Regional white matter hyperintensity volume in Parkinson's disease and associations with the motor signs

Abstract

Objective: To determine whether white matter hyperintensity (WMH) volumes in specific regions are associated with Parkinson's disease (PD) compared to non-PD controls, and to assess their impact on motor signs through cross-sectional and longitudinal analyses.

Methods: A total of 50 PD participants and 47 age- and gender-matched controls were enrolled. All PD participants were followed up for at least 2 years. To detect regions of greater WMH in the PD, the WMH volume of each region was compared with the corresponding region in the control group. Linear regression and linear mixed effects models were respectively used for cross-sectional and longitudinal analyses of the impact of increases in WMH volume on motor signs.

Results: The PD group had greater WMH volume in the occipital region compared with the control group. Cross-sectional analyses only detected a significant correlation between occipital WMH volume and motor function in PD. Occipital WMH volume positively correlated with the severity of tremor, and gait and posture impairments, in the PD group. During the follow-up period, the participants' motor signs progressed and the WMH volumes remained stable, no longitudinal association was detected between them. The baseline occipital WMH volume cannot predict the progression of signs after adjustment for baseline disease duration and the presence of vascular risk factors.

Interpretation: PD participants in this study were characterized by greater WMH at the occipital region, and greater occipital WMH volume had cross-sectional associations with worse motor signs, while its longitudinal impact on motor signs progression was limited.

Figure 1

Flowchart illustrating how the white matter hyperintensities (WMHs) were divided into six regions on the basis of atlas. Fluid‐attenuated inversion recovery (FLAIR) images of each participant were converted into Mayo Clinic Adult Lifespan space by reference to their associated T1 images; transformation matrices were then obtained (steps 1 and 2). Inverse transformation matrices were used to transform the atlas images back into FLAIR images (steps 3–5). Finally, the WMH volume of each region was determined by overlaying the WMH map (segmented from FLAIR images) onto the atlas image in native space (steps 6 and 7). DGW, deep gray and white matter.

Figure 2

Comparison of baseline white matter hyperintensity (WMH) volumes between Parkinson's disease (PD) participants and controls. All WMH volumes were normalized by the intracranial volume and log‐transformed. The bar shows mean and standard deviation values for the PD (blue) and controls (yellow) across six regions. PD participants had greater WMH volumes in the occipital region compared with controls (marked with “*”). No significant differences were detected between the two groups in other regions. DGW, deep gray and white matter.

Figure 3

Association between motor signs and occipital white matter hyperintensity (WMH) volume in Parkinson's disease participants at baseline. All WMH volumes were normalized by the intracranial volume and log‐transformed. The occipital WMH volume was significantly correlated with tremor (A) and gait/posture impairment (D) after adjustment for age, gender, vascular risk factors and disease duration, but not with bradykinesia (B) or rigidity (C). All p‐values were corrected by the false discovery rate.

Figure 4

Longitudinal changes in motor function and global white matter hyperintensity (WMH) volume. The red, blue, and green lines represent the changes in motor function (A), global WMH volume (B), and absolute global WMH volume (before normalization and log transformation, C), respectively, in the Parkinson's disease group. The black dots and connected lines indicate the changes of variables during follow‐up at the subject level. Motor function was evaluated using the Unified Parkinson's Disease Rating Scale Part III (UPDRS III).