Abstract
Background/purpose
Drug-eluting stents (DES) reduce in-stent restenosis and repeat revascularization in comparison to bare metal stents. Individual DES vary, however, in regard to rates of restenosis and stent thrombosis; they also differ in regard to their platform and physical characteristics. The Promus Premier was designed to improve the performance of the Promus Element, with respect to conformability, trackability, and avoidance of longitudinal stent deformation; there is little published data, however, on clinical outcomes with Promus Premier.
Methods
We performed a registry study that compared 952 patients who underwent percutaneous coronary intervention with Promus Premier to 595 patients who received Taxus Liberte and 600 patients who received Xience V for a variety of indications. The primary endpoint was a composite of all-cause mortality, definite or probable stent thrombosis, myocardial infarction, and target vessel revascularization (TVR-MACE). Kaplan–Meier analysis and Cox proportional hazards regression were performed in order to compare the three stents in regard to outcomes at 1 year.
Results
Procedural success was highest with Premier (99.4%) when compared to Xience V (98.0%) and Taxus Liberte (97.3%; p < 0.001). Unadjusted survival analysis showed that TVR-MACE was less frequent with Premier in comparison to Taxus Liberte (p = 0.003), and similar frequency in comparison to Xience V (p = 0.16). Following multivariable adjustment, and using Xience V as the reference, there was only a borderline association of Promus Premier and lower rates of TVR-MACE (HR 0.69, 95% CI 0.45–1.04; p = 0.075).
Conclusion
Promus Premier demonstrates excellent procedural success rates and real-world outcomes that are similar to Xience V.
1
Introduction
Drug-eluting stents (DES) reduce the incidence of in-stent restenosis and repeat revascularization when compared to bare metal stents. There are subtle but important differences, however, in regard to the relative incidence of restenosis and stent thrombosis among individual DES platforms . Further complicating the interpretation of clinical data for DES is the continuing evolution of stent technology, with incremental changes in stent construction, alloy composition, antiproliferative drug delivery, and polymer. The Promus Premier was designed to increase procedural confidence by axial strength conformability and vessel straightening. Although the PLATINUM (Prospective, Randomized, Multicenter Trial to Assess an Everolimus-Eluting Coronary Stent System (Promus Element) for the Treatment of up to Two de Novo Coronary Artery Lesions) trial showed that Promus Element was noninferior to Xience V in regard to safety and efficacy , subsequent study showed that the Element stent was more prone to longitudinal stent deformation . The Promus Premier stent was therefore re-engineered to include four connectors for the proximal three hoops, compared to two connectors for the Element. Although this appeared to ameliorate the problem of longitudinal stent deformation in bench testing , there is no clinical data available regarding outcomes with the newer stent design. We therefore sought to compare clinical and procedural outcomes with Promus Premier to first and second generation DES. We hypothesized that the incidence of major adverse cardiovascular events at one year after receiving a platinum chromium everolimus-eluting Promus Premier DES (Boston Scientific, Marlborough, MA) would be similar to that of the cobalt chromium everolimus-eluting Xience V (Abbott Vascular, Abbott Park, IL) and lower than that of the platinum chromium paclitaxel-eluting Taxus Liberte (Boston Scientific).
2
Methods
The registry experience at the Washington Hospital Center, DES: Premier versus Taxus Liberte and Xience V (REWARDS Premier TLX) was designed as a multicenter, retrospective, nonrandomized study of 952 patients who underwent percutaneous coronary intervention (PCI) between December 2014 and October 2015 with a Promus Premier everolimus-eluting stent. These patients were compared to the previously enrolled REWARDS TLX cohort, which included 595 patients who underwent PCI with Taxus Liberte and 600 patients who received Xience V . Patients were enrolled from 10 hospitals in the United States. These centers were selected to obtain a representative cross-section of contemporary interventional cardiology practice. Sites were eligible if they demonstrated a volume of at least 750 percutaneous coronary interventions (PCI) a year that included a significant proportion of Promus Premier DES. Sites also were required to demonstrate adequate infrastructure and staff to enable compliance with the study requirements and procedures.
Study subjects were identified retrospectively by querying local catheterization laboratory or research databases for PCI that included use of Promus Premier after November 25, 2013. The only inclusion criteria were that patients were at least 18 years of age and underwent PCI with Promus Premier alone. Patients were excluded if they underwent PCI with any other DES during the index procedure or if they were unable to take dual antiplatelet therapy at the time of PCI. Given the retrospective design of the study, periprocedural pharmacotherapy and stent selection were at the discretion of the operator. All data were deidentified and collected using a secure, password-protected electronic data capture system. The study was conducted at each site in accordance with the regulations of the local or central institutional review board.
Clinical follow-up for each subject was collected in-hospital and at 9 to 12 months after PCI. For patients that were already part of a dedicated research database, this data was collected in a retrospective manner utilizing a HIPAA waiver of consent. For patients who underwent follow-up in a prospective manner, documentation of informed consent was required. A standardized telephone questionnaire was utilized for the ascertainment of clinical events at the 9 to 12 month follow-up.
