In this study, de Castro and colleagues wisely made use of a “natural experiment.” Some regions of Spain altered standard immunization practices in a way that left some still giving routine neonatal BCG vaccination and others forgoing that vaccine. Carefully controlling for as many other variables as they could, these investigators compared hospitalization rates for non-tuberculous respiratory infections and sepsis between Basque Country (where BCG was still routinely given) and the rest of Spain.

Nearly half a million hospitalizations from 1992 to 2011 were analyzed. BCG-vaccinated children had significantly fewer respiratory hospitalizations for all age groups than did other children (preventive fraction 41%, p < 0.001). The hospitalization rate for sepsis during the first year of life was also lower in BCG-vaccinated children (preventive fraction 53%, p < 0.001). (The hospitalization rate for tuberculosis-related illnesses was also 74% lower in the Basque Country where BCG was still used.)

The authors point out that BCG vaccine seems to stimulate a subsequently enhanced release of monocyte-derived cytokines when a child is exposed to non-tuberculous bacterial and fungal infections. Thus, they viewed their data as supporting the conclusion that BCG vaccination may provide heterologous protection against non-tuberculous infections/illnesses.


BCG is one of the safest and most widely used vaccines. For decades, however, most American global health professionals have advised against the use of routine BCG vaccination with good reason. Now, mounting evidence suggests that perhaps there is still value in using BCG for newborns.

Why have we said not to use BCG?

First, the incidence of TB in the United States is relatively low. Some assume the risk is low enough not to bother vaccinating. They leave the vaccine for use in developing countries where there is more tuberculosis and where early diagnosis and treatment of active infection might not be feasible.

Second, though relatively safe, the vaccine is not without risk. Some children develop adenitis after the vaccine, and children with hereditary immunodeficiencies can develop disseminated disease from the BCG germs.

Third, use of the vaccine can complicate the interpretation of subsequent tuberculosis tests. While we never really took previous BCG use into consideration when interpreting tuberculosis skin tests, we did know that BCG can “falsely” increase induration around the site of purified protein derivative (PPD) skin tests. Some argued that future diagnosis post-exposure to tuberculosis would be complicated by previous receipt of BCG.

Finally, and most importantly, the vaccine doesn’t help much. There is good evidence that BCG reduces the risk of miliary and disseminated tuberculosis. But, BCG is not consistently effective in preventing pulmonary tuberculosis. Some studies have shown favorable effects, but others have not; this is perhaps due to variations if efficacy between various lots/strains of BCG.

Now, though, new data prompt us to challenge each of these aspects of our sage BCG-avoiding advice.

First, it is still true that tuberculosis is uncommon in the United States. In fact, the incidence has been steadily decreasing for two decades. However, there are still new cases each year about 3 per 100,000 people. About 35% of the new cases are in people who were born in the United States and, thus, potentially could have been helped by an effective tuberculosis vaccine. There were more than 500 deaths due to tuberculosis in the United States in 2011 (the last year for which final data are available).1

Second, the risks are actually minor or very rare. Non-suppurative adenitis following BCG vaccination usually resolves without treatment.2 HIV-exposed newborns have similar responses to BCG vaccine as do HIV-unexposed babies.3

Third, new interferon-based tuberculosis tests are not susceptible to confusion due to previous BCG vaccination. These tests have not yet been perfected, but they obviate much of the challenge of interpreting tuberculosis skin tests in BCG-exposed children.4

And, there is some value in giving BCG. The vaccine does reduce disseminated TB, might actually decrease acquisition of infection following known exposure,5 and, as noted by de Castro, might serve as a general immune stimulant. Other recent work also affirmed the reduced risk of non-tuberculous lower respiratory infection in children who had received BCG.6 Thoughts that BCG might also permanently reduce the risk of asthma have not be confirmed.7

The use of immune stimulants is not new. The “hygiene hypothesis” suggests that some inflammatory and infectious problems are less common in children who are exposed to immunity-altering poor hygiene during infancy. Silver nitrate has been put in newborns’ eyes for decades, not to kill contaminating gonococci, but, rather, to stimulate the child to develop a local inflammatory response that will prevent the gonococci from initiating an infection. It is not novel to suggest that the immune stimulation of exposure to BCG might be useful in general ways beyond whatever effect it has on subsequent tuberculosis disease.

Of course, it would be premature to alter immunization strategies simply based on de Castro’s study. It could well be that there are genetic, nutritional, and environmental variations between residents of Basque Country and the residents of the rest of Spain. Other as-yet unidentified factors might have confounded the interpretation of de Castro’s results in ways that overestimate the value of BCG.

Nonetheless, it is possible that the international community was correct in continuing with BCG vaccination. Perhaps with further data, we Americans might reconsider our resistance to the use of BCG (and advocate for its availability in the United States). Or, as suggested by an editorial accompanying de Castro’s paper, perhaps we will use these new data as further impetus to develop a new vaccine that will be more effective against tuberculosis than is BCG and that might also provide other helpful immune enhancements.8


  1. CDC. Tuberculosis – Fact Sheet. 2015., accessed July 2015.
  2. Cuello-Garcia CA, Perez-Gaxiola G, Jiminez Gutierrez C. Treating BCG-induced disease in children. Cochrane Database Syst Rev 2013;Jan 31:1:CD008300.
  3. Jones CE, Hesseling AC, Tena-Coki NG, Scriba TJ, Chegou NN, Kidd M, Wilkinson RJ, Kampmann B. The impact of HIV exposure and maternal Mycobacterium tuberculosis infection on infant immune responses to bacille Calmette-Guerin vaccination. AIDS 2015;29:155-165.
  4. Starke JR, Committee on Infectious Diseases. Interferon-gamma release assays for diagnosis of tuberculosis infection and disease in children. Pediatrics 2014;134:e1763-e1773.
  5. Roy A, Eisenhut M, Harris RJ, Rodrigues LC, Sridhar S, Habermann S, Snell L, Mangtani P, Adetifa I, Lalvani A, Abubakar I. Effect of BCG vaccination against Mycobacterium tuberculosis infection in children: Systemic review and meta-analysis. BMJ 2014;349:g4643.
  6. Hollm-Delgado MG, Stuart EA, Black RE. Acute lower respiratory infection among Bacille Calmette-Guerin (BCG)-vaccinated children. Pediatrics 2014;133:e73-e81.
  7. Linehan MF, Nurmatov U, Frank TL, Niven RM, Baxter DN, Sheikh A. Does BCG vaccination protect against childhood asthma? Final results from the Manchester Community Asthma Study retrospective cohort study and updated systematic review and meta-analysis. J Allergy Clin Immunol 2014;133:688-695.
  8. Iglesias MJ, Martin C. Nonspecific beneficial effects of BCG vaccination in high-income countries, should we extend recommendation of BCG vaccination? Clin Infect Dis 2015;60:1620-1621.