This study compares clopidogrel vs aspirin monotherapy in patients with high bleeding risk and/or percutaneous coronary intervention complexity.
Key Points
Question
What is the comparative effectiveness of clopidogrel vs aspirin monotherapy in the chronic maintenance period after percutaneous coronary intervention (PCI) according to bleeding risk and procedural complexity?
Findings
In this post hoc analysis of the HOST-EXAM Extended clinical trial, clopidogrel monotherapy was associated with a reduction in thrombotic and bleeding risks compared with aspirin, regardless of the presence of high bleeding risk and/or PCI complexity.
Meaning
The study suggests that clopidogrel monotherapy can be a preferable choice over aspirin for secondary chronic maintenance therapy, irrespective of the bleeding risk and/or PCI complexity.
Abstract
Importance
Antiplatelet monotherapy in the chronic maintenance period for patients with high bleeding risk (HBR) and those who have undergone complex percutaneous coronary intervention (PCI) has not yet been explored.
Objective
To compare clopidogrel vs aspirin monotherapy in patients with HBR and/or PCI complexity.
Design, Setting, and Participants
This post hoc analysis of the multicenter HOST-EXAM Extended study, an open-label trial conducted across 37 sites in South Korea, enrolled patients from 2014 to 2018 with up to 5.9 years of follow-up. The analysis was conducted from February to November 2023. Patients who maintained dual antiplatelet therapy (DAPT) event-free for 6 to 18 months following PCI were included.
Interventions
Patients were randomized to receive either clopidogrel or aspirin in a 1:1 ratio. Those with sufficient data to assess HBR or complex PCI were analyzed.
Main Outcomes and Measures
Coprimary end points were thrombotic composite end point (cardiovascular death, nonfatal myocardial infarction, stroke, readmission due to acute coronary syndrome, and definite/probable stent thrombosis) and any bleeding (Bleeding Academic Research Consortium type 2 to 5).
Results
Of 3974 patients included (mean [SD] age, 63.4 [10.7] years; 2976 male [74.9%]), 866 had HBR (21.8%), and 849 underwent complex PCI (21.4%). Clopidogrel as compared with aspirin was associated with lower rates of thrombotic and bleeding events regardless of HBR and/or PCI complexity. For the thrombotic composite end point, the hazard ratio (HR) was 0.75 (95% CI, 0.53-1.04) among HBR vs 0.62 (95% CI, 0.48-0.80) among patients without HBR (P for interaction = 0.38) and 0.49 (95% CI, 0.32-0.77) among patients with complex PCI vs 0.74 (95% CI, 0.59-0.92) among patients with noncomplex PCI (P for interaction = 0.12). The reduction in bleeding by clopidogrel compared with aspirin was consistent among both patients with HBR (HR, 0.82; 95% CI, 0.56-1.21) and patients without HBR (HR, 0.58; 95% CI, 0.40-0.85; P for interaction = 0.20) and among patients undergoing complex PCI (HR, 0.79; 95% CI, 0.47-1.33) vs noncomplex PCI (HR, 0.68; 95% CI, 0.50-0.93; P for interaction = 0.62).
Conclusions and Relevance
In this study, in patients who experienced PCI and were event-free during 6 to 18 months of DAPT, the beneficial impact of clopidogrel monotherapy over aspirin monotherapy was consistent, regardless of bleeding risk and/or PCI complexity.
Trial Registration
ClinicalTrials.gov Identifier: NCT02044250
Introduction
Current guidelines recommend individual assessment of ischemic and bleeding risks when determining antiplatelet therapy and various combinations are recommended during the first year following percutaneous coronary intervention (PCI).1,2,3,4 However, we lack evidence and data regarding the optimal strategy for the chronic period across various clinical situations. Although aspirin monotherapy is recommended as the antiplatelet treatment of choice, recently published studies have highlighted the superiority of P2Y12 inhibitor monotherapy for long-term secondary prevention in patients post–PCI.1,2,3 An increasing number of patients at high bleeding risk (HBR) are undergoing PCI. HBR, defined by the Bleeding Academic Research Consortium (BARC) as features that elevate the risk of major bleeding (BARC score of 3 or 5 bleeding), has recently been delineated based on clinical evidence and expert consensus.4 Subsequent studies have validated this definition, reporting that patients with HBR show higher rates of not only bleeding but also other major adverse cardiovascular events.5,6 On the other hand, the number of patients undergoing complex procedures is on the rise due to evolution of devices and procedural techniques. Complex PCI is known to significantly increase the risk of future ischemic events, prompting proposals for various antithrombotic strategies to mitigate this risk.7 Patients with HBR and/or undergoing complex PCI constitute the most complicated and challenging subsets of patients to manage, with a considerable portion of the PCI population falling into these high-risk categories.7,8,9 Despite the need for lifelong antiplatelet monotherapy in these patients, to our knowledge, no study to date has specifically addressed this issue during the chronic maintenance period. Hence, in this post hoc analysis of the HOST-EXAM Extended study, we aimed to evaluate the comparative efficacy and safety of clopidogrel vs aspirin for chronic maintenance in patients characterized by HBR and/or PCI complexity.
Methods
Study Design and Patient Population
The design of the HOST-EXAM trial and the results of the end points from its extended follow-up study, the HOST-EXAM Extended, have been previously reported.1,2 Briefly, HOST-EXAM was a randomized, open-label trial conducted at 37 sites in South Korea. Patients who underwent PCI with drug-eluting stent and remained on dual antiplatelet therapy (DAPT) for 6 to 18 months without any ischemic or major bleeding events after PCI were enrolled (Figure 1). Patient race was determined by trial investigators and all enrolled patients were East Asian. They were randomly assigned to receive either clopidogrel (75 mg once daily) or aspirin (100 mg once daily) in a 1:1 ratio. HOST-EXAM published the primary analysis results at 24 months postrandomization.1 In the HOST-EXAM Extended study, the antiplatelet prescription was at the discretion of the treating physician during the posttrial period. We reported long-term results from the trial.2 The HOST-EXAM study was conducted in accordance with Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines.
Figure 1. CONSORT Study Flow Diagram.
Patients who maintained dual antiplatelet therapy without ischemic or bleeding events during 6 to 18 months after percutaneous coronary intervention (PCI) with a drug-eluting stent were randomized to treatment arms, aspirin or clopidogrel. Coprimary end points, which were thrombotic composite end point (comprising cardiovascular death, nonfatal myocardial infarction, stroke, readmission due to acute coronary syndrome, and definite or probable stent thrombosis) and any bleeding (comprising Bleeding Academic Research Consortium type 2 or more), were analyzed in the per-protocol population stratified by the presence of high bleeding risk (HBR) or PCI complexity.
