MR angiography of internal carotid arteries: breath-hold Gd-enhanced 3D fast imaging with steady-state precession versus unenhanced 2D and 3D time-of-flight techniques

PURPOSE: The purpose of this work was to compare Gd-enhanced breath-hold fast imaging with steady-state precession (Gd-FISP) with unenhanced time-of-flight (TOF) sequences in evaluating internal carotid arteries (ICAs). METHOD: Thirty patients underwent three unenhanced TOF sequences [2D traveling saturation (Travelsat); 3D tilted optimized nonsaturated excitation (TONE); TOF 3D Multislab] and two breath-hold 3D Gd-FISP sequences with automated intravenous contrast agent injection (axial and coronal). ICAs were classified as normal (no stenosis); with mild (<30%), moderate (30-70%), or severe stenosis; or occluded (100%). Digital subtraction angiography (DSA) with aortic arch injection was used as a reference technique. RESULTS: DSA revealed 20 normal ICAs; 11 mild, 9 moderate, and 14 severe stenoses; and 2 occlusions. DSA and all MR angiography (MRA) sequences diagnosed the occlusion of four common carotid arteries. The TOF 2D overestimated 10 stenoses, TOF 3D TONE 9, and TOF 3D Multislab 5; Gd-FISP 3D overestimated only 2 of them, reaching the highest sensitivity and specificity for severe stenoses. Significant differences were found between the overestimation of Gd-FISP and each of the three unenhanced sequences (0.0020 < p < 0.0313, Wilcoxon and McNemar tests). Severe artifacts were observed with TOF techniques only. CONCLUSION: Gd-FISP is an interesting, largely artifact-free improvement for MRA of ICAs.