Introduction: The novel multi-parametric ‘DEGUM’ duplex ultrasonography (DUS) criteria may improve accuracy of extracranial carotid artery grading but have not been compared with the established Society of Radiologists in Ultrasound (SRU) consensus criteria yet. We sought to investigate the diagnostic agreement between these two widely used DUS approaches for grading of carotid artery steno-occlusive disease.
Methods: This was a sub-analysis of the complete dataset from a prospective multicenter study, where a total of 120 patients underwent both DUS and catheter angiography for validation of the multi-parametric DEGUM ultrasound approach. In this analysis, two ultrasonography experts independently interpreted documented B-image, color-flow and Doppler spectra findings utilizing DEGUM and SRU criteria. Carotid arteries were categorized into clinically relevant NASCET strata: moderate (50-69%), severe (70-99%) and occlusion. On angiography, ICA was graded using the NASCET approach. Inter-rater and between-methods reliability was assessed through weighted Cohen’s kappa. Overall accuracy for both ultrasound approaches was computed using catheter angiography as the gold standard.
Results: We analyzed a total of 162 carotid artery pairs on catheter angiography and DUS. There was almost perfect agreement between both ultrasonography diagnostic approaches in describing all clinically relevant grades of the disease (kw 0.91, CI95%: 0.89-0.94). Inter-rater reliability was found to be comparably high for both the DEGUM (kw 0.97, CI95%: 0.94-0.98) and the SRU (kw 0.98, CI95%: 0.97-0.99) ultrasonography approaches. Compared with catheter angiography, overall accuracy for moderate and severe stenosis ranges did not differ between the DEGUM (73% and 73%, respectively) and the SRU (73% and 70%, respectively) approaches. There was also no difference for detection of carotid artery occlusion (96% and 96%, respectively).
Conclusions: Our analysis showed almost perfect diagnostic agreement between the DEGUM and the SRU approaches for ultrasonography grading of carotid artery steno-occlusive disease. Therefore, our data do not support superiority of one approach over the other in the evaluation of clinically relevant grades of the disease.