Accuracy of transcutaneous bilirubin determination in neonatal hyperbilirubinemia: A meta-analysis

ABSTRACT

BACKGROUND@#Timely initiation of therapy for neonatal hyperbilirubinemia is routinely made based on total serum bilirubin levels. However, serial samplings by invasive needle pricks are needed for laboratory analyses. Studies comparing the correlation between serum bilirubin and transcutaneous bilirubin have yielded diverse results. A meta- analysis was done to find out the relationship between transcutaneous bilirubin measurements and serum bilirubin values.@*OBJECTIVE@#This study aims to analyze scientific articles regarding the accuracy of transcutaneous bilirubin measurements among healthy neonates as an alternative screening for hyperbilirubinemia.@*STUDY DESIGN@#Diagnostic Accuracy meta- analysis.@*METHODS@#Studies on the accuracy of transcutaneous bilirubin measurements were identified through intensive literature search. Local studies were confirmed thru personal communication.@*RESULTS@#Three hundred eighteen studies were identified through literature search. Ten studies met the eligibility criteria. Eight of the ten studies reported results as correlation coefficients. The pooled estimates of correlation coefficients is high at r = 0.85 (95% CI = 0.84 to 0.857). Five studies reported results with data for diagnostic accuracy. The pooled analysis for sensitivity and specificity are high at 0.84 (95% CI 0.8-0.88) and 0.79 (95% CI 0.77-0.81) respectively. The pooled likelihood ratio has a significant difference with a pooled positive LR of 4.19 (95% CI 2.98-5.9, P<0.01) while the negative likelihood ratio is 0.23 (95% CI 0.17 to 0.29). The AUC for transcutaneous bilirubinometry is 0.89.@*CONCLUSIONS@#Transcutaneous bilirubin measurement can be an alternative in monitoring the risk of healthy neonates for hyperbilirubinemia based on the pooled analysis of correlation coefficient and diagnostic accuracy.