MR breast imaging: A comparative analysis of conventional and parallel imaging acquisition [RM delle mammelle: Confronto tra tecnica convenzionale ed imaging parallelo]

Purpose. The objective of this study was to compare conventional breast magnetic resonance imaging (MRI) with breast MRI acquired with the sensitivity-encoding (SENSE) technique on a 1.5-T MRI scanner in the same patient, on the basis of image quality and kinetics analysis. Materials and methods. Thirty-one patients with suspicious mammography and US findings were included in the study. Conventional breast MRI consisted of the following sequences: T1 (matrix, 288x512); T2 (matrix 225x512); short tau inversion recovery (STIR) (matrix 320x224) and dynamic T1 [2D fast-field echo (FFE)] (matrix 256x512; temporal resolution =80 s). The SENSE technique included the following sequences: T1 (matrix 512x512); T2 (matrix 512x512); short-tau inversion recovery (STIR) (matrix 320x224); dynamic T1 (3D FFE) (matrix 512x512, with a temporal resolution ≤70 s). Image quality was graded on a four-point scale, and the mean scores given to each sequence were compared between the two protocols. The relative enhancement rates and the qualitative features of the signal intensity (SI)/time curves were also compared between the two protocols. Results. The readers found 64 contrast-enhanced lesions in 31 patients. Nineteen patients had a total of 27 malignant lesions. In the remaining 12 patients, 37 benign lesions were found. No significant differences between the two protocols were observed with regard to the mean relative enhancement rates and the qualitative features of the SI/time curves. In detail, the mean image quality scores were higher for SENSE imaging (p<0.05). The mean image quality score for the T1 and T2 morphological sequences were comparable. In contrast, the quality scores for the STIR images differed significantly between the two protocols (p<0.001), and a significant difference was also observed when comparing the T1 postcontrast images (p<0.001). Conclusions. Our data suggest that the SENSE imaging protocol applied in our study is superior to conventional imaging with regard to image quality, especially for T1 postcontrast and STIR images. SENSE imaging protocols may provide an alternative to conventional sequences for contrast-enhanced MRI of the breast using 1.5-T MR scanners. © 2008 Springer-Verlag.