Author + information
- Received May 18, 2010
- Revision received July 14, 2010
- Accepted July 25, 2010
- Published online December 1, 2010.
- Adriano Caixeta, MD, PhD⁎,
- Alexandra J. Lansky, MD⁎,
- Patrick W. Serruys, MD, PhD†,
- James B. Hermiller, MD‡,
- Peter Ruygrok, MD§,
- Yoshinobu Onuma, MD†,
- Paul Gordon, MD∥,
- Manejeh Yaqub, MD¶,
- Karine Miquel-Hebert, PhD#,
- Susan Veldhof, RN#,
- Poornima Sood, MD, MBA¶,
- Xiaolu Su, MS¶,
- Lalitha Jonnavithula, MBA¶,
- Krishnankutty Sudhir, MD, PhD¶,
- Gregg W. Stone, MD⁎,⁎ (, )
- SPIRIT II and III Investigators
- ↵⁎Reprint requests and correspondence:
Dr. Gregg W. Stone, Columbia University Medical Center and the Cardiovascular Research Foundation, 111 East 59th Street, 11th Floor, New York, New York 10022
Objectives The purpose of this study was to investigate long-term 3-year clinical outcomes of an everolimus-eluting stent (EES) versus a paclitaxel-eluting stent (PES).
Background Compared with PES, EES reduced target vessel failure and major adverse cardiac events at 2 years. Whether the benefits of EES are sustained at 3 years has not been reported.
Methods In the SPIRIT II (A Clinical Evaluation of the XIENCE V Everolimus Eluting Coronary Stent System in the Treatment of Patients With De Novo Native Coronary Artery Lesions) and SPIRIT III (A Clinical Evaluation of the Investigational Device XIENCE V Everolimus Eluting Coronary Stent System [EECSS] in the Treatment of Subjects With De Novo Native Coronary Artery Lesions) trials, 1,302 patients were randomly assigned to EES (n = 892) or PES (n = 410). We report the 3-year clinical follow-up of this patient-level pooled analysis.
Results At 3 years, EES compared with PES resulted in a significant reduction in myocardial infarction (3.8% vs. 6.7%; relative risk [RR]: 0.56; 95% confidence interval [CI]: 0.34 to 0.94; p = 0.04), and target lesion revascularization (6.8% vs. 12.7%; RR: 0.53; 95% CI: 0.37 to 0.77; p = 0.001). Everolimus-eluting stents resulted in a significant reduction in target vessel failure (13.7% vs. 19.5%; RR: 0.70; 95% CI: 0.54 to 0.92; p = 0.01), and major adverse cardiac events (9.1% vs. 16.3%; RR: 0.56; 95% CI: 0.41 to 0.76; p = 0.0004). The cumulative rates of Academic Research Consortium–defined definite or probable stent thrombosis were 1.2% in EES patients and 1.9% in PES patients (RR: 0.64; 95% CI: 0.25 to 1.68; p = 0.43).
Conclusions In this patient-level pooled analysis, EES compared with PES resulted in a significant and persistent reduction in target vessel failure and major adverse cardiac events at 3 years due to fewer myocardial infarction and ischemic target lesion revascularization events, which is consistent with superior safety and efficacy of the EES platform. (A Clinical Evaluation of the XIENCE V Everolimus Eluting Coronary Stent System in the Treatment of Patients With De Novo Native Coronary Artery Lesions [SPIRIT II]; NCT00180310) (SPIRIT III: A Clinical Evaluation of the Investigational Device XIENCE V Everolimus Eluting Coronary Stent System [EECSS] in the Treatment of Subjects With De Novo Native Coronary Artery Lesions [SPIRIT III]; NCT00180479)
First-generation stainless steel drug-eluting stents (DES), including sirolimus-eluting stents (SES) and paclitaxel-eluting stents (PES), have significantly reduced rates of restenosis and target lesion revascularization (TLR) compared with bare-metal stents (1–3). However, delayed re-endothelialization and suppression of vascular healing with both SES and PES have been associated with increased rates of late and very late stent thrombosis (4–7), as well as myocardial infarction (MI) and death (8). Furthermore, in-stent restenosis after DES can still occur, especially in patients with complex coronary artery disease (9). Additional limitations of first-generation DES include enhanced platelet aggregation and inflammation, polymer hypersensitivity reactions, high rates of late-acquired incomplete stent apposition, and strut fractures (10–14). Improved stent designs with thinner struts and more biocompatible polymer coatings and optimized drug elution profiles might favorably influence DES performance.
