Philips’ new augmented reality system has been used in clinical practice for the first time at the Karolinska University Hospital, Solna, Sweden.
“This new technology allows us to use augmented reality in combination with 3D imaging for intraoperative surgical planning and navigation of our devices. We have now treated four patients using the system and placed 44 pedicle screws with satisfactory results. We have been able to check the overall result in 3D in the operating room without the need to move the patient to a CT scanner. The radiation dose to the staff is zero and minimal to the patient.” said Adrian Elmi-Terander, principal investigator at Karolinska University Hospital, Solna, Sweden. “We tested this workflow pre-clinically for complex thoracic and cervical spine surgeries with very convincing results and look forward to extending it to complex cerebral neurosurgical procedures.”
A pre-clinical study published in Spine has demonstrated superior accuracy for thoracic pedicle screws placement using an augmented-reality navigation system (Philips) in comparison to those placed freehand.
The system consists of four high resolution optical video cameras mounted to in the frame of a motorised ceiling-mounted C-arm flat detector (Allura Clarity FD20, Philips Healthcare). Between eight and 10 markers were placed on the skin around the surgical site on each of four human cadavers for tracking by the system.
Each cadaver was placed in the prone position and opened along the whole dorsal thoracic region, exposing the spine “down to the spinous processes and tips of the transverse processes”. Pedicle screws were inserted unilaterally using the augmented-reality system, following freehand placement on the contralateral side by a different surgeon.
Postoperative 3D cone beam computed tomography (CT) images were evaluated by two independent neurosurgeons and two radiologists. Using Gertzbein grading, the observers found grade zero placement (without cortical breach) 71% more frequently with the augmented reality system then with placement by freehand method. Significantly (p<0.05) fewer incidences of grade three placement (>4mm breach, severe displacement) were found with augmented-reality placement. Difference between rates of grades one (0-2mm breach, minor perforation including cortical encroachment) and two (2-4mm breach, moderate breach) breaches were not significantly different. Excellent interrater agreement was observed upon statistical analysis.
The team’s research continues, with published results expected in Autumn, following study completion in June. Elmi-Terander told Spinal News International that the Karolinska team is continuing research and development of the system in collaboration with Philips, with applications in cervical and cranial surgery expected in the near future. According to Elmi-Terander, at least four further clinical trials have been planned.