Definition of HBR and Complex PCI
Patients were designated as HBR if they met at least 1 major or 2 minor criteria from the HBR–Academic Research Consortium (ARC) definition according to the gathered data.4 Complex PCI was defined by the presence of at least 1 of the following features: 3 or more stents implanted, 3 or more lesions treated, bifurcation with 2 stents implanted, total stent length more than 60 mm, or chronic total occlusion. After excluding patients with insufficient data to assess HBR or complex PCI, they were categorized into distinct groups depending on the presence of HBR and PCI complexity: HBR, non–HBR, complex PCI, and noncomplex PCI.
Study Oversight and Ethical Considerations
The study was overseen by the Seoul National University Hospital Clinical Trial Center and Medical Research Collaborating Center. An independent clinical event committee, blinded to trial-group assignments, adjudicated all events. Patient vital status was cross-checked through the National Health Insurance Service System of South Korea and the South Korea National Statistics System. Institutional review board at each participating site granted protocol approval. All patients provided written informed consent at enrollment. The study was conducted in accordance with the International Council for Harmonization Good Clinical Practice Guideline standards and the Declaration of Helsinki principles.
Outcome Evaluation
The coprimary end points were thrombotic composite end point (defined as cardiac death, nonfatal myocardial infarction, ischemic stroke, readmission due to acute coronary syndrome, and definite or probable stent thrombosis) and any bleeding (defined as BARC bleeding type 2 or higher). Secondary end points were analyzed, including individual components of the coprimary end points and revascularization.
Statistical Analysis
The analysis included patients who maintained their allocated single antiplatelet therapy during the follow-up and those assessable for HBR and PCI complexity. Demographic, clinical, and procedural characteristics were summarized according to HBR and PCI complexity. Continuous variables and categorical variables were expressed as means with SD and absolute values with their proportions, respectively. Differences between variables were compared using Student t test or χ2 test, as per variable type. Cumulative incidences of the coprimary and secondary end points were compared across different groups and between treatment arms using Kaplan-Meier censoring estimates and log-rank test. Corresponding hazard ratios (HRs) with 95% CIs were derived via Cox proportional hazards models, including interaction testing for the presence of HBR and/or PCI complexity. The analyses were reiterated by stratifying patients based on the number of HBR criteria or the number of complex PCI features. Absolute risk differences and corresponding 95% CIs for the clinical end points were determined following the Greenwood formula. For sensitivity analyses, we compared clinical outcomes across different groups and treatment arms by assigning patients excluded from HBR or complex PCI assessment to the non–HBR and noncomplex PCI groups, respectively. All tests were 2-sided and P < .05 was considered statistically significant. Statistical tests were carried out using R version 4.2.1 (The R Project for Statistical Computing).
Results
Baseline Characteristics
Of the 5438 patients initially randomized in the HOST-EXAM trial, 4717 were included in the per-protocol population after excluding 9 patients with consent withdrawal, 94 with follow-up loss, and 618 with use of a different antiplatelet regimen during the nonrandomized extension period. For this post hoc analysis, a total of 3974 patients were analyzed after excluding 743 patients with insufficient information to assess HBR and PCI complexity from the per-protocol population. Among the population, 2064 were in the clopidogrel arm (51.9%) and 1910 in the aspirin arm (48.1%), and 866 had HBR (21.8%) (468 vs 398 in the clopidogrel and aspirin arms, respectively), while 849 received complex PCI (21.4%) (441 vs 408 in the clopidogrel and aspirin arms, respectively). Baseline characteristics were similarly balanced between those randomized to clopidogrel vs aspirin for both HBR and non–HBR groups (Table 1). Compared with the non–HBR group, the HBR group was older and had a higher prevalence of chronic kidney disease, previous cerebrovascular accident, and lower mean hemoglobin levels, consistent with the attributes that make up the HBR-ARC criteria. Moreover, patients with HBR were more likely to be female and had a higher prevalence of hypertension, diabetes, and dyslipidemia (eTable 1 in Supplement 1). The prevalence of the individual HBR criterion is detailed in eTable 2 in Supplement 1.
Table 1. Baseline Patient Characteristics According to High Bleeding Risk Status (HBR) and Percutaneous Coronary Intervention (PCI) Complexity.
| Characteristic | No. (%) | |||
| Non-HBR | HBR | |||
| Clopidogrel arm (n = 1596) | Aspirin arm (n = 1512) | Clopidogrel arm (n = 468) | Aspirin arm (n = 398) | |
| Age, y, mean (SD) | 60.84 (9.68) | 60.81 (9.63) | 72.71 (9.09) | 72.67 (9.14) |
| BMI,a mean (SD) | 25.10 (3.10) | 24.97 (3.39) | 24.15 (3.16) | 23.96 (3.21) |
| Sex | ||||
| Female | 327 (20.5) | 309 (20.4) | 200 (42.7) | 162 (40.7) |
| Male | 1269 (79.5) | 1203 (79.6) | 268 (57.3) | 236 (59.3) |
| Diabetesb | 484 (30.3) | 469 (31.0) | 218 (46.6) | 185 (46.5) |
| Insulin-dependent diabetes | 24 (1.5) | 19 (1.3) | 19 (4.1) | 29 (7.3) |
| Hypertension | 934 (58.5) | 878 (58.1) | 338 (72.2) | 286 (71.9) |
| Dyslipidemia | 1096 (68.7) | 1059 (70.0) | 346 (73.9) | 296 (74.4) |
| Current smoker | 356 (22.3) | 379 (25.1) | 44 (9.4) | 53 (13.3) |
| Chronic kidney disease | 79 (4.9) | 78 (5.2) | 224 (47.9) | 185 (46.5) |
| Previous myocardial infarction | 270 (16.9) | 242 (16.0) | 92 (19.7) | 68 (17.1) |
