Metformin + Insulin vs. Insulin for GDM and T2DM during pregnancy: systematic review and meta-analysis.
Evaluate the effectiveness and safety of adding metformin to insulin (M+I) versus insulin alone for pregnant women with type 2 diabetes mellitus (T2DM) or gestational diabetes mellitus (GDM), focusing on stillbirth as the primary outcome.
PubMed, Embase, and Cochrane Central were searched. No date limits. Last search: January 2025.
Randomized clinical trials including women with T2DM or GDM were eligible. Trials restricted to type 1 diabetes were excluded.
Two reviewers extracted maternal and neonatal outcomes and assessed risk of bias with the Cochrane RoB 2 tool. Evidence certainty was graded using GRADE. Data were pooled with random-effects models and reported as risk ratios (RRs) or mean differences (MDs) with 95% confidence intervals.
Nine RCTs (2,420 women) were included, most with GDM and some with T2DM. Moderate-quality evidence indicated reduced stillbirth risk with M+I (6 RCTs, 2,196 participants; RR 0.36, 95% CI 0.14-0.90; NNT 111). Low-quality evidence suggested lower risks of gestational hypertension (4 RCTs; RR 0.68, 95% CI 0.48-0.97) and neonatal hypoglycemia (7 RCTs; RR 0.49, 95% CI 0.30-0.80). No significant differences were found for cesarean section, preterm delivery, or other neonatal outcomes. Heterogeneity, baseline imbalances, and small samples limited certainty.
M+I may reduce stillbirth and some adverse outcomes compared with insulin alone, but most evidence remains low certainty. Further high-quality RCTs are needed.Registered in PROSPERO: CRD42024617330.
PubMed, Embase, and Cochrane Central were searched. No date limits. Last search: January 2025.
Randomized clinical trials including women with T2DM or GDM were eligible. Trials restricted to type 1 diabetes were excluded.
Two reviewers extracted maternal and neonatal outcomes and assessed risk of bias with the Cochrane RoB 2 tool. Evidence certainty was graded using GRADE. Data were pooled with random-effects models and reported as risk ratios (RRs) or mean differences (MDs) with 95% confidence intervals.
Nine RCTs (2,420 women) were included, most with GDM and some with T2DM. Moderate-quality evidence indicated reduced stillbirth risk with M+I (6 RCTs, 2,196 participants; RR 0.36, 95% CI 0.14-0.90; NNT 111). Low-quality evidence suggested lower risks of gestational hypertension (4 RCTs; RR 0.68, 95% CI 0.48-0.97) and neonatal hypoglycemia (7 RCTs; RR 0.49, 95% CI 0.30-0.80). No significant differences were found for cesarean section, preterm delivery, or other neonatal outcomes. Heterogeneity, baseline imbalances, and small samples limited certainty.
M+I may reduce stillbirth and some adverse outcomes compared with insulin alone, but most evidence remains low certainty. Further high-quality RCTs are needed.Registered in PROSPERO: CRD42024617330.
Authors
Hallack Hallack, Nogueira Nogueira, Bomfim Bomfim, Meireles Meireles, Martins Martins
View on Pubmed