Eleven randomised controlled trials (RCTs) were included, nine of which were double-blind (n=14,833)
[A: Notice of correction. It has come to our attention that the data in 2 of the 11 references (nos.4 and 12) that met our predetermined meta-analysis criteria may not be valid. We would note that the senior author of these reports did answer our questions, and he provided precise data in written correspondence to us. However, because of this concern, we repeated our meta-analyses without these two publications, and added data from a recent randomised trial in Brazil that also met our criteria. We found results similar to those in our published report. The incidence of abruptio placentae in these 10 trials of low-dose aspirin was 119 out of 7,845 women (1.52%) assigned to aspirin and 102 out of 7,694 women (1.32%) assigned to placebo (x squared, p=0.314 Fischer exact test, two-tailed, p=0.343). Perinatal death occurred in 215 out of 8,038 cases (2.7%) assigned to aspirin and in 212 out of 7,940 cases (2.7%) assigned to placebo (x squared, p=0.0946; Fischer exact test, two-tailed , p=0.961).]
The incidence of abruptio placentae in the 7,493 women randomised to aspirin treatment (1.67%) was similar to that in the 7,430 women randomised to a placebo or no treatment group (1.43%) (p=0.24). Only 2 of the 11 studies reported a statistically significant difference, one showing a reduction in abruptions associated with aspirin and the other an increase with aspirin use.
Review of perinatal mortality in the 11 randomised trials of low-dose aspirin in pregnancy confirmed that the mortality was similar in both groups (p=0.42), with only 1 of the 11 trials reporting a significant benefit in the aspirin group.