Critical appraisal of the Carotid Duplex Consensus criteria in the diagnosis of carotid artery stenosis

BackgroundClinicians have relied on published institutional experience for interpreting carotid duplex ultrasound studies (CDUS). This study will validate the ultrasound imaging consensus criteria published in 2003.MethodsThe CDUS and angiography results of 376 carotid arteries were analyzed. Receiver-operating characteristic (ROCs) curves were used to compare peak systolic velocities (PSVs), end-diastolic velocities (EDVs) of the internal carotid artery (ICA), and ICA/common carotid (CCA) ratios in detecting <50%, 50% to 69% (ICA PSV of 125-230 cm/s), and 70% to 99% (PSV of ≥230 cm/s) stenosis according to the consensus criteria.ResultsThe consensus criteria uses a PSV of 125 to 230 cm/s for detecting angiographic stenosis of 50% to 69%, which has a sensitivity of 93%, specificity of 68%, and overall accuracy of 85%. A PSV of ≥230 cm/s for ≥70% stenosis had a sensitivity of 99%, specificity of 86%, and overall accuracy of 95%. ROC curves showed that the ICA PSV was significantly better (area under the curve [AUC], 0.97) than EDV (AUC, 0.94) or ICA/CCA ratio (AUC, 0.84; P = .036) in detecting ≥70% stenosis and ≥50% stenosis. Pearson correlations showed a statistical difference between the correlation of PSV with angiography (0.833; 95% confidence interval [CI], 0.8-0.86), EDV with angiography (0.755; 95% CI, 0.71-0.80), and ICA/CCA systolic ratio with angiography (0.601; 95% CI, 0.53-0.66; P < .0001) in detecting 70% to 99% stenosis. Adding the EDV values or the ratios to the PSV values did not improve accuracy. The consensus criteria for diagnosing 50% to 69% stenosis can be significantly improved by using an ICA PSV of 140 to 230 cm/s, with a sensitivity of 94%, specificity of 92%, and overall accuracy of 92%.ConclusionsThe consensus criteria can be accurately used for diagnosing ≥70% stenosis; however, the accuracy can be improved for detecting 50% to 69% stenosis if the ICA PSV is changed to 140 to <230 cm/s.