# Falling Third Trimester Insulin Requirements and Adverse Pregnancy Outcomes in Individuals with Pre-Existing Diabetes: A Retrospective Cohort Study

## TL;DR

This study found no link between a drop in insulin needs during the third trimester and adverse pregnancy outcomes in people with pre-existing diabetes.

## Contribution

The study is the first to investigate third-trimester insulin drops as a potential predictor of adverse pregnancy outcomes in pre-existing diabetes.

## Key findings

- A 15.4% of participants experienced a 15% or more drop in third-trimester insulin requirements.
- No significant association was found between the insulin drop and adverse pregnancy outcomes.
- The study highlights the need for larger, prospective research to confirm these findings.

## Abstract

Objective: To determine whether a third-trimester drop in insulin requirements in pregnant people with pre-existing diabetes is associated with a subsequent occurrence of adverse pregnancy outcomes. Research Design and Methods: We conducted a retrospective cohort study of patients with type 1 and 2 diabetes who were followed at a tertiary referral center in Toronto, Canada. We collected data on insulin dosing in the third trimester (after 28 weeks of pregnancy) and compared outcomes in those with and without a third-trimester drop of 15% or more in their total insulin requirements. Our primary outcome was a composite of stillbirth, spontaneous preterm birth or preterm premature rupture of membranes, and iatrogenic preterm birth or cesarean birth for fetal wellbeing concerns, occurring following the drop in insulin requirements. We conducted regression analyses controlling for early pregnancy glycosylated hemoglobin, body mass index, and diabetes-related microvascular disease, and presented results as odds ratios (OR) with 95% confidence intervals (95%CI). Results: We included 350 pregnant people—146 with type 1 and 204 with type 2 diabetes. Of these, 54 (15.4%) had a third-trimester drop of 15% or more in their total insulin requirements. There was no difference in the primary outcome between groups (OR 0.97; 95% CI 0.41–2.10). Conclusions: Based on this single-center study, limited by sample size and analytic constraints, in people with pre-existing diabetes, a third-trimester drop of ≥15% in total insulin requirements was not associated with subsequent occurrence of adverse pregnancy outcomes. Larger prospective studies looking at associations between a drop in insulin requirements and subsequent occurrence of adverse pregnancy outcomes are necessary to inform meta-analyses and clinical decision making.