Friday, April 20, 2012

Obesity, Systemic Inflammation, and Bipolar Disorder

Mental illness is not invented, and psychiatry is not all in your head.  Today an exploration of the link between bipolar disorder and obesity.  And there is a link, despite the tremendous confounding fact that most of the medicines used to treat bipolar disorder definitely cause obesity, researchers separate out the medication component (and also look at reports from history, prior to medications even being available) to find a strong correlation.  But why, and how?

Cool song: Howler, Back of Your Neck. (right click to open in new tab)

Well, I don't have any definitive answers for you today as to the ultimate cause of bipolar disorder and why it is linked with obesity.  What we do have is a little recent observational study, Increased Levels of Adipokines in Bipolar Disorder that can serve as a handy review of some of our obesity hormones.  

Everyone, I'm assuming, has heard of insulin, and probably leptin.  Leptin is an adipokine, a chemical mediator produced by the fat tissue that can help tell the body how much fat is on board.  Most of you will know that leptin tends to be increased in obese folks, suggesting that obesity may be a function of leptin resistance.  There are other adipokines besides leptin, however, including resistin and and adiponectin.  Adiponectin is of particular interest, because it is known to be anti-inflammatory, the body makes quite a bit of it relative to other hormones, and its levels inversely correlate with obesity in adults.  Leptin is thought to be pro-inflammatory, and may be responsible for the activation of an inflammatory cytokine, TNF-alpha.  Adiponectin will help to decrease the production of TNF-alpha.  So, in general, an obese person will have lower than normal adiponectin and higher than normal leptin, with associated increases in inflammation.

In the study, 30 bipolar (type I) patients were compared with 30 matched (age, BMI, education level, and gender) controls.  All the bipolar patients were on medication (a huge weakness of the study).  Compared to the matched controls, the bipolar patients had higher levels of adiponectin, leptin, and one of the receptors for TNF-alpha.  The kind of medication (as the patients were on several different classes) and medical co-morbities did not correlate with the hormone levels.  Since, again, adiponectin is generally lower in obese individuals, it is interesting that the bipolar patients had higher levels than the controls of the same age, gender, and BMI.  

One previous study of obese bipolar patients showed the same increase in adiponectin.  A study of non-obese depressed bipolar patients had adiponectin levels lower than the controls.  Leptin has been more vigorously studied, and the levels are elevated in some and not in others. The only other study of previously manic but now normal mood patients had the same leptin levels (elevated).  

It's interesting, though we don't quite know what it means.  We get a hint of a large puzzle of a systemic illness, affecting mood, sleep, appetite, thought, immunity, and the adipose tissue.  The connecting process of inflammation is not controversial.  What causes the inflammation… that we don't yet know, and it is likely a confusing combination of factors.  There are a number of papers exploring, for example, peripheral biomarkers and different moods in bipolar disorder, helping us to figure out other pieces of the puzzle.  No one has all the answers. 


  1. I saw a 41 year old man, who was bipolar and committed suicide by hanging, in my practice as a forensic pathologist yesterday. Not only was he morbidly obese, but he also had metastatic carcinoma of the rectum and cardiomegaly due to hypertension. I see combos like this frequently, and I always have wondered if hyperinsulinemia, which is likely the cause of these non-mental illnesses, also plays a role in the mental illnesses.

    1. There is a correlation between metabolic syndrome and bipolar disorder. Also between cholesterol dropping and manic or psychotic episodes (cholesterol levels return to baseline when euthymic). I don't know what it all means.

  2. I read the post but am still confused. How did any of these studies isolate the pharmaceutical component (i.e. the obesity "effect" caused by medicine effect seems to be there, even though there is 'supposedly' no link between medications and hormone levels). We give meds to attenuate Bipolar disorder, patients get fat, they get better--and your argument is obesity hormones(inflammation) are the source of the problem? Very confusing ;P

    1. The meds definitely cause obesity. But prior to meds being available (back in the 40s and 50s), bipolar I patients as a group were known to be more obese than the general population (there are, of course, plenty of slender people with bipolar disorder also).

      In my practice experience the meds that cause massive weight gain (olanzapine and clozaril, for example) tend to work in a fairly consistent manner, whereas meds like ziprasidone (which tends to cause weight loss, particularly in people who are obese) are very tricky.

      I think it is pretty well understood that obesity in the case of insulin resistance/type II diabetes is a way to protect the body from the excess circulating energy by stuffing it into the fat cells. Those who don't get obese will be more like the Diabetes type I.5, with sudden adult onset ketoacidosis. Here we have folks with bipolar disorder who have high BMIs and high adiponectin -- it's like the body/brain doesn't see the adiponectin. What does that mean? Not sure, but it is interesting.

  3. Whenever someone tells me of their psychiatric disorder, they always tell me that it's "genetic". I'm believing more and more that it's down to a bad diet and inflammation.. although part of me thinks genetics affects it a little bit.. but mainly diet.

