Is Premorbid Glenoid Anatomy Altered in Patients with Glenohumeral Osteoarthritis?

Abstract

Restoration of native, premorbid glenoid anatomy may be a goal in component placement during total shoulder arthroplasty. However, if patients with unilateral glenohumeral osteoarthritis are predisposed to the development of arthritis owing to abnormal native glenoid anatomy, this recommendation may be inappropriate. The purpose of this study was to determine if patients with glenohumeral osteoarthritis have abnormal premorbid glenoid version or inclination, thereby predisposing them to subsequent glenoid disorders. We specifically tested whether: (1) premorbid glenoid version or inclination in the pathologic shoulder of patients with unilateral osteoarthritis, as determined by the glenoid vault model, is different from glenoid version or inclination in the contralateral nonpathologic shoulder of these patients; (2) there are differences between glenoid version or inclination in normal cadaver shoulders and the nonpathologic side of patients with unilateral osteoarthritis; and (3) there are differences between glenoid version or inclination in normal cadaver shoulders and the premorbid glenoid version and inclination in the pathologic shoulder of patients with unilateral osteoarthritis, as determined by the glenoid vault model. Bilateral CT scans were obtained in 27 patients with unilateral glenohumeral osteoarthritis. Thirty normal cadaver control shoulders also underwent CT scans. Premorbid glenoid version and inclination in the pathologic shoulder, as measured by the glenoid vault model, were compared with the contralateral nonpathologic shoulder and the normal cadaver control shoulders. Glenoid version and inclination of the normal shoulders were compared with the nonpathologic side from patients with unilateral osteoarthritis. Measurements were made by two different methods using three-dimensional surgical simulation software: (1) a direct measurement technique and (2) measurements derived from placement of a glenoid vault model. Mean differences in these parameters were compared between shoulder groups using paired and unpaired Student's t-tests. Premorbid glenoid version and inclination in the pathologic shoulder as measured by the vault model averaged -7° (SD, 5) and 10° (SD, 6), respectively, compared with -7° (SD, 5) and 12° (SD, 6) as directly measured on the nonpathologic side, and -7° (SD, 4) and 12° (SD, 5) as directly measured in the normal cadaver control shoulders. There were no differences in glenoid version or inclination between the normal shoulders and the nonpathologic side of patients with unilateral osteoarthritis or between these shoulders and the premorbid version and inclination of the arthritic shoulder as measured by the vault model. Patients with glenohumeral osteoarthritis do not appear to have abnormal premorbid glenoid retroversion or inclination. The glenoid vault model can be used to determine premorbid glenoid version and inclination. The glenoid vault model may be a clinically useful tool to estimate patient-specific premorbid glenoid anatomy, which may help in preoperative or intraoperative surgical planning for total shoulder arthroplasty.