Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Filter by Categories
Case Report
Case Report with Review of literature
Case Series
Commentary
Current Issue
Editorial
Guest Editorial
Guided Interventions
Invited Editorial
Letter to the Editor
Media and News
Original Article
Pictorial Review
Review Article
Technical Note
The Joint Effort: MSK Quiz
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Filter by Categories
Case Report
Case Report with Review of literature
Case Series
Commentary
Current Issue
Editorial
Guest Editorial
Guided Interventions
Invited Editorial
Letter to the Editor
Media and News
Original Article
Pictorial Review
Review Article
Technical Note
The Joint Effort: MSK Quiz
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Filter by Categories
Case Report
Case Report with Review of literature
Case Series
Commentary
Current Issue
Editorial
Guest Editorial
Guided Interventions
Invited Editorial
Letter to the Editor
Media and News
Original Article
Pictorial Review
Review Article
Technical Note
The Joint Effort: MSK Quiz
View/Download PDF

Translate this page into:

Review Article
8 (
1
); 59-72
doi:
10.25259/IJMSR_50_2025

Unraveling the masquerade: Decoding mimics of spinal tuberculosis

Department of Radiodiagnosis and Interventional Radiology, VMMC and Safdarjung Hospital, New Delhi, India.
Department of Radiology, Holy Family Hospital, New Delhi, India.
Central Institute of Orthopaedics, VMMC and Safdarjung Hospital, New Delhi, India.
Department of Radiology, Medanta-The Medicity, Gurugram, Haryana, India.
Department of Radiology, MGM Seven Hills Hospital, Visakhapatnam, Andhra Pradesh, India.
Author image
Corresponding author: Neha Nischal, Department of Radiology, Holy Family Hospital, New Delhi, India. neha.nischal@gmail.com
Licence
This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-Share Alike 4.0 License, which allows others to remix, transform, and build upon the work non-commercially, as long as the author is credited and the new creations are licensed under the identical terms.

How to cite this article: Narwani A, Nischal N, Sreenivasan R, Boruah T, Singh J, Bugata S, et al. Unraveling the masquerade: Decoding mimics of spinal tuberculosis. Indian J Musculoskelet Radiol. 2026;8:59-72. doi: 10.25259/IJMSR_50_2025

Abstract

Spinal tuberculosis (TB) is the most common form of skeletal TB and a significant cause of morbidity in endemic regions. Although culture, histopathology, and molecular assays provide definitive diagnosis, empirical anti-tubercular therapy is frequently initiated due to cost and accessibility constraints. While convenient, this practice risks serious consequences, including toxicity, emergence of drug resistance, and delayed recognition of alternative conditions that radiologically resemble TB. A wide spectrum of disorders can mimic spinal TB, which include a myriad of infective etiologies include pyogenic and fungal spondylodiscitis and brucellosis, as well as non-infective mimics, such as inflammatory disorders, degenerative changes, and neoplastic processes. Atypical manifestations of TB itself further complicate interpretation, reinforcing the limitations of relying solely on imaging. In such contexts, empirical therapy not only masks critical alternative diagnoses but also fosters multidrug-resistant and extensively drug-resistant strains of Mycobacterium tuberculosis. Tissue diagnosis, preferably through computed tomography-guided biopsy, remains the cornerstone of accurate characterization and should precede therapy whenever feasible. This review article aims to highlight the diverse spectrum of spinal TB mimickers, emphasizing their radiological features, diagnostic challenges, and the indispensable role of tissue confirmation in guiding appropriate management.

Keywords

Biopsy
Mimics
Radiology
Spinal tuberculosis
Spondylodiscitis

INTRODUCTION

Tuberculosis (TB) of the spine is the most common form of skeletal TB, accounting for approximately 50% of all cases.[1] In endemic countries like India, TB constitutes a staggering 80% cases of all infective spondylodiscitis.[2] Isolation of Mycobacterium tuberculosis bacilli through culture of tissue samples obtained through biopsy remains the gold standard for the definitive diagnosis of TB. Additional diagnostic methods include Ziehl–Neelsen staining and histopathological examination, which involve the detection of caseous necrosis and epithelioid granulomas. The emergence of polymerase chain reaction-based techniques, such as Cartridge-Based Nucleic Acid Amplification Test and GeneXpert, has gained increased recognition due to high sensitivity, specificity, and the ability to assay for simultaneous drug resistance. However, in resource-limited settings, the norm of cost-cutting, coupled with logistical restraints, poses a significant challenge in establishing histopathological confirmation of TB. In developing countries such as the South Asian subcontinent, where TB is omnipresent, the institution of empirical anti-tubercular therapy (ATT) without histopathological or microbiological confirmation is a fairly common practice.

However, in many scenarios, it has proven to be counterproductive as it leads to unnecessary ATT administration in radiologically similar, but nontubercular etiologies, leading to both misdiagnosis, as well as ATT toxicity and adverse effects. This also leads to potentially hazardous diagnostic and therapeutic delays in the management of these entities - the tubercular mimics. These include infectious conditions such as pyogenic and fungal spondylodiscitis, granulomatous lesions of the spine, and primary and secondary malignancies.[3] The chances of misdiagnosis are further accentuated due to the increasing incidence of atypical lesions, Human Immunodeficiency Virus (HIV) co-infection, and multidrug-resistant TB.[4] Furthermore, the blind administration of ATT has led to the emergence of multidrug-resistant (MDR) and extensively drug-resistant (XDR) strains of M. tuberculosis that are notoriously difficult to treat, worsening the situation.[5] The literature focusing on non-tubercular spondylodiscitis, both infectious and non-infectious, is surprisingly inadequate.

On the opposite side of the spectrum, tubercular spondylodiscitis has variable and unique presentations that can appear confusing. The incidence of atypical lesions is on an upward trend, which may lead to grave misdiagnosis. Atypical imaging features combined with indolent and non-specific symptoms often lead to a delay in diagnosis of something as ubiquitous and readily treatable as TB, leading to irreversible deformity and neurological symptoms. This is particularly indispensable in the background of TB-endemic countries, where TB can cause anything and everything, and should always be considered a valid first differential.

