Changes from baseline in patient-reported outcomes and patient satisfaction do not vary significantly between 1 and 2 years postoperatively after shoulder arthroplasty: a multicenter analysis of 2580 patients

Background

At the present time, most orthopedic journals require 2 years of follow-up data for clinical studies. However, it is unclear if there are significant changes in patient-reported outcomes (PROs) between 1 and 2 years postoperatively. The current study sought to evaluate changes in PROs between 1 and 2 years postoperatively after shoulder arthroplasty.

Methods

A prospective, multicenter registry was queried for all patients that underwent primary total shoulder arthroplasty, primary reverse shoulder arthroplasty (RSA), revision RSA, and RSA for fracture. Patients with preoperative, 6-month, 1-year, and 2-year postoperative American Shoulder and Elbow Surgeons (ASES) scores were included. We evaluated mean ASES scores, Δ (improvement from preoperative) ASES, and Percent Maximum Possible Improvement. We also evaluated achievement of Clinically Significant Outcomes (CSOs) for the ASES score, including the Minimal Clinically Important Difference, Substantial Clinical Benefit, and Patient-Acceptable Symptom State. Patient satisfaction questionnaires were also evaluated.

Results

A total of 2580 patients were identified that underwent shoulder arthroplasty for all indications with full follow-up data. Overall, changes between 1 and 2 years postoperatively were minimal: ASES score 0.8, Δ ASES 0.8, and % Maximum Possible Improvement 1.6%. Similarly, the percentage of patients who achieved CSOs was minimally different at 1 and 2 years: Minimal Clinically Important Difference −0.4%, Substantial Clinical Benefit 1.1%, and Patient-Acceptable Symptom State 2.4%. Subgroup analyses of all primary arthroplasties, primary total shoulder arthroplasty, primary RSA, revision RSA, and RSA for proximal humerus fracture revealed similar, minimal changes from 1 to 2 years postoperatively. Overall, less than 5% of patients changed their responses to any of the satisfaction questions between 1 and 2 years: 4.1% of patients changed their answer with respect to pain, 2.8% with respect to function, 3.9% with respect to activities of daily living, and 2.5% with respect to ability to return to sports.

Conclusion

PROs, achievement of CSOs, and patient satisfaction are minimally different at 1 and 2 years after shoulder arthroplasty. This data suggests that accepting 1-year, rather than 2-year, outcomes for short-term reporting of shoulder arthroplasty may facilitate improvement in follow-up rates with less risk of bias due to loss to follow-up, more timely dissemination of information, and decreased administrative burden. This does not change the need for midterm and long-term studies.