Abstract: Ｏｂｊｅｃｔｉｖｅ：To explore the risk factors of left ventricular diastolic dysfunction (LVDD) in maintenance hemodialysis (MHD) patients and establish a predictive model of LVDD.
Ｍｅｔｈｏｄｏｌｏｇｙ：A total of 75 MHD patients,treated in the Blood Purification Center of the affiliated Hospital of Xuzhou Medical University from 2019 to 2020,were selected as subjects. The clinical indicators,laboratory and imaging data of the patients were collected. Univariate analysis and binary logistic regression analysis were used to screen the independent risk factors of LVDD in patients with MHD. The predictive model of LVDD was established by combining  independent risk factors and its predictive value was tested.
Ｒｅｓｕｌｔｓ：Univariate analysis and binary Logistic regression analysis showed that highsensitive cardiac troponin T(hscTnT) and left ventricular mass index (LVMI) were independent risk factors for LVDD. According to the independent risk factors of LVDD,a predictive model was established. When the total score of risk factors was more than 0406,the sensitivity and specificity of diagnosis were 9024% and 7353% respectively. The area under the receiver operator characteristic (ROC) was 0871 (P<001).
Ｃｏｎｃｌｕｓｉｏｎ：
HscTnT and LVMI are independent risk factors for LVDD in  MHD patients. The prediction model based on independent risk factors is better than a single index,and has a good prediction efficiency for LVDD. 

Key words: maintenance hemodialysis, high-sensitive cardiac troponin T, left ventricular diastolic dysfunctionrisk factors, prediction