Accuracy of a patient-specific template for pedicle screw placement compared with a conventional method: a meta-analysis

Accurate placement of pedicle screws in spine surgery is a challenge for surgeons. Patient-specific template techniques have the potential for improving the accuracy of screw placement. The target of this analysis was to investigate differences in terms of accuracy of pedicle screw insertion between patient-specific template assistance and the conventional free-hand method for reconstruction of spinal stability. The Cochrane Library, Ovid, Web of Science, PubMed, EMBASE and CNKI database were searched until February 2017 for a systematic review, and several comparative studies were screened for comparisons of accuracies of pedicle screw insertion with patient-specific assistance and conventional methods. Primary outcomes extracted from papers that met the selection criterion were expressed as odds ratios for dichotomous outcomes with a 95% confidence interval. A chi (2) test and I (2) statistics were used to evaluate heterogeneity. A total of ten RCTs and two prospective cohort studies were finally chosen for the analysis of accuracy rates. Study quality was assessed using the Cochrane Collaboration's Tool and Newcastle-Ottawa Quality Assessment Scale. There were obvious differences between them, and the accuracy rate of screw implantation among a patient-specific template assistance set was statistically significantly higher than the conventional free-hand set (OR 95% CI 3.78-6.41, P < 0.01); in vitro: OR 95% CI 3.93-7.42, P < 0.01; in vivo: OR 95% CI 2.49-6.44, P < 0.01. The template-assisted technique is superior to the conventional method for the reduction of pedicle violation. The template-assisted technique is a promising technique that should be considered as another available navigation tool for surgeons to improve the accuracy of pedicle screw placement. As an available technique for emerging applications in spine surgeries, this technique will face challenges but ultimately prove successfully.