Indian Journal of Research in Homeopathy

: 2020  |  Volume : 12  |  Issue : 2  |  Page : 129--133

Infective Spondylodiscitis: A Comparative Analysis of Magnetic Resonance Imaging Findings with Etiology

Ibad I Sha1, Ajin Edwin1, Namitha Shah2, SR Roshna3,  
1 Department of Orthopedics, Government Medical College, Thiruvananthapuram, India
2 Department of Radiology, KIMS, Thiruvananthapuram, India
3 Department of Radiology, Government Medical College, Kozhikode, Kerala, India

Correspondence Address:
Dr. Ibad I Sha
Department of Orthopedics, Government of Medical College, Thiruvananthapuram, Kerala


Introduction: Spondylodiscitis in the Indian population is mostly tubercular and pyogenic. Gadolinium-enhanced magnetic resonance imaging (MRI) is currently most suited for radiological evaluation of the spine in patients with suspected vertebral osteomyelitis. Histology is highly sensitive in detecting the presence or absence of spondylodiscitis but fails in addressing the causative organism. The identification of organism is mainly through microbiological methods such as culture and tuberculosis (TB) polymerase chain reaction. The use of both pathologic examination and culture of the specimen along with imaging findings is paramount in establishing the diagnosis. Methodology: We performed prospective analysis of MR images obtained from patients with suspected spondylodiscitis who have undergone spinal biopsy for confirmation of diagnosis. T2- and T1-weighted images with and without contrast enhancement were subjected to assessment in coronal, axial, and sagittal planes. The sample obtained from biopsy was sent for microbiological and pathological analyses. Based on the final results, MRI findings and histopathological findings were correlated. Results: Out of 45 studied cases, 35 were confirmed to be infective spondylodiscitis, 5 were degenerative disc disease, 1 lymphoma, and 4 metastatic disease of the spine. Out of 35 confirmed cases, 17 (48.6%) were caused by TB and 18 (51.4%) by pyogenic organisms. The sensitivity and specificity of MRI for correctly diagnosing tuberculous spondylodiscitis were 70.6% and 77.8%, respectively, whereas the sensitivity and specificity of MRI for correctly diagnosing pyogenic spondylodiscitis were 44.4% and 85.2%, respectively. The statistically significant features specific for tuberculous spondylodiscitis for MRI were thin and smooth abscess wall, involvement of >2 vertebral bodies (52.9% in tuberculous vs. 44.4% in pyogenic), skip lesions (23.5% in tuberculous vs. 0% in pyogenic), and cold abscess (17.6% in tuberculous vs. 0% in pyogenic). Conclusion: It has been found that MRI is highly sensitive in detecting infective spondylodiscitis and has been widely accepted as the imaging modality of choice for the same. Although many imaging features have been identified to differentiate pyogenic and tuberculous etiologies, due to overlapping of these features, the sensitivity of MRI to conclusively diagnose the etiology needs further studies.

How to cite this article:
Sha II, Edwin A, Shah N, Roshna S R. Infective Spondylodiscitis: A Comparative Analysis of Magnetic Resonance Imaging Findings with Etiology.J Orthop Traumatol Rehabil 2020;12:129-133

How to cite this URL:
Sha II, Edwin A, Shah N, Roshna S R. Infective Spondylodiscitis: A Comparative Analysis of Magnetic Resonance Imaging Findings with Etiology. J Orthop Traumatol Rehabil [serial online] 2020 [cited 2021 Apr 15 ];12:129-133
Available from:

Full Text


Infective spondylodiscitis is an inflammatory process involving the disc and adjacent vertebrae of the intervertebral disc along with osteomyelitis of the adjacent vertebrae caused by an infectious pathogen.[1] With ever-expanding elderly and immunocompromised populations (e.g., from diabetes, immunosuppression therapy, acquired immunodeficiency syndrome, or drug abuse), together with greater availability and use of magnetic resonance imaging (MRI), the detection of spondylodiscitis is on the rise.[2],[3] Even though SD can occur at any age, it is more commonly seen in patients under 20 years and between 50 and 70 years, with higher predominance in males (1.5:1–2:1).[3] Spinal infections are caused mainly by three major agents: pyogenic infections caused by bacteria, granulomatous infections caused by tuberculosis (TB) or fungi, and parasitic infections which constitute a less common etiology.[4] The incidence of SD is more common in developing countries where TB is more prevalent than other pyogenic organisms, but over the past years, there has been a rise in incidence of pyogenic spondylodiscitis.[5]

