Reinfection Tuberculosis
How Important Is It?

Joseph H. Bates

University of Arkansas College of Medicine, Little Rock, Arkansas

	ARTICLE

TOP ARTICLE REFERENCES

How often does exogenous reinfection with Mycobacterium tuberculosis occur among immunocompetent persons who have been treated and cured of tuberculosis? Does exogenous reinfection occur among healthy persons who are tuberculin skin test positive and have latent tuberculosis infection? Are there host genetic factors that come into play in this equationare some healthy or cured subjects more resistant to challenge by M. tuberculosis as a result of genetic mechanisms that control native immunity? Are some strains of M. tuberculosis more virulent or more likely for other reasons to infect subjects who have native or acquired immunity? These questions have been raised over and over by investigators, clinicians, and tuberculosis controllers because of their importance to those who seek to understand immunity/resistance to tuberculosis, and to those working toward the development of an effective vaccine. It is also of interest to those who prepare prediction models for tuberculosis control and eradication in geographic areas with widely varying tuberculosis case rates. In this issue of AJRCCM (pp. 717-720), the article entitled "Exogenous Reinfection with Tuberculosis on a European Island with a Moderate Incidence of Disease" by Caminero and coworkers provides valuable information on these questions (1). Caminero and colleagues evaluated persons with recurrent tuberculosis living on the Spanish Island of Gran Canaria between 1991 and 1996. The island population of almost 714,000 persons had a case rate that varied from 28 to 32 per 100,000 during the study period. This rate is about five times higher than in the United States and about 15-20 times less than in countries where tuberculosis is out of control. Among 912 patients with positive cultures during the study period, 23 (2.4%) became culture positive again after being culture negative for at least 12 mo postchemotherapy. Cultures obtained before therapy and 12 mo or more after ending therapy were available for 18 patients. For all 18 patients DNA fingerprints of the pretreatment and recurrent isolates were obtained and for eight of them the genotype of the recurrent isolate showed a different pattern as compared with the pretreatment isolate. The authors conclude that these 8 patients are examples of exogenous reinfection. Thus about half the patients in this population who experienced a second episode of tuberculosis were apparently infected exogenously. And this occurred in a setting where the case rate is modestly low, meaning that chance inhalation of viable tubercle bacilli would be infrequent as compared with an area with a much higher case rate. This observation runs counter to commonly held concepts and must be scrutinized carefully.

The first step is to examine how each reinfection was documented. Did laboratory error such as mislabeling or laboratory cross-contamination of specimens occur? I think not. At least two isolates each from the pre- and posttreatment period were genotyped in each instance and the cases were evaluated clinically as well for signs and symptoms of recurrent disease. In every case there seemed to be no other explanation except exogenous reinfection. However, there is one troubling point: three of the exogenous reinfecting isolates were of the same genotype, a finding that seems unlikely to occur by chance alone. Could laboratory cross-contamination explain it? The authors state, without giving any data, that this particular genotype is "frequently isolated on the island" and no epidemiological links were found among these patients. The authors state that they are not surprised to find multiple isolates of this genotype in this set. This is a concern that remains unresolved. Two of the eight patients with exogenous reinfection had acquired immunodeficiency syndrome (AIDS) and it is well known that these patients are easily reinfected.

In my view, reinfection of individuals who have been cured of tuberculosis remains an uncommon explanation for recurrent tuberculosis among immunocompetent patients living in geographic areas of low incidence. The chance of anyone inhaling viable tubercle bacilli in these areas is growing increasingly less likely. In addition, these episodes of recurrent tuberculosis after an effective therapy regimen has been completed are clustered in the first 2 yr after ending treatment. One would expect exogenous reinfection to be a more random event and thus should occur sporadically over the remaining lifetime of the subject.

Caminero and colleagues cite a number of prior publications that show strong evidence of exogenous reinfection and give emphasis to a report by van Rie and coworkers, who studied this question among persons living in a very high incidence area of Cape Town, South Africa (2). In this setting it was found that exogenous reinfection is the major cause of recurrent disease. I have serious reservations about this particular report and advise caution about extending these observations as being true and applicable to other high incidence areas. Laboratory cross-contamination as an explanation for some of their observations was not effectively excluded by van Rie and associates, nor was the clinical evidence of relapse convincingly set out. More details about these concerns are published in another journal (3).

Cured patients who live in high incidence areas are at risk of inhaling tubercle bacilli on a relatively frequent basis and if their native and acquired immunity to tuberculosis is inadequate, recurrent disease with a new infecting strain will result. No doubt this does occur. The larger question is how often. More studies of this question are needed, not only to help those planning for tuberculosis control, but for those interested in tuberculosis immunity and variation in virulence among isolates of M. tuberculosis. Human subjects who have firmly demonstrated reinfection should be studied for genetic markers of resistance to tuberculosis (as these markers become known) and the infecting isolates should be evaluated for any features they might exhibit that promote their ability to cause disease in a subject with presumed acquired immunity.

	References

TOP ARTICLE REFERENCES

1. Caminero JA, Pena MJ, Campos-Herrero MI, Rodriguez JC, Alfonso O, Martin C, Pavon JM, Torres MJ, Burgos M, Cabrera P, Small PM, Enarson DA. Exogenous reinfection with tuberculosis on a European island with a moderate incidence of disease. Am J Respir Crit Care Med 2001; 163: 717-720 [Abstract/Free Full Text].

2. van Rie A, Warren R, Richardson M, Vector TC, Gie RP, Enarson DA, Beyers N, van Helden PD. Exogenous reinfection as a cause of recurrent tuberculosis after curative treatment. N Engl J Med 1999; 341: 1174-1179 [Abstract/Free Full Text].

3. Stead WW, Bates JH. Recurrent tuberculosis due to exogenous reinfection. N Engl J Med 2000; 341: 1050 .