Susannah Cahalan’s Transient Madness

Brain on Fire - My Month of Madness - by Susannah Cahalan

Brain on Fire
– My Month of Madness –
by Susannah Cahalan

I recently visited the fabulous country of Japan (more on that in a future post). On the plane to Tokyo, I read Susannah Cahalan’s book on her transient craziness, Brain on Fire: My Month of Madness. It’s a long flight to Tokyo and the book is highly engaging. As it turned out, with my reading speed (fairly slow), first to last page was a perfect fit for Chicago-to-Tokyo.

Ms Cahalan tells the story of her experience with anti-NMDA receptor encephalitis. In this post, I will describe this disease briefly and then delve into the aspects of Ms Cahalan’s personal experience of the disease that particularly struck me.

Anti-NMDA receptor encephalitis

Recognition and naming of anti-NMDA receptor encephalitis is recent, having occurred in 2007 by Josep Dalmau and colleagues. The disease results when antibodies against the NMDA receptor gain access to the brain. Sometimes these antibodies are made in response to an ovarian tumor called a teratoma.

Teratomas are weird tumors that contain derivatives from 2 or 3 germ layers rather than only a single tissue type as is true of most tumors. Some teratomas have teeth, hair, bone, or miniature and not wholly formed portions of an organ.

The remainder of patients with anti-NMDA receptor encephalitis, possibly even a majority (jury appears to still be out on this question which is understandable given the disease’s very recent recognition), do not have a tumor and start making anti-NMDA receptor antibodies idiopathically, or for no discernible reason. The antibodies are typically against one particular subunit of the NMDA receptor (the NR1 subunit). The antibodies cause harm by 1) stimulating receptor endocytosis and/or 2) competing with glutamate, the endogenous ligand for the NMDA receptor. In either case, the antibodies reduce the number of NMDA receptors available for glutamate binding and thereby work against and depress glutamatergic transmission at the NMDA receptor.

The NMDA receptor

The NMDA receptor, while perhaps not as numerous as the other major ionotropic glutamate receptor (AMPA receptor), is very important. Unlike the AMPA receptor, the NMDA receptor passes Ca++ as well as K+ and Na+. The NMDA receptor is also special because it serves as a one-molecule-coincidence-detector of glutamate binding and cell depolarization. This role in coincidence detection makes the NMDA absolutely critical to learning and forgetting.

How can the NMDA receptor act as a coincidence detector? Well, at rest, the pore of the NMDA receptor is blocked by a magnesium ion (Mg++). Cell depolarization is required to dislodge the Mg++ from the NMDA receptor pore. If the cell is depolarized but no glutamate is around, no current passes through the NMDA receptor. Further, if glutamate is around but the cell is at a rest potential, no current passes through the NMDA receptor. Only when the coincidence of glutamate binding to a NMDA receptor on a depolarized cell occurs does the pore of the NMDA receptor pass current, including current carried by the all important Ca++ ions. In this way, the NMDA receptor detects the coincidence of cell depolarization and glutamate release from a presynaptic input.

Symptoms resulting from NMDA receptor antagonism

Let’s return to anti-NMDA receptor encephalitis. In this disease, antibodies against the NMDA receptor are not only produced but they access the brain. The blood-brain-barrier or BBB should prevent big and bulky antibodies from passing into the brain. But for reasons that are just beginning to be explored and understood, the BBB fails at its job. So, we end up with a situation where antibodies against a very important brain protein are present in the central nervous system.

What would we expect from antagonizing glutamatergic transmission through NMDA receptors? Since glutamatergic transmission is involved in every neural process that I am aware of and NMDA receptors are present at most glutamatergic synapses, changes in any and every neural function would be predicted. Not a very satisfying answer. But it turns out that ketamine, an NMDA receptor antagonist, has been administered to human volunteers and so the answer is at hand.

Ketamine is a dissociative anesthetic, meaning that it disconnects subjective experience and external reality. It is used for veterinary sedation or anesthesia (typically in combination with xylazine, a parasympathomimetic) for short surgical procedures. A former graduate student “volunteered” (meaning was paid as a human subject) to receive ketamine for a study of its psychotropic effects. He described the experience to me as crazy-making and completely dissociative. My student’s experience matches that described in multiple studies. Hallucinations are also reported after administration of a moderate to high dose of ketamine.

Because of its dissociative and hallucinatory effects, ketamine is, not surprisingly, used by researchers to model a schizophrenia-like state. Bingo! Blocking NMDA receptor-mediated glutamatergic transmission appears to produce a break from reality, a state where paranoia, disordered thought, and hallucinations rein free.

Ketamine is abused as is phencylidine or PCP, a drug with efficacy as an NMDA receptor antagonist. Feeling separated from my body and hallucinations does not sound like a good time to me. Thus I do not understand why some people find the effects of ketamine or PCP so desirable that they seek out and eventually come to reorganize their lives around the drug (=addiction).

