Article ID: CJ-24-0173
Background: Low-dose prasugrel (3.75 mg) is used as maintenance therapy for percutaneous coronary intervention; however, data on long-term outcomes are scarce.
Methods and Results: We analyzed 5,392 participants in the KiCS-PCI registry who were administered low-dose prasugrel or clopidogrel at discharge between 2008 and 2018 and for whom 2-year follow-up data were available. We adjusted for confounders using matching weight analyses and multiple imputations. Similarly, we used inverse probability- and propensity score-weighted analyses. We also performed instrumental variable analyses. The primary outcomes were acute coronary syndrome (ACS) and bleeding requiring readmission. Secondary outcomes were all-cause death and a composite outcome of ACS, bleeding, heart failure, stroke, coronary bypass requiring admission, and all-cause death. In this cohort, 12.2% of patients were discharged with low-dose prasugrel. Compared with clopidogrel, low-dose prasugrel was associated with a reduced risk of ACS (hazard ratio [HR] 0.58; 95% confidence interval [CI] 0.39–0.85), bleeding (HR 0.62; 95% CI 0.40–0.97), and the composite outcome (HR 0.71; 95% CI 0.59–0.86). Inverse probability-weighted analysis yielded similar results; however, matching weight analysis without multiple imputations and propensity score-matched analyses showed similar outcomes in both groups. Instrumental variable analyses showed reduced risks of ACS and composite outcome for those on low-dose prasugrel. All-cause mortality did not differ in all analyses.
Conclusions: Low-dose prasugrel demonstrates comparable outcomes to clopidogrel in terms of ACS and bleeding.
Dual antiplatelet therapy (DAPT) with aspirin and P2Y12 inhibitors is the standard treatment for patients undergoing percutaneous coronary intervention (PCI).1–5 The American College of Cardiology/American Heart Association and European Society of Cardiology guidelines advocate for a standard dose of prasugrel of 10 mg for patients with acute coronary syndrome (ACS),6–8 following insights from the TRITON TIMI 38 (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel-Thrombolysis in Myocardial Infarction 38) study.6 This study, published in 2007, highlighted prasugrel’s effectiveness in reducing ischemic events, but was conducted during an era dominated by bare-metal or first-generation drug-eluting stents.6 However, the standard dose of prasugrel has been linked with an elevated risk of major bleeding.6
Consequently, Japan has introduced a low dose of prasugrel (3.75 mg), considering the unique genetic profile of the East Asian population, particularly the cytochrome P450 family 2 subfamily C member 19 (CYP2C19) loss-of-function variant. This genetic variation can increase ischemic event risk due to reduced clopidogrel efficacy and increased platelet reactivity.9 The PRASFIT-ACS (Prasugrel Compared with Clopidogrel for Japanese Patients With ACS Undergoing PCI) study echoed the findings of the TRITON TIMI 38 study, suggesting a diminished risk of ischemic events with low-dose prasugrel. This is particularly relevant given the higher average age and lower body weight in the Japanese population compared with Western demographics.10,11 Notably, although the PRASFIT-ACS trial did not demonstrate statistically significant differences between prasugrel and clopidogrel in terms of benefits regarding ischemic events, the authors inferred positive outcomes from low-dose prasugrel.10
In light of these comprehensive findings from large-scale randomized control trials (RCTs), low-dose prasugrel has been getting attention in Japan for patients with ACS and those undergoing elective PCI since its introduction in 2014.10,12 Despite its widespread use, observational and real-world studies have indicated that low-dose prasugrel is still linked to an increased risk of in-hospital or short-term bleeding compared with clopidogrel, specifically within the Japanese context.13,14
Internationally, the use of low-dose prasugrel is drawing increasing attention, particularly due to its potential benefits in short-term DAPT or de-escalation strategies for patients at high risk of bleeding. The correlation between its use and long-term outcomes remains an important area for investigation, including the legacy effect beyond 1 year of DAPT.15,16 Therefore, in our study, using a contemporary multicenter all-comer PCI registry, we aimed to explore the relationship between low-dose prasugrel and long-term outcomes at 2 years after PCI compared with clopidogrel.
