- Email: [email protected]
Three-Year Patient-Related and Stent-Related Outcomes of Second-Generation Everolimus-Eluting Xience V Stents Versus Zotarolimus-Eluting Resolute Stents in Real-World Practice (from the Multicenter Prospective EXCELLENT and RESOLUTE-Korea Registries)
Three-Year Patient-Related and Stent-Related Outcomes of Second-Generation Everolimus-Eluting Xience V Stents Versus Zotarolimus-Eluting Resolute Stents in Real-World Practice (from the Multicenter Prospective EXCELLENT and RESOLUTE-Korea Registries) Joo Myung Lee, MD, MPHa, Kyung Woo Park, MD, PhDa, Jung-Kyu Han, MD, PhDa, Han-Mo Yang, MD, PhDa, Hyun-Jae Kang, MD, PhDa, Bon-Kwon Koo, MD, PhDa, Jang-Whan Bae, MD, PhDb, Sung-Il Woo, MD, PhDc, Jin Sik Park, MD, PhDd, Dong-Kyu Jin, MD, PhDe, Dong Woon Jeon, MD, PhDf, Seok Kyu Oh, MD, PhDg, Jong-Seon Park, MD, PhDh, Doo-Il Kim, MD, PhDi, Min Su Hyon, MD, PhDj, Hui-Kyung Jeon, MD, PhDk, Do-Sun Lim, MD, PhDl, Myeong-Gon Kim, MD, PhDm, Seung-Woon Rha, MD, PhDn, Sung-Ho Her, MD,PhDo, Jin-Yong Hwang, MD, PhDp, Sanghyun Kim, MD, PhDq, Young Jin Choi, MD, PhDd, Jin Ho Kang, MD, PhDr, Keon-Woong Moon, MD, PhDs, Yangsoo Jang, MD, PhDt, and Hyo-Soo Kim, MD, PhDa,* Long-term outcomes are imperative to confirm safety of drug-eluting stents. There have been 2 randomized controlled trials comparing everolimus-eluting stents (EESs) and Resolute zotarolimus-eluting stents (ZES-Rs). To date, long-term clinical outcomes of these stents were limited to only 1 report, which has recently reported 4-year comparisons of these stents. Therefore, more evidence is needed regarding long-term clinical outcomes of the second-generation stents. This study compared the long-term clinical outcomes of EES with ZES-R in “all-comer” cohorts up to 3-year follow-up. The EXCELLENT and RESOLUTEKorea registries prospectively enrolled 3,056 patients treated with EES and 1,998 with ZESR, respectively, without exclusions. Stent-related composite outcomes (target lesion failure) and patient-related composite events up to 3-year follow-up were compared in crude and propensity scoreematched analyses. Of 5,054 patients, 3,830 patients (75.8%) had off-label indication (2,217 treated with EES and 1,613 treated with ZES-R). The stent-related outcome (189 [6.2%] vs 127 [6.4%], p [ 0.812) and the patient-related outcome (420 [13.7%] vs 250 [12.5%], p [ 0.581) did not differ between EES and ZES-R, respectively, at 3 years, which was corroborated by similar results from the propensity scoreematched cohort (hazard ratio [HR] 0.92, 95% confidence interval [CI] 0.70 to 1.20, p [ 0.523 and 0.85, 95% CI 0.70 to 1.02, p [ 0.081, for stent- and patient-related outcomes, respectively). The rate of definite or probable stent thrombosis up to 3 years (22 [0.7%] vs 10 [0.5%], p [ 0.370) was also similar. The rate of very late definite or probable stent thrombosis was very low and comparable between the 2 stents (3 [0.1%] vs 1 [0.1%], p [ 0.657). In multivariate analysis, chronic renal failure (adjusted HR 3.615, 95% CI 2.440 to 5.354, p <0.001) and off-label indication (adjusted HR 1.782, 95% CI 1.169 to 2.718, p [ 0.007)
a Department of Internal Medicine and Cardiovascular Center, Seoul National University Hospital, Seoul, Republic of Korea; bChungbuk National University, Cheongju, Republic of Korea; cInha University Hospital, Incheon, Republic of Korea; dSejong General Hospital, Sejong Heart Institute, Bucheon, Republic of Korea; eSoonchunhyang University Cheonan Hospital, Cheonan, Republic of Korea; fNational Health Insurance Corporation Ilsan Hospital, Goyang, Republic of Korea; gWonkwang University School of Medicine, Iksan, Republic of Korea; hYeungnam University Medical Center, Daegu, Republic of Korea; iInje University, Haeundae Paik Hospital, Busan, Republic of Korea; jSoonchunhyang University Hospital, Seoul, Republic of Korea; kUijeongbu St Mary’s Hospital, Uijeongbu, Republic of Korea; lKorea University Anam Hospital, Seoul, Republic of Korea; mKyunghee University College of Medicine, Seoul, Republic of Korea; nKorea University Guro, Hospital, Seoul, Republic of Korea; oDaejeon St. Mary’s Hospital, Daejeon, Republic of Korea; pGyeongsang National University Hospital, Jinju,
0002-9149/14/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2014.07.065
Republic of Korea; qBoramae Medical Center, Seoul, Republic of Korea; Kangbuk Samsung Hospital, Seoul, Republic of Korea; sSt. Vincent’s Hospital, Suwon, Republic of Korea; and tYonsei University College of Medicine, Seoul, Republic of Korea. Manuscript received July 6, 2014; revised manuscript received and accepted July 30, 2014. Dr. Lee and Dr. K.W. Park contributed equally to this study. This study was supported by grant A102065 from the Korea Healthcare technology R&D Project, Ministry of Health and Welfare, Republic of Korea, and grant A062260 from the Innovative Research Institute for Cell Therapy, Seoul National University Hospital, Seoul, Korea, sponsored by the Ministry of Health, Welfare & Family, Republic of Korea. See page 1336 for disclosure information. *Corresponding author: Tel: (þ82) 2-2072-2226; fax: (þ82) 2-7668904. E-mail address: [email protected] (H.-S. Kim). r
www.ajconline.org
1330
The American Journal of Cardiology (www.ajconline.org)
were the strongest predictors of target lesion failure at 3 years. In conclusion, both stents showed comparable safety and efficacy at 3-year follow-up in this robust real-world registry with unrestricted use of EES and ZES-R. Overall incidences of target lesion failure and definite stent thrombosis, including very late stent thrombosis, were low, even in the patients with off-label indications, suggesting excellent long-term safety and sustained efficacy of both types of second-generation drug-eluting stents. Ó 2014 Elsevier Inc. All rights reserved. (Am J Cardiol 2014;114:1329e1338) Second-generation drug-eluting stents (DESs), developed with the purpose of improving long-term safety, have almost completely replaced the first-generation DESs in the treatment of coronary artery disease. This has mostly been due to the representative second-generation DES, Xience V/Promus everolimus-eluting stent (EES; Abbott Vascular, Santa Clara, California), which showed sustained safety and efficacy up to 3 years in comparisons with first-generation DESs.1e4 Another second-generation DES, the Resolute zotarolimus-eluting stent (ZES-R; Medtronic Cardiovascular, Minneapolis, Minnesota) showed comparable clinical outcomes up to 2-year follow-up, compared with EES,5,6 and is also used often in daily routine practice. Although 4-year comparison of these stents was recently published,7 long-term clinical outcomes of these stents were limited only to this report despite its wide use in routine clinical practice, especially regarding the ZES-R. Thus, more data about their everyday use are needed. Previously, we published the 1-year comparison of EES versus ZES-R in 5,054 patients from 2 multicenter all-comers registries (the EXCELLENT and RESOLUTE-Korea registries).8 The purpose of this study was to evaluate the long-term safety and efficacy of the Xience V/Promus EES and Resolute ZES up to 3-year follow-up in everyday real-world use from large prospective registries of these DESs. Methods Extended description of study methods including statistical analysis is presented in Supplementary Data. This study evaluated 3-year clinical outcomes of Xience V/Promus EES and Resolute ZES from 2 prospective multicenter registries: the EXCELLENT and RESOLUTE-Korea, which enrolled all-comers treated with 1 EESs or ZES-R (3,056/ 29 and 1,998/25 patients/participating centers, respectively) without exclusions (Supplementary Figure 1).8 The study protocol was approved by the ethics committee at each participating center and was conducted according to the principles of the Declaration of Helsinki. All patients provided written informed consent. During the enrollment period of each registry, the EES was available in diameters of 2.25, 2.50, 2.75, 3.00, 3.50, and 4.00 mm, with each available in lengths of 8, 12, 15, 18, 23, and 28 mm. In contrast, the ZES was available in diameters of 2.25, 2.50, 2.75, 3.00, 3.50, and 4.00 mm and in lengths of 8 and 14 mm for stents with a diameter of 2.5 mm; 9, 15, and 38 mm for stents with a diameter of 3.00 mm; and 12, 14, 18, 24, and 30 mm for all available stent diameters. Coronary interventions were performed according to current standard techniques. The choice of the stent, predilatation, poststenting adjunctive balloon inflation, and the use of intravascular ultrasound or glycoprotein
IIb/IIIa inhibitors were left to the operators’ discretion. All patients received a loading dose of aspirin or were on longterm therapy before the procedure. A loading dose of 300 to 600 mg of clopidogrel was administered to all patients who were not taking clopidogrel before the procedure. After procedure, all patients were given aspirin (at least 100 mg/ day) indefinitely and clopidogrel (75 mg/day) for at least 6 months after the index procedure. The decision to obtain cardiac biomarkers and electrocardiograms after procedure, as well as maintenance of dual antiplatelet therapy (DAPT) beyond 6 months, was left to the treating physician. After the index PCI, follow-ups were performed at 1, 3, 9, 12 months and annually thereafter; angiography was optional at 9 months. Clinical follow-up data were obtained at outpatients visits or if not feasible, by telephone and/or medical questionnaire. At each follow-up session, regardless of outpatient visits, telephone, and/or medical questionnaire, detailed medication data including aspirin, clopidogrel, or other angina medications were collected. All relevant medical records were reviewed for any clinical event and adjudicated by an external clinical event committee. Using the Korean health system’s unique identification numbers, the vital status of all patients was crosschecked. Therefore, even in patients who were lost to follow-up, the occurrence of mortality was confirmed. All the medical records, followup, and events data were collected through electronic case report forms with strict security measures (T&W software, Seoul, Korea). The primary outcome (stent related) was target lesion failure (TLF), a composite of cardiac death, target vessel myocardial infarction (MI, not clearly attributed to a nonetarget vessel), or a clinically indicated target lesion revascularization (TLR) at 3 years. The major secondary outcome, the patient-oriented composite events (POCE), included all-cause mortality, any MI (including nonetarget vessel territory), and any revascularization. Other secondary outcomes included individual components of TLF and POCE, and stent thrombosis (ST) defined as definite, probable, or possible. All clinical outcomes, including ST, were defined according to the Academic Research Consortium, including the addendum to the definition of MI.9,10 The indication of PCI was considered “off label” if any of the following features were present: serum creatinine concentration 140 mmol/L (1.6 mg/dl); left ventricular ejection fraction <30%; an acute MI within the previous 72 hours; >1 lesion per vessel; 2 vessels treated with a stent; a lesion length 28 mm; or a bifurcated lesion, bypass graft, instent restenosis, unprotected left main coronary artery, presence of thrombus, or total occlusion, as with the previous randomized controlled trials.5,6 The statistical analysis was performed in 2 parts. First, analysis of primary and secondary clinical outcomes was
