Treatment of thoracolumbar fractures by temporary posterior instrumentation with selective fusion schemes

Objective: This retrospective study investigated the clinical and radiographic outcomes following temporary transpedicular posterior instrumentation between two cohorts of patients with thoracolumbar fractures (TLF) who underwent selective or bi-segments intervertebral articular process fusion. Materials and Methods: Patients with TLF who underwent the temporary posterior fixation with selective fusion ( Group SF ), or bi-segments fusion ( Group BF ) were studied. Superior intervertebral articular process and interlaminar fusion were performed in Group SF, whereas in Group BF, the patients underwent bi-segments fusion in both superior and inferior articular processes, as well as interlaminar fusion. We measured the distal and proximal intervertebral mobility, regional kyphotic angle, and vertebral height before and after surgery in both groups. Greenough Low-Back Outcome Score was used to assess the clinical outcomes. Results: Sixty-five patients with TLF from T12 to L2 fractures were enrolled in the study period: 23 patients in the Group SF and 22 patients in the Group BF. All the patients experienced fracture healing (mean follow-up time: 19.7 months). The mean postoperative functional outcomes were 65.0±2.0 points for the Low-Back Outcome Score in the Group SF and 65.2±1.8 for the Group BF. A progressive regional kyphotic angle was observed with time regardless of fusion but was not significantly different between the two groups. There was a statistical difference between unfused inferior proximal adjacent and inferior distal adjacent segment regardless of fracture segments. Conclusion: The strategy of selective fusion is reported to be useful for the treatment of patients with TLF. The motion in the un-fused and adjacent segment could be better regained after instrumentation removal in the selectively fusion group.


Background
Fractures of the thoracolumbar junction and upper lumbar spine (T12-L2) are common injuries as a result of high-energy trauma with potentially devastating consequences including pain, neurological injury, and loss of function [1]. The main strategy of surgical treatment for thoracolumbar fractures (TLF) has been neurologic decompression, anatomic reduction, stabilization, and fusion [2,3]. However, the best treatment and surgical indications of TLF, especially for the fusion procedure are still controversial [4][5][6]. Fusion generates definite immediate benefits and late stability, such as vertebral body height loss beyond 40%, an abnormal signal in the intervertebral disc, and rupture of the inter-spinal ligament [7]. Although the fusion of mobile segments solves the problem of instability, it restricts the movement of the spine. As adopted by reconstructive surgery of spine trauma, the protection of mobile segments is possible by restricting the number of fusion segment [8,9]. In the present study, patients who suffered from unstable TLF with kyphotic angle above 20 degrees and vertebral body height loss more than 40% were enrolled. Typically, the fused schemes are performed in both superior and inferior articular processes [10,11]. As a result, we proposed the selective fusion schemes for TLF with the superior intervertebral articular process and interlaminar fusion, regardless of bi-segments fusion. This approach aimed at post-operative, preservation of spine motion, stabilization of fracture, and restoration of spinal alignment. In order to validate the effectiveness of selective fusion schemes, we compared the regional kyphotic angles, intervertebral mobility to the fractured segment, and the distal adjacent segments following temporary posterior fixation in TLF patients who underwent either selective (superior) intervertebral articular process or bi-segments (superior-inferior) articular process fusion.

Patient Population
The institutional review boards at two trauma centers (grade III-A) approved this retrospective clinical investigation. All the participants were over 16 years old, and their written informed consent was obtained. Data from 102 patients who were diagnosed with TLF at two-grade III-A hospital (spinal trauma centers) were retrospectively reviewed between January 2011 and January 2018. Fractures were classified according to the AO classifications of TLF as follows: The inclusion criteria were: 1) A1 A3 B1 and B2 AO type TLF from T12 to L2 fractures with one injured spinal segment. 2) TLF with kyphotic deformity > 20 degrees, loss of vertebral body height by more than 40%, and posterior ligamentous complex disruption. 3) The injured spinal segment with only injured upper endplate or corresponding abnormal intervertebral disc signal. 4) Fractures without or partial neurological deficit.
The exclusion criteria were: 1) Patients who suffered severe neurological deficits, and required laminectomy for neural decompression. 2) Fractures that involved more than one vertebras. 3) Patients with severe osteoporosis. 4) Patients that did not receive hardware removal after fracture healing.
From 102 reviewed cases, 37 patients were excluded because of the exclusion criteria and incomplete radiographic data. Sixty-five patients with fractures of the thoracolumbar junction (45 male, 20 female, 17-58 years old) were enrolled in this study. They were all surgically stabilized with the transitory and transpedicular posterior instrumentation, involving the injured vertebrae above and below the fractured vertebrae with superior intervertebral articular process fusion or superior-inferior articular process fusion. Thus, the patients were divided into two groups by the intra-operational intervertebral articular process fusion segment and three subgroups by the injured vertebral segments (T12, L1, or L2 fractures). The patients' general information are listed in Table 1 ( no significant difference was observed).

