Drug-Eluting versus Bare-Metal Stents in Large Coronary Arteries

Transcription

1 T h e n e w e ngl a nd j o u r na l o f m e dic i n e original article Drug-Eluting versus Bare-Metal Stents in Large Coronary Arteries Christoph Kaiser, M.D., Soeren Galatius, M.D., Paul Erne, M.D., Franz Eberli, M.D., Hannes Alber, M.D., Hans Rickli, M.D., Giovanni Pedrazzini, M.D., Burkhard Hornig, M.D., Osmund Bertel, M.D., Piero Bonetti, M.D., Stefano De Servi, M.D., Hans-Peter Brunner-La Rocca, M.D., Ingrid Ricard, Ph.D., and Matthias Pfisterer, M.D., for the BASKET PROVE Study Group* A bs tr ac t From University Hospital Basel, Basel (C.K., H.-P.B.-L., I.R., M.P.); State Hospital, Lucerne (P.E.); Triemli Hospital, Zurich (F.E.); State Hospital, St. Gallen (H.R.); Cardiocentro, Lugano (G.P.); Clara Hospital, Basel (B.H.); Cardiovascular Center Zurich, Zurich (O.B.); and State Hospital, Chur (P.B.) all in Switzerland; Gentofte University Hospital, Hellerup, Denmark (S.G.); University Hospital, Innsbruck, Austria (H.A.); Ospedale Civile di Legnano, Milan, Italy (S.D.S.); and Maastricht University Medical Center, Maastricht, the Netherlands (H.-P.B.-L.). Address reprint requests to Dr. Kaiser at the Department of Cardiology, University Hospital, CH 4031 Basel, Switzerland, or at ckaiser@ uhbs.ch. *Members of the Basel Stent Kosten- Effektivitäts Trial Prospective Validation Examination (BASKET PROVE) Study Group are listed in the Supplementary Appendix, available at NEJM.org. This article ( /NEJMoa ) was published on November 16, 2010, at NEJM.org. N Engl J Med 2010;363: Copyright 2010 Massachusetts Medical Society. Background Recent data have suggested that patients with coronary disease in large arteries are at increased risk for late cardiac events after percutaneous intervention with firstgeneration drug-eluting stents, as compared with bare-metal stents. We sought to confirm this observation and to assess whether this increase in risk was also seen with second-generation drug-eluting stents. Methods We randomly assigned 2314 patients needing stents that were 3.0 mm or more in diameter to receive sirolimus-eluting, everolimus-eluting, or bare-metal stents. The primary end point was the composite of death from cardiac causes or nonfatal myocardial infarction at 2 years. Late events (occurring during months 7 to 24) and targetvessel revascularization were the main secondary end points. Results The rates of the primary end point were 2.6% among patients receiving sirolimuseluting stents, 3.2% among those receiving everolimus-eluting stents, and 4.8% among those receiving bare-metal stents, with no significant differences between patients receiving either drug-eluting stent and those receiving bare-metal stents. There were also no significant between-group differences in the rate of late events or in the rate of death, myocardial infarction, or stent thrombosis. Rates of target-vessel revascularization for reasons unrelated to myocardial infarction were 3.7% among patients receiving sirolimus-eluting stents, 3.1% among those receiving everolimus-eluting stents, and 8.9% among those receiving bare-metal stents. The rate of target-vessel revascularization was significantly reduced among patients receiving either drugeluting stent, as compared with a bare-metal stent, with no significant difference between the two types of drug-eluting stents. Conclusions In patients requiring stenting of large coronary arteries, no significant differences were found among sirolimus-eluting, everolimus-eluting, and bare-metal stents with respect to the rate of death or myocardial infarction. With the two drug-eluting stents, similar reductions in rates of target-vessel revascularization were seen. (Funded by the Basel Cardiovascular Research Foundation and the Swiss National Foundation for Research; Current Controlled Trials number, ISRCTN ) 2310 n engl j med 363;24 nejm.org december 9, 2010

