Are pedicle screw perforation rates influenced by distance from the reference frame in multilevel registration using a computed tomography-based navigation system in the setting of scoliosis?

Abstract

Pedicle screw fixation is commonly employed for the surgical correction of scoliosis but carries a risk of serious neurovascular or visceral structure events during screw insertion. To avoid these complications, we have been using a computed tomography (CT)-based navigation system during pedicle screw placement. As this could also prolong operation time, multilevel registration for pedicle screw insertion for posterior scoliosis surgery was developed to register three consecutive vertebrae in a single time with CT-based navigation. The reference frame was set either at the caudal end of three consecutive vertebrae or at one or two vertebrae inferior to the most caudal registered vertebra, and then pedicle screws were inserted into the three consecutive registered vertebrae and into the one or two adjacent vertebrae. This study investigated the perforation rates of vertebrae at zero, one, two, three, or four or more levels above or below the vertebra at which the reference frame was set. This is a retrospective, single-center, single-surgeon study. One hundred sixty-one scoliosis patients who had undergone pedicle screw fixation were reviewed. Screw perforation rates were evaluated by postoperative CT. We evaluated 161 scoliosis patients (34 boys and 127 girls; mean±standard deviation age: 14.6±2.8 years) who underwent pedicle screw fixation guided by a CT-based navigation system between March 2006 and December 2015. A total of 2,203 pedicle screws were inserted into T2-L5 using multilevel registration with CT-based navigation. The overall perforation rates for Grade 1, 2, or 3, Grade 2 or 3 (major perforations), and Grade 3 perforations (violations) were as follows: vertebrae at which the reference frame was set: 15.9%, 6.1%, and 2.5%; one vertebra above or below the reference frame vertebra: 16.5%, 4.0%, and 1.2%; two vertebrae above or below the reference frame vertebra: 20.7%, 8.7%, and 2.3%; three vertebrae above or below the reference frame vertebra: 23.8%, 7.9%, and 3.5%; and four vertebrae or more above/below the reference frame vertebra: 25.4%, 9.5%, and 4.1%, respectively. Fisher exact test was performed to detect significant differences among the above five groups. With regard to Grade 1, 2, or 3 perforations, the rates of screw perforation for three and four vertebrae or more above or below the reference frame vertebra were significantly larger than that for vertebrae at the reference frame (both p<.01). No significant differences were found for Grade 3 perforations (violations) among the groups. In multilevel registration of three consecutive vertebrae, the accuracy of screw insertion into vertebrae at which the reference frame was not set was not significantly inferior to that in vertebrae at which the reference frame was set with regard to major perforation rate. Including minor perforations, however, a distance of three vertebrae or more above or below the reference frame vertebra produced significantly more frequent perforations.