Increased glenoid baseplate retroversion improves internal rotation following reverse shoulder arthroplasty

Internal rotation after reverse total shoulder arthroplasty is often unchanged or minimally improved. The primary purpose of this study was to investigate the effects of glenoid baseplate version on postoperative internal rotation. The secondary purpose to investigate the effects of baseplate retroversion on external rotation (ER) and patient-reported outcomes (PROs).

A retrospective review was performed on a prospectively maintained multicenter database of patients who underwent primary reverse shoulder arthroplasty using a 135° humeral prosthesis and lateralized glenoid with minimum 2-year clinical follow-up. Preoperative and postoperative radiographs were reviewed by 2 independent observers who assessed preoperative glenoid version and postoperative glenoid baseplate version. Patients were stratified by postoperative retroversion (<10°, 10°-19° or >20°) and change in version from preoperative to postoperative (ΔRV). Primary outcomes were internal rotation with the arm at 90° (IR90) and internal rotation estimated to nearest spinal level (IRspine). Secondary outcomes were active ER in adduction (ER0), active ER with arm at 90° (ER90), forward flexion (FF), and PROs. Linear regression analyses and 1-way analysis of variance analyses were used for comparisons.

Two hundred seventy-four patients with a mean of 71 years of age were included in the study. Patients with >10° of postoperative baseplate retroversion gained 20° of IR90 (P = .005) without loss of ER90 (P < .001) compared to patients with <10° of baseplate retroversion. More than 10° of postoperative baseplate retroversion was associated with significantly improved Constant-Murley scores (41.5, P = .007) and Single Assessment Numeric Evaluation scores (45.4, P = .047) compared to patients with less than <10° of baseplate retroversion. Patients with a ΔRV increase of >10° had significantly improved IR90 (P = .031) without loss of ER90 (P = .019). There was no correlation between ΔRV and IRspine, ER0 or FF, or PROs.

With a 135° and lateralized glenoid, postoperative baseplate retroversion of >10° was associated with significantly improved IR90, ER90, Constant-Murley, and Single Assessment Numeric Evaluation scores at 2-year follow-up compared to <10° retroversion. Additionally an increased ΔRV from preoperative to postoperative appears to improve IR90 without limiting ER0 or FF. While baseplate retroversion does not improve IRspine, overall function appears to be improved and therefore consideration may be given to accepting retroversion or intentionally retroverting the baseplate if fixation allows.