I'd like to thank my co-authors and AOSSM for allowing us to present our research. These are our disclosures. Though recognized as a risk factor, posterior glenoid bone loss has only recently been characterized and is distinctly different than anterior glenoid bone loss patterns. It is manifested by repetitive activities such as bench press, swimming, wrestling, and blocking in football with loading of the shoulder in a forward flexed and internally rotated position. Existing retrospective and biomechanical studies have identified a range of critical posterior bone loss values, but are limited by employment of anterior bone loss models, which differ in both orientation and morphology to posterior glenoid bone loss, which has been characterized as occurring at a 30-degree arc off of the long axis of the glenoid in a posterior-inferior direction and at a 26.8-degree slope relative to the glenoid fossa. Positioning in a single neutral arm position does not fully account for the contributions of the capsule ligamentous structures at various at-risk arm positions. We sought to evaluate the biomechanical effectiveness of a posterior labor repair in the setting of a clinically relevant bone loss model using three-dimensional CT modeling of patients with recurrent posterior shoulder instability in various at-risk positions. Ten fresh frozen cadaveric shoulders were prepared by removing all tissue except the capsule and distal rotator cuff insertions. The joint coordinate frame was established after the specimens were potted and mounted to a 6-degree of freedom robot. A posterior labral tear was created and then repaired with three horizontal mattress sutures through transosseous drill holes. Bone loss models were created off of a cohort of 3D CT data on patients who had undergone revision posterior labral repair surgery to develop clinically relevant models. Mean bone loss represented a posterior-inferior arc of 81.5 degrees, or 7% bone loss calculated off of a central angle geometric method. Large bone loss represented the 2-plus standard deviation, which was an arc of 139.4 degrees, or a calculated 28% bone loss. The bone loss models were created using a 3-millimeter burr manually to match the respective 3D printed templates. Specimens were tested in seven consecutive states to include native, posterior labral tear, and posterior labral repair, then with mean posterior glenoid bone loss or small bone loss with a labral tear and repair, and large posterior glenoid bone loss with a labral tear and repair. A 75-newton compressive force was placed on the glenohumeral joint to simulate rotator cuff compressive forces, while a 75-newton posterior-inferior force was placed on all specimens for testing. The specimens were tested at 60 and 90 degrees in both flexion and scaption, and we measured posterior-inferior translation, lateral translation, and peak dislocation forces. Labral repair alone, independent of bone loss state or arm position, resulted in a significant increase in stability, with a mean increase in peak dislocation force of 12.1 newtons across all arm positions. Peak dislocation force significantly decreased between no bone loss and small bone loss, and between small bone loss and large bone loss, regardless of labral state in all arm positions. To look at this graphically, there was a significant reduction in peak dislocation force and stability with a labral tear in the setting of both small and large bone loss. However, when the labral tear was repaired in the setting of small bone loss, it restored the shoulder to near native state. But when a labral repair was performed in the setting of large bone loss, a significant reduction in peak dislocation force resulted. When we looked at posterior-inferior translation, we found that independent of bone loss states, the humeral head translated by a mean of 1.71 millimeters. When we looked at lateral translation, predictably, it progressively decreased with increasing bone loss in all arm positions. However, labral repair increased lateral translation of the humeral head by a mean of 1.05 millimeters. When we looked at native state stability, the humerus was found to be most unstable in the plane of scaption at both 60 and 90 degrees. And the most unstable position of the shoulder was at 60 degrees of scaption with 29.9 millimeters of posterior-inferior translation. Study was not without limitations. All shoulders were tested with only static stabilizers as there was no rotator cuff activation. Bone loss models were created by a single surgeon, which may have resulted in error across the specimens. This was a cadaveric study with older average ages of the specimens and tissue compliance that is different than live human tissue. And we did not take into account glenoid version. In conclusion, independent of bone loss, posterior labral repair improves stability of the glenohumeral joint by increasing peak forces by 12.1 newtons and decreasing posterior-inferior translation by 1.71 millimeters. In small amounts of bone loss, 7% in this model, labral repair alone may be enough to restore shoulder stability in most individuals. However, larger bone loss states, somewhere between 20 and 30% or greater, may require bony augmentation for adequate stability. Thank you.

characterization

biomechanical effectiveness

posterior labral repair

posterior glenoid bone loss

cadaveric shoulders