Diagnostic Performance Analysis of the Point-of-Care Bilistick System in Identifying Severe Neonatal Hyperbilirubinemia by a Multi-Country Approach

Abstract

Importance The real prevalence and clinical burden of severe neonatal jaundice are undefined due to difficulties in measuring total serum bilirubin (TSB) outside secondary and tertiary clinical centers. Objective To assess the diagnostic performance of the point-of care Bilistick System (BS) in identifying neonatal jaundice patients requiring treatment. Design Between April 2015 and November 2016, 1911 neonates, were recruited to participate in the study. Blood samples were simultaneously collected for the TSB determination by BS and by hospital laboratory (Lab). Data were collected and sent to the Bilimetrix headquarter in Trieste where statistical analysis was performed. Newborns with neonatal jaundice were treated with phototherapy according to each center's guidelines. Setting 17 hospitals from Nigeria, Egypt, Indonesia, and Viet Nam. Participants 1911 newborns were included, of which 1458 (76·3%) fulfilled the inclusion criteria. Results TSB level measured by BS agreed (p < .0001) with the lab result in all four countries. The diagnostic performance of BS showed a positive predictive value (PPV) of 92·5% and a negative predictive value (NPV) of 92·8%. Conclusions and Relevance BS is a reliable system to detect neonatal jaundice over a wide range of bilirubin levels. Since Bilistick is a point-of-care test, its use may provide appropriate and timely identification of jaundiced newborns requiring treatment.

Laboratory validation and field usability assessment of a point-of-care test for serum bilirubin levels in neonates in a tropical setting

“…This population contrasts with that of the three previous studies that tested the BS under ideal conditions: in the first study, neonates had a lower mean HCT value of 41.5% 19 , those enrolled in the second study were of an older median age of 6 days with venous blood 20 and in the most recent multi-centre study, the median HCT was 42.7% IQR (38.0-48.0) at a median age of 4 days 21 .…”

“…The study concluded that BS had comparable accuracy to the TcB JM-103. A recent multi-centre study reported the BS to slightly underestimate SBR value with maximum limits of agreements of −128.3 to 102.6 µmol/l 21 .…”

“…BS in its current form does not appear to provide any benefit over the non-invasive TCB devices as its reliability diminishes with increasing haematocrit level above 55%. Perhaps, more crucially, the study has highlighted a major drawback and context for evaluating a recent multi-centre study 1 that seems to have exaggerated the utility of BS in developing countries. Whereas Greco et al reported technical problems in only 6.8% of the tests, the current study showed a significantly higher percentage (48.6%) of error messages.…”

Laboratory validation and field usability assessment of a point-of-care test for serum bilirubin levels in neonates in a tropical setting

“…This population contrasts with that of the three previous studies that tested the BS under ideal conditions: in the first study, neonates had a lower mean HCT value of 41.5% 22 , those enrolled in the second study were of an older median age of 6 days with venous blood 23 and in the most recent multi-centre study, the median HCT was 42.7% IQR (38.0-48.0) at a median age of 4 days 24 .…”

“…The study concluded that BS had comparable accuracy to the TcB JM-103. A recent multi-centre study reported the BS to slightly underestimate SBR value with maximum limits of agreements of −128.3 to 102.6 µmol/l 24 .…”