Thoracolumbar Burst Fractures

A Systematic Review and Meta-Analysis Comparing Posterior-Only Instrumentation Versus Combined Anterior-Posterior Instrumentation

Hannah Hughes, BScArch, MB, BCh, BAO, MRCSI; Andrea Mc Carthy, BSc; Gerard Anthony Sheridan, MB, BCh, BAO, FRCSI; Jake Mc Donnell, MB, BCh, BAO; Frank Doyle, PhD; Joseph Butler, MB, BCh, BAO, FRCSI


Spine. 2021;46(15):E840-E849. 

In This Article

Materials and Methods

The methodology of this systematic review and meta-analysis followed the "Preferred Reporting Items for Systematic Reviews and Meta-Analyses" (PRISMA) guidelines.[15] Institutional review board approval was not required. A protocol for this systematic review was not registered in "PROSPERO."[16] Provisional data collection was required to determine the feasibility of this meta-analysis. PROSPERO registration is not permitted for reviews for which data collection has commenced.[16]

Search Strategy

An electronic search of the literature was performed on the following databases: MEDLINE (1948–May 2020), EMBASE (1947–May 2020), and The Cochrane Library (1991–May 2020) (Appendix 1, Search terms included "spinal injuries, spinal fractures, lumbar vertebrae, thoracic vertebrae, randomized control trial, single-blind-method, double-blind-method." Keywords included "thoracolumbar, burst, fracture, injury, fusion, arthrodesis, spondylodesis, fixation, internal-fixation, instrumentation, stabilisation." Further databases searched included, Grey Literature Report and the International Clinical Trials Registry Platform (ICTRP), and The International Prospective Register of Systematic Reviews (PROSPERO). The reference section of each relevant article was screened for other potentially suitable publications. Authors of trials were contacted for additional information where necessary.

Inclusion and Exclusion Criteria

Studies were included based on the following criteria: (i) the study must be a RCT; (ii) report on patients over the age of 18 with TLBFs; (iii) compare surgical approaches for TLBFs (posterior-only instrumentation vs. combined instrumentation); (iv) report on at least one of the clinical, functional, or radiological outcomes outlined below. Publications in all languages were considered for inclusion. Studies were excluded from the analysis if: (i) they were not RCTs, (ii) they did not analyze TLBFs, (iii) the surgical approaches of interest were not compared, (iv) the outcomes of interest were not reported.

Outcomes of Interest

The outcomes of interest were divided into three categories: radiological, functional, and clinical. The radiological outcomes of interest included the degree of kyphosis correction postoperatively and the loss of kyphosis correction at subsequent follow-up. The functional outcomes investigated included two patient reported outcome measures (PROMS); visual analogue scale (VAS) score; and Oswestry Disability Index (ODI). The clinical outcomes analyzed included intraoperative blood loss, length of stay (LOS), operative time, and the number and type of postoperative complications.

Data Extraction

Two authors (H.H. and A.M.) reviewed appropriate titles and abstracts and extracted the following data: study details, population characteristics, details of the surgical intervention performed, and data pertaining to the outcomes of interest (Appendix 2, Where there was an additional treatment arm in a trial comparing a surgical approach not of interest to this meta-analysis, the treatment arm was not included in the analysis. To limit selection bias, the independently extracted databases were compared by the reviewers after collection.

Two authors (H.H. and A.M.) independently reviewed the risk of bias for each included study using the Cochrane "Risk of Bias 2" (RoB2) tool and deemed to have a "high," "low," or "unclear" risk of bias.[17] Each study included in the meta-analysis was also subject to critical analysis via the "Grading of Recommendations, Assessment, Development and Evaluation" (GRADE) approach, as per the "Cochrane Handbook for Systematic Reviews of Interventions" guidance.[18] The GRADE approach assesses five criteria in relation to internal and external validity of outcomes; study design, risk of bias, inconsistency, indirectness, imprecision, and the detection of publication bias.[18]

Data Analysis

Statistical analysis was performed using Statistical Package for Social Sciences (SPSS) statistical analysis software [IBM SPSS Statistics for Mac, version 26 (STATA, Texas, USA)] and Review Manager software (RevMan, version 5.3, Cochrane Training).[19] Descriptive statistics were used to describe study characteristics. Dichotomous outcomes were expressed as relative risk (RR) and continuous outcomes as weighted mean difference (WMD) with associated 95% confidence intervals (CI).

A random-effects model was used to perform meta-analyses. Data was pooled for each outcome, the results of which are illustrated on forest plot graphs. A result was deemed to be statistically significant when the P-value was <0.05 and when the value of one was not included in the 95% CI. Heterogeneity was determined by the I 2 statistic, with significance set at a P-value of ≤1.0.