ABSTRACT
PURPOSE: Duplex examination is today the principal initial method of assessing extracranial carotid or vertebral artery disease. However, varying haemodynamic criteria have been described to categorize the degree of internal carotid artery stenosis, and similarly the degree of stenosis detected with angiography has been assessed with different methods as highlighted in studies performed by the North American Symptomatic Carotid Endarterectomy Trial and the European Carotid Surgery Trial. This study looks at the correlation between two commonly used methods of assessing carotid artery stenosis with duplex criteria and the two methods of angiographic interpretation used in these trials. Duplex parameters are also identified to determine the greater than 70% stenosis group identified as at risk in these studies. METHODS: A total of 120 carotid bifurcations were studied in patients who underwent both carotid duplex and angiography. Correlations of duplex with angiography were assessed with duplex criteria described by Zwiebel and by Strandness and the angiographic methods used in studies performed by the North American Symptomatic Carotid Endarterectomy Trial and the European Carotid Surgery Trial. Receiver operator curves were constructed from the duplex data for the detection of stenosis greater than 70% based on the angiographic assessment used in the study performed by the North American Symptomatic Carotid Endarterectomy Trial. RESULTS: The duplex criteria described by Zwiebel and Strandness differed in their accuracy depending on which of the two methods was used to report the angiograms. Zwiebel's criteria agreed more with the angiographic method used in the study performed by the European Carotid Surgery Trial (sensitivity 98%, specificity 81%, accuracy 88%), whereas Strandness' criteria agreed more with the angiographic method used in the study performed by the North American Symptomatic Carotid Endarterectomy Trial (sensitivity 96%, specificity 85%, accuracy 89%). For the detection of a stenosis greater than 70%, a peak systolic velocity greater than 270 cm/sec and end diastolic velocity greater than 110 cm/sec provided a sensitivity of 96%, specificity of 91%, and accuracy of 93%. CONCLUSIONS: The accuracy of duplex studies compared with angiography in the assessment of extracranial vascular disease depends on the method of angiographic determination of carotid stenosis. Vascular laboratories should validate the duplex criteria they use against a standard method of angiographic assessment of carotid artery stenosis, with special reference to the recently reported studies noting the significance of a stenosis greater than 70% in patients with symptoms.
