1C-03 - Oral Corticosteroids During Pregnancy and Gestational Diabetes Mellitus: A Sequential Landmark Analysis

Background: The utilization of oral corticosteroids (OCS) during pregnancy has increased as they are used to manage a wide range of conditions for their immunosuppressive and anti-inflammatory effects. However, OCS use could lead to insulin resistance that might contribute to the onset of gestational diabetes mellitus (GDM). To date, only a few studies have investigated the association between OCS and the risk of GDM.

Objectives: To examine the risk of GDM associated with OCS usage in pregnancy.

Methods: This population-based cohort study included pregnant women with live births (2010–2021), using Korea’s nationwide healthcare database. OCS exposure was defined as at least one prescription during early pregnancy (LMP to 19 weeks) or late pregnancy (20–27 weeks). The outcome of interest was GDM, assessed from LMP+140 days to delivery. Primary analyses compared OCS-exposed pregnancies to two reference groups: (1) unexposed and (2) past-exposed. Secondly, to assess OCS effects beyond early pregnancy, we performed a sequential landmark analysis at 20–27 weeks. For each week, we included women who had not previously initiated OCS nor diagnosed with GDM up to that week to mitigate reverse causation and immortal time bias, and the results were pooled, with further adjustment for the specific week of inclusion. Propensity score-based fine stratification was applied for controlling for confounders and generalized linear models estimated relative risk (RR) with 95% CIs for GDM. Furthermore, additional analyses were conducted according to OCS categories (short, intermediate, and long-acting) and main indications of OCS.

Results: In the cohort of 3,848,270 pregnancies, 207,974 (5.4%) were exposed to OCS during early pregnancy. The absolute risk of GDM was 101.3 and 92.0 per 1,000 pregnancies in the exposed and unexposed group. The PS-adjusted RR of OCS exposed group was 1.02 (95% CI 1.00–1.03) for GDM, compared with unexposed, and 1.03 (1.01–1.05), compared with past-exposed. Further, exposure during late pregnancy, weeks 20 to 27, resulted in a null association, with a pooled aRR of 1.00 (0.97–1.04). Short- and intermediate-acting OCS showed no association with GDM, while long-acting OCS had a marginally increased risk (1.08, 1.01–1.15), also no association was observed among pregnancies with asthma or autoimmune disease.

Conclusions: The findings of this cohort study indicate that there is no association between OCS exposure during pregnancy and GDM. Given the benefits of treating inflammatory symptoms during pregnancy, the current findings suggest that OCS use may be considered without concern for GDM risk, while further investigation into other potential risks associated with prenatal OCS use is warranted.