Abstract
Acquired brain injury remains common in very preterm infants and is associated with significant risks for short- and long-term morbidities. Cranial ultrasound has been widely adopted as the first-line neuroimaging modality to study the neonatal brain. It can reliably detect clinically significant abnormalities that include germinal matrix and intraventricular hemorrhage, periventricular hemorrhagic infarction, post-hemorrhagic ventricular dilatation, cerebellar hemorrhage, and white matter injury. The purpose of this article is to provide a consensus approach for detecting and classifying preterm brain injury to reduce variability in diagnosis and classification between neonatologists and radiologists. Our overarching goal with this work was to achieve homogeneity between different neonatal intensive care units across a large country (Canada) with regards to classification, timing of brain injury screening and frequency of follow up imaging. We propose an algorithmic approach that can help stratify different grades of germinal matrix-intraventricular hemorrhage, white matter injury, and ventricular dilatation in very preterm infants.