The article of S. Thambu David et al "Detecting mycobacteraemia for diagnosing tuberculosis'" in this issue of the Journal is an opportune time to comment on this approach for improving our capabilities to confirm the diagnosis of tuberculosis. There are several important issues that need to be considered.
The first question is whether this approach is needed? It is known that the characteristic histology, demonstration of acid fast bacilli (AFB), and isolation of Mycobacterium tuberculosis are important conventional approaches for confirming the diagnosis of tuberculosis. However, all these have some limitations. While histology may not be characteristic in all cases, demonstration of AFB has problems of sensitivity and also specificity. Same is true for isolation of M. tuberculosis even though this approach is much more specific with the added advantage of having the isolate for carrying out drug susceptibility. It is well known that tuberculosis involves almost all organs of human body and disease manifestations may vary from a localized to disseminated disease. While a significant proportion of pulmonary specimens are negative for AFB, because of the inaccessibility of afflicted sites, it is difficult to get appropriate sample in most of extrapulmonary cases. Further, pulmonary cavitory disease may not be present in a section of AIDS cases. But on the other hand, the presence of mycobacteraemia has been increasingly reported in cases with disseminated disease. Because of these factors, blood becomes another 'easy to collect and analyze' specimen for confirming the diagnosis of tuberculosis.
While blood is a convenient specimen to collect like in subacute bacterial endocarditis (SABE) and typhoid, mycobacteria in many cases of tuberculosis may be present in low numbers and intermittently in the blood. Thus in clinical practice it is important to determine the clinical profile of cases in which this approach will be beneficial. As also brought out in the article1, mycobacteraemia has been more commonly associated with disseminated disease which is found in situations like miliary tuberculosis and AIDS cases2'5. It is reasonable to think that it will be more meaningful to correlate mycobacteraemia with immune status rather than HIV positivity status. Mycobacteraemia has also been reported in non-HIV TB cases as seen in some cases of present series and also observed by others. For instance in a study carried out at our Institute (unpublished observations), in nearly a third of cases of HlV negative histologically proven cases of lymph gland tuberculosis, M. tuberculosis could be isolated from blood. As required for diagnosis of SABE and typhoid etc., it would be important to have multiple blood samples to achieve better sensitivity. Appropriate numbers need to be determined by actual trials.
Several methods have been described to demonstrate and culture mycobacteria from blood. Lysiscentrifugation strategy has been commonly followed1,4,6,7 and has shown good results. It may be desirable to keep on evaluating and improving these methods learning from experience of other diseases like leprosy8.
It is easy and convenient to demonstrate AFB in smears. But, this approach has several limitations which include low sensitivity and identity of AFB. It has been reported that smears have much lower sensitivity than culture2. Further, mycobacteria other than M. tuberculosis could be responsible for mycobacteraemia in AIDS cases2. It would thus be more important to rely on isolation of mycobacteria and their identification.
For the cultivation of mycobacteria from blood, conventional media like Lowenstein- Jensen (LJ)7 as well as rapid systems like BACTEC5, BACTEC 13 A, BactT/ALERT MB9 have been used. There are very few studies available on the comparative usefulness of these systems. Major advantage of newer radiometric (BACTEC) and non-isotopic systems (BactT/ALERT MB) over conventional approaches is time taken for getting the result. For culture on LJ media, time taken for growth of M. avium and M. tuberculosis ranged from 4-6 wk7 whereas the average time taken was 2-3 wk with the newer assay systems9. As no major differences in the sensitivity are apparent, it would be important to take into consideration economic aspects as well.
As expected, cultivation methods have significant lower sensitivity than gene amplification even for blood specimens. PCR has been used by many investigators for the detection of mycobacteraemia6,10,11. While a high positive rates in HIV-AIDS cases have been observed, this approach has been found to be much more useful in non-HIV cases. In a study on HIV negative pulmonary tuberculosis cases, 44 per cent positivity was reported11. in another study of tubercular lymphadenitis 46 percent positivity by PCR as compared to 36 per cent by culture in blood was recorded (unpublished observation). The results of PCR are often doubted due to fear of false positivity but these apprehensions are often not correct. However, there is a long way before the approach is accepted by end users.
