The Statistical Fragility of Functional Outcomes for Arthroscopic Rotator Cuff Repair With and Without Acromioplasty: A Systematic Review and Meta-analysis.

BACKGROUND Views surrounding acromioplasty at the time of arthroscopic rotator cuff repair (RCR) have shifted dramatically over time. In recent years, various studies have argued against acromioplasty, citing equivocal functional outcomes after arthroscopic RCR with or without acromioplasty. PURPOSE To assess the statistical fragility of functional outcomes after arthroscopic RCR with and without acromioplasty using the reverse continuous fragility index (RCFI). STUDY DESIGN Systematic review and meta-analysis; Level of evidence, 3. METHODS A systematic review and meta-analysis was performed including all randomized controlled trials through February 5, 2024 investigating arthroscopic RCR with and without acromioplasty. The RCFI, defined as the number of qualifying data points required to be moved from the lower mean group to the higher mean group to alter the significance, was calculated for the Welch t test, Student t test, and Wilcoxon rank-sum test under various data assumptions. The reverse continuous fragility quotient (RCFQ) was determined by dividing the RCFI by the sample size. RESULTS A total of 6 clinical trials consisting of 609 patients with functional outcome scores were analyzed. Using the Welch t test, the median RCFI across all study outcomes was 20 (interquartile range [IQR], 17-24). For the Student t test, the median RCFI across all study outcomes was 14 (IQR, 13-19), with a median RCFQ of 0.18 (IQR, 0.15-0.20). For the Wilcoxon rank-sum test, the median RCFI was 14 (IQR, 13-17), with a median RCFQ of 0.17 (IQR, 0.13-0.19). While using the Welch t test, 64% of study outcomes had an RCFI greater than the loss to follow-up (LTFU). When using the other tests, 32% of study outcomes had an RCFI greater than the LTFU. CONCLUSION The fragility of these studies was largely dependent on the statistical test used to analyze the results. The Wilcoxon rank-sum test and Student t test appeared to be most appropriate to find differences in treatment arms. When using these tests, we found the results to be fragile. This, in combination with a small number of studies and the LTFU close to or exceeding 20%, indicates an overall lack of strong evidence to support previously accepted conclusions.