This study was conducted as part of the Japan Cardiovascular Database-Keio Interhospital Cardiovascular Study (JCD-KiCS) PCI registry, a multicenter prospective registry that includes data of consecutive patients who have undergone PCI since 2008 at 15 institutions within the Tokyo metropolitan area. The details of this registry have been published previously.17,18 The participating hospitals were instructed to document and register patient data from consecutive hospital visits for PCI using an internet-based data collection system. Registered data were reviewed for completeness and internal consistency.
Quality assurance of the data was achieved through automatic system validation, reporting of data completeness, and the education and training of clinical research coordinators specifically trained to use the present PCI registry. The senior study coordinator (I.U.) and the conduction of exclusive onsite auditing by the investigator (S.K.) ensured the appropriate registration of each patient. All participants provided written informed consent.
The JCD-KiCS registry has been registered with the University Hospital Medical Information Network (UMIN) Clinical Trials Registry (ID: UMIN000004736). The study protocol was approved by the institutional review board of each participating hospital.
Studied PatientsOf the 26,224 consecutive PCI patients registered between September 2008 and March 2021, we initially selected 8,930 who underwent PCI by December 2018 and had 2-year outcomes without duplicates (such as staged PCI). The patients included in this study were those who: (1) received drug-eluting stents; (2) were discharged on DAPT with aspirin and low-dose prasugrel or clopidogrel; and (3) did not require oral anticoagulants (i.e., those with atrial fibrillation). We excluded patients with atrial fibrillation or those taking oral anticoagulants at discharge (n=1,138), those without drug-eluting stents (n=2,141), those who did not receive aspirin at discharge (n=49), and those who did not receive low-dose prasugrel or clopidogrel at discharge (n=210). The final study cohort included 5,392 patients (Figure 1), divided into low-dose prasugrel (n=659; 12.2%) and clopidogrel (n=4,733; 87.8%) groups. The administration of a standard dose of ticagrelor was approved by the Japanese Pharmaceutical Agency in 2016; however, ticagrelor is not often prescribed by Japanese physicians primarily because of concerns about the increased risk of bleeding events observed after its administration to patients from East Asia.19 Therefore, none of the patients in our study cohort received ticagrelor.
Flow chart of the analysis. PCI, percutaneous coronary intervention.
Definitions of Outcomes and Variables
The clinical variables and outcomes of the JCD-KiCS registry were aligned using data from the National Cardiovascular Data Registry (CathPCI Registry version 4.1). ACS was defined as ST-segment elevation myocardial infarction (STEMI), non-STEMI, or unstable angina. Stable coronary artery disease was defined as stable angina, previous myocardial infarction, or silent ischemia. The presence of heart failure was defined as left ventricular ejection fraction ≤35% or documentation of heart failure by the attending physician regardless of left ventricular ejection fraction. Multivessel disease was defined as 2 or more major coronary arteries with ≥75% stenosis. The estimated glomerular filtration rate was calculated using the Modification of Diet in Renal Disease Equation for Japanese Patients proposed by the Japanese Society of Nephrology.20–22
All major procedural complications (e.g., death, bleeding complications, and cardiac and cerebrovascular events) were defined by the clinical research coordinator, and details have been published elsewhere.23 Procedural complications were initially reviewed by a trained clinical research coordinator, supervised by the project coordinator, and categorized as those requiring adjudication or those exempted. A separate member of the event committee reviewed the abstract records. A second or third adjudicator was needed concerning a disagreement between the opinions of the project coordinator and the first adjudicator. During follow-up, antithrombotic therapy was administered according to hospital guidelines and established standards of care. Regarding drug-eluting stent implantation, DAPT was required for 12 months during the study period.
Participants were followed up for 2 years after hospital discharge to identify hospitalizations for cardiovascular or bleeding events and all-cause deaths via medical records, telephone calls, or mail. All follow-up data were collected and recorded using a secure internet-based electronic data capture system by dedicated clinical research coordinators trained by the primary investigator and project coordinators. The primary efficacy and safety outcomes of this study were ACS and bleeding events requiring readmission, respectively. Secondary outcomes were all-cause death and a composite outcome of ACS, bleeding, heart failure, stroke, coronary bypass requiring readmission, and all-cause death.