2 Drug-Eluting vs. Bare-Metal Stents in Large Arteries First-generation drug-eluting coronary-artery stents (which release sirolimus or paclitaxel) have been shown to reduce the risk of restenosis after percutaneous coronary intervention, as compared with bare-metal stents. However, these drug-eluting stents have also been associated with a risk of late stent thrombosis, which in turn may be associated with death from cardiac causes or nonfatal myocardial infarction. 1,2 For patients with stenoses in large coronary arteries, requiring stents of 3.0 mm or more in diameter, the benefits of the use of drug-eluting stents have been considered especially uncertain for two reasons. First, in these patients, the benefit of reducing the rate of restenosis is relatively small, as compared with the benefit seen in patients with small-vessel stents. 3,4 Second, acute thrombotic occlusion of large coronary arteries usually leads to sudden death or myocardial infarction, whereas small-vessel occlusions may remain clinically undetected. Unfortunately, there have been only a few retrospective subgroup analyses or registries of largevessel stenting, 5-7 even though the majority of patients in need of stent placement have lesions in large coronary arteries. In contrast, in smallvessel stenting, a particular benefit of drug-eluting stents has been repeatedly shown In the 18-month analysis of the Basel Stent Kosten- Effektivitäts Trial (BASKET), 15 among patients with stenoses in large coronary arteries who received drug-eluting stents, as compared with patients receiving bare-metal stents, there was no significant reduction in the rate of target-vessel revascularization and there was an increased rate of death from cardiac causes or nonfatal myocardial infarction 6 months after the procedure. In contrast, a beneficial effect of drug-eluting stents, as compared with bare-metal stents, was seen in the small-vessel subgroup. This difference was still present 3 years after stenting. 16 Similar findings were reported in a study from Canada. 17 However, these observations were based on retrospective analyses. We evaluated the relative efficacy of first-generation drug-eluting stents and bare-metal stents placed in large coronary arteries, in a large, prospective, randomized, multicenter trial, called the BASKET Prospective Validation Examination (BASKET PROVE). The study was also designed to evaluate whether the risk benefit balance seen with first-generation drug-eluting stents would be similar with second-generation drug-eluting stents. Me thods Study Design The study design and methods have been described in detail previously. 18 The protocol was designed by the steering committee and approved by the ethics committee at each center. The study protocol and a complete list of study group members in the Supplementary Appendix are available with the full text of this article at NEJM.org. The study was sponsored by the Basel Cardiovascular Research Foundation and the Swiss National Foundation for Research. The sponsors had no role in the design or conduct of the study, the analysis or interpretation of the data, or the decision to submit the manuscript for publication. All stents were purchased by the participating hospitals, and reimbursement by health insurers was exactly the same as that for patients not enrolled in the trial. The authors vouch for the accuracy and completeness of the data and all analyses. Patients Patients who presented at participating centers in Switzerland, Denmark, Austria, and Italy between March 5, 2007, and May 15, 2008, were evaluated for enrollment in the trial. Eligible patients were those who presented with chronic or acute coronary disease, who underwent angioplasty with stenting, and who required only stents that were 3.0 mm or more in diameter. No restrictions were placed on the number of treated lesions or vessels, the length of treated lesions, or the number of stents placed. Patients were excluded if they had cardiogenic shock; in-stent restenosis or thrombosis of stents placed before the study; unprotected left main coronary artery (i.e., with no functioning bypass graft) or substantial stenosis in a bypass graft; plans for any surgery within 12 months; a need for oral anticoagulation, an increased risk of bleeding, or known intolerance to or suspected noncompliance with long-term antiplatelet therapy; or circumstances that would have made follow-up impossible. In addition, patients requiring stents larger than 4.0 mm in diameter were excluded because no sirolimuseluting stents of this size were available. All patients provided written informed consent. Study Procedures Patients were randomly assigned in a 1:1:1 ratio to receive a first-generation sirolimus-eluting stent (Cypher Select, Cordis), a bare-metal (cobalt chro- n engl j med 363;24 nejm.org december 9,

