Bacillus Calmette-Guérin (or Bacille Calmette-Guérin, BCG) is a vaccine against tuberculosis that is prepared from a strain of the attenuated (weakened) live bovine tuberculosis bacillus, Mycobacterium bovis, that has lost its virulence in humans by being specially cultured in an artificial medium for years. The bacilli have retained enough strong antigenicity to become a somewhat effective vaccine for the prevention of human tuberculosis. At best, the BCG vaccine is 80% effective in preventing tuberculosis for a duration of 15 years, however, its protective effect appears to vary according to geography.
The history of BCG is tied to that of smallpox
. Jean Antoine Villemin
first recognised bovine tuberculosis in 1854 and transmitted it, and Robert Koch
first distinguished Mycobacterium bovis
from Mycobacterium tuberculosis
. After the success of vaccination
in preventing smallpox, scientists thought to find a corollary in tuberculosis
by drawing a parallel between bovine tuberculosis and cow pox
: It was hypothesised that infection with bovine tuberculosis might protect against infection with human tuberculosis. In the late 19th century, clinical trials using M. bovis
were conducted in Italy with disastrous results, because M. bovis
was found to be just as virulent as M. tuberculosis
Albert Calmette, a French bacteriologist, and his assistant and later colleague, Camille Guérin, a veterinarian, were working at the Pasteur Institute in Lille in 1908. Their work included subculturing virulent strains of the tubercule bacillus and testing different culture media. They noted that a glycerin-bile-potato mixture grew bacilli that seemed less virulent, and changed the course of their research to see if repeated subculturing would produce a strain that was attenuated to be considered for use as a vaccine. Throughout World War I, the research continued until 1919, when the now non-virulent bacilli were unable to cause tuberculosis disease in research animals. They transferred to the Paris Pasteur Institute in 1919. The BCG vaccine was first used in humans in 1921.
Public acceptance was slow and one disaster in particular did much to harm public acceptance of the vaccine. In Lubeck, 240 infants were vaccinated in the first 10 days of life; almost all developed tuberculosis and 72 infants died. It was subsequently discovered that the BCG administered had been contaminated with a virulent strain that was being stored in the same incubator, and led to legal action being taken against the manufacturers of BCG.
In 1928, BCG was adopted by the Health Committee of the League of Nations (predecessor to the WHO). However, because of opposition, it did not become widely used until after World War II. From 1945 to 1948, relief organizations (International Tuberculosis Campaign or Joint Enterprises) vaccinated over 8 million babies in eastern Europe and prevented the predicted increase of TB after a major war.
BCG is the safest and the most widely used vaccine in the world. It is very efficacious against tuberculous meningitis in the pediatric age group, but its efficacy against pulmonary tuberculosis appears to be variable. As of 2006, only a few countries do not use BCG for routine vaccination, and the USA and the Netherlands have never used it routinely. In the United States, BCG vaccination is not routinely given to adults because it is felt that having a reliable Mantoux test and being able to accurately detect active disease is more beneficial to society than vaccinating against a relatively rare (in the US) condition.
The most controversial aspect of BCG is the variable efficacy found in different clinical trials that appears to depend on geography. Clinical trials conducted in the UK have consistently shown a protective effect of 60 to 80%, but trials conducted elsewhere have shown no protective effect, and efficacy appears to fall the closer one gets to the equator.
The first large scale trial evaluating the efficacy of BCG was conducted from 1956 to 1963 and involved almost 60,000 school children who received BCG at the age of 14 or 15; this study showed an efficacy of 84% up to 5 years after immunization. However, a US Public Health Service trial of BCG in Georgia and Alabama published in 1966 showed an efficacy of only 14%, and did much to convince the US that it did not want to implement mass immunisation with BCG. A further trial conducted in South India and published in 1979 (the "Chingleput trial"), showed no protective effect.
The duration of protection of BCG is not clearly known. In those studies that have shown a protective effect, the data is inconsistent. The MRC study showed that protection waned to 59% after 15 years and to zero after 20 years; however, a study looking at native Americans immunised in the 1930s found evidence of protection even 60 years after immunisation with only a slight waning in efficacy.
