Pulsed Doppler frequency and carotid stenosis.

Two hundred and forty-two internal carotid arteries (ICA) were evaluated by independently interpreted arteriography and pulsed Doppler spectrum analysis using ultrasonic arteriography to evaluate the ability of peak systolic frequency (PSF) to predict the degree of internal carotid stenosis. Mean PSF in the 129 (53.3%) high grade ICA stenoses of greater than 50% diameter reduction was 6.55 +/- 0.14 (SEM) khz, while mean PSF in the 113 (46.7%) low grade (less than 50% diameter reduction) stenoses was 3.38 +/- 0.12 (SEM) kHz (P less than 0.0001). Receiver-operator characteristic (ROC) analysis revealed that PSFs of 4.5 kHz (sensitivity 87%, specificity 88%) and 5.0 kHz (sensitivity 83%, specificity 93%) were best for identifying a 50% diameter stenosis. Positive predictive value of 5.0 kHz was 93% (107/115) and negative predictive value was 82.7% (105/127). Linear regression analysis of PSF in kHz versus percentage diameter reduction yielded the equation: % stenosis = 10.7 (PSF) - 4.1 (r = 0.76). A nonlinear equation was also derived: % stenosis = 61.9 - 33.5 (PSF) + 8.7 (PSF)2 - 0.5 (PSF)3 (r = 0.77). Based on this analysis peak systolic frequency criteria measured by pulsed Doppler spectrum analysis appear to be useful for distinguishing high grade from low grade stenoses. Both the linear and nonlinear equations further suggest that PSF can more precisely quantitate the degree of ICA luminal narrowing.

Correlation of hemodynamically significant internal carotid stenosis with pulsed Doppler frequency analysis.

Systolic and mean pressure gradients across internal carotid stenoses were measured at the time of carotid endarterectomy in the arteries of 90 patients, all of whom underwent angiography. Eighty-two of these patients also had pulsed Doppler ultrasonic arteriography with real-time spectrum analysis. There were 71 (79%) high grade stenoses of greater than 50% diameter reduction by angiography. Significant systolic pressure gradients (greater than or equal to 10 mmHg) were identified in 41 patients (46%), 38 (46%) of whom underwent ultrasonic evaluation. A pulsed Doppler frequency measured within the stenosis equal to or greater than 6.5 kiloHertz had a sensitivity of 94.7% (36/38) in identifying pressure reducing lesions with a specificity of 47.7% (21/44). Positive predictive value was 61% (36/59). Angiographic criteria (50% diameter reduction) exhibited a sensitivity of 97.6% (40/41), a specificity of 36.7% (18/49) and a positive predictive value of 56.3% (40/71). Negative predictive value was 94.7% for angiography and 91.3% for ultrasonic arteriography. A pulsed Doppler frequency equal to or greater than 6.5 kiloHertz appears to accurately identify lesions that are at risk to reduce distal internal carotid pressure under operative conditions with a sensitivity similar to angiography. This criterion has a positive predictive value and specificity that is slightly superior to angiography and a high negative predictive value. Pulsed Doppler spectrum analysis provides physiologic information relative to blood flow velocity that is complimentary to the anatomic data provided by angiography for assessing the potential for hemodynamic significance of internal carotid stenoses.