3 T h e n e w e ngl a nd j o u r na l o f m e dic i n e mium) stent (Vision, Abbott Vascular), or a secondgeneration everolimus-eluting stent (Xience V, Abbott Vascular). Randomization was carried out in permuted blocks of 12 for each center with the use of sealed envelopes. Angioplasty and stenting were performed according to standard techniques chosen at the discretion of each interventional cardiologist. The same type of stent was used in patients with multiple lesions and in those undergoing staged procedures. All patients were prescribed aspirin at a daily dose of 75 to 100 mg indefinitely and clopidogrel at a daily dose of 75 mg for 1 year, after a loading dose of 300 mg or 600 mg, regardless of stent type. Therapeutic agents for secondary prevention, such as statins, were prescribed according to current guidelines. Clinical follow-up by means of questionnaires developed for this study was performed at 12 months and 24 months. Follow-up angiography and revascularization were performed only if clinically indicated. Data on all patients were collected by local investigators, recorded on electronic casereport forms, and transmitted over the Internet to a central database at the University of Basel. A central monitoring team verified all submitted information. End Points and Definitions The primary end point was death from cardiac causes or nonfatal myocardial infarction at 2 years. The main secondary end points included the efficacy end point of target-vessel revascularization and the safety end point of late death from cardiac causes or nonfatal myocardial infarction. Late events were defined as those occurring 7 to 24 months after the intervention, on the basis of analyses previously performed in the BASKET trial 19 and on angiographic studies of the timing of restenosis. 20 Other secondary analyses included rates of death from cardiac causes or any cause, nonfatal myocardial infarction, stent thrombosis, and the composite of death from any cause, nonfatal myocardial infarction, or target-vessel revascularization. Death from cardiac causes was defined as any death without a clear extracardiac cause. Myocardial infarction was defined as a clinical event with typical electrocardiographic or enzymatic changes. 21 Repeat revascularizations were subdivided into those related to myocardial infarction and those not related to myocardial infarction in order to avoid double-counting of infarct-related events. Stent thrombosis was defined according to the criteria of the Academic Research Consortium. 22 Bleeding events were defined according to Thrombolysis in Myocardial Infarction (TIMI) criteria. 23 Off-label use of stenting was defined as stenting in patients who presented with an acute coronary syndrome, stenting of more than one lesion per vessel, stenting of two or more vessels, stenting of lesions that were 27 mm or longer, stenting of bifurcation lesions, or stenting of chronic total occlusions. An independent critical events committee adjudicated all clinical end points. This assessment was conducted in a blinded fashion for the initial two thirds of events. However, as a result of illness of one of the study monitors, there was a delay in adjudicating the final one third of events. Because of time constraints for the completion of the analysis, these files were adjudicated without blinding. Statistical Analysis The sample size for the trial was estimated on the basis of findings from the 18-month results of BASKET. 15 Among the 558 patients who received stents that were 3 mm or more in diameter, the composite end point of death from cardiac causes or nonfatal myocardial infarction occurred in 6.0% of recipients of bare-metal stents and in 11.2% of recipients of drug-eluting stents during months 7 to 18 (P<0.05). On the basis of a slightly more conservative assumption (6.0% vs. 10.5% after 24 months), it was determined that 753 patients per group would provide a power of 80% to show the expected difference between either drugeluting stent and bare-metal stents, given an alpha level of in a two-sided test. Baseline characteristics are reported as counts and percentages or means ±SD. Baseline comparability of the study groups was examined with the use of an F-test for quantitative variables and a chi-square test for qualitative variables. All outcome analyses were performed according to the intention-to-treat principle. Data for patients who withdrew or were lost to follow-up were censored at the time of the last study contact. Prespecified comparisons were carried out for patients who were assigned to receive sirolimus-eluting stents versus those assigned to receive bare-metal stents, and for patients assigned to receive everolimus-eluting stents versus those assigned to receive bare-metal stents. In addition, post hoc comparisons were made between patients 2312 n engl j med 363;24 nejm.org december 9, 2010

