Abstract
We report a case of infectious spondylitis in a 52-year-old woman who presented with progressive difficulty in walking. The patient had a 2-month long history of neurological symptoms, which progressed rapidly to paraplegia, following her admission. Imaging studies demonstrated the presence of vertebral lesions as well as additional tissue with inflammatory elements in the spinal canal, which caused a mass effect. In combination with the presence of increased cells and protein in the cerebrospinal fluid (CSF), the differential was steered towards causes of infectious spondylitis, primarily tuberculosis. However, brucellosis was also considered, as it is endemic in our area. Prompt surgical decompression produced biopsy samples, which confirmed the presence of granulomatous inflammation. The patient was started on an empiric regimen covering both for tuberculosis and brucellosis, and gradually regained full mobility in her lower limbs. The differential of infectious spondylitis is discussed, with an emphasis on the differentiation between tuberculosis and brucellosis.
Keywords: bone and joint infections, drugs: infectious diseases, TB and other respiratory infections, spinal cord
Background
Tuberculous spondylitis or Pott’s disease comprises around 5% of cases of extrapulmonary tuberculosis cases.1 It has an insidious onset with a rather long duration of symptoms (a mean time of 23.2 months), and can lead to the development of severe complications, such as kyphosis, paraplegia or even death.2 Therefore, prompt diagnosis and treatment are essential. The differential includes other causes of vertebral osteomyelitis, such as brucellosis and osteomyelitis from common bacteria like Staphylococcus aureus and Escherichia coli.3 While a positive culture is required to confirm the aetiology,3 the combination of clinical, laboratory and radiological findings can often help distinguish the most probable cause, as the pathobiological effects of the infecting organisms differ and can be characteristic.4 We report a case of granulomatous infective spondylitis, in a 52-year-old woman who presented with a progressive difficulty in walking, who responded to combined antibiotic treatment for brucellosis and tuberculosis. The treatment for brucellosis was discontinued after 6 weeks, whereas the treatment for tuberculosis was continued for 12 months. The patient continued to improve while on antituberculous treatment, suggesting that tuberculosis was the most probable diagnosis, without, however, excluding a possible coinfection by Brucella spp.
Case presentation
A 52-year-old woman presented with a week-long history of numbness of the abdominal wall and difficulty in walking. Her medical history commenced 50 days prior to presentation, at which time she developed neck pain, that radiated to the left upper limb. An MRI examination of the cervical spine at that time demonstrated herniation of intervertebral discs at level of C5–C6 and C6–C7, and stenosis of the left vertebral canal at the level of C5–C6 vertebra, with probable irritation of C6 root. No interventions were performed. On account of the continuing pain, 10 days prior to presentation, the patient visited a neurosurgeon, who prescribed pregabalin and ibuprofen for neuropathic pain and analgesia, respectively. The patient reported that the numbness manifested soon after she began taking pregabalin.
Her prior medical history was remarkable for osteoporosis and hypothyroidism for which she was taking calcium supplements, risendronic acid and thyroxin. She had no prior medical or family history of tuberculosis or brucellosis, nor did she have any predisposing conditions or known risk factors for either disease.
On physical examination, the patient was found to have hypoaesthesia of the T5–T10 dermatomes with diminished pinprick sensation of the left upper limb and diminished muscle strength to the trunk and lower limbs, while she retained normal muscle strength to the rest of her body. Patellar reflexes were very brisk, whereas plantar reflexes were normal. Muscle strength of the lower limbs was diminished but adequate, scoring 4/5 as per the Medical Research Council scale. Therefore, her difficulty walking was attributed to trunk muscle weakness rather than lower limb weakness.
Investigations
On admission, the patient had a normal leucocyte count (white blood cells (WBC)=7480 cells/mL, neutrophils=4290 cells/mL, lymphocytes=2230 cells/mL, monocytes=740 cells/mL) and an elevated erythrocyte sedimentation rate (ESR) of 43 mm/hour.
