Author + information
- Received October 10, 2010
- Revision received December 6, 2010
- Accepted January 2, 2011
- Published online May 24, 2011.
- Lorenz Räber, MD⁎,
- Peter Jüni, MD†,‡,
- Eveline Nüesch, PhD‡,
- Bindu Kalesan, MSc‡,
- Peter Wenaweser, MD⁎,
- Aris Moschovitis, MD⁎,
- Ahmed A. Khattab, MD⁎,
- Maryam Bahlo, MD⁎,
- Mario Togni, MD⁎,
- Stéphane Cook, MD⁎,
- Rolf Vogel, MD, PhD⁎,
- Christian Seiler, MD⁎,
- Bernhard Meier, MD⁎ and
- Stephan Windecker, MD⁎,†,⁎ ()
- ↵⁎Reprint requests and correspondence:
Prof. Stephan Windecker, Department of Cardiology, Bern University Hospital, 3010 Bern, Switzerland
Objectives This study sought to compare the unrestricted use of everolimus-eluting stents (EES) with sirolimus-eluting stents (SES) in patients undergoing percutaneous coronary intervention.
Background It is unclear whether there are differences in safety and efficacy between EES and SES during long-term follow-up.
Methods Using propensity score matching, clinical outcome was compared among 1,342 propensity score–matched pairs of patients treated with EES and SES. The primary outcome was a composite of death, MI, and target vessel revascularization.
Results The median follow-up was 1.5 years with a maximum of 3 years. The primary outcome occurred in 14.9% of EES- and 18.0% of SES-treated patients up to 3 years (hazard ratio [HR]: 0.83, 95% confidence interval [CI]: 0.68 to 1.00, p = 0.056). All-cause mortality (6.0% vs. 6.5%, HR: 0.92, 95% CI: 0.68 to 1.25, p = 0.59) was similar, risks of myocardial infarction (MI) (3.3% vs. 5.0%, HR: 0.62, 95% CI: 0.42 to 0.92, p = 0.017), and target vessel revascularization (7.0% vs. 9.6%, HR: 0.75, 95% CI: 0.57 to 0.99, p = 0.039) were lower with EES than SES. Definite stent thrombosis (ST) (HR: 0.30, 95% CI: 0.12 to 0.75, p = 0.01) was less frequent among patients treated with EES. The reduced rate of MI with EES was explained in part by the lower risk of definite ST and the corresponding decrease in events associated with ST (HR: 0.25, 95% CI: 0.08 to 0.75, p = 0.013).
Conclusions The unrestricted use of EES appears to be associated with improved clinical long-term outcome compared with SES. Differences in favor of EES are driven in part by a lower risk of MI associated with ST.
Early generation drug-eluting stents (DES) releasing sirolimus (sirolimus-eluting stents [SES]) or paclitaxel (paclitaxel-eluting stents [PES]) have reduced the need of repeat revascularization compared with bare-metal stents (1,2). Although the rate of mortality and myocardial infarction (MI) was similar for DES and bare-metal stents (3), very late stent thrombosis (ST) emerged as a distinct entity complicating the use of early generation DES (4). Moreover, restenosis still occurs after DES implantation with evidence of an erosion of antirestenotic efficacy over time (5). Newer generation DES have been developed with the aim to improve the safety and efficacy of early generation devices (6). The newer generation everolimus-eluting stent (EES) has been shown to improve outcome compared with PES (7–10). However, data comparing EES with SES are limited. Since SES have been shown to be superior compared with PES (3,11) as well as with a new-generation stent eluting zotarolimus from a phosphorylcholine polymer (12,13), it is relevant to determine whether EES provide therapeutic benefit over SES. We therefore compared the outcomes of the unrestricted use of EES and SES in a large, consecutively enrolled patient population followed for up to 3 years in a propensity-matched analysis.
Study population and data collection
A total of 1,532 consecutive patients were treated with SES (Cypher, Cordis, Miami Lakes, Florida) between May 2004 and January 2006, whereas 1,601 consecutive patients underwent treatment with EES (XIENCE V, Abbott Vascular, Santa Clara, California; or PROMUS, Boston Scientific, Natick, Massachusetts) between November 2006 and March 2009. Patients included in the SIRTAX (Sirolimus-Eluting Versus Paclitaxel-Eluting Stents for Coronary Revascularization) trial were not eligible in view of mandated angiographic follow up (14). The study complied with the Declaration of Helsinki and was approved by the institutional ethics committee at Bern University Hospital, Switzerland. Patients gave written informed consent to be prospectively followed.