Statistical AnalysesContinuous variables are presented as the mean±SD or median with interquartile range, as appropriate for data distribution. Categorical variables are expressed as percentages. The significance of differences in continuous variables were evaluated using Student’s t-test or the Mann-Whitney U test. Categorical variables were analyzed using the χ2 test or Fisher’s exact t-test.
A propensity score analysis was performed to adjust for differences in baseline characteristics and post-procedural conditions between those receiving low-dose prasugrel and those treated with clopidogrel. The following variables were used to estimate propensity score: age, sex, body mass index, diabetes, dyslipidemia, hypertension, chronic lung disease, cerebrovascular disease, cancer, history of PCI, history of coronary artery bypass, history of myocardial infarction, history of heart failure, smoking, indication for PCI, urgency of PCI, heart failure symptoms, cardiogenic shock at presentation, cardiopulmonary arrest at presentation, diseased vessels, PCI lesions, lesion characteristics (bifurcation, Type C lesion, and chronic total occlusion), estimated glomerular filtration rate, hemoglobin at discharge, puncture site, and bleeding complications after PCI within 72 h. Subsequently, we performed a matching-weighted analysis to estimate the effect of low-dose prasugrel on a complete case analysis basis;24 baseline characteristics were assessed and balanced if the standardized mean difference was <0.1.25 Furthermore, we created Kaplan-Meier estimates from hospital discharge to the outcome of interest and performed Cox proportional hazard model analysis among the matching-weighted data. Of 5,392 patients, 403 had missing values (n=5 for dyslipidemia, n=71 for cancer, n=7 for smoking, n=5 for heart failure symptoms, n=7 for puncture site, n=99 for body mass index, n=222 for hemoglobin, n=81 for estimated glomerular filtration rate). To reduce bias owing to missing data and to increase the precision of the estimates, multiple imputations for missing data were applied.26 For multiple imputation, we imputed missing data with 10 datasets assuming missing data at random,27 estimated the hazard ratio (HR) of prasugrel vs. clopidogrel for the outcomes by applying matching weights in each imputed dataset, and pooled the estimates using Rubin’s rule.
For sensitivity analyses, we conducted: (1) an inverse probability-weighted analysis with 5% truncated weight to mitigate the influence of extreme weight;27 (2) a 1:1 propensity score matching analysis using the nearest neighbor with a caliper equal to 0.2 of the standard deviation of the logit of the propensity score;28 and (3) a matching weight analysis using 1-year outcomes.
Finally, we used an instrumental variable method to demonstrate the robustness of our analysis.29 The study period (pre-prasugrel approval period [September 2008–February 2014] and post-prasugrel approval period [March 2014–December 2018]) was used as an instrumental variable, which has been used in previous studies in the field of cardiology.5,30–32 A valid instrumental variable should fulfill several assumptions. First, the instrumental variable should be strongly correlated with the exposure variable. We evaluated the strength of this assumption using F-statistics; F>10 is a commonly used indicator for a valid instrumental variable. The second assumption was that the instrumental variable affected the relevant outcomes only through its effects on the exposure variable. The third assumption was that the instrumental variable did not affect possible confounders. To indirectly assess this hypothesis, we compared the characteristics of patients in the earlier study period with those of patients in the later study period.32 Subsequently, 2-stage residual inclusion instrumental variable estimates were determined. The first was a multivariable model that included age, sex, and the instrumental variable (the binary indicator of the study period: early or late). The second stage was a multivariable Cox regression, including the residual from the first-stage model and the measured confounding predictors and exposure variables.32
We also added several subgroup analyses: patients since prasugrel approval (March 2014–); patients with ACS; and patients with stable coronary artery disease.
All statistical calculations and analyses were performed using R 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria), using ‘MatchIt’, ‘VGAM’, ‘ivtools’, and ‘mice’ packages. Two-tailed P values <0.05 was considered statistically significant.
