Visual Abstract
Keywords: drug eluting stents, maintenance dialysis, prasugrel, ticagrelor myocardial infarction
Abstract
Background and objectives
Prasugrel and ticagrelor have superior efficacy compared with clopidogrel in moderate CKD but have not been studied in kidney failure. The study objective is to determine the effectiveness and safety of prasugrel and ticagrelor in kidney failure.
Design, setting, participants, & measurements
This retrospective cohort study used United States Renal Data System data from 2012 to 2015. We identified all patients on dialysis who received a drug-eluting stent and were alive at 90 days after stent implantation. Inverse probability–weighted Cox proportional hazard models were used. Weights were estimated with propensity scores for multiple treatments.
Results
This cohort included 6648 patients on clopidogrel, 621 on prasugrel, and 449 on ticagrelor. A total of 3279 primary composite (cardiovascular death, myocardial infarction, or stroke) and 2120 clinically relevant bleeding events were observed. The incidence of the primary composite outcome of cardiovascular death, myocardial infarction, or stroke at 12 months was similar across the three treatment groups. The absolute event rate in the unweighted cohort was 144 events per 100 patient-years for clopidogrel, 126 for prasugrel, and 161 for ticagrelor. For prasugrel versus clopidogrel, the weighted hazard ratio was 0.96 (95% confidence interval, 0.82 to 1.11; P=0.58). For ticagrelor versus clopidogrel, the hazard ratio was 1.00 (95% confidence interval, 0.83 to 1.20; P=0.98). A numerically higher incidence of clinically relevant bleeding was seen with prasugrel or ticagrelor compared with clopidogrel (weighted hazard ratio, 1.15; 95% confidence interval, 0.95 to 1.38 and weighted hazard ratio, 1.13; 95% confidence interval, 0.91 to 1.40, respectively).
Conclusions
Prasugrel or ticagrelor does not seem to be associated with significant benefits compared with clopidogrel in patients with kidney failure treated with drug-eluting stents.
Podcast
This article contains a podcast at https://www.asn-online.org/media/podcast/CJASN/2021_04_02_CJN12120720.mp3
Introduction
Patients who receive a drug-eluting stent (DES) require dual antiplatelet therapy with low-dose aspirin and a P2Y12 ADP receptor blocker, with clopidogrel being the standard of treatment for many years (1). The Study of Platelet Inhibition and Patient Outcomes (PLATO) and the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel-Thrombolysis in Myocardial Infarction 38 randomized trials established superiority of ticagrelor and prasugrel over clopidogrel for the prevention of cardiovascular death, myocardial infarction, or stroke in patients with acute coronary syndromes (2,3). Therefore, these newer antiplatelet agents are now preferred to clopidogrel, unless unavailable or contraindicated.
Although patients with CKD are at higher risk for adverse cardiovascular events (4), they are typically under-represented in randomized trials in coronary disease with underuse of most recent therapeutic advances (5). The Clopidogrel for the Reduction of Events During Observation study failed to confirm the overall benefit of long-term clopidogrel (beyond 1 month postpercutaneous coronary intervention) compared with placebo in the subgroup of participants with CKD and was the first trial to challenge the efficacy of clopidogrel in this population, raising the question of the antiplatelet agent of choice in these patients (6). Randomized data from the PLATO trial in 3237 patients with an estimated creatinine clearance <60 ml/min were reassuring, with ticagrelor being more effective than clopidogrel without any significant increase in major bleeding event rates, regardless of baseline kidney function (7). However, most patients had moderate CKD, with patients on dialysis excluded from this study.
Three cohort studies investigated the effectiveness and safety of ticagrelor and prasugrel stratified by kidney function using the SWEDEHEART, PROMETHEUS, and RENAMI-BleeMACS registries (8 –10). These three studies enrolled 210 patients on ticagrelor with an eGFR<30 ml/min per 1.73 m2 (8), 617 patients on prasugrel with an eGFR<60 ml/min per 1.73 m2 (9), or 86 patients with an eGFR<30 ml/min per 1.73 m2 on any antiplatelet agent (10), respectively. However, the number of patients on KRT was not available, and the clinical outcomes with the newer antiplatelet agents compared with clopidogrel were not reported for this subgroup. Therefore, there is a paucity of data for ticagrelor or prasugrel in patients with kidney failure.
We designed this retrospective cohort study to determine whether prasugrel or ticagrelor is associated with improved clinical outcomes compared with clopidogrel in patients with kidney failure on maintenance dialysis who are treated with DES.
Materials and Methods
This retrospective cohort study used United States Renal Data System data from 2012 to 2015 (11). The Partners Healthcare Institutional Review Board approved the study (2016P001613/BWH) and waived need for informed consent. This study was carried out in accordance with the Declaration of Helsinki. Restrictions apply to the availability of these data, which were used under license for this study, and therefore are not publicly available.
