Thursday, July 29, 2010

Low Cholesterol and Suicide 2

In my last post on the link between low cholesterol and suicide, I made note of some general trends between low cholesterol, suicide (particularly violent suicide), accidents, and violence, and raised some questions about the safety of cholesterol-lowering drugs. I didn't find any researched link between statin therapy and suicide, though one study showed that a statin reduced the ability of a certain serotonin receptor to do its job (linked below). My takeaway point from the post was that, hey, cholesterol is important and needed in the brain. Obliterating the ability of our liver to make cholesterol may have some untoward mental health side effects.

Since then, I've kept an eye out for more information, and a few interesting snippets have come up. Current Psychiatry has a decent article this month, "Cholesterol, mood, and vascular health. Untangling the relationship."

Some interesting facts from the article:

1) 1/4 of the body's free cholesterol is found in the central nervous system
2) Depleting cholesterol impairs the function of the serotonin 1A receptor and the serotonin 7 receptor, and reduces the ability of the membrane serotonin transporter to do its thing. (Serotonin is made within nerve cells and needs to be transported outside into the synapse between the nerve cells to work. If the transporter isn't functioning, we have a Big Problem).
3)Cholesterol is also needed for forming a nerve synapse (also Important) and making myelin.
4) Cholesterol may be involved in GABA and NMDA receptor signaling, opioid signaling, and the transport of excitatory amino acids.

Just to be crystal clear - low serotonin is associated with violent suicide, impulsive acts, hostility, and aggression. We need plenty of cholesterol in the brain to have all our serotonin machinery work properly. Low cholesterol is also associated with suicide and violence. If you have low cholesterol, of course it does not mean you will be suicidal. Suicide is, fortunately, rare, and will have multiple predisposing causes.

So the paragraph above, with its caveat, brings up an interesting and actionable hypothetical question - does lowering cholesterol with medication predispose you to suicide or violence? The first cholesterol-lowering drugs were not statins. And an early analysis of the primary prevention trials of the non-statins showed a doubling of the risk of violent death or suicide. Oops. (I also linked the J-LIT trial in my previous post, which showed a 3-fold increase in suicide or accidents with statin therapy, though the increase was not statistically significant).

A later case-controlled study showed that statin users had a lower risk of depression than patients on non-statin lipid-lowering drugs. The LIPID study followed 1130 patients on pravastatin for 4 years, and found no changes in (self-reported) anger, impulsivity, anxiety, or depression. Pravastatin doesn't cross the blood-brain barrier very well. Simvastatin, a very commonly used statin, crosses it quite readily - but why this would be important may be interesting. HMG Co-A reductase inhibitors (statins) do most of their work in the liver, after all. But it turns out we have HMG Co-A reductase all sorts of places. These researchers found it in Chinese hamster ovary cells. And in these cells, administration of a statin reduced the ability of the serotonin IA receptor to work. Getting rid of the statin restored the serotonin IA receptor function.

But there's another complication in examining the literature for statin side effects. Some studies excluded patients with psychiatric problems (1). And due to the ability of statins to cause birth defects, many trials have excluded any women of childbearing age. Just something to keep in mind.

We are left with... well, a clinical trial is apparently underway to study the effects if pravastatin, simvastatin, or placebo on mood, sleep and aggression. We still don't know if low cholesterol causes suicide and aggression, or if it is a biomarker of depression. I'm convinced high cholesterol is just a biomarker for heart disease, after all, rather than a cause. Thus the whole question of why treat high cholesterol at all (though the magical anti-inflammatory statin effect may help younger men. With known heart disease.)

My brain needs cholesterol! So does yours.


  1. cholesterol AND saturated fat...the stuff the brain needs. I like your list of like minded types. Please let me suggest Gary Taubes's "Good Calories, Bad Calories" and "Life w/o Bread" by Wolfgang Lutz. Lutz is an MD, an internist.

  2. Great posts Emily. Interesting and thinking back to my time as a 20-something when I was most depressed, most anxious, most suicidal... I was semi-vegetarian (I ate a small amount of skinless chicken and fish) (actually I was a hopeless vegetarian - it never felt right so I gave up on it) and being a skinny cyclist, I ate mostly wholegrains (lots of bread & shredded wheat) and tried to minimise my fat intake from all sources. My cholesterol levels, by NZHF standards were fantastic (low total & low LDL). There is a lot in your posts that resonate with me. Though I do also need to factor in being on Roaccutane for a couple of courses & there are some question marks over how that affects people. I had shocking acne, no doubt due to a shocking pro-inflammatory diet. The Roaccutane took away the acne, but it didn't change my diet - it was still very inflammatory when I look back. Perhaps Roaccutane becomes an innocent bystander when it comes to the psychological effects it has? It is just something that many people take when they have a lot of systemic inflammation going on?

