Constant-Murley score categorization to assess outcomes after reverse shoulder arthroplasty following complex proximal humeral fracture.

Abstract

The Constant-Murley (CM) score is considered the gold standard of shoulder scoring systems. However, it does not include scoring categorization, and in some acute traumatic processes, a paired comparative analysis is not feasible. The purpose of this study was to determine cutoff values for the CM score that indicate a specific category on the outcome score following reverse total shoulder arthroplasty (rTSA) for fracture. Consecutive patients with complex proximal humeral fracture who underwent rTSA from 2010 to 2022, were included in the study. Two years of postoperative follow-up was required. Postoperative clinical evaluations were conducted using the CM and University of California, Los Angeles (UCLA) scores. The correlation between CM and UCLA scores was analyzed. Then, the receiver-operating characteristic (ROC) curve was utilized to determine the cutoff points in CM scores that most effectively distinguished between categories. The UCLA categories were used as the external criterion. 132 patients completed the evaluations at the 2-year follow-up. Among them, 112 (84.8 %) were females and 20 (15.2 %) were males, with a mean age of 73.5 years. The UCLA score showed a strong correlation with the CM score (r = 0.936; 95% CI 0.911-0.954; p = 0.000). For the CM score, the cutoff for an excellent outcome was 67.5 (AUC 0.963, 95% CI 0.925-1.000), 56.5 (AUC 0.920, 95% CI 0.872-0.969) for good, 41.5 (AUC 0.701, 95% CI 0.617-0.784) for fair, and <41.5 (AUC 0.965, 95% CI 0.939-0.990) for poor. According to the cut-off values in the CM score used to categorize the outcomes after rTSA in proximal humeral fractures, an excellent outcome was considered a score equal to or greater than 68, good between 57 and 67 points, fair between 41 and 56 points, and a poor outcome with a score equal to or less than 40 points. III.