The incidence of spinal tuberculosis accounts for approximately 50% of osteoarticular tuberculosis. Of all spinal tuberculosis, the lumbar spine has the highest incidence, followed by the thoracic spine, with an incidence of about 39.6%. Destruction of the thoracic vertebral body can lead to thoracic kyphosis and paraplegia, so thoracic spine tuberculosis has a greater risk than lumbar spine tuberculosis. The main goals of surgical treatment of thoracic TB are to remove the lesion, decompress the spinal cord, stabilize the spine, and correct the kyphotic deformity. However, the choice of surgical modality for thoracic spinal tuberculosis is still controversial. The author retrospectively analyzed the surgical methods and efficacy of 81 cases of non-jumping thoracic spinal tuberculosis in adults and explored the indications for thoracic spinal tuberculosis surgery. 1. Data and methods (1) General data From January 2001 to December 2010, a total of 112 adult patients with thoracic spinal tuberculosis were treated surgically at the author’s hospital. 81 cases of non-jumping tuberculosis obtained an average follow-up of 37 months (17-72 months). There were 32 males and 49 females with a mean age of 38 years (17-68) years. The duration of the disease was 4 months to 3 years, with a mean of 11 months. There was 1 patient with 7 consecutive vertebral involvement, 3 with 5 consecutive vertebral involvement, 11 with 4 consecutive vertebral involvement, 23 with 3 vertebral involvement, 34 with 2 vertebral involvement, and 9 with 1 vertebral involvement. The highest incidence was double vertebral tuberculosis, followed by 3 vertebral tuberculosis, followed by 4 vertebral tuberculosis, and the least was above 5 vertebral and single vertebral tuberculosis. Thoracic kyphosis (Cobb’s angle) was less than 30° in 42 cases, 30°-60° in 31 cases, and 60°-70° in 8 cases. There were 76 cases (93.8%) with elevated hematocrit. 67 cases had significant dead bone on CT or MRI, 73 cases had significant paravertebral abscesses, and 49 cases had spinal cord or dural compression. The 23 patients with preoperative spinal cord compression symptoms were graded according to Frankel, with 3 cases in grade A, 4 cases in grade B, 7 cases in grade C, and 9 cases in grade D. There were 5 cases with combined sinus tracts. There were 29 cases with a history of pulmonary tuberculosis, 21 cases with a history of tuberculous pleurisy, 23 cases with combined cardiovascular and cerebrovascular system diseases, 34 cases with diabetes mellitus, 4 cases with chronic bronchitis, 1 case with cirrhosis, and 9 cases with mildly reduced pulmonary function tests. Five types of surgical procedures were used according to the location and extent of lesions: group A (18 cases), with intrathoracic extrapleural lesion removal and internal fixation with bone graft fusion; group B (21 cases), with transthoracic lesion removal and internal fixation with bone graft fusion; group C (10 cases), with extrapleural transverse costal approach lesion removal and internal fixation with bone graft fusion; group D (27 cases), with posterior fixation in stage I or stage II anterior In group D (27 cases), the lesion removal and internal fixation with bone graft fusion was performed by posterior fixation in the first-stage or second-stage anterior transthoracic or extrapleural approach; in group E (5 cases), the lesion removal and internal fixation with bone graft fusion was performed by sternal stalk or sternal splitting approach for upper thoracic segment tuberculosis. (2) Preoperative preparation After admission, the relevant departments were consulted to lower blood pressure, improve pulmonary function, control blood sugar, and correct anemia or hypoproteinemia by blood transfusion, albumin and plasma. In 23 patients with spinal cord compression symptoms, surgery was performed after 6 to 18 hours of antituberculosis treatment. The surgery was performed after 6 to 18 hours of anti-tuberculosis medication. (3) Surgical method: Removal of extra-thoracic pleural lesions and internal fixation with bone graft fusion. Patients were placed under general anesthesia with tracheal intubation. The patient is entered from the severe side, and the gauze wrapped fingers are used to make blunt dissection in the extra pleural fat layer. The abscess wall is incised longitudinally, the pus is aspirated, the damaged vertebral body and intervertebral disc tissue are scraped away, the tuberculous lesion is completely removed, and the spinal canal is decompressed if it is accompanied by spinal canal encroachment. The lesioned segment is first propped open to correct the posterior convexity deformity, followed by implantation of autologous ribs or titanium cage in the bone defect, local application of streptomycin 2g, and completion of plate or nail fixation. Normal drainage was placed after surgery. Transthoracic lesion removal and internal fixation with bone graft fusion General anesthesia with tracheal intubation. For T1-4 tuberculosis, the scapular retraction approach with resection of the 3rd rib is used, and for T4-12 tuberculosis, the standard open thoracic approach is used, and the rest of the procedure is the same as the intrathoracic extrapleural lesion removal and internal fixation with bone graft fusion. Closed chest drainage was placed after the operation. Extra-thoracic transverse costal approach for lesion removal and internal fixation with bone graft fusion Tracheal intubation with general anesthesia, lateral oblique position, with the thoracoabdominal plane at 60° to the operating table. With the diseased vertebral body as the center, the spinous process of the vertebral body is revealed from above, and the spinous process is revealed by using a curved incision 6-8 cm from the spinous process of the diseased vertebral body to reach the inferior vertebral body, and the flap is opened medially to routinely reveal the spinous process of the diseased vertebral body, the vertebral plate, the articular process and one normal vertebral body above and below, and according to the severity of the posterior convexity deformity, the angle and direction of the nail is determined with the assistance of a “C” arm X-ray machine. According to the severity of the kyphosis, the angle and direction of the nail is determined with the assistance of a “C” arm X-ray machine, and the pedicle screw is