4 Drug-Eluting vs. Bare-Metal Stents in Large Arteries receiving sirolimus-eluting stents and those receiving everolimus-eluting stents. Time-dependent occurrence of events was investigated with Kaplan Meier curves and Cox proportional-hazard models, stratified according to center. The overall treatment effect was tested with the use of a likelihood-ratio test. Since the assumption of constant hazard may not be consistent with the data, a time-independent analysis was also performed. Since large-sample approximations do not hold when dealing with rare events, odds ratios with 95% confidence intervals were estimated and Fisher s exact test was applied. 24 Landmark analyses were performed for the major trial end points by dividing the entire follow-up period into the initial 6 months and the subsequent 18 months. P values were adjusted for multiple comparisons with the use of the step-up procedure, according to the method of Benjamini and Hochberg, 25 in order to maintain the false discovery rate at the 0.05 level within each table. All calculations were performed with the statistical software system R, version R esult s Patients A total of 2323 patients were enrolled in the trial, but 9 withdrew their consent before randomization. The remaining 2314 patients were randomly assigned to receive sirolimus-eluting stents (775 patients), everolimus-eluting stents (774 patients), or bare-metal stents (765 patients) (Fig. 1 in the Supplementary Appendix). Eighteen patients who underwent randomization received the wrong stent or no stent or should have been excluded on the basis of enrollment criteria but were included in the intention-to-treat analysis. There were no significant between-group differences in baseline clinical characteristics (Table 1). Two thirds of patients presented with acute coronary syndromes, and half of these patients had myocardial infarction with ST-segment elevation. Five percent of patients received unplanned treatment with stents that were less than 3.0 mm in diameter for distal dissections or side-branch obstructions, in addition to the larger stent or stents that were implanted according to the protocol. A total of 76% of patients were treated for off-label indications. Follow-up and Clinical Outcomes The median follow-up for surviving patients was 745 days, without significant differences between groups. A total of 21 patients withdrew consent at some time after the initial intervention, and 38 were lost to follow-up before the 2-year follow-up visit. At the 1-year visit, 97% of patients were taking at least one antiplatelet drug, and at the 2-year visit, 95% of patients were taking at least one antiplatelet drug; the proportions of patients taking aspirin and clopidogrel at 1 and 2 years were 82% and 19%, respectively. At 2 years of follow-up, the event rate for the primary end point (death from cardiac causes or nonfatal myocardial infarction) was 2.6% in the group receiving sirolimus-eluting stents, 3.2% in the group receiving everolimus-eluting stents, and 4.8% in the group receiving bare-metal stents. There were no significant differences between either of the drug-eluting stent groups and the bare-metal stent group or between the two drugeluting stent groups, with respect to the primary end point. This finding was consistent with the use of both time-dependent proportional-hazards analysis (Table 2 and Fig. 1) and time-independent analysis (by means of Fisher s exact test) (Table 1 in the Supplementary Appendix). Similarly, no significant differences were seen for the rate of death from cardiac causes or the rate of nonfatal myocardial infarction when considered separately, either by time-dependent or time-independent analysis. There were also no significant betweengroup differences in rates of stent thrombosis (Table 2 and Fig. 2, and Table 1 in the Supplementary Appendix). The rate of target-vessel revascularization that was not related to myocardial infarction was 3.7% in the group receiving sirolimus-eluting stents, 3.1% in the group receiving everolimus-eluting stents, and 8.9% in the group receiving bare-metal stents. Thus, the event rate was significantly reduced, by more than 50%, in both groups receiving drug-eluting stents, as compared with the group receiving bare-metal stents, with no significant difference between the two groups receiving drug-eluting stents (Table 2 and Fig. 3, and Table 1 in the Supplementary Appendix). As a result, the composite end point of major adverse cardiac events (i.e., death from any cause, nonfatal myocardial infarction, or target-vessel revascularization not related to myocardial infarction) was n engl j med 363;24 nejm.org december 9,

