Determination of the most appropriate velocity threshold for applying hemispheric flow convergence equations to calculate flow rate: selected according to the transorifice pressure gradient. Digital computer analysis of the Doppler color flow convergence region.

BACKGROUND

While flow convergence methods have been promising for calculating volume flows from color Doppler images, it appears that the velocity threshold used and the transorifice pressure gradient dramatically influence the accuracy of application of the simple hemispheric flow convergence equation for calculation of flow rate. The present in vitro study was performed to determine whether the value of velocity threshold at which the shape of proximal isovelocity surface best fits given shape assumptions with different orifice sizes and flow rates is predictable as a function independent of orifice size from clinically measurable peak velocity or transorifice pressure gradient information.

METHODS AND RESULTS

In an in vitro model built to facilitate ultrasound imaging, steady flow was driven through circular discrete orifices with diameters of 3.8, 5.5, and 10 mm. Flow rates ranged from 2.88 to 8.28 L/min with corresponding driving pressure gradients from 14 to 263 mm Hg. At each flow rate, Doppler color-encoded M-mode images through the center of the flow convergence region were obtained and transferred into the microcomputer (Macintosh IIci) in their original digital format. Then, the continuous wave Doppler traces of maximal velocity through the orifice were derived for the calculation of driving pressure gradient. Direct numerical spatial velocity measurements were obtained from the digital color encoded M-mode velocities with computer software. For each flow rate, we could calculate flow volume from any number of velocity distance combinations with a number of assumptions and use the results to assess expected flow convergence shape based on a priori knowledge of the progression from oblate hemispheroid to hemisphere to prolate hemispheroid changes observed previously. Our results showed that for a given ratio of calculated flow rate to actual flow rate (0.7 and 1), the velocity threshold that could be used for the calculation of flow rate with a hemispheric flow convergence equation correlated well with the pressure gradient for a given orifice size, and the differences in velocity threshold that could be used this way among different orifice sizes once they were adjusted for the covariate pressure gradients were not statistically significant (P = .79 for ratio = 0.7, and P = .81 for ratio = 1).

CONCLUSIONS

Our present study provides an orifice size-independent quantitative method that can be used to select the most suitable velocity threshold for applying a simple hemispheric flow convergence equation based on clinically predictable pressure gradients ranging from 40 to 200 mm Hg, and it offers a correction factor that can be applied to the hemispheric flow convergence equation when the pressure gradient is less than 40 mm Hg.