All patients were followed up for major adverse cardiac events using patient-administered postal questionnaires. Vital status was ascertained from hospital records and municipal civil registries. All suspected events were independently adjudicated by a clinical event committee whose members were unaware of the type of stent implanted. Baseline clinical and procedural characteristics and all follow-up data were entered into a dedicated database, held at an academic clinical trials unit (CTU Bern, Bern University Hospital, Switzerland) responsible for central data audits and maintenance of the database.
The treatment guidelines, including periprocedural and post-procedural medication regimen, were performed according to current practice guidelines and did not change between the inclusion of the first patient into the SES and inclusion of the last patient into the EES cohort. All patients received a loading dose of clopidogrel 300 to 600 mg during the procedure and were prescribed aspirin once daily lifelong and clopidogrel for 12 months. The use of glycoprotein IIb/IIIa antagonists was left to the discretion of the operator. Creatinine kinase (CK), CK-MB, and troponin T were routinely assessed at baseline and 12 to 24 h after percutaneous coronary intervention as was a 12-lead electrocardiogram. Biomarkers were sampled every 6 to 8 h in patients with signs of ischemia until identification of peak levels.
The primary endpoint was the composite of death, MI, and target vessel revascularization (TVR) up to a maximum follow-up of 3 years. The definition of cardiac death included any death due to immediate cardiac cause, procedure-related deaths, and death of unknown cause. The diagnosis of Q-wave MI required ischemic signs or symptoms and new pathological Q waves in ≥2 contiguous electrocardiogram leads. In the absence of Q waves, the diagnosis of MI was based on an elevation in CK to ≥2× upper limit of normal and an elevation of CK-MB or troponin to ≥3× upper limit of normal. TVR was defined as repeat revascularization of any segment within the entire major coronary vessel proximal and distal to a target lesion. Target lesion revascularization (TLR) was defined as revascularization for a stenosis within the stent or the 5-mm borders adjacent to the stent. ST was defined according to Academic Research Consortium definitions (15).
This was a propensity score (PS)-matched superiority analysis. Sample size considerations were based on an updated pooled analysis of trials comparing EES with PES (16), suggesting a relative risk (RR) of 0.60 for the composite of death, MI, or TVR, and a network analysis comparing SES with PES (3), which suggested a RR of 0.80 in favor of SES. Taken together, these data suggested a RR of 0.60/0.80 = 0.75 in favor of EES. With an expected crude event rate of 18% at a median follow-up of 1.5 years with SES, a sample size of 1,400 matched pairs would provide 90% power to detect a RR of 0.75 in favor of EES. Assuming that 90% of patients treated with EES could be matched to patients treated with SES, 1,560 patients treated with EES were necessary for this study.
We compared baseline characteristics between patients treated with EES and SES using a chi-square test for categorical variables and an unpaired t test for continuous variables. Then, we used PS matching to account for differences in baseline characteristics. PS for receiving EES were estimated using a probit model including age, gender, and pre-treatment variables associated with stent selection in the multivariable model at p < 0.10 as independent variables (arterial hypertension, hypercholesterolemia, clinical manifestation of coronary artery disease at baseline, and ejection fraction below 50%). An automated matching procedure randomly selected a patient treated with EES and a randomly selected patient treated with SES from the pool of patients with PS within a caliper of ±0.05 on the propensity score. For each pair, we ensured equal follow-up times. We used Cox proportional hazards models that accounted for the 1:1 matching to calculate hazard ratios (HR) comparing the 2 stent types. In a sensitivity analysis, we adjusted procedural characteristics that differed between stent types in the PS-matched sample at p < 0.10. For ST, we performed landmark analyses according to time points specified in Academic Research Consortium definitions (15). Then, we compared the 2 stent types separately on clinical outcomes associated with ST (defined as events occurring within a 1-day time window of ST) and not associated with ST. Finally, we used univariable Cox models to determine whether procedural characteristics were associated with the primary composite endpoint, because procedural characteristics were different between stent types. All p values and 95% confidence intervals (CIs) are 2-sided.
A total of 3,133 patients (98.7%) completed the last follow-up (EES = 98.7%, SES = 98.7%). A comparison of patients before PS matching is provided in Table 1. One thousand three hundred forty-two patients treated with EES could be matched to 1,342 patients treated with SES. The median follow-up duration was 1.3 years in both groups (range 1.0 year to 2.2 years), with an accumulated 2,221 and 2,238 patient-years, respectively. Table 1 shows pre-treatment characteristics at baseline after matching, which were comparable between groups. Table 2 presents procedural characteristics after matching. Implantation of EES appeared more complex, with a higher proportion of patients with multivessel disease and higher number of lesions and vessels treated per patient. Discharge medications were comparable for both groups, and the median length of clopidogrel prescription duration was 12 months (Table 2).
