# Association between white matter hyperintensities and gray matter volume in cerebral small vessel disease: insights from periventricular and deep white matter lesions

## TL;DR

This study finds that periventricular white matter lesions are linked to reduced gray matter volume in brain disease patients.

## Contribution

The study identifies periventricular white matter hyperintensities as an independent predictor of gray matter loss in cerebral small vessel disease.

## Key findings

- PWMH severity is independently associated with lower normalized gray matter volume.
- PWMH mediates 39.1% of the age-related decline in gray matter volume.
- DWMH severity does not show an independent association with gray matter volume.

## Abstract

To investigate the relationship between the severity of periventricular white matter hyperintensities (PWMH) and deep white matter hyperintensities (DWMH) and gray matter volume in cerebral small vessel disease (CSVD) patients.

Clinical and imaging data from 125 CSVD patients, collected between April and December 2022, were analyzed. The severity of PWMH and DWMH was assessed using the Fazekas scale. 3D T1-weighted images were processed with FSL software to segment gray matter and intracranial volume. The normalized gray matter volume (GM_n) was computed as the proportional ratio of gray matter volume relative to total intracranial volume.

Group-level analysis revealed significant differences in GM_n were observed across various Fazekas scores for both DWMH (p = 0.041) and PWMH (p < 0.001). Post hoc analysis with false discovery rate (FDR) correction showed that PWMH severity was associated with significantly lower GM_n in scores 2 and 3 compared to score 1 (FDR-adjusted p = 0.001 and 0.004, respectively), though no difference was observed between scores 2 and 3 (p > 0.05). For DWMH, no pairwise comparisons survived FDR correction, though a non-significant trend toward reduced GM_n was noted in score 3 versus scores 0–1 (FDR-adjusted p = 0.08). After multivariate analysis adjusting for vascular risk factors, PWMH severity was found to be independently associated with GM_n [p < 0.05, β (95% CI): Age −0.212 (−0.412, −0.011), PWMH score −0.408 (−0.605, −0.210)], while DWMH severity showed no independent association with GM_n (p > 0.05). Mediation analysis revealed PWMH significantly mediated 39.1% of the total effect of age on GM_n (indirect effect = −0.136; 95% CI: −0.265 to −0.053).

PWMH, but not DWMH, is independently associated with gray matter volume and partially mediates the age-related decline in GM volume. These findings highlight the topographical and mechanistic heterogeneity of WMH subtypes and underscore PWMH as a potential imaging marker for early intervention in CSVD-related neurodegeneration.