Background: The assessment of fatty degeneration of rotator cuff muscles with a reliable grading system is very important. However, there is no generally accepted tool to measure it quantitatively.
Purpose: This work was undertaken to introduce a new method to evaluate fatty degeneration by calculating the occupation ratio of the supraspinatus, to determine the correlation between this occupation ratio and the degree of Goutallier's fatty degeneration or tear size of rotator cuff muscles and to assess the correlation between multidetector computed tomography (MDCT) and magnetic resonance imaging (MRI) using this new measuring tool.
Study design: Cohort study (diagnosis); Level of evidence, 2.
Methods: Sixty-one patients with full-thickness rotator cuff tears who had undergone arthroscopic rotator cuff repair were included. The occupation ratio was measured using the magic selection tool in Photoshop on MDCT and both T1- and T2-weighted oblique sagittal magnetic resonance images of Y-view. One-way analysis of variance was used to determine the association between this occupation ratio and the degree of fatty degeneration as assessed by Goutallier's method. The correlation between the occupation ratio and tear size, and the correlation of the occupation ratio between MDCT and MRI, were evaluated using the Pearson correlation coefficient. In addition, the interobserver and intraobserver reliabilities of this measuring tool were assessed on each image by 2 orthopaedic surgeons and analyzed using interclass correlation coefficients.
Results: Comparison of the occupation ratio measured by this new method with the Goutallier grade of fatty degeneration on MDCT and MRI revealed a highly significant correlation, with all P values < .001. Also, comparison of this new occupation ratio and cuff tear size showed a reverse correlation on MDCT, T1-weighted magnetic resonance images, and T2-weighted images, with Pearson correlation coefficients of -.61, -.56, and -.50 (all P < .001) by rater 1, and -.57, -.53, and -.43 (all P < .001) by rater 2. Comparison of the occupation ratio between MDCT and MRI demonstrated significant correlation, with Pearson correlation coefficients of .78 and .87 between MDCT and T1-weighted magnetic resonance images and .73 and .82 between MDCT and T2-weighted images. The interobserver and intraobserver reliabilities of this new measuring tool on MDCT and both T1- and T2-weighted magnetic resonance images were excellent in all images, with interobserver correlation of .89, .92, and .91 and intraobserver correlation of .89 and .96, .90 and .98, and .85 and .97, respectively.
Conclusion: This new quantitative measuring tool to evaluate the fatty degeneration of rotator cuff muscles was reliable and correlated well with both the preexisting grade of fatty degeneration and cuff tear size. Also, there was a significant correlation between oblique sagittal images of MDCT and MRI using this method; therefore, MDCT can be another option for imaging modality, comparable with MRI, for the evaluation of fatty degeneration of rotator cuff muscles.