Coronary Artery Disease/EES Versus ZES-R in Real-World Patients
1331
Table 1 Baseline clinical characteristics of patients in crude population* Variable Age (years) Male BMI (kg/m2) Diabetes mellitus Hypertension Dyslipidemia Peripheral artery disease Chronic renal failure† Current smoker Previous percutaneous coronary intervention Previous coronary bypass Previous MI Previous heart failure Previous stroke Family history of CAD LV ejection fraction LVEF <30% Clinical indication for PCI Stable angina pectoris Unstable angina pectoris Acute myocardial infarction NSTEMI STEMI Silent ischemia No. of coronaries involved 1 2 3 No. of treated lesions/patient At least 1 ISR At least 1 bifurcation At least 1 thrombotic totalz At least 1 small vesselx At least 1 long lesion{ Multivessel PCI GP IIb/IIIa antagonist use At least one off-label indicationk Medication at discharge Aspirin Clopidogrel Statin ACE inhibitor Angiotensin-II receptor blocker Beta-blocker Calcium-channel blocker
Total (N ¼ 5054)
EES (N ¼ 3056)
ZES-R (N ¼ 1998)
p Value 0.897 0.389 0.333 0.068 0.195 <0.001 0.909 0.321 0.344 0.184 0.507 0.079 0.919 0.181 0.053 <0.001 0.283 <0.001 <0.001 0.881 <0.001 0.005 0.001 0.505 <0.001
63.9 10.8 3419/5054 24.9 9.32 1855/5029 3251/5025 3268/5017 80/4989 186/5017 1506/4971 757/5035 87/5039 326/5034 102/4992 395/4996 263/4898 58.8 11.4 75/4453
(5054) (67.6%) (4892) (36.9%) (64.7%) (65.1%) (1.6%) (3.7%) (30.3%) (15.0%) (1.7%) (6.5%) (2.0%) (7.9%) (5.4%) (4453) (1.7%)
63.9 10.8 2053/3056 25.0 11.8 1149/3031 1980/3027 1850/3019 47/2991 105/3019 893/2998 440/3041 56/3041 212/3036 62/2994 250/2998 171/2900 59.3 11.4 41/2714
(3056) (67.2%) (2935) (37.9%) (65.4%) (61.3%) (1.6%) (3.5%) (29.8%) (14.5%) (1.8%) (7.0%) (2.1%) (8.3%) (5.9%) (2714) (1.5%)
63.9 10.9 1366/1998 24.8 3.1 706/1998 1271/1998 1418/1998 33/1998 81/1998 613/1973 317/1998 31/1998 114/1998 40/1998 145/1998 92/1998 58.0 11.4 34/1739
(1998) (68.4%) (1957) (35.3%) (63.6%) (71.0%) (1.7%) (4.1%) (31.1%) (15.9%) (1.6%) (5.7%) (2.0%) (7.3%) (4.6%) (1739) (2.0%)
1696/5036 1856/5036 1330/5036 624/5036 706/5036 154/5036
(33.7%) (36.9%) (26.4%) (12.4%) (14.0%) (3.1%)
1095/3038 1117/3038 729/3038 344/3038 385/3038 97/3038
(36.0%) (36.8%) (24.0%) (11.3%) (12.7%) (3.2%)
601/1998 739/1998 601/1998 280/1998 321/1998 57/1998
(30.1%) (37.0%) (30.1%) (14.0%) (16.1%) (2.9%)
2207/5037 1597/5037 1233/5037 1.49 0.77 373/5054 832/5054 561/5054 1033/5054 2215/5054 1569/5054 133/4759 3830/5054
(43.8%) (31.7%) (24.5%) (5024) (7.4%) (16.5%) (11.1%) (20.4%) (43.8%) (31.0%) (2.8%) (75.8%)
1424/3046 923/3046 699/3046 1.47 0.74 231/3056 388/3056 293/3056 612/3056 1240/3056 930/3056 61/2763 2217/3056
(46.7%) (30.3%) (22.9%) (3038) (7.6%) (12.7%) (9.6%) (20.0%) (40.6%) (30.4%) (2.2%) (72.5%)
783/1991 674/1991 534/1991 1.53 0.80 142/1998 444/1998 268/1998 421/1998 975/1998 639/1998 72/1996 1613/1998
(39.3%) (33.9%) (26.8%) (1986) (7.1%) (22.2%) (13.4%) (21.1%) (48.8%) (32.0%) (3.6%) (80.7%)
0.009 0.548 <0.001 <0.001 0.368 <0.001 0.250 0.004 <0.001
4929/5018 4937/5017 4335/4998 1843/4966 1562/4939 3159/4970 1343/4931
(98.2%) (98.4%) (86.7%) (37.1%) (31.6%) (63.6%) (27.2%)
2969/3030 2974/3027 2613/3023 1113/3011 939/3016 1853/3009 830/3016
(98.0%) (98.2%) (86.4%) (37.0%) (31.1%) (61.6%) (27.5%)
1960/1988 1963/1990 1722/1975 730/1955 623/1923 1306/1961 513/1915
(98.6%) (98.6%) (87.2%) (37.3%) (32.4%) (66.6%) (26.8%)
0.126 0.301 0.468 0.810 0.363 <0.001 0.577
ACE ¼ angiotensin-converting enzyme; BMI ¼ body mass index; GP ¼ glycoprotein; ISR ¼ in-stent restenosis; LV ¼ left ventricle; LVEF ¼ left ventricular ejection fraction; MI ¼ myocardial infarction; NSTEMI ¼ noneST-segment elevation myocardial infarction; STEMI ¼ ST-segment elevation myocardial infarction. * Data are number (%), unless otherwise indicated. Plus-minus values are means SDs. † The chronic renal failure was defined as the presence of kidney damage or estimated glomerular filtration rate of <60 ml/min/1.73 m2 calculated by means of the Cockcroft-Gault formula for <3 months, as with the KDOQI guideline. z Thrombotic total occlusion denotes total occlusion (TIMI flow grade 0) in the angiography with visible thrombus. x Small vessel denotes lesion with reference diameter 2.75 mm. { Long lesion denotes lesion with length 28 mm. k Off-label indication: the indication of PCI was considered “off label” if any of the following features were present: serum creatinine concentration 140 mmol/L (1.6 mg/dl); left ventricular ejection fraction (LVEF) <30%; an acute MI within the previous 72 hours; >1 lesion per vessel; 2 vessels treated with a stent; a lesion length 28 mm; or a bifurcated lesion, bypass graft, in-stent restenosis, unprotected left main coronary artery, presence of thrombus, or total occlusion.
1332
The American Journal of Cardiology (www.ajconline.org)
Table 2 Baseline angiographic characteristics of lesions in crude population* Variable Target coronary artery Left main LAD LC Right ACC/AHA lesion class A B1 B2 C Type B2 or C lesions† In-stent restenosis Severe calcification Bifurcationz Bifurcation treatment Thrombus present Small vesselx Long lesion{ Max pressure deployment (atm) Mean stent diameter/lesion (mm) Total stent length (mm) Per patient Per lesion Number of stents Per patient Per lesion IVUS guided stenting Device success Lesion success Procedure success
Total (N ¼ 7084)
EES (N ¼ 4248)
ZES-R (N ¼ 2836)
258/7084 3179/7084 1567/7084 2071/7084
(3.6%) (44.9%) (22.1%) (29.2%)
178/4248 1907/4248 976/4248 1182/4248
(4.2%) (44.9%) (23.0%) (27.8%)
80/2836 1272/2836 591/2836 889/2836
(2.8%) (44.9%) (20.8%) (31.3%)
564/7084 1705/7084 1650/7084 2285/7084 3935/7084 424/7084 623/7084 919/7084 394/7084 633/7084 1200/7084 2671/7084 13.56 4.63 3.13 3.39
(8.0%) (24.1%) (23.3%) (32.3%) (55.5%) (6.0%) (8.8%) (13.0%) (5.6%) (8.9%) (16.9%) (37.7%) (6487) (7084)
247/4248 1064/4248 987/4248 1358/4248 2345/4248 257/4248 388/4248 419/4248 194/4248 336/4248 704/4248 1504/4248 13.45 4.79 3.16 4.31
(5.8%) (25.0%) (23.2%) (32.0%) (55.2%) (6.0%) (9.1%) (9.9%) (4.6%) (7.9%) (16.6%) (35.4%) (3790) (4248)
317/2836 641/2836 663/2836 927/2836 1590/2836 167/2836 235/2836 500/2836 200/2836 297/2836 496/2836 1167/2836 13.72 4.40 3.09 0.85
(11.2%) (22.6%) (23.4%) (32.7%) (56.1%) (5.9%) (8.3%) (17.6%) (7.1%) (10.5%) (17.5%) (41.1%) (2697) (2836)
38.97 26.01 (5054) 27.97 14.34 (7084) 1.67 0.97 1.19 0.49 2695/7084 6908/7084 6903/7084 6912/7084
(5054) (7084) (38.0%) (97.5%) (97.4%) (97.6%)
37.41 25.50 (3056) 26.90 14.06 (4248) 1.65 0.97 1.19 0.48 1601/4248 4147/4248 4145/4248 4140/4248
41.35 26.58 (1998) 29.61 14.61 (2836)
(3056) (4248) (37.7%) (98.2%) (98.1%) (98.1%)
1.70 0.98 1.19 0.51 1094/2836 2761/2836 2758/2836 2772/2836
(1998) (2836) (38.6%) (98.5%) (98.5%) (98.5%)
p Value 0.001 0.003 0.981 0.035 0.002 <0.001
0.479 0.798 0.231 <0.001 <0.001 <0.001 0.316 <0.001 0.024 0.363 <0.001 <0.001 0.091 0.467 0.454 0.484 0.399 0.479
IVUS ¼ intravascular ultrasound; LAD ¼ left anterior descending artery; LC ¼ left circumflex artery. * Data are number (%), unless otherwise indicated. Plus-minus values are means SDs. † Type B2 or C lesions according to American College of Cardiology/American Heart Association classification. z Bifurcation means bifurcated lesion that has been treated solely by drug-eluting stents. x Small vessel denotes lesion with reference diameter 2.75 mm. { Long lesion denotes lesion with length 28 mm.
performed in the crude population. Second, a propensity scoreematched population was selected to adjust for uneven distribution of baseline characteristics. Kaplan-Meier analysis was used to calculate cumulative incidence of primary and secondary clinical outcomes, and the log-rank test was used to compare between-group differences. A logistic regression model was conducted to generate propensity score, which was the probability that a patient received a ZES-R. The adjusted covariates that were used to calculate the propensity score are listed in the Supplementary Methods. For matching, nearest neighbor matching with a caliper width of 0.6 SD was used because this value has been shown to eliminate >90% of the bias in the observed confounders.11 The discrimination and calibration abilities of propensity score model were assessed by C statistics (0.637) and the Hosmer-Lemeshow statistics (p ¼ 0.515). To reduce possible confounding factors from the differences in baseline characteristics, multivariateadjusted stratified Cox proportional hazard regression to find independent predictors of TLF and exploratory subgroup analysis was performed in propensity scoreematched cohorts. All probability values were 2-sided, and p values <0.05 were considered statistically significant.
