Predictive model of small vessel vascular dementia based on comorbidities, cytokines and claudin-1 genetic polymorphisms

Although small vessel vascular dementia (SmVD) remains a significant contributor to global disease burden, its aetiology and pathogenesis are not fully understood. Past research has highlighted an association of SmVD with blood-brain barrier (BBB) dysfunction and neuroinflammation. Studies of claudin-1 polymorphisms and serum cytokines have been found to be useful biomarkers and offer possible mechanistic relationship. However, the pathophysiology and primary factors that contribute to the development and onset of SmVD remain to be further elucidated. This study aimed to evaluate the association of these biomarkers with SmVD through the construction of a logistic regression model. Two datasets from a study group, sourced from the publications of “ Association of genetic polymorphisms of claudin-1 with small vessel VD” and “ Macrophage and microglia related chemokines are associated with small vessel (white matter) VD: A case-control study ”, were merged with multiple imputation chain equations to allow for the construction of a unified logistic regression model. The model incorporated an L2 regularisation penalty to improve the interpretability of the findings. The results showed a significant link between disease comorbidities and SmVD, particularly those with a history of stroke and Parkinson’s diseases, while cytokines displayed a weaker association with SmVD. Overall, this study supported the mechanistic theories linking SmVD with BBB dysfunction and neuroinflammation.