Pascal's Pensées

Few FDA approved drugs have a reputation as controversial as Ketamine. This reputation is well earned. Originally developed in the 1960s as a short-acting anesthetic for battlefield use, in recent decades it has become notorious as a date-rape drug (‘Special K’), a club drug (‘Vitamin K’) and for its use in veterinary medicine (‘horse tranquilizer’).

However, I venture to bet that Ketamine is about to be rehabilitated for legitimate human uses. Here is why.

The reason for this belief is a most severe crisis in mental health and our approach to treating mental illness. It is not polite to say this, but given the gravity of the situation it needs to be said: It’s not working out. Not really.

Certainly a strong statement, but this sobering reality is becoming increasingly apparent. Every year, upwards of 10 million people experience a major depressive episode in the US alone. The number of people who never see a professional – and are thus never diagnosed – is likely much higher. Each of these episodes can be expected to last for months to years and is characterized by utter misery; a substantial number is terminal (the number of suicides in the US outnumbers the number of violent non-self gun deaths by about 3:1). Currently available “antidepressants” (a term I would use only in quotation marks, as most struggle to beat placebo except for the most extreme cases and might appear statistically more effective than they actually are) – these days usually a selective serotonin reuptake inhibior (SSRI) – can be expected to help one in ten people, and only after a trial period of 4 weeks to several months. Of course, if the depression happens to coincide with an underlying bipolar tendency, treatment with an SSRI will likely trigger a dysphoric mania. In other words, one will never reach the end of the interval that the SSRI needs to take to work (if it ever does). Instead, one will – at best – need to be hospitalized. So-called “treatment resistant depression” (in reality, depression that didn’t respond to treatment with a few SSRIs or SNRIs as MAO-inhibitors and tricyclics have largely fallen out of favor due to their side effect profile) is either allowed to take its course (with devastating consequences for the individual) or treated with electroshock therapy (ECT). ECT is remarkably effective in addressing this kind of depression, but the “side effects” (more aptly called treatment effects) like deleterious memory loss are a steep price to pay.

This state of affairs is obviously unacceptable. The good news is that it is increasingly being recognized as unacceptable. The National Institute of Mental Health (NIMH) recently announced that it would no longer fund research based on diagnostic criteria as outlined in the DSM. Psychiatric disorders are the only medical conditions that – for historical reasons – are diagnosed entirely based on subjectively reported symptoms. Doing this for any other condition, e.g. cancer or infectious diseases would obviously be absurd. Molecular biology has seen to that. In the 21st century, this simply will no longer do and the NIMH is essentially willing to start over from scratch.

Enter Ketamine. When Ketamine was used as a battlefield anesthetic in Vietnam and the first Gulf war, it was anecdotally noted that wounded soldiers such treated developed far fewer cases of PTSD than soldiers with similar wounds that were treated with other anesthetics. A couple of years ago, systematic studies showed that the vast majority of patients suffering from major depressive disorders made a rapid recovery when given a low, sub-anesthetic dose (most studies show a dose of 0.5 mg/kg to be effective) of Ketamine.

Ketamine

The clinical effectiveness of Ketamine in these scientific reports sounds too good to be true, particularly when compared with anything else on the market. The effects take hold within days (if not hours, compared with weeks to months for SSRIs), they seem to work for most people, there seem to be few discernible side effects (in stark contrast to – for instance – ECT), and it seems to be equally effective in the treatment of bipolar depression (which is notoriously hard to treat).

So what is the catch? Given the crushing disease burden inflicted by major depressive disorders, the question why this treatment is not readily available arises immediately.

The issue does not seem to be primarily medical in nature. Ketamine *is* a dissociate anesthetic, so the immediate effects on the conscious experience are rather extreme. Based on reports from people who have received low sub-anesthetic doses of Ketamine, the Kantian a priori categories of space and time seem to unravel shortly after the injection. They report that it becomes obvious that shared reality is a construct – brought about by the normal operation of the brain – but that Ketamine suspends this normal construction process, allowing for different reconstructions of reality. At higher doses, the reports speak of “leaving flatland” and becoming aware of higher dimensional objects that only appear to be separate when projected onto a low-dimensional space (such as the one we commonly perceive). Whether these experience reports sound scary or intriguing, there is no question that these experiences do not last very long. Given the short half-life of Ketamine and depending on individual metabolism and route of administration (IM or IV), these dissociative effects last for an hour or two, not longer.

