Repair Integrity and Clinical Outcomes Following Arthroscopic Rotator Cuff Repair: A Prospective, Randomized Trial of Early and Delayed Motion Protocols

OBJECTIVES: To prospectively evaluate the effect of early versus delayed motion on repair integrity on 6-month postoperative magnetic resonance imaging (MRI) scans following rotator cuff repair, and to correlate repair integrity with clinical and functional outcomes. We hypothesized that repair integrity would differ between the early and delayed groups and that patients with repair failures would have worse clinical and functional outcomes. METHODS: This was a prospective, randomized, single blinded clinical trial comparing an early motion (post-op day 2-3) to a delayed motion (post-op day 28) rehabilitation protocol following arthroscopic repair of isolated supraspinatus tears. All patients underwent MRI at 6 months post-operatively as part of the study protocol. A blinded board-certified and fellowship-trained orthopaedic surgeon (not part of the surgical team) reviewed operative photos and video to confirm the presence of a full thickness supraspinatus tear and to ensure an adequate and consistent repair. The same surgeon along with a blinded sports medicine fellowship-trained musculoskeletal radiologist independently reviewed all MRIs to determine whether the repair was intact at 6 months. Outcome measures were collected by independent evaluators who were also blinded to group assignment. These included the Western Ontario Rotator Cuff (WORC) index, Single Assessment Numeric Evaluation (SANE) ratings, pain scores, sling use, and physical exam data. Enrolled patients were followed at 6 weeks, 6 months, and 1 year. RESULTS: From October 2008 to April 2012, 73 patients met all inclusion criteria and were willing to participate. 36 patients were randomized to delayed motion and 37 were randomized to early motion. The final study group at 6 months consisted of 58 study participants. Postoperative MRIs were obtained on all of these patients at 6 months regardless of whether or not they were progressing as expected. These MRIs demonstrated an overall failure rate of 29%. This did not differ significantly based on early or late motion, with 9 (26%) tears occurring in the delayed motion group compared to 10 (32%) in the early motion group (p=0.70). When patients with repair failures were compared to those with intact cuffs, no significant differences were detected in range of motion, strength, or WORC, SANE, or pain scores at 6 months or 1 year. Interestingly, however, at 6 weeks, both WORC and SANE scores were significantly better in patients who were later found to have repair failures on MRI (p<0.05). When evaluated irrespective of rehabilitation protocol, non-compliance with sling use was associated with higher failures rates (p<0.05). CONCLUSION: Repair failure rates do not significantly differ between patients randomized to early and delayed motion protocols following arthroscopic single tendon rotator cuff repair. Repair failure on MRI does not correlate with clinical outcome at 6 months or 1 year. However, better subjective outcome scores at 6 weeks are associated with higher rates of repair failure at 6 months.