Predicting outstanding results after anatomic total shoulder arthroplasty using percentage of maximal outcome improvement

Abstract

The purpose of this study was to determine whether thresholds in the percentage of maximal improvement in the Simple Shoulder Test (SST) or American Shoulder and Elbow Surgeons (ASES) score exist for predicting "excellent" patient satisfaction after total shoulder arthroplasty (TSA). A retrospective query identified patients who underwent TSA with a minimum of 2 years' follow-up. Preoperative and postoperative SST and ASES scores and postoperative patient satisfaction were recorded. Receiver operating characteristic curve analyses were performed to determine thresholds in the percentage of maximal improvement in the SST and ASES scores that predict excellent satisfaction. Univariate and multivariate analyses determined preoperative factors that predicted achievement of these thresholds. A total of 301 and 319 patients had at least 2 years' follow-up for the SST score and ASES score, respectively. We determined 72.1% of maximal improvement in the SST score to be the threshold for excellent satisfaction (area under the curve, 0.777; 95% confidence interval, 0.712-0.841; P < .001). We determined 75.6% of maximal improvement in the ASES score to be the threshold for excellent satisfaction (area under the curve, 0.799; 95% confidence interval, 0.743-0.856; P < .001). Both groups showed significant positive correlations between percentage of maximal score achieved and excellent satisfaction (r = 0.396 for SST score [P < .001] and r = 0.325 for ASES score [P < .001]). Younger age was the only independent predictor for achieving the SST score threshold. No independent predictors existed for the ASES score threshold. Achievement of 72.1% of maximal SST score improvement and achievement of 75.6% of maximal ASES score improvement represent thresholds for achievement of excellent satisfaction after TSA. Most preoperative factors did not have an impact on the likelihood of achieving these thresholds.