The clinical safety and efficacy of the everolimus-eluting stent (EES) is currently being tested in the SPIRIT family of randomized controlled trials. In both the SPIRIT II (A Clinical Evaluation of the XIENCE V Everolimus Eluting Coronary Stent System in the Treatment of Patients With De Novo Native Coronary Artery Lesions) (15) and SPIRIT III (A Clinical Evaluation of the Investigational Device XIENCE V Everolimus Eluting Coronary Stent System [EECSS] in the Treatment of Subjects With De Novo Native Coronary Artery Lesions) (16) trials, the Xience V EES was superior to the Taxus PES for primary angiographic end points in reducing in-stent late loss at 6 months and in-segment late loss at 8 months, respectively. In the SPIRIT III trial, EES also demonstrated superiority to PES for the coprimary clinical end point of target vessel failure (TVF) and major adverse cardiac events (MACE) at 2 years. The 2-year outcomes of the pooled SPIRIT II and SPIRIT III trials demonstrated the benefit of the EES at 2 years (17). Whether the benefits of EES are sustained or increase with longer-term follow-up remains unknown. We, therefore, examined the 3-year outcomes in a pooled analysis of the SPIRIT II and III trials.
Patient population and study design
This study represents a patient-level pooled analysis of the SPIRIT II and SPIRIT III prospective, multicenter, single-blind, controlled randomized clinical trials, in which patients with de novo lesions were treated with either polymer-based EES (Xience V, Abbott Vascular, Santa Clara, California) or with polymer-based PES (Taxus Express2, Boston Scientific, Natick, Massachusetts). The protocol and principal results of each trial have been published previously (15,16). In brief, in the SPIRIT II trial, 300 patients in Europe, India, and New Zealand were randomized in a 3:1 ratio to receive EES (Xience V, n = 223) or PES (n = 77 [Taxus Express2, n = 59 or Taxus Liberté, n = 17, with 1 patient receiving a nonstudy stent]) (15). In SPIRIT III, 1,002 patients were randomized at 65 U.S. sites in a 2:1 ratio to receive the Xience V EES (n = 669) or the Taxus Express2 PES (n = 333). The studies were approved by the institutional review board or ethics committee at each participating center, and all patients provided written informed consent.
Medication and follow-up
The trial protocols recommended indefinite use of aspirin ≥75 mg daily and clopidogrel 75 mg daily for a minimum of 6 months. In SPIRIT II, protocol-specified angiographic follow-up was planned in all patients at 6 months and in a subgroup of 152 patients at 2 years. In SPIRIT III, angiographic follow-up was planned in a subset of 564 patients at 8 months. Clinical follow-up was performed at 1, 6, and 9 months, and 1 year and then annually through 5 years in both studies. Three-year clinical follow-up was available in 1,224 patients (94%), including 839 EES patients and 385 PES patients.
Definitions and clinical end points
The safety end points for this analysis included rates of cardiac death, MI, stent thrombosis defined as definite and probable by Academic Research Consortium (ARC) (18), as well as the composite of cardiac death or MI. Efficacy end points included target vessel revascularization (TVR) and TLR. Combined safety and efficacy end points included TVF, defined as the composite of cardiac death, MI, or ischemia-driven TVR by either percutaneous coronary intervention or bypass graft surgery, and MACE, defined as the composite of cardiac death, MI, or ischemia-driven TLR.
Data management and statistical analysis
Study monitors verified all case report form data on site. All MACE events were adjudicated by independent clinical event committees blinded to treatment assignment with review of original source documentation. Retrospectively, all stent thrombosis events were readjudicated using the ARC definitions (18).
