Problems With Ascribing Between-Trial Differences in BCG Effectiveness to Sensitization With Environmental Mycobacteria

TO THE EDITOR—Mangtani et al recently analyzed the results of 21 trials of BCG vaccination [1], finding that “latitude and age at vaccination/tuberculin testing stringency could explain all of the betweentrial heterogeneity.” They conclude that the “absence of prior M. tuberculosis infection or sensitization with environmental mycobacteria is associated with higher efficacy of BCG against pulmonary tuberculosis and possibly against miliary and meningeal tuberculosis.” The first test of the merit of a hypothesis is how well it explains prior observations. If the work of Mangtani et al survives scrutiny, then they will have settled a long-standing controversy regarding whether observed differences in BCG efficacy are due to sensitization with environmental mycobacteria or changes in BCG. However, there are problems with their analysis. For example, although the authors support their theory by citing a retrospective analysis involving children in the Chingleput trial with low reactivity toMycobacterium intracellulare in whom BCG was modestly effective, they fail to mention that, in the Puerto Rico, trial a large tuberculin dose (100 TU) was used to identify sensitized children and that BCG was as effective in this group as in nonsensitized children [2]. Their model also does not explain why BCG was ineffective in Muscogee County, Georgia, despite the use of a stringent skin test to exclude sensitized children. It is also unclear why 60% of adults 25–44 years old were protected