| Previous cerebrovascular accident | 47 (2.9) | 49 (3.2) | 48 (10.3) | 55 (13.8) |
| Previous peripheral artery disease | 11 (0.7) | 12 (0.8) | 9 (1.9) | 8 (2.0) |
| Previous heart failure | 39 (2.4) | 53 (3.5) | 34 (7.3) | 30 (7.5) |
| Hemoglobin, g/dL, mean (SD) | 14.21 (1.31) | 14.21 (1.28) | 12.03 (1.77) | 12.02 (1.69) |
| Clinical indication of PCI | ||||
| Silent ischemia | 30 (1.9) | 42 (2.8) | 9 (1.9) | 14 (3.5) |
| Stable angina | 393 (24.6) | 385 (25.5) | 136 (29.1) | 115 (28.9) |
| Unstable angina | 558 (35.0) | 491 (32.5) | 171 (36.5) | 135 (33.9) |
| NSTEMI | 315 (19.7) | 303 (20.0) | 94 (20.1) | 79 (19.8) |
| STEMI | 300 (18.8) | 291 (19.2) | 58 (12.4) | 55 (13.8) |
| Angiographic data | ||||
| Extent of CAD | ||||
| 1-Vessel disease | 855 (53.6) | 795 (52.6) | 191 (40.8) | 153 (38.4) |
| 2-Vessel disease | 480 (30.1) | 461 (30.5) | 166 (35.5) | 139 (34.9) |
| 3-Vessel disease | 261 (16.4) | 256 (16.9) | 111 (23.7) | 106 (26.6) |
| Left main disease | 70 (4.4) | 62 (4.1) | 40 (8.5) | 33 (8.3) |
| PCI for bifurcation lesion | 174 (10.9) | 141 (9.3) | 54 (11.5) | 51 (12.8) |
| 2-Stenting for bifurcation PCI | 26 (1.6) | 18 (1.2) | 9 (1.9) | 12 (3.0) |
| PCI for CTO lesion | 146 (9.1) | 140 (9.3) | 39 (8.3) | 44 (11.1) |
| No. of treated lesions, mean (SD) | 1.29 (0.57) | 1.27 (0.54) | 1.39 (0.65) | 1.39 (0.65) |
| Mean diameter of implanted stents, mm, mean (SD) | 3.11 (0.43) | 3.12 (0.43) | 3.01 (0.40) | 3.02 (0.40) |
| Minimum diameter of implanted stents, mm, mean (SD) | 3.03 (0.46) | 3.05 (0.46) | 2.91 (0.42) | 2.93 (0.42) |
| Total length of implanted stents, mm, mean (SD) | 34.60 (22.04) | 33.99 (21.64) | 41.90 (30.52) | 39.05 (26.40) |
| Total No. of implanted stents, mean (SD) | 1.43 (0.76) | 1.41 (0.75) | 1.66 (1.04) | 1.60 (0.89) |
| Baseline medication | ||||
| β-Blockers | 809 (50.7) | 757 (50.1) | 231 (49.4) | 190 (47.7) |
| Angiotensin-converting enzyme inhibitors | 271 (17.0) | 221 (14.6) | 54 (11.5) | 43 (10.8) |
| Angiotensin II receptor blockers | 532 (33.3) | 488 (32.3) | 199 (42.5) | 162 (40.7) |
| Calcium-channel blockers | 417 (26.1) | 369 (24.4) | 142 (30.3) | 117 (29.4) |
| Diuretics | 134 (8.4) | 130 (8.6) | 63 (13.5) | 90 (22.6) |
| Nitrates | 150 (9.4) | 125 (8.3) | 49 (10.5) | 28 (7.0) |
| Statins | 1355 (84.9) | 1293 (85.5) | 394 (84.2) | 332 (83.4) |
| Proton pump inhibitors | 175 (11.0) | 173 (11.4) | 57 (12.2) | 58 (14.6) |
| Characteristic | Noncomplex PCI | Complex PCI | ||
| Clopidogrel arm (n = 1623) | Aspirin arm (n = 1502) | Clopidogrel arm (n = 441) | Aspirin arm (n = 408) | |
| Age, y, mean (SD) | 63.40 (10.84) | 63.08 (10.65) | 64.02 (10.48) | 64.00 (10.74) |
| BMI,a mean (SD) | 24.84 (3.17) | 24.81 (3.44) | 25.07 (3.01) | 24.62 (3.17) |
| Sex | ||||
| Female | 422 (26.0) | 382 (25.4) | 105 (23.8) | 89 (21.8) |
| Male | 1201 (74.0) | 1120 (74.6) | 336 (76.2) | 319 (78.2) |
| Diabetesb | 511 (31.5) | 504 (33.6) | 191 (43.3) | 150 (36.8) |
| Insulin-dependent diabetes | 32 (2.0) | 34 (2.3) | 11 (2.5) | 14 (3.4) |
| Hypertension | 1000 (61.6) | 910 (60.6) | 272 (61.7) | 254 (62.3) |
| Dyslipidemia | 1108 (68.3) | 1043 (69.4) | 334 (75.7) | 312 (76.5) |
| Current smoker | 319 (19.7) | 344 (22.9) | 81 (18.4) | 88 (21.6) |
| Chronic kidney disease | 228 (14.0) | 196 (13.0) | 75 (17.0) | 67 (16.4) |
| Previous myocardial infarction | 262 (16.1) | 240 (16.0) | 100 (22.7) | 70 (17.2) |
| Previous cerebrovascular accident | 71 (4.4) | 75 (5.0) | 24 (5.4) | 29 (7.1) |
| Previous peripheral artery disease | 18 (1.1) | 17 (1.1) | 2 (0.5) | 3 (0.7) |
| Previous heart failure | 60 (3.7) | 59 (3.9) | 13 (2.9) | 24 (5.9) |
| Hemoglobin, g/dL, mean (SD) | 13.78 (1.66) | 13.82 (1.59) | 13.62 (1.73) | 13.59 (1.74) |
| Clinical indication of PCI | ||||
| Silent ischemia | 20 (1.2) | 35 (2.3) | 19 (4.3) | 21 (5.1) |
| Stable angina | 426 (26.2) | 410 (27.3) | 103 (23.4) | 90 (22.1) |
| Unstable angina | 586 (36.1) | 491 (32.7) | 143 (32.4) | 135 (33.1) |
| NSTEMI | 312 (19.2) | 299 (19.9) | 97 (22.0) | 83 (20.3) |
| STEMI | 279 (17.2) | 267 (17.8) | 79 (17.9) | 79 (19.4) |
| Angiographic data | ||||
| Extent of CAD | ||||
| 1-Vessel disease | 943 (58.1) | 861 (57.3) | 103 (23.4) | 87 (21.3) |
| 2-Vessel disease | 478 (29.5) | 424 (28.2) | 168 (38.1) | 176 (43.1) |
| 3-Vessel disease | 202 (12.4) | 217 (14.4) | 170 (38.5) | 145 (35.5) |
| Left main disease | 73 (4.5) | 53 (3.5) | 37 (8.4) | 42 (10.3) |
| PCI for bifurcation lesion | 142 (8.7) | 115 (7.7) | 86 (19.5) | 77 (18.9) |
| 2-Stenting for bifurcation PCI | 0 (0.0) | 0 (0.0) | 35 (7.9) | 30 (7.4) |
| PCI for CTO lesion | 0 (0.0) | 0 (0.0) | 185 (42.0) | 184 (45.1) |
| No. of treated lesions, mean (SD) | 1.15 (0.35) | 1.13 (0.34) | 1.93 (0.84) | 1.88 (0.80) |
| Mean diameter of implanted stents, mm, mean (SD) | 3.11 (0.44) | 3.12 (0.44) | 2.99 (0.34) | 3.03 (0.37) |
| Minimum diameter of implanted stents, mm, mean (SD) | 3.07 (0.46) | 3.08 (0.46) | 2.78 (0.34) | 2.83 (0.40) |
| Total length of implanted stents, mm, mean (SD) | 28.09 (11.73) | 27.27 (11.40) | 66.32 (33.68) | 63.64 (30.26) |
| Total No. of implanted stents, mean (SD) | 1.21 (0.41) | 1.19 (0.39) | 2.48 (1.19) | 2.39 (1.08) |
| Baseline medication | ||||
| β-Blockers | 802 (49.4) | 720 (47.9) | 238 (54.0) | 227 (55.6) |
| Angiotensin-converting enzyme inhibitors | 265 (16.3) | 208 (13.8) | 60 (13.6) | 56 (13.7) |
| Angiotensin II receptor blockers | 554 (34.1) | 518 (34.5) | 177 (40.1) | 132 (32.4) |
| Calcium-channel blockers | 441 (27.2) | 384 (25.6) | 118 (26.8) | 102 (25.0) |
| Diuretics | 148 (9.1) | 167 (11.1) | 49 (11.1) | 53 (13.0) |
| Nitrates | 157 (9.7) | 123 (8.2) | 42 (9.5) | 30 (7.4) |
| Statins | 1362 (83.9) | 1265 (84.2) | 387 (87.8) | 360 (88.2) |
| Proton pump inhibitors | 181 (11.2) | 172 (11.5) | 51 (11.6) | 59 (14.5) |
Abbreviations: BMI, body mass index; CAD, coronary artery disease; CTO, chronic total occlusion; NSTEMI, non–ST elevation myocardial infarction; STEMI, ST elevation myocardial infarction.
