Editorial: Pedicle screw navigation

Abstract

In their paper, Mr. Shin and colleagues present a metaanalysis comparing the accuracy of computer-navigated pedicle screw insertions versus nonnavigated techniques.1 The authors only included studies in which both techniques were performed. A total of 18 cohort studies and 2 randomized controlled trials, performed between 2000 and 2011, were evaluated. Approximately similar numbers of pedicle screws were placed in the 2 groups evaluated. The authors found that navigated screws violated the pedicle in 6% of instances, while nonnavigated screws had a 15% violation rate (p < 0.001). Despite this discrepancy, the revision surgery rates were not statistically different between the 2 groups. Additionally, the operative times and the blood loss were not significantly different between the groups. While the numbers appear to favor superior accuracy in the navigation group, the details of some of these studies lead to some problematic issues. Bias cannot be excluded in this meta-analysis. Articles that compared the 2 methods are very likely written by advocates of navigation. In fact, the authors of this meta-analysis state, “...the published literature is likely to exhibit a publication bias.” Issues that raise concern in these articles include the complication rates. With more than 8000 pedicle screws placed, the navigation group had a single wound infection, and the nonnavigation group had 3 wound infections in only 1 study in this group. These figures are at dramatic odds with the published wound infection rates for thoracic spinal surgery and cause the reader to be concerned that an underreporting of complications exists in these reports. The well-described in-out-in technique, used predominantly in the thoracic spine for placing larger screws into smaller pedicles, is ignored in this meta-analysis. No attempt was made to assess screw size. The clinical significance of placing a screw entirely within a pedicle is not clear from the wealth of published data on this topic. If a navigation advocate places smaller screws to remain intrapedicular, will maneuvers such as deformity correction be helped or hindered? This topic is not covered, and it bears much clinical import. While the authors of the meta-analysis state that radiographic evidence was required, it is not clear whether plain films or CT scans (much more accurate for this type of analysis) were assessed. The quoted survey of more than 3000 spine surgeons, with a 10% response rate, is not adequate to make scientific statements regarding this topic. Furthermore, it is likely that advocates of navigation were more likely to respond to a survey on its usage than spine surgeons who do not use, or possibly do not believe in, navigation. Additional information that would be useful for spine surgeons in determining the utility of navigation would be clinical outcomes, cost data, and radiation exposure. None of these was addressed in this meta-analysis. Some highly experienced spine surgeons who do not use navigation have reported better screw placement accuracy, using CT scanning data, than either of the groups in this meta-analysis. As such, it is possible that this meta-analysis has inadvertently skewed the argument in favor of surgeons who utilize navigation and may not be reflective of the spine surgical community at large. While this meta-analysis clearly shows that navigation is able to improve the accuracy of pedicle screw placement in the hands of surgeons who perform both navigation and nonnavigation techniques, this review is unable to determine whether there is a clinical advantage to the use of navigation for the reasons outlined above. (http://thejns.org/doi/abs/10.3171/2012.3.SPINE111098)