As the main diagnostic consideration was infectious spondylitis, with the main suspected microorganism being M. tuberculosis, a Mantoux test, sputum staining as well as culture for tuberculosis and common aerobes, anaerobes and fungi were requested. Moreover, blood cultures were drawn with an emphasis on the isolation of Brucella spp, which is endemic in our area, in addition to Wright and Wright-Coombs tests. A serum 1βD glucan assay was also ordered. Other tests requested included urine acid-fast staining and culture for M. tuberculosis and Brucella spp. However, all of the above tests quickly resulted as negative excepting for M. tuberculosis cultures which were pending. An interferon-γ release assay (IGRA) was not available and therefore was not requested.
A lumbar puncture was performed, the results of which were: cells: 22/μl (lymphocytes: 19, monocytes: 3), red blood cells: 3/μL, glucose: 58 mg/dL (serum: 89 mg/dL), protein: 470 mg/dL, Gram stain: negative, acid-fast stain: negative. A PCR assay was negative for M. tuberculosis and while culture of the CSF was negative for common bacteria. After an adequate incubation period, CSF culture for M. tuberculosis, as well as cultures from other bodily fluids were reported to be negative.
A chest X-ray did not reveal any evidence of prior pulmonary disease. A new MRI examination of the spine revealed the presence of significant disruption of C6 and C7 vertebrae and of the intervening disk, as well as compression of the spinal cord by the thickened dura matter at these levels (figure 1).
Figure 1.

MRI of the cervical spine of the patient on admission, T1 sequence, with (A) and without (B) gadolinium enhancement, as well as follow-up examination 2 years postdischarge (C). Abnormal signal intensity can be seen in the bodies of the C6 and C7 vertebrae, as well as exudative elements at the posterior soft tissues, at the level of C4–T1 (A). Gadolinium administration showed abnormal enhancement of the C5, C6 and C7 vertebrae and the C6–C7 intervertebral disc, all of which showed signs of significant deterioration. Additionally, from the level of C3 to T1, the dura matter appeared thickened, while the posterior paravertebral space displayed inflammatory elements (B). The presence of abnormal additional tissue is depicted within the spinal canal in the epidural space, which embraces the spinal cord, causing a mass effect at the levels of C7–T1, and appears to project over the adjacent intervertebral foramina. Additionally, abnormal signal intensity was observed at the levels of C4–T1 with concomitant elimination of the posterior and anterior subarachnoid space at the same level. Follow-up MRI 2 years postdischarge (C) showed regression of the inflammatory elements, decompression of the spinal cord and laminectomy of the C3–C7 vertebrae.
Differential diagnosis
The differential diagnosis mainly focused on causes of chronic infectious spondylodiscitis, such as tuberculosis, brucellosis, syphilis, as well as common bacterial and fungal infections. This was based on the finding of increased cells and protein in the CSF, in combination with the results of the imaging studies, which demonstrated the presence of vertebral lesions as well as additional tissue with inflammatory elements in the spinal canal with subsequent mass effect.
Tuberculosis was the principal diagnosis. The pattern of vertebral involvement, which included the posterior elements of the spine, with continuous, disruptive lesions of the C6 and C7 vertebral bodies as well as of the C6–C7 disc, were accordant with the pattern typically observed in tuberculous spondylitis.4 5 However, as patients with tuberculous spondylitis usually have significantly elevated ESR levels, with some series reporting an average ESR of 101 mm/hour (75–115)6 and 79.4 mm/hour (43.8–115),7 the patient’s ESR on admission of 43 mm/hour prompted the consideration of alternative diagnoses. Nevertheless, it should be noted that the levels of ESR alone are not sufficiently sensitive nor specific to either include or exclude the diagnosis of tuberculosis or brucellosis.
