A Comparative Study on Ecacies of Posterior Microscopic Mini-open and Open Technique for Thoracolumbar Burst Fractures Accompanied With Severe Traumatic Spinal Stenosis

Purpose To compare the ecacies of minimal invasive decompression by posterior microscopic mini-open technique combined with percutaneous pedicle xation (hereafter MOT) and traditional open surgeries in patients with severe traumatic spinal canal stenosis resulting from AO Type A3 or A4 thoracolumbar burst fractures and provide references for clinical treatment. Methods The clinical materials of 133 patients with severe traumatic spinal canal stenosis caused by AO Type A3 or A4 thoracolumbar burst fractures who underwent MOT (group A) or traditional open surgery (group B) were retrospectively enrolled. The patient demographic and radiological data were analyzed between the two groups.


Introduction
90% of all spinal fractures are related to the thoracolumbar region, and burst fracture is a common form in this region, accounting for approximately 60% [1]. Thoracolumbar fractures can make serious spinal cord injury [2]. Decompression surgery can help patients limit their secondary spinal cord injury and improve their neurological recovery after acute spinal cord injury. Short-segment posterior xation, especially percutaneous minimally invasive xation, for thoracolumbar fracture are well-accepted [3][4][5][6], but it is di cult to decompression and bone grafting for patients with severe traumatic spinal canal stenosis resulting from thoracolumbar burst fractures, when mid-sagittal canal diameter compression ratio (MSDCR)>50% [3,7]. Clinical and radiographic records All patient clinical and radiographic data were recorded at admission, postoperative, and last follow-up (12-24months after surgery). Clinical records included general date, follow-up time, injury mechanism, hospital stay, operative time, intraoperative bleeding volume, visual analogue scale (VAS) score[8], American Spinal Injury Association (ASIA) scores [9], AO spine injury classi cation [10].

MOT methods
After general anesthesia, the patient was placed in a prone position and the abdomen was suspended.
Percutaneous pedicle screw xation was conducted under uoroscopic guidance, where percutaneous pedicle screws were placed in the superior and inferior vertebra adjacent to the injured vertebra. An approximately 3cm posterior midline incision, centered by the injured vertebra, was performed to expose lamina space of the injured vertebra. Then fenestration of the vertebral lamina was performed via the 3cm incision under microscopy, followed by observing and pushing the fracture fragments into the vertebral body using the L-shaped operative tool, followed by longitudinal distraction of the vertebral body anterior margin. Subsequently, using a curette to nd the fracture lines at the posterior wall of the injured vertebral body and achieve the anterior and middle column reduction and manufacture a diameter approximately 5mm bone grafting channel. After con rmation of satisfactory reduction using uoroscopy, autologous bones or allograft bones were implanted in the injured vertebral body via bone grafting channel, until 3-4mm deep from the posterior wall of the injured vertebral body. After adequate hemostasis, the wound was ushed, and the drainage tube was placed as necessary, incisions were sutured layer by layer. See gure 1.

Statistical analysis
Statistical analysis SPSS20.0 statistical software (IBM Corp., Armonk, NY, USA) was used for data analysis. Quantitative data were presented as mean ± standard deviation. Within-group comparisons of PHR, Cobb angle, MSDCR and VAS score at multiple time points were used repeated measurement analysis of variance combined with post Bonferroni test. Mann-Whitney U-test was used for ASIA scores within-group comparisons. Age, follow-up time, hospital stay, operative time, intraoperative bleeding volume, PHR, Cobb angle, MSDCR and VAS score between two groups were compared by independentsample t tests. Gender, injury mechanism, injured level, AO classi cation between two groups were compared by Chi-square test. A probability less than 0.05 was considered statistically signi cant.

Comparison of general conditions
A total of 64 patients were nally recruited, 69 patients with incomplete clinical records or lost to follow up were excluded. These 64 patients included 28 patients had treatment with MOT were classi ed as group A (21 males and 7 females), and 36 patients had treatment with traditional open surgeries were classi ed as group B (25 males and 11 females). Statistical analysis revealed that there were no signi cant intergroup differences in gender, age, follow-up time, injury mechanism, injured level, AO classi cation, ASIA score, and MSDCR between the two groups (all P>0.05, Table 1). All procedures were completed successfully. No signi cant difference was found concerning duration of hospital stay between both groups (P>0.05, Table 2). The mean operative time was longer and the mean intraoperative bleeding volume was less in group A when compared to that in group B. The difference was statistically signi cant (all P<0.05).

Radiographic ndings
The PHR, the Cobb angle, and the MSDCR in the two groups at post-operation and last follow-up were signi cantly improved when compared with the preoperative value (all P<0.05, Table 3). The PHR, the Cobb angle in the two groups at the post-operation and last follow up without signi cantly different (P>0.05), but the MSDCR was improved at last follow up when compared with the value at post-operation (P<0.05).
The PHR in group B was heavier than in group A at the pre-operation (P<0.05), but there was no signi cantly difference at post-operation and last follow up (P>0.05). The Cobb angle in group B was heavier than in group A at the pre-operation (P<0.05), same as above, there was no signi cantly difference at post-operation (P>0.05), but the Cobb angle in group A was well maintained than in group B at last follow up (P<0.05). The MSDCR in two groups without signi cantly difference at pre-and postoperation (P>0.05), however, the MSDCR in group B at last follow up improved better than in group A (P<0.05). Visual analogue scale (VAS) score There was no signi cant difference in VAS score between the groups before operation (P>0.05, Table 4).
The VAS score on postoperative day 1 and at the last follow up were signi cantly lower than the respective preoperative values in both groups (P < 0.05). However, the improvement in the VAS score was signi cantly more favorable in Group A than in Group B (P < 0.05).

