Assessment of the optimal shoulder outcome score for reverse shoulder arthroplasty.

BACKGROUND: With growing attention being paid to quality and cost-effectiveness in health care, outcome evaluations are becoming increasingly important. This determination can be especially difficult in reverse shoulder arthroplasty (RSA) given the complex pathology and extensive disabilities in this patient population. This study evaluated the use of 3 validated questionnaires used to assess outcomes for RSA. METHODS: Using a database of patients treated with RSA, we assessed preoperative and postoperative Constant-Murley Scores, American Shoulder and Elbow Surgeons Scores, and Subjective Shoulder Values in 148 shoulders. The outcomes at each scoring period were described, and the scores were compared with one another as well as with active range of motion. RESULTS: There were no significant differences in the mean improvement of any of the scores. Improvements in all of the outcome scales were correlated with each other and with improvement in forward elevation but not with external rotation. Multivariate regression analysis the 3 outcome measures was able to predict 38.9% of the variation in improvement in functional outcomes (forward elevation). This was only slightly greater than that provided by improvements in the outcome variable Constant-Murley score alone (36.7%). CONCLUSIONS: The 3 shoulder outcome scores evaluated, regardless of whether they were patient reported or physician based, appear to appropriately reflect improvements after RSA with equal validity. The objective physician-assessed Constant-Murley score had the strongest correlation with function of the arm, and use of a combination of all 3 outcome scores did not increase the ability to predict range of motion compared with using the Constant-Murley score alone.