The purpose of the surgical treatment for thoracolumbar infectious spondylitis in the elderly is to facilitate early mobilization after the operation. One-stage posterior focus debridement, bone graft fusion, and internal fixation are widely accepted methods for the treatment of spinal infections in the elderly , . For elderly patients in poor health and with poor surgical tolerance, this method can effectively improve the safety of the operation. Our results show that the surgical treatment of infectious spondylitis was effective, enabling 91% of the patients who received surgical treatment to walk independently during the follow-up period. Although we only included patients over 65 years of age, our results are consistent with previous reports [23–26].
In our study, we found no significant differences between the PS and TS groups for mean blood loss and the operation duration. However, hospitalization and ICU stay duration in the PS group were significantly shorter than in the TS group. Okada et al.  reported similar results. TS is a long-term chronic wasting disease that is usually associated with systemic diseases such as tuberculosis, which may help to explain this phenomenon . Thus, a post-operative stay in the ICU and delays in the recovery process would be inevitable. In particular, the onset of tuberculosis is typically unknown, the early symptoms are atypical and it takes longer to diagnose, which partially explains why the hospitalisation duration for patients in the TS group was longer than in the PS group. In our study, 44.7% of patients had post-operative complications. In a related study of elderly patients with spinal tuberculosis, Luo et al.  reported a 97% incidence of complications, which may be due to the absence of statistics on the complications of water-electrolyte imbalance. However, we did not find significant differences in complication rates between the two groups.
In terms of the clinical follow-up results, the VAS, ODI and the MCS and PCS scores of the SF-36 for the TS and PS groups were significantly improved from the pre-operative evaluation to the final follow-up. However, the VAS score for lower back pain in the TS group was significantly lower than that in PS group six months post-operatively (P < 0.05) (Fig. 1). Okada et al.  reported that the duration of achieving a CRP-negative result in a TS group was longer than for a PS group (105.9 ± 16.3 vs 52.6 ± 20.2 days), and patients with tuberculous spondylitis experienced more severe inflammatory stimulation within a short period following an operation. Therefore, before inflammation subsides, PCS scores may demonstrate significant differences. Correspondingly, the PCS scores in our TS group were lower than the PS group at the six-month follow-up. At the one-year follow-up, the MCS score of the TS group was worse than the PS group.
The literature suggests that tuberculosis is usually associated with poor living standards and social factors, which can more strongly affect the elderly. Therefore, we speculate that surgical trauma has a greater impact on patients with TS than PS; although, the difference may become indistinct over time. Chun Kim et al.  reported that at the final follow-up of 485 people, 85% demonstrated excellent or good results in lumbar fusion in the elderly according to the MacNab criteria. Similarly, approximately 83% of the patients in our study were satisfied with the post-operative results, indicating that the positive effects of surgical treatment for the elderly should be expected if the operation is safe.
Similar results from other studies , , , have shown that solid bone fusion occurs in over 90% of cases. Likewise, the solid fusion rates in the PS and TS groups in our study also reached 90.5% and 92.4%, respectively, at the final follow-up. Although there was no statistical difference between the PS and TS groups, both achieved good fusion rate results. We assumed TS might hamper the maintenance of alignment due to necrotic disease and osteoporosis , , and it has been reported that TS is more frequently associated with greater deformity than PS . However, we found no significant difference in the correcting angle loss between the PS and TS groups. This is similar to that reported by Okada et al. ,however, definitive results may require longer follow-up observation times.
There were some limitations in our study. First, the sample sizes in the two groups were small. A sample size of 47 patients is insufficient and was due to the small number of elderly surgical patients. Second, our average follow-up period of 26 months was not enough to observe long-term effects. Finally, our study focused on specific sub-groups and did not make comparisons with conservative treatments in the elderly, which may have led to inaccurate results. Further, multi-centre, randomised, long-term follow-up studies are needed to overcome these issues.