BCG vaccination in developing countriesBMJ 2010; 340 doi: http://dx.doi.org/10.1136/bmj.c809 (Published 15 March 2010) Cite this as: BMJ 2010;340:c809
- Frank Shann, professor of critical care medicine
BCG vaccine is given to more than 100 million infants each year, making it one of the most widely used vaccines.1 In the linked randomised controlled trial (doi:10.1136/bmj.c671), Roth and colleagues assess whether revaccinating children at 19 months reduces overall child mortality.2
A single dose of BCG provides useful protection against systemic mycobacterial infections, such as tuberculous meningitis, miliary tuberculosis, and leprosy.1 3 It is less effective against pulmonary tuberculosis—effectiveness varies greatly between trials, but geographical latitude and study validity explained 66% of this variance in one meta-analysis.4 BCG was less effective near the equator, in lower quality trials, and in rural areas, but effectiveness was not influenced by age at vaccination or the strain of BCG used.4 5 BCG may be of less benefit to people living in tropical or rural areas because they have greater contact with environmental mycobacteria. These organisms may have heterologous (non-specific) immunological effects that provide some protection against tuberculosis but also reduce the effectiveness of BCG.5
Because BCG provides useful protection against tuberculous meningitis and miliary tuberculosis in children, the World Health Organization recommends that it be given as soon as possible after birth to all infants living in areas where tuberculosis is highly endemic.1 WHO also recommends that BCG be given only once, largely because of the lack of evidence that revaccination is beneficial.2 3 6 7 8 Two randomised controlled trials, a case-control study, and a cohort study have suggested that revaccination with BCG does not reduce the incidence of pulmonary tuberculosis.3 6 7 8 However, all these studies were performed in countries near the equator, where BCG is likely to have little effect on pulmonary tuberculosis.5 The two randomised trials and the case-control study were confined to people with at least one BCG scar, so we do not know whether a second dose of BCG is helpful in people with no scar, although the presence of a scar and the reaction to purified protein derivative are probably not a reliable guide to immunity to tuberculosis.6
Revaccination with BCG may still be beneficial even if it does not provide extra protection against pulmonary tuberculosis. Attenuated Mycobacterium bovis (BCG) has potent heterologous effects on the immune response to many organisms other than M tuberculosis, including other mycobacteria (M avium intracellulare, M leprae, M microti, M ulcerans), other bacteria (Brucella, Coxiella, Listeria, Salmonella), protozoa (Babesia, Leishmania, Plasmodium, Toxoplasma), and vaccinia virus.9 These heterologous effects have an important influence on human infections other than tuberculosis. Randomised trials of a single dose of BCG in the United States and the United Kingdom published in the 1940s and 1950s showed that it substantially reduced mortality even after excluding deaths from tuberculosis and accidents.10
Two randomised trials have shown that revaccination with BCG provides protection against non-tuberculous infections.3 11 One trial of 121 020 people in Malawi showed that revaccination increased protection against leprosy by about 50% compared with a single BCG vaccination, even though it did not protect against pulmonary tuberculosis.3 In the other trial, 41 302 children born in Algeria between 1935 and 1947 were given BCG orally at birth and then alternately allocated to receive no further doses or three extra doses at 1, 3, and 7 years of age; revaccination reduced total mortality between 1 and 11 years of age by 27% (95% confidence interval 22% to 31%).2 11
Although BCG reduces mortality from diseases other than tuberculosis, diphtheria-tetanus-pertussis (DTP) vaccine may increase mortality from infections other than diphtheria, tetanus, and pertussis in some circumstances.2 12 To assess whether BCG can counteract these harmful non-specific effects of DTP, Roth and colleagues randomised children to receive either no vaccine or a second dose of BCG at 19 months of age, one month after they should have received a booster dose of DTP (DTP4).2 All the children had been given BCG in infancy. No significant difference in mortality was seen between the groups (hazard ratio 1.20, 0.77 to 1.89).
Unfortunately, the trial was inconclusive for several reasons. Firstly, as discussed in the paper, monitoring of the progress of the trial was not ideal. Secondly, mortality during the trial was lower than expected, which is a common finding when free medical care is provided during a study in an area with high mortality. Thirdly, and most importantly, major interactions occurred between BCG, vitamin A, and DTP. Sixty per cent of the children had not received the DTP4 booster at the time of enrolment, and importantly many of these children were given DTP after they had been vaccinated with BCG at 19 months of age. Children who received BCG had a lower mortality than the controls if they had received DTP4 before enrolment (hazard ratio 0.36, 0.13 to 0.99) and a higher mortality if they had not received DTP4 before enrolment (hazard ratio 1.78, 1.04 to 3.04). Mortality was 0.36 per 100 person years if DTP4 had been given before BCG revaccination, 1.02 in controls who were not revaccinated with BCG (mortality was not affected by DTP4 status at enrolment), and 1.83 if DTP4 had not been given before BCG revaccination. These findings suggest that it may be beneficial to give BCG after a booster dose of DTP, but that it may be harmful to revaccinate with BCG if this is followed by a dose of DTP. Children were not randomised to receive DTP4 before or after BCG revaccination in this study, so the apparent interaction between DTP and BCG may have been the result of confounding, although this seems unlikely. Similar positive and negative heterologous interactions are common in the immunological literature.12
The study adds to the existing evidence that BCG, DTP, and measles vaccines have important heterologous interactions with each other as well as with the pathogens that cause fatal pneumonia in children in developing countries, and that the order in which vaccines are given is important.2 12 This raises the exciting prospect that we may be able to reduce child mortality substantially by revising the current immunisation schedule, and it suggests that new vaccines should be tested for their effects on total mortality and not just their effects on the target diseases.
Cite this as: BMJ 2009;340:c809
Competing interests: The author has completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: (1) No financial support for the submitted work from anyone other than his employer; (2) No financial relationships with commercial entities that might have an interest in the submitted work; (3) No spouse, partner, or children with relationships with commercial entities that might have an interest in the submitted work; (4) No non-financial interests that may be relevant to the submitted work.
Provenance and peer review: Commissioned; not externally peer reviewed.