Abstract
The safety of immunizing HIV-infected infants and children with live vaccines, including Mycobacterium bovis, bacille Calmette Guérin (BCG), was first addressed by the World Health Organization in 1987. The advisory group concluded that the benefits of BCG immunization for all children outweighed the risks among those with HIV infection, and the group recommended that routine childhood immunization be continued and that BCG vaccine only be withheld from infants with symptomatic HIV infection [1]. A 2004 World Health Organization review came to the same conclusion, adding that BCG-immunized infants of HIV-infected mothers should undergo long-term follow-up to monitor for possible adverse effects associated with BCG vaccine [2]. Two developments warrant reconsideration of these recommendations: the results of a study of BCG complications from South Africa published in this issue of the journal [3] and the identification of BCG-induced immune reconstitution inflammatory syndrome (IRIS) in children receiving antiretroviral therapy.