Introduction

Dilated perivascular spaces (dPVS) are visible on MRI and thought to reflect dysfunction in the fluid circulation of the glymphatic system. Glymphatic system in the brain is a system that plays an important role in waste clearance. In adults, dPVS is known to increase with advanced age and is considered an emerging biomarker for various diseases^1,2.

In contrast, dPVS in the young population is poorly understood. Some reports showed a very high prevalence of dPVS in children and other reports showed a higher incidence of dPVS in children with epilepsy³. In neonates, the glymphatic system may play a role in the development and pathogenesis of the brain. However, there is limited information on the severity and regional distribution of dPVS in neonates. Therefore, the objective of our study was to perform a visual and volumetric evaluation of dPVS in neonates.

Method

We collected 2D T2-weighted images of neonates from public data set, the Developing Human Connectome project (http://www.developingconnectome.org)⁴. We included images of both data release 1 and 2. After excluding duplicated cases and the cases that were obtained before term-equivalent age, total of 244 neonates’ images were included (Figure 1).

The 2D turbo spin-echo T2-weighted images were acquired using a 3T scanner (Philips Achieva, Best, Netherlands) using the following scanning protocol: TR, 12 s; TE, 156 ms; voxel size, 0.8 x 0.8 x 1.6 mm³; and reconstruction size, 0.5 x 0.5 x 0.5 mm³.

One radiologist performed visual analysis of dPVS in basal ganglia (BGdPVS) and white matter (WMdPVS), using five-point scales: grade 0, 0 dPVS; grade 1, 1-10 dPVS; grade 2, 11-20 dPVS; grade 3, 21 – 40 dPVS; and grade 4 , > 40 dPVS. For BGdPVS, we excluded perforated substances at the level and below the anterior commissure.

For volumetric analysis, we first segmented the brain regions using Developing Human Connectome project structural pipeline, and then segmented dPVS using Frangi filtering with experimentally determined thresholding value⁵. Fraction of dPVS was calculated by dividing dPVS volume by the region's brain volume (Figure 2).

We compared dPVS extent according to gender and GA at birth (preterm group, < 37 weeks; term group, ≥ 37 weeks) using Mann-Whitney test. We evaluated the association between GA and dPVS volume and GA and dPVS fraction using linear regression. Both GA at birth and GA at MRI scan were used for the analysis.

Result

A total of 244 neonates (median GA at birth, 39 weeks; 145 boys and 99 girls) were evaluated including 88 neonates in the preterm group (median GA at birth, 33 weeks) and 156 neonates in the term group (median GA at birth, 40 weeks). On demographic data comparison, there was no significant difference between the two groups regarding gender (P = 0.892) and GA at MRI scan (P = 0.172).

On visual analysis, most neonates showed WMdPVS grade 0 (227/244) and all neonates showed BGdPVS grade 0 (90/244) or grade 1 (154/244). Preterm neonates showed lower BGdPVS visual grade compared to term neonates (median grade, 0 vs. 1, P < 0.001).

Since most neonates did not show WMdPVS, only BGdPVS were further evaluated for volumetric analysis. Overall median (interquartile range) of BGdPVS volume and fraction were 48.2 mm³ (37.8 mm³ – 59.0 mm³) and 0.189 % (0.155 %– 0.232 %), respectively. There was no significant difference of BGdPVS volume (P = 0.069) and fraction (P = 0.282) according to gender (Figure 3). When preterm and term groups were compared, BGdPVS volume in preterm neonates was significantly lower than that in term neonates (45.3 mm³ vs. 49.9 mm³, P = 0.040) (Figure 3). However, BGdPVS fraction did not show significant difference between the two groups (0.2 % vs. 0.2 %, P = 0.622) (Figure 3). When association with GA at birth was evaluated, there was a significant positive relationship between GA at birth and BGdPVS volume (coefficient = 0.603, P = 0.012). GA at MRI scan and BGdPVS fraction showed a negative relationship (coefficient = - 0.008, P < 0.001) (Figure 4).

Representative cases are shown in Figure 5. A preterm neonate (upper row) showed lower visual grade and volume of BGdPVS compared to those of a term neonate (lower row). Both neonates did not show WMdPVS.

Discussion

Both visual grades and volume of dPVS increased with age in previous studies, however, a recent report showed dPVS decrease with age in a young population⁶. Our results showing lower dPVS extent in preterm neonates than that in term neonates needs careful interpretation. Since preterm neonates’ brains in term-equivalent age generally show underdevelopment compared to that in term neonates, lower dPVS extent in preterm neonates can also be interpreted as an underdeveloped glymphatic system in this group. Smaller dPVS fraction according to GA at MRI scan suggests that the regional brain volume increase exceeds the dPVS volume increase with age in neonates.

Conclusion

In neonates, dPVS were rarely observed in the white matter but commonly observed in the basal ganglia. Gender did affect dPVS extent in neonates and preterm neonates showed lower dPVS extent than term neonates. We showed that dPVS extent, which reflects the glymphatic system in the brain, was affected by preterm birth.

Acknowledgements

Hyun Gi Kim has received funding from the National Research Foundation of Korea (NRF-2021R1A2C1007831)