Darren R. Lebl, MD [Credit: Hospital for Special Surgery]
“Compared to conventional free-hand techniques, both robotic-assisted navigation and the use of augmented reality have demonstrated superior accuracy,” said Darren R. Lebl, MD, MBA, director of research at HSS, and principal investigator of the study, in a press release. “Our study is the first to directly compare RAN to AR with respect to screw placement precision.”
Study design
The investigation protocol used the Gertzbein and Robbins classification system for post-operative assessment, with dual independent surgical evaluation of screw placement accuracy. In the RAN cohort, 108 patients received 827 screws, while in the AR group, 104 patients received 384 screws.
The RAN system achieved a 99.6% Grade A/B placement rate, with 95.2% (787 screws) achieving Grade A classification. The AR system demonstrated a 98.7% Grade A/B placement rate, with 89.8% (345 screws) achieving Grade A classification. Statistical analysis revealed a significant difference in Grade A placement rates between the two systems (p=0.001).
Overall, Grade A placement was achieved in 92.6% of all screws across both technologies. The study documented minimal critical errors, with RAN showing a Grade C/D rate of 0.4% (3 screws) and AR showing a rate of 1.3% (5 screws).
Technical implementation
The study’s methodology incorporated intraoperative 3D fluoroscopic scans for accuracy verification and employed a generalized linear mixed model for statistical analysis. The research design focused specifically on lumbosacral (L1-S1) pedicle screw placement procedures for degenerative conditions.
The research methodology included certain constraints that warrant consideration. The study used a non-randomized cohort assignment approach, and there was potential variation in surgeon experience between modalities. Additionally, as a single-center study, the findings reflect the experience and protocols of one institution.
