© 2006 American Thoracic Society
Reinfection with Mycobacterium tuberculosisTo the Editor:Verver and colleagues in their recent article provide valuable information on reinfection with Mycobacterium tuberculosis, but it is not correct to conclude that this excellent study "challenges the hypothesis that in immunocompetent persons, infection with one strain of M. tuberculosis protects against disease due to subsequent reinfection with another strain" (1). First, HIV testing of all study subjects would have been required to conclude that the data are relevant to immunocompetent persons. The authors' estimated HIV coinfection rate of 11% would represent the lowest rate reported in sub-Saharan Africa, typically 40 to 70%. The relative risk for reinfection with tuberculosis for HIV-infected persons has been estimated to be 18.7 (2). Applying this risk to the authors' observed rate of 2.2/100 patient-years (p-yr), the rate would be 0.75/100 p-yr in the HIV-negatives and 14.0/100 p-yr in the 11% estimated to be HIV-positive. Thus, nearly all of the effect could be attributed to HIV infection. Second, as many as 50% of patients who require treatment for active tuberculosis may be culture-negative but meet a clinical definition of tuberculosis (3). We do not presently have the means to obtain microbiological or molecular characterization of the strains causing these culture-negative cases and therefore cannot determine their risk of reinfection. It is possible that culture- positive cases represent a group genetically susceptible to reinfection, but that culture-negative cases reflect a stronger adaptive immune response with reduced susceptibility to reinfection. Third, extrapolation of the protection from tuberculosis disease to protection from tuberculosis infection is quite speculative. Tuberculosis disease represents only a small part of the burden of tuberculosis in an endemic region, where 50 to 80% may have tuberculosis infection. Persons who are infected and do not develop disease are likely to have natural host protective mechanisms that exceed those of persons who are infected and develop disease. Therefore, their risk for reinfection disease could well be less that that of persons with previous tuberculosis disease. While molecular studies of reinfection are making major contributions to our understanding of immunity to tuberculosis, it is important to recognize that it is not presently possible to know the denominator for the numerator data that would allow estimation of rates of reinfection disease among all persons with tuberculosis infection or tuberculosis disease. A substantial body of both experimental and epidemiologic evidence supports the concept that immunocompetent persons with tuberculosis infection have relative protection against reinfection tuberculosis disease (4, 5). This evidence underlies current efforts to develop effective tuberculosis vaccines (6). The report by Verver and coworkers should not be construed to suggest that immunocompetent hosts do not develop immunity against tuberculosis after tuberculosis infection, nor to imply that vaccines that mimic host immune responses to M. tuberculosis might not be effective.
Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
Boston University School of Public Health, Boston, Massachusetts FOOTNOTES Conflict of Interest Statement: Neither of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