Study Population and Exposure
We identified all adult patients who were on hemodialysis or peritoneal dialysis on January 1, 2012 or who were started on hemodialysis or peritoneal dialysis after this date. We next identified all patients who received a DES on the basis of at least one inpatient claim using the procedural diagnostic code 36.07 or 027x34x (International Classification of Diseases, ninth or tenth revision, respectively) or the Current Procedural Terminology codes G0290, G0291, and C960x (12). We excluded (1) patients who had a coronary artery bypass grafting during the index hospitalization; (2) patients without at least 6 months of Medicare Part A and B coverage in the year prior to DES implantation, in order to ensure appropriate capture of comorbid conditions; (3) patients without at least 3 months of Part D enrollment post-DES implantation, in order to ensure appropriate capture of filled prescriptions; (4) patients who died in the first 3 months post-DES implantation; and (5) patients who had recovered kidney function or with unknown dialysis status at 3 months post-DES implantation.
Relevant baseline characteristics and medication prescriptions were retrieved for each patient from the institutional and physician claims datasets or Medicare Part D data (Supplemental Table 1). For baseline medication, the 90-day period post-DES implantation was examined. Of note, use of aspirin is not available in this dataset. Exposure (clopidogrel, ticagrelor, or prasugrel) was assessed in the 90 days post-DES implantation. The last agent prescribed in this time period was used. We also conducted an additional analysis using a 45-day window for exposure assessment.
Outcomes
The primary outcome was a composite of cardiovascular mortality, myocardial infarction, or stroke in the first year post-DES implantation. The secondary outcomes were the components of the primary outcome; a composite outcome of cardiovascular mortality, myocardial infarction, stroke, or revascularization in the first year post-DES implantation; revascularization (defined as percutaneous coronary intervention or coronary artery bypass grafting); all-cause mortality at 12 months; and clinically relevant bleeding, which was defined as any bleeding resulting in death or occurring at a critical site (intracranial, intraocular, retroperitoneal, intra-articular, pericardial, or airway) or any gastrointestinal, urinary tract, or gynecologic bleeding requiring hospitalization. Diagnostic codes for these outcomes in any position were considered and are shown in Supplemental Table 2.
The follow-up period (time 0) started 90 days after DES implantation to allow enough time for patients to fill a prescription. Patients were followed from day 90 post-DES implantation up to the date of death, kidney transplantation, loss of Medicare A and B coverage, or 12 months after DES implantation. A total of 3350 patients (43%) were followed up for <12 months due to loss of Medicare coverage (Parts A, B, or D). These patients were censored upon loss of Medicare coverage.
Statistical Analyses
For the main analysis, we examined exposure (prasugrel, ticagrelor, or clopidogrel) as a time-varying covariate (“as-treated” analysis), and we further censored patients upon loss of Part D Medicare coverage or at the last available antiplatelet prescription plus drug supply days. In case of temporary treatment interruption, a 90-day grace period was used during which patients were assumed to still be on the last drug filled. In case of more than one prescription of different antiplatelet agents for overlapping days, patients were censored on the day of overlap.
Inverse probability–weighted Cox proportional hazard models were used to examine the association between antiplatelet agents and the clinical outcomes. Weights were estimated with propensity scores for multiple treatments using generalized boosted models (13). The model included 27 parameters at baseline selected for their eventual association with cardiovascular or bleeding outcomes (shown in Table 1). Standardized mean differences were calculated to assess balance of the weighted cohorts, with a cutoff of ≤0.10 considered to indicate a negligible difference (14). Propensity score distribution per treatment group is shown in Supplemental Figure 1.
Table 1.
Baseline characteristics of patients on dialysis included in the United States Renal Data System survey who received a drug-eluting stent from 2012 to 2015 and had Parts A, B, and D Medicare coverage
| Characteristic | Clopidogrel | Prasugrel | Ticagrelor | Standardized Mean Differences (prior to Weighting) | Standardized Mean Differences (after Weighting) | ||
