A child survival tool

Print edition : November 27, 1999

The BCG vaccine is still efficacious in protecting children against some serious forms of tuberculosis, though it affords no such protection to adults.

A CONTROVERSY over the use of the BCG (bacillus calmette-guerin) vaccine in the prevention and control of tuberculosis (TB) was recently set off by a newspaper report that questioned the usefulness of the entire BCG immunisation programme for children an d called it "the Great BCG Hoax". Since the administration of the vaccine forms part of the expanded programme of immunisation (EPI) of children in the country, such a controversy could harm the programme by creating doubts in the minds of parents about the efficacy of BCG vaccination.

The news item was based on a report of the 15-year follow-up trials conducted by the Tuberculosis Research Centre (TRC), an institution in Chennai under the Indian Council of Medical Research (ICMR), in Chingleput district of Tamil Nadu, not far from Che nnai. The trials were conducted in order to evaluate the efficacy of the BCG vaccine as a public health measure in preventing pulmonary TB. The TRC study, a well-managed large-scale, double-blind randomised trial, concluded that the vaccine provided no p rotection at all in adults but a low level of overall protection in children against pulmonary TB.

The results were published in the August 1999 issue of the Indian Journal of Medical Research, an ICMR journal. The findings confirmed the results at the end of the first phase of the trials in 1979 (it lasted seven and a half years), which had be en published in the same journal in July 1980, as well as the results of other studies. The Chingleput trials were designed to study the vaccine's efficacy against only the pulmonary form of TB.

Mycobacterium tuberculosis, the tubercle bacillus, belongs to the family of organisms called mycobacteria and is the deadliest pathogen in the group afflicting humans. In recent times, it has had a resurgence in association with Acquired Immune De ficiency Syndrome (AIDS). The similar M. leprae causes leprosy. TB in humans occurs in different forms and can affect almost every system. The most common form is pulmonary TB. Extra-pulmonary forms such as meningitis and skeletal, abdominal or ge nital TB are relatively uncommon. When TB occurs in more than one system simultaneously, it is termed disseminated TB, and miliary TB is the most extreme such form. (When it was discovered, the X-ray image of the lungs of the affected persons showed mill et shaped shadows; hence the name.) In children, however, these forms are more common, and some of them, like tuberculous meningitis and miliary TB, show high mortality rates.

There is only one vaccine against all mycobacterial diseases - the BCG. It is a derived from the attenuated bovine tubercle bacillus called M. bovis. The vaccine was first used against human tuberculosis in 1921. The protective effect of the BCG v accine against TB has been well demonstrated in animal systems, but a number of controlled trials conducted since the 1930s have produced conflicting results with regard to its protective efficacy against TB in humans. Its efficacy (against pulmonary TB ) has been found to vary from zero to 80 per cent! (In Brazilian trials on TB meningitis, the efficacy is nearly 90 per cent.) From a medical point of view this is baffling, and in the face of the divergent results, it has not been possible to quantify, or even identify, the factors that govern the level of protection needed.

The variation has been debated over the last four decades, but there is as yet no consensus on the possible causes. One explanation attributes it to differences among the various strains. Another explanation is that the differences reflect the prevalence of infections with "environmental" mycobacteria. Numerous species of mycobacteria are found in water and soil, for example, and many human populations are sensitised to these. This is particularly true of India. Animal-based studies have shown that infe ctions with some of these species can confer a certain degree of protection against TB, as can BCG. Therefore, in populations where there is natural infection with "environmental" and "atypical" mycobacteria (such as M. avium), or natural immunity derived from exposure load to M. tuberculosis itself, it can mask the effect of BCG.

Another view is that different immunological mechanisms act against different stages of mycobacterial infection and disease, and that BCG is more effective in stopping bacterial dissemination in blood, than in containing the growth in specific parts of t he body. This view seems consistent with the finding that BCG generally provides a high level of protection against tuberculous meningitis even in trials in which the vaccine offered little protection against other forms of TB. Yet another explanation fo r BCG's inconsistent behaviour is geographic variations in the TB bacilli. In the context of the Chingleput trials, it has been pointed out that the common South Indian strain of M. tuberculosis is less virulent than other strains. Perhaps several factors, biological and environmental, have conspired to varying levels in the contexts of different trials, resulting in the wide variations noted.

EVEN though the use of the BCG vaccine as a public health measure in checking the pulmonary form of TB has thus been controversial for long, there is conclusive evidence from studies around the world, including India, that it is very effective in prevent ing childhood forms of TB. It is for this reason that the World Health Organisation (WHO) has recommended its continued use in immunisation programmes in TB-endemic areas. And it is for the same reason that, based on the recommendations of the ICMR, the vaccine has been retained in the EPI to immunise children under one year of age in India where TB remains widely endemic and a major cause of death.