One of the most remarkable figures I've ever seen. Dissociative symptoms over time. x-axis is nonlinear. Differences at all time points other than the 40 minute mark are not significant. Adapted from Diazgranados, Nancy, et al. "A randomized add-on trial of an N-methyl-D-aspartate antagonist in treatment-resistant bipolar depression." Archives of general psychiatry 67.8 (2010): 793.

It is also reassuring that the patients could *answer* the CADSS, so they couldn’t have been too far gone. Intriguingly, while there was no significant difference between placebo and ketamine group expect for the 40 minute mark, *all* the mean scores of the placebo group seem to be slightly above the ketamine group. The CADDS features items like: “Do objects look different than you would expect?”. It is not inconceivable that, the experience properly calibrated the scale of the Ketamine group.

Obviously, there are no studies on long term effects in humans at this point, but similar studies on monkeys are encouraging. Animals which had received similar doses for long periods of time on a daily basis show that the dose has to be relatively high and the period of time relatively long to demonstrate impairment. Moreover, the doses involved in the treatment of pain are usually considerably higher, without reports of long term ill effects. To be clear: *Any* potential neurotoxicity is obviously cause for serious concern. However, all medical decisions involve tradeoffs. Depression it itself is increasingly linked to neurotoxicity. Moreover, the spectre of neurotoxicity can lurk where one least expects it, e.g. from antibiotic treatments. In addition, Ketamine seems to potentially *reverse* depression (or stress) induced brain damage via synaptogenesis. The moral of the story is that one should not embark on a course of medical treatment unless the expected upside (far) exceeds the expected downside.

Of course, there is a rub. More than one, actually. The two biggest and uncontested issues seem to be unclear mechanism of action as well as sustainability.

The first issue is that we do not really know or understand how Ketamine seems to bring about its antidepressant magic. Many theories are currently being explored in active research. Some of the hottest trails are NMDA antagonism, neurogenesis and dentritic sprouting. Personally, I believe there might be something to the notion that one is “growing a prefrontal forest”, strengthening prefrontal networks that are in turn able to better quench aberrant activity originating in evolutionary older structures (amygdala, limbic system, etc.)  One way or the other, glutamate seems to be involved. While it may sound unsettling that we do not understand the mechanism of a treatment, this situation is far from unique. As a matter of fact, we do not understand the mechanism of action of *any* anesthetic drug. Similarly, there is plenty of evidence that even SSRIS don’t work the way we thought they do. There is mounting empirical support for the notion that the Serotonin action is largely incidental, and that neurogenesis is really behind their therapeutic effects.

The second issue – that of sustainability – is more serious. Curiously, the effects don’t seem to last. While Ketamine can rapidly pull someone out of a serious depression, the depression seems to return in time, requiring a “booster” shot to banish the demons again, even for a short while. The time until relapse is different from individual to individual, ranging from weeks to months, but it is intriguing that there is a time constant at all.

However, there are many chronic diseases that require daily administration of medications, including injections. In this regard, Ketamine isn’t even all that different from most other antidepressants, which are – more often that not – a pretty permanent deal. Many have to keep taking them, for fear of relapse. The real issue why these treatments are not more readily available seems to be economic in nature.

Ketamine is an extremely cheap drug, as it has been off patent for almost 40 years. Put differently, there is no money to be made here. The FDA is tasked with protecting the public from harmful treatments. Thus, the approval process is lengthy and costly. In reality, only a major pharmaceutical company has the financial resources to spearhead the approval of a drug. In the case of Ketamine, this is unlikely to happen, as these companies could never realistically expect to recover their expenses. To be clear, Ketamine already *is* FDA approved, but not for the treatment of depression. There are already some courageous pioneers who will administer Ketamine today (for a king’s ransom) in its off-label use for depression. This is not in itself unusual. Many drugs – once FDA approved – are prescribed for off-label uses. For instance, Modafinil was initially approved for the treatment of narcolepsy. Today, the vast majority of prescriptions are not issued by neurologists for narcoleptics, but rather by primary care physicians for people who feel a little tired – or simply because people want to use it. So in principle, Ketamine already is available for the treatment of depression, off-label. However, the overwhelming majority of psychiatrists is unlikely to touch a drug that is a PCP derivative and needs to be injected, no matter how effective. This problem is not unique to Ketamine. Once a drug has acquired a certain infamy, it is hard to change minds. For instance, Thalidomide is now being explored as an effective cancer treatment, precisely because its such a potent inhibitor of angiogenesis. But that’s a hard sell, given its historical record.

