Studies have indicated varying mortality risks with timing of stent thrombosis (ST), but few have been adequately powered with prospective late follow-up. PROTECT randomized 8,709 subjects to either Endeavor zotarolimus-eluting or Cypher sirolimus-eluting stents. PROTECT Continued Access enrolled 1,018 patients treated with Endeavor zotarolimus-eluting stents. Subjects completed at least 4 and 3 years of follow-up, respectively. ARC-defined definite and probable ST events were stratified by time from index procedure: early (≤30 days), late (>30 and ≤360 days), and very late (>360 days). Rates of death and myocardial infarction were analyzed by ST timing. Median follow-up was 4.1 years. There were 184 ST events (1.9%): 61 early, 27 late, and 96 very late. Patient and procedural characteristics were similar between timing groups. There was no difference in dual-antiplatelet therapy use at discharge (97%) or 1 year (84%). Cardiac death in patients with ST at 4 years occurred in 32.1% compared with 2.5% in patients without ST (p <0.001). Combined rates of cardiac death and myocardial infarction did not differ according to ST timing, yet early ST was more commonly associated with cardiac death at 4 years than later ST (50.8% for early vs 18.5% for late vs 24.0% for very late; p <0.001). The relation between ST timing and outcomes did not differ between stent types. In conclusion, in prospective data, cardiac death was more common after early ST than later ST. Although ST remains infrequent, continued efforts to determine how to reduce ST, particularly within the first 30 days, are warranted. (The PROTECT trial is registered with ClinicalTrials.gov , number NCT00476957 .)
The mechanistic diversity of stent thrombosis (ST) across time intervals may correspond with differing clinical outcomes. However, because of the low frequency of ST, this relationship has been poorly characterized. Previous prospective studies have often been underpowered, used restricted inclusion criteria, or lacked long-term follow-up. Alternatively, large registry studies have excluded patients with out-of-hospital events, lacked standardization in the definition and adjudication of ST, and varied in their results. For instance, data from the CathPCI registry demonstrated that early ST was associated with the highest risk of inpatient mortality, whereas the REgistry of Stent Thrombosis for review And Re-evaluaTion (RESTART) found similar rates of mortality at 30 days and 1 year for patients with early and late ST. As the relation between timing of ST and adverse clinical outcomes remains unclear, the present study was conducted to address this question using prospective, pooled patient-level data from Patient Related Outcomes with Endeavor versus Cypher Stenting Trial (PROTECT) and PROTECT Continued Access trials. The PROTECT trials enrolled a broad patient population with both stable angina and acute coronary syndromes in an attempt to represent real-world practices, while including protocol-mandated long-term clinical follow-up with central event adjudication for the primary end point of ST.
Methods
We examined patients from the previously conducted PROTECT and PROTECT Continued Access trials, totaling 9,727 subjects. The PROTECT trial was a large, broadly inclusive, multicenter randomized controlled trial comparing the long-term safety of the Endeavor zotarolimus-eluting stent (E-ZES) and the Cypher sirolimus-eluting stent (C-SES). Patients were consented for 5 years of follow-up. From May 2007 to December 2008, 8,709 subjects were randomized 1:1 to receive either E-ZES (4,357 patients) or C-SES (4,352 patients). The PROTECT Continued Access trial was a single-arm study that paralleled the PROTECT trial inclusion/exclusion criteria. PROTECT Continued Access enrolled 1,018 patients who received an E-ZES and were also consented for 5 years of follow-up. Both trials included only patients who underwent percutaneous coronary intervention (PCI) with DES implantation and excluded patients with previous bare-metal stents in the past 12 months, previous DES, previous brachytherapy, or need for oral anticoagulation. Standard stent implantation procedures were followed. Dual-antiplatelet therapy consisting of aspirin and a thienopyridine was recommended for a minimum of 3 months and up to 12 months or longer according to guidelines and local practices. The present analysis included at least 4 years of follow-up for patients in the PROTECT trial or 3 years of follow-up for patients in the PROTECT Continued Access trial. All study patients were included in the analysis; however, only patients from the PROTECT trial who attended their 4-year visit and the PROTECT Continued Access trial who attended their 3-year visit were considered as having completed follow-up, unless death occurred.
