Antibiotic Use in Early Pregnancy and Risk of Gestational Diabetes Mellitus: A Nationwide Cohort Study of 3 Million Pregnancies

Background: Antibiotics can alter the gut microbiome and metabolic health of the host. Accumulating studies have suggested an increased risk of metabolic disorders, including obesity and type 2 diabetes, associated with antibiotic use; however, it is largely unknown in the pregnant population.

Objectives: To examine the association between antibiotic use in early pregnancy and the risk of gestational diabetes mellitus (GDM).

Methods: We conducted a retrospective cohort study using nationwide Korea's healthcare database from 2008 to 2019. We included all pregnancies aged 20-44 years who gave live birth during 2010-2019. We excluded women with multifetal pregnancies and diagnosed with pre-existing DM and obesity. Exposure was defined as having ≥1 systemic antibiotic (ATC code J01) prescription in early pregnancy (first 19 weeks of gestation); the reference group was defined as having no antibiotic prescription for one month before and during early pregnancy. We further classified the exposure by 1) antibiotic classes (restricting women receiving each class alone), 2) cumulative duration of antibiotic use (1-6, 7-14, and 15-29, 30+ days), and 3) the number of antibiotic classes prescribed (1, 2, 3, 4, and 5+ classes) in early pregnancy. Study outcome was GDM, ascertained from week 20 of gestation until the delivery date. We estimated relative risks (RR) with 95% confidence intervals (CIs) using a weighted generalized linear model with propensity score (PS) stratification to adjust for potential confounders.

Results: Of the 3,045,589 pregnancies, 737,970 (24.2%) were exposed to antibiotics in early pregnancy. The PS-adjusted RR for GDM was 1.04 (95% CI 1.03-1.04) in antibiotic users vs. non-users (incidence: 10.9% vs. 10.0%). The RRs were significantly increased in fosfomycin (1.16, 1.06-1.26), tetracyclines (1.15, 1.09-1.21), macrolides (1.14, 1.12-1.17), and cephalosporins (1.04, 1.03-1.05). However, there was no substantial relationship between the GDM risk and cumulative duration of antibiotic use (1-6 days: 1.04 [1.03-1.05], 7-14 days: 1.04 [1.03-1.06], 15-29 days: 1.06 [1.01-1.10], and 30+ days: 0.97 [0.82-1.14]); there was also no relationship to GDM risk of the number of antibiotic classes.

Conclusions: In this large nationwide cohort study, antibiotic use in early pregnancy was associated with a small increased risk of the GDM. However, the magnitude of association was modest, and there was no quantity-response relationship. Taken together, antibiotic use in early pregnancy in real-world practice was not largely responsible for the development of GDM.