Abstract
Purpose
The surgical treatment of scaphoid fractures consists of reduction of the fracture followed by stable internal fixation using a headless compression screw. Proper positioning of the screw remains technically challenging and therefore computer assisted surgery may have an advantage.
Navigation assisted surgery requires placement and registration of stable reference markers which is technically impossible in a small bone like the scaphoid. Custom made wrist-positioning devices with built-in reference markers have been developed for this purpose. The purpose of this study was to evaluate a different method of navigation assisted scaphoid fracture fixation. Temporary stabilisation with a pin of the scaphoid to the radius enables placement of the reference markers on the radius. Our hypothesis was that this method will achieve precise fracture fixation, superior to the standard free hand technique.
Methods
In 20 identical saw bone models with mobile scaphoids, the scaphoid was stabilised to the radius using one Kirschner wire (KW). An additional KW representing the fixating screw was placed either using the Mazor Renaissance Robotic System (MAZOR Surgical Technologies, Israel) or standard free hand technique. CT scans were performed prior to fixation and after fixation in order to plan the location of the KW and compare this planned location with the final result.
Results
No significant difference was found between the measures of KW location between groups and in comparison with the planned location, including entry and exit points of the KW, length of KW through the scaphoid (mean axis length of 28.7 mm [SD 1.5] with the robot system versus 29.6 mm [SD 2.1] with the free hand technique) and difference in angle of fixation with the planned axis of fixation (mean of 1.7 degrees [SD 5.5] with the robot versus 3.8 degrees [SD 5.6] free hand). Significant differences were found between exposure to radiation (mean of 0.07 Rad [SD 0.04] with the robot system versus 13.9 Rad [SD 18.4] with the free hand technique; p=0.04) and the number of attempts in placing the KW (mean of 1.1 attempts [SD 0.32] with the robot versus 8 attempts [SD 6.65] free hand; p=0.01).
Conclusion
Computer assisted fixation of a scaphoid fracture was found to be as accurate as the free hand technique, after fixation of the scaphoid to the radius, without the need for a custom splint. It was also shown to be superior by decreasing the exposure to radiation and number of attempts of KW placement.