Surgical methods
Patients were operated by a posterior approach, using a standard surgical technique. Transpedicular instrumentation was performed under C-arm X-ray. We performed short-segment instrumentation that included the vertebrae above and below the fractured vertebrae. According to literature, fixation was recommended to be placed in the fractured vertebrae instead of fractured pedicle or Magerl type A3.3 fractures with extensive vertebral body comminution. Consequently, fracture reduction was obtained by indirect manipulation. After the implantation of transpedicular instrumentation, careful dissection was carried out, and articular anatomy in those non-fused segments was preserved. In the fused spinal segment, the corresponding articular cartilage surface was removed and posterolateral arthrodesis beds were prepared. Autologous bone graft or allograft was added into the superior intervertebral articular process and interlaminar space. Moreover, a drainage tube was inserted and the wound was enclosed in layers.

Postoperative conductions
After surgery, external support was used (Jewett brace) for 2-3 months. Every patient was kept in a periodical clinical and radiological follow-up. After fracture healing was detected in the radiographs reports, the hardware removal process was initiated at about 6-12 months postoperatively in both groups.

Evaluations
Radiological measurements were carried out before and after surgery, as well as after the removal of instrumentation, and during late follow-up periods. All measurements were performed by two authors of this study, who took the mean value of the two measurements.
The heights of both anterior (AH) and posterior vertebral body (PH) were measured through lateral thoracolumbar radiography. The vertebral compression ratio (VCR) was calculated (VCR=[PH -AH]TT/PH).
The regional kyphotic angle was measured between the inferior, superior adjacent vertebra. The radiographic measurements of kyphosis were taken preoperatively, i.e., before the removal of instrumentation, 3 months after removal, and at the time of the latest follow-up.

Statistical Analysis
The data were presented as the mean ± standard error of the mean. Continuous data (compression Ratio, kyphosis angle, intervertebral mobility) were analyzed with the independent-samples t-tests.
The Greenough Low-Back score in the two groups were analyzed with the Mann-Whitney U tests based on assumptions of normality. Non-categorical data were analyzed with the χ2 test. A p-value <0.05 was used to define statistical significance. The IBM SPSS statistical software (Version 19.0) was used to complete the statistical analyses.

Results
Sixty-five patients with TLF from T12 to L2 fractures were enrolled in the study period: 33 in the Group SF and 32 in the Group BF. All the patients had an injury at the upper endplate or corresponding IVD in the vertebral column. No significant differences (p >0.05) were present in the demographic data between these two groups ( Table 1).
All patients were treated by transitory and transpedicular posterior instrumentation with a selective segment or bi-segments intervertebral fusion. However, the patients experienced fracture healing within 12 months after temporary fixation (mean follow-up time: 19.7 months) (Figure 2, 3).
The mean preoperative Greenough Low-Back Outcome Scores were 10.0±1.9 points and 10.5±1.9 points in group SF and BF, respectively; whereas, the post-operational scores were 65.0±2.0 and 65.2±1.8 points in group SF and BF, respectively. The clinical outcomes between the two groups were not significantly different ( Table 2). Nevertheless, one patient in Group BF suffered from a superficial infection after surgery, which was successfully treated with antibiotics. Another patient in the group SF and 3 patients in group BF reported pain at the bone graft donor site. No cerebrospinal fluid leakage, neurological injuries, vascular injuries or any other complications were identified during or after operations. During the follow-up of the patients, no cases of adjacent-segment disease were reported.
Radiological outcomes from 65 patients indicated that following the removal of the implant, the regional kyphotic angle progressed with time regardless of fusion. However, the results were not significantly different between the two groups (p>0.05) ( Table 3). One patient (50 female, T12 Fracture) had a progressive kyphosis less than 30 degrees without any clinical symptom in Group SF after implant removal. Therefore, we suggested a prolonged external support usage (Jewett brace 3-6 months) and anti-osteoporosis drug.
The residual intervertebral mobility of four segments (superior distal adjacent segment, superior proximal adjacent segment, inferior proximal adjacent segment, and inferior distal adjacent segment) after implant removal of patients in the SF and BF groups are shown in Table 4. No statistical differences were observed for superior proximal and distal adjacent segments between the two groups (P > 0.05, t test). However, statistical differences were observed for unfused inferior proximal and distal adjacent segments regardless of in T12, L1,or L2 fracture segments between the groups (P < 0.05, t test) ( Table 4).

