Association between stroke and memory diseases: evidence from a prospective national cohort study in China

Kai Zhou¹Zhongyue Lv^2*Jiehua Gan²Guomin Xie²Chen Xiao²

• ¹Department of Neurology, Ningbo Medical Centre Li Huili Hospital, Ningbo, Zhejiang Province, China
• ²Ningbo Medical Centre Lihuili Hospital, Ningbo, China

Background: Previous studies had shown stroke played an important role in the pathogenesis of memory diseases. Thus, this study investigated the correlation between stroke and memory diseases (Alzheimer's disease (AD) and brain atrophy) and provided a new theoretical basis for the diagnosis of stroke disease. Methods: A total of 15,904 total participants were obtained based on the China Health and Retirement Longitudinal Study (CHARLS), including 322 stroke subjects and 15582 control subjects. Stroke was outcome variable, after the questionnaire, the subjects were divided into stroke and control groups. Meanwhile, various categorical variables, such as memory diseases (AD, brain atrophy), gender and medical insurance were included in this study. The weighted chi-square test was used to analyze whether there were differences in covariates between stroke and control groups. In addition, the correlation between memory diseases and stroke were analyzed by weighted logistic regression. Receiver operating characteristic (ROCs) curves were used to assess the accuracy and reliability of the Model III. Results: Stroke and control subjects differed significantly in a variety of clinical characteristics, and variables such as healthy status, patient service and memory diseases were significantly correlated with stroke prevalence. According to the three models constructed in this study, memory diseases was significantly associated with stroke in all three models (Model I, odds ratio (OR) = 7.33, p < 0.001, 95% Confidence interval (CI) = 5.31-9.94; Model II, OR = 7.33, p < 0.001, 95% CI = 5.31-9.95; Model III, OR = 3.84, p < 0.001, 95% CI = 2.73-5.30). Weighted logistic regression analysis showed the stability of the relationship between memory diseases and stroke, further suggested the correlation between memory diseases and stroke. Finally, the area under curve (AUC) of 0.778 indicated that the prediction accuracy of Model III was better. Conclusions: According to the results in this study, there was a significant association between memory diseases and stroke. It is worthwhile to further study the mechanisms between stroke and memory diseases.