<$BlogRSDURL$>

Saturday, May 20, 2006

Serotonin and Depression: A Disconnect


This is sort of  a follow-up to a recent post, in which I discussed the correlation between administration of fluoxetine, and neurogenesis in the hippocampus.  In that post, I mentioned that skeptics of psychopharmacology are fond of pointing out the lack of a well-defined connection between the inhibition of serotonin reuptake, and the clinical effects of the serotonin reuptake inhibitors.  

The hypothesis that there is a connection between serotonin and depression is a refinement of the monoamine hypothesis of depression.  That hypothesis states that there may be a connection between relative underactivity of brain monoamines (serotonin, norepinepherine, dopamine) and depression.  It is an old idea, derived in part from studies such as this one:  Antagonism to reserpine induced depression by imipramine, related psychoactive drugs, and some autonomic agents.  Sigg EB, Gyermek L, Hill RT. Psychopharmacologia. 1965 Feb 15;7(2):144-9.

Obviously, we have learned a lot since 1965, so the monoamine hypothesis is significant mainly as an historical relic.  Even the refinement -- the serotonin hypothesis -- is ancient.  Since then, ample evidence has been found, showing that the serotonin hypothesis is, at best, a gross oversimplification.  There is no strict correlation between serotonin activity and any defined mental illness.

The lack of a strict correlation means that the action of antidepressant medication is not fully understood; nobody I know will dispute that. But the thing is, harping on that string is pointless.  It is like pointing out that there is not a strict correlation between the noise of an automobile engine, and the motion of he automobile.  It is true that sometimes an automobile moves without the engine making noise; and sometimes the engine makes noise without moving.  Sometimes there is a significant lag between the time the engine starts making noise, and the time the automobile starts to move.  With the correct instruments, it would even be possible to show that none of the energy that goes into the production of noise has anything to do with the movement of the automobile.  True, all of it.  But can we conclude that there is no connection between the noise of the automobile, and the motion?  

Perhaps the serotonin boost is not at all related to the clinical effect.  It could be an .  Physicians generally know about epiphenomena, and are cautious about overinterpreting them.  Pointing out the fact, that there is not necessarily a causal connection, is tiresome.  Thus, when a skeptic takes pains to point it out, it merely shows that the skeptic is not aware of how obvious it is to medically-educated people.

As an aside, it occurs to me that epiphenomena are roughly analogous to spandrels.  If I felt like getting tediously philosophical, I could write a post outlining that analogy.  Another time, perhaps.  (A hint: just as a feature that evolves as a spandrel can turn out to be useful, sometimes side effects of mediation can be useful.)

Sometimes it is difficult to avoid getting tediously philosophical.  Oh well.  It occurs to me also that this is an example of how a good understanding of evolutionary theory can be helpful in understanding medicine, as Orac is fond of pointing out.  (As is my former residency director, Randy Nesse.)  This is true regardless of whether you believe that evolution is responsible for the origin of species.

Putting that aside aside, and getting back to the point: Because it is tiresome to hear skeptics harp on that string (the lack of a fully-delineated connection between serotonin and depression) I was surprised to see this article in a well-regarded journal:
Serotonin and Depression: A Disconnect between the Advertisements and the Scientific Literature, by Jeffrey R. Lacasse and Jonathan Leo.

Illustration: Margaret Shear, Public Library of Science
Illustration: Margaret Shear, Public Library of Science

I was surprised to see this, because it is, to me, a new twist on the old theme.  The authors do not imply that antidepressant medication does not work; nor do they argue that psychopharmacology is suspect.  
To equate the impressive recent achievements of neuroscience with support for the serotonin hypothesis is a mistake.
Rather, they argue that it is inappropriate for pharmaceutical companies to base advertising campaigns upon the presumed link between serotonin and depression.  
In the US, the FDA monitors and regulates DTCA. The FDA requires that advertisements “cannot be false or misleading” and “must present information that is not inconsistent with the product label” [27]. Pharmaceutical companies that disseminate advertising incompatible with these requirements can receive warning letters and can be sanctioned. The Irish equivalent of the FDA, the Irish Medical Board, recently banned GlaxoSmithKline from claiming that paroxetine corrects a chemical imbalance even in their patient information leaflets [29]. Should the FDA take similar action against consumer advertisements of SSRIs?
Curiously, the authors devote several paragraphs to debunking the serotonin hypothesis, even though, in my opinion, their purpose could have been served with one or two citations.  What is even more curious is that they do not go on to draw a specific conclusion that answers the question they pose: they never answer directly the question about whether the FDA should take action against the pharmaceutical companies that refer to the serotonin hypothesis in their advertisements.  

In fact, they do make a good case against the pharmaceutical companies, and I happen to agree.  It appears to be true, based upon their argument, that pharmaceutical companies are not following FDA regulations regarding DTC advertisements.  Moreover, the authors make several other good points, such as this one:
Patients who are convinced they are suffering from a neurotransmitter defect are likely to request a prescription for antidepressants, and may be skeptical of physicians who suggest other interventions, such as cognitive-behavioral therapy [48], evidence-based or not.
I personally have spent a lot of time in the office, with patients, trying to undo the misinformation contained in DTCA.  It bothers me that I have to do that.  I would much rather spend the time providing good education, not undoing bad education.  I would prefer to not have to deal with direct-to-consumer advertising at all; but if we have to have it, companies really ought to be held to the standards that exist to ensure balance and accuracy.