The primary endpoint of the study was major adverse cardiac events (MACE), a composite of all-cause mortality, Q wave myocardial infarction (MI), target vessel revascularization (TVR), and stent thrombosis. MI was defined as the appearance of new pathological Q waves in the distribution of the target vessel with an increase in creatine kinase MB levels to ≥2 reference values. TVR was defined as revascularization occurring in any area along the previously treated vessel. Stent thrombosis was defined as Academic Research Consortium definite or probable stent thrombosis occurring within the target vessel. Secondary endpoints included the individual components of MACE, in addition to cardiac mortality (death from a cardiovascular cause) and target lesion revascularization (TLR, defined as the need for repeat PCI within the target lesion or 5 mm margins proximal and distal). Acute coronary syndrome (ACS) was defined as presentation with ST elevation myocardial infarction (STEMI) or non-ST elevation myocardial infarction (NSTEMI).
Continuous variables are presented as mean ± standard deviation; categorical variables are presented as percentages. Differences in continuous variables between groups were compared using Student’s t test. Categorical variables were compared using the χ 2 test or Fisher’s exact test when appropriate. Means for different groups were compared with an analysis of variance (ANOVA). A p value of <0.05 was considered statistically significant. Kaplan–Meier analysis was used to analyze the association between stent type (Promus Premier, Xience V, and Taxus Liberte) and MACE at up to 1 year after PCI. The Bonferroni method was used to adjust log-rank p values for multiple comparisons. To test the independent association of stent type and MACE, a multivariable Cox proportional hazards regression was performed. Covariables were selected based upon significant univariable p values and overall clinical relevance. Covariables included in the model, in addition to stent type (with Xience V as reference), included age (per 10 years), female gender, African-American race, presentation with STEMI, total stented length (per 10 mm), and a history of: diabetes mellitus, hypertension, previous myocardial infarction, coronary artery bypass grafting (CABG), chronic renal insufficiency (CRI), and peripheral vascular disease (PVD). Covariables in the model are expressed as hazard ratios (HR) with 95% confidence intervals (CIs). Statistical analyses were performed using SAS version 9.1 (SAS Institute, Cary, NC).
2
Methods
The registry experience at the Washington Hospital Center, DES: Premier versus Taxus Liberte and Xience V (REWARDS Premier TLX) was designed as a multicenter, retrospective, nonrandomized study of 952 patients who underwent percutaneous coronary intervention (PCI) between December 2014 and October 2015 with a Promus Premier everolimus-eluting stent. These patients were compared to the previously enrolled REWARDS TLX cohort, which included 595 patients who underwent PCI with Taxus Liberte and 600 patients who received Xience V . Patients were enrolled from 10 hospitals in the United States. These centers were selected to obtain a representative cross-section of contemporary interventional cardiology practice. Sites were eligible if they demonstrated a volume of at least 750 percutaneous coronary interventions (PCI) a year that included a significant proportion of Promus Premier DES. Sites also were required to demonstrate adequate infrastructure and staff to enable compliance with the study requirements and procedures.
Study subjects were identified retrospectively by querying local catheterization laboratory or research databases for PCI that included use of Promus Premier after November 25, 2013. The only inclusion criteria were that patients were at least 18 years of age and underwent PCI with Promus Premier alone. Patients were excluded if they underwent PCI with any other DES during the index procedure or if they were unable to take dual antiplatelet therapy at the time of PCI. Given the retrospective design of the study, periprocedural pharmacotherapy and stent selection were at the discretion of the operator. All data were deidentified and collected using a secure, password-protected electronic data capture system. The study was conducted at each site in accordance with the regulations of the local or central institutional review board.
Clinical follow-up for each subject was collected in-hospital and at 9 to 12 months after PCI. For patients that were already part of a dedicated research database, this data was collected in a retrospective manner utilizing a HIPAA waiver of consent. For patients who underwent follow-up in a prospective manner, documentation of informed consent was required. A standardized telephone questionnaire was utilized for the ascertainment of clinical events at the 9 to 12 month follow-up.
The primary endpoint of the study was major adverse cardiac events (MACE), a composite of all-cause mortality, Q wave myocardial infarction (MI), target vessel revascularization (TVR), and stent thrombosis. MI was defined as the appearance of new pathological Q waves in the distribution of the target vessel with an increase in creatine kinase MB levels to ≥2 reference values. TVR was defined as revascularization occurring in any area along the previously treated vessel. Stent thrombosis was defined as Academic Research Consortium definite or probable stent thrombosis occurring within the target vessel. Secondary endpoints included the individual components of MACE, in addition to cardiac mortality (death from a cardiovascular cause) and target lesion revascularization (TLR, defined as the need for repeat PCI within the target lesion or 5 mm margins proximal and distal). Acute coronary syndrome (ACS) was defined as presentation with ST elevation myocardial infarction (STEMI) or non-ST elevation myocardial infarction (NSTEMI).
