Accuracy of maximum velocity estimates made using Doppler ultrasound systems.

Abstract

This study was performed in order to provide quantitative data on the estimation of maximum velocity made using modern Doppler ultrasound systems. This is important since the degree of stenosis within arteries is commonly assessed from the maximum velocity. A string phantom was used as the source of Doppler signals. This enables direct comparison between the Doppler estimated maximum velocity and the true filament velocity. Six modern commercial Doppler systems were used. Measurements were made under standard conditions for each probe. In addition a number of factors were varied in turn (beam-filament angle, filament depth, filament velocity and Doppler aperture position). Under standard conditions the maximum velocity was overestimated in all cases (0-29% error). For all measurements maximum velocity errors ranged from -4% to 47%. There was a large intraprobe variation in maximum velocity estimation (mean variation of 25%), and a large interprobe variation (mean variation of 25%), and a large interprobe variation (mean variation of 18%). These results indicate that, at present, errors in maximum velocity estimation may be directly translated into significant errors in the estimate of the degree of arterial stenosis made from velocity measurements. As a consequence, some patients may be incorrectly categorized. Consideration should be given to applying angle dependent correction factors to maximum velocity measurements, and to the use of conversion from Doppler frequency shift to velocity using the angle derived from the edge of the Doppler aperture.