Spreading knowledge about non-tubercular spondylodiscitis mimicking TB on imaging is crucial to enable radiologists and treating clinicians to accurately identify subtle differentiating cues on imaging and to inculcate histopathological and microbiological confirmation in mainstream practice. We present stimulating cases that were hitherto diagnosed with TB on imaging, which proved to be non-tubercular conditions. The most common mimickers included neoplastic conditions (both benign and malignant) and aseptic inflammatory conditions (such as rheumatoid arthritis and ankylosing spondylitis). Other entities encompassed degenerative disc disease, pyogenic spondylodiscitis, and even Paget’s disease. This article aims to summarize the conditions that mimic Pott’s spine and clues to radiological diagnosis.

MIMICKERS OF SPINAL TB

Non-tubercular infectious spondylodiscitis

Pyogenic spondylodiscitis usually presents as a result of hematogenous dissemination from a septic focus [Figure 1]. There has been an increasing incidence of spondylodiscitis in recent years due to increasing numbers of older patients with chronic diseases such as diabetes mellitus and renal failure, steroid and other immunosuppressive therapies, sickle cell disease (SCD), HIV infection, and other immunocompromised states, and intravenous drug abuse.[6,7] Escherichia coli and methicillin-resistant Staphylococcus aureus are isolated from blood cultures in most of the patients of pyogenic spondylodiscitis.[8] Bacterial infections, particularly from S. aureus, are the most common cause of vertebral osteomyelitis, accounting for 40–67% of cases.[9] Although most of these Staphylococcus infections respond to methicillin, Methicillin-resistant S. aureus strains are becoming more prevalent.[10] Other common bacterial culprits include Gram-negative bacteria such as E. coli, Pseudomonas, and Proteus species.[11] Certain conditions increase the risk of specific infections: Streptococcus is more common in patients with infectious endocarditis, while Salmonella is a frequent cause in individuals with SCD. An isolated psoas abscess is often caused by bacteria such as S. aureus, rather than TB.[12] Common risk factors for the same include immunocompromised states such as diabetes or intravenous drug use, and the infection can spread from intra-abdominal sources. Due to the endemicity of TB, radiologists and clinicians often tend to misdiagnose it as TB even when there is no obvious evidence of Pott’s spine.

Pyogenic spondylodiscitis. (a) Sagittal T1-weighted and (b) short tau inversion recovery magnetic resonance images demonstrate edema involving the L1-L2 as well as the L4-L5 intervertebral discs and the adjacent vertebral bodies, suggesting infective spondylodiscitis. This was reported as Pott’s spine, and the patient started on ATT with inadequate clinical response after 2 months. (c and d) Axial computed tomography images show the transpedicular biopsy of the affected vertebra. Cultures were positive for Escherichia coli with no histopathological or microbiological evidence of tuberculosis. ATT: Anti-tubercular therapy
Figure 1:
Pyogenic spondylodiscitis. (a) Sagittal T1-weighted and (b) short tau inversion recovery magnetic resonance images demonstrate edema involving the L1-L2 as well as the L4-L5 intervertebral discs and the adjacent vertebral bodies, suggesting infective spondylodiscitis. This was reported as Pott’s spine, and the patient started on ATT with inadequate clinical response after 2 months. (c and d) Axial computed tomography images show the transpedicular biopsy of the affected vertebra. Cultures were positive for Escherichia coli with no histopathological or microbiological evidence of tuberculosis. ATT: Anti-tubercular therapy

Approximately 30% of cases are caused by M. tuberculosis in non-endemic countries,[9] whereas in endemic countries, TB constitutes a staggering 80% cases of all infective spondylodiscitis.[2] Other bacterial infections cause most remaining cases, and a definitive organism is never identified in 21–34% of cases.[13] The closest imaging mimicker for tubercular spondylodiscitis would be spinal brucellosis, in a relevant clinical scenario, and thus needs to be differentiated intricately. It should be considered in patients from rural or endemic regions with occupational or dietary exposure to livestock or unpasteurized dairy products presenting with prolonged fever and musculoskeletal symptoms.[14] While spinal TB tends to affect the thoracic or thoracolumbar junction, brucellosis shows a distinct predilection for the lumbar spine [Figure 2]. Brucellosis primarily affects the anterior aspect of the vertebral endplates, leading to a characteristic serrated or irregular outline. The vertebral body itself is often relatively preserved. In contrast, TB frequently causes significant, widespread destruction of the vertebral bodies, often leading to their collapse. The rarity of large paravertebral abscesses and the absence of extensive subligamentous spread and multi-level contiguous vertebral involvement in brucellosis are a few hallmark features that can differentiate it from tubercular spondylodiscitis.

Spinal Brucellosis. (a) Lateral radiograph of the lumbar spine in a 34-year-old female with backache for about 3 months, shows reduced L3-L4 disc space with erosion of the inferior endplate of L3. (b) Sagittal T1-weighted, (c) T2-weighted, and (d) coronal short tau inversion recovery magnetic resonance images demonstrate fluid signal in the L3-L4 intervertebral disc and adjacent endplate erosions (white arrows in d), suggesting infective spondylodiscitis. The patient also had lesions in the lungs and liver. Based on imaging, empirical anti-tubercular therapy was given with inadequate clinical response after 2 months. (e) A biopsy was performed from the liver lesion, and the high-power photomicrograph (hematoxylin & eosin, ×400) shows the presence of non-caseating granulomas. Enzyme-linked immunosorbent assay immunoglobulin G for Brucella was positive in this patient, which confirmed the diagnosis.
Figure 2:
Spinal Brucellosis. (a) Lateral radiograph of the lumbar spine in a 34-year-old female with backache for about 3 months, shows reduced L3-L4 disc space with erosion of the inferior endplate of L3. (b) Sagittal T1-weighted, (c) T2-weighted, and (d) coronal short tau inversion recovery magnetic resonance images demonstrate fluid signal in the L3-L4 intervertebral disc and adjacent endplate erosions (white arrows in d), suggesting infective spondylodiscitis. The patient also had lesions in the lungs and liver. Based on imaging, empirical anti-tubercular therapy was given with inadequate clinical response after 2 months. (e) A biopsy was performed from the liver lesion, and the high-power photomicrograph (hematoxylin & eosin, ×400) shows the presence of non-caseating granulomas. Enzyme-linked immunosorbent assay immunoglobulin G for Brucella was positive in this patient, which confirmed the diagnosis.