American Spinal Injury Association (ASIA) scores
The pre-and last follow-up differences in the two groups were statistically signi cant for ASIA scores comparisons (P < 0.05, Table 5).  A 47-year-old male patient was diagnosed AO type A4 fractures at the L2 level with a large posterior wall retropulsed fragment into the spinal canal causing signi cant spinal canal encroachment, and had ASIA E neurological status. After MOT treatment, the prevertebral height and Cobb angle improved, and the fragment in the spinal canal had good reduction. At the time of last follow-up, the patient had ASIA E neurological status, and the radiographic results showed good fracture union. Although partial bone graft resorption was observed, the prevertebral height, Cobb angle and the spinal canal patency were wellmaintained, see Figure 2.

Discussion
The standard of severe traumatic spinal stenosis caused by thoracolumbar burst fractures The standard of severe traumatic spinal stenosis of thoracolumbar fractures remains controversial.
Wolter [12] considered that retropulsed fragment into the spinal canal cause MSDCR > 2/3 as severe spinal stenosis. Meves et al. [13] believed that there is a positive correlation between narrowing of the spinal canal and the severity of the incomplete neurological de cit. When patients with 25, 50, and 75% narrowing of the thoracolumbar spinal canal, the probability of neurological de cit may be 12, 41, and 78%, and in the lumbar spinal canal it was 8, 30, and 68%, respectively. Based on our previous experience, when MSDCR > 50%, patients have a high probability of neurological de cit, coincided with the ndings of previous studies of Mohanty et al [14]. We therefore recommend that MSDCR > 50% is a standard of severe traumatic spinal stenosis, because of di culty in fracture reduction and high probability of The signi cance of using a surgical microscope in MOT Because of the incision for decompression was small and deep, insu cient view of the surgical area and inadequate light results the decompression is so di cult and not safe that clinicians may choose a traditional open surgery. The surgical microscope can compensate for the above limitations. The major advantages of the microscope include better illumination, magni cation and coaxial vision, these can avoid spinal cord and nerve root injuries as well as dural lesions [15]. The intraoperative bleeding volume of MOT was less than traditional open surgeries (P<0.05). One major reason stem from using surgical microscope, as the perivertebral venous plexus anatomy be identi ed intraoperatively and controlled with compression hemostasis and electrocautery accurately [16].

Details of decompression and reduction
Depending on the severity compression of the nerve root and spinal cord, different order of decompression and reduction should be considered [17]. When neural tissue be compressed seriously or stuck by bony fragments in the spinal canal or fractured lamina, especially at the T11-12 vertebral level, the decompression and reduction should in the order of: fenestration → longitudinal distraction → reduction by using a curette. When neural tissue without serious compression or incarceration, the decompression and reduction should in the order of: longitudinal distraction → fenestration → reduction by using a curette. Fenestration should be performed on the side of neural tissue be compressed or injured seriously. Unilateral fenestration is recommended unless bilateral lamina or lateral wall of the spinal canal were fractured seriously.

Bone graft skills
In our passed study, we found sagittal fracture lines existed over the pedicle horizontal in thoracolumbar fractures accompanied with loss of vertebral body height. This region is overlap with the interlaminar space, so a 3cm incision at this region is su cient for decompression, reduction and bone graft. Because of the adjacent discs and PLC were not severe damaged, the MOT does not require posterolateral fusion or interbody fusion, the xation by posterior approach and bone graft can make the intervertebral spaces does not change signi cantly, then anterior column will be spontaneous fusion [6,[18][19][20]. In this study, the postoperative PHR and Cobb angles improved signi cantly (P<0.05) and remained stable at last followup (P<0.05). These results coincide with the points of view above.

Treatment strategy for the spinal canal
It is not necessary to achieve the anatomic reduction if reduction is di cult, to avoid damage to the spinal cord or nerves. Miyashita et al. [21]found that there is no signi cant effect on the recovery of neurological function when MSDCR < 30%. When decompression removed the fragments or soft tissues which compress the neural structures, bone resorption can complete spinal canal remodeling. This is related to intraspinal neurilemmoma and venous pulsation [22].

Declarations
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
This study was conducted in accordance with the declaration of Helsinki. This study was conducted with the approval from the Ethic committee of A liated Hospital of Zunyi Medical University.
Consent for publication written informed consent to publish the clinical details and images of the patient was obtained. Figure 1 MOT methods a & b. Four guide needles were inserted into the adjacent vertebral pedicles under uoroscopic guidance; c. the positions of the percutaneous pedicle screws were con rmed radiographically after implantation; d. fenestration of the vertebral lamina was performed via the 3cm incision under microscopy; e. observing and pushing the fracture fragments into the vertebral body under microscopy; f. using a curette to achieve the anterior and middle column reduction; g. the satisfactory reduction was con rmed by lateral radiograph; h. autologous bones and allograft bones were implanted in the injured vertebral body under microscopy; i. a drainage tube was placed and incision was intradermal sutured.