Detection and Outcome Associations of Low-Grade Intraventricular Hemorrhage in Very Preterm Infants: Influence of Neuroimaging Modality and Grading System

PURPOSE: The clinical relevance of low-grade intraventricular hemorrhage (IVH) in very preterm infants remains debated. We investigated how cMRI and cranial ultrasound (cUS) influence IVH detection and grading and how differences between the Papile and Volpe classifications affect interpretation and outcome stratification. METHODS: This retrospective, single-center cohort study included 507 consecutively born infants at < 32 weeks' gestation with serial cUS and cMRI at term-equivalent age (2009-2018). IVH was graded according to the Papile and Volpe classifications. Clinical parameters and neurodevelopmental outcomes at 24 months of corrected age were assessed (Bayley Scales of Infant Development). Univariable and multivariable regression analyses were performed. RESULTS: Among 507 preterm infants, IVH was identified in 85 (16.8%; IVH I° n = 33, IVH II° n = 35, IVH III° n = 13, PVHI n = 4). CMRI and cUS differed significantly in IVH detection (p = 0.025). IVH was not detected on cUS in 32 cases (37.6%, IVH I° n = 21, IVH II° n = 9, PVHI n = 2). Additionally, 48.6% of IVH II° according to Papile were classified as IVH I° on cUS due to non-detection of intraventricular blood. IVH II° was associated with significantly lower cognitive and motor outcomes scores than IVH I° (p = 0.035/p = 0.016). Differences in outcome stratification between grading systems were significant (cognitive p = 0.047/motor p = 0.015) and driven by differences in classification of intraventricular blood. CONCLUSION: Detection of intraventricular blood was associated with worse cognitive and motor outcomes. Differences between IVH grading systems and the choice of neuroimaging modality may influence risk stratification in infants with low-grade hemorrhage, particularly when small amounts of intraventricular blood are present.