The mean age of the analyzed patients was 69.0±10.9 years. Tables 1 and 2 present the baseline characteristics and in-hospital and long-term outcomes of patients receiving low-dose prasugrel and those on clopidogrel. The low-dose prasugrel group was younger and had significantly lower proportions of comorbidities, such as hypertension, diabetes, dialysis, peripheral artery disease, cerebrovascular disease, and a history of heart failure, than the clopidogrel group (Table 1). In contrast, the low-dose prasugrel group had a significantly higher proportion of ACS and emergency PCIs than the clopidogrel group. The procedural complications after PCI were similar in the 2 groups (Table 2).
Baseline Characteristics of All Patients
| Clopidogrel (n=4,733) |
Low-dose prasugrel (n=659) |
P value | |
|---|---|---|---|
| Age (years) | 69.00 [61.00–76.00] | 67.00 [59.00–75.00] | 0.001 |
| Male sex | 3,657 (77.3) | 531 (80.6) | 0.063 |
| Body mass index (kg/m2) | 23.97 [21.91–26.19] | 23.99 [22.07–26.28] | 0.407 |
| eGFR (mL/min/1.73 m2) | 63.12 [50.25–75.07] | 64.05 [52.07–74.73] | 0.345 |
| Hemoglobin (g/dL) | 12.50 [11.20–13.70] | 12.80 [11.60–13.80] | <0.001 |
| Previous MI | 750 (15.8) | 66 (10.0) | <0.001 |
| Previous HF | 289 (6.1) | 24 (3.6) | 0.014 |
| Diabetes | 1,889 (39.9) | 233 (35.4) | 0.028 |
| Cerebrovascular disease | 394 (8.3) | 16 (2.4) | <0.001 |
| Peripheral artery disease | 379 (8.0) | 30 (4.6) | 0.002 |
| Chronic lung disease | 144 (3.0) | 14 (2.1) | 0.236 |
| Hypertension | 3,615 (76.4) | 478 (72.5) | 0.035 |
| Dyslipidemia | 3,115 (65.9) | 410 (62.2) | 0.07 |
| Smoking | 1,585 (33.5) | 212 (32.2) | 0.513 |
| Dialysis | 178 (3.8) | 7 (1.1) | 0.001 |
| Cancer | 162 (3.5) | 38 (5.8) | 0.005 |
| Previous PCI | 992 (21.0) | 79 (12.0) | <0.001 |
| Previous coronary bypass | 207 (4.4) | 15 (2.3) | 0.015 |
| HF on admission | 451 (9.5) | 33 (5.0) | <0.001 |
| CS on admission | 107 (2.3) | 26 (3.9) | 0.013 |
| CPA on admission | 51 (1.1) | 16 (2.4) | 0.006 |
| Puncture site | <0.001 | ||
| Femoral artery approach | 2,498 (52.9) | 88 (13.4) | |
| Radial artery approach | 2,163 (45.8) | 568 (86.2) | |
| Brachial artery approach | 65 (1.4) | 3 (0.5) | |
| Significant lesions | |||
| Right coronary artery | 2,246 (47.5) | 316 (48.0) | 0.843 |
| Left main | 397 (8.4) | 50 (7.6) | 0.533 |
| Left anterior descending artery | 3,558 (75.2) | 486 (73.7) | 0.457 |
| Left circumflex artery | 2,159 (45.6) | 271 (41.1) | 0.033 |
| Multivessel disease | 2,789 (58.9) | 378 (57.4) | 0.469 |
| Culprit lesions | |||
| Right coronary artery | 1,363 (28.8) | 205 (31.1) | 0.239 |
| Left main | 196 (4.1) | 27 (4.1) | 1 |
