1 Background & Study Overview

There are three main P2Y12 inhibitors used in clinical practice:

Drug Brand Name Potency
Clopidogrel Plavix Weakest (prodrug, variable response)
Prasugrel Effient Potent
Ticagrelor Brilinta Potent

1.1 How did we measure kidney function?

We used the CKD-EPI equation (Chronic Kidney Disease Epidemiology Collaboration), the current gold-standard formula for estimating GFR from serum creatinine, age, and sex. Patients were divided into three groups:

  • eGFR > 60 ml/min/1.73m² — normal to mildly reduced kidney function (reference group)
  • eGFR 30–60 ml/min/1.73m² — moderately reduced (CKD stage 3)
  • eGFR < 30 ml/min/1.73m² — severely reduced (CKD stage 4–5)

1.2 Study population

We analyzed all patients who presented with ACS and underwent PCI at our center, recorded in the PCI Registry. The primary outcome was all-cause mortality, analyzed using Cox proportional hazards models — a standard survival analysis method that accounts for the time until an event occurs and for patients who were lost to follow-up.

2 Baseline Characteristics

The table below describes the characteristics of the study population, stratified by which P2Y12 inhibitor they received. This helps us understand whether the three groups were similar at baseline (before treatment), or whether there were important differences that might confound our comparisons.

How to read this table: Continuous variables (like age) are shown as mean ± SD. Categorical variables (like sex) are shown as count (%). The p-value column indicates whether the differences between the three P2Y12 groups are statistically significant (p < 0.05).

