Evaluation of Screw Placement Accuracy in Circumferential Lumbar Arthrodesis Using Robotic Assistance and Intraoperative Flat-Panel Computed Tomography.

BACKGROUND: Neuronavigation has become a common means of placing pedicle screws in vertebral arthrodesis, because it reduces the incidence of complications related to poor screw positioning.

OBJECTIVE: The To evaluate the accuracy of pedicle screw placement during lumbar arthrodesis performed with the new ROSA Spine robot and intraoperative flat-panel computed tomography (fpCT) guidance.

METHODS: We performed a descriptive, observational, retrospective, single-center study of patients having undergone transforaminal lumbar interbody fusion (TLIF) for degenerative pathologies with the ROSA Spine robot and intraoperative fpCT guidance. The primary endpoint was the presence or absence of pedicle wall breach, as assessed from the fpCT scan. Screw placement was graded according to Ravi's classification.

RESULTS: Twenty-five patients were operated on between November 2014 and July 2016; 21 underwent minimally invasive TLIF, and 4 underwent open TLIF. The mean ± standard deviation operating time was 241 ± 49.8 minutes. Of the 110 screws placed, 101 (91.8%) were completely within the pedicle (Ravi grade A), 5 (4.5%) had a pedicle wall breach <2 mm (grade B), 2 (1.8%) had a pedicle wall breach of 2-4 mm (grade C), and 2 had pedicle wall breach >4 mm (grade D) (1.8%). One screw (a grade D) was replaced during surgery. None of the breaches were symptomatic.

CONCLUSIONS: When coupled with fpCT, the ROSA Spine robot is a reliable, accurate means of performing lumbar pedicle screwing.