Monday, November 28, 2011

Depression - Beyond the Chemical Imbalance (Part 1)

Today we go back to the basics of depression.  Borodin's Nocturne (right click to open in new tab).

I would say there are three main theories held by the general public about the causes of depression:

1) Bootstrap theory:  you are a lazy good-for-nothing who just needs to snap out of it and get up and get yourself better.

2) Trauma theory: too much stress, death, trauma, etc.

3) The chemical imbalance:  You have an SSRI deficiency and your serotonin needs to be regulated (see this memorable old zoloft commercial)

Of course, I don't subscribe to any of these theories entirely, though there are elements to each of them that hold a kernel of truth - my belief and one that is largely supported by the literature is that stress and genetic susceptibility leads to depressive symptoms, which are mediated by inflammatory means in the brain.  And certainly if one is capable, getting up and getting out and exercising and eating right can be very helpful, but sometimes asking a depressed person to wake up early and exercise is like asking someone with a broken ankle to go for a run.  The frontal lobe isn't firing on all cylinders.  There's no motivation, no zazz.

The scientifically minded probably are most familiar with theory number three.  In medical terms, the "chemical imbalance" theory is called the "monoamine hypothesis" of depression.  The monoamine theory is (I would say) largely accepted by doctors of a certain age (even psychiatrists), but it holds about as much water as the carbohydrate-insulin theory of obesity.  Back in the day there was a medication for blood pressure called reserpine.  Among other things it depletes the brain of serotonin, and does indeed tend to cause depression (it is rarely used nowadays).

The first antidepressant, a drug used to treat tuberculosis, was found serendipitously.  One of its actions was to change the concentrations of monoamines (such as serotonin and norepinephrine) in the synapse between nerves.   And thus, the monoamine hypothesis of depression was born along with a billion dollar antidepressant industry.  All the antidepressants affect the monoamines one way or another, and they work… if you are lucky, often with side effects, and maybe they protect your brain during one episode of depression, but they don't seem to protect you from the next episode if you go off the medicine when you feel better (talk therapy when compared to medicines seems to have more long term benefit, not surprisingly).

Along the way, the monoamine theory picked up a bunch of other diseases (called the affective spectrum disorders) including major depressive disorder's anxious twin, generalized anxiety disorder, migraines, irritable bowel syndrome, bipolar disorders, social phobia, PTSD, OCD fibromyalgia, and chronic fatigue syndrome among others (1).  All of these diseases have been shown to respond (somewhat) to three or more different classes of antidepressant medication.

Problem is, when you measure serotonin in depressed people, the levels are often all over the map.  In fact, low serotonin doesn't really correlate with depression very well at all, though low serotonin in the central nervous system does correlate with suicide, violence, and insomnia.   Brain researchers quickly figured out that the monoamine hypothesis has some pretty big holes, and the mechanism of antidepressants is not about increasing serotonin and other monoamines in the synapse but rather changing the efficiency with which monoamine signals are transmitted.

Instead, the current literature-supported theory of the brain pathology of depression and the other affective spectrum disorders leads us to two things going awry - the immune system (inflammation) and mitochondrial dysfunction.

How messy is the study of depression?  Consider these facts - if we look at the modern criteria, the classic unipolar major depression is a smallish subset of the whole.  31-62% of people with depression have symptoms of "atypical depression" (leaden feelings in the arms and legs, increased appetite, increased sleep, as opposed to the classic weight loss and insomnia).  64-72% of those with atypical depression meet criteria for bipolar spectrum disorders.  Depressive disorders are often comorbid with ADHD, anxiety, and substance abuse disorders.  You can see if we try to study a group of patients with "major depressive disorder" by criteria that represents the typical clinical outpatient, we will get a mix of people with various complicating neuropsych problems, and any studies of so-called "pure" major depressive disorders where other problems are excluded (which is typical for pharmaceutical studies) will not necessarily be generalizable to the actual population.

Add in frequent comorbid medical conditions, and you have a whole soup of pathology.  92% of depressed inpatients have pain, typically headaches or muscle aches.  Irritable bowel and migraines are often found, along with metabolic syndrome, pre-diabetes and diabetes, and obesity.

However, rather than be taken aback by the complexities, theories of mitochondrial dysfunction and inflammation can scoop up the entire variable pathology (which makes these theories very pleasing to me).

