Purpose
To compare a balanced steady-state free-precession sequence with a radial k-space trajectory and alternating repetition time fat-suppression (Radial-ATR) with other currently used fat-suppressed three-dimensional sequences for evaluating the articular cartilage of the knee joint at 3.0T.
Methods
Radial-ATR, FSE-Cube, GRASS, and SPGR sequences with similar voxel volumes and identical scan times were performed at 3.0T on both knee joints of 5 volunteers. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) measurements were performed for all sequences using a double acquisition method and compared using Mann-Whitney Wilcoxon tests. Radial-ATR sequences with 0.3 mm and 0.4 mm isotropic resolution were also performed on the knee joints of 7 volunteers and 3 patients with osteoarthritis.
Results
Average SNR values for cartilage, synovial fluid, and bone marrow were 54.7, 153.3, and 12.9 respectively for Radial-ATR, 30.8, 44.1, and 1.9 respectively for FSE-Cube, 13.3, 46.9, and 3.3 respectively for GRASS, and 19.1, 8.1, and 2.1 respectively for SPGR. Average CNR values between cartilage and synovial fluid and between cartilage and bone marrow were 98.6 and 41.8 respectively for VIPR-ATR, 13.4 and 28.8 respectively for FSE-Cube, 33.6 and 10.0 respectively for GRASS, and 11.0 and 16.9 respectively for SPGR. Radial-ATR had significantly higher (p<0.001) cartilage, synovial fluid, and bone marrow SNR and significantly higher (p<0.01) CNR between cartilage and synovial fluid and between cartilage and bone marrow than FSE-Cube, GRASS, and SPGR. Radial-ATR provided excellent visualization of articular cartilage at high isotropic resolution with no image degradation due to off-resonance banding artifacts.
Conclusion
Radial-ATR had superior SNR efficiency to other fat-suppressed three-dimensional cartilage imaging sequences and produced high isotropic resolution images of the knee joint which could be used for evaluating articular cartilage at 3.0T.