Results A total of 5,054 patients (7,084 lesions) were enrolled (EES group: 3,056 patients/4,248 lesions and ZES-R group: 1,998 patients/2,836 lesions). The flow of the study patients is shown in Supplementary Figure 1. Through 3 years, 110 (3.5%) and 120 patients (6.0%) were lost to follow-up in the EES and ZES-R groups, respectively. However, all were confirmed to be alive. The distribution of cardiac risk factors was similar, except for dyslipidemia, lesion complexity, and left main disease (Tables 1 and 2). High-risk patients and lesions were frequent, implying that our registries were an enriched PCI population, reflecting real-world practice in Korea. The device, lesion, and procedure success rates were excellent and similar for both stents (Table 2). At 3 years, the incidence of the primary clinical outcome, TLF, was 6.2% for EES and 6.4% for ZES-R, which was not significantly different (p ¼ 0.812). The rate of the individual components of TLF (cardiac death, TLR, and target vessel MI) was not statistically different between the 2 groups. POCE were also similar, as were the individual components (all-cause death, any revascularization, and any MI). About 1/2 of the target vesselerelated MI was due to ST (Table 3).
Coronary Artery Disease/EES Versus ZES-R in Real-World Patients
1333
Table 3 Clinical outcomes in crude population at 3-year follow-up* Total (N ¼ 5054) All cause death Cardiac death Any myocardial infarction Target vessel Non target vessel MI due to ST Any revascularization Clinically driven revascularization Target lesion revascularization Target vessel revascularization Cerebrovascular accident Target lesion failure† Target vessel failurez Patient-oriented composite eventsx
292 188 37 25 12 12 386 296 123 176 73 361 369 670
(5.8%) (3.7%) (0.7%) (0.5%) (0.2%) (0.2%) (7.6%) (5.9%) (2.4%) (3.5%) (1.4%) (6.3%) (7.3%) (13.3%)
EES (N ¼ 3056) 181 109 25 18 7 9 243 190 76 105 52 189 218 420
ZES-R (N ¼ 1998)
(5.9%) (3.6%) (0.8%) (0.6%) (0.2%) (0.3%) (8.0%) (6.2%) (2.5%) (3.4%) (1.7%) (6.2%) (7.1%) (13.7%)
111 79 12 7 5 3 143 106 47 71 21 127 151 250
(5.6%) (4.0%) (0.6%) (0.4%) (0.3%) (0.2%) (7.2%) (5.3%) (2.4%) (3.6%) (1.1%) (6.4%) (7.6%) (12.5%)
RR (95% CI) 0.94 1.11 0.73 0.59 1.09 0.51 0.90 0.85 0.95 1.03 0.62 1.03 1.06 0.91
(0.75e1.18) (0.83e1.47) (0.39e1.46) (0.25e1.42) (0.35e3.44) (0.14e1.88) (0.74e1.10) (0.68e1.07) (0.66e1.35) (0.77e1.39) (0.37e1.02) (0.83e1.28) (0.87e1.29) (0.79e1.05)
p Value 0.622 0.494 0.404 0.306 0.880 0.385 0.304 0.178 0.780 0.875 0.070 0.812 0.581 0.219
CI ¼ confidence interval; MACE ¼ major adverse cardiovascular events; MI ¼ myocardial infarction; RR ¼ relative risk; ST ¼ stent thrombosis. * Data are number (%), unless otherwise indicated. † Target lesion failure defined as a composite of cardiac death, myocardial infarction (not clearly attributed to a nontarget vessel), or clinically indicated target lesion revascularization by percutaneous or surgical methods at 1 year. z Target vessel failure defined as a composite of cardiac death, myocardial infarction (not clearly attributed to a nontarget vessel), or clinically indicated target vessel revascularization by percutaneous or surgical methods at 1 year. x Patient-oriented composite outcomes included all-cause mortality, any myocardial infarction (includes nontarget vessel territory), and any revascularization (includes all target and nontarget vessel, regardless of percutaneous or surgical methods).
In survival analysis, there were no differences between the 2 groups regarding cumulative incidence of TLF (hazard ratio [HR] 1.03, 95% confidence interval [CI] 0.82 to 1.29, log-rank p ¼ 0.797) or POCE (HR 0.91, 95% CI 0.78 to 1.06, log-rank p ¼ 0.237; Figure 1) as well as the individual components of TLF or POCE (Supplementary Figure 2). The rates of Academic Research Consortiumedefined definite ST and definite or probable ST at 3-year follow-up were low and comparable between the 2 stents (HR 0.70, 95% CI 0.33 to 1.47, p ¼ 0.344, log-rank p ¼ 0.341). All 33 patients with ST had at least 1 off-label indication. Among these, 20 events (60.6%) initially presented as sudden cardiac death. All but 4 patients were on DAPT at the time of the event, and the main reason for discontinuation was bleeding. Definite or probable ST after 1 year occurred in 4 patients (0.1%) without between-group differences (Table 4 and Figure 2). Two of the 4 very late ST events were associated with mortality. A detailed description of all ST cases is presented in Supplementary Table 1. About 1/2 of the patients were on DAPT up to 3-year follow-up, and the mean duration of DAPT was not different between the 2 groups (Table 4). Matching by propensity score yielded 1,698 pairs of patients in both stent groups. Baseline clinical and angiographic characteristics of the matched groups are listed in Supplementary Table 2 and were more balanced than the crude population. The standard differences of the adjusted variables were <10% (Supplementary Figure 3). The cumulative incidences of TLF and POCE up to 3 years and its individual components were comparable between the 2 groups (Figure 1 and Supplementary Table 3). Definite or probable ST was also similar (Supplementary Table 4). In univariate analysis, several underlying co-morbidities and lesion characteristics were significantly different
between those with and without events (Supplementary Table 5). In multivariate analysis, chronic renal failure (adjusted HR 3.615, 95% CI 2.440 to 5.354, p <0.001) and off-label indication (adjusted HR 1.782, 95% CI 1.169 to 2.718, p ¼ 0.007) were the strongest predictors of TLF at 3 years. In addition, other significant predictors of TLF included diabetes mellitus, history of cerebrovascular accident, and increasing age (Table 5). The Harrell’s C-index of the multivariate model was 0.686. Exploratory subgroup analysis at 3-year follow-up with regard to TLF showed consistent results across different subgroups without significant differences in clinical outcomes between EES and ZES-R. The interaction p value was nonsignificant across all the subgroups (Figure 3). Discussion At 3-year follow-up of this large cohort of enriched patients undergoing PCI who received EES or ZES-R, both groups showed excellent and comparable safety and efficacy results. These results extend previous 2-year findings from the RESOLUTE ALL Comers and TWENTE trials. In both crude and propensity scoreematched analysis, 3-year rates of stent- and patient-related composite events were comparable along with its individual components. The absence of a difference in TLF at 3-year follow-up was consistent across various subgroups. The rates of definite or probable ST were very low in both stents especially after 1 year considering the complexity of the lesions treated and were balanced between the 2 stents, suggesting excellent safety. Finally, off-label use of the stent and important co-morbidities, such as chronic renal failure and diabetes mellitus, were significantly associated with TLF at 3 years, rather than the stent type.
1334
The American Journal of Cardiology (www.ajconline.org)
Figure 1. Survival analysis: primary and major secondary outcomes. Kaplan-Meier curves are shown for each outcome and cohort combination (A to D).
The 2 previous randomized controlled trials that compared these DESs reported 2-year TLF rates of 10.7% for EESs and 11.2% for ZES-R (p ¼ 0.74) and 9.8% for EESs and 10.5% for ZES-R (p ¼ 0.68). In the present study, the 3-year TLF rate was relatively lower (6.2%/ 6.4%, p ¼ 0.812), mainly due to the relatively lower rates of target vesselerelated MI (RESOLUTE ALL Comers 4.6%, TWENTE 5.5% at 2 years, EXCELLENT and RESOLUTE-Korea 0.5% at 3 years). The main reason for relatively lower rates of target vessel MI might be nonsystematic collection of cardiac biomarkers after index procedures in each participating centers because it was not mandated and is not a routine practice in Korea. In
addition, TWENTE 2-year results reported significant lower rates of TLR in the EES group than the ZES-R group (4.9% vs 2.6%, p ¼ 0.03); however, 3-year rates of TLR were comparable in the present study (2.5% vs 2.4%, p ¼ 0.780). Both second-generation DESs enhanced biocompatibility of the polymers and incorporated a thinner cobalt-chromium stent platform to enhance endothelial coverage of stent struts and reduce risk of ST. Our results support the previous 2-year results from the RESOLUTE ALL Comers and TWENTE trials and further show that the enhanced safety and efficacy profiles of these stents are sustained up to 3-year follow-up. Notably, ST only occurred after off-label use, and >1/2 of the events were
Coronary Artery Disease/EES Versus ZES-R in Real-World Patients
1335
Table 4 Stent thrombosis and use of dual antiplatelet agents in crude population at 3-year follow-up* Total (N ¼ 5054) Definite Acute (0e1 day) Subacute (2e30 days) Late (31e360 days) Very late (361 days) Probable Acute (0e1 day) Subacute (2e30 days) Late (31e360 days) Very late (361 days) Possible Late (31e360 days) Stent thrombosis Definite Definite or probable Definite, probable, or possible Very late definite or probable Clinical presentation at time of ST Sudden cardiac death ST-segment elevation MI NSTEMI/unstable angina Use of dual anti-platelet agent DAPT at 1-year No DAPT user at 1-year Aspirin single Clopidogrel single None of DAPT DAPT at 2-year No DAPT user at 2-year Aspirin single Clopidogrel single None of DAPT DAPT at 3-year No DAPT user at 3-year Aspirin single Clopidogrel single None of DAPT Mean duration of DAPT
EES (N ¼ 3056)
12 4 3 2 3 20 4 14 1 1 1 1
(0.2%) (0.1%) (0.1%) (<0.1%) (0.1%) (0.4%) (0.1%) (0.3%) (<0.1%) (<0.1%) (<0.1%) (<0.1%)
8 3 2 1 2 14 1 11 1 1 1 1
(0.3%) (0.1%) (0.1%) (<0.1%) (0.1%) (0.5%) (<0.1%) (0.4%) (<0.1%) (<0.1%) (<0.1%) (<0.1%)
12 32 33 4
(0.2%) (0.6%) (0.7%) (0.1%)
8 22 23 3
(0.3%) (0.7%) (0.8%) (0.1%)
20/33 (60.6%) 9/33 (27.3%) 4/33 (12.1%) 3762/4684 922/4684 640/922 173/922 109/922 2259/4359 2100/4359 1517/2100 427/2100 156/2100 1807/4027 2220/4027 1550/2220 510/2220 160/2220 737.4 332.8
(80.3%) (19.7%) (69.4%) (18.8%) (11.8%) (51.8%) (48.2%) (72.2%) (20.3%) (7.4%) (44.9%) (55.1%) (69.8%) (23.0%) (7.2%) (4586)
14/23 (60.9%) 7/23 (30.4%) 2/23 (8.7%) 2312/2856 544/2856 355/544 116/544 73/544 1383/2650 1267/2650 893/1267 270/1267 104/1267 1110/2479 1369/2479 932/1369 324/1369 113/1369 743.9 331.9
(81.0%) (19.0%) (65.3%) (21.3%) (13.4%) (52.2%) (47.8%) (70.5%) (21.3%) (8.2%) (44.8%) (55.2%) (68.1%) (23.7%) (8.3%) (2793)
ZES-R (N ¼ 1998)
p Value
4 1 1 1 1 6 3 3 0 0 0 0
(0.2%) (0.1%) (0.1%) (0.1%) (0.1%) (0.3%) (0.2%) (0.2%) (0%) (0%) (0%) (0%)
0.774 0.657 1.000 1.000 1.000 0.494 0.307 0.185 1.000 1.000 1.000 1.000
4 10 10 1
(0.2%) (0.5%) (0.5%) (0.1%)
0.774 0.370 0.292 0.657 0.739
6/10 (60.0%) 2/10 (20.0%) 2/10 (20.0%) 1450/1828 378/1828 285/378 57/378 36/378 876/1709 833/1709 624/833 157/833 52/833 697/1548 851/1548 618/851 186/851 47/851 727.3 334.0
(79.3%) (20.7%) (75.4%) (15.1%) (9.5%) (51.3%) (48.7%) (74.9%) (18.8%) (6.2%) (45.0%) (55.0%) (72.6%) (21.9%) (5.5%) (1793)
0.175
0.555
0.896
0.100
DAPT ¼ dual antiplatelet agent therapy; MI ¼ myocardial infarction; NSTEMI ¼ noneST-segment elevation myocardial infarction; ST ¼ stent thrombosis. * Data are number (%), unless otherwise indicated.
