One-stage posterior debridement, titanium mesh cage bone grafting and single-segment �xation for treating mono-segmental lumbar and lumbosacral spinal tuberculosis in adult patients following minimum 5-year follow-up

Background To evaluate the mid-long term outcomes of surgical management of mono-segmental lumbar and lumbosacral spinal tuberculosis (TB) by one-stage posterior debridement, titanium mesh cage bone grafting and single-segment �xation. Methods A total of 62 patients with mono-segmental lumbar or lumbosacral spinal tuberculosis were enrolled. One-stage posterior debridement, titanium mesh cage bone grafting and single-segment �xation was performed. Clinical and radiographic outcomes were compared and analyzed.


Background
According to the global TB report in 2019 by World Health Organization, TB is the second most common fatal infectious diseases.There are more than 10 million new TB cases worldwide in 2019, causing approximately 1.2 million deaths [1].As the most common extrapulmonary TB, spinal TB accounts for approximately 50% cases of skeletal TB [2].With the increasing rate of HIV infection and TB drug-resistant strains, the prevalence of spinal TB continues to rise recently.Lumbar and lumbosacral spines are the most common sites for spinal TB.This severe disease is accompanied by a high refractory, disability, and recurrence rate, which seriously in uences the quality of patients' life [3,4].
Effective anti-TB drugs are still the basic therapy throughout the entire treatment process for most patients with lumbar and lumbosacral spinal TB [5].However, appropriate surgical treatment is the key to improve the cure rate.
The aim of surgery is to remove lesions, improve neurological function and reconstruct spinal stability [6].With the introduction of the spinal pedicle screw system, one-stage posterior approach has been increasingly adopted to treat lumbar and lumbosacral spinal TB by surgeons [7][8][9].The rigid internal xation system is able to prevent the kyphosis progression and severe back pain caused by spinal instability.For the treatment of mono-segmental lumbar and lumbosacral spinal TB in adults, the xation including one or more normal motor units can be performed via posterior decompression and instrumentation.Although this procedure provides rm temporary stability, it also limits the activity of normal spinal motor unit and accelerates the degeneration of adjacent vertebral body [10][11][12].Therefore, selection of appropriate xed segment will determine the mid-long-term effect.In this study, we evaluated the effect of surgical treatment of mono-segmental lumbar and lumbosacral spinal TB with one-stage posterior debridement, titanium mesh cage bone grafting and single-segment xation in adult patients following minimum 5-year follow-up.

Basic information
This study enrolled 62 patients with mono-segmental lumbar and lumbosacral spinal TB, who were treated by onestage posterior debridement, titanium mesh cage bone grafting and single-segment xation between January 2010 and December 2014.The inclusion criteria were as follows: (1) the lesion mainly involved a functional unit of the lumbar and lumbosacral spine (L2-S1); (2) vertebral body damage was less than 1/2 of the vertebral height so that the pedicle screw can be implanted; and (3) The paravertebral abscesses were limited to the diseased vertebrae.The exclusion criteria were as follows: (1) severe kyphosis caused by vertebral bone destruction; (2) severe osteoporosis in senile patients; (3) bone healed tuberculous spondylitis associated with associated with kyphosis deformity; and (4) huge paravertebral abscess or psoas abscess.The initial diagnosis of spinal TB was made based on the clinical presentations, hematologic examinations, and imaging ndings, and the nal diagnosis was made based on the pathological examination and tubercle bacillus culture.Patients who met the inclusion criteria were followed up for at least 5 years.
The demographic data and disease characteristics of patients are summarized in Table 1.All patients presented with constitutional symptoms such as night sweats, anorexia, weight loss, fatigue, and low back pain.Some patients also had symptoms of neurological impairment including sensory impairment and muscular weakness.All patients had elevated ESR and CRP values.Pain severity was assessed by VAS.The JOA and ODI were used to evaluate neurological function and the quality of life, respectively.
All patients underwent routine imaging examinations.Plain radiograph, computed tomography (CT), and magnetic resonance imaging (MRI) were used to detect vertebral body collapse, spinal instability, bone destruction, epidural and paravertebral abscess formation, and narrowing of the intervertebral space.Lesions with segments L2/3 L3/4, L4/5, and L5/S1 occurred in 11 cases, 17 cases, 21 cases, and 13 cases, respectively.The University of California at Los Angeles (UCLA) grading scale [13] was applied to assess the ASD on radiograph.