As brucellosis is endemic in our region, brucellar spondylitis was the major alternative diagnosis. The patient’s ESR was more in accordance with the levels observed in brucellar spondylitis, which have been reported to average between 44 mm/hour (28–94 mm/hour)8 and 35 mm/hour (6–94 mm/hour).9 However, in contrast to the findings of this case (figure 1), brucellar spondylitis usually does not disrupt the architecture of affected vertebrae.4
Treatment
The patient’s clinical condition gradually deteriorated, as she developed difficulty buttoning her shirt, worsening faecal incontinence and paraparesis. Because of her worsening symptoms, she was scheduled urgently for surgery, where she underwent a posterior laminectomy of the C3–C7 vertebrae, along with a removal of an epidural abscess.
The pathology report of the excised tissue noted the presence of granulomatous inflammation with necrosis. The differential diagnosis provided by the pathology report included infectious diseases, with first and foremost tuberculosis; atypical mycobacteria, fungi and Brucella spp were also suspected. However, a culture performed on a specimen obtained from the abscess was negative for M. tuberculosis or common bacteria. Tissue PCR was also requested but was not available at the time.
Overall, due to the severe presentation and worsening neurological signs of the patient, it was decided to administer a regimen covering for both tuberculous and brucellar spondylitis empirically, as an urgent treatment, in order to inhibit the disease evolution and neurological deterioration. The treatment regimen consisted of rifampicin, isoniazid, ethambutol and pyrazinamide, administered for tuberculosis; linezolid, doxycycline and ceftazidime were administered for coverage of brucellosis and other common infectious agents of spondylitis. This regimen also concomitantly covered for S. aureus and P. aeragunosa, given that this patient had also undergone a neurosurgical operation, pending final diagnosis.
Outcome and follow-up
Following the procedure and the empiric antibiotic treatment, the patient’s condition substantially improved, and subsequent MRI of the cervical spinal cord showed regression of the radiological findings. Subsequently, linezolid and ceftazidime were discontinued, although linezolid also provided coverage for tuberculosis and brucellosis.10 11 The patient was discharged from the hospital a few days later, with antibiotic regimens covering only for tuberculosis and brucellosis, which were to be continued for 12 months and 6 weeks, respectively.
On a follow-up appointment 1 month postdischarge, the patient was adhering to the antibiotic treatment and her mobility and clinical condition continued to show recovery, while the elevated ESR continued to decline. Her ability to walk gradually progressed, initially walking with the use of a walker and later with the use of a cane. On a follow-up appointment at 12 months, the patient was able to walk normally, without any assistance.
Discussion
The spine is the most common location of musculoskeletal tuberculosis, accounting for almost 5% of extrapulmonary cases and 50% of musculoskeletal cases.1 12 The infection reaches the spine either hematogenously or through the lymphatics, typically originating from the lung (33%–50% of cases).13 The dense network of blood vessels of the cancellous bone of the vertebral bodies facilitates the dissemination of the bacilli to the spine. It has been suggested that vascular dissemination into paradiskal regions occurs through the anterior and posterior spinal arteries, whereas the involvement of the central vertebral body occurs through the valve-less Batson’s paravertebral venous plexus.4
The disease is characterised by the destruction of the intervertebral disk space and the adjoining vertebral bodies, typically affecting more than one vertebra, and by gibbus formation, which results from the anterior wedging of the spine, following the collapse of the anterior spinal elements. The most commonly involved regions are the upper lumbar and lower thoracic vertebrae, with the vertebral body affected more commonly than the posterior elements.4 However, in the case that the posterior elements are involved, tuberculous and fungal spondylitis are more probable than typical bacterial osteomyelitis.5
On presentation, patients report (in order of frequency) back pain, fever, weight loss, neurological deficits and night sweats. The duration of symptoms varies, with patients presenting from 2 weeks to several years after the onset of symptoms. Left untreated, spinal tuberculosis can lead to complications such as syringomyelia, permanent neurological and spinal osseous deficits.14 The most common neurological deficits observed are numbness, decreased sensation and paralysis.2
