The Clinical Effect of a Rotator Cuff Retear: A Meta-analysis of Arthroscopic Single-Row and Double-Row Repairs

Abstract

Background:

The clinical effect of a retear after rotator cuff repair remains unclear. While some studies have indicated clinical deficits due to a retear, others have stated that a retear does not detrimentally affect outcomes.

Purpose:

To conduct a meta-analysis comparing clinical outcomes between intact and retorn rotator cuffs after arthroscopic repair.

Study Design:

Meta-analysis.

Methods:

A literature search using the terms “arthroscopic,” “rotator cuff,” “repair,” “retear,” “re-tear,” “defect,” “single-row,” “double-row,” “clinical outcomes,” and “functional outcomes” was conducted. Article inclusion criteria were an adequate description of the surgical technique, stratification of outcomes by intact rotator cuff versus retear with a minimum of 1 year of follow-up, and documentation of the presence/absence of a full-thickness retear using imaging. Exclusion criteria were isolated subscapularis tears/repairs, labral repairs, infections, postoperative fractures, insufficient data or statistical indications, and postoperative data not stratified by retear versus intact rotator cuff. A meta-analysis was performed using a random-effects model on variables that had comparisons from at least 3 studies. Single-row (SR) and double-row (DR) studies were analyzed both separately and together in an “all arthroscopic repairs” (AAR) comparison. The calculated effect was considered significant at a P value <.05.

Results:

Within the SR group, patients with a rotator cuff retear had a significantly lower Constant score (mean difference [95% CI], −6.79 [–8.94 to −4.65]; P < .001) and lower University of California, Los Angeles (UCLA) score (−3.21 [–5.27 to −1.15]; P = .002) but not higher pain (0.071 [–0.34 to 0.49]; P = .739). Within the DR group, patients with a rotator cuff retear had a significantly lower Constant score (mean difference [95% CI], −9.35 [–12.2 to −6.50]; P < .001), lower American Shoulder and Elbow Surgeons (ASES) score (−12.1 [–17.1 to −7.26]; P < .001), lower UCLA score (−3.07 [–4.85 to −1.29]; P < .001), higher pain (0.622 [0.19 to 1.05]; P = .005), and lower abduction strength (P < .001). In the AAR comparison, patients with a retear had a significantly lower Constant score (mean difference [95% CI], −7.56 [–9.55 to −5.57]; P < .001), lower ASES score (−10.1 [–15.5 to −4.64]; P < .001), lower UCLA score (−3.00 [–4.47 to −1.53]; P < .001), and lower abduction strength (in kg·f) (−3.32 [–4.53 to −2.12]; P < .001) but not higher pain (0.332 [–0.014 to 0.680]; P = .060).

Conclusion:

Patients with a full-thickness rotator cuff retear exhibited significantly lower clinical outcome scores and strength compared with patients with an intact or partially torn rotator cuff.