RESUMO
PURPOSE: To assess the use of 3-dimensional (3D) MR imaging of the ankle with the 3D-turbo-spin-echo-sequence 3D-"Sampling Perfection with Application optimized Contrast using different flip angle Evolutions" (SPACE), as compared with 2-dimensional-turbo-spin-echo-sequence. MATERIAL AND METHODS: After internal review board's approval and informed consent, 15 healthy volunteers and 45 consecutive patients were examined at 3 T with isotropic fat-saturated moderately T2-weighted 3D-SPACE (voxel size: 0.6(3) mm(3)/acquisition time: 6:43 minutes) featuring radial k-space reordering for optimized contrast. Signal- and contrast-to-noise ratios (SNR; CNR, respectively) were calculated with the subtraction method. Using free 3D reconstructions, 2 radiologists independently assessed depiction of cartilage, ligaments, and tendons, as well as detection and grading of abnormalities of these structures (5-point Likert scale) compared with conventional 2-dimensional-TSE-sequences (voxel size: 0.4 × 0.4 × 3 mm(3)/total acquisition time: 11 minutes). Statistical analysis was performed with Wilcoxon signed rank tests, 95% and 99% confidence intervals and weighted κ coefficients. RESULTS: SNR and CNR of fluid/cartilage were significantly higher for 3D-SPACE (P < 0.05). The isotropic voxel size facilitated improved depiction of the medial and lateral ankle ligaments with significant differences for the calcaneofibular ligament and the anteromedial ligament complex (P < 0.05). In the patient cohort, cartilage and spring ligaments were also significantly better depicted (P < 0.05). However, there were no significant differences in the number or in the diagnostic confidence of detected cartilage, ligament, or tendon abnormalities. Interreader correlation was good (κ = 0.69-0.71) for both sequences. The correlation between the 2 sequences was excellent (κ = 0.84-0.85). CONCLUSION: 3D-SPACE allows 3D acquisition and assessment of the ankle and facilitates depiction of the complex ankle anatomy at sufficient SNR and CNR.