Paraplegia is one of the most alarming symptoms of spinal tuberculosis. Depending on the activity of the tuberculous infection, paraplegia has been classified into early and late onset. Early onset paraplegia occurs <2 years after the onset of disease. Causes include external pressure of the spinal cord by a paravertebral abscess, penetration of the dura by the tuberculous infection, involvement of the neural arch or, rarely, acute thrombosis of arteries supplying the spinal cord. If the cause is mechanical pressure the prognosis is favourable and, provided that the pressure is removed soon after the onset of symptoms, the patient can recover fully. Nevertheless, if the infection involves the spinal cord then the recovery will always be incomplete.15
In contrast, late-onset paraplegia occurs when the disease has healed, resulting from fibrosis induced by tissue repair mechanisms, following remission of the active lesion. Late-onset paraplegia results from the transection of the spinal cord by a bony bridge or by pressure on the cord by the fibrosis that surrounds granulation tissue. The prognosis is poor and recovery is not anticipated, even with adequate decompression.15
Diagnosis of spinal tuberculosis requires a high degree of clinical suspicion. The diagnosis is generally based on clinical and neuroimaging findings. CT can reveal the discovertebral abscess and MRI demonstrates the involvement of soft tissue and the extend of spinal cord affected. Further imaging of the entire spine should be done to look for noncontiguous vertebral lesions.4
Other important investigations are acid-fast stain on microscopy and PCR or culture obtained from a biopsy of the lesion.4 The sensitivity of acid-fast microscopy is low, due to the presence of only a few bacilli in the tubercular spondylodiscitis lesion.16 The sensitivity and specificity of PCR in tuberculous spondylitis, has been reported to range between 84.6%–98.7% and 98.7%–100%, respectively.17 Up to 67% of patients with tuberculous spondylitis have a prior medical history or radiological evidence of pulmonary tuberculosis,4 therefore warranting the examination and culture of sputum for the presence of bacilli. Although these tests are considered the gold standard for the diagnosis to be established, they are less sensitive for spinal tuberculosis than they are for pulmonary tuberculosis.16 Other tests pointing towards tuberculosis include the Mantoux test, which is reported to be positive (≥15 mm) in 54.2% of patients and, the IGRA, which is characterised by higher sensitivity and specificity.16 18
The sensitivity of several diagnostic modalities for spinal tuberculosis has been shown to be 33% for microscopic analysis and 43% for culture, with a combined sensitivity of 59%. In another report of spinal tuberculosis, IGRA was found to have 84% sensitivity and 95% specificity, in contrast to 30% sensitivity estimated in the combination of smear and culture. Additionally, the combination of IGRA, smear and culture raises the sensitivity to 88%.16
The parameters generally used to evaluate the therapeutic response of osteomyelitis during follow-up are clinical symptoms, radiographic images, C reactive protein (CRP) and ESR.12 Specifically, the ESR is a helpful index of the activity of the disease, the average values of which have been reported to range between 79.4 and 101 mm/hour.6 7 The majority of patients with spinal tuberculosis typically present with elevated ESR, as well as CRP, which decline when the infection has been controlled.4 12 The elevated ESR is usually accompanied with normal WBC count, in contrast with pyogenic spondylitis where both the WBC number and ESR are elevated.4 Therefore, inflammatory markers can be used for diagnosis, assessment of response to treatment and prognosis of spinal tuberculosis.12
Brucellosis should be considered as an alternative diagnosis of spinal tuberculosis, especially in regions with high prevalence of brucellosis. Brucellosis presents as spondylitis in 7.5%–30% of cases, with the most common genus of Brucella spp causing spondylitis being B. melitensis.5 Another common presentation of brucellosis is sacroiliitis.4 The typical patient with brucellar spinal disease presents with a 2- to 6-month long history of fever, perfuse sweating, weight loss, malaise and backache, is usually middle aged and has been exposed on contaminated animal products or unpasteurised milk.4 19 20 The lesion is usually localised at the lumbar region, likely due to its rich vascular network or the frequency of degeneration in this region.19 It only involves the vertebral bodies and the intervening disk, sparing the paravertebral soft tissue.4 19 Radiology reveals that the entire vertebral body is involved; however, its architecture remains intact.4 In addition, patients may have an elevated ESR, CRP and transaminases levels.19
In order for the diagnosis to be validated, at least two of the following should occur:
The isolation of Brucella spp in cultures from blood and/or bone marrow aspirate.
