Correspondence

Recurrent Tuberculosis Due to Exogenous Reinfection

N Engl J Med 2000; 342:1050-1051April 6, 2000

Article

To the Editor:

The report by van Rie and colleagues (Oct. 14 issue)1 ascribes 12 of 16 cases of recurrent tuberculosis in patients who were negative for the human immunodeficiency virus (HIV) to exogenous reinfection. However, we are concerned about several points.

In Table 2 of the article, it appears that Patients 1, 2, 3, 5, 12, and 13 had a negative sputum smear and only a single positive culture in the first episode of tuberculosis, as did Patients 7 and 9 in the subsequent episode. More information should be provided in order to make a convincing case for exogenous reinfection. Presumably, “no” indicates stronger evidence of disease. However, Burman et al. have shown that even a positive smear and heavy growth on culture do not rule out laboratory cross-contamination.2 Were any of these sputum specimens processed in the same laboratory at or near the time that any other smear or culture-positive specimen was processed? If so, did any of these isolates show a matching pattern on restriction-fragment–length polymorphism (RFLP) analysis?

Why were the RFLP patterns for Patients 15 and 16 not included in Figure 1? These cases are not mentioned in the Discussion section. The RFLP patterns for organisms from the “previous” or “subsequent” episode are the same for Patients 7, 8, and 12, and the pattern for Patient 3 differs from these by a single hybridizing fragment. A secondary method of RFLP typing might help show whether these strains are truly different.

Finally, were the chest roentgenograms and clinical pictures of the patients with a single positive culture reviewed to distinguish active tuberculosis from dormant scars? This is a common source of confusion with cross-contamination.

William W. Stead, M.D.
Joseph H. Bates, M.D.
Arkansas Department of Health, Little Rock, AR 72205-3867

2 References

1. 1

   van Rie A, Warren R, Richardson M, et al. Exogenous reinfection as a cause of recurrent tuberculosis after curative treatment. N Engl J Med 1999;341:1174-1179
   Full Text | Web of Science | Medline

2. 2

   Burman WJ, Stone BL, Reves RR, et al. The incidence of false-positive cultures for Mycobacterium tuberculosis. Am J Respir Crit Care Med 1997;155:321-326
   Web of Science | Medline

To the Editor:

In the Netherlands in 1997, the incidence of tuberculosis was 8.5 per 100,000 population for new cases and 1.1 per 100,000 for cases of relapse. We performed RFLP analysis with IS6110 on isolates of Mycobacterium tuberculosis from 546 patients with two or more isolates, received between mid-1986 and December 1996, excluding cases of laboratory cross-contamination.1 Of these, 543 (99 percent) had a repeat isolate with an RFLP pattern that was identical or almost identical to that of a prior isolate. Only two had isolates that were completely different, suggesting reinfection, and one had lost one band and gained three bands.

For a proper comparison with the study by van Rie et al., we excluded 501 patients (92 percent) with less than six months between isolates. Of the remaining 45 patients, information was available for 15. Of these, eight had a positive culture while still receiving treatment, one had a relapse after defaulting, and six had been declared cured. For all 15 patients, the isolates of M. tuberculosis from the second episode had RFLP patterns that were identical to those of the isolates from the first episode. Thus, all six patients with a relapse after cure had reactivation (100 percent; 95 percent confidence interval, 60 to 100).

In our study, reinfection was rare. This provides direct evidence, as suggested in the editorial2 accompanying the article by van Rie et al., that in a country with a low incidence of tuberculosis, cases of relapse are probably due to endogenous reactivation.

Annette S. de Boer, M.Sc.
Dick van Soolingen, Ph.D.
National Institute of Public Health and the Environment, 3720 BA Bilthoven, the Netherlands

Martien W. Borgdorff, M.D., Ph.D.
Royal Netherlands Tuberculosis Association, 2501 CC The Hague, the Netherlands

2 References

1. 1

   de Boer AS, Borgdorff MW, de Haas PEW, Nagelkerke NJ, van Embden JDA, van Soolingen D. Analysis of rate of change of IS6110 RFLP patterns of Mycobacterium tuberculosis based on serial patient isolates. J Infect Dis 1999;180:1238-1244
   CrossRef | Web of Science | Medline

2. 2

   Fine PEM, Small PM. Exogenous reinfection in tuberculosis. N Engl J Med 1999;341:1226-1227
   Full Text | Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: Stead and Bates are concerned about the accuracy of the clinical diagnosis of tuberculosis and the molecular classification of isolates of M. tuberculosis collected from patients with recurrence of tuberculosis after curative treatment. As they correctly point out, incorrect diagnosis or laboratory error may bias the interpretation of these results. All 16 patients in our study needed treatment for tuberculosis on both occasions on the basis of their clinical symptoms, chest radiograph, and positive culture for M. tuberculosis. Of the six patients who, during their first episode of tuberculosis, had a negative smear and a single positive culture, three had cavities, two had infiltration of the upper lobe, and one had symptoms consistent with the presence of tuberculosis. Both patients with a negative smear and a single positive culture during the subsequent episode of tuberculosis were found to have large cavities in the upper lobe on chest radiography. We are therefore confident that all patients with a negative smear and a single positive culture had active tuberculosis at the time of the positive culture and not just dormant scars.