“Neuropathic” paranoia

One of the very first symptoms of Ms Cahalan’s disease was a fear of bugs that led her to uncharacteristically clean house. Although normally a pack-rat, Ms Cahalan found herself bagging or throwing out all of her belongings. She called an exterminator and ignored the exterminator’s assurances that no infestation existed, insisting on a spray treatment. At the end of all this anti-bug activity, Ms Cahalan “still didn’t feel any better”. This is because the problem was not external; rather it was a product of her brain. A similar phenomenon can happen with persistent pain. Some chronic pain conditions are produced by the brain; no actual stimulus exists. An example of this is amputation pain where a body part which does not even exist is perceived as painful. This type of pain is termed neuropathic. All indications are that Ms Cahalan’s paranoia was neuropathic. Indeed her entire psychotic experience was brain-produced.

Paranoia is a frequent neuropsychiatric symptom with an interesting blend of biology and culture. Paranoia is intriguing because of this blend: an underlying motivation that is evolutionarily conserved and an expression that is highly particular to cultural circumstances. What I mean is that paranoia occurs in virtually every individual who suffers from certain diseases such as schizophrenia. Paranoia also frequently occurs in elderly individuals with poor memory. I have a friend who is 101 years old and has very limited ability to make new memories. Not infrequently, my friend (who lives in a secure assisted living home) tells me that she woke up the previous night and saw three men in her apartment. One of the men had a gun and the group robbed her. “What could I do?” she asks plaintively. Others express paranoia by talking about the CIA controlling their actions or the TV spying on them and the like. It makes sense to me that overblown fear – paranoia – is a common result of a brain built to feel and show fear when advantageous. The particulars of the fears involved in paranoia are clearly more malleable and environmentally influenced than innately determined.

Complete loss of normality

There are so many experiences that Ms Cahalan shares that are essentially unthinkable to anyone in a healthy state of mind. Here are just a few:

  • Emotional lability: “I’m crying for no reason….the crying spell took over my whole body… The tears continued down my face, but I was surprised to realize that instantly I was no longer sad. I was fine… Happy. No, not happy, sublime, better than I had ever felt in my entire life.”
  • Failure to stick to one action or have one thought: She had “to physically restrain myself from pacing around the apartment. I couldn’t stay with one thought; my mind was flooded with different desires.”
  • Capgras syndrome: She sees her father and recognizes that the man looks like her father but she does not feel that he is her father. She concludes that he is a kidnapper: “my family members were turning into other people….I also insisted that my father was an impostor.”
  • Amnesia: Ms Cahalan watched the same DVD of Ryan Adams over and over again on loop: “…every time I watched this DVD, it was as if for the first time.”
  • Hemispatial neglect: A recreation of the clock that Ms Cahalan drew is shown below. This is classic of hemispatial neglect which is typically described as a failure to recognize that the left half of the world exists. People with hemispatial neglect don’t turn to the left, don’t look to the left, and don’t see the left half of the world. Ms Cahalan refers to this condition more as indifference to than as blindness to the left world. Very interesting.
A re-creation of my clock drawing" from Susannnah Cahalan's book, Brain on Fire: My Month of Madness (p.132, Simon and Schuster Paperbacks, 2012).

“A re-creation of my clock drawing” from Susannnah Cahalan’s book, Brain on Fire: My Month of Madness (p.132, Simon and Schuster Paperbacks, 2012).

Life after recovery

Ms Cahalan recovered fully or nearly so. Yet she lives with fear of a relapse. Since relapses unfortunately happen, this is not an idle fear. Yet she closes her book [spoiler alert, spoiler alert!!!!] by addressing whether she would undo the experience if she could. She answers, “I wouldn’t take that terrible experience back for anything in the world. Too much light has come out of my darkness.” Eloquent.

A book “dedicated to those without a diagnosis”

I close this post with a thread that starts at the Dedication and continues through virtually every page of the book. That thread is the importance of having a diagnosis. Ms Cahalan must endure appalling misdiagnosis after insulting misdiagnosis. She endures. Her family pushes and does not give up. In some strange way, the fact that she has seizures, an objective sign of brain dysfunction, works to her advantage as it makes it harder for the physicians to brush her off as hysterical and commit her to psychiatric treatment. And yet some physicians did just that. Ultimately, she is relieved to learn of her diagnosis, even though the diagnosed disease is bad news.

Since 2007, we have known of anti-NMDA receptor encephalitis. What about all those who suffered with this disease prior to 2007 or even those who suffer today, without anyone recognizing this newly described disease? What about those suffering from as yet un-described diseases that manifest in psychiatric symptoms? A hundred years ago, a person with neurosyphilis was considered crazy. In 2006, a person with anti-NMDA receptor encephalitis would similarly be considered crazy. Tragically, anti-NMDA receptor encephalitis primarily strikes young people, including young children. It is unspeakably tragic to contemplate the suffering of undiagnosed children and young sufferers of anti-NMDA receptor encephalitis and all the as of yet un-recognized diseases with psychiatric manifestations.

Final word

This book is worth reading. Two thumbs up. It is thought-provoking, fascinating, disturbing, and will make you think new thoughts about brain function.