A Brucella spp agglutination titer at or over 1:160.
A bone scan, radiographs and/or a CT scan showing skeletal involvement features of osteomyelitis.
A biopsy revealing non-caseating granulomas, indicative of brucellosis.21
Diagnosis of chronic brucellosis can be difficult, as the diagnostic yield of cultures and serological tests depended on the stage of disease. Specifically, the diagnostic yield of blood and bone marrow cultures is highest in acute brucellosis (<8 weeks) and declines significantly in subacute (8–52 weeks) and chronic (>52 weeks) disease. To the contrary, the diagnostic yield of serological tests either improves or remains adequate as the disease progresses, as the yield of ELISA IgM and IgG improves in chronic brucellosis, whereas tube agglutination assays are usually positive in acute disease and remain positive in chronic disease.22 In our case, all tests pertaining to brucellosis were negative, making this diagnosis less likely. Regardless, the rapid deterioration of the patient’s clinical condition and high prevalence of brucellosis in our region prompted the empiric treatment of the disease.
Coinfection by M. tuberculosis and Brucella spp can occur but is a rare occasion. A literature review identified only one case of culture confirmed coinfection.23 As the clinical signs and symptoms of brucellosis do not differ significantly from those of tuberculosis, misdiagnosis can easily occur.24 Therefore, in order to establish a correct diagnosis, the clinical findings should be correlated with laboratory results, as well as radiological and pathological characteristics.25 Due to financial limitations, not all examinations could be performed in this case; therefore, empirical treatment regimens were utilised to ensure that the critical condition of the patient was improved.
In summary, tuberculosis usually occurs in younger age groups than brucellosis. While both diseases typically affect the lumbar region, tuberculosis also shows predilection for the thoracic vertebrae. Brucellosis does not affect the paraspinal soft tissue or the architecture of the vertebral bodies, despite causing diffuse osteomyelitis. In contrast, tuberculosis can destroy both the vertebral bodies and the disk spaces, leading to the development of gibbus deformity. Furthermore, tuberculosis can cause significant soft tissue inflammation and can lead to the formation of cold abscesses.4 Finally, while both diseases can lead to an elevation of the ESR,4 in tuberculosis this elevation is more pronounced6–9 and is not accompanied by a concomitant elevation of WBC, a finding can distinguish tuberculosis from both brucellar and bacterial osteomyelitis, both of which are characterised by leucocytosis.4 Furthermore, brucellosis rarely affects the pulmonary system26 and is often accompanied by foul-smelling perspiration, described in the literature as reminiscent of wet hay or mould27 or, in our clinical experience, of freshly baked bread.
Considering that the definitive distinction between the two conditions is difficult and requires prolonged cultures,23 28 in cases with rapid clinical deterioration, it may be prudent to provide empirical antibiotic treatment covering for both infections. Furthermore, due to the overlap between the treatment regimens of the two diseases, the sole addition of doxycycline to the standard quadruple treatment for tuberculosis can provide adequate coverage for brucellosis, demonstrated as well in the only identified case of culture confirmed coinfection.23
Learning points.
Tuberculous spondylitis can lead to significant complications and should always be considered in patients with difficulty walking or peripheral neurological deficits.
The neurological deficits brought on by tuberculous spondylitis are reversible, especially if are due to mass effects. Prompt decompression of the spinal cord can lead to complete recovery.
Other causes of vertebral osteomyelitis include common bacteria and, in endemic areas, Brucella spp. They can be distinguished from tuberculous spondylitis with the combination of clinical and radiological findings.