BCG seems to have its greatest effect in preventing miliary TB or TB meningitis, for which reason, it is still extensively used even in countries where efficacy against pulmonary tuberculosis is negligible.
Reasons for variable efficacy
The reasons for the variable efficacy of BCG in different countries is difficult to understand. A number of possible reasons have been proposed but none have been proven.
- Background frequency of exposure to tuberculosis It has been hypothesised that in areas with high levels of background exposure to tuberculosis, every susceptible individual is already exposed prior to BCG, and that the natural immunising effect of background tuberculosis then appears to wipe out any benefit of BCG.
- Genetic variation in BCG strains There is genetic variation in the BCG strains used and this may explain the variable efficacy reported in different trials.
- Genetic variation in populations Difference in genetic make-up of different populations may explain the difference in efficacy. The Birmingham BCG trial was published in 1988. The trial was based in Birmingham, England, and examined children born to families who originated from the Indian subcontinent (where vaccine efficacy had previously been shown to be zero). The trial showed a 64% protective effect, which is very similar to the figure derived from other UK trials, thus refuting the genetic variation hypothesis.
- Interference by non-tuberculous mycobacteria Exposure to environmental mycobacteria (especially M. avium, M. marinum and M. intracellulare) results in a non-specific immune response against mycobacteria. Administering BCG to someone who already has a non-specific immune response against mycobacteria does not augment the response that is already there. BCG will therefore appear not to be efficacious, because that person already has a level of immunity and BCG is not adding to that immunity. This effect is called masking, because the effect of BCG is masked by environmental mycobacteria. There is clinical evidence for this effect from a series of studies performed in parallel in adolescent school children in the UK and Malawi. In this study, the UK school children had a low baseline cellular immunity to mycobacteria which was increased by BCG; in contrast, the Malawi school children had a high baseline cellular immunity to mycobacteria and this was not significantly increased by BCG. Whether this natural immune response is protective is not known. This hypothesis was first made by Palmer and Long. An alternative explanation is suggested by mouse studies: immunity against mycobacteria stops BCG from replicating and so stops it from producing an immune response. This is the called the blocking hypothesis.
- Interference by concurrent parasitic infection Another hypothesis is that simultaneous infection with parasites changes the immune response to BCG, making it less effective. A Th1 response is required for an effective immune response to tuberculous infection; one hypothesis is that concurrent infection with various parasites produces a simultaneous Th2-response which blunts the effect of BCG.
The main use of BCG is for vaccination against tuberculosis. It is recommended that the BCG vaccination be given intradermally by a nurse skilled in the technique. Having had a previous BCG vaccination is a cause of a false positive Mantoux test
, although a very high-grade reading is usually due to active disease.
The age and frequency that BCG is given has always varied from country to country.
- WHO BCG policy The WHO recommend that BCG be given to all children born in countries highly endemic for TB because it protects against miliary TB and TB meningitis.
- United States The US has never used mass immunisation of BCG, relying instead on the detection and treatment of latent tuberculosis.
- United Kingdom The UK introduced universal BCG immunisation in 1953 and until 2005, the UK policy was to immunise all school children at the age of 13, and all neonates born into high risk groups. BCG was also given to protect people who had been exposed to tuberculosis. The peak of tuberculosis incidence is in adolescence and early adulthood, and the evidence from the MRC trial was that efficacy lasted only 15 years at most. Styblo and Meijer argued that neonatal immunisation protected against miliary TB and other non-contagious forms of TB and not pulmonary TB which was a disease of adults, and that mass immunisation campaigns with BCG would therefore not be expected to have a significant public health impact. For these and other reasons, BCG was therefore given to time with the peak incidence of pulmonary disease. Routine immunisation with BCG was withdrawn in 2005 because of falling cost-effectiveness: whereas in 1953, 94 children would have to be immunised to prevent one case of TB, by 1988, the annual incidence of TB in the UK had fallen so much that 12,000 children would have to be immunised to prevent one case of TB.