Fungal infections are rare, making up only about 0.5% of cases, and viral and parasitic causes are even less common. However, when fungal infections do occur, they most often affect the vertebrae. Of these, about a third are due to Aspergillus and another third are due to Candida.[15,16]

Multiple contiguous vertebral body involvement with subligamentous, intraosseous, and paraspinal abscesses is a hallmark feature of tubercular spondylodiscitis,[17] which characteristically differentiates it from pyogenic spondylodiscitis, which often shows affection of a single vertebral body with absence of subligamentous and intraosseous abscesses. However, sometimes involvement of multiple contiguous vertebral bodies can also be seen in bacterial spondylodiscitis. In such a scenario, other corroborative findings, such as small, non-loculated abscesses with thick shaggy walls of pyogenic spondylodiscitis as compared to thin-walled, smooth, large, loculated abscesses in tubercular spine, should be given importance.[18] A diagnosis is easily confirmed when a patient’s blood cultures are positive. However, if they are negative, an image-guided biopsy is essential [Figure 3].

Flowchart for workup of suspected infective spondylodiscitis.
Figure 3:
Flowchart for workup of suspected infective spondylodiscitis.

The emerging role of quantitative magnetic resonance imaging (MRI) techniques, such as T2 mapping, offers a powerful, novel dimension. The T2 value of the involved vertebra in bacterial etiology was found to be markedly higher than that in Pott’s spine.[19] The different inflammatory processes, such as a caseating granuloma in TB versus an acute, suppurative infection in pyogenic spondylodiscitis, result in different fluid and tissue micromilieu within the affected vertebrae, affecting the T2 relaxation time.

Axial spondyloarthropathy (AS)

Inflammatory arthritic conditions such as ankylosing spondylitis/AS can present diagnostic challenges when differentiating from spinal TB [Figure 4]. Elevated inflammatory markers, including erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), add complexity to the diagnostic process, as they can be seen in both. Sacroiliitis, which is the hallmark of ankylosing spondylitis, may also resemble tuberculous infection patterns. When sacroiliac joint inflammation is associated with spinal changes consistent with infection, ankylosing spondylitis and other seronegative spondyloarthropathies should be considered.

Axial spondyloarthropathy (AS). (a) Sagittal and (b) coronal magnetic resonance images in a 28-year-old male with backache for 2–3 months showed intense edema in the left-sided posterior elements in the lumbar region (dashed circles in a and b). Repeat imaging after 6 months of anti-tubercular therapy was done in view of inadequate clinical response and to assess treatment response. Follow-up sagittal images of (c) thoracic and (d) lumbar spine revealed new areas of edema at the posterior corners of the thoracic spine (white arrows in c) with resolution of signal abnormalities in the lumbar spine (dashed circle in d). (e) Dedicated coronal T1-weighted and (f) short tau inversion recovery (STIR) images of the sacro-iliac joints demonstrate changes of chronic bilateral sacroiliitis (yellow circles in e) with active edema on the left (block arrow in f). Re-assessment of the earlier images revealed that these changes could be seen on the coronal STIR image (dashed arrow in b) with fatty metaplasia and edema in the SI joints. Further work-up confirmed AS with positive HLA-B27.
Figure 4:
Axial spondyloarthropathy (AS). (a) Sagittal and (b) coronal magnetic resonance images in a 28-year-old male with backache for 2–3 months showed intense edema in the left-sided posterior elements in the lumbar region (dashed circles in a and b). Repeat imaging after 6 months of anti-tubercular therapy was done in view of inadequate clinical response and to assess treatment response. Follow-up sagittal images of (c) thoracic and (d) lumbar spine revealed new areas of edema at the posterior corners of the thoracic spine (white arrows in c) with resolution of signal abnormalities in the lumbar spine (dashed circle in d). (e) Dedicated coronal T1-weighted and (f) short tau inversion recovery (STIR) images of the sacro-iliac joints demonstrate changes of chronic bilateral sacroiliitis (yellow circles in e) with active edema on the left (block arrow in f). Re-assessment of the earlier images revealed that these changes could be seen on the coronal STIR image (dashed arrow in b) with fatty metaplasia and edema in the SI joints. Further work-up confirmed AS with positive HLA-B27.

These patients typically demonstrate hypointensity in subchondral bone on T1-weighted imaging and hyperintensity on T2-weighted sequences, resembling paradiscal TB. Andersson lesions may also mimic tuberculous discitis, as they exhibit bright signals on T2 sequences.[20] When ankylosing spondylitis affects multiple spinal levels, it can be mistaken for disseminated TB.

Several distinguishing features that can aid in accurate diagnosis include the absence of paravertebral abscesses and the presence of syndesmophytes in ankylosing spondylitis. Andersson lesions characteristically affect all three spinal columns, which serves as another differentiating factor. Sacroiliitis is usually bilateral in inflammatory conditions, although affliction may be asymmetrical. In the sacroiliac joints, active lesions are characterized by bone marrow edema, synovitis, capsulitis, and enthesitis, all of which present as high-signal intensity on short tau inversion recovery (STIR) sequences.[21] Symmetrical edema in the lower and posterior joint sections is typical, while unilateral edema may suggest an infectious cause. In the spine, active inflammation includes spondylitis, which appears as high-signal intensity at the vertebral corners (Romanus lesions), and spondylodiscitis (Andersson lesions), which affects the cortical plates and intervertebral discs. Other active lesions include arthritis in the facet joints and costovertebral joints, and enthesitis of the spinal ligaments.

Beyond active inflammation, MRI can also reveal features of long-standing damage. In the sacroiliac joints, this includes low-signal intensity areas representing subchondral sclerosis, bony defects from erosions, and bony fusion (ankylosis). Fat depositions, which appear as high-signal intensity, are also a sign of past inflammation. Similarly, chronic changes in the spine are characterized by new bone formation at the vertebral corners, known as syndesmophytes, and bony bridging, or ankylosis, both of which show low-signal intensity.[22] Fat deposits on the vertebral corners are also considered a sign of previous inflammation and are a predictor of future syndesmophyte formation.

