Short Segment Pedicle Instrumentation Versus Short Segment Pedicle Instrumentation With Intermediate Screws For Thoraco-lumbar Burst Fracture:a Meta-analysis

Objective: At present,in the treatment of thoracolumbar burst fractures has not yet formed a consensus,full of challenge and controversy.Whether need the intermediate screws or not for thoraco-lumbar burst fracture patients who want to undergo the posterior short segment pedicle instrumentation surgery? In order to offer the theoretical basis,the researchers search and analyze the relevant literatures in the nearly past 30 years;assess the quality of literatures completely;select the appropriate criteria which can systematically evaluate the efficacy of short segment pedicle instrumentation(SSPI)versus short segment pedicle instrumentation with intermediate screws(SSPI+IS)for thoraco-lumbar burst fracture.Methods: Based on the specific searching strategies that for different database,studies of short segment pedicle instrumentation versus short segment pedicle instrumentation with intermediate screws for thoraco-lumbar burst fracture were electronically searched from January 1990 to October 2016 using The Cochrane Library、Pub Med、Embase、Web of science、Wan Fang、CNKI、VIP database.According to the established inclusion and exclusion criteria,two independent researchers browsed literature,extracted data and assessed the risk of bias of studies,Meta-analysis was performed using Rev Men5.3software.Results: Seventeen literatures which composed of fourteen Chinese literatures and three English literatures involving 1303 cases,of whom 622 had short segment pedicle instrumentation and 681 had short segment pedicle instrumentation with intermediate screws,were included.Two of them are randomized control trial,fifteen of them are retrospective.The Meta analysis showed: the canal compromise rate of SSPI+IS group less than the SSPI group after the operation(P<0.001,MD=-3.69,95%CI=[-5.03,-2.36])、 the visual analog scale score of SSPI+IS group less than the SSPI group at the last follow-up(P<0.001,MD=-1.58,95%CI=[-2.26,-0.90])、the Cobb′s angle of SSPI+IS group smaller than the SSPI group after the operation(P<0.001,MD=-1.74,95%CI=[-2.54,-0.94])、the Cobb′s angle of SSPI+IS group smaller than the SSPI group at the last follow-up(P<0.001,MD=-2.44,95%CI=[-3.37,-1.51])、the anterior vertebra body height of SSPI+IS group lower than the SSPI group before the operation(P=0.09,MD=-1.22,95%CI为[-2.64,0.19])、the anterior vertebra body height of SSPI+IS group higher than the SSPI group after the operation(P=0.0009,MD=1.55,95%CI=[0.64,2.47])、the anterior vertebra body height of SSPI+IS group higher than the SSPI group at the last follew-up(P<0.001,MD=2.83,95%CI为[1.84,3.82]).There is no statistical significance in sides of operation time(P=0.52,MD=1.17,95%CI=[-2.37,4.70])、 intraoperative blood loss(P=0.36,MD=8.71,95%CI=[-9.83,27.25])、hospitalization time(P=0.38,MD=0.83,95%CI=[-1.04,2.70])、 canal compromise rate before operation(P=0.41,MD=0.61,95%CI=[-0.83,2.04])、Cobb′s angle before operation(P=0.26,MD=0.32,95%CI=[-0.24,0.88])、visual analog scale score before operation(P=0.69,MD=0.08,95%CI=[-0.33,0.50])、visual analog scale score after operation(P=0.16,MD=-0.26,95%CI=[-0.62,0.1])。Conclusion: Compared to short segment pedicle instrumentation,short segment pedicle instrumentation with intermediate screws can better correct the kyphosis and maintain the reduction.Inclusion of the intermediate screws did not lengthen the operation or hospitalization,nor did it increase the amount of blood loss or post-operative pain.SSPI+IS for the treatment of thoracolumbar burst fractures came into being as a supplementary means when SSPI unable to better correct the kyphosis and maintain the reduction.According to the current studies,people who have indication ought to placement intermediate screw to reduce the failure rate of internal fixation.However,SSPI+IS was insufficient for the patient with disc rupture or three column injury.Maybe,long segment fixation or fusion was need to increase the stability of the spine.In addition,there are few randomized controlled trials in this study,which are not blind,and the sample size is not enough.We need to design and implement more large sample,multi center RCT to verify the conclusions...