Prevalence and predictors of persistent pain 2 years after total shoulder arthroplasty

Abstract

BackgroundElective total shoulder arthroplasty is highly successful, but there is a subset of patients who continue to experience pain after this procedure. The purposes of this study were to elucidate the prevalence and patient characteristics predictive of persistent pain after shoulder arthroplasty. MethodsWe identified patients who had received an elective primary shoulder arthroplasty (anatomic or reverse) between 2016 and 2017 from our prospectively maintained, single-surgeon registry. Patients were stratified by the presence or absence of persistent pain at 2 years postoperatively, as defined by a Visual Analog Pain score of ≥2. This value was chosen in accordance with previously reported values for minimal clinically important differences. Multivariable logistic regression was used to identify factors associated with persistent pain. ResultsOf the 244 patients included for analysis, 46(18.9%) had persistent pain at 2 years postoperatively. The average pain score in this group was found to be significantly greater than the patients without pain (3.6 ± 1.6 vs 0.14 ± 0.35, P< .001). Patients with persistent pain had a higher incidence of preoperative opioid use (26.1% vs 13.6%, P= .038), diabetes (28.3% vs 11.6%, P= .004), number of self-reported allergies (3 ± 4.8 vs 1.9 ± 2.3,P= .03), peak postoperative inpatient pain (8.1 ± 1.9 vs 6.9 ± 2.1,P< .001), incidence of prior ipsilateral shoulder surgery (50% vs 30%, P= .011), and a higher percentage received a reverse shoulder arthroplasty (87% vs 70.7%, P= .024). The patient characteristics that were independently predictive of persistent pain were history of prior ipsilateral shoulder surgery (odds ratio 2.44; 95% confidence interval 1.21-4.91; P = .013) and greater postoperative inpatient pain intensity (odds ratio, 1.26 per 1 unit increase; 95% confidence interval 1.03-1.55; P = .024). ConclusionAbout 1 in 5 patients report persistent pain after elective primary shoulder arthroplasty. Prompt identification of at-risk patients (eg, those with severe postoperative inpatient pain and a history of prior shoulder surgery) may prove effective in optimizing the pain experience after shoulder arthroplasty. Level of evidenceLevel III; Cohort Study with Prospective Data and Retrospective Study Design.