Key messages

• Operative treatment of thoracolumbar burst fractures may reduce the risk of neurologic impairment, but in turn, could increase the risk of general complications.
• It is unclear whether there are differences in pain reduction, improvement in function and quality of life, need for subsequent surgery or radiographic progression of kyphosis in both groups because the certainty of the evidence is very low.
• The findings of this summary partially agree with the results of identified systematic reviews and the recommendations of the main guidelines.

What is the evidence.
See evidence matrix  in Epistemonikos later

We found six systematic reviews [3],[4],[5],[6],[7],[8], including four randomized controlled trials [9],[10],[11],[12].

What types of patients were included

All studies included adult patients (age 18-65 years) with no pathologic thoracolumbar burst fractures without neurological deficit [9],[10],[11],[12]. One study included patients with CT concordant with retropulsion of the vertebral body into the medullar canal [11], one study included patients with CT that ruled out dislocations, with pedicles and facet joints intact despite the fracture of the pedicle [10], one study included patients with CT concordant with residual spinal canal > 50% of normal [9], and one study did not discriminate by type of canal compromise [12].

What types of interventions were included

All studies compared conservative versus surgical management independent of the type of treatment [9],[10],[11],[12]. As conservative treatment two studies used brace [9],[10], one used brace or cast [11] and one used corset [12]. As surgical treatment two studies used fixation with instrumentation [10],[11], one study used posterior stabilization with transpedicular graft or posterolateral fusion [12] and one study did not specify [9].

What types of outcomes
were measured

Reduction of pain by visual analog scale (VAS), Oswestry back pain questionnaire, Short Form-36 and Greenough Low Back Outcome Score (GLBOS); improvement in function and quality of life by Roland Morris Disability Questionnaire (RMDQ); need for subsequent surgery; neurologic impairment and general complications were measured. Other outcomes measured were time to return to work, progression of spinal canal stenosis, radiographic progression of kyphosis (degrees), treatment costs and duration of hospitalization.

We found three systematic reviews [5], [6], [7], including 14 studies   reported in 21 references [8], [9], [10], [11], [12], [13], [14], [15], [16],   [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28]. Eight   studies correspond to randomized controlled trials (15 references [9], [10],   [11], [12], [13], [14], [16], [17], [19], [21], [22], [23], [24], [26], [27]).   This table and the summary in general are based on the latter. One study [15]   did not contribute data to any of the outcomes of interest.

To whom this evidence does and does not apply

• The included studies evaluated adult patients with traumatic thoracolumbar burst fractures without neurological deficit. Patients with pathological vertebral fracture in osteoporotic or tumor bone (metastasis or primary) or neurological deficit at admission are therefore excluded. There was no restriction of patients according to comorbidities or medication use.

• Outcomes considered critical for decision making in the view of the authors of this summary were included. These coincide with those presented in the main systematic reviews identified.

• The evidence shows a possible benefit of operative treatment regarding neurologic impairment, but an increased risk of general complications.
• The main advantage of conservative treatment is to prevent perioperative complications, including infection, iatrogenic neurological injury, non-union, instrumentation failure and anesthesia-related complications.
• There is not an option clearly preferable, so it is essential to stimulate shared decision-making with the patient to assess the risks and benefits of surgical versus conservative treatment.

• It is possible that many patients would give more value to the prevention of neurological impairment and prefer operative treatment despite the low certainty of the evidence.
• Risk-averse patients or those at higher operative risk may prefer a conservative management. 

• Due to the high level of uncertainty about the benefits and risks, it is not possible to make an adequate balance. Whereas both alternatives have superiority over the other in terms of a critical outcome for decision-making, a formal cost-benefit assessment is needed, as well as a better evidence base, to determine which option is superior.

Differences between this summary and other sources

• The findings of this summary partially agree with the results of most recent systematic reviews.
• The findings of this summary are in partial agreement with the recommendations of the main clinical guidelines [13]. The guideline of the American Academy of Orthopaedic Surgeons (AAOS) suggests conservative treatment should be considered as the initial method of choice even in borderline cases where the absence of neurological injury or instability is not entirely clear, on the basis that the benefits of surgery are not well defined. It is important to note this guide includes a low proportion of the evidence identified in this summary, in part because it was not available at the time of its publication.

• The probability that future evidence change the conclusions of this summary is moderate due to the low certainty of the evidence.
• There are no new or ongoing studies regarding this matter.