Morphological analysis of the scapula in healthy and osteoarthritic subjects.

BACKGROUND: While glenoid morphology in primary glenohumeral osteoarthritis (OA) has been extensively studied, less is known about differences in global scapular morphology. The purpose of this study was to identify three-dimensional morphologic differences of the scapula between healthy shoulders and shoulders with primary osteoarthritis, and between Walch type A and type B glenoids.

METHODS: We retrospectively analyzed 619 shoulders, including 134 healthy controls older than 60 years and 485 shoulders with primary osteoarthritis (133 type A, 352 type B). Automated three-dimensional analysis measured glenoid version, inclination, and rotation; humeral subluxation; critical shoulder angle (CSA); sagittal acromial tilt (SAT); posterior acromial coverage (PAC); posterior acromial height (PAH); and scapular spine angle (SSA). Group comparisons were performed between healthy and osteoarthritic scapulae and between Walch subtypes.

RESULTS: All morphological parameters differed significantly between healthy and OA scapulae (p < 0.05). Several parameters demonstrated graded differences across healthy, type A, and type B groups, including CSA, SAT, PAC, PAH, glenoid version, and humeral subluxation (all p < 0.001). Within the osteoarthritic cohort, correlations between acromial or scapular spine parameters and glenoid orientation or humeral subluxation were generally weak to moderate; only glenoid version was strongly correlated with posterior humeral subluxation. Glenoid inclination, glenoid rotation, and SSA differed between healthy and OA scapulae but did not differ between type A and type B glenoids, indicating morphologic features associated with osteoarthritis independent of posterior humeral subluxation.

CONCLUSION: Primary glenohumeral osteoarthritis is associated with distinct global scapular morphologic characteristics. Scapular spine angle, glenoid inclination, and glenoid rotation differ between healthy and osteoarthritic shoulders regardless of Walch subtype, whereas acromial parameters vary progressively with increasing posterior humeral subluxation, consistent with a shorter, flatter, and higher posterior acromion. These findings are associative and hypothesis-generating, and future longitudinal studies are required to determine whether these morphologic differences contribute to disease development or progression.