ABSTRACT
The objective of this paper is to determine the importance of perinatal factors predicting occurrence of severe intracranial hemorrhage (ICH) in premature newborns. A post-hoc analysis of a previously published randomized, double-blinded, prospective trial was performed. Logistic regression analysis was used to assess the importance of obstetrical variables and umbilical cord blood coagulation studies and acid-base status in predicting severe ICH (grades 3 and 4). The trial lasted 42 months; 401 pregnant women were invited to participate and 48 declined. The most important predictors of severe ICH in order of decreasing significance were: gestational age at delivery (p=0.0001), duration of painful labor (p=0.0077), duration of antenatal antibiotic therapy (p=0.0203), maternal age (p=0.0247), and factor X activity in umbilical cord blood (p=0.0302). Mode of delivery, cord blood acid-base, and coagulation status did not correlate with severe ICH. The majority of severe ICHs were late in onset; only two of 31 were known to have occurred in the first 24 hr of life. Gestational age, duration of painful labor, duration of antibiotic therapy, and maternal age were the most important predictors of severe ICH.