| Left anterior descending artery | 2,646 (55.9) | 361 (54.8) | 0.615 |
| Left circumflex artery | 1,010 (21.3) | 105 (15.9) | 0.002 |
| Use of IABP | 196 (4.1) | 27 (4.1) | 1 |
| PCI indication | <0.001 | ||
| STEMI | 959 (20.3) | 227 (34.4) | |
| UA/NSTEMI | 1,263 (26.7) | 174 (26.4) | |
| Elective | 2,496 (52.7) | 256 (38.8) | |
| PCI urgency | <0.001 | ||
| Salvage | 30 (0.6) | 9 (1.4) | |
| Emergency | 929 (19.6) | 218 (33.1) | |
| Urgent | 1,081 (22.8) | 163 (24.7) | |
| Elective | 2,693 (56.9) | 269 (40.8) | |
| Chronic total occlusion | 228 (4.8) | 22 (3.3) | 0.111 |
| Bifurcation lesion | 1,319 (27.9) | 146 (22.2) | 0.002 |
| Type C lesion | 1,395 (29.5) | 254 (38.5) | <0.001 |
| Use of RA | 162 (3.4) | 23 (3.5) | 1 |
| Use of IVUS | 4,121 (87.1) | 621 (94.2) | <0.001 |
| Total no. stents | 1.00 [1.00–2.00] | 1.00 [1.00–1.00] | <0.001 |
| Total stent length (mm) | 24.00 [18.00–38.00] | 26.00 [19.00–38.00] | 0.191 |
Unless indicated otherwise, data are given as the median [interquartile range], or n (%). CPA, cardiopulmonary arrest; CS, cardiogenic shock; eGFR, estimated glomerular filtration rate; HF, heart failure; IABP, intra-aortic balloon pump; IVUS, intravascular ultrasound; PCI, percutaneous coronary intervention; RA, rotational atherectomy; STEMI, ST-elevation myocardial infarction; UA/NSTEMI, unstable angina/non-ST-elevation myocardial infarction.
In-Hospital Outcomes of All Patients
| Clopidogrel (n=4,733) |
Low-dose prasugrel (n=659) |
P value | |
|---|---|---|---|
| All complications | 279 (5.9) | 38 (5.8) | 0.966 |
| Coronary dissection | 29 (0.6) | 1 (0.2) | 0.226 |
| Coronary perforation | 30 (0.6) | 4 (0.6) | 1 |
| Myocardial infarction | 52 (1.1) | 2 (0.3) | 0.087 |
| Cardiogenic shock | 38 (0.8) | 9 (1.4) | 0.218 |
| Heart failure | 48 (1.0) | 9 (1.4) | 0.533 |
| Cerebral infarction | 8 (0.2) | 2 (0.3) | 0.788 |
| Dialysis | 22 (0.5) | 2 (0.3) | 0.787 |
| Intracranial hemorrhage | 0 (0.0) | 0 (0.0) | |
| Cardiac tamponade | 8 (0.2) | 0 (0.0) | 0.606 |
| Transfusion | 65 (1.4) | 7 (1.1) | 0.638 |
| Bleeding (all types) | 85 (1.8) | 13 (2.0) | 0.871 |
| Puncture site bleeding | 31 (0.7) | 3 (0.5) | 0.731 |
| Puncture site hematoma | 29 (0.6) | 1 (0.2) | 0.226 |
| Peritoneal bleeding | 4 (0.1) | 0 (0.0) | 1 |
| Gastrointestinal bleeding | 6 (0.1) | 3 (0.5) | 0.154 |
| Genitourinary bleeding | 4 (0.1) | 0 (0.0) | 1 |
| Other bleeding | 25 (0.5) | 4 (0.6) | 1 |
Unless indicated otherwise, data are given as n (%).
Crude and Adjusted Outcomes
The incidences of ACS, heart failure requiring hospitalization, and the composite outcome during follow-up were lower in the low-dose prasugrel than clopidogrel group (Table 3).
