Doppler Ultrasound-Derived Blood Flow Parameters Associated with Significant Arteriovenous Fistula Stenosis in a Large Vietnamese Hemodialysis Cohort.

PURPOSE: Arteriovenous fistula (AVF) stenosis is a major cause of vascular access dysfunction in hemodialysis patients. This study evaluated the associations between Doppler ultrasound-derived blood flow parameters measured at different AVF sites and Doppler-defined significant AVF stenosis.

METHODS: This retrospective cohort study included adult hemodialysis patients with native AVFs who underwent Doppler ultrasound evaluation at Cho Ray Hospital, Vietnam. Blood flow parameters were measured at the brachial artery, anastomosis, 5 cm from the anastomosis, and 10 cm from the anastomosis. Doppler-defined significant stenosis was defined as ≥50% luminal narrowing compared with the adjacent normal vessel segment. Univariable and multivariable logistic regression analyses were performed to evaluate associations between Doppler flow parameters and significant stenosis.

RESULTS: Among 1,541 patients, 585 (38.0%) had Doppler-defined significant AVF stenosis. Patients with significant stenosis had lower blood flow values at all evaluated sites, including Qmin and Qmean (all p<0.001). After adjustment for demographic and AVF characteristics, higher blood flow values remained associated with lower odds of significant stenosis. The adjusted odds ratios (95% confidence intervals) per 500 mL/min increase were 0.73 (0.67-0.79) for brachial artery flow, 0.76 (0.71-0.83) for anastomosis flow, 0.83 (0.78-0.89) for flow measured 5 cm from the anastomosis, 0.81 (0.75-0.87) for flow measured 10 cm from the anastomosis, 0.63 (0.56-0.71) for Qmin, and 0.76 (0.70-0.82) for Qmean. Significant stenosis was still observed in some patients with relatively high flow values.

CONCLUSION: Lower Doppler ultrasound-derived blood flow values at multiple AVF sites were associated with Doppler-defined significant AVF stenosis and may support comprehensive vascular access assessment. Further prospective studies using independent reference standards and clinically relevant outcomes are needed before these parameters can be used as validated predictive tools.