Sri Lankan deaths following Pentavalent vaccine: Acceptable collateral damage?

Jacob Puliyel, KB Saxena, Debabar Banerji, Imrana Qadeer, NJ Kurian, Ritu Priya , Mira Shiva and Gopal Dabade.

Sri Lankan deaths following Pentavalent vaccine:
Acceptable collateral damage?

As a group of pediatricians, health care activists, teachers in Public Health and bureaucrats who have championed the cause of universal immunization in India all our working lives, we were taken aback at being described as ‘anti-vaccine’, in the print version of the BMJ (1). We plead for the right to respond in our defense, in the print version.

India is a country where 50% of the population, (mostly poor living in remote rural areas) do not receive the 6 basic vaccines in the EPI, namely BCG, DPT, OPV and measles vaccine. The incremental cost of complete immunization is less than US $1 (Rupees 30) per child. Every Rupee we can mobilize, we need to invest to ensure that all babies in India get this basic right. The push to include newer vaccines into the routine immunization schedule must be viewed in this context. The ‘pentavalent vaccine’ that is being canvassed, on its own, will increase cost of DPT immunization twenty-fold. It stands to reason that these new vaccines will be taken up mostly by the urban privileged and there will be even less available for the rural poor.

The thrust for including Hib vaccine in India is based on 2 arguments. The first is that there is anecdotal evidence of the existence of Hib disease and Hib meningitis in India and that Hib meningitis may lead to long term morbidity. The second argument is that the well-to-do parents sometimes buy Hib vaccine in the open market to vaccinate their children. The Government of India must therefore provide it free for the poor, on the grounds of equity and fair play (2).

The problem with the anecdotal reports is that they do not specify the size of the universe from which the samples are drawn and public health policies cannot be based on these figures without a denominator. The many systematic surveys done to look at the magnitude of the problem of Hib disease in India have nearly always shown that the incidence of Hib disease is much lower than what was projected for India. Most of these studies have been funded by the WHO and these have been reviewed recently in an open access journal (3).

Over and above this, is the fact that ‘probe studies’ and ‘probe-like studies’ done in Asia have all shown the vaccine does not result in significant reduction in disease burden compared to placebo (4). The poor need a modicum of equity in terms of food for survival, potable water and basic sanitation but they are certainly not hankering after the useless vaccines the rich may be taking. In this situation, to foist this vaccine program on them in the name of equity is at once insensitive and cruel. It merely siphons-off the limited funds available for the poor, and in its place provides them a service they do not need and which does them no good.

Crucially, there is another reason why the authors decided to approach the Law Courts in India to block introduction of the Pentavalent vaccine and this is the safety issue. The pentavalent vaccine combines DPT, Hib and Hepatitis B. A Cochrane meta-analysis has found that the combination vaccine is not as effective as the vaccines administered separately (5). As such, the combined vaccine is not used widely in the West – in areas which have a good system for reporting adverse events.

Vaccines are generally given to healthy children and therefore the tolerance in the public for vaccine-related adverse-reactions is low. If a vaccine results in death it is a very serious concern. A vaccine may have been used successfully in millions and with no adverse effects, yet in the next patient it may cause a fatal anaphylaxis reaction. The adverse effect following immunization (AEFI) investigation is set up to study if the reaction in the given child was related to the vaccine or is unrelated. The AEFI report does not comment on the likelihood of reaction if the vaccine is given to other children in the future. If the same reaction occurs twice it is defined as a cluster.

Pentavalent vaccine was introduced in the national immunization programme in Sri Lanka in January 2008 but after several thousand doses were administered, it was withdrawn in April 2008 because of 25 serious adverse reactions that included 5 deaths. A WHO expert panel investigated the adverse effects and deaths and in its report classified 3 deaths (identified as cases D1, D3, and D6) as ‘Unlikely’ (to have been caused by the vaccine). Pentavalent vaccine was then introduced in national immunization programme of Bhutan in July 2009. Within 2 months, after 8 deaths, the vaccine was withdrawn in that country (6). At the behest of the WHO, the vaccination programme was restarted in Sri Lanka some 4 months ago and it is claimed that there is no change in the death rate in Sri Lanka – as if deaths in 3 infants from AEFI will alter the countries mortality statistics.

In response to our petition in the High Court in Delhi, the detailed AEFI report on the deaths in Sri Lanka was made available to the Court and thereby, in the public domain. As only a summary of this report is available on the internet (7), we have scanned the report and put it on the net and it can be accessed freely (8). The report quotes an aide-memoire on the standard WHO classification on AEFI (9).

The standard WHO classification of AEFI is best understood in the form of an algorithm. The first question is whether the adverse events have a plausible time relationship to vaccine administration. All such reactions are classified in one of three categories: ‘Very likely/Certain’, ‘Probable’ or ‘Possible’.

