Study Design. Human cadaveric.
Objectives. Compare the accuracy of electromagnetic field (EMF)-based image-guided thoracic pedicle screw insertion to conventional techniques using anatomic landmarks and fluoroscopy.
Background. Image-guided surgical systems that aid in spinal instrumentation seek to minimize radiation exposure and improve accuracy. EMF image guidance was developed as an alternative to optical tracking to eliminate potential line of sight issues.
Materials and Methods. Four fresh-frozen human cadavers were randomly allocated into two groups. Pedicle screws were inserted from T1 to T12 using anatomic landmarks and fluoroscopy in group 1 and image guidance in group 2. Insertion and fluoroscopy time were recorded. Anatomic dissections were performed to assess screw placement.
Results. Image guidance placed 92% of thoracic pedicle screws safely, and conventional fluoroscopy placed 90% safely. The average degree of perforation was 2.4 mm with conventional fluoroscopy and 1.7 mm with image guidance (P = 0.055). Fluoroscopic time per screw was 5.9 seconds for conventional fluoroscopy and 3.6 seconds for image guidance (P = 0.045). Insertion time per screw was 4.35 minutes for conventional fluoroscopy and 2.98 minutes for image guidance (P = 0.007). However, when set-up time and image capture time were taken into account, the total insertion time per screw was not significantly different between the two groups.
Conclusions. Our study has shown that EMF image-guided thoracic pedicle screw placement results in a similar incidence of safely placed screws as does conventional fluoroscopy. When set-up time and image-capture time were factored in for image guidance, the average time to insert a pedicle screw was equal for both techniques. The use of EMF image guidance significantly reduced fluoroscopic time and thus radiation exposure per screw compared with conventional fluoroscopic techniques.
From *University of California San Francisco-Fresno, Fresno, CA;
2Naval Medical Center, San Diego, CA;
3Kwandong University, Korea;
4GE Surgical Navigation, Lawrence, MA; and
5SUNY Upstate Medical University, Syracuse, NY.
Acknowledgment date: November 19, 2002.
Revision date: February 21, 2003.
Acceptance date: May 1, 2003.
Address correspondence to H. Claude Sagi, MD, University Medical Center, Fourth Floor, Department of Orthopaedics, 445 South Cedar Avenue, Fresno, CA 93702; e-mail: firstname.lastname@example.org
Partial funding and assistance was provided by Visualization Technology, Inc. and Synthes, USA.