Table 3 presents clinical outcomes up to 3 years. The primary outcome occurred in 14.9% of EES- and 18.0% of SES-treated patients up to 3 years (p = 0.056) (Fig. 1). The trend in favor of EES was driven by a significantly lower rate of MI (3.3% vs. 5.0%, p = 0.017) and TVR (7.0% vs. 9.6%, p = 0.039). Rates of all-cause and cardiac mortality were similar, whereas Q-wave MI (0.5% vs. 1.6%, p = 0.010) and the composite of cardiac death or MI (6.8% vs. 8.9%, p = 0.030) were less frequent with EES. Table 4 shows associations of procedural characteristics with the primary outcome stratified by stent type and overall. The presence of more complex procedural characteristics was generally associated with worse outcome for both stent types and overall. A sensitivity analysis of the primary outcome adjusted for procedural characteristics yielded similar results: HR: 0.78, 95% CI: 0.63 to 0.97, p = 0.029.
Results on Academic Research Consortium–defined ST are summarized in Table 5. Definite ST was less frequent with EES than SES (0.5% vs. 1.6%, HR: 0.30, 95% CI: 0.12 to 0.75, p = 0.010) as was definite or probable ST (Fig. 2). Clinical outcomes associated with definite ST (left) and outcomes occurring in the absence of ST (right) are shown in Figure 3. ST-associated MI (HR: 0.25, 95% CI: 0.08 to 0.75, p = 0.013) and TVR (HR: 0.33, 95% CI: 0.12 to 0.92, p = 0.033) were less frequent with EES. These differences were less pronounced for MI (HR: 0.79, 95% CI: 0.51 to 1.21, p = 0.28) and TVR (HR: 0.84, 95% CI: 0.63 to 1.12, p = 0.24) occurring in the absence of ST (Fig. 3).
In this observational, PS-matched study, nested in a prospective registry, the use of EES was associated with a trend toward a lower risk of the patient-oriented safety and efficacy endpoint of death, MI, and TVR as compared with SES during follow-up to 3 years. The risk of MI was reduced by 38%, and differences in rates of MI were driven by a 70% reduction in the risk of Q-wave MI.
The results of the present study contribute to a mechanistic explanation of differences in clinical outcome between EES and SES. The lower risk of MI with EES was explained in part by the lower rate of ST (Fig. 2), whereas differences in the risk of MI occurring in the absence of ST were less pronounced (Fig. 3). This observation is important because the unrestricted use of early generation DES was associated with an ongoing risk of ST during long-term follow-up and stirred a debate regarding the need of prolonged dual antiplatelet therapy (17–20). Our long-term data provide novel evidence that ST beyond 1 year is less frequent with EES compared with SES (p = 0.007), circumventing an important shortcoming of early generation DES. The mechanisms underlying the lower risk of ST with EES remain speculative but may be related to the lower strut thickness with less arterial injury and more rapid and complete endothelialization, a biocompatible polymer less prone to hypersensitivity reactions, and a lower dose of the antiproliferative drug.
The risk of TVR was 25% lower with EES than SES, and the majority of revascularization procedures were related to the target lesion. Of note, the risk of TVR associated with ST was lower with EES, whereas differences between stent types were less pronounced for revascularization procedures performed in the absence of ST. This suggests that differences in revascularization in favor of EES were related in part to a lower predisposition for ST rather than restenosis. One clinical registry and 2 randomized clinical trials have compared EES with SES. The X-SEARCH registry (21) showed similar safety and efficacy outcomes in both EES and SES at 6 months of follow-up after multivariate adjustment of the 2 sequential cohorts. The ISAR-TEST 4 (Intracoronary Stenting and Angiographic Results: Test Efficacy of 3 Limus-Eluting STents 4) trial (22) observed a trend toward lower TLR (9.9% vs. 13.5%, p = 0.06) and a significant reduction of binary restenosis at 2 years (12.7% vs. 16.9%, p = 0.03) in favor of EES in the absence of differences for safety endpoints among 1,304 patients randomly assigned treatment with EES or SES. The SORT OUT IV (Randomized Clinical Comparison of the Xience V and the Cypher Coronary Stents in Non-selected Patients With Coronary Heart Disease) trial (23) reported noninferior outcomes of EES compared with SES in terms of major adverse cardiac events and TLR at 9 months among 2,774 patients randomly assigned treatment with EES or SES. The investigators noted a trend toward a lower rate of definite ST with EES (0.1% vs. 0.7%, HR = 0.22, 95% CI: 0.05 to 1.02, p = 0.05).