inserted. To correct the kyphosis, the spinous process with severe kyphosis is removed, and an autologous iliac bone is placed on the vertebral plate to close the subcutaneous “sterile” incision. In the same skin incision, the muscles of the thoracic back are cut along the outer edge of the sacrospinous muscle, the ribs connected to the diseased vertebrae are removed, the transverse processes are occluded, the intercostal vessels are ligated, a section of the ribs and the corresponding transverse processes superior to the center of the lesion are removed in the same way to enlarge the surgical field, the pus is aspirated, the damaged vertebral body and intervertebral disc tissue are scraped away, the tuberculous lesion is completely removed, the spinal canal is decompressed if it is accompanied by spinal canal encroachment, the bone graft bed is chiseled out, and the autologous iliac bone is implanted rib or titanium cage in the vertebral bone defect, local application of streptomycin 2g, placement of drainage and closure of the incision. In 22 patients, posterior fixation and correction of kyphosis were performed with a posterior pedicle nail system, followed by laminectomy and decompression for those with spinal cord compression, followed by anterior lesion removal and fusion with bone grafting. 5 patients underwent anterior lesion removal and fusion in the second stage due to poor general condition. A transthoracic approach was used in 21 cases of thoracic spinal tuberculosis and an extrapleural approach was used in 6 cases. Tuberculosis lesions were completely removed and spinal canal decompression was performed in those with spinal canal encroachment. Closed chest drains were placed in the transthoracic approach, and normal drains were placed in the extrapleural approach. Sternal stalk or sternal split approach for upper thoracic tuberculosis removal and internal fixation with bone grafting General anesthesia with tracheal intubation. From the midpoint of the anterior border of the right sternocleidomastoid muscle obliquely downward to the midpoint of the sternal stalk incision, longitudinally to below the sternal angle. The sternum was incised longitudinally to slightly below the sternal stalk, and the sawn sternal stalk was transected and then propped open. The separation was continued downward between the carotid sheath and the visceral sheath. For T1,2 nodules, a medial cephalic trunk approach was used, with the cephalic trunk and right common carotid artery drawn to the right, and the trachea and esophagus drawn to the left; for T3,4 nodules, a lateral cephalic trunk approach was used, with the cephalic trunk, trachea, and esophagus drawn to the left, and the base of the right cephalic vein drawn to the right, and the left cephalic vein pressed downward. The prevertebral fascia is incised, the lesion is removed, and the autologous iliac bone or titanium cage filled with allograft bone is implanted into the bone defect, and the titanium plate is fixed. (4) Postoperative management A standard chemotherapy regimen of anti-tuberculosis drugs (3HRZE/9HRE) was used postoperatively, i.e., isoniazid, rifampin, pyrazinamide, and ethambutol were given orally for 3 months, after which pyrazinamide was discontinued and other drugs continued to be given orally for 9 months. Patients wore a brace for 1 to 2 weeks postoperatively to get out of bed and the brace was worn for 3 months. Postoperative follow-up ranged from 22 to 72 months, with a mean of 37 months. 2. Results (A) General conditions (1) Operative time and bleeding volume: group A, average operative time 3.5h, average intraoperative bleeding volume 350ml; group B, average operative time 3.0h, average intraoperative bleeding volume 350ml; average intraoperative bleeding volume 450ml; group C, average operative time 3.0h, average intraoperative bleeding volume 300ml; group D, average operative time 4.5h (2) Correction rate of kyphosis: (47.5±11.8)% in group A, (46.9±10.2)% in group B, (59.9±17.4)% in group C, (61.7±18.6)% in group D, and (44.1±8.7 ) %. (3) The rate of corrected angle loss at the final follow-up: (64.2±19.1)% in group A, (63.8±18.1)% in group B, (56.9±11.8)% in group C, (53.6±15.6)% in group D, and (63.5±17.1)% in group E. All patients had significant reduction of local symptoms 1~3 weeks after surgery, and the erythrocyte sedimentation rate returned to normal 8~12 weeks after surgery. no serious complications occurred in any of the 5 groups. one case in each of groups A, C and D had sinus tract formation in the incision half month, 1 month and 1.5 months after surgery, respectively. no abscess or dead bone was found by CT and MRI examination, and was cured after 1~2 months of drug replacement treatment. one case in group A was found to have loose internal fixation at 1 year postoperative follow-up. One case in group A was found to have loosened internal fixation at 1 year postoperative follow-up, but the bone graft had fused. One patient in group A (5.6%), two in group B (9.1%) and two in group D (7.4%) had drug reactions. four cases had liver function impairment, three of which had gastrointestinal reactions; one case had renal function impairment. After changing rifampin to rifapentine (2 times/week) orally, the drug reactions gradually disappeared. (B) Correction of kyphosis and bone graft fusion The postoperative radiographs showed that the kyphosis was partially corrected in all patients, and the correction rate was higher in groups C and D with posterior fixation than in groups A, B and E with anterior fixation (P<0.05). Group A, Group B and Group E with anterior fixation had a higher rate of angle loss than Group C and Group D with posterior fixation (P<0.05). One case in group A was found to have loosened internal fixation at the 1-year postoperative follow-up, but the bone graft had fused, and the corrected angle of lordosis was lost by 16°, and there was no increase in lordosis at the 2-year postoperative follow-up. There was no recurrence in all cases by the time of the last follow-up. In conclusion, the choice of surgical plan for thoracic spine tuberculosis should be based on the principle of individualized treatment and comprehensive consideration without bias. A reasonable anti-tuberculosis chemotherapy regimen remains the key to the treatment of spinal tuberculosis and is also a guarantee of successful surgery.