Neurodevelopmental Outcomes in Term Infants with Neonatal Hyperbilirubinaemia in China: A Retrospective Cohort Study...

Background: Neonatal hyperbilirubinaemia (NHB) affects approximately 60% of term infants and is a recognised cause of bilirubin-induced neurologic dysfunction (BIND); however, its subclinical neurodevelopmental sequelae have not been well characterised at the level of specific developmental domains. Objectives: We aimed to use the Griffiths Developmental Scales-Chinese Edition (GDS-C) to characterise the domain-specific neurodevelopmental profile of term infants with NHB and to identify clinical risk factors for adverse outcomes. Methods: We conducted a single-centre, retrospective cohort study of 123 term newborns delivered between September 2019 and August 2023 at a tertiary hospital in Shanghai, China; 77 had NHB and 46 were healthy controls. Neurodevelopmental outcomes were assessed using the GDS-C at a median age of 46.9 months (interquartile range [IQR], 36.4-59.7) by a single certified examiner who was blinded to bilirubin status. A developmental quotient (DQ) below 85 (>1 SD below the standardised mean of 100) or at or below the 10th percentile was classified as below cutoff; otherwise, performance was classified as within the normal range. To account for testing across six domains, a Bonferroni-adjusted significance threshold of p < 0.0083 was applied. Results: Compared to the control group, the jaundice group had higher proportions of below-cutoff performance in the Locomotor (26.0% vs. 10.9%; p = 0.045), Personal-Social (28.6% vs. 10.9%; p = 0.022), Eye and Hand Coordination (29.9% vs. 10.9%; p = 0.015) and Performance (35.1% vs. 15.2%; p = 0.018) domains. Among infants with severe hyperbilirubinaemia (total serum bilirubin (TSB) ≥ 342 μmol/L; n = 19), the proportions of below-cutoff performance in the Locomotor (47.4% vs. 19.0%; p = 0.032), Personal-Social (57.9% vs. 19.0%; p = 0.003) and Performance (57.9% vs. 27.6%; p = 0.034) domains exceeded those in the non-severe subgroup. Jaundice lasting ≥ 14 days was associated with poorer personal-social outcomes (p = 0.007). In multivariable logistic regression, both a peak TSB ≥ 342 μmol/L (adjusted odds ratio [aOR], 5.59; 95% confidence interval [CI], 1.64-19.02; p = 0.006) and a jaundice duration ≥ 14 days (aOR, 5.68; 95% CI, 1.61-20.04; p = 0.007) were independently associated with below-cutoff personal-social performance. Conclusions: Among term infants, NHB was associated with an increased risk of below-cutoff performance across several GDS-C domains, particularly those reflecting gross motor, personal-social and visual-spatial functions. Severe hyperbilirubinaemia and prolonged jaundice were independent risk factors. The GDS-C may serve as a sensitive, domain-specific instrument for the early identification of infants at risk of adverse neurodevelopmental outcomes.