Our purpose was to analyze the impact of the speed of recovery of ROM on tendon healing in patients undergoing an isolated arthroscopic SSN repair and the same rehabilitation protocol. We hypothesized that a faster recovery of ROM would lead to a better functional outcome without compromising tendon healing.

This was a prospective monocentric study. All primary isolated arthroscopic SSN repairs for small to medium tears, without retraction (Patte 1), significant fatty infiltration (Goutallier<2) and glenohumeral osteoarthritis were eligible. Patients who did not complete all follow-ups were excluded. An experienced orthopedic surgeon performed all procedures using a double-row technique. All patients followed the same rehabilitation protocol. It included a sling and progressive passive overhead stretches and ER with the elbow at side during the first 6 weeks followed by active mobilization. An independent observer assessed all patients before and at 6 weeks, 3 and 6 months after surgery. Collected data at each follow-up included passive and active EA and ER as well as the VAS for pain. The Constant score was obtained before and at 6 months after surgery. A single radiologist examined the healing of the repair by ultrasound at 6 months postoperatively according to Sugaya.

1323 consecutive patients between 2010 and 2020 were eligible. 169 were excluded according to the aforementioned criteria. Finally, 1154 arthroscopic SSN repairs were included. The healing rate was 87.3 %. Preoperative characteristics of healed and non-healed repairs were similar in terms of passive and active ROM, VAS pain, and Constant score. Compared to the non-healed repairs, the healed ones were slightly younger (57.8±8.0 vs. 61.5±8.5 years; p<0.001) and had a lower passive AE and ER at 6 weeks and 3 months postoperatively. However, this difference faded by 6 months after surgery. There was no difference in Constant score and VAS pain between healed and non-healed repairs. In both cases, the SSN repair resulted in an improvement of the Constant score at 6 months postoperatively and a decrease in the VAS pain already from the 6th postoperative week.

Our study shows that the speed of recovery of passive ROM influences tendon healing after isolated arthroscopic SSN repair. Repairs that resulted in healing had lower AE and ER up to 3 months after surgery. This difference did not affect the level of pain and/or shoulder function, which were similar.