Collected patient and procedural characteristics included demographic variables (age, gender, and body mass index), coronary artery disease risk factors (hypertension, diabetes, hyperlipidemia, and smoking status), cardiovascular history (previous PCI, previous coronary artery bypass graft surgery, previous myocardial infarction [MI], previous stroke, baseline left ventricular ejection fraction, and baseline serum creatinine level), presentation for index procedure (silent ischemia, stable angina, unstable angina, non–ST-elevation MI, and ST-elevation MI), and angiographic/procedural data (lesion location by vessel, multivessel PCI, lesion length, total stent length, vessel diameter, bifurcation lesion, saphenous vein graft intervention, in-stent restenosis, type of stent implanted, and device success). Device success was defined as attaining <50% residual stenosis of the target lesion with only the assigned device.
All patients who experienced probable or definite ST by the end of follow-up were identified and categorized by the timing of the event. ST was a prespecified outcome for both PROTECT studies, was adjudicated by an independent events committee with angiographic review for all definite ST cases, and was classified according to the Academic Research Consortium (ARC) definitions for definite or probable ST. Timing of ST relative to the index procedure was measured in days and stratified into ARC-defined timing strata: early (≤30 days), late (>30 and ≤360 days), and very late (>360 days). Acute and subacute ST were categorized as early ST for this analysis. The primary outcome was cardiac death in patients with ST through the end of follow-up. Secondary outcomes in patients with ST included (1) cardiac death occurring on the same day as ST event, (2) MI occurring on the same day as ST event, (3) MI occurring through the end of follow-up, (4) cardiac death and/or MI occurring on the same day as ST event, (5) cardiac death and/or MI occurring through the end of follow-up, (6) all-cause mortality occurring through the end of follow-up, and (7) recurrent ST through the end of follow-up. Cardiac death was a prespecified outcome for both PROTECT studies, was adjudicated by an independent events committee whenever a patient death was identified, and was classified according to the ARC definition as cardiac death unless an unequivocal noncardiac cause could be established. MI and all-cause mortality were also prespecified, adjudicated outcomes. MIs were classified on the basis of ARC definitions as troponin or creatinine kinase MB more than 3 times the upper limits of normal for periprocedural MIs and either creatinine kinase MB or troponin greater than the upper limits of normal for spontaneous MIs.
Clinical and procedural characteristics with dichotomous and categorical variables were reported as counts and percentages and continuous variables as means with standard deviations. Between-group differences were assessed using Fisher’s exact and the chi-square tests for binary and categorical variables and analysis of variance tests for continuous variables. Temporal distribution of the primary and secondary outcomes was reported as rates (number of events in patients at risk) for each of the 3 time strata. Rates were calculated for all events occurring on the day of ST and through 4 years of follow-up after the index PCI. Time to cardiac death after ST for each timing group was calculated by the Kaplan–Meier method, and log-rank tests were used to examine for significance between groups. Patient follow-up for the Kaplan–Meier analysis began at the time of the ST event and ended at the time of cardiac death or when follow-up ended. Patients were then grouped into timing intervals corresponding with the period in which the ST event occurred, with the day of the ST event assigned day 0. A 2-sided p value <0.05 was considered significant. Statistical analyses were done using SAS software versions 8.2 or higher (SAS Institute, Cary, NC).
Results
Complete follow-up was available for 96.7% (9,370 of 9,727 patients) at a median duration of 4.1 years (interquartile range [IQR] 4.0 to 5.0 years). A total of 184 ST events (1.9% of patients) occurred during the study period: 61 early (0.6% of patients), 27 late (0.3% of patients), and 96 very late (1.0% of patients). The annualized rate of very late ST was 0.33%. Median time from index PCI to ST was 5 days (IQR 3 to 11 days) for early ST, 151 days (IQR 88 to 270 days) for late ST, and 1,018 days (IQR 660 to 1241 days) for very late ST.
