P119. Are biplanar projection radiographs sufficient to confirm intraoperative pedicle screw placement?

Abstract

<h3>BACKGROUND CONTEXT</h3> Two-dimensional BPR (AP and lateral fluoroscopy or x-ray) has been the gold standard in spine fusion surgery for many years, however, 3D intraoperative imaging systems are now available. When utilizing BPR for pedicle screw confirmation, significant misplacement rates have been reported in the lumbar and thoracic spine. <h3>PURPOSE</h3> Biplanar projection radiographs (BPR) for confirmation of intraoperative pedicle screw placement are inferior to three-dimensional fluoroscopic (3D) imaging. <h3>STUDY DESIGN/SETTING</h3> Retrospective review of prospectively collected data. <h3>PATIENT SAMPLE</h3> A total of 100 patients underwent degenerative spinal surgery by one spine surgeon. <h3>OUTCOME MEASURES</h3> Screw placement was compared between postoperative CT, intraoperative x-rays( BPR) and intraoperative 3D fluoroscopic imaging. <h3>METHODS</h3> One hundred patients underwent degenerative spinal surgery by one spine surgeon and screw placement was compared from preoperative CT images to intraoperative 3D fluoroscopy. The M-line and Gertzbein-Robbins system (GRS) was used to confirm false positives or negatives in BPR. M-line was defined as a line connecting the upper and lower spinous processes through the fixed vertebrae within BPR. GRS measures pedicle screw accuracy and categorizes positions into 4 groups based on a clinical positioning grading used in 3D imaging. <h3>RESULTS</h3> A total of 604 instrumented pedicles in 100 patients were planned robotically, 497 (82.2%) were placed robotically. Each patient received a postoperative CT, intraoperative X-rays (BPR), and intraoperative 3D fluoroscopic imaging. M-Lines were measured for each patient vias intraoperative X-ray postoperatively. Patient postoperative CTs were compared to intraoperative 3D fluoroscopy by the GRS method. Postoperative CTs were then compared to M-line findings in BPR, to confirm whether false positives or negatives were detected. Based on BPR imaging, 5 (5%) patients exhibited 5 (0.8%) screws which were all robotically placed screws crossing the M-line in BPR. However, when compared to their postoperative CT imaging, breach was not detected and was classified as fully within the pedicle (category A of GRS). A BPR in a separate patient (1%) did exhibit one screw (0.2%) with a false negative BPR compared to 3D imaging. This could be possibly due to anatomical rotation in BPR. <h3>CONCLUSIONS</h3> A clinically significant number of patients had false-negative and false-positive screw confirmation on BPR. In 5% of patients and 0.8% of screws, a false-negative breaching within robotically placed screws was seen. Furthermore, 1% of patients and 0.2% of screws showed a false negative of pedicle screw mal-positioning. BPR may yield a false sense of security for misplaced screws and conversely may raise concern over screws that are in a satisfactory position. These findings may make 3D imaging a more desirable post-instrumentation gold standard. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.