  1. Thank you Pr. Mason for all the interesting topics you present to us with your amazing way of explaining them!!!


    • The treatment depends on the cause. If the source of the anti-NMDA receptor antibodies is a teratoma, then the tumor is removed. In such cases, the disease is unlikely to return. However if anti-NMDA receptor antibodies arise idiopathically (= for no reason that we know about), then the person is treated with immunotherapy aimed at suppressing autoimmune antibody production. In idiopathic cases, the disease may come back. I don’t know what the return rate is and I suspect that we need some more time and cases before we have anything close to an accurate estimate of this rate. I also think that any chance of return, in the case of such a horrific disease, is a major burden to carry.

      No way to prevent this disease. In general we can’t prevent too many diseases except by vaccination… Get vaccinated! And brush your teeth!


  2. “Ms Cahalan must endure appalling misdiagnosis after insulting misdiagnosis.” ; and unfortunately, such is not uncommon. Here’s but one example of how physicians can be mislead or fail to understand what test results are indicating:

    Far too often patients forget physicians are just fallible human beings as they put the physician on a pedestal and also put the responsibility for their healthcare on anyone but themselves. NOT that that is what Ms. Calahan did or even was capable of. Just an observation.

    You surprised me me in writing “Thus I do not understand why some people find the effects of ketamine or PCP so desirable that they seek out” given that the urge/desire to alter consciousness has existed for all of documented history and has been traced to certain brain areas. I suspect you meant not an understanding of the biology but the functional aspects of such desire. Ah, the semantic prosody of web communication 🙂

    Thanks again for the ongoing education; it is truly appreciated.


  3. One thing that keeps me puzzled is how a substance or an effector (such as these antibodies) with so widespread target as NMDA receptors can have a specific, reproducible effect different from just shutting down the brain altogether.

    Or in the case of serotonin – this substance has effects throughout the CNS, from motor neurons to cognition with pretty much everything in between, and there are more classes of serotonin receptors than I can possibly remember, and yet increasing serotonin availability by quite indisciminatively blocking its reuptake by SSRIs produces a specific, anti-depresssant action.

    I am a computer professional by trade; with computers, if you start messing with circuits like that, nothing works. Period. The computer system would simply not start.

    Thanks Prof. Mason for your wonderfully accessible education


    • Viktor,

      It is a very good question. Note that this a loss of a receptor NOT a neurotransmitter. I do not believe that anyone could survive without either glutamate or GABA. But loss of one of several receptors for these two critical neurotransmitters can be tolerated. And it is not even loss of a receptor; rather one subunit is lost. So there will still be NMDA receptors but they will not include the NR1 subunit. While the NR1 subunit is ordinarily necessary, it may be that 1) not all NR1 splice variants are recognized by the antibodies; and 2) compensatory mechanisms result in at least some functional NMDA receptors with an atypical subunit composition.

      I agree that the efficacy of SSRIs belies the importance of specific neuronal connections. It is as though a pump is as functional as a synaptic connection. Another realm where this disturbs me is in the treatment of Parkinson’s Disease where essentially pumping in dopamine rescues function to a large degree. I think that it is notable that in both cases, neurotransmission is through metabotropic receptors (the role of the ionotropic 5ht3 receptor in psychotropic effects is extremely limited to nil) and synapses are primarily of the non-classical, paracrine, or volume-transmission type. I don’t know whether there is a computer analog to such a type of communication.

      Thanks for the thoughtful comment,


      • Thanks for your answer (it packs quite a lot of info for me 🙂
        So the antibodies being active only against one variant of NR1 subunit of NDMA receptor makes sense. It can also at least begin to explain the specificity, as different variants might be expressed differently in different parts of the brain.

        However, one of the turning points was this clock test, which has shown a hemispatial neglect. For this condition to occur, the effect has to be strongly lateralized to the right parietal lobe, hasnt’ it? Can antibody effect be localized like that? (I assume it can, as the upshot in BBB permeability can be localized, unlike systemic plasma cells circulation. But it presents a chicken and egg problem, as it implies that inflammation is already present before lymphocites start infiltrating it with their pesky anti-NMDAR antibodies).


      • Dear Viktor,

        Good question! Hemispatial neglect results from right inferior parietal (near temporal-parietal-occipital junction) cortical damage. The damage does not have to be exclusive to the right side but simply has to be inclusive of the right side. There are a number of models for how this might work but the one that I believe is favored currently is that the left cortex is responsible for attention to the right side of the world only whereas the right cortex supports attention to both sides of the world, albeit with a bias toward the left (contralateral) side. So if the left parietal cortex were damaged, the right cortex could take over adequately but if the right side were injured, the left side is only capable of supporting attention to the right side of the world. Alas, the left side of the world would be neglected or put another way, people would become indifferent, to one or another degree to the left side of the world.

        I will say that I do not think that this story is written in stone. Lateralized functions (eg semantic language) and lateralized biases (eg attention) are found in non-human animals but the details (side, functions affected, strength of laterality) differ in different species. So, hemispatial neglect and lateral bias in attention are studied in humans where we have to count on nature’s experiments – principally brain tumors and strokes – for data. So progress is slow and hypothesis testing limited to prospