Carotid arterial ultrasound scan imaging: A direct approach to stenosis measurement.

PURPOSE Management decisions regarding carotid artery disease are critically dependent on stenosis but have been made difficult because of conflicting methods used to determine such stenosis. The increasing use of duplex ultrasound scanning has conventionally depended on Doppler velocity measurement, an indirect method for calculating carotid stenosis. Recent technical advances have improved the quality of B-mode/color-flow ultrasound scan imaging (USI). We tested prospectively whether USI was clinically effective as the primary criterion for estimating carotid stenosis. METHODS Transverse and longitudinal USI, Doppler velocity, and arteriography data were obtained sequentially and independently for 713 carotid bifurcations. The internal carotid artery (ICA) residual lumen, the local outer diameter at the stenotic site, and the diameter distal to the bulb were measured in a representative USI longitudinal section. The peak systolic velocity and the end diastolic velocity (EDV) were measured at the stenosis. Local stenosis as determined with USI was compared with the x-ray arteriographic clinical radiology interpretation (XRI). As the primary method, radiologists compared the residual lumen with the distal ICA diameter, as recommended by the North American Symptomatic Carotid Endarterectomy Trial and the Asymptomatic Carotid Atherosclerosis Study. Analysis was by means of the USI positive predictive value (PPV) and negative predictive value (NPV) of the XRI findings, with the assumption that 80%, 70%, and 60% local stenosis with USI related to 70%, 60%, and 50% stenosis with XRI, respectively. RESULTS All 56 ICA occlusions as determined with USI were confirmed with XRI. When the USI showed 80% to 99% stenosis, the PPV of the XRI showing 70% to 99% stenosis was 94% (116/123). Two ICAs that were shown to be severely diseased with USI appeared to be occluded with XRI. For <50% stenosis shown with USI, the prediction of <50% stenosis shown with XRI was 94% (253/269). For borderline stenosis in the 50% to 79% range with USI, the addition of velocity criteria to USI data improved both the PPV and the NPV. In the range of 70% to 79% stenosis with USI, the PPV improved from 82% (76/93) to 91% (53/58) for the subgroup with an EDV of more than 80 cm/s. For the range of 60% to 69% stenosis with USI, the PPV improved from 75% (71/95) to 95% (21/22) for the subgroup with an EDV of more than 80 cm/s. In the range of 50% to 59% stenosis with USI, the NPV improved from 69% (53/77) to 93% (14/15) for the subset with a peak systolic velocity of less than 100 cm/s. CONCLUSION On the basis of the USI data alone, a prediction of arteriographic findings was possible at the 95% level for occlusion and severe stenosis and for ruling out hemodynamically significant stenosis. The addition of velocity data improved prediction in borderline degrees of stenosis. USI was effective for quantifying clinically significant degrees of stenosis.