Tuberculous spondylitis leads to significant destruction of the involved vertebrae, whereas brucellar spondylitis, despite affecting the entire vertebral body, does not disrupt its architecture. While most bacterial causes of spondylitis lead to an elevated ESR and leucocytosis, in tuberculous spondylitis the elevation of the ESR is not accompanied by leucocytosis.
Empiric antibiotic treatment should be considered in cases with neurological involvement and rapid deterioration, which can consist of the sole addition of doxycycline to the standard quadruple regimen for tuberculosis.
Footnotes
Contributors: EP evaluated the patient, conducted the appropriate literature review and drafted the manuscript. LK evaluated the patient, contributed to the drafting of the manuscript and obtained informed consent from the patient. SL evaluated the patient and contributed to the drafting of the manuscript. GP was the attending physician of the patient, guided the rest of the team in the drafting of the manuscript and reviewed the manuscript.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Patient consent for publication: Obtained.
Provenance and peer review: Not commissioned; externally peer reviewed.
References
- 1. Gautam MP, Karki P, Rijal S, et al. Pott's spine and paraplegia. JNMA J Nepal Med Assoc 2005;44:106–15. [PubMed] [Google Scholar]
- 2. Yao Y, Song W, Wang K, et al. Features of 921 patients with spinal tuberculosis: a 16-year investigation of a general Hospital in Southwest China. Orthopedics 2017;40:e1017–23. 10.3928/01477447-20171012-03 [DOI] [PubMed] [Google Scholar]
- 3. Nickerson EK, Sinha R. Vertebral osteomyelitis in adults: an update. Br Med Bull 2016;117:121–38. 10.1093/bmb/ldw003 [DOI] [PubMed] [Google Scholar]
- 4. Garg RK, Somvanshi DS. Spinal tuberculosis: a review. J Spinal Cord Med 2011;34:440–54. 10.1179/2045772311Y.0000000023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Gouliouris T, Aliyu SH, Brown NM. Spondylodiscitis: update on diagnosis and management. J Antimicrob Chemother 2010;65:iii11–24. 10.1093/jac/dkq303 [DOI] [PubMed] [Google Scholar]
- 6. Kostov K, Petrov I. Tuberculous spondylitis--analysis of 22 cases. Acta Neurol Belg 2009;109:127–31. [PubMed] [Google Scholar]
- 7. Guo L-X, Ma Y-Z, Li H-W, et al. [Variety of ESR and C-reactive protein levels during perioperative period in spinal tuberculosis]. Zhongguo Gu Shang 2010;23:200–2. [PubMed] [Google Scholar]
- 8. Yang B, Hu H, Chen J, et al. The evaluation of the clinical, laboratory, and radiological findings of 16 cases of brucellar spondylitis. Biomed Res Int 2016;2016:1–6. 10.1155/2016/8903635 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Solera J, Lozano E, Martinez-Alfaro E, et al. Brucellar spondylitis: review of 35 cases and literature survey. Clinical Infectious Diseases 1999;29:1440–9. 10.1086/313524 [DOI] [PubMed] [Google Scholar]
- 10. Singh B, Cocker D, Ryan H, et al. Linezolid for drug-resistant tuberculosis. Cochrane Database Syst Rev;46 10.1002/14651858.CD012836 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Sezgin Sayiner H, Akgun S. In vitro effectiveness of linezolid in brucellosis. Biomed Res 2017;28:6234–6. [Google Scholar]