    1. Hi Dr. Dean, look at . There is lots of articles on cholesterol and violence, and especially "road rage". Especially see:

  3. Thanks for writing about this important issue, Emily.

    Statins and SSRIs have both been candidates for addition to the public water supply. [I just remembered...they're in the sewers already.]

    We need more info on long-term side effects.


  4. SSS - love Gary Taubes! That book by Lutz looks interesting and I will have to check it out. Thanks!

    Jamie - That's a very interesting point about accutane. None of my patients on antidepressants were depressed on it - antidepressants are anti-inflammatory in the CNS. But I'm going to look into that more. Was reading an interesting little article about the teenaged Masai warriors who were considered unpredictable and dangerous. There are obviously cultural factors at play too - but I wonder how impulsive and hormoneal are teenagers and young adults regardless of diet? The Masai are known for their gorgeous skin - all that saturated fat!

    Steve - congrats on the second printing of your book! I think both SSRIs and statins are lousy candidates for the water supply - though if you look at my last post, a little lithium seems to mellow out El Paso :). One morsel for thought is that the older cholesterol lowering drugs seem to have more mental health effects and more other mortality-problematic effects, though they are less successful at lowering the actual cholesterol number. It makes me wonder if the anti-inflammatory statin is cooling off the CNS too, and it might even be anti depressant if it weren't dismantling the serotonin machinery at the same time. Total speculation. Agree we need more data, and we oughtn't do ridiculous things like treat a c reactive protein level with a statin.

  5. Cholesterol lowering drugs = shoot the messenger.

  6. Hi Emily, please define low total cholesterol and low LDL. Like Mr. Scott, I'm prone to anxiety / nervousness, and have made a concerted effort to increase my total cholesterol while preserving a "healthy" LDL/HDL ratio. Anyhow, phrases like "low total cholesterol" are often quite meaningless. Can you please give it a concrete sense? Thank you, best, Kevin

  7. Hi Kevin. In the J-LIT trial, the group with the issues (p=.09) was the TC <160, which tended to correspond with LDL<80. In the Paris prospective study (, the group with low cholesterol was just "below average" cholesterol, which in American numbers was TC <185 or so. In a meta-analysis of studies between 1965 and 1995 of low cholesterol and violence - including suicide - ( the "low cholesterol" group was also defined as TC<160. It is important to know that the association is not strong enough to be an independent risk factor for suicide. Though suicide is rare, and somewhat difficult to study and predict.

  8. Jamie - this paper is LONG and I only skimmed it, but seems to suggest accutane may have its own mechanism reducing neuronal plasticity in the hippocampus and frontal cortex separate from other dietary inflammation. Though if all rats were on their dismal inflammatory rat chow and the humans were on their dismal SAD, maybe it only accentuates an inflammatory diet problem.

  9. Oh, Kevin, one more thing - at least in the J-LIT data, the cholesterol groups of total cholesterol between 160 and 280 the risk of accidents/suicides were all about the same, then above 280 the risk went up again (p=.08) This suggests to me (though this is a bit of speculation, as the study enrolled some 4000 people, and the accidents and suicides were rare, so the P values for the middle groups were all quite high) that the brain is happy with a certain amount of cholesterol, and as long as it has enough, it is more or less happy, up until levels >280 when all cause and cardiac mortality definitely goes up for everyone, maybe even women :)

  10. Thanks for that Emily! I certainly don't think Roaccutane therapy did me too many favours (aside from clearing my skin), but there are so many other factors involved in terms of diet, etc. I was a chronic cardio 60kg (20kg lighter than I am now) cyclist living off shredded wheat, pasta, bread, cookies, soft drink, etc whilst undertaking my first degree.

    All of this would have had an effect on both my skin and my head (though the dermatologist rattled off the old line that diet has nothing to do with skin health). So hard to separate drug from diet. And even harder to separate out typical early 20-something angst from anything something more pathological!

  11. Interesting, thank you. Before I revised my diet, my LDL was 33, no joke. I now eat quite a bit of pastured meat and eggs, and I'm interested to see how my total cholesterol profile changes, when I do bloodwork later this winter.

  12. Not sure if you saw Dr. Harris' post on statins and cholesterol, but if not, you should definitely take a look, as he talks a little about some of the mechanisms behind statins and why they cause problems.

    I suspect that the cholesterol and suicide link is more about correlation than it is about causation. It's quite difficult to lower cholesterol below normal levels, and takes a very unhealthy diet to do so. I suspect that it is some of the other dietary factors that are causing the suicide issues.

  13. Has anyone else noticed a decrease in psychological side effects of accutane when it is given in combination with physiologically appropriate doses of Vitamin D? It could be that D tends to increase total cholesterol and balances out accutane's tendency to decrease it or it could be one of the many other actions of D at work. Interesting anyway.


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

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