fatal. Even among very late ST cases, 2 of 4 eventually led to mortality. Interestingly, 87.9% of the all STs, including 3 of 4 very late definite or probable STs, occurred while on DAPT. The TWENTE trial showed that the rate of very late ST was still low (0.3% in each group) despite a low rate of continued DAPT usage beyond 12 months, which was in accordance with the current guidelines.12 In the present study, the rate of DAPT use at 2-year follow-up was 48.2%, which was relatively higher than that in other 2 trials (RESOLUTE ALL Comers trial: 18%, TWENTE trial: 6.9%). Although it might be difficult to directly compare the rates of clinical outcomes from different trials because of different patient populations; nevertheless, the overall incidence of very late ST was <1% across all 3 studies and was lower than that reported for the first-generation DESs, which showed continuous risk of very late ST.13 With the use of second-generation DESs, whether prolonged DAPT can reduce the
possibility of very late ST is still unclear and need to be clarified in future trials. In multivariate analysis, chronic renal failure was the strongest predictor of TLF, concordant with previous literature.14 Other independent predictors included off-label DES use and diabetes mellitus. ZES-R showed significantly lower incidence of target vessel failure at 1 year (7.8%) than predefined DES performance goal of 14.5% in patients with diabetes mellitus and achieved the first Food and Drug Administration approval for PCI in patients with diabetes.15 However, it should be noticed that the pooled analysis of the international global RESOLUTE program included only the patients with on-label indication of the stent; the high-risk patients with off-label indication in the Resolute All Comers trial and the Resolute International registry were excluded from the pooled analysis.15 Despite previously reported safety and efficacy of the secondgeneration stents in diabetes population, our study
1336
The American Journal of Cardiology (www.ajconline.org)
Figure 2. Survival analysis: definite or probable ST. Arrow indicates the patients not on DAPT at the time of ST.
Table 5 Independent predictors of target lesion failure at 3-year follow-up in propensity scoreematched group*
Chronic renal failure Off label indication Diabetes mellitus Previous history of CVA Age
Hazard Ratio
95% CI
p Value
3.615 1.782 1.719 1.599 1.035
2.440e5.354 1.169e2.718 1.304e2.266 1.078e2.372 1.021e1.049
<0.001 0.007 <0.001 0.020 <0.001
CI ¼ confidence interval; CVA ¼ cerebrovascular accident; LVEF ¼ left ventricular ejection fraction; NSTEMI ¼ noneST-segment elevation myocardial infarction; STEMI ¼ ST-segment elevation myocardial infarction. * Identification of independent predictors was done with stratified Cox proportional hazard regression model, and the variables are presented with multivariate-adjusted hazard ratios, 95% confidence intervals, and p values. Included variables to the final model are presented in Supplementary Table 5. The individual components of off-label indication (i.e., STEMI, NSTEMI, in-stent restenosis, bifurcation, thrombotic total occlusion, long lesion, multivessel PCI, severe left ventricular dysfunction [LVEF <30%], and left main procedure) were not included individually to the final model because of significant correlation with off-label indication itself (i.e., colinearity between these covariates). The Harrell’s c-index of the model was 0.686.
confirmed that it is still an important risk factor in the treatment of coronary artery disease.16 Along with the excellent safety and efficacy profiles of the secondgeneration DES up to 3-year follow-up, the results of the multivariate analysis again show the importance of risk factor control, which is not directly related to the stent itself, when considering the prognosis of individual patients.
The present study has several limitations. First, inherent limitations of nonrandomized comparisons such as allocation bias and uneven distribution of risk factors should be considered. Although rigorous risk adjustment model including propensity score matching and multiple imputation of missing values was performed, unmeasured variables could not be controlled. Second, relatively lower incidence of clinical events, mainly driven by lower incidence of target vessel MI compared with previous randomized controlled trials and pooled analysis of EES and ZES-R,5,6,15,17 should be acknowledged. Because postprocedural collection of cardiac biomarkers to detect periprocedural MI was not routinely undertaken in all the participating centers, this might be the reason for the lower incidence of MI in our registries. Therefore, we believe the true rate of MI would have been much higher if the cardiac enzymes were followed systematically. Third, although vital status of all patients including patients lost to follow-up was confirmed with the Korean national database using a citizen registration number that is unique to each individual, we cannot exclude the possibility of under-reporting of clinical outcomes other than death such as nonfatal MI, TLR, or most importantly, ST that did present as sudden cardiac death, in the patients who were lost to follow-up. Disclosures Dr. H.S. Kim received research grants, lecture fees, and honoraria from Medtronic Cardiovascular, Minneapolis, Minnesota, Abbott Vascular, Santa Clara, California, and Boston Scientific, Natick, Massachusetts. The other authors have no conflicts of interest to disclose.
Coronary Artery Disease/EES Versus ZES-R in Real-World Patients
1337
Figure 3. Subgroup analysis for TLF in propensity scoreematched cohort. Exploratory subgroup analysis at 3-year follow-up with regard to TLF is shown in the figure. AMI ¼ acute myocardial infarction; STEMI ¼ ST-segment elevation myocardial infarction.
Supplementary Data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. amjcard.2014.07.065. 1. Stone GW, Rizvi A, Sudhir K, Newman W, Applegate RJ, Cannon LA, Maddux JT, Cutlip DE, Simonton CA, Sood P, Kereiakes DJ, Investigators SI. Randomized comparison of everolimus- and paclitaxeleluting stents. 2-year follow-up from the SPIRIT (Clinical Evaluation of the XIENCE V Everolimus Eluting Coronary Stent System) IV trial. J Am Coll Cardiol 2011;58:19e25. 2. Byrne RA, Kastrati A, Massberg S, Wieczorek A, Laugwitz KL, Hadamitzky M, Schulz S, Pache J, Fusaro M, Hausleiter J, Schomig A, Mehilli J; Investigators I-T. Biodegradable polymer versus permanent polymer drug-eluting stents and everolimus- versus sirolimus-eluting stents in patients with coronary artery disease: 3year outcomes from a randomized clinical trial. J Am Coll Cardiol 2011;58:1325e1331. 3. Jensen LO, Thayssen P, Christiansen EH, Tilsted HH, Maeng M, Hansen KN, Kaltoft A, Hansen HS, Botker HE, Krusell LR, Ravkilde J, Madsen M, Thuesen L, Lassen JF; Investigators SOI. 2-Year patientrelated versus stent-related outcomes: the SORT OUT IV (Scandinavian Organization for Randomized Trials With Clinical Outcome IV) Trial. J Am Coll Cardiol 2012;60:1140e1147.
4. Smits PC, Kedhi E, Royaards KJ, Joesoef KS, Wassing J, RademakerHavinga TA, McFadden E. 2-year follow-up of a randomized controlled trial of everolimus- and paclitaxel-eluting stents for coronary revascularization in daily practice. COMPARE (Comparison of the everolimus eluting XIENCE-V stent with the paclitaxel eluting TAXUS LIBERTE stent in all-comers: a randomized open label trial). J Am Coll Cardiol 2011;58:11e18. 5. Silber S, Windecker S, Vranckx P, Serruys PW. Unrestricted randomised use of two new generation drug-eluting coronary stents: 2-year patient-related versus stent-related outcomes from the RESOLUTE All Comers trial. Lancet 2011;377:1241e1247. 6. Tandjung K, Sen H, Lam MK, Basalus MW, Louwerenburg JH, Stoel MG, van Houwelingen KG, de Man FH, Linssen GC, Said SA, Nienhuis MB, Lowik MM, Verhorst PM, van der Palen J, von Birgelen C. Clinical outcome following stringent discontinuation of dual antiplatelet therapy after 12 months in real-world patients treated with second-generation zotarolimus-eluting Resolute and everolimus-eluting Xience V stents: 2-year follow-up of the randomized TWENTE trial. J Am Coll Cardiol 2013;61:2406e2416. 7. Taniwaki M, Stefanini GG, Silber S, Richardt G, Vranckx P, Serruys PW, Buszman PE, Kelbaek H, Windecker S. Four year clinical outcomes and predictors of repeat revascularization in patients treated with new generation drug-eluting stents in the RESOLUTE all comers randomized trial. J Am Coll Cardiol 2014;63:1617e1625. 8. Park KW, Lee JM, Kang SH, Ahn HS, Yang HM, Lee HY, Kang HJ, Koo BK, Cho J, Gwon HC, Lee SY, Chae IH, Youn TJ, Chae JK,
1338
9.
10.
11. 12.
13.
The American Journal of Cardiology (www.ajconline.org)
Han KR, Yu CW, Kim HS. Safety and efficacy of second-generation everolimus-eluting Xience V stents versus zotarolimus-eluting resolute stents in real-world practice: patient-related and stent-related outcomes from the multicenter prospective EXCELLENT and RESOLUTE-Korea registries. J Am Coll Cardiol 2013;61:536e544. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A, Hamon M, Krucoff MW, Serruys PW. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation 2007;115:2344e2351. Vranckx P, Cutlip DE, Mehran R, Kint PP, Silber S, Windecker S, Serruys PW. Myocardial infarction adjudication in contemporary allcomer stent trials: balancing sensitivity and specificity. Addendum to the historical MI definitions used in stent studies. EuroIntervention 2010;5:871e874. Gu XS, Rosenbaum PR. Comparison of multivariate matching methods: structures, distances, and algorithms. J Comput Graphical Stat 1993;2:405e420. Levine GN, Bates ER, Blankenship JC, Bailey SR, Bittl JA, Cercek B, Chambers CE, Ellis SG, Guyton RA, Hollenberg SM, Khot UN, Lange RA, Mauri L, Mehran R, Moussa ID, Mukherjee D, Nallamothu BK, Ting HH; American College of Cardiology F; American Heart Association Task Force on Practice G; Society for Cardiovascular A, Interventions. 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Society for Cardiovascular Angiography and Interventions. J Am Coll Cardiol 2011;58:e44ee122. Daemen J, Wenaweser P, Tsuchida K, Abrecht L, Vaina S, Morger C, Kukreja N, Juni P, Sianos G, Hellige G, van Domburg RT, Hess OM,
14.
15.
16.
17.
