Quantitative and Qualitative MRI Assessment of Perivascular Spaces in Parkinson’s Disease Patients

Background and Objectives: Enlarged perivascular spaces (ePVS) demonstrated by MRI have recently been associated with cerebral small vessel disease and glymphatic dysfunction, implicated in Parkinson’s disease (PD) pathophysiology. This study aimed to quantify the burden of ePVS in PD patients versus healthy controls and to examine associations with cognitive performance. Materials and Methods: A total of 51 participants underwent 3T MRI, including a T2-weighted sequence. Twenty-one patients with Parkinson’s disease and 21 age-matched healthy controls were included in the final analysis. The ePVS burden was assessed quantitatively by counting visible PVS in the basal ganglia and centrum semiovale, and qualitatively using Potter and Heier rating scales. Cognitive function was measured with the Montreal Cognitive Assessment (MoCA). Statistical analyses used Mann–Whitney U tests and Spearman correlations. Results: PD patients had significantly higher total PVS counts in the basal ganglia (84.8 vs. 48.0; p < 0.001) and centrum semiovale (290.6 vs. 143.9; p < 0.001). Potter scale ratings were higher in PD across regions (p ≤ 0.025). Largest per-slice PVS counts negatively correlated with MoCA scores in right basal ganglia (ρ = −0.362, p = 0.012) and bilateral centrum semiovale (right: ρ = −0.421, p = 0.003; left: ρ = −0.431, p = 0.002). Heier scale differences were significant only in the right centrum semiovale (p = 0.023). PVS diameters were larger in PD only in the centrum semiovale (right: p = 0.010; left: p = 0.040). Conclusions: In this cohort, increased ePVS burden in the basal ganglia and centrum semiovale was associated with cognitive impairment in PD patients. Qualitative and quantitative PVS assessment, notably the largest-per-slice counts, may serve as a sensitive, non-invasive imaging biomarker for neurodegeneration and cognitive decline in PD. Larger group studies and longitudinal data are needed to assess their prognostic value in the long term, as well as the development of automatic quantification applications for better reproducibility.