This study aims to investigate the efficacy of posterior short-segment pedicle instrumentation without fusion in curing thoracolumbar burst fracture. All of the 53 patients were treated with short-segment pedicle instrumentation and laminectomy without fusion, and the restoration of retropulsed bone fragments was conducted by a novel custom-designed repositor (RRBF). The mean operation time and blood loss during surgery were analyzed; the radiological index and neurological status were compared before and after the operation. The mean operation time was 93 min (range: 62–110 min) and the mean intraoperative blood loss was 452 mL in all cases. The average canal encroachment was 50.04% and 10.92% prior to the surgery and at last followup, respectively (
Owing to the fulcrum of increased motion at the T12-L1 junction, approximately 90% of spine fractures are located in the thoracolumbar region, and the burst fractures account for nearly 10–20% of spine injuries [
The aims of surgical treatment for thoracolumbar burst fracture are as follows: (1) promoting neurological recovery by decompression of spinal canal and nerve roots, (2) obtaining rigid fixation to prevent delayed neural injuries, (3) restoration and maintenance of anatomic alignment, (4) relieving pain and facilitating early rehabilitation, and (5) limiting the number of instrumented centrums [
Compared with long-segment posterior fixation (LSPF), the short-segment posterior fixation (SSPF) is easy to implement, offering the advantage of preserving spinal motion segments, shorter operative time, and less blood loss. With the improvement of rigidity and stiffness of pedicle screw-based posterior spinal instrumentation systems, the short-segment has become more reliable [
The inclusion criteria of this study were a single-level thoracolumbar burst fracture (T11-L1) with neurological deficit and the intraspinal bone fragments confirmed by CT. The exclusion criteria were as follows: patients with pathological fractures, multiple level fracture, polytraumatized patients, and a history of spine surgery and preexisting neurological deficit. From July 2009 to Aug 2011, 53 patients were in accordance with the inclusive criteria. Among the 53 patients, there are 19 females and 34 males, ranging from 18 to 63 years (average 34.5 years). The major mechanism of injury was fall and traffic accidents. According to the classification described by Magerl et al. [
Anteroposterior and lateral radiographs, CT, and magnetic resonance imaging (MRI) of the thoracolumbar region were performed in all patients on admission. The segmental kyphotic angle of fractured vertebra was measured as the angle between the upper margin of the vertebral body and the lower margin of the vertebral body. The vertebral body height was calculated by the formula adopted by Haas et al. [
(a)-(b) The diagram of kyphotic deformity and vertebral body height calculated in current study. Vertebral body height
All patients were operated in prone position, with a midline skin incision extending above and below 1 or 2 levels of the fractured levels. The paraspinal musculature was dissected bluntly. Aiming to achieve temporary stability of the spine, the short-segment pedicle instrumentation was fixed before posterior decompression was implemented. The fixation was achieved by inserting four screws into the pedicles of the adjacent vertebra above and below the injured level, and then rods with appropriate length were inserted and connected to the screws. Some reduction techniques such as distraction of anterior and posterior column and in situ bending of the rods were used, and then the screws were fixed tightly. After conventional laminectomy, the reduction of the retropulsed bone fragments was conducted by using a novel custom-designed instrument.
This novel custom-designed repositor for retropulsed bone fragments (NO. ZL201020673890.X, CN PAT) was designed by the corresponding author and manufactured by Dragonbio (Hubei, China). It consisted of two parts, the supporting apparatus and reduction apparatus (Figures
All patients were immobilized by a thoracolumbar sacral orthosis brace for preventing the implant failure and promoting neurological recovery. After appropriate 4 weeks, patients with sufficient motor movement were gradually mobilized according to personal neurological status and radiological review. The restoration of vertebral height and correction of the kyphotic angle were assessed by anterioposterior and lateral film for several times which included 1 week after surgery, regular interval of 3 months, and the final follow-up examination. CT scans were also obtained to assess the percentage of postoperative spinal canal compromise using the similar formula described in radiological evaluation.
The whole statistical analysis was performed with SPSS for windows version 18.0 (SPSS Inc, Chicago, USA). The paired
The mean operative time in all cases was 93 min (ranging from 62 to 110 min). No patient needed blood transfusion with mean 452 mL intraoperative blood loss. There were no major complications such as death, cerebrospinal fluid leak, or epidural hematoma. As for the other complications, 2 cases suffered pneumonia which was cured by antibiotics, and one patient developed sacrococcygeal bedsore which was cured by applying vacuum-assisted closure. Patients were monitored 14–49 months (average 25.4 months) for follow-up evaluation, the implant failure (broken screws and rods) was observed after operation in two patients. The fixation was removed for all patients between 12 and 16 months after surgery.
