Single-stage transverse process resection, debridement, interbody fusion, and internal xation for the treatment of lumbar spinal tuberculosis via posterior-only approach

Background: Spinal tuberculosis (TB) is a less frequently reported infectious spinal pathology. There are controversies on the surgical intervention of lumbar spinal TB with neurological damage and paraspinal abscess. This retrospective study was conducted to determine the effectiveness of single-stage transverse process resection, debridement, interbody fusion, and internal xation for the treatment of lumbar spinal TB via posterior-only approach. Methods: From January 2015 to June 2018, 32 consecutive patients (19 males and 13 females) with lumber spinal TB complicated with neurological damage and paraspinal abscess treated by single-stage transverse process resection, debridement, interbody fusion, and internal xation were enrolled. Medical records, imaging studies, laboratory data were collected and summarized. Anti-TB drugs with HREZ chemotherapy regimen was administered to all patients. Surgical outcomes were evaluated based on visual analogue scale (VAS), American Spinal injury Association (ASIA) classication. The changes in C-reactive protein (CRP) levels, erythrocyte sedimentation rate (ESR), clinical symptoms and complications were investigated. Graft fusion was evaluated using Bridwell grading criteria. Results: The mean follow-up period was 20.41 ± 5.19 months. No implant failures were observed in any patients. Wound infection was observed in one patient. Solid bony fusion was achieved in 9 cases at 6 months and 23 cases at 12 months after operation. Kyphosis angle was 11.28 ± 4.01° at nal follow-up. The levels of ESR and CRP were returned to normal at the nal follow-up. VAS scores were signicantly improved (P < 0.05). According to ASIA classication, 6 cases were classied as with grade D and 26 cases were classied as grade E at the last follow-up. Conclusion: Single-stage transverse process resection, debridement, interbody fusion, and internal xation via posterior-only approach is a feasible and effective surgical therapy for lumbar spinal TB with neurological damage and paraspinal abscess.


Background
Tuberculosis (TB), a potentially serious infectious disease, remains a severe impact on human health, particularly in developing countries [1,2]. Skeletal TB is 10% of the extrapulmonary TB, of which spinal TB accounts for approximately 50% [3,4]. Spinal TB, the most common pattern of extrapulmonary TB, is a severe spinal disease characterized by spinal cord compression, abscess formation, and kyphotic deformity, frequently causes neurologic de cit and even paraplegia [5][6][7]. Spinal TB often have evident collapse of vertebra because of TB destruction, and it occurs most commonly in the lumbar region [8,9].
Although there has been a signi cant evolution in spinal TB treatment during the past several decades, anti-TB therapy and external immobilization remain the irreplaceable treatment options in most patients of spinal TB [10,11]. However, cases with neurological dysfunction, spinal instability, abscess formation, kyphotic deformity and failed response to conservative treatment may require surgical intervention [12]. Diverse surgical approaches have been performed for the treatment of spinal TB patients, including an anterior-only, posterioronly or combined anterior-posterior approaches [13]. The posterior-only approach with the instrumentation system has been proven to be effective for treating various thoracic and lumbar spinal disorders with the strategy of more conservative and less invasive in recent years [14,15].
Despite the fact that posterior-only approach has been widely used in spinal TB treatment, there is no report on the feasibility and safety of single-stage transverse process resection, debridement, interbody fusion, and internal xation for the treatment of lumbar spinal TB. The objectives of our study are: 1) to present the indications of single-stage transverse process resection, debridement, interbody fusion and pedicle screw xation in the treatment of lumbar spinal TB via posterior-only approach; 2) to evaluate the outcomes of singlestage transverse process resection, debridement, interbody fusion and pedicle screw xation in the treatment of lumbar spinal TB via posterior-only approach.