Table 1. Baseline Characteristics by P2Y12 Inhibitor
Overall Brilinta (Ticagrelor) Effient (Prasugrel) Plavix (Clopidogrel) p Missing
n 18392 3054 2247 10521
Demographics
age (mean (SD)) 64.94 (12.42) 64.78 (11.89) 56.91 (9.26) 66.11 (12.59) <0.001 0.0
gender_f_1 = 1 (%) 4123 (22.4) 634 ( 20.8) 266 ( 11.8) 2641 ( 25.1) <0.001 0.0
Labs & eGFR
hdl (mean (SD)) 40.34 (11.72) 40.77 (12.04) 40.11 (11.40) 40.45 (11.60) 0.155 25.8
a1c (mean (SD)) 6.99 (1.77) 6.92 (1.80) 6.70 (1.89) 7.13 (1.72) <0.001 80.4
hgb (mean (SD)) 13.44 (1.84) 13.68 (1.79) 14.56 (1.46) 13.24 (1.82) <0.001 23.1
plt (mean (SD)) 236.04 (75.00) 233.28 (72.94) 249.68 (73.45) 234.14 (74.94) <0.001 4.6
creatinine_1 (mean (SD)) 1.11 (0.90) 1.07 (0.77) 0.95 (0.39) 1.14 (0.98) <0.001 0.0
tc (mean (SD)) 170.15 (46.02) 168.57 (46.98) 181.15 (43.68) 169.74 (45.57) <0.001 23.4
tg (mean (SD)) 161.18 (115.81) 163.14 (128.95) 163.39 (111.20) 159.04 (111.74) 0.149 24.4
ua (mean (SD)) 6.10 (1.74) 6.01 (1.65) 5.87 (1.48) 6.15 (1.77) <0.001 10.8
wbc (mean (SD)) 9.00 (6.02) 8.65 (4.00) 10.78 (3.70) 8.83 (6.87) <0.001 4.6
ldl (mean (SD)) 98.69 (39.26) 96.70 (39.51) 109.82 (38.34) 98.42 (38.74) <0.001 29.3
base_albumin (mean (SD)) 4.07 (0.45) 4.14 (0.43) 4.26 (0.40) 3.97 (0.44) <0.001 57.6
egfr (mean (SD)) 77.25 (24.04) 78.20 (23.09) 87.38 (17.96) 75.85 (24.59) <0.001 0.0
gfr_class (%) <0.001 0.0
eGFR > 60 14328 (77.9) 2433 ( 79.7) 2076 ( 92.4) 7965 ( 75.7)
eGFR 30-60 3193 (17.4) 512 ( 16.8) 153 ( 6.8) 1992 ( 18.9)
eGFR < 30 871 ( 4.7) 109 ( 3.6) 18 ( 0.8) 564 ( 5.4)
Comorbidities
dm = 1 (%) 8776 (47.7) 1442 ( 47.2) 672 ( 29.9) 5308 ( 50.5) <0.001 0.0
htn = 1 (%) 13951 (75.9) 2206 ( 72.2) 1151 ( 51.2) 8577 ( 81.5) <0.001 0.0
smok_hx = 1 (%) 7425 (40.4) 1306 ( 42.8) 1355 ( 60.3) 4065 ( 38.6) <0.001 0.0
chf = 1 (%) 5152 (28.0) 668 ( 21.9) 275 ( 12.2) 3509 ( 33.4) <0.001 0.0
prior_chf = 1 (%) 2525 (13.7) 277 ( 9.1) 167 ( 7.4) 1854 ( 17.6) <0.001 0.0
prior_copd = 1 (%) 1504 ( 8.2) 192 ( 6.3) 101 ( 4.5) 982 ( 9.3) <0.001 0.0
pvd = 1 (%) 1211 ( 6.6) 155 ( 5.1) 62 ( 2.8) 825 ( 7.8) <0.001 0.0
prior_af = 1 (%) 1541 (44.5) 159 ( 63.3) 61 ( 61.0) 1175 ( 44.2) <0.001 81.2
prior_malignancy = 1 (%) 1987 (10.8) 308 ( 10.1) 105 ( 4.7) 1252 ( 11.9) <0.001 0.0
subseq_malignancy = 1 (%) 1283 ( 7.0) 125 ( 4.1) 69 ( 3.1) 982 ( 9.3) <0.001 0.0
prior_stroke = 1 (%) 1257 ( 6.8) 215 ( 7.0) 24 ( 1.1) 845 ( 8.0) <0.001 0.0
subsequent_stroke = 1 (%) 1458 ( 7.9) 171 ( 5.6) 76 ( 3.4) 1060 ( 10.1) <0.001 0.0
Cardiac Presentation
m2s_lv = 1 (%) 2806 (15.3) 433 ( 14.2) 290 ( 12.9) 1648 ( 15.7) 0.002 0.0
stemi = 1 (%) 802 ( 4.4) 0 ( 0.0) 0 ( 0.0) 747 ( 7.1) <0.001 0.0
shock_yn (%) <0.001 88.0
0.5 1 ( 0.0) 1 ( 0.8) 0 ( 0.0) 0 ( 0.0)
NO 2099 (95.5) 120 ( 98.4) 669 ( 99.3) 1239 ( 95.6)
YES 99 ( 4.5) 1 ( 0.8) 5 ( 0.7) 57 ( 4.4)
cath4mi = 1 (%) 10445 (56.8) 2568 ( 84.1) 2156 ( 96.0) 4727 ( 44.9) <0.001 0.0
cath4acs = 1 (%) 8244 (44.8) 544 ( 17.8) 107 ( 4.8) 5985 ( 56.9) <0.001 0.0
sp_cabg = 1 (%) 2263 (12.3) 269 ( 8.8) 66 ( 2.9) 1571 ( 14.9) <0.001 0.0
prox_lad = 1 (%) 4074 (22.2) 765 ( 25.0) 585 ( 26.0) 2203 ( 20.9) <0.001 0.0
unprot_lm = 1 (%) 615 ( 3.3) 160 ( 5.2) 41 ( 1.8) 331 ( 3.1) <0.001 0.0
Vessels Treated
cx_treated (mean (SD)) 1.30 (0.55) 1.31 (0.55) 1.31 (0.58) 1.31 (0.56) 0.979 68.8
lad_treated (mean (SD)) 1.39 (0.61) 1.43 (0.62) 1.46 (0.65) 1.39 (0.61) 0.001 54.1
lm_treated (mean (SD)) 1.02 (0.14) 1.02 (0.15) 1.00 (0.00) 1.02 (0.15) 0.583 94.9
rca_treated (mean (SD)) 1.37 (0.57) 1.44 (0.57) 1.41 (0.56) 1.36 (0.58) 0.001 67.6
Procedure Details
sum_stnt (mean (SD)) 1.71 (1.10) 1.93 (1.20) 1.79 (1.09) 1.69 (1.06) <0.001 0.0
bif = 1 (%) 1779 (13.1) 543 ( 18.0) 246 ( 11.2) 744 ( 12.0) <0.001 26.2
calcif = 1 (%) 1425 (10.5) 337 ( 11.2) 108 ( 4.9) 740 ( 11.9) <0.001 26.2
diffc = 1 (%) 2152 (15.9) 438 ( 14.5) 338 ( 15.4) 1035 ( 16.7) 0.021 26.2
i_ib3a = 1 (%) 1352 (10.0) 163 ( 5.4) 371 ( 16.9) 767 ( 12.4) <0.001 26.2
long = 1 (%) 2557 (18.8) 634 ( 21.0) 330 ( 15.0) 1226 ( 19.8) <0.001 26.2
radial = 1 (%) 8614 (46.8) 2448 ( 80.2) 1670 ( 74.3) 2772 ( 26.3) <0.001 0.0
Scores
cha2ds2vasc (mean (SD)) 3.61 (1.67) 3.51 (1.62) 2.37 (1.18) 3.86 (1.68) <0.001 0.0
Medications
anti_coag = 1 (%) 1595 ( 8.7) 34 ( 1.1) 11 ( 0.5) 1425 ( 13.5) <0.001 0.0
aspirin = 1 (%) 15307 (95.6) 3006 ( 98.4) 2243 ( 99.8) 9887 ( 94.0) <0.001 13.0
other_anti_platelet = 1 (%) 15831 (98.9) 3054 (100.0) 2246 (100.0) 10521 (100.0) 0.049 13.0
oac = 1 (%) 1225 ( 7.7) 22 ( 0.7) 4 ( 0.2) 1183 ( 11.2) <0.001 13.0
ace_arb = 1 (%) 13510 (84.4) 2597 ( 85.0) 2043 ( 90.9) 8729 ( 83.0) <0.001 13.0
bb = 1 (%) 13238 (82.7) 2500 ( 81.9) 1972 ( 87.8) 8619 ( 81.9) <0.001 13.0
statin = 1 (%) 15651 (97.8) 3019 ( 98.9) 2236 ( 99.5) 10229 ( 97.2) <0.001 13.0
diuretics = 1 (%) 4085 (25.5) 641 ( 21.0) 370 ( 16.5) 3009 ( 28.6) <0.001 13.0
vka = 1 (%) 736 ( 4.6) 8 ( 0.3) 4 ( 0.2) 710 ( 6.7) <0.001 13.0
noac = 1 (%) 490 ( 3.1) 14 ( 0.5) 0 ( 0.0) 474 ( 4.5) <0.001 13.0
insulin = 1 (%) 2083 (13.0) 543 ( 17.8) 190 ( 8.5) 1322 ( 12.6) <0.001 13.0
clopidogrel = 1 (%) 10536 (65.8) 14 ( 0.5) 1 ( 0.0) 10521 (100.0) <0.001 13.0
ticagrelor = 1 (%) 3054 (19.1) 3054 (100.0) 0 ( 0.0) 0 ( 0.0) <0.001 13.0
prasugrel = 1 (%) 2250 (14.1) 3 ( 0.1) 2247 (100.0) 0 ( 0.0) <0.001 13.0
thyroxin = 1 (%) 952 ( 5.9) 162 ( 5.3) 68 ( 3.0) 710 ( 6.7) <0.001 13.0
arni = 1 (%) 62 ( 0.4) 22 ( 0.7) 13 ( 0.6) 27 ( 0.3) <0.001 13.0
sglt2 = 1 (%) 584 ( 3.6) 258 ( 8.4) 135 ( 6.0) 190 ( 1.8) <0.001 13.0
Outcomes
death = 1 (%) 5673 (30.8) 467 ( 15.3) 146 ( 6.5) 4311 ( 41.0) <0.001 0.0
any_mace = 1 (%) 9258 (50.3) 1021 ( 33.4) 560 ( 24.9) 6552 ( 62.3) <0.001 0.0
Note:
Values are mean (SD) for continuous variables and n (%) for categorical variables. Missing % shown where applicable. P-values reflect comparison across all three P2Y12 groups.

