This review assessed the effects of bracing therapies in patients with traumatic thoracolumbar fractures on clinical and functional outcomes, including complications and length of hospital stay, compared with non-bracing therapies. The authors concluded that the value of bracing for these patients remained unclear. Given the level of evidence presented, these findings are likely to be reliable.

To assess the effects of bracing on clinical and functional outcomes (based on the International Classification of Functioning, Disability and Health (ICF)) and length of hospital stay compared with non-bracing therapies in patients with traumatic thoracolumbar fractures and to ascertain the incidence of complications associated with bracing.

MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), CINAHL, Cochrane Database of Systematic Reviews and DARE were searched. Search terms were reported, but search dates were not. Languages were restricted to English, German, French and Dutch. All reference lists were examined for additional studies.

Systematic reviews, randomised or non-randomised controlled trials and observational studies with a control group that compared bracing (three-point corset, Jewett, extension) following non-operative or operative treatment with non-bracing therapies in adult patients admitted to hospital with traumatic (single or multiple, stable or unstable) thoracolumbar fractures were eligible for inclusion in the review.

To be eligible, studies needed to report at least one of a list of outcome measures classified according to ICF in three categories: body functions and structure; activities; and participation or quality of life (QOL) or length of hospital stay (LOS). Studies that included patients with neurological deficit, fractures due to osteoporosis, non-traumatic fractures or patients wearing a cast were excluded. Only full publications were included.

Retrospective studies that evaluated the effect of bracing in non-operatively treated patients with stable thoracolumbar compression fractures (range from 11% to 25% compression on admission) were included in the review. Most fractures were located at T12 and L1. Length of time wearing a brace ranged from 32 days to nine months. Reported indications for bracing, cointerventions, duration of bed rest, mean follow up time (11 months to nine years) and reported outcome measures varied across included studies. Incidence of bracing complications was reported in one of the seven studies.

Two reviewers independently selected studies for inclusion in the review.

Study quality was assessed according to 11 pre-defined criteria for internal validity as published by the Back Group (van Tulder et al. 2003), which addressed methodological issues that related to randomisation and allocation concealment, baseline characteristics, blinding, cointerventions, compliance, drop-out rate, timing of outcome assessment and whether the study included intention-to-treat analysis. Criteria for blinding of patients and providers to the intervention were considered not to be applicable to bracing. The level of agreement between reviewers was assessed using a Kappa (κ) score. A percentage achievement out of the nine remaining relevant criteria was calculated for each study.

Two reviewers independently assessed study validity. Disagreements were resolved through discussion and the use of a third reviewer.

In addition to summary study details, a range of radiographic and clinical outcome measures were extracted. These ranged from subjective measures, such as complaints and pain scores, to those focusing on radiological outcomes, such as percentage compression, gibbus and scoliosis angles measured (as described by Cobb). Individual study characteristics (participants, interventions and follow-up mean) and results for each outcome measure were summarised in a table.

Data were extracted independently by two reviewers.

Due to variations across studies with regard to the diversity of fractures, interventions used and reported outcomes, the studies were combined narratively.

Seven retrospective observational studies with a control group (n=647) and one observational study (without a control group n=48) were included.

The methodological quality of included studies was very low. No studies achieved compliance with 50% or more of the predefined criteria. Five studies met none of the quality criteria and the remaining two studies met only one criteria each.

No significant effects of bracing were found in any of the studies for any of the outcomes reported. Findings for each study were presented in a table. One additional observational study (not summarised in the table) without a control group reported a complication of bracing where two of 48 patients experienced skin problems when treated non-operatively with a brace for thoracolumbar fracture.

There was no evidence for the effectiveness of bracing in patients admitted to hospital with traumatic thoracolumbar fractures.

This review had clearly stated inclusion and exclusion criteria in terms of participants, study design, interventions and outcomes. The authors searched relevant databases and efforts were made to identify additional studies by reviewing reference lists from relevant literature. There were some language restrictions and unpublished studies were excluded, however, it was unlikely that these restrictions would have prevented sufficient robust evidence from being detected to change the conclusions of the review. Search dates were not reported. Appropriate measures to reduce reviewer bias and error were taken at all stages of the review process. The methodological quality of included studies was formally assessed and reported, but the assessment tool used was not appropriate for observational studies and was more relevant to randomised controlled trials. Due to the small number of studies and their heterogeneity, the authors' decision to present the findings in a brief narrative synthesis was appropriate. Given the level of evidence presented the authors' conclusions are likely to be reliable.

Practice: The authors stated that a careful and critical case-to-case decision-making approach to bracing should be undertaken giving consideration to individual patient benefits and the burden and cost of care.

Research: The authors stated that there was a need for high-quality well-reported large randomised controlled trials with sufficient power to detect clinically relevant differences. In addition to radiological and functional outcomes, studies should include patient-centered outcomes such as pain, anxiety, activity status, return to work and quality of life. Long-term follow-up, intention-to-treat analysis, economic evaluation and adequate reporting of complications, drop-out rates and cointerventions such as physiotherapy or medications were strongly recommended.

Not stated.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.