Given the similarity in the inclusion and exclusion criteria and test arms in the 2 trials, the patient-level databases from the 2 trials were pooled comparing EES and PES. Continuous variables are expressed as mean ± SD and compared by t test. Binary variables are presented as counts and percentages and were compared using the Fisher exact test. All p values are 2-tailed and not from formal hypothesis testing and are shown for descriptive purposes only. All analyses are by intention-to-treat using all patients randomized in the study, regardless of the treatment actually received. Patients lost to follow-up in whom no event had occurred before the follow-up windows were not included in the denominator for calculations of binary end points. Relative risk was calculated as the event rate of the EES arm divided by the event rate of the PES arm. Three-year outcomes are summarized as Kaplan-Meier estimates and compared using log-rank test and hazard ratios. Time-to-event hazard curves also are presented and compared by log-rank test. To assess events occurring between 1 and 3 years, a landmark analysis was performed. Cox proportional hazard models were used to calculate the hazard ratios of the EES versus PES groups. To explore the interaction between subgroups and the treatment effect for TVF and MACE, a post hoc logistic regression analysis was performed and p values were calculated by Wald chi-square statistics. All statistical tests were 2-tailed and a p value <0.05 denoted significance. All statistical analyses were performed by SAS software (version 9.1.3, SAS Institute, Cary, North Carolina).
Baseline and procedural characteristics
In the SPIRIT II and III trials, 1,302 patients at 93 international centers were randomly assigned to receive either EES (n = 892) or PES (n = 410). The baseline clinical and angiographic characteristics were well matched between the 2 groups, except for a slightly lower incidence of unstable angina in patients randomized to EES (Table 1). The overall follow-up rate at 3 years for SPIRIT II and III patients was 94.1% in the EES arm and 93.9% in the PES arm. Aspirin use was high in both groups throughout the 3-year follow-up, whereas thienopyridine use declined progressively over time. Compliance with dual antiplatelet therapy was also similar between the EES and PES groups at 3-year follow-up (43.4% vs. 47.0%; p = 0.27).
Clinical outcomes at 3 years
Table 2 displays the major clinical end points at 3 years, and between 1 and 3 years. At 3-year follow-up, treatment with EES resulted in a 30% decrease in the rate of TVF (13.7% vs. 19.5%; p = 0.01) and a 44% decrease in MACE (9.1% vs. 16.3%; p = 0.0004). The reduction in TVF and MACE with EES at 3 years was driven by both a 45% relative reduction in non–Q-wave MI and a 41% relative reduction in ischemia-driven TLR (Table 2). In addition, there is a 41% decrease in the composite of death or MI (6.6% vs. 11.2%; p = 0.009). As shown in Figure 1, the absolute benefits of EES compared with PES achieved by 1 year continued to spread through the 3-year follow-up. Between 1 and 3 years of follow-up, EES resulted in a 36% decrease in the rate of TVF (6.3% vs. 9.8%, p = 0.04) and in a 42% decrease in MACE (3.9% vs. 6.8%, p = 0.04) (Fig. 2) (Table 2).
The rates of ARC-defined definite and probable stent thrombosis were not significantly different between the 2 stent types (1.2% with EES and 1.9% with PES; p = 0.43) (Table 3). The rate of very late (>1 year) definite or probable stent thrombosis was numerically lower but not significantly different in the EES group (0.5% of EES and 1.1% of PES; p = 0.26) (Table 3).
Logistic regression analysis was performed to assess whether the reduction in TVF and MACE in EES-treated compared with PES-treated patients was consistent among different subgroups. Treatment benefit of EES was seen in all subgroups tested for TVF (Fig. 3) and for MACE (Fig. 4), except for patients with diabetes. Among the 326 patients with diabetes, the 3-year rate of MACE was 12.0% in the EES group compared with 10.8% in the PES group (relative risk [RR]: 1.12, 95% confidence interval [CI]: 0.57 to 2.21), whereas, in those without diabetes, the rate was 8.0% in the EES group compared with 18.3% in the PES group (RR: 0.43, 95% CI: 0.30 to 0.63). The relative reduction in MACE with EES compared with PES was comparable in both patients undergoing and not undergoing angiographic follow-up.
This patient-level pooled analysis demonstrates the significant and sustained reduction in TVF and MACE among patients with de novo coronary artery lesions treated with EES compared with PES to 3-year follow-up. Everolimus-eluting stents demonstrated improved efficacy and safety over time, with robust reductions in ischemia-driven TLR and TVR as well as significant and sustained reductions in MI and composite death or MI.
Long-term safety and efficacy of EES
This report provides information on the latest (3-year) follow-up of patients treated with EES. After the completion of 1-year follow-up, patients treated with EES compared with those treated with PES had fewer episodes of MI, TLR, TVR, and death. The time-to-event curves for TVF and MACE between 1 and 3 years have continued to diverge, resulting in a significant 30% relative reduction in TVF and a 44% relative reduction in MACE at 3-year follow-up. The improved clinical outcomes demonstrated at 1 year and 2 years (17) in patients receiving EES compared with those receiving PES have become even more pronounced at 3-year follow-up.
