Pilot study of association between neonatal brain perivascular space volume and neurodevelopmental outcomes at 24 months.

Introduction: Perivascular space (PVS) has recently gained attention as a neurological indicator. However, there is still limited research on the relationship between basal ganglia PVS (BG-PVS) volume and neurodevelopmental outcome of neonates. Therefore, this pilot study aimed to investigate the association between BG-PVS volume of neonates at term-equivalent age and their neurodevelopmental outcome at 24 months.

Methods: This single-center retrospective pilot study included neonates who underwent brain MRI between 2019 and 2022 at term-equivalent age and had neurodevelopmental assessment at 24 months using Bayley Scales of infant development third (Bayley-III). Sample size was determined by feasibility constraints with recruitment of available neonates from our clinical population. BG-PVS volume was extracted from brain MRI using 3D T2-weighted images through a combination of computational processing and manual refinement. Multiple linear regression models were used to examine the associations between BG-PVS volume and three Bayley-III scores (cognitive, language, and motor), adjusting for postmenstrual age. BG-PVS volumes between neonates with normal and delayed development were compared using Wilcoxon rank-sum tests with False Discovery Rate correction for multiple comparisons.

Results: A total of 14 neonates were included (8 [57%] males; median gestational age, 246 [243-254] days). There were negative associations between BG-PVS volume and cognitive (coefficient = -0.70, P = 0.04), language (coefficient = -0.69, P = 0.01), and motor (coefficient = -0.71, P = 0.02) development scores. Neonates with delayed development showed larger BG-PVS volumes compared to those with normal development for language (38 [27-45] mm³ vs. 18 [17-25] mm³, P = 0.02) and motor domains (50 [49-51] mm³ vs. 26 [20-36] mm³, P = 0.02). BG-PVS volume was larger in neonates with delayed cognitive development compared to those with normal development, but the difference was not significant (49 [40-50] mm³ vs. 25 [20-36] mm³, P = 0.06).

Conclusion: In this pilot study, larger BG-PVS volume in neonates at term-equivalent age was associated with poorer developmental outcomes at 24 months. These preliminary findings suggest that BG-PVS volume may warrant further investigation as a potential early imaging biomarker for neurodevelopmental risk assessment in neonates, though larger studies are needed to confirm these associations.