Where does this leave us? In an uncomfortable (as millions are suffering right now), but hopeful position. The evidence for the antidepressant effectiveness of Ketamine (for whatever, yet to be understood reason) is so overwhelming that quite a few pharmaceutical companies are feverishly working on Ketamine analogues and delivery methods (e.g. nasal sprays) that *can* be patented and thus would be worthwhile to put through the highly demanding FDA approval process. Preliminary results are so promising that one can reasonably hope to have truly effective antidepressants available within another decade or so. If this happens, a mental health revolution will be at hand. And it will be sorely needed. Having rapidly acting and unequivocally effective antidepressants widely available and covered by insurance (akin to antibiotics) will make all the difference.

Note: There is no question that Ketamine is a crude drug when it comes to addressing the pathology that underlies depression. Nevertheless, it is a promising and encouraging start, not necessarily the end. Further drug development will need to hone in on the underlying biological target systems (which is why a mental illness classification based on biomarkers is so sorely needed). Moreover, it has not escaped my notice that this discussion has focused on psychopharmacological aspects of depression. There are other aspects that are social, psychological and nutritional in nature, among others. Etiology is likely complex. Breakdown in social coping structures? Sedentary lifestyles? Overfeeding? Intense and chronic stress? Extreme social competition and comparison? Sleep deprivation? Light pollution? Hormone disruption? There certainly is a discussion to be had about these aspects, but not now. It will take a while for research to disentangle these causal links. Meanwhile, it is important to lighten the burden of disease. Thus, the focus was deliberate and we will save a deliberation of other factors for later.

On a final note, it is quite unsettling that virtually all truly effective treatments for mental disorders (e.g. Lithium, ECT, Ketamine, etc.) and psychoactive substances in general (e.g. LSD, Benzos, etc.) were discovered entirely by chance, by pure serendipity. Conversely, all mental health treatments *designed* to do a certain thing, based on our current understanding of the nervous system (e.g. SSRIs, but not just SSRIs – one can always do worse) basically failed to deliver. This suggests that we do not currently understand it very well. Recognizing this should break a lance for pioneering (some call it basic, but there is nothing basic about it) neuroscience research.

PS: This is another installation in an ongoing series on how language really does matter. Most people suffering from depression would reasonably turn to antidepressants, not dissociative anesthetics for help. But just because marketing calls them that doesn’t make it so (whatever one’s position, they are on average so ineffective that a vigorous debate whether or not they are more effective than placebo is even possible. A debate we didn’t see regarding the effectiveness of penicillin. Regardless, there are so many vested interests involved here that the debate is likely to go on. So it is certainly premature to write “Listening to Ketamine” even as people are sick of listening to overhyped BS. Yet, the suffering of the suffering is so severe that I remain hopeful that reason will prevail in the struggle to make depression history. The notion to have effective antidepressants available is a powerful one). As it turns out, the antidepressant effects of some dissociative anesthetics are in all likelihood much more potent than that of current “antidepressants”. Go figure.

Briefly, if you look at a large moving scene for a while, you will experience things moving in the opposite direction afterwards. This “motion aftereffect” was already known to Aristotle, presumably from the visual inspection of waterfalls. It was rediscovered by Purkinje in the 19th century, on the occasion of witnessing a cavalry parade. Now, we were able to show that listening to ascending or descending musical scales produces a visual aftereffect in the expected (opposite) direction.

A cavalry parade, much like the one that inspired Purkinje and launched a thousand papers.

I anticipate to get a lot of grief for this, but I consider the study to be executed in a valid and rigorous way and it contributes to the often neglected study of multimodal effects. Let alone multimodal higher-order effects. But they do exist.

Hedger SC, Nusbaum HC, Lescop O, Wallisch P, Hoeckner B (2013). Music can elicit a visual motion aftereffect. Attention, Perception & Psychophysics.

What is a motion aftereffect (MOA)? Have a look here or below.

To see a motion aftereffect, look at the red dot (Warning: Browser-dependent, the effect might be a little subtle if the animation renders in a choppy fashion.)

Here are two figures that didn’t make it into the paper. Consider them as supplementary material. For an explanation of what the axes mean, etc. – see the paper itself. It won’t make much sense to go into that here without the context of the experimental setup.

The basic effect

Psychometric functions

It is somewhat of a cheap shot for a scientist to come down harshly on astrology. As a matter of fact, it is probably the lowest hanging fruit there is.

