Spinal tuberculosis is not seen frequently in the western world. Most of the patients with Extra Pulmonary Spinal Tuberculosis (EPSTB) in developed countries are originally from other countries where tuberculosis is seen to be a more prevalent.  In our study around 75% of the patients were of non-British origins.
A review of data of musculoskeletal tuberculosis reviewed by Talbot et al. over the period of 6-year was performed from 1999 to 2004, including 729 patients with tuberculosis. Approximately 8% (61) of the cases had demonstrated musculoskeletal involvement; nearly 50% of these patients demonstrated spinal involvement . Similar results have been obtained in our study over a 5-year period between 2015 and 2020 with total number of patients were 515 who were newly registered under our care, 6% (n = 31) have been diagnosed with Extra Pulmonary Spinal TB (EPSTB).
The most common symptom seen in patients with spinal TB is back pain [6, 7]. This has been documented in 71%(n = 22) of the patients distributed equally between CP EPSTB and CN EPSTB. Other associated symptoms include tenderness, stiffness, muscular spasms, kyphotic changes, and cold abscess. The latter was found in 58%(n = 18) of the patients, with the number of the patients noticed to be double in the CP EPSTB group (n = 12). These symptoms tend to progress gradually, with an average disease duration ranging between 4 to 11 months. 
Constitutional symptoms are not frequently seen. including malaise, loss of weight, night sweats, and fatigue. This was presented in 6.25% (n = 1) of CPSTB patients and 40% (n = 6) with CNSTB in our study.
Neurological compromise including weakness and numbness, which can progress to complete paraplegia if left untreated. The incidence of neurological complications has been reported to vary from as low as 10% to as high as 76% in previous literature . Which is consistent the results of our study in 38.7% (n = 12) of the patients equally in both groups.
The diagnosis of EPSTB was based on clinical findings, laboratory studies and Radiological Findings. Laboratory samples included Sputum to Acid Fast Bacilli (AFB) on microscopy or isolating the bacilli from Sputum cultures. Biopsies has been also obtained under CT/Ultrasound guidance from pathological lesions in vertebrae or para vertebral collections were also sent for detecting AFB on microscopy and isolating the mycobacterium from biopsy cultures. The latter classified our sample into Culture Positive Extra Pulmonary Spinal TB (CP EPSTB) and Culture Negative Extra Pulmonary Spinal TB (CN EPSTB). PCR and IGRA were also used in our studies but they were not the main parameters for the group identification as shown in Table   and 
It was noticed that EPSTB lesions have a lower number of Mycobacterium bacilli, making it less reliable for microscopical studies  which can be clearly seen in the samples taken from our biopsies to detect AFB on microscopy resulting more than 90% of sample had no mycobacterium detected.
Samples taken for Liquid culture has shown positive results in 51.6% (n = 16). Combining the microbiological and histological results together, the diagnostic yield of biopsy in EPSTB ranges between 42 and 76% [10–13] In our study positive results has been detected in our biopsies only were 51.6% (n = 16) and when combined with our histological results, showing caseating necrotising granulomas which were found in 16.1% (n = 5) of the patients whose cultures were negative. The diagnostic yield for the biopsies was therefore, 67.7%. This lies within the ranges of the current literature reported by different authors  which can be demonstrated in Table 
The diagnosis of the EPSTB is usually needs to be corelated to radiological findings. MRI is considered the most sensitive radiological study as it allows for identification of not just the bone destruction but also the presence of granulomatous tissues, which may not be detected on plain radiographs or CT scans. EPSTB is characterised by some radiological findings including decreased signal intensity of affected bone and soft tissues on T2 images with an associated thin rim enhancement of signal intensity is believed to be a pathognomonic sign for caseating necrosis or a cold abscess  The sensitivity of the MRI findings in our study was more than 80% in both groups as shown in table 
PCR has been an effective diagnostic tool for pulmonary TB and is now thought to have high sensitivity and specificity for extrapulmonary TB as well. Compared to culture, PCR allows for a more rapid diagnosis and greater sensitivity even when small amounts of bacilli are present in the samples taken, as is the case with vertebral biopsies . However, our study wasn’t able to focus on PCR studies as it wasn’t available to all patients and couldn’t be obtained as this is a retrospective study however, we were able to calculate the sensitivity of the PCR tests done to a number of the patients and the results were still able to demonstrate a higher sensitivity (55.5%) compared to the culture results as demonstrated in table 
