AUTHOR=Ye Zhaolan , Chen Xiang , Wu Daoming , Lin Jiantao , Zhang Lihua 

TITLE=Diagnostic prediction of neonatal hyperbilirubinemia combined with germinal matrix-intraventricular hemorrhage based on cranial ultrasound hemodynamics: a retrospective case-control study

JOURNAL=Frontiers in Medicine

VOLUME=Volume 12 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2025.1606892

DOI=10.3389/fmed.2025.1606892

ISSN=2296-858X

ABSTRACT=ObjectiveTo study the diagnostic prediction of cranial ultrasound hemodynamics in children with neonatal hyperbilirubinemia combined with germinal matrix-intraventricular hemorrhage (GMH-IVH).MethodsWe included 148 hyperbilirubinemic neonates who underwent cranial ultrasound to obtain hemodynamic parameter indexes, and constructed a nomogram visual prediction model through unifactorial and multifactorial analyses to study the role of cranial ultrasound hemodynamic parameters in the diagnostic prediction of neonatal hyperbilirubinemia combined with GMH-IVH.ResultsA total of 148 patients eligible for enrollment were included in this study, of which 40 children developed GMH-IVH, with an incidence rate of 27.03%. Multifactorial logistic stepwise regression analysis showed that mothers suffering from gestational hypertension, total bilirubin ≥15 mg/dl, anterior cerebral artery third day to first day resistance index ratio of ≥1, and middle cerebral artery third day to first day resistance index ratio of ≥1 were the independent risk factors for the development of GMH-IVH in neonatal hyperbilirubinemic infants (P < 0.05); and ROC analysis showed that the area under the ROC curve (AUC) of the prediction model was 0.821 (95% CI: 0.746–0.897, P < 0.001), indicating good predictive efficacy of the model (discrimination), and the Hosmer-Lemeshow test (χ2 = 7.779, P = 0.255) and the calibration curve showed that the model had a good goodness-of-fit (calibration). The predictive model was visualized by plotting nomogram.ConclusionCraniocerebral ultrasound hemodynamics-related parameters combined with clinical features to construct a predictive model for early and effective prediction of the occurrence and prognosis of GMH-IVH in neonates with hyperbilirubinemia.