ABSTRACT
BACKGROUND: This study aimed to determine the accuracy of a point-of-care Bilistick method for measuring total serum bilirubin (TSB) and its turn-around-time (TAT) against hospital laboratory methods. METHODS: This prospective study was carried out on 561 term-gestation jaundiced neonates in two Malaysian hospitals. Venous blood sample was collected from each neonate for contemporary measurement of TSB by hospital laboratories and Bilistick. TAT was the time interval between specimen collection and TSB result reported by each method. RESULTS: The mean laboratory-measured TSB was 194.85 (±2.844) µmol/L and Bilistick TSB was 169.37 (±2.706) µmol/L. Pearson's correlation coefficient was: r = 0.901 (p < 0.001). The mean difference of [laboratory TSB- Bilistick TBS] was 26.48 (±29.41) µmol/L. The Bland-Altman plots show that the 95% limits of agreement (-31.1577, 84.11772) contain 94.7% (=531/561) of the difference in TSB readings. Bilistick has a 99% accuracy and 100% sensitivity to predict laboratory TSB levels of ≥80 µmol/L and ≥360 µmol/L at lower Bilistick TSB levels of ≥55 and ≥315 µmol/L, respectively. TAT of Bilistick TSB (2.0 min) was significantly shorter than TAT (105 min) of laboratory TSB (p < 0.001). CONCLUSIONS: Bilistick has shorter TAT. The accuracy and sensitivity of Bilistick TSB for predicting laboratory TSB is high at lower cutoff levels.