3 Survival Analysis: P2Y12 Comparisons

3.1 How to read the Cox regression tables

Each table below shows results from a Cox proportional hazards model — a statistical model that estimates the Hazard Ratio (HR) of death between two drug groups over the follow-up period.

  • HR < 1: the drug listed is associated with lower mortality compared to the reference
  • HR > 1: the drug listed is associated with higher mortality compared to the reference
  • 95% CI: the confidence interval — if it crosses 1.0, the result is not statistically significant
  • p-value: < 0.05 is considered statistically significant

The reference drug (the comparator) varies by section and is noted in each subsection.

3.2 Effient (Prasugrel) vs. Brilinta (Ticagrelor)

Clinical context: Prasugrel and ticagrelor are both considered more potent than clopidogrel. Guidelines generally favor these agents in ACS patients who can tolerate them. This comparison asks: among the two more potent agents, does one perform better than the other — particularly in patients with reduced kidney function?

Reference group: Brilinta (ticagrelor). An HR < 1 means Effient is associated with lower mortality.

3.2.1 Interaction Model (Full Cohort)

This model tests whether eGFR modifies the relationship between drug choice and mortality — i.e., does the relative benefit of Effient vs. Brilinta change depending on kidney function? A significant interaction term (p2y12:egfr) would suggest yes.

