Abstract 1481 Poster Session III, Monday, 5/3 (poster 6)

Background: In VLBW infants, maternal treatment with betamethasone is associated with lower mortality and a lower risk of intraventricular hemorrhage. However, postnatal treatment of VLBW infants with another corticosteroid - dexamethasone - has been associated with an increased risk of neurodevelopmental impairment (Yeh et al. Pediatrics 1998). Objective: The objective of this case-control study was to analyze the association between treatment of mothers with betamethasone and the risk, in their infants, of cerebral palsy (CP) at one year adjusted age. Methods: Cases were 81 consecutive infants with the following attributes: 1) born 1/1/86 to 12/31/96, 2) birth weight 500 to 1500 grams, 3) no major congenital anomaly, 4) born to a mother residing in a 17-county perinatal region, 5) born at our regional obstetric referral center, 6) diagnosed with CP at one year adjusted age. Controls (n=81) had these same attributes, except that they were free from signs of CP at one year, and were the VLBW infants born closest in time after the birth of the respective case. A research nurse, who was not aware of which infants had CP, obtained maternal and neonatal data from hospital records, including data about pregnancy complications, obstetric interventions, estimated gestational age, Apgar scores, first arterial pH, lowest systolic blood pressure in the first 12 hours, receipt of a fluid bolus in the first 12 hours, and the diagnosis of respiratory distress syndrome, pneumothorax, pulmonary interstitial emphysema (PIE), or cranial ultrasound (CUS) abnormalities. Associations with CP were expressed as the odds ratio (OR); the precision of these estimates, as 95% confidence limits (CL). Multivariate analyses were performed with logistic regression. Results: 19% of cases (15/81) and 42% of controls (34/81) were born to mothers treated with betamethasone [OR: 0.31 (0.15-0.64)]. A complete course of betamethasone (2 doses given at least 48 hours and no more than 7 days before delivery) was given to 7% of mothers of cases, as compared to 17% of mothers of controls. The association between betamethasone and a lower risk of CP persisted after adjusting for the two other prenatal factors associated with CP (gestational age and chorioamnionitis) [multivariate OR for betamethasone: 0.26 (0.12-0.58)]. When adjusted for these factors, as well as postnatal events that were associated with an increased risk (systolic blood pressure < 30 mm Hg in the first 12 hours, PIE, and CUS abnormalities), the association between betamethasone and a lower risk of CP remained [multivariate OR: 0.27 (0.10-0.70)]. Among 44 cases and 80 controls who did not have a major CUS abnormality (post-hemorrhagic hydrocephalus, persistent ventricular dilatation, or periventricular echodensity or echolucency) the multivariate OR was 0.20 (0.07-0.59). Conclusions: Among mothers delivering a VLBW infant, antenatal treatment with betamethasone is associated with a reduced risk of CP.