Background and Aims: Several flow velocity thresholds have been proposed as diagnostic criteria for carotid stenosis >50% using color-coded duplex (CCD) in patients with contralateral occlusion, assuming that compensatory flow acceleration may overestimate the grade of stenosis. Our aim was to evaluate the performance of different criteria in our neurosonology lab. Methods: Retrospective observational study of patients attended at our neurosonology lab for CCD with confirmed contralateral carotid occlusion by angio-CT in the preceding 6 months. We compared classic criteria for carotid stenosis >50%: Peak Systolic Velocity (PSV) >125cm/s, End Diastolic Velocity (EDV) > 40cm/s and Systolic Index (SI) >2 with those proposed by Preiss et al.: PSV >250 cm/s, EDV >90 cm/s and SI >2.3. We built ROC curves to determine the best cut-off points (defined by highest Sensitivity + Specificity) in our sample. Measurements in angio-CT using NASCET criteria were used as the gold standard. Results: Of 86 patients with carotid occlusion, 25 (29%) had a contralateral carotid stenosis >50%. Classic criteria showed better performance (88-84-76% Sensitivity, 85-82-87% Specificity, according to PSV-EDV-SI) than those proposed than Preiss et al (60-32-64% Sensitivity, 100-100-92% Specificity). ROC curves set the best cut-off points at PSV>114cm/s, EDV >38cm/s and SI >1.46, with 96-88-88% Sensitivity and 82-82-84% Specificity. Conclusions: Classic criteria for defining carotid stenosis by CCD showed a good performance in patients with contralateral occlusion. Use of more specific criteria may result in under detection of significant stenosis. Even lower cut-offs seem to achieve the best performance in our cohort
No datasets are available for this submission.
No license information is available for this submission.