Load-sharing at the wrist following radial head replacement with a metal implant. A cadaveric study.

BACKGROUND Surgical excision of the radial head is frequently required after a comminuted fracture of the radial head. The outcome of this procedure is often unpredictable, with some patients experiencing ulna-sided pain in the wrist secondary to proximal migration of the radius. Insertion of a radial head prosthesis could prevent proximal radial migration and restore normal load-sharing at the wrist. The thickness of the radial head implant is an important variable in restoring anatomical radial length; however, the effects of varying the length of implants that were used to reconstruct the radius on load-sharing at the wrist have not been studied biomechanically, to our knowledge. METHODS A miniature load cell was attached to fifteen fresh-frozen cadaveric forearms to record force in the distal part of the ulna as the wrist was axially loaded to 134 N of compression force. Proximal displacement of the radius relative to the capitellum was also recorded. Loading tests on intact forearms were performed with the elbow in valgus and varus alignment and with three positions of wrist rotation (neutral, 45 degrees of pronation, and 45 degrees of supination). Loading tests were then repeated, with the same positions of varus and valgus elbow alignment and wrist rotation as had been used in the tests of the intact forearm, after radial head excision and subsequent insertion of metal radial head implants that restored anatomical length, implants that produced a radial length that was longer than the anatomical length, and implants that produced a radial length that was shorter than the anatomical length. Testing of these different implant thicknesses was repeated after sectioning of the interosseous membrane. RESULTS The mean distal ulnar forces and mean proximal radial displacements following insertion of an implant that restored anatomical length were not significantly different from the corresponding values for the intact forearm. At neutral wrist rotation, replacing that implant with an implant that increased the radial length by 4 mm (after sectioning of the interosseous membrane) decreased the mean distal ulnar force from 13.4% to 3.3% of the applied wrist force with the elbow in valgus alignment and from 29.1% to 8.6% with the elbow in varus alignment. Replacing the implant that restored anatomical length with one that decreased the length by 4 mm (after sectioning of the interosseous membrane) significantly increased the mean distal ulnar force from 13.4% of the applied wrist load to 33.3% with the elbow in valgus alignment and from 29.1% to 51.6% with it in varus alignment. The mean distal ulnar forces were not significantly affected by the position of wrist rotation when the elbow was in valgus alignment. However, when the elbow was in varus alignment, the mean distal ulnar forces associated with all reconstructed radial lengths were significantly higher when the wrist was placed in 45 degrees of supination. CONCLUSIONS In this cadaveric model, insertion of a metal implant maintained distal ulnar forces at normal levels, at all three positions of wrist rotation, when the radius had been restored to its original anatomical length. Distal ulnar forces and proximal radial displacements were significantly affected by the reconstructed length of the radius. CLINICAL RELEVANCE Radial head implants are utilized to prevent proximal migration of the radius as the wrist is loaded; this is especially important when the interosseous membrane has been ruptured and thus cannot help to limit radial displacement. At the time of surgery, comminution and displacement of a radial head fracture may make estimation of the original radial length difficult. Our results demonstrate that, in terms of distal ulnar loading, it is preferable to insert an implant that is too thick rather than too thin.