Joint effusion and extra-articular fluid collections/abscesses (e.g., extracapsular fluid collections, periarticular muscle edema) are significantly more common in infective sacroiliitis on MRI compared to inflammatory sacroiliitis. These findings help distinguish infection from spondyloarthropathy-related sacroiliitis. In contrast, inflammatory sacroiliitis due to seronegative spondyloarthropathy typically shows bone marrow edema and capsulitis without pronounced fluid collections extending into the surrounding soft tissues.[23]

Paget’s disease

Paget’s disease can present as a hypointense lesion on T1-weighted images and a hyperintense lesion on T2-weighted images, which can be confused with spinal TB. However, hematological parameters and the characteristic absence of soft-tissue components in Paget’s disease can offer additional clues to differentiate between them. In the rare instance where vertebrae are affected by the lytic phase of Paget’s disease, radiographs reveal a marked reduction in bone density, leading to a “ghost vertebra” appearance. This can be challenging to distinguish from other causes of osteolysis and vertebral collapse, such as TB of the spine.[24] However, computed tomography scans can confirm the diagnosis by demonstrating classic features of Paget’s disease, such as cortical thickening, trabecular hypertrophy, and vertebral expansion, findings that are typically absent in tubercular spondylitis. Another presentation of Paget’s disease, which often can mimic Pott’s spine, is “ivory vertebra.” It presents as a diffusely sclerotic vertebral body on radiograph and non-contrast computed tomography, with a corresponding hypointense lesion on T1- and T2-weighted images. While positron emission tomography and bone scans can aid in diagnosis, a histopathological examination remains the definitive mode to diagnose Paget’s disease.[25]

Degenerative changes

Type 1 modic changes, characterized by hypo-intense lesions on T1-weighted MRI and hyper-intense lesions on T2-weighted MRI, can present a diagnostic dilemma due to their potential to involve paraspinal soft-tissue changes [Figure 5]. This similarity often leads to confusion with spinal TB. Asymmetric disc space narrowing with dehydrated disc, presence of osteophytes, and preserved end plates point toward a diagnosis of degenerative spine. Diffusion-weighted and T2-weighted/fluid-sensitive sequences are instrumental in differentiating infectious/inflammatory processes from degenerative spondylosis. Osteomyelitis, discitis, and abscess reveal hyperintensity on T2-weighted and fluid-sensitive sequences, which can be critical to clinch a diagnosis. The presence of a characteristic “claw sign” on sagittal diffusion-weighted images is a hallmark of degenerative spine disease [Figure 6]. It appears as well-defined, linear, and often paired areas of high-signal intensity located at the margins of adjacent vertebral bodies, near the affected disc.[26] Recognition of this sign is instrumental in avoiding unnecessary biopsies in these patients. Believed to represent a physiological reactive response, this sign is particularly useful for distinguishing a degenerative condition with Modic type I changes from an infection.[27]

(a-d) Degenerative spondylosis. (a and d) Sagittal T1 and (b and e) sagittal short tau inversion recovery (STIR) and (c and f) coronal STIR magnetic resonance (MR) images in a 49-year-old female with backache. (a-c) Initial MR images at the time of presentation show intense edema adjoining the L2, L3 vertebral end-plates, predominantly on the right side, with subtle adjacent soft-tissue edema. The intervening disc shows fluid signal intensity (white arrow in b). This was reported as Pott’s spine, and the patient started on empirical anti-tubercular therapy. (d-f) Follow-up imaging after 9 months shows subsidence of edema and loss of intervertebral disc space on the right with increased scoliosis. However, the prominent marginal osteophytes are clearly visualized (dashed circle in f), favoring degenerative etiology. Furthermore, there is an absence of end-plate erosions/osseous destruction, bony ankylosis, or paravertebral collections, which would be expected in a tubercular disease. Degenerative edema would also be expected to subside on conservative management.
Figure 5:
(a-d) Degenerative spondylosis. (a and d) Sagittal T1 and (b and e) sagittal short tau inversion recovery (STIR) and (c and f) coronal STIR magnetic resonance (MR) images in a 49-year-old female with backache. (a-c) Initial MR images at the time of presentation show intense edema adjoining the L2, L3 vertebral end-plates, predominantly on the right side, with subtle adjacent soft-tissue edema. The intervening disc shows fluid signal intensity (white arrow in b). This was reported as Pott’s spine, and the patient started on empirical anti-tubercular therapy. (d-f) Follow-up imaging after 9 months shows subsidence of edema and loss of intervertebral disc space on the right with increased scoliosis. However, the prominent marginal osteophytes are clearly visualized (dashed circle in f), favoring degenerative etiology. Furthermore, there is an absence of end-plate erosions/osseous destruction, bony ankylosis, or paravertebral collections, which would be expected in a tubercular disease. Degenerative edema would also be expected to subside on conservative management.
Degenerative spondylosis. (a) Sagittal T1 and (b) sagittal T2, and (c) diffusion-weighted imaging (DWI) magnetic resonance images in a 47-year-old male with backache for few months and low-grade fever for 15 days. Reduced L5-S1 intervertebral disc height is seen with mixed signal within the disc (white arrow in b). Mild marrow edema is present at the adjoining end-plates. Note that the rest of the intervertebral discs appear normal in morphology (black stars in a and b). The involvement of a single disc and a history of fever raised suspicion of a low-grade infection. DWI (c) showed well-defined paired bands of hyperintensity at the margins of vascularized and normal marrow (black arrows), suggesting the presence of “claw-sign.” This was managed as a case of degenerative spondylosis.
Figure 6:
Degenerative spondylosis. (a) Sagittal T1 and (b) sagittal T2, and (c) diffusion-weighted imaging (DWI) magnetic resonance images in a 47-year-old male with backache for few months and low-grade fever for 15 days. Reduced L5-S1 intervertebral disc height is seen with mixed signal within the disc (white arrow in b). Mild marrow edema is present at the adjoining end-plates. Note that the rest of the intervertebral discs appear normal in morphology (black stars in a and b). The involvement of a single disc and a history of fever raised suspicion of a low-grade infection. DWI (c) showed well-defined paired bands of hyperintensity at the margins of vascularized and normal marrow (black arrows), suggesting the presence of “claw-sign.” This was managed as a case of degenerative spondylosis.

The presence of a vacuum phenomenon on CT is highly suggestive of a degenerated rather than infected disc. The end-plate changes in degeneration appear more irregular with the presence of end-plate sclerosis, whereas end-plate destruction or erosion favors infection.[28] However, the presence of gas does not rule out infection, and in the appropriate clinical context, this may be well seen in more severe cases of infective spondylodiscitis, as was shown by Schömig et al.[29]

While constitutional symptoms and hematological markers such as ESR and CRP can offer supportive evidence to help differentiate between degenerative spine and spinal TB, none of these indicators are definitive.

