(815) Treatment effect heterogeneity of antenatal steroids among late preterm births

Objective: To determine if there is treatment effect heterogeneity for late preterm steroids among factors previously reported to be associated with a neonate’s risk of respiratory morbidity.

Study Design: This is an exploratory secondary analysis of the Antenatal Late Preterm Steroid (ALPS) Trial. The primary outcome was the same composite of respiratory morbidity used in the trial. The primary exposure was antenatal steroid exposure vs. placebo. Treatment heterogeneity of antenatal steroids was assessed across 3 factors: gestational age at trial enrollment, fetal sex, planned cesarean delivery. A logistic regression model with interaction terms between the exposure and each factor was fit. Predicted probabilities of the primary outcome were compared across all potential possible combinations, and the absolute risk difference (ARD) between ALPS vs. placebo was plotted across days of gestation, stratified by sex and mode of delivery.

Results: There were 2,825 participants included. The distribution of gestational age, fetal sex, and planned mode of delivery was not different between the groups. The Figure shows the ARD for respiratory morbidity over the late preterm period by fetal sex and planned mode of delivery. The absolute risk of the outcome in the untreated group is shown below the x-axis for each week of gestation. Patients with a planned cesarean delivery and presenting at 34+0 weeks had the highest baseline risk of respiratory morbidity (49.6% (95% CI 39.1, 59.9%)) and demonstrated the greatest ARD with ALPS. In comparison, the absolute risk for patients presenting at 36+5 weeks and having planned vaginal delivery was 5.9% (95% CI 4.1, 7.6%) with no observed ARD with steroid exposure, though the analysis was limited by small sample sizes.

Conclusion: These findings suggest there is treatment heterogeneity for ALPS depending on fetal sex, gestational age, and mode of delivery, with the largest benefit for patients with presenting at 34 weeks having a planned CD. Studies with larger sample sizes are needed to better inform shared decision-making around the practice of late preterm steroids.