Brace or Surgery for the Neurologically Intact Burst Fracture
A wrong call either puts a patient through a fusion and its complications for a fracture that would heal without one, or leaves an unstable spine to collapse into kyphosis and nerve injury.
Key terminology
- Thoracolumbar junction
- This region is where the stiff thoracic spine meets the mobile lumbar spine, roughly T11 to L2. Most burst fractures happen here because load concentrates at the transition.
- Middle column
- This model splits each spinal segment into anterior, middle and posterior columns. The middle column is the back wall of the vertebral body and the ligament and disc just behind it.
- Burst fracture
- The vertebral body is crushed by an axial load and its posterior wall breaks. Bone fragments can be pushed back toward the spinal canal.
- Posterior ligamentous complex (PLC)
- These are ligaments behind the spinal canal, including the supraspinous and interspinous ligaments, ligamentum flavum and facet capsules. They act as the posterior tension band.
- TLICS
- This score adds points for fracture pattern, posterior ligament injury and neurologic deficit. The total sorts patients toward nonoperative care or surgery.
- Retropulsed fragment
- A fragment of the posterior vertebral wall is driven backward into the spinal canal. It is seen best on axial CT.
- TLSO
- This rigid brace wraps the trunk from chest to pelvis to limit bending. It is the classic nonoperative treatment for a stable burst fracture.
- Equivalence trial
- This trial asks whether 2 treatments give outcomes within a preset margin of each other. Landing inside the margin means the cheaper or simpler option is as good.
Takeaways
- Neurologic status and the posterior ligaments decide whether a burst fracture needs surgery. TLICS scores fracture morphology, PLC integrity and neurologic status. A total of ≤3 points to nonoperative care, 4 is the surgeon's call and ≥5 points to surgery. A burst in an intact patient with an intact PLC scores 2, and the same fracture with a torn PLC scores 5. (Vaccaro et al, Spine 2005;30:2325-2333, PMID 16227897; Shen et al, PLoS One 2015;10:e0135735, PMID 26284373; Rainone et al, Cureus 2022;14:e29492, PMID 36299980).
- A burst fracture breaks the anterior and middle columns under an axial load. Denis defined the middle column as the posterior wall of the vertebral body, the posterior longitudinal ligament and the posterior annulus. A compression fracture spares the middle column, so the posterior wall stays intact on CT. A broken posterior wall is what lets fragments retropulse into the canal. (Denis, Spine 1983;8:817-831, PMID 6670016; Abudou et al, Cochrane Database Syst Rev 2013;(6):CD005079, PMID 23740669).
- MRI is the test for the posterior ligamentous complex, and kyphosis on CT is a poor stand-in. On fat-suppressed T2 or STIR, a discontinuous supraspinous ligament marks PLC rupture, with sensitivity 91% and specificity 100% in 58 fractures. Kyphosis >20° and height loss >50% did not predict PLC injury in 46 burst fractures. (Pizones et al, Spine 2013;38:745-751, PMID 23089929; Radcliff et al, Spine 2012;37:1142-1150, PMID 22146278).
- In the longest randomized trial, nonoperative care did as well as surgery early and better in the long run. Wood randomized 47 intact patients with stable burst fractures to fusion or a cast or brace. At 44 months outcomes were similar and complications were more frequent after surgery. At 16 to 22 years the nonoperative arm had a lower median ODI (2 vs 20) and less pain. (Wood et al, J Bone Joint Surg Am 2003;85:773-781, PMID 12728024; Wood et al, J Bone Joint Surg Am 2015;97:3-9, PMID 25568388).
- Two small randomized trials of surgery for the intact burst fracture reached opposite conclusions. Siebenga randomized 34 intact AO A3 fractures and found less kyphosis, less pain, better function and more return to work after posterior fixation. Cochrane judged the 2 trials too small and biased to pick a winner. Surgery had more complications (21/41 vs 6/38) and cost over 4 times more in one trial. (Siebenga et al, Spine 2006;31:2881-2890, PMID 17139218; Abudou et al, Cochrane Database Syst Rev 2013;(6):CD005079, PMID 23740669).
- The brace is optional for the stable burst fracture in an intact patient. In an equivalence trial of 96 intact patients with AO A3 fractures from T11 to L3, 3-month disability was the same with a TLSO or with no orthosis and immediate walking (RMDQ 6.8 vs 7.7). Six patients needed surgery, 5 of them before first discharge. A meta-analysis of 3 trials found no difference in pain, function or kyphosis. (Bailey et al, Spine J 2014;14:2557-2564, PMID 24184649; Wallace et al, JBJS Rev 2019;7:e9, PMID 31567716).
- A retropulsed fragment in an intact patient is no reason on its own to decompress. In 41 intact patients treated without surgery, canal compromise averaged 37% and about two-thirds of the fragment resorbed, mostly within 1 year. Another series saw mean stenosis fall from 50% to 25%. Canal compromise was still the most consistent predictor of conversion to surgery in a 2026 meta-analysis. (Mumford et al, Spine 1993;18:955-970, PMID 8367783; de Klerk et al, Spine 1998;23:1057-1060, PMID 9589546; Taddei et al, Eur Spine J 2026, PMID 42811137).
- A thoracolumbar fracture means imaging the whole spine. A noncontiguous cervical fracture turned up in 10.49% of blunt trauma patients with a thoracolumbar fracture. L1 at the thoracolumbar junction is the most commonly fractured vertebra, and burst is the most common pattern. Motor vehicle collisions and high-energy falls cause most of these injuries. (Katsuura et al, J Orthop 2016;13:383-388, PMID 27504058).