Should ob/gyns use metformin to treat pregnant women with PCOS?

©Sherry Young - stock.adobe.com

A new study in Lancet Diabetes & Endocrinology that leveraged nearly two decades of research has tested the hypothesis that metformin prevents late miscarriage and preterm birth (PTB) in women with polycystic ovary syndrome (PCOS). 

From 2000 to 2017, the research group oversaw three controlled studies comparing the impact in pregnant women with PCOS of metformin versus placebo. PregMet2, the largest of the studies, took place between October 19, 2012 and September 1, 2017. The authors also performed a post-hoc analysis of pregnancy outcomes in which they pooled data from PregMet2 with individual participant data from two previous trials, which had shown a significant reduction in late miscarriage and PTB in women with PCOS taking metformin. 

Administered at 14 hospitals in Norway, Sweden, and Iceland, PregMet2 was a randomized, double-blind trial. Participants were women aged 18 to 45 with PCOS who were pregnant with singletons. They were randomly assigned (1:1) to receive oral metformin 500 mg twice a day or placebo during the first week of the study and metformin was increased to 1000 mg twice a day from week 2 until delivery. 

The composite primary outcome of the study-late miscarriage and PTB-occurred in 12 (5%) of 238 women in the metformin group and 23 (10%) of 240 women in the placebo group (P= 0.08). No significant differences were found for secondary endpoints, including gestational diabetes (60 [25%] women in the metformin group and 57 [24%] women in the placebo group were diagnosed). No significant between-group differences in serious adverse events were observed in either mothers or offspring. 

In the post-hoc pooled analysis, 18 (5%) of 397 women in the metformin group had late miscarriage or PTB compared with 40 (10%) of 399 in the placebo group (P= 0.004). 

Based on their findings, the authors believe that metformin treatment reduces risk of late miscarriage and PTB in pregnant women with PCOS. However, they were surprised that metformin did not prevent gestational diabetes in this population, given that metformin is an insulin sensitizer and is predominantly used in treatment of type 2 diabetes.