Gadolinium-Enhanced Liver Magnetic Resonance Imaging Using a 2-Point Dixon Fat-Water Separation Technique

PURPOSE: To compare image quality and lesion detection in postcontrast liver magnetic resonance imaging (MRI) between volumetric interpolated breath-hold examination (VIBE) sequences that achieve fat suppression via chemically selective fat saturation (FS-VIBE) and a 2-point Dixon water-fat separation method (Dixon-VIBE). MATERIALS AND METHODS: Thirty patients underwent contrast-enhanced liver MRI at 1.5 T in which Dixon-VIBE was performed immediately after a delayed FS-VIBE. Two radiologists in consensus reviewed the sequences for a variety of qualitative and quantitative image quality measures and for lesion detection. RESULTS: Dixon-VIBE received nearly perfect scores for strength and homogeneity of fat suppression that were significantly better than scores for FS-VIBE, with an associated significant improvement in liver-fat contrast (P < 0.0001 for all comparisons). Dixon-VIBE also received significantly better scores for sharpness of intrahepatic vessels (P = 0.0029) and overall image quality (P < 0.0001). Despite a slightly longer acquisition time for Dixon-VIBE, there was no significant difference in motion artifact (P = 0.3877). There was no significant difference for sensitivity, positive predictive value, or contrast relative to background liver for focal lesions (P = 0.448, P = 0.347, and P = 0.2312, respectively). CONCLUSIONS: For postcontrast liver MRI, Dixon-VIBE demonstrated significantly improved fat suppression. Various assessments of lesion detection showed no significant difference between sequences.