Common carotid artery peak systolic velocity ratio predicts high-grade common carotid stenosis

Abstract

Screening for common carotid artery (CCA) stenosis with duplex ultrasound (DUS) velocity criteria alone can be limited by within-patient and between-patients hemodynamic variability. This study aimed to evaluate inter-CCA velocity ratio criteria to predict high-grade CCA stenosis. This was a retrospective review of consecutive patients who underwent computed tomography angiography and DUS peak systolic velocity (PSV) measurements of bilateral CCAs, independently recorded, between 2008 and 2014. Patients with dampened CCA waveforms on DUS composed group B. The remainder without dampened waveforms constituted group A. Inter-CCA PSV ratios were calculated by dividing the higher CCA PSV by the lower one of the other side, so the ratios would always be ≥1. Ratios were subsequently paired with each respective unilateral CCA diameter stenosis and differential bilateral CCA diameter stenosis. A quadratic regression model was fitted to predict unilateral and differential stenosis. Receiver operating characteristic curve was used to determine optimal ratios for ≥50% and ≥80% CCA stenosis. The study excluded patients with carotid artery occlusion. From a total of 201 patients, 193 patients were included in group A and 8 in group B. Within group A, 31 patients had ≥50% unilateral stenosis and 17 had ≥50% differential stenosis. All stenoses ≥50% were identified on the same side with the higher PSV. Inter-CCA PSV ratio predicted ≥50% unilateral (r(2) = 0.536; P < .001) and differential stenosis (r(2) = 0.581; P < .001). In group B, all patients had ≥60% stenosis that was near or involved the vessel origin. An increasing inter-CCA PSV ratio showed a trend toward contralateral high-grade stenosis (r(2) = 0.596; P = .1). Receiver operating characteristic curves showed an optimal threshold CCA ratio ≥2.16 for ≥50% unilateral stenosis with 92% accuracy, 62% sensitivity, and 98% specificity (area under curve = 0.854; 95% confidence interval, 0.759-0.948) and a ratio ≥2.62 for ≥50% differential stenosis with 97% accuracy, 83% sensitivity, and 98% specificity (area under curve = 0.94; 95% confidence interval, 0.835-1). DUS-based CCA PSV ratio can accurately predict unilateral and differential high-grade CCA stenosis. Also, in patients with unilateral dampened waveforms, it implied contralateral severe proximal stenosis. This parameter should be further validated in prospective studies and may serve as an adjunct screening tool to detect high-grade CCA stenosis.