Anterior column reconstruction in thoracolumbar injuries utilizing a computer-assisted navigation system.

BACKGROUND: Discectomy, corpectomy, and resection of isolated posterior wall fragments are technically demanding steps requiring maximum surgical precision during anterior reconstruction of the unstable thoracolumbar spine.
PURPOSE: This study investigates the feasibility of computer-aided guidance for these steps. It also analyzes the precision, advantages, and disadvantages of the procedure. STUDY
DESIGN: Controlled clinical trial. PATIENT SAMPLE: 21 patients were included in the trial group; the control group consisted of 10 patients. OUTCOME MEASURES: Total time for surgery was noted. To assess surgical precision, decentralization of the cage was measured in postoperative X-rays. Additionally, parallel alignment of vertebral body endplates with the cage was evaluated in postoperative CT scans.
METHODS: Vertebral body fractures of the thoracolumbar spine addressed by disc-/corpectomy and subsequent cage interposition for anterior reconstruction were included. All surgical steps were performed under endoscopic assistance. In the trial group, disc- and corpectomy were performed under computer-aided guidance; in the control group, no computer navigation was utilized. In cases of initial neurological deficit after trauma, the patients underwent emergency laminectomy during the initial posterior stabilization procedure. During the second-stage anterior procedure, resection of the posterior wall fragment with the aid of computer-aided navigation was performed.
RESULTS: Fractures were localized between Th9 and L1 in the trial group, and Th10 and L1 in the control group. Time for surgery was significantly shorter in the control group: 1.7 h ± 0.5, as opposed to 3.8 h ± 1.0 in the trial group (p < 0.0005). In contrast, data on surgical precision did not show statistically significant differences between both groups for either decentralization or parallel endplate alignment of cages. Remarkably, we noted two cases of subsidence in bilevel cages in the control group, whereas this was only noted in one case in the trial group. However, this difference was not statistically significant. There were five patients with initial neurological deficits. At the time of follow-up, the neurological statuses of all five had improved by at least one Frankel grade.
CONCLUSIONS: Computer-aided guidance in anterior reconstruction of the thoracolumbar spine is a technically feasible option that may aid in the performance of disc- and corpectomy, as well as the resection of isolated posterior wall fragments in cases with initial neurological compromise. However, total time for surgery is significantly prolongated by this technique. There were no differences in the precision of cage positioning between groups. However, during discectomy, the use of computer navigation may aid in the protection of adjacent endplates, as there was a trend towards fewer cases with cage subsidence in the navigated group.