A Reappraisal of CT Angiography Derived Duplex Ultrasound Velocity Criteria With a Comparison to Digital Subtraction Angiography in Patients With Carotid Artery Stenosis

Background: Traditionally, carotid duplex ultrasound (CDUS) velocity criteria have been derived from angiography. Recent studies support a shift toward computed tomography angiography (CTA) derived velocity criteria; however, they lack a comparison to angiography. The purposes of this study are to validate CTA derived measurements with digital subtraction angiography (DSA) and to update our previous CTA-derived velocity criteria for 50% and 80% stenosis. Methods: All patients between 2010 and 2019 who underwent CDUS and a neck CTA within 6 months were identified for a retrospective review. Vessel diameter and corresponding CDUS data were recorded. Additional DSA measurements were recorded for a subset of patients. Data from this cohort were added to a previously reported deidentified data set from patients between 2000 and 2009. Receiver operating characteristic (ROC) curves were generated to determine optimal velocity thresholds. Spearman rank correlation was used to correlate measurements obtained by CTA to those obtained by DSA. Results: A total of 1139 vessels from 636 patients were analyzed. ROC analysis to identify >= 50% stenosis resulted in optimized thresholds of 143 cm/sec, 46.2 cm/sec, and 2.15 for peak systolic velocity (PSV), end-diastolic velocity (EDV), and PSV to common carotid artery PSV ratio (PSVR), respectively. ROC analysis to identify >= 80% stenosis resulted in optimized thresholds of 319 cm/sec, 87.2 cm/sec, and 3.49 for PSV, EDV, and PSVR, respectively. The degree of carotid artery stenosis for a subset of 124 vessels on CTA correlated well with that of DSA (rho = 0.89, P < 0.0001). Conclusions: These data demonstrate a high correlation between measurements obtained on CTA and DSA while forming reliable CTA-derived CDUS velocity criteria.