Early Versus Delayed Passive Range of Motion Exercise for Arthroscopic Rotator Cuff Repair: A Meta-analysis of Randomized Controlled Trials

• Published in: The American journal of sports medicine 2015-05, Vol.43 (5), p.1265-1273
• Main Authors: Chang, Ke-Vin, Hung, Chen-Yu, Han, Der-Sheng, Chen, Wen-Shiang, Wang, Tyng-Guey, Chien, Kuo-Liong
• Format: Article
• Language: English
• Subjects: Arthroscopy; Arthroscopy - methods; Clinical trials; Exercise; Exercise Therapy - methods; Female; Humans; Meta-analysis; Randomized Controlled Trials as Topic; Range of motion; Range of Motion, Articular; Recovery of Function; Rehabilitation; Rotator cuff; Rotator Cuff - surgery; Shoulder; Shoulder Joint - surgery; Sports medicine; Systematic review; Treatment Outcome; Wound Healing
• ISSN: 0363-5465; 1552-3365
• DOI: 10.1177/0363546514544698

Background: Postoperative shoulder stiffness complicates functional recovery after arthroscopic rotator cuff repair. Purpose: To compare early passive range of motion (ROM) exercise with a delayed rehabilitation protocol with regard to the effectiveness of stiffness reduction and functional improvements and rates of improper healing in patients undergoing arthroscopic repair for torn rotator cuffs. Study Design: Systematic review and meta-analysis. Methods: Randomized controlled trials (RCTs) comparing both rehabilitation approaches were identified in PubMed and Scopus. Between-group differences in shoulder function were transformed to effect sizes for comparisons, whereas the effectiveness against stiffness and the risk of tendon failure were reported using standardized mean differences of ROM degrees and odds ratios (ORs) of recurrent tears, respectively. Results: Six RCTs were included, consisting of 482 patients. No significant difference in shoulder function existed across both protocols. The early ROM group demonstrated more improvement in shoulder forward flexion than the delayed rehabilitation group, with a standardized mean difference of 7.45° (95% CI, 3.20°-11.70°) at 6 months and 3.51° (95% CI, 0.31°-6.71°) at 12 months. Early ROM exercise tended to cause a higher rate of recurrent tendon tears (OR, 1.43; 95% CI, 0.90-2.28), and the effect became statistically significant (OR, 1.93; 95% CI, 1.04-3.60) after excluding 2 RCTs that recruited only those patients with small to medium-sized tears. Conclusion: Early ROM exercise accelerated recovery from postoperative stiffness for patients after arthroscopic rotator cuff repair but was likely to result in improper tendon healing in shoulders with large-sized tears. The choice of either protocol should be based on an accommodation of the risks of recurrent tears and postoperative shoulder stiffness.