    1. It's definitely genetic. Probably diet. Sleep, stress, traumatic exposures. Lots of things.

  4. I think we can probably say that we have a good handle on inflammation, but I like your modesty and lack of woo.

    The nagging "it's all in your mind" psychologist voice in my head is saying "obese people are often ridiculed and develop low self esteem!" and that is true. A confounder.

    However I'm in the "inflammation of the brain plays a role" camp. And I may just be able to pull of few studies to suggest it.

    Adipose tissue secretes adipokines depending on various factors. One of them is when adipocytes are under stress, and die, particularly large ones associated with obesity, there is an inflammatory reaction, and it manifests itself in the rest of the body. Preventing damage and death of the cells can greatly reduce this in mice. If they are just fed a lot of concentrated blueberry they don't exhibit the metabolic abnormalities as much TNF-a gets slashed almost down to the control!

    And they increase adiponectin levels

    And it just so happens that blueberries can reduce depression somewhat, amongst other things Maybe due to its actions directly inside the brain too.

    And maybe it would be even more pronounced in clinically depressed obese subjects. That would be an interesting trial. Depression/obesity blueberry extravaganza!

    Curcumin too, although in animal models

    And it does the same, reducing adipose tissue inflammation

    The case for the "inflammation counts" hypothesis is strengthening. That's telling!


    1. Thanks, Stabby, yes, there are very thorough reviews of inflammation and depression and bipolar disorder and schizophrenia. Psychoneuroimmunoligy is the future of psychiatry. To the extent that we can affect inflammation via diet and lifestyle and other evolutionary means (say helminths and pseudocommensals -- needs to be studied more with respect to psychiatric dx before we know much) with few side effects… well, that seems like a good thing.

    2. It was my understanding that chronic stress actually attenuates inflammation? Acute stressors increase the immune response, then it comes down to baseline, and then below baseline if the effect is still in play. Who needs helminths when you can liberally apply some stress?

    3. When I say stress, I mean oxidative stress and endoplasmic reticulum stress of the adipocytes, causing them to die and explode all sorts of cytokines all over.

      And cortisol may be anti-inflammatory but I certainly wouldn't use chronic stress as an anti-inflammatory! Chronic stress is definitely a contributor to all sorts of pathologies, there are so many aspects to it. Elevated adrenaline levels can damage the arteries, cortisol can interfere with thyroid hormone function which can make people depressed. And eventually chronic stress produces resistance to cortsol, causing fatigue. I would rather have anti-inflammatory eicosanoids and other molecules and nutrients that modulate the immune system, keeping it in check. Cortisol is the last anti-inflammatory I would want.

      Emily: I'll agree there. I can think of ways to reduce systemic inflammation, and I'm confident that if I suddenly had a CRP of 4 mg/dl I could get it way down (mine's 0.3 mg/dl and I cheat frequently now) but medicine needs a proven protocol for getting someone from the high risk zone to the low risk zone without giving them a handful of COX inhibitors. That would be very helpful. And that's the conservative doctor's bah humbug, that there is no standardized protocol that is reliable.

  5. Emily, I agree that mental illness is not invented--wait what's wrong with invention?--how about artificial? I also think that oppositional defiant disorder (ODD) is essentially bullshit. I'm happy to believe in a middle ground somewhere, like here.

  6. Everything to do with the mind-body whatever is complicated. I'm still fascinated by the circadian rhythm connection with bipolar disorder and the action of lithium on the SCN. Biochemisty is awesome.

  7. There are so many conditions connected through a systemic inflammation! When I addressed an inflammation with my diet, long list of illnesses disappeared, so I could start thinking that frequent urinary tract infections were associated with migraines or asthma could be connecter with a leg edema.

  8. The elevated adiponectin is quite interesting, will have to read how this is regulated as adiponectin increases whole insulin sensitivity and centrally will promote a thrifty metabolic state; adiponectin is a marker of starvation. There may be an adiponectin link to BP.

    When I investigated the relationship of leptin levels and mood states, it was found that untreated severely mentally ill people had abnormally low leptin. Experiments in animals suggest lack of leptin allows for the onset of mental illness, as pretreatment with leptin protects animals from declining into stress induced depression. Leptin is important to brain health by regulating cortisol activity and promoting BDNF, and normalizing energy use in the body.

    As stated in my journal while taking leptin I cracked out a bit, but I think this was related to a combination of stress + reproductive hormone changes, + prior years of significant leptin insufficiency...and, ironically, many years of experimental treatment with leptin seems to have profoundly benefited my mental condition.

    Though leptin is inflammatory, evidence I have found supports a unanimous conclusion that this stuff is awesome for mental health, possibly curative/preventative, and seriously mentally ill people will typically have an abnormally decreased leptin.

  9. Regarding bipolar I and metabolic syndrome/diabetes, I have found research that suggests hyperglycemia, centrally, may trigger mania in vulnerable individuals. It may be not that bipolar people are obese and diabetic, but rather that mania is a symptom of hyperglycemia for some individuals who otherwise have diabetes. The relationship would then appear superficially that diabetes causes bipolar when in reality some manic individuals only had that condition because of central hyperglycemia.

