Abstract
INTRODUCTION: The recovery patterns following arthroscopic rotator cuff repairs are influenced by several factors, including the size of the tear and the degree of patient compliance. This study aimed to ascertain whether a correlation exists between the size of the rotator cuff tear before surgery and the subsequent recovery of the rotator cuff following arthroscopic repair. MATERIALS AND METHODS: We retrospectively reviewed 744 patients who underwent arthroscopic rotator cuff repair. Based on the inclusion and exclusion criteria, 132 patients were finally included. Muscle strength, range of motion (ROM), pain, and clinical scores were assessed preoperatively and postoperatively. Demographic and perioperative factors were assessed, and factors significantly correlated with muscle strength recovery were entered into multivariate binary logistic regression to obtain odds ratios. RESULTS: In small and medium tears, muscle strength from preoperative levels significantly recovered at 6 and 12 months, respectively. In contrast, no significant improvement in muscle strength was observed in large-to-massive tears. Pain scores initially improved at the 6-month mark for medium and large-to-massive tears, whereas small tears demonstrated improvement from the 12-month mark. A history of more than three shoulder injections may have an adverse effect on abduction strength recovery following arthroscopic rotator cuff repair. Furthermore, the potential risk of poor abduction and external rotation strength recovery exists when the requisite tension for rotator cuff repair > 35 N. CONCLUSIONS: The study demonstrated that the tear size exhibited a robust correlation with the pace and prospective outcome of muscle strength recovery. Conversely, pain, ROM, and clinical score were not significantly associated with the tear size. A history of more than three shoulder injections prior to surgery and increased requisite tension power for cuff repair are associated with a higher likelihood of not recovering as much muscle strength as the uninjured shoulder. LEVEL OF EVIDENCE: Level III. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00402-026-06218-2.