Boersma E, Meier B, Windecker S, Serruys PW. Early and late coronary stent thrombosis of sirolimus-eluting and paclitaxel-eluting stents in routine clinical practice: data from a large two-institutional cohort study. Lancet 2007;369:667e678. Saltzman AJ, Stone GW, Claessen BE, Narula A, Leon-Reyes S, Weisz G, Brodie B, Witzenbichler B, Guagliumi G, Kornowski R, Dudek D, Metzger DC, Lansky AJ, Nikolsky E, Dangas GD, Mehran R. Long-term impact of chronic kidney disease in patients with ST-segment elevation myocardial infarction treated with primary percutaneous coronary intervention: the HORIZONS-AMI (Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction) trial. JACC Cardiovasc Interv 2011;4: 1011e1019. Silber S, Serruys PW, Leon MB, Meredith IT, Windecker S, Neumann FJ, Belardi J, Widimsky P, Massaro J, Novack V, Yeung AC, Saito S, Mauri L. Clinical outcome of patients with and without diabetes mellitus after percutaneous coronary intervention with the resolute zotarolimus-eluting stent: 2-year results from the prospectively pooled analysis of the international global RESOLUTE program. JACC Cardiovasc Interv 2013;6:357e368. Park KW, Lee JM, Kang SH, Ahn HS, Kang HJ, Koo BK, Rhew JY, Hwang SH, Lee SY, Kang TS, Kwak CH, Hong BK, Yu CW, Seong IW, Ahn T, Lee HC, Lim SW, Kim HS. Everolimus-eluting xience v/ promus versus zotarolimus-eluting resolute stents in patients with diabetes mellitus. JACC Cardiovasc Interv 2014;7:471e481. Stone GW, Kedhi E, Kereiakes DJ, Parise H, Fahy M, Serruys PW, Smits PC. Differential clinical responses to everolimus-eluting and Paclitaxel-eluting coronary stents in patients with and without diabetes mellitus. Circulation 2011;124:893e900.
a Department of Internal Medicine and Cardiovascular Center, Seoul National University Hospital, Seoul, Republic of Korea; bChungbuk National University, Cheongju, Republic of Korea; cInha University Hospital, Incheon, Republic of Korea; dSejong General Hospital, Sejong Heart Institute, Bucheon, Republic of Korea; eSoonchunhyang University Cheonan Hospital, Cheonan, Republic of Korea; fNational Health Insurance Corporation Ilsan Hospital, Goyang, Republic of Korea; gWonkwang University School of Medicine, Iksan, Republic of Korea; hYeungnam University Medical Center, Daegu, Republic of Korea; iInje University, Haeundae Paik Hospital, Busan, Republic of Korea; jSoonchunhyang University Hospital, Seoul, Republic of Korea; kUijeongbu St Mary’s Hospital, Uijeongbu, Republic of Korea; lKorea University Anam Hospital, Seoul, Republic of Korea; mKyunghee University College of Medicine, Seoul, Republic of Korea; nKorea University Guro, Hospital, Seoul, Republic of Korea; oDaejeon St. Mary’s Hospital, Daejeon, Republic of Korea; pGyeongsang National University Hospital, Jinju,
0002-9149/14/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2014.07.065
Republic of Korea; qBoramae Medical Center, Seoul, Republic of Korea; Kangbuk Samsung Hospital, Seoul, Republic of Korea; sSt. Vincent’s Hospital, Suwon, Republic of Korea; and tYonsei University College of Medicine, Seoul, Republic of Korea. Manuscript received July 6, 2014; revised manuscript received and accepted July 30, 2014. Dr. Lee and Dr. K.W. Park contributed equally to this study. This study was supported by grant A102065 from the Korea Healthcare technology R&D Project, Ministry of Health and Welfare, Republic of Korea, and grant A062260 from the Innovative Research Institute for Cell Therapy, Seoul National University Hospital, Seoul, Korea, sponsored by the Ministry of Health, Welfare & Family, Republic of Korea. See page 1336 for disclosure information. *Corresponding author: Tel: (þ82) 2-2072-2226; fax: (þ82) 2-7668904. E-mail address: [email protected] (H.-S. Kim). r
www.ajconline.org
1330
The American Journal of Cardiology (www.ajconline.org)
were the strongest predictors of target lesion failure at 3 years. In conclusion, both stents showed comparable safety and efficacy at 3-year follow-up in this robust real-world registry with unrestricted use of EES and ZES-R. Overall incidences of target lesion failure and definite stent thrombosis, including very late stent thrombosis, were low, even in the patients with off-label indications, suggesting excellent long-term safety and sustained efficacy of both types of second-generation drug-eluting stents. Ó 2014 Elsevier Inc. All rights reserved. (Am J Cardiol 2014;114:1329e1338) Second-generation drug-eluting stents (DESs), developed with the purpose of improving long-term safety, have almost completely replaced the first-generation DESs in the treatment of coronary artery disease. This has mostly been due to the representative second-generation DES, Xience V/Promus everolimus-eluting stent (EES; Abbott Vascular, Santa Clara, California), which showed sustained safety and efficacy up to 3 years in comparisons with first-generation DESs.1e4 Another second-generation DES, the Resolute zotarolimus-eluting stent (ZES-R; Medtronic Cardiovascular, Minneapolis, Minnesota) showed comparable clinical outcomes up to 2-year follow-up, compared with EES,5,6 and is also used often in daily routine practice. Although 4-year comparison of these stents was recently published,7 long-term clinical outcomes of these stents were limited only to this report despite its wide use in routine clinical practice, especially regarding the ZES-R. Thus, more data about their everyday use are needed. Previously, we published the 1-year comparison of EES versus ZES-R in 5,054 patients from 2 multicenter all-comers registries (the EXCELLENT and RESOLUTE-Korea registries).8 The purpose of this study was to evaluate the long-term safety and efficacy of the Xience V/Promus EES and Resolute ZES up to 3-year follow-up in everyday real-world use from large prospective registries of these DESs. Methods Extended description of study methods including statistical analysis is presented in Supplementary Data. This study evaluated 3-year clinical outcomes of Xience V/Promus EES and Resolute ZES from 2 prospective multicenter registries: the EXCELLENT and RESOLUTE-Korea, which enrolled all-comers treated with 1 EESs or ZES-R (3,056/ 29 and 1,998/25 patients/participating centers, respectively) without exclusions (Supplementary Figure 1).8 The study protocol was approved by the ethics committee at each participating center and was conducted according to the principles of the Declaration of Helsinki. All patients provided written informed consent. During the enrollment period of each registry, the EES was available in diameters of 2.25, 2.50, 2.75, 3.00, 3.50, and 4.00 mm, with each available in lengths of 8, 12, 15, 18, 23, and 28 mm. In contrast, the ZES was available in diameters of 2.25, 2.50, 2.75, 3.00, 3.50, and 4.00 mm and in lengths of 8 and 14 mm for stents with a diameter of 2.5 mm; 9, 15, and 38 mm for stents with a diameter of 3.00 mm; and 12, 14, 18, 24, and 30 mm for all available stent diameters. Coronary interventions were performed according to current standard techniques. The choice of the stent, predilatation, poststenting adjunctive balloon inflation, and the use of intravascular ultrasound or glycoprotein
IIb/IIIa inhibitors were left to the operators’ discretion. All patients received a loading dose of aspirin or were on longterm therapy before the procedure. A loading dose of 300 to 600 mg of clopidogrel was administered to all patients who were not taking clopidogrel before the procedure. After procedure, all patients were given aspirin (at least 100 mg/ day) indefinitely and clopidogrel (75 mg/day) for at least 6 months after the index procedure. The decision to obtain cardiac biomarkers and electrocardiograms after procedure, as well as maintenance of dual antiplatelet therapy (DAPT) beyond 6 months, was left to the treating physician. After the index PCI, follow-ups were performed at 1, 3, 9, 12 months and annually thereafter; angiography was optional at 9 months. Clinical follow-up data were obtained at outpatients visits or if not feasible, by telephone and/or medical questionnaire. At each follow-up session, regardless of outpatient visits, telephone, and/or medical questionnaire, detailed medication data including aspirin, clopidogrel, or other angina medications were collected. All relevant medical records were reviewed for any clinical event and adjudicated by an external clinical event committee. Using the Korean health system’s unique identification numbers, the vital status of all patients was crosschecked. Therefore, even in patients who were lost to follow-up, the occurrence of mortality was confirmed. All the medical records, followup, and events data were collected through electronic case report forms with strict security measures (T&W software, Seoul, Korea). The primary outcome (stent related) was target lesion failure (TLF), a composite of cardiac death, target vessel myocardial infarction (MI, not clearly attributed to a nonetarget vessel), or a clinically indicated target lesion revascularization (TLR) at 3 years. The major secondary outcome, the patient-oriented composite events (POCE), included all-cause mortality, any MI (including nonetarget vessel territory), and any revascularization. Other secondary outcomes included individual components of TLF and POCE, and stent thrombosis (ST) defined as definite, probable, or possible. All clinical outcomes, including ST, were defined according to the Academic Research Consortium, including the addendum to the definition of MI.9,10 The indication of PCI was considered “off label” if any of the following features were present: serum creatinine concentration 140 mmol/L (1.6 mg/dl); left ventricular ejection fraction <30%; an acute MI within the previous 72 hours; >1 lesion per vessel; 2 vessels treated with a stent; a lesion length 28 mm; or a bifurcated lesion, bypass graft, instent restenosis, unprotected left main coronary artery, presence of thrombus, or total occlusion, as with the previous randomized controlled trials.5,6 The statistical analysis was performed in 2 parts. First, analysis of primary and secondary clinical outcomes was
