We treated 32 consecutive patients with thoracic or thoracolumbar (T12-L1) tuberculosis with one-stage posterior debridement, allograft bone graft using titanium mesh, posterior instrumentation, and fusion from May 2011 to September 2015.
The spinal tuberculosis diagnosed based on the clinical symptoms, imaging results (anteroposterior and lateral radiography, computerized tomography (CT), and magnetic resonance imaging (MRI)(Fig.1a-d), and hematologic and pathological examinations. The presence of bacillus confirmed the diagnosis. Preoperative or postoperative pathological exams were utilized to make a definite diagnosis.
The American Spinal Injury Association (ASIA) neurological classification was used to evaluate neurological dysfunction and the visual analog scale (VAS) and Oswestry Disability Index (ODI) scores  were used to evaluate pain before surgery, before discharge and at final follow-up. The degree of kyphosis (sagittal Cobb angle) was recorded in the lateral radiograph of full spine preoperatively, postoperatively and at final follow-up to evaluate the kyphosis correction and the loss of the correction. X-ray and CT images were used to assess bony fusion or incorporation of allograft (Fig.1f-j) according to the classification of Tan et al.
Anti-tuberculosis drugs (rifampicin, 0.3 g; isoniazid, 0.45 g; ethambutol, 0.75 g) were taken for 2 to 3 weeks before the surgery. Operation was scheduled, when the erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were stable or started to decrease and other examination and condition were suitable for operation. The operation was performed immediately in case of aggravated neurological status. The surgical plans were created according to the focus of infection, the involved spinal segment, CT, and MRI. In the patients with evident vertebra damage, two vertebrae above and below the involved segment were fixed.
After inducing general anesthesia, the patients were placed in the prone position with forelimbs held upward. Intraoperative C-arm fluoroscopy was used to position the destroyed vertebras. The mid-spinal incision of appropriate length was made, and the spinous process, bilateral lamina, facets joints, transverse process, or part of the ribs needed to be excised were exposed. Transpedicular screws were placed in the vertebrae according to the preoperative design. After the screws were placed, a temporary rod devoid of any rib excision was placed on the side to avoid movement during the debridement; in addition, it also helped in the placement of titanium mesh. After excision of the part of the transverse process and ribs, the collapsed vertebrae, necrotic disc, and prevertebral or paravertebral abscesses were completely removed. The posterior roots were cut in the patients with the excision of ribs. After the prepared bone trough was cleaned by saline irrigation, a suitable titanium mesh with allogenic bone was inserted into the designed preoperative place to reconstruct the stability. Intraoperative fluoroscopy was used to confirm the position of titanium mesh. The intact posterior fixation instrument was placed. Isoniazid or streptomycin was administered locally, and two drainage tubes were placed before the incision was sutured. The excised vertebrae and necrosis tissues were sent for pathological examination, and the abscess was sent for smear examination.
The drain was removed when the drainage flow was < 50ml per 24 hours. All patients were treated with anti-tuberculosis drugs same as given preoperatively for 12 to 18 months. In osteoporosis patients, thoracolumbar brace was continually used for 3 to 6 months postoperatively. All patients were evaluated with physical examination, radiograph (X-ray and CT), ESR, CRP and hepatorenal function at follow-up.
The preoperative and postoperative Cobb angle, VAS and ODI scores were analyzed by ANOVA and t-tests (Version 25.0, SPSS, Chicago, Illinois, USA). P values < 0.05 were considered statistically significant.