Medial Position of the Bone Block by Computed Tomography Assessment Is Associated With Excessive Bone Resorption After the Modified Eden-Hybinette Procedure for Recurrent Anterior Shoulder Instability.

PURPOSE: To quantitatively evaluate the effects of initial medial-lateral position of bone block on bone remodeling.

METHODS

Patients who underwent the modified Eden-Hybinette procedure for the treatment of recurrent anterior shoulder instability between May 2017 and July 2022 were included. Follow-up of at least 2 years and computed tomography scans preoperatively, immediately after surgery, and at a minimum of 12 months after surgery were necessary for inclusion. The medial-lateral as well as the superior-inferior position of bone block after surgery and diameter, depth, version, and surface area of the glenoid at the 3 time points were measured. Bivariate analyses and multivariate logistic regression analyses were used to determine variables on excessive bone resorption, which was defined as a final glenoid ratio less than 100%. The cutoff value was confirmed using receiver operating characteristic curve.

RESULTS

Excessive bone resorption was detected in 17.7% (11/62) of patients. Medial-lateral position of the bone block, glenoid depth and its increment immediate after surgery differed significantly between patients with and without excessive bone resorption. Multivariate logistic regression analyses revealed that excessive bone resorption was significantly correlated with medial-lateral position of bone block (odds ratio 0.191; P = .002). The cutoff value was -0.5 mm (area under the curve, 0.934; 95% confidence interval 0.870-0.998; P < .001; sensitivity, 84%; specificity, 100%). In total, 81.8% (9/11) of cases with excessive bone resorption were detected with graft surfaces distant from the extrapolated glenoid curvature.

CONCLUSIONS

Medial-lateral position of bone block was a critical determinant for bone remodeling after glenoid reconstruction and excessive bone resorption was associated with too medially placed graft. To achieve a more stable glenohumeral joint with physiologically pear-shaped glenoid surface after remodeling, the graft placing at no more than 0.5 mm medial to bony surface of the glenoid as well as with a well-matched articular surface to the extrapolated glenoid curvature was proved to be preferred.

LEVEL OF EVIDENCE: Level IV, retrospective case series.