A three dimensional model-based simulation of the wrist joint during radioulnar deviation

Abstract

The primary objective of this study was to numerically simulate the carpal bone kinematics during radioulnar deviation of the wrist. Based on computed tomography (CT) images of the human arm, we constructed a biomechanical model comprising the arm and wrist (humerus, ulna, radius, and the bones of the hand). The carpal bones were fused into a single functional unit, while the humerus, ulna, and radius were rigidly fixed, allowing only the wrist joint to move. Further, the ligaments connecting the radius and the ulna to the carpal bones were approximated by wire models. In this study, five palmar and four dorsal ligaments were included. For the muscles controlling radial deviation, the flexor carpi radialis and extensor carpi radialis longus were selected. For the muscles responsible for ulnar deviation, the flexor carpi ulnaris and extensor carpi ulnaris were selected. These muscles were represented with muscle spring models. Two types of muscle spring models were employed: a linear muscle spring model and tendon-modulated muscle spring model. In the linear muscle spring model, the points of muscle origin and insertion were linked with linear springs. These springs simulated muscle stretching and relaxation, without allowing for deformations. The tendon-modulated muscle spring model comprised elements simulating contractible the muscle fibers and elastic tendons that transmitted muscle force to the bone. To compare with the simulation results, CT images of the human wrist joint were obtained at different radioulnar deviation angles. The results of having carried out radioulnar deviation in both models, the linear muscle spring model yielded a maximum ulnar deviation angle of 10° and a maximum radial deviation angle of 3.5°. The tendon-modulated muscle spring model provided a maximum ulnar deviation angle of no less than 30° and a maximum radial deviation angle of no less than 20°. We used the tendon-modulated muscle spring model to estimate the translation of anatomic landmarks (the trapezium and scaphoid) at different degrees of radioulnar deviation, and these simulation results were compared with the human imaging data. For the trapezium, the traveled distances were comparable, whereas the results were incongruent for the scaphoideum. These discrepancies were probably attributable to the improper modeling of the proximal carpal row movement associated with ulnar deviation.