society of Cardiology and documented by the attending physician40 Any documented new-onset or symptomatic tachy- or bradyarrhythmia requiring new/extended treatment Thromboembolic events: deep vein thrombosis, pulmonary embolism, intracardiac thrombosis, arterial thrombosis, systemic arterial embolisms, or transient ischemic attack Myocardial infarction Cardiac arrest Cardiac death Endocarditis: according to the Duke criteria41 Aortic dissection Secondary cardiovascular events during pregnancy, up to 6 m postpartum NYHA class deterioration: decline of 2 points in NYHA functional class during pregnancy or within 6 m postpartum, compared with pre-pregnancy NYHA class or a persisting deterioration in NYHA functional class postpartum Primary obstetric events Assisted delivery: use of forceps, use of a vacuum extractor, or the performance of a cesarean section for delivery on maternal cardiac and medical indication, on maternal obstetric indication, or on fetal indication PIH: systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg or an increase of systolic (N30 mm Hg) or diastolic (N15 mm Hg) blood pressure, in the absence of proteinuria, occurring after ≥20 wk of gestation PE: PIH with ≥0.3 g/24 h proteinuria Eclampsia: PE with grand mal seizures Mild gestational diabetes mellitus: a fasting blood glucose b140 mg/dL (7.8 mmol/L) and 2-h postprandial 140-198 mg/dL (7.8-11 mmol/L) Severe gestational diabetes mellitus: a random serum glucose value N200 mg/dL (11.1 mmol/L) or a fasting blood glucose value N126 mg/dL (7.0 mmol/L) HELLP: hemolysis (LDH N 250 U/L), elevated liver enzymes (ASAT N40 U/L and ALAT N45 U/L), low platelets (b1.0 × 10.6/mm3) syndrome Hyperemesis gravidarum: severe, intractable nausea and vomiting, leading to dehydration, loss of weight, metabolic disorders, and hospitalization Noncardiac death: all cause mortality, except cardiac mortality Postpartum hemorrhage: blood loss N500 mL (vaginal delivery) or N1000 mL (cesarean section), requiring transfusion or leading to a drop in hemoglobin N20 g/L (1.24 mmol/L) Premature labor: spontaneous onset of labor b37 wk of gestation Preterm premature rupture of membranes: spontaneous rupture of membrane before the onset of uterine contractions and before 37 wk of gestation Abruptio placentae: premature detachment of the placenta from the wall of the uterus Secondary obstetric events Amniotomy: mechanical/artificial rupture of membranes Induction of labor Prolongation of cervix ripening: omitted dilatation of the portio vaginalis during ≥20 h (nullipara) or ≥14 h (multipara), despite adequate and regular uterine contractions Prolongation of second stage of delivery (primipara N2 h; multipara N1 h) Placenta previa: localization of the placenta partially or completely above the internal ostium of the cervix Offspring events Fetal death: intrauterine death N20 wk of gestation Extended perinatal death: the number of stillbirths from 20 wk of gestation and neonatal death up to 28 d postpartum Early neonatal death: within 6 d after birth Late neonatal death: within 7-28 days after birth Infant death: N28 d and within 1 y after birth Perinatal death: the total number of stillbirths from 20 wk of gestation and death up to 7 days postpartum Offspring death: the total number of stillbirths from 20 wk of gestation and death up to 1 y postpartum Intraventricular hemorrhage: bleeding in the fetal cerebral ventricles Neonatal respiratory distress syndrome: respiratory insufficiency caused by a developmental insufficiency of surfactant production and structural immaturity in the lungs in premature infants Infections leading to hospital admission Premature birth: birth b37 wk gestation Occurrence of CHD Occurrence of other congenital disease Small for gestational age: birth weight below the 10th percentile adjusted for gestational age and based on population values Low birth weight: birth weight b2500 g Meconium stained amniotic fluid General events Anemia: between 18 wk of gestation until 1 wk postpartum: 6.5 mmol/L or 10.5 g/dL42 Hospitalization: all cause hospitalization for more than 1 night Fever: ≥38.5°C during pregnancy up to 6 m postpartum requiring medical treatment Infection: infections during pregnancy up to 6 m postpartum requiring medical treatment ALAT, Alanine aminotransferase; ASAT, aspartate aminotransferase; LDH, lactate dehydrogenase; NYHA, New York Heart Association; PIH, pregnancy-induced hypertension.


272 Balci et al

rule), atrial size, valvular function (quantification of regurgitation and stenosis of all valves) as well as disease-specific evaluation (ie, presence and location of intracardiac shunts, evaluation of conduits, or baffles) are performed according to current recommendations and guidelines.14-20 Blood pressure and heart rate are measured during echocardiography to calculate CO and TVR as described previously.15,17,19 Prepregnancy routinely performed echocardiograms (available in most of the patients) are analyzed for comparison.

Obstetric evaluation Fetal biometry is assessed by ultrasound and uteroplacental perfusion is studied by Doppler flow measurements of uterine arteries. Umbilical artery pulsatility index, resistance index, and uterine artery flow and early diastolic notching are evaluated at 20 and 32 weeks of pregnancy according to the guidelines of the International Perinatal Doppler Society.5,21 Evaluation of the digitally stored (Doppler) ultrasound registrations is performed in the University Medical Center Groningen.

Blood and urinalysis Hematologic parameters, renal and hepatic function, hemoglobin A1c as well as nonfasting glucose, and N-terminal prohormone brain natriuretic peptide (NT-proBNP) are assessed from blood samples at 20 and 32 weeks of gestation and 1 year postpartum and compared with the values before pregnancy, if available. Proteinuria is quantitatively assessed at 20 and 32 weeks of gestation.

Cardiopulmonary aerobic capacity testing Cardiopulmonary aerobic capacity testing is performed 1 year postpartum in patients who underwent this test b2 years before pregnancy.22

Adverse events Clinical adverse events in ZAHARA II are subdivided in cardiac, obstetric, offspring, and general adverse events (Table II). We define cardiovascular events as described in previous studies (Table II). In addition, we assess changes in NTpro-BNP levels and in echocardiographic findings in patients with CHD compared to those in healthy controls. We define abnormal echocardiographic changes in pregnancies as a significant deterioration in size or function of subpulmonary or sub-aortic ventricle; new onset or aggravation of valve regurgitation ≥1 grade (mild to moderate or severe, or moderate to severe) during pregnancy and/or persisting 1 year postpartum; persistent (≥1 year) significant aggravation of valve stenosis (mild to moderate or severe, or moderate to severe); significant increase in aortic dimensions (≥5 mm) during pregnancy and/or 1 year postpartum. Furthermore, pulsatility index N95th centile in umbilical artery, uterine artery resistance index N95th centile, umbilical artery resistance index N0.58 or N90th centile, or early diastolic notching in uterine artery is considered abnormal in obstetric ultrasound evaluation.23 Finally, a significant deterioration in functional capacity and/or exercise capacity 1 year postpartum in all patients compared with pre-pregnancy values is considered abnormal.

Statistical and ethical considerations Sample size calculation. One of the primary aims of the ZAHARA II study is to compare the uteroplacental Doppler flow, expressed as pulsatility index in the umbilical artery, during pregnancy between women with CHD and healthy controls. A sample size of 160 patients and 60 controls achieves a power of 80% at a significance level of .05 to detect a difference of 5% higher mean pulsatility index in women with CHD. Statistical analysis. Continuous variables with normal distribution will be presented as mean (±SD), nonnormally distributed variables as median (with 25th and 75th percentile), and dichotomous variables will be presented as absolute numbers and percentages. Comparison of continuous variables between groups will be made by independent t tests or the Mann-Whitney U test, depending on their distribution. For the comparison of dichotomous variables, we will use the χ2 test or Fisher exact test, where applicable. Uni- and multivariable logistic regression analyses will be performed to identify predictors of adverse pregnancy outcome. For the analysis of repeated measures within the women with CHD, between the women with CHD, as well as between the women with CHD and the healthy controls, we will use multivariate general linear models. Ethical considerations. The study is conducted according to the principles of the Declaration of Helsinki and in accordance with the Medical Research Involving Human Subjects Act (Wet medisch-wetenschappelijk onderzoek met mensen). The study design, all research aims, and the specific measurements in the ZAHARA II Study have been approved by the medical ethical committee of all participating hospitals. New measurements will only be embedded in the study after approval of the medical ethical committee. All participants are asked for their written informed consent after having received written and oral information about the study. Data management and privacy protection. Data are directly entered onto written case record forms (CRF) and manually entered into an electronic database by 2 researchers. Random samples of all entered data are double checked by other research members to monitor the quality of this manual data entry process. Open text fields are copied into the electronic database exactly as they are filled in on the CRF. All measurements will be checked by examination of the data including their ranges, distributions, means, SD, outliers, and logical errors. Data outliers and missing values will be checked on the original CRF. All information in these data sets that enables identification of a participant will be excluded. The CRF as well as the data sets include subject unique identification numbers that enable feedback about one subject to the data manager but do not enable identification of that particular subject.

Discussion In the present study, we assess whether changes in cardiovascular, hemodynamic, neurohumoral parameters, and uteroplacental Doppler flow patterns during pregnancy of women with CHD differ from age- and parity-matched healthy controls. We also assess the interaction of these changes with the occurrence of cardiovascular, obstetric, and offspring events. In addition, we evaluate the incidence of permanent changes


in cardiovascular parameters 1 year postpartum in women with CHD and compare these with matched healthy controls. During normal pregnancy, considerable hemodynamic changes occur.24 Total vascular resistance decreases, mainly due to peripheral arterial vasodilatation, mediated by progesterone and vasodilators such as nitric oxide as well as a low-resistance flow in the uteroplacental circulation.25,26 In normal pregnancy, a gradual widening of the maternal spiral arteries occurs early in the first trimester due to the invasion of endovascular and interstitial trophoblasts that convert maternal spiral arteries closer to the intervillous space. As a consequence, plasma volume, heart rate, and CO gradually increase in the first 2 trimesters of pregnancy to maintain adequate organ and uteroplacental perfusion while TVR decreases.24,26 Blood pressure drops in the first trimester of pregnancy, reaching its lowest point at the end of the second trimester, around 20 weeks of gestation, and returns to near pre-pregnancy levels around term.24,27 Together with the rise in plasma volume, the CO rises, which is reflected by an increase in both atrial and ventricular volumes, ventricular wall thickness, and the rise of heart rate.24 In normal pregnancy, an increased NT-proBNP level can be measured.28 Cardiovascular hemodynamic state returns back to pre-pregnancy state in most women within 6 months postpartum.24,29 In women with CHD, hemodynamic changes in pregnancy can exceed the compensatory possibilities of their compromised circulation, resulting in cardiac complications such as heart failure, arrhythmias, and other cardiovascular events. Known predictors of maternal cardiovascular events are related to underlying disease as well as to pre-pregnancy hemodynamic and functional status.2,3,9 However, detailed information about changes in ventricular and valvular function as well as in CO and TVR and their relationship to the occurrence of cardiovascular events in pregnancy is not available. Echocardiographic evaluation before, during, and after pregnancy in our population with CHD and in matched healthy pregnant controls will clarify some of the confusion. NT-proBNP is a well-known marker of heart failure severity.30,31 In women with CHD, NT-proBNP has been incompletely evaluated, even outside pregnancy. However, it has been shown that NT-proBNP correlates positively with New York Heart Association functional class deterioration as well as with cyanosis and inversely with ventricular ejection fraction, even in asymptomatic women with CHD.32,33 As these parameters predict pregnancy complications in women with CHD, it is plausible to hypothesize that NT-proBNP may be an easy and useful method for stratification of cardiovascular risk in pregnancy.1-3,9 Moreover, NT-proBNP levels are divergent in hypertensive disorders of pregnancy, with a graded increase from normal pregnancy to

Balci et al 273

gestational hypertension and PE.34 The frequency of these disorders is increased in women with CHD, and NTproBNP levels may provide further insight in the relation of cardiovascular status and the occurrence of these complications.4,35 Hypertensive disorders of pregnancy, especially PE, are characterized by impaired trophoblast invasion and failure of dilatation of spiral arteries, resulting in high uteroplacental vascular resistance, leading to inadequate uteroplacental perfusion and adverse obstetric and offspring outcome, including IUGR and PE.6,7 Compromised uterine perfusion with placental dysfunction is reflected by abnormal uteroplacental Doppler waveform patterns.5 Abnormal uterine artery pulsatility index and notching in early diastolic phase predicts PE, is related to high TVR and low CO, and is associated with IUGR and small for gestational age.5 In CHD, a higher incidence of PE and IUGR is seen.1-4,35,36 This may be related to a lower CO and higher incidence of heart failure in these women that may lead to inadequate uteroplacental perfusion. Therefore, uteroplacental flow patterns may differ in women with CHD compared to healthy women and abnormal uteroplacental perfusion may be associated with cardiovascular status as well as with offspring outcome.

Late effects of pregnancy in CHD Long-term effects of pregnancy in women with CHD are incompletely described. In a cohort of women with aortic stenosis, requirement of cardiac intervention was the most important late cardiac event after pregnancy.10 In a small group of women after Mustard correction, permanent deterioration in functional class, dilatation of right ventricle, right ventricular dysfunction, and tricuspid valve regurgitation have been demonstrated.11 In women with atrioventricular septal defect, persisting deterioration of atrioventricular valve regurgitation was found as well as persisting deterioration of functional class.37 Unfortunately, most data are retrospective, and for many defects, no data on mid- and late-term outcome after pregnancy are available. In the present study, echocardiographic and clinical follow-up will be at least 1 year post pregnancy for all women with CHD. In normal pregnancy, exercise capacity declines in the early postpartum period.38,39 Despite the improvement in the following 6 months, the pre-pregnancy level is not reached. Whether exercise capacity in women with CHD declines more than in healthy pregnant women is uncertain and will be studied in our cohort.

Conclusion The current ZAHARA II study is the first “in vivo” study in women with CHD to evaluate the effect of compromised cardiac performance on the uteroplacental circulation and its relationship with the occurrence of obstetric events and adverse offspring outcome. By

274 Balci et al

identifying the components responsible for pregnancyrelated events in women with CHD, we will refine risk stratification that will lead to better pre-pregnancy counseling and may eventually improve treatment of these women.

References 1. Drenthen W, Pieper PG, Roos-Hesselink JW, et al. Outcome of pregnancy in women with congenital heart disease: a literature review. J Am Coll Cardiol 2007;49:2303-11. 2. Siu SC, Sermer M, Colman JM, et al. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation 2001; 104:515-21. 3. Drenthen W, Boersma E, Balci A, et al. Predictors of pregnancy complications in women with congenital heart disease. Eur Heart J 2010 doi:10.1093/eurheartj/ehq200. 4. Drenthen W, Pieper PG, Ploeg M, et al. Risk of complications during pregnancy after Senning or Mustard (atrial) repair of complete transposition of the great arteries. Eur Heart J 2005;26: 2588-95. 5. Aardema MW, Lander M, Oosterhof H, et al. Doppler ultrasound screening predicts recurrence of poor pregnancy outcome in subsequent pregnancies, but not the recurrence of PIH or preeclampsia. Hypertens Pregnancy 2000;19:281-8. 6. Bosio PM, McKenna PJ, Conroy R, et al. Maternal central hemodynamics in hypertensive disorders of pregnancy. Obstet Gynecol 1999;94:978-84. 7. Gerretsen G, Huisjes HJ, Elema JD. Morphological changes of the spiral arteries in the placental bed in relation to pre-eclampsia and fetal growth retardation. Br J Obstet Gynaecol 1981;88:876-81. 8. Madazli R, Somunkiran A, Calay Z, et al. Histomorphology of the placenta and the placental bed of growth restricted foetuses and correlation with the Doppler velocimetries of the uterine and umbilical arteries. Placenta 2003;24:510-6. 9. Khairy P, Ouyang DW, Fernandes SM, et al. Pregnancy outcomes in women with congenital heart disease. Circulation 2006;113:517-24. 10. Tzemos N, Silversides CK, Colman JM, et al. Late cardiac outcomes after pregnancy in women with congenital aortic stenosis. Am Heart J 2009;157:474-80. 11. Guedes A, Mercier LA, Leduc L, et al. Impact of pregnancy on the systemic right ventricle after a Mustard operation for transposition of the great arteries. J Am Coll Cardiol 2004;44:433-7. 12. Thorne S, MacGregor A, Nelson-Piercy C. Risks of contraception and pregnancy in heart disease. Heart 2006;92:1520-5. 13. Warnes CA, Liberthson R, Danielson GK, et al. Task force 1: the changing profile of congenital heart disease in adult life. J Am Coll Cardiol 2001;37:1170-5. 14. Baumgartner H, Hung J, Bermejo J, et al. Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice. J Am Soc Echocardiogr 2009;22:1-23. 15. Easterling TR, Carlson KL, Schmucker BC, et al. Measurement of cardiac output in pregnancy by Doppler technique. Am J Perinatol 1990;7:220-2. 16. Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography's Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18: 1440-63.