Early diagnosis and organism-targeted antibiotics are essential to prevent disease progression.[6] The diagnosis of pyogenic SD is made on the basis of clinical features, imaging, histopathology, and microbiological investigations. The early clinical features of SD include back pain, spinal tenderness, and stiffness which are relatively nonspecific.[3],[5],[6] Other symptoms, such as fever, chills, night sweats, weight loss, and malaise, may also be present.[1],[3],[5] Neurologic compromise may occur in late stages. Although a raised white blood cell count, erythrocyte sedimentation rate, and C-reactive protein provide useful information, the sensitivity and specificity of these markers for SD are limited.[7]

Targeted microbiological investigations to isolate organisms include Gram stain and culture, acid-fast bacillus stain and culture, fungal culture, TB polymerase chain reaction, and GeneXpert.[1] Histology provides useful information regarding granulomatous change, but it lacks specific morphological criteria for detecting SD as well as causative agents.[1],[8] The tissue for histology as well as culture can be either through image-guided biopsy or specimen during surgery.[9],[10]

Gadolinium-enhanced MRI is the investigation of choice for early detection, treatment, and follow-up of suspected cases of spondylodiscitis.[11] Infections of the spine commonly demonstrate low T1 signal intensity, high T2 signal intensity, and enhancement within the affected marrow after gadolinium administration.[12],[13] Most of the time, MRI patterns can be used to differentiate infectious causes of marrow abnormalities from various neoplastic, inflammatory, degenerative, systemic, metabolic, and traumatic causes.[13],[14] However, the sensitivity of MRI to differentiate pyogenic from tuberculous etiology is not yet properly studied in the literature.

Our study is to correlate the MRI findings with histopathology and microbiology along with laboratory findings and the final clinical diagnosis so as to identify the sensitivity of MRI in differentiating TB from pyogenic etiology and also to identify the overall sensitivity of MRI in detecting spondylodiscitis. In this study, the sample for microbiological and histological findings is mainly obtained through biopsies taken in patients suspected of pyogenic and mycobacterial spondylodiscitis.


We retrospectively reviewed clinical, laboratory tests, imaging, histopathology, and microbiology results of 45 consecutive patients who underwent percutaneous (n = 38) or open (n = 7) spinal biopsy for suspected infective spondylodiscitis over a 3-year period. Computed tomography (CT)-guided biopsy was done routinely except in patients with suspected abscess who need operative stabilization or drainage where open biopsy was preferred.

All patients with clinical suspicion of infective spondylodiscitis who underwent MRI and tissue diagnosis were selected for the study. We excluded all patients who had claustrophobia or other contraindications for MRI, patients who did not undergo spinal biopsy following MRI, and patients with associated malignancies and postoperative spondylodiscitis. Those whose final diagnosis was not spondylodiscitis were excluded from final statistical evaluation except for MRI sensitivity and specificity.

Informed consent was taken from all patients. All patients who presented in outpatient or emergency departments with clinical features suggestive of infective spondylodiscitis as suspected by orthopedic surgeons were evaluated by X-ray, MRI, and image-guided/open biopsy. Image-guided biopsy was taken by the principal investigator under CT image guidance. The biopsy is also taken as a part of surgical procedure under direct visualization. The sample obtained from biopsy was sent for microbiological and pathological analyses. All data were collected and filled in the format of predefined pro forma by the principal investigator. Based on the final results, MRI findings and histopathological findings were correlated.

Magnetic resonance imaging protocol

The MRI protocol consisted of sagittal T1-weighted time of echo (TE 10/time to repeat [TR] 550), sagittal T2-weighted (TE 108/TR 3600), fat-suppressed T2-weighted inversion recovery (TE 84/TR 3800/inversion time 160 m), and transaxial T2-weighted (TE 99/TR 5550) spin-echo sequences. In addition, axial and sagittal fat-suppressed T1-weighted images (TE 30/TR 500) were obtained after intravenous administration of gadolinium. A 350-mm field of view and a 448 × 336 matrix with three acquisition averages were used. The slice thickness was 3 mm. The contrast was administered at 0.1 mmol/kg body weight guided by 5cc saline flush wherever possible.

Statistical analysis

Data collected were entered in spreadsheet and analyzed using SPSS version 21.0 (Armonk, NY: IBM Corp., 2012). All categorical variables were expressed as percentage (%) and all continuous variables as mean ± standard deviation (SD). The diagnostic accuracy of the test was found, and the sensitivity and specificity of MRI findings were obtained. MRI findings were correlated with biopsy/microbiological results.

The following statistical tests were used for analyzing the data:

Chi-square analysis/Fisher's exact testSensitivity, specificity, positive predictive value, negative predictive value, and accuracyConfidence interval was considered as 95%. P <0.05 was considered as statistically significant.


