Novel Augmented Reality System: Accuracy of Pedicle Screw Placement with Two Different Cone Beam CT Scanners

Introduction: Augmented Reality (AR) offers improved hand eye coordination with the potential to improve surgical accuracy in spinal surgery. We evaluated a novel AR guidance system, assessing the accuracy of pedicle screw placement using different intra-operative 3D scanners.

Methods: Pedicle screws were implanted in thoracic and lumbar spines using a novel AR system (OnPoint AR-AI, OnPoint Surgical, Bedford, MA): 80 screws, 25 open, 55 MIS implanted with the 1st cone beam CT scanner (O-arm, Medtronic); 10 screws, 5 open, 5 MIS implanted with the 2nd cone beam CT scanner (GE OEC 3D, GE Healthcare). Heary grading was assessed by a radiologist for degrees of pedicle breaches. 3D measurements of positional error (PE) and angular error (AE) were performed by comparing post-operative CT scans of screws with planned trajectories. Statistical analyses compared PE and AE using an unequal variance t-test method.

Results: PE and AE for the 1st cone beam CT scanner were 1.1 mm (1 SD 0.5 mm) and 1.3 degrees (1 SD 0.5 degrees), respectively. PE and AE for the 2nd cone beam CT scanner were 1.2 mm (1 SD 0.2 mm) and 1.1 degrees (1 SD 0.3 degrees), respectively. There was no statistically significant difference in screw placement accuracy between the two scanners.

Conclusion : PE and AE for placing pedicle screws are statistically significantly smaller for this novel AR system compared to published data for currently available navigation, robotic and AR systems. A positional error of 1.1 – 1.2 mm and angular error of 1.1 - 1.3 degrees makes this technology particularly suited for small pedicles and challenging anatomy, as is frequently encountered in thoracic and cervical spine.