Factors affecting the continuous wave Doppler spectrum for the diagnosis of carotid arterial disease

Spectral analysis of continuous wave (CW) Doppler signals is used for the diagnosis of carotid arterial disease. Previous clinical and in vitro studies have documented that the peak Doppler frequency is increased in recordings made directly over a stenosis and that spectral broadening is observed beyond a stenosis in the region of disturbed flow. However, certain hemodynamic and technical factors can effect the Doppler spectrum and in particular cause spectral broadening although they are not related to the severity of the arterial stenosis. In this in vitro study, Doppler spectra were quantified by (1) measurements of the peak, mean and mode frequencies, and (2) measurements that quantify changes in the shape of the spectra and thus can potentially detect the presence of spectral broadening. The latter measurements included the spectral broadening index (SBI), coefficient of variation (CV), coefficient of skewedness (CS), and coefficient of kurtosis (CK). Using straight tubes without a bifurcation in a steady flow model, we found that the peak frequency and the extent of spectral broadening were dependent upon the severity of the stenosis, the relation of the recording site to the stenosis or bulb, and the flow rate. Comparison of the severity of Doppler spectral broadening from bulb and stenosis recordings allowed us to conclude that any observed changes in spectral broadening measurements are probably due to a significant stenosis and not to the presence of a normal bulb. If the tube is not completely insonated by the CW Doppler beam, an error of between 4 and 35% can be read in the spectral broadening measurements. The peak frequency, mean frequency, and SBI are not altered significantly by the automatic gain control or dynamic range and noise level settings usually chosen by the spectrum analyzer. Because of the variability of individual Doppler spectra, one should be cautious about deriving quantitative data from one individual spectrum. The results of quantitative analysis of the amplitude spectrum are different from the power spectrum. In conclusion, this in vitro study identified several hemodynamic and technical factors that affect the CW Doppler spectrum; however, in the clinical setting, their influence on quantitative measurements of the extent of spectral broadening likely can be minimized by a skilled technologist who uses a standardized technique.