SI conversion factor: To convert hemoglobin from g/dL to g/L, multiply by 10.
Calculated as weight in kilograms divided by height in meters squared.
Diabetes was defined as any type of diabetes.
Regarding PCI complexity, baseline characteristics were evenly distributed between the clopidogrel and aspirin arms in both complex and noncomplex PCI groups (Table 1). The complex PCI group, when compared with their noncomplex counterparts, exhibited a higher prevalence of multivessel disease, 2-stenting for bifurcation lesions, and chronic total occlusions, as well as increased number of treated lesions and stents, and longer stent lengths. The patients with complex PCI also had a higher prevalence of diabetes, dyslipidemia, and chronic kidney disease, and displayed lower hemoglobin levels (eTable 3 in Supplement 1). The distribution of each complex PCI feature is outlined in eTable 4 in Supplement 1.
Association of HBR Status and PCI Complexity on Clinical Outcomes
Over the follow-up period, the coprimary end points occurred more frequently in the HBR group than the non–HBR group (eFigure 1 in Supplement 1). However, no statistical difference was found in the incidence of the coprimary end points between the complex and noncomplex PCI groups (eFigure 1 in Supplement 1). Comparisons of the coprimary end points using the intention-to-treat population also showed consistent results (eFigure 2 in Supplement 1). Additionally, the risk of coprimary end points was correlated with the number of HBR criteria, while this linear trend was not observed with the number of complex PCI features (eFigure 3 in Supplement 1).
Comparison of Clopidogrel vs Aspirin According to HBR Status and PCI Complexity
Among the study population, clopidogrel monotherapy exhibited lower risks compared with aspirin monotherapy for the coprimary end points (eFigure 4 in Supplement 1). No significant interactions were found between the treatment arms and the presence of HBR or PCI complexity for the coprimary end points and individual components of the composite end points or revascularization (P interaction for all > .05; Figure 2 and Table 2). These results were consistent in the analyses with the intention-to-treat population (eFigure 5 in Supplement 1). Causes of death were specified in eTable 5 in Supplement 1.
Figure 2. Risk of Coprimary End Points by Treatment Arms in Relation to High Bleeding Risk (HBR) or Complex Percutaneous Coronary Intervention (PCI).
Risks of thrombotic composite end point by HBR (A) and complex PCI (B), and those of any bleeding by high bleeding risk (C) and complex PCI (D) by treatment arms. ASA indicates aspirin; CPD, clopidogrel.
Table 2. Clinical Outcomes According to the High Bleeding Risk (HBR) or Percutaneous Coronary Intervention (PCI) Complexity Status.
| Characteristica | No. (%) | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Non-HBR | HBR | P value for interaction | |||||||
| CPD | ASA | HR (95% CI)b | P value | CPD | ASA | HR (95% CI)b | P value | ||
| No. | 1596 | 1512 | NA | NA | 468 | 398 | NA | NA | NA |
| Thrombotic composite end pointc | 102 (6.8) | 153 (11.0) | 0.62 (0.48-0.80) | <.001 | 67 (16.4) | 73 (21.6) | 0.75 (0.53-1.04) | .08 | .38 |
| Any bleeding (BARC type ≥2) | 44 (2.9) | 71 (5.1) | 0.58 (0.40-0.85) | .01 | 52 (12.7) | 53 (15.3) | 0.82 (0.56-1.21) | .32 | .20 |
| All-cause death | 59 (4.0) | 44 (3.3) | 1.27 (0.86-1.88) | .23 | 76 (17.3) | 73 (20.9) | 0.89 (0.64-1.22) | .46 | .16 |
| Cardiovascular death | 22 (1.5) | 20 (1.5) | 1.04 (0.57-1.91) | .89 | 39 (9.3) | 36 (11.5) | 0.92 (0.58-1.45) | .72 | .74 |
| Noncardiovascular death | 37 (2.6) | 24 (1.8) | 1.46 (0.87-2.44) | .15 | 37 (8.7) | 37 (10.6) | 0.85 (0.54-1.35) | .50 | .12 |
| Nonfatal myocardial infarction | 25 (1.6) | 37 (2.4) | 0.64 (0.38-1.06) | .08 | 7 (1.5) | 9 (2.3) | 0.66 (0.25-1.77) | .41 | .95 |
| Stroke | 19 (1.2) | 33 (2.2) | 0.54 (0.31-0.95) | .03 | 13 (2.8) | 22 (5.5) | 0.49 (0.25-0.98) | .04 | .82 |
| Ischemic stroke | 14 (0.9) | 27 (1.8) | 0.49 (0.26-0.93) | .03 | 9 (1.9) | 9 (2.3) | 0.83 (0.33-2.09) | .69 | .36 |
| Hemorrhagic stroke | 6 (0.4) | 6 (0.4) | 0.94 (0.30-2.92) | .92 | 4 (0.9) | 13 (3.3) | 0.26 (0.08-0.79) | .02 | .11 |
| Readmission due to ACS | 72 (4.5) | 113 (7.5) | 0.60 (0.44-0.80) | <.001 | 23 (4.9) | 35 (8.8) | 0.54 (0.32-0.92) | .02 | .77 |
| Major bleeding (BARC type ≥3) | 25 (1.6) | 40 (2.6) | 0.59 (0.36-0.97) | .04 | 32 (6.8) | 39 (9.8) | 0.69 (0.43-1.10) | .12 | .66 |
| Any revascularization | 77 (4.8) | 101 (6.7) | 0.72 (0.53-0.96) | .03 | 26 (5.6) | 29 (7.3) | 0.76 (0.45-1.28) | .30 | .86 |
