ABSTRACT
Objective:To establish the reference values and neurological intervention cutoffs for cerebral ventricular size in neonates born at 33 +0-41 +6 weeks of gestation and to investigate the influential factors and reliability of the related indices. Methods:This study prospectively recruited 1 370 1-to 7-day neonates born or hospitalized at the Hunan Provincial Maternal and Child Health Care Hospital from February to August 2021. All the neonates, who were born between 33 +0 and 41 +6 weeks of gestation, were subjected to ultrasound scanning to obtain the indices, including ventricular index (VI), anterior horn width (AHW), thalamo-occipital distance (TOD), and ventricular height (VH). The reference value and neurological intervention cutoff for each index were set. Quantile regression was used to estimate the correlation between each index and continuous covariates [gestational age at birth (GA) and birth weight (BW)]. Mann-Whitney U test was used to analyze the differences in the medians of indices in different categorical covariates groups (males/females, left/right lateral ventricles, vaginal delivery/cesarean section, and singleton/multiple births). Intraclass correlation coefficient (ICC) calculated by a two-way mixed effect model and absolute agreement was used to access intra-rater reliability; ICC via a two-way random effect model and absolute agreement was utilized to rate inter-rater reliability (pool reliability: ICC below 0.50; moderate reliability: ICC between 0.50 and 0.75; good reliability: ICC between 0.75 and 0.90; excellent reliability: ICC exceeding 0.90). Results:The upper limits of reference values for AHW, TOD, VI, and VH in 555 (40.5%) preterm neonates were 2.7-3.5 mm, 20.9-22.5 mm, 12.6-13.7 mm, and 3.8-4.9 mm, and in 815 (59.5%) term newborns were 3.4-4.3 mm, 18.6-21.3 mm, 14.2-14.7 mm, and 3.4-3.8 mm, respectively. The cutoff of neurosurgical intervention for each index was the upper limit of reference value plus 4 mm. AHW median was positively correlated with GA [partial regression coefficient (PRC): 0.12, P<0.05], while TOD and VH medians were negatively correlated with GA (PRC:－0.31 and－0.06, both P<0.05). VI, AHW, and TOD medians were positively associated with BW (PRC: 0.46, 0.23, and 0.97, all P<0.05). The medians of VH, AHW, and TOD in the left cerebral ventricular exceeded those in the right cerebral ventricular, respectively (VH: 2.0 vs 1.8 mm, U=836 071.50; AHW: 1.8 vs 1.7 mm, U=874 141.50; TOD: 13.6 vs 12.5 mm, U=738 409.00, all P<0.05). The medians of AHW and VI in male neonates were greater than those in female newborns, respectively (AHW: 1.8 vs 1.7 mm, U=834 124.00; VI: 11.1 vs 10.8 mm, U=884 156.50, both P<0.05). The neonates delivered vaginally had greater AHW median, but smaller TOD median than those delivered by cesarean section (AHW: 2.0 vs 1.6 mm, U=685 546.00, P<0.001; TOD: 13.1 vs 12.9 mm, U=850 797.00, P=0.010). The AHW median in singleton newborns exceeded that in multiple births (1.9 vs 1.4 mm, U=356 999.00, P<0.001). The lower limits of 95% confidence intervals for intra-rater and inter-rater ICCs exceeded 0.75 and 0.50, respectively. Conclusion:Reference values and surgical intervention thresholds for VI, AHW, TOD, VH of newborns with a gestational age of 33 +0-41 +6 weeks were preliminarily established, and the reliability of these indicators were verified.