UPDATE July 14: It turns out that the BMJ study actually did NOT prove there was no increased incidence of GBS. It only showed that the increased risk, if any, was at most 2.7 times higher in those vaccinated with Pandemrix, while back in 1976 the increased rate of GBS in the six weeks following swine flu vaccination was about 8. But it sure got a lot of publicity, practically all of which incorrectly claimed there was no increased risk of GBS. According to CIDRAP:
In the accompanying editorial, four specialists with the US Centers for Disease Control and Prevention (CDC) describe the findings as being in accord with studies on the GBS risk associated with nonadjuvanted pandemic H1N1 vaccines but as puzzling in some respects.Another Irish website (JOE) today noted instead that narcolepsy was associated with the same vaccine, and that over 6 times as many cases developed in those who received Pandemrix as in those who did not. Narcolepsy is another life-threatening condition, and those who develop it do not recover.
"Overall, the results suggest that if there was an increased risk associated with the adjuvanted 2009 H1N1 vaccines studied, it was considerably smaller than that seen with the 1976 flu vaccines," they write. "Whether there was an increased risk, however, is not clear."
[UPDATE July 15: Finland's Minister of Health and Social Services has promised to help compensate victims of the swine flu vaccine program who developed narcolepsy. So far, 96 cases have been diagnosed in Finland, resulting in 70 claims for compensation. (Finland has only 5.3 million people.)]
There has been a suggestion that other neurologic conditions were associated with new narcolepsy cases, such as changes in personality*, but such changes are harder to quantify and little information has been made publicly available about them. Because falling asleep involuntarily and unexpectedly is hard to mask or ignore, it is relatively easy to get a reliable count of severe narcolepsy cases. Mild cases may be notoriously difficult to diagnose.
How much more likely are you to get narcolepsy if vaccinated? Reports of preliminary studies indicated the risk was high in children and teens but not adults. For an epidemiologist, once you know who is in a high risk goup, you can calculate how much more likely it is for members of the group to develop narcolepsy. However, if you include in your calculations all those who got the vaccine, when only those between, say, ages 10 and 20 were at increased risk, you will dilute the degree of risk if you compare all age groups en masse. In other words, the rate of new narcolepsy cases in all vaccinated people divided by the rate in all unvaccinated people will be much lower than the same comparison using only the smaller, at-risk group. The cited article does not make clear which groups were compared, but earlier data suggested the risk in adolescents was approximately tenfold higher in the vaccinated. The peak age of onset in all narcolepsy cases is 15.
Another issue to be clarified is the role of a gene that is known to predispose to narcolepsy. While a genetic predisposition is important, the gene is commonly found, but the disease itself is rare. So you cannot blame the gene for cases, although it may be a necessary prerequisite. The vast majority of people with this gene will never develop narcolepsy. Instead, narcolepsy is an autoimmune disorder, according to researchers at Stanford, where a lot of the genetic work was done. Autoimmune disorders are believed to be due to attack by one's immune system on a particular tissue type: by this reasoning, the vaccine likely triggers autoimmune destruction of a small number of cells in the hypothalamus that make a substance which helps regulate sleep, termed hypocretin.
This ABC News story gives a good idea about what it is like to live with narcolepsy, and also explains some of the science.
BTW, the baseline incidence of narcolepsy in Finland, where the association with Pandemrix was first noted, was no greater than in the USA, prior to use of vaccine for swine flu.
It is very hard to predict vaccine side effects in advance of vaccinating large numbers of people, but the most well-known (and probably the most common) serious side effects are both neurologic and autoimmune, from all vaccines. And so that is where we should cast our net when trying to learn more about the safety of new vaccines in a post-marketing environment. Surveillance for just Guillain-Barre is a road to disaster, as it is likely to miss the much bigger picture, and alone tells us nothing about the overall safety of a vaccine.
J Neurol Neurosurg Psychiatry. 2009 Jun;80(6):636-41. Epub 2009 Feb 11.
Psychological health in central hypersomnias: the French Harmony study.
Département de Neurologie, Hôpital Gui de Chauliac, 80 av Augustin Fliche, Cedex 5, Montpellier 34295, France. email@example.com
A large observational French study of central hypersomnia, including narcolepsy with cataplexy (C+), without cataplexy (C-) and idiopathic hypersomnia (IH), was conducted to clarify the relationships between the severity of the condition, psychological health and treatment response.
601 consecutive patients over 15 years of age suffering from central hypersomnia were recruited on excessive daytime sleepiness, polysomnography and Multiple Sleep Latency Test (MSLT) results. 517 (47.6% men, 52.4% women) were finally included: 82.0% C+, 13.2% C- and 4.8% IH. Face to face standardised clinical interviews plus questionnaires (Epworth Sleepiness Scale (ESS), short version Beck Depression Inventory (S-BDI), Pittsburgh Sleep Quality Index (PSQI) and 36-item Short Form Health Survey (SF-36)) were performed. Patients affected with a different diagnosis and with and without depressive symptoms were compared.
Mean ESS and body mass index were higher in C+ compared with C-/IH patients. Half of the patients (44.9%) had no depressive symptoms while 26.3% had mild, 23.2% moderate and 5.6% severe depressive symptoms. C+ patients had higher S-BDI and PSQI and lower SF-36 scores than C-/IH patients. Depressed patients had higher ESS scores than non-depressed patients, with no difference in age, gender, duration of disease or MSLT parameters. Finally, C+ patients treated with anticataplectic drugs (38.7%) had higher S-BDI and lower SF-36 scores than C+ patients treated with stimulants alone.
Our data confirmed the high frequency of depressive symptoms and the major impact of central hypersomnias on health related quality of life, especially in patients with cataplexy. We recommend a more thorough assessment of mood impairment in central hypersomnias, especially in narcolepsy-cataplexy.