Development and validation of a prediction model for estimating patency rates after treatment of haemodialysis access dysfunction.

Abstract

Objective: This study aims to develop a risk assessment model for predicting haemodialysis access dysfunction and to construct a nomogram.

Method: The clinical data of patients with haemodialysis access dysfunction treated at our hospital from October 2020 to January 2024 were retrospectively analysed. The least absolute shrinkage and selection operator regression method was used to filter variables and select predictors, while Cox regression was applied to filter variables and construct a nomogram. The discriminatory ability of the model was determined by calculating the area under the curve (AUC). Calibration was evaluated using bootstrap internal validation and the Hosmer-Lemeshow test. The clinical utility and applicability of the model were assessed through decision curve analysis (DCA) and the clinical impact curve (CIC). Subgroup analysis of risk factors for haemodialysis access dysfunction was performed using Kaplan-Meier survival curves.

Result: The study included 423 patients, and seven variables were used to construct the risk prediction model and nomogram for haemodialysis access dysfunction. The C-index of the prediction model was 0.783, and the time-dependent AUC (>0.8) at 6, 12, 18 and 24 months post-surgery indicated strong discriminatory ability. The calibration curve and Hosmer-Lemeshow test demonstrated good agreement between the prediction of the nomogram and the observed values. The DCA and CIC curves further confirmed the clinical practicability of the model.

Conclusion: A risk assessment model and nomogram for haemodialysis access dysfunction based on seven variables were successfully constructed. This model demonstrates good discrimination and calibration, offering valuable guidance for clinical decision-making and significant clinical utility.