INTRINSIC CONSTRAINT OF UNLINKED TOTAL ELBOW REPLACEMENTS—THE ULNOTROCHLEAR JOINT

Abstract

Background: Many unlinked total elbow replacement designs with radically differing articular geometries exist, suggesting that there is no consensus regarding an optimal design. A feature inherent to the articular design is the intrinsic constraint afforded to the joint by the implant. Our aim was to compare the intrinsic constraints of unlinked implants with that of the normal ulnotrochlear joint. Methods: We tested twelve cadaveric ulnotrochlear joints with a custom-made multiple-axis materials testing machine. With compressive loads ranging from 10 to 100 N, the joints were moved in either valgus or varus directions at 90° of flexion. The ulnotrochlear components from a single example of five medium-sized unlinked elbow replacements (Ewald, Kudo, Pritchard ERS, Sorbie-Questor, and Souter-Strathclyde) were also tested. The recorded measurements included the torques and forces, angular displacement, and axial displacement of the humerus relative to the ulna. Results: In general, the peak torque and the constraint ratio significantly increased with increasing compressive load for the implants as well as for the normal elbow. In valgus displacement, the Souter-Strathclyde implant had the highest and the Sorbie-Questor had the smallest peak torque and the Souter-Strathclyde had the highest and the Ewald had the smallest constraint ratio. In varus displacement, the Kudo had the highest and the Ewald had the smallest peak torque and constraint ratio. Conclusions: The constraint ratio is a characteristic that is useful for describing elbow joint behavior and for comparing the behavior of implants with that of the human elbow. Of the unlinked implants tested, the Souter-Strathclyde and Kudo prostheses most closely approximated the behavior of the human elbow joint. Implants that resemble the human elbow in appearance do not replicate normal behavior consistently, whereas other implants that do not resemble the human elbow closely do not deviate markedly from human behavior. Thus, much basic information about elbow form and function is needed to improve the performance of total elbow prostheses.