Postnatal Steroid Exposure in Extremely Low Gestational Age Newborns and Kidney Function at 24 Months Corrected Age

Key Points Overall, steroid exposure was not associated with GFR at follow-up. Hydrocortisone duration of 1–7 days had 2.8 times increased odds of reduced GFR at 24-month corrected age. Dexamethasone given at ≤25 weeks postmenstrual age had 9.4 times increased odds of reduced GFR compared with those given dexamethasone at ≥29 weeks. Background Preterm birth is associated with low nephron endowment, with an increased risk of CKD later in life. Almost all pregnant women at risk for preterm delivery are given antenatal corticosteroids to accelerate lung maturity in preterm neonates. Similar to antenatal steroids, postnatal corticosteroids are also given to improve lung function, but the effect on kidney function is unknown. The objective of this study was to determine if duration and timing of postnatal corticosteroids in preterm infants influences kidney function at 24-month corrected age. Methods A secondary analysis of the Preterm Erythropoietin Neuroprotection Trial (neonates <28 weeks' gestation) was performed and included surviving participants with serum creatinine measured at 22–26 months corrected gestational age. Exposure included the presence, type, start date, and duration of postnatal steroids (duration and start date based on postmenstrual age). The primary outcome was reduced eGFR (GFR<90 ml/min per 1.73 m 2 at the 24-month corrected gestational age time point). Outcomes were adjusted for perinatal/neonatal exposures, and neonatal outcomes were compared. Results Of 838 surviving infants, 397 (47%) were exposed to any postnatal steroid. Dexamethasone was the most common exposure ( n =238, median start date at day of life 27, median duration of 8 days), followed by hydrocortisone ( n =232, median start day of life 13, median duration 17 days). In total, 348 infants were evaluated at the 22–26-month follow-up, 61 of whom had reduced GFR. Hydrocortisone duration of 1–7 days had 2.8 (95% confidence interval, 1.06 to 7.62) times increased odds of reduced GFR at 24-month corrected age. Although overall steroid exposure was not associated with GFR at follow-up, initiation of dexamethasone at ≤25 weeks postmenstrual age was associated with 9.39 (95% confidence interval, 1.61 to 54.71) increased odds of reduced GFR compared with those given dexamethasone at ≥29 weeks. Conclusions Although observational, this study supports an association between postnatal steroid timing, duration, and reduced GFR. Further studies are warranted to better understand this association to protect long-term kidney health.