Methods
Inclusion and exclusion and general information Written informed consent was obtained from all patients, and this study protocol was approved by the Ethics Committee of the First A liated Hospital of Xinjiang Medical University. From January 2015 to June 2018, 32 patients with lumber spinal TB complicated with neurological damage and paraspinal abscess received surgery by the same surgical team. Among these patients, 19 of them were males and the other 13 were females, aging from 16 to 65 years old with an average of 41.21 years old. Surgery was considered in the presence of the following indications: (1) patients with lumbar TB were clearly diagnosed, and the lesions were located in L1-5; (2) progressive neurological de cit and paraspinal abscess that were unresponsive to chemotherapy for 2 months; (3) patients with paraspinal abscess without infusing into the presacral and iliac fossa; (4) with moderate and severe kyphosis (kyphosis angle < 60°); (5) multilevel vertebrae were involved (less than 3 levels) with signi cant vertebra destruction or collapse (≤ the intact vertebral height). Patients that presented with the following conditions were excluded: (1) complicated with open pulmonary TB and acute miliary TB; (2) cervicothoracic, thoracic and lumbosacral spinal TB; (3) multilevel lesions that require anterior long-segment bone fusion; (4) patients with severe cardiopulmonary diseases and could not tolerate surgery. Preoperative Xray, CT and MRI showed different degrees of bone destruction, and the lesions were located in L1-3 (6 cases), L2-3 (12 cases), L3-4 (10 cases), and L4-5 (4 cases). All the patients had different degrees of low back pain as the rst symptoms, of which 18 cases were accompanied with TB poisoning symptoms such as low fever, night sweats and fatigue, and 10 cases had di culty walking due to pain. Of the 32 patients, 22 cases were complicated with paraspinal abscess, 22 cases had symptoms of spinal cord injury. According to the American Spinal injury Association (ASIA) classi cation, 4 cases were classi ed as with grade C and 18 cases were classi ed as grade D. Twenty-one cases had different degrees of kyphosis with a preoperative Cobb angle of 28.97 ° ± 8.28 °, (range, 12 °-46 °). The preoperative Visual analogue scale (VAS) score was 5.81 ± 1.03, (range, [4][5][6][7][8]. The erythrocyte sedimentation rate (ESR) was 44.44 ± 10.24 mm/h, and the C-reactive protein (CRP) was 35.75 ± 22.19 mg/L.
Preoperative procedure A chemotherapy regimen was administered to all patients. Anti-TB drugs with HREZ chemotherapy regimen that consisted of isoniazid (300 mg/day), rifampicin (300 mg/day), ethambutol (500 mg/day), and pyrazinamide (750 mg/day) was administered 2-4 weeks before operation. The ESR, CRP, and temperature needed to decrease signi cantly before the surgical intervention was performed.

Operative technique
After administration of general anesthesia with endotracheal intubation, all the patients were placed in a prone position on the spinal surgeon table. The paraspinal muscles were stripped from the spinous process to the outer margin of the articular processes after a standard dorsal midline incision was performed. In general, the pedicle screws were inserted into the two superior and two inferior healthy vertebrae. The correct position of screws was evaluated with the assistance of intraoperative C-arm uoroscopy. The internal xation can also be adjusted according to the vertebral lesions. A pre-bent long rod was installed and tightened on the side with less abscess and mild lesions to ensure the spinal stability and avoid spinal cord injury during decompression and debridement. Debridement and bone grafting were performed on the side where the paraspinal abscess, lesions, and vertebral collapse were relatively severe. The transverse process of vertebral body on the lesion space was exposed and removed ( Fig. 1). Furthermore, the articular processes above intervertebral foramen were removed to expose the nerve roots. The psoas major muscle was exposed carefully, and the abscesses were debrided. After routine exposure, the caseous necrosis and granulation tissues, necrotic discs and infected endplates, sequestrated bone within the vertebral body and collapsed vertebrae were debrided with curettes as thoroughly as possible via paravertebral space. A large amount of normal saline was utilized for surgical area irrigation to clear the residual tuberculous tissue following completed hemostasis. Afterwards, a pre-bent rod was temporarily installed on the other side and the rod installed previously was removed. The same operation was performed on the opposite side of the lesions if necessary. The suitable autologous iliac bone was mixed with 0.2 g streptomycin and the posterolateral fusion using autograft was performed. The compression and stretch of the internal xation instruments were used to rectify the kyphosis deformity cautiously and slowly.
Sequentially, the rods were tightened under pressure and checked the nerve root. Thereafter, 1.0 g streptomycin was administered accurately into the operative area. The drainage tube with negative pressure was routinely placed in the operative region before the incision sutures performed nally. All the resected specimens were sent for histopathologic examination.
Postoperative care Vital signs, motor, and sensory functions of both lower limbs were observed after the operation. When the drainage volume was less than 100 ml in 24 hours, the drainage tube could be pulled out. One day after the drainage tube was removed, the X-ray of the anterior and lateral position of the lumbar vertebrae were reexamined. Ambulation was allowed gradually with the protection of custom-made thoracolumbar brace. The anti-TB therapy was continued for 12 months after operation according to the liver function.

Results
All the 32 patients completed the operation successfully. No nerve and macrovascular injury were observed during the operation. There was no cerebrospinal uid leakage and aggravation of neurological function after operation. The operation time was 191.91 ± 28.85 min (range 150-260 min), blood loss was 527.93 ± 156.53 ml (range 260-950 ml) ( Table 1). Follow-up duration of all the 32 patients was 20.41 ± 5.19 months (range 12-33 months). Kyphosis angle was 11.28 ± 4.01° at nal follow-up with a loss of correction of only 0.25 ± 0.12°. This continued to signi cantly improve compared to preoperative measurements (P < 0.05).
Neurologic de cits in all patients improved at the nal follow-up examination. Results were evaluated by ASIA classi cation during the nal follow-up, 6 cases were classi ed as with grade D and 26 cases were classi ed as grade E (Table 2). Statistical analysis revealed that there was a signi cant difference between pre-operation and the nal follow-up (P < 0.05). Six patients revealed incomplete neurological function attributed to delayed diagnosis. Intervertebral bone graft and intertransverse fusions were performed in all patients. Lateral X-ray or CT was used to assess the fusion and formation of the bone bridge according to the criteria of Lee et al [16]. All patients achieved bone fusion within 5.3 ± 2.8 months after surgery (Fig. 2). The bone fusion was performed in 9 cases at 6 months and 23 cases at 12 months after operation. Average pre-treatment for ESR and CRP was 44.44 ± 10.24 mm/h and 35.75 ± 22.19 mg/L, respectively; which returned to 24.09 ± 6.03 mm/h and 10.59 ± 4.13 mg/L postoperatively, and 9.47 ± 3.16 mm/h and 5.13 ± 2.83 mg/L at the time of the nal follow-up. The difference in ESR and CRP between the preoperative period, postoperative period, and during nal follow-up was signi cant (P < 0.05). In 31 patients, the incision healed in one stage, and the bone graft in the lesion space achieved bony fusion. One patient recurred on the 40th day after the operation, the incision had chronic sinus formation, and the cause of recurrence was anti-TB drug resistance. Complete recovery was achieved after adjusting the anti-TB therapy and the reoperation of debridement. No failure of internal xation was found in 32 patients during the follow-up period.

