Regional Burden of Enlarged Perivascular Spaces and Cognition and Neuropsychiatric Symptoms in Drug-Naive Patients With Parkinson Disease.

Abstract

Although the potential role of enlarged perivascular spaces (EPVSs) in Parkinson disease (PD) is increasingly recognized, whether EPVSs located in different anatomical regions exert differential effects on clinical manifestation remains uncertain. We investigated the regional EPVS burden and its association with cognition and neuropsychiatric symptoms (NPSs) in newly diagnosed PD population. In this retrospective, cross-sectional study, EPVS in the temporal lobe (T-EPVS), centrum semiovale (CS-EPVS), and basal ganglia (BG-EPVS) were visually rated in drug-naive patients with PD who underwent magnetic resonance imaging, dopamine transporter (DAT) scans, neuropsychological assessments, and Neuropsychiatric Inventory Questionnaire at baseline. Cognitive performance, NPS burden, vascular risk factors, small vessel disease (SVD) imaging markers, and DAT availability were compared across groups dichotomized by their regional EPVS burden (cutoff for high-degree vs low-degree: >10 for T-EPVS/BG-EPVS and >20 for CS-EPVS). A total of 480 patients with PD (123 without cognitive impairment, 291 with mild cognitive impairment, and 66 with dementia) were included. The proportion of high-degree T-EPVS (p for trend <0.001) and BG-EPVS (p for trend = 0.001) exhibited an increasing trend across the cognitive spectrum, corresponding to worsening cognition. Compared with the low-degree group, the high-degree BG-EPVS group showed higher SVD burden (moderate-to-severe white matter hyperintensity [14.8% vs 40.5%, p < 0.001], lacune [10.3% vs 30.7%, p < 0.001], and cerebral microbleeds [8.1% vs 22.2%, p < 0.001]), greater atrophy in cortical gray matter (40.73% ± 1.09% vs 39.96% ± 1.20% of intracranial volume, p < 0.001), and lower cognitive performance (in language [-0.22 ± 1.18 vs -0.53 ± 1.29, p = 0.013], and visual memory domains [-0.24 ± 0.97 vs -0.61 ± 0.96, p = 0.009]). The high-degree T-EPVS group presented with greater NPS burden in decreased motivation (0.61 ± 1.78 vs 1.35 ± 2.36, p = 0.007), affective dysregulation (0.88 ± 2.13 vs 2.36 ± 3.53, p < 0.001), and impulse dyscontrol (0.43 ± 1.67 vs 1.74 ± 4.29, p < 0.001), compared with the low-degree T-EPVS group. Meanwhile, the burden of CS-EPVS did not reveal any differences in cognition or NPS. BG-EPVS and T-EPVS seem to exert differential effects on cognition and NPS in patients with PD. Investigating the EPVS profile in distinct anatomical regions may be useful in disentangling the heterogeneity within PD.