In this study, all patients have neurological deficit at admission. The mean spinal canal compromise in patients with complete paraplegia and incomplete paraplegia was
The relationship between the level of initial neurological status with the extent of canal encroachment, the loss of vertebral body height, and the kyphotic angle, respectively.
Neurological status |
|
Canal compromise (%) | Loss of vertebral body height (%) | kyphotic angle (degree) |
---|---|---|---|---|
A | 11 |
|
|
|
B | 16 |
|
|
|
C | 16 |
|
|
|
D | 10 |
|
|
|
Total |
|
|
|
|
|
|
|
The average canal encroachment of retropulsed bone fragments was 50.04%, 20.04%, and 10.92% preoperatively, postoperatively, and at last followup, respectively. The difference had a statistical significance (
(a)–(e) The preoperative CT scan image demonstrating a significant spinal canal encroachment by retropulsed bone fragments. (f)-(g) The intraoperative application of RRBF and the fluoroscopy of restoration of fractured vertebral body.
(a)–(d) The postoperative CT scan confirming satisfactory canal clearance and restoration of vertebral body height. (e) After removal of implants for one months, the lateral radiograph showed that the kyphotic angle and vertebral body height were maintained acceptably.
There was no deterioration in patients’ neurological status. Among the 11 patients with Frankel’s grade A, 6 showed no improvement in neurological status, and the other 5 showed 1-2 grades improvement in Frankel’s scale at last followup. Especially, 40 out of 42 patients with incomplete paraplegia (Grade B–D) achieved significantly neurological recovery (Table
The neurological recovery of 53 patients with thoracolumbar burst fracture.
Neurological status at admission | Neurological status at last followup | ||||
---|---|---|---|---|---|
A | B | C | D | E | |
A | 6 | 3 | 2 | 0 | 0 |
B | 0 | 1 | 6 | 8 | 1 |
C | 0 | 0 | 0 | 12 | 4 |
D | 0 | 0 | 0 | 1 | 9 |
The treatment of thoracolumbar burst fracture remains challenging and debatable. There are middle column injuries in thoracolumbar burst fracture, so a lot of surgeons think that it is an unstable spinal fracture [
Even it is not clear that there is correlation between the neurologic recovery and the surgical approach, the pedicle screw fixation via posterior approach has been widely used for most thoracolumbar fractures owing to its 3-column fixation and satisfactory clinical outcomes [
Several studies have reported good outcomes in thoracolumbar burst fracture patients treated by only SSPF without fusion and laminectomy [
There is still no consensus about the relationship between canal encroachment and initial neurological status. Studies have revealed that the outcome of neurological injury depends on the severity and extent of damage to the neural elements at the time of injury, so the radiological static image of encroachment of the spinal canal after trauma is unable to represent the dynamic process and predict the neurological status [
In thoracolumbar burst fracture, the spinal canal narrowing is caused by the fracture bone fragments. Yan et al. [
There are several limitations in this study, such as it was a retrospective study and mean followup was 25.4 months. The long-term effectiveness of this technique still needs to be evaluated. The small sample of patients precludes absolute conclusion with regard to the advantages of this technique. Although this technique seems to be a potential therapy for thoracolumbar burst fracture, it is still unknown whether this technique is superior to others because no control group was available in this study. The outcome of this technique was satisfactory in this study; however, it by no means indicated that all patients with thoracolumbar burst fracture were appropriate for this treatment. The number of patients with AO 3.3 thoracolumbar burst fracture was too small to assure that this technique was proper for severe and comminuted fracture. The implant failure reminds us to be cautious when SSPF was implemented in patients with osteoporosis. It is inadequate to conclude that this technique could be a first line treatment for thoracolumbar burst fracture. However, mastering this technique will allow surgeon to be more flexible in specific situation, for example, patients who are not ideal surgical candidates for anterior approach, long-segment instrumentation, fusion, and vertebroplasty.
The authors declare that they have no conflict of interests.
This study was supported by Science and Technology Key Project of Wuhan (no. 201260523173-1), Innovation Seed Fund of Wuhan University School of Medicine, and the Fundamental Research Funds for the Central Universities (no. 2012303020207).