So let's start with mitochondria.  As we know, these are the energy factories of the cells, and their primary mission is to make the cellular equivalent of gasoline, ATP.  Problems with the mitochondria tend to show up as symptoms with the most energetically hungry cells of the muscle and nerves.   Nutritionally, CoQ10, carnitine, B-vitamin, and selenium deficiencies can also cause mitochondrial dysfunction directly.  Mitochondria desperately need these micronutrients to do their work efficiently.

Symptoms of mitochondrial dysfunction can be non-specific, but the cognitive symptoms are very similar to those found in depression, including impairments in attention and executive function and memory.  Tellingly, in families with known genetic problems with mitochondria, the symptoms worsen around times of stress - overwork, fasting, over-exercise, and environmental temperature extremes.  Children with mitochondrial disorders are more likely to be depressed than control children, and among adults, folks with known mitochondrial disorders are more likely to have depression, chronic fatigue, major depressive disorders, bipolar disorder, and panic disorder than the general population.

But all of that is the typical chicken and egg clinical stuff.  Maybe people with genetic mitochondrial problems have a lot of stress, and are thus more depressed.  What's the biochemical evidence for mitochondrial dysfunction in major depressive disorders (and bipolar disorder)?  Autopsies show all sorts of interesting problems with mitochondrial proteins, unusual mitochondrial DNA mutations, and poor mitochondrial complex activity (2).  ATP production rates and respiratory chain enzyme ratios seem to be decreased in the muscle of biopsied patients with major depressive disorder and pain.  In fact, several studies have shown that patients with a high degree of somatic complaints (typically muscle aches) have much lower ATP production than average in the muscle.  In chronic fatigue patients, some similar abnormalities have been found.

And finally, in some mouse models of mitochondrial dysfunction, the mice have bipolar-like symptoms and altered levels and turnover of the monoamines.  The researchers worked out that the mitochondrial dysfunction was the cause of the monoamine depletion, leading to the mouse mood disorders.

So mitochondrial problems (which can be brought about genetically, but also by micronutrient deficiencies) can cause oxidative stress, and eventually lead to nerve damage and psychiatric symptoms.  More on the specifics of this pathology and the role of inflammation in the next post.


  1. How can one be tested for mitochondrial dysfunctions?

  2. Very nice post and I can´t wait to read the part 2.

    Thyroid function could be an interesting subtopic of the chemical imbalance theory.

    Have you read Peter Kramer´s book Against depression? For the general public it is a good read on the biological basis of depression. He used to blog at PT like you too.

    It is surprising how the treatment of HBP changed in so little time. Let´s hope the same goes for depression.

  3. from my experience, i would support the assumtions you make in this post (immune system, micronutrient deficiency, and mitochondrial dysfunction). most of this can be caused by gluten (or simply food) sensitivities and/or chronic infections. pro tip: if you suspect someone to have m. disf., give them NADH (sublingual 10mg) and see how they react. i checked out NADH several years ago when i was pretty healthy, and it did exactly nothing for me. now, in the recent 2-3 years during my "CFM/ME" episode, it pretty much was a life safer for me (and still is). there are other supplements that are extremely effectice in directly boosting the m., like R-LA, GTE, resveratrol and others. but of course they don't fix the underlying causes. in my case i susbect a fucked-up digestive flora (all this started after heavy duty oral and IV antibiotics), chronic viral (throat) infection, and food sensitivities to mainly gluten, dairy and eggs. if i take all this into account and take a state of the art stack of (around 25 different) supplements in the morning, i'm back to 100% overall performance (=symptom free and fully functional)! no strict AI paleo diet and no supplements = no able to work. it's that simple for me.

  4. I can accept that affective disorders, some forms of depression and bipolar, are mitochondrial in origin. However, I think it stands to reason that the mitochondrial dysfunction is exerting its mood disorder effects by altering either the level, or sensitivity, to monoamines in the brain. If this was not true, we could not, for example, induce symptoms almost exactly like mania by using meth. A person high on a strong stimulant looks manic, and people who are manic are accused of using drugs. Both are treated acutely with strong central inhibitory medications if they are unmanageable/dangerous.

    I've often suspected that my obesity, glucose processing issues are clearly biologically related to my depression; they get better together, they get worse together. It seems another symptom. I believe depression (or perhaps other mood disorders like bipolar) is an emotional/cognitive version of diabetes and obesity. I agree with this.

