A Cadaveric Study Validating in vitro Monitoring Techniques to Measure the Failure Mechanism of Glenoid Implants against Clinical CT

Definite glenoid implant loosening is identifiable on radiographs, however, identifying early loosening still eludes clinicians. Methods to monitor glenoid loosening in vitro have not been validated to clinical imaging. This study investigates the correlation between in vitro measures and CT images. Ten cadaveric scapulae were implanted with a pegged glenoid implant and fatigue tested to failure. Each scapulae were cyclically loaded superiorly and CT scanned every 20,000 cycles until failure to monitor progressive radiolucent lines. The superior and inferior rim displacements were also measured. A finite element (FE) model of one scapula was used to analyse the interfacial stresses at the implant/cement and cement/bone. All ten implants failed inferiorly at the implant-cement interface, two also failed at the cement-bone interface inferiorly, and three showed superior failure. Failure occurred at of 80,966 ± 53,729 (mean ± SD) cycles. CT scans confirmed failure of the fixation, and in most cases, was observed either before or with visual failure, indicating its capacity to detect loosening earlier for earlier intervention if needed. Significant correlations were found between both increasing inferior rim displacement (ASTM standard F2028-14), increasing vertical head displacement and failure of the glenoid implant. The FE model showed peak tensile stresses inferiorly and high compressive stresses superiorly, corroborating experimental findings. Similar failure modes have been cited in clinical and in vitro studies. In vitro monitoring methods correlated to failure progression in clinical CT images.
Clinical Significance: The study highlights failure at the implant-cement interface and early signs of failure are identifiable in CT images.