|---|---|---|---|---|---|---|---|
| Prasugrel versus clopidogrel | Ticagrelor versus clopidogrel | Prasugrel versus clopidogrel | Ticagrelor versus clopidogrel | ||||
| No. of patients | 6648 | 621 | 449 | ||||
| Follow-up, d | 170 (41–275) | 171 (39–275) | 93 (17–258) | ||||
| Demographics | |||||||
| Age | 64±11 | 60±11 | 64±12 | 0.37 | 0.05 | 0.03 | 0.01 |
| Men | 3634 (55%) | 357 (58%) | 226 (50%) | 0.06 | 0.09 | 0.03 | 0.06 |
| Black race | 1940 (29%) | 153 (25%) | 134 (30%) | 0.10 | 0.02 | 0.04 | 0.02 |
| Hemodialysis | 6168 (93%) | 550 (89%) | 406 (90%) | 0.16 | 0.09 | 0.02 | 0.04 |
| Dialysis vintage, mo | 37 (20–63) | 37 (18–62) | 38 (21–61) | 0.01 | 0.01 | 0.06 | 0.09 |
| Comorbidities | |||||||
| Hypertension | 6645 (99%) | 620 (99%) | 449 (100%) | 0.05 | 0.02 | 0.00 | 0.02 |
| Diabetes | 5912 (89%) | 548 (88%) | 395 (88%) | 0.02 | 0.03 | 0.01 | 0.05 |
| Dyslipidemia | 6255 (94%) | 584 (94%) | 427 (95%) | 0.00 | 0.04 | 0.07 | 0.06 |
| MI | 4520 (68%) | 437 (70%) | 344 (77%) | 0.05 | 0.19 | 0.03 | 0.04 |
| Prior PCI | 1023 (15%) | 133 (21%) | 55 (12%) | 0.17 | 0.09 | 0.00 | 0.02 |
| Prior CABG | 331 (5%) | 28 (5%) | 22 (5%) | 0.02 | 0.00 | 0.04 | 0.00 |
| Stroke history | 2780 (42%) | 171 (28%) | 198 (44%) | 0.29 | 0.05 | 0.02 | 0.01 |
| Peripheral vascular disease | 4785 (72%) | 400 (64%) | 309 (69%) | 0.17 | 0.07 | 0.03 | 0.01 |
| Heart failure | 5675 (85%) | 493 (79%) | 373 (83%) | 0.17 | 0.06 | 0.04 | 0.02 |
| Bleeding history | 3620 (55%) | 314 (51%) | 256 (57%) | 0.08 | 0.05 | 0.01 | 0.02 |
| Malignancy | 1461 (22%) | 104 (17%) | 102 (23%) | 0.13 | 0.02 | 0.01 | 0.00 |
| Alcohol-related disease | 752 (11%) | 57 (9%) | 50 (11%) | 0.07 | 0.01 | 0.02 | 0.01 |
| Liver disease | 2023 (30%) | 180 (29%) | 127 (28%) | 0.03 | 0.05 | 0.01 | 0.07 |
| COPD | 2949 (44%) | 220 (35%) | 187 (42%) | 0.18 | 0.06 | 0.05 | 0.01 |
| Medication | |||||||
| ACEI | 2789 (42%) | 270 (44%) | 181 (40%) | 0.03 | 0.03 | 0.02 | 0.03 |
| ARB | 1355 (20%) | 125 (20%) | 96 (21%) | 0.01 | 0.03 | 0.00 | 0.03 |
| β-blocker | 5784 (87%) | 550 (89%) | 388 (86%) | 0.05 | 0.02 | 0.02 | 0.00 |
| Statin | 5369 (81%) | 509 (82%) | 365 (81%) | 0.03 | 0.01 | 0.02 | 0.03 |
| MRA | 96 (1%) | <11 (2%) | <11 (2%) | 0.01 | 0.01 | 0.03 | 0.01 |
| P2Y12 a | 3008 (45%) | 262 (42%) | 149 (33%) | 0.06 | 0.24 | 0.02 | 0.02 |
| Anticoagulation | 727 (11%) | 61 (10%) | 22 (5%) | 0.04 | 0.20 | 0.00 | 0.09 |
| Setting | |||||||
| Presenting with ACS | 4761 (72%) | 468 (75%) | 350 (78%) | 0.08 | 0.14 | 0.07 | 0.02 |
Results are presented as mean ± SD, median (interquartile range), or number (percentage). Standardized mean differences in the unweighted and the weighted cohorts are depicted. MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting; COPD, chronic obstructive pulmonary disease; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; MRA, mineralocorticoid receptor antagonists; P2Y12, P2Y12 receptor inhibitor (antiplatelet); ACS, acute coronary syndrome.
Only 3187 of these patients had continuous Part D coverage in the 90 d prior to stent implantation.
We performed three sensitivity analyses: (1) an analysis including only patients <75 years old without history of stroke; (2) an analysis including only patients presenting with an acute coronary syndrome; and (3) an analysis including only patients who remained on the same antiplatelet agent throughout the 90-day exposure assessment period (nonswitchers). We also calculated interaction terms for incident versus prevalent users of P2Y12 inhibitors.
In a secondary analysis, we matched 1:1 clopidogrel with prasugrel users or clopidogrel with ticagrelor users using a propensity score with nearest neighbor matching without replacement within a caliper of 0.03. The propensity score for each comparison was estimated using 27 parameters at baseline (shown in Table 1). For this analysis, death from any cause was considered as a competing risk for the secondary outcomes of myocardial infarction, stroke, revascularization, and clinically significant bleeding, as described by Fine and Gray (15). Standardized mean differences were calculated to assess balance of the matched cohorts. In addition, an “intention-to-treat” analysis was performed.
Statistical analyses were performed in SPSS Statistics (version 24.0, released 2016; IBM Corp., Armonk, NY), R (version 3.4.4; R Foundation for Statistical Computing, Vienna, Austria), or Stata (version 14 IC; College Station, TX). Propensity scores for multiple treatments were calculated with the Toolkit for Weighting and Analysis of Nonequivalent Groups package (16). P values of <0.05 were considered statistically significant.