Coronary Artery Disease/EES Versus ZES-R in Real-World Patients
1331
Table 1 Baseline clinical characteristics of patients in crude population* Variable Age (years) Male BMI (kg/m2) Diabetes mellitus Hypertension Dyslipidemia Peripheral artery disease Chronic renal failure† Current smoker Previous percutaneous coronary intervention Previous coronary bypass Previous MI Previous heart failure Previous stroke Family history of CAD LV ejection fraction LVEF <30% Clinical indication for PCI Stable angina pectoris Unstable angina pectoris Acute myocardial infarction NSTEMI STEMI Silent ischemia No. of coronaries involved 1 2 3 No. of treated lesions/patient At least 1 ISR At least 1 bifurcation At least 1 thrombotic totalz At least 1 small vesselx At least 1 long lesion{ Multivessel PCI GP IIb/IIIa antagonist use At least one off-label indicationk Medication at discharge Aspirin Clopidogrel Statin ACE inhibitor Angiotensin-II receptor blocker Beta-blocker Calcium-channel blocker
Total (N ¼ 5054)
EES (N ¼ 3056)
ZES-R (N ¼ 1998)
p Value 0.897 0.389 0.333 0.068 0.195 <0.001 0.909 0.321 0.344 0.184 0.507 0.079 0.919 0.181 0.053 <0.001 0.283 <0.001 <0.001 0.881 <0.001 0.005 0.001 0.505 <0.001
63.9 10.8 3419/5054 24.9 9.32 1855/5029 3251/5025 3268/5017 80/4989 186/5017 1506/4971 757/5035 87/5039 326/5034 102/4992 395/4996 263/4898 58.8 11.4 75/4453
(5054) (67.6%) (4892) (36.9%) (64.7%) (65.1%) (1.6%) (3.7%) (30.3%) (15.0%) (1.7%) (6.5%) (2.0%) (7.9%) (5.4%) (4453) (1.7%)
63.9 10.8 2053/3056 25.0 11.8 1149/3031 1980/3027 1850/3019 47/2991 105/3019 893/2998 440/3041 56/3041 212/3036 62/2994 250/2998 171/2900 59.3 11.4 41/2714
(3056) (67.2%) (2935) (37.9%) (65.4%) (61.3%) (1.6%) (3.5%) (29.8%) (14.5%) (1.8%) (7.0%) (2.1%) (8.3%) (5.9%) (2714) (1.5%)
63.9 10.9 1366/1998 24.8 3.1 706/1998 1271/1998 1418/1998 33/1998 81/1998 613/1973 317/1998 31/1998 114/1998 40/1998 145/1998 92/1998 58.0 11.4 34/1739
(1998) (68.4%) (1957) (35.3%) (63.6%) (71.0%) (1.7%) (4.1%) (31.1%) (15.9%) (1.6%) (5.7%) (2.0%) (7.3%) (4.6%) (1739) (2.0%)
1696/5036 1856/5036 1330/5036 624/5036 706/5036 154/5036
(33.7%) (36.9%) (26.4%) (12.4%) (14.0%) (3.1%)
1095/3038 1117/3038 729/3038 344/3038 385/3038 97/3038
(36.0%) (36.8%) (24.0%) (11.3%) (12.7%) (3.2%)
601/1998 739/1998 601/1998 280/1998 321/1998 57/1998
(30.1%) (37.0%) (30.1%) (14.0%) (16.1%) (2.9%)
2207/5037 1597/5037 1233/5037 1.49 0.77 373/5054 832/5054 561/5054 1033/5054 2215/5054 1569/5054 133/4759 3830/5054
(43.8%) (31.7%) (24.5%) (5024) (7.4%) (16.5%) (11.1%) (20.4%) (43.8%) (31.0%) (2.8%) (75.8%)
1424/3046 923/3046 699/3046 1.47 0.74 231/3056 388/3056 293/3056 612/3056 1240/3056 930/3056 61/2763 2217/3056
(46.7%) (30.3%) (22.9%) (3038) (7.6%) (12.7%) (9.6%) (20.0%) (40.6%) (30.4%) (2.2%) (72.5%)
783/1991 674/1991 534/1991 1.53 0.80 142/1998 444/1998 268/1998 421/1998 975/1998 639/1998 72/1996 1613/1998
(39.3%) (33.9%) (26.8%) (1986) (7.1%) (22.2%) (13.4%) (21.1%) (48.8%) (32.0%) (3.6%) (80.7%)
0.009 0.548 <0.001 <0.001 0.368 <0.001 0.250 0.004 <0.001
4929/5018 4937/5017 4335/4998 1843/4966 1562/4939 3159/4970 1343/4931
(98.2%) (98.4%) (86.7%) (37.1%) (31.6%) (63.6%) (27.2%)
2969/3030 2974/3027 2613/3023 1113/3011 939/3016 1853/3009 830/3016
(98.0%) (98.2%) (86.4%) (37.0%) (31.1%) (61.6%) (27.5%)
1960/1988 1963/1990 1722/1975 730/1955 623/1923 1306/1961 513/1915
(98.6%) (98.6%) (87.2%) (37.3%) (32.4%) (66.6%) (26.8%)
0.126 0.301 0.468 0.810 0.363 <0.001 0.577
ACE ¼ angiotensin-converting enzyme; BMI ¼ body mass index; GP ¼ glycoprotein; ISR ¼ in-stent restenosis; LV ¼ left ventricle; LVEF ¼ left ventricular ejection fraction; MI ¼ myocardial infarction; NSTEMI ¼ noneST-segment elevation myocardial infarction; STEMI ¼ ST-segment elevation myocardial infarction. * Data are number (%), unless otherwise indicated. Plus-minus values are means SDs. † The chronic renal failure was defined as the presence of kidney damage or estimated glomerular filtration rate of <60 ml/min/1.73 m2 calculated by means of the Cockcroft-Gault formula for <3 months, as with the KDOQI guideline. z Thrombotic total occlusion denotes total occlusion (TIMI flow grade 0) in the angiography with visible thrombus. x Small vessel denotes lesion with reference diameter 2.75 mm. { Long lesion denotes lesion with length 28 mm. k Off-label indication: the indication of PCI was considered “off label” if any of the following features were present: serum creatinine concentration 140 mmol/L (1.6 mg/dl); left ventricular ejection fraction (LVEF) <30%; an acute MI within the previous 72 hours; >1 lesion per vessel; 2 vessels treated with a stent; a lesion length 28 mm; or a bifurcated lesion, bypass graft, in-stent restenosis, unprotected left main coronary artery, presence of thrombus, or total occlusion.
1332
The American Journal of Cardiology (www.ajconline.org)
Table 2 Baseline angiographic characteristics of lesions in crude population* Variable Target coronary artery Left main LAD LC Right ACC/AHA lesion class A B1 B2 C Type B2 or C lesions† In-stent restenosis Severe calcification Bifurcationz Bifurcation treatment Thrombus present Small vesselx Long lesion{ Max pressure deployment (atm) Mean stent diameter/lesion (mm) Total stent length (mm) Per patient Per lesion Number of stents Per patient Per lesion IVUS guided stenting Device success Lesion success Procedure success
Total (N ¼ 7084)
EES (N ¼ 4248)
ZES-R (N ¼ 2836)
258/7084 3179/7084 1567/7084 2071/7084
(3.6%) (44.9%) (22.1%) (29.2%)
178/4248 1907/4248 976/4248 1182/4248
(4.2%) (44.9%) (23.0%) (27.8%)
80/2836 1272/2836 591/2836 889/2836
(2.8%) (44.9%) (20.8%) (31.3%)
564/7084 1705/7084 1650/7084 2285/7084 3935/7084 424/7084 623/7084 919/7084 394/7084 633/7084 1200/7084 2671/7084 13.56 4.63 3.13 3.39
(8.0%) (24.1%) (23.3%) (32.3%) (55.5%) (6.0%) (8.8%) (13.0%) (5.6%) (8.9%) (16.9%) (37.7%) (6487) (7084)
247/4248 1064/4248 987/4248 1358/4248 2345/4248 257/4248 388/4248 419/4248 194/4248 336/4248 704/4248 1504/4248 13.45 4.79 3.16 4.31
(5.8%) (25.0%) (23.2%) (32.0%) (55.2%) (6.0%) (9.1%) (9.9%) (4.6%) (7.9%) (16.6%) (35.4%) (3790) (4248)
317/2836 641/2836 663/2836 927/2836 1590/2836 167/2836 235/2836 500/2836 200/2836 297/2836 496/2836 1167/2836 13.72 4.40 3.09 0.85
(11.2%) (22.6%) (23.4%) (32.7%) (56.1%) (5.9%) (8.3%) (17.6%) (7.1%) (10.5%) (17.5%) (41.1%) (2697) (2836)
38.97 26.01 (5054) 27.97 14.34 (7084) 1.67 0.97 1.19 0.49 2695/7084 6908/7084 6903/7084 6912/7084
(5054) (7084) (38.0%) (97.5%) (97.4%) (97.6%)
37.41 25.50 (3056) 26.90 14.06 (4248) 1.65 0.97 1.19 0.48 1601/4248 4147/4248 4145/4248 4140/4248
41.35 26.58 (1998) 29.61 14.61 (2836)
(3056) (4248) (37.7%) (98.2%) (98.1%) (98.1%)
1.70 0.98 1.19 0.51 1094/2836 2761/2836 2758/2836 2772/2836
(1998) (2836) (38.6%) (98.5%) (98.5%) (98.5%)
p Value 0.001 0.003 0.981 0.035 0.002 <0.001
0.479 0.798 0.231 <0.001 <0.001 <0.001 0.316 <0.001 0.024 0.363 <0.001 <0.001 0.091 0.467 0.454 0.484 0.399 0.479
IVUS ¼ intravascular ultrasound; LAD ¼ left anterior descending artery; LC ¼ left circumflex artery. * Data are number (%), unless otherwise indicated. Plus-minus values are means SDs. † Type B2 or C lesions according to American College of Cardiology/American Heart Association classification. z Bifurcation means bifurcated lesion that has been treated solely by drug-eluting stents. x Small vessel denotes lesion with reference diameter 2.75 mm. { Long lesion denotes lesion with length 28 mm.
performed in the crude population. Second, a propensity scoreematched population was selected to adjust for uneven distribution of baseline characteristics. Kaplan-Meier analysis was used to calculate cumulative incidence of primary and secondary clinical outcomes, and the log-rank test was used to compare between-group differences. A logistic regression model was conducted to generate propensity score, which was the probability that a patient received a ZES-R. The adjusted covariates that were used to calculate the propensity score are listed in the Supplementary Methods. For matching, nearest neighbor matching with a caliper width of 0.6 SD was used because this value has been shown to eliminate >90% of the bias in the observed confounders.11 The discrimination and calibration abilities of propensity score model were assessed by C statistics (0.637) and the Hosmer-Lemeshow statistics (p ¼ 0.515). To reduce possible confounding factors from the differences in baseline characteristics, multivariateadjusted stratified Cox proportional hazard regression to find independent predictors of TLF and exploratory subgroup analysis was performed in propensity scoreematched cohorts. All probability values were 2-sided, and p values <0.05 were considered statistically significant.
Results A total of 5,054 patients (7,084 lesions) were enrolled (EES group: 3,056 patients/4,248 lesions and ZES-R group: 1,998 patients/2,836 lesions). The flow of the study patients is shown in Supplementary Figure 1. Through 3 years, 110 (3.5%) and 120 patients (6.0%) were lost to follow-up in the EES and ZES-R groups, respectively. However, all were confirmed to be alive. The distribution of cardiac risk factors was similar, except for dyslipidemia, lesion complexity, and left main disease (Tables 1 and 2). High-risk patients and lesions were frequent, implying that our registries were an enriched PCI population, reflecting real-world practice in Korea. The device, lesion, and procedure success rates were excellent and similar for both stents (Table 2). At 3 years, the incidence of the primary clinical outcome, TLF, was 6.2% for EES and 6.4% for ZES-R, which was not significantly different (p ¼ 0.812). The rate of the individual components of TLF (cardiac death, TLR, and target vessel MI) was not statistically different between the 2 groups. POCE were also similar, as were the individual components (all-cause death, any revascularization, and any MI). About 1/2 of the target vesselerelated MI was due to ST (Table 3).
