[Clinical efficiency of computer-assisted pedicle screw placement versus conventional method: a meta-analysis]

To compare the clinical efficacy of computer-assisted pedicle screw placement and conventional placement in the treatment of spinal disease.

A systematic search of studies published between Jan. 1990 and Feb. 2012 was conducted using Medline, Embase, OVID, ScienceDirect and Cochrane Review databases. Randomized controlled trials (RCTs) and prospective controlled trials of comparing computer-assisted pedicle screw placement to conventional method performed at one center or multi-centers providing data on accuracy of placement and clinical effects were identified. Two study authors independently reviewed the 93 articles originally identified and selected 10 for analysis. Study title,demographic characteristics,number of pedicle screw,anatomical area and outcomes were extracted manually from all selected studies. RevMan 5.1 software was used for meta-analysis.

Ten studies encompassing 2813 pedicle screws met the inclusion criteria. Overall,the result of meta-analysis indicated that there were significant differences between two groups in accuracy in placement of pedicle screw [OR = 2.58, 95% CI (1.18, 5.63)], insertion time [WMD = -2.15, 95% CI (-2.36, -1.94)]. However, there was only one study reported preparation time of pedicle screw of navigation group was longer than conventional group. No neurological complication in navigation group was reported.

As a safety supplementary measure, computer navigation provide better accuracy in placement of pedicle screw and insertion time. The preparation time of pedicle screw may prolong due to the complexity of navigation system. Further reseach should include randomized controlled trials with well-planned methodology to limit bias.