Preoperative procedure
All patients were treated with anti-TB drugs for two to four weeks prior to surgery, including isoniazid (300 mg/day), rifampicin (450 mg/day), and pyrazinamide (750 mg/day), and ethambutol (750 mg/day).Preoperative examinations were performed to exclude contraindications.The poor conditions with anemia and hypoproteinemia were corrected preoperatively, the targets were blood pressure below 140/90 mmHg and random blood glucose lower than 10.0 mmol/L.Surgery was performed when the ESR value returned to normal or signi cant decreased.Patients with progressive radiculopathy or syndrome of cauda equine were treated with adequate anti-TB drugs.If the patients had no absolute contraindications, surgery was performed as soon as possible.

Operation procedure
The patients were placed in a prone position under general endotracheal anesthesia.A posterior midline incision was made over the diseased vertebra, and the spinous processes, lamina, facet joints and transverse processes were exposed.With the assistance of C-arm uoroscopy, pedicle screws were implanted in the affected vertebrae close to the endplates in order to preserve enough debridement space.A temporary rod was stabilized on the side of less bone destruction or neurologic manifestation to avoid nerve injury during focal debridement.Then, hemilaminectomy or laminectomy was performed on the severely damaged side of the lesion segment to expose the diseased vertebral bodies.Curettes with different angles were used to remove the lesion focus, including sequestrum, necrotic intervertebral disc, tuberculous granuloma and caseous necrosis.The pus and necrotic tissue were eliminated by negative pressure washing via a soft silicone tube which was placed deeply into the lesion.The same procedure was performed on the other side if necessary.The bone surfaces of the vertebral body were repaired as bone graft beds, and autogenous bone particles from healthy lamina and spinous process packed in one or two shaped titanium mesh cages were implanted into the front 2/3 of the intervertebral to reconstruct the anterior middle column.Then, the opposite rod was installed and tightened.Streptomycin (0.1 g) and isoniazid (0.3 g) were applied locally in the focus area, and the incision was closed in layers after a drainage tube was placed.The specimens were collected for mycobacterium culture and histopathological examination.

Postoperative procedure
Routine antibiotics were used to prevent infection.All patients received nutritional improvement and support treatment.Dressing was changed regularly in the wound, and the drainage tube was removed when the volume of drainage was less than 20 ml in 24 hours.The patients continued with the anti-TB treatment regimen for 12 to 18 months.Routine blood test, liver function test, ESR and CRP were performed to monitor adverse reactions and to evaluate drug e cacy.After strict bed rest for four weeks, the patients gradually began to walk with the help of external braces for three months.Clinical and radiologic examinations were performed every three months in the rst year after surgery and then every six months thereafter.

Follow-up evaluation
Lumbar anterolateral radiograph and CT were performed to assess the placement of graft and internal xation.CT was used to evaluate bone healing according to the radiologic criteria by Lee et al. [14].The following indexes were recorded preoperatively, postoperatively, and during the follow-up:

Statistical Analysis
Statistical analyses were performed using the SPSS 20.0 software.The clinical and radiographic data between preoperative and postoperative were compared using paired t test.Discrepancy of the normal distribution was evaluated using the rank sum test.P value of less than 0.05 was considered statistically signi cant.