- 12. Sudprasert W, Piyapromdee U, Lewsirirat S. Neurological recovery determined by C-reactive protein, erythrocyte sedimentation rate and two different posterior decompressive surgical procedures: a retrospective clinical study of patients with spinal tuberculosis. J Med Assoc Thai 2015;98:993–1000. [PubMed] [Google Scholar]
- 13. Schirmer P, Renault CA, Holodniy M. Is spinal tuberculosis contagious? Int J Infect Dis 2010;14:e659–66. 10.1016/j.ijid.2009.11.009 [DOI] [PubMed] [Google Scholar]
- 14. Chen C-H, Chen Y-M, Lee C-W, et al. Early diagnosis of spinal tuberculosis. J Formos Med Assoc 2016;115:825–36. 10.1016/j.jfma.2016.07.001 [DOI] [PubMed] [Google Scholar]
- 15. Hodgson AR, Yau A. Pott's paraplegia: a classification based upon the living pathology. Spinal Cord 1967;5:1–16. 10.1038/sc.1967.2 [DOI] [PubMed] [Google Scholar]
- 16. Delogu G, Zumbo A, Fadda G. Microbiological and immunological diagnosis of tuberculous spondylodiscitis. Eur Rev Med Pharmacol Sci 2012;16:73–8. [PubMed] [Google Scholar]
- 17. Rasit AH, Ibrahim SF, Wong C, et al. The role of Polymerase Chain Reaction (PCR) in diagnosis of spine tuberculosis after pre-operative anti-tuberculosis treatment. Malaysian Orthop J 2011;5:8–12. 10.5704/MOJ.1103.002 [DOI] [Google Scholar]
- 18. Jalleh RD, Kuppusamy I, Mahayiddin AA, et al. Spinal tuberculosis: a five-year review of cases at the National tuberculosis centre. Med J Malaysia 1991;46:269–73. [PubMed] [Google Scholar]
- 19. Bozgeyik Z, Ozdemir H, Demirdag K, et al. Clinical and MRI findings of brucellar spondylodiscitis. Eur J Radiol 2008;67:153–8. 10.1016/j.ejrad.2007.07.002 [DOI] [PubMed] [Google Scholar]
- 20. Leong KN, Chow TS, Wong PS, et al. Outbreak of human brucellosis from consumption of raw goats' milk in Penang, Malaysia. Am J Trop Med Hyg 2015;93:539–41. 10.4269/ajtmh.15-0246 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Hantzidis P, Papadopoulos A, Kalabakos C, et al. Brucella cervical spondylitis complicated by spinal cord compression: a case report. Cases J 2009;2:6698 10.4076/1757-1626-2-6698 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Al Dahouk S, Nöckler K. Implications of laboratory diagnosis on brucellosis therapy. Expert Rev Anti Infect Ther 2011;9:833–45. 10.1586/eri.11.55 [DOI] [PubMed] [Google Scholar]
- 23. Zou D, Zhou J, Jiang X. Diagnosis and management of spinal tuberculosis combined with brucellosis: a case report and literature review. Exp Ther Med 2018;15:3455–8. 10.3892/etm.2018.5812 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Dewan E, John B, Mathias A, et al. Brucellar spondylodiscitis mimicking tuberculosis. Chrismed J Heal Res 2015;2 10.4103/2348-3334.158721 [DOI] [Google Scholar]
- 25. Kis N, Kapmaz M, Duzkalir A, et al. Brucellar and tuberculous spondylodisitis: comparison of magnetic resonance imaging findings an unusual etiological cause of discitis: case report view project. J Turkish Spinal Surg 2015;26:285–90. [Google Scholar]
- 26. Hatipoglu C, Bilgin G, Tulek N, et al. Pulmonary involvement in brucellosis. J Infect 2005;51:116–9. 10.1016/j.jinf.2004.10.004 [DOI] [PubMed] [Google Scholar]
- 27. Smith M, Bhimji S. Brucellosis - StatPearls. StatPearls Publishing LLC, 2017. https://www.ncbi.nlm.nih.gov/books/NBK441831/ [Google Scholar]
- 28. Dasari S, Naha K, Prabhu M. Brucellosis and tuberculosis: clinical overlap and pitfalls. Asian Pac J Trop Med 2013;6:823–5. 10.1016/S1995-7645(13)60145-5 [DOI] [PubMed] [Google Scholar]