##                               exp(coef) [confint] p     
## p2y12Effient (Prasugrel)      0.37 [0.20, 0.71]    0.003
## egfr                          0.97 [0.96, 0.97]   <0.001
## p2y12Effient (Prasugrel):egfr 1.00 [1.00, 1.01]    0.389
Table 2a. Effient vs. Brilinta — Interaction Model (CKD-EPI, primary analysis)
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.37 [0.20, 0.71] 0.003
egfr 0.97 [0.96, 0.97] <0.001
p2y12Effient (Prasugrel):egfr 1.00 [1.00, 1.01] 0.389
Note:
The interaction term (p2y12:egfr) tests whether the effect of drug choice on mortality differs across levels of kidney function. eGFR calculated using CKD-EPI.

3.2.2 Sensitivity: MDRD Interaction Model

This replicates the interaction model using MDRD-derived eGFR instead of CKD-EPI, as a sanity check to confirm the code is producing the same result as the original analysis. If the interaction p-value here matches what you previously obtained, the code is correct and any difference in the CKD-EPI model above reflects a genuine formula effect.

##                                    exp(coef) [confint] p     
## p2y12Effient (Prasugrel)           0.22 [0.11, 0.43]   <0.001
## egfr_mdrd                          0.97 [0.97, 0.97]   <0.001
## p2y12Effient (Prasugrel):egfr_mdrd 1.01 [1.00, 1.02]    0.041
Table 2a (Sensitivity). Effient vs. Brilinta — Interaction Model (MDRD, for verification only)
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.22 [0.11, 0.43] <0.001
egfr_mdrd 0.97 [0.97, 0.97] <0.001
p2y12Effient (Prasugrel):egfr_mdrd 1.01 [1.00, 1.02] 0.041
Note:
Sensitivity check only. eGFR calculated using MDRD4. Compare interaction p-value to previously obtained result to verify code integrity. Primary analysis uses CKD-EPI (table above).

3.2.3 Overall — Unadjusted

Unadjusted comparison across the entire cohort, without stratifying by kidney function.

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.35 [0.29, 0.43]   <0.001
Table 2b. Effient vs. Brilinta — Overall, Unadjusted
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.35 [0.29, 0.43] <0.001
Note:
Reference: Brilinta (Ticagrelor). HR < 1 favors Effient. No adjustment for confounders.

3.2.4 Overall — Multivariable

Adjusted for key confounders: age, sex, diabetes, and eGFR (continuous). These variables were selected because they are established predictors of mortality in ACS.

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.75 [0.61, 0.92]    0.007
## age                      1.05 [1.04, 1.06]   <0.001
## gender_f_1               0.87 [0.71, 1.05]    0.148
## dm                       1.88 [1.59, 2.23]   <0.001
## egfr                     0.98 [0.97, 0.98]   <0.001
Table 2c. Effient vs. Brilinta — Overall, Multivariable
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.75 [0.61, 0.92] 0.007
age 1.05 [1.04, 1.06] <0.001
gender_f_1 0.87 [0.71, 1.05] 0.148
dm 1.88 [1.59, 2.23] <0.001
egfr 0.98 [0.97, 0.98] <0.001
Note:
Reference: Brilinta (Ticagrelor). Adjusted for age, sex, diabetes, and eGFR (continuous).

3.2.5 Subgroup: eGFR > 60 — Unadjusted

Patients with near-normal kidney function.

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.48 [0.38, 0.60]   <0.001
Table 2d. Effient vs. Brilinta — eGFR > 60, Unadjusted
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.48 [0.38, 0.60] <0.001
Note:
Reference: Brilinta (Ticagrelor).

3.2.6 Subgroup: eGFR > 60 — Multivariable

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.84 [0.66, 1.08]    0.171
## age                      1.07 [1.06, 1.08]   <0.001
## gender_f_1               0.83 [0.63, 1.09]    0.172
## dm                       2.20 [1.78, 2.72]   <0.001
Table 2e. Effient vs. Brilinta — eGFR > 60, Multivariable
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.84 [0.66, 1.08] 0.171
age 1.07 [1.06, 1.08] <0.001
gender_f_1 0.83 [0.63, 1.09] 0.172
dm 2.20 [1.78, 2.72] <0.001
Note:
Reference: Brilinta (Ticagrelor). Adjusted for age, sex, and diabetes. eGFR excluded from subgroup models as patients are already stratified by eGFR class.

3.2.7 Subgroup: eGFR 30–60 — Unadjusted

Patients with moderate CKD (stage 3).

