Background: Aspirin is believed to improve the outcome of IVF, but previous conventional meta-analyses on the subject are conflicting. Therefore, we performed a meta-analysis with individual patient data (IPD MA) of randomized clinical trials (RCTs) on the subject.
Methods: A systematic literature search was conducted to identify RCTs assessing the effectiveness of aspirin in IVF. Authors were asked to share their original data. In a one step meta-analytic approach, the treatment effect of aspirin was estimated with odds ratios (ORs) and 95% confidence intervals (CIs) using logistic regression, based on the intention to treat principle.
Results: Ten studies fulfilled the inclusion criteria. Authors of six studies provided IPD, including 1119 patients (562 placebo and 557 aspirin). There were 160 clinical pregnancies in the aspirin (28.8%) and 179 (31.9%) in the placebo group [OR 0.86, 95% CI (0.69-1.1)]. There were 129 ongoing pregnancies in the aspirin (23.6%) and 147 in the placebo group (26.7%) [OR 0.85, 95% CI (0.65-1.1)]. Whereas the conventional meta-analysis limited to studies that could provide IPD showed an OR of 0.89 (95% CI 0.69-1.2), the conventional meta-analysis limited to the eight studies of which method of randomization could be confirmed showed an OR of 0.94 (95% CI 0.76-1.17) and the conventional meta-analysis including all 10 eligible RCTs identified with our search changed the OR to 1.07 (95% CI 0.81-1.41). This difference in direction of effect, derived from the studies not able to share IPD of which quality of randomization could not be confirmed.
Conclusions: Aspirin does not improve pregnancy rates after IVF.