Long-Term Outcomes of All Patients
| Clopidogrel (n=4,733) |
Low-dose prasugrel (n=659) |
P value | |
|---|---|---|---|
| Acute coronary syndrome | 150 (3.2) | 10 (1.5) | 0.027 |
| Bleeding | 109 (2.3) | 9 (1.4) | 0.162 |
| Heart failure | 140 (3.0) | 7 (1.1) | 0.008 |
| Stroke | 72 (1.5) | 7 (1.1) | 0.456 |
| Coronary artery bypass | 41 (0.9) | 5 (0.8) | 0.956 |
| Death | 147 (3.1) | 16 (2.4) | 0.406 |
| Composite endpoint | 571 (12.1) | 47 (7.1) | <0.001 |
Unless indicated otherwise, data are given as n (%).
After matching weight analysis, the baseline characteristics were mostly well balanced (Supplementary Tables 1,2). Kaplan-Meier curves of the composite endpoint showed that low-dose prasugrel was associated with a decreased risk compared with clopidogrel (Figure 2). The HRs of these events with multiple imputations are presented in Table 4, demonstrating that low-dose prasugrel was associated with reduced risks of ACS, bleeding, and the composite endpoint. However, complete case analyses with matching weights did not reveal significant differences in ACS or bleeding between the 2 groups.
Kaplan-Meier event curve for the composite endpoint with matched weighted analysis.
Cox Proportional Hazard Model for Each Outcome; Low-Dose Prasugrel vs. Clopidogrel 2008–2018
| ACS | Bleeding | Death | Composite | |||||
|---|---|---|---|---|---|---|---|---|
| Matching weight (imputation) |
0.58 (0.39–0.85) | 0.62 (0.40–0.97) | 0.97 (0.69–1.38) | 0.71 (0.59–0.86) | ||||
| Matching weight |
0.58 (0.29–1.16), P=0.12 |
0.62 (0.29–1.32), P=0.22 |
0.97 (0.55–1.72), P=0.92 |
0.71 (0.51–0.99), P=0.041 |
||||
| Truncated weight |
0.42 (0.21–0.86), P=0.017 |
0.42 (0.20–0.87), P=0.020 |
0.98 (0.56–1.73), P=0.95 |
0.67 (0.48–0.94), P=0.019 |
||||
| Propensity score matched |
0.77 (0.34–1.76), P=0.54 |
0.80 (0.32–2.03), P=0.64 |
1.16 (0.55–2.44), P=0.69 |
0.80 (0.54–1.19), P=0.27 |
||||
| Instrumental variable |
0.18 (0.065–0.53), P=0.0015 |
0.97 (0.33–2.84), P=0.95 |
1.82 (0.76–4.38), P=0.18 |
0.55 (0.34–0.89), P=0.015 |
||||
| Matching weight | ||||||||
| 1 year outcome |
0.67 (0.27–1.63), P=0.38 |
0.55 (0.19–1.61), P=0.27 |
0.97 (0.43–2.20), P=0.94 |
0.68 (0.44–1.06), P=0.086 |
||||
| After prasugrel approval |
0.80 (0.36–1.75), P=0.60 |
0.57 (0.25–1.3), P=0.18 |
0.87 (0.48–1.60), P=0.7 |
0.73 (0.51–1.05), P=0.09 |
||||
| ACS | 0.62 (0.28–1.37), P=0.24 |
P interaction =0.39 |
0.85 (0.30–2.38), P=0.76 |
P interaction =0.56 |
1.15 (0.48–2.76), P=0.75 |
P interaction =0.87 |
0.73 (0.46–1.15), P=0.17 |
P interaction =0.75 |
| Stable CAD |
0.26 (0.035–1.94), P=0.19 |
0.47 (0.14–1.56), P=0.21 |
1.06 (0.50–2.23), P=0.89 |
0.77 (0.47–1.25), P=0.29 |
||||
Data show hazard ratios with 95% confidence intervals in parentheses and P values as indicated. ACS, acute coronary syndrome; CAD, coronary artery disease.
Baseline characteristics for the sensitivity analysis using inverse probability-weighted analysis are presented in Supplementary Tables 3 and 4. In this analysis, low-dose prasugrel was associated with reduced risks of ACS, bleeding, and the composite endpoint, and the baseline characteristics were not well balanced (Table 4). With propensity score-matched analysis, the baseline characteristics were well balanced (Supplementary Tables 5,6); however, there were no significant differences in the long-term outcomes between the 2 groups. In a sensitivity analysis using a matching weight with 1-year outcomes, the results were mostly similar, but there were no significant differences between the 2 groups, including in the composite outcome (Table 4).