| Target lesion revascularization | 30 (1.9) | 41 (2.7) | 0.69 (0.43-1.11) | .12 | 11 (2.4) | 14 (3.5) | 0.66 (0.30-1.46) | .31 | .94 |
| Target vessel revascularization | 45 (2.8) | 62 (4.1) | 0.68 (0.47-1.00) | .05 | 16 (3.4) | 18 (4.5) | 0.75 (0.38-1.48) | .41 | .81 |
| Definite or probable stent thrombosis | 7 (0.4) | 10 (0.7) | 0.66 (0.25-1.75) | .41 | 1 (0.2) | 4 (1.2) | 0.21 (0.02-1.89) | .17 | .35 |
| Characteristic | Noncomplex PCI | Complex PCI | P value for interaction | ||||||
| CPD | ASA | HR (95% CI)b | P value | CPD | ASA | HR (95% CI)b | P value | ||
| No. | 1623 | 1502 | NA | NA | 441 | 408 | NA | NA | NA |
| Thrombotic composite end pointc | 139 (9.2) | 172 (12.7) | 0.74 (0.59-0.92) | .01 | 30 (7.7) | 54 (14.7) | 0.49 (0.32-0.77) | .002 | .12 |
| Any bleeding (BARC type ≥2) | 69 (4.6) | 93 (6.8) | 0.68 (0.50-0.93) | .02 | 27 (6.5) | 31 (8.2) | 0.79 (0.47-1.33) | .38 | .62 |
| All-cause death | 109 (7.2) | 84 (6.4) | 1.21 (0.91-1.61) | .19 | 26 (6.4) | 33 (8.9) | 0.72 (0.43-1.21) | .22 | .09 |
| Cardiovascular death | 52 (3.5) | 42 (3.3) | 1.15 (0.77-1.73) | .50 | 9 (2.2) | 14 (3.9) | 0.59 (0.26-1.36) | .22 | .16 |
| Noncardiovascular death | 57 (3.8) | 42 (3.2) | 1.26 (0.85-1.88) | .25 | 17 (4.3) | 19 (5.1) | 0.82 (0.43-1.58) | .55 | .27 |
| Nonfatal myocardial infarction | 26 (1.6) | 34 (2.3) | 0.70 (0.42-1.17) | .18 | 6 (1.4) | 12 (2.9) | 0.46 (0.17-1.22) | .12 | .44 |
| Stroke | 24 (1.5) | 40 (2.7) | 0.55 (0.33-0.91) | .02 | 8 (1.8) | 15 (3.7) | 0.49 (0.21-1.15) | .10 | .82 |
| Ischemic stroke | 16 (1.0) | 26 (1.7) | 0.56 (0.30-1.05) | .07 | 7 (1.6) | 10 (2.5) | 0.64 (0.24-1.68) | .37 | .83 |
| Hemorrhagic stroke | 8 (0.5) | 14 (0.9) | 0.52 (0.22-1.25) | .15 | 2 (0.5) | 5 (1.2) | 0.37 (0.07-1.89) | .23 | .71 |
| Readmission due to ACS | 78 (4.8) | 115 (7.7) | 0.62 (0.46-0.82) | .001 | 17 (3.9) | 33 (8.1) | 0.47 (0.26-0.84) | .01 | .39 |
| Major bleeding (BARC type ≥3) | 40 (2.5) | 61 (4.1) | 0.60 (0.40-0.90) | .01 | 17 (3.9) | 18 (4.4) | 0.87 (0.45-1.70) | .69 | .35 |
| Any revascularization | 83 (5.1) | 99 (6.6) | 0.77 (0.58-1.03) | .08 | 20 (4.5) | 31 (7.6) | 0.59 (0.33-1.03) | .06 | .40 |
| Target lesion revascularization | 31 (1.9) | 44 (2.9) | 0.65 (0.41-1.03) | .07 | 10 (2.3) | 11 (2.7) | 0.84 (0.36-1.98) | .69 | .60 |
| Target vessel revascularization | 49 (3.0) | 58 (3.9) | 0.78 (0.53-1.14) | .20 | 12 (2.7) | 22 (5.4) | 0.50 (0.25-1.01) | .05 | .27 |
| Definite or probable stent thrombosis | 8 (0.5) | 13 (0.9) | 0.57 (0.24-1.38) | .21 | 0 (0.0) | 1 (0.3) | NA | NA | NA |
Abbreviations: ACS, acute coronary syndrome; ASA, aspirin; BARC, Bleeding Academic Research Consortium; CPD, clopidogrel; HR, hazard ratio; NA, not applicable.
Data are expressed as No. (cumulative incidence). Clinical end points were assessed in the per-protocol population at 5.5 years after randomization.
The 95% CIs have not been adjusted for multiple comparisons, and therefore inferences from these intervals may not be reproducible.
Composite of cardiac death, nonfatal myocardial infarction, ischemic stroke, readmission due to ACS, and definite or probable stent thrombosis.
The total population was classified into 4 groups based on the presence of HBR and PCI complexity: 2486 patients in the non–HBR and noncomplex PCI group (62.6%), 622 in the non–HBR and complex PCI group (15.7%), 639 in the HBR and noncomplex PCI group (16.1%), and 227 in the HBR and complex PCI group (5.7%). In all 4 groups, clopidogrel monotherapy compared with aspirin monotherapy was associated with lower risk of the coprimary end points. For the thrombotic composite end point, the HR was 0.64 (95% CI, 0.49-0.85) among the non–HBR and noncomplex PCI group, 0.53 (95% CI, 0.30-0.95) among the non–HBR and complex PCI group, 0.88 (95% CI, 0.60-1.29) among the HBR and noncomplex PCI group, and 0.46 (95% CI, 0.23-0.91) among the HBR and complex PCI group (P for interaction = .28). The reduction in bleeding by clopidogrel compared with aspirin was consistent across the 4 groups with an HR of 0.50 (95% CI, 0.32-0.77) in the non–HBR and noncomplex PCI group, 0.98 (95% CI, 0.45-2.14) in the non–HBR and complex PCI group, 0.91 (95% CI, 0.57-1.44) in the HBR and noncomplex PCI group, and 0.69 (95% CI, 0.35-1.38) in the HBR and complex PCI group (P for interaction = .23; Table 3). However, the greatest risk reduction of clopidogrel monotherapy occurred in the group with both HBR and complex PCI. The impact of clopidogrel compared with aspirin monotherapy for secondary end points across 4 groups is described in eTable 6 in Supplement 1.
