Reproducibility of duplex ultrasound in the measurement of venous reflux

Objective: To determine the reproducibility of duplex ultrasound in the measurement of reflux duration in lower limb veins. Design: Repeatability study. Setting: Vascular clinic in a tertiary referral centre. Participants: Twenty-one patients with severe venous disease. Interventions: Patients were scanned using duplex ultrasound by two of three observers, then rescanned by a different pair of observers after a mean interval of 51 days. Main outcome measures: Duration of venous reflux. Results: On a Wilcoxon signed rank test, observers 2 and 3 showed no significant interobserver variability. The other pairs of observers agreed at the majority of segments, but differed at the popliteal vein (p less than or equal to 0.001), and superficial femoral and common femoral veins (p less than or equal to 0.05). Observer 2 showed no significant intraobserver variability, but observer 1 differed at the common femoral and superficial femoral veins (p less than or equal to 0.05), and observer 3 differed at the short saphenous vein (p less than or equal to 0.05). Conclusions: Reproducibility was reasonable at certain sites, but appeared to be influenced by position of the vein.