Comparative analysis of early versus delayed rehabilitation protocols following rotator cuff repair: A randomized controlled trial

Background: The optimal timing for initiating shoulder motion after rotator cuff repair remains controversial, balancing re-tear risk against stiffness from immobilization. This trial compared clinical and structural outcomes of early versus delayed rehabilitation. Methods: From March 2024 to February 2025, 120 patients undergoing arthroscopic (single-row) or open (double-row) repair of full-thickness tears were randomized to an early-motion group (EMG; n = 60) or a delayed-motion group (DMG; n = 60). The EMG began supervised passive forward flexion and abduction within 24 h postoperatively. The DMG underwent strict immobilization for 6 weeks before starting identical rehabilitation. Primary outcomes at 6 months included pain (VAS), function (UCLA score), active range of motion (ROM), and re-tear rate on MRI (Sugaya IV/V). Secondary outcomes covered satisfaction, complications, and return to activities. Demographic, clinical, and surgical factors were analyzed. Results: At 6 months, the EMG showed significantly lower pain (VAS: 1.7 ± 0.8 vs. 2.9 ± 1.1; p = 0.02), higher function (UCLA: 33.7 ± 2.1 vs. 29.4 ± 3.3; p = 0.01), greater ROM, and lower stiffness (5 % vs. 15 %; p = 0.03). Re-tear rates did not differ significantly (EMG: 3.3 % vs. DMG: 1.7 %; p = 0.56). All three re-tears occurred in patients with large tears (>3 cm) and high-grade fatty infiltration (Goutallier ≥3). Multivariate analysis confirmed early motion as an independent predictor of better outcomes. Surgical technique did not alter the primary findings. Conclusions: Supervised early passive motion is safe and superior to delayed immobilization, providing better functional recovery and lower stiffness without increasing re-tear risk. However, patients with large tears and advanced fatty infiltration represent a high-risk subgroup, necessitating personalized, cautious rehabilitation. Level of evidence: Therapeutic Level I.