17. Robson SC, Dunlop W, Moore M, et al. Combined Doppler and echocardiographic measurement of cardiac output: theory and application in pregnancy. Br J Obstet Gynaecol 1987;94:1014-27. 18. Rudski LG, Lai WW, Afilalo J, et al. Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 2010;23:685-713. 19. Vasapollo B, Novelli GP, Valensise H. Total vascular resistance and left ventricular morphology as screening tools for complications in pregnancy. Hypertension 2008;51:1020-6. 20. Vahanian A, Baumgartner H, Bax J, et al. Guidelines on the management of valvular heart disease: the Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology. Eur Heart J 2007;28:230-68. 21. Barnett SB, Maulik D. Guidelines and recommendations for safe use of Doppler ultrasound in perinatal applications. J Matern Fetal Med 2001;10:75-84. 22. ATS/ACCP. ATS/ACCP statement on cardiopulmonary exercise testing. Am J Respir Crit Care Med 2003;167:211-77. 23. Gomez O, Figueras F, Fernandez S, et al. Reference ranges for uterine artery mean pulsatility index at 11-41 weeks of gestation. Ultrasound Obstet Gynecol 2008;32:128-32. 24. Duvekot JJ, Peeters LL. Maternal cardiovascular hemodynamic adaptation to pregnancy. Obstet Gynecol Surv 1994;49:S1-14. 25. Weiner CP, Knowles RG, Moncada S. Induction of nitric oxide synthases early in pregnancy. Am J Obstet Gynecol 1994;171: 838-43. 26. Duvekot JJ, Cheriex EC, Pieters FA, et al. Early pregnancy changes in hemodynamics and volume homeostasis are consecutive adjustments triggered by a primary fall in systemic vascular tone. Am J Obstet Gynecol 1993;169:1382-92. 27. Wilson M, Morganti AA, Zervoudakis I, et al. Blood pressure, the renin-aldosterone system and sex steroids throughout normal pregnancy. Am J Med 1980;68:97-104. 28. Hameed AB, Chan K, Ghamsary M, et al. Longitudinal changes in the B-type natriuretic peptide levels in normal pregnancy and postpartum. Clin Cardiol 2009;32:E60-2. 29. van Oppen AC, Stigter RH, Bruinse HW. Cardiac output in normal pregnancy: a critical review. Obstet Gynecol 1996;87:310-8. 30. Hogenhuis J, Jaarsma T, Voors AA, et al. BNP and functional status in heart failure. Cardiovasc Drugs Ther 2004;18:507. 31. Palazzuoli A, Gallotta M, Quatrini I, et al. Natriuretic peptides (BNP and NT-proBNP): measurement and relevance in heart failure. Vasc Health Risk Manag 2010;6:411-8. 32. Tulevski II, Groenink M, van der Wall EE, et al. Increased brain and atrial natriuretic peptides in patients with chronic right ventricular pressure overload: correlation between plasma neurohormones and right ventricular dysfunction. Heart 2001;86:27-30. 33. Giannakoulas G, Dimopoulos K, Bolger AP, et al. Usefulness of natriuretic peptide levels to predict mortality in adults with congenital heart disease. Am J Cardiol 2010;105:869-73. 34. Moghbeli N, Srinivas SK, Bastek J, et al. N-terminal pro-brain natriuretic peptide as a biomarker for hypertensive disorders of pregnancy. Am J Perinatol 2010;27:313-9. 35. Drenthen W, Pieper PG, Roos-Hesselink JW, et al. Non-cardiac complications during pregnancy in women with isolated congenital pulmonary valvar stenosis. Heart 2006;92:1838-43. 36. Yap SC, Drenthen W, Meijboom FJ, et al. Comparison of pregnancy outcomes in women with repaired versus unrepaired atrial septal defect. BJOG 2009;116:1593-601.


37. Drenthen W, Pieper PG, van der Tuuk K, et al. Cardiac complications relating to pregnancy and recurrence of disease in the offspring of women with atrioventricular septal defects. Eur Heart J 2005;26: 2581-7. 38. Treuth MS, Butte NF, Puyau M. Pregnancy-related changes in physical activity, fitness, and strength. Med Sci Sports Exerc 2005;37: 832-7. 39. South-Paul JE, Rajagopal KR, Tenholder MF. Exercise responses prior to pregnancy and in the postpartum state. Med Sci Sports Exerc 1992;24:410-4. 40. Dickstein K, Cohen-Solal A, Filippatos G, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the Diagnosis and Treatment of Acute

Balci et al 275

and Chronic Heart Failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur Heart J 2008;29: 2388-442. 41. Durack DT, Lukes AS, Bright DK. New criteria for diagnosis of infective endocarditis: utilization of specific echocardiographic findings. Duke Endocarditis Service. Am J Med 1994;96:200-9. 42. Verstappen WHJM, Jans SMPJ, Van Egmond N, et al. Nationwide Primary Care Cooperation Agreement on anemia during pregnancy and puerperium. (Landelijke Eerstelijns Samenwerkings Afspraak Anemie tijdens zwangerschap en kraamperiode). Huisarts Wetenschap 2007;50:S17-20.


Appendix. Supplementary Data The ZAHARA II investigators From the Departments of 1Cardiology, 2Obstetrics and 3 Epidemiology, University Medical Centre Groningen, University of Groningen, Groningen, The Netherlands; 4 Interuniversity Cardiology Institute of the Netherlands (ICIN)/Royal Dutch Academy of Science (KNAW), Utrecht, The Netherlands: Ali Balci, MD, MSc1,4, Willem Drenthen, MD, PhD1, Joost van Melle, MD, PhD1, Elke Hoendermis, MD, PhD1, Adriaan A. Voors, MD, PhD1, Dirk J. van Veldhuisen, MD, PhD1, Petronella G. Pieper, MD, PhD1, Krystyna M. Sollie, MD2, Jan G. Aarnoudse, MD, PhD2, Hans L. Hillege, MD, PhD1,3 From the Departments of 5Cardiology and 6Obstetrics, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands: Barbara J.M. Mulder, MD, PhD5, Berto J. Bouma, MD, PhD5, Maarten Groenink, MD, PhD5, Michiel M. Winter, MD, PhD5, Jeroen C. Vis, MD, PhD5, Paul Luijendijk, MD5, Zelia Koyak, MD5, Piet de Witte, MD5, A. Carla Zomer, MD5, Monique W. M. de Laat, MD, PhD6, Manja H.T.L Bunschoten, RN6 From the Departments of 7Cardiology and 8Obstetrics, Erasmus Medical Centre, Erasmus University, Rotterdam, The Netherlands: Jolien W. Roos-Hesselink, MD, PhD7,

Balci et al 275.e1

Maarten Witsenburg, MD, PhD7, Judith A.A.E. Cuypers, MD7, Eric A.P. Steegers, MD, PhD8, J. Cornette, MD, PhD8 From the Departments of 9Cardiology and 10Obstetrics University Medical Centre Nijmegen St Radboud, Radboud University Nijmegen, Nijmegen, The Netherlands: Arie P.J. van Dijk, MD, PhD9, W. Marc Waskowsky, MD, PhD9, Marc Spaanderman, MD, PhD10 From the Department of 11Cardiology, Medical Spectrum Twente, Enschede, The Netherlands: Elly M.C.J. Wajon, MD11, Lodewijk J. Wagenaar, MD, PhD, Jeannine A.J.M. Hermens, MD11 From the Departments of 12Cardiology and 13Obstetrics, Leiden University Medical Centre, University of Leiden, Leiden, The Netherlands: Hubert W. Vliegen, MD, PhD12, Monique R.M. Jongbloed, MD, PhD12, Marjolein S. Verhart, RN13, Jos J.M. van Roosmalen, MD, PhD13 From the Departments of 14Cardiology and 15Obstetrics, University Medical Centre Utrecht, The Netherlands: Gertjan T. Sieswerda, MD, PhD14, A. Carla C van Oppen, MD, PhD15, Martijn A. Oudijk, MD, PhD15 From the Departments of 16Cardiology and 17Obstetrics, Maastricht University Medical Centre, University of Maastricht, Maastricht, The Netherlands: Jan L.M. Stappers, MD, PhD16, Jos P.M. Offermans, MD, PhD17

Clinical Investigations

Acute Ischemic Heart Disease

Prehospital triage in the ambulance reduces infarct size and improves clinical outcome Sonja Postma, MSc, a Jan-Henk E. Dambrink, MD, PhD, b Menko-Jan de Boer, MD, PhD, b A. T. Marcel Gosselink, MD, PhD, b Gerrit J. Eggink, MD, c Henri van de Wetering, MANP, b,c Frans Hollak, RN, c Jan Paul Ottervanger, MD, PhD, b Jan C. A. Hoorntje, MD, PhD, b Evelien Kolkman, MSc, a Harry Suryapranata, MD, PhD, a,b and Arnoud W. J. van 't Hof, MD, PhD b Zwolle, The Netherlands

Background We evaluated the effect of prehospital triage (PHT) in the ambulance on infarct size and clinical outcome and studied its relationship to the distance of patient's residence to the nearest percutaneous coronary intervention (PCI) center. Methods All consecutive ST-segment elevation myocardial infarction patients who were transported to the Isala klinieken from 1998 to 2008 were registered in a dedicated database. Of these, 2,288 (45%) were referred via a spoke center and 2.840 (55%) via PHT. Results PHT patients were more often treated within 3 hours after symptom onset (46.2% vs 26.8%, P b .001), more often had a post-procedural thrombolysis in myocardial infarction (TIMI) 3 flow (93.0% vs 89.7%, P b .001) had a smaller infarct size (peak creatine kinase 2,188 ± 2,187 vs 2,575 ± 2,259 IU/L, P b .001) and had a lower 1-year mortality (4.9% vs 7.0%, P = .002). After multivariate analysis, PHT was independently associated with ischemic time less than 3 hours (OR 2.45, 95% CI 2.13-2.83), a peak creatine kinase less than the median value (OR 1.19, 95% CI 1.04-1.36) and a lower 1-year mortality (OR 0.67, 95% CI 0.50-0.91). The observed differences between PHT patients and the spoke group were more pronounced in the subgroup of patients living N38 km from the PCI center. Conclusion PHT in the ambulance is associated with a shorter time to treatment, a smaller infarct size and a more favorable clinical outcome, especially with longer distance from the patient's residence to the nearest PCI center. Therefore, PHT in the ambulance may reduce the negative effect of living at a longer distance from the PCI center. (Am Heart J 2011;161:276-82.)

It is well known that rapid restoration of coronary blood flow in patients with ST-segment elevated myocardial infarction (STEMI) is crucial1 to limit myocardial damage.2,3 There are 2 established therapeutic strategies to achieve this: fibrinolysis and primary percutaneous coronary intervention (pPCI), of which the latter is considered more effective.2,4–7 However, there is still debate about the time interval during which pPCI can be recommended. According to the American College of Cardiology/American Heart Association (ACC/AHA), pPCI is preferred if the delay between first medical

From the aDiagram, Zwolle, Netherlands, bIsala Klinieken, Zwolle, Netherlands, and cRAV IJsselland, Zwolle, The Netherlands. Submitted January 21, 2010; accepted October 18, 2010. Reprint requests: Arnoud W.J. van't Hof, MD, PhD, FESC, Isala klinieken, location Weezenlanden, Department of Cardiology, Groot Wezenland 20, 8011JW Zwolle, The Netherlands. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.10.028

contact and pPCI does not exceed 90 minutes8 and according to the European Society of Cardiology (ESC), this delay should not exceed 120 minutes.9 If the anticipated delay exceeds 90 minutes (ACC/AHA) or 120 minutes (ESC), fibrinolytic therapy may be an alternative because it can be administered with a shorter delay in the ambulance or in a nearby spoke center.10,11 One of the possibilities to shorten delays and maximize the number of patients eligible for pPCI is to transport STEMI patients directly to a PCI center after prehospital triage (PHT) in the ambulance. Several studies have shown that this strategy can significantly reduce ischemic times compared to patients being referred via a spoke center.12–15 Recently, Pedersen et al15 have demonstrated that PHT also improves outcome. However, a recent analysis from the HORIZONS AMI study did not show a benefit for patients immediately transported to a PCI center.16 In addition, it is less well known if PHT is beneficial for patients living at a longer distance from a PCI center. Therefore, we addressed both the question on efficacy and on distance in a large cohort of nonselected STEMI patients who


were directed to our center either via a referral center (spoke) or via PHT.

Methods Population Since the early nineties, STEMI patients referred to the Isala Klinieken were treated by pPCI. The PHT project was initiated to improve the logistics of STEMI patients. This has gradually been implemented in the region, starting in 1998. At that time 13 spoke centers referred their STEMI patients to our PCI center however, this number decreased to 8 spoke centers since new PCI centers opened. During the project, all consecutive STEMI patients who were transported to our PCI center from 1998 until 2008 and underwent pPCI were prospectively registered in a dedicated database. Criteria for the diagnosis of STEMI were: (1) history of cardiac symptoms of at least 10 minutes in the last 24 hours before presentation at the spoke or PCI center, (2) elevated levels of creatine kinase (CK) or CK-MB, and (3) concurrent electrocardiogram (ECG) changes: ST-segment elevation of N1 mV in at least 2 adjacent electrocardiogram leads.17 Information whether patients were transferred via referral centers (spoke group) in the network or via PHT (PHT group) was recorded. In addition, information on infarct size, angiographic outcome, and short- and long-term clinical outcome were registered as well. Infarct size was calculated as the peak level of CK (peak CK) within 48 hours after admission.18 The distance via motorway to the nearest PCI center was computerized using the postal codes of the patient's residence and the PCI center. Subsequently, the percentiles (25-75) of the computed distances were calculated. The PHT and spoke groups were subdivided in short (≤38 km) and long distance (N38 km) based on the median distance from patient's residence to the PCI center. Patients were excluded if they did not have a confirmed diagnosis of acute myocardial infarction (CK b200 and no evidence of unstable plaque or culprit lesion at coronary angiography), the distance from patient's residence to the PCI center was not available, or if the distance from patient's residence to the PCI center was ≥120 km (outer bound of referring area).

Triage for pPCI PHT: The algorithm of PHT has been described previously.19 In brief, after patients dialed the emergency number, patients were triaged in the ambulance, if the ambulance was equipped with the PHT equipment. If STEMI was suspected, an ECG was made by highly trained paramedics and the computerized algorithm revealed an outcome. If a diagnosis of STEMI was made, the ambulance went straight to the catheterization laboratory of the PCI center, bypassing the emergency departments of the spoke center. Spoke: If the ambulance was not equipped with the PHT equipment, the ambulance went to the nearest spoke center where diagnosis and triage was performed. For diagnosis, an ECG was made immediately upon arrival. In case of a STEMI diagnosis, patients were transported to the catheterization laboratory of the PCI center as soon as possible. Walk-ins at the PCI center were excluded because they did not receive PHT.

Postma et al 277

pPCI procedure In both situations (PHT and spoke), the staff of the catheterization laboratory of the PCI center was preinformed about the estimated time of arrival of the patient and was activated well before the arrival of the patient. In case the staff lived more than 30 minutes away from the PCI center, they had to stay in the PCI center when being on-call. Patients were pretreated with an intravenous bolus of 5000 IU unfractionated heparin, 500 mg aspirin intravenously, and subsequently with 600 mg clopidogrel and/or tirofiban (25 μg/kg bolus, 0.15 μg kg−1 min−1 maintenance infusion). Prehospital triage patients were pretreated in the ambulance and spoke patients were pretreated at the spoke center and/or in the ambulance that transferred the patient form the spoke center to the PCI center.

Time intervals Six different time intervals were evaluated: (1) time from symptom onset (SO) to infarct diagnosis (time diagnostic ECG) either in the ambulance or at a spoke center (SO-diagnosis); (2) time from diagnosis till arrival at the PCI center (diagnosis-door PCI); (3) time from diagnosis to balloon inflation (BI) (diagnosisBI); (4) time from arrival spoke center to arrival PCI center (door-to-door, or D2D); (5) time from arrival at the PCI center to BI for PHT patients and time from arrival spoke hospital to BI for spoke patients (door-to-balloon, or D2B); and (6) total ischemic time defined as the time from SO to BI.

Statistical analysis Statistical analysis was performed with SPSS 17.0 (SPSS, Chicago, IL). Continuous data were expressed as mean ± SD or median and interquartile range. Categorical data were presented as percentage. Analysis of variance was used for continuous data and Pearson χ2 test was used for the categorical data, respectively. We tested the associations between the variable “PHT” and other baseline characteristics using univariate logistic regression. The Mann-Whitney test has been used to calculate the time intervals between the PHT group and the spoke group, because they were non-Gaussian distributed. To assess independent predictors of PHT, multivariate analysis was performed using a logistic regression analysis. In this analysis, univariate variables with a P value b.10 were included. In all the statistical analyses, P values ≤.05 were considered as statistical significant. The study was conducted according to the principles of the Declaration of Helsinki and the protocol was approved by the local institutional review board. No extramural funding was used to support this work. The authors are solely responsible for the design and conduct of this study, all study analyses and the drafting and editing of the paper.

Results Ambulance triage versus referral by spoke center From 1998, the first year of implementation of PHT in the ambulance, until 2008, 5,674 patients were referred to our institution with the intention to perform pPCI. Three hundred twenty-three patients (5.7%) were false positives (PHT and spoke patients) and 223 patients


278 Postma et al

Table I. Baseline characteristics Characteristic

Total (N = 5128)

PHT (n = 2840)

Spoke (n = 2288)

P

Age (y) ± SD Male gender Risk factors Hypertension DMII Smoking Hypercholesterolemia Family history Killip class N1 Previous MI Previous PCI Previous CABG Previous CVA TIMI risk score

62.17 ± 12.14 3800/5125 (74.1%)

62.80 ± 12.12 2110/2838 (74.3 %)

61.40 ± 12.12 1690/2287 (73.9%)

b.001 .713

1668/5063 (32.9%) 557/5108 (10.9%) 2433/5015 (48.5%) 1085/4852 (22.4%) 2058/4928 (41.8%) 346/5098 (6.8%) 499/5109 (9.8%) 364/5074 (7.2%) 132/5119 (2.6%) 146/5114 (2.9%) 2.16 ± 1.82 2 (1-3)

926/2811 (32.9%) 296/2833 (10.4%) 1310/2785 (47.0%) 563/2682 (21.0%) 1151/2731 (42.1%) 171/2831 (6.0%) 232/2830 (8.2%) 214/2821 (7.6%) 84/2837 (3.0%) 78/2834 (2.8%) 2.17 ± 1.84 2 (1-3)

742/2252 (32.9%) 261/2275 (11.5%) 1123/2230 (50.4%) 522/2170 (24.1%) 907/2197 (41.3%) 175/2267 (7.7%) 267/2279 (11.7%) 150/2253 (6.7%) 48/2282 (2.1%) 68/2280 (3.0%) 2.15 ± 1.80 2 (1-3)

.996 .243 .019 .011 .542 .018 b.001 .203 .054 .623 .948

CABG, Coronary artery bypass grafting; CVA, cerebrovascular accident; DMII, diabetes mellitus type II; MI, myocardial infarction.

Figure 1

The different time intervals are shown for the total group, for patients living ≤38 km from a PCI center and for patients living N38 km from a PCI center. These groups were subdivided for PHT and spoke patients.

