Incidence of neonatal hyperbilirubinemia: a population-based prospective study in Pakistan

Abstract

To estimate the incidence of neonatal jaundice and hyperbilirubinemia in a poor urban community in Karachi, where 70% of births occur at home. Home-based pregnancy and newborn surveillance were conducted from September 2004 to July 2006 in a multi-ethnic population by trained community health workers. Newborns were visited several times at scheduled intervals until 59 days of life; any baby with jaundice was referred to the local clinic. Clinical assessments of jaundice were assigned by a physician and recorded using an adapted Kramer scale. Blood for plasma bilirubin was obtained if parents consented. Of a birth cohort of 1690 young infants during the study period, 466 infants (27.6%) were referred to our centre with jaundice. Of these, 64% were 0-6 days old. Bilirubin was measured in 125 of 466 (27%) jaundiced newborns. Overall detected rate of hyperbilirubinemia (bilirubin >5 mg/dl) among 1690 newborns was 39.7/1000 live births (95% CI 29.3-47.6). Rate of plasma bilirubin levels in the range of 15-20 mg/dl was 13/1000 live births (95% CI 7.6-18.4); levels >20 mg/dl were observed in 3.5/1000 live births (95% CI 0.4-5.5). The proportion of newborns with bilirubin > or =15 mg/dl was significantly higher among those assigned a Kramer score of 4-5 compared to those receiving a score of 1-3 (P-value 0.00004). A significant burden of untreated severe neonatal jaundice, causing potential neurological sequelae, exists in developing countries such as Pakistan. WHO guidelines are needed for screening and appropriate management of neonatal jaundice in developing countries.