Validation of a transcutaneous bilirubin meter in Mongolian neonates: comparison with total serum bilirubin

Abstract

Background: Neonatal hyperbilirubinemia, especially kernicterus, can be prevented by screening for neonatal jaundice. The transcutaneous bilirubin (TcB) meter is a non-invasive medical device for screening neonates. The study aimed to investigate the validity of a TcB meter in a resource-limited setting such as Mongolia.

Methods: Term and late preterm neonates from the National Center for Maternal and Child Health of Ulaanbaatar in Mongolia who met the inclusion criteria (gestational age ≥35 weeks, birth weight ≥2000 g, postnatal age ≤ 1 month) were enrolled in the study. We used a TcB meter, JM-103 to screen for neonatal jaundice. TcB measurements at the infant's forehead and midsternum were performed within 3 h of obtaining samples for total serum bilirubin (TSB) measurement. We analyzed the correlation between TcB measurements and TSB measurements to validate the meter.

Results: A total of 47 term and six late preterm neonates were included in the study. TcB measured by the meter at both the forehead and the midsternum showed a strong correlation with TSB measured in the laboratory. The correlation equations were TSB = 1.409+0.8655 × TcB (R2=0.78871) at the forehead, and TSB = 0.7555+0.8974 × TcB (R2=0.78488) at the midsternum. Bland-Altman plots and the Bradley-Blackwood test showed no significant differences between the two methods at all measured ranges of bilirubin. The mean areas under the curves of TcB at the forehead and midsternum at three TSB levels (>10 mg/dL, >13 mg/dL, >15 mg/dL) of TcB were greater than 0.9, and all had high sensitivity and specificity.

Conclusions: This study established the validity of the JM-103 meter as a screening tool for neonatal jaundice in term and late preterm infants in Mongolia. Future studies are needed, including the establishment of a TcB hour-specific nomogram, for more effective clinical practice to prevent severe hyperbilirubinemia.

Figure 1

Linear regression plots (solid lines) of JM-103 TcB versus TSB measurements at different measurement sites (A: forehead, B: midsternum) in the neonate.

Figure 2

Bland-Altman plots of TSB and TcB in the neonates. The graphs show the difference in mean values between TSB and TcB (forehead) (A), TSB and TcB (midsternum) (B), and TcB (midsternum) and TcB (forehead) (C). The P-value by the Bradley-Blackwood test was 0.315 in (A), 0.073 in (B) and 0.247 in (C). FH: forehead; CH: midsternum.

Figure 3

Receiver operating characteristics curves for JM-103 cutoff values (mg/dL) at TSB >10 mg/dL, >13 mg/dL, or >15 mg/dL. The areas under the curves were 0.962 (A), 0.945 (B), 0.908 (C), 0.987 (D), 0.913 (E), and 0.903 (F).