Sunday, October 9, 2011

Infections and Schizophrenia Risk

Whew.  Quite a response over my disappointment in Wheat Belly.  And Melissa gives us an informed and reasoned review (mine was more of a visceral reaction).  It's hard for me to see the merit in a book just because the idea of wheat not being ideal for human consumption agrees with my own views.

On the interesting papers front, a couple of new articles shine more light on the relationship between infections and mental disorders.  I consider this type of thing "evolutionary psychiatry" as it brings us closer to finding the true pathology of illness, inflammation, and disease.  In addition, the typical evolutionary prescription of nutrient-rich and anti-inflammatory diets and appropriate amounts of vitamin D ought to increase resistance to infection and resilience to the inflammation and autoimmune issues that may be spurred on by such infections.

The first paper is from Denmark, Toxoplasma Infection and Later Development of Schizophrenia in Mothers, from August's American Journal of Psychiatry.  There is also an enlightening editorial in the same issue.

Toxoplasma gondii is a parasite that one can pick up from contaminated cat feces, from eating undercooked meat containing the infectious cysts or contaminated vegetables, and from being a fetus whose mother is infected.  Infection in pregnant women can cause major birth defects in offspring, and this fact is the origin of the common advice for pregnant women not to clean litter boxes.  Studies have linked infection with toxoplasma with schizophrenia since the 1950s.  More recently the association was confirmed in a 2008 study of the US military.  All these early studies were retrospective and observational, which is the weakest sort of experimental data this side of the anecdote.  Meaning folks with schizophrenia were compared with normal controls, and it turned out that people with schizophrenia have a higher rate of previous exposure to toxoplasma.

A step up from the retrospective, after the fact sort of study is the observational cohort study.  In this type of study, a group of people are followed for many years to see what develops.  This type of experiment presumably takes away what sorts of bias can be introduced by finding cases after the fact. (For example, do people with prodromal symptoms of schizophrenia engage in behavior that makes it more likely for them to be infected with toxoplasma - washing hands less, or not cooking meat as thoroughly?)  Scandinavian countries are hotbeds of these studies, as they've collected all sorts of medical data on pretty much all of their citizens for a generation now.

In Denmark, 45,609 women were followed from childbirth, when antibodies indicating prior prenatal exposure (or not) to toxoplasma in their babies were measured by heel-stick 5-10 days after being born.   All these antibodies circulating in the baby's bloodstream were made by the mother, not the baby, as a baby won't make too much in the way of these sorts of antibodies (IgG) until the immune system is a bit more mature, by 3-6 months.  Therefore mothers with babies who were positive for T gondii exposure were presumed infected themselves.  Some, but not all, of the mothers had been tested for IgG levels in the first trimester of pregnancy - these levels correlated with the newborn levels that were available for all the mothers.

Over the following years (the women were followed from 1992-2008), 80 of the mothers developed schizophrenia.  The ones whose babies had the highest IgG levels had a higher risk of developing schizophrenia than those who had babies with the lowest levels (the risk was increased by 1.73-fold, which was statistically significant - though with a population risk of approximately 1%, toxoplasma seems to increase the risk to about 1.7%).   Other meta-analysis have shown odds ratios of around 2.54-2.73 (odds ratios above two are considered a significant finding).  In the Danish study, adjustments were made for confounders (such as age, urban or rural, and other known risk factors), and women already diagnosed with schizophrenia at the beginning of the study were obviously excluded.

Why would infection with T gondii increase the risk of schizophrenia?  Active infection in the central nervous system can certainly cause huge problems (such as seizures) and inflammation.  In addition, our immune reactions to these infections can cause problems, especially if something on the infectious particle looks a bit like something in our own cells.  The classic example of this type of problem is rheumatic heart disease, most likely caused by our own antibodies attacking heart tissue after a strep infection.  It is also thought that neurological symptoms of lupus are caused by these neuro-specific auto-antibodies.  In the case of toxoplasma, it is a possibility that the anti-toxo IgG antibodies react with neural tissue and might help the immune system attack the NMDA receptors in particular.

The paper appeared in the same August, 2011 issue of the American Journal of Psychiatry.  In this study, there were four groups of folks - schizophrenic patients (none of whom were on antipsychotic medication) who were seropositive for HSV1 infection or not positive, and normal healthy controls who were also positive for HSV1 infection or not.  All these folks were followed with some cognitive testing and neuroimaging at the beginning of the study and at one year.  It was found that the schizophrenic folks with HSV1 infection had significant worsening of certain measures of cognitive functioning and shrinking of gray matter in certain regions (meaning the brain cells are dying off).  The other three groups of people didn't have these changes.

Here is what the researchers had to say about the possibilities:

There are several plausible explanations for the observed changes. In the rodent and rabbit models of CNS HSV1 exposure, latent infection and reactivation directly affected functioning through neuronal death or dysfunction. Neuronal death resulted from apoptosis. Neuronal dysfunction during reactivation and latency resulted from modulation of apoptosis and autophagy, host cell translational shutoff, oxidative damage, and/ or neurotransmitter alterations. Even with peripheral infections, HSV1 could alter neurotransmission through release of cytokines, especially chemokines, which may be elevated in HSV1-exposed individuals. Human studies support some of these observations. These processes occur throughout the life of an infected person.
In short, infection (even a smoldering latent infection without obvious active signs) in the brain or periphery can lead to all sorts of changes in the way the cell handles energy and self-destruction and general inflammatory badness.

