Rotator cuff tears are seen in 40% of subjects over age 50 57, with a prevalence known to increase with each decade of life 51. Each year rotator cuff disorders lead to 4.5 million physician visits, 40,000 inpatient surgeries, and 250,000 outpatient surgeries with costs of $17,427 per patient in 2004. As the population ages, the number of rotator cuff repair surgeries is increasing; in New York State, rotator cuff repairs increased by 50% over a 5 years span 48. Unfortunately, surgically repaired rotator cuff tears fail in at least 20% and up to 94% of individuals 3, 9, 13, 20, 28, 40, 52, 58. While many non-modifiable variables have been associated with failure of repair, postoperative rehabilitation is a modifiable variable that has received little attention. The investigators propose a multicenter randomized controlled trial to study one important strategy for postoperative rehabilitation: early versus delayed onset of physical therapy. This pilot study will enroll patients with isolated supraspinatus tears who undergo a standard surgical repair technique. The investigators hypothesize that delaying the onset of physical therapy will improve healing and patient outcomes. Our primary outcome variable is the Western Ontario Rotator Cuff (WORC) Index (a disease specific validated outcome measure). Because outcome measures may not always correlate with healing of rotator cuff repairs 50, our secondary outcome measure will be healing based on MRI scans 12 months after surgery.

Surgically repaired rotator cuff tears fail at surprisingly high rates, approaching 20-94% as determined by MRI 3, 9, 13, 20, 28, 40, 52, 58. Non-modifiable variables associated with rotator cuff repair failure include: duration of symptoms 9, 19, 39, fatty infiltration of the atrophied muscle 6, 15, 17, larger rotator cuff tears 6, 9, 19, 20, 21, 26, 39, 47, older age 9, 20, 22, 48, 55, co-morbidities 48, and workers compensation claims 2, 22, 23. Potentially modifiable variables include: tobacco use 11, 38, limited preoperative range of motion 45, and surgeon volume 48.

Two other modifiable variables are surgical technique and postoperative management. As advances in arthroscopic techniques have evolved, a variety of studies have evaluated different surgical techniques in clinical series and trials 3, 4, 20, 12, 18, 31, 36. In 2006, Park et al described the "transosseous equivalent" repair technique characterized by sutures secured medially, passed through and over the top of the rotator cuff, and then secured laterally 44. This method of repair has been shown to have better biomechanical properties and produce superior healing rates compared to other techniques in randomized clinical trials 5, 12. These data suggest this technique is currently the best available method to repair rotator cuff tears.

Postoperative management of patients who have had rotator cuff repair has received little attention in the literature, yet is likely one of the most important modifiable variables that can influence rotator cuff repair integrity. It has been argued that early physical therapy may increase failure rates of rotator cuff repair 1, 14. This has led some to recommend delaying therapy for up to 6 weeks after surgery.

Three groups have studied delaying the onset of physiotherapy in patients with full thickness rotator cuff tears. Klintberg et al 29 performed a randomized trial on patients who underwent open rotator cuff repair comparing two postoperative physical therapy protocols: a "Progressive Therapy Group" ¬ (started the day after surgery with passive range of motion and dynamic muscle activation of the rotator cuff, a sling was removed after 4 weeks and rotator cuff loading was progressive through the rehabilitation), and a "Traditional Group" (passive range of motion the day after surgery with immobilization in a sling for 6 weeks during which time no rotator cuff loading was applied). In this trial no significant differences in outcome were detected.

In contrast to this, Deutsch et al 7 reported on 70 patients who had a single row arthroscopic rotator cuff repair with simple suture configuration randomized to one of two physical therapy protocols that were different only in that the "Standard" protocol began passive forward elevation on postoperative day 7 (N=37), whereas the "Delayed" protocol began passive forward elevation after 4 weeks (N=33). At 6 months range of motion was not different, however ultrasound examination demonstrated 19% of the repairs failed in the "Standard" group, whereas only 9% failed in the "Delayed" group (p<0.05).

In a prospective cohort study presented by Accousti et al 1, 56 patients were all managed by 6 weeks of immobilization following rotator cuff repair. When immobilization was removed, 13 patients were considered "stiff" with limited motion, and 43 had "good" passive motion. By one year none of the patients were considered stiff. In the "stiff" group rotator cuff repairs failed in 30%, whereas in the group with "good" motion at six weeks 62% of repairs failed.

These data, while limited, suggest that some period of immobilization may improve healing of rotator cuff repairs. The effect of immobilization on other measures of outcome is unclear. Unfortunately none of these studies used the transosseous equivalent repair technique.

The purpose of this study is to compare two different strategies of postoperative management in patients with isolated supraspinatus rotator cuff tears repaired with a standardized "SpeedBridge" arthroscopic transosseous equivalent technique, using a disease specific validated measure of outcome (WORC index) as the primary outcome variable. The design will be a multicenter randomized trial comparing immediate to delayed physical therapy. As patient oriented outcomes frequently do not correlate with rotator cuff repair healing 50, the second aim is to study the effect of these rehabilitation paradigms on rotator cuff healing by MRI.

Inclusion Criteria:

• Patients for inclusion based on MRI documented full thickness supraspoinatus tear with acute tears, or chronic tears that fail non-operative therapy.
• Tears that involve the supraspinatus and are minimally displaced (Grade I) or displaced to the humeral head (Grade II) will be included

Exclusion Criteria:

1. Age < 18 years
2. Evidence of major joint trauma, infection, avascular necrosis, chronic dislocation, inflammatory or degenerative glenohumeral arthropathy, frozen shoulder, or previous surgery of the affected shoulder,
3. Evidence of significant cuff arthropathy with superior humeral translation and acromial erosion diagnosed by x-ray or other investigations,
4. Major medical illness (life expectancy < 2 years or unacceptably high operative risk),
5. Unable to speak or read English,
6. Psychiatric illness that precludes informed consent,
7. Unwilling to be followed for 2 years,
8. Large, massive, or irreparable cuff tears extending into the subscapularis or teres minor,
9. Inelastic and immobile tendon which cannot be advanced to articular margin,
10. Co-existing labral pathologies requiring repair (SLAP II-IV), Bankart lesions requiring repair, partial tears of biceps (more than 60% of thickness) requiring surgical treatment,
11. Acromioclavicular pathology requiring a distal clavicle excision.