Development and validation of an inflammatory-based nomogram for predicting arteriovenous fistula maturation failure in end-stage renal disease patients
摘要
Arteriovenous fistula (AVF) is regarded as the most effective vascular access for hemodialysis in patients with end-stage renal disease (ESRD).However, research investigating the relationships between the systemic inflammation response index (SIRI), the triglyceride-glucose index combined with body mass index (TyG-BMI), and the high-sensitivity C-reactive protein to albumin ratio (HRR) with AVF maturation is still limited.
MethodsWe included all ESRD patients undergoing first-time AVF creation between March and August 2024 at Haidian Hospital, Beijing (n = 249). To evaluate predictors of AVF maturation failure, we analyzed preoperative clinical/laboratory data and follow-up ultrasound assessments using restricted cubic spline models for dose-response relationships (SIRI, TyG-BMI, HRR). Multivariable Cox models identified independent risk factors, and a predictive nomogram was developed and validated through C-index and calibration plots.
ResultsThis study included 249 patients (73.5% AVF maturation rate). Restricted cubic spline analysis revealed linear associations between AVF maturation and SIRI (p-overall = 0.047), TyG-BMI (p-overall = 0.039), and HRR (p-overall = 0.026). Multivariable Cox regression identified three independent predictors: SIRI (HR = 1.69, 95%CI:1.45–1.96), TyG-BMI (HR = 2.68, 95%CI:1.62–4.41), and HRR (HR = 1.44, 95%CI:1.28–1.61) (all p < 0.001). The biomarker-based nomogram showed strong predictive accuracy, with calibration curves demonstrating excellent observed-versus-expected agreement.
ConclusionNon-maturation of AVF was independently associated with three inflammation-based biomarkers, namely SIRI, TyG-BMI, and HRR. High predictive accuracy was demonstrated by the nomogram incorporating these markers, supporting its use for the early identification of high-risk patients.