Interventions in preconception and pregnant women at risk of gestational diabetes; a systematic review and meta-analysis of randomised controlled trials

Women at risk of gestational diabetes mellitus (GDM) need preventative interventions. To evaluate targeted interventions before and during pregnancy for women identified as being at risk of developing GDM. Systematic review and meta-analysis conducted following PRISMA guidelines. MEDLINE, EMBASE and the Cochrane Library in addition to reference and citation lists were searched to identify eligible randomised controlled trials (RCTs) utilising risk stratification during the preconception period or in the first/early second trimester. Screening and data extraction were carried out by the authors independently. Quality assessment was conducted based on the Cochrane risk-of-bias tool. Random effects meta-analysis and narrative synthesis were performed. Eighty-four RCTs were included: two during preconception and 82 in pregnancy, with a pooled sample of 22,568 women. Interventions were behavioural (n = 54), dietary supplementation (n = 19) and pharmacological (n = 11). Predictive factors for risk assessment varied; only one study utilised a validated prediction model. Gestational diabetes was reduced in diet and physical activity interventions (risk difference - 0.03, 95% CI 0.06, - 0.01; I 58.69%), inositol (risk difference - 0.19, 95% CI 0.33, - 0.06; I 92.19%), and vitamin D supplements (risk difference - 0.16, 95% CI 0.25, - 0.06; I 32.27%). Subgroup analysis showed that diet and physical activity interventions were beneficial in women with ≥ 2 GDM risk factors (risk difference - 0.16, 95% CI 0.25, - 0.07; I 11.23%) while inositol supplementation was effective in women with overweight or obesity (risk difference - 0.17, 95% CI 0.22, - 0.11; I 0.01%). Effectiveness of all other interventions were not statistically significant. This review provides evidence that interventions targeted at women at risk of GDM may be an effective strategy for prevention. Further studies using validated prediction tools or multiple risk factors to target high-risk women for intervention before and during pregnancy are warranted.