A total of 45 patients clinically suspected to have spondylodiscitis were enrolled for the study; MRI diagnosis of spondylodiscitis was made in 40 patients, and the remaining diagnoses were degenerative disc disease (n = 1), lymphoma of the spine (n = 1), and metastatic disease to the spine (n = 3). Of the 45 cases, there were 35 confirmed cases of spondylodiscitis, in which 17 (48.6%) were caused by TB and 18 (51.4%) by pyogenic organisms. Among the cases who were pyogenic spondylodiscitis, the organisms isolated were Staphylococcus aureus (10), Escherichia coli (2), Klebsiella pneumoniae (2), and Pseudomonas aeruginosa (2). Of the remaining 10 cases, 5 had degenerative disc disease, 1 had lymphoma of spine, and 4 had metastatic disease of the spine. Twenty-five out of 35 confirmed spondylodiscitis patients were above the age of 50 years which constituted 71.4% of the study population. Regarding the clinical presentation, back pain 86.7% was the most common complaint present, followed by fever 55.6%. Neurological deficit was present in 31.1% at the time of presentation.

In our study, the sensitivity, specificity, and accuracy of MRI in detecting spondylodiscitis were 100%, 50%, and 88.9%, respectively. Contrast-enhanced MRI in our study was able to identify tuberculous infection with a sensitivity of 70.6% and specificity of 77.8%. Pyogenic infection was identified with a sensitivity of 44.4% and specificity of 85.2%. The MRI findings were evaluated to differentiate tuberculous spondylitis and pyogenic spondylitis.

In our study, we noticed the following findings higher with tuberculous spondylodiscitis: a thin and smooth abscess wall (92.9% in tuberculous vs. 9.1% in pyogenic), paraspinal/intraspinal abscess (82.4% in tuberculous vs. 61.1% in pyogenic), heterogeneous vertebral body enhancement (52.9% in tuberculous vs. 44.4% in pyogenic), <50% degree of disc destruction (47.1% in tuberculous vs. 22.2% in pyogenic), involvement of vertebral body >2 (41.2% in tuberculous vs. 5.6% in pyogenic), skip lesion (23.5% in tuberculous vs. 0% in pyogenic), and cold abscess (17.6% in tuberculous vs. 0% in pyogenic) [Table 1]. Even though these features are noticed higher in tuberculous spondylodiscitis, all of these were not statistically significant. The statistically significant features specific for tuberculous spondylodiscitis for MRI were thin and smooth abscess wall, involvement of >2 vertebral bodies (52.9% in tuberculous vs. 44.4% in pyogenic), skip lesions (23.5% in tuberculous vs. 0% in pyogenic), and cold abscess (17.6% in tuberculous vs. 0% in pyogenic) [Figure 1].{Figure 1}{Table 1}

The MRI features that were more visualized in pyogenic spondylodiscitis were >50% degree of disc destruction (52.9% in tuberculous vs. 44.4% in pyogenic), involvement of vertebral body ≤2 (52.9% in tuberculous vs. 44.4% in pyogenic), homogeneous vertebral body enhancement (47.1% in tuberculous vs. 55.6% in pyogenic), and thick irregular abscess (7.1% in tuberculous vs. 90.9% in pyogenic) [Figure 2]. Among these, only statistically significant feature was thick irregular abscess (7.1% in tuberculous vs. 90.9% in pyogenic).{Figure 2}

In tuberculous spondylitis, thoracic or thoracolumbar involvement was observed in 40% of the cases and lumbar involvement in 50% of the cases, whereas in pyogenic spondylitis, thoracic involvement was observed in 10% of the cases and lumbar involvement was observed in 70% of the cases. The longest contiguous involvement of the spine in tuberculous spondylitis was associated with four vertebral bodies. Four patients had skip lesions, and all were in tuberculous spondylitis patients.


Infectious spondylitis accounts for 2%–4% of the cases of skeletal infection.[4] Among these, the incidence of TB of the spine is higher in endemic regions, but recent studies have documented that the incidence of pyogenic infections is on the rise.[2],[3],[15] This may be attributed to the proper imaging and microbiological and histologic correlation studies. Early detection and targeted antibiotic treatment is overemphasized because of the higher incidence neurological and skeletal complications that may ensue as disease progress.[14],[15],[16]

Clinical profile of spondylodiscitis includes symptoms such as back or neck pain, neck pain, weight loss, anorexia, and neurological deficit. Among these, the most common symptom is back pain which is seen in 90% of the cases.[9] The primary complaint in our study population was backpain with only 55% of them having associated fever. Since the clinical profile of tuberculous and pyogenic spondylodiscitis is similar, it is unable to differentiate them based on that, especially in the early stages. In the late stages of presentation, it has been documented that tuberculous spondylodiscitis is more associated with spine deformities compared to pyogenic reflecting the destructive nature of caseating granulomata.[14]

MRI has evolved as the gold standard for imaging of spinal infections helping in the early detection of infection when other imaging modalities are usually normal (radiography) or nonspecific (nuclear medicine).[17],[18],[19] Even radiographically also differentiation between tuberculous and pyogenic spondylitis is difficult mainly because of the overlap of features.[13],[14] There have been few reports in the past which elaborated features to differentiate pyogenic from tuberculous pathology, but there have not been enough studies on the validity of these features.