In another sensitivity analysis performed using the instrumental variable method, we confirmed that the F value was >10 (F=2,061, P<0.01). The 2-stage residual inclusion instrumental variable estimate demonstrated associations between the post-discharge antiplatelet regimen and ACS and a decreased risk of the composite outcome (Table 3).
In the subgroup analysis among patients in the post-prasugrel approval period, the trend is similar, but there were no significant differences between the 2 groups, including in the composite endpoint (Table 4). In the subgroup analysis among patients with ACS or stable coronary artery disease, P interaction values were not significant, suggesting that the effect size is similar to the principal findings (Table 4).
The principal findings of our study are as follows. First, in the matching-weighted with multiple imputation and inverse probability weight analyses, the use of low-dose prasugrel was associated with the risk of ACS and bleeding requiring readmission; however, this was not observed in matching weight on a complete case analysis basis or in propensity score matched analyses. Second, analysis using the instrumental variable method demonstrated a decreased risk of ACS. Overall, most analyses consistently showed a favorable outcome in terms of the composite endpoint in the low-dose prasugrel group compared with the clopidogrel group, and our results suggest that low-dose prasugrel can be similarly effective, safe, or more effective than clopidogrel in the East Asian population.
Low-Dose Prasugrel and Ischemic EventsA prior multicenter registry study in Japan with long-term follow-up investigated the effects of low-dose prasugrel on patients with acute myocardial infarction.33 That study found that the incidence of ischemic events was similar between patients receiving low-dose prasugrel and those treated with standard-dose clopidogrel.33 In addition, the risk of bleeding was found to be lower in the low-dose prasugrel group.33 Although these findings on bleeding align with our findings in the present study, our study diverges by demonstrating a reduction in ACS events among patients in the low-dose prasugrel group. Our analysis was methodologically rigorous, evaluating the appropriateness of propensity score analysis through standardized mean differences between the groups, thus offering a more robust comparison than methods that triangulate data, including the use of instrumental variables.24,29 The collective data on low-dose prasugrel have consistently shown favorable outcomes, surpassing those observed in RCTs. Moreover, transitioning to low-dose prasugrel after standard DAPT has revealed promising results.34 Together, these findings support low-dose prasugrel as a viable option for patients undergoing PCI.
Low-Dose Prasugrel and Bleeding EventsBleeding in patients undergoing PCI is more prevalent in the East Asian population than in Western countries. The PHILO trial highlighted an increased risk of bleeding with ticagrelor among East Asian patients, leading to a widespread reluctance to use ticagrelor in many countries within this region.19,35 Conversely, low-dose prasugrel, rather than ticagrelor, is frequently used in Japan, supported by findings from the PRASFIT-ACS and PRASFIT-Elective trials.10,12 The standard dose of prasugrel (10 mg) used in other East Asian countries, such as Korea, raises concerns regarding post-PCI bleeding.10,12,15 These trials notably excluded patients with comorbidities, such as bleeding history, significant history of heart failure, or severe kidney conditions. Furthermore, immediate postprocedural concerns about bleeding with low-dose prasugrel persist in the Japanese population,13,14 demonstrated the effectiveness of reduced platelet reactivity with low-dose prasugrel compared with clopidogrel, suggesting efficacy, but also highlighting bleeding risks. Therefore, there was a significant need for long-term data on the efficacy and safety of low-dose prasugrel.36
To answer this critical question, we conducted an analysis of all-comer PCI patients with adjudicated 2-year outcomes. Our data showed that low-dose prasugrel offers comparable or better efficacy and safety for Japanese patients undergoing PCI, presenting valuable insights for East Asian countries and their diaspora in the US. From our analysis, low-dose prasugrel appears to be a viable option for a wide range of patients, aiming to reduce the risk of bleeding. This is crucial, because bleeding can lead to secondary ischemic events and fatalities, a particularly important consideration among East Asian patients.3,37
