Abstract
Background: Severe neonatal hyperbilirubinemia, with consequent encephalopathy, remains a common cause of morbidity and death in many regions of the world. Poor access to clinical laboratory resources and screening programs to measure plasma bilirubin levels is a major contributor to delayed treatment in developing countries, and the cost of existing point-of-care screening instruments precludes their dissemination. Objectives: We are evaluating the accuracy of a low-cost, minimally invasive point-of-care system (Bilistick) requiring a 25-µl blood sample that could be used in low-resource environments to evaluate patients with neonatal jaundice. Methods: We compared plasma bilirubin levels in divided blood samples by clinical laboratories and by Bilistick at two medical centers serving term and near-term newborns from ethnically different populations. Results: 118 neonates with bilirubin levels ranging from 24.8 to 501.0 µmol/l were analyzed. The mean bilirubin concentration (±SD) was 215.6 ± 85.5 µmol/l for Bilistick and 226.1 ± 86.4 µmol/l by laboratory determination. Pearson’s correlation coefficient between all paired results was 0.961, and the Bland-Altman analysis showed a mean difference of 10.3 µmol/l with a 95% interval of agreement of –38.0 to 58.7 µmol/l. Conclusion: Bilistick is a minimally invasive method for measuring total bilirubin concentration over a wide range of values and should provide an affordable and accurate system for pre-discharge and follow-up screening of jaundiced infants, particularly in low-resource environments.