This is an individual-participant meta-analysis, the strongest design available short of a single large trial. Ten double-blind placebo-controlled trials of metformin in pregnancies without established diabetes were identified; seven supplied individual data, and 2,297 pregnancies were analysed — 1,159 assigned metformin, 1,138 placebo. Models were adjusted for maternal age, body mass index, gestational age at starting, and baseline glucose.
Metformin did not prevent gestational diabetes. Using World Health Organization 1999 criteria the adjusted odds ratio was 1.00 (95% CI 0.71 to 1.41); using the 2015 National Institute for Health and Care Excellence criteria it was also 1.00 (0.71 to 1.41). Only under International Association of Diabetes and Pregnancy Study Groups thresholds did the adjusted estimate reach significance, at 0.71 (0.52 to 0.98). Fasting glucose fell by 0.06 mmol/L (-0.10 to -0.01). What did move: gestation was 0.30 weeks longer (0.06 to 0.54), preterm birth was less likely (adjusted odds ratio 0.64, 0.47 to 0.89), and neonatal head circumference was 2.43 percentiles larger (0.13 to 4.72). Gastrointestinal side effects were commoner on metformin.
This closes a question that has stayed open because individual trials were too small. Metformin given prophylactically to high-risk pregnant women — for obesity, polycystic ovary syndrome, or previous gestational diabetes — does not stop them developing gestational diabetes. The preterm birth reduction is interesting and biologically plausible, but it was not the question the trials were designed to answer.
In practice: do not start metformin in pregnancy to prevent gestational diabetes. Continue to use it to treat gestational diabetes once diagnosed, where the evidence is entirely different, and keep the focus on early screening in the high-risk women rather than pre-emptive drug treatment.
- Do not prescribe metformin in early pregnancy with the aim of preventing gestational diabetes
- Continue metformin where it is treating diagnosed gestational diabetes or pre-existing type 2 diabetes — this analysis says nothing against that
- In women with obesity or polycystic ovary syndrome, screen earlier rather than medicating pre-emptively
- Warn about gastrointestinal side effects, which were commoner on metformin and are a real reason women stop it in pregnancy
- Treat the preterm birth signal as hypothesis-generating; it should not by itself drive a prescription
The statistics, in plain English
An adjusted odds ratio of exactly 1.00 with an interval of 0.71 to 1.41 is a clean null: the data are most consistent with no effect at all, and the interval is tight enough to exclude a large benefit. The one positive result, 0.71 under International Association of Diabetes and Pregnancy Study Groups criteria, appeared only after adjustment and only under one of three diagnostic definitions tested. When one of several outcome definitions turns significant and the others do not, the safest reading is chance rather than a real, criteria-specific effect.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for diabetes & endocrinology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free