A Cochrane review included 24 studies with a total of 2132 subjects. 10 studies examined the effect of early vs delayed (by about one week) implementation of post-operative exercise. Early exercise was more effective than delayed exercise in the short term recovery of shoulder range of motion (ROM) (Weighted Mean Difference (WMD): 10.6 degrees, 95% CI 4.51 to 16.6; 3 trials, n=677) but not at 4-6 weeks or at 6 months. However, early exercise also resulted in significant increase in wound drainage volume and duration (WMD: 1.15 days, 95% CI 0.65 to 1.65). 14 studies (6 post-operative, 3 during adjuvant treatment and 5 following cancer treatment) examined the effect of structured exercise compared to usual care/comparison. Structured exercise programs in the post-operative period significantly improved shoulder flexion ROM in the short-term (WMD: 12.92 degree, 95% CI: 0.69 to 25.16; 6 trials, n=324) and at 6 months (WMD 11.86, 95% CI 4.25 to 19.46; 3 trials, n=149). Physical therapy treatment yielded additional benefit for shoulder function post-intervention and at six-month follow-up. There was no evidence of increased risk of lymphedema from exercise at any time point.
Comment: The quality of evidence is downgraded by study quality (several shortcomings in most of the studies) and upgraded by large magnitude of effect.