In vitro comparison of alternative methods for quantifying the severity of Doppler spectral broadening for the diagnosis of carotid arterial occlusive disease.

Quantitative analysis of continuous wave Doppler recordings is of clinical value in the noninvasive diagnosis of carotid arterial disease. Peak frequency measurements are useful and accurately detect severe stenoses but do not reliably diagnose minor or moderate stenoses because the measurement is dependent upon the probe to vessel angle, which cannot be measured accurately. Recent investigations have focused on efforts to overcome this limitation by quantifying the degree of spectral broadening that occurs as the result of flow disturbances downstream from a stenosis. In this study, an in vitro model was used to determine the optimum method for quantifying the instantaneous Doppler spectrum. The model generates blood flow velocity waveforms that are virtually identical to those found in the human internal carotid artery. Doppler recordings were made from normal tubes and distal to stenoses (39-87% cross-sectional area reduction). The spectra were quantified by the following angle-independent measurements: spectral broadening index and three standard statistical shape descriptors, namely the coefficients of variation, skewedness and kurtosis. Using this model, the results demonstrate an excellent relationship between the severity of the stenosis and each of spectral broadening index (r = 0.99), coefficient of variation (r = 0.96), and coefficient of skewedness (r = 0.99). The calculation of each of the measurements can be implemented quite easily, and a prospective trial is warranted to evaluate their clinical diagnostic accuracy.