Results
Cohort and Patient Characteristics
There were 1,142,353 prevalent or incident patients on dialysis on or after January 1, 2012. Among them, 32,968 patients had a DES while on dialysis and after January 1, 2012, without coronary artery bypass grafting during the index hospitalization. Of them, 15,813 had at least 6 months of Parts A and B Medicare coverage in the year prior to DES implantation, and 8883 also had Part D coverage through 3 months after DES implantation. Among them, 8566 were alive at 90 days after DES implantation. There was no antiplatelet agent prescription in the 90 days post-DES implantation for 834 patients, whereas 14 patients had recovered kidney function to be off dialysis by that date or their dialysis status was unknown. The final cohort included 6648 patients on clopidogrel, 621 on prasugrel, and 449 on ticagrelor (Figure 1).
Figure 1.
Study flow chart: population selection from the United States Renal Data System. CABG, coronary artery bypass grafting; DES, drug-eluting stent. *No antiplatelet agent at all in 469 patients; antiplatelet prescription after day 90 in 365 patients.
Baseline characteristics of the final cohort are shown in Table 1. The weighted cohort was well balanced for all baseline characteristics. Median follow-up was 166 days (interquartile range, 38–275 days). Four hundred thirty-two patients switched to another antiplatelet agent between day 90 and the end of follow-up, whereas 7286 patients (94%) stayed on the same treatment during the whole follow-up period. A total of 5579 patients (72%) received a DES in the setting of an acute coronary syndrome.
Primary Outcome
A high incidence of the primary outcome, a composite of cardiovascular death, myocardial infarction, or stroke at 12 months, was identified among patients on maintenance dialysis (Table 2). The incidence of the primary outcome was similar across the three treatment groups. For prasugrel versus clopidogrel, the weighted hazard ratio (HR) was 0.96 (95% confidence interval [95% CI], 0.82 to 1.11; P=0.58), corresponding to six fewer events per 100 patient-years (95% CI, 26 fewer to 16 more). For ticagrelor versus clopidogrel, the HR was 1.00 (95% CI, 0.83 to 1.20; P=0.98), corresponding to zero fewer events per 100 patient-years (95% CI, 24 fewer to 29 more). When competing risk survival models were used (with death from any cause as the competing risk), results were qualitatively similar.
Table 2.
Clinical outcomes in the “as-treated” population (main analysis)
| Outcome | Clopidogrel | Prasugrel | Ticagrelor | Weighted Hazard Ratio (95% Confidence Interval) | ||||
|---|---|---|---|---|---|---|---|---|
| Incidence Rate | Events | Incidence Rate | Events | Incidence Rate | Events | Prasugrel versus Clopidogrel | Ticagrelor versus Clopidogrel | |
| CV death, MI, stroke | 144 | 2868 | 126 | 244 | 161 | 167 | 0.96 (0.82 to 1.11) | 1.00 (0.83 to 1.20) |
| Cardiovascular death | 16 | 467 | 15 | 41 | 22 | 36 | 0.94 (0.63 to 1.40) | 1.17 (0.75 to 1.82) |
| MI | 72 | 1707 | 77 | 163 | 89 | 108 | 1.01 (0.83 to 1.23) | 1.04 (0.83 to 1.31) |
| Stroke | 67 | 1563 | 46 | 110 | 74 | 95 | 0.91 (0.73 to 1.14) | 1.04 (0.82 to 1.32) |
| Revascularization | 33 | 877 | 46 | 106 | 41 | 55 | 1.29 (1.00 to 1.67) | 1.19 (0.88 to 1.62) |
| Death from any cause | 27 | 782 | 25 | 68 | 34 | 55 | 0.94 (0.69 to 1.28) | 1.06 (0.75 to 1.50) |
| Clinically relevant bleeding | 79 | 1812 | 84 | 182 | 100 | 126 | 1.15 (0.95 to 1.38) | 1.13 (0.91 to 1.40) |
| MI, stroke, CV death, or revascularization | 153 | 2983 | 141 | 261 | 171 | 172 | 1.00 (0.86 to 1.15) | 1.01 (0.84 to 1.20) |
Incidence rates (unweighted cohort) are presented as number of events per 100 patient-years. Patients were followed from day 90 after drug-eluting stent (DES) implantation up to the date of death, kidney transplantation, loss of Medicare A and B coverage, end of antiplatelet agent prescription or loss of Part D coverage, or 12 month after DES implantation. Exposure was examined as a time-varying covariate. Inverse probability weighted hazard ratios are shown. CV, cardiovascular; MI, myocardial infarction.
Secondary Outcomes
A similar risk of stroke was observed with prasugrel or ticagrelor compared with clopidogrel (Table 2). A numerically higher incidence of revascularization was seen in patients treated with prasugrel compared with clopidogrel: weighted HR, 1.29 (95% CI, 1.00 to 1.67; P=0.05). No significant difference in the incidence of revascularization was observed in patients treated with ticagrelor or clopidogrel: HR, 1.19 (95% CI, 0.88 to 1.62; P=0.26). However, there was no difference in myocardial infarction or cardiovascular death with prasugrel or ticagrelor compared with clopidogrel (Table 2). All-cause mortality at 12 months was also similar with prasugrel or ticagrelor and clopidogrel in this dataset (Table 2).