Monitoring, modulating, and avoiding these infections seem like different ways to decrease the risk of central nervous system symptoms.


  1. Glad to see you exploring these topics, Emily! Re its relevance to evolutionary psychiatry, since the pathogens are evolving with us, and quite rapidly, if there's any branch of medicine that needs an evolutionary focus it's infectious diseases.

  2. Candace Pert postulates (in Everything You Need to Know To Feel Good) that exposure to influenza virus in the third trimester is associated with the development of schizophrenia in the offspring once they reach their early 20s when the frontal cortex is completing its development.

    Her theory is too complex to repeat here but it certainly throws more and more light on the effects of inflammation in utro. Danish (I think) studies have shown that dairy consuming mothers produce babies that can be predisposed to diabetes later on.

  3. Dear Dr Deans, in the good old literature on schizofrenia, very low prevalence in Papua New Guinea was variously attributed either to the diet or to the alleged very low number of cats hanging around among the locals. Now it seems that prevalence has skyrocketed in the region. This sounds as interesting news. Since their traditional diet has obviously remained the same over the last, say, fifty years, shall we conclude that kitties have become more common down there ? If so, should we urgently classify veterinarians and butchers as populations at a serious professional risk of going insane? Well, this can be easily tested I suppose. In science everything is plausible including the toxoplasma theory, but I am sure that poor Occam is sometimes tempted to use his razor to commit suicide, after taking a mega dose of artemisin in a last attempt to save his mind of course.

  4. Two thoughts came to mind as I read this post: 1) Paul Jaminet, and 2) medical school, when I was told that half of everything I learned would be obsolete in five years.

    Thanks for covering this topic, Emily. American Journal of Psychiatry isn't on my reading list, but Evolutionary Psychiatry is.


  5. Everything is interconnected, isn't it?

    T. Gondii could be one of the reasons Hashimoto's thyroiditis and schizophrenia are associated commorbities (e.g., pubmed 17026726, 19648193).

    Two more or less recent studies have associated T. Gondii with thyroid autoimmunity (pubmed 19811261 and 18700170).

    Not to mention gluten associations with both schizophrenia and Hashimoto's.

  6. Steve - Paul Jaminet has the irritating propensity to be on target ;-)… I do think infections have a lot to do with hastening or worsening inflammation, particularly in nerves and in brain where infections can hide, perhaps, better than in other areas. I don't know that aggressively treating latent infections will do much good, which is Paul's tack… but it seemed to work for him. I'm totally on board with making the immune system happy using food, sleep, play, and maybe probiotics and the like.

    Bolalbi - I do think mental illness is increasing in general, but, man, is it hard to definitively prove. Awareness is fairly modern, diagnosis every modern. Always hard to know if the (definitely) increasing prevalences in studies are real or artifacts of awareness and changing definitions of psychopathology. I'm more interested in the changing morphology of the diseases (reading classical descriptions of depression vs. modern ones, or schizophrenia at the beginning of the 20th century vs, now) - though medicine throws a HUGE monkey wrench into those sorts of comparisons too.

    Mario - there you go again, reminding me how I'm neglecting the thyroid. I wish I had 30 hours in the day. I really do. Also, that someone would give me a scholarship so I could take a year to do some intense reading…

  7. Cavegirl - studies show that exposure to influenza B (but not A) is associated with worsening suicide risk and psychosis in bipolar disorder. Infections are big, I have no doubt.

  8. Emily, do you have a link for these studies, I had an appalling flu which resulted in pneumonia and pleurisy and can date my worsening mental health from that period. Fascinating. I must add that I am now well since a change of diet! But we are now talking 18 years on.

  9. Cavegirl - there are many if you search pubmed - here is a more recent one:

  10. Thank you, a connection I'd never made before, seems in many ways I was subject to the perfect storm ... high carb vegetarian diet, a family history of manic depression and then a bad case of the flu. The most interesting thing is that by removing the highly inflammatory diet I've recovered, an interesting thought train makes me wonder if by allowing my 'system' to attain a low inflammatory status I have enabled the re-establishment of an efficient immune response which has rectified the imbalances/dysfunctions.

  11. Dear Dr Deans, what is the confounder?

  12. Dear Emily,

    Very good post. How is a latent brain or CNS infection diagnosed? By symptoms or by some other measure like a spinal tap? I have read Paul Jaminet's work in this area and wonder if there is a way to better diagnose the presence of such an infection. I've been suffering from a mysterious illness for a few months that causes daily headaches, ear pressure/pain, and sleep difficulties. The illness onset one week after my cat was sick with a very serious upper respitory infection. An MRI and CT Scan were done and nothing showed up. Extensive blood work shows some sort of viral infection, but beyond that nothing shows conclusively. I am wondering if there are other diagnostic measures that might be able to help. It's been three months.


Tired of receiving spam comments! Sorry, no new comments on the blog

Note: Only a member of this blog may post a comment.