An easy target

Nevertheless, the undying popularity of astrology in general and horoscopes in particular cannot be denied. Most newspapers in most countries won’t do without a horoscope page. Most can do without a science and technology page. This is not a conspiracy. Generally speaking, the media aims to please – it will give people whatever they want to read or hear (in order to survive against its competition, until a stable state is reached). This should tell us something about the relative popularity of these topics in the general population. Of course, most hardcore scientists take this only as further evidence that the mass of men are – at best – stupid sheep and that their opinions can be dismissed out of hand. There might be something to that, but it does not take away from the fact that the continuing popularity of astrology – not to be confused with astronomy – is a fascinating phenomenon in its own right. As a matter of fact, this has fascinated me at least since high-school. Incredulous at the fact that one of my friends was looking up her horoscope, I asked her why she believes this stuff in this day and age. The response was as pragmatic as it was telling: “Oh, I only believe it, if it is good!” Recently, it was pointed out that due to the precession of the axis of the Earth, the ancient star signs are no longer in alignment with their current positions. In other words, the dates of the star signs of the Zodiac, familiar to all have to be adjusted. In addition, a new sign – Ophiuchus was introduced. Outrage predictably followed. Of course, this whole affair shows just how preposterous the whole notion of zodiac signs and zodiac based horoscopes really is. Axial precession is a phenomenon that did not just happen yesterday. It has been continuously going on for millions of years, including the couple of thousand years since the introduction of the Zodiac system. The value of these personality characterizations and predictions of the future can adequately be assessed by the fact that the fundamental misalignment was never noticed. Of course, this makes a perfect foil for people who always noticed some slight inaccuracies, but let’s be serious. Obviously, there is nothing to horoscopes. More on this later. But how could the people who came up with this stuff be so wrong? What were they thinking? Contrary to popular belief, people are not stupid. At least those who advance civilization by creating culture typically aren’t (as controversial as the zodiac might be, it is a cultural achievement; let’s be fair). How is this possible? Historically speaking, this is actually a quite common or even typical phenomenon: Something that once made sense in the original context of discovery/creation no longer makes sense in the context of justification. Between initial discovery and the current need for justification, millenia have passed. During this time, civilization has accumulated a lot more knowledge. In other words, the caravan of culture has moved on. This happens all the time. It is called progress. Note that if astrology and horoscopes didn’t already exist, it is very unlikely that – given what we know about the nature of stars, incomplete as our knowledge might be – we would introduce such a system today. So how and why did it come about way back when? Because it made sense. Carefully watching the sky – as the ancients are known to have done – can reveal a tremendous amount of real information about issues of tremendous importance to the civilization. For instance, observing an equinox allows a civilization to decide when to harvest and when to plant. This is absolutely crucial information for an agrarian society. Similarly, as the motion of heavenly bodies is closely tied to the progression of time, watching lights in the sky can provide quite a bit of information about the general kind of weather (e.g. storms, snow) one is likely to encounter. Moreover, the annual recurrence of the seasons and the close watch thereof some amount of Planungssicherheit to early societies. The progression of lights in the sky seemed even to directly influence events on earth, such as the tides and their relation to lunar cycles. In other words, the sky abounds with information, and ancient cultures spend tremendous efforts extracting it systematically, which led to the creation of calendars, etc. From this, it is a very short leap to astrology. In a world largely untouched by scientific knowledge, a world shrouded in mysticism, a world in which the night sky loomed much brighter than today (due to the lack of electric illumination), it might not be a leap at all. On the contrary. Given all the demonstrably useful information about the future that the sky does hold, the burden of proof that the sky does not hold more personally relevant information about the future might be on the skeptic (in the ancient world). As a matter of fact, it might have been tempting to link the constellations of the smaller night lights to personal matters, in contrast to the motions of the brighter and larger lights, which determine the future of the entire society. Also remember that there was plenty of time for the ancients to ponder these matters, in the absence of any electronic night entertainment whatsoever. Given what we know about human perception, it is no surprise that they were able to recognize patterns in the starry arrangements of the night sky.

The classical zodiac. Can you spot your sign?