The present analysis is consistent with the clinical results from 2 recent randomized studies demonstrating superior safety and efficacy of EES compared with PES. In the large-scale SPIRIT IV (Clinical Evaluation of the Xience V Everolimus Eluting Coronary Stent System in the Treatment of Patients With De Novo Native Coronary Artery Lesions IV) trial (19), randomization of 3,687 patients—including a more complex patient subgroup with more than 1,100 patients with diabetes—to EES (Xience V) or PES (Taxus Express2) resulted in a significant 38% relative reduction in the 1-year end points of target lesion failure (a composite of cardiac death, target vessel MI, or ischemia-driven TLR), cardiac death, MI, or ischemia-driven TLR at 1-year follow-up. Likewise, in the COMPARE (Second-Generation Everolimus-Eluting and Paclitaxel-Eluting Stents in Real-Life Practice) trial (20), a randomization of 1,800 unselected patients to EES (Xience V) or PES (Taxus Liberté) also resulted in a significant 33% relative reduction in the 1-year primary end point (all-cause mortality, nonfatal MI, and TVR) and nonfatal MI. The current analysis confirms and extends the results from the SPIRIT IV and COMPARE trials in demonstrating that EES continues to be both safer and more efficacious than PES through 3-year follow-up.
Use of first-generation SES and PES has been associated with increased rates of stent thrombosis compared with bare-metal stents, particularly after 1 year of follow-up (4–7). In this pooled analysis, the rates of stent thrombosis were comparable between EES and PES within the first year after implantation. Nonetheless, there was a trend for fewer ARC-defined definite or probable very late stent thrombosis episodes with EES compared with PES (0.5% vs. 1.1%, respectively). However, the SPIRIT II and III trials were underpowered to evaluate differences in stent thrombosis between these 2 devices. Longer-term follow-up from the larger SPIRIT IV (19) and COMPARE (20) trials are required to determine whether very late stent thrombosis is less common with EES compared with PES, especially in complex coronary lesions.
The reduced rate in the composite safety and efficacy end points of TVF and MACE with EES compared with PES in the present analysis was attributable to fewer TLR and periprocedural non–Q-wave MI events. The lower rate of TLR (i.e., superior efficacy) with EES compared with PES may be directly attributed to reductions in late lumen loss, binary restenosis, and neointimal hyperplasia in the target lesion, as previously described (15,16). Importantly, the early reduction in TLR with EES is sustained through 3 years, although longer-term (5 years) follow-up is required to assess the late durability of these clinical advantages. In addition to efficacy, EES was safer than PES, with fewer MIs and composite death or MI, as also seen in the SPIRIT IV (19) and COMPARE (20) trials. In the SPIRIT II and III trials, the increased rates of MI were due mostly to an increase in periprocedural non–Q-wave MIs. The likely cause is less side-branch compromise due to thinner polymer/stent interface of EES compared with PES (89 μm vs. 148 μm), which was also shown to be the possible cause of lower rates of 30-day MI with the Express bare-metal stent compared with PES in the TAXUS V trial (9). Lansky et al. (21) have recently demonstrated less side branch occlusion with EES compared with PES in the SPIRIT III trial, a finding correlated with lower non–Q-wave MIs in the EES treatment group.
The 3-year reduction in TVF with EES compared with PES was consistent across all subgroups except patients with diabetes. A significant interaction was noted between diabetic status and stent type with respect to MACE, similar to findings reported in SPIRIT IV at 1 year (19). Conflicting findings from prior studies comparing SES and PES in diabetic patients also have been reported; some studies suggested that stents eluting everolimus or sirolimus might offer advantages over PES (22,23), whereas others reported disadvantages (24,25). These drugs (e.g., sirolimus and everolimus) may be less effective in inhibiting neointimal formation in diabetic compared with nondiabetic patients. In the SPIRIT IV trial, target lesion failure rates were similar in diabetic patients receiving an EES or PES at 1 year, though stent thrombosis rates were nonsignificantly lower in the EES treatment group. Differences in the mechanisms of action of the drugs may explain the inconsistent outcomes with EES compared with PES in the diabetic cohort (26). In the pooled analysis, the performance of the EES in the diabetic subgroup was similar to nondiabetics patients, whereas PES performance was significantly better than in nondiabetics patients. However, it should be noted that post hoc subgroup analyses often are statistically underpowered, which can lead to overstated and misleading results (27).
