Key points are not available for this paper at this time.
Objective To compare homocysteine (Hcy) and uric acid (UA) levels between patients with Parkinson’s disease with cognitive impairment (PD-CI) and those with post-stroke cognitive impairment (PS-CI), and to analyze the correlation of these biomarkers with cognitive function and their heterogeneity across different disease types. Methods Patients diagnosed with cognitive impairment who were admitted to the Department of Neurology at Yangjiang Hospital of Traditional Chinese Medicine between January 2025 and December 2025 were included (60 PD-CI; 60 PS-CI). Blood Hcy and UA levels were uniformly measured, and cognitive function was assessed using the Mini-Mental State Examination (MMSE). Covariance analysis compared biomarker level differences between groups. Spearman correlation analysis and multivariate linear regression models with interaction terms explored the relationship between biomarkers and MMSE scores. Binary logistic regression identified factors independently associated with disease type. Results After adjusting for confounding factors, the PD-CI group exhibited significantly higher Hcy levels (adjusted mean difference: 2.55 μmol/L, 95% CI: 0.23–4.87, p = 0.032), while UA levels were significantly lower (adjusted mean difference: −34.66 μmol/L, 95% CI: −63.87 to −5.46, p = 0.020). In the overall sample, Hcy negatively correlated with MMSE ( r = −0.309, p = 0.001). This association was numerically stronger in the PD-CI group ( r = −0.456, p 0.001) and weaker in the PS-CI group ( r = −0.175, p = 0.180). However, the difference in correlation coefficients between the two groups did not reach statistical significance ( p = 0.093). UA levels showed no significant linear or nonlinear association with MMSE scores in either group. Multivariate regression analysis confirmed that elevated Hcy levels maintained a persistent negative correlation with MMSE scores in both groups. Low educational attainment, absence of hypertension, and lower UA levels were identified as independently associated with PD-CI (compared to PS-CI). Conclusion Patients with PD-CI and PS-CI exhibit distinct metabolic profiles, with the PD-CI group characterized by elevated Hcy and low UA levels. Although elevated Hcy is a common factor associated with cognitive impairment in both conditions, UA levels in this cohort were independently associated with PD-CI (relative to PS-CI), not with MMSE scores. These findings suggest potential differences in metabolic profiles and may offer new insights for clinical differential diagnosis. As an exploratory cross-sectional analysis, these findings require validation in prospective studies.
Zhan et al. (2026) studied this question.