Coronary Artery Disease/EES Versus ZES-R in Real-World Patients
1333
Table 3 Clinical outcomes in crude population at 3-year follow-up* Total (N ¼ 5054) All cause death Cardiac death Any myocardial infarction Target vessel Non target vessel MI due to ST Any revascularization Clinically driven revascularization Target lesion revascularization Target vessel revascularization Cerebrovascular accident Target lesion failure† Target vessel failurez Patient-oriented composite eventsx
292 188 37 25 12 12 386 296 123 176 73 361 369 670
(5.8%) (3.7%) (0.7%) (0.5%) (0.2%) (0.2%) (7.6%) (5.9%) (2.4%) (3.5%) (1.4%) (6.3%) (7.3%) (13.3%)
EES (N ¼ 3056) 181 109 25 18 7 9 243 190 76 105 52 189 218 420
ZES-R (N ¼ 1998)
(5.9%) (3.6%) (0.8%) (0.6%) (0.2%) (0.3%) (8.0%) (6.2%) (2.5%) (3.4%) (1.7%) (6.2%) (7.1%) (13.7%)
111 79 12 7 5 3 143 106 47 71 21 127 151 250
(5.6%) (4.0%) (0.6%) (0.4%) (0.3%) (0.2%) (7.2%) (5.3%) (2.4%) (3.6%) (1.1%) (6.4%) (7.6%) (12.5%)
RR (95% CI) 0.94 1.11 0.73 0.59 1.09 0.51 0.90 0.85 0.95 1.03 0.62 1.03 1.06 0.91
(0.75e1.18) (0.83e1.47) (0.39e1.46) (0.25e1.42) (0.35e3.44) (0.14e1.88) (0.74e1.10) (0.68e1.07) (0.66e1.35) (0.77e1.39) (0.37e1.02) (0.83e1.28) (0.87e1.29) (0.79e1.05)
p Value 0.622 0.494 0.404 0.306 0.880 0.385 0.304 0.178 0.780 0.875 0.070 0.812 0.581 0.219
CI ¼ confidence interval; MACE ¼ major adverse cardiovascular events; MI ¼ myocardial infarction; RR ¼ relative risk; ST ¼ stent thrombosis. * Data are number (%), unless otherwise indicated. † Target lesion failure defined as a composite of cardiac death, myocardial infarction (not clearly attributed to a nontarget vessel), or clinically indicated target lesion revascularization by percutaneous or surgical methods at 1 year. z Target vessel failure defined as a composite of cardiac death, myocardial infarction (not clearly attributed to a nontarget vessel), or clinically indicated target vessel revascularization by percutaneous or surgical methods at 1 year. x Patient-oriented composite outcomes included all-cause mortality, any myocardial infarction (includes nontarget vessel territory), and any revascularization (includes all target and nontarget vessel, regardless of percutaneous or surgical methods).
In survival analysis, there were no differences between the 2 groups regarding cumulative incidence of TLF (hazard ratio [HR] 1.03, 95% confidence interval [CI] 0.82 to 1.29, log-rank p ¼ 0.797) or POCE (HR 0.91, 95% CI 0.78 to 1.06, log-rank p ¼ 0.237; Figure 1) as well as the individual components of TLF or POCE (Supplementary Figure 2). The rates of Academic Research Consortiumedefined definite ST and definite or probable ST at 3-year follow-up were low and comparable between the 2 stents (HR 0.70, 95% CI 0.33 to 1.47, p ¼ 0.344, log-rank p ¼ 0.341). All 33 patients with ST had at least 1 off-label indication. Among these, 20 events (60.6%) initially presented as sudden cardiac death. All but 4 patients were on DAPT at the time of the event, and the main reason for discontinuation was bleeding. Definite or probable ST after 1 year occurred in 4 patients (0.1%) without between-group differences (Table 4 and Figure 2). Two of the 4 very late ST events were associated with mortality. A detailed description of all ST cases is presented in Supplementary Table 1. About 1/2 of the patients were on DAPT up to 3-year follow-up, and the mean duration of DAPT was not different between the 2 groups (Table 4). Matching by propensity score yielded 1,698 pairs of patients in both stent groups. Baseline clinical and angiographic characteristics of the matched groups are listed in Supplementary Table 2 and were more balanced than the crude population. The standard differences of the adjusted variables were <10% (Supplementary Figure 3). The cumulative incidences of TLF and POCE up to 3 years and its individual components were comparable between the 2 groups (Figure 1 and Supplementary Table 3). Definite or probable ST was also similar (Supplementary Table 4). In univariate analysis, several underlying co-morbidities and lesion characteristics were significantly different
between those with and without events (Supplementary Table 5). In multivariate analysis, chronic renal failure (adjusted HR 3.615, 95% CI 2.440 to 5.354, p <0.001) and off-label indication (adjusted HR 1.782, 95% CI 1.169 to 2.718, p ¼ 0.007) were the strongest predictors of TLF at 3 years. In addition, other significant predictors of TLF included diabetes mellitus, history of cerebrovascular accident, and increasing age (Table 5). The Harrell’s C-index of the multivariate model was 0.686. Exploratory subgroup analysis at 3-year follow-up with regard to TLF showed consistent results across different subgroups without significant differences in clinical outcomes between EES and ZES-R. The interaction p value was nonsignificant across all the subgroups (Figure 3). Discussion At 3-year follow-up of this large cohort of enriched patients undergoing PCI who received EES or ZES-R, both groups showed excellent and comparable safety and efficacy results. These results extend previous 2-year findings from the RESOLUTE ALL Comers and TWENTE trials. In both crude and propensity scoreematched analysis, 3-year rates of stent- and patient-related composite events were comparable along with its individual components. The absence of a difference in TLF at 3-year follow-up was consistent across various subgroups. The rates of definite or probable ST were very low in both stents especially after 1 year considering the complexity of the lesions treated and were balanced between the 2 stents, suggesting excellent safety. Finally, off-label use of the stent and important co-morbidities, such as chronic renal failure and diabetes mellitus, were significantly associated with TLF at 3 years, rather than the stent type.
1334
The American Journal of Cardiology (www.ajconline.org)
Figure 1. Survival analysis: primary and major secondary outcomes. Kaplan-Meier curves are shown for each outcome and cohort combination (A to D).
The 2 previous randomized controlled trials that compared these DESs reported 2-year TLF rates of 10.7% for EESs and 11.2% for ZES-R (p ¼ 0.74) and 9.8% for EESs and 10.5% for ZES-R (p ¼ 0.68). In the present study, the 3-year TLF rate was relatively lower (6.2%/ 6.4%, p ¼ 0.812), mainly due to the relatively lower rates of target vesselerelated MI (RESOLUTE ALL Comers 4.6%, TWENTE 5.5% at 2 years, EXCELLENT and RESOLUTE-Korea 0.5% at 3 years). The main reason for relatively lower rates of target vessel MI might be nonsystematic collection of cardiac biomarkers after index procedures in each participating centers because it was not mandated and is not a routine practice in Korea. In
addition, TWENTE 2-year results reported significant lower rates of TLR in the EES group than the ZES-R group (4.9% vs 2.6%, p ¼ 0.03); however, 3-year rates of TLR were comparable in the present study (2.5% vs 2.4%, p ¼ 0.780). Both second-generation DESs enhanced biocompatibility of the polymers and incorporated a thinner cobalt-chromium stent platform to enhance endothelial coverage of stent struts and reduce risk of ST. Our results support the previous 2-year results from the RESOLUTE ALL Comers and TWENTE trials and further show that the enhanced safety and efficacy profiles of these stents are sustained up to 3-year follow-up. Notably, ST only occurred after off-label use, and >1/2 of the events were
Coronary Artery Disease/EES Versus ZES-R in Real-World Patients
1335
Table 4 Stent thrombosis and use of dual antiplatelet agents in crude population at 3-year follow-up* Total (N ¼ 5054) Definite Acute (0e1 day) Subacute (2e30 days) Late (31e360 days) Very late (361 days) Probable Acute (0e1 day) Subacute (2e30 days) Late (31e360 days) Very late (361 days) Possible Late (31e360 days) Stent thrombosis Definite Definite or probable Definite, probable, or possible Very late definite or probable Clinical presentation at time of ST Sudden cardiac death ST-segment elevation MI NSTEMI/unstable angina Use of dual anti-platelet agent DAPT at 1-year No DAPT user at 1-year Aspirin single Clopidogrel single None of DAPT DAPT at 2-year No DAPT user at 2-year Aspirin single Clopidogrel single None of DAPT DAPT at 3-year No DAPT user at 3-year Aspirin single Clopidogrel single None of DAPT Mean duration of DAPT
EES (N ¼ 3056)
12 4 3 2 3 20 4 14 1 1 1 1
(0.2%) (0.1%) (0.1%) (<0.1%) (0.1%) (0.4%) (0.1%) (0.3%) (<0.1%) (<0.1%) (<0.1%) (<0.1%)
8 3 2 1 2 14 1 11 1 1 1 1
(0.3%) (0.1%) (0.1%) (<0.1%) (0.1%) (0.5%) (<0.1%) (0.4%) (<0.1%) (<0.1%) (<0.1%) (<0.1%)
12 32 33 4
(0.2%) (0.6%) (0.7%) (0.1%)
8 22 23 3
(0.3%) (0.7%) (0.8%) (0.1%)
20/33 (60.6%) 9/33 (27.3%) 4/33 (12.1%) 3762/4684 922/4684 640/922 173/922 109/922 2259/4359 2100/4359 1517/2100 427/2100 156/2100 1807/4027 2220/4027 1550/2220 510/2220 160/2220 737.4 332.8
(80.3%) (19.7%) (69.4%) (18.8%) (11.8%) (51.8%) (48.2%) (72.2%) (20.3%) (7.4%) (44.9%) (55.1%) (69.8%) (23.0%) (7.2%) (4586)
14/23 (60.9%) 7/23 (30.4%) 2/23 (8.7%) 2312/2856 544/2856 355/544 116/544 73/544 1383/2650 1267/2650 893/1267 270/1267 104/1267 1110/2479 1369/2479 932/1369 324/1369 113/1369 743.9 331.9
(81.0%) (19.0%) (65.3%) (21.3%) (13.4%) (52.2%) (47.8%) (70.5%) (21.3%) (8.2%) (44.8%) (55.2%) (68.1%) (23.7%) (8.3%) (2793)
ZES-R (N ¼ 1998)
p Value
4 1 1 1 1 6 3 3 0 0 0 0
(0.2%) (0.1%) (0.1%) (0.1%) (0.1%) (0.3%) (0.2%) (0.2%) (0%) (0%) (0%) (0%)
0.774 0.657 1.000 1.000 1.000 0.494 0.307 0.185 1.000 1.000 1.000 1.000
4 10 10 1
(0.2%) (0.5%) (0.5%) (0.1%)
0.774 0.370 0.292 0.657 0.739
6/10 (60.0%) 2/10 (20.0%) 2/10 (20.0%) 1450/1828 378/1828 285/378 57/378 36/378 876/1709 833/1709 624/833 157/833 52/833 697/1548 851/1548 618/851 186/851 47/851 727.3 334.0
(79.3%) (20.7%) (75.4%) (15.1%) (9.5%) (51.3%) (48.7%) (74.9%) (18.8%) (6.2%) (45.0%) (55.0%) (72.6%) (21.9%) (5.5%) (1793)
0.175
0.555
0.896
0.100
DAPT ¼ dual antiplatelet agent therapy; MI ¼ myocardial infarction; NSTEMI ¼ noneST-segment elevation myocardial infarction; ST ¼ stent thrombosis. * Data are number (%), unless otherwise indicated.
fatal. Even among very late ST cases, 2 of 4 eventually led to mortality. Interestingly, 87.9% of the all STs, including 3 of 4 very late definite or probable STs, occurred while on DAPT. The TWENTE trial showed that the rate of very late ST was still low (0.3% in each group) despite a low rate of continued DAPT usage beyond 12 months, which was in accordance with the current guidelines.12 In the present study, the rate of DAPT use at 2-year follow-up was 48.2%, which was relatively higher than that in other 2 trials (RESOLUTE ALL Comers trial: 18%, TWENTE trial: 6.9%). Although it might be difficult to directly compare the rates of clinical outcomes from different trials because of different patient populations; nevertheless, the overall incidence of very late ST was <1% across all 3 studies and was lower than that reported for the first-generation DESs, which showed continuous risk of very late ST.13 With the use of second-generation DESs, whether prolonged DAPT can reduce the
possibility of very late ST is still unclear and need to be clarified in future trials. In multivariate analysis, chronic renal failure was the strongest predictor of TLF, concordant with previous literature.14 Other independent predictors included off-label DES use and diabetes mellitus. ZES-R showed significantly lower incidence of target vessel failure at 1 year (7.8%) than predefined DES performance goal of 14.5% in patients with diabetes mellitus and achieved the first Food and Drug Administration approval for PCI in patients with diabetes.15 However, it should be noticed that the pooled analysis of the international global RESOLUTE program included only the patients with on-label indication of the stent; the high-risk patients with off-label indication in the Resolute All Comers trial and the Resolute International registry were excluded from the pooled analysis.15 Despite previously reported safety and efficacy of the secondgeneration stents in diabetes population, our study
1336
The American Journal of Cardiology (www.ajconline.org)
Figure 2. Survival analysis: definite or probable ST. Arrow indicates the patients not on DAPT at the time of ST.
