Comparison between the accuracy of percutaneous and open pedicle screw fixations in lumbosacral fusion

Background context: In pedicle screw fixation, accurate insertion is essential to avoid neurological injury or weak stability. The percutaneous pedicle screw system was developed for minimally invasive spine surgery, and its safety has already been reported. However, the accuracy of percutaneous pedicle screw fixation (PPF) has not been compared with that of the open system to date. Purpose: To compare the accuracy of PPF with that of open pedicle screw fixation (open PF) and to investigate the risk factors associated with pedicle wall penetration. Study design/setting: A retrospective case series. Patient sample: The study group included 237 patients who underwent posterior pedicle screw fixation between January 2008 and October 2010 at a single institute with a total of 1,056 pedicle screw fixations completed. One hundred and twenty-six patients with 558 screws underwent open PF and 111 patients with 498 screws underwent PPF. Outcome measures: Postoperative computerized tomography, including sagittal and coronal reformatted images. Methods: Consecutive surgeries with either conventional open PF or PPF for anterior lumbar interbody fusion or transforaminal lumbar interbody fusion were performed. The open pedicle screw employed was from the WSH system (Winova, Seoul, Korea), and the two percutaneous pedicle screw systems were the Sextant (Medtronics, Minneapolis, MN, USA) and the Viper systems (DePuy Spine, Raynham, MA, USA). Computed tomography images were evaluated to determine pedicle wall penetration after operation. Severity was classified as mild (<3 mm), moderate (3-6 mm), and severe (≥6 mm), and the direction was assessed as medial, lateral, inferior, and superior. Results: Pedicle wall penetration occurred in 75 patients (13.4%) in the open PF group and 71 patients (14.3%) in the PPF group and was not statistically different between the groups (p=.695). Assessment of the severity of the pedicle wall penetration revealed that minor penetration was the most common (open PF group, 9.7%; PPF group, 10.6%), although the distribution of the degree of severity was not statistically different between the groups (p=.863). A relatively higher incidence of lateral penetration was observed in the open PF group (66.7% vs. 43.7%), whereas medial, superior, and inferior penetrations were higher in the PPF group (p=.033). Other parameters such as age, sex, surgical method, and surgeon factors did not influence the penetration rate, but bone mineral densitometry negatively correlated with the penetration. Conclusions: Pedicle wall penetration during screw fixation was not different between the open PF and PPF groups. The lateral, paraspinal, muscle-splitting approach seems to lessen medial wall penetration, especially in the S1 vertebra. Distribution of the direction of penetration differs between the groups, with lateral wall penetration being more prominent in the open PF group. Careful placement of pedicle screws is necessary for a stronger construct because of the high incidence of penetration.