Abstract

Objective Metformin has been reported to reduce the risk of pre‐eclampsia. It is also known to influence soluble fms‐like tyrosine kinase‐1 level, which correlates significantly with the gestational age at onset and severity of pre‐eclampsia. The main aim of this systematic review and meta‐analysis of randomized trials was to determine whether metformin use is associated with the incidence of hypertensive disorders of pregnancy (HDP). Methods MEDLINE (1947 to September 2017), Scopus (1970 to September 2017) and the Cochrane Library (inception to September 2017) were searched for relevant citations in the English language. Only randomized controlled trials on metformin use, reporting the incidence of pre‐eclampsia or pregnancy‐induced hypertension, were included. Studies on populations with a high probability of metformin use prior to randomization (those with type II diabetes or polycystic ovary syndrome) were excluded. Random‐effects models with the Mantel–Haenszel method were used for subgroup analyses. Bayesian random‐effects meta‐regression was used to summarize the evidence. Results In total, 3337 citations matched the search criteria. After evaluating 2536 abstracts and performing full‐text review of 52 studies, 15 were included in the review. In women with gestational diabetes, metformin use was associated with a reduced risk of pregnancy‐induced hypertension when compared with insulin (relative risk (RR), 0.56; 95% CI, 0.37–0.85; I2 = 0%; 1260 women) and a non‐significantly reduced risk of pre‐eclampsia (RR, 0.83; 95% CI, 0.60–1.14; I2 = 0%; 1724 women). In obese women, when compared with placebo, metformin use was associated with a non‐significant reduction in risk of pre‐eclampsia (RR, 0.74; 95% CI, 0.09–6.28; I2 = 86%; 840 women). In women with gestational diabetes, metformin use was also associated with a non‐significant reduction in risk of any HDP (RR, 0.71; 95% CI, 0.41–1.25; I2 = 0%; 556 women) when compared with glyburide. When studies were combined using Bayesian random‐effects meta‐regression, with treatment type as a covariate, the posterior probabilities of metformin having a beneficial effect on the prevention of pre‐eclampsia, pregnancy‐induced hypertension and any HDP were 92.7%, 92.8% and 99.2%, respectively, when compared with any other treatment or placebo. Conclusions There is a high probability that metformin use is associated with reduced HDP incidence when compared with other treatments or placebo. The small number of studies included in the analysis, the low quality of evidence and the clinical heterogeneity preclude generalization of these results to broader populations. Given the clinical importance of this topic and the magnitude of effect observed in this meta‐analysis, further prospective trials are urgently needed.