Abstract: Objective  To analyze the risk factors for complications of autologous arteriovenous fistula (AVF) in maintenance hemodialysis (MHD) patients, and to construct a risk prediction model for the complications.  Method  A retrospective analysis was conducted on the clinical data of 178 MHD patients admitted to Department of Nephrology, The Second Medical Center, PLA General Hospital from March 2021 to March 2023. Patients were divided into a AVF complication group (n=46) and a non-AVF complication group (n=132) based on the complications of stenosis and/or thrombosis in AVF. General data was collected. The independent risk factors for AVF complications in MHD patients were analyzed using univariate and multivariate logistic regression methods, from which a prediction model (R software) was constructed. The receiver operating characteristic (ROC) curve and calibration curve were plotted to evaluate the discrimination and accuracy of the prediction model in predicting the risk of AVF complications in MHD patients.   Results  Among 178 MHD patients, 46 cases were in the AVF complication group accounted for 25.84%, while 132 cases were in the non-AVF complication group accounted for 74.16%. Multivariate regression showed that age >65 years old (OR=1.658, OR 95% CI: 1.267～3.025, P<0.001), education level of junior high school or below (OR=1.543, OR 95% CI: 1.068～2.597, P<0.001), hypotension during dialysis (OR=5.492, OR 95% CI: 2.541～10.597, P<0.001), button puncture method (OR=1.843, OR 95% CI: 1.056～2.819, P<0.001), puncture failure or hematoma formation (OR=2.641, OR 95% CI: 1.273～4.320, P<0.001), platelet count >227.84×109/L (OR=4.413, OR 95% CI: 3.171～9.492, P<0.001) and albumin levels ＜32.83g/L (OR=3.981, OR 95% CI: 2.316～8.873, P<0.001) were the risk factors for AVF complications in MHD patients. ROC curve showed that the area under the curve of nomogram model to predict AVF complications in MHD patients was 0.823 (95% CI: 0.763～0.887), the calibration curve slope was close to 1, and there was no significant difference between the predicted probability and the actual probability (χ2=6.521, P=0.573).  Conclusion  Older age, lower education level, hypotension during dialysis, button puncture method, puncture failure or hematoma formation, higher platelet count, and lower albumin level are the risk factors for AVF complications in MHD patients. The column chart prediction model constructed based on the above risk factors has better discrimination and accuracy abilities.

Key words: Maintenance hemodialysis, Autologous arteriovenous fistula, Complication, Risk factors, Risk prediction model