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.25 [0.15, 0.41]   <0.001
Table 2f. Effient vs. Brilinta — eGFR 30–60, Unadjusted
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.25 [0.15, 0.41] <0.001
Note:
Reference: Brilinta (Ticagrelor). Moderate CKD.

3.2.8 Subgroup: eGFR 30–60 — Multivariable

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.42 [0.24, 0.72]    0.002
## age                      1.05 [1.03, 1.06]   <0.001
## gender_f_1               0.85 [0.61, 1.18]    0.333
## dm                       1.22 [0.91, 1.65]    0.182
Table 2g. Effient vs. Brilinta — eGFR 30–60, Multivariable
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.42 [0.24, 0.72] 0.002
age 1.05 [1.03, 1.06] <0.001
gender_f_1 0.85 [0.61, 1.18] 0.333
dm 1.22 [0.91, 1.65] 0.182
Note:
Reference: Brilinta (Ticagrelor). Adjusted for age, sex, and diabetes.

3.2.9 Subgroup: eGFR < 30 — Unadjusted

Patients with severely reduced kidney function.

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.93 [0.48, 1.80]    0.829
Table 2h. Effient vs. Brilinta — eGFR
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.93 [0.48, 1.80] 0.829
Note:
Reference: Brilinta (Ticagrelor). Severe CKD.

3.2.10 Subgroup: eGFR < 30 — Multivariable

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 1.22 [0.62, 2.39]    0.572
## age                      1.03 [1.00, 1.06]    0.020
## gender_f_1               0.72 [0.43, 1.20]    0.206
## dm                       1.99 [1.03, 3.86]    0.041
Table 2i. Effient vs. Brilinta — eGFR
exp(coef) [confint] p
p2y12Effient (Prasugrel) 1.22 [0.62, 2.39] 0.572
age 1.03 [1.00, 1.06] 0.020
gender_f_1 0.72 [0.43, 1.20] 0.206
dm 1.99 [1.03, 3.86] 0.041
Note:
Reference: Brilinta (Ticagrelor). Adjusted for: age, gender_f_1, dm. Covariates with no variation in this subgroup were automatically excluded. Interpret with caution — small sample size in the eGFR < 30 subgroup limits statistical power.

Note for abstract writing: The primary analysis for this paper is the Effient vs. Brilinta comparison above (Tables 2a–2i). The comparisons below (Plavix vs. Brilinta, Plavix vs. Effient) are secondary analyses included for completeness.

3.3 Plavix (Clopidogrel) vs. Brilinta (Ticagrelor)

Reference group: Brilinta (ticagrelor). An HR < 1 means Plavix is associated with lower mortality.

3.3.1 Interaction Model (Full Cohort)

##                                exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel)      1.32 [1.03, 1.69]    0.027
## egfr                           0.97 [0.96, 0.97]   <0.001
## p2y12Plavix (Clopidogrel):egfr 1.00 [1.00, 1.00]    0.820
Table 3a. Plavix vs. Brilinta — Interaction Model (eGFR as continuous variable)
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.32 [1.03, 1.69] 0.027
egfr 0.97 [0.96, 0.97] <0.001
p2y12Plavix (Clopidogrel):egfr 1.00 [1.00, 1.00] 0.820
Note:
The interaction term tests whether the drug effect on mortality differs by kidney function level.

3.3.2 Subgroup: eGFR > 60

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 1.38 [1.21, 1.58]   <0.001
Table 3b. Plavix vs. Brilinta — Subgroup: eGFR > 60 ml/min/1.73m²
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.38 [1.21, 1.58] <0.001
Note:
Reference: Brilinta (Ticagrelor).

3.3.3 Subgroup: eGFR 30–60

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 1.29 [1.10, 1.52]    0.002
Table 3c. Plavix vs. Brilinta — Subgroup: eGFR 30–60 ml/min/1.73m²
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.29 [1.10, 1.52] 0.002
Note:
Reference: Brilinta (Ticagrelor). Moderate CKD.

3.3.4 Subgroup: eGFR < 30

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 1.54 [1.16, 2.04]    0.003
Table 3d. Plavix vs. Brilinta — Subgroup: eGFR
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.54 [1.16, 2.04] 0.003
Note:
Reference: Brilinta (Ticagrelor). Severe CKD — this is the most clinically relevant subgroup for this comparison.

3.4 Plavix (Clopidogrel) vs. Effient (Prasugrel)

Reference group: Effient (prasugrel). An HR < 1 means Plavix is associated with lower mortality.