Isolation of M. tuberculosis and/or detection by PCR may be the only evidence in some cases. This is important help in improving the diagnosis. But the results need to be cautiously interpreted especially when PCR is being taken as sole evidence.
Taking into consideration the above selected reports and the current article of Thambu David et al1, it is clear that detection of mycobacteraemia should be considered as one of the potentially useful approaches to arrive at the diagnosis of tuberculosis. While positive evidence is helpful, negative result cannot be considered significant. Whether mycobacteria will be occasionally circulating transiently in a few healthy individuals needs to be investigated and how often this happens also should be analyzed in prospective studies. More studies are necessary to determine the relative usefulness of this approach and also find out/improvise existing methods in the diagnosis of TB.
References
1. Thambu David S, MuKundan U, Brahmadathan KN, John TJ. Detecting mycobacteraemia for diagnosing tuberculosis. Indian J Med Res 2004; 119 : 259-66.
2. Stone BL, Cohn, Kane MS, HIldred MV, Wilson ML, Reves RR. Utility of paired blood cultures and smears in diagnosis of disseminated Mycohacterium avnim complex infections in AIDS patients. J Clin Microbiol 1994. 32 : 841-2.
3. Barber TW, Craven DE, McCabe WR. Bacteremia due to Mycobacterium tuberculosis in patients with human immunodeficiency virus infection. A report of 9 cases and a review of the literature. Medicine (Baltimore) 1990; 69 : 375-83.
4. Rosas RC. Salomao R, da Malta DA, Lopes HV, Pignatari AC, Colombo AL. Blood stream infections in late-stage acquired immuno-deficiency syndrome patients evaluated by a lysis centrifugation system. Mem Inst Oswaldo Cruz, Rio de Janereiio 2003; 98 :529-33.
5. Ramachandran R, Svvaminathan S, Somasundaram S, Asgar VN, Paramesh P, Paramasivan CN. Mycobacteremia in tuberculosis patients with HIV infection. Indian J Tuberc 2002; 50:29-31.
6. Kiehn TE, Gold JWM, Timberger RJ, Armstrong D. Mycobacterium tuberculosis bacteremia detected by isolator lysis-centrifugation blood culture system. JClin Microbiol 1985; 21: 647-8.
7. Esteban J, MoIIeJaA, Fernandez-Roblas R, Soriano F. Number of days required for recovery of mycobacteria from blood and other samples. J Clin Microbiol 1998; 36 : 1456-7.
8. Raval N, Sengupta U, Ramu G, Prabhu PV, Desikan KV. A study of continuous bacillemia in borderline and lepromatous type of leprosy. Lepr India 1982; 54 : 623-33.
9. Crump JA, Tanner DV, Mirret TS, McKnight CM, Barth Relier L. Controlled comparison of BACTEC 13 A, MYCO/F LYTIC, BactT/ALERT MB and ISOLATOR 10 systems for detection of mycobacteremia. J Clin Microbiol 2003; 41 : 1987-90.
10. Folgueria I, Delgado R, Palenque E, Aguado JM, Noreiga AR. Rapid detection of Mycobacterium tuberculosis bacteremia by PCR. J Clin Microbiol 1996; 34 : 512-5.
11. Ahmed N, Mohanty AK, Mukhopadhyyay U, Batish VK, Grover S. PCR-based rapid detection of Mycobacterium tuberculosis in blood from immunocompetent patients with pulmonary tuberculosis. J Clin Microbiol 1998; 36 : 3094-5.
V.M. Katoch
Director
Central JALMA Institute for Leprosy & Other Mycobacterial Diseases (ICMR)
Tajganj, Agra202001, India
e-mail: rohinik@sancharnet.in
Copyright Indian Council of Medical Research Jun 2004
Provided by ProQuest Information and Learning Company. All rights Reserved
Contact
Home
A
Babesiosis