Compared with patients without ST, patients with ST had a greater prevalence of diabetes mellitus, history of smoking, previous MI, and previous stroke ( Table 1 ). In addition, patients with ST were more likely to present for their index PCI with MI, whereas patients without ST more often presented with unstable angina or silent ischemia. Baseline characteristics and procedure indications were similar between patients with early, late, and very late ST, with the exception of diabetes occurring more commonly in patients with late ST ( Table 1 ). Baseline lesion, procedure, and post-procedure characteristics are provided in Table 2 . Lesion characteristics differed minimally according to ST timing, whereas patients with ST compared to those without ST had a greater prevalence of index lesion calcification, thrombus present and longer lesion lengths. Similarly, procedural characteristics were comparable according to ST timing; however, patients with ST compared to patients without ST had a greater number of lesions treated, received a greater number of stents, and were more often treated with glycoprotein IIb/IIIa inhibitors. Patients with ST received more C-SES stents, whereas patients without ST received more E-ZES stents. In addition, patients with early and late ST received more E-ZES stents compared with patients with very late ST who received more C-SES stents. Post-procedurally, no significant differences were observed for patients with and without ST. However, among patients with ST, those with late ST were less likely to achieve TIMI 3 flow compared with patients with early and very late ST ( Table 2 ). In addition, lack of device success was associated more commonly with patients with early ST. Use of dual-antiplatelet therapy in patients with ST was 97.3% at discharge, 96.3% at 30 days, 84.0% at 1 year, and 44.3% at 2 years. There were no significant differences in dual-antiplatelet therapy (DAPT) use at all time points between patients with and without ST and between ST timing groups.
| Characteristic | Early ST n = 61 | Late ST n = 27 | Very Late ST n = 96 | p value | All ST n = 184 | No ST n = 9543 | p value |
|---|---|---|---|---|---|---|---|
| Age (years), mean ± SD | 64.5 ± 12.6 | 63.9 ± 10.9 | 60.8 ± 11.1 | 0.12 | 62.5 ± 11.7 | 62.3±10.7 | 0.78 |
| Male | 50/61 (82%) | 18/27 (67%) | 77/96 (80%) | 0.24 | 145/184 (78.8%) | 7247/9543 (75.9%) | 0.37 |
| Body mass index (kg/m 2 ), mean ± SD | 29.2 ± 7.2 | 29.2 ± 4.6 | 27.7 ± 4.5 | 0.18 | 28.4 ± 5.5 | 281 ± 4.7 | 0.35 |