(3.9%) were walk-ins at the PCI center. From the remaining 5,128 patients who actually underwent pPCI, 2,288 patients (45%) were referred via a spoke center and 2,840 patients (55%) via PHT. Patients from the spoke group were younger, more often smoked, more often had hypercholesteremia, more often had a previous myocardial infarction, more often presented in Killip class N1 and less often had previous coronary artery bypass grafting (CABG) (Table I). The median distance from patient's residence to the nearest PCI center was 38 km (25-49), 28 km (16-41) for the PHT

group and 43 km (37-60) for the spoke group (P b .001). The different time intervals for the 2 groups are presented in Figure 1. Total median ischemic time was 209 minutes, 184 minutes for the PHT group and 260 minutes for the spoke group (P b .001). The percentage of patients treated within 3 hours of SO was 46.2% in the PHT group compared with 26.8% in the spoke group (P b .001) (Table II). The time from SO-diagnosis was shorter for the PHT group as compared to that for the spoke group (82 vs 129 minutes, P b .001). The median D2D for the spoke group was 85 minutes. The D2B was longer for the spoke


Postma et al 279

Table II. Distance, total time delays, indication of pPCI, and distance of patient's residence to PCI center PHT Total Distance (km)⁎ Ischemic time (median, IQR, n = 2728) Ischemic time b3 h (%, n = 5028) SO-diagnosis (median, IQR, n = 2531) Diagnosis-door PCI (median, IQR, n = 1982) Diagnosis-BI (median, IQR, n = 2625) D2D (median, IQR, n = 984) D2B (median, IQR, n = 2664) Indication pPCI ACC/AHA (%, n = 2625) Indication pPCI ESC (%, n = 2625) Distance home-PCI center ≤38 km Distance (km)⁎ Ischemic time (median, IQR, n = 1308) Ischemic time b3 h (n = 2522) SO-diagnosis (median, IQR, n = 1206) Diagnosis-door PCI (median, IQR, n = 974) Diagnosis-BI (median, IQR, n = 1240) D2D (median, IQR, n = 285) D2B (median, IQR, n = 1266) Indication pPCI ACC/AHA (%, n = 1245) Indication pPCI ESC (%, n = 1245) Distance home-PCI center N38 km Distance (km)⁎ Ischemic time (median, IQR, n = 1420) Ischemic time b3 h (%, n = 2506) SO-diagnosis (median, IQR, n = 1325) Diagnosis-door PCI (median, IQR, n = 1326) Diagnosis-BI (median, IQR, n = 1358) D2D (median, IQR, n = 699) D2B (median, IQR, n = 1398) Indication pPCI ACC/AHA (%, n = 1385) Indication pPCI ESC (%, n = 1385)

28.1 184 1285/2784 82 52 94

(15.9-14.8) (140-274) (46.2%) (47-150) (37-68) (77-119)

44 (30-69) 719/1637 (43.9%) 1230/1637 (75.1%) 19.0 187 850/1908 87 46 90

(13.1-29.0) (139-282) (44.5%) (49-158) (33-61) (74-118)

46 (31-74) 484/962 (49.7%) 731/962 (76.0%) 45.9 180 435/876 76 60 101

(41.0-63.0) (142-256) (49.7%) (45-136) (45-79) (82-122)

40 (27-61) 241/675 (35.7%) 488/675 (73.9%)

Spoke

P

42.6 (37.3-59.9) 260 (187-393) 601/2244 (26.8%) 129 (68-252) 83 (59-116) 123 (98-160) 85 (60-123) 132 (101-189) 173/988 (17.5%) 466/988 (47.2%)

b.001 b.001 b.001 b.001 b.001 b.001

32.6 (26.8-36.0) 255 (177-407) 171/614 (27.9%) 132 (69-259) 78 (54-114) 118 (93-159) 78 (53-120) 127 (98-198) 56/283 (20.1%) 149/283 (52.7%)

b.001 b.001 b.001 b.001 b.001 b.001

49.6 261 430/1630 125 85 124 89 133 117/710 322/710

(41.6-66.7) (190-391) (26.4%) (66-251) (61-117) (100-160) (62-126) (103-188) (16.5%) (45.4%)

b.001 b.001 b.001

b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001

IQR, Interquartile range. ⁎ Data expressed as median and IQR.

group as compared to that for the PHT group (132 vs 44 minutes, P b .001). Significantly more patients of the PHT group were treated according to the most recent guidelines of the ACC/AHA and the ESC as compared to the spoke group (ACC/AHA: PHT 43.9% vs spoke 17.5%, ESC: PHT 75.1% vs spoke 47.2%, P b .001 for both comparisons) (Table II).8,9 PHT patients had also better angiographic and clinical outcome. They more often had post-procedural TIMI 3 flow (93.0% vs 89.7%, P b .001), had a smaller infarct size (peak CK: 2188 ± 2187 vs 2575 ± 2259 IU/L, P b .001) and a lower 1-year mortality (4.9% vs 7.0%, P = .002) (Table III). After correction for differences in baseline characteristics, including the difference in distance, PHT was independently associated with an ischemic time of less than 3 hours (OR 2.45, 95% CI 2.13-2.83), infarct size of less than the median value (OR 1.19, 95% CI 1.04-1.36) and a lower 1-year mortality (OR 0.67, 95% CI 0.50-0.91) (Table IV).

Distance to PCI center Figure 1 illustrates the time intervals in association with the distance from the patient's residence to the PCI center. PHT patients had a shorter time from SO-diagnosis irrespective of distance to the PCI center. Surprisingly, total ischemic time was even somewhat shorter in the PHT patients living at a distance N38 km as compared to ≤38 km. For both groups, however, the time from diagnosis-door PCI was longer for patients living at a longer distance from the PCI center (PHT from 44 to 55 minutes, spoke from 78 to 89 minutes, P b .001 for both comparisons). A longer distance led to a decrease in the percentage of patients treated according to the guidelines of the ACC/AHA (≤38 km: 540/1245 [43.4%] vs N38 km: 358/1385 [25.6%], P b .001) and the ESC (≤ 38 km: 880/1245 [70.7%] vs N38 km: 810/1385 [58.5%], P b .001) (Table II).8,9 In patients living ≤38 km from the PCI center, apart from a higher post-procedural TIMI 3 flow in the PHT group, no significant differences in angiographic


280 Postma et al

Table III. Angiographic and clinical outcome PHT (n = 2840)

Spoke (n = 2288)

Total TIMI 3 flow pre-PCI 570/2826 (20.2%) 423/2286 (18.5%) TIMI 3 flow post-PCI 2625/2822 (93.0%) 2042/2276 (89.7%) Infarct size ± SD⁎ 2188±2187 2575±2259 (IU/L, n = 4986) 30-d mortality 86/2798 (3.1%) 90/2249 (4.0%) 1-y mortality 134/2729 (4.9%) 151/2164 (7.0%) Distance residence-PCI center ≤38 km TIMI 3 flow pre-PCI 359/1935 (18.6%) 125/626 (20.0%) TIMI 3 flow post-PCI 1792/1932 (92.8%) 561/623 (90.0%) Infarct size ± SD⁎ 2154 ± 2161 2359±2146 (IU/L, n = 2482) 30-d mortality 60/1913 (3.1%) 28/615 (4.6%) 1-y mortality 98/1878 (5.2%) 41/581 (7.1%) Distance residence-PCI center N38 km TIMI 3 flow pre-PCI 211/891 (23.7%) 298/1660 (18.0%) TIMI 3 flow post-PCI 833/890 (93.6%) 1481/1653 (89.6%) Infarct size ± SD⁎ 2261 ± 2241 2656±2295 (IU/L, n = 2504) 30-d mortality 26/885 (2.9%) 62/1634 (3.8%) 1-y mortality 36/851 (4.2%) 110/1583 (6.9%)

P

.134 b.001 b.001 .074 .002 .432 .030 .054 .096 .094 b.001 b.001 b.001 .264 .007

⁎ Peak CK.

parameters, infarct size, or clinical outcome were found. However, in patients living N38 km away from the PCI center, PHT patients more often had a higher initial and post-procedural TIMI 3 flow, had a smaller infarct size and had a lower 1-year mortality compared to the spoke group (Table III).

Discussion So far, this is the largest study showing that PHT in the ambulance with immediate transportation to the nearest PCI center result in a shorter time to treatment, a reduction in infarct size and a better angiographic and clinical outcome as compared to referral via a spoke center. This was most evident for patients living at a longer distance from the PCI center. In addition, PHT significantly increased the number of patients treated according to the ACC/AHA and the ESC guidelines.8,9 These results suggest that living at a longer distance from a PCI center may not negatively influence angiographic and clinical outcomes when PHT is available. Our study showed that 43.9% and 75.1% of the patients, respectively, fulfill the ACC/AHA and the ESC criteria for pPCI when PHT is available. When diagnosis and triage was performed at a spoke center, only 17.5% (ACC/AHA) and 47.2% (ESC) of the patients were treated according to the guidelines. Figure 1 shows that the largest difference between the groups is seen in the time from SO-diagnosis. Because the ambulance that brought the patient initially to a spoke center was not able to make the infarct diagnosis (no trained personnel, no

ECG equipment), diagnosis was only made after arrival of the patient in the spoke center at a median of 129 minutes after SO, whereas after PHT, diagnosis was made 47 minutes (36%) earlier. This earlier diagnosis, together with the early initiation of potent antiplatelet and antithrombotic agents in the ambulance (unfractionated heparin, aspirin, and clopidogrel), may have led to the higher initial patency and better angiographic outcome in PHT patients.20 Despite the fact that all efforts were taken to arrange further or a new transport to the PCI center as soon as possible, the diagnosis-door PCI time was considerably longer in the spoke group as compared to the diagnosis-door PCI time in the PHT group (83 vs 52 minutes, P b .001) (Figure 1). As a consequence, the D2B time was significantly longer for the spoke group as compared to the PHT group (132 vs 44 minutes, P b .001). These findings correspond with the results of Le May et al.21

Effect of distance Our study shows that time to treatment is substantially reduced by PHT in the ambulance and shows that incorporating this triage significantly increases the number of patients who are candidates for primary angioplasty instead of thrombolysis according to the guidelines, especially for patients who live further than 38 km away from a PCI center. These findings suggest that the logistics of arranging diagnosis and immediate transportation is more important than the distance of the patient's residence from a PCI center: for PHT patients, total ischemic times remained very short despite a longer distance from the PCI center. A longer distance from patients' residence to the PCI center had very limited effect on total ischemic time despite a significantly increased diagnosis-door PCI time. This was due to a shorter SO-diagnosis time. Other investigators have also studied the effect of distance on clinical outcome in STEMI patients. In a cohort study, Wei et al22 demonstrated that patients with a first myocardial infarction who were living at more than 9 miles (14.5 km) from the PCI center had a higher mortality compared to patients who lived closer to the PCI center. Nevertheless, in our study, the effect of distance on outcome was related to the type of patient triage, which was not reported in the study from Wei et al. According to a recent study, D2B may also be reduced by performing pPCI at a PCI center without on-site cardiac surgery.23 However, this study was stopped prematurely and a third arm with routine ambulance triage was lacking.24 It might therefore be true that the opening of extra PCI centers is not necessary when routine ambulance triage systems are being developed. Because PHT patients have better outcomes compared to spoke patients, more centers should implement the


Postma et al 281

Table IV. Output multivariate analyses Treatment within 3 h Variables Type triage Gender Hypertension Family history DMII Age TIMI risk score Previous CABG Previous PCI IRV/LAD Distance

OR (95% CI) 2.45 1.21 1.20 1.04 1.01 1.03 0.61 0.70 1.55 1.27 1.00

(2.13-2.83) (1.04-1.41) (1.04-1.41) (0.91-1.18) (0.81-1.26) (1.02-1.04) (0.57-0.65) (0.46-1.08) (1.21-1.99) (1.13-1.47) (1.00-1.00)

Infarct size < median Variables Type triage Gender Hypertension Hypercholesterolemia Smoking Age Previous MI Previous CABG Previous PCI Killip class N1 IRV/CX IRV/graft IRV/LAD IRV/RCA Distance

OR (95% CI) 1.19 0.67 0.96 0.99 0.86 1.01 1.29 1.03 1.47 0.61 2.10 3.98 1.85 4.05 0.99

(1.04-1.36) (0.58-0.78) (0.84-1.10) (0.85-1.16) (0.75-0.98) (1.00-1.01) (1.01-1.65) (0.58-1.83) (1.12-1.93) (0.47-0.78) (0.67-6.64) (1.02-15.6) (0.59-5.81) (1.29-12.7) (0.99-1.00)

One-year mortality Variables Type triage Gender Hypertension Hypercholesterolemia DMII Age TIMI risk score Family history Previous MI Pervious CVA Killip class N1 Distance IRV/LAD

OR (95% CI) 0.67 0.90 1.42 0.63 1.06 1.04 1.21 0.70 1.23 2.29 3.07 1.00 1.05

(0.50-0.91) (0.66-1.21) (1.06-1.92) (0.43-0.93) (0.72-1.57) (1.02-1.05) (1.09-1.35) (0.51-0.96) (0.90-1.67) (1.28-4.10) (1.99-4.72) (0.99-1.01) (0.80-1.39)

CABG, Coronary artery bypass grafting; CX, circumflex; DMII, diabetes mellitus type II; IRV, infarct related vessel; LAD, left anterior descendents; MI, myocardial infarction; RCA, right coronary artery.

PHT for STEMI patients and use ECG equipment with a computerized electrographic algorithm or with telemedicine. It is also important to make patients and general practitioners aware of the fact that PHT with ambulance transport has better outcomes for STEMI patients instead of self-referring, referring via general practitioners, or referring via a spoke center.

Limitations Several limitations of this study need to be acknowledged. First, because the project was not randomized and dispersed over 10 years (the percentage of PHT patients increased from 6.0% in 1998, 51.6% in 2004, to 68.7% in 2008), consequently, the risk of unknown confounders exists. Some selection of patients has occurred. PHT patients were older and more often lived closer to the PCI center. During the PHT project, more remote ambulance services started participating, while at the beginning these patients all were transported via a spoke center. Second, we did not investigate the effect of fibrinolytics for patients living at great distance. Therefore, we cannot state that PHT is the best way to treat all STEMI patients living at a great distance however, in our population PHT patients have better outcomes when living at N38 km from a PCI center compared to patients referred via a spoke center. Third, we have no exact numbers of patients who were self-referred or came in by ambulance at the spoke center. Most patients came in by ambulance at the spoke center however, these ambulances were not operational with ECG equipment and the EMS personnel were not trained to make STEMI diagnosis. Subsequently, diagnosis was made in the spoke hospital however, first medical contact took place in the ambulance.

Fourth, information of the time of first medical contact is lacking. We expect these times to be the same in both groups however, we do not have solid evidence. Fifth, although peak CK has been shown to be a reliable parameter for the estimation of infarct size, data on the area under the CK release curve are lacking. Sixth, we have used the postal codes of patient's residence to calculate the distance to the PCI center because the exact location of the place where the patient was picked up by an ambulance was not available. Finally, more research has to be performed to the favor of PHT for patients living at a long distance from the PCI center, so more understanding and verification of this improvement can be achieved.

Conclusion In conclusion, PHT in the ambulance with immediate transportation to the nearest PCI center is associated with a significantly shorter time to treatment, reduced infarct size, and better angiographic and clinical outcome when compared to referral via a nearby spoke center. This beneficial effect is more apparent with longer distance from the patient's residence to the PCI center. PHT also significantly increased the percentage of patients that fall within the time window in which pPCI is the preferred treatment according to the ACC/AHA and ESC guidelines. Therefore, PHT may reduce transportation delays in patients who live at a longer distance from the PCI center.

Disclosures Conflict of interest: none declared.

282 Postma et al

References 1. Keeley EC, Hillis LD. Primary PCI for myocardial infarction with STsegment elevation. N Engl J Med 2007;356:47-54. 2. Boersma E. The Primary Coronary Angioplasty vs. Thrombolysis (PCAT)-2 Trialists' Collaborative Group. Does time matter? A pooled analysis of randomized clinical trials comparing primary percutaneous coronary intervention and in-hospital fibrinolysis in acute myocardial infarction patients. Eur Heart J 2006;27:779-88. 3. De Luca G, Suryapranata H, Ottervanger JP, et al. Time delay to treatment and mortality in primary angioplasty for acute myocardial infarction. Every minute of delay counts. Circulation 2004;109: 1223-5. 4. Dalby M, Bouzamondo A, Lechat P, et al. Transfer for primary angioplasty versus immediate thrombolysis in acute myocardial infarction. A meta-analysis. Circulation 2003;108:1809-14. 5. Henry TD, Unger BT, Sharkey SW, et al. Design of a standardized system for transfer of patients with ST-elevation myocardial infarction for percutaneous coronary intervention. Am Heart J 2005;150: 373-84. 6. Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. Lancet 2003;361:13-20. 7. Andersen HR, Nielsen TT, Rasmussen K, et al. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Eng J Med 2003;349:733-42. 8. ACC/AHA. 2007 focussed update of the ACC/AHA 2004 guidelines for the management of patients with ST elevated myocardial infarction. JACC 2008;51:210-47. 9. The task force on the management of ST-segment elevation acute myocardial infarction of the European Society of Cardiology: Task force members. Management of acute myocardial infarction in patients presenting with persistent ST-segment elevation. Eur Heart J 2008;29:2909-45. 10. Pinto DS, Kirtane AJ, Nallamothu BK, et al. Hospital delays in reperfusion for ST-elevation myocardial infarction: implications when selecting a reperfusion strategy. Circulation 2006;114: 2019-25. 11. Nallamothu BK, Bates ER, Herrin J, et al. Times to treatment in transfer patients undergoing primary percutaneous coronary intervention in the United States: National Registry of Myocardial Infarction (NRMI)3/4 analysis. Circulation 2005;111:761-7. 12. Terkselsen CJ, Lassen JF, Nørgaard BJ, et al. Reduction of treatment delay in patients with ST-elevated myocardial infarction: impact of pre hospital diagnosis and direct referral to primary coronary intervention. Eur Heart J 2005;26:770-7.