Common MRI findings in infectious spondylitis are hypointensity of the involved tissue on T1-weighted images, hyperintensity on T2-weighted images, destruction of one or more adjacent vertebral bodies with involvement of the intervening disc, epidural and paraspinal extension, and/or abscesses.[14],[19],[20] MRI tends to overestimate the amount of tissue involvement by the infection as some of the signal changes are reactive.[21] Previous studies have reported characteristic MRI findings in TB spondylodiscitis, which include loss of cortical definition, intraosseous abscesses with rim enhancement, large paraspinal abscesses with thick rim enhancement, well-defined paraspinal area of abnormal signal intensity, subligamentous spread of inflammatory tissue, and relative disc preservation.[14],[19],[20]

In our study rim enhancement of abscess on contrast enhanced MRI was seen in both pyogenic and tuberculous abcessess as opposed to few previous studies which considered it be specific for tuberculous etiology.[13] The most reliable MRI findings suggesting tuberculous spondylitis in our study were thin and smooth enhancement of the abscess wall compared to thick and irregular enhancement of abscess wall, which was more suggestive of pyogenic spondylitis. Hence, these findings noticed in contrast-enhanced MRI reiterate the use of contrast study needed for differentiation.

Chang et al. observed a greater extent of vertebral body destruction in tuberculous infection, heterogeneous signals from vertebral bodies following contrast administration in TB, and homogeneous contrast enhancement in purulent inflammation.[13] Jain et al. observed that contrast enhancement in 75% of tuberculous spondylodiscitis was focal and heterogeneous, whereas in 70% of the pyogenic group, the pattern was diffuse and homogeneous.[22] Compared to other studies, our study observed that vertebral enhancement pattern was focal and heterogeneous in 52.9% of the tuberculous group compared to 44.4% of the pyogenic group, whereas the homogeneous pattern of enhancement in the pyogenic group was 55.6% compared to 47.1% in TB (P = 0.615).

Another clinical feature more suggestive of tuberculous spondylodiscitis is the subligamentous spread and involvement of three or more vertebral levels.[13],[14],[21] This is explained as the lack of proteolytic enzymes in the mycobacterium as compared with pyogenic infection of caseations they barely penetrate the anterior longitudinal ligament causing subligamentous spread. In our study among tuberculous spondylodiscitis, 41.2% of the cases had involvement of three or more vertebral bodies (P = 0.012) and were consistent with previous reports. In tuberculous spondylodiscitis, it found that the involvement of the thoracic spine was higher compared to lumbar than in patients with pyogenic spondylitis where more of the lumbar spine was involved, but these were not statistically significant.

The presence of skip lesions has been reported as suggestive of tuberculous spondylitis.[9] Four cases with skip lesions were observed in our study. In a similar study by Chang et al., there was only one case of skip lesion out of twenty cases they studied.[13] We also had three cases of cold abscess present in our study. Both the presence of skip lesions and cold abscess were statistically significant and specific to tuberculous spondylodiscitis. Recent studies reported that the involvement of posterior elements in tuberculous spondylitis is less common in tuberculous spondylodiscitis as well as pyogenic etiology.[23] In our study, abnormal signal of the posterior element was observed in 30% of the patients with tuberculous spondylitis and 10% with pyogenic spondylitis, but this difference was not statistically significant.

There are few limitations associated with our study. Our study showed an overall specificity of 50% for MRI which can be attributed to a lesser number of true negative cases. In the future, this can be overcome by increasing the sample size. Most of our patients were in either subacute or chronic stage of presentation; hence, imaging features in acute cases could not be efficiently delineated. The interobserver variability and accuracy associated with MR images were not evaluated as the images were interpreted by a single musculoskeletal radiologist.


It has been found that MRI is highly sensitive in detecting infective spondylodiscitis and has been widely accepted as the imaging modality of choice for the same. Although many imaging features have been identified to differentiate pyogenic and tuberculous etiologies, due to overlapping of these features, the sensitivity of MRI to conclusively diagnose the etiology needs further studies.


