Abstract
Objective: To report a series of Granulomatous Spondylodiscitis (GS) with focus on the histopathological features of the different forms of GS.
Design: Case series.
Setting: Pathology department of Charles Nicolle’s Hospital of Tunisia
Participants: This study included 57 patients diagnosed with GS. There were 44 (77.2%) female patients and 13 (22.8%) male patients (sex ratio = 0.28).
Intervention: Not applicable.
Outcome measures: Clinical, microbiological and histopathological features were assessed in this study.
Results: Fifty-seven patients with GS were enrolled: 51 tuberculous spondylodiscitis (TS), 2 fungal spondylodiscitis (FS), 3 brucellar spondylodiscitis (BS) and 1 case of sarcoidosis. Granulomas with necrosis were seen in 38 (66.6%) cases: 36 TS and 2 FS, while granulomas without necrosis were observed in the remaining 19 cases: 15 TS, 3 BS and 1 sarcoidosis. In all cases of TS, granulomas were epithelioid type, associated with histiocytic type granulomas in 7 cases. Caseous necrosis was seen in 35 cases of TS and suppurative granuloma in one case. The 3 cases of BS exhibited non-necrotizing and histiocytic type granulomas. The 2 cases of FS showed histiocytic, epithelioid and necrotizing granulomas. Necrosis was mixed: suppurative and caseous in both cases of FS. Sarcoidosis was characterized with epithelioid type granulomas without necrosis.
Conclusion: Granuloma with caseous necrosis is highly suggestive of TS but does not rule out FS. Certain fungi can exhibit this type of necrosis as do tuberculosis species. Suppurative inflammation, although rare in TS, does exist. Histiocytic type granuloma without necrosis is suggestive of brucellosis.
Keywords: Granuloma, Spondylodiscitis, Necrosis, Histopathology, Tuberculosis
Introduction
Histopathology is a widely used method to distinguish granulomatous spondylodiscitis (GS) from pyogenic spondylodiscitis.1 GS is classified according to causative microorganisms. Tuberculous spondylodiscitis (TS), brucellar spondylodiscitis (BS) and fungal spondylodiscitis (FS) are the main forms of GS.2,3 Sarcoidosis is a granulomatous disease that rarely affects the spine resulting in GS.4
The frequency of different GS forms varies from one country to another throughout the world. Tuberculosis represents the most common cause of GS in high-income countries. Whereas, BS is endemic in several Mediterranean regions.5 However, the frequency of BS granulomatous forms is not well known. Fungal spine infections are uncommon; but their incidence is increasing in recent year.6
Clinical presentation of GS is usually nonspecific resulting in diagnostic and treatment delays that may lead to severe neurological complications and spinal deformities. Distinguishing between various forms of GS is mandatory to initiate appropriate treatment.7 The etiological diagnosis of GS requires a multidisciplinary approach that relies on clinical, laboratory and radiological features.8 The histopathological diagnosis of GS is challenging due to the lack of specific histological markers. The aim of this study was to report a series of GS with focus on the histopathological features of its forms.
Methods
Participants and setting
This study included 57 GS among 122 percutaneous biopsies collected in our pathology department from January 2014 to May 2018 and obtained from 121 patients suspected to have infective spondylodiscitis. CT-guided percutaneous biopsies were performed in a center specialized in bone diseases under local anesthesia. All spine biopsies were carried out using an 11 Gauge co-axial core-needle system. Two to seven specimens from the intervertebral disc and the adjacent vertebral endplates were obtained from each patient. The medical records of patients were reviewed.
The etiological diagnosis of TS was based on positive bacteriology. When microbiological investigations were negative, the diagnosis of TS was based on the combination of clinical features, imaging findings, histopathology and response to antituberculous treatment. The diagnosis of BS was based on positive culture and/or positive serology. FS was diagnosed based on positive culture.
Microbiological procedure
Core samples were washed with sterile saline solution. The wash solution was collected in sterile dry tubes and processed at the Microbiology department. Direct examination and mycobacterial culture on both solid (Lowenstein Jensen medium) and liquid medium (Mycobacterial Growth Indicator Tube) were performed to identify acid fast bacilli. Positive cultures were analyzed at the National Reference Laboratory for Mycobacteria of Ariana for mycobacterial identification and drug susceptibility testing. Identification of mycobacterial species was performed using conventional biochemical tests and was based on detection of MPT64 antigen immunochromatography test (SD Bioline). Drug susceptibility testing was performed according to the proportion method. Aerobic and anaerobic bacterial culture as well as fungal culture were performed. Polymerase chain reaction was not performed.
