Abstract
Objective. Previous studies have suggested that Mycobacterium tuberculosis (MT) could be involved in the pathogenesis of Takayasu’s arteritis (TA). The search for MT in arterial lesions of TA has never been assessed directly by sensitive methods. Our aim was to assess the presence of MT in arterial samples obtained in patients with TA.
Methods. Fresh arterial samples were collected from 10 consecutive patients (9 women and 1 man, median age 42 yrs, range 19–67 yrs) with a diagnosis of TA according to the American College of Rheumatology criteria who underwent vascular surgical procedures for their disease. Three patients had recent onset of TA and 7 had longstanding disease. No patient had evidence of active tuberculosis. Arterial biopsies were collected during vascular surgical procedures, and were systematically studied by a pathologist specializing in vascular diseases. Presence of MT was assessed in the biopsies by acid-fast and auramine-fluorochrome stainings, mycobacterial cultures, and direct amplification test (DAT) for MT.
Results. Histological examination showed active (n = 5) and inactive (n = 5) arterial lesions. MT was not detected in arterial lesions of either active or inactive TA, by acid-fast and auramine-fluorochrome staining, mycobacterial cultures, or DAT. No DAT inhibitors were found.
Conclusion. Our study does not support a direct role of MT in the pathogenesis of arterial lesions in either recent or longstanding TA, but does not exclude the possibility of a cross-reaction between mycobacterial and arterial antigens.
Takayasu’s arteritis (TA) is a granulomatous arteritis predominantly affecting the elastic arteries such as the aorta, its major division branches, and the pulmonary arteries. The etiology of TA is still unclear more than 100 years after its first description by Takayasu1. However, most available data suggest that both cellular and humoral immune responses are involved in pathogenesis2.
A link between TA and tuberculosis has been suggested for a long time, as TA is more common in individuals originating from Asia, Africa, and South America, where the incidence of tuberculosis is high, and because granulomatous lesions may be observed in both diseases3,4. Interestingly, Serratrice et al5 and Duzova, et al6 have reported cases of patients with simultaneous occurrence of TA and tuberculosis, further supporting a possible relationship between the 2 diseases.
Several laboratory findings support the idea of an antigen-specific immune reaction against M. tuberculosis in TA. Both humoral and cell-mediated immune responses directed towards M. tuberculosis antigens have been reported in patients with TA7–11. Aggarwal, et al7 and Moraes, et al8 have reported the existence of an increased humoral immune response directed towards the mycobacterial 65-kDa heat-shock protein (mHSP65) in patients with TA. Kumar Chauhan, et al10 have demonstrated a significant correlation between T cells reactive against mHSP65 and its human homolog (hHSP60), as well as between anti-mHSP65 and anti-hHSP60 IgG antibodies, suggesting that M. tuberculosis may have a role in the immunopathogenesis of TA through a molecular mimicry mechanism.
Surprisingly, the presence of M. tuberculosis in arterial lesions of patients with TA has not been assessed directly using sensitive microbial assays12. The aim of our study was to assess the presence of M. tuberculosis in arterial lesions of patients with TA by 3 different methods: acid-fast and auramine-fluorochrome stainings, mycobacterial cultures, and nucleic acid amplification, which is a sensitive method that may detect nonviable bacteria13. Presence of M. tuberculosis in arterial lesions of patients with TA would provide a clue to direct involvement of this infectious agent in the pathogenesis of TA.
MATERIALS AND METHODS
Patients
Our case series is based on the prospective analysis of arterial biopsies obtained in 10 consecutive patients with TA followed in the internal medicine department of our tertiary care center (Pitié-Salpêtrière Hospital, Paris, France). The diagnosis of TA was made according to the American College of Rheumatology (ACR) criteria14. Presence of a temporal artery abnormality, new-onset headache, visual loss, jaw claudication, or polymyalgia rheumatica was carefully excluded in the oldest patients, who had negative temporal artery biopsies and did not fulfill ACR criteria for giant cell arteritis15. These 10 consecutive patients were operated on in the vascular surgery department of our center, which specializes in surgical care of patients with TA16. All these patients underwent a vascular surgical procedure for either a clinically and/or hemodynamically significant arterial stenosis, or for an aortic aneurysm or stenosis. Fresh arterial biopsies were obtained during surgery, and samples of these biopsies were systematically analyzed by a pathologist with experience in vascular pathology (IB) and for presence of M. tuberculosis as a routine diagnosis procedure performed in our center. Local ethics commitee approval and patient informed consent were obtained before the study began.
Microbiological processing
Preparation of arterial samples was performed using the technique described by Shepard for leprosy biopsies17. Fresh arterial biopsies were washed in sterile phosphate buffered saline to remove the remaining blood. These samples were minced, crushed on stainless steel mesh, and mixed with 1 ml of sterile water under sterile conditions. The total volume was split into 3 aliquots. The first was used to prepare smears for acid-fast and auramine-fluorochrome stainings, and cultures on Lowenstein-Jensen culture media according to the usual procedures. The second and third aliquots were used for direct amplification testing (DAT), which is a very sensitive method to detect M. tuberculosis.
Amplification method
Nucleic acid amplification was performed using the sensitive Amplified Mycobacterium tuberculosis direct assay (AMTD; Gen-Probe Inc., San Diego, CA, USA), following the manufacturer’s instructions. This test amplifies the 16S ribosomal RNA of the M. tuberculosis complex by a transcription-mediated amplification. The resulting amplicons are detected by a hybridization protection assay using a specific probe for M. tuberculosis13.
