Quantification of glenoid dysplasia utilizing vault depth: a classification and treatment guide

Glenoid dysplasia (GD) is an infrequent cause of glenohumeral osteoarthritis. However, significant variability in the characteristics and severity of pathology may impact treatment choice and outcomes. The purpose of this study was to subclassify GD and determine if the severity of dysplasia affected the type of reconstruction and its clinical outcomes in our patient population.

This is a retrospective, single-surgeon cohort study where patients with GD (Walch C) who underwent primary anatomic total shoulder arthroplasty (aTSA) or reverse shoulder arthroplasty (rTSA) were identified. Their radiographs and axial computed tomography cuts were analyzed subjectively using retroversion and bone stock and classified as structural (Walch C1) or combined (Walch C2) glenoid deformity and graded as either mild or severe by at least 2/3 agreement and validated using reliability analysis. To quantify morphology, three-dimensional models were created from computed tomography scans. The primary outcome measure was American Shoulder and Elbow Surgeons score at final follow-up and was used to compare the type of arthroplasty (aTSA vs. rTSA) and severity grade. The threshold delineating mild from severe deformity was also assessed.

Forty-five patients (54 shoulders) were included. Median age was 66.5 years (interquartile range 59.0-72.1 years). The cohort comprised 27 aTSAs and 27 rTSAs with median follow-up of 58 months (interquartile range 33-84 months). Thirty-seven glenoids were classified as C1 and 17 as C2. There were 36 mild and 18 severe deformities. There was no significant difference between dysplasia groups by severity in median preoperative glenoid inclination (mild: 6.0° vs. severe: 7.1°, P = .353) or version (mild: 28.8° vs. severe: 36.0°, P = .060). Vault depth was less in the severe group (mild: 14.6 mm vs. severe: 7.6 mm, P < .001). No differences found in median American Shoulder and Elbow Surgeons score at final follow-up between severity groups (mild: 95.0 vs. severe: 93.5, P = .597) or arthroplasty types (aTSA: 92.6 vs. rTSA: 95.8, P = .529). The mild cohort comprised 23 aTSAs (63.9%) and 13 rTSAs, whereas the severe cohort comprised 4 aTSAs and 14 rTSAs (77.8%, P = .008).

GD can be differentiated into mild and severe deformity. aTSAs and rTSAs did not have significant differences in outcomes despite differences in severity. Vault depth differentiates mild and severe GD and provides a threshold that helps guide reconstruction type selection.