>

027. Natural history of post-hemorrhagic ventricular dilatation and prediction of persistent symptomatic ventriculomegaly in preterm infants

Authors: Grace Yee Yan Lai, MD

Introduction:
There are currently no empirically derived thresholds to predict need for surgical intervention in preterm infants with severe intraventricular hemorrhage (IVH). The purpose of this study is to characterize the progression of ventriculomegaly in preterm infants with severe IVH and derive optimal thresholds to predict need for neurosurgical intervention.Methods: Retrospective review of neuroimaging and neurosurgical outcomes in preterm infants with grade III/IV IVH admitted between 2007-2020. Frontal-occipital horn ratio (FOHR), frontal-temporal horn ratio (FTHR), anterior horn width (AHW), and ventricular index (VI) were measured from all cranial imaging studies during hospitalization. Generalized estimating equations were used to compare time-course of ventricular measurements between patients who had neurosurgical intervention, those who had neurosurgery consult but no intervention, and those who did not have neurosurgery consult. Area under the receiver operating curve (AUC) for predicting intervention was calculated for measurements on diagnostic scan, scans during weeks 1-4, and maximum measurement prior to intervention. Threshold values that optimized sensitivity and specificity were derived for each index.
Results:
1254 scans across 132 patients with grade III/IV IVH (mean 26.3±2.97 weeks gestational age, 82 male) were measured. Time course over the first 16 weeks of life differed across groups (p < 0.001). 37 patients had neurosurgical intervention and 19 had consult but no intervention. All indices differed between those with and without intervention starting from the first diagnostic scan (p < 0.001). For prediction of need for intervention, AUC of maximum measurement was 97.1%(95%CI94.6-99.7) for FOHR, 97.7%(95%CI95.6-99.8) for FTHR, 96.6%(95%CI93.9-99.4) for AHW, and 96.8%(95%CI94.0-99.5) for VI. Optimal threshold of maximum measurement was 0.66 for FOHR, 0.62 for FTHR, 15.5mm for AHW, and 8.4mm>p97 for FOHR, 97.7%(95%CI95.6-99.8) for FTHR, 96.6%(95%CI93.9-99.4) for AHW, and 96.8%(95%CI94.0-99.5) for VI. Optimal threshold of maximum measurement was 0.66 for FOHR, 0.62 for FTHR, 15.5mm for AHW, and 8.4mm>p97 for VI (sensitivity>86.8%, specificity>90.1%).
Conclusion:
Ventriculomegaly indices were greater for patients who required intervention from the first diagnostic scan and had excellent prediction for need of neurosurgical intervention. Ideal thresholds should minimize intervention and optimize long-term neurodevelopmental outcomes.