Trends toward more frequent bleeding episodes were seen with both prasugrel and ticagrelor compared with clopidogrel (Table 2).
Acute coronary syndrome at presentation was independently associated with worse prognosis for the outcomes of myocardial infarction (weighted HR, 2.17; 95% CI, 1.72 to 2.75), revascularization (weighted HR, 1.57; 95% CI, 1.17 to 2.09), and the composite outcome (weighted HR, 1.45; 95% CI, 1.23 to 1.71).
Sensitivity Analyses
Because prasugrel is not indicated in patients aged ≥75 years old or with a history of stroke, we performed a sensitivity analysis including only patients <75 years old without history of stroke (prasugrel candidates) (Table 3). This subgroup included 3159 patients on clopidogrel, 427 on prasugrel, and 211 on ticagrelor. There were no significant differences between prasugrel or ticagrelor and clopidogrel in any outcomes tested (Table 3).
Table 3.
Clinical outcomes in patients aged <75 without history of stroke (“as-treated” analysis; n=3797)
| Outcome | Clopidogrel | Prasugrel | Ticagrelor | Weighted Hazard Ratio (95% Confidence Interval) | ||||
|---|---|---|---|---|---|---|---|---|
| Incidence Rate | Events | Incidence Rate | Events | Incidence Rate | Events | Prasugrel versus Clopidogrel | Ticagrelor versus Clopidogrel | |
| CV death, MI, stroke | 104 | 1126 | 105 | 152 | 102 | 56 | 0.99 (0.81 to 1.21) | 0.90 (0.67 to 1.23) |
| Cardiovascular death | 16 | 227 | 16 | 30 | 16 | 12 | 0.85 (0.51 to 1.39) | 0.68 (0.33 to 1.41) |
| MI | 71 | 808 | 73 | 111 | 78 | 45 | 0.94 (0.74 to 1.19) | 0.94 (0.66 to 1.33) |
| Stroke | 24 | 317 | 26 | 47 | 22 | 15 | 1.15 (0.81 to 1.64) | 0.79 (0.43 to 1.46) |
| Revascularization | 36 | 454 | 47 | 77 | 38 | 24 | 1.16 (0.85 to 1.57) | 1.08 (0.68 to 1.70) |
| Death from any cause | 25 | 346 | 24 | 46 | 20 | 15 | 0.90 (0.61 to 1.33) | 0.60 (0.32 to 1.11) |
| Clinically relevant bleeding | 74 | 828 | 83 | 125 | 97 | 56 | 1.21 (0.97 to 1.50) | 1.17 (0.86 to 1.59) |
| MI, stroke, CV death, or revascularization | 114 | 1200 | 120 | 166 | 109 | 58 | 1.01 (0.83 to 1.23) | 0.91 (0.67 to 1.22) |
Incidence rates (unweighted cohort) are presented as number of events per 100 patient-years. There were 3159 patients on clopidogrel, 427 patients on prasugrel, and 211 patients on ticagrelor. Patients were followed from day 90 after drug-eluting stent (DES) implantation up to the date of death, kidney transplantation, loss of Medicare A and B coverage, end of antiplatelet agent prescription or loss of Part D coverage, or 12 months after DES implantation. CV, cardiovascular; MI, myocardial infarction.
The two sensitivity analyses in patients presenting with acute coronary syndromes or patients who remained on the same antiplatelet agent throughout the 90-day exposure assessment period yielded qualitatively similar results compared with the main analysis with respect to the primary and secondary outcomes (Supplemental Tables 3 and 4). Test of interaction between the antiplatelet agent used and the presence of acute coronary syndrome for the primary outcome were nonsignificant (P=0.71 for prasugrel and P=0.43 for ticagrelor). When a 45-day window was used for exposure assessment, results were qualitatively similar (Supplemental Table 5).
There were 7118 patients with continuous Part D data in the 90 days prior to stent implantation. Among them, there were 3187 patients with a prescription for a P2Y12 inhibitor in the 90 days prior to stent implantation (“prevalent” population) and 3931 patients without a P2Y12 inhibitor in the same time period (“incident” population). The choice of antiplatelet agent was separately examined in the prevalent and incident populations. P values for interaction were >0.05 or =0.05 for all outcomes (Supplemental Table 6).
Intention-to-Treat Analysis
The “intention-to-treat” analysis yielded qualitatively similar results compared with the main analysis (Table 4). The intention-to-treat analysis in the propensity score–matched cohorts (see below) yielded qualitatively similar results (Supplemental Table 8).
Table 4.
