Clinical and Mechanistic Research on Autogenous Arteriovenous Fistula in Hemodialysis Patients

This observational study aims to investigate the long-term clinical outcomes and molecular mechanisms underlying autogenous arteriovenous fistula (AVF) maturation failure in uremic patients, with the primary goal of developing a precision prediction model integrating clinical, imaging, and biomarker data while also identifying key molecular targets regulating AVF maturation. Conducted as a single-center prospective study at Shanghai Tenth People's Hospital, the research will enroll 300 ESRD patients undergoing first-time AVF creation, collecting comprehensive multi-omics data including clinical parameters (demographics, comorbidities, laboratory tests), ultrasound imaging (vessel diameter, blood flow measurements), and molecular biomarkers (miRNAs, cytokines, vascular remodeling proteins). Using advanced machine learning techniques like Random Forest and XGBoost, the study will construct a predictive model for AVF maturation failure based on KDOQI criteria (blood flow <600 mL/min, diameter <6 mm, or depth >6 mm at 6 weeks), while parallel mechanistic investigations will employ WGCNA and Cox regression analyses to elucidate critical molecular pathways such as VEGF/TGF-β signaling and potential therapeutic targets including ALPL and Eph-B4. The study adheres to strict ethical guidelines (Helsinki Declaration, GCP standards) and includes rigorous statistical planning with a sample size calculated to ensure adequate power (300 patients yielding 110 expected events), employing AUC analysis for model discrimination and standard statistical tests for group comparisons. Expected outcomes include the development of a clinically applicable prediction tool (targeting software copyright), 2-3 high-impact SCI publications, and identification of novel therapeutic targets, with the entire project spanning from patient recruitment (2025.07-2026.07) through data analysis and manuscript preparation (2026.11-2027.01), ultimately aiming to significantly improve AVF management through its innovative combination of multi-omics integration and machine learning approaches.