To the Editor:
We read with great interest the article by Li et al. (Diagn Mol Pathol 2000;9:67–74) regarding molecular detection of Mycobacterium tuberculosis (M.tbc) in tissues showing granulomatous inflammation. The detection of deoxyribonucleic acid (DNA) from M.tbc by polymerase chain reaction (PCR) has been described as a suitable method for the rapid and sensitive diagnosis of M.tbc infection. However, the authors did not evaluate sufficiently the benefits and limits of this technique (5).
For detection of M.tbc, Li et al. used PCR with oligonucleotide primers, described previously by Eisenach et al. (2), at slightly modified reaction conditions. These primers amplify a 123-base pair (bp) fragment of the IS 986/6110 insertion sequence specific for M.tbc complex and M. bovis Bacille Calmette-Guerin (BCG) as shown by sequencing the PCR product (2,4). They studied paraffin-embedded tissues of 115 patients showing granulomatous inflammation without demonstrable acid-fast bacilli by Ziehl–Neelsen staining. Sixty-eight of these tissue samples yielded positive results in the 123-bp PCR analysis. In addition, culture results using corresponding tissue samples were available in a different cohort of 68 of the 115 patients studied. Fifty-three of these patients were diagnosed clinically with tuberculosis (TB), but only 22 of them were M.tbc-culture positive, and 38 of them were 123-bp PCR positive. Thirteen patients were diagnosed clinically as having TB without detection of M.tbc or M.tbc DNA. With reference to this clinical diagnosis, Li et al. reported a low sensitivity of both M.tbc culture and 123-bp PCR (42% and 72% respectively) in their study, although other pulmonary diseases such as infection with atypical mycobacteria may not be fully excluded. Interestingly, the percentage of 123-bp PCR-positive samples determined from culture-positive TB patients (20 of 22) was comparable with the PCR sensitivity in our study (91% vs. 95%) (4). To estimate the detection limit of 123-bp PCR for M.tbc complex DNA extracted from paraffin-embedded samples, we analyzed geometrically diluted BCG–vaccine (strain Kopenhagen 1331; Chiron-Behring, Mannheim, Germany) in clotted blood plasma after formalin fixation and embedding in paraffin. In comparison with Ziehl–Neelsen staining, we observed a 1,000 to 5,000-fold higher sensitivity of 123-bp PCR in paraffin-embedded samples (Fig. 1). Therefore, we fully agree with the conclusion by Li et al. that 123-bp PCR is much more sensitive than conventional acid-fast bacilli staining in detection of M.tbc infection.
In addition, Li et al. found M.tbc DNA by 123-bp PCR in 27% samples from patients not verified as TB patients (four of 15) and in 55% samples from patients without any available clinical diagnosis (26 of 47). However, they did not explain these discrepancies sufficiently. Using 123-bp PCR, we investigated paraffin-embedded lung tissue from 50 patients with clinically proved sarcoidosis, and found in 64% of patients the amplification product of M.tbc DNA (4). These findings are in accordance with the results from a number of other studies (1,3,6), which also detected M.tbc DNA in approximately half of their examined sarcoidosis patients using PCR based on the same sequence. In addition, we could demonstrate that M.tbc may play a pathogenetic role, at least in a certain group of sarcoidosis patients—namely, in those patients with elevated soluble interleukin-2 receptor values.
Thus, an unambiguous discrimination between sarcoidosis and TB as well as a “molecular diagnosis of TB” by detection of M.tbc DNA remains controversial. The 123-bp PCR, however, is a suitable method for sensitive and fast detection of M.tbc DNA in histologic material from patients with TB or sarcoidosis.
M. Grosser Ph.D.
D. D. Dittert M.D.
T. Luther M.D., Ph.D.
1. Degnitz K. Detection of mycobacterial DNA in the skin. Arch Dermatol 1996; 132:71–5.
2. Eisenach KD, Cave MD, Bates JH, Crawford JT. Polymerase chain reaction amplification of repetitive DNA sequence specific for Mycobacterium tuberculosis
. J Infect Dis 1990; 161:977–81.
3. Fidler HM, Rook GAW, Johnson N, McFadden JJ. Mycobacterium tuberculosis
DNA in tissue affected by sarcoidosis. BMJ 1993; 306:546–9.
4. Grosser M, Luther T, Müller J, et al. Detection of M. tuberculosis
DNA in sarcoidosis: correlation with T-cell response. Lab Invest 1999; 79:775–84.
5. Noordhoeck GT, Kolk AHJ, Bjune G, et al. Sensitivity and specificity of PCR for detection of Mycobacterium tuberculosis:
a blind comparison study among seven laboratories. J Clin Microbiol 1994; 32:277–84.
6. Saboor SA, Johnson NM, McFadden J. Detection of mycobacterial DNA in sarcoidosis and tuberculosis with polymerase chain reaction. Lancet 1992; 339:1012–5.