Clinical outcomes in the “intention-to-treat” analysis
| Outcome | Clopidogrel | Prasugrel | Ticagrelor | Weighted Hazard Ratio (95% Confidence Interval) | ||||
|---|---|---|---|---|---|---|---|---|
| Incidence Rate | Events | Incidence Rate | Events | Incidence Rate | Events | Prasugrel versus Clopidogrel | Ticagrelor versus Clopidogrel | |
| CV death, MI, stroke | 150 | 3926 | 142 | 365 | 166 | 272 | 1.08 (0.96 to 1.21) | 1.06 (0.93 to 1.21) |
| Cardiovascular death | 26 | 1087 | 28 | 110 | 28 | 76 | 1.12 (0.90 to 1.39) | 1.11 (0.86 to 1.44) |
| MI | 71 | 2243 | 79 | 231 | 84 | 165 | 1.15 (0.99 to 1.34) | 1.12 (0.95 to 1.33) |
| Stroke | 68 | 2140 | 50 | 159 | 69 | 141 | 0.99 (0.84 to 1.18) | 1.02 (0.85 to 1.22) |
| Revascularization | 32 | 1132 | 40 | 133 | 39 | 88 | 1.27 (1.03 to 1.56) | 1.22 (0.96 to 1.54) |
| Death from any cause | 49 | 2045 | 46 | 180 | 56 | 154 | 1.01 (0.85 to 1.19) | 1.17 (0.98 to 1.40) |
| Clinically relevant bleeding | 81 | 2489 | 85 | 241 | 92 | 181 | 1.15 (0.99 to 1.33) | 1.11 (0.95 to 1.31) |
| MI, stroke, CV death, or revascularization | 159 | 4058 | 154 | 381 | 176 | 281 | 1.09 (0.98 to 1.22) | 1.06 (0.93 to 1.21) |
Incidence rates (unweighted cohort) are presented as number of events per 100 patient-years. Patients were followed from day 90 after drug-eluting stent (DES) implantation up to the date of death, kidney transplantation, loss of Medicare A and B coverage, or 12 months after DES implantation. Patients were assumed to remain on the same antiplatelet agent between day 90 and the end of follow-up. CV, cardiovascular; MI, myocardial infarction.
Propensity Score–Matched Cohorts
After propensity score matching, the first cohort included 619 patients on clopidogrel and 619 on prasugrel, while the second cohort included 449 patients on clopidogrel and 449 on ticagrelor. The matched cohorts were well balanced for most baseline characteristics (Supplemental Table 7). There was no significant difference between prasugrel or ticagrelor and clopidogrel for most outcomes (Supplemental Tables 8 and 9). A higher risk of revascularization was observed among prasugrel-treated patients compared with patients who received clopidogrel. A numerically higher risk of clinically relevant bleeding was seen with ticagrelor compared with clopidogrel.
Discussion
To the best of our knowledge, this is the first study to examine the effect of prasugrel or ticagrelor compared with clopidogrel in patients with kidney failure on maintenance hemodialysis who receive a DES. A similar incidence of a composite outcome of cardiovascular death, myocardial infarction, or stroke was observed with prasugrel compared with clopidogrel and ticagrelor compared with clopidogrel in this population. A numerically higher risk of clinically relevant bleeding was observed with prasugrel and ticagrelor, compared with clopidogrel.
The higher incidence of revascularization in patients treated with prasugrel compared with clopidogrel may be explained, at least in part, by selection bias, with patients suffering from more severe and anatomically complex coronary disease potentially receiving more potent antiplatelet agents. If this was the case, the similar incidence of cardiovascular death, myocardial infarction, or stroke with either prasugrel/ticagrelor or clopidogrel is a reassuring finding, suggesting that the newer, more potent antiplatelet drugs could eventually be used in selected patients with kidney failure and anatomically complex coronary disease. The potential benefit, if any, should be balanced against a higher risk of clinically relevant bleeding observed with both prasugrel and ticagrelor, compared with clopidogrel.
Three cohort studies investigated the effectiveness and safety of ticagrelor and prasugrel stratified by kidney function (8 –10). In the SWEDEHEART registry, there were 1986 patients with myocardial infarction on ticagrelor with an eGFR of 30–60 ml/min per 1.73 m2 and 210 patients with an eGFR<30 ml/min per 1.73 m2. In patients with moderate CKD, ticagrelor was superior to clopidogrel with respect to a composite outcome of death, readmission with myocardial infarction, or stroke, but it was also associated with higher incidence of in-hospital bleeding. However, no significant difference was detected between ticagrelor and clopidogrel for all tested clinical outcomes in the smaller subgroup of patients with advanced CKD (8). The PROMETHEUS registry enrolled 617 patients on prasugrel and 4966 on clopidogrel who underwent percutaneous coronary intervention for an acute coronary syndrome and had an eGFR<60 ml/min per 1.73 m2. Major adverse cardiovascular and bleeding event rates at 90 days and 1 year were similar with both agents in the CKD subgroup (9). The RENAMI-BleeMACS registry enrolled 540 patients on ticagrelor, 192 on prasugrel, and 1758 on clopidogrel with an eGFR<60 ml/min per 1.73 m2 who underwent percutaneous coronary angiography for an acute coronary syndrome. Ticagrelor and prasugrel were associated with lower mortality, reinfarction, and major bleeding event rates compared with clopidogrel in patients with CKD, although no significant difference was detected between the three agents in the subgroup of patients with an eGFR<30 ml/min per 1.73 m2 (10). In all three registries presented above, the number of patients on KRT and the outcomes in this subgroup were not reported.
