Cross-sectional reference values of cerebral ventricle for Chinese neonates born at 25–41 weeks of gestation

To establish the cross-sectional reference values of cerebral ventricular size for the Chinese newborns by the most correlated explanatory variables. The anterior horn width (AHW), thalamo-occipital distance (TOD), and ventricular index (VI) were collected prospectively from 1- to 7-day neonates without potential neurological problems. All neonates were delivered or treated at the Hunan Provincial Maternal and Child Health Care Hospital or Second Xiangya Hospital of Central South University between February and August 2021. The most correlated explanatory variables were identified with the max-min normalization and multiple regression. The reference values were then established based on the above variables. Additionally, intraclass correlation coefficients (ICC) were applied to evaluate the reliability of the overall data collection process. This prospective study consisted of 1848 neonates. The AHW was most highly correlated with GA; the TOD and VI were most strongly correlated with birth weight. All the foregoing correlations were positive ones. Heteroscedasticity and influential points existed in both TOD and VI. The ICCWe recommend using the GA-based AHW reference values and birth weight-based TOD and VI ones. We also present a comparison of GA-based upper limits from all available reference intervals. Moreover, we determine that measurement errors are the primary cause of influential points and heteroscedasticity in TOD and VI studies and infer that the studies of TOD and VI are vulnerable to them.• Reference values of infantile cerebral ventricles are vital in diagnosing and treating cerebral ventricular dilatation. • Precursors established gestational age-based reference values subjectively.• We set cross-sectional reference values based on the most correlated variables for Chinese neonates and compared all available gestational age-based upper limits. • Influential points and heteroscedasticity mainly caused by measurement errors are common in TOD and VI studies.