The strategy of routine childhood vaccination has been adopted in most other nations (with the notable exceptions of the Netherlands and the United States). In fact, it is the world's most extensively used vaccine. Because of the controversy over its eff icacy, it also has been the most studied one. The news report in question reflected an improper understanding of the TRC trials and their outcome. It missed the fact that there are other severe forms of TB that affect children, against which the BCG vacc ine has been demonstrated to be effective. The TRC study was not designed to evaluate the efficacy of the vaccine against these extra-pulmonary forms of TB, however.

The Chingleput trials were started in July 1968 and the vaccine/placebo intake phase was completed in March 1971. The 15-year follow-up of the effect of the vaccine on the study population was completed in 1987. (It is, however, not clear why it took 12 years for the results to be published.) A study conducted during 1950-55 on the rural population of Madanapalle in Andhra Pradesh had shown that BCG could only protect to the level of 31 per cent. In 1963, the Government of India felt that a reliable est imate of the efficacy of BCG as a public health measure should be obtained. This was the rationale for the Chingleput trials. The study was expected to provide a conclusive answer with regard to the protective efficacy of BCG, especially in the light of the Madanapalle findings and the conflicting results of earlier trials.

SINCE their isolation between 1908 and 1918 at the Pasteur Institute in France, BCG organisms have been maintained in several laboratories around the world. It has never been cloned but several strains of BCG are used in the manufacture of the vaccine, t he most potent ones being the Paris and the Copenhagen strains. The vaccine is cheap and stable and comes in the form of freeze-dried powder which is reconstituted in injectible liquid form before use. The vaccine used in the Indian EPI is based on the C openhagen strain. The Chingleput trials, however, used both the strains among the study population so that any intrinsic strain-dependent disparity in the protective efficacy of the vaccine does not affect the results. The strains were evaluated against placebo control, and within each strain two doses (high and low) were used.

The study population included a largely rural community with some semi-urban people as well. The trial area covered 209 panchayats and nine town blocks. The entire resident population 3,66,625 (except children under one month of age) was included. Of thi s, 281,161 persons were vaccinated with BCG or placebo by random allocation in such a manner that the two strains were administered to one-third of the population each and the placebo to the remaining one-third. The study area is characterised by high in fection rates in children and high disease rates among the middle aged and the old. The trial was organised by the ICMR, in cooperation with the WHO and the Centre for Disease Control (CDC), U.S., during the first phase and subsequently by the ICMR by it self.

At the end of seven and a half years, the preliminary results showed that BCG did not protect against pulmonary TB. When the results of Phase I were published, several reasons, including possible methodological flaws, were postulated for the lack of prot ection. The follow-up period was extended to 15 years chiefly in order to test these hypotheses and settle the issue. Among the 109,873 persons classified as "uninfected" at intake phase, 560 TB cases had arisen over the 15-year period. The number of cas es from the high-dose, low-dose and placebo groups were 189, 191 and 180 respectively. The similar rates in the three groups indicate lack of efficacy, confirming the results at the end of seven and a half years. The study has also ruled out the various other hypothesised reasons for the null result.

The design of the trials was to "test the efficacy of BCG as a public health measure; that is, in cutting down transmission of TB". The disease is spread by prolonged contact with an infected individual through airborne droplets from sputum. Accordingly, the TB cases in the study were identified by means of the demonstration of the TB bacilli by sputum smear analysis as well as culture. This kind of bacteriology, however, does not detect all forms of TB, particularly extra-pulmonary forms or childhood f orms, for persons with such forms do not excrete bacilli as those with the pulmonary form do. But from the perspective of limiting transmission, it is enough to look at forms that excrete bacilli through sputum and this is rare among children. Indeed, mo re than 95 per cent of the TB cases in children are usually smear negative. Thus, owing to the nature of the design of identifying TB cases, the Chingleput trials did not investigate the impact of BCG vaccination on childhood forms of TB.

Tuberculosis patients in a government hospital in Chennai. BCG vaccination confers a high level of protection against childhood forms of TB, and provides additional protection against leprosy in endemic areas, although it is a different story when it comes to its efficacy in an adult.-N. BALAJI

However, from the limited sample of children in the trial population, the study examined the data by age group to estimate the possible protection offered by BCG against pulmonary TB in children. In children aged one month to nine years, it found a moder ate protective efficacy of 27 per cent and 21 per cent with high and low doses of BCG respectively. However, the report pointed out that these levels of protection were not statistically significant. Also, even if BCG does confer protection, it is limite d in time, and the vast majority of newly developed smear positive cases of pulmonary TB among the general population cannot be prevented by mass BCG vaccination, particularly in the developing countries, where a substantial number of the TB cases is cau sed through re-infection or reactivation.

