Thin‐beam ultrasound overestimation of blood flow: How wide is your beam?

In theory, mean blood velocity is overestimated by current ultrasound systems using thin‐beam ultrasound. However, if beam and vessel width are known, this overestimation may be correctable. To test this, we developed a string‐phantom based method to measure the effective beam width on a standard vascular ultrasound system (GE Vingmed System 5, Horton, Norway) and used a flow phantom to compare beam‐width estimates of mean velocity to half the peak velocity. We assumed that during steady‐state flow in the flow phantom, mean velocity should equal half the peak velocity. When insonating at a depth of 2 cm, effective beam width was 2.449±0.005 mm (mean±SE). From this, the calculated error correction factor was 0.786 (to correct for a predicted overestimation of 27.2%) for a flow phantom diameter of 6.5 mm based on a parabolic velocity model. We found that, across a range of velocities (8 to 60 cm/s), uncorrected mean velocity overestimated half the peak velocity by 31.7±0.7%, whereas corrected mean velocity overestimated by 3.5±0.6% (P<0.05 vs uncorrected). These findings suggest that while overestimation of mean velocity by thin‐beam ultrasound is a practical concern, it can be minimized if beam and vessel width are known.