Inset glenoid component stability and bone removal by implantation depth

Aseptic loosening of on-lay all-polyethylene cemented glenoid components in total shoulder arthroplasty is a leading cause of revision surgery and may be accelerated by the rocking-horse phenomenon. In response to this, inset glenoid implants have been developed in hopes that they will lead to lower loosening rates. Presently, little literature exists on the ideal depth of insetting to minimize micromotion while also considering glenoid bone preservation. This study, therefore, evaluated a generic circular inset glenoid component implanted at 4 depths in osteoarthritic glenoids using finite element models. Micromotion under simulated joint loading, bone removal, and underlying bone density were compared. The goal was to determine an optimal inset depth that minimizes micromotion while preserving glenoid bone.

Finite element models of 7 male osteoarthritic scapulae were generated from pre-operative computed tomography scans. Circular inset glenoid components were virtually implanted at 4 depths: 25%, 50%, 75%, and 100% (inlay). Glenohumeral joint loading was simulated in 5 directions. Tangential (parallel) and normal (perpendicular) micromotions were measured across 6 backside regions of the component for each loading direction. The volume of bone removed for implantation and bone density within a 5-mm depth region beneath each component were also evaluated.

No significant relationship was found between inset depth and tangential micromotion across load directions or locations (P > .05). The 25% depth showed the greatest median tangential micromotion in 12 of 42 cases (29%), the 50% and 100% depths in 11 cases each (26%), and the 75% depth in 8 cases (19%). Similarly, no significant relationship was observed between inset depth and normal micromotion. The 25% depth had the greatest median normal micromotion in 15 of 42 cases (36%), followed by 75% in 13 cases (31%), 50% in 8 cases (19%), and 100% in 6 cases (14%).Bone removal increased significantly with increasing inset depth (P < .05). On average, the 100% depth required approximately 4 times more bone removal than the 25% depth. Bone density within the 5-mm depth region beneath the component was highest at the 25% depth. Significant differences were observed between the 25% depth and the 50% (P = .03), 75% (P = .03), and 100% depths (P = .049), while no significant differences were found among the deeper depths.

Inset depth was not significantly associated with glenoid component micromotion, although a trend toward reduced micromotion with greater depth was observed. However, deeper insetting required substantially more bone removal and was associated with lower underlying bone density. Considering implant stability, bone preservation, and supporting bone density, these findings support glenoid component insetting at approximately 25-50% depth.