The patient’s age may be an important criterion, as degenerative spine disease is less likely in the younger patient. However, in older patients, differentiation is often challenging and therefore, a short imaging follow-up is recommended for excluding spinal TB in equivocal cases.

SCD

SCD is a hereditary hemoglobinopathy characterized by “sickling” of red blood cells (RBCs). This leads to a myriad of musculoskeletal manifestations secondary to anemia as well as end-organ damage due to vaso-occlusion and infarction. Clinically, this can result in a migratory pattern of “bone pains.” SCD also makes the patients prone to infections with resultant septic arthritis, osteomyelitis with Gram-negative bacteria, especially Salmonella, being isolated as the causative organism. In the spine, characteristic changes of “H-shaped” vertebral bodies are seen as a result of infarction with consequent end-plate depression. In the early stage, however, the episodes of bone ischemia may present as multiple areas of bone marrow edema without a definite infarct. This can lead to confusion with multifocal infections, with TB being a preferred diagnosis in endemic areas. Furthermore, the presence of secondary bacterial infection with the development of abscesses may confuse the clinical picture [Figure 7].

Sickle cell disease (SCD). A 17-year-old male with multiple episodes of fever, long-standing history of bone pains, and anemia. (a) Initial axial, (b) coronal short tau inversion recovery (STIR) magnetic resonance (MR) images of the pelvis and (c) sagittal STIR image of the lumbosacral spine show a large right gluteal abscess (black star) with multiple bone lesions (white arrows in a-c) suggesting multifocal infections. Note that the disc is preserved. A differential of multifocal tuberculosis was considered along with SCD. Work-up confirmed SCD, and the cultures from the gluteal abscess were positive for Salmonella. (d and e) Coronal T2-weighted and (f) sagittal STIR MR images performed 1 year later, reveal well-defined infarcts in the right femoral head (red arrow in d), a residual fistulous tract from the right femur (black star in e), and the presence of end-plate depressions of the L5 vertebral body (yellow arrows in f).
Figure 7:
Sickle cell disease (SCD). A 17-year-old male with multiple episodes of fever, long-standing history of bone pains, and anemia. (a) Initial axial, (b) coronal short tau inversion recovery (STIR) magnetic resonance (MR) images of the pelvis and (c) sagittal STIR image of the lumbosacral spine show a large right gluteal abscess (black star) with multiple bone lesions (white arrows in a-c) suggesting multifocal infections. Note that the disc is preserved. A differential of multifocal tuberculosis was considered along with SCD. Work-up confirmed SCD, and the cultures from the gluteal abscess were positive for Salmonella. (d and e) Coronal T2-weighted and (f) sagittal STIR MR images performed 1 year later, reveal well-defined infarcts in the right femoral head (red arrow in d), a residual fistulous tract from the right femur (black star in e), and the presence of end-plate depressions of the L5 vertebral body (yellow arrows in f).

In such a scenario, it is always prudent to get blood work up in terms of peripheral smear to look for abnormal RBC morphology. Other signs of SCD should be looked, for example, presence of bone infarcts elsewhere, small calcified spleen (autosplenectomy), extramedullary hematopoiesis. In case of doubt, image-guided biopsy should be done whenever possible to isolate the offending organism.

Metastasis

Metastasis in the spine can be difficult to distinguish from central spinal TB, as both can present with similar symptoms such as pain, weight loss, and appetite loss. The presence of a soft-tissue mass further complicates differentiation. Metastatic deposits typically spare the disc space and end plates, while spinal TB often shows irregular endplates. Furthermore, involvement of the posterior elements and pedicles is more indicative of a malignant process and is rarely seen in infective spondylodiscitis. Contrast-enhanced MRI is useful for differentiating between an abscess (peripheral rim enhancement), which is more common in TB, and a solid soft-tissue mass (homogenous enhancement) seen in metastasis.[30]

This should be given consideration even in patients with a history of TB and/or presence of mediastinal lymphadenopathy [Figure 8]. The role of image-guided biopsy and histopathological evaluation cannot be overemphasized in such scenarios.

Spinal metastasis. A 47-year-old male with a past history of anti-tubercular therapy intake for pulmonary Koch’s 7 years ago, presenting with backache after a trivial trauma. (a) Sagittal T1, (b) short tau inversion recovery (STIR) and (c) post-contrast T1 fat-suppressed magnetic resonance images reveal contiguous involvement of mid dorsal vertebral bodies (dashed circle in a-c) as well as posterior elements (white arrows in a-c). Prevertebral soft tissue is evident (white star in b and c). (d) Coronal STIR and (e) post-contrast images show the presence of extensive conglomerated necrotic mediastinal lymphadenopathy (dashed white arrows). With the given clinical history, tuberculosis was given as a likely diagnosis. Biopsy from the supraclavicular lymph node, however, confirmed metastatic disease from small cell neuroendocrine carcinoma of the lung with nodal as well as spinal metastasis.
Figure 8:
Spinal metastasis. A 47-year-old male with a past history of anti-tubercular therapy intake for pulmonary Koch’s 7 years ago, presenting with backache after a trivial trauma. (a) Sagittal T1, (b) short tau inversion recovery (STIR) and (c) post-contrast T1 fat-suppressed magnetic resonance images reveal contiguous involvement of mid dorsal vertebral bodies (dashed circle in a-c) as well as posterior elements (white arrows in a-c). Prevertebral soft tissue is evident (white star in b and c). (d) Coronal STIR and (e) post-contrast images show the presence of extensive conglomerated necrotic mediastinal lymphadenopathy (dashed white arrows). With the given clinical history, tuberculosis was given as a likely diagnosis. Biopsy from the supraclavicular lymph node, however, confirmed metastatic disease from small cell neuroendocrine carcinoma of the lung with nodal as well as spinal metastasis.

