Abstract
BACKGROUND: Glenoid implant loosening is the most common complication of anatomical total shoulder arthroplasty. It is caused by high glenohumeral shear forces and by an eccentric loading of the glenoid implant provoking its "rocking". The critical shoulder angle (CSA) varies with the glenoid inclination and the acromion length. A higher CSA has been correlated with earlier radiological signs of glenoid loosening. However, the reliability of the CSA in predicting the risk factors of glenoid loosening has yet to be determined since the same CSA can result from multiple scapular anatomies. METHODS: An inverse-dynamic musculoskeletal model in Anybody Modeling System of the shoulder with anatomical implants allowing glenohumeral translations was used. The acromion length and the glenoid implant inclination were varied to create multiple CSA configurations. Muscle forces, the force, and the moment applied to the glenoid implant were simulated during a shoulder abduction to compare the risks of glenoid loosening. RESULTS: Increasing the CSA with an upward-tilted glenoid and a longer acromion led to more eccentric forces applied to the glenoid. The moment and shear applied to the glenoid implant increased with a higher CSA and were minimal for the smaller CSAs. Depending on the combination of inclination and acromion length, the shear and the moment were highly variable for the same CSA. CONCLUSION: Measuring the CSA as a global indicator may be insufficient to accurately predict the risk of glenoid loosening. It suggests that the acromion length could be considered during surgical planning to determine the adequate glenoid implant inclination.