Table 3. Impact of Antiplatelet Monotherapy on Coprimary End Points Based on High Bleeding Risk (HBR) and/or Percutaneous Coronary Intervention (PCI) Complexity.
| End pointc | Clopidogrel, No./total No. (%) | Aspirin, No./total No. (%) | HR (95% CI)a | P value | ARD (95% CI) | P value for interaction |
|---|---|---|---|---|---|---|
| Thrombotic composite end point b | ||||||
| Non–HBR and noncomplex PCI | 84/1269 (7.0) | 123/1217 (11.1) | 0.64 (0.49 to 0.85) | .002 | 4.1 (1.7 to 6.5) | .28 |
| Non–HBR and complex PCI | 18/327 (6.3) | 30/295 (10.9) | 0.53 (0.30 to 0.95) | .03 | 4.6 (−0.1 to 9.3) | |
| HBR and noncomplex PCI | 55/354 (17.8) | 49/285 (20.1) | 0.88 (0.60 to 1.29) | .50 | 2.3 (−4.6 to 9.2) | |
| HBR and complex PCI | 12/114 (11.8) | 24/113 (25.6) | 0.46 (0.23 to 0.91) | .03 | 13.8 (2.6 to 25.0) | |
| Any bleeding (BARC type ≥2) | ||||||
| Non–HBR and noncomplex PCI | 31/1269 (2.6) | 59/1217 (5.2) | 0.50 (0.32 to 0.77) | .002 | 2.6 (0.8 to 4.4) | .23 |
| Non–HBR and complex PCI | 13/327 (4.1) | 12/295 (4.3) | 0.98 (0.45 to 2.14) | .95 | 0.2 (−2.9 to 3.3) | |
| HBR and noncomplex PCI | 38/354 (12.4) | 34/285 (13.7) | 0.91 (0.57 to 1.44) | .68 | 1.3 (−4.4 to 7.0) | |
| HBR and complex PCI | 14/114 (13.5) | 19/113 (19.1) | 0.69 (0.35 to 1.38) | .30 | 5.6 (−4.8 to 16.0) | |
Abbreviations: ACS, acute coronary syndrome; ARD, absolute risk difference; BARC, Bleeding Academic Research Consortium; HR, hazard ratio.
The 95% CIs have not been adjusted for multiple comparisons, and therefore, inferences from these intervals may not be reproducible.
Composite of cardiac death, nonfatal myocardial infarction, ischemic stroke, readmission due to ACS, and definite or probable stent thrombosis.
Data of treatment arms are expressed as No. (cumulative incidence). Clinical end points were assessed in the per-protocol population at 5.5 years after randomization.
Sensitivity Analysis
Baseline characteristics and clinical outcomes of patients excluded from the analysis due to insufficient data to assess HBR or complex PCI (743 patients) were compared with those included (3974 patients). Most baseline characteristics were well balanced between the populations, except for insulin-dependent diabetes, chronic kidney disease, previous myocardial infarction, previous cerebrovascular accident, previous peripheral artery disease, and the stent diameter (eTable 7 in Supplement 1). Despite these differences, they did not lead to differences in clinical outcomes or influence the impact of treatment arms (eFigure 6 in Supplement 1). Among the excluded patients, the predominantly unmeasured variables were glomerular filtration rate and hemoglobin (eFigure 7 in Supplement 1). Assigning patients excluded from HBR or complex PCI assessment to the respective non–HBR and noncomplex PCI groups yielded consistent results (eFigures 8 and 9 in Supplement 1).
Discussion
In this post hoc analysis of the HOST-EXAM Extended study with a median follow-up of 5.9 years postrandomization, we investigated the long-term prognosis of patients with HBR and/or those who underwent complex PCI. We also compared the outcomes of clopidogrel vs aspirin monotherapy in these patients. It is important to remember that the HOST-EXAM trial population consisted of patients without adverse clinical events for 6 to 18 months post-PCI while receiving DAPT, and thus, were stabilized patients requiring long-term antiplatelet monotherapy. The main findings can be summarized as follows: (1) patients with HBR showed higher long-term rates of both coprimary thrombotic and bleeding end points compared with non–HBR counterparts; (2) in contrast, the rates of the thrombotic composite end point and any bleeding were comparable irrespective of PCI complexity; (3) no significant interactions were found between the treatment arms (advantageous long-term impact of clopidogrel monotherapy over aspirin monotherapy) and HBR and/or PCI complexity; and (4) the beneficial impact of clopidogrel monotherapy was more pronounced in patients at higher risk and largest in those with the highest risk; patients with HBR who underwent complex PCI.
HBR and Procedural Complexity During the Chronic Maintenance Period After PCI
Among the various scores to evaluate bleeding risk, the most recently developed HBR-ARC criteria have been validated to predict adverse outcomes, influenced not only by its presence or absence, but by the increasing number of fulfilled criteria.4,5,8,10,11 However, most studies have focused on the risk during the first year after PCI or have used outdated patient groups to derive long-term outcomes, complicating the application to contemporary patients.12 Our results provide additional support for HBR as a long-term prognostic factor. Notably, even in our study population who were those without any clinical events for 6 to 18 months post–PCI, patients with HBR showed an increased risk of adverse clinical events, not only bleeding, but ischemic events during the long-term follow-up period. A risk continuum was also observed with the number of HBR criteria met for both bleeding and ischemic outcomes, confirming the importance of HBR on long-term outcomes. This simultaneous elevation of bleeding and thrombotic risks in patients with HBR can be partially explained by the overlap of clinical factors linked to both event types and by the heightened thrombotic risk following poor antiplatelet adherence after minor bleeding events.2,13
Though a universally accepted definition for complex PCI does not yet exist,14 previous studies have consistently reported an elevated ischemic risk within the first year following PCI in those undergoing complex PCI, highlighting the necessity for an extended duration of DAPT in these patients.7,15 In our study, complex PCI was defined according to the European Society of Cardiology guideline–endorsed criteria.16 We demonstrated that the long-term risk was not so elevated in those who underwent complex PCI compared with noncomplex PCI group, nor was it related to the degree of complex PCI (represented as the number of complex PCI feature), in patients who had a stable course over the first year after PCI. These results are consistent with previous studies that reported the elevated thrombotic risk associated with complex PCI was confined to the first year post–PCI13,14,17 and that long-term risk was more determined by clinical rather than procedural factors.13 Notably, our results detected a marginal increase in the bleeding risk in patients who underwent complex PCI, suggesting potential hazards associated with strategies aimed at reducing ischemic risk beyond 1 year of stabilization, as they may lead to further bleeding events.