Primary tumors-benign

Osteoid osteoma, a benign bone tumor, can be mistaken for spinal TB because it often presents with significant surrounding inflammation in MRI[31] [Figure 9]. When located eccentrically within a vertebral body, it can appear as an abscess-like lesion. Several clues can help distinguish osteoid osteoma from TB. A key indicator for osteoid osteoma is the presence of pain that worsens at night, which typically responds well to nonsteroidal anti-inflammatory drugs. A CT scan is usually diagnostic, revealing a characteristic nidus surrounded by a sclerotic rim. Osteoblastoma is similar to osteoid osteoma but with a larger nidus (>1.5cm). These have a propensity to occur in posterior elements, which is a less favored location for TB.[32]

Osteoid osteoma. (a) Short tau inversion recovery coronal and (b) sagittal magnetic resonance images in a 21-year-old male with severe back pain, show edema in the left transverse process and costo-transverse joint (white solid arrow in a and b), and a suspicious round hypointensity (dashed arrow in b). (c) Axial computed tomography image shows a characteristic nidus (white arrow) surrounded by a sclerotic rim (black arrow), pathognomonic of osteoid osteoma.
Figure 9:
Osteoid osteoma. (a) Short tau inversion recovery coronal and (b) sagittal magnetic resonance images in a 21-year-old male with severe back pain, show edema in the left transverse process and costo-transverse joint (white solid arrow in a and b), and a suspicious round hypointensity (dashed arrow in b). (c) Axial computed tomography image shows a characteristic nidus (white arrow) surrounded by a sclerotic rim (black arrow), pathognomonic of osteoid osteoma.

Hence, the presence of unilateral edema on MRI, especially in an unusual location like posterior elements, accompanied by normal ESR and CRP levels, which are typically elevated in infections like TB, should raise suspicion for osteoid osteoma or osteoblastoma instead, and further imaging with CT should always be performed. CT usually clinches the diagnosis and can be used to guide ablation of the nidus as well.[33]

An extremely important differential in the pediatric population that should always be considered is Langerhans cell histiocytosis (LCH). Involvement of the vertebra is known to occur in 10–18% of pediatric LCH.[34] Collapsed vertebral body-vertebra plana can be found in both TB of the spine and LCH [Figure 10].[35,36] However, the absence of paravertebral soft-tissue favors LCH. Nonetheless, biopsy should always be performed as a thumb rule in children to rule out LCH, for better survival and prognosis.

Langerhans cell histiocytosis (LCH). (a) Lateral cervical spine lateral radiograph in an 8-year-old male with neck pain and restricted movement demonstrates complete collapse of C4 vertebral body- vertebra plana (white arrow in a-c). (b) Sagittal T2-weighted, (c) T1-weighted and (d) T1-weighted fat saturation contrast-enhanced magnetic resonance images show complete collapse of the solitary cervical vertebral body with anterior epidural enhancement (dashed arrow in d). Biopsy proved it to be LCH.
Figure 10:
Langerhans cell histiocytosis (LCH). (a) Lateral cervical spine lateral radiograph in an 8-year-old male with neck pain and restricted movement demonstrates complete collapse of C4 vertebral body- vertebra plana (white arrow in a-c). (b) Sagittal T2-weighted, (c) T1-weighted and (d) T1-weighted fat saturation contrast-enhanced magnetic resonance images show complete collapse of the solitary cervical vertebral body with anterior epidural enhancement (dashed arrow in d). Biopsy proved it to be LCH.

Primary tumors-malignant

Primary malignant spinal tumors of the spine, such as myeloma and plasmacytoma, lymphoma, and Ewing’s sarcoma, can also mimic spinal TB, owing to hyperintensity on T2-weighted imaging and hypointensity on T1-weighted imaging [Figures 11 and 12].[37] Primary malignancy of the spine can mimic central TB. Spondylitis without discitis, flattened vertebra with sclerosis, or lesions within the spinal canal and epidural space are seen in tumors more often than TB. However, the presence of paravertebral abscesses with subligamentous spread, sparing of the posterior elements of the vertebra, and presence of skip lesions often favor the diagnosis of spinal TB over primary malignancy.[38,39]

Spinal lymphoma. (a) Sagittal and (b) coronal short tau inversion recovery magnetic resonance images in a 44-year-old male who was referred from another centre with large paravertebral soft-tissue collections along with marrow edema involving multiple vertebral bodies as well as posterior elements. (c) Intraoperative image shows large paravertebral soft tissues, which proved to be lymphomatous deposits.
Figure 11:
Spinal lymphoma. (a) Sagittal and (b) coronal short tau inversion recovery magnetic resonance images in a 44-year-old male who was referred from another centre with large paravertebral soft-tissue collections along with marrow edema involving multiple vertebral bodies as well as posterior elements. (c) Intraoperative image shows large paravertebral soft tissues, which proved to be lymphomatous deposits.
Plasmacytoma. (a) Sagittal T2-weighted and (b) T1-weighted Magnetic resonance images in a 44-year-old male with a history of pain over the lower back for 3 months, show vertebral body destruction, with serpiginous intraosseous soft tissue with herniation of disc into the L4 vertebra (white arrow in a and b). (c) C-arm-guided vertebral biopsy was done. (d) High power (hematoxylin & eosin, ×400) photomicrograph shows diffuse sheets of mature and atypical plasma cells with eccentric nuclei, coarse “clock-face” chromatin, prominent nucleoli, and perinuclear hof, consistent with plasmacytoma.
Figure 12:
Plasmacytoma. (a) Sagittal T2-weighted and (b) T1-weighted Magnetic resonance images in a 44-year-old male with a history of pain over the lower back for 3 months, show vertebral body destruction, with serpiginous intraosseous soft tissue with herniation of disc into the L4 vertebra (white arrow in a and b). (c) C-arm-guided vertebral biopsy was done. (d) High power (hematoxylin & eosin, ×400) photomicrograph shows diffuse sheets of mature and atypical plasma cells with eccentric nuclei, coarse “clock-face” chromatin, prominent nucleoli, and perinuclear hof, consistent with plasmacytoma.

However, misdiagnosis is still common in areas that lack experienced radiologists, scenarios being complicated by the fact that endemicity of TB may induce a cognitive bias in misinterpretation. Delayed diagnosis in these cases can lead to disease progression and poorer patient outcomes. This further highlights the importance of a timely biopsy for accurate characterization of indeterminate lesions. Table 1 summarises the common mimickers of Pott’s spine and radiological clues for differentiating these.