Clinical outcomes
Tuberculous granulomas or caseous necrosis was con rmed by pathological examination of the surgical specimens.The average follow-up time was 75.0 ± 11.5 months.All patients were diagnosed with lumbar or lumbosacral spinal TB, and achieved clinically cure.Preoperative CRP and ESR were 43.5 ± 14.9 mg/L and 69.1 ± 17.8 mm/h, respectively.The values returned to normal at three months postoperatively.For patients with preoperative neurological dysfunction, neurological function was improved to varying degrees after surgery.The JOA was improved from preoperative 18.3 ± 3.7 to 26.9 ± 2.2 at the nal follow-up (P 0.05).The VAS and ODI values were 6.9 ± 1.1 and 42.6 ± 6.2 preoperatively and signi cantly decreased to 1.0 ± 0.8 and 10.2 ± 1.6 at the last follow-up, respectively, (P 0.05) (Table 2).

Radiographic outcomes
The lordosis angle and lumbosacral angle were increased from preoperative 20.4 ± 2.9° and 14.7 ± 3.4°to 32.8 ± 3.6°a nd 22.4 ± 5.5° immediately postoperative, and resulted in 1.0 ± 0.7° and 0.8 ± 0.7° of angle loss at the nal followup, respectively.The lordosis angle or lumbosacral angle was signi cantly improved immediately postoperative and at the last visit compared with preoperatively (P 0.05).The mean fusion time was 9.8 ± 2.6 months (Table 3).There was no signi cant difference between preoperative and postoperative adjacent segment DH (Table 4).According to the UCLA grading scale, 52 patients had grade I ASD and 10 patients had grade II ASD preoperatively.At the nal follow-up, 8 patients progressed from grade I ASD to grade II ASD, and 2 patients from grade II ASD to grade III ASD.No nonunion, pseudoarthrosis, loosening or fracture of instruments occurred at the last visit (Figs. 1, 2).

Complications
Super cial wound infection occurred in nine patients and was cured by antibiotics.Four patients experienced hypostatic pneumonia and were treated with sensitive antibiotics according to the results of sputum culture.Local abscess recurrence due to irregular administration of anti-TB drugs occurred in one patient, who was treated by catheter drainage through minimally invasive incision and regular chemotherapy.No operative mortality or permanent nerve injury occurred.

