Objective: To validate carotid duplex sonography (CDS) in diagnosis of intracranial dural arteriovenous fistulas (AVFs) against the standard of cerebral catheter angiography.
Methods: We investigated 35 patients with dural AVFs and 64 patients without dural AVFs confirmed by the catheter angiographic studies. Twenty CDS parameters in 4 categories, including resistive index (RI), flow volume, peak systolic velocity, and end-diastolic velocity, were evaluated. Abnormal CDS findings were defined as the data above 95th percentile or below 5th percentile values from 180 control subjects. We determined the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy in each CDS parameter.
Results: The parameter of RI of the external carotid artery (ECA; cutoff points: right, 0.72; left, 0.71) yielded the highest sensitivity (74%), specificity (89%), positive predictive value (79%), negative predictive value (86%), and accuracy (84%) for predicting dural AVFs. All other ECA-related parameters yielded sensitivity lower than 70%, and those related to the internal carotid artery were lower than 30%. The sensitivity values for the parameter of RI of the ECA in different locations of dural AVFs were 54% (7 of 13 patients) in cavernous sinus dural AVFs and 86% (19 of 22 patients) in non-cavernous sinus dural AVFs (P = .05).
Conclusions: The RI of the ECA is the best CDS parameter for predicting intracranial dural AVFs. Carotid duplex sonography can be used as the initial screening tool for diagnosis in patients having symptoms related to dural AVFs.