Detection and treatment of early gestational diabetes mellitus: a systematic review and meta-analysis of randomized controlled trials.

Abstract

To estimate the impact of detection and treatment of early gestational diabetes mellitus on short-term maternal, fetal, and neonatal outcomes. We defined 2 maternal (gestational diabetes prevalence and cesarean section) and 2 neonatal (preterm birth and macrosomia) primary outcomes. We also defined 5 maternal and 12 fetal-neonatal secondary outcomes. Ovid Medline, Cochrane CENTRAL, and Embase since inception. The search was updated in November 2024. Inclusion criteria: randomized controlled trials addressing detection and treatment of early gestational diabetes (diagnosed before 20 completed weeks). pregestational diabetes or overt diabetes in pregnancy. The Cochrane Handbook was used to guide data extraction and interpretation including risk of bias assessment (Risk of Bias 2 tool). Aggregation and comparison of results were performed with Revman 5.4.1. Pooled relative risk and mean differences were calculated with 95% confidence intervals using random-effects models. The quality of the evidence for primary outcomes was summarized using Grading of Recommendations Assessment, Development and Evaluation criteria. We identified 1221 unique references. Seven articles addressing early gestational diabetes met the eligibility criteria with a total of 30,791 participants. These studies used 2 strategies: (1) treatment vs usual care of women with a diagnosis of early gestational diabetes and (2) population-based approaches, either performing screening (vs not) or using different cutoffs for diagnosis. In studies comparing treatment vs usual care, differences were observed only in secondary outcomes: more drug treatment, less maternal weight gain, lower birthweight, and less respiratory distress. In studies comparing different population-based strategies, primary outcomes differed for a higher rate of early and overall gestational diabetes (relative risk, 5.50; 95% confidence interval, 3.56-8.48 and 1.83; 95% confidence interval, 1.41-2.38, respectively) and a lower rate of primary cesarean section (relative risk, 0.88; 95% confidence interval, 0.84-0.93); as to secondary outcomes, differences were observed in terms of higher total pregnancy-induced hypertension and preeclampsia. The quality of evidence for most outcomes was low/very low. Detection and treatment of early gestational diabetes mellitus do not offer indisputable benefits either in treated women or at the population level. More studies are required to elucidate this issue.