Is ketamine really a plausible treatment for depression?



Last week, a publication in the Journal of Psychopharmacology made international news by reporting that patients with treatment-resistant depression (TRD) showed improvement after being given the dissociative hallucinogenic drug ketamine. Ketamine, which is traditionally used as an anesthetic in humans and other animals, is probably better known for its use as a party drug (in this context it is often called "special K"). However, a growing body of evidence has begun to suggest that ketamine may be effective (at least in the short-term) in treating depression.

I'm a bit surprised by the headlines prompted by this recent publication, though, for a number of reasons. The study, conducted by a group of scientists at Oxford, didn't really present any groundbreaking--or extremely convincing--data. The group explored the effects of ketamine infusions over a period of three weeks. Similar protocols of ketamine administration have been tested in the past (with similar results). However, the recently-published study had some shortcomings that make it a bit less convincing than some prior ketamine studies. First, there was no control group. All patients received ketamine and, although 29% of the participants showed improvement (a modest effect but relevant because these patients experienced little benefit from other treatments in the past), there is not a group whom their improvement can be compared to in order to gauge the true effects of the drug. Additionally, this was an open-label study, which means that the investigators and patients all knew that ketamine was being administered. In other words, there was no possibility that a placebo might be given. This could create expectancy effects in the patients and investigators, making the need for a control group all the more important.

The investigators were aware of these shortcomings in the study design; they initiated the study as an exploratory venture. They were interested in knowing how ketamine infusions over a prolonged period affected memory when patients also continued to take other antidepressant medications. So, they were mostly concerned with examining safety and effects on memory (they did not observe any detrimental effects on memory), not with assessing the benefit of the treatment.

But the fact remains that, despite the headlines, this study was not a huge advancement in depression research or even research into the use of ketamine to treat TRD. There is some intrigue (especially in the media) surrounding the use of ketamine as an antidepressant because of its notoriety as a taboo recreational substance. I assume this is why a relatively minor study was reported on in major media outlets across the world.

Ketamine is also an intriguing treatment for depression in the eyes of scientists, but its abuse status has nothing to do with that. It's interesting because ketamine is thought to work as an antagonist at receptors for glutamate called NMDA receptors. Since hypotheses regarding the mechanism of depression have historically focused on monoamines like serotonin, ketamine's unique (although as yet not fully elucidated) mechanism suggests there may be other valid approaches to treating depression.

However, any publicly-available ketamine treatment is at best far off and at worst improbable. 29% (the same percentage that saw a benefit) of the participants in the Oxford experiment withdrew, either due to lack of perceived benefit or adverse reactions. The adverse reactions ranged from anxiety and panic to a vasovagal reaction that caused a "reduced level of consciousness" and lasted for 10 minutes. Two of the patients vomiting during infusions. So, although the reported improvements in a minority of patients are dramatic, there are also significant adverse effects that would make treatment undesirable for other patients. Additionally, very little is known known about the potential long-term effects of ketamine treatment; there are some indications ketamine has the potential to be neurotoxic.

Ketamine may have a role to play in helping us to understand depression. But right now it is very unclear if this drug will ever be of real use in treating patients with TRD on a large scale. So, these media reports about the excitement surrounding ketamine should be taken with a grain of salt.


Diamond, P., Farmery, A., Atkinson, S., Haldar, J., Williams, N., Cowen, P., Geddes, J., & McShane, R. (2014). Ketamine infusions for treatment resistant depression: a series of 28 patients treated weekly or twice weekly in an ECT clinic. Journal of Psychopharmacology. DOI:10.1177/0269881114527361

Cocaine and Glutamate, Part Two

Ten years ago, if you had asked a neuroscientist what neurotransmitter is most important to the development of an addiction, nine out of ten times they would have said “dopamine”. Ask the same question today, however, and you’ll probably be told that it is impossible to pin such a complex process on one neurotransmitter, as clearly (at least) both dopamine and glutamate are integral to the addiction process.

In hindsight, it is not surprising that glutamate be involved in addiction. Glutamate is the most abundant excitatory neurotransmitter in the brain. It is utilized in a number of cognitive processes, but essential to synaptic plasticity, and thus to learning and memory. And addiction is really just a type of learning—perhaps learning gone haywire, but learning nonetheless. It involves the association of a positive experience with the drug that was taken to induce it, resulting in a seeking of the drug to reproduce the experience. In addiction, however, unlike other learning processes, this seeking becomes obsessive and compulsive.

It is now thought that cocaine use causes glutamatergic synapses on dopamine neurons in the ventral tegmental area (VTA), a midbrain region of the reward system, to become stronger—even after just a single use. This makes the dopamine neurons there more sensitive to glutamate, causing a hyper-sensitivity to cocaine that results in addiction. It is believed the strengthening of these glutamatergic synapses involves changes in the composition of subunits of glutamate receptors.

