Towards a Broader Classification of Deep Venous Cerebral Anatomy in Extremely Preterm Infants: A Literature-Based Approach and Its Relevance to Intraventricular Hemorrhage

BACKGROUND AND PURPOSE: Variability in the deep cerebral venous system is hypothesized to contribute to intraventricular hemorrhage (IVH) in preterm infants. Existing classifications do not adequately capture the range of subependymal venous patterns seen on SWI in clinical practice. This study aimed to adapt and expand current classification systems to provide a more comprehensive representation of subependymal vein anatomy in the preterm brain. A secondary aim was to assess maturational changes in venous patterns between 30 and 40 weeks’ postmenstrual age (PMA) and explore their potential association with germinal matrix hemorrhage (GMH-IVH). MATERIALS AND METHODS: We retrospectively analyzed a cohort of extremely preterm infants (born,28 weeks’ gestation) who underwent cranial MRI with SWI at both 30 and 40 weeks’ PMA. In total, 88 infants (176 hemispheres) were included. Images were reviewed to determine the subependymal venous configuration and record GMH, IVH, or periventricular hemorrhagic venous infarction (PHVI). On the basis of observed anatomic variations—and informed by historical anatomic descriptions—a novel classification system was developed. Changes in anatomy between the 2 time points were also recorded. RESULTS: The most frequently observed anatomic pattern (new classification system) was type 1A at 30 weeks’ PMA (31.3%) and types 1A (26.1%) and 2A (25.6%) at 40 weeks’ PMA. The change of deep venous anatomy seen on SWI between 30 and 40 weeks’ PMA was statistically significant (x² ¼ 28.11; P ¼ .005). There was no significant relationship between the subependymal vein type and the change in the anatomic pattern. Hemorrhage occurred in 79 hemispheres (48.1% GMH, 39.2% IVH, and 12.7% PHVI). There was no significant correlation between GMH-IVH-PHVI and the anatomic pattern according to the new classification at 30 weeks’ PMA, or between GMH-IVH-PHVI and the change of anatomic patterns between 30 and 40 weeks’ PMA. CONCLUSIONS: This study presents a detailed classification of deep cerebral venous anatomy in extremely preterm infants using SWI at 30 and 40 weeks’ PMA. Despite observed anatomic variability and maturation, no consistent link was found between venous patterns and GMH-IVH or PHVI. This finding suggests that anatomic variation alone may not strongly influence hemorrhagic injury risk, though limited statistical power may affect this conclusion.