| Diabetes mellitus | 30/61 (49%) | 15/27 (56%) | 31/96 (32%) | 0.030 | 76/184 (41.3%) | 2666/9543 (27.9%) | <0.01 |
| Insulin dependent | 11/61 (18%) | 5/27 (19%) | 10/96 (10%) | 0.32 | 26/184 (14.1%) | 691/9543 (7.2%) | <0.01 |
| Hypertension | 44/61 (72%) | 19/27 (70%) | 64/96 (67%) | 0.76 | 127/184 (69.0%) | 6252/9543 (65.5%) | 0.32 |
| Hyperlipidemia | 37/61 (61%) | 19/27 (70%) | 70/96 (73%) | 0.27 | 126/184 (68.5%) | 6099/9543 (63.9%) | 0.20 |
| History of smoking | 39/61 (64%) | 14/27 (52%) | 69/96 (72%) | 0.14 | 122/184 (66.3%) | 5533/9542 (58.0%) | 0.023 |
| Current smoker | 21/61 (34%) | 6/27 (22%) | 37/96 (39%) | 0.29 | 64/184 (34.8%) | 2377/9542 (24.9%) | <0.01 |
| Premature coronary artery disease in first-degree relative | 22/51 (43%) | 6/22 (27%) | 31/80 (39%) | 0.44 | 59/153 (38.6%) | 2906/8211 (35.4%) | 0.42 |
| Prior myocardial infarction | 22/61 (36%) | 6/27 (22%) | 27/96 (28%) | 0.37 | 55/184 (29.9%) | 1874/9543 (19.6%) | <0.01 |
| Prior coronary artery bypass graft surgery | 4/61 (7%) | 2/27 (7%) | 6/96 (6%) | 0.98 | 12/184 (6.5%) | 510/9543 (5.3%) | 0.48 |
| Prior percutaneous coronary intervention | 11/61 (18%) | 3/27 (11%) | 12/96 (13%) | 0.56 | 26/184 (14.1%) | 1212/9543 (12.7%) | 0.56 |
| Prior stroke | 6/61 (10%) | 0/27 (0%) | 5/96 (5%) | 0.18 | 11/184 (6.0%) | 303/9543 (3.2%) | 0.033 |
| Left ventricular ejection fraction (%), mean ± SD | 50.8 ± 15.3 | 53.1 ± 13.7 | 56.7 ± 11.3 | 0.17 | 53.6 ± 13.6 | 58.1 ± 12.2 | <0.01 |
| Serum creatinine (umol), mean ± SD | 105.3 ± 82.7 | 85.8 ± 22.7 | 95.9 ± 31.4 | 0.30 | 97.6 ± 54.1 | 88.5 ± 38.4 | <0.01 |
| Procedure indication | 0.058 | <0.01 | |||||
| Unstable angina | 10/61 (16%) | 1/27 (4%) | 15/96 (16%) | 26/184 (14.1%) | 1820/9542 (19.1%) | ||
| Stable angina | 27/61 (44%) | 19/27 (70%) | 49/96 (51%) | 95/184 (51.6%) | 4657/9542 (48.8%) | ||
| Silent ischemia | 0/61 (0%) | 2/27 (7%) | 2/96 (2%) | 4/184 (2.2%) | 683/9542 (7.2%) | ||
| Acute myocardial infarction | 24/61 (39%) | 5/27 (19%) | 30/96 (31%) | 59/184 (32.1%) | 2382/9542 (25.0%) | ||
| ST-elevation myocardial infarction | 10/61 (16%) | 2/27 (7%) | 9/96 (9%) | 0.31 | 21/184 (11.4%) | 750/9542 (7.9%) | 0.077 |
| Non-ST-elevation myocardial infarction | 14/61 (23%) | 3/27 (11%) | 21/96 (22%) | 0.41 | 38/184 (20.7%) | 1632/9542 (17.1%) | 0.21 |
| Variable | Early ST n = 61 patients; n = 101 lesions | Late ST n = 27 patients; n = 40 lesions | Very Late ST n = 96 patients; n = 159 lesions | p value | All ST n = 184 patients; n= 300 lesions | No ST n = 9543 patients; n= 13495 lesions | p value |
