(A-283) Second-Line Antidiabetic Medications in Early Pregnancy: Do They Increase Miscarriage Risk?

Background: For pregnant or pregnancy-planning patients with type 2 diabetes (T2D), insulin has traditionally been the guideline-recommended treatment due to limited safety data on non-insulin antidiabetic medications (ADMs) for the fetus. However, unintended exposure to second-line ADMs during early pregnancy has increased over time. While previous studies have explored associations with congenital malformations, evidence on miscarriage remains limited.

Objectives: To assess the periconceptional exposure to second-line ADMs and the risk of miscarriages

Methods: We conducted a nested case-control study using nationwide claims data of Korea (2009 to 2022). The study included patients diagnosed with T2D and prescribed at least one ADM from six months before the pregnancy until the pregnancy end. Pregnancy episodes resulting in miscarriage (cases) were matched to those resulting in live birth (controls) using risk-set sampling, up to 1:4, on age [±1 year], last menstrual period (LMP) [±30 days], and follow-up duration [±7 days]. The LMP was identified using a internally validated algorithm with high accuracy, and the index date was defined as the date of miscarriage for cases and the corresponding date for controls. The exposure of interest was second-line non-insulin ADMs, including SGLT2 inhibitors (SGLT2i), DPP4 inhibitors (DPP4i), GLP-1 receptor agonists (GLP-1RA), and sulfonylureas (SU), assessed during the exposure window (LMP-30 days to the index date). We estimated adjusted odds ratios (ORs) with 95% confidence intervals (CIs) using multivariable conditional logistic regression, adjusting for potential confounders.

Results: From 18,024 pregnancy episodes with T2D and prescribed with ADMs, 3,207 cases (mean [SD] age, 35.1 [5.2]) were matched with 12,612 controls, where the balance of most baseline characteristics was achieved. From the LMP to the index date, 367 cases (11.4%) and 1,256 controls (10.0%) were exposed to SGLT2i, 864 cases (26.9%) and 3,038 controls (24.1%) to DPP4i, 60 cases (1.9%) and 175 controls (1.4%) to GLP-1RA, and 654 cases (20.4%) and 2,279 controls (18.1%) to SU, respectively. Overall, there were no increased risks observed for SGLT-2i (adjusted OR 1.07, 95% CI 0.93-1.24), DPP4i (1.10, 0.99-1.22), GLP-1 RA (1.05, 0.75-1.48), SU (1.07, 0.96-1.20), when we adjusted for demographic, clinical, general and T2D-specific healthcare utilization.

Conclusions: In this nationwide study, we found no evidence that second-line non-insulin ADMs in patients with T2D increase the risk of miscarriage. Our findings offer reassurance to patients with unintended exposure of second-line ADM use during the periconceptional period. Of note, further research is warranted to include treatment options beyond insulin and metformin during pregnancy.