Toward Further Reductions in Event RatesOur findings, alongside previous meta-analyses, indicate that ischemic and bleeding outcomes in patients with ACS are similar after treatment with low-dose prasugrel and clopidogrel.34,38 The CYP2C19 loss-of-function variant, prevalent in the East Asian population, suggests that low-dose prasugrel could be particularly advantageous for this group. Notably, in our registry, as is typical in Japan, the vast majority of PCIs were guided by intravascular ultrasound, which is associated with fewer ischemic events, including stent thrombosis.18,39 Hence, we consider that the higher use of intravascular ultrasound in Japan compared with Western countries may obviate the need for high-potency antiplatelet agents, including standard doses of prasugrel.40 In addition, considering Japan’s relatively older population, low-dose prasugrel may be more appropriate than clopidogrel for these patients.22 However, the use of intravascular imaging is on the rise in the US, supported by numerous randomized trials highlighting its beneficial effects. In addition, real-world data suggest that patients, particularly patients who are older than those typically enrolled in randomized trials, may fare better with clopidogrel than with prasugrel or ticagrelor, especially elderly ACS patients.6,41–43 Therefore, there is a compelling case for low-dose prasugrel to receive greater consideration, even within Western populations, as we strive for even lower rates of cardiovascular events.
Study LimitationsOur study has some limitations. Because this was an observational study, there may be unknown and unmeasurable factors that are confounding the relationship between access site and postprocedural stroke. However, our sensitivity analysis showed that our results are robust in almost all plausible scenarios. First, our study did not capture specific details regarding the duration of DAPT, including any transitions between low-dose prasugrel and clopidogrel. However, during the study period of 2008–2018, it was standard practice among Japanese physicians to prescribe DAPT for 1 year after PCI, in line with prevailing guidelines at the time. Notably, significant studies such as the MASTER DAPT (The Management of High Bleeding Risk Patients Post Bioresorbable Polymer Coated Stent Im- plantation with an Abbreviated versus Standard DAPT Regimen) trial, which explored the implications of a shortened DAPT duration, and the HOST-EXAM (the Harmonizing Optimal Strategy for Treatment of coronary artery stenosis-Extended Antiplatelet Monotherapy) trial, which investigated clopidogrel monotherapy, were published subsequent to our study period.44–46 Regardless, the inadequate information may have affected the study results. Second, we did not have bleeding outcomes with standardized definitions, such as Bleeding Academic Research Consortium Types 2, 3, or 5; however, Type 2 includes bleeding events requiring hospitalization.47 In addition, we did not have data on stent thrombosis, which could be a crucial outcome in DAPT studies. Third, prasugrel became available in Japan from 2014, which contributes to the better outcomes with prasugrel than clopidogrel because there have been improvements in patient management after PCI. However, we present outcomes among patients since prasugrel approval, demonstrating a similar effect, but not statistically significant due to the study being underpowered. Finally, it is important to note that, in Japan, prasugrel monotherapy was not sanctioned until a guideline update in 2020. Nevertheless, the lack of detailed information on DAPT duration may have influenced our study outcomes.
In conclusion, low-dose prasugrel demonstrates comparable outcomes to clopidogrel in terms of post-PCI ACS and bleeding in a real-world, all-comer registry in Japan. This finding is particularly significant given the higher bleeding risk among Japanese patients compared with their Western counterparts, even after rigorous statistical adjustments.
The authors thank all the investigators, clinical coordinators, and institutions involved in the JCD-KiCS.
This research was supported by a grant from the Ministry of Education, Culture, Sports, Science, and Technology, Japan (KAKENHI No. 20H03915).
S.K. has received research grants from Daiichi Sankyo Co., Ltd., Novartis, and AstraZeneca. The funders did not play any role in study design, data collection, data analysis, decision to publish, or manuscript preparation. The remaining authors declare that they have no conflicts of interest.
The present study was approved by the IRB Committee of Keio University (Reference no. 200873).
Please find supplementary file(s);
https://doi.org/10.1253/circj.CJ-24-0173
