Extent of spontaneous cross-flow via the anterior communicating artery in steno-occlusive carotid artery disease

PURPOSE: Our purpose was to evaluate the agreement of transcranial color-coded duplex sonography (TCCS) measurements and intra-arterial digital subtraction angiography (DSA) findings in determining the extent of spontaneous cross-flow via the anterior communicating artery (AcoA) in patients with internal carotid artery (ICA) stenosis. METHODS: Thirty adult patients with suspected uni- or bilateral high-grade carotid artery stenosis were prospectively examined by DSA and angle-corrected TCCS. The extent of cross-flow was determined considering retrograde flow in the ipsilateral anterior cerebral artery (ACA) and sideto-side differences of the A1-segments of the ACA and middle cerebral arteries (MCAs) by both techniques. Cross-flow was angiographically categorized by means of a four-step scale. DSA findings were correlated with side-to-side differences in mean blood flow velocity as well as flow direction measured by TCCS. RESULTS: Twenty-seven of 30 patients had a uni- or bilateral ICA stenosis of >49%. Excellent agreement between TCCS and DSA was evaluated for the detection of lack (grades 0 and 1) or presence (grades 2 and 3) of reversed flow in the ACA (sensitivity 100%, specificity 93%, positive predictive value 94%). Post hoc analysis of the mean velocities in the ACA and MCA revealed a side-to-side difference of 25 cm/s as a cutting point allowing for definition of a corresponding four-grade scale for ultrasound. However, full agreement, i.e.same grade of cross-flow detected by both techniques, was only found in 17(57%) of 30 cases. CONCLUSIONS: Non-invasive TCCS is reliable for detecting reversed flow in the ACA in patients with ICA stenosis. However, there is only a moderate agreement between angiography and TCCS in quantifying the extent of spontaneous anterior cross-flow because different information on the intracranial hemodynamics may be obtained.