Analysis of Risk Factors for Leukoaraiosis: A Multicenter Retrospective Study

Abstract Background: Leukoaraiosis is a common imaging marker of cerebral small vessel disease. There is now increasing evidence that it is associated with cognitive impairment and a high risk of death after stroke, among others. The aim of this study was to analyze and screen the risk factors clinically associated with the development of leukoaraiosis, and to explore clinical biomarkers that may predict leukoaraiosis. Methods: Inpatients were continuously recruited from July 2014 to October 2020. After admission, the cranial MRI examination was evaluated, and the severity of leukoaraiosis were evaluated and graded. Vascular risk factors and relevant clinical data were collected. Univariate analysis was used to analyze the parameters, and multivariate logistic regression analysis was used to analyze the statistically significant parameters. The analysis results were plotted as ROC curve to find out the diagnostic accuracy of the model. Results: 1), 327 patients meeting the study criteria were included. Univariate analysis showed that 13 factors were statistically significantly (P < 0.05). 2), Multivariate logistic regression model showed that age, elevated systolic blood pressure, ischemic stroke and FT4 were independent risk factors for leukoaraiosis. 3), The ROC curve indicated the accuracy of diagnosis on leukoaraiosis is 0.906, and the positive rate and negative rate are both 85.2%. Conclusions: 1), Factors affecting the development of leukoaraiosis include age, systolic blood pressure, ischemic stroke, and FT4 level. 2), The model of "age, systolic blood pressure, ischemic stroke, FT4" have relatively ideal sensitivity and specificity in predicting the development of leukoaraiosis.