Discussion
The local anatomical structures of lumbar region are complex and adjacent to many vital organs.Lumbar stability is maintained by the combined effects of the vertebrae, intervertebral discs, muscle groups, and ligaments against the strong pressure and shear stress.Among all spinal regions, lumbar and lumbosacral segments subject to the largest load and exhibit the greatest mobility.Lumbar and lumbosacral spinal TB has the characteristics of insidious onset and atypical symptoms at early stages.Patients often present with low back pain and are easily misdiagnosed with degenerative spinal diseases such as lumbar disc herniation, lumbar spinal stenosis and osteoporosis [15].With the progression of the disease, the sequestrum and abscesses can lead to vertebral collapse, spinal instability, kyphosis deformity, and changes in normal physiological curvature and load biomechanics, or invade the spinal canal to compress the nerves, resulting in nerve damage or even paralysis [16].
There are many surgical approaches for treating lumbar and lumbosacral spinal TB [17][18][19][20].The posterior approach is associated with some advantages such as lesion removal, and simultaneous intervertebral bone graft and xation through one incision without position changing.Additionally, pedicle screws can effectively correct the kyphosis via the posterior approach, thus restoring the normal physiological lordosis of the lumbar segment.Furthermore, the posterior approach is less invasive and avoids possible damage to the large vessels, nerves or vital anatomical structures.Therefore, many surgeons recommend posterior-only approach for lumbar and lumbosacral spinal TB, which achieves good curative effects [7][8][9]20].The rationality of the posterior approach is to remove the sequestrum that prevents anti-TB drugs from entering the lesions.The small amount of residual lesions and abscesses can be absorbed following postoperative standardized anti-TB chemotherapy.
Previous studies have shown that the range of xation for mono-segmental lumbar and lumbosacral spinal TB in adults involves one to multiple normal motion units above and below the lesion via posterior debridement, decompression and instrumentation.Although multi-segment xation provides strong temporary stability, it sacri ces the motion of the xed segments and accelerates degeneration of adjacent segments [21,22].Since the lumbar region requires greater mobility than the rest of vertebral column, adopting a surgical approach with minimally invasion and less damage to the vertebral function is the major goal for treating mono-segmental lumbar and lumbosacral spinal TB.Previous studies have reported that the spinal motor unit remains essentially unchanged following single-segment xation with pedicle screws in the treatment spinal fractures.Spinal stability in patients with fractures of spine can be achieved by mono-segment xation [23].For the treatment of lumbar and lumbosacral spinal TB, single-segment xation with pedicle screw is feasible because of the reactive new bone formation in the affected vertebrae with TB.The involved vertebrae often presents with sclerotic bones, resulting in higher bone mineral density than normal vertebral body.This pathological process results in stronger holding forces for pedicle screws in the involved vertebrae compared with fracture [24].Liu et al. [25] compared the results of mono-segment xation versus short-segment xation for the treatment of single-segment lumbar spinal TB and found that mono-segment xation was more suitable since the normal motion was preserved.Xu et al. [26] reported that single-segment xation and bone fusion was effective in maintaining the stability of the spine and retaining normal motion units.In the present study, the lordosis and lumbosacral angle was increased from preoperative 20.4 ± 2.9° and 14.7 ± 3.4° to postoperative 32.8 ± 3.6° and 22.4 ± 5.5.Our ndings suggest that deformity was effectively corrected.The characteristic of lumbar and lumbosacral lordosis and longitudinal arrangement of facet joints, which takes buffering effect of kyphosis caused by bone destruction.Therefore, mono-segment xation is su cient for relatively small orthopedics.In addition, there was no signi cant difference between the preoperative and nal adjacent segment DH and the ASD rate was lower in this study (16.1%) than the incidence of ASD (range 21.3-31.9%)after lumbar fusion reported by the recent meta-analysis [27].These results suggest that monosegment xation can maintain the normal motion units and to some extent, retard the degeneration of adjacent segment.
Intervertebral bone grafting is critical for maintaining long-term stability of the spine after surgery according to the three columns theory of Denis [28].Adopting posterior single-segment xation limited exposure space, which is not suitable for large area bone grafting.One or two shaped titanium mesh cages lled with autogenous bone particles autogenous bone particles from healthy lamina and spinous process were implanted to reconstruct the anterior and middle column.This intervertebral bone grafting method by implanting ideal titanium cage increases the contact area between the bone particles and the bone grafting surface, thus facilitating the penetration of various cytokines to promote bone metabolism and accelerate the osteogenesis.Furthermore, titanium mesh cage has the characteristics of high strength, strong support, and great friction on the contact surface, which can withstand compressive force well to prevent the bone from fracture and displacement.Furthermore, according to the speci c shape of the bone defect between vertebrae, the titanium mesh can be trimmed and shaped to match the shape of the bone defect.Thus, more healthy bone can be retained, thus avoiding the decrease in spinal stability due to larger bone defects after the removal of the lesions [29].Biomechanical studies have indicated that TB bacilli have weak adhesion to the titanium materials and do not affect the bactericidal effect of anti-TB drugs [30].In this study, we found that bone graft fusion was achieved in all patients at 4 months after surgery.No deep infection, internal xation loosening, fracture and pseudoarticular formation occurred.
This work was supported by the National Natural Science Foundation of China (No.81672191).No bene t in any form has been or will be received from a commercial party related directly or indirectly to the subject of this manuscript.
2 .Denis F. The three column spine and its signi cance in the classi cation of acute thoracolumbar spinal injuries.

Figures
Figures

Figure 1 A
Figure 1

Figure 2 A
Figure 2 Liu HZ, Wang GP, Pang XY, Luo CK, Zeng H, Xu ZQ, Liu XY, Wang XY.The Role of Single-stage posterior debridement, interbody fusion with titanium mesh cages and short-segment instrumentation in