Between January 2005 and January 2015, 302 patients with thoracolumbar junction (T12-L1) TB from six hospitals across China were hospitalized; 125 were excluded because of conservative therapy, complicated spinal tumours or active pulmonary TB, poor tolerance or compliance, HIV co-infection and lost to follow-up (Figure 1). The remaining 177 patients were included in the study, comprising 88 men and 89 women with a mean age of 35.2 ± 10.0 years (range 14–62). A total of 45 patients were treated by an anterior-only procedure (Group A), 52 by a combined anterior and posterior procedure (Group B) and 80 by a posterior-only procedure (Group C) (Table 1).
The diagnosis of spinal TB was made on the basis of clinical symptoms, physical signs, laboratory findings, radiological evidence and histopathology. The American Spinal Injury Association (ASIA) score was used to evaluate neurological function. Six patients were grade A, 14 were grade C, 47 were grade D and 108 were grade E. The visual analogue scale (VAS) was used to evaluate back pain for all patients. The Cobb technique was used to assess the local kyphotic angle.
All patients were administered a chemotherapy regimen (isoniazid 300 mg/day, rifampicin 450 mg/day, ethambutol 750 mg/day and pyrazinamide 750 mg/day) for 2–4 weeks prior to the operation.
Operations at each centre were performed by senior surgeons. All patients underwent general endotracheal anaesthesia, after which they were placed in the appropriate position on the spinal table. (1) The anterior-only approach was a thoracoabdominal approach. The exposure focus was complete debridement. The defect area in the vertebral body was packed with an appropriately sized allograft or autologous iliac bones or cage. The screw-plate or screw-rods were placed in the lateral anterior, then streptomycin (1.0 g) and isoniazid (0.3 g) were administered locally, before the incision was closed. (2) In the combined anterior–posterior approach, the patient was placed in a prone position, a standard dorsal midline incision was made, the lamina and articular process were exposed, a pedicle screw was implanted, the kyphosis angle was corrected, a bone graft was performed, and the incision was closed. Then, patients were transferred to the lateral position, and a correctly placed incision was made. The thoracoabdominal approach was used to complete debridement, spinal cord decompression and bone graft fusion. (3) In the posterior-only approach, patients were placed in the prone position on the spinal table. A standard dorsal midline incision was performed, and the posterior tissues were exposed. Abscesses, granulation tissue, sequestra, caseous necrosis, necrotic endplates and discs were debrided as thoroughly as possible via the transpedicular space. Then, two pre-bent rods were installed. A bone autograft of suitable size or a titanium cage containing cancellous bone from the iliac crest was placed within the intervertebral space. Compression and expansion of the internal fixation instrument was used to rectify the kyphosis and scoliosis gradually and carefully, then the contoured rods were tightened. Finally, streptomycin (1.0 g) and isoniazid (0.3 g) were administered locally, and a local drainage tube was placed in the operation site before the incision was closed.
Preventive antibiotic treatment was used within 48 hours after the operation. All patients were advised to use a bracing apparatus until bony fusion was observed on radiography. Patients resumed oral HREZ chemotherapy postoperatively, then Pyrazinamide was discontinued at 6 months. Patients continued to receive 9–12 months regimens of HRE chemotherapy. If a drug sensitivity test suggested resistant TB, the chemotherapy regimen was adjusted. Patients’ liver function and ESR rates were monitored carefully at regular intervals. Follow-up examinations were conducted at 1, 3, 6, 12 and 18 months. Subsequent follow-ups were performed at 12-month intervals.
Continuous data are expressed as 一 X ± S.D. The LSD or Dunnett T3 test was used to evaluate differences in operation time, blood loss, kyphosis angle, ESR, VAS score. Statistical analyses were performed using SPSS version 22 (SPSS, Inc., Chicago, USA). Values of 𝑃 less than 0.05 were considered to indicate significant differences.