Several limitations of the present study should be acknowledged. First, the SPIRIT II and SPIRIT III trials were originally designed and powered to evaluate a reduction in angiographic in-stent and in-segment late loss and noninferiority for the clinical end point TVF for the entire study population as a whole. Thus, the results of the present patient-pooled analysis combining the 2 studies are hypothesis-generating only. A few minor differences exist between the 2 studies, including the percentage of patients undergoing follow-up angiography and the timing of follow-up, as well as differing proportions of patients receiving PES. Second, although 2 larger, randomized trials with more complex and previously unstudied lesions demonstrated a significant reduction in the primary clinical end point of target lesion failure and MACE in EES- compared with PES-treated patients (19,20), the results of the present study cannot be generalized for the use of EES beyond the approved indications (off-label use). Third, the present study is underpowered to definitively examine low-frequency adverse events such as stent thrombosis, MI, and death. Finally, comparative trials with emerging DES are required to determine the optimal platform for specific patient and lesion subtypes, particularly patients with diabetes.
In this patient-level pooled analysis, EES compared with PES resulted in significant and sustained reductions in TVF and MACE at 3 years due to fewer MI and ischemic TLR events, which is consistent with superior safety and efficacy of the EES platform.
Sponsored by Abbott Vascular. Dr. Lansky has received research grants from The Medicines Co., Cordis, Boston Scientific, Medtronic, and Abbott. Dr. Hermiller is a research consultant for Abbott Vascular. Dr. Gordon has received research support from Abbott Vascular. Dr. Yaqub, Dr. Miquel-Hebert, Susan Veldhof, Dr. Sood, Xiaolu Su, Lalitha Jonnavithula, and Dr. Sudhir are employed by and hold equity in Abbott Vascular. Dr. Stone serves on the scientific advisory boards for and has received honoraria from Abbott Vascular and Boston Scientific. All other authors have reported that they have no relationships to disclose.
- Abbreviations and Acronyms
- Academic Research Consortium
- drug-eluting stent(s)
- everolimus-eluting stent(s)
- major adverse cardiac events
- myocardial infarction
- paclitaxel-eluting stent(s)
- sirolimus-eluting stent(s)
- target lesion revascularization
- target vessel failure
- target vessel revascularization
- Received May 18, 2010.
- Revision received July 14, 2010.
- Accepted July 25, 2010.
- American College of Cardiology Foundation
- Kastrati A.,
- Dibra A.,
- Spaulding C.,
- et al.
- Finn A.V.,
- Joner M.,
- Nakazawa G.,
- et al.
- Pfisterer M.,
- Brunner-La Rocca H.P.,
- Buser P.T.,
- et al.,
- BASKET-LATE Investigators
- Nebeker J.R.,
- Virmani R.,
- Bennett C.L.,
- et al.
- Hofma S.H.,
- van der Giessen W.J.,
- van Dalen B.M.,
- et al.
- Hoffmann R.,
- Morice M.C.,
- Moses J.W.,
- et al.
- Maehara A.,
- Mintz G.S.,
- Lansky A.J.,
- et al.
- Nakazawa G.,
- Finn A.V.,
- Vorpahl M.,
- et al.
- Onuma Y.,
- Serruys P.W.,
- Kukreja N.,
- et al.,
- SPIRIT II and III Investigators
- Cutlip D.E.,
- Windecker S.,
- Mehran R.,
- et al.,
- Academic Research Consortium
- ↵Lansky AJ, Applegate RJ, Hermiller JB, et al. Side-branch occlusion with Xience V everolimus-eluting and Taxus Express paclitaxel-eluting stents: 2-year results from the SPIRIT III randomized trial. Paper presented at: Transcatheter Cardiovascular Therapeutics Conference; September 23, 2009; San Francisco, CA.
- Billinger M.,
- Beutler J.,
- Taghetchian K.R.,
- et al.
- Stettler C.,
- Allemann S.,
- Wandel S.,
- et al.
- Caixeta A.,
- Leon M.B.,
- Lansky A.J.,
- et al.