A meta-analysis of low-dose aspirin for the prevention of pregnancy-induced hypertensive disease

Abstract

Background. —Pregnancy-induced hypertension (PIH), defined as either isolated hypertension after the 20th week of gestation or hypertension with proteinuria (preeclampsia), occurs in 5% to 15% of pregnancies and is associated with maternal and neonatal morbidity. Previous clinical trials with small numbers of patients have suggested that aspirin in doses of 60 to 150 mg/d during the second and third trimesters reduces the risk of PIH and improves maternal and neonatal outcomes. Objective. —We performed a meta-analysis of the six published controlled trials to estimate more precisely (1) the magnitude of protection of aspirin from PIH; (2) the effect of aspirin on severe low-birth-weight infants, cesarean section, and perinatal mortality; and (3) the risk of adverse effects. Methods. —We critically and independently evaluated study methods, assigned a quality score to each trial, and abstracted quantitative outcomes data. For each outcome, both relative risk (RR) and the number needed to be treated were calculated. Results. —Among 394 subjects from six trials, the RR of PIH among women who took aspirin was 0.35 (95% confidence interval [Cl], 0.22 to 0.55) and the number needed to be treated was 4.4, meaning that between four and five highrisk women would need to be treated with aspirin to prevent one case of PIH. Aspirin reduced the risk of severe low birth weight among newborns by 44% (RR = 0.56; 95% Cl, 0.36 to 0.88) and reduced the risk of cesarean section by 66% overall (RR = 0.34; 95% Cl, 0.25 to 0.48), although the specific indications for cesarean section were generally not described. There was no effect on fetal and neonatal death (RR=0.88; 95% Cl, 0.32 to 2.46), and there were no maternal or neonatal adverse effects associated with taking aspirin. Conclusion. —This meta-analysis suggests that low-dose aspirin reduces the risks of PIH and severe low birth weight, with no observed risk of maternal or neonatal adverse effects. (JAMA. 1991;266:261-265)