Patients With Mild Osteoarthritis Are Less Likely to Achieve a Clinically Important Improvement in Pain or Function After Anatomic Total Shoulder Arthroplasty

Background The severity of glenohumeral osteoarthritis (OA) as demonstrated by preoperative radiographs and patient-reported pain plays an important role in the indication for anatomic total shoulder arthroplasty (aTSA). In hip and knee research, data about the effect of the severity of preoperative radiographic OA on the outcome of total joint arthroplasty have been mixed. For shoulder replacement, we are unsure of the effects of radiographic severity on outcomes. Questions/purposes This study investigated whether the preoperative radiographic severity of glenohumeral OA is associated with improvement in pain and function after aTSA. We asked, (1) does the severity of glenohumeral OA correlate with improvement in patient-reported outcomes after TSA (delta American Shoulder and Elbow Surgeons score [postoperative-preoperative], delta Single Assessment Numeric Evaluation, delta Simple Shoulder Test, and delta VAS)? (2) Is having mild osteoarthritis associated with not meeting the minimum clinically important differences in preoperative and postoperative American Shoulder and Elbow Surgeons scores? Methods An institutional query of patients who underwent aTSA for OA was performed between January 2015 and December 2018. A total of 1035 patients were eligible; however, only patients with adequate preoperative radiographs and patient-reported outcome measures collected preoperatively and at a minimum of 2 years postoperatively were included. Patients with proximal humerus fractures, inflammatory arthropathy, cuff tear arthropathy, prior ipsilateral rotator cuff repair, brachial plexus injury or neuromuscular disorder, workers compensation, periprosthetic joint infection, or revision surgery within 2 years were excluded. Patient characteristics, comorbidities, and prior shoulder surgery were recorded. The severity of OA was classified based on the modified Samilson-Prieto and Walch classification. The association between Samilson-Prieto grade and patient-reported outcome measures (American Shoulder and Elbow Surgeons Score, Single Assessment Numeric Evaluation, Simple Shoulder Test, and VAS score) was evaluated. Radiographic characteristics, patient demographics, comorbidities, and prior surgery were also evaluated for the potential risk of not achieving improvement in the minimum clinically important difference (16.1) with respect to the American Shoulder and Elbow Surgeons score. The American Shoulder and Elbow Surgeons score is scored 0 to 100, with higher scores representing less pain and better function. A total of 206 patients (20% of those eligible) with a mean follow-up of 2.3 years were included. Twenty-three patients had Samilson-Prieto Grade I, 38 had Grade II, 57 had Grade III, and 88 had Grade IV. Results There were no differences in improvements (delta) between the groups and between patient-reported outcome scores (American Shoulder and Elbow Surgeons score, Single Assessment Numeric Evaluation, Simple Shoulder Test, and VAS). Compared with patients with more severe osteoarthritis (Samilson-Prieto Grades II, III, and IV), a higher proportion of patients with less severe osteoarthritis (Grade I) did not exceed the minimum clinical important difference for the American Shoulder and Elbow Surgeons score (22% [five of 23] versus 4% [seven of 183]; odds ratio 0.14 [95% confidence interval 0.04 to 0.520]; p = 0.006). Conclusion The improvement in patient-reported outcome measure scores was similar regardless of radiographic severity after aTSA. Surgeons should use caution when recommending surgery to patients with less severe OA because a higher percentage did not improve, based on the minimum clinically important difference.