Abstract
Background: The use of sacro-iliac screws to restore the stability of posterior pelvic ring dissociations has become a standard technique. Several methods are described including fluoroscopy, CT and computer assisted techniques. Fluoroscopically assisted insertion is the standard technique. Multiple exposures of ionising radiation permit acquisition of a target in sequential planes, requiring a process of interpolation for 3-D localisation. A computer assisted technique facilitates the simultaneous visualisation of multiple planes following a single image intensifier acquisition and registration process in each plane. The purpose of this study is to demonstrate the accuracy of a computer assisted surgery technique and quantify the predicted reduction in radiation exposure.
Methods: 10 embalmed human cadavers were used. In each specimen, a sacro-iliac screw was simulated by passing a 5mm reamer over a 3.2mm guide wire. The index track was formed with a closely sheathed 4.8mm drill and was inserted with the standard fluoroscopically assisted technique in the left hemipelvis and a computer assisted technique(Vector Vision trauma) on the right. Registration of the system is achieved by the placement of infra red reflective arrays on all tracked objects. These include the right and left hemi-pelvis, the fluoroscope, the drill guide and the driver unit. The system is an open platform which registers arrays of known geometry whilst permitting the registration of instruments by means of an instrument calibration matrix. The pelvic T and Y pattern fiducials are fixed rigidly to inter-table threaded pins at the level of the gluteal tuberosities. The standard acquisition projections are inlet and outlet views for both techniques with the addition of lateral projections although the latter were not directly used for navigation. Postoperative CT scans demonstrate the actual tracks and analysis is facilitated by means of a digital mapping technique.
Results: The accuracy of the fluoroscopically assisted and computer assisted techniques is identical. The procedure time was significantly longer for the computer assisted technique although most of the additional time was accounted for by the “line of sight” registration process. There was a reduction in both the total screening time and the measured radiation dose in the case of the computer assisted technique although this did not reach statistical significance as the sample size is relatively small.
Conclusions: The navigation of sacro-iliac screw tracks by means of both fluoroscopically assisted and computer assisted techniques proves equally accurate in a human cadaveric model.
Theses abstracts were prepared by Professor Roger Lemaire. Correspondence should be addressed to EFORT Central Office, Freihofstrasse 22, CH-8700 Küsnacht, Switzerland.