Table 1: Imaging features differentiating spinal tuberculosis from its common mimics
Entity Typical level Disc involvement Vertebral body pattern Paravertebral/epidural soft tissue Subligamentous spread Key distinguishing imaging clues
Spinal tuberculosis (Pott’s) Thoracic, thoracolumbar Late Multilevel contiguous destruction, anterior wedging, collapse Large, thin-walled, smooth, often multiloculated abscesses Common Skip lesions, subligamentous spread, preserved disc early, gibbus deformity
Pyogenic spondylodiscitis Lumbar >cervical Early Usually single level; endplate erosion Small, thick-walled, shaggy abscesses Rare Prominent discitis, rapid destruction, high T2 signal in vertebra (T2 mapping)
Brucellar spondylitis Lumbar Early Anterior endplate erosions with serrated margins; vertebral body preserved Minimal or absent Rare Absence of large abscess, lack of multilevel contiguous disease
Fungal spondylodiscitis Thoracic, lumbar Variable Vertebral body osteomyelitis Paraspinal collections possible Variable Immunocompromised host, indolent course, biopsy required
Axial spondyloarthropathy SI joints→lumbar spine Pseudodiscitis Corner erosions, sclerosis Absent Absent Bilateral sacroiliitis, syndesmophytes, Romanus and Andersson lesions
Paget’s disease Lumbar No Vertebral expansion, sclerosis or lysis Absent Absent Cortical thickening, trabecular hypertrophy, ivory or ghost vertebra
Degenerative Modic-1 Lumbar Dehydrated Endplate irregularity with sclerosis Minimal Absent DWI “claw sign”, vacuum phenomenon, osteophytes
Metastases Thoracic >lumbar Spared Vertebral body+posterior element involvement Solid enhancing mass Absent Disc preserved, pedicle involvement, homogeneous enhancement
Osteoid osteoma/osteoblastoma Posterior elements No Focal nidus with sclerosis Reactive edema Absent Night pain, CT-visible nidus
LCH (children) Cervical, thoracic No Vertebra plana Minimal Absent Solitary collapse without abscess

CT: Computed tomography, LCH: Langerhans cell histiocytosis, DWI: Diffusion-weighted imaging, SI: Sacroiliac joint

CONCLUSION

In regions where TB is endemic, the empirical initiation of ATT for suspected spinal TB is a common practice. However, this approach carries significant risks, as it can inadvertently delay the accurate diagnosis of other critical conditions, particularly malignancies, which may present with similar clinical and radiological features. Second, the empirical use of ATT contributes to the global challenge of drug-resistant TB, as inappropriate or incomplete treatment can foster the emergence and selection of MDR-TB and XDR-TB strains of M. tuberculosis. This not only renders treatment more complex and prolonged for individual patients but also exacerbates the public health threat posed by increasingly difficult-to-treat forms of TB.

Many studies and guidelines strongly advocate for a definitive tissue diagnosis of TB whenever feasible. A heightened index of suspicion is paramount, especially for lesions exhibiting a central location within the spine or those with minimal to no associated soft-tissue component. For cases where a tissue diagnosis cannot be obtained, rigorous patient follow-up is essential. A lack of clinical response to empirical treatment in such instances should immediately prompt a re-evaluation and consideration of alternative diagnoses.

Biopsy remains the cornerstone for differentiating these conditions, offering a confirmatory diagnosis that guides appropriate management. Specifically, CT-guided biopsy has emerged as the standard of care and the gold standard for tissue diagnosis in the management of infective spondylodiscitis. Its universal adoption is strongly recommended before initiating treatment for this condition, ensuring diagnostic precision and preventing potentially adverse outcomes associated with misdiagnosis or delayed definitive therapy.

Ethical approval:

Institutional Review Board approval is not required.

Declaration of patient consent:

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patients have given their consent for their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Conflict of interest:

Neha Nischal and Dharmendra Kumar Singh are on the Editorial Board of the Journal.

Use of artificial intelligence (AI)-assisted technology for manuscript preparation:

The authors confirm that there was no use of artificial intelligence (AI)-assisted Technology for assisting in the writing or editing of the manuscript, and no images were manipulated using AI.

Financial support and sponsorship: Nil.

