Color Duplex Ultrasound Profiling of Native Hemodialysis Arteriovenous Fistula.
Prashant Sarda, Rohit Sharma, Sovinder Baisoya, Vivek Ruhela, Rajendra Srivastava, Sandeep Kumar, Kiran Kumar Shetty
Abstract
Open AccessBackground Arteriovenous fistulas (AVFs) remain the preferred vascular access for hemodialysis patients due to their superior long-term patency and lower complication rates. However, AVF dysfunction due to stenosis and thrombosis remains a significant clinical challenge. Color duplex ultrasound (CDUS) has emerged as a valuable, non-invasive tool for AVF assessment, but its accuracy compared to digital subtraction angiography (DSA) remains debated. This study aims to evaluate the diagnostic reliability of CDUS in detecting AVF dysfunction and compare its findings with DSA. Methods A prospective observational study was conducted over 18 months at Shri Guru Ram Rai Institute of Medical & Health Sciences, Dehradun, including 57 hemodialysis patients undergoing AVF evaluation. CDUS parameters, including peak systolic velocity (PSV), flow volume (FV), and draining vein diameter, were assessed and compared with DSA findings. Diagnostic performance, inter-method agreement, and key hemodynamic determinants of AVF function were analyzed. Results CDUS demonstrated high sensitivity (93.35%) and specificity (92.86%) in detecting stenosis, with 100% agreement with DSA in identifying stenosis location and severity. Functional AVFs had larger draining vein diameters (6±1.1 mm vs. 2.9±0.7 mm, p=0.001) and lower FVs, emphasizing the importance of hemodynamic balance in AVF maturation. Stenotic length significantly impacted FV, with longer stenotic segments (>6 cm) associated with reduced flow (p=0.023). Conclusion CDUS is a highly accurate and non-invasive modality for AVF surveillance, demonstrating strong agreement with DSA. Given the absence of standardized Doppler criteria, further research should focus on refining hemodynamic thresholds and integrating AI-assisted ultrasound analysis for improved diagnostic precision.