13. Terkelsen CJ, Nørgaard BJ, Lassen JF, et al. Prehospital evaluation in ST-elevation myocardial infarction patients treated with primary percutaneous coronary intervention. J of Electrocard 2005;38: 187-92. 14. Clemmensen P, Sejersten M, Sillesen M, et al. Diversion of STelevation myocardial infarction patients for primary angioplasty based on wireless prehospital 12-lead electrocardiographic transmission directly to the cardiologist's handheld computer: a progress report. J of Electrocard 2005;38:194-8. 15. Pedersen SH, Galatius SG, Hanse PR, et al. Field triage reduces treatment delay and improves long-term clinical outcome in patients with acute ST-segment elevation myocardial infarction treated with primary percutaneous coronary intervention. J Am Coll Card 2009; 54:2296-302. 16. O'Riordan M. Should patients go directly to a PCI-capable hospital? Two studies, different results. Paper presented at: Transcatheter Cardiovascular Therapeutics; September 23, 2009, San Fransico, CA. 17. Van ‘t Hof AWJ, Liem A, de Boer MJ, et al. Clinical value of 12-lead electrocardiogram after successful reperfusion therapy for acute myocardial infarction. The Lancet 1997;350:615-9. 18. Nienhuis MB, Ottervanger JP, de Boer MJ, et al. Prognostic importance of creatine kinase and creatine kinase–MB after primary percutaneous coronary intervention for ST-elevation myocardial infarction. Am Heart J 2008;155:673-9. 19. Van't Hof AWJ, Rasoul S, Wetering H, et al. Feasibility and benefit of prehospital diagnosis, triage, and therapy by paramedics only in patients who are candidates for primary angioplasty for acute myocardial infarction. Am Heart J 2006;151:1255.e1-5. 20. Zijlstra F, Patel A, Jones M, et al. Clinical characteristics and outcome of patients with early (b2 h), intermediate (2-4 h) and late (N4 h) presentation treated by primary coronary angioplasty or thrombolytic therapy for acute myocardial infarction. Eur Heart J 2002;23:550-7. 21. Le May MR, Derek YS, Dionne R, et al. A citywide protocol for primary PCI in ST-segment elevated myocardial infarction. N Eng J Med 2008;358:231-40. 22. Wei L, Lang CC, Sullivan FM, et al. Impact on mortality following first acute myocardial infarction of distance between home and hospital: cohort study. Heart 2008;94:1141-6. 23. Peels HO, Swart de H, van der Ploeg T, et al. Percutaneous coronary intervention with off-site cardiac surgery backup for acute myocardial infarction as a strategy to reduce door-to-balloon time. Am J Card 2007;100:1353-8. 24. de Boer MJ, Bronzwaer JGF, Boers M. Percutaneous coronary intervention with off-site cardiac surgical backup. Am J Card 2008; 101:1522.

The influence of time from symptom onset and reperfusion strategy on 1-year survival in ST-elevation myocardial infarction: A pooled analysis of an early fibrinolytic strategy versus primary percutaneous coronary intervention from CAPTIM and WEST Cynthia M. Westerhout, PhD, a,d Eric Bonnefoy, MD, b,d Robert C. Welsh, MD, a,d Philippe Gabriel Steg, MD, c,d Florent Boutitie, PhD, b,d and Paul W. Armstrong, MD a,d Edmonton, Canada; and Lyon and Paris, France

Background The CAPTIM trial suggested a survival benefit of prehospital fibrinolysis (FL) compared to primary percutaneous coronary intervention (PCI) in patients with ST-elevation myocardial infarction (STEMI) with a presentation delay of b2 hours. We examined the relationship between reperfusion strategy and time from symptom onset on 1-year mortality in a combined analysis of 1,168 patients with STEMI. Methods Individual patient data from CAPTIM (n = 840, 1997-2000) and the more recent WEST trial (n = 328, 20032005) were pooled. Results Median age was 58 years, 81% were men, and 41% had anterior myocardial infarction; 640 patients were randomized to FL versus 528 patients to PCI. Both arms received contemporary adjunctive medical therapy. Presentation delay (ie, symptom onset to randomization) was similar in FL and PCI patients (median 105 [72-158] vs 106 [74-162] minutes, P = .712). Rescue PCI after FL occurred in 26% and 27%, and 30-day PCI, in 70% and 71% in CAPTIM and WEST, respectively. Mortality was not different between FL and PCI (4.6% vs 6.5%, P = .263); however, the interaction between presentation delay and treatment was significant (P = .043). Benefit with FL was observed with time b2 hours (2.8% [FL] vs 6.9% [PCI], P = .021, hazard ratio [HR] 0.43, 95% CI 0.20-0.91), whereas beyond 2 hours, no treatment difference was observed (6.9% [FL] vs 6.0% [PCI], P = .529, HR 1.23, 95% CI 0.61-2.46). Conclusions A strategy of early FL demonstrated a reduction in 1-year mortality compared to primary PCI in early presenters. Time from symptom onset should be a key consideration when selecting reperfusion therapy for STEMI. (Am Heart J 2011;161:283-90.)

The optimal reperfusion therapy for patients with STelevation myocardial infarction (STEMI) has attracted great interest, stimulated vigorous controversy, and led to constructive enhancement of health care systems aimed at abbreviating time from symptom onset to reperfusion. Although contemporary STEMI guidelines recommend primary percutaneous coronary intervention (PCI) as the

From the aUniversity of Alberta, Edmonton, Canada, bUniversite Lyon, Lyon, France, and c Universite Paris, Paris, France. ClinicalTrials.gov Identifier: NCT00121446 (WEST). d For the CAPTIM/WEST Investigators. Submitted August 25, 2010; accepted October 15, 2010. Reprint requests: Paul W. Armstrong, MD, 251 Medical Sciences Building, University of Alberta, Edmonton, Canada T6G 2H7. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.10.033

preferred strategy, provided it can be delivered promptly in an expert facility, this is often not feasible.1,2 However these guidelines also emphasize the desirability of a total ischemic time of b2 hours and support the use of fibrinolytic therapy if primary PCI cannot be performed within 90 minutes of first medical contact. It follows that patient and situational specific reperfusion strategies, that is, fibrinolysis (FL) or primary PCI, require consideration, and the optimal approach remains an active source of debate in many circumstances. Although some trials and a systematic overview conclude that PCI is the preferred therapy, there are notable caveats arguing against a “one size fits all” strategy favoring a more nuanced approach.3,4 An important and well-recognized modulator of prognosis after STEMI is time from symptom onset until reperfusion,5,6 which is now a required consideration in therapeutic decision making.1,2 Indeed, recent STEMI guidelines indicate a goal of curtailing total ischemic time


284 Westerhout et al

to b2 hours.1 Major advancements in the delivery of STEMI care have been aimed at minimizing time to treatment, particularly in prehospital diagnosis, early delivery of reperfusion,7 and timely triage to an appropriate hospital. The CAPTIM trial was the first large-scale trial to evaluate the efficacy of prehospital FL versus PCI8 and, importantly, provided evidence of an early window of benefit with prehospital FL when time from symptom onset was within 2 hours and mechanical coronary cointervention was frequently employed.9 However, the CAPTIM trial was hypothesis-generating in this regard and concluded enrollment (after 840 patients) before its target sample of 1,200 patients. The current study extends this observation through collaboration between CAPTIM and the WEST trialists, the latter adding over 300 similarly randomized contemporaneous patients with STEMI, thereby approximating the initial CAPTIM enrollment.10 Our primary goal was to examine the influence of reperfusion choice on 1-year mortality and the extent to which it is modulated by time from symptom onset. In addition, we explored this issue on the prespecified events of recurrent myocardial infarction (MI), cardiogenic shock, and safety.

Methods The details and primary results of the CAPTIM and WEST trials have been previously published.8,10 In brief, the CAPTIM trial randomized eligible patients at the site of initial management to prehospital FL or direct transfer for primary PCI. All patients received an intravenous bolus of 5,000 U heparin and 250 to 500 mg aspirin. Patients assigned to prehospital FL received an intravenous bolus of alteplase followed by an infusion over 90 minutes. Coronary angiography and subsequent revascularization were allowed in the FL group at the discretion of the responsible physicians and, when appropriate, rescue angioplasty was done. Patients assigned to primary PCI were transported immediately to the hospital for coronary angiography and angioplasty, if indicated. Angioplasty was done according to local standards with the intention of restoring blood flow in the infarct-related artery as soon as possible. After randomization, heparin was continued for at least 48 hours. Those receiving stents were treated with a thienopyridine for 1 month. In addition to aspirin, the protocol recommended use of atenolol in all patients and lisinopril in those with anterior infarcts. The WEST trial had a parallel-group design that randomized patients into one of the following 3 treatment arms, at the earliest point of care, including the prehospital setting and participating study hospitals: (1) usual care: optimal pharmacologic therapy (prompt administration of tenecteplase [TNK], aspirin, and enoxaparin); (2) early invasive strategy: identical pharmacologic therapy and early invasive strategy including mandatory rescue PCI; or (3) primary PCI (after aspirin, enoxaparin, and 300 mg clopidogrel). Abciximab was recommended for all PCI procedures unless performed within 3 hours of fibrinolytic therapy, and clopidogrel was used in patients in the pharmacologic therapy groups according to American College of Cardiology/ American Heart Association PCI guidelines.

Patients The CAPTIM trial enrolled patients if they presented within 6 hours of symptom onset (ie, characteristic pain lasted for at least 30 minutes, not responsive to nitrates, with electrocardiographic [ECG] ST-segment elevation of at least 0.2 mV in ≥2 contiguous leads, or left bundle branch block) in mobile emergency care units (Service d'Aide Médicale d'Urgence) with 27 affiliated tertiary hospitals in France. Patients could be excluded if the transfer time to hospital was expected to be N60 minutes. Like CAPTIM, the WEST trial enrolled patients with STEMI within 6 hours of symptom onset. Specifically, patients were sought in whom primary PCI could not be delivered within 60 minutes but for whom primary PCI and/or transfer for rescue PCI was feasible within 3 hours of randomization.11 Symptoms presumed secondary to STEMI lasting at least 20 minutes were required, accompanied by ECG evidence of high risk, which included ≥2 mm of ST-elevation in ≥2 contiguous precordial leads or limb leads or ≥1-mm STelevation in ≥2 limb leads coupled with ≥1-mm STdepression in ≥2 contiguous precordial leads (total STdeviation ≥4 mm) or presumed new left bundle branch block. An emphasis on the earliest possible randomization was a cardinal feature of WEST and prehospital ECG; randomization and initiation of therapy (using paramedics vs physicians in CAPTIM) were strongly encouraged for those patients using 911/ambulance access to health care facilities. Forty-four percent of patients were randomized prehospital: 45.7% with FL and 41.1% with primary PCI. The enrollment period was extended beyond the prespecified sample of 304 patients to include an additional 24 patients to expand the prehospital randomization. Fifteen sites within 4 metropolitan areas in Canada (Edmonton, Halifax, Montreal, and Vancouver) were involved in enrollment. The maximum follow-up in WEST was 1 year.

Statistical analysis Discrete variables are reported as counts and percentages of nonmissing cases; median and 25th and 75th percentiles are reported for continuous variables. Differences between studies and between study treatments were tested by χ2, WilcoxonMann-Whitney U, and Kruskal-Wallis tests as appropriate. Analyses were conducted according to the intention-to-treat principle, which included all patients who gave informed consent and were randomized to study treatment, irrespective of whether treatment was actually received. In the WEST trial, the 2 FL arms (TNK + usual care and TNK + early angiography) were not statistically different with respect to the primary end point of the trial10; thus, these 2 arms were combined with the fibrinolytic arm in CAPTIM. The primary end point of this study was all-cause mortality within 1 year. Eight patients were lost to follow-up in the WEST study, whereas CAPTIM had a complete 1-year followup. Kaplan-Meier survival curves were constructed to illustrate the time to death within 1 year after randomization, with between-group differences evaluated by the log-rank test (and accounting for the between-trial variation by stratification). Via stratified Cox proportional hazards regression, the adjusted HR and corresponding 95% CI for 1-year mortality were estimated for study treatment, time from symptom onset, and the


Westerhout et al 285

Figure 1

Derivation of study cohort.

Table I. Selected baseline characteristics CAPTIM

n Age, y Male, % Systolic BP, mm Hg Pulse, beat/min Weight, kg Height, cm Killip Class N1, % Hypertension, % History of diabetes, % History of angina, % Prior MI, % Previous PCI, % Current smoker, % Anterior MI, %

WEST

CAPTIM-WEST

FL

Primary PCI

FL

Primary PCI

All

FL

Primary PCI

419 58 (49-68.5) 82.5 125 (110-140) 75 (64-84) 75 (68-85) 170 (165-175) 8.7 34.1 11.1 13.5 8.2 5.3 52.7 40.1

421 58 (50-68) 81.5 128 (111-140) 75 (66-88) 75 (67-84) 170 (165-175) 12.2 34.8 13.6 14.6 6.7 4.3 49.2 42.7

221 58 (50-68.5) 77.4 140 (122.3-160) 72 (62-85) 83 (73-94) 173 (167-178) 5.4 45.2 11.8 28.5 12.7 8.1 49.5 38.9

107 60 (49-71) 80.4 143 (124-158.5) 74 (62-88) 82 (73-93) 173 (167.5-179) 6.5 34.6 15.0 18.7 13.1 4.7 39.3 42.1

1168 58 (49-69) 81.0 131 (23.5) 75 (65-86) 77 (69-86) 170 (165-176) 9.1 36.5 12.5 17.2 9.0 5.4 49.5 41.0

640 58 (49-68.3) 80.7 130 (115-147) 74 (63-84) 78 (70-88) 171 (165-176) 7.5 38.0 11.3 18.7 9.8 6.3 51.6 39.7

528 58.5 (50-69) 81.3 130 (115-145) 75 (65-88) 75 (68-85) 170 (165-176) 11.0 34.8 13.9 15.4 8.0 4.4 47.1 42.9

Continuous variables as median (25th to 75th percentile).

interaction of these 2 factors, again accounting for betweentrial variation. Time from symptom onset was treated as a discrete variable (ie, time from symptom onset to randomization b2 vs ≥2 hours), which was prespecified based on previous analyses of time from symptom onset and is in accord with current STEMI guidelines relating to reperfusion options and total ischemic time.1,2,5,9 The relationship between time from symptom onset, treatment, and 1-year mortality was also expressed with time in a continuous fashion. The linearity and proportional hazard assumptions

were evaluated. Adjustment of the relationship was performed via backward selection using an inclusion criterion of α = .05 and exclusion criterion of α = .10 (and confirmed with forward selection). Baseline patient characteristics considered in adjustment included age, sex, systolic blood pressure, heart rate, Killip class, hypertension, history of diabetes, history of angina, prior MI, previous PCI, current smoking status, and anterior MI. Because the DANAMI-2 study reported that the benefit observed for PCI in their study was restricted to patients with



Table II. Timing intervals CAPTIM

n Symptom onset to randomization, min b2 h, % ≥2 h, % Symptom onset to first medical contact, min First medical contact to randomization, min Symptom onset to FL, min Symptom onset to primary PCI, min Randomization to FL, min Randomization to primary PCI, min Length of stay, d

WEST

CAPTIM-WEST

FL

Primary PCI

FL

Primary PCI

All

FL

Primary PCI

419 107 (75.8-158.3) 55.3 44.7 78 (48-135)

421 108 (76-162) 55.0 45.0 77 (48-135)

221 105 (66-159) 60.2 39.8 52.5 (29-101.5)⁎

107 101 (70-163.8) 56.1 43.9 54 (29.3-108.8)⁎

1168 105.5 (73-161) 56.1 43.9 72 (42-130)

640 105 (72.5-158.5) 57.0 43.0 70 (40-125)

528 106 (74-161.5) 55.3 44.7 73 (44.8-131)

25 (18-34)

26 (19-35)

38 (28-59)⁎

39 (26-60)⁎

29 (20-40)

30 (21-41)

28 (20-38)

130 (95-180)

–

117 (75-179.3)⁎

–

–

122 (85-180)

–

–

–

–

9 (6-15)⁎

189 (150.5-293) –

–

16 (10-23)

190 (148.5-255) –

–

13 (8-21)

189 (148.3-256.8) –

–

72 (60-88)

–

90 (64.5-97.3)⁎

–

–

76 (62-94)

8 (6-11)

8 (6-10)

3 (2-5)

3 (2-5)

7 (3-10)

7 (3-10)

7 (4-9)

⁎ P b .05 for comparison between studies within treatment.

high Thrombolysis In Myocardial Infarction (TIMI) risk, we undertook a secondary analysis to examine this in high- versus low-risk patients as defined by the TIMI risk score (high risk ≥5 points, low risk b5 points).12,13 Additional analysis was undertaken on prespecified events, that is, recurrent MI and cardiogenic shock, and the incidence of major systemic bleeding and nonfatal intracranial hemorrhage (ICH) inhospital. All statistical comparisons were done at the 5% level of significance using a 2-sided alternative hypothesis, unless stated otherwise. Analyses were performed using SAS (version 9.2, SAS Institute, Cary, NC).

in both trials with 56% of patients randomized within 2 hours of symptom onset. Although there was a shorter interval from symptom onset to first medical contact in WEST than in CAPTIM (median 53 vs 78 minutes), CAPTIM's time from first medical contact to randomization was shorter at 26 minutes versus 38 minutes in WEST. Overall, time from symptom onset to treatment with FL was shorter in WEST compared to CAPTIM (P = .006); however, time to PCI was comparable between the studies (Table II). The median length of stay was 8 days in CAPTIM and 3 days in WEST.

No extramural funding was used to support this work. The authors are solely responsible for the design and conduct of this study, all study analyses, and the drafting and editing of the paper and its final contents.

One-year mortality Overall, 63 (5.4%) of 1,160 patients with complete follow-up died within 1 year of randomization. There was no difference in overall 1-year mortality between FL and primary PCI ([29/633] 4.6% vs [34/527] 6.5%, unadjusted HR 0.75 [FL vs PCI], 95% CI 0.46-1.24, P = .546) (Figure 2). When 1-year mortality was further examined according to the TIMI risk score (high risk ≥5 points [17.8% of patients], low risk b5 points [82.2%]), mortality was comparable in low-risk patients (FL [12/513] 2.3% vs PCI [9/405] 2.2%, P = .899). Oneyear mortality was nominally higher in PCI than in FL in high-risk patients, but this did not reach statistical significance (FL [13/99] 13.1% vs PCI [18/101] 17.8%, P = .373). However, when mortality was examined according to time from symptom onset to randomization (Figures 3 and 4), the interaction was statistically significant (P = .043). Patients randomized within 2 hours of symptom onset had improved survival with FL compared to those receiving primary PCI ([10/358] 2.8% vs [20/288] 6.9%, P = .021, HR 0.43, 95% CI 0.20-

Results Individual patient data from the CAPTIM, which enrolled 840 patients with STEMI between 1997 and 2000, and WEST trials, which enrolled 328 patients with STEMI between 2003 and 2005, were obtained for the current study (Figure 1). Baseline patient characteristics according to study treatment are presented in Table I for the individual trials as well as the pooled cohort. Overall, the patients enrolled in the CAPTIM and WEST trials were comparable across the trials and between the study treatments: median age of 58 years, 81% men, 9% with prior MI, and median weight of 77 kg.

