Falling Third-Trimester Insulin Requirements in Diabetic Pregnancies and Adverse Pregnancy Outcomes: A Systematic...

Background: There is conflicting evidence on whether falling insulin requirements (FIRs) in the third trimester are associated with adverse pregnancy outcomes. We synthesized published evidence to address this knowledge gap. Methods: We conducted a systematic review and meta-analysis, wherein we searched four bibliographic databases until September 04 2025 for articles describing third-trimester FIR and pregnancy outcomes. We assessed the risk of bias using the Quality In Prognosis Studies (QUIPS) tool, performed meta-analysis with pooled odds ratios (ORs) and 95% confidence intervals (95% CIs) for maternal and perinatal outcomes and assessed certainty of evidence (CoE) using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach. Results: We identified 2044 articles, of which nine fulfilled the eligibility criteria. Third-trimester FIR has an important association with preeclampsia [OR 3.0 (95% CI 1.41-6.38, absolute event rate (AER) 15.9%, high CoE], probably has an important association with neonatal respiratory distress [OR 2.03 (95% CI 1.27-3.26), AER 15.8%, moderate CoE]; may have an important association with a composite of outcomes reflecting placental dysfunction [OR 2.32 (95% CI 1.07-5.03), AER 13.5%, low CoE] and preterm birth [OR 2.0 (95% CI 0.51-7.85), AER 9.3%, very low CoE]; and may not have an important association with stillbirth [OR 1.5 (95% CI 0.27-8.40), AER 0.05%, low CoE], small-for-gestational-age [OR 1.29 (95% CI 0.77-2.15), AER 1.8%, low CoE], or low Apgar score at 5 minutes [OR 1.68 (95% CI 0.68-4.14), AER 2.3%, low CoE]. Conclusions: The CoE regarding associations between third-trimester FIR and adverse pregnancy outcomes varies considerably, and it remains uncertain whether these associations reflect cause or effect. Therefore, a solitary finding of third-trimester FIR does not warrant early delivery and maternal-foetal surveillance should be based on the primary clinical diagnosis.