In order to shed more light on the specifics of this subunit restructuring, a study published last week in the journal Neuron investigates the behavioral results of changes in glutamate receptor structure. The authors created genetically engineered mice that lacked one of three types of glutamate receptor subunits: GluR1, GluR2, or NR1.

As expected, they found that cocaine-induced strengthening of synapses on dopamine neurons was dependent on the functionality of glutamate receptor subunits, specifically the GluR1 and Nr1 subunits. They also, however, made two major discoveries. First, deletion of the GluR1 subunit caused the extinction of cocaine-seeking behavior to be slowed. Thus, these mice continued to seek cocaine long after cocaine had been withheld from them, when normal mice had already “forgotten” about the drug. By extension, this might mean that pharmacological stimulation of this receptor could have potential as a treatment for addiction.

Additionally, they found that the NR1 receptor subunit was necessary for the reinstatement of drug-seeking behavior after extinction. This is analogous to relapse behavior in humans. Once again, this could have pharmacological potential in addiction treatment.

Of course, these pharmacological applications, if viable, will take some time to work out. As you can imagine, it will not be easy to create a treatment that can selectively inhibit specific subunits on glutamate receptors in a particular brain region (although this can and has been done with other receptor subunits). And, with how important glutamate is to learning in general, there is potential that a treatment aimed at glutamate receptors could disrupt other cognitive processes. So, if you’re waiting for a pill to solve your cocaine problem, you may have to wait a while longer. A cocaine vaccine (see this post about vaccinations to treat drug abuse) may be available first.


ENGBLOM, D., BILBAO, A., SANCHISSEGURA, C., DAHAN, L., PERREAULENZ, S., BALLAND, B., PARKITNA, J., LUJAN, R., HALBOUT, B., MAMELI, M. (2008). Glutamate Receptors on Dopamine Neurons Control the Persistence of Cocaine Seeking. Neuron, 59 (3), 497-508. DOI:10.1016/j.neuron.2008.07.010

Cocaine's Addictive Influence Begins Even Before Euphoria

It has long been known in the addiction field that exposure to drug-associated stimuli, commonly referred to as relapse triggers, is one of the primary causes of relapse in abstinent addicts. Neuroscience studies have added evidential support for this perspective by providing a molecular explanation for it. It is thought to principally involve two neurotransmitters: dopamine and glutamate, and a region of the reward system called the ventral tegmental area (VTA).

The VTA is part of the midbrain, and two major dopamine pathways—the mesolimbic and mesocortical—run through it. It is chock full of dopamine, glutamate, and GABA neurons. When a subject who has acquired the self-administration of a drug like cocaine is exposed to environmental stimuli they have associated with the drug, glutamate and dopamine are released from the VTA. This rush of neurotransmitters activates another area of the reward system, the nucleus accumbens, and usually leads to an attempt to reinstate drug-using behavior.

As might be expected, cocaine use itself also results in increased dopamine and glutamate transmission in the VTA. Interestingly, however, this increased neurotransmitter activity begins before the pharmacological effects of cocaine can occur. While it takes about 10 seconds for cocaine to cross the blood-brain barrier and exert its psychotropic influence, dopamine levels rise almost immediately. Thus, it would seem that the reinforcing qualities of the drug may not be solely attributable to the euphoria it produces.

Roy Wise, Bin Wang, and Zhi-Bing You published an article last week in PloS One that investigates this phenomenon. They injected cocaine methiodide (MI)—an analogue to cocaine that does not cross the blood-brain barrier to have a psychotropic effect—into rats and measured the resultant changes in neurotransmitter levels.

In rats that had never been exposed to cocaine, the MI had no effect. But in those that had previously acquired cocaine self-administration, the MI caused VTA glutamate release. It was also enough to cause these rats to reacquire cocaine-seeking behavior that had been rendered extinct.

This study speaks to the complexity and potency of the inclination toward relapse. While it has been known that external cues can cause changes in brain chemistry that predispose one toward relapse, this is the first evidence that internal cues (besides the actual rewarding mental influences of the drug) may also play a role in reinstating drug use. Fortunately, these added influences can be avoided by continued abstinence from the drug. But once a drug is used, how pleasurable the resultant experience is may have little to do with the re-emergence of drug cravings.


Roy A. Wise, Bin Wang, Zhi-Bing You, Antonio Verdejo García (2008). Cocaine Serves as a Peripheral Interoceptive Conditioned Stimulus for Central Glutamate and Dopamine Release. PLoS ONE, 3 (8) DOI:10.1371/journal.pone.0002846