Pathological procedure
For histopathological evaluation, the biopsy samples were fixed using formaldehyde solution. Ethylenediaminetetraacetic acid was used to decalcify biopsy specimens. After decalcification, tissues were embedded in paraffin and sectioned (4 µm thickness). The sections were stained with hematoxylin and eosin. Ziehl-Neelsen (ZN) stain was performed in 40 cases on formalin-fixed paraffin-embedded tissues with an automated slide staining machine (Leica). For each case, a slide representative of the lesions was selected for the ZN staining. An external positive control of the technique (ZN positive pulmonary tuberculosis) was tested simultaneously. All slides were retrospectively reviewed by one pathologist. Board certified pathologist assessed the following pathological features: granuloma, necrosis and disc inflammation. Granulomas were classified as follows: epithelioid granuloma (well-formed granuloma composed of aggregates of epithelioid cells with or without giant cells) and histiocytic granuloma (less well-formed granuloma composed of aggregates of histiocytes without epithelioid appearance). The number of granulomas was evaluated for each case on all cores as follows: 1–3 granulomas was considered low (+), from 4 to 6 granulomas was considered medium (++) and≥7 granulomas was considered high (+++). Necrosis was classified as suppurative (liquefactive with numerous neutrophils) or caseous. Two types of caseous necrosis were distinguished: typical caseous necrosis with a granular and homogeneous eosinophilic appearance and atypical one with basophilic appearance containing some neutrophils and nuclear debris. The extent of necrosis was considered as “limited” (+) when necrosis affected one medullary space, “moderate” (++) when it occupied 2 medullary spaces and “extensive” (+++) when more than 2 medullary spaces were affected.
Results
Based on histopathological features, the 122 percutaneous biopsies were classified as GS in 57 (46.7%) cases and non-GS in the 65 (53.3%) others. The 57 GS corresponded to: 51 TS (89.5%), 2 FS (3.5%), 3 BS (5.3%) and 1 (1.7%) sarcoidosis. The 65 non-GS corresponded to 59 (90.8%) pyogenic spondylodisctis, 1(1.5%) BS, and 5 (7.7%) TS. These 5 TS exhibited necrosis without granuloma.
Microbiological findings
Among the 51 TS, mycobacterial culture was positive in 18 (35.3%) cases. Direct examination performed on core samples wash was negative in all cases. Fifteen isolates were identified as M. tuberculosis while the remaining 3 cases were identified as M. bovis. No resistance to rifampicin or ethambutol was identified. Resistance to isoniazid was tested in 17 cases and was positive in 1 case. Resistance to pyrazinamide was tested in 5 cases of M. tuberculosis and in 2 cases of M. bovis and was positive in 1 isolate and 2 isolates respectively. In the remaining 33 (64.7%) cases, mycobacterial culture was negative and the diagnosis of TS was based on the combination of clinical features, imaging findings and histopathology.
BS was diagnosed in 3 cases. Brucella spp was isolated in 2 cases. The diagnosis of FS was made in 2 cases. Fungal culture was positive to Aspergillus flavus (1 case) and Candida tropicalis (1 case).
Clinical findings
In the GS subgroup (57 cases), there were 44 (77.2%) females and 13 (22.8%) males (sex ratio = 0.28). The age of patients ranged from 19 to 84 years. Mean and median ages were 47 and 52 years respectively. The location of infected spine was available in 46 (80.7%) cases. Among the 41 TS with available location, the thoracic spine was affected in 20 (48.8%) cases, the lumbar spine in 14 (34.1%) cases and the cervical spine in 1 (2.4%) case. Multiple spine levels were affected in 6 (14.6%) cases. Thoracic spine was involved in the 2 FS.
For BS, thoracic spine and lumbar spine were affected in one case and 2 cases respectively. Clinical data of GS were available in 39 (68.4%) cases. Chronic back pain was the most common symptom, seen in 37 (94.9%) cases. Fever was noted in 21 (53.8%) patients, and night sweat in 14 (35.9%) patients. Neurologic deficit was noted in 11 cases. Clinical features are summarized in Table 1. C-reactive protein level was elevated (> 6 mg/L) in 28 patients among the in 34 patients with GS tested. Thirty-one patients with TS received purified protein derivative skin test, which was positive (≥5 mm) in 17 (54.8%) cases.
