Objectives: To evaluate the role of the end-diastolic velocity (EDV) in the common carotid artery (CCA) as a marker of internal carotid artery (ICA) occlusion.
Design: Validation of retrospective data in a prospective clinical study.
Methods: The EDV in 94 patients with total ICA occlusion and in 24 patients with high grade (95-99%) unilateral ICA stenosis identified on extracranial carotid colour-flow Duplex imaging (CFDI) and arteriography was reviewed, and was retrospectively compared to the EDV of 176 normal individuals. Identification of patients with ICA occlusion was most accurate (99.3%) with an ipsilateral EDV > or = 12 cm/s and a DIFF > or = 10 cm/s (DIFF = contralateral EDV -- ipsilateral EDV). These values were then prospectively applied to all 886 patients (67 with high grade stenosis or occlusion) who underwent CFDI at our institution during 1994.
Results: The EDV > or = 12 had a 92% sensitivity, a 99.4% negative predictive value (NPV) and a 85% specificity in distinguishing between occluded and patent ICA's. In combination with a DIFF > or = 10 was 80.4% sensitive and 97.5% specific. The positive predictive value of the EDV > or = 12 in the distinction between 95-99% ICA stenosis and ICA occlusion was 78.3%, and that of the combination was 85.4%. The EDV was rarely zero and 10% of patients with normal or minimally diseased ICA's had an EDV > or = 12 and/or a DIFF > or = 10.
Conclusions: The EDV < or = 12 cm/s is a sensitive marker of ICA occlusion with a high NPV and in combination with the DIFF > or = 10 cm/s, is specific. Nevertheless, EDV parameters are inaccurate in the distinction of 95-99% ICA stenosis from occlusion. Low EDV can be found in a number of patients with minor or no ICA disease, particularly in those with a stroke or silent cerebral infarct.