Given this fact, the report makes this pertinent observation: "The public health value of BCG can only be in preventing childhood mortality caused by disease resulting from haemotogenous spread (through blood)". It has thus reiterated the WHO/ICMR ration ale for recommending continued use of BCG in EPI, the objective of which is to protect children from serious forms of TB and ensure childhood survival. Indeed, based on the data of Phase I itself, the immunisation policy of the Government of India had ch anged. Earlier, everybody below the age of 20 years was vaccinated with BCG. After the Chingleput trials since 1979, only infants below one year of age are vaccinated under the EPI. According to the ICMR, inclusion of BCG in EPI (whose coverage today is over 90 per cent) has proved beneficial as evidenced by the low incidence of TB meningitis and miliary TB in children in recent years.

"BCG protects children against serious forms of TB. BCG does not protect adults," emphasises Dr Thomas R. Frieden of the WHO in New Delhi. He adds: "What protects adults is the stopping of TB at the source and that is done by treatment with multi-drug th erapy based on the DOTS (direct observation short course) strategy of the WHO. In the case of TB, the clinical programme is a public health measure and treatment is prevention. BCG is a child survival tool whereas DOTS is a TB control tool." (In the cont ext of HIV/AIDS, Dr. Frieden points out that children who are simply infected with HIV (to be mere carriers of infection) can be given BCG safely. However, children who have symptoms of AIDS should not be administered the vaccine because it may lead to c omplications.)

Since the Chingleput trials did include children, it is unclear why the efficacy of BCG against childhood forms of TB was not simultaneously investigated. Since these forms do not excrete bacilli, diagnosis is difficult. For example, TB meningitis would require extraction of spinal fluid for analysis and this procedure cannot obviously be carried out on a large scale. The incidence of these forms is low and, for a sufficient sample size, a huge population would be required to conduct randomised trials, and also the trials will be expensive, points out Dr. S. P. Tripathy, former Director-General of the ICMR, who was closely associated with the Chingleput trials. "While it may be difficult to do such a study today, nobody is prepared to take the risk. Th e BCG vaccine itself is inexpensive. So why drop it?" he asks.

But there are several case-control studies now available which have demonstrated that BCG confers consistently a high level of protection against childhood forms of TB. The results of the Brazilian trials on meningeal TB are a case in point. The protecti ve efficacy found in other studies has ranged from 52 per cent to 84 per cent. The better done the study, the higher the evidence of efficacy, points out Dr. Frieden.

A case-control study carried out at the Kalawati Saran Children Hospital, New Delhi, with 37 cases and 74 controls, found the efficacy of BCG in preventing TB meningitis to be 84 per cent. A similar study at the Government Medical College, Nagpur, with 9 2 cases and 92 controls in the 0-12 age group, found the efficacy to be 86.54 per cent. Here the efficacy was greater in the 0-6 age group than in the 7-12 age group. A study carried out at the Institute of Child Health and Hospital for Children, Chennai , during 1990-1992 found the protective efficacy of BCG against TB meningitis in children in the 0-12 age group children to be 77 per cent. A British study factored out the meningeal and miliary TB cases from the results of randomised controlled trials a nd case-control trials on the effectiveness of BCG against all forms of TB. It found the efficacy against meningeal/miliary forms to be 86 per cent in randomised controlled studies and 75 per cent in case-control studies.

There is also a fringe benefit from BCG vaccination which is particularly relevant in the Indian context. The Chingleput area is also endemic to leprosy. The first phase of the TRC trials also looked at the protection that the BCG vaccine conferred again st leprosy and found its efficacy to be about 30 per cent. Evidence that the BCG vaccine may provide a higher level of protection against leprosy than against TB also comes from a 1979-89 study in the African country of Malawi, which found 50 per cent pr otection whereas there was no significant protection against pulmonary TB.

Even though the BCG vaccine is a less than ideal vaccine for the purpose because of the unpredictable nature of its protective efficacy in TB or in preventing transmission, the fact that it is very effective against childhood forms of TB in most populati ons and that it provides additional protection against leprosy in endemic areas makes it a useful prophylactic tool in countries where the diseases are prevalent. Above all, it is inexpensive, safe and stable. Nothing much is lost in continuing its use i n EPI. So why complain?