Antituberculous chemotherapy (ATT) is considered the mainstay of treatment for spinal TB. However, there is still no standardized regimen or known optimal duration of treatment. The induction of therapy should include four main medications, Isoniazid, Rifampin, Pyrazinamide, and either Ethambutol or Streptomycin and can be modified based on results of susceptibility testing, Similar studies have shown no superior effect for Streptomycin in better osseous penetration as was previously believed . Varying treatment durations ranging from 6 to 18 months have been reported . Our study has been following for TB treatment which states, initial treatment of Rifampicin 600mg od PO and Isoniazid 300mg od PO should be given for the first 6 months combined with Pyrazinamide 2g od PO and Ethambutol 15mg/kg od PO for the first 2 months. Treatment was usually allowed to extend for more than 6 months and up to 12 months in cases of Spinal or CNS involvement as seen in our cases. Treatment were able step down to a fewer drug combination when sensitivities are back. Treatment was allowed to be terminated in case of full or partial resolution of the radiology findings.
Surgical intervention can be a treatment option to prevent progressing to neurological complications, and can be used for correction of deformities caused by the progression of the disease, evacuating abscess drainage, spinal cord decompression, or maintaining stabilization for the vertebrae. These indications has been met with a favourable outcome  However, the benefits for surgical intervention can be controversial up to this date .
Two systematic reviews compared medical treatment alone versus medical treatment combined with surgery. Two trials between the period of 1970 and 1980 who fulfilled the inclusion criteria, concluded that there was no statistical difference in outcomes between both groups [21, 22]. They also concluded that surgery had no effect on the resultant kyphosis angle, bearing in mind the incidence of kyphosis for all study subjects was high at the onset (> 30°), which is agreed as an indication for operative intervention by vast majority. However, these reviews were limited by a very small sample size and tremendous surgical and medical advancements have followed these trials.
Recently, more evidence has emerged to back the outcomes of the cases managed by highly efficacious combination of Anti-Tuberculous medications with minimally invasive surgeries for stabilisation or percutaneous fixation. [23–25]
Cold abscess is thought to resolve with Anti-Tuberculous medications only . However, controversy still exists around that its efficacy alone to prevent vertebral destruction, hence immediate drainage combined with medical therapy is usually needed. Epidural abscesses in particular are more likely to cause neurological issues and require urgent drainage to prevent cord compression . Some suggests later surgical intervention may increase the risks of failure. However, estimates of these failures were noted to be variating between studies leading to uncertainty about actual outcomes . Surgical drainage is therefore reserved for worsening abscess or the ones risking causing mechanical pressure related symptoms owing to their size or location [28, 29].
There are limited guidelines to help determine the route of management and choosing between medical or combined medical and surgical approaches . Gulhane Askeri Tip Akademisi (GATA) classification, divides spinal TB into three main types (Types IA/IB, II, and III) based on several clinical and radiological findings. Type IA is the mildest while Type III the severest type. Surgery is recommended for all types except Type IA
In our study surgical intervention was reserved for patients who suffered from severe bone destruction which renders the spine stability. The presence of a large collection of cold abscesses which failed to be aspirated under CT guidance or presence of neurological symptoms were also surgical candidates.
Major Limitations in this study were presented in the number of patients included in this study.
Data gathered through paper and electronic medical records based on the clinical letters in the clinic, records of the cultures and blood tests found on the system and imaging which has been reported by our radiological team. Some of the imaging has been addressed by the infectious disease team with no official report were documented. However, the initial and final imaging of the patients were done and reported in our hospital.
Modifications in the medical treatment has been done during the study, which were still following the guidelines of TB treatment in St. George’s, however the step down of treatment due to culture sensitivity or other causes as allergic reactions against them were not accurately documented
The study was done on a retrospective basis, hence there was no definite structure done by the corresponding authors, which restricted the research team from comparing the results of other tests done for some patients and omitted in others.