Consent has been taken from the patient and can be provided if requested.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Sobottke R, Seifert H, Fätkenheuer G, Schmidt M, Gossmann A, Eysel P. Current diagnosis and treatment of spondylodiscitis. Dtsch Arztebl Int 2008;105:181-7.
2Gary A. Infectious spondylodiscitis: Diagnosis and treatment. Mo Med 2013;110:80-4.
3Tsiodras S, Falagas ME. Clinical assessment and medical treatment of spine infections. Clin Orthop Relat Res 2006;444:38-50.
4Turunc T, Demiroglu YZ, Uncu H, Colakoglu S, Arslan H. A comparative analysis of tuberculous, brucellar and pyogenic spontaneous spondylodiscitis patients. J Infect 2007;55:158-63.
5Mylona E, Samarkos M, Kakalou E, Fanourgiakis P, Skoutelis A. Pyogenic vertebral osteomyelitis: A systematic review of clinical characteristics. Semin Arthritis Rheum 2009;39:10-7.
6Fantoni M, Trecarichi EM, Rossi B, Mazzotta V, Di Giacomo G, Nasto LA, et al. Epidemiological and clinical features of pyogenic spondylodiscitis. Eur Rev Med Pharmacol Sci 2012;16 Suppl 2:2-7.
7Tyrrell PN, Cassar-Pullicino VN, McCall IW. Spinal infection. Eur Radiol 1999;9:1066-77.
8Turgut M. Spinal tuberculosis (Pott's disease): Its clinical presentation, surgical management, and outcome: A survey study on 694 patients. Neurosurg Rev 2001;24:8-13.
9Le HB, Lee ST, Munk PL. Image-guided musculoskeletal biopsies. Semin Intervent Radiol 2010;27:191-8.
10Peh W. CT-guided percutaneous biopsy of spinal lesions. Biomed Imaging Interv J 2006;2:e25.
11Ledermann HP, Schweitzer ME, Morrison WB, Carrino JA. MR imaging findings in spinal infections: Rules or myths? Radiology 2003;228:506-14.
12Yasar K, Pehlivanoglu F, Cicek G, Sengoz G. The evaluation of the clinical, laboratory and the radiological findings of the fifty-five cases diagnosed with tuberculous, brucellar and pyogenic spondylodiscitis. J Neurosci Rural Pract 2012;3:17-20.
13Chang MC, Wu HT, Lee CH, Liu CL, Chen TH. Tuberculous spondylitis and pyogenic spondylitis: Comparative magnetic resonance imaging features. Spine (Phila Pa 1976) 2006;31:782-8.
14Lee KY. Comparison of pyogenic spondylitis and tuberculous spondylitis. Asian Spine J 2014;8:216-23.
15Sur A, Tsang K, Brown M, Tzerakis N. Management of adult spontaneous spondylodiscitis and its rising incidence. Ann R Coll Surg Engl 2015;97:451-5.
16Roblot F, Besnier JM, Juhel L, Vidal C, Ragot S, Bastides F, et al. Optimal duration of antibiotic therapy in vertebral osteomyelitis. Semin Arthritis Rheum 2007;36:269-77.
17Cheng VC, Yam WC, Hung IF, Woo PC, Lau SK, Tang BS, et al. Clinical evaluation of the polymerase chain reaction for the rapid diagnosis of tuberculosis. J Clin Pathol 2004;57:281-5.
18Sharif HS. Role of MR imaging in the management of spinal infections. AJR Am J Roentgenol 1992;158:1333-45.
19Dagirmanjian A, Schils J, McHenry M, Modic MT. MR imaging of vertebral osteomyelitis revisited. AJR Am J Roentgenol 1996;167:1539-43.
20Smith AS, Weinstein MA, Mizushima A, Coughlin B, Hayden SP, Lakin MM, et al. MR imaging characteristics of tuberculous spondylitis vs. vertebral osteomyelitis. AJR Am J Roentgenol 1989;153:399-405.
21Diehn FE. Imaging of spine infection. Radiol Clin North Am 2012;50:777-98.
22Galhotra RD, Jain T, Sandhu P, Galhotra V. Utility of magnetic resonance imaging in the differential diagnosis of tubercular and pyogenic spondylodiscitis. J Nat Sci Biol Med 2015;6:388-93.
23Frel M, Białecki J, Wieczorek J, Paluch Ł, Dąbrowska-Thing A, Walecki J. Magnetic resonance imaging in differentatial diagnosis of pyogenic spondylodiscitis and tuberculous spondylodiscitis. Pol J Radiol 2017;82:71-87.