To compare the diagnostic performance of three-dimensional (3D) intermediate-weighted FSE (IW-3D) and 3D hybrid T1-weighted sequences (Hy-3D) and 2D fast-spin-echo sequences (FSE) in diagnosing chondral and labral lesions at 1.5 Tesla hip MR arthrography (MRA).
Materials and methods
Institutional review board approval was obtained and informed consent was waived. Ninety-two consecutive patients were evaluated. Chondral and labral lesions were retrospectively and independently evaluated by two radiologists. Intra-operative findings were used as the reference standard (arthroscopy = 73, open surgery = 19). Sensitivity (Se), specificity (Sp), and accuracy (Acc) values that obtained were compared using McNemar test. A value of p < 0.05 was considered statistically significant. Inter-observer agreement was calculated using kappa statistics.
Surgeons revealed 81 labrum and 44 chondral lesions, respectively. The highest Se, Sp, and Acc for Reader 1 were 96.3, 90.9, and 95.6%, respectively, in evaluating labral lesions (by reading 2D data set) and 90.9, 100, and 95.7% in evaluating chondral lesions (by reading IW-3D images). The highest Se, Sp, and Acc for Reader 2 were 93.8, 81.8, and 92.4% in evaluating labral lesions (using 2D images) and 88.6, 97.9, and 93.5%, respectively, in evaluating chondral lesions (using Hy-3D). The difference of diagnostic accuracy achieved was not significant (p > 0.05). A near-perfect inter-observer agreement was achieved by reading 2D data set (k = 0.88) and Hy-3D (k = 0.83) and IW-3D (k = 0.85).
At 1.5 Tesla hip MRA, the accuracy of IW-3D and Hy-3D images was not significantly higher than the 2D sequences in evaluating acetabular labrum and chondral lesions.