Time from symptom onset In Table II, the overall median time from symptom onset to randomization was 106 minutes and was similar


Figure 2

Kaplan-Meier curve of 1-year survival according to study treatment (P = .263) (FL, dashed; PCI, solid line).


Figure 4

Association between time from symptom onset and study treatment according to 1-year mortality.

Figure 3 Figure 5

Kaplan-Meier curve of 1-year survival according to study treatment and time from symptom onset (P = .021, FL b2 hours vs PCI b2 hours; P = .529, FL ≥2 hours vs PCI ≥2 hours) (FL b2 hours, dotted; FL ≥2 hours, dashed; PCI b2 hours, thick solid; PCI ≥2 hours, thin solid line).

0.91). Beyond 2 hours, however, no treatment difference in 1-year mortality was observed (FL [19/274] 6.9% vs PCI [14/234] 6.0%, P = .529, HR 1.23, 95% CI 0.61-2.46). After adjustment for age, systolic blood pressure, heart rate, diabetes, and Killip class, a significant interaction between time from symptom onset and study treatment remained (P = .037) such that FL patients had a (relative) 42% lower hazard of 1year mortality than PCI patients with symptom onset within 2 hours (ie, in time b2 hours, FL vs PCI, adjusted HR 0.58, 95% CI 0.26-1.29). In patients randomized beyond 2 hours, there appeared to be an excess hazard associated with FL versus PCI (ie, FL vs PCI, adjusted HR 1.81, 95% CI 0.87-3.77). When examining time from symptom onset in a continuous fashion (Figure 5), the point estimates for FL versus PCI suggest a survival benefit for FL until approximately 127 minutes from

Fibrinolysis versus PCI on 1-year mortality according to increasing time from symptom onset. The hazard of mortality for FL approximated that of PCI at 127 minutes (note that time b127 minutes, 59.1% of all patients; time b240 minutes, 91.6% of all patients). Hazard ratio indicated by thick solid line and 95% CI in thick gray lines.

when symptom onset had elapsed; thereafter, a reversal occurred with improved survival with PCI.

Inhospital and 30-day events and interventions Table III provides the 30-day occurrence of cardiogenic shock and re-MI and inhospital safety events. Although 30-day cardiogenic shock appeared to be higher in primary PCI patients than in FL patients, this did not achieve statistical significance (5.3% vs 3.8%, P = .203); re-MI, however, was over 2-fold higher in FL patients (2.0% vs 5.3%, P = .004). Intracranial hemorrhage was rare in this patient population. Although major systemic bleeds were nominally higher in primary PCI than in FL patients, this did not reach statistical significance. In FL



Table III. Cardiogenic shock, re-MI, ICH, and major systemic bleeding CAPTIM

n 30-day cardiogenic shock, n (%) Onset to randomization b2 h, n (%) Onset to randomization ≥2 h, n (%) 30-day re-MI, n (%) Onset to randomization b2 h, n (%) Onset to randomization ≥2 h, n (%) Inhospital ICH, n (%) Inhospital major systemic bleeding, n (%)

WEST

CAPTIM-WEST

FL

Primary PCI

FL

Primary PCI

All

FL

Primary PCI

419 12 (2.9) 5 (2.2) 7 (3.7) 15 (3.7) 9 (4.0) 6 (3.4) 2 (0.5) 2 (0.5)

421 19 (4.5) 11 (4.8) 8 (4.3) 7 (1.7) 3 (1.4) 4 (2.2) 0 (0.0) 8 (1.9)

221 12 (5.4) 9 (6.8) 3 (3.4) 18 (8.1) 11 (8.3) 7 (8.0) 0 (0.0) 3 (1.4)

107 9 (8.4) 6 (10.0) 3 (6.4) 3 (2.8) 3 (5.0) 0 (0.0) 0 (0.0) 1 (0.9)

1168 52 (4.5) 31 (4.7) 21 (4.1) 43 (3.8) 26 (4.1) 17 (3.5) 2 (0.2) 14 (1.2)

640 24 (3.8) 14 (3.8) 10 (3.6) 33 (5.3) 20 (5.6) 13 (4.9) 2 (0.3) 5 (0.8)

528 28 (5.3) 17 (5.9) 11 (4.7) 10 (2.0)⁎ 6 (2.2)⁎ 4 (1.8) 0 (0.0) 9 (1.7)

⁎ P b .05.

patients, rescue PCI occurred in 26% of patients and PCI within 30 days in 70% of patients.

Discussion Our novel findings not only support the critical relationship between time from symptom onset and 1year survival after reperfusion, but also provide new evidence about the impact of this relationship on the relative efficacy of the 2 standard modes of therapy. A survival advantage existed for patients treated with FL within 2 hours of symptom onset relative to those treated with primary PCI (P interaction [unadjusted] = .043). When time from symptom onset was explored in a continuous fashion, a progressive attenuation of the benefit of FL was observed until approximately 127 minutes had elapsed; thereafter, the survival advantage of PCI appeared and increased with time (Figure 5). What might account for the difference in our findings and those previously reported in a systematic overview and meta-analysis?3,5 A key distinguishing characteristic of the current versus prior trials comparing pharmacologic and mechanical reperfusion was the short time between symptom onset and randomization (including prehospital randomization) such that more than one half of the patients achieved this within 2 hours. The influence of time to treatment in myocardial salvage in patients treated with primary PCI versus FL has been examined by Schomig et al.14 Using myocardial scintigraphy to quantify the salvage index in 264 patients from 2 randomized trials, this group demonstrated that, although a similar salvage index occurred within the first tertile of time from symptom onset (approximately 165 minutes), there was a progressive decrease in salvage with FL versus PCI. Noteworthy and perhaps accounting for these findings was that median door-to-needle time was 35 minutes, that is, beyond current recommendations, whereas the door-to-balloon time was a remarkably short 65 minutes, thus resulting in an unusually brief half-hour difference between reperfusion strategies. Important additional distinguishing features of patients

in the CAPTIM-WEST analysis include systematic use of fibrin-specific pharmacologic therapy, rescue PCI in 26% of patients, and frequent cointervention with PCI in 70% of the fibrinolytic-treated patients by 30 days. The earlier time-to-treatment and more frequent cointervention likely account for the better outcome of fibrinolytictreated patients in the current study versus the DANAMI-2 trial; thereafter, mechanical rescue and cointervention were discouraged and infrequent in the fibrinolytictreated patients. Although prior data and guidelines suggest an advantage for PCI in high-risk patients, they represented a minority of the overall STEMI population and our combined cohort (17.8% of patients), whereas they represented 26% of the DANAMI-2 trial.12,13 Importantly, both the CAPTIM and WEST trials also used contemporary adjunctive antithrombotic and antiplatelet therapy in patients assigned to the primary PCI arms to ensure optimal overall results in both treatment arms. Our findings are in accordance with those of the GRACIA-2 investigators who demonstrated that routine stent/angioplasty within 3 to 12 hours of FL proved both safe and equivalent to primary stenting in preserving myocardial function.15 Although these investigators found better myocardial perfusion in the fibrinolytictreated patients, their study of 212 patients was underpowered to address whether there was associated clinical improvement and called for “a larger clinical outcome study for confirmation.” We observed an increase in the occurrence of inhospital re-MI among fibrinolytic-treated patients. This finding could relate not only to surveillance bias and the challenge of defining and detecting periprocedural MI after PCI for STEMI, but also to the possibility that early fibrinolytic-treated patients had more salvaged but subsequently jeopardized myocardium. Fibrinolytic therapy was well tolerated by this patient population with low ICH rates and minimal major bleeding that was numerically greater in the PCI group as has been observed previously.3 The low rate of ICH in the current report likely relates the relatively young age of our patients and the careful exclusion of prior


stroke, a feature that has not been uniformly applied in other studies. Our data are also consistent with recent observational reports from the French registry of acute ST-elevation myocardial infarction and respond to the recent call for confirmation of those findings in a randomized design.16,17 They are also aligned with the 3-state regional approach from the Mayo Clinic and the Vienna STEMI registry, which both attest to the excellent results after fibrinolytic administration in the early post–symptomonset period.18,19 The primacy of time from symptom onset as a modulator of outcome is underscored by the recent data of Francone et al,20 which showed marked attenuation of the potential for myocardial salvage in patients undergoing primary PCI N90 minutes after symptom onset. Hence, the relatively flat relationship between survival and time from symptom onset that we observed in primary PCI-treated patients is perhaps not surprising, given that, even under the optimal circumstances existing for timely access to expert PCI in both CAPTIM and WEST, the median time to achieve it was 189 minutes; hence, only one quarter of our patients underwent PCI within 148 minutes of symptom onset.

Limitations and strengths Some limitations of our analysis should be noted. Although combining these 2 trials was not prespecified, we believe it to be warranted given the demonstrated similar baseline characteristics and times to randomization and reperfusion. The populations studied were part of clinical trials; hence, caution regarding the generalizability of our results to a broader clinical population would be prudent. However, they are in accordance with substantial data acquired in registries.21 Although the interaction between treatment and delay was evaluated on 5-year mortality in the CAPTIM trial, this relationship did not achieve statistical significance.22 By merging the WEST and CAPTIM trials, we increased the size of the study population and events resulting in a statistically significant interaction for 1year mortality, indicating that we had sufficient power to test this association.

Conclusion Given persisting evidence of failure to meet guidelinesuggested times to PCI, especially among patients presenting to non-PCI hospitals, our data provide additional evidence to support the efficacy of an alternative reperfusion strategy, that is, fibrinolytic therapy (in patients without contraindications, coupled with contemporary adjunctive therapy, timely rescue PCI, and subsequent revascularization), especially in those presenting early after symptom onset either in the prehospital setting or in the non-PCI hospital setting. Such an approach is also likely to be welcome in


countries and regions where there is a relative paucity of PCI centers and where climate and geography may complicate immediate transfer for primary PCI.

Acknowledgements The authors would like to acknowledge the expert editorial assistance of Jo-An Padberg.

Disclosures The original WEST study was supported by an unrestricted grant from Hoffmann-LaRoche Limited; Aventis Pharma, a member of the Sanofi-Aventis group; as well as Eli-Lilly Canada. The original CAPTIM study was supported by a grant from the French Ministry of Health (Projet Hospitalier de Recherche Clinique, 96/045), by the Hospices Civils de Lyon, and by an unrestricted research grant from AstraZeneca France. Biotronic GmbH provided balloons and guidewires free of charge. Drs Armstrong and Welsh received Clinical Trial funding from Boehringer-Ingelheim.

References 1. Antman EM, Hand M, Armstrong PW, et al. 2007 focused update of the ACC/AHA 2004 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: developed in collaboration with the Canadian Cardiovascular Society endorsed by the American Academy of Family Physicians. Circulation 2008;117:296-329. 2. The Task Force on the management of ST-segment elevation acute myocardial infarction of the European Society of Cardiology. Management of acute myocardial infarction in patients presenting with persistent ST-segment elevation. Eur Heart J 2008;29:2909-45. 3. Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomized trials. Lancet 2003;361:13-20. 4. Willerson JT. Editor's commentary: one size does not fit all. Circulation 2003;107:2543-4. 5. Boersma E and the Primary Coronary Angioplasty versus Thrombolysis (PCAT)-2 trialists' collaborative group. Does time matter? A pooled analysis of randomized clinical trials comparing primary percutaneous coronary intervention and in-hospital fibrinolysis in acute myocardial infarction patients. Eur Heart J 2006;27:779-88. 6. Armstrong PW, Westerhout CM, Welsh RC. Duration of symptoms is the key modulator of the choice of reperfusion for ST-elevation myocardial infarction. Circulation 2009;119:1293-303. 7. Solis P, Amsterdam EA, Bufalino V, et al. Development of systems of care for ST-elevation myocardial infarction patients. Policy recommendations. Circulation 2007;116:e73-6. 8. Bonnefoy E, Lapostolle F, Leizorovicz A, et al, on behalf of the Comparison of Angioplasty and Prehospital Thrombolysis in Acute Myocardial Infarction (CAPTIM) study group. Primary angioplasty versus prehospital fibrinolysis in acute myocardial infarction: a randomized study. Lancet 2002;360:825-9. 9. Steg PG, Bonnefoy E, Chabaud S, et al, for the Comparison of Angioplasty and Prehospital Thrombolysis in acute Myocardial infarction (CAPTIM) Investigators. Impact of time to treatment on



10.

11.

12.

13.

14.

15.

16.

mortality after prehospital fibrinolysis or primary angioplasty. Data from the CAPTIM randomized clinical trial. Circulation 2003;108:2851-6. Armstrong PW, and the WEST steering committee. A comparison of pharmacologic therapy with/without timely coronary intervention vs primary percutaneous intervention early after ST-elevation myocardial infarction: the WEST (which early ST-elevation myocardial infarction therapy) study. Eur Heart J 2006;27:1530-8. Buller CE, Welsh RC, Westerhout CM, et al. Guideline adjudicated fibrinolytic failure: incidence, findings, and management in a contemporary clinical trial. Am Heart J 2008;155:121-7. Thune JJ, Hoefsten DE, Lindholm MG, et al, for the Danish Multicenter Randomized Study on Fibrinolytic Therapy Versus Acute Coronary Angioplast in Acute Myocardial Infarction (DANAMI)-2 Investigators. Simple risk stratification at admission to identify patients with reduced mortality from primary angioplasty. Circulation 2005;112:2017-21. Morrow DA, Antman EM, Parsons L, et al. Application of the TIMI risk score for ST-elevation MI in the National Registry of Myocardial Infarction 3. JAMA 2001;286:1356-9. Schomig A, Ndrepepa G, Mehilli J, et al. Therapy-dependent influence of time-to-treatment interval on myocardial salvage in patients with acute myocardial infarction treated with coronary artery stenting or thrombolysis. Circulation 2003;108:1084-8. Fernandez-Aviles F, Alonso JJ, Pena G, et al, for the GRACIA-2 Investigators. Primary angioplasty vs early routine post-fibrinolysis angioplasty for acute myocardial infarction with ST-segment elevation: the Gracia 2 non-inferiority, randomized, controlled trial. Eur Heart J 2007;28:949-60. Danchin N, Coste P, Ferrieres J, et al, for the FAST-MI Investigators. Comparison of thrombolysis followed by broad use of percutaneous

17.

18.

19.

20.

21.

22.

coronary intervention with primary percutaneous coronary intervention for ST-segment elevation acute myocardial infarction: data from the French registry in acute ST-elevation myocardial infarction (FASTMI). Circulation 2008;118:268-76. Nielsen PH, Maeng M, Busk M, et al, for the DANAMI-2 Investigators. Primary angioplasty versus fibrinolysis in acute myocardial infarction: long-term follow-up in the Danish acute myocardial infarction 2 trial. Circulation 2010;121:1484-91. Ting HH, Rihal CS, Gersh BJ, et al. Regional systems of care to optimize timeliness of reperfusion therapy for ST-elevation myocardial infarction: the Mayo Clinic STEMI protocol. Circulation 2007; 116:729-36. Kalla K, Christ G, Karnik R, et al, for the Vienna STEMI Registry Group. Implementation of guidelines improves the standard of care: the Viennese Registry on reperfusion strategies in ST-elevation myocardial infarction (Vienna STEMI Registry). Circulation 2006; 113:2398-405. Francone M, Bucciarelli-Ducci C, Carbone I, et al. Impact of primary coronary angioplasty delay on myocardial salvage, infarct size, and microvascular damage in patients with ST-segment elevation myocardial infarction: insight from cardiovascular magnetic resonance. J Am Coll Cardiol 2009;54:2145-53. Goldberg RJ, Spencer FA, Fox KAA, et al. Prehospital delay in patients with acute coronary syndromes (from the Global Registry of Acute Coronary Events [GRACE]). Am J Cardiol 2009;103: 598-603. Bonnefoy E, Steg PG, Boutitie F, et al, for the CAPTIM investigators. Comparison of primary angioplasty and pre-hospital fibrinolysis in acute myocardial infarction (CAPTIM) trial: a 5-year follow-up. Eur Heart J 2009;30:1598-606.