Antiplatelet Therapy According to HBR and Procedural Complexity
Recent evidence has shown that the P2Y12 inhibitor may be superior to aspirin monotherapy for secondary prevention following PCI. With an emphasis on intensive lipid-lowering and guideline-directed medical therapy for secondary prevention, there seems to be a shift toward a new equilibrium, with diminishing concerns over ischemic risk relative to the past and increasing concerns over bleeding risk.18 Following the CAPRIE trial, which first reported the superiority of clopidogrel over aspirin in reducing ischemic events,19 the HOST-EXAM trial and the HOST-EXAM Extended study further demonstrated the superiority of clopidogrel in reducing the composite thrombotic and bleeding risk.1,2 A recent meta-analysis encompassing all these studies corroborated that P2Y12 inhibitors reduce ischemic risk without increasing bleeding risk compared with aspirin.3 In the current study, we aimed to verify whether the persistent benefit of clopidogrel over aspirin was consistent in high-risk patients who underwent PCI; those with HBR and/or undergoing complex PCI.
Overall, no significant interaction was found between treatment arms and risk groups, denoting that the beneficial impact of clopidogrel monotherapy was consistent regardless of HBR or PCI complexity. However, the magnitude of thrombotic benefit of clopidogrel appeared to be greater numerically in the complex PCI group, although statistically insignificant. Greater adherence to clopidogrel, as compared with aspirin, may have contributed to such numerical difference,2 but additional evidence from other studies will be required to confirm these findings.
Neither the impact of clopidogrel over aspirin on mortality across risk groups, nor the comparisons of specific causes of death between the treatment arms in any risk group, were statistically significant. Although the analysis may not be powered to detect statistical differences in mortality, these results may support the safe use of clopidogrel compared with aspirin, even in the presence of clinical and procedural risk factors.
Limitations
Our study has several limitations that warrant consideration. First, the study is a post hoc analysis, and as such, its findings must be considered primarily hypothesis generating. However, the narrow confidence intervals suggest a consistent beneficial impact of clopidogrel monotherapy over aspirin monotherapy for clinical end points in patients with HBR and/or complex PCI and underscore the net clinical advantage in these specific populations. Second, the definitions of HBR and complex PCI were not prespecified in the study protocol, as the HOST-EXAM trial was designed in 2014. Nevertheless, this post-hoc analysis used well-established definitions of HBR and complex PCI.4,16 All patients not meeting HBR or complex PCI criteria were classified as non–HBR or noncomplex PCI, but again, these were not prespecified definitions. Third, this study was conducted in an East Asian population, which requires cautious interpretation when generalizing these findings to other ethnic groups. Fourth, certain criteria that constitute HBR were not studied because they were among the exclusion criteria of the HOST-EXAM trial or were not collected in the study case report form. This might have contributed to the observed lower incidence of major bleeding events (8.2% of major bleeding over median 5.9 years in this study) than the expected 4% per year or above, as defined by the HBR-ARC. Additionally, the absence of HBR criteria, such as glomerular filtration rate and hemoglobin in some patients, could have potentially introduced bias. Fifth, since this study is a subanalysis of an open-label trial, the medication prescriptions of nonblinded physicians could have influenced the clinical outcomes of the risk groups. Specifically, the higher use of statins in the complex PCI group compared with the noncomplex PCI group might have contributed to a lower clinical risk. Moreover, factors such as medication adherence, alterations in medication, and the use of additional therapies during the follow-up period could have affected the outcomes, but this information was not collected. Sixth, most of the interaction testing was underpowered, especially for the individual components of the composite end points or revascularization, and therefore should be interpreted with caution. Lastly, the study population consisted of stabilized patients who underwent and remained event-free after 6 to 18 months of DAPT. As such, the results of the current study may not directly be applicable to patients outside this DAPT duration.
Conclusions
In this study, patients who underwent PCI and were event-free during 6 to 18 months of DAPT, the beneficial impact of continued clopidogrel monotherapy over aspirin monotherapy was consistent regardless of high bleeding risk and/or PCI complexity.
eTable 1. Baseline patient characteristics of patients with high bleeding risk
eTable 2. High bleeding risk criteria
eTable 3. Baseline patient characteristics of patients undergone complex PCI
eTable 4. Complex PCI feature
eTable 5. Causes of Death by treatment arm according to high bleeding risk or complex PCI
eTable 6. Impact of clopidogrel versus aspirin in relation to the presence of high bleeding risk and/or complex PCI
eTable 7. Baseline patient characteristics of patients included and excluded in the analysis
eFigure 1. Cumulative incidence of co-primary endpoints according to high bleeding risk or complex PCI
eFigure 2. Cumulative incidence of co-primary endpoints according to high bleeding risk or complex PCI from intention-to-treat population
eFigure 3. Cumulative incidences of the clinical endpoints by number of HBR criteria or complex PCI feature
eFigure 4. Cumulative incidence of the clinical outcomes according to treatment arms
eFigure 5. Cumulative incidence of co-primary endpoints by treatment arms in relation to high bleeding risk or complex PCI among intention-to-treat population
eFigure 6. Cumulative incidence of co-primary endpoints between included population and excluded population
eFigure 7. Number and proportion of unmeasured data for high bleeding risk or complex PCI
eFigure 8. Cumulative incidence of co-primary endpoints according to high bleeding risk or complex PCI from sensitivity analysis
eFigure 9. Cumulative incidence of co-primary endpoints by treatment arms in relation to high bleeding risk or complex PCI from sensitivity analysis
Trial protocol
Data sharing statement
References
- 1.Koo BK, Kang J, Park KW, et al. ; HOST-EXAM investigators . Aspirin versus clopidogrel for chronic maintenance monotherapy after percutaneous coronary intervention (HOST-EXAM): an investigator-initiated, prospective, randomised, open-label, multicentre trial. Lancet. 2021;397(10293):2487-2496. doi: 10.1016/S0140-6736(21)01063-1 [DOI] [PubMed] [Google Scholar]