3.4.1 Interaction Model (Full Cohort)

##                                exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel)      3.28 [1.80, 5.96]   <0.001
## egfr                           0.97 [0.96, 0.98]   <0.001
## p2y12Plavix (Clopidogrel):egfr 1.00 [0.99, 1.01]    0.527
Table 4a. Plavix vs. Effient — Interaction Model (eGFR as continuous variable)
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 3.28 [1.80, 5.96] <0.001
egfr 0.97 [0.96, 0.98] <0.001
p2y12Plavix (Clopidogrel):egfr 1.00 [0.99, 1.01] 0.527
Note:
The interaction term tests whether the drug effect on mortality differs by kidney function level.

3.4.2 Subgroup: eGFR > 60

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 2.87 [2.38, 3.47]   <0.001
Table 4b. Plavix vs. Effient — Subgroup: eGFR > 60 ml/min/1.73m²
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 2.87 [2.38, 3.47] <0.001
Note:
Reference: Effient (Prasugrel).

3.4.3 Subgroup: eGFR 30–60

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 4.80 [3.02, 7.65]   <0.001
Table 4c. Plavix vs. Effient — Subgroup: eGFR 30–60 ml/min/1.73m²
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 4.80 [3.02, 7.65] <0.001
Note:
Reference: Effient (Prasugrel). Moderate CKD.

3.4.4 Subgroup: eGFR < 30

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 1.38 [0.80, 2.40]    0.249
Table 4d. Plavix vs. Effient — Subgroup: eGFR
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.38 [0.80, 2.40] 0.249
Note:
Reference: Effient (Prasugrel). Severe CKD — interpret with caution given small sample sizes expected in this subgroup.

4 Secondary Outcome: MACE

What is MACE? Major Adverse Cardiovascular Events (MACE) is a composite outcome. It is a broader measure of cardiovascular harm than mortality alone, and is the standard primary endpoint in meny cardiology trials. The analyses below mirror the mortality analyses exactly, using any_mace as the event and f_umace as the follow-up time.

4.1 Effient (Prasugrel) vs. Brilinta (Ticagrelor) — MACE

Reference group: Brilinta (ticagrelor). An HR < 1 means Effient is associated with lower MACE.

4.1.1 Interaction Model (Full Cohort)

##                               exp(coef) [confint] p     
## p2y12Effient (Prasugrel)      0.35 [0.22, 0.55]   <0.001
## egfr                          0.99 [0.98, 0.99]   <0.001
## p2y12Effient (Prasugrel):egfr 1.01 [1.00, 1.01]    0.001
Table M-2a. Effient vs. Brilinta — MACE Interaction Model (CKD-EPI)
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.35 [0.22, 0.55] <0.001
egfr 0.99 [0.98, 0.99] <0.001
p2y12Effient (Prasugrel):egfr 1.01 [1.00, 1.01] 0.001
Note:
The interaction term (p2y12:egfr) tests whether the effect of drug choice on MACE differs across levels of kidney function.

4.1.2 Overall — Unadjusted

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.65 [0.59, 0.73]   <0.001
Table M-2b. Effient vs. Brilinta — MACE Overall, Unadjusted
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.65 [0.59, 0.73] <0.001
Note:
Reference: Brilinta (Ticagrelor). HR < 1 favors Effient.

4.1.3 Overall — Multivariable

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.81 [0.72, 0.90]   <0.001
## age                      1.01 [1.00, 1.01]    0.039
## gender_f_1               0.86 [0.75, 0.98]    0.024
## dm                       2.01 [1.82, 2.23]   <0.001
## egfr                     0.99 [0.99, 0.99]   <0.001
Table M-2c. Effient vs. Brilinta — MACE Overall, Multivariable
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.81 [0.72, 0.90] <0.001
age 1.01 [1.00, 1.01] 0.039
gender_f_1 0.86 [0.75, 0.98] 0.024
dm 2.01 [1.82, 2.23] <0.001
egfr 0.99 [0.99, 0.99] <0.001
Note:
Reference: Brilinta (Ticagrelor). Adjusted for age, sex, diabetes, and eGFR (continuous).

4.1.4 Subgroup: eGFR > 60 — Unadjusted

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.75 [0.67, 0.84]   <0.001
Table M-2d. Effient vs. Brilinta — MACE, eGFR > 60, Unadjusted
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.75 [0.67, 0.84] <0.001
Note:
Reference: Brilinta (Ticagrelor).

