Does time interval between antenatal steroid administration and delivery affect neonatal outcomes in very low birth weight neonates—A cohort study

Abstract

DELIVERY AFFECT NEONATAL OUTCOMES IN VERY LOW BIRTH WEIGHT NEONATESDA COHORT STUDY ALLISON GOEDEN, JEFFERY GARLAND, GAIL PECKMAN, MARGARET CARR, RICHARD PIRCON, Medical College of Wisconsin Affiliated Hospitals, OB/GYN, Wauwatosa, Wisconsin, St. Joseph Regional Medical Center, Pediatrics, Milwaukee, Wisconsin, St. Joseph Regional Medical Center, Center for Obstetrical Research and Education, Milwaukee, Wisconsin, Perinatal Medicine, Ltd., Milwaukee, Wisconsin, St. Joseph Regional Medical Center, Director-Center for Obstetrical Research and Education, Milwaukee, Wisconsin OBJECTIVE: To determine whether the neonatal benefit of a single course of antenatal steroids diminishes beyond fourteen days from administration. STUDY DESIGN: This retrospective cohort study included all women who received a complete course of antenatal steroids and delivered a living singleton between 26 and 34 weeks’ gestation between January 1999 and December 2004. Maternal charts were reviewed to obtain histories and confirm treatment. Neonatal demographic data and outcomes were collected prospectively and retrospectively reviewed. Individuals reviewing maternal data were blinded to neonatal outcomes. Comparisons were made between those delivering within seven days of receiving antenatal steroids and those beyond fourteen days using Student’s t-test, chi-square analysis, Fisher’s exact test and regression analysis. RESULTS: Two hundred and fifty-one patients met inclusion criteria. Patients were grouped based upon time from administration of steroids to delivery. In univariate analysis, neonates delivering greater than 14 days from antenatal steroids had higher Scores for Acute Neonatal Physiology (SNAP) (measures the severity of illness at 24 hours of life) (9.7 G 5.6 vs 7.3 G 4.4, p = 0.001). This difference remained in regression models that controlled potential confounders [RR = 1.76 (95 percent CI = 1.2, 2.5), p=0.006]. Risk of death or chronic lung disease was statistically significant in univariate analysis (17% vs 10%, p = 0.02). When corrected for potential confounders in regression analysis, statistical significance was not achieved [RR = 2.1 (95 percent CI = 0.98, 6.2), p=0.054]. There was no difference in IVH risk in either univariate (11% vs 12%, p = 0.8) or regression analysis [RR = 0.88 (95 percent CI = 0.38, 2.0)]. S118 SMFM Abstracts