Has the ClOpidogrel and Metoprolol in Myocardial Infarction Trial (COMMIT) of early β-blocker use in acute coronary syndromes impacted on clinical practice in Canada? Insights from the Global Registry of Acute Coronary Events (GRACE) Jeremy Edwards, MD, a,i Shaun G. Goodman, MD, MSc, a,b,i Raymond T. Yan, MD, b,i Robert C. Welsh, MD, c,i Jan M. Kornder, MD, d,i J. Paul DeYoung, MD, e,i Denis Chauret, MD, f,i Jean-Pierre Picard, MD, g,i Kim A. Eagle, MD, h,i and Andrew T. Yan, MD a,b ,i Ontario, Alberta, British Columbia, and Quebec, Canada; and Ann Arbor, MI

Background The COMMIT/CCS-2 trial, published in 2005, demonstrated no net benefit of early β-blocker (BB) therapy in acute coronary syndromes (ACS). We sought to assess the short-term impact of this landmark trial by comparing the use of early BB therapy in patients with a broad spectrum of ACS before and after 2005. Methods Using data from the Global Registry of Acute Coronary Events and Canadian Registry of Acute Coronary Events, we compared the rates of BB use within the first 24 hours of presentation in the periods 1999 to 2005 and 2006 to 2008, after stratifying patients by the type of ACS (ST-segment elevation myocardial infarction [STEMI] and non–ST-segment elevation ACS [NSTEACS]) and clinical presentation. Results Of the 14,231 patients with ACS, 77.7% received BB therapy within 24 hours of presentation (78.5% and 77.4% in the STEMI and NSTEACS groups, respectively). The early use of BB declined in the STEMI group (80.3% to 76.7%, P = .005) but increased in the NSTEACS group (75.4% to 78.9%, P b .001) after 2005. Long-term BB use, higher systolic blood pressure, and higher heart rate were independent predictors of early BB use. Conversely, patients who were female, older, Killip class N1, and had cardiac arrest at presentation were less likely to receive early BB. Multivariable analysis showed a trend toward lower use of BB among patients with STEMI (adjusted odds ratio 0.76, 95% CI 0.57-1.00, P = .055) and a trend toward more frequent BB use among patients with NSTEACS (adjusted odds ratio 1.22, 95% CI 0.96-1.55, P = .11) after 2005. The temporal trends in the early use of BB differed between patients with STEMI and patients with NSTEACS (P for interaction with period b.001). Conclusions

Most patients with STEMI or NSTEACS were treated with early BB therapy. In accordance with the COMMMIT/CCS-2 trial, patients with lower systolic blood pressure and higher Killip class in the “real world” less frequently received early BB therapy. Since the publication of COMMIT/CCS-2, there has been no significant change in the use of BB in patients with STEMI or NSTEACS after controlling for their clinical characteristics. (Am Heart J 2011;161:291-7.)

a

From the Division of Cardiology, Terrence Donnelly Heart Centre, St. Michael's Hospital, University of Toronto, Toronto, Ontario, Canada, bCanadian Heart Research Centre, Toronto, Ontario, Canada, cMazankowksi Alberta Heart Institute, Edmonton, Alberta, Canada, dSurrey Memorial Hospital, Surrey, British Columbia, Canada, eCornwall Community Hospital, Cornwall, Ontario, Canada, fUniversity of Ottawa, Ottawa, Ontario, Canada, gHopital Hotel-Dieu de Sorel, Sorel-Tracy, Quebec, Canada, and hUniversity of Michigan Health System, Ann Arbor, MI. i For the Canadian Global Registry of Acute Coronary Events (GRACE) and Canadian Registry of Acute Coronary Events (CANRACE) investigators. Submitted March 3, 2010; accepted October 18, 2010. Reprint requests: Andrew T. Yan, MD, Division of Cardiology, St. Michael's Hospital, 30 Bond Street, Room 6-030 Queen, Toronto, Ontario, Canada M5B 1W8. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.10.034

β-Blocker (BB) therapy has been used in the management of acute coronary syndromes (ACS) for decades. The proposed mechanisms of benefit of BBs in the setting of acute myocardial infarction include decreased myocardial oxygen consumption due to effects on heart rate (HR), contractility, and afterload; antiarrhythmic effects; and improved myocardial oxygen supply secondary to a prolonged diastolic filling interval.1 Long-term beneficial effects include protection against adverse remodeling and declining ventricular function.2,3 In late 2005, the COMMIT/CCS-2 trial was published.4 Largely in response to the results of COMMIT/CCS-2, the American College of Cardiology (ACC) and the American Heart Association (AHA) updated their guidelines on the


292 Edwards et al

management of both ST-segment elevation myocardial infarction (STEMI) and non–ST-segment elevation ACS (NSTEACS) in 2007.5,6 The updated guidelines caution against the use of early BB therapy in patients with evidence of or risk factors for hemodynamic compromise. However, there are limited data regarding the impact of COMMIT/CCS-2 on the clinical practice in the real world.7 The primary goal of our study was to assess the shortterm impact of this landmark trial by comparing the patterns of early BB use in Canadian patients with ACS before and after publication of COMMIT/CCS-2 in 2005. The secondary objectives were to investigate (a) what patient factors may be associated with the early use of BB therapy and (b) the relationship between early BB therapy and in-hospital cardiovascular outcomes among unselected patients with ACS in the real world.

Methods Registry design and data collection The full details of the Global Registry of Acute Coronary Events (GRACE) design and methods have been previously published.8 Briefly, GRACE is a multicenter, multinational, prospective registry of a broad spectrum of patients with ACS that was expanded (GRACE2) in 2003.9 The inclusion criteria were age N18 years, alive at hospital presentation, and ACS was the provisional diagnosis with at least one of the following: elevated biomarkers, electrocardiographic changes indicative of ischemia or infarction, and/or known history of coronary artery disease. Patients were excluded if the ACS was precipitated by surgery or other comorbidity. The final ACS diagnosis was classified as STEMI, non–ST-segment elevation myocardial infarction (NSTEMI), or unstable angina. At each site, a trained coordinator collected data on standardized case report forms. Enrollment in GRACE/GRACE2 was completed in December 2007, but Canadian enrollment was extended in 2008 in the Canadian Registry of Acute Coronary Events (CANRACE), which had identical inclusion and exclusion criteria. Across Canada, 53 hospitals participated in GRACE/GRACE2/CANRACE. Of these hospitals, 35% had on-site cardiac catheterization laboratory and 21% had on-site coronary bypass grafting surgery capability.

Temporal trends of early BB therapy in ACS Early BB therapy was defined as (intravenous [IV] or oral) drug administration within 24 hours of ACS presentation. Of the 14,435 patients in GRACE/GRACE2/CANRACE, data on early BB use were available for 14,231 patients (98.6%). We compared BB use in 1999 to 2005 with use in 2006 to 2008, as the COMMIT/ CCS-2 trial was published in late 2005. Recognizing that there could have been a change in BB use over time, independent of the publication of COMMIT/CCS-2, we also repeated the comparison of BB use excluding the patients enrolled from 1999 to 2002. The COMMIT/CCS-2 study and the ACC/AHA guidelines for the management of STEMI and NSTEACS have identified the following as risk factors for hemodynamic and/or respiratory compromise with early BB use4-6: age N70 years, systolic blood pressure (SBP) b120 mm Hg, HR b60 or N110 beat/min, Killip class N1,10 and delayed presentation N13 hours. We specifically

looked at BB use in these subgroups, which were at risk for harm associated with early BB therapy.

Outcomes in relation to early BB use for ACS We compared patients who did and did not receive early BB therapy with respect to these prespecified outcomes: in-hospital mortality, re-infarction, cardiogenic shock, pulmonary edema and/or acute congestive heart failure (CHF), and sustained ventricular tachycardia or fibrillation. The definition of reinfarction was limited to a re-elevation in cardiac biomarkers 24 hours after presentation.11 We also sought to investigate cardiovascular outcomes in high-risk patients by stratifying patients by hemodynamic status, as mentioned above.

Statistical analysis Discrete variables are presented as frequencies or percentages. Continuous variables are presented as medians with interquartile ranges. We compared group differences in discrete and continuous variables using the χ2 and Mann-Whitney U tests, respectively. We developed a multivariable model to examine factors associated with early BB use, based on clinical considerations and prior studies, including the study by Emery et al7 on the use of BB in NSTEACS published in 2006. Generalized estimating equations were used to control for the clustering of patients within hospitals. The variables considered in the multivariable model were age, gender, myocardial infarction, CHF, long-term BB use, SBP, HR, Killip class, cardiac arrest, creatinine, type of ACS, ST-segment deviation, and initial biomarker status.7,12,13 We also included the period (1999-2005 vs 2006-2008) of enrollment in the multivariable model to assess whether the publication of COMMIT/CCS-2 might have impacted upon the use of early BB therapy. Because of the different trends in the use of BB in the NSTEACS and STEMI groups, we tested for their interaction with period of enrollment in the model. To assess for an independent association between early BB use and outcome, we adjusted for other evidence-based therapies and for components of the GRACE risk score. The GRACE score is based upon age, HR, SBP, serum creatinine, Killip class, cardiac arrest at presentation, the presence of STsegment deviation, and elevated cardiac enzymes. Statistical analysis was performed using SPSS v. 15.0 (SPSS, Inc, Chicago, IL), and 2-sided P b .05 was considered to be significant. This research was supported by an unrestricted grant from Sanofi-Aventis, Paris, France, and Laval, Quebec, Canada, and by Bristol Myers Squibb Canada, Montreal, Quebec, Canada. The industrial sponsors had no involvement in the study conception or design; collection, analysis, and interpretation of data; writing, review, or approval of the manuscript; and decision to submit the manuscript for publication. The authors are solely responsible for the design and conduct of this study, all study analyses, and the drafting and editing of the paper and its final contents.

Results Patient characteristics Between 1999 and 2008, there were 14,231 Canadian patients with ACS in GRACE/GRACE2/CANRACE with complete data on early BB use. Overall, 77.7% of patients received early BB therapy. The rates of early BB use were


Edwards et al 293

Table I. Baseline demographic and clinical characteristics

Age, y⁎ Female Medical history Current smoker Diabetes Hypertension Angina Dyslipidemia Peripheral vascular disease TIA/stroke Renal insufficiency MI PCI CABG Preexisting heart failure Atrial fibrillation Long-term BB use Clinical presentation Systolic BP, mm Hg⁎ Diastolic BP, mm Hg⁎ HR, beat/min⁎ Body mass index⁎ Killip class Killip I Killip II Killip III Killip IV Cardiac arrest ST elevation ST depression Significant Q wave Positive initial cardiac biomarker Creatinine, μmol/L⁎ GRACE risk score⁎

Table II. In-hospital management

No early BB therapy (n = 3175)

Early BB therapy (n = 11056)

P

71 (59-80) 38%

66 (56-76) 32%

b.001 b.001

25% 28% 60% 42% 50% 10%

27% 27% 60% 44% 54% 8.4%

.031 .80 .93 .060 b.001 .004

11% 13% 32% 15% 11% 14%

8.6% 10% 33% 18% 13% 10%

b.001 b.001 .20 b.001 .032 b.001

12% 16%

8.6% 40%

b.001 b.001

140 (120-160) 77 (65-89)

144 (126-162) 80 (70-92)

b.001 b.001

78 (65-95) 26.7 (23.7-30.5)

78 (66-92) 27.5 (24.8-31.1)

.39 b.001 b.001

77% 14% 8.6% 0.7% 2.9% 24% 28% 8.7% 46%

86% 9.6% 4.4% 0.3% 1.1% 25% 30% 11% 48%

b.001 .064 .021 .003 .018

89 (69-116) 135 (109-169)

87 (70-107) 125 (101-153)

b.001 b.001

CABG, Coronary artery bypass graft surgery; MI, myocardial infarction; PCI, percutaneous coronary intervention; BP, blood pressure. ⁎ Median (25th and 75th percentiles).

78.5% and 77.4% in the STEMI and NSTEACS groups, respectively (P = .15). Table I summarizes the relevant demographics and clinical features of the study population.

In-hospital management Patients who received early BB therapy were more likely to receive other evidence-based therapies and to undergo cardiac catheterization and percutaneous coronary intervention during index hospitalization (Table II). They were less likely to be treated with an IV inotropic agent, calcium-channel blocker, and diuretic during the first 24 hours of presentation.

Medication use within first 24 h Aspirin Clopidogrel Warfarin Any heparin Glycoprotein IIb/IIIa inhibitor Thrombolytic ACE inhibitor Angiotensin receptor blocker Calcium-channel blocker IV inotropic agent Statin Nitrates Diuretic In-hospital management Cardiac catheterization PCI CABG LVEF assessment LV systolic function Normal Mildly diminished Moderately to severely diminished Length of stay, d⁎

No early BB therapy (n = 3175)

Early BB therapy (n = 11056)

P

83 52 5.6 81 6.9 11 37 12 27 6.6 46 59 33

94 67 5.0 89 10 14 60 11 20 2.1 73 72 28

b.001 b.001 .20 b.001 b.001 b.001 b.001 .092 b.001 b.001 b.001 b.001 b.001

53 31 2.7 59

62 34 3.5 67

b.001 .001 .042 b.001 .015

49 29 22

52 30 19

5 (3-9)

5 (3-8)

.075

Data shown in percentages, unless otherwise indicated. LVEF, Left ventricular ejection fraction; LV, left ventricular. ⁎ Median (25th and 75th percentiles).

Early BB use before and after the publication of COMMIT/CCS-2 Overall, early BB use was slightly higher among patients who were enrolled in 2006 to 2008 than among those enrolled in 1999 to 2005 (78.4% vs 76.9%, P = .039) (Fig. 1). The frequency of early BB use in patients with STEMI decreased in the years 2006 to 2008 compared with 1999 to 2005, whereas the frequency of early BB use rose in patients with NSTEACS (Fig. 1). There were no significant temporal changes in early BB use in subgroups identified in COMMIT/CCS-2 to be at risk for an adverse event associated with early BB therapy: age N70 years, SBP b120 mm Hg, Killip class N1, HR ≥100 beat/min, or time to presentation N13 hours (Fig. 2). There was no significant change in the results when the analysis was repeated excluding the patients from 1999 to 2002. Comparing the period 2003 to 2005 versus 2006 to 2008, there were significant increases in the early use of aspirin (92% to 93%, P = .014), clopidogrel (54% to 74%, P b .001), angiotensin receptor blockers (9% to 12%, P b .001), and statins (64% to 74%, P b .001). There was no significant change in angiotensin-converting enzyme inhibitor use between the 2 periods.


294 Edwards et al

Figure 1

Early BB use by ACS type before and after publication of COMMIT/ CCS-2.

Outcomes according to early BB use The group of patients who did not receive early BB therapy had significantly higher unadjusted in-hospital mortality rate than those who received early BB therapy (8.4% vs 2.2%, P b .001). After adjusting for components of the GRACE score, early BB use was not independently associated with higher in-hospital mortality (odds ratio [OR] 0.54, 95% CI 0.40-0.72, P b .001). Multivariable analysis Long-term BB use, higher HR and SBP, positive biomarkers, and ST-segment deviation at the time of presentation were independent predictors of early BB use (Table III). Conversely, female gender, increasing age, cardiac arrest, and Killip class N1 were independent negative predictors of early BB therapy. After adjusting for patient characteristics in multivariable analysis, there was no significant change in the overall use of early BB therapy in the years 2006 to 2008 compared to 1999 to 2005. However, the temporal trends in the early use of BB differed between patients with STEMI and patients with NSTEACS (P for interaction with period b.001)—there was a trend toward increased use of early BB therapy in patients with NSTEACS and a trend toward decreased use in patients with STEMI. The multivariable analysis was repeated excluding the patients enrolled between 1999 and 2002. For the NSTEACS group, there was no significant change in the use of early BB therapy between the periods 2003 to 2005 and 2006 to 2008 (adjusted OR 1.11, 95% CI 0.87-1.42, P = .40). Similarly, in the STEMI group, there was no significant change either in early BB use between the 2 periods (adjusted OR 0.75, 95% CI 0.56-1.01, P = .057).

Discussion This study demonstrates that most patients with ACS received early BB therapy in the past decade. In accordance with the updated ACC/AHA guidelines for the management of patients with STEMI, patients who were hypotensive or had evidence of heart failure at presentation were less often treated with early BB therapy. Since the publication of COMMIT/CCS-2 in 2005, there has been no significant change in early BB use among patients with STEMI or NSTEACS independent of their clinical characteristics. There have been multiple randomized controlled trials investigating BB use in patients with ACS.14-16 Yet, the optimal timing to initiate BB therapy has been controversial. The effect of early BB therapy on mortality was directly assessed in the COMMIT/CCS-2 trial, which was published in 2005.4 The COMMIT/CCS-2 trial randomized 45,852 patients with suspected ACS and ST-segment deviation or left bundle branch block to receive metoprolol or placebo. Patients in the metoprolol arm received up to 15 mg of IV metoprolol in 3 divided doses at 2- to 3-minute intervals, followed by 200 mg of metoprolol daily. There was no significant difference in mortality between the metoprolol and placebo groups. In the metoprolol arm, there was a trend toward increased mortality in the subgroup of patients who had evidence of hemodynamic instability at presentation, which was opposite to a trend toward decreased mortality in patients who were hemodynamically stable at presentation.4,17 In 2007, the ACC/AHA updated the 2004 STEMI management guidelines to caution against the early use of both oral and IV BB therapy in patients with signs of heart failure, evidence of a low output state, increased risk for cardiogenic shock, or other contraindications to BB therapy.5,18 American College of Cardiology/American Heart Association included a similar caution in their guidelines for the management of NSTEACS.6

Factors associated with early BB therapy Our study found that between 1999 and 2008, 78% of the Canadian patients with ACS enrolled in GRACE/ GRACE2/CANRACE received early BB therapy. This is comparable to the results of Emery et al7 who examined the early use of BB in patients with NSTEMI in the international GRACE up to the year 2004 (before the publication of COMMIT/CCS-2) and found that 76% received early BB therapy. However, Emery et al7 did not specifically investigate the use of early BB therapy in hypotensive patients or in patients with STEMI. In our study, patients who did not receive early BB therapy were older and more likely to be female, similar to other studies investigating the use of evidence-based therapies in patients with ACS.7,19-25 Long-term BB use was much higher in the group that received early BB therapy, suggesting that prior BB use may have


Edwards et al 295

Figure 2

Early BB use according to baseline characteristics, before and after publication of COMMIT/CCS-2.

Table III. Independent predictors of early BB use

Age, per decade increase Female HR 60-109 beat/min b60 beat/min ≥110 beat/min SBP b120 mm Hg Positive initial biomarker History of MI History of CHF Killip class I II III or IV Creatinine, per 10 μmol/L increase Cardiac arrest Long-term BB use ST-segment deviation Long-term BB use and history of MI⁎ Long-term BB use and history of CHF⁎ STEMI† 1999-2005 2006-2008 NSTEACS† 1999-2005 2006-2008

OR

95% CI

P

0.82 0.78

0.78-0.86 0.67-0.91

b.001 .002

reference 0.62 0.93 0.69 1.36 0.72 0.65

0.55-0.71 0.79-1.10 0.60-0.79 1.20-1.54 0.61-0.85 0.52-0.82

b.001 .41 b.001 b.001 b.001 b.001

reference 0.68 0.54 0.98 0.46 4.17 1.15 1.71 1.37

0.57-0.82 0.40-0.71 0.97-0.99 0.30-0.70 2.58-6.74 1.00-1.31 1.29-2.26 1.06-1.77

b.001 b.001 b.001 b.001 b.001 .047 b.001 .017

reference 0.76

0.57-1.00

.055

reference 1.22

0.96-1.55

.11

⁎ Interaction terms. † P for interaction with period b.001.

influenced physicians' decision to administer early BB therapy. Possibly, physicians felt reassured that the patient was tolerating BB therapy or concerned that discontinuing therapy may lead to worse outcomes, with

higher reinfarction rates and larger infarct sizes. Increasing HR and SBP were both found to be independent predictors of early BB use, a result that may be intuitive given that BBs lower both HR and SBP and may result in hemodynamic compromise in bradycardic and hypotensive patients. Conversely, cardiogenic shock and increasing Killip class were both negative predictors of early BB use.

