Reply

Indian Pediatrics 1999;36: 944-946

Drs. Kaur, Faridi and Ramachandran have done an excellent job of reviewing the pros and cons of early BCG vaccination based on many published papers. The authors highlight several problems in documenting or determining the protective efficacy of BCG vaccination in individual children. The absence of scar formation is not a reliable evidence of the lack of cell mediated immunity. A small dimension of delayed hypersensitivity skin test response or even negative response to PPD is also not a reliable evidence of the lack of CMI. So far we agree. The authors assert that, "therefore, there is no need to repeat BCG inoculation in babies who do not develop BCG scar, as advocated in the guidelines of IAP, 1996". Here I disagree and abide by the stand taken by the Immunization Committee.

In a baby given BCG but without scar, it is possible that the vaccination was unsuccessful due to defective vaccine or faulty procedures. The absence of scar does not help us to differentiate between the baby with no immune response and the one with immune response in the absence of a scar. One could use the PPD test to try to ascertain the immune status. If PPD reaction is of reasonable dimensions (5 mm or more), one could attribute it to BCG response. Even if PPD response was negative, there could have been some immunological response to BCG. Thus, if BCG was given, the baby developed no scar, and the PPD re-action was negative, the above logic would suggest that the baby could have developed immune response. That is correct, but not always. Suppose the BCG vaccine contained only dead bacilli; there would be no scar and no PPD response. In other words, it is simple enough to repeat the BCG vaccination rather than remain in the dark about the response. We only assume that the BCG vaccination might have been unsuccessful in the asbence of scar when we recommend repeat BCG. The opposite assumption is more risky than the assumption that leads to one more attempt. Scar is a better surrogate of successful immunization than the history of having given BCG.

The authors have correctly pointed out many problems with the PPD test, the measurment of response and also the inter-pretation of the result. These are all very true. What they have not addressed is the fact that even the development of CMI is not necessari-ly a reliable surrogate of protection. A lot of skepticism about the need or value of BCG vaccination had arisen due to this observation. As there is good evidence that BCG vaccination does offer good protection against those forms of tuberculosis that devleop largely in the absence of previous experience with Mycobacteria (CMI being the simplest surrogate), and not against those forms of disease that develop in spite of, or actually due to, previous experience and immune response (such as CMI), the Academy recommends BCG vaccination routinely. The Academy does not promote BCG for public health, that is to reduce the transmission or incidence of infection or disease. We have seen a decline in the frequency with which we admit children for tuberculous meningitis but the incidence of adult tuberculosis has not fallen. We believe that this dichotomy is partly at least due to the successful BCG program and the unsuccessful TB control program. It may also be partly due to better nutrition of young children.

The authors reviewed the literature and concluded that "it is not known whether the vaccine is as immunogenic or protective in infants who suffer from prematurity and/or IUGR as compared to appropriate for gestational age babies". What surprises me is that instead of using their analytical skill and knowledge to suggest how best to go about giving BCG vaccination to premature and IUGR babies, they ask for "evidence based scientific reasons" for deciding when to give BCG in situations of low gestational age and low birth weight. As Chairman of the Committee on Immunization I would have very much welcomed the opinions of the authors themselves in this matter. Since evidence is said to be unavailable, I shall attempt a logical approach to develop guidelines for practical purposes rather than for debate.

Why is BCG given to the neonate? The authors seem to be under the impression that BCG is given to the neonate because the neonate is vulnerable to infection during the early weeks of life. They state that "if BCG is given at 6 weeks, to minimize hospital visits, the baby remains vulnerable to acquire tuberculosis for longer period, as 8-12 weeks are normally required for development of immunity". The neonate and young infant are vulnerable but not exposed under ordinary circumstances. Only when the mother has cavitory tuberculosis is the young infant exposed and clear guidelines have been formulated to manage this situation(1). Otherwise, the cumulative burden of infection (measured by the surrogate of PPD reaction) and of disease (primary tuberculosis) are of the order of about 2% by age 5 years. So then, why neonatal BCG?

In the case of neonatal OPV, I had made the original recommendation since we were seeing poliomyelitis in young infants before they could be given sufficient number of doses of OPV. But, I had to show that OPV would be immunogenic in the newborn and that was my contribution to show that the neonate could be immunized as successfully as any older infant. In the case of Hepatitis B immunization, the neonate needs protection either from vertical or from early horizontal transmission from a virus carrier mother and it has been shown that the vaccine is immunogenic in the neonate even in spite of maternal antibody or administered immune globulin. In India, on the average, 4-5% mothers are virus carriers. In the case of BCG, the need for protection is not urgent, but the vaccine takes in the neonate as well as in the older infant. Since the vaccine can be given at this age (unlike DPT, measles vaccine, etc.), the convenience of easy contact with the health worker is the main reason for the recommendation of neonatal BCG. Therefore, in situations of pre- maturity or IUGR, BCG may be delayed until the infant is stabilized, is growing well and weight gain is reasonable. Instead of evidence based answers, I shall answer the questions using common sense and the above background information, as follows.

In general, any low birth weight baby may be given BCG as the baby is ready to be sent home. In terms of minimum gestational age, about 38 weeks may be satisfactory provided the baby has reasonable weight (such as about 1800 g or more) and the baby is ready for going home. With less gestational age, we can wait for chronological age and weight to catch up to those necessary for sending the baby home. Babies weighing less than 1200 g are usually not sent home until they have gained some weight and thus the chronological age would be accordingly considered. When they are ready to be sent home, the chronological age would have been 38 weeks equivalent and weight at least 1800 g. When the gestational age was 26-32 weeks, babies post conception are kept in the hospital along with the mothers, until they become stable and growing well. When they are ready to be sent home, that is when weight is about 1800 g and age has reached about 38 weeks equivalent, they could be given BCG. Some neonatologists might consider sending a stable baby home at weight as low as 1500 g but perhaps for such babies the priority is not to give early BCG but to wait until the weight has caught up to about 1800 g. In all these suggestions I have tried to balance the need to wait until the baby's immune system is robust enough, for which purpose it is better to delay vaccination and not to miss the opportunity of the baby being available for vaccination just before sending home.

The authors seem to be under the impression that once BCG is given, the other vaccines must be suitably delayed. They state: "If BCG/OPV zero dose is delayed by 3-4 weeks (for babies with very low weight), further vaccination gets progressively delayed, thus keeping the baby at risk to acquire all the dangerous infections for a longer period". Irrespective of when BCG was given any baby who is doing well can be given DPT about 6-8 weeks after birth. The reason for 6-8 weeks interval is for the maternal antibody to decline, and for no other reason. DPT and OPV may be given concurrent with BCG or at any convenient interval after (or even before if need be) giving BCG.

T. Jacob John,

2/91 E2 Kamalakshipuram,

Vellore, Tamil Nadu 632 002, India.

Reference

1. Consensus statement of IAP working group. Treatment of childhood tuberculosis. Indian Pediatr 1997; 34: 1093-1096.