This was not a randomized trial, and results may be biased. However, we used appropriate PS matching to ensure comparability of groups. Propensity scores were defined as the probability to receive EES conditional on pre-treatment covariates. These covariates summarize what is known about that patient prior to treatment. By definition, it is not possible to include procedural characteristics of the compared interventions in the PS. Procedural characteristics after PS matching were different between groups, however. To determine whether this could explain some of the observed differences between stent types, we examined the association between markers of increased procedural complexity and clinical outcome, and performed a sensitivity analysis adjusted for procedural characteristics. Our results indicate that, if anything, EES was put at a disadvantage by the observed higher procedural complexity (Table 4) and that results remained robustly in favor of EES after adjusting for procedural characteristics (p = 0.029). Another limitation is the sequential enrollment period. It cannot be excluded that changes in treatment may have had a favorable impact on clinical outcome. However, results were obtained at a single institution with similar patient profiles during sequential enrollment periods, thus minimizing the risk of institutional heterogeneity. Treatment protocols did not change during enrollment, and we observed no differences with respect to discharge medications. The sequential enrollment of SES and EES minimizes the potential of confounding by indication because there was no competition between stent types. Finally, the number of pairs successfully matched was lower than assumed in the sample size considerations. This resulted in somewhat lower power and may explain that we formally missed the pre-specified alpha level for the primary endpoint. However, the consistent findings in clinical outcomes during long-term follow-up, with robust reductions in MI and ST, make it unlikely that estimates of safety and efficacy would differ when studied in a larger population.
The clinical implications of our study are 3-fold: First, DES efficacy can be further advanced beyond the level of the previous gold standard of SES without compromising, but even improving their safety profile. Second, the phenomenon of very late ST may be less frequent with EES. Third, our results suggest that the lower rate of MI was driven at least in part by a lower risk of ST. This has important implications for the duration of dual antiplatelet therapy.
The analysis was funded by intramural grants provided by CTU Bern, Bern University Hospital, the Institute of Social and Preventive Medicine, University of Bern, and the Swiss National Science Foundation (Grant 33CM30-124112). Dr. Räber is the recipient of a research fellowship (SPUM) funded by the Swiss National Science Foundation. Dr. Jüni is an unpaid steering committee member of Medtronic, Abbott Vascular, and Johnson & Johnson (clinical trials). Dr. Meier has received research grants from and served on the Speakers' Bureaus of Abbott and Cordis. Prof. Windecker has received consulting and lecture fees from Abbott, Boston Scientific, Biosensors, Cordis, and Medtronic. All other authors have reported that they have no relationships to disclose. Drs. Räber and Jüni contributed equally to this work.
- Abbreviations and Acronyms
- confidence interval
- drug-eluting stent(s)
- everolimus-eluting stent(s)
- hazard ratio
- myocardial infarction
- paclitaxel-eluting stent(s)
- propensity score
- relative risk
- sirolimus-eluting stent(s)
- stent thrombosis
- target lesion revascularization
- target vessel revascularization
- Received October 10, 2010.
- Revision received December 6, 2010.
- Accepted January 2, 2011.
- American College of Cardiology Foundation
- Byrne R.A.,
- Iijima R.,
- Mehilli J.,
- et al.
- Räber L.,
- Windecker S.
- Stone GW, Rizvi A, Newman W, et al. Everolimus-eluting versus paclitaxel-eluting stents in coronary artery disease. N Engl J Med 362:1663–74.
- Schomig A.,
- Dibra A.,
- Windecker S.,
- et al.
- Kandzari D.E.,
- Leon M.B.,
- Popma J.J.,
- et al.
- Rasmussen K, Maeng M, Kaltoft A, et al. Efficacy and safety of zotarolimus-eluting and sirolimus-eluting coronary stents in routine clinical care Lancet 375:1090–9.
- Cutlip D.E.,
- Windecker S.,
- Mehran R.,
- et al.
- Windecker S.,
- Juni P.
- Daemen J.,
- Serruys P.W.
- Wenaweser P.,
- Daemen J.,
- Zwahlen M.,
- et al.
- Stettler C.,
- Allemann S.,
- Wandel S.,
- et al.
- Onuma Y.,
- Kukreja N.,
- Piazza N.,
- et al.
- ↵Byrne RA. Two-year outcomes after everolimus- or sirolimus-eluting stents in patients with coronary artery disease in the ISAR TEST 4 Trial. Paper presented at: Transcatheter Cardiovascular Therapeutics; September 24, 2010; Washington, DC.
- ↵Okkels Jensen L. Randomised comparison of everolimus-eluting and sirolimus-eluting stents in patients treated with percutaneous coronary intervention. Paper presented at: Transcatheter Cardiovascular Therapeutics; September 24, 2010; Washington, DC.