Validating common carotid stenosis by duplex ultrasound with carotid angiogram or computed tomography scan

Background No consensus exists for duplex ultrasound criteria in the diagnosis of significant common carotid artery (CCA) stenosis. In general, peak systolic velocity (PSV) >150 cm/s with poststenotic turbulence indicates a stenosis >50%. The purpose of our study is to correlate CCA duplex velocities with angiographic findings of significant stenosis >60%. Methods We reviewed the carotid duplex records from 2008 to 2011 looking for patients with isolated CCA stenosis and no ipsilateral internal or contralateral carotid artery disease who received either a carotid angiogram or a computed tomography scan. We identified 25 patients who had significant CCA disease >60%. We also selected 74 controls without known CCA stenosis. We performed receiver operating characteristics analysis to correlate PSV and end-diastolic velocity (EDV) with angiographic stenosis >60%. The degree of stenosis was determined by measuring the luminal stenosis in comparison to the proximal normal CCA diameter. Results Most patients had a carotid angiogram (21/25), four only had a computed tomography angiography and four had both. Eighteen patients had history of neck radiation. The CCA PSV ≥250 cm/s had a sensitivity of 98.7% (81.5%-100%) and a specificity of 95.7% (92.0%-99.9%), CCA PSV ≥300 cm/s had a sensitivity of 90.9% (69.4%-98.4%) and a specificity of 98.7% (92.0%-99.9%). The CCA EDV ≥40 cm/s had a sensitivity of 95.5% (95% confidence interval of 75.1-99.8%) and specificity of 98.7% (92.0%-99.9%), EDV ≥60 cm/s had a sensitivity of 100% (75.1%-99.8%) and specificity of 87% (94.1-100%), and EDV ≥70 cm/s had a sensitivity of 86.4% (64.0%-96.4%) and specificity of 100% (94.1%-100%). The presence of both PSV