Multidirectional depiction of internal carotid arterial stenosis: Three-dimensional time-of-might MR angiography versus rotational and conventional digital subtraction angiography

PURPOSE: To evaluate whether and to what extent greater number of projectio n images obtained at three-dimensional (3D) time-of-fright (TOF) magnetic r esonance (MR) angiography versus conventional digital subtraction angiograp hy (DSA) causes overestimation of internal carotid arterial (ICA) stenosis. MATERIALS AND METHODS: DSA (two or three projections), rotational angiograp hy (16 or 32 projections), and 3D TOF MR angiography (12 projections) were performed in 47 stenotic ICAs of 38 symptomatic patients. Two observers ind ependently measured maximum stenosis, and the mean differences among MR ang iography, DSA, and rotational angiography were compared. RESULTS: Three rotational and five MR angiograms were nondiagnostic. Seven MR aniograms of ICA stenoses showed a signal void and were excluded from an alysis. On the remaining 32 angiograms, mean differences in maximum stenosi s for observers 1 and 2, respectively, were 7% (95% Cl: 3%, 12%) and 8% (95 % Cl: 3%, 13%) at MR angiography versus DSA and 2% (95% Cl: -2%, 7%) and -1 % (95% Cl: -5%, 3%) at MR angiography versus rotational angiography. ICA st enosis was graded significantly higher at MR angiography versus DSA, wherea s, it was not, overestimated at MR angiography versus rotational angiograph y. The difference in in maximum stenosis at MR angiography versus DSA was s ignificantly different from that of MR angiography versus rotational angiog raphy. CONCLUSION: Apparent overestimation of ICA stenosis at 3D TOF MR angiograph y versus conventional DSA may be partly explained by the greater number of projection images available at 3D TOF MR angiography.