The clinical impact of retears after repair of posterosuperior rotator cuff tears: a systematic review and meta-analysis

Background: Structural integrity after rotator cuff repair is frequently not achieved, but the clinical consequences of a retear remain disputed. The purpose of this meta-analysis was to analyze the relationships between postoperative cuff integrity and shoulder pain and function. Methods: We searched the literature for studies of surgical repair of full-thickness rotator cuff tears published after 1999 describing rates of retear and clinical outcomes and providing sufficient data to estimate effect sizes (standardized mean differences [SMDs]). Baseline and follow-up data of healed and failed repairs were extracted, including shoulder-specific scores, pain, muscle strength, and health-related quality of life. Pooled SMDs, mean differences, and overall change from baseline to follow-up according to structural integrity at follow-up were calculated. Subgroup analysis was performed to assess the influence of study quality on differences. Results: This analysis included 43 study arms with 3350 participants. The average age of the participants was 62 years (range, 52-78 years). The median number of participants per study was 65 (interquartile range, 39-108). At a median of 18 months' follow-up (interquartile range, 12-36 months), 844 repairs (25%) were described as retorn on imaging. The pooled SMD between healed re-pairs and retears at follow-up was 0.49 (95% confidence interval [CI], 0.37 to 0.61) for the Constant-Murley score, 0.49 (95% CI, 0.22 to 0.75) for the American Shoulder and Elbow Surgeons score, 0.55 (95% CI, 0.31 to 0.78) for other shoulder-specific out-comes combined, 0.27 (95% CI, 0.07 to 0.48) for pain, 0.68 (95% CI, 0.26 to 1.11) for muscle strength, and -0.001 (95% CI, -0.26 to 0.26) for health-related quality of life. The pooled mean differences were 6.12 (95% CI, 4.65 to 7.59) for the Constant-Murley score, 7.13 (95% CI, 3.57 to 10.70) for the American Shoulder and Elbow Surgeons score, and 0.49 (95% CI, 0.12 to 0.87) for pain, all below the commonly suggested minimal clinically important differences. The differences were not significantly affected by study quality and were generally modest compared with overall improvements from baseline to follow-up for both healed and failed repairs. Conclusions: The negative impact of retears on pain and function was statistically significant but judged to be of minor clinical importance. The results indicate that most patients may expect satisfactory outcomes despite retears. (c) 2023 The Author(s). This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).