Impact of the publication of COMMIT/CCS-2 on the use of early BB therapy We demonstrated that, overall, there was no significant change in the use of early BB therapy after the publication of COMMIT/CCS-2. In COMMIT/CCS-2, the investigators found no net benefit of early BB therapy. Although the trial included both patients with STEMI and NSTEMI with ST-segment deviation, most (93%) of the patients in COMMIT/CCS-2 had STEMI.4 Thus, the publication of COMMIT/CCS-2 may affect early BB use in patients with STEMI differently compared to patients with NSTEACS. Indeed, we found that there has been a nonsignificant increasing trend in early BB use among patients with NSTEACS, which is opposite to the nonsignificant decreasing trend in BB use among patients with STEMI. This temporal difference between the 2 groups was significant (P for interaction b.001). In COMMIT/CCS-2, the authors suggest that patients with hemodynamic compromise were most susceptible to adverse outcomes with early BB therapy. They found that patients who had an SBP b120 mm Hg or Killip class III had an increased risk of developing cardiogenic shock when allocated to the metoprolol group.4 In our study, we stratified patients based upon age N70 years, SBP b120 mm


296 Edwards et al

Hg, HR b60 or N110 beat/min, Killip class N1,10 and delayed presentation N13 hours. We found that, despite the publication of COMMIT/CCS-2 in 2005, there was no significant change in the use of early BB therapy in these subgroups who may be at increased risk for an adverse outcome associated with early BB therapy. Somewhat surprisingly, there may have been a trend toward an increased use of early BB therapy in these subgroups. Whether this trend to increase use of early BB therapy in high-risk patients is inappropriate deserves further study.

Outcomes of early BB use One of the criticisms about COMMIT/CCS-2 is that the dose of the metoprolol may have been excessive, and all patients in the BB arm received IV metoprolol in addition to oral metoprolol. In contrast, in GRACE/GRACE2/ CANRACE, physicians were not constrained by predefined medication route or dosage, as might be in a randomized controlled trial. As such, if physicians elected to use BB within the first 24 hours, they would have been able to tailor the dosage and route to individual patients. This may have attenuated any significant increase in the risk of adverse events associated with the early and aggressive use of BB in COMMIT/CCS-2. Our observational study was not designed to assess the treatment efficacy or risk of harm of early BB therapy. Such assessment is best accomplished via randomized controlled trials such as COMMIT/CCS-2. Nevertheless, randomized controlled trials may be less generalizable to unselected patients and are less useful for assessing treatment effectiveness (eg, with a titrated dose and different route of administration) in the real world. In our study, we found that patients who received early BB therapy had a significantly lower in-hospital mortality rate, even in patients who had an SBP b120 mm Hg or Killip class N1. There was also a significantly lower incidence of cardiogenic shock, sustained ventricular arrhythmia, and acute CHF in patients who received early BB therapy. Miller et al12 also reported an independent association between BB use and better clinical outcomes in the large CRUSADE database, and similar results have been shown in other observational studies.26-29 Based upon these observational data, early BB therapy did not appear to be associated with worse outcomes, suggesting that physicians may have carefully selected patients for early BB therapy use and tailored treatment appropriately. Importantly, our observational study serves as an illustrative example of how even large and wellconducted pragmatic trials may have limited impact on real-world practice. Several large pragmatic trials have revolutionized patient care in ACS.30,31 However, occasionally, when pragmatic trials (such as COMMIT) fail to demonstrate treatment efficacy, it may merely reflect failure to appropriately tailor therapy, which mandates careful patient selection and dose titration to

be effective. It is plausible that physicians consider the findings of the COMMIT inapplicable to their selected patients with ACS who are treated with different regimens of BB.

Study limitations Several limitations in this study should be mentioned. First, our study was a retrospective analysis using data from large registries. As such, this observational study cannot assess treatment efficacy and establish any causeand-effect relationship. However, it does allow for accurate assessment of real-world practice patterns. Second, we did not collect data on the route of administration and the dose of early BB therapy, which might vary according to the physician's practice and the patient's clinical presentation. Third, we did not specifically explore the reasons for administering or withholding BB therapy (eg, history of asthma or chronic obstructive pulmonary disease). Fourth, there may be a longer time lag beyond 2008 from the publication of COMMIT/CCS-2 to a potential change in clinical practice. Nevertheless, our study may serve as an important benchmark for future analysis of usage trends. Finally, although the registries aimed to recruit unselected consecutive patients, we could not determine how successful the registries were at attaining this goal. For example, patients dying early (before or shortly after admission) were more likely to be excluded because of insufficient time to obtain informed consent, leading to a selection bias toward less sick patients.

Conclusion Most Canadian patients with ACS were treated with early BB therapy over the past decade, irrespective of the type of ACS. In accordance with the ACC/AHA NSTEACS and STEMI management guidelines, we found that patients with lower SBP and higher Killip class in the real world less frequently received early BB therapy. Patients who were already on BB therapy at the time of presentation were more likely to receive early BB therapy. Since the publication of COMMIT/CCS-2 in 2005, there has been no significant change in the use of BB in patients with STEMI or NSTEACS after controlling for clinical characteristics.

Acknowledgements We thank Sue Francis for her assistance with manuscript preparation and all the study investigators, coordinators, and patients who participated in GRACE/ GRACE-2 and CANRACE. Dr. Andrew Yan is supported by a New Investigator Award from the Heart and Stroke Foundation of Canada.


Edwards et al 297

References 1. Lopez-Sendon J, Swedberg K, McMurray J, et al. Expert consensus document on beta-adrenergic receptor blockers. Eur Heart J 2004; 25:1341-62. 2. Doughty RN, Whalley GA, Walsh HA, et al. Effects of carvedilol on left ventricular remodeling after acute myocardial infarction: the CAPRICORN echo substudy. Circulation 2004; 109:201-6. 3. Galcerá-Tomás J, Castillo-Soria FJ, Villegas-García MM, et al. Effects of early use of atenolol or captopril on infarct size and ventricular volume: a double-blind comparison in patients with anterior acute myocardial infarction. Circulation 2001;103:813-9. 4. Chen ZM, Pan HC, Chen YP, et al. COMMIT (ClOpidogrel and Metoprolol in Myocardial Infarction Trial) collaborative group. Early intravenous then oral metoprolol in 45,852 patients with acute myocardial infarction: randomised placebo-controlled trial. Lancet 2005;366:1622-32. 5. Antman EM, Hand M, Armstrong PW, et al. 2007 focused update of the ACC/AHA 2004 guidelines for the management of patients with ST-elevation myocardial infarction. Circulation 2008; 117:296-329. 6. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/ non–ST-elevation myocardial infarction. J Am Coll Cardiol 2007;50: e1-e157. 7. Emery M, López-Sendón J, Steg PG, et al. Patterns of use and potential impact of early beta-blocker therapy in non–ST-elevation myocardial infarction with and without heart failure: the Global Registry of Acute Coronary Events. Am Heart J 2006;152: 1015-21. 8. 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;141:190-9. 9. Goodman SG, Huang W, Yan AT, et al. The expanded Global Registry of Acute Coronary Events: baseline characteristics, management practices, and hospital outcomes of patients with acute coronary syndromes. Am Heart J 2009;158:193-201. 10. Killip T, Kimball JT. Treatment of myocardial infarction in a coronary care unit. A two-year experience with 250 patients. Am J Cardiol 1967;20:457-64. 11. Yan AT, Steg PG, FitzGerald G, et al. Recurrent ischemia across the spectrum of acute coronary syndromes: prevalence and prognostic significance of (re-)infarction and ST-segment changes in a large contemporary registry. Int J Cardiol 2010;145:15-20. 12. Miller CD, Roe MT, Mulgund J, et al. Impact of acute beta-blocker therapy for patients with non–ST-segment elevation myocardial infarction. Am J Med 2007;120:685-92. 13. Silvet H, Spencer F, Yarzebski J, et al. Community-wide trends in the use and outcomes associated with beta-blockers in patients with acute myocardial infarction: the Worcester Heart Attack Study. Arch Intern Med 2003;163:2175-83. 14. Freemantle N, Cleland J, Young P, et al. Beta-blockade after myocardial infarction: systematic review and meta regression analysis. BMJ 1999;318:1730-7. 15. Roberts R, Rogers WJ, Mueller HS, et al. Immediate versus deferred beta-blockade following thrombolytic therapy in patients with acute myocardial infarction. Results of the Thrombolysis In

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

Myocardial Infarction (TIMI) II-B study. Circulation 1991;83: 422-37. Randomised trial of intravenous atenolol among 16027 cases of suspected acute myocardial infarction: ISIS-1. First International Study of Infarct Survival collaborative group. Lancet 1986;2: 57-66. Roe MT, Chen AY, Riba AL, et al. Impact of congestive heart failure in patients with non–ST-segment elevation acute coronary syndromes. Am J Cardiol 2006;97:1707-12. Salpeter SR, Ormiston TM, Salpeter EE. Cardioselective beta-blockers for chronic obstructive pulmonary disease. Cochrane Database of Syst Rev 2005:CD003566, doi:10.1002/14651858.CD003566. pub2. Yan AT, Yan RT, Tan M, et al. Management patterns in relation to risk stratification among patients with non–ST-elevation acute coronary syndromes. Arch Intern Med 2007;167:1009-16. Mehta RH, Roe MT, Chen AY, et al. Recent trends in the care of patients with non–ST-segment elevation acute coronary syndromes: insights from the CRUSADE initiative. Arch Intern Med 2006;166: 2027-34. Spencer F, Scleparis G, Goldberg RJ, et al. Decade-long trends (1986 to 1997) in the medical treatment of patients with acute myocardial infarction: a community-wide perspective. Am Heart J 2001;142: 594-603. Eagle KA, Kline-Rogers E, Goodman SG, et al. Adherence to evidence-based therapies after discharge for acute coronary syndromes: an ongoing prospective, observational study. Am J Med 2004;117:73-81. Yan AT, Yan RT, Tan M, et al. Optimal medical therapy at discharge in patients with acute coronary syndromes: temporal changes, characteristics, and 1-year outcome. Am Heart J 2007;154: 1108-15. Gurwitz JH, Goldberg RJ, Chen Z, et al. Beta-blocker therapy in acute myocardial infarction: evidence for underutilization in the elderly. Am J Med 1992;93:605-10. Rathore SS, Mehta RH, Wang Y, et al. Effects of age on the quality of care provided to older patients with acute myocardial infarction. Am J Med 2003;114:307-15. Cuculi F, Radovanovic D, Pedrazzini G, et al. Is pretreatment with beta-blockers beneficial in patients with acute coronary syndrome? Cardiology 2009;115:91-7. Granger CB, Steg PG, Peterson E, et al. Medication performance measures and mortality following acute coronary syndromes. Am J Med 2005;118:858-65. Krumholz HM, Radford MJ, Wang Y, et al. Early beta-blocker therapy for acute myocardial infarction in elderly patients. Ann Intern Med 1999;131:648-54. Allen LaPointe NM, Chen AY, Roe MT, et al. Relation of patient age and mortality to reported contraindications to early beta-blocker use for non–ST-elevation acute coronary syndrome. Am J Cardiol 2009; 104:1324-9. Tunis SR, Stryer DB, Clancy CM. Practical clinical trials: increasing the value of clinical research for decision-making in clinical and health policy. JAMA 2003;290:1624-32. Baigent C, Collins R, Appleby P, et al. ISIS-2: 10-year survival among patients with suspected acute myocardial infarction in randomised comparison of intravenous streptokinase, oral aspirin, both, or neither. The ISIS-2 (Second International Study of Infarct Survival) collaborative group. BMJ 1998;316:1337-43.

Incidence and clinical consequences of acquired thrombocytopenia after antithrombotic therapies in patients with acute coronary syndromes: Results from the Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) trial Adriano Caixeta, MD, PhD, a,c,j George D. Dangas, MD, PhD, a,b,j Roxana Mehran, MD, a,b,j Frederick Feit, MD, e,j Eugenia Nikolsky, MD, PhD, a,c,j Alexandra J. Lansky, MD, d,j Jiro Aoki, MD, PhD, a,c,j Jeffrey W. Moses, MD, a,c,j Steven R. Steinhubl, MD, f,j Harvey D. White, DSc, g,j E. Magnus Ohman, MD, h,j Steven V. Manoukian, MD, i,j Martin Fahy, MSc, c,j and Gregg W. Stone, MD a,c,j New York, NY; New Haven, CT; Lexington, KY; Auckland, New Zealand; Durham, NC; and Nashville, TN

Background The aim of the study was to investigate the incidence and clinical consequences of acquired thrombocytopenia in patients with acute coronary syndromes (ACS) in the ACUITY trial. Methods We examined 10,836 patients with ACS randomized to receive heparin plus glycoprotein (GP) IIb/IIIa inhibitor, bivalirudin plus GP IIb/IIIa inhibitor, or bivalirudin monotherapy. Results

Acquired thrombocytopenia developed in 740 (6.8%) patients; mild (100,000-150,000 platelets/mm3), moderate (50,000-100,000 platelets/mm3), and severe (b50,000 platelets/mm3) developed in 656 (6%), 51 (0.5%), and 33 (0.3%) patients, respectively. Patients with acquired thrombocytopenia, compared with those without, were more likely to develop major bleeding (14% vs 4.3%, P b .0001) at 30 days and had higher rates of mortality (6.5% vs 3.4%, P b .0001) at 1 year. By multivariate analysis, acquired thrombocytopenia was an independent predictor of major bleeding at 30 days (hazard ratio [HR] 1.68, 95% CI 1.04-2.72, P = .03). Moderate and severe acquired thrombocytopenia were predictors of mortality at 1 year (HR 2.89, 95% CI 0.92-9.06, P = .06, and HR 3.41, 95% CI 1.09-10.68, P = .03, respectively). Compared to heparin plus GP IIb/IIIa inhibitor, bivalirudin monotherapy was associated with less declines in platelet count by N25% (7.6% vs 5.6%, P = .0009) and N50% (1.4% vs 0.7%, P = .004) from baseline.

Conclusions Acquired thrombocytopenia occurs in approximately 1 in 14 patients with ACS treated with antithrombin and antiplatelet medications and is strongly associated with hemorrhagic and ischemic complications. Compared to an anticoagulant regimen including a GP IIb/IIIa inhibitor, administration of bivalirudin monotherapy appears to be associated with less frequent declines in platelet count. (Am Heart J 2011;161:298-306.e1.)

From the aCardiovascular Research Foundation, New York, NY, bMount Sinai Medical Center, New York, NY, cColumbia University Medical Center, New York, NY, dYale University Medical Center, New Haven, CT, eNew York University School of Medicine, New York, NY, fUniversity of Kentucky, Lexington, KY, gAuckland City Hospital, Auckland, New Zealand, hDuke University Medical Center, Durham, NC, and iSarah Cannon Research Institute and the Hospital Corporation of America, Nashville, TN. j For the ACUITY trial investigators. RCT reg #NCT00093158. Submitted March 8, 2010; accepted October 29, 2010. Reprint requests: George D. Dangas, MD, PhD, Cardiovascular Institute (Box 1030), Mount Sinai Medical Center, One Gustave L. Levy Place, New York, NY 10029. E-mails: [email protected], [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2010.10.035

Current guidelines for patients with moderate- or high-risk acute coronary syndromes (ACS) recommend early invasive management with concomitant antithrombotic therapy, including aspirin, clopidogrel, heparin plus glycoprotein (GP) IIb/IIIa inhibitor, or, as an alternative, bivalirudin.1,2 However, the antithrombotic therapy combination of heparin and GP IIb/IIIa inhibitor use may cause acquired thrombocytopenia and has been strongly associated with increased risks of hemorrhagic and ischemic complications, as well as early and late mortality.3-9 The direct thrombin inhibitor bivalirudin (Angiomax; The Medicines Company, Parsippany, NJ) is an antithrombotic drug, indicated for


Caixeta et al 299

Table I. Baseline and procedural characteristics of patients with and without thrombocytopenia

Age (y), median (IQR) Male Renal insufficiency⁎ Diabetes mellitus Current smoking Previous MI Previous PCI Previous CABG Hypertension Hyperlipidemia Platelet count (×103 platelets/mm3), median (IQR) Creatinine clearance (mL/min), median (IQR) Cardiac biomarker elevation† TIMI risk score Low (0-2) Intermediate (3-4) High (5-7) Medications before hospitalization Aspirin Thienopyridines Antithrombin medications Prerandomization UFH LMWH Antiplatelet medication – Preintervention Aspirin Thienopyridine Randomization Heparin plus GP IIb/IIIa inhibitor Bivalirudin plus GP IIb/IIIa inhibitor Bivalirudin monotherapy Duration from first study drug to first actual PCI (h), median (IQR) Treatment strategy triage PCI Medical therapy

Thrombocytopenia (n = 740)

No thrombocytopenia (n = 10096)

P

67.00 (57.50-74.00) 76.5% 25.1% 27.5% 27.4% 36.9% 43.5% 25.9% 71.1% 60.3% 173.00 (160.00-192.50) 80.11 (59.95-106.48) 59.5%

62.00 (53.00-71.00) 67.6% 18.5% 26.6% 29.8% 31.1% 40.5% 18.4% 66.4% 57.2% 236.00 (202.00-278.00) 88.25 (66.30-114.10) 57.5%

b.0001 b.0001 b.0001 .57 .18 .001 .11 b.0001 .009 .11 b.0001 b.0001 .31

10.1% 54.5% 35.4%

17.2% 54.9% 27.9%

b.0001 .87 b.0001

72.3% 24.2%

69.7% 25.1%

.13 .63

62.0% 39.9% 24.3%

64.2% 40.3% 25.9%

.25 .81 .34

97.7% 67.7%

98.0% 65.6%

.49 .26

34.7% 34.5% 30.8% 4.93 (1.35-20.47)

32.9% 33.4% 33.8% 4.00 (1.33-19.23)

.29 .54 .11 .07

75.9% 24.1%

62.8% 37.2%

b.0001 b.0001

IQR indicates interquartile range; heparin, unfractionated heparin (UFH) or low molecular weight heparin (LMWH) at site discretion. ⁎ Renal insufficiency was defined as a calculated creatinine clearance rate of b60 mL/min as determined by the Cockcroft-Gault equation. † Creatine kinase–MB/troponin I or T elevated.

use in patients with heparin-induced thrombocytopenia undergoing angioplasty,1,2 that has been recently used in patients with ACS and those undergoing percutaneous coronary intervention (PCI).10 The incidence of thrombocytopenia after bivalirudin treatment was significantly lower than heparin plus GP IIb/IIIa inhibitor treatment in REPLACE-2 trial (0.7% vs 1.7%), which included relatively low-risk patients undergoing PCI.10 It is unknown whether bivalirudin reduces the incidence of acquired thrombocytopenia in moderateto high-risk patients with ACS, for whom early invasive therapy is planned, and whether the severity of acquired thrombocytopenia impacts on clinical outcomes. Furthermore, in addition to possible direct effects, declines in platelet count during ACS may also have implications for antiplatelet medication, prescription, and adherence.

