Effectiveness of nonoperative treatment in patients with glenohumeral osteoarthritis: a prospective cohort study

Abstract

Background

There is limited evidence supporting the use of nonoperative strategies in the treatment of glenohumeral osteoarthritis (GHOA). Recent clinical practice guidelines have stated that it is unclear whether nonoperative management of GHOA would produce a clinically important difference in pain or function. Therefore, the purpose of this study was to determine the effectiveness of nonoperative treatment on patient-reported outcomes (PROs) and to identify factors that could predict which patients would undergo total shoulder arthroplasty (TSA).

Methods

62 patients with primary GHOA were recruited. Patients could choose to receive or refuse different nonoperative modalities, including physical therapy (PT) and corticosteroid injections, based on their preference. American Shoulder and Elbow Surgeons (ASES) score were administered at baseline, 3, 6, and 12 months to evaluate treatment response. Demographic, clinical, and radiographic characteristics were compared between patients who failed and did not fail nonoperative management. Failure was defined as having undergone TSA.

Results

14 (23%) patients who initially attempted nonoperative management underwent TSA at 7.7 months (range, 1.6-25.2 months). In patients who continued nonoperative management, only 19 (31%) patients met the minimum clinical important difference and 26 (42%) patients achieved patient acceptable symptom state. There was no significant difference in the change in ASES score between patients who did and did not undergo PT (P = .524). A lack of belief in PT (HR = 33.6 [95% CI: 5.26-214], P < .001), decrease in ASES score (HR = 6.25 [95% CI: 2.04-20.0], P = .001]), female sex (HR = 5.38 [95% CI: 1.31-22.1], P = .020), and lower resilience (HR = 7.14 [95% CI: 1.78-33.3], P = .006) were independently associated with failure of nonoperative treatment. Patients who received at least one glenohumeral corticosteroid injection (HR = 0.16 [95% CI: 0.04-0.67], P = .012) or had more joint space remaining (HR = 0.22 [95% CI: 0.06-0.80], P = .021) had a decreased risk of failure.

Conclusions

Approximately, 30% of patients with GHOA who chose their nonoperative treatment regimen had clinically meaningful improvements in symptoms. Despite this, patients elected to undergo TSA less than 25% of the time at short-term follow-up. PT was not beneficial in the treatment of GHOA. Screening questionnaires that evaluate a patient’s belief in PT and resilience could potentially be used to identify which patients will fail nonoperative treatment.