Discussion
Our study shows that single-stage transverse process resection, debridement, interbody fusion and pedicle screw xation via posterior-only approach is an effective and feasible approach for the treatment of lumbar spinal TB with kyphosis deformity and neurological de cits. However, standard anti-TB therapy, strict bed rest, and supportive therapy remain the fundamental approaches for treating spinal TB. Surgical intervention is only recommended for lumbar spinal TB patients with abscess formation, spinal cord compression, signi cant kyphosis deformity, and neurological dysfunction.
Spinal TB accounts for almost half of the bone and joint TB, mainly affecting the anterior and middle column of the spine and leading to vertebral bone defects, collapse, compression, and kyphosis deformity [17]. The anterior-only approach is preferred for decompression and debridement in spinal TB as it allows direct access to the lesion site, complete debridement, su cient decompression, and reduces muscle trauma [18]. However, the high frequency of complications such as pseudarthrosis, ineffective correction of kyphosis and maintenance of the correction, unsatisfactory neurological function, and vascular injuries overwhelm its advantages [19]. Many researchers reported that patients treated by the anterior-only approach experience more signi cant blood loss, longer duration of the operation, and the hospitalization period than that of the posterioronly approach [20]. Combined anterior and posterior surgery has become popular due to its bene cial clinical outcomes [21]. However, when poor conditions complicate the aged, it would be di cult to tide over the serious trauma such as more signi cant loss of blood, longer operation time, and approach-related complications [22].
Furthermore, no literature reports in the surgical management of lumbar spinal TB by single-stage transverse process resection, debridement, interbody fusion, and internal xation via a posterior-only approach.
Controversies on the strategy of a posterior-only approach in treating lumbar spinal TB mainly concentrated in whether surgeons can throughly perform debridement and anterior decompression in such a limited visual eld, whether it would achieve the anterior bony fusion, and whether it would maintain the spinal stability [23,24]. A number of advantages to the posterior-only approach were highlighted: reduced bleeding, shorter operation and hospitalization durations, released the nerve compression, corrected kyphosis deformity, regained spinal stability, and improved the quality of life of these patients [25]. These results re ect those of Abulizi et al. who also found that single-stage transforaminal decompression, debridement, interbody fusion, and posterior instrumentation is an effective and safe surgical for the treatment of spinal infection [14]. Additionally, as far as the patients with less involved spinal TB for the anterior column that is mainly affected by TB achieving spontaneous fusion are concerned, the posterior-only approach may be a better strategy [26,27].
Although previous studies have demonstrated that the translaminar debridement and transforaminal bone graft are feasible with minimal damage to the posterior column, the operation still results in a certain degree of damage to the lamina or the superior and inferior articular processes [28]. Therefore, some investigators have several concerns: (1)  be achieved [29]. Additionally, this method of long-segment bone graft with interlaminar decompression and posterior lumbar interbody fusion (PLIF) approach is infeasible [30]. Therefore, OLIF fusion technique can only be used for interbody fusion. According to these two ideas, the problems of lesion debridement, bone grafting in the anterior and middle column can be treated poster laterally. Combine the absolute advantages of posterior internal xation and correction of the deformity, we believe that this surgical approach can be used as a surgical choice for lumbar spinal TB.
This study holds several limitations. The potential risk of TB spreading to the healthy posterior regions is concerned in this surgical therapy, as posterior debridement can result in diffusion of infection and stulas.
Fortunately, this complication was not found in our research. Our study included 32 patients, future study should include multicentre studies and larger sample size to con rm our results. Furthermore, some patients with missing data or lost to follow-up were excluded from the analysis. The future management of these patients should be enhanced with further long-term follow-up data to closely monitor the effect of this surgical option.

Conclusion
The present study demonstrated that single-stage transverse process resection, debridement, interbody fusion, and internal xation via posterior-only approach could be a feasible and effective treatment therapy for most patients with lumbar spinal TB and associated with great neurologic recovery, correction of kyphosis deformity.  Preoperative coronal and sagittal CT-scan showed bone destruction, vertebral collapse, and disc space