The present study evaluated the incidence and clinical consequences of acquired thrombocytopenia in the largescale, contemporary, randomized ACUITY trial.

Methods ACUITY was a prospective, open-label, randomized, multicenter trial. The study design and principal results have been previously described in detail.11 In brief, 13,819 patients with moderate- and high-risk ACS were assigned by a primary randomization to 1 of 3 antithrombin regimens started before angiography: heparin (unfractionated heparin or enoxaparin) plus GP IIb/IIIa inhibitor, bivalirudin plus GP IIb/IIIa inhibitor, or bivalirudin monotherapy. In the bivalirudin monotherapy group, provisional GP IIb/IIIa inhibitor was used in 7% of the patients.11 Angiography was performed in all patients within 72 hours after randomization. Patients were then triaged to PCI, coronary artery bypass grafting (CABG), or medical


300 Caixeta et al

Table II. Thirty-day and 1-year adverse clinical events Thrombocytopenia (n = 740) 30 Days NACE Composite ischemia Death/MI Death Cardiac death Noncardiac death MI Q wave MI Non–q wave MI Unplanned revascularization Non-CABG major bleeding Non-CABG major bleeding (excluding hematoma ≥ 5 cm) Non-CABG related blood transfusion Non-CABG minor bleeding TIMI non-CABG major bleeding TIMI non-CABG minor bleeding Definite/probable stent thrombosis 1 Year Composite ischemia Death/MI Death Cardiac death Noncardiac death MI Q-wave MI Non–Q-wave MI Unplanned revascularization Definite/probable stent thrombosis

No thrombocytopenia (n = 10096)

P

21.7% 12.5% 9.3% 3.1% 2.2% 0.7% 7.5% 1.9% 5.6% 5.3% 14.0% 12.4% 8.9% 30.2% 5.6% 15.2% 1.6%

(160) (92) (69) (23) (16) (5) (55) (14) (41) (39) (103) (91) (66) (223) (41) (112) (12)

9.7% 6.3% 4.9% 1.1% 0.8% 0.1% 4.1% 0.6% 3.5% 2.4% 4.3% 3.4% 1.3% 18.7% 1.3% 4.6% 0.8%

(975) (632) (494) (106) (83) (14) (408) (63) (347) (239) (429) (339) (135) (1886) (127) (463) (78)

b.0001 b.0001 b.0001 b.0001 .0002 .0007 b.0001 b.0001 .0025 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001 .01

22.8% 14.2% 6.5% 3.5% 1.9% 10.0% 2.5% 7.6% 13.8% 2.4%

(162) (100) (43) (23) (13) (72) (17) (55) (96) (17)

15.1% 9.1% 3.4% 1.9% 1.0% 6.4% 1.2% 5.3% 9.1% 1.2%

(1446) (872) (309) (177) (95) (622) (108) (521) (860) (112)

b.0001 b.0001 b.0001 .0070 .02 b.0001 .002 .006 b.0001 .003

NACE, Net adverse clinical events, indicating composite ischemia or major bleeding; composite ischemia, death from any cause, MI, or unplanned revascularization for ischemia. Results are given as Kaplan Meier rates (n). Stent thrombosis was defined according to Academic Research Consortium criteria.

management at the discretion of the physician. All antithrombotic agents were discontinued at the completion of angiography or PCI according to the protocol. Aspirin 300 to 325 mg/d orally or 250 to 500 mg/d intravenously was administered during the index hospitalization, and 75 to 325 mg/d was prescribed indefinitely after discharge. Clopidogrel 75 mg/d was recommended for 1 year in all patients with coronary artery disease. The study was approved by the institutional review board or ethics committee at each participating center, and all patients provided written informed consent.

Acquired thrombocytopenia Patients with baseline thrombocytopenia (platelet count b150,000 platelets/mm3) and patients triaged to CABG, for having a high prevalence of acquired thrombocytopenia secondary to the use of extracorporeal circulation, were excluded. Acquired thrombocytopenia was defined as an inhospital nadir platelet count of b150,000 platelets/mm3 (referenced lower limit of normal).12 Thrombocytopenia was classified as mild (100,000-150,000 platelets/mm3), moderate (50,000-100,000 platelets/mm3), or severe (b50,000 platelets/ mm3). A second definition of acquired thrombocytopenia, using a reduction platelet count from baseline (either N25% or N50% reduction)13 was also reported in patients after assignment to antithrombin therapy. Platelet count was analyzed before

angiography and every 24 hours up to 24 hours after PCI or after angiography in patients triaged to medical therapy.

Clinical end points Clinical end points were evaluated at 30 days and 1 year as previously described.11 Major bleeding was defined as the cumulative occurrence within 1 year after randomization of intracranial or intraocular bleeding, hemorrhage at the access site requiring intervention, hematoma with a diameter of at least 5 cm, a reduction in hemoglobin level of at least 4 g/dL without an overt bleeding source or at least 3 g/dL with such a source, reoperation for bleeding, or transfusion of a blood product. Bleeding was also classified according to the criteria of the TIMI study group.14 A clinical events committee blinded to treatment assignment adjudicated all 30-day and 1-year primary end point events using original source documents.

Statistical analysis Continuous variables were expressed as mean and SD and compared using the Student t test or Wilcoxon rank sum test if applicable. Discrete variables were presented as numbers and percentages and compared with the χ2 test, unless the observation in any cell was b5, in which case Fisher exact test was used. Univariable and multivariable analyses by logistic regression were performed to identify the predictors of absolute (nadir platelet count of b150,000 platelets/mm3)


Caixeta et al 301

Figure 1

Kaplan-Meier event curves for patients with and without thrombocytopenia. One-year cumulative event curves for death (A), myocardial infarction (B), and composite ischemic events (C); 30-day cumulative event curves for non–CABG-related major bleeding (D).

and relative acquired thrombocytopenia (drop in platelet count N25% or N50%). Time-dependent Cox regression models, where thrombocytopenia is included in the model as a time-dependent covariate, were performed to identify the predictors of non-CABG major bleeding and death. The multivariable model was built by stepwise variable selection with entry and exit criteria set at the P = .2 and P = .1 levels, respectively. All variables in Table I were considered for the

univariate selection. Time-to-event distributions were displayed according to the Kaplan-Meier method and were compared with the use of the log-rank test. All statistical tests were 2-tailed. Probability was considered significant at a level of b.05. No extramural funding was used to support this work. The ACUITY trial was sponsored by the Medicines Company and


302 Caixeta et al

Figure 1

(continued).

Nycomed. The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the manuscript, and its final contents.

Results After exclusion of 1,625 patients with thrombocytopenia at baseline (platelet count b150,000 platelets/mm3) and 1,358 patients triaged to CABG, 10,836 patients from

the ACUITY trial were included in the present analysis. During hospitalization, acquired thrombocytopenia developed in 740 patients (6.8%). Mild, moderate, and severe acquired thrombocytopenia developed in 656 (6%), 51 (0.5%), and 33 (0.3%) patients, respectively. Patients who developed thrombocytopenia were more likely to be older; men; and have impaired creatinine clearance, lower platelet count at baseline, prior myocardial infarction (MI), prior CABG, and higher TIMI risk


Caixeta et al 303

Table III. Incidence of acquired thrombocytopenia (a) Heparin plus (b) Bivalirudin plus GP IIb/IIIa GP IIb/IIIa inhibitor inhibitor In-hospital Any thrombocytopenia 7.2% Mild 6.2% Moderate 0.7% Severe 0.4% 30-Day follow-up Any thrombocytopenia 7.3% Mild 6.1% Moderate 0.8% Severe 0.4% Reduction in platelet count from baseline N25% 7.6% (271/3574) N50% 1.4% (50/3574)

(c) Bivalirudin monotherapy

P (all groups) P (a) vs (b) P (a) vs (c) P (b) vs (c)

7.0% 6.2% 0.4% 0.4%

6.3% 5.8% 0.3% 0.2%

.25 .74 .07 .17

.79 .93 .19 .88

.12 .53 .02 .1

.19 .47 .33 .07

7.2% 6.4% 0.5% 0.4%

6.4% 5.9% 0.4% 0.2%

.26 .71 .06 .12

.89 .68 .12 .83

.13 .67 .02 .04

.17 .40 .46 .07

7.4% (270/3626) 1.1% (41/3626)

5.6% (205/3636) 0.7% (26/3636)

.001 .018

.83 .31

.0009 .004

.002 .06

In-hospital outcomes are summarized as % and compared between groups using χ2 or Fisher exact tests when appropriate. “30 Days” is summarized as Kaplan-Meier % and compared between groups with log-rank tests.

score. The use of aspirin before randomization was similar between the 2 groups. Patients undergoing PCI were more likely to develop thrombocytopenia than those triaged to medical therapy (Table I).

Clinical outcomes stratified by acquired thrombocytopenia Compared to patients without acquired thrombocytopenia, those with this complication had higher rates of mortality, MI, unplanned revascularization for ischemia, stent thrombosis, and composite ischemic events at 30 days and 1 year (Table II). In addition, the severity of acquired thrombocytopenia strongly correlated with adverse clinical outcomes at 30-day and 1-year followup (Figure 1). The 30-day mortality rate in patients without acquired thrombocytopenia was 1.1% versus 2.5%, 7.8%, and 9.1% in those who developed mild, moderate, and severe thrombocytopenia, respectively (log-rank P b .001). The differences in mortality among the groups were maintained at 1-year follow-up: 3.4% in patients without acquired thrombocytopenia versus 6.1%, 9.9%, and 9.1% in those who developed mild, moderate, and severe thrombocytopenia, respectively (log-rank P b .001) (Figure 1A). Similarly, compared with patients without acquired thrombocytopenia, the rates of MI and composite events were approximately 2-, 3-, and 4-fold higher in patients who developed mild, moderate, and severe thrombocytopenia, respectively (Figures 1B and 1C). Patients who had acquired thrombocytopenia were more likely to develop protocol-defined and TIMI-criteria major bleeding complications and to receive blood product transfusions (Table II). In addition, the severity of acquired thrombocytopenia strongly correlated with non-CABG major bleeding at 30 days (Figure 1D). Patients with mild or moderate acquired thrombocytopenia had a 3-fold higher incidence of non-CABG major bleeding, and

those who developed severe acquired thrombocytopenia had a 7-fold higher rate of non-CABG major bleeding at 30-day follow-up.

Incidence of acquired thrombocytopenia by randomized antithrombin therapy As shown in Table III, the incidence of acquired thrombocytopenia was comparable between bivalirudin plus GP IIb/IIIa inhibitor therapy and heparin plus GP IIb/IIIa inhibitor. Compared to heparin plus GP IIb/ IIIa inhibitor, any acquired thrombocytopenia after bivalirudin monotherapy was numerically reduced, with significant reductions in moderate acquired thrombocytopenia (0.7 vs 0.3, P = .02). Furthermore, compared to heparin plus GP IIb/IIIa inhibitors, the reduction of platelet count from baseline (either N25% or N50%) occurred significantly less often after bivalirudin monotherapy. By multivariable analysis, development of mild thrombocytopenia was an independent predictor of major bleeding at 30 days. Moderate and severe acquired thrombocytopenia were significantly associated with 1-year mortality (Table IV). Bivalirudin monotherapy was significantly associated with less frequent declines in platelet by N25% or N50% from baseline (Table V).

Discussion In this large-scale prospective, randomized trial of patients with moderate- or high-risk ACS, the principal findings are (1) occurring in approximately 1 in 14 patients, acquired thrombocytopenia develops frequently with an early invasive strategy and antithrombotic treatment; (2) the 30-day incidence of death, MI, and major bleeding doubled in patients with even mild acquired thrombocytopenia compared to patients without thrombocytopenia; (3) rates of mortality (9.1%) and


304 Caixeta et al

Table IV. Multivariable predictors of non-CABG major bleeding and death Hazard ratio (95% CI) Non-CABG major bleeding at 30 days Male 0.49 (0.40-0.60) PCI triage (vs medical treatment) 2.30 (1.81-2.93) Randomized to bivalirudin alone 0.49 (0.39-0.63) Age 1.02 (1.01-1.03) Renal insufficiency 1.60 (1.26-2.04) Prior PCI 0.68 (0.53-0.87) Diabetes 1.35 (1.09-1.66) ST-segment deviation ≥ 1 mm 1.45 (1.12-1.88) Mild thrombocytopenia (vs none) 1.68 (1.04-2.72) Moderate thrombocytopenia 1.09 (0.15-7.75) (vs none) Death at 1 year Age 1.07 (1.06-1.09) Diabetes 2.00 (1.55-2.60) PCI triage (vs medical treatment) 0.63 (0.49-0.81) Male 1.53 (1.16-2.01) Current smoker 1.64 (1.19-2.26) Severe thrombocytopenia (vs none) 3.41 (1.09-10.68) ST-segment deviation ≥1 mm 1.57 (0.99-2.51) Moderate thrombocytopenia 2.89 (0.92-9.06) (vs none) TIMI risk score high 1.28 (0.98-1.67) Elevated biomarkers 1.41 (0.96-2.08) Renal insufficiency 1.29 (0.96-1.74) Mild thrombocytopenia (vs none) 1.30 (0.84-2.02)

χ2

P

47.88 b.0001 45.96 b.0001 34.07 b.0001 16.36 .0001 15.07 .0001 9.37 .006 7.78 .005 7.75 .005 4.42 .03 0.01 .93

76.24 b.0001 27.60 b.0001 12.8 .0003 9.21 .002 9.19 .002 4.42 .03 3.60 .06 3.31 .06 3.26 2.99 2.79 1.37

.07 .08 .09 .24

Table V. Multivariable predictors of absolute and relative acquired thrombocytopenia during hospitalization Odds ratio (95% CI)

χ2

P

Predictors of acquired thrombocytopenia (

American Heart Journal, Volume 161, Issue 2, Feb 2011

Microwave Journal, February 2011, Volume. 54, Issue No.2

Popular Science Feb 2011

Superboy (Vol 4) #2, Feb 2011

New Avengers (Vol 2) #7 Feb 2011

Journal of Semantics, Volume 2, Issue 2, 1983

Israel Journal of Mathematics, Volume 2, Issue 2 (1964)

Computer Power User – 2011 Feb

Nutraceuticals World Jan Feb 2011

The Unwritten #20, Feb 2011

Brand Packaging Jan-Feb 2011

Israel Journal of Mathematics, Volume 5, Issue 2 (1967)

Israel Journal of Mathematics, Volume 2, Issue 3 (1964)

Israel Journal of Mathematics, Volume 1, Issue 2 (1963)

Journal of Semantics 1983 Volume 2 Issue 1

International Journal of Powder Metallurgy Volume 44 Issue 2

Cinema Journal, Volume 8, Issue 2, Spring, 1969

International Journal of Contemporary Iraqi Studies Volume 5 Issue 2

Journal of Semantics, Volume 2, Issue 3-4, 1983

International Journal of Powder Metallurgy Volume 46 Issue 2

Israel Journal of Mathematics, Volume 2, Issue 1 (1964)

Israel Journal of Mathematics, Volume 2, Issue 4 (1964)

Israel Journal of Mathematics, Volume 3, Issue 2 (1965)

International Journal of Powder Metallurgy Volume 45 Issue 2

Cultural Studies: Volume 3, Issue 2 (Cultural Studies Journal)

The Flash #7 Feb 2011

Israel Journal of Mathematics, Volume 4, Issue 2 (1966)

sports medicine Volume 41 Issue 2 February 2011

Journal of Bodywork and Movement Therapies, Volume 15, issue 2 (2011)

Energy Prices and Taxes, Volume 2011 Issue 2

American Heart Journal, Volume 161, Issue 2, Feb 2011

Microwave Journal, February 2011, Volume. 54, Issue No.2

Popular Science Feb 2011

Superboy (Vol 4) #2, Feb 2011

New Avengers (Vol 2) #7 Feb 2011

Journal of Semantics, Volume 2, Issue 2, 1983

Israel Journal of Mathematics, Volume 2, Issue 2 (1964)

Computer Power User – 2011 Feb

Nutraceuticals World Jan Feb 2011

The Unwritten #20, Feb 2011

Brand Packaging Jan-Feb 2011

Israel Journal of Mathematics, Volume 5, Issue 2 (1967)

Israel Journal of Mathematics, Volume 2, Issue 3 (1964)

Israel Journal of Mathematics, Volume 1, Issue 2 (1963)

Journal of Semantics 1983 Volume 2 Issue 1

International Journal of Powder Metallurgy Volume 44 Issue 2

Cinema Journal, Volume 8, Issue 2, Spring, 1969

International Journal of Contemporary Iraqi Studies Volume 5 Issue 2

Journal of Semantics, Volume 2, Issue 3-4, 1983

International Journal of Powder Metallurgy Volume 46 Issue 2

Israel Journal of Mathematics, Volume 2, Issue 1 (1964)

Israel Journal of Mathematics, Volume 2, Issue 4 (1964)

Israel Journal of Mathematics, Volume 3, Issue 2 (1965)

International Journal of Powder Metallurgy Volume 45 Issue 2

Cultural Studies: Volume 3, Issue 2 (Cultural Studies Journal)

The Flash #7 Feb 2011

Israel Journal of Mathematics, Volume 4, Issue 2 (1966)

sports medicine Volume 41 Issue 2 February 2011

Journal of Bodywork and Movement Therapies, Volume 15, issue 2 (2011)

Energy Prices and Taxes, Volume 2011 Issue 2

Recommend Documents