The reliability of duplex derived haemodynamic measurements in the assessment of femoro-distal grafts

Abstract

The aim of this study was to assess the reliability of flow and velocity measurements obtained in the postoperative period by duplex scan monitoring of femoro-distal bypass grafts. Three points along each of 24 grafts were studied by 2 independent observers using a Diasonics 300 DRF duplex scanner. Results obtained for diameter, peak systolic velocity (PSV) and flow were compared. The mean PSV for all points studied was 70 cm/s (S.D. = 14.5): comparing observers the calculated experimental error (EE) for all points was 11 cm/s (coefficient of EE = 15%). In contrast, the experimental error for flow was 44.5 ml/min (coefficient of EE = 31). The mean flow for all grafts was 142 ml/min (S.D. = 66.5). Flow at the different points of individual grafts could also be compared since routine postoperative intravenous angiography showed all grafts to be devoid of branches. The coefficient of experimental error for flow was 58%. Four of the grafts studied were found to have asymptomatic stenoses. These 4 grafts had low flows (mean = 42 ml/min) but so did 7 other non-stenotic grafts. Although PSV measurements entailed a 15% error, the mean change in PSV in relation to the stenoses was 350%. These data suggest that flow is an unreliable measurement. PSV is more reliable, changes exponentially with luminal diameter and can detect, and perhaps grade, asymptomatic stenoses.