    However, I disagree that "mitochondrial in origin" means "not monoamine". I think that mitochondrial level errors in energy use cause mood disorders explictly by screwing up monoamine systems, either sensitivity or level. These are not mutually exclusive. The mitochondrial problems cause the symptoms of mood disorder by leading to abnormal monoamine system function. This would be analogous to mitochondrial problems causing the symptom of obesity by leading to gross insulin hypersecretion when the person tries to fuel their body on glucose primarily. Yes, I know this is controversial, but lets assume for a second this hypothesis is correct for arguments sake. Whatever the mechanism for obesity, most would agree that the body does not use energy normally.

    The argument I am making is that the same root cause (e.g. mitochondria) can affect many different systems (Mitochondrial defect umbrellas both monoamines -> depression/mania/anxiety in brain; as well as insulin signalling -> diabetes/obesity in body). That there is an even deeper root cause than the system dysfunction does not invalidate the disorder in the system, and it does not mean we should NOT try to medicate the system independent of the root cause.

    This means, for a person who has both obesity and depression, they may require a low carbohydrate/insulin suppressive eating program so as to stabilize blood glucose / weight, and they also may need drugs to boost / inhibit their monoamines so as to stabilize their mood. Each disordered system must be treated independently even if the root cause is something else. We can't touch broken mitochondria, much. The best we can do is control the disorder in individual systems.

    I disagree with the idea that monoamines do not produce mood symptoms. Prolactin levels and blink rate, and metabolites of NTs, all of these change in mood states, and these speak of NT action in the CNS. Furthermore, depletion of monoamines can reverse theraputic drugs, and taking drugs of abuse can help dial back mood states. People self medicate with opiates, and stimulants, and benzos, and alcohol, because they affect NTs which are abnormal in mood states.

    I suppose I am saying that, I do very much agree that mitochondrial defects are probably at root of many mood disorders, but this does not mean monoamine systems are not involved, and it does not mean the energy defects are not affecting monoamine systems.

  5. Since you raised Chronic Fatigue Syndrome and a person has commented on it, I thought I'd share my opinion. First, Chronic Fatigue Syndrome should be distinguished from chronic fatigue. The latter improves after lowering stress levels and getting several night's good sleep. If it persists after making those changes, my understanding is that it may be a sign of clinical depression.

    Chronic Fatigue Syndrome appears to be several discrete disorders or illnesses that are currently being categorized under this umbrella term of CFS. This is why it's hard to find two people with a CFS diagnosis who have the same symptoms. It's also why some people improve with a particular treatment, while others do not. (I've tried NADH, as qualia suggests, and it gives me a wired fatigue—like jet lag on steroids. Some people with a CFS diagnosis improve on anti-depressants at a low dose. I've tried four different ones and just get that same wired fatigue.) As for clinical depression, I believe there are two or three studies showing that people diagnosed with CFS were not clinically depressed before they got sick, but some become so after months or years of illness.

    Until this umbrella term "CFS" (which I consider to be relatively meaningless) is broken down into the separate illnesses or disorders it represents, little progress will be made on the treatment front. For example, I've yet to find anyone who presents with the same symptoms I have: my fatigue is not a tired fatigue—I don't get sleepy—but I feel like I have the flu all the time so I refer to it as a "sickly fatigue."

    I do believe (and people disagree with me here—and even get hostile toward me which is why I don't usually talk about it) that one of the subsets of CFS is a form of somatization where the body breaks down because of mental factors in a person's life (sorry, I'm not a therapist so I may not have expressed this using the proper terms of art).

    I say this because I have no other explanation for why some people seem to recover after undergoing "mental" therapy, either in the form of talk therapy or in the form of some drastic behavioral change in their lives, such as following one of the promised cures that are popular in the U.K. where people are told to never use the term ME/CFS again, get up off the bed and force themselves to resume the lifestyle of a healthy person. This is a recipe for disaster for most people (and, unfortunately, empties their pocket books), but I know several people who have completely recovered by doing this. (I like to say, "I don't know what they had, but it's not what I have!")

    I hope this is helpful to your readers, Emily. Thanks for the interesting post.

  6. It seems the original version of Procol Harum's Whiter Shade of Pale started with "and we skipped the mitochondria".

    Because turning to mitochondria seems an irresistable temptation these days. The paleo movement started with the right intuition that environmental factors explain most of the causes of disease and ends with the need to recreate a fast-track genetic armageddon through mitochondrial degeneration affecting 40-50-60% of the population. I am afraid this will prove a road to nowhere.