- 2.Kang J, Park KW, Lee H, et al. Aspirin versus clopidogrel for long-term maintenance monotherapy after percutaneous coronary intervention: the HOST-EXAM Extended Study. Circulation. 2023;147(2):108-117. doi: 10.1161/CIRCULATIONAHA.122.062770 [DOI] [PubMed] [Google Scholar]
- 3.Gragnano F, Cao D, Pirondini L, et al. ; PANTHER Collaboration . P2Y12 inhibitor or aspirin monotherapy for secondary prevention of coronary events. J Am Coll Cardiol. 2023;82(2):89-105. doi: 10.1016/j.jacc.2023.04.051 [DOI] [PubMed] [Google Scholar]
- 4.Urban P, Mehran R, Colleran R, et al. Defining high bleeding risk in patients undergoing percutaneous coronary intervention. Circulation. 2019;140(3):240-261. doi: 10.1161/CIRCULATIONAHA.119.040167 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cao D, Mehran R, Dangas G, et al. Validation of the academic research consortium high bleeding risk definition in contemporary PCI patients. J Am Coll Cardiol. 2020;75(21):2711-2722. doi: 10.1016/j.jacc.2020.03.070 [DOI] [PubMed] [Google Scholar]
- 6.Lee J, Kim MJ, Kim M, et al. ; TICAKOREA Study Investigators . Validation of academic research consortium for high bleeding risk definition in East-Asian patients. JACC Asia. 2023;3(3):390-399. doi: 10.1016/j.jacasi.2022.11.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Giustino G, Chieffo A, Palmerini T, et al. Efficacy and safety of dual antiplatelet therapy after complex PCI. J Am Coll Cardiol. 2016;68(17):1851-1864. doi: 10.1016/j.jacc.2016.07.760 [DOI] [PubMed] [Google Scholar]
- 8.Nakamura M, Kadota K, Nakao K, et al. High bleeding risk and clinical outcomes in East Asian patients undergoing percutaneous coronary intervention: the PENDULUM registry. EuroIntervention. 2021;16(14):1154-1162. doi: 10.4244/EIJ-D-20-00345 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Chiarito M, Kini A, Roumeliotis A, et al. Prevalence and impact of high bleeding risk in patients undergoing left main artery disease PCI. JACC Cardiovasc Interv. 2021;14(22):2447-2457. doi: 10.1016/j.jcin.2021.08.056 [DOI] [PubMed] [Google Scholar]
- 10.Corpataux N, Spirito A, Gragnano F, et al. Validation of high bleeding risk criteria and definition as proposed by the academic research consortium for high bleeding risk. Eur Heart J. 2020;41(38):3743-3749. doi: 10.1093/eurheartj/ehaa671 [DOI] [PubMed] [Google Scholar]
- 11.Ueki Y, Bär S, Losdat S, et al. Validation of the Academic Research Consortium for High Bleeding Risk (ARC-HBR) criteria in patients undergoing percutaneous coronary intervention and comparison with contemporary bleeding risk scores. EuroIntervention. 2020;16(5):371-379. doi: 10.4244/EIJ-D-20-00052 [DOI] [PubMed] [Google Scholar]
- 12.Natsuaki M, Morimoto T, Shiomi H, et al. Application of the Academic Research Consortium high bleeding risk criteria in an all-comers registry of percutaneous coronary intervention. Circ Cardiovasc Interv. 2019;12(11):e008307. doi: 10.1161/CIRCINTERVENTIONS.119.008307 [DOI] [PubMed] [Google Scholar]
- 13.Kang J, Park KW, Lee HS, et al. Relative impact of clinical risk versus procedural risk on clinical outcomes after percutaneous coronary intervention. Circ Cardiovasc Interv. 2021;14(2):e009642. doi: 10.1161/CIRCINTERVENTIONS.120.009642 [DOI] [PubMed] [Google Scholar]
- 14.Serruys PW, Takahashi K, Chichareon P, et al. Impact of long-term ticagrelor monotherapy following 1-month dual antiplatelet therapy in patients who underwent complex percutaneous coronary intervention: insights from the Global Leaders trial. Eur Heart J. 2019;40(31):2595-2604. doi: 10.1093/eurheartj/ehz453 [DOI] [PubMed] [Google Scholar]
- 15.Hwang D, Lim YH, Park KW, et al. ; HOST-RP-ACS investigators . Prasugrel dose de-escalation therapy after complex percutaneous coronary intervention in patients with acute coronary syndrome: a post hoc analysis from the HOST-REDUCE-POLYTECH-ACS Trial. JAMA Cardiol. 2022;7(4):418-426. doi: 10.1001/jamacardio.2022.0052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Neumann FJ, Sousa-Uva M, Ahlsson A, et al. ; ESC Scientific Document Group . 2018 ESC/EACTS guidelines on myocardial revascularization. Eur Heart J. 2019;40(2):87-165. doi: 10.1093/eurheartj/ehy394 [DOI] [PubMed] [Google Scholar]
- 17.Yeh RW, Kereiakes DJ, Steg PG, et al. ; DAPT Study Investigators . Lesion complexity and outcomes of extended dual antiplatelet therapy after percutaneous coronary intervention. J Am Coll Cardiol. 2017;70(18):2213-2223. doi: 10.1016/j.jacc.2017.09.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Capodanno D, Baber U, Bhatt DL, et al. P2Y12 inhibitor monotherapy in patients undergoing percutaneous coronary intervention. Nat Rev Cardiol. 2022;19(12):829-844. doi: 10.1038/s41569-022-00725-6 [DOI] [PubMed] [Google Scholar]
- 19.CAPRIE Steering Committee . A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). Lancet. 1996;348(9038):1329-1339. doi: 10.1016/S0140-6736(96)09457-3 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Baseline patient characteristics of patients with high bleeding risk
eTable 2. High bleeding risk criteria
eTable 3. Baseline patient characteristics of patients undergone complex PCI
eTable 4. Complex PCI feature
eTable 5. Causes of Death by treatment arm according to high bleeding risk or complex PCI
eTable 6. Impact of clopidogrel versus aspirin in relation to the presence of high bleeding risk and/or complex PCI
eTable 7. Baseline patient characteristics of patients included and excluded in the analysis
eFigure 1. Cumulative incidence of co-primary endpoints according to high bleeding risk or complex PCI
eFigure 2. Cumulative incidence of co-primary endpoints according to high bleeding risk or complex PCI from intention-to-treat population
eFigure 3. Cumulative incidences of the clinical endpoints by number of HBR criteria or complex PCI feature
eFigure 4. Cumulative incidence of the clinical outcomes according to treatment arms
eFigure 5. Cumulative incidence of co-primary endpoints by treatment arms in relation to high bleeding risk or complex PCI among intention-to-treat population
eFigure 6. Cumulative incidence of co-primary endpoints between included population and excluded population
eFigure 7. Number and proportion of unmeasured data for high bleeding risk or complex PCI
eFigure 8. Cumulative incidence of co-primary endpoints according to high bleeding risk or complex PCI from sensitivity analysis
eFigure 9. Cumulative incidence of co-primary endpoints by treatment arms in relation to high bleeding risk or complex PCI from sensitivity analysis
Trial protocol
Data sharing statement