Table 1. Clinical features of granulomatous spondylodiscitis patients.
| Clinical features | TS* patients (n = 34) | BS**patients (n = 3) | FS*** patients (n = 2) |
|---|---|---|---|
| Clinical symptoms Pain Fever (>38°C) Night sweats Anorexia Asthenia Weight loss Spinal stiffness Paraparesis Motor deficit Lameness Paresthesia |
32 (94.1%) 17 (50%) 13 (38.2%) 10 (29.4%) 12 (35.3%) 16 (47%) 9 (26.5%) 4 (11.8 %) 3 (8.8%) 1 (2.9 %) 3 (8.8%) |
3 3 1 1 1 1 0 0 0 0 0 |
2 1 0 0 0 0 0 0 0 0 0 |
| Predisposing factors and comorbidities Diabetes Chronic kidney failure Urinary tract infection Sickle cell anemia Previous tuberculosis Actual tuberculosis Chemotherapy Degenerative disc disease History of brucellosis |
4 (11.8%) 1 (2.9%) 2 (5.9%) 1 (2.9%) 4 (11.8%) 2 (5.9%) 1 (2.9%) 1 (2.9%) 0 |
0 0 0 0 0 0 0 0 1 |
0 0 0 1 0 0 0 0 0 |
*TS, Tuberculous Spondylodiscitis; **BS, Brucellar Spondylodiscitis; ***FS, Fungal Spondylodiscitis.
Histopathological findings
A total of 322 core biopsies were taken from the 57 GS, with a mean of 4 core biopsies per patient. The histopathological diagnosis of GS was obtained in 168 (52.3%) core biopsies. Granulomas with necrosis were seen in 38 (66.7%) GS: 36 TS and 2 FS. Granulomas without necrosis were observed in the remaining 19(33.3%) cases: 15 TS, 3 BS, and 1 sarcoidosis. Ziehl-Neelsen staining was negative in the 40 tested specimens. In all cases of TS, granulomas were epithelioid type, associated with histiocytic type granulomas in 7 cases. The number of granulomas in TS was high (≥ 7) in 18(35.3%) cases and low (≤ 3) in 24 (47%) others. Necrotizing granulomas were observed in 36 (65.5%) TS cases (Figure 1). Necrosis was extensive in 19 cases. Typical caseous necrosis was seen in 31 cases while, atypical caseous necrosis with basophilic appearance containing some neutrophils and nuclear debris was observed in 4 cases. Suppurative necrosis was seen in one case (1.9%).
Figure 1.
Tuberculous spondylodiscitis: (A) epithelioid and histiocytic granuloma types with caseous necrosis (HE ×100); (B) Epithelioid granuloma with giant cell (HE ×400).
When the intervertebral disc was visualized (25 cases), inflammatory infiltrate of the disc was identified in 5 cases. In these cases, the disc tissue contained non-necrotizing granulomas (4 cases) and neutrophilic infiltration (1 case).
All 3 cases of BS exhibited non-necrotizing and histiocytic type granulomas. The number of granulomas was low in all 3 cases.
The 2 cases of FS showed histiocytic, epithelioid and necrotizing granulomas. Granulomas were numerous (≥7) in the 2 cases. Necrosis was mixed suppurative and caseous and extensive in these 2 cases.
The case of sarcoidosis showed epithelioid type granulomas without necrosis. The histopathological features of GS are summarized in Table 2.
Table 2. Histopathological features of tuberculous, brucellar and fungal spondylodiscitis.
| Pathological features | TS* (n = 51) | BS** (n = 3) | FS*** (n = 2) | |
|---|---|---|---|---|
| Type of granulomas | Epithelioid |
44(86.3%) | 0 | |
| Histiocytic |
0 | 3/3 | 0 | |
| Epithelioid + histiocytic | 7(13.7%) | 0 |
2/2 | |
| Number of granuloma | + | 24(47%) | 3/3 | 0 |
| ++ | 9(17.6%) | 0 | 0 | |
| +++ | 18(35.3%) | 0 | 2/2 | |
| Type of necrosis | – Caseous – Atypical caseous – Suppurative – Mixed suppurative and caseous |
31(60.8%) 4(7.8%) 1 (1.9%) 0 |
0 0 0 0 |
0 0 0 2/2 |
| Extension of necrosis | + | 14 | 0 | 0 |
| ++ | 3 | 0 | 0 | |
| +++ | 19 | 0 | 2/2 | |
*TS, Tuberculous Spondylodiscitis; **BS, Brucellar Spondylodiscitis; ***FS, Fungal Spondylodiscitis.