Table 5 Independent predictors of target lesion failure at 3-year follow-up in propensity scoreematched group*
Chronic renal failure Off label indication Diabetes mellitus Previous history of CVA Age
Hazard Ratio
95% CI
p Value
3.615 1.782 1.719 1.599 1.035
2.440e5.354 1.169e2.718 1.304e2.266 1.078e2.372 1.021e1.049
<0.001 0.007 <0.001 0.020 <0.001
CI ¼ confidence interval; CVA ¼ cerebrovascular accident; LVEF ¼ left ventricular ejection fraction; NSTEMI ¼ noneST-segment elevation myocardial infarction; STEMI ¼ ST-segment elevation myocardial infarction. * Identification of independent predictors was done with stratified Cox proportional hazard regression model, and the variables are presented with multivariate-adjusted hazard ratios, 95% confidence intervals, and p values. Included variables to the final model are presented in Supplementary Table 5. The individual components of off-label indication (i.e., STEMI, NSTEMI, in-stent restenosis, bifurcation, thrombotic total occlusion, long lesion, multivessel PCI, severe left ventricular dysfunction [LVEF <30%], and left main procedure) were not included individually to the final model because of significant correlation with off-label indication itself (i.e., colinearity between these covariates). The Harrell’s c-index of the model was 0.686.
confirmed that it is still an important risk factor in the treatment of coronary artery disease.16 Along with the excellent safety and efficacy profiles of the secondgeneration DES up to 3-year follow-up, the results of the multivariate analysis again show the importance of risk factor control, which is not directly related to the stent itself, when considering the prognosis of individual patients.
The present study has several limitations. First, inherent limitations of nonrandomized comparisons such as allocation bias and uneven distribution of risk factors should be considered. Although rigorous risk adjustment model including propensity score matching and multiple imputation of missing values was performed, unmeasured variables could not be controlled. Second, relatively lower incidence of clinical events, mainly driven by lower incidence of target vessel MI compared with previous randomized controlled trials and pooled analysis of EES and ZES-R,5,6,15,17 should be acknowledged. Because postprocedural collection of cardiac biomarkers to detect periprocedural MI was not routinely undertaken in all the participating centers, this might be the reason for the lower incidence of MI in our registries. Therefore, we believe the true rate of MI would have been much higher if the cardiac enzymes were followed systematically. Third, although vital status of all patients including patients lost to follow-up was confirmed with the Korean national database using a citizen registration number that is unique to each individual, we cannot exclude the possibility of under-reporting of clinical outcomes other than death such as nonfatal MI, TLR, or most importantly, ST that did present as sudden cardiac death, in the patients who were lost to follow-up. Disclosures Dr. H.S. Kim received research grants, lecture fees, and honoraria from Medtronic Cardiovascular, Minneapolis, Minnesota, Abbott Vascular, Santa Clara, California, and Boston Scientific, Natick, Massachusetts. The other authors have no conflicts of interest to disclose.
Coronary Artery Disease/EES Versus ZES-R in Real-World Patients
1337
Figure 3. Subgroup analysis for TLF in propensity scoreematched cohort. Exploratory subgroup analysis at 3-year follow-up with regard to TLF is shown in the figure. AMI ¼ acute myocardial infarction; STEMI ¼ ST-segment elevation myocardial infarction.
Supplementary Data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. amjcard.2014.07.065. 1. Stone GW, Rizvi A, Sudhir K, Newman W, Applegate RJ, Cannon LA, Maddux JT, Cutlip DE, Simonton CA, Sood P, Kereiakes DJ, Investigators SI. Randomized comparison of everolimus- and paclitaxeleluting stents. 2-year follow-up from the SPIRIT (Clinical Evaluation of the XIENCE V Everolimus Eluting Coronary Stent System) IV trial. J Am Coll Cardiol 2011;58:19e25. 2. Byrne RA, Kastrati A, Massberg S, Wieczorek A, Laugwitz KL, Hadamitzky M, Schulz S, Pache J, Fusaro M, Hausleiter J, Schomig A, Mehilli J; Investigators I-T. Biodegradable polymer versus permanent polymer drug-eluting stents and everolimus- versus sirolimus-eluting stents in patients with coronary artery disease: 3year outcomes from a randomized clinical trial. J Am Coll Cardiol 2011;58:1325e1331. 3. Jensen LO, Thayssen P, Christiansen EH, Tilsted HH, Maeng M, Hansen KN, Kaltoft A, Hansen HS, Botker HE, Krusell LR, Ravkilde J, Madsen M, Thuesen L, Lassen JF; Investigators SOI. 2-Year patientrelated versus stent-related outcomes: the SORT OUT IV (Scandinavian Organization for Randomized Trials With Clinical Outcome IV) Trial. J Am Coll Cardiol 2012;60:1140e1147.
4. Smits PC, Kedhi E, Royaards KJ, Joesoef KS, Wassing J, RademakerHavinga TA, McFadden E. 2-year follow-up of a randomized controlled trial of everolimus- and paclitaxel-eluting stents for coronary revascularization in daily practice. COMPARE (Comparison of the everolimus eluting XIENCE-V stent with the paclitaxel eluting TAXUS LIBERTE stent in all-comers: a randomized open label trial). J Am Coll Cardiol 2011;58:11e18. 5. Silber S, Windecker S, Vranckx P, Serruys PW. Unrestricted randomised use of two new generation drug-eluting coronary stents: 2-year patient-related versus stent-related outcomes from the RESOLUTE All Comers trial. Lancet 2011;377:1241e1247. 6. Tandjung K, Sen H, Lam MK, Basalus MW, Louwerenburg JH, Stoel MG, van Houwelingen KG, de Man FH, Linssen GC, Said SA, Nienhuis MB, Lowik MM, Verhorst PM, van der Palen J, von Birgelen C. Clinical outcome following stringent discontinuation of dual antiplatelet therapy after 12 months in real-world patients treated with second-generation zotarolimus-eluting Resolute and everolimus-eluting Xience V stents: 2-year follow-up of the randomized TWENTE trial. J Am Coll Cardiol 2013;61:2406e2416. 7. Taniwaki M, Stefanini GG, Silber S, Richardt G, Vranckx P, Serruys PW, Buszman PE, Kelbaek H, Windecker S. Four year clinical outcomes and predictors of repeat revascularization in patients treated with new generation drug-eluting stents in the RESOLUTE all comers randomized trial. J Am Coll Cardiol 2014;63:1617e1625. 8. Park KW, Lee JM, Kang SH, Ahn HS, Yang HM, Lee HY, Kang HJ, Koo BK, Cho J, Gwon HC, Lee SY, Chae IH, Youn TJ, Chae JK,
1338
9.
10.
11. 12.
13.
The American Journal of Cardiology (www.ajconline.org)
Han KR, Yu CW, Kim HS. Safety and efficacy of second-generation everolimus-eluting Xience V stents versus zotarolimus-eluting resolute stents in real-world practice: patient-related and stent-related outcomes from the multicenter prospective EXCELLENT and RESOLUTE-Korea registries. J Am Coll Cardiol 2013;61:536e544. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A, Hamon M, Krucoff MW, Serruys PW. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation 2007;115:2344e2351. Vranckx P, Cutlip DE, Mehran R, Kint PP, Silber S, Windecker S, Serruys PW. Myocardial infarction adjudication in contemporary allcomer stent trials: balancing sensitivity and specificity. Addendum to the historical MI definitions used in stent studies. EuroIntervention 2010;5:871e874. Gu XS, Rosenbaum PR. Comparison of multivariate matching methods: structures, distances, and algorithms. J Comput Graphical Stat 1993;2:405e420. Levine GN, Bates ER, Blankenship JC, Bailey SR, Bittl JA, Cercek B, Chambers CE, Ellis SG, Guyton RA, Hollenberg SM, Khot UN, Lange RA, Mauri L, Mehran R, Moussa ID, Mukherjee D, Nallamothu BK, Ting HH; American College of Cardiology F; American Heart Association Task Force on Practice G; Society for Cardiovascular A, Interventions. 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Society for Cardiovascular Angiography and Interventions. J Am Coll Cardiol 2011;58:e44ee122. Daemen J, Wenaweser P, Tsuchida K, Abrecht L, Vaina S, Morger C, Kukreja N, Juni P, Sianos G, Hellige G, van Domburg RT, Hess OM,
14.
15.
16.
17.
Boersma E, Meier B, Windecker S, Serruys PW. Early and late coronary stent thrombosis of sirolimus-eluting and paclitaxel-eluting stents in routine clinical practice: data from a large two-institutional cohort study. Lancet 2007;369:667e678. Saltzman AJ, Stone GW, Claessen BE, Narula A, Leon-Reyes S, Weisz G, Brodie B, Witzenbichler B, Guagliumi G, Kornowski R, Dudek D, Metzger DC, Lansky AJ, Nikolsky E, Dangas GD, Mehran R. Long-term impact of chronic kidney disease in patients with ST-segment elevation myocardial infarction treated with primary percutaneous coronary intervention: the HORIZONS-AMI (Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction) trial. JACC Cardiovasc Interv 2011;4: 1011e1019. Silber S, Serruys PW, Leon MB, Meredith IT, Windecker S, Neumann FJ, Belardi J, Widimsky P, Massaro J, Novack V, Yeung AC, Saito S, Mauri L. Clinical outcome of patients with and without diabetes mellitus after percutaneous coronary intervention with the resolute zotarolimus-eluting stent: 2-year results from the prospectively pooled analysis of the international global RESOLUTE program. JACC Cardiovasc Interv 2013;6:357e368. Park KW, Lee JM, Kang SH, Ahn HS, Kang HJ, Koo BK, Rhew JY, Hwang SH, Lee SY, Kang TS, Kwak CH, Hong BK, Yu CW, Seong IW, Ahn T, Lee HC, Lim SW, Kim HS. Everolimus-eluting xience v/ promus versus zotarolimus-eluting resolute stents in patients with diabetes mellitus. JACC Cardiovasc Interv 2014;7:471e481. Stone GW, Kedhi E, Kereiakes DJ, Parise H, Fahy M, Serruys PW, Smits PC. Differential clinical responses to everolimus-eluting and Paclitaxel-eluting coronary stents in patients with and without diabetes mellitus. Circulation 2011;124:893e900.