We are also confident that the molecular results are correct. As described, the accumulated laboratory error for the complete RFLP data base was found to be 3.4 percent. It is highly unlikely that this error rate would be overrepresented among the patients who were classified as having been reinfected after clinical cure. We acknowledged that a similar error rate would apply to the isolates collected from these patients; however, this would not influence the conclusion that the predominant pathogenic mechanism of recurrent tuberculosis in this community with a high incidence of the disease is reinfection. To eliminate the possibility of laboratory cross-contamination further, we reanalyzed the IS6110 RFLP patterns from all the isolates cultured on the same day and could not identify any strains with identical RFLP patterns; cross-contamination was therefore highly unlikely.

We have published further support for the conclusion that reinfection is a primary cause of recurrent tuberculosis in a high-incidence community.1 RFLP analysis of isolates of M. tuberculosis collected from 16 patients with tuberculosis that is resistant to multiple drugs, in which four DNA fingerprinting probes were used, showed that the isolates were genotypically identical. Eleven of these patients had prior episodes of tuberculosis with different and sensitive strains. Since it is highly improbable that the strains individually acquired identical drug-resistance mutations, we concluded that these patients were reinfected with a single outbreak strain. In contrast, de Boer et al. reported that recurrent tuberculosis occurred only by reactivation in the Netherlands. This is to be expected, given the differences in incidence rates for the communities studied. In a community with a high incidence of tuberculosis, the chance of reinfection would be expected to be high. As the incidence rate declines so does the reinfection rate, until it becomes almost nonexistent.

Annelies van Rie, M.D.
Robin Warren, Ph.D.
Paul D. van Helden, Ph.D.
University of Stellenbosch, Tygerberg 7505, South Africa

1 References

1. 1

   van Rie A, Warren RM, Beyers N, et al. Transmission of a multidrug-resistant Mycobacterium tuberculosis strain resembling “strain W“ among noninstitutionalized, human immunodeficiency virus-seronegative patients. J Infect Dis 1999;180:1608-1615
   CrossRef | Web of Science | Medline

Citing Articles (7)

Citing Articles

1. 1

   G.D. van der Spuy, P.D. van Helden, R.M. Warren. (2009) Effect of study duration on the interpretation of tuberculosis molecular epidemiology investigations. Tuberculosis 89:3, 238-242
   CrossRef

2. 2

   Seung Soo Yoo, Jee Suk Kwon, Yeh Rim Kang, Jeong Woo Lee, Seung Ick Cha, Jae Yong Park, Tae Hoon Jung, Chang Ho Kim. (2008) The Clinical Characteristics and Outcomes of Short-term Treatment in Patients with Recurrent Pulmonary Tuberculosis. Tuberculosis and Respiratory Diseases 64:5, 341
   CrossRef

3. 3

   T. Mark Doherty. (2004) New vaccines against tuberculosis. Tropical Medicine and International Health 9:7, 818-826
   CrossRef

4. 4

   Eline L. Korenromp, Fabio Scano, Brian G. Williams, Christopher Dye, Paul Nunn. (2003) Effects of Human Immunodeficiency Virus Infection on Recurrence of Tuberculosis after Rifampin‐Based Treatment: An Analytical Review. Clinical Infectious Diseases 37:1, 101-112
   CrossRef

5. 5

   Annika Krüüner, Lea Pehme, Solomon Ghebremichael, Tuija Koivula, Sven E. Hoffner, Marika Mikelsaar. (2002) Use of Molecular Techniques to Distinguish between Treatment Failure and Exogenous Reinfection with Mycobacterium tuberculosis. Clinical Infectious Diseases 35:2, 146-155
   CrossRef

6. 6

   Pamela Sonnenberg, Jill Murray, Judith R Glynn, Stuart Shearer, Bupe Kambashi, Peter Godfrey-Faussett. (2001) HIV-1 and recurrence, relapse, and reinfection of tuberculosis after cure: a cohort study in South African mineworkers. The Lancet 358:9294, 1687-1693
   CrossRef

7. 7

   D. Van Soolingen. (2001) Molecular epidemiology of tuberculosis and other mycobacterial infections: main methodologies and achievements. Journal of Internal Medicine 249:1, 1-26
   CrossRef