Abstract
BACKGROUND: The efficacy of antenatal corticosteroids (ACS) in late-preterm birth remains uncertain, with benefits of repeated courses not well established. The aim of this study was to assess real-world effectiveness of antenatal dexamethasone in late-preterm births and identify optimal dosing, timing, and beneficiary subgroups. METHODS: This retrospective cohort study utilized data from all deliveries occurring between January 1, 2018, and June 30, 2025 at a Chinese tertiary center. 3,703 individuals with singleton or multiple gestations who delivered during the late preterm period were included. The primary outcome was the incidence of neonatal respiratory distress syndrome (RDS, defined as requiring clinical signs, radiography, ≥ 24-hour persistence, and respiratory support or oxygen). The association between antenatal dexamethasone exposure and pregnancy outcomes was evaluated using propensity score matching (PSM) to balance baseline characteristics, and multivariate logistic regression to adjust for potential confounders. RESULTS: Of 3,703 eligible late-preterm births, 1,279 (34.5%; mean [standard deviation] maternal age, 30.6 [4.0] years) were exposed to ACS. After PSM (n = 820 per group), ACS showed no significant association with neonatal RDS [OR (odds ratio) 0.97; 95% CI (confidence interval): 0.70-1.35; P = 0.867] but was associated with an increased risk of neonatal hypoglycemia (OR 1.49; 95% CI: 1.17-1.91; P = 0.011). Administration of a complete course 36 h to 7 days before delivery was associated with a reduced risk of RDS (OR 0.44; 95% CI: 0.24-0.80; P = 0.008). Repeated courses were also associated with reduced RDS risk (OR 0.40; 95% CI: 0.18-0.90; P = 0.026) but with an increased risk of hypoglycemia (OR 2.93; 95% CI: 1.56-5.49; P = 0.001). In singletons, treatment prevented 24 RDS cases per 1000 but resulted in 26 additional chorioamnionitis and 110 hypoglycemia cases. No benefit was observed in twins. The relative excess risk due to interaction test was - 1.08 (95% CI: -3.91, 0.23), suggesting a potential non-significant antagonistic interaction between ACS and twin pregnancy. CONCLUSIONS: Among late-preterm births, ACS exposure was associated with a reduction in neonatal RDS for singleton pregnancies when administered in optimally timed or repeated courses; however, it was also associated with an increased risk of neonatal hypoglycemia. In twin pregnancies delivering late-preterm, no significant respiratory benefit was observed. These findings highlight that both treatment timing and pregnancy type are critical considerations when administering ACS.