The following is a summary of “Development and validation of a nomogram to predict intracranial hemorrhage in neonates,” published in the March 2024 issue of Pediatrics by Xu et al.
This study aimed to establish and validate a predictive model for neonatal intracranial hemorrhage (ICH) based on susceptibility-weighted imaging (SWI).
A retrospective analysis enrolled 1,190 neonates suspected of ICH after cranial ultrasound screening in a tertiary hospital. The cohort was randomly divided into training and internal validation cohorts (7:3 ratio). Univariate analysis identified risk factors for ICH, and a predictive model was developed using multivariate logistic regression based on the minimum Akaike information criterion (AIC). External validation was performed in another tertiary hospital with 91 neonates. Model performance was assessed using the area under the curve (AUC), calibration curve, and decision curve analysis (DCA).
Variables including platelet count (PLT), gestational diabetes, mode of delivery, amniotic fluid contamination, and 1-minute Apgar score were selected to establish the predictive model for neonatal ICH. The AUCs were 0.715, 0.711, and 0.700 for the training, internal validation, and external validation cohorts. Calibration curves demonstrated good agreement between predicted and observed ICH rates. DCA indicated the clinical utility of the nomogram.
This study presents an easily applicable nomogram for predicting neonatal ICH, facilitating individualized risk assessment and healthcare decisions.
Source: sciencedirect.com/science/article/pii/S1875957224000378