Glenohumeral joint translations before and after total shoulder arthroplasty. A study in cadavera

The purpose of the present study was to examine the motions of natural and prosthetically reconstructed glenohumeral joints without capsular contracture in the laboratory to obtain a better understanding of joint motion as it may relate to failure of the implant. Seven joints from fresh-frozen human cadavera were instrumented with a six-degrees-of-freedom magnetic tracking device to study patterns of translation and rotation with and without the application of simulated muscle forces (active and passive models). The specimens were tested before and after reconstruction of the joint with use of modified operative techniques and implants that had from zero to five millimeters of radial mismatch between the humeral head and glenoid components. The natural and reconstructed joints had similar patterns of translation, with larger rotations and translations observed in the passive model. On the average, the active translations of the natural joints were best reproduced by the reconstructed joints with less conforming articulations. The mean active translation in the natural joints was 1.5 millimeters along the anterior-posterior axis and 1.1 millimeters along the superior-inferior axis. The active translations in the reconstructed joints were observed to increase consistently as the conformity of the components decreased: the mean active anterior-posterior translations ranged from 0.3 millimeter for conforming components to 1.7 millimeters for components with a five-millimeter radial mismatch, and the mean active superior-inferior translations ranged from 0.4 to 1.1 millimeters. The patterns of translation during passive motions were less consistent, presumably because of the influence of capsular ligaments and because the translations were large enough for the articular surface of the humeral head to lose contact with the articular surface of the glenoid component.

Glenohumeral translations during active motions were found to depend on articular conformity, which indicates that the choice of implants may have important consequences for normal motions. Totally conforming designs may impose a degree of restraint to translations that is higher than that in the natural joint. The patterns of translation may influence the longevity of the joint replacement with respect to stability, loosening of the glenoid component, and wear of the components.