Systematic review of risk prediction models for autogenous arteriovenous fistula dysfunction in maintenance hemodialysis patients.

OBJECTIVE

To systematically evaluate existing risk prediction models for autogenous arteriovenous fistula (AVF) dysfunction in maintenance hemodialysis (MHD) patients, in order to provide a reference for clinical practice.

MATERIALS AND

METHODS

A systematic search was conducted in China National Knowledge Infrastructure (CNKI), Wanfang Data Knowledge Service Platform, VIP Chinese Science and Technology Journal Database, Chinese Biomedical Literature Database (CBM), PubMed, Web of Science, Embase, and The Cochrane Library ror studies on the development and validation of risk prediction models for autogenous arteriovenous fistula (AVF) dysfunction in MHD patients. The search time was from database inception to November 1, 2025. Two researchers independently screened the literature and extracted data. The Prediction Model Risk of Bias Assessment Tool (-PROBAST) was used to evaluate the risk of bias and applicability of the included studies.

RESULTS

A total of 20 articles and 21 prediction models were included. The incidence of autogenous AVF dysfunction in MHD patients ranged from 3.9 to 49.5%, and the area under the ROC curve (AUC) of the models ranged from 0.530 to 0.949. Age, diabetes mellitus, intradialytic hypotension, calcium-phosphorus product, fibrinogen, and platelet count were the most common predictive factors. The quality assessment showed that most studies had a relatively high risk of bias.

CONCLUSION

Research on risk prediction models for autogenous AVF dysfunction in MHD patients is still in its early stage. Overall, the risk of bias is relatively high, and clinical application is lacking. In the future, researchers may adopt multicenter prospective designs based on machine learning methods to develop high-quality risk prediction models with high accuracy and strong generalizability.