Briefly, 50 μl of the second aliquot were added to a tube containing 25 μl of amplification reagent and 200 μl of oil. The tube was incubated at 42°C for 5 min in a heating block. The amplification enzyme (25 μl) was added, and the tube was incubated at 42°C for 2 h. Then, 20 μl of termination reagent were added, and the sample was incubated at 60°C for 10 min. After addition of the detection reagent (300 μl), the tube was incubated at 60°C for 11 min, and then remained at room temperature for 5 min. The samples were read in a Gen-Probe luminometer (Gen-Probe Inc.) and results were recorded in relative light units (RLU). According to the manufacturer’s instructions, a positive test result was defined as a luminescence reading of 30,000 RLU or greater.
Positive and negative controls
Each test was run in duplicate and included positive (5 × 103 CFU/ml M. tuberculosis) and negative (5 × 103 CFU/ml M. terrae) extraction and amplification controls, as well as positive and negative hybridization controls provided by Gen-Probe. To rule out the presence of polymerase chain reaction inhibitors in the prepared arterial samples, additional positive controls were obtained by adding 100 bacilli of M. tuberculosis into the third aliquots, which were further submitted to the amplification procedure.
RESULTS
Patient characteristics
Arterial samples were obtained from 10 patients (9 women and 1 man, median age 42 yrs, range 19–67 yrs). All these patients were treated by corticosteroids and/or other immunosuppressive agents. Median delay between TA diagnosis and surgical procedures was 5.5 years (range 0–22 yrs). No patient had evidence of active tuberculosis at the time of surgery, as assessed by Mantoux test, chest radiograph, mycobacterial cultures of urine samples, bacilloscopy, and sputum cultures. Patient characteristics and surgical data are presented in Table 1.
Patient characteristics and pathological features.
Pathological analysis
Histological examination of arterial samples revealed active granulomatous arteritis in 4 patients, active nonspecific arteritis in 1 patient, and inactive arterial lesions with intimal and adventitial fibrosis highly suggestive of past arteritis in 5 patients (Table 2). Thus, 5 (50%) of 10 patients were classified as having pathologically active TA, and 5 patients (50%) as having pathologically inactive TA.
Results of pathological and microbiological analyses.
Microbiological analysis
Smears for acid-fast and auramine-fluorochrome staining, as well as mycobacterial cultures on Lowenstein-Jensen culture media were negative for all arterial samples (Table 2). M. tuberculosis was not detected by the AMTD DAT assay in patients with either pathologically active or inactive TA (Table 2).
DISCUSSION
Previous clinical and laboratory studies have suggested that M. tuberculosis could be involved in the pathogenesis of TA3–11. However, the search for M. tuberculosis in arterial lesions of patients with TA has never been examined directly by sensitive methods12. In this study, we have shown that M. tuberculosis could not be detected in arterial samples from 10 consecutive patients with TA of both Caucasian and non-Caucasian origins, with both pathologically active and inactive arteritis, by acid-fast and auramine-fluorochrome staining, mycobacterial cultures, or the sensitive AMTD amplification testing.
Increased prevalence of tuberculosis and of positive tuberculosis skin tests has been reported in Asian, African, and South American series of patients with TA3,4. However, these findings may be only fortuitous, as tuberculosis is endemic in these regions.
One interesting aspect of our study is that the search for M. tuberculosis was performed using microbiological techniques that may detect not only live M. tuberculosis bacilli, as the reference culture method, but also nonviable or dormant bacilli, and even nucleic acid that might have remained in the tissue. Arterial samples were prepared using Shepard’s method, known to be one of the most sensitive methods to prepare slow-growing mycobacteria-containing samples17. Although acid-fast microscopy of smears followed by cultural confirmation remains the cornerstone of the diagnosis of active tuberculosis13, sensitivity of smears ranges only from 22% to 78%18, and culture is dependent on the viability of the bacilli19. All the tests used for active tuberculosis diagnosis yielded negative results in our study, which was expected as all the patients had undergone immunosuppressive treatments and showed no tuberculosis symptoms. Any cryptic but active mycobacterial infection would have probably resulted in an active and patent tuberculosis in these patients.
We did not detect the presence of M. tuberculosis complex RNA. Since this method was shown to be 97% sensitive and up to 100% specific20, we could reasonably conclude that arterial samples from patients with TA did not and had not contained tuberculous bacilli.
Interestingly, the search for M. tuberculosis yielded negative results in patients with both recent and active arteritis and in patients with longstanding and pathologically inactive TA. Vascular surgical procedures were performed within 2 years from the diagnosis of TA in Patients 1, 2, and 3, suggesting that M. tuberculosis is not present in the vascular wall even at an early and active stage of the arterial lesions.
Among the limitations of our study was the relatively small number of biopsies analyzed. However, taking into account the very low prevalence of the disease in European and North American populations21,22, and the limited indications for vascular surgery in TA (clinical and/or hemodynamically significant arterial stenoses, and large aneurysms), our series may be considered as a valuable exploratory work.
Because we did not find any evidence of M. tuberculosis in the arterial samples from patients with TA, a direct role of this infectious agent in the pathogenesis of TA seems unlikely. Our study excludes neither the possibility of another infectious etiology of the disease, nor the presence of an extraarterial antimycobacterial immune response triggering a cross-reaction directed towards an antigen of the arterial wall10. Future studies focused on the identification of cross-reacting arterial antigens would be important to delineate the exact pathogenesis of the disease.
Acknowledgments
We are grateful to Claudine Wichlacz for technical assistance.
Footnotes
- Accepted for publication March 3, 2009.