If the time of the adverse event and the time of vaccine administration make a causal connection improbable, it is classified as ‘Unlikely to be related’. If the time of the event and the time of vaccine administration make a causal connection incompatible, it is to be classified as ‘Unrelated’

For adverse events that have a plausible time relationship to vaccine administration, the next level of the algorithm is used to distinguish Probable from Possible. If the death cannot conclusively be attributable to another cause, it is classified as ‘Probable related’. If the death can be attributable to another cause then the association with vaccine is said to be ‘Possible’

The report presented to the Delhi High Court shows that the experts on AEFI in Sri Lanka deleted the categories Probable and Possible from the standard classification. All adverse events which could not be classified as Very likely/Certain were classified as Unlikely. Using this new classification, they declared that 3 of the Sri Lanka reactions were ‘unlikely’ to be related to vaccine ‘although it could not be conclusively attributable to another cause’. It is evident from the discussion above that the three deaths (D1, D3, D6) would have been classified as ‘Very likely/certain AEFI’ or ‘Probable AEFI’ using the standard WHO classification.

When the news of this change in classification caught the attention of the press, a WHO spokesperson defended the change saying that the WHO used ‘independent experts’ and they were free to make up their own classification. Furthermore the WHO spokesperson clarified that even if the standard classification were used; the three deaths would not be classed as probable or possible as there was ‘non-conclusive evidence of other potentially attributable causes for the adverse events’. The potentially attributable conditions unearthed by the experts were the following: one had malnutrition (a condition which defines a large proportion of the population in rural areas), the second case had autopsy findings of milk aspiration (often seen as a terminal event in many cases of sudden death from any cause) and the third case had ‘autopsy findings suggestive of Reye’s syndrome’. The expert body did not suggest that these findings qualified as plausible explanations for the deaths and the temporal relation of the deaths to vaccination has not been disputed. Interestingly the experts say that they discussed the possibility that the vaccine may have unmasked an underlying condition. One is left wondering whether they considered malnutrition, milk in the trachea, or Reye ’s syndrome as conditions that would have remained ‘masked’ had it not been for the vaccine.

For a document produced to a court of law, the document is not even complete. The names of the experts and their declaration of conflict of interests were left out. We have learnt that at least one of experts has previously been accused of not declaring his conflict of interest arising from research funded by pharmaceutical companies including GlaxoSmithKline.

Classification of AEFI as ‘Certain/ Very likely’, often needs de-challenge or re-challenge evidence. This is not possible when the initial reaction results in death. The question the larger public (outside of India and Asia) needs to answer is whether they will permit this new classification to replace the standard WHO classification of AEFI. If this is allowed, deaths which occur as reaction to vaccine will nearly always be classified as ‘unlikely’ because de-challenge and re-challenge will not be possible post-mortem.

The opprobrium of the BMJ stems from our raising these issues. Using the same yard-stick the Editor of the BMJ can be castigated as being ‘anti-vaccine’ for exposing the N1H1 vaccine scam (10).

Dr KB Saxena
Former Union Health Secretary, Government of India.

Dr Debabar Banerji
Professor Emeritus, Centre of Social Medicine and Community Health, Jawaharlal Nehru University Delhi.

Dr Imrana Qadeer
Professor (Retired)
Centre of Social Medicine and Community Health, Jawaharlal Nehru University Delhi.

Dr NJ Kurian
Former advisor, Union Ministry of Finance, Government of India.

Dr Ritu Priya
Professor of Community Health, Jawaharlal Nehru University Delhi.

Dr Mira Shiva
Coordinator Initiative for Health and Equity in Society/ Member, Third World Network/ Co Convener All India Drug Action Network (AIDAN).

Dr Jacob Puliyel
Head of Paediatrics, St Stephens Hospital, Delhi.

Dr Gopal Dabade
President, Drug Action Forum- Karnataka/ Co-convener, All India Drug Action Network.

1. Ganapati Mudur. Antivaccine lobby resists introduction of Hib vaccine in India, BMJ 2010;340:c3508

2. ICMR Expert Group. Minutes of the Expert group meeting on Hepatitis B and Hib vaccines. accessed on 4/7/10

3. Lone Z, Puliyel JP. Introducing pentavalent vaccine in the EPI in India: A counsel for caution. Indian J Med Res 2010; 132:1-3

4. Puliyel JM, Mathew JL, Priya R. Incomplete reporting of research in press releases: Et tu, WHO? Indian J Med Res 2010;131:588-9

5. Bar-On ES, Goldberg E, Fraser A, Vidal L, Hellmann S, Leibovici L. Combined DTP-HBV-HIB vaccine versus separately administered DTP-HBV and HIB vaccines for primary prevention of diphtheria, tetanus, pertussis, hepatitis B and Haemophilus influenzae B (HIB). Cochrane Database Systematic Rev 2009, Issue 3. Art. No.: CD005530.

6. Pelden S. Pentavalent vaccine suspended. Kuensel Online 27. 28 October 2009. Available at : accessed on 4/7/2010.

7. WHO Investigation of adverse events following immunization with liquid pentavalent vaccine in Sri Lanka. accessed on 4/7/10
8. WHO. Report of an ad-hoc WHO expert panel to review reports of serious AEFI following administration of pentavalent and other vaccines in Sri Lanka 2008 available at:
9. WHO. Adverse events following immunization (AEFI): causality assessment. accessed on 4/7/2010
10. Godlee F. Conflicts of interest and pandemic flu. BMJ 2010;340:c2047