Our study has important limitations. Aspirin use could not be captured in this dataset. In addition, information on the specific type of DES was not available. The sample sizes for prasugrel and ticagrelor were small, limiting statistical power and making it difficult to derive a definite interpretation. Inclusion of “prevalent” patients is a limitation of the main analysis: it may have underestimated the incidence of adverse events and might explain numerically higher incidence of bleeding among prasugrel and ticagrelor versus clopidogrel. However, the effect of prasugrel and ticagrelor versus clopidogrel was found to be similar in a sensitivity analysis in the incident and prevalent populations. In this observational cohort study, diagnostic codes were used for clinical outcomes that were not independently adjudicated. In addition, residual confounding and selection bias may explain some of the results. Nevertheless, inclusion of patients who were treated with DES and similar results in the subgroup analyses are important strengths of our study.
In conclusion, the newer, more potent antiplatelet agents, prasugrel and ticagrelor, were of similar effectiveness compared with clopidogrel in patients on maintenance dialysis with DES, and they were relatively well tolerated with no significant increase in clinically significant bleeding. Clinical trials in this complex population are needed. In the meantime, both prasugrel and ticagrelor could potentially be considered in selected patients on maintenance dialysis with anatomically complex coronary disease who are treated with DES. The potential benefit should be balanced against a potentially higher risk of clinically relevant bleeding.
Disclosures
D.M. Charytan reports employment with New York University School of Medicine; consultancy agreements with Allena Pharmaceuticals (data and safety monitoring board), Amgen, AstraZeneca (data and safety monitoring board), Eli Lilly/Boehringer Ingelheim, Fresenius, Gilead, GSK, Janssen (steering committee), Medtronic, NovoNordisk, and PLC Medical (clinical events committee); receiving consulting fees and travel support from Amgen, Daichi Sankyo, Fresenius, Medtronic/Covidien, and Merck; receiving research funding from Amgen, Bioporto (clinical trial support), Gilead, Medtronic (clinical trial support), and NovoNordisk; receiving research support and consulting fees related to services as a national investigator, on the trial steering committee, or on the data monitoring committee for Allena Pharmaceuticals, AstraZeneca (modest), Gilead Pharmaceuticals, Janssen Pharmaceuticals, NovoNordisk, and Zoll Medical; serving as an Associate Editor of CJASN; and receiving expert witness fees related to proton pump inhibitors and expert witness fees related to dialysate composition (Fresenius). All remaining authors have nothing to disclose.
Funding
T.A. Mavrakanas received salary support from Université de Genève grant 2178281.02.0001 and received McGill University Department of Medicine salary support.
Supplementary Material
Acknowledgments
Preliminary results from this work were presented at the American Society of Nephrology Kidney Week on November 8, 2019 in Washington, DC.
The data reported here have been supplied by United States Renal Data System. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy or interpretation of the US Government.
Because Dr. David M. Charytan is an Associate Editor of CJASN, he was not involved in the peer review process for this manuscript. Another editor oversaw the peer review and decision-making process for this manuscript.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
Supplemental Material
This article contains the following supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.12120720/-/DCSupplemental.
Supplemental Table 1. Comorbidity diagnostic codes.
Supplemental Table 2. Outcome diagnostic codes.
Supplemental Table 3. Clinical outcomes in patients with acute coronary syndromes (“as-treated” analysis; N=5579).
Supplemental Table 4. Clinical outcomes in patients who remained on the same antiplatelet agent throughout the 90-day exposure assessment period (“as-treated” analysis; N=7331).
Supplemental Table 5. Clinical outcomes in the “as-treated” population when exposure to antiplatelet agents was assessed in the first 45 days after drug-eluting stent implantation.
Supplemental Table 6. Interaction terms for the prevalent and incident population.
Supplemental Table 7. Baseline characteristics in the propensity score–matched cohorts.
Supplemental Table 8. Clinical outcomes of the “intention-to-treat” analysis in the matched cohorts.
Supplemental Table 9. Clinical outcomes of the “as-treated” analysis in the matched cohorts.
Supplemental Figure 1. Propensity score distribution per treatment group.
