The Impact of Telemedicine Interventions on Maternal and Fetal Outcomes in Women With Gestational Diabetes: Systematic Review and Meta-Analysis of Randomized Controlled Trials.
Hongyue He, Juan Deng, Jingxian Liu et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis pooled 23 randomized controlled trials involving 3055 women with gestational diabetes to assess telemedicine and maternal and fetal outcomes. Telemedicine was associated with lower fasting glucose (MD -0.26, 95% CI -0.42 to -0.09) and 2-hour postprandial glucose (MD -0.50, 95% CI -0.87 to -0.14), as well as lower gestational weight gain. It was also associated with lower risks of several cesarean and neonatal outcomes. However, heterogeneity was very high for key continuous outcomes, prediction intervals crossed the null, and evidence certainty was very low to low.
Key findings
- Across 23 RCTs and 3055 women, fasting glucose was lower with telemedicine (MD -0.26, 95% CI -0.42 to -0.09).
- Telemedicine was associated with lower 2-hour postprandial glucose (MD -0.50, 95% CI -0.87 to -0.14) and gestational weight gain (MD -0.78, 95% CI -1.40 to -0.17).
- Lower risks were reported for cesarean section, emergency cesarean section, macrosomia, neonatal hypoglycemia, and neonatal respiratory distress.
- No significant differences were found for hemoglobin A1c, preterm birth, or neonatal intensive care unit admission.
Why it matters
Remote care may help address access barriers to gestational diabetes follow-up and could support glycemic management and selected perinatal outcomes. Yet the pooled improvements in glucose and weight gain were highly heterogeneous, with prediction intervals including no benefit, and certainty was very low to low. These findings therefore suggest potential rather than established benefit and underscore the need to interpret pooled averages cautiously.
What to keep in mind
Evidence certainty was very low to low across outcomes. Fasting glucose, postprandial glucose, and gestational weight gain had very high unexplained heterogeneity, and their prediction intervals crossed the null.