- India India introduced BCG mass immunisation in 1948, the first non-European country to do so.
- Brazil Brazil introduced universal BCG immunization in 1967-1968, and the practice continues until the present day. According to Brazilian law, BCG is given again to professionals of the health sector and to people close to patients with tuberculosis or leprosy.
- Other countries In the UK, BCG was only ever given once (as there is no evidence of additional protection from more than one vaccination), but in some countries such as the former USSR, BCG was given regularly throughout life. In Singapore, Taiwan and Malaysia, BCG was given at birth and again at the age of 12. But in Singapore, from 2001, this policy was changed to once only at birth.
Method of administration
Except in neonates, a tuberculin skin test should always be done before administering BCG. A reactive tuberculin skin test is a contraindication to BCG. If someone with a positive tuberculin reaction is given BCG, there is a high risk of severe local inflammation and scarring. It is a common misconception that tuberculin reactors are not offered BCG because "they are already immune" and therefore do not need BCG. People found to have reactive tuberculin skin tests should be screened for active tuberculosis.
BCG is given as a single intradermal injection at the insertion of the deltoid. If BCG is accidentally given subcutaneously, then a local abscess may form (a BCG-oma) that may ulcerate and often requires treatment with antibiotics. However, it is important to note that an abscess is not always associated with incorrect administration, and it is one of the more common complications that can occur with the vaccination. Numerous medical studies on treatment of these abscesses with antibiotics have been done with varying results, but the general consensus of opinion is that once pus is aspirated and analysed, providing there are no unusual bacilli present, the abscess will generally heal spontaneously in a matter of weeks.
BCG immunization leaves a characteristic raised scar that is often used as proof of prior immunization. The scar of BCG immunization must be distinguished from that of small pox vaccination which it may resemble.
- Leprosy: BCG has a small protective effect against leprosy of around 26%, although it is not used specifically for this purpose.
- Buruli ulcer: It is possible that BCG may protect against or delay the onset of Buruli ulcer.
- Cancer Immunotherapy: BCG is useful in the treatment of superficial forms of bladder cancer. Since the late 1980s evidence has become available that instillation of BCG into the bladder is an effective form of immunotherapy in this disease. While the mechanism is unclear, it appears that a local immune reaction is mounted against the tumor. Immunotherapy with BCG prevents recurrence in up to ⅔ of cases of superficial bladder cancer. BCG also finds use for immunotherapy of colorectal cancer and for the treatment of equine sarcoid in horses.
- Diabetes, Type I: Clinical trials based on the work of Denise Faustman use BCG to induce production of TNF-α which can kill the T-cells responsible for Type 1 diabetes. Studies using mice have shown that a similar treatment results in a permanent cure for about a third of the test subjects.
- Interstitial Cystitis (IC) / Painful Bladder Syndrome (PBS): BCG has been useful in treating some people with IC and/or PBS, which are chronic inflammatory bladder problems with unknown etiology. It is instilled directly into the bladder. It is not clear how it works, but the mechanism is likely immunotherapeutic, as the chronic inflammation could be the result of an autoimmune problem.
BCG is one of the most widely used vaccines in the world, with an unparalleled safety record. BCG immunization causes pain and scarring at the site of injection. The main adverse effects are keloids
or large, ugly scars. The insertion of deltoid is most frequently used because the local complication rate is smallest when that site is used. If given subcutaneously, BCG causes a local skin infection that may spread to the regional lymph nodes causing a suppurative lymphadenitis.
If BCG is accidentally given to an immunocompromised patient (e.g., an infant with SCID), it can cause disseminated or life-threatening infection. The documented incidence of this happening is less than 1 per million immunisations given. In 2007, The WHO stopped recommending BCG for infants with HIV, even if there is a high risk of exposure to TB, because of the risk of disseminated BCG infection (which is approximately 400 per 100,000).
Other tuberculosis vaccines
- See: tuberculosis vaccines