Discussion
The thoracolumbar junction is susceptible to injury as it is the fulcrum for increased motion between Lumbar Spine [17,18]. Randomized trials indicated that surgically treated TLF burst fractures with reduction, and posterior fixation were sufficient; and hence, the use of bone grafting for definitive fusion was not necessary [16]. Furthermore, the restitution of intervertebral mobility of such an unfused segment after fracture healing could unload adjacent parts of the spine and reduce the risk of degeneration of these segments [8,9].
However, surgeons still recommend grafting for TLF, particularly for distraction injuries, fractures with extensive vertebral body comminution, and abnormal intervertebral disc signal [15]. Qian et al. In the present study, the enrolled patients had unstable TLF with kyphotic angle above 20 degrees and vertebral body height loss of more than 40%. In our study, the non-fusion procedures were not an optimal option for their treatment. Also, the literature suggested fused schemes only after an abnormal signal of intervertebral disc was prominent [7, 11]. Hence, we proposed the selective fusion or limited fusion schemes for the TLF with short-segment instrumentation, i.e., the vertebrae above and below the fractured vertebrae that preserved not only the motion of the spine postoperatively but also, stabilized the fracture, and restored the spinal alignment. The selection of the fused segment relied on the abnormal signal in the intervertebral disc and the injured endplate. Mostly, the upper endplate was susceptible to injury in TLF. Thereofore, we chose the fusion of upper endplate as the content of this study. According to clinical practice, it was found that the group with selective fusion had low surgical complications, blood loss, and surgical time. Significance progression of regional kyphosis could be observed along with the immediate postoperative kyphotic angle in both groups, however, these were not significantly different between the two groups. Based on previous reports, progressive kyphosis might be inevitable despite fusion. For our study, the residual deformity did not correlate with the symptoms of the patients at their time of follow-up [1,19]. The result showed that the selective fusion procedures achieved comparable clinical and radiographic outcomes with bisegment fusion groups, which not only validated the selectively fused schemes but also is appropriate for TLF treatment.
Recently, the researchers were curious to understand if unfused segments regained motion after instrumentation removal [9,11,20]. Some reports stated that the potential return to the mobile, non- In the present study, we compared the intervertebral mobility not only in the proximal adjacent fractured segment but also in the distal adjacent segments following temporary posterior fixation between the two cohorts of patients with single articular process fusion or bi-segments articular process fusion. The intervertebral mobility of superior distal and proximal adjacent segments had no significant difference between the two groups. By selective fusion of superior intervertebral articular process, the inferior distal and proximal adjacent segments in the selective fusion group regained more motion after instrumentation removal. The selectively fused scheme restored the spine and alleviated the adjacent vertebral degeneration [21].

Limitation
This study had several limitations. Firstly, patients with complicated injuries, such as progressive neurologic deficits, and anterior surgery, were excluded. Secondly, an increase of retained motion was observed in patients after hardware removal. However, a long-term comparative study is necessary to assess the functional and radiographic outcomes of such patients.

Conclusions
The strategy of selective fusion and short-segment instrumentation is reported to be reliable and useful for the treatment of the vast majority of TLF. The surgeon could fuse only the injured segments, and reliably avoid the bi-segment fusion. Motion in the un-fused inferior proximal and distal adjacent segment in the selective fusion group could be regained after instrumentation removal.

Competing interests
Conflict of interest The authors declare that they have no conflict of interest.

Funding
The research did not receive any specific grant from funding agencies.

Informed consent
Informed consent was obtained from all individual participants included in the study.

Acknowledgements
We are grateful to all study participants for their participation in the study.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Ethics approval and consent to participate
The study was approved by the ethical committee of institutional review board. All the patients gave written consent to for research applications of their clinical data. The patient data was anonymized in this study.

Consent for publication
The patients (all adults) all gave their explicit written consent to have all their outcome data, images and demographics included in this article write-up and for publication.