  7. Heather, if one has very prominent symptoms certain tests can be ordered, however, it is a fringe specialty at this point and I am not familiar with the latest research on the testing for it.

    Gabriel - it sounds interesting - I haven't read it!

    Woo - I never said monoamines had nothing to do with the affective spectrum disorders

    Toni - I agree that CFS is a spectrum of diseases, many of which are very poorly understood. Even its classification as an "affective spectrum disorder" is controversial, and like the depressive or anxiety disorders the advice to "snap out of it" works for some lucky individuals and backfires in most. It really becomes an issue as most doctors are trained to rule out dangerous things and when nothing drastic is found, and the person isn't responding to help, it becomes too easy to blame the patient for not responding.

    Bolalbi - mitochondria are a piece of the problem.

  8. I'm interested in how the HPA axis and adrenal glands fit into this mitochondrial malfunction picture.

    My cortisol levels are flat-lined all day (but I suspect were elevated for many years), my thyroid values are normal, and I am a poor metabolizer of many drugs. I don't fit the atypical form of depression but the more mixed anxious/depressed kind plus very specific inflammatory/autoimmune symptoms (postpartum celiac disease triggered but now six years GF and two years grain free/paleo) -- chronic muscle pain and other CFS symptoms. My mood symptoms are entirely linked to my inflammatory symptoms and both are stress-mediated to a large degree. The diet and supplements have gotten me about 80% better but there's a level of inflammation or fatigue that I can't seem to recover from. Maybe if I slept 11 hours every night and had no external sources of stress...

    So how do cortisol and adrenal function relate to mitochondrial function in this picture?

  9. Very nice post! Eagerly awating for the second part!

  10. Like many people, I too believe that it is a good idea to address the mitochondrial dysfunction. This can be a tricky thing though, if one is talking about mitochondrial death, and therefore less mitochondria to work with. It is then important to work to maintain the current status of one's mDNA. Like Woo stated, the effects of mitochondria on monamines still determines one's health at either the metabolic and/or affective level. It is still beneficial to work at the systems level while similtaneously try to correct (or at least maintain) functional health at the root problem.

    Likening to a computer can be helpful in understanding and even applying more effective treatment, however it can be a bit depressing to think about. See, when one has affective disorders, that is when having a 'software' problem is often used as an analogy. Reformat the system (through which ever monoamines are particular to your specific problem---which goes beyond seratonin, but GABA, dopamine (as Woo strongly advocates for) and ESPECIALLY endorphin levels). Now to look at the hardware level, such as mitochondrial functioning, this is where it can get a bit depressing. Eventually, no matter how many times you address the systems level, ultimately the hardware will malfunction and die. This is inevitable. If you see in the example of Parkinsons (or any other neuro-degenerative disease), of course that is a hardware problem, where it affects the systems level (which in the case of Parkinson's) is dopamine in particular. So you help the systems level until the hardware becomes non-functional.

    So yes, it is necessary to address hardware and software problems (if at all possible). I just wanted to remind everyone that ultimately all living things die---which is normal and natural (even if depressing).

    I also agree with the commentor who stated that many people don't benefit from cognitive-behavioural interventions. Of course it can help to a degree, but ultimately that has got to be one of the most misleading treatments out there. What is also misleading is if one tries to treat someone with the wrong treatment. If someone doesn't have TB, why would you treat them with a TB medication? So yes, please let's use more common sense. I'm so glad that Woo has discovered an effective intervention that works for her (no thanks to 'experts' in the medical field. She is only lucky that she is scientifically minded enough to help herself).

    In the meantime, I truly believe in minimizing pain and suffering as much as possible. This is opposite of trying to 'maximize' happiness and pleasure. I know that hedonic motives can be fallacious, however, by not addressing 'happiness' or 'pleasure' directly (such as through the intentional activity of 'removing that which makes one uncomfortable') this can therefore end up in more 'happiness' or 'pleasure'.

    So that is my two cents worth.

  11. Dr. Deans, I'm curious to hear your take on the apparent inability of antidepressants to demonstrate substantial efficacy in clinical trials, since you seem to find that they can be effective in your own practice. One (albeit controversial) review of that literature is here:

    One idea I find interesting is that any benefit of antidepressants may reflect nothing more than role as "active placebos." We know how impressively powerful even passive placebos can be, so this seems at least plausible.