  10. I have been following your posts on bipolar and the connection to metabolic factors with great interest.

    I have a sister who has been diagnosed bipolar and BPD (and a variety of other mental alphabet soup disorders at different times).

    Another sister is diagnosed ADHD with apparent sub-bipolar symptoms at different times (manic phases and crashes/depressed phases).

    Both have had problems with weight and have cycled through crash dieting and being overweight.

    I myself have never been "fat" per se -- I've always exercised regularly and lifted weights, even to the point of being a "gym rat" -- but in the last two years I had started to get a noticeable paunch. I am 32 now. I couldn't figure this result out because I was still exercising quite frequently (3+ days a week), living a walking-centric lifestyle, and eating generally "healthy" (in a standard sense).

    I also tend towards slight depression, but never medicated for it. In high school I was prescribed prozac but never took it. I chalked it more up to my parents' divorce and took St. John's Wort to cope.

    A couple years ago, sister B (the ADHD one) self-diagnosed as gluten sensitive. I didn't think much of it, but have noticed since then she has had much better control over her weight. She also became much more productive and successful in her life.

    Last fall I started a paleo/primal diet and began rapidly losing weight. I went from 5'10" 182lbs to 160-161 lbs today. I am much leaner yet just as strong, if not stronger, so there was no muscle wasting

    I also noticed that chronic joint pains (wrists and feet in particular) and lower back problems (a degenerated L4-l5, and as of last August, a herniated L5-S1), as well as mysterious persistent skin rashes, all started to clear up or have totally disappeared.

    The "break" in connecting the dots for me came in response to a few episodes of deviating from paleo, and eating a significant amount of wheat or rice. In response to both, skin rashes re-appeared, and to some extent, joint pains flared up again.

    As a result of this I came to the conclusion that I had gluten/grain protein-induced psoriatic arthritis, with symptoms going back at least ten years. I even developed the fingernail "ridges" characteristic of the condition in the last year. This of course cannot be officially explained, because according to NIH, psoriatic arthritis has "no known cause and no known cure".

    But now it is quite clear to me that a gradually-worsening grain protein intolerance led to my complex disorder.

    Connecting the dots back to the topic of this post -- depression -- I also noticed a huge mood lift when I made the dietary shift. I do not know if this is a direct effect, or a secondary effect of my much greater apparent energy on a low-carb, essentially grain-free diet.

    So it seems, at least in my case, we have poorly-tolerated grain proteins -> inflammation -> skin, joint, and mood disorders. This is a virtually-certain connection.

    Now, what are the odds that the same does not apply to both siblings as well, one of which has an extremely-stubborn, drug-resistant bipolar?

    Very low, I think.

    I am telling them all to try a grain-free diet (preferably low carb as well) to see what happens.

    I think there is really something big here regarding latent inflammation caused by dietary and possibly other lifestyle factors.

    I just wanted you to hear yet another anecdote (and a few indirect ones) supporting this line of inquiry.

    I will report back my own and my family's results.

    Keep up the good work.

  11. I was diagnosed last month as bipolar II and have been diagnosed with ADHD in the past (though now it's not quite clear if it was really ADHD or just symptoms of bipolar). I was on a strict paleo diet last fall and winter and felt great; I didn't know I was bipolar, but my usual seasonal depression and periodic mysterious difficulty sleeping (i.e. hypomanic episodes) were greatly reduced, along with the constant brain fog and all manner of other little aches and pains and complaints. I fell off the paleo wagon (eating tons of grains again) earlier this year and my mental symptoms worsened, which was what finally prompted me to seek out the eventual bipolar diagnosis. I've gone back to a grain-free diet right now, and I'm really hoping it will help, as I have yet to start on any meds and am going through the worst episode I've ever had. This blog is providing some fascinating reading while I wait for results =)

    Oh, also relevant to this particular post, I am now obese, and my weight has slowly but steadily gone up over the years as my symptoms have worsened. It doesn't seem to matter what I try (low-cal, low-carb, exercise of all sorts, etc), I can't lose the weight. I never had trouble staying at a healthy weight when I was younger and in a better mental state.

  12. I'm a 64 year old male with treatment resistant ultradian cycling/mixed state bipolar. Recently, because of chronically low body temps and loss of response to thyroid meds, I've been trying Matt Stone's recommendations that come from 'Eat for Heat', which is lots of carbs, salt/salty snacks & sugar, all of which seem to reduce cortisol.(Ray Peat, PhD makes similar recommendations). I do feel warmer and my libido is reappearing but, aside from adding a few more pounds to my small frame, my mood, which, popped up like a jack-in-the-box when my wife and I got to Florida for the winter, has now miserably crashed and all I'm feeling is hopeless.

    Then I began revisiting the inflammation/mood connection, focusing on the huge amount of wheat I've been eating, everyday, since starting this carb loading diet in the beginning of December, when we got to Florida. My sleep has also been poor(non-restorative)for several years and I'm weaning off an eight year addiction to clonazepam(Klonopin).

    Maybe diet, for me, won't be the whole solution, but if I don't take some positive action, I will crash and burn.


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

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