LAUSR

OBJECTIVE  Severe neonatal hyperbilirubinemia can cause neurological disability or mortality if not effectively managed. Exchange transfusion (ET) is an efficient treatment to prevent bilirubin neurotoxicity. The purpose of this study was to evaluate outcomes in severe neonatal hyperbilirubinemia with ET and to identify the potential risk factors for poor outcomes. METHODS  Newborns of ≥28 weeks of gestational age with severe hyperbilirubinemia who underwent ET from January 2015 to August 2019 were included. Demographic data were recorded and analyzed according to follow-up outcomes at 12 months of corrected age. Poor outcomes were defined as death due to bilirubin encephalopathy or survival with at least one of the following complications: cerebral palsy, psychomotor retardation (psychomotor developmental index < 70), mental retardation (mental developmental index < 70), or hearing impairment. RESULTS  A total of 524 infants were eligible for recruitment to the study, and 62 infants were lost to follow-up. The outcome data from 462 infants were used for grouping analysis, of which 398 cases (86.1%) had normal outcomes and 64 cases (13.9%) suffered poor outcomes. Bivariate logistic regression analysis showed that peak total serum bilirubin (TSB) (odds ratio [OR] = 1.011, 95% confidence interval [CI] = 1.008-1.015, p = 0.000) and sepsis (OR = 4.352, 95% CI = 2.013-9.409, p < 0.001) were associated with poor outcomes of hyperbilirubinemia. Receiver operator characteristic curve analysis showed that peak TSB ≥452.9 µmol/L could predict poor outcomes of severe hyperbilirubinemia. CONCLUSION  Peak TSB and sepsis were associated with poor outcomes in infants with severe hyperbilirubinemia, and peak TSB ≥452.9 µmol/L could predict poor outcomes.