4.1.5 Subgroup: eGFR > 60 — Multivariable

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.85 [0.76, 0.96]    0.010
## age                      1.01 [1.00, 1.01]    0.025
## gender_f_1               0.82 [0.70, 0.97]    0.020
## dm                       2.12 [1.89, 2.38]   <0.001
Table M-2e. Effient vs. Brilinta — MACE, eGFR > 60, Multivariable
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.85 [0.76, 0.96] 0.010
age 1.01 [1.00, 1.01] 0.025
gender_f_1 0.82 [0.70, 0.97] 0.020
dm 2.12 [1.89, 2.38] <0.001
Note:
Reference: Brilinta (Ticagrelor). Adjusted for age, sex, and diabetes.

4.1.6 Subgroup: eGFR 30–60 — Unadjusted

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.50 [0.36, 0.68]   <0.001
Table M-2f. Effient vs. Brilinta — MACE, eGFR 30–60, Unadjusted
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.50 [0.36, 0.68] <0.001
Note:
Reference: Brilinta (Ticagrelor). Moderate CKD.

4.1.7 Subgroup: eGFR 30–60 — Multivariable

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.62 [0.43, 0.89]    0.009
## age                      1.02 [1.00, 1.03]    0.010
## gender_f_1               0.82 [0.62, 1.08]    0.158
## dm                       1.40 [1.09, 1.79]    0.008
Table M-2g. Effient vs. Brilinta — MACE, eGFR 30–60, Multivariable
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.62 [0.43, 0.89] 0.009
age 1.02 [1.00, 1.03] 0.010
gender_f_1 0.82 [0.62, 1.08] 0.158
dm 1.40 [1.09, 1.79] 0.008
Note:
Reference: Brilinta (Ticagrelor). Adjusted for age, sex, and diabetes.

4.1.8 Subgroup: eGFR < 30 — Unadjusted

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.70 [0.37, 1.33]    0.278
Table M-2h. Effient vs. Brilinta — MACE, eGFR
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.70 [0.37, 1.33] 0.278
Note:
Reference: Brilinta (Ticagrelor). Severe CKD.

4.1.9 Subgroup: eGFR < 30 — Multivariable

##                          exp(coef) [confint] p     
## p2y12Effient (Prasugrel) 0.83 [0.43, 1.60]    0.580
## age                      1.02 [0.99, 1.04]    0.169
## gender_f_1               0.78 [0.49, 1.24]    0.289
## dm                       1.83 [1.02, 3.29]    0.043
Table M-2i. Effient vs. Brilinta — MACE, eGFR
exp(coef) [confint] p
p2y12Effient (Prasugrel) 0.83 [0.43, 1.60] 0.580
age 1.02 [0.99, 1.04] 0.169
gender_f_1 0.78 [0.49, 1.24] 0.289
dm 1.83 [1.02, 3.29] 0.043
Note:
Reference: Brilinta (Ticagrelor). Adjusted for: age, gender_f_1, dm. Covariates with no variation in this subgroup were automatically excluded. Interpret with caution — small sample size limits statistical power.

Note: The MACE analyses below (Plavix vs. Brilinta, Plavix vs. Effient) are secondary. The primary focus remains the Effient vs. Brilinta comparison above.

4.2 Plavix (Clopidogrel) vs. Brilinta (Ticagrelor) — MACE

Reference group: Brilinta (ticagrelor). An HR < 1 means Plavix is associated with lower MACE.

4.2.1 Interaction Model

##                                exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel)      1.60 [1.30, 1.99]   <0.001
## egfr                           0.99 [0.98, 0.99]   <0.001
## p2y12Plavix (Clopidogrel):egfr 1.00 [0.99, 1.00]    0.060
Table M-3a. Plavix vs. Brilinta — MACE Interaction Model
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.60 [1.30, 1.99] <0.001
egfr 0.99 [0.98, 0.99] <0.001
p2y12Plavix (Clopidogrel):egfr 1.00 [0.99, 1.00] 0.060
Note:
The interaction term tests whether the drug effect on MACE differs by kidney function level.

4.2.2 Subgroup: eGFR > 60

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 1.32 [1.22, 1.43]   <0.001
Table M-3b. Plavix vs. Brilinta — MACE, eGFR > 60
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.32 [1.22, 1.43] <0.001
Note:
Reference: Brilinta (Ticagrelor).

4.2.3 Subgroup: eGFR 30–60

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 1.35 [1.18, 1.55]   <0.001
Table M-3c. Plavix vs. Brilinta — MACE, eGFR 30–60
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.35 [1.18, 1.55] <0.001
Note:
Reference: Brilinta (Ticagrelor). Moderate CKD.

4.2.4 Subgroup: eGFR < 30

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 1.22 [0.95, 1.57]    0.111
Table M-3d. Plavix vs. Brilinta — MACE, eGFR
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.22 [0.95, 1.57] 0.111
Note:
Reference: Brilinta (Ticagrelor). Severe CKD.

