Frozen shoulder: correlation between the response to physical therapy and follow-up shoulder arthrography

To study the correlation between improvement of shoulder motion and shoulder joint space capacity determinated by arthrography.

Case series.

General community hospital.

Twelve patients with clinically diagnosed frozen shoulder without rotator cuff tear. All subjects were divided as "primary" and "secondary" according to spontaneous onset or not, and "acute" or "chronic" depending on whether duration of disease was less than 2 months or longer.

Outpatient rehabilitation programs, including physical modalities, exercise intervention, and regular weekly outpatient clinic follow-up.

Shoulder range of motion (ROM) and joint space capacity in shoulder arthrography.

In acute patients, the joint space capacity increased significantly after treatment (t = 2.82; p < .05). Increased joint space capacity was most significantly correlated with improvement in external rotation (r = .77, p < .05), followed by abduction (r = .43, p > .05), but was poorly correlated with flexion and internal rotation. In chronic patients, both primary and secondary groups, there was no obvious joint space capacity increase despite significant shoulder motion improvement. Follow-up arthrograms showed the reappearance and/or enlargement of the axillary recess and smoother capular margins in all the patients except one chronic case (disease duration for 1 year). These findings were more obvious in acute than in chronic patients.

For frozen shoulder, generally described as "adhesive capsulitis," the adhesion was reversible in the acute stage. The increase of joint space capacity was significant and was correlated with improvement of external rotation. In chronic patients, ROM restoration occurred independent of change in joint space capacity, which increased slightly. The stretching of other contracted soft tissues around the shoulder, in addition to the adhesive capsule, may contribute to the recovery of chronic frozen shoulder.