A follow-up study of infants with intracranial hemorrhage at full-term.

OBJECTIVE To determine physical and cognitive outcomes of full-term infants who suffered intracranial hemorrhage (ICH) at birth. METHODS A retrospective hospital-based, follow-up study of infants treated in London, Ontario between 1985 and 1996. Follow-up was conducted by telephone interviews and clinic visits. Outcome was measured according to physical and cognitive scales. Perinatal risk factors and hemorrhage characteristics were correlated with final outcome. RESULTS For this study 66 infants with ICH were identified, of which seven died during the first week of life. We obtained follow-up in all but ten cases (median = 3-years; range 1.0 to 10.9 years). Overall, 57% of infants had no physical or cognitive deficits at follow-up. Death occurred most frequently among those with primarily subarachnoid hemorrhage (19%) and the most favorable outcomes occurred among those with subdural hemorrhage (80% had no disability). In univariate models, thrombocytopenia (platelet count < or = 70 x 10(9)/L), increasing overall hemorrhage severity, frontal location and spontaneous vaginal delivery as opposed to forceps-assisted delivery increased risk for poor outcome. In multivariate models, all these factors tended towards increased risk, but only thrombocytopenia remained significant for physical disability (OR = 7.6; 95% CI = 1.02 - 56.6); thrombocytopenia was borderline significant in similar models for cognitive disability (OR = 4.6; 95% CI = 0.9 - 23.9). CONCLUSION Although forceps-assisted delivery may contribute to ICH occurrence, our study found better outcomes among these infants than those who had ICH following a spontaneous vaginal delivery. Hemorrhage in the frontal lobe was the most disabling hemorrhage location and if multiple compartments were involved, disability was also more likely to occur. However, in this report we found that the factor that was most likely to contribute to poor outcome was thrombocytopenia and this remained important in multivariate analysis.