The wrist has often been modelled as an orthogonal universal joint, based upon past kinematic investigations. These past planar analyses did not provide enough information to completely characterize a universal joint, however. This study was an attempt io determine whether an axis of flexion-extension was fixed to the forearm, and an axis of radio-ulnar deviation was fixed to the hand, or whether this situation was reversed, thereby completing the description of the orthogonal universal joint model of the wrist.

Three dimensional motion data of planar flexion-extension, planar radio-ulnar deviation, and unconstrained circumduction motions were collected from the dominant hand of six healthy right-handed volunteers. Helical screw axes of the planar motions were determined and were used to construct two possible universal joint mathematical models of the wrist. These models were then fit to the unconstrained circumduction motion data, and a measure of the closeness of fit was evaluated for each model. This same method provided a simple and clear method for describing the position and orientation of the hand with respect to the forearm.

The universal joint model consisting of a radio-ulnar deviation axis fixed to the forearm, and a flexion-extension axis fixed to the hand was the best orthogonal universal joint model for describing wrist kinematics. However, it was found that this one model did not fit well for the complete range of wrist motion, and that the kinematics of the human wrist are not as simple as a universal joint.

This information has implications for wrist arthroplasty design. An accurate model of wrist motion would also be useful in kinetic modelling of wrist motion, and may aid in better understanding various pathologies of the wrist.