|---|---|---|---|---|---|---|---|
| Lesion characteristics | |||||||
| Coronary vessel location (by lesion) | 0.62 | 0.013 | |||||
| Left anterior descending | 50/101 (50%) | 19/40 (48%) | 67/159 (42%) | 136/300 (45.3%) | 6171/13492 (45.7%) | ||
| Left circumflex | 19/101 (19%) | 8/40 (20%) | 33/159 (21%) | 60/300 (20.0%) | 3088/13492 (22.9%) | ||
| Right | 27/101 (27%) | 13/40 (33%) | 56/159 (35%) | 96/300 (32.0%) | 4077/13492 (30.2%) | ||
| Left main | 3/101 (3%) | 0/40 (0%) | 1/159 (1%) | 4/300 (1.3%) | 118/3492 (0.9%) | ||
| Bypass graft | 2/101 (2%) | 0/40 (0%) | 2/159 (1%) | 4/300 (1.3%) | 38/3492 (0.3%) | ||
| Restenosis after previous percutaneous coronary intervention | 0/101 (0%) | 0/40 (0%) | 0/159 (0%) | 0.055 | 0/300 (0.0%) | 48/13491 (0.4%) | 0.69 |
| In-stent restenosis | 0/101 (0%) | 1/40 (3%) | 1/159 (1%) | 0.055 | 2/300 (0.7%) | 140/3491 (1.0%) | 0.69 |
| Chronic total occlusion ∗ | 0/101 (0%) | 1/40 (3%) | 4/159 (3%) | 0.28 | 5/300 (1.7%) | 381/13486 (2.8%) | 0.23 |
| Bifurcation | 11/101 (11%) | 8/40 (20%) | 22/159 (14%) | 0.36 | 41/300 (13.7%) | 2208/13489 (16.4%) | 0.21 |
| Moderate or severe calcification (vs none or mild) | 43/101 (43%) | 17/40 (43%) | 51/159 (32%) | 0.17 | 111/300 (37.0%) | 3785/13487 (28.1%) | <0.001 |
| Moderate or severe tortuosity (vs mild) | 30/101 (30%) | 14/40 (35%) | 30/159 (19%) | 0.038 | 74/300 (24.7%) | 3037/13477 (22.5%) | 0.38 |
| Pre-Procedure TIMI Flow | 0.008 | 0.27 | |||||
| 0 | 2/101 (2%) | 2/40 (5%) | 10/159 (6%) | 14/300 (4.7%) | 914/13488 (6.8%) | ||
| 1 | 16/101 (16%) | 1/40 (3%) | 9/159 (6%) | 26/300 (8.7%) | 1001/13488 (7.4%) | ||
| 2 | 18/101 (18%) | 2/40 (5%) | 27/159 (17%) | 47/300 (15.7%) | 1790/13488 (13.3%) | ||
| 3 | 65/101 (64%) | 35/40 (88%) | 113/159 (71%) | 213/300 (71.0%) | 9783/13488 (72.5%) | ||
| Thrombus | 17/101 (17%) | 4/40 (10%) | 13/159 (8%) | 0.097 | 34/300 (11.3%) | 1055/13488 (7.8%) | 0.026 |
| ACC/AHA lesion class B2/C | 60/101 (59%) | 28/40 (70%) | 102/159 (64%) | 0.48 | 190/300 (63.3%) | 7218/13477 (53.6%) | <0.001 |
| Reference vessel diameter (mm), mean ± SD | 2.9 ± 0.5 | 2.9 ± 0.5 | 2.9 ± 0.5 | 0.78 | 2.9 ± 0.5 | 3.0 ± 0.5 | 0.002 |
| Minimum luminal diameter (mm), mean ± SD | 0.5 ± 0.4 | 0.5 ± 0.3 | 0.5 ± 0.4 | 0.92 | 0.5 ± 0.4 | 0.5 ± 0.5 | 0.74 |
| Diameter stenosis (%), mean ± SD | 81.0 ± 12.6 | 82.7 ± 9.8 | 82.4 ± 13.5 | 0.64 | 82.0 ± 12.8 | 83.0 ± 12.7 | 0.20 |
| Lesion length (mm), mean ± SD | 18.5 ± 9.9 | 21.6 ± 12.8 | 19.5 ± 10.0 | 0.28 | 19.5 ± 10.4 | 17.5 ± 9.1 | <0.001 |
