Risk of Thrombosis With the Use of Sirolimus-Eluting Stents for Percutaneous Coronary Intervention (from Registry and Clinical Trial Data) Anthony A. Bavry, MD, MPH, Dharam J. Kumbhani, MD, SM, Thomas J. Helton, DO, and Deepak L. Bhatt, MD We conducted a meta-analysis on 6 studies in 2,963 patients who had coronary artery disease and re- ceived a sirolimus-eluting stent or a bare metal stent for revascularization. Compared with bare metal stents, sirolimus-eluting stents did not appear to in- crease the risk for thrombosis up to 13.5 months after coronary intervention (risk ratio 0.49, 95% confi- dence interval 0.22 to 1.12, p  0.09). 2005 by Excerpta Medica Inc. (Am J Cardiol 2005;95:1469 –1472) D rug-eluting coronary stents dramatically decrease coronary restenosis and the subsequent need for repeat revascularization compared with bare metal stents. 1,2 One year after the release of an advisory that cautioned about an association between sirolimus- eluting stents (Cypher, Cordis Corp., Johnson & John- son, Miami Lakes, Florida) and the risk for thrombo- sis, lingering concerns remain. 3 A recently published case series documented the occurrence of late stent thrombosis after discontinuation of antiplatelet thera- py. 4 Accordingly, we sought to quantify systemati- cally the risk of thrombosis from the use of sirolimus- eluting stents using registry observations and randomized clinical trial data. ••• We searched the MEDLINE database for random- ized clinical trials from 2001 to 2004 using the Med- ical Subject Heading terms “angioplasty,” “translumi- nal,” “percutaneous coronary,” “stents,” “sirolimus,” and “thrombosis.” We also obtained recently pre- sented data at national cardiology conferences, corre- sponded with investigators and experts in the field, and used the Science Citation Index to cross-reference any articles that met our selection criteria. The inclusion criteria were (1) a randomized clin- ical trial that assigned patients to a sirolimus-eluting stent or a bare metal stent or (2) a cohort or registry that collected information on the performance of pa- tients who received sirolimus-eluting stents compared with those who received bare metal stents, and (3) angiographic or clinical thrombosis data that were available for 30 days of follow-up. Stent thrombosis was defined as an angiographically documented thrombosis or a cardiac event when the study investi- gators clinically presumed that a stent thrombosis had occurred, although angiographic data were unavail- able. Studies were excluded if both groups received drug-eluting stents (i.e., high- vs low-dose sirolimus). Two independent reviewers abstracted the follow- ing variables by intention-to-treat analysis in each study: (1) number of angiographically documented or clinical thromboses and time of the event after the index procedure, (2) number of patients at risk for stent-associated thromboses, (3) duration of dual an- tiplatelet therapy, and (4) extent of angiographic and clinical follow-up. Baseline demographic and angio- graphic characteristics of percutaneous coronary inter- vention were also abstracted. Discrepancies were re- solved through a third reviewer. To use data from all available studies, we consid- ered stent-associated thrombosis at any point in time up to the extent of clinical follow-up after the index procedure as the primary end point. We obtained a summary estimate of the risk ratio (RR) and 95% confidence interval (CI) for sirolimus- eluting versus bare metal stents. We used the auto- matic “zero cell” correction so that studies with no events in a given arm would still be included for analysis. The p values were 2-tailed, with statistical significance set at 0.05 (CIs were calculated at the 95% level). All analyses were performed with STATA 8.0. (STATA Corp., College Station, Texas). Baseline characteristics of participants in the 6 studies were similar (Table 1). 5–11 A total of 2,963 patients was enrolled from 1996 to March 2000. Of these, 1,515 received sirolimus-eluting stents (140 g of sirolimus per square centimeter) and 1,448 received bare metal stents. During the extent of clinical follow-up, 24 thromboses were identified, resulting in an overall incidence of 0.8% for stent-associated thrombosis. There were 8 and 16 thromboses in the sirolimus- eluting and bare metal stent groups, for incidences of stent-associated thrombosis 0.5% and 1.1%, re- spectively. No individual study showed a significant difference in risk of thrombosis. All studies provided angiographic follow-up data 6 to 8 months after the index procedure, and 66% to 95% of participants underwent follow-up angiogra- phy. In the Rapamycin-Eluting Stent Evaluated At Rotterdam Cardiology Hospital (RESEARCH) regis- try, angiographic follow-up was performed in 38% of participants who received a drug-eluting stent; routine angiographic follow-up was not performed in patients From the Department of Cardiovascular Medicine, Cleveland Clinic Foundation, Cleveland, Ohio; and the Department of Internal Medi- cine, University of Pennsylvania, Philadelphia, Pennsylvania. Dr. Bhatt’s address is: Department of Cardiovascular Medicine, Desk F25, Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44195. E-mail: bhattd@ccf.org. Manuscript received December 13, 2004; revised manuscript received and accepted February 10, 2005. 1469 ©2005 by Excerpta Medica Inc. All rights reserved. 0002-9149/05/$–see front matter The American Journal of Cardiology Vol. 95 June 15, 2005 doi:10.1016/j.amjcard.2005.02.015