5 T h e n e w e ngl a nd j o u r na l o f m e dic i n e Table 1. Baseline Characteristics of the Patients and Stenting Information.* Characteristic Sirolimus- Eluting Stent (N = 775) Everolimus- Eluting Stent (N = 774) Bare-Metal Stent (N = 765) P Value Age yr 66±11 66±11 67± Male sex no. (%) 576 (74) 587 (76) 586 (77) 0.80 Cardiac risk factors no. (%) Diabetes mellitus 136 (18) 119 (15) 108 (14) 0.76 Systemic arterial hypertension 483 (62) 469 (61) 485 (63) 0.80 Hyperlipidemia 475 (61) 498 (64) 495 (65) 0.76 Current smoker 235 (30) 267 (34) 261 (34) 0.76 Previous myocardial infarction 96 (12) 82 (11) 103 (13) 0.76 Previous coronary intervention 97 (13) 93 (12) 88 (12) 0.92 Previous coronary-artery bypass grafting 27 (3) 20 (3) 20 (3) 0.80 Clinical presentation no. (%) Stable angina 266 (34) 271 (35) 285 (37) 0.80 Unstable angina 244 (31) 264 (34) 246 (32) 0.80 Myocardial infarction with ST-segment elevation Treated coronary arteries no. (%) 265 (34) 239 (31) 234 (31) 0.76 Left main with bypass graft 5 (1) 7 (1) 9 (1) 0.80 Left anterior descending 404 (52) 412 (53) 400 (52) 0.92 Left circumflex 201 (26) 202 (26) 203 (27) 0.96 Right coronary 321 (41) 310 (40) 325 (42) 0.80 Complexity of coronary artery disease no. (%) Multivessel disease 343 (44) 319 (41) 327 (43) 0.80 Bifurcation lesion 63 (8) 58 (7) 68 (9) 0.80 Chronic total occlusion 35 (5) 34 (4) 39 (5) 0.92 Additional placement of stents <3.0 mm in diameter 27 (3) 44 (6) 45 (6) 0.36 Use of glycoprotein IIb/IIIa inhibitor 181 (23) 168 (22) 176 (23) 0.92 Procedural characteristics No. of treated segments per patient 1.4± ± ± No. of stents per patient 1.6± ± ± No. of stents Total stent length per patient mm 30.0± ± ± Stent length per lesion mm 22±9 22±11 21± Staged procedures no. (%) 34 (4) 41 (5) 33 (4) 0.80 Segments with angiographic success no./total no. (%) 1045/1066 (98) 1070/1092 (98) 1080/1116 (97) 0.92 Off-label stent use no. (%) 608 (78) 582 (75) 575 (75) 0.76 * Plus minus values are means ±SD. P values were calculated with the use of an F-test for quantitative variables and a chi-square test for qualitative variables and were adjusted for multiple comparisons. This category includes myocardial infarction without ST-segment elevation n engl j med 363;24 nejm.org december 9, 2010

6 Drug-Eluting vs. Bare-Metal Stents in Large Arteries Table 2. Clinical Outcomes at 24 Months in the Intention-to-Treat Population.* Outcome Sirolimus- Eluting Stent (N = 775) Everolimus- Eluting Stent (N = 774) Bare-Metal Stent (N = 765) Sirolimus-Eluting Stent vs. Bare-Metal Stent Hazard Ratio (95% CI) P Value Everolimus-Eluting Stent vs. Bare-Metal Stent Hazard Ratio (95% CI) P Value Sirolimus-Eluting Stent vs. Everolimus-Eluting Stent Hazard Ratio (95% CI) P Value Death no. of patients (%) From any cause 28 (3.6) 25 (3.2) 34 (4.4) 0.82 ( ) ( ) ( ) 0.85 From cardiac causes 13 (1.7) 13 (1.7) 22 (2.9) 0.59 ( ) ( ) ( ) 0.93 Nonfatal myocardial infarction 7 (0.9) 13 (1.7) 20 (2.6) 0.37 ( ) ( ) ( ) 0.43 Death from cardiac causes or nonfatal myocardial infarction Total 20 (2.6) 25 (3.2) 37 (4.8) 0.54 ( ) ( ) ( ) mo 11 (1.4) 10 (1.3) 21 (2.7) 0.52 ( ) ( ) ( ) mo 9 (1.2) 15 (1.9) 16 (2.1) 0.56 ( ) ( ) ( ) 0.51 Target-vessel revascularization Any 33 (4.3) 29 (3.7) 79 (10.3) 0.47 ( ) ( ) ( ) 0.85 Not related to myocardial infarction 29 (3.7) 24 (3.1) 68 (8.9) 0.46 ( ) ( ) ( ) 0.82 Related to myocardial infarction 4 (0.5) 5 (0.6) 11 (1.4) 0.40 ( ) ( ) ( ) 0.90 Death, myocardial infarction, or targetvessel revascularization Stent thrombosis 61 (7.9) 59 (7.6) 99 (12.9) 0.59 ( ) ( ) ( ) 0.90 Definite 3 (0.4) 2 (0.3) 6 (0.8) 0.50 ( ) ( ) ( ) 0.85 Definite or probable 6 (0.8) 5 (0.6) 9 (1.2) 0.75 ( ) ( ) ( ) 0.90 Definite, probable, or possible 11 (1.4) 12 (1.6) 13 (1.7) 0.92 ( ) ( ) ( ) 0.93 * Patients may have had more than one event. All P values have been adjusted for multiple comparisons. The difference between groups is significant after adjustment for multiple comparisons by means of the step-up procedure. n engl j med 363;24 nejm.org december 9,

