Quantitative Auswertung der Vermessungen der neonatalen Seitenventrikel bei posthämorrhagischem Hydrozephalus - Vergleich der Levene-Messung und der Längen-Breiten-Messung

Background: Posthemorrhagic ventricular dilatation of the brain is a common complication following intraventricular hemorrhage (IVH) in preterm infants. The Levene index is commonly used to monitor ventricular size, but alternative ultrasound measurements such as ventricular length, width, and the cross sectional area may provide additional information. Aim: In this study, the Levene index will be compared with length–width–based ventricular measurements to assess how strongly they correlate at different time points and whether they reflect ventricular enlargement in a similar or different way. A central question is whether there is indication that the length width method can detect changes earlier or more sensitively than the Levene index. Methods: Ultrasound data from 137 mostly preterm infants with IVH were analyzed retrospectively. Length, width, area, and Levene measurements were collected from up to ten predefined time points at the same coronal plane. Spearman correlation coefficients were calculated for each time point. Results were additionally presented as spaghetti- and scatterplots. Results: The Levene index showed strong correlations with the length–width–based measurements (length, width, and cross-sectional area). The additional parameters followed a similar general pattern of ventricular enlargement. With length, the Levene index showed the strongest and most consistent correlation. Width and area also correlated mostly strong with the Levene index but showed greater variability, resulting in (mostly) slightly lower correlation coefficients compared to length. Especially in the early posthemorrhagic phase, the difference in width was more pronounced at the first time points, with only moderate correlation, but soon became stronger and then showed greater stability over time. Conclusion: Overall, the alternative measurements did not differ substantially from the Levene index in terms of their informative value with regard to ventricular enlargement. These results confirm the Levene index as a reliable and practical method for routine monitoring of ventricular size, as it is also less complex and therefore easier to perform. The results of width in this study raise the question of whether there is a potential difference in sensitivity in the early posthemorrhagic phase, as the correlations are rather moderate at this period.