Influence of preoperative rotational shoulder stiffness on rate of motion restoration after anatomic and reverse total shoulder arthroplasty for glenohumeral osteoarthritis with an intact rotator cuff

Abstract

BackgroundA subset of patients undergoing anatomic and reverse total shoulder arthroplasty (aTSA and rTSA) lag behind their peers in regaining overhead range of motion (ROM) after surgery. The primary purpose of this study was to compare the rate of recovery of ROM after aTSA and rTSA performed in stiff (preoperative passive external rotation [ER] ≤0 °) versus non-stiff (preoperative passive ER >0°) shoulders with RCI-GHOA. HypothesisWe hypothesized that preoperatively stiff shoulders (preoperative passive ER ≤0 °) would have slower recovery in ROM postoperatively with lower postoperative motion compared to non-stiff shoulders. Methods and materialsA retrospective review of a multi-institution shoulder arthroplasty database was performed between 2001 and 2021. We identified 1,164 aTSAs and 539 rTSAs performed for RCI-GHOA with a minimum of 2-year clinical follow-up along with follow-up between 3–6 months and a third visit at any other time point. Primarily, the rate of recovery in ROM and time to maximum ROM was evaluated. Secondarily, we assessed six outcome scores and the influence of subscapularis repair during rTSA. Recovery in each outcome was modeled using continuous two-phase segmented linear regression models with random effects. Rate of recovery was defined as the slope of the first segment. Patients were considered to have recovered after surgery at the timepoint corresponding to the inflection point between piecewise segments. ResultsOf the 1,164 aTSAs and 539 rTSAs included, 172 aTSAs (15%) and 80 rTSAs (15%) were stiff preoperatively, respectively. Compared to preoperatively stiff aTSAs, non-stiff aTSAs regained ER, abduction, internal rotation (IR), and forward elevation (FE) faster over a shorter duration. Similarly, non-stiff rTSAs regained ER, abduction, and FE faster and over a shorter duration compared to stiff rTSAs, but regained IR more slowly over a longer duration. Stiff rTSAs performed with subscapularis repair did not have any appreciable gain in ER after the immediate postoperative period. Although non-stiff and stiff rTSAs performed without subscapularis repair regained ER at a similar rate (4.4 vs. 4.2 °/month), stiff rTSAs continued to regain ER 1.9-times longer (11.9 vs. 6.4 months). When the subscapularis was repaired, non-stiff rTSAs regained abduction and IR faster over a short duration compared to stiff rTSAs. ConclusionsPreoperative stiffness is associated with slower recovery of active ROM over a longer duration in patients undergoing shoulder arthroplasty for RCI-GHOA. Level of evidenceIII; Retrospective Cohort Comparison; Treatment Study.