7 T h e n e w e ngl a nd j o u r na l o f m e dic i n e 1.0 Probability of Death or Myocardial Infarction No. at Risk Months P= Figure 1. Kaplan Meier Estimates of Death from Cardiac Causes or Nonfatal Myocardial Infarction (Composite Primary End Point) at 24 Months. There were no significant differences between bare-metal stents and either everolimus-eluting stents or sirolimus-eluting stents for the composite primary end point of death from cardiac causes or nonfatal myocardial infarction. A likelihood test that was performed to assess the overall treatment effect in a time-dependent analysis (with the use of a stratified Cox model) showed a trend toward significance (P = 0.06 without adjustment for multiple comparisons). also significantly reduced in both groups receiving drug-eluting stents, as compared with the group receiving bare-metal stents, with no significant difference seen between the two groups receiving drug-eluting stents (Table 2, and Table 1 in the Supplementary Appendix). Major bleeding events were reported in 2.4% of patients, of whom 76% had the events within the first year, when all patients were advised to take aspirin and clopidogrel (Table 2 in the Supplementary Appendix). To assess the differences in outcome during the first 6 months, as compared with the subsequent 18 months, landmark analyses were performed. During the initial 6 months after stenting, there was a trend toward a lower rate of the primary end point with both drug-eluting stents, as compared with bare-metal stents, but this trend was not significant after adjustment for multiple testing (Fig. 2 in the Supplementary Appendix). A similar nonsignificant trend was observed when death from cardiac causes and nonfatal myocardial infarction were considered separately (Fig. 3 in the Supplementary Appendix). There were no significant between-group differences in either early or late rates of stent thrombosis. The reduction in the rate of target-vessel revascularization that was not related to myocardial infarction occurred during the first 6 months after the intervention, without a significant divergence of the corresponding event curves thereafter (Fig. 3 in the Supplementary Appendix). Discussion In this trial, we compared two types of drugeluting stents with bare-metal stents in patients requiring large coronary stents (3.0 mm or more in diameter). The likelihood of a benefit of drugeluting stents is relatively small in this group of patients, since the rate of restenosis is low and the risk of adverse cardiac events due to late stent thrombosis may be greater than the risk among patients with small-vessel stents. 15,19 We also compared the first-generation sirolimus-eluting stent with the second-generation everolimus-eluting stent. Both types of drug-elut n engl j med 363;24 nejm.org december 9, 2010

8 Drug-Eluting vs. Bare-Metal Stents in Large Arteries Probability of Stent Thrombosis No. at Risk P= Months Figure 2. Kaplan Meier Estimates of Definite or Probable Stent Thrombosis at 24 Months. There were no significant differences between bare-metal stents and either everolimus-eluting stents or sirolimuseluting stents for the secondary end point of definite or probable stent thrombosis. A likelihood test that was performed to assess the overall treatment effect in a time-dependent analysis (with the use of a stratified Cox model) also showed no significant difference (P = 0.68 without adjustment for multiple comparisons). ing stents were superior to bare-metal stents in reducing the clinical need for target-vessel revascularization after large-vessel stenting. In addition, at 2 years of follow-up, the incidence of late death from cardiac causes or nonfatal myocardial infarction was similar for drug-eluting stents and bare-metal stents. Among patients receiving drugeluting stents, there was a nonsignificant reduction in such events, suggesting that if the sample had been larger, even a reduction in the rate of death from cardiac causes or nonfatal myocardial infarction might have been observed. Finally, in the analyses of drug-eluting stents, we could not detect significant differences in event rates between the first-generation stents and the second-generation stents when they were used in large coronary arteries. In our study, rates of cardiac events were substantially lower than anticipated. We predicted a 24-month rate of 6.0% for the primary end point in the group receiving bare-metal stents, but the observed rate was only 4.8%. The predicted event rate among patients receiving drug-eluting stents was 10.5%, but the observed rates were 2.6% among patients receiving sirolimus-eluting stents and 3.2% among those receiving everolimus-eluting stents. A similar decrease in the rate of cardiac events over time has been observed in Sweden. 