Use of BCG in high prevalence areas for HIV.

Recommendations state that, where the risk of tuberculosis is high, BCG should be administered to infants as early in life as possible, even if the mother is known to be HIV-infected. BCG should be withheld from individuals with symptomatic HIV infections. However, continuing reports from sub-Saharan Africa and elsewhere of BCG complications in HIV-infected persons call for a re-assessment of current vaccination policies. For HIV-infected infants any benefit of BCG vaccination may be marginal because the prognosis is very poor. It is however not possible to exclude HIV-infected children from BCG vaccination at birth. HIV-uninfected infants born to HIV-infected mothers are at great risk of tuberculosis infection, which justifies routine vaccination. BCG rarely causes serious complications. Theoretically, persons with asymptomatic HIV infection may be at greater risk of complications from BCG vaccines, but available data are inconclusive in that respect. To vaccinate children with BCG at one year of age does not seem feasible and would increase the risk of tuberculosis especially for uninfected infants of HIV seropositive mothers. Available data seem to indicate that routine vaccination of newborns is indeed safe, even in areas with high prevalence of HIV infection.