4.3 Plavix (Clopidogrel) vs. Effient (Prasugrel) — MACE

Reference group: Effient (prasugrel). An HR < 1 means Plavix is associated with lower MACE.

4.3.1 Interaction Model

##                                exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel)      4.43 [2.95, 6.66]   <0.001
## egfr                           1.00 [0.99, 1.00]    0.041
## p2y12Plavix (Clopidogrel):egfr 0.99 [0.98, 0.99]   <0.001
Table M-4a. Plavix vs. Effient — MACE Interaction Model
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 4.43 [2.95, 6.66] <0.001
egfr 1.00 [0.99, 1.00] 0.041
p2y12Plavix (Clopidogrel):egfr 0.99 [0.98, 0.99] <0.001
Note:
The interaction term tests whether the drug effect on MACE differs by kidney function level.

4.3.2 Subgroup: eGFR > 60

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 1.72 [1.56, 1.88]   <0.001
Table M-4b. Plavix vs. Effient — MACE, eGFR > 60
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.72 [1.56, 1.88] <0.001
Note:
Reference: Effient (Prasugrel).

4.3.3 Subgroup: eGFR 30–60

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 2.65 [1.97, 3.56]   <0.001
Table M-4c. Plavix vs. Effient — MACE, eGFR 30–60
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 2.65 [1.97, 3.56] <0.001
Note:
Reference: Effient (Prasugrel). Moderate CKD.

4.3.4 Subgroup: eGFR < 30

##                           exp(coef) [confint] p     
## p2y12Plavix (Clopidogrel) 1.58 [0.91, 2.75]    0.102
Table M-4d. Plavix vs. Effient — MACE, eGFR
exp(coef) [confint] p
p2y12Plavix (Clopidogrel) 1.58 [0.91, 2.75] 0.102
Note:
Reference: Effient (Prasugrel). Severe CKD.

5 Visualization: Hazard Ratio Across the eGFR Spectrum

The following plot shows how the Hazard Ratio between Effient and Brilinta changes continuously across the range of eGFR values. Rather than just comparing fixed subgroups, this approach uses a sliding window (each point represents patients within ±10 ml/min/1.73m² of that eGFR value), providing a more nuanced view of how kidney function modifies the drug effect.

  • The blue line is the estimated HR at each eGFR value
  • The shaded blue ribbon is the 95% confidence interval
  • The dashed red line at HR = 1.0 is the null (no difference)
  • Green points indicate statistically significant differences (p < 0.05); grey points are non-significant
Figure 1. Hazard Ratio for Effient vs. Brilinta across the eGFR spectrum. Each point represents a localized Cox model fit within a ±10 ml/min/1.73m² window around that eGFR value.

Figure 1. Hazard Ratio for Effient vs. Brilinta across the eGFR spectrum. Each point represents a localized Cox model fit within a ±10 ml/min/1.73m² window around that eGFR value.

6 Abstract Draft — ESC Format

Title: Prasugrel versus Ticagrelor in Acute Coronary Syndrome Patients with Chronic Kidney Disease: A Registry-Based Analysis

Background:

Chronic kidney disease (CKD) affects platelet function, drug metabolism, and cardiovascular outcomes. Among patients with acute coronary syndrome (ACS) undergoing percutaneous coronary intervention (PCI), the optimal P2Y12 inhibitor in the setting of reduced renal function remains uncertain. Current guidelines provide limited guidance on the comparative effectiveness of prasugrel versus ticagrelor specifically in CKD patients.

Purpose:

To compare all-cause mortality and mace between prasugrel and ticagrelor across strata of renal function in a real-world ACS-PCI population.

Methods:

We analyzed [N] consecutive ACS patients who underwent PCI, identified from a prospective institutional PCI registry. Estimated GFR was calculated using the CKD-EPI equation and patients were stratified into three groups: eGFR >60, 30–60, and <30 ml/min/1.73m². The primary outcome was all-cause mortality. Cox proportional hazards models were used to estimate hazard ratios (HR) with 95% confidence intervals, comparing prasugrel to ticagrelor (reference). Multivariable models were adjusted for age, sex, and diabetes mellitus.

Results:

Among [N] patients receiving prasugrel or ticagrelor, [N prasugrel] received prasugrel and [N ticagrelor] received ticagrelor. Median follow-up was [X] months….

Conclusion(s):

In this real-world ACS-PCI registry, prasugrel was associated with lower mortality compared to ticagrelor….

ESC submission checklist before you submit:

Report generated on February 20, 2026 from the PCI Registry database.