26 Factors contributing to the reduced event rates may have included longer experience in the use of drug-eluting stents, differences in patient selection, the fact that no paclitaxel-eluting stents were used, technical factors during the stenting procedure, improvements in perioperative management, and the recommendation that dual-antiplatelet therapy be used for 12 months rather than 6 months. Second-generation drug-eluting stents were developed in an effort to address concerns about the late safety of first-generation drug-eluting stents. Modifications in the second-generation stents include different drugs, lower drug doses, and newer stent designs particularly, thinner struts and newer, more biocompatible or even bioabsorbable polymers. 27 To test their efficacy and safety, these stents have most often been compared with the n engl j med 363;24 nejm.org december 9,

9 T h e n e w e ngl a nd j o u r na l o f m e dic i n e Probability of Revascularization No. at Risk P< Months Figure 3. Kaplan Meier Estimates of Target-Vessel Revascularization Not Related to Myocardial Infarction at 24 Months. The rate of the secondary end point of target-vessel revascularization not related to myocardial infarction was significantly higher among patients receiving bare-metal stents than among those receiving either drug-eluting stent. A likelihood test that was performed to assess the overall treatment effect in a time-dependent analysis (with the use of a stratified Cox model) showed a significant difference (P<0.001 with adjustment for multiple comparisons). first-generation paclitaxel-eluting stent However, one head-to-head comparison showed the superiority of the sirolimus-eluting stent over the paclitaxel-eluting stent, 32 suggesting that the sirolimus-eluting stent should be the standard for comparison with newer stents. Our comparison of the everolimus-eluting stent with the sirolimuseluting stent was a post hoc analysis, but it did suggest that the two stents are similar with respect to rates of major cardiovascular events. In view of the unexpectedly low event rates, our trial was underpowered to detect small differences in events. However, given the 95% confidence intervals for the hazard ratios, our results rule out a substantial increase in the risk of death from cardiac causes or nonfatal myocardial infarction with drug-eluting stents versus bare-metal stents. In our trial, the patients, but not the interventional cardiologists, were unaware of which stents they were receiving. As noted, the criticalevents committee adjudicated the final one third of events without blinding, which could have influenced decisions about those events. Finally, it should be noted that the findings of this trial are not applicable to patients receiving smaller-diameter stents. In summary, our comparison of sirolimuseluting stents and everolimus-eluting stents with bare-metal stents for percutaneous intervention in large coronary arteries showed no evidence of an increased risk associated with drug-eluting stents. A clinically relevant reduction in the rate of target-vessel revascularization was confirmed, even though patients with stents in large vessels are at reduced risk for clinically relevant restenosis. In this patient population, we found no significant differences in rates of cardiovascular events between the first-generation sirolimus-eluting stent and the second-generation everolimuseluting stent at 2 years of follow-up. Supported by the Basel Cardiovascular Research Foundation and a grant from the Swiss National Foundation for Research (32003B ). Disclosure forms provided by the authors are available with the full text of this article at NEJM.org n engl j med 363;24 nejm.org december 9, 2010

10 Drug-Eluting vs. Bare-Metal Stents in Large Arteries References 1. Farb A, Boom AB. Stent thrombosis redux the FDA perspective. N Engl J Med 2007;356: Serruys PW. FDA panel, 7 and 8 December 2006 the impact on our practice and research. EuroIntervention 2007;2: Elezi S, Kastrati A, Neumann FJ, Hadamitzky M, Dirschinger J, Schömig A. Vessel size and long-term outcome after coronary stent placement. Circulation 1998; 98: Akiyama T, Moussa I, Reimers B, et al. Angiographic and clinical outcome following coronary stenting of small vessels: a comparison with stenting of large vessels. J Am Coll Cardiol 1998;32: Steinberg DH, Mishra S, Javaid A, et al. Comparison of effectiveness of bare metal stents versus drug-eluting stents in large ( 3.5mm) coronary arteries. Am J Cardiol 2007;99: Cox D, Hermiller JB, Cannon L, et al. Are DES beneficial in large vessels compared to BMS? Data from the TAXUS IV and V randomized clinical trials. J Am Coll Cardiol 2008;51:B109. abstract. 