Preoperative factors associated with loss of range of motion after reverse shoulder arthroplasty

Reverse total shoulder arthroplasty (RTSA) is a successful procedure, often allowing patients to achieve better range of motion (ROM) compared with their preoperative baseline. However, there is a subset of patients who either fail to improve or lose ROM postoperatively. These patients are at increased risk of poor satisfaction and patient-reported outcomes. To date, characteristics of this subset of patients have not been well described. The purpose of this study is to determine risk factors associated with loss of ROM after primary RTSA.

A retrospective review using a commercial international RTSA database (Exactech Inc., Gainesville, FL, USA) of patients who underwent primary RTSA between 2007 and 2017 was performed. A total of 123 (7.7%) shoulders lost ≥10° of forward elevation (FE) (group 1, P₁) and 183 (11.4%) lost ≥10° of external rotation (ER) (group 2, P₂). Univariate and multivariate analyses were performed comparing these patients with control cohorts to evaluate risk factors for loss of motion.

Better preoperative abduction, FE, ER, and internal rotation were each associated with greater loss of FE (P₁ < .001) and ER (P₂ < .001) postoperatively. Higher preoperative Simple Shoulder Test (P₁ < .001, P₂ < .001), Constant (P₁ < .001, P₂ < .001), Shoulder Pain and Disability Index (P₁ < .001, P₂ < .001), American Shoulder Elbow Surgeons (P₁ < .001), and University of California at Los Angeles (P₁ < .001) scores were also strongly associated with loss of ROM postoperatively. Other factors associated with a higher risk of losing ROM included a diagnosis of irreparable rotator cuff tear (P₁ = .038), rotator cuff arthropathy (P₁ = .017, P₂ ≤ .001), and inflammatory arthropathy (P₁ = .021). After multivariate analysis, higher preoperative FE (P₁ < .001), internal rotation (P₁ = -.018), and weight (P₁ = .008) remained significant predictors of loss of FE. Better preoperative FE (P₂ = .003), ER (P₂ < .001), and University of California at Los Angeles score (P₂ < .001) remained significant predictors of loss of ER. Patients who lost FE or ER were more likely to report lower satisfaction scores than their counterparts who did not lose ROM (P₁ < .001, P₂ < .001).

Patients with greater preoperative shoulder ROM or higher patient-reported outcomes are at higher risk of losing ROM after primary RTSA. They are also at higher risk of reporting lower postoperative satisfaction, though the majority were still satisfied. Surgeons should strongly counsel patients with well-preserved preoperative function on the risk of loss of ROM.