Discussion
The present study detailed the histopathological features of different forms of GS. To the best of our knowledge, these features were not previously detailed in the literature. The main limitation of this study is the small number of non-tuberculous forms due to their rarity. These forms, especially BS, are exceptionally biopsied because their diagnosis is based on serological data and blood culture.
In the present series, about 90% of GS were caused by tuberculosis. Ten per cent of GS were caused by brucellosis, mycosis and sarcoidosis. Necrotizing granulomas are highly suggestive of TS.9–10 This type of granuloma was observed in about two-thirds of the cases. Caseous necrosis is a characteristic but not pathognomonic feature of TS. Some fungi can exhibit this type of necrosis. In the present study, typical caseous necrosis associated with suppurative necrosis was observed in the two cases of FS. Suppurative necrosis was observed in about 2% of TS. Pure suppurative forms of TS without epithelioid granuloma are histologically misdiagnosed. Only the isolation of the causative organism by bacteriological examination allows the diagnosis.
Due to the paucibacillary character of the tuberculosis in osseous location,11 direct examination and ZN staining on formalin-fixed paraffin-embedded tissue are frequently negative. In the present study, both tests were negative in all tested cases. In the literature, positive culture rates of TS vary from 50% to 70%.12 In the present study this rate was 35.3%. This relatively low rate can be explained by the fact that culture was done on core wash and not on core specimen. The diagnosis of TS with negative bacteriology is based on multidisciplinary approach and relies essentially on histopathology.
Polymerase chain reaction (PCR) is a reliable technique for a rapid diagnosis of TS. However, the access to this molecular method is limited in endemic areas with limited resources. The sensitivity and specificity of PCR in fresh tissue, used in the diagnosis of TS, vary from 61%13 to 91%14 and from 63.7%14 to 93.7% respectively.13 Fresh tissues are sometimes unavailable. PCR in formalin fixed paraffin embedded (FFPE) tissues represents an alternative approach for diagnosing TS thus avoiding biopsy repetition. PCR in FFPE tissues in the diagnosis of TS was not widely reported in the literature. Berk et al.15 reported in a small series (19 cases) a good effectiveness of PCR in FFPE tissues with sensitivity of 94.7%, and specificity of 83.3%.
BS is typically characterized by small granulomas composed of aggregates of histiocytes without epithelioid appearance (histiocytic type granulomas).16 Necrosis is typically absent.9 This is consistent with the findings of the 3 cases of granulomatous brucellosis of the present series.
Differentiating the different forms of GS is mandatory and requires isolation of the causative pathogen and/or detection of specific antibody by serological tests (Wright, Rose Bengal, immunofluorescence).7 PCR is a reliable tool allowing a rapid diagnosis and earlier treatment. Various targets have been employed by PCR such as the sequence IS6110 for the diagnosis of TS13 and, the 16S ribosomal RNA for the diagnosis of BS.2 Unfortunately, the access to molecular methods is limited in endemic areas. Histopathology establish the diagnosis of GS and allows a presumptive diagnosis of the causative microorganism. Special stains on formalin-fixed paraffin-embedded tissue such as Ziehl–Neelsen for Mycobacteria and periodic acid–Schiff and Grocott for fungi help to establish the causative agent.
Conclusion
The histopathological diagnosis of GS evokes firstly TS but also BS and FS. The distinction of different forms of GS is challenging if based on histopathological criteria only. The etiological diagnosis of GS requires a multidisciplinary approach. Granuloma with caseous necrosis is highly suggestive of TS but must rule out FS. In these cases if FS is microbiologically eliminated, the diagnosis of TS can be retained even if positive cultures of mycobacterium have not been obtained and molecular analyzes are not available. Non-necrotizing histiocytic granuloma is highly suggestive of BS. But the definite diagnosis of BS must be based on culture and/or demonstration of specific antibody by serology. We propose the diagnostic diagram below of GS (Figure 2).
Figure 2.
Flow chart depicting the diagnosis of granulomatous spondylodiscitis (GS). GS, Granulomatous Spondylodiscitis; TS, Tuberculous Spondylodiscitis; FS, Fungal Spondylodiscitis; BS, Brucellar Spondylodiscitis; ESV, Erythrocyte Sedimentation Velocity; CPR, C-Reative Protein; MRI, Magnetic Resonance Imaging; CT, computed tomography.
Acknowledgement
The authors would like to thank Doctor Noureddine Litaiem for the revision of the English language of the paper.
Disclaimer statements
Contributors None
Funding This research received no specific grant from any funding agency, commercial or not-for profit sectors.
Conflicts of interest The authors report no conflicts of interest.