7. Gordon PC, Applegate J, Hermiller JB, et al. Clinical and angiographic outcomes with an everolimus-eluting stent in large coronary arteries: the SPIRIT III 4.0 mm registry. Catheter Cardiovasc Interv 2010; 75: Moreno R, Fernández C, Alfonso F, et al. Coronary stenting versus balloon angioplasty in small vessels: a meta-analysis from 11 randomized studies. J Am Coll Cardiol 2004;43: Ardissino D, Cavallini C, Bramucci E, et al. Sirolimus-eluting vs. uncoated stents for prevention of restenosis in small coronary arteries: a randomized trial. JAMA 2004;292: Kastrati A, Schömig A, Dirschinger J, et al. Randomized trial comparing stenting with balloon angioplasty in small vessels in patients with symptomatic coronary artery disease. Circulation 2000;102: Park SW, Lee CW, Hong MK, et al. Randomized comparison of coronary stenting with optimal balloon angioplasty for the treatment of lesions in small coronary arteries. Eur Heart J 2000;21: Hermiller JB, Fergues T, Pierson W, et al. Clinical and angiographic comparison of everolimus-eluting and paclitaxel-eluting stents in small coronary arteries: a post hoc analysis of the SPIRIT III randomized trial. Am Heart J 2009;158: Bartorelli AL, Serruys PW, Miquel- Hébert K, et al. An everolimus-eluting stent versus a paclitaxel-eluting stent in small vessel coronary artery disease: a pooled analysis from the SPIRIT II and SPIRIT III trials. Catheter Cardiovasc Interv 2010;76: Wykrzykowska JJ, Serruys PW, Onuma Y, et al. Impact of vessel size on angiographic and clinical outcomes of revascularization with biolimus-eluting stent with biodegradable polymer and sirolimuseluting stent with durable polymer. JACC Cardiovasc Interv 2009;2: Brunner-La Rocca HP, Kaiser C, Pfisterer M. Targeted stent use in clinical practice based on evidence from the Basel Stent Cost Effectiveness Trial (BASKET). Eur Heart J 2007;28: Pfisterer M, Brunner-La Rocca H-P, Rickenbacher P, et al. Long-term benefitrisk balance of drug-eluting stents in daily practice: does stent diameter matter? Threeyear follow-up of BASKET. Eur Heart J 2009;30: Tu JV, Bowen J, Chiu M, et al. Effectiveness and safety of drug-eluting stents in Ontario. N Engl J Med 2007;357: Pfisterer M, Bertel O, Bonetti PO, et al. Drug-eluting or bare-metal stents for large coronary vessel stenting? The BASKET- PROVE (PROspective Validation Examination) trial: study protocol and design. Am Heart J 2008;155: Pfisterer M, Brunner-La Rocca H-P, Buser PT, et al. Late clinical events after clopidogrel discontinuation may limit the benefit of drug-eluting stents: an observational study of drug-eluting versus baremetal stents. J Am Coll Cardiol 2006;48: Kimura T, Abe K, Shizuta S, et al. Long-term clinical and angiographic follow-up after coronary stent placement in native coronary arteries. Circulation 2002; 105: Myocardial infarction redefined a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. Eur Heart J 2000;21: Cutlip DE, Windecker S, Mehran R, et al. Clinical end points in coronary stent trials: a case for standard definitions. Circulation 2007;115: 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: Agresti A. Categorical analysis. 2nd ed. New York: Wiley, Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 1995;57: James SK, Stenestrand U, Lindbäck J, et al. Long-term safety and efficacy of drug-eluting versus bare-metal stents in Sweden. N Engl J Med 2009;360: Garg S, Serruys PW. Coronary stents: current status. J Am Coll Cardiol 2010;56: Suppl:S1-S Stone GW, Midei M, Newman W, et al. Comparison of an everolimus-eluting stent and a paclitaxel-eluting stent in patients with coronary artery disease: a randomized trial. JAMA 2008;299: Leon MB, Mauri L, Popma JJ, et al. A randomized comparison of the ENDEAVOR zotarolimus-eluting stent versus the TAXUS paclitaxel-eluting stent in de novo native coronary lesions: 12 month outcomes from the ENDEAVOR IV trial. J Am Coll Cardiol 2010;55: Stone GW, Rizvi A, Newman W, et al. Everolimus-eluting versus paclitaxel-eluting stents in coronary artery disease. N Engl J Med 2010;362: Kedhi E, Joesoef KS, McFadden E, et al. Second-generation everolimus-eluting and paclitaxel-eluting stents in real-life practice (COMPARE): a randomised trial. Lancet 2010;375: Windecker S, Remondino A, Eberli FR, et al. Sirolimus-eluting and paclitaxeleluting stents for coronary revascularization. N Engl J Med 2005;353: Copyright 2010 Massachusetts Medical Society. n engl j med 363;24 nejm.org december 9,