Optimal timing of delivery in pregnant individuals with pregestational diabetes mellitus.

Introduction: SMFM and ACOG recommend delivery of gravidae with pregestational diabetes at 36-39 6/7 weeks based on glycemic control and vascular complications. The optimal gestational age within this wide range is unknown. Our objective was to evaluate the risk of adverse outcomes with delivery versus expectant management at increasing gestational ages.

Methods: Retrospective cohort study of gravidae with pregestational diabetes who delivered a non-anomalous singleton at 36 weeks (2012-2022). The primary outcome was composite neonatal morbidity: hypoglycemia, hyperbilirubinemia, shoulder dystocia, and perinatal death. Secondary outcomes included composite components, composite severe neonatal morbidity, LGA, SGA, NICU admission and cesarean. Poisson regression with robust error variance estimated the association between delivery at 36, 37, and 38 weeks and outcomes, compared with expectant management.

Results: 843 gravidae met inclusion criteria: 235 (28%) type 1 diabetes and 602 (71%) type 2 diabetes. Overall, 146 (17%) delivered at 36 weeks, 283 (34%) at 37 weeks, 217 (26%) at 38 weeks, and 197 (23%) at 39 weeks. Compared with expectant management, delivery at 36 weeks was associated with higher odds of composite morbidity (aRR 1.31, 95% CI 1.11-1.55) as well as hypoglycemia, hyperbilirubinemia, SGA and NICU admission. At 37 and 38 weeks, there was no significant difference in composite morbidity among those delivered versus expectantly managed. However, delivery at 37 weeks was associated with higher odds of hyperbilirubinemia, compared with expectant management. No other outcomes differed between with delivery versus expectant management at 37 or 38 weeks. Few associations differed by diabetes type.

Conclusion: Based on these results and supporting literature, elective delivery at 36 weeks should be avoided unless necessary. Although the data are inconclusive regarding delivery at 37 weeks, delivery at 38 weeks should be evaluated further for gravidae with pregestational diabetes. Confirmation in a large, contemporary cohort or a randomized trial are needed.