Page 342 - Read Online

Page 8 of 12 Maalouly et al. Mini-invasive Surg 2021;5:35 https://dx.doi.org/10.20517/2574-1225.2021.57

Figure 5. Axial CT scan showing: (A) Grade A on the left pedicle screw; and (B) Grade B on the right pedicle screw. CT: Computed
tomography.

DISCUSSION
Robotics in spinal surgery has been developing rapidly over the last decade, designed to augment and
enhance the surgeon’s abilities. Freehand placement of pedicle screws requires detailed knowledge of
anatomical landmarks and surgeon experience. The use of imaging technology such as fluoroscopy and
intraoperative navigation has improved the safety and accuracy of pedicle screw placement over freehand
[6]
[3]
techniques . The meta-analysis of 30 studies analyzing 9000 pedicle screws by Mason et al. concluded that
traditional fluoroscopy reached an accuracy of 63.1%, two-dimensional navigation had 84.3% accuracy, and
3D navigation was most accurate at 95.5% . Many new robotic spine guidance systems are being developed,
[7]
improving the safety and accuracy of pedicle screw placement. However, the learning curve of these systems
has not been studied in detail. There was insufficient evidence to conclude the effectiveness of robot-assisted
over conventional fluoroscopy-guided pedicle screw insertion in a systematic review by Marcus et al. . On
[8]
the contrary, another study reported equivalent accuracy with reduced radiation exposure in robot-assisted
cases compared to conventional fluoroscopy-guided surgery .
[9]

Multiple studies have reported on the accuracy of robot-assisted pedicle screws using a variety of
classification systems. In multiple systematic meta-analyses, it was concluded that the 2 mm incremental
classification system developed by Gertzbain and Robbins has been widely accepted and used for the
assessment of pedicle screw placement accuracy on CT scans [10,11,12] . Theologou et al. reported good inter-
[13]
observer reliability and commented on the ease of using this system. Hu et al. reported a high accuracy of
[14]
98.9% using the Renaissance system but used postoperative radiographs for the assessment which was a
major limitation of the study. Similarly, Pechlivanis et al. used the Renaissance system and reported that
[15]
91.5%-98.3% of screws were placed in an acceptable position as per the Gertzbein and Robbins classification
[16]
system. Keric et al. also reported a higher accuracy of 96.7% in the placement of 2067 screws using the
assessment system described by Wiesner et al. . In this study, the overall accuracy was 98% using the 2 mm
[17]
incremental system of CT scan. Out of 250 screws placed using the robot, five screws (2%) were revised due
to unsatisfactory placement immediately after fluoroscopic confirmation. These screws were considered
inaccurate. According to our assessment, 90.4% of screws (Grade A + B) were placed entirely inside the
pedicle and 9.6% of screws were acceptable, being Grade C. None of the screws had any major pedicle
breach or associated clinical symptoms. This was probably because the surgeon could modify the planned
trajectory if there was excessive force or skiving of the drill causing malposition of screws. This helped