Continuous variables are presented as mean ± standard deviation; categorical variables are presented as percentages. Differences in continuous variables between groups were compared using Student’s t test. Categorical variables were compared using the χ 2 test or Fisher’s exact test when appropriate. Means for different groups were compared with an analysis of variance (ANOVA). A p value of <0.05 was considered statistically significant. Kaplan–Meier analysis was used to analyze the association between stent type (Promus Premier, Xience V, and Taxus Liberte) and MACE at up to 1 year after PCI. The Bonferroni method was used to adjust log-rank p values for multiple comparisons. To test the independent association of stent type and MACE, a multivariable Cox proportional hazards regression was performed. Covariables were selected based upon significant univariable p values and overall clinical relevance. Covariables included in the model, in addition to stent type (with Xience V as reference), included age (per 10 years), female gender, African-American race, presentation with STEMI, total stented length (per 10 mm), and a history of: diabetes mellitus, hypertension, previous myocardial infarction, coronary artery bypass grafting (CABG), chronic renal insufficiency (CRI), and peripheral vascular disease (PVD). Covariables in the model are expressed as hazard ratios (HR) with 95% confidence intervals (CIs). Statistical analyses were performed using SAS version 9.1 (SAS Institute, Cary, NC).
3
Results
Overall, 2147 patients were included in the study: 952 received Promus Premier, 600 received Xience V, and 595 received Taxus Liberte. Average age was 64.5 ± 11.2 years, 69.4% of patients were male, 86.4% were Caucasian, 35.3% had a history of diabetes mellitus, 9.9% presented with STEMI, and 17.5% presented with NSTEMI. There were significant differences between the three groups in regard to baseline characteristics ( Table 1 ): patients receiving Premier tended to be older, have a history of CRI or hypertension, and present with STEMI or NSTEMI, whereas patients receiving Xience were more likely to present because of a positive stress test and patients receiving Taxus were more likely Caucasian. P2Y12 inhibition varied significantly according to the stent utilized, largely reflecting the different time periods under study: patients receiving Taxus or Xience almost exclusively received clopidogrel, whereas 17.3% of the Premier group received prasugrel and 12.2% received ticagrelor ( Table 2 ). There were also important differences in procedural anticoagulation: patients receiving a Premier stent were more likely to receive bivalirudin, whereas patients receiving Taxus were more likely to receive both heparin and glycoprotein IIb/IIIa inhibitors. Stented length also varied by group, with 20.3 ± 8.6 mm for Xience, 21.6 ± 10.2 for Premier, and 22.7 ± 12.2 for Taxus (p < 0.001 for trend; Table 3 ). Procedural success also varied by stent platform; it was lowest for Taxus at 97.3%, intermediate for Xience at 98.0%, and highest for Premier at 99.4% (p < 0.001 for trend.) Major adverse in-hospital outcomes were uncommon, with no difference between stents in regard to all-cause mortality, stent thrombosis, TVR, or TLR ( Table 4 ). There was, however, a higher rate of myocardial infarction with Taxus Liberte (1.3%) and Xience V (1.0%) in comparison to Promus Premier (0.1%; p = 0.002).
| Taxus Liberte (n = 595) | Xience V (n = 600) | Promus Premier (n = 952) | p Value | |
|---|---|---|---|---|
| Age (years) | 64.6 ± 11.3 | 63.2 ± 11.2 | 65.3 ± 11.1 | <0.001 |
| Male | 416 (69.9%) | 416 (69.3%) | 658 (69.1%) | 0.95 |
| Caucasian | 542 (92.5%) | 445 (75.3%) | 852 (89.6%) | <0.001 |
| Hispanic | 17 (3.0%) | 3 (1.2%) | 17 (1.8%) | 0.15 |
| History of smoking | 302 (50.8%) | 272 (45.3%) | 444 (46.3%) | 0.13 |
| Diabetes mellitus | 199 (33.4%) | 211 (35.2%) | 351 (36.6%) | 0.44 |
| Previous myocardial infarction | 238 (24.8%) | 142 (23.9%) | 138 (23.0%) | 0.70 |
| Previous coronary artery bypass grafting | 130 (21.8%) | 125 (20.8%) | 213 (22.2%) | 0.81 |
| Hypertension | 446 (75.0%) | 506 (84.3%) | 812 (84.8%) | >0.001 |
| Hyperlipidemia | 469 (78.8%) | 515 (85.8%) | 791 (82.6%) | 0.006 |
| Chronic renal insufficiency | 33 (5.5%) | 40 (6.7%) | 84 (8.8%) | 0.047 |
| Peripheral vascular disease | 55 (9.2%) | 71 (11.8%) | 110 (11.5%) | 0.28 |
| ST segment elevation myocardial infarction | 61 (10.3%) | 37 (6.2%) | 115 (12.1%) | <0.001 |
| Non-ST segment elevation myocardial infarction | 92 (15.5%) | 75 (12.6%) | 208 (21.8%) | <0.001 |
| Cardiogenic shock | 1 (0.2%) | 4 (0.7%) | 3 (0.3%) | 0.43 |
| Positive stress test | 193 (32.4%) | 258 (43.2%) | 201 (21.1%) | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
