Skeptical Raptor's Blog hunting pseudoscience in the internet jungle

Pandemic flu vaccine and narcolepsy–an analysis


H1N1 vaccineToday, the British Medical Journal published a retrospective study, by Elizabeth Miller, that analyzes the risk of narcolepsy in children and adolescents in England who received the 2009 A/H1N1 pandemic flu vaccine (Pandemrix) from October 2009 through mid-2010. This study followed up on the observations seen in Finland and other countries that there was some increased risk of narcolepsy in children a few months after receiving the vaccine.

As background, narcolepsy is a chronic neurological disorder caused by the brain’s inability to regulate sleep-wake cycles normally. Individuals with narcolepsy often experience daytime sleep patterns, but the disorder should not be confused with insomnia. It is not caused by a mental illness or psychological problems. It is most likely affected by a number of genetic mutations and abnormalities that affect specific biologic factors in the brain, combined with an environmental trigger during the brain’s development, such as a virus. It may also be a result of an autoimmune disorder. At this time, there is no cure for narcolepsy, but it can be successfully treated by medications and other therapies.

Roughly 70% of individuals who have narcolepsy have a comorbidity of cataplexy,  a sudden and transient episode of loss of muscle tone, often triggered by strong emotions. 

The initial observation about a potential link between narcolepsy (and cataplexy) and the 2009 pandemic flu vaccine was made in Finland and other Scandinavian countries, but it was not observed in other countries. This finding was perplexing to researchers, so they considered the possibility of another confounding environmental variable that was not uncovered. It was originally assumed that it might be related to some genetic susceptibility to narcolepsy, although other countries, such as Canada and the United Kingdom, had the same risk of narcolepsy as Scandinavian countries.

The new article from Miller uncovers the risk of narcolepsy in children who received the vaccine in England, though it is an extremely rare occurrence. This gives more evidence that populations in Europe that have similar genetic predisposition to narcolepsy may have the same risk of developing the disorder after this specific vaccination.

The key data from this study are:

  • The researchers extracted information from 160,400 individuals, and found that 14,400 could be at risk of developing narcolepsy. They found that 75 met the criteria of narcolepsy (according to international classification of sleep disorders).
  • They then obtained the vaccination history of the 75 study cases. Of those, 60 were never vaccinated, 4 were vaccinated after initial symptoms, and 11 vaccinated before first symptoms.
  • The most important statistic: the estimated risk attributable to the pandemic flu vaccine (in this study only) is between 1 in 57,500 to 1 in 52,000 vaccine doses, a significantly lower risk than was found in the Finland study (which was about 1 in 16,000 vaccine doses). Miller suggests that the either the population in Finland is more susceptible to narcolepsy, or the risk increased because of the higher vaccine use in Finland’s adolescents than in the UK.

The authors, as is typical with these type of retrospective studies, self-critique their work (as all good scientists should). Their points were:

  • Some data from sleep centers (where narcolepsy is usually diagnosed) was not included because there were no positive cases of adolescent narcolepsy. It is possible that some patients may have been referred to those centers after this study stopped adding cases. If more unvaccinated children had been diagnosed with narcolepsy, it could have reduced the risk attributable to the vaccine. On the other hand, if more patients who were vaccinated during that time period, it could have increased the risk.
  • Because there is no data to determine how long narcolepsy can be diagnosed after vaccination (if, in fact, that is a reasonable hypothesis), the authors chose an arbitrary, but reasonable, date to exclude patients, which is the month when the H1N1 vaccine-narcolepsy link became a public news item.
  • They did not identify another confounding variable (which there may be, just not identified), which could make these results nothing more than random.
  • The authors admit to the possibility of ascertainment bias (known usually as sampling bias), in which they collect data that may oversample one group or another. The risk may have been overestimated, because of more rapid referral of vaccinated patients to sleep centers, which would then create a biased statistical measurement.

The evidence presented by both the Miller study and the earlier Finnish study do provide a substantial, evidence-based link between the pandemic vaccines (used in Europe) and narcolepsy (and the associated cataplexy). The statistical analysis, vaccine attributable risk, appears to give a valid relative risk number.

Even though the actual number of cases is extremely low (less than 100 cases), it is still statistically significant, though the statistical variance is sufficient that we could conclude that this study has shown us randomness, nothing more. Miller concludes that “further study” is warranted, and supports the change in indication for use of the Pandemrix vaccine. Furthermore, a recently published study concludes that there was no increase in narcolepsy diagnoses in countries that did not receive the vaccine, or had low vaccine coverage in the affected age group.

So what does this all mean:

  • The article fails to answer two important questions: do we know for certain that these children would not have eventually contracted the disorder (independent of the vaccine), and have we missed some confounding variable? This concerns me when this study becomes overblown by the vaccine denialists (just a prediction).
  • In the United States, the infant mortality from the H1N1 pandemic in 2009-10 resulted in 344 deaths. Though it is impossible to compare the statistics between the narcolepsy study and pediatric mortality and morbidity in the US, the rate of hospitalization in children for H1N1 was around 83 per 100,000, a risk far above the narcolepsy risk. The risk of death of children who caught H1N1 was around 3 per 100,000, still way above the potential risk of narcolepsy.
  • It is possible that H1N1 itself may cause narcolepsy. A study in China indicated that there was no correlation between narcolepsy and H1N1 vaccination, but it was correlated with H1N1 itself.
  • Pandemrix contains a proprietary adjuvant called AS03 (see Pandemrix product information, pdf, page 26), which contains a compound called squalene. Immunologic adjuvants are substances, administered in conjunction with a vaccine, that stimulate the immune system and increase the response to the vaccine. Squalene is not itself an adjuvant, but it has been used in conjunction with surfactants in certain adjuvant formulations. Some have pointed to this compound as a potential cause of autoimmune disorders, but a large meta-analysis have not uncovered any link (until the more current potential link between the vaccine and narcolepsy). The FDA has never approved squalene in vaccines in the USA, so if there is a link, the risk doesn’t exist in the USA. Furthermore, like a lot of other ingredients in vaccines, squalene is manufactured by the liver of every human being, it is found in numerous foods, medications, and “natural” health supplements. Squalene itself is extracted from fish oil

I am also reluctant to give excessive weight to primary retrospective studies that have not been repeated and analyzed by others. For example, I am not a genius about statistics, and it is possible that the data, if analyzed in a different manner could lead to a wholly different conclusion. Although the authors appear to want us to follow the precautionary principle, that it’s better to be safe than sorry, the evidence of benefit of the H1N1 vaccines is so solid, and the evidence of narcolepsy seems to rely upon statistical averages without much analysis of a causal link that relies upon known and verifiable biological processes. I’ll wait until this data is repeated and reviewed by others before jumping on the H1N1 vaccine causes narcolepsy bandwagon.

It is clear to me that there are no simple answers here, and there might be a link between narcolepsy and this particular vaccination for H1N1 influenza. At the time of the 2009-10 H1N1 Pandemic, and given the mortality rate from the H1N1 flu, vaccinating children with the H1N1 vaccine appears to have provided a benefit far in excess of the risk. And no, washing your hands alone was not going to keep your children from being infected by this flu. The only evidence based way to keep from getting the flu is vaccination. Or living in isolation from the rest of the world.

Vaccines Save Lives.

Visit the Science-based Vaccine Search Engine.

 

Key citations:

Comments (1)
Powered by WordPress 3.9.1
Don't forget to subscribe to this blog through any the services in the right sidebar.