Quantifying success after reverse total shoulder arthroplasty: the substantial clinically important percentage of maximal possible improvement

Background: Changes in preoperative to postoperative outcome scores are often used to quantify success after reverse total shoulder arthroplasty (rTSA). However, ceiling effects associated with many outcome scores limit the ability to differentiate success among high-functioning patients. The percentage of maximal possible improvement (%MPI) was introduced to simplify and better stratify pa-tient success. The primary purpose of this study was to define the %MPI thresholds associated with substantial clinical improvement following primary rTSA and compare the rates of success as defined by those achieving the substantial clinical benefit (SCB) compared to the 30% MPI for different outcome scores. Methods: A retrospective review was performed of an international shoulder arthroplasty database between 2003 and 2020. All primary rTSAs performed using a single implant system with a minimum 2-year follow-up were reviewed. Preoperative and postoperative outcome scores were evaluated for all patients to calculate improvement. Six outcome scores were assessed: the Simple Shoulder Test (SST), Constant, American Shoulder and Elbow Surgeons (ASES), University of California Los Angeles (UCLA), Shoulder Pain and Disability Index (SPADI), and Shoulder Arthroplasty Smart (SAS) scores. The proportion of patients achieving the SCB and 30% MPI was determined for each outcome score. Thresholds for the substantial clinically important %MPI (SCI-%MPI) were calculated using an anchor-based method for each outcome score and stratified by age and sex. Results: Of total, 2573 shoulders with a mean follow-up of 47 months were included. Outcome scores with known ceiling effects (SST, ASES, UCLA, SPADI) had higher rates of patients achieving the 30% MPI compared to scores without ceiling effects (Constant, SAS). However, scores without ceiling effects had higher rates of patients achieving the SCB. The SCI-%MPI differed among outcome scores, and mean values were 47% for the SST, 35% for the Constant score, 50% for the ASES score, 52% for the UCLA score, 47% for the SPADI score, and 45% for the SAS score. The SCI-%MPI increased in patients older than 60 years (P < .001) except for the SAS and Constant scores. SCI-%MPI was greater in females for all scores assessed except the Constant and SPADI scores (P < .001 for all). The higher SCI-%MPI thresholds in these populations mean that these patients required a greater fraction of the MPI to have substantial improvement.