Serial Changes of Fatty Degeneration and Clinical Outcomes after Repair of Medium-Sized Rotator Cuff Tears

Abstract

Background: This study was designed to longitudinally analyze quantitative intramuscular and perimuscular fat and evaluate clinical outcomes according to healing degree after rotator cuff repair.

Methods: From June 2013 through October 2018, patients who had undergone repair due to medium-sized rotator cuff tears and serial chest computed tomography (CT) preoperatively and at early (6-12 months) and late (at least 3 years) postoperative follow-ups were included. Supraspinatus (SST) intramuscular fat fraction ratio (IFFR) and perimuscular fat fraction ratio (PFFR) were calculated using chest CT. The rotator cuff integrity was categorized as healed, smaller retear (SRT), and larger retear (LRT) by comparing the preoperative tear size and retear size in shoulder CT arthrography at postoperative follow-ups. Clinical outcomes were evaluated using the American Shoulder and Elbow Surgeons (ASES) score, the University of California at Los Angeles (UCLA) shoulder rating scale, and the Constant score preoperatively and at early and late postoperative follow-ups.

Results: In the LRT group, compared with the preoperative values, there were increases in the SST IFFR and PFFR at the early (p = 0.002 and p = 0.006, respectively) and late (p < 0.001 and p < 0.001, respectively) postoperative time points. Late postoperative clinical scores (UCLA and Constant scores) were not improved compared to preoperative scores (p = 0.156 and p = 0.094, respectively). In the SRT group, there was no difference in the mean SST IFFR and PFFR between preoperative and early postoperative time points (p = 0.766 and p = 0.180, respectively), but the late postoperative values were higher than preoperative values (p = 0.009 and p = 0.049, respectively). Late postoperative clinical scores (ASES, UCLA, and Constant scores) in the SRT group improved compared to preoperative time (p < 0.001, p < 0.001, and p = 0.016, respectively). In the healed group, compared with the preoperative values, there was no difference in the mean SST IFFR and PFFR at postoperative time points; however, the late postoperative clinical scores (ASES, UCLA, and Constant scores) were improved (all p < 0.001).

Conclusions: In the SRT group, IFFR and PFFR progressed in the late postoperative period and clinical scores improved over time. However, in the LRT group, IFFR and PFFR progressed in the early and late postoperative periods and clinical scores did not improve at the late postoperative follow-up.

Keywords: Arthroscopy; Computed tomography; Rotator cuff; Rotator cuff injury.

Fig. 1. Patient flowchart. CT: computed tomography, ASES: American Shoulder and Elbow Surgeons, UCLA: Shoulder Rating Scale of the University of California at Los Angeles.

Fig. 2. Reconstruction of a chest computed tomography image with reference to the scapular plane by an image reconstruction program (Radiant DICOM viewer 2020.1). (A) First, three landmarks (scapular trigonum, inferior angle, and center of glenoid) were located in the coronal, axial, and sagittal views. (B) The plane that passes through the three landmarks was made using a plane reconstruction tool.

Fig. 3. Finding the threshold of fat material that makes fat values the same in Digital Imaging and Communications in Medicine (DICOM) and JPG files. (A) In DICOM files, –190 to –30 Hounsfield units estimated in a previous study were entered for calculating fat material. (B) In JPG files, we found the threshold that makes the fat value the same. In this figure, a threshold of 70–90 was used to calculate an equivalent fat value of 693.899 mm³ in the DICOM file.

Fig. 4. Schematic explanation of the perimuscular and intramuscular fat areas. Perimuscular fat is out of muscle in the fossa. Intramuscular fat is inside muscle.

Fig. 5. A reconstructed parasagittal image of the scapular plane in a fat threshold according to a previously calculated value using ImageJ software.

Fig. 6. Measurement of intramuscular and perimuscular fat fraction ratios in the supraspinatus (SST). (A) To measure the intramuscular fat fraction ratio, the region of interest was drawn manually around the border of the SST muscle. (B) For the perimuscular fat fraction ratio, the region of interest was drawn manually around the border of the SST muscle fossa. After the border was set, the specified area and fat values in that area were automatically calculated.

Fig. 7. Radiologic evaluation of healing degree in shoulder computed tomography arthrography at 6 months postoperatively. (A) In oblique coronal and oblique sagittal images, there is no contrast leakage in the distal portion of the supraspinatus (SST) tendon (healing group). (B) There is a defect in the distal portion of the SST tendon filled with contrast material and opacification of the subacromial bursa in the oblique coronal view (red arrow) and oblique sagittal view (white arrow), which means that there is a full-thickness retear that is smaller than the initial tear (smaller retear group). (C) There is a defect at the distal portion of the SST tendon, and the defect is larger than the initial tear in the oblique coronal view (red arrow) and oblique sagittal view (white arrow) (larger retear group).