Risk Assessment Strategy for Prediction of Pathological Hyperbilirubinemia in Neonates

Abstract

Objective

To evaluate combined ability of clinical risk factors and transcutaneous bilirubin (TcB) in predicting pathological hyperbilirubinemia (PHB) needing treatment during first week of life in healthy term and late preterm neonates.

Methods

This prospective cohort study included healthy neonates with gestation ≥35 wk and birth weight ≥2000 g. TcB was measured with a multi-wavelength transcutaneous bilirubinometer (Bilichek®) at 30 ± 12 h of postnatal age. Follow-up was conducted as per American Academy of Pediatrics guidelines. For diagnosis of PHB, TcB was measured at each follow-up visit. Serum bilirubin was measured if TcB was >15 mg/dL or within 2 mg/dL of phototherapy cut-off.

Results

Among 462 neonates [birth weight (g; mean ± SD): 2711 ± 431, gestation (wk; median, IQR): 38 (37–39), male: 52%] enrolled in the study, 392 (84.9%) completed followup and PHB was observed in 65 (16.6%) neonates. Discriminant ability of risk model, including both clinical risk factors and TcB, was better than the risk models with clinical risk factors or TcB alone (c-statistic: 0.86 vs. 0.74 vs. 0.77). On logistic regression analysis risk factors found significant were TcB (OR: 1.65, 95% CI: 1.4–1.9), gestation at birth (OR: 0.6, 95% CI: 0.50–0.77) and primiparity (OR: 2.1, 95% CI: 1.1–3.9). A risk prediction score was developed with these three risk factors as ordinal/dichotomous variables. Negative and positive predictive values for score <8 and >12 were 97% and 46%, respectively.

Conclusions

Risk score consisting of TcB, gestation at birth and parity status was able to accurately predict pathological hyperbilirubinemia in derivation cohort of healthy term and late preterm north Indian neonates.