References
- 1. Bonello L, Angiolillo DJ, Aradi D, Sibbing D: P2Y12-ADP receptor blockade in chronic kidney disease patients with acute coronary syndromes. Circulation 138: 1582–1596, 2018. [DOI] [PubMed] [Google Scholar]
- 2. Wallentin L, Becker RC, Budaj A, Cannon CP, Emanuelsson H, Held C, Horrow J, Husted S, James S, Katus H, Mahaffey KW, Scirica BM, Skene A, Steg PG, Storey RF, Harrington RA, Freij A, Thorsén M; PLATO Investigators: Ticagrelor versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 361: 1045–1057, 2009. [DOI] [PubMed] [Google Scholar]
- 3. Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, Gottlieb S, Neumann FJ, Ardissino D, De Servi S, Murphy SA, Riesmeyer J, Weerakkody G, Gibson CM, Antman EM; TRITON-TIMI 38 Investigators: Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 357: 2001–2015, 2007. [DOI] [PubMed] [Google Scholar]
- 4. Mathew RO, Bangalore S, Lavelle MP, Pellikka PA, Sidhu MS, Boden WE, Asif A: Diagnosis and management of atherosclerotic cardiovascular disease in chronic kidney disease: A review. Kidney Int 91: 797–807, 2017. [DOI] [PubMed] [Google Scholar]
- 5. Charytan D, Kuntz RE: The exclusion of patients with chronic kidney disease from clinical trials in coronary artery disease. Kidney Int 70: 2021–2030, 2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Best PJ, Steinhubl SR, Berger PB, Dasgupta A, Brennan DM, Szczech LA, Califf RM, Topol EJ; CREDO Investigators: The efficacy and safety of short- and long-term dual antiplatelet therapy in patients with mild or moderate chronic kidney disease: Results from the Clopidogrel for the Reduction of Events During Observation (CREDO) trial. Am Heart J 155: 687–693, 2008. [DOI] [PubMed] [Google Scholar]
- 7. James S, Budaj A, Aylward P, Buck KK, Cannon CP, Cornel JH, Harrington RA, Horrow J, Katus H, Keltai M, Lewis BS, Parikh K, Storey RF, Szummer K, Wojdyla D, Wallentin L: Ticagrelor versus clopidogrel in acute coronary syndromes in relation to renal function: Results from the Platelet Inhibition and Patient Outcomes (PLATO) trial. Circulation 122: 1056–1067, 2010. [DOI] [PubMed] [Google Scholar]
- 8. Edfors R, Sahlén A, Szummer K, Renlund H, Evans M, Carrero JJ, Spaak J, James SK, Lagerqvist B, Varenhorst C, Jernberg T: Outcomes in patients treated with ticagrelor versus clopidogrel after acute myocardial infarction stratified by renal function. Heart 104: 1575–1582, 2018. [DOI] [PubMed] [Google Scholar]
- 9. Baber U, Chandrasekhar J, Sartori S, Aquino M, Kini AS, Kapadia S, Weintraub W, Muhlestein JB, Vogel B, Faggioni M, Farhan S, Weiss S, Strauss C, Toma C, DeFranco A, Baker BA, Keller S, Effron MB, Henry TD, Rao S, Pocock S, Dangas G, Mehran R: Associations between chronic kidney disease and outcomes with use of prasugrel versus clopidogrel in patients with acute coronary syndrome undergoing percutaneous coronary intervention: A report from the PROMETHEUS study. JACC Cardiovasc Interv 10: 2017–2025, 2017. [DOI] [PubMed] [Google Scholar]
- 10. De Filippo O, D’Ascenzo F, Raposeiras-Roubin S, Abu-Assi E, Peyracchia M, Bocchino PP, Kinnaird T, Ariza-Solé A, Liebetrau C, Manzano-Fernández S, Boccuzzi G, Henriques JPS, Templin C, Wilton SB, Omedè P, Velicki L, Xanthopoulou I, Correia L, Cerrato E, Rognoni A, Fabrizio U, Nuñez-Gil I, Iannaccone M, Montabone A, Taha S, Fujii T, Durante A, Song X, Gili S, Magnani G, Varbella F, Kawaji T, Blanco PF, Garay A, Quadri G, Alexopoulos D, Caneiro Queija B, Huczek Z, Cobas Paz R, González Juanatey JR, Cespón Fernández M, Nie SP, Muñoz Pousa I, Kawashiri MA, Gallo D, Morbiducci U, Conrotto F, Montefusco A, Dominguez-Rodriguez A, López-Cuenca A, Cequier A, Iñiguez-Romo A, Usmiani T, Rinaldi M, De Ferrari GM: P2Y12 inhibitors in acute coronary syndrome patients with renal dysfunction: An analysis from the RENAMI and BleeMACS projects. Eur Heart J Cardiovasc Pharmacother 6: 31–42, 2020. [DOI] [PubMed] [Google Scholar]
- 11.United States Renal Data System: Annual Data Report. Available at: https://www.usrds.org/annual-data-report/. Accessed December 30, 2020
- 12. Shroff GR, Solid CA, Herzog CA: Long-term survival and repeat coronary revascularization in dialysis patients after surgical and percutaneous coronary revascularization with drug-eluting and bare metal stents in the United States. Circulation 127: 1861–1869, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. McCaffrey DF, Griffin BA, Almirall D, Slaughter ME, Ramchand R, Burgette LF: A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Stat Med 32: 3388–3414, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Normand ST, Landrum MB, Guadagnoli E, Ayanian JZ, Ryan TJ, Cleary PD, McNeil BJ: Validating recommendations for coronary angiography following acute myocardial infarction in the elderly: A matched analysis using propensity scores. J Clin Epidemiol 54: 387–398, 2001. [DOI] [PubMed] [Google Scholar]
- 15. Fine JP, Gray RJ: A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 94: 496–509, 1999. [Google Scholar]
- 16. Cefalu M, Buenaventura M: Propensity scores for multiple treatments: A tutorial on the MNPS command for Stata users, 2017. Available at: https://www.rand.org/pubs/tools/TL170z1.html. Accessed October 1, 2020
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.