Accuracy and reproducibility of duplex ultrasound imaging in a phantom model of femoral artery stenosis.

Purpose: The improvement of management strategies in patients with intermittent claudication might depend on a better understanding of the natural history of femoral atherosclerosis. The grading of stenoses, the monitoring of their progression, and the assessment of response to treatment are critically dependent on a method's accuracy and variability. Duplex ultrasound imaging provides a noninvasive way of measuring localized disease, but there has been relatively little objective evaluation of its accuracy and reproducibility. The aim of this study was to evaluate the accuracy and variability of duplex velocity ratio measurements of stenosis.

Methods: In a laboratory flow model of the femoral artery, 14 concentric and eccentric stenoses were examined five times by three sonographers. Measurements were then repeated with a standardized technique in which Doppler angle and aperture position were fixed, giving a total of 420 measurements.

Results: Velocity ratio showed good correlation with degree of stenosis, R2 = 0.996. Intraobserver variability was low, but interobserver variability was significant with more severe stenosis (p = 0.002, analysis of variance). Standardization of the technique did not improve accuracy or variability. The 95% confidence limit was +/- 20% for a single reading of velocity ratio for stenoses of > 50% diameter reduction.

Conclusions: We conclude that duplex ultrasound imaging can be used to accurately grade arterial stenosis in this range, and the potential exists for noninvasive monitoring of the progression of preocclusive femoral atherosclerosis and its response to treatment. In addition, repeated measurements of velocity ratio over time should be made by the same observer.