References

  1. , , , . Spinal tuberculosis: cur-rent concepts. Glob Spine J. 2018;8:96S-108S.
    [CrossRef] [PubMed] [Google Scholar]
  2. World Health Organization Global Tuberculosis Report 2020. Geneva: World Health Organization; .
    [Google Scholar]
  3. , , , , . Establishing the diagnosis of tuberculous vertebral osteomyelitis. European Spine Journal. 2012;22(Suppl 4):579-586.
    [CrossRef] [PubMed] [Google Scholar]
  4. , et al. Role of percutaneous transpedicular biopsy in diagnosis of spinal tuberculosis and its correlation with the clinico-radiological features. Indian J Tuberc 2018
    [CrossRef] [PubMed] [Google Scholar]
  5. , , , . Drug-resistant spinal tuberculosis. Indian J Orthop. 2018;52(02):100-107.
    [CrossRef] [PubMed] [Google Scholar]
  6. , , , . Incidence and treatment trends of infectious spondylodiscitis in South Korea: A nationwide population-based study. PLoS ONE. 2023;18:e0287846.
    [CrossRef] [PubMed] [Google Scholar]
  7. . Spinal infections. J. Bone Joint Surg Br. 2005;87:1454-1458.
    [CrossRef] [PubMed] [Google Scholar]
  8. , , , , , , , , . Concordance of results of blood and tissue cultures from patients with pyogenic spondylitis: a retrospective cohort study. Clin Microbiol Infect. 2018;24(3):279-282.
    [CrossRef] [PubMed] [Google Scholar]
  9. , , , , , , , . A systematic review: Characteristics, complications and treatment of spondylodiscitis. Eur. Rev. Med. Pharmacol. Sci. 2019;23(Suppl. S2):117-128.
    [Google Scholar]
  10. , , , , , , , . Epidemiological and clinical features of pyogenic spondylodiscitis. Eur. Rev. Med. Pharmacol. Sci. 2012;16(Suppl. S2):2-7.
    [Google Scholar]
  11. , , , , , , , , , , et al. Pyogenic Spondylitis: Clinical Features, Diagnosis and Treatment. Kurume Med. J. 2019;65:83-89.
    [CrossRef] [PubMed] [Google Scholar]
  12. , , , et al. Psoas abscess: Clinical features, microbiology and treatment. Postgrad Med J. 2025;101(1185):482-488.
    [Google Scholar]
  13. , , . Spontaneous Spondylodiscitis-Epidemiology, Clinical Features, Diagnosis and Treatment. Folia Med. 2017;59:254-260.
    [CrossRef] [PubMed] [Google Scholar]
  14. , , , , . Brucellosis Masquerading as Disseminated Tuberculosis: A Clinical Case Report. Perm J. 2023;27(3):110-115.
    [CrossRef] [PubMed] [Google Scholar]
  15. , , , , , , , , , , et al. Fungal Osteomyelitis: A Systematic Review of Reported Cases. Microorganisms. 2023;11:1828.
    [CrossRef] [PubMed] [Google Scholar]
  16. , , , , , , . Spondylodiscitis Caused by Aspergillus Species. Diagnostics. 2021;11:1899.
    [CrossRef] [PubMed] [Google Scholar]
  17. , , , , . Imaging findings of Pott's disease. Eur Spine J. 2013;22:567-78.
    [CrossRef] [PubMed] [Google Scholar]
  18. , et al. "Unmasking the Great ImitatorsNoninfectious Conditions Masquerading as Spinal Tuberculosis in a Developing Country: A Single-Center Case Series Analysis" North American Spine Society Journal. 2023;16:100245. 1 Dec
    [CrossRef] [PubMed] [Google Scholar]
  19. , et al. "Differentiating Brucella Spondylitis from Tuberculous Spondylitis by the Conventional MRI and MR T2 Mapping: A Prospective Study" European Journal of Medical Research. 2021;26(1)
    [CrossRef] [PubMed] [Google Scholar]
  20. , , , , . Dis-covertebral (Andersson) lesions of the spine in ankylosing spondylitis revisited. Clin Rheumatol. 2009;28:883-92.
    [CrossRef] [PubMed] [Google Scholar]
  21. , , , et al. The Assessment of Spondyloarthritis International Society (ASAS) handbook: a guide to assess spondyloarthritis. Ann Rheum Dis. 2009;68(Suppl 2):1-44.
    [CrossRef] [PubMed] [Google Scholar]
  22. , , , , . Magnetic resonance tomography of sacroiliitis: anatomy, histological pathology, MR-morphology, and grading [in German] Radiologe. 2004;44:217-228.
    [CrossRef] [PubMed] [Google Scholar]
  23. , , , , , , . Unilateral Sacroiliitis: Differential Diagnosis Between Infectious Sacroiliitis and Spondyloarthritis Based on MRI Findings. AJR Am J Roentgenol. 2015;205(5):1048-55.
    [CrossRef] [PubMed] [Google Scholar]
  24. , , , , . The spine in Paget's disease. Skeletal Radiol. 2007;36(7):609-26.
    [CrossRef] [PubMed] [Google Scholar]
  25. . Graham TS. The ivory vertebra sign. Radiology. 2005;235:614-15.
    [CrossRef] [PubMed] [Google Scholar]
  26. , , , , . Diffusion-Weighted MRI “Claw Sign” Improves Differentiation of Infectious from Degenerative Modic Type 1 Signal Changes of the Spine. American Journal of Neuroradiology Aug. 2014;35(8):1647-1652.
    [CrossRef] [PubMed] [Google Scholar]
  27. , , , , , , . Utility of the claw sign in spine magnetic nuclear resonance with diffusion to differentiate Modic type I changes for degenerative disease versus infection. J Spine Surg. 2018;4(3):616-623.
    [CrossRef] [PubMed] [Google Scholar]
  28. , , , et al. Spinal disorders mimicking infection. Insights Imaging. ;12(2021):176.
    [CrossRef] [PubMed] [Google Scholar]
  29. , , , , , . Gas within the Intervertebral Disc Does Not Rule Out Spinal Infection-A Case Series of 135 Patients with Spontaneous Spondylodiscitis. Diagnostics. 2022;12(5):1089.
    [CrossRef] [PubMed] [Google Scholar]
  30. , , , . Differentiation of tuberculosis and metastatic cancer in the spine using dynamic contrast-enhanced MRI. Eur Spine J. 2015;24:1729-37.
    [CrossRef] [PubMed] [Google Scholar]
  31. , , . Osteoid osteoma masquerading tubercular arthritis or osteomyelitis on MRI: case series and review of literature. Indian J Radiol Imaging. 2015;25:261-8.
    [CrossRef] [PubMed] [Google Scholar]
  32. , , , , , , , , , , . Imaging algorithm and multimodality evaluation of spinal osteoblastoma. BMC Musculoskelet Disord. 2020;21(1):240.
    [CrossRef] [PubMed] [Google Scholar]
  33. , et al. "Computed Tomography of Axial Skeletal Osteoid Osteomas" American Journal of Roentgenology. 1984;142(4):769-772. 1 Apr
    [CrossRef] [PubMed] [Google Scholar]
  34. , , . Eosinophilic granuloma of the cervical spine. Spine (Phila PA 1976). 2002;27:1408-1413.
    [CrossRef] [PubMed] [Google Scholar]
  35. , . Benign pediatric bone tumor. Evaluation and treatment. Pediatr Clin North Am. 1996;43:949-966.
    [CrossRef] [PubMed] [Google Scholar]
  36. , , . Langerhans cell histiocytosis of the spine in children: Long-term follow-up. J Bone Joint Surg Am. 2004;86:1740-1750.
    [CrossRef] [PubMed] [Google Scholar]
  37. , , , . Primary Ewing's sarcoma of the second cervical vertebra: a rare entity. J Pediatr Orthop B. 2011;20(6):408-12.
    [CrossRef] [PubMed] [Google Scholar]
  38. , and . "Tissue Diagnosis in Infective Spondylodiscitis by CT-Guided Spinal Biopsy: A Standard Practice before Starting Treatment" Indian Journal of Radiology and Imaging. 2023;33(03):287-288.
    [CrossRef] [PubMed] [Google Scholar]
  39. , et al. "Clinico-Radio-Histopathological Correlation by C-Arm Image-Guided Biopsy in SpinalTuberculosis" Indian Journal of Orthopaedics. 2021;55(4):1028-1036.
    [CrossRef] [PubMed] [Google Scholar]
Show Sections