Are You Puzzled When Credentialing Entities Ask You for Criteria for External Carotid Artery Stenosis? A Study Correlating Peak Systolic Velocities and Degree of External Carotid Artery Stenosis Based on CT Angiography

Abstract

External carotid artery (ECA) disease has been understated in the current literature, likely secondary to the lack of clinical neurological symptoms when this vessel is stenotic. However, vascular ultrasound credentialing entities request the adoption of specific criteria for the assessment of the ECA. Our purpose is to determine the ultrasound criteria for the detection of hemodynamically significant stenosis (50%-99%) using computed tomography angiography (CTA) as the reference standard. We included patients who had a carotid ultrasound and a CTA of the neck performed both within a 12-month period. Ultrasound measurements were reviewed for technical accuracy. CTAs were reassessed to determine the degree of stenosis of the bilateral ECAs adopting the North American Symptomatic Carotid Endarterectomy Trial criteria. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were calculated for different peak systolic velocity (PSV) cutoff points to determine the most accurate PSV for detecting 50% to 99% stenosis of the ECA. Receiver operating curve analysis was used to assess diagnostic performance. Of the 536 ECAs included in this study, 13 were noted to be occluded by CTA. Six of these occlusions were accurately detected by ultrasound. There were 82 ECAs with 50% to 99% stenosis on CTA; these vessels had a significantly higher mean PSV compared to ECAs with <50% stenosis (294 vs 122 cm/s) ( P < .0001). A cutoff velocity of ⩾200 cm/s demonstrated a sensitivity, specificity, and overall accuracy of 0.84, 0.93, and 0.91, respectively. Using univariate analysis, smoking was significantly associated with the presence of ⩾50% ECA stenosis. We determined that a PSV cutoff of ⩾200 cm/s showed high sensitivity and specificity for detecting 50% to 99% ECA stenosis based on CTA.