Journal of Infectious Diseases
Volume 211 Issue 3 February 1, 2015
http://jid.oxfordjournals.org/content/current

EDITORIAL COMMENTARY
Delayed BCG Vaccination—Time to Take a Shot
Alexander W. Kay1 and Catherine A. Blish2,3
Author Affiliations
1Department of Pediatrics
2Department of Medicine
3Stanford Immunology, Stanford University School of Medicine, California
(See the major article by Toukam Tchakoute et al on pages 338–46.)
[Excerpt]
The BCG vaccine is often derided for the lack of efficacy in preventing Mycobacterium tuberculosis infection and pulmonary disease in adults. However, BCG vaccine remains a highly effective and cost-efficient intervention to prevent tuberculous meningitis and miliary tuberculosis in infants, reducing the incidence of these life-threatening and debilitating infections by approximately 75% [1, 2]. In addition, BCG vaccine coverage rates typically exceed those of other vaccines because it can be administered at birth as a single vaccination [3].

However, this strength of the BCG vaccination strategy has become a liability because of the risks of administering BCG vaccine to human immunodeficiency virus (HIV)–infected infants. The HIV diagnosis is typically not made until the second or third month of life in resource-limited settings, and BCG vaccination in this population results in unacceptably high rates of disseminated BCG disease of 417–992 cases per 100,000 vaccinations, with a mortality of approximately 75% [4–6]. To put this in perspective, this rate of disseminated BCG disease exceeds the rate of disseminated disease due to M. tuberculosis in the same South African population of HIV-infected infants, which is estimated to be 241 cases per 100,000 [7]. In light of this significant risk for the vaccine to cause harm, the World Health Organization (WHO) now identifies known HIV infection in infants, or HIV exposure and symptoms concerning for HIV, as contraindications to BCG vaccination [8, 9]. The rationale for this recommendation is augmented by the unknown clinical efficacy of BCG vaccination in HIV-infected infants and the immunologic data suggesting that BCG given at birth is unlikely to be efficacious in this population [10]….