Humans typically have a desperate desire to reduce uncertainty. Of course, the future introduces a tremendous amount of uncertainty. Hindsight is famously 20/20, whereas foresight is far from it. The insidious thing about the nature of reality is that decisions have to be made in the present – under uncertainty, whereas the consequences of these decisions occur in the future. Once the future reveals the outcome of the decision, there is no way to go back and change it, however desirable. Obviously, this is unacceptable and terribly unfair. Contemplating these matters, the desire to predict the future from things we can observe right now becomes overpowering. And so, Astrology is born. Of course, nothing changed about the nature of reality, nor the human desire to know the future since those days. And so, it lives on. It it can reduce the uncertainty even a little bit, it seems worth doing. Now, make no mistake. There is absolutely zero empirical evidence in favor of astrology or horoscopes. Nothing, nada, null. Conversely, there is plenty of positive evidence showing that it actually does not work. In scientific terms, this is a closed case. Open and shut. No ambiguity about it. If anything, the situation is eerily similar to homeopathy, which works by the power of the placebo effect. Thus, from now on, I will refer to astrology and horoscopes as an “epistemological placebo“. Reviewing the empirical evidence that astrology and horoscopes are really just instantiations of an epistemological analogue to a placebo effect would require a book-length contribution. That won’t and can’t happen today. Luckily, that is not necessary. A sampler will do. a) Horoscopes “work” (insofar as a sizable proportion of people feel themselves adequately described) because of Barnum statements. These are so general in their characteristics and so vague in their predictions that they could essentially fit anyone and anything. The remarkable thing is that people still believe that they apply specifically to them, particularly if the purported characteristics or events are positive. The psychological literature clearly shows that horoscopes work via Barnum statements, even across cultures. On another note, there is also no internal consistency. This is something you can try at home. Get 10 different horoscopes from 10 different newspapers for a given day and see if there is any overlap that goes beyond that expected by chance. Actually, this has now been done. Given the empirical evidence there is no merit in belaboring this point. To be perfectly clear, the situation truly is remarkable. To use an analogy: If people come to me throughout the day and ask me for the time and I always tell them the exact same time, and they always believe me. That is what it means to believe in astrology. This suggests the feedback mechanism is broken in those who do believe. Presumably, the want to believe overrides the feedback from reality, which is often more ambiguous as simply knowing what time it actually is. b) There is no plausible mechanism. I will make this one short. Even if one is generous, no known or conceivable mechanism exists that could explain why a particular constellation of heavenly bodies at the time of birth matters. And none has been put forward. That is only consistent. Any that is put forward will be shot down quickly. One example: Gravity. Even if one is generous, the gravitational force of the people in the room at birth is substantially larger than that of most distant heavenly bodies. Unless we live in a completely mystical universe, this one is a tough nut to crack for astrologists. The fact that science works, that nature is lawful, is evidence enough (for me) that we do not live in that place. c) Personality psychologists have worked hard, with sophisticated statistical methods, for a long period of time to quantitatively characterize the similarities and dissimilarities of people. What we came up with so far is different “inventories” with reliabilities and validities of varying respectability. It is a tough problem. This is not surprising. The brain is complex and is known to generate behavior that is not necessarily consistent in various temporal and situational contexts. What we did not find was that people could be classified in 12 (or 13) distinct categories so that the characteristics are distinct between categories but exhibit low variance within a category. Such a simple scheme is alluring, but none ever worked. People are just not that simple. It is preposterous to suggest – despite every evidence to the contrary – that they are. To be sure, timing of birth within the year does have an impact on personality and behavior (e.g. likelihood and method of suicide). There is no question about that. But there are many good reasons for these effects, none of which have to do with astrology: Amount of sunlight in the first life months, levels and type of airborne pollen, temperature, general mood, and so on. There is also a tremendous amount of artificial categorization which is imposed by society itself, which can have a considerable influence on personal future behavior. School enrollment cutoffs are typically between June and September (depending on the locality). For instance, it has been shown that – due to the fact that grouping kids by cohorts that span one year gives some that are among the oldest a considerable developmental advantage. This has been named the relative age effect or birth date effect. The final outcomes are dramatic: Elite athletes are much more likely to have birthdays in the first half of the year. Similar effects have been reported for intellectual outcomes. In short, there is absolutely no need to invoke mystical forces, however appealing, to account for such effects. So why do some people still believe this, even if there is not a shred of evidence in favor of it? I don’t think ignorance can account for everything. Motivational components have to be considered. To be sure, motives might differ. Some people might personally and financially profit from astrology. That is a pretty mighty motive. Others might be comforted by sharing beliefs in something that has been believed by many people for very long, a kind of cultural bedrock. The rest? Probably because they really, really want to. And it doesn’t seem to cost much. On the surface of it. But there is hope. I think that if anything, the “new” zodiac system will weaken some dearly held beliefs about astrology. After all, I’m no Pisces, come on now…