  12. But what would be the mechanism? To me, it seems like this is a case of Occam's razor. Instead, what if it is all due to glucose levels? Glucose affects serotonin levels, and mitochondria run on glucose. So when glucose drops after a peak, serotonin levels will drop and mitochondria will have less fuel - leading to depression. In my experience, I feel that this is the case. Because I can feel my blood sugar rise above a baseline and then drop way below the baseline, leaving me feeling depressed, after eating a carbohydrate-rich food.

  13. @Elisa

    Irritability/foul moods after hypoglycemia are not depression. Depression is a medical condition where a person is in a state of low motivation, pleasure, pain, almost every day, almost all day, for a prolonged period of time.

    I am a hypoglycemic. For example, when I was lifting heavy boxes, I began feeling dizzy and numb, and very very tired. I sat down for 10 minutes and could barely think or move. I realized sooner or later that my sugar was probably in the high 50s/low 60s and I needed to eat. Later in the day I felt a bit irritable and uncomfortable, which was probably the post hypoglycemia epinephrine reaction. I typically feel irritable and uncomfortable after acute hypoglycemia. It feels like crap but this isn't depression. This is the response to adrenal gland freak out. Foul moods.

    I am also depressive. To contrast, 2 years ago I became quite depressed in october. I had gone through a stressor in the summer (stressors always worsen my mood states), and my seasonal mood shifts resulted in me falling into a pretty significant depression. It onset gradually and got progressively worse, by november I was very miserable and down in the pit. I was functioning at a much lower level, crying every evening, my body was very heavy, my thoughts were very narrow and dim (like you can't see more than a few feet in front of you). I was in a lot of pain physically and emotionally. I was thinking of suicide often.
    I started light therapy and within a week my depression dialed back considerably, within a month it was entirely gone.

    This is the difference between depression and hypoglycemia moodiness. Hypoglycemia moodiness is a response to hypoglycemia and it is temporary. Depression is something that starts, gets worse, and ends when it ends, and it is much more chronic and consuming than hypoglycemia irritability. Hypoglycemia irritability generally doesn't rob me of will to live and normal motivation/ hope the way depression does.

  14. William S - I don't think that is particularly controversial! The poor performance of antidepressants in clinical trials (and the shameful difference between the positive results of the published trials vs. the iffy results of the unpublished trials) should be on the forefront of the mind of every psychopharmacologist. In my practice I very, very rarely even see anyone without any history of medication use. I'm sort of the clean-up person. I've seen antidepressants not do squat, and I've seen them turn things around and help people a great deal. It is thought the placebo effect tends to wear off after 3-6 months - but at that point have the antidepressants themselves altered the brain landscape so that coming off of them causes a relapse of the depression?

    In my clinical population, I tend to think of antidepressants as a cast. It can hold things in place while therapy, nutrition, lifestyle changes, and acute stressors are taken care of. Sometimes they seem to hold things together enough to enable the person to make use of the other modalities of treatment (sometimes people seem to be way too depressed to use therapy or make lifestyle changes).

    I really don't think they should be used without attempting non-medical interventions first, though in more serious cases (where depression threatens a job or one's life) they can be jumped to sooner rather than later. Depression complicated by anxiety is another tricky situation. I honestly view antidepressants very similarly to how I view statins - the anti-inflammatory effects are likely their saving graces. They have unknown long term side effects and in general have many drawbacks that make them a pain in the a**. At least antidepressants aren't used (yet) to treat mere numbers without knowing a little more about what is going on with the patient.

    But, as I said, I hardly ever have the "initial antidepressant" conversation with people. 99% of my patients come to me already on meds, and the horse is out of the barn.

  15. You can get a Miochondrial Function Frofile here in the UK from:

    Jay Bryant

  16. Emily, great clarification on AD use in last comment.
    What do you think about benzodiazepines for anxiety? I think many times Drs try to use SSRIs and they do not help much. Besides, there also give discontinuation and side effects problems.

  17. I read you do not like comments in old posts but I think it is better to put here for context.

    Here is a good video of a study using fluoxetine. it improves vision in previously closed eye through neurogenesis

    Eero Castrén: Why do antidepressants work?
    A Stockholm Psychiatry Lecture held September 18 2012 at Nobel Forum, Karolinska Institutet.

    This seems like a good argument against SSRIs are not more than placebo.


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

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