Static posterior humeral head subluxation and total shoulder arthroplasty

Static posterior subluxation of the humeral head (PSH) is often associated with glenohumeral arthritis. It may persist following total shoulder arthroplasty (TSA) and lead to accelerated polyethylene wear and glenoid component loosening. The factors which lead to PSH are poorly understood. The purpose of this study was to test the hypothesis that operative correction of glenoid version during shoulder arthroplasty re-centers the glenohumeral joint; therefore, glenoid replacement may be considered even in cases of osteoarthritis associated with posterior humeral head subluxation.

Thirty-three of 124 (27%) consecutive shoulders undergoing primary TSA had static preoperative PSH with a subluxation index of at least 65% determined on standardized computer tomographic scans. Twenty-three of these 33 shoulders were available for clinical and computed tomography follow-up after a minimum of 24 and average of 42 months. Mean preoperative glenoid retroversion was -18 [range, 0 degrees - (-40 degrees)], the subluxation index averaged 71% (range, 65-81%). Glenoid morphology, according to Walch et al, was type B1 in 9 patients, type B2 in 5 patients, and type C in 9 patients. A conventional total shoulder replacement was performed through a deltopectoral interval. Using corrective glenoid reaming, restoration of glenoid version to between 0 degrees and 10 degrees of retroversion was attempted in addition to standard soft tissue release. Humeral head retroversion was replicated from the diseased humeral head as closely as possible.

PSH was reversed in 21/23 patients following TSA with an average final subluxation index of 50% (range, 40-68%; P = .001). There was no significant correlation statistically between PSH and preoperative or postoperative glenoid version, humeral torsion, glenoid morphology, or acromio-humeral distance. Mean absolute Constant scores improved from 39 to 78 points, age-adjusted Constant scores improved from 49% to 95% and subjective shoulder values improved from 40% to 89%, which were all statistically significant (P < .0001).

PSH is frequently present in shoulders with osteoarthritis. It can be corrected in the majority of shoulders undergoing total shoulder replacement; however, re-centering is not correlated with glenoid version or its correction.

Level 4; Case series, treatment study.