References
- 1.Duarte RM, Vaccaro AR.. Spinal infection: state of the art and management algorithm. Eur Spine J. 2013;22(12):2787–99. doi: 10.1007/s00586-013-2850-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gouliouris T, Aliyu SH, Brown NM.. Spondylodiscitis: update on diagnosis and management. J Antimicrob Chemother. 2010;65(supp 3):S11–S24. [DOI] [PubMed] [Google Scholar]
- 3.Sharif H. Granulomatous spinal infections: MR imaging. Radiology. 1990;177:101–7. doi: 10.1148/radiology.177.1.2399306 [DOI] [PubMed] [Google Scholar]
- 4.Garg RK, Somvanshi DS.. Spinal tuberculosis: a review. J Spinal Cord Med. 2011;34:440–54. doi: 10.1179/2045772311Y.0000000023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Erdem H, Elaldi N, Batirel A.. Comparison of brucellar and tuberculous spondylodiscitis patients: results of the multicenter ‘backbone-1 study’. Spine J. 2015;15(12):2509–17. doi: 10.1016/j.spinee.2015.09.024 [DOI] [PubMed] [Google Scholar]
- 6.Caldera G, Cahueque M, Cobar A, Gómez G, Rodríguez O.. Fungal spondylodiscitis: review. J Spine. 2016;5:302. doi: 10.4172/2165-7939.1000302 [DOI] [Google Scholar]
- 7.Sans N, Faruch M, Lapègue F, Ponsot A, Chiavassa H, Railhac JJ.. Infections of the spinal column-spondylodiscitis. Diagn Interv Imaging. 2012;93(6):520–9. doi: 10.1016/j.diii.2012.04.003 [DOI] [PubMed] [Google Scholar]
- 8.Cottle L, Riordan T.. Infectious spondylodiscitis. J Infect. 2008;56:401–12. doi: 10.1016/j.jinf.2008.02.005 [DOI] [PubMed] [Google Scholar]
- 9.Turunc T, Demiroglu YZ, Uncu H, Colakoglu S, Arslan H.. A comparative analysis of tuberculous, brucellar and pyogenic spontaneous spondylodiscitis patients. J Infect. 2007;55(2):158–63. doi: 10.1016/j.jinf.2007.04.002 [DOI] [PubMed] [Google Scholar]
- 10.Xinming Y, Xianyong M, Wei S, Yakun D, Lei Z, Yaoyi W.. The comparison of the manifestation of the clinical imageology and pathology between the brucellar spondylitis and the spine turberculosis. Surgical Science. 2014;5:60–9. doi: 10.4236/ss.2014.52014 [DOI] [Google Scholar]
- 11.Trecarichi EM, Meco EDI, Mazzotta V, Fantoni M.. Tuberculous spondylodiscitis: epidemiology, clinical features, treatment, and outcome. Eur Rev Med Pharmacol Sci. 2012;16:58–72. [PubMed] [Google Scholar]
- 12.Kim CJ, Kim EJ, Song KH, Choe PG, Park WB, Bang JH, et al. Comparison of characteristics of culture-negative pyogenic spondylitis and tuberculous spondylitis: a retrospective study. BMC Infect Dis. 2016;16:560. doi: 10.1186/s12879-016-1897-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Pandey V, Chawla K, Acharya K, Rao S, Rao S.. The role of polymerase chain reaction in the management of osteoarticular tuberculosis. Int Orthop. 2009;33:801–5. doi: 10.1007/s00264-007-0485-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Arockiaraj J, Michael JS, Amritanand R, David KS, Krishnan V.. The role of Xpert MTB/RIF assay in the diagnosis of tubercular spondylodiscitis. Eur Spine J. 2017;26:3162–9. doi: 10.1007/s00586-017-5076-9 [DOI] [PubMed] [Google Scholar]
- 15.Berk RH, Yazici M, Atabey N, Ozdamar OS, Pabuccuoglu U, Alici E.. Detection of Mycobacterium tuberculosis in formaldehyde solution-fixed, paraffin-embedded tissue by polymerase chain reaction in Pott’s disease. Spine. 1996;21:1991–5. doi: 10.1097/00007632-199609010-00011 [DOI] [PubMed] [Google Scholar]
- 16.Nas K, Gür A, Kemaloğlu MS, Geyik MF, Cevik R, Büke Y, et al. Management of spinal brucellosis and outcome of rehabilitation. Spinal Cord. 2001;39(4):223–7. doi: 10.1038/sj.sc.3101145 [DOI] [PubMed] [Google Scholar]