A Critical Reassessment of Inflammatory and Nutritional Indices (PNI, GNRI, SII, NLR, PLR) for Predicting Arteriovenous Fistula Maturation and Long-Term Patency: A 945-Patient Cohort

Background: While recent literature emphasizes the predictive value of composite inflammatory and nutritional indices for vascular outcomes, this study evaluates the actual predictive capacity of preoperative indices (PNI, GNRI, SII, NLR, PLR) for de novo arteriovenous fistula (AVF) maturation and 1-year primary patency. Methods: We retrospectively analyzed 945 end-stage renal disease patients who underwent strictly radio-cephalic autologous AVF creation. Preoperative indices were calculated from routine parameters. Diagnostic accuracy for predicting 1-year patency loss was assessed using receiver operating characteristic (ROC) curves, and a multivariate logistic regression model was constructed to adjust for baseline anatomical and clinical variables. Targeted subgroup analyses evaluated high-risk populations, including those with diabetes, coronary, and peripheral artery disease. Results: The 1-year primary and secondary patency rates were 73.3% and 93.1%, respectively. In contrast to prevalent reports, no significant differences in preoperative PNI, GNRI, NLR, PLR, or SII scores existed between patients with patent and thrombosed fistulas (p > 0.05). ROC analyses showed no predictive utility (AUC: 0.476–0.518). Crucially, multivariate logistic regression revealed that preoperative arterial (OR: 0.58, p < 0.001) and venous diameters (OR: 0.51, p < 0.001) were the strongest independent predictors of AVF failure, whereas all systemic biomarkers lacked independent predictive significance. Subgroup analyses confirmed these indices failed to predict AVF outcomes even in high-risk settings with severe endothelial dysfunction. Conclusions: Preoperative composite nutritional and inflammatory indices do not independently predict AVF maturation or long-term patency when adjusted for local anatomy. Local anatomical features and hemodynamics heavily dominate vascular outcomes, indicating that systemic biomarkers have limited standalone clinical utility for guiding preoperative vascular access planning.