Abstract
OBJECTIVE: Because positron emission tomography/computed tomography (PET/CT) using fluorine-18-fluorodeoxyglucose ((18)F-FDG) can be used to visualize inflammation of the musculoskeletal system, it may help elucidate the pathophysiology of frozen shoulder (FS). The purpose of this study was to characterize the uptake pattern on (18)F-FDG PET/CT in patients with idiopathic FS and to determine if there is a correlation between its metabolic parameters and clinical findings. METHODS: (18)F-FDG PET/CT was conducted to 35 patients with unilateral idiopathic FS. Clinical data including pain, functional scores, and passive range of motion (ROM) were collected. Maximum standardized uptake values (SUVmax) were measured at the four regions of interest (ROIs): rotator interval (RI), anterior joint capsule (AJC), axillary recess (AR), and posterior joint capsule (PJC) from the attenuation-corrected axial images. RESULTS: Mean SUVmax values for four ROIs of the affected shoulder were significantly higher than those of the unaffected shoulder. Mean SUVmax values of RI and AR were significantly higher than those of AJC and PJC and mean SUVmax of AJC was significantly higher than that of PJC in the affected side. Three recognizable patterns of increased uptake were noted: (1) AR dominant type (15 patients); (2) RI dominant type (9 patients); (3) both RI and AR dominant type (11 patients). The SUVmax of AR showed negative correlation with abduction and forward flexion. The SUVmax of RI showed negative correlation with external rotation and internal rotation. The SUVmax of AJC showed negative correlation with all ROMs. However, there was no significant correlation between the SUVmax of PJC and any ROM. CONCLUSION: Our study demonstrates that the anterior-inferior capsular portion, including RI and AR, is the main pathologic site of idiopathic FS and reveals significant correlations between ROM and metabolic parameters on (18)F-FDG PET/CT. These results imply that AR and RI lesions are related to elevational limitations and rotational limitations, respectively.