One-year outcomes
| Prasugrel | Ticagrelor | ||
| Patients, n | 2073 | 35 917 | |
| Events, n (%) | HR (95% CI) | ||
| MACCE | |||
| Crude | 127 (6.1) | 2196 (6.1) | 1.00 (0.84 to 1.20) |
| MV analysis | 1.03 (0.86 to 1.24) | ||
| IPTW weighting | 1.11 (0.87 to 1.40) | ||
| PSM cohort* | 127 (6.1) | 122 (5.9) | 1.04 (0.81 to 1.33) |
| NACCE | |||
| Crude | 174 (8.4) | 3130 (8.7) | 0.96 (0.82 to 1.12) |
| MV analysis | 1.03 (0.88 to 1.20) | ||
| IPTW weighting | 1.12 (0.91 to 1.37) | ||
| PSM cohort* | 174 (8.4) | 169 (8.2) | 1.02 (0.83 to 1.27) |
| All-cause mortality | |||
| Crude | 48 (2.3) | 1056 (2.9) | 0.79 (0.59 to 1.05) |
| MV analysis | 0.89 (0.67 to 1.20) | ||
| IPTW weighting | 1.04 (0.69 to 1.56) | ||
| PSM cohort* | 48 (2.3) | 59 (2.8) | 0.81 (0.55 to 1.19) |
| Myocardial infarction | |||
| Crude | 85 (4.1) | 1123 (3.2) | 1.32 (1.06 to 1.64) |
| MV analysis | 1.26 (0.98 to 1.58) | ||
| IPTW weighting | 1.26 (0.95 to 1.67) | ||
| PSM cohort* | 85 (4.1) | 66 (3.2) | 1.26 (0.91 to 1.77) |
| Stroke | |||
| Crude | 18 (0.9) | 385 (1.1) | 0.81 (0.50 to 1.30) |
| MV analysis | 0.94 (0.58 to 1.53) | ||
| IPTW weighting | 0.93 (0.52 to 1.67) | ||
| PSM cohort* | 18 (0.9) | 14 (0.7) | 1.28 (0.64 to 2.58) |
| Major bleeding | |||
| Crude | 51 (2.5) | 1124 (3.2) | 0.78 (0.59 to 1.03) |
| MV analysis | 0.92 (0.69 to 1.22) | ||
| IPTW weighting | 1.02 (0.72 to 1.47) | ||
| PSM cohort* | 51 (2.5) | 51 (2.5) | 0.99 (0.67 to 1.46) |
MACCE including all-cause death, myocardial infarction or stroke (ischaemic and haemorrhagic); NACCE including MACCE and major bleeding during follow-up.
HR with 95% CI was derived from Cox regression analysis.
In the unadjusted model (crude) only treatment was included as covariate. In the multivariable model 34 additional covariates were included. Using the same covariates, the individual propensity score, reflecting the probability to be treated with prasugrel, and propensity score weights (IPTW) were calculated. IPTW Cox regression models were constructed.
*Propensity matching resulted in a population of 4142 patients (PSM cohort), 2071 in each group.
IPTW, inverse probability of treatment weighting; MACCE, major adverse cardiac and cerebrovascular events; MV, multivariable model; NACCE, net adverse cardiac and cerebrovascular events; PSM, propensity score matched.