Treatment practices and implementation of guidelines for hyperbilirubinemia and rebound hyperbilirubinemia

Assessing accuracy of BiliPredics algorithm in predicting individual bilirubin progression in neonates-results from a prospective multi-center study

Background

Neonatal jaundice affects more than half of neonates. As bilirubin values usually peak few days after hospital discharge, jaundice remains a leading cause of rehospitalization. The recently developed BiliPredics algorithm, integrated in the first CE-approved bilirubin prediction tool, predicts individual bilirubin progression for up to 60 h into the future. Goal of the prospective study was to assess accuracy of this algorithm in predicting individual bilirubin prior to hospital discharge in neonates.

Methods

A prospective multi-center study was conducted in 2021 at the University Children's Hospitals in Tübingen and Regensburg, Germany. Various scenarios differing in type and number of bilirubin measurements and in prediction horizon were tested. Primary objective was prediction accuracy of the BiliPredics algorithm based on total serum bilirubin (TSB) measurements or based on transcutaneous bilirubin (TcB) measurements alone. Secondary objective was prediction accuracy based on combinations of TSB and TcB measurements. For assessment of accuracy, two validation metrics, absolute prediction error ( a P E ) and relative prediction error ( r P E ) , and two clinical acceptance conditions, margin of error of the 95%-confidence interval (95%-CI) and percentage of clinically relevant mis-predictions defined as a P E > 85 μ mol / L , were investigated.

Results

Out of 455 enrolled neonates, 276 neonates met bilirubin inclusion criteria and were included in the analyses. Irrespective from tested prediction horizons, median r P E was small (8.5% to 9.5%) utilizing TSB measurements for up to 30 and 60 h and slightly higher (13.8%) utilizing TcB measurements for up to 48 h. The same applied for median a P E . Both clinical acceptance conditions were fulfilled across tested scenarios. Results for combined TSB-TcB scenarios up to a prediction horizon of 48 h without adjustment for type of measurement were comparable to TSB and TcB scenarios fulfilling both clinical acceptance conditions.

Conclusion

Results from this prospective study in neonates confirm that the BiliPredics algorithm accurately predicts bilirubin progression up to 60 h with TSB measurements and up to 48 h with TcB or combined TSB-TcB measurements. As such, prediction tools utilizing this algorithm are expected to facilitate and safely optimize jaundice risk assessment at hospital discharge with the potential to reduce jaundice-related rehospitalizations.

Leveraging Predictive Pharmacometrics-Based Algorithms to Enhance Perinatal Care-Application to Neonatal Jaundice

The field of medicine is undergoing a fundamental change, transforming towards a modern data-driven patient-oriented approach. This paradigm shift also affects perinatal medicine as predictive algorithms and artificial intelligence are applied to enhance and individualize maternal, neonatal and perinatal care. Here, we introduce a pharmacometrics-based mathematical-statistical computer program (PMX-based algorithm) focusing on hyperbilirubinemia, a medical condition affecting half of all newborns. Independent datasets from two different centers consisting of total serum bilirubin measurements were utilized for model development (342 neonates, 1,478 bilirubin measurements) and validation (1,101 neonates, 3,081 bilirubin measurements), respectively. The mathematical-statistical structure of the PMX-based algorithm is a differential equation in the context of non-linear mixed effects modeling, together with Empirical Bayesian Estimation to predict bilirubin kinetics for a new patient. Several clinically relevant prediction scenarios were validated, i.e., prediction up to 24 h based on one bilirubin measurement, and prediction up to 48 h based on two bilirubin measurements. The PMX-based algorithm can be applied in two different clinical scenarios. First, bilirubin kinetics can be predicted up to 24 h based on one single bilirubin measurement with a median relative (absolute) prediction difference of 8.5% (median absolute prediction difference 17.4 μmol/l), and sensitivity and specificity of 95.7 and 96.3%, respectively. Second, bilirubin kinetics can be predicted up to 48 h based on two bilirubin measurements with a median relative (absolute) prediction difference of 9.2% (median absolute prediction difference 21.5 μmol/l), and sensitivity and specificity of 93.0 and 92.1%, respectively. In contrast to currently available nomogram-based static bilirubin stratification, the PMX-based algorithm presented here is a dynamic approach predicting individual bilirubin kinetics up to 48 h, an intelligent, predictive algorithm that can be incorporated in a clinical decision support tool. Such clinical decision support tools have the potential to benefit perinatal medicine facilitating personalized care of mothers and their born and unborn infants.