关键词: font-family:Inter, -apple-system, BlinkMacSystemFont, ", font-size:16px, background-color:#FFFFFF, ">肌少症、自体动静脉内瘘、成熟不良、预测模型

Abstract: Objective: To establish a prediction model for the risk of arteriovenous fistula (AVF) maturation failure based on sarcopenia assessment indicators. Methodology: A total of 105 patients with end-stage kidney disease (ESKD) who underwent AVF surgery in Sir Run Run Hospital, Nanjing Medical University between February 2021 and January 2025 were enrolled as study subjects. Sarcopenia assessment indicators and clinical data were collected. The maturation status of AVF was evaluated at 8-12 weeks after surgery. The relationship between sarcopenia assessment indicators and poor AVF maturation was analyzed via regression analysis, followed by nomogram construction and efficacy evaluation. Results: After 12 weeks of follow-up post-AVF surgery, among the 105 patients, 32 experienced AVF maturation failure, resulting in an incidence rate of 30.48%. Logistic regression analysis revealed that reduced skeletal muscle index, reduced hand grip strength, slow gait speed, and low SPPB scores were associated with AVF maturation failure (all P<0.01). The validation of the nomogram constructed based on this shows that the total score of 0-220 corresponds to a risk range of 0.1-0.7, and the calibration curve is in good agreement with the ideal curve, with good model calibration and a C-index value of 0.85; the ROC curve analysis shows that the AUC of the nomogram for predicting poor internal fistula maturation is 0.852 (95%CI 0.778-0.925). Conclusion: The nomogram model constructed based on sarcopenia assessment indicators can effectively predict the risk of AVF maturation failure. The model exhibits good discrimination and calibration, providing a quantitative tool for the early clinical identification of high-risk patients.

Key words: font-family:Inter, -apple-system, BlinkMacSystemFont, ", font-size:16px, background-color:#FFFFFF, ">sarcopenia, arteriovenous fistula, maturation failure, prediction model