| Procedure characteristics | |||||||
| Number of vessels treated per patient, mean ± SD | 1.3 ± 0.6 | 1.2 ± 0.4 | 1.3 ± 0.5 | 0.41 | 1.3 ± 0.45 | 1.2 ± 0.5 | 0.007 |
| Number of lesions treated per patient, mean ± SD | 1.6 ± 1.0 | 1.5 ± 1.0 | 1.6 ± 0.8 | 0.71 | 1.6 ± 0.9 | 1.4 ± 0.7 | <0.001 |
| Number of stents per patient, mean ± SD | 2.1 ± 1.2 | 1.8 ± 1.6 | 1.9 ± 1.0 | 0.49 | 2.0 ± 1.2 | 1.6 ± 1.0 | <0.001 |
| Total stent length per patient (mm), mean ± SD | 40.6 ± 27.7 | 38.6 ± 41.3 | 40.2 ± 23.6 | 0.95 | 40.1 ± 28.0 | 31.1 ± 20.7 | <0.001 |
| Number of stents per lesion, mean ± SD | 1.3 ± 0.7 | 1.2 ± 0.5 | 1.2 ± 0.5 | 0.36 | 1.2 ± 0.6 | 1.1 ± 0.5 | 0.054 |
| Type of drug-eluting stent | |||||||
| Cypher sirolimus-eluting stent | 26 (43%) | 5 (19%) | 75 (78%) | <0.001 | 106 (57.6%) | 4246 (44.5%) | <0.001 |
| Endeavor zotarolimus-eluting stent | 35 (57%) | 22 (82%) | 21 (22%) | 78 (42.4%) | 5297 (55.5%) | ||
| Lesions with predilatation | 77/101 (76%) | 28/40 (70%) | 121/159 (76%) | 0.70 | 226/300 (75.3%) | 9197/13492 (68.2%) | 0.008 |
| Periprocedure medication | |||||||
| Unfractionated heparin | 54/61 (89%) | 23/27 (85%) | 88/96 (92%) | 0.58 | 165/184 (89.7%) | 8437/9541 (88.4%) | 0.60 |
| Low-molecular-weight heparin | 5/61 (8%) | 0/27 (0%) | 3/96 (3%) | 0.15 | 8/184 (4.3%) | 464/9541 (4.9%) | 0.75 |
| Direct thrombin inhibitor | 4/61 (7%) | 3/27 (11%) | 5/96 (5%) | 0.55 | 12/184 (6.5%) | 734/9541 (7.7%) | 0.55 |
| Glycoprotein IIb/IIIa inhibitor | 23/61 (38%) | 6/27 (22%) | 20/96 (21%) | 0.056 | 49/184 (26.6%) | 1852/9541 (19.4%) | 0.014 |
| Aspirin | 60/61 (98%) | 27/27 (100%) | 95/96 (99%) | 0.79 | 182/184 (98.9%) | 9178/9541 (96.2%) | 0.055 |
| Clopidogrel | 59/61 (97%) | 27/27 (100%) | 92/96 (96%) | 0.56 | 178/184 (96.7%) | 9218/9541 (96.6%) | 0.93 |
| Post-procedure characteristics | |||||||
| Residual stenosis (%), mean ± SD | 2.7 ± 11.1 | 0.8 ± 2.7 | 1.7 ± 7.7 | 0.45 | 1.9 ± 8.6 | 2.0 ± 9.5 | 0.83 |
| Post-procedure TIMI Flow | 0.008 | 0.097 | |||||
| 0/1 | 0/101 (0%) | 0/40 (0%) | 0/159 (0%) | 0/300 (0%) | 72/13489 (0.5%) | ||
| 2 | 1/101 (1%) | 3/40 (8%) | 1/159 (1%) | 5/300 (1.7%) | 86/13489 (0.6%) | ||
| 3 | 100/101 (99%) | 37/40 (93%) | 158/159 (99%) | 295/300 (98.3%) | 13331/13489 (98.8%) | ||
| Lesion success † | 98/99 (99%) | 40/40 (100%) | 157/159 (99%) | 0.78 | 295/298 (99.0%) | 13230/13301 (99.5%) | 0.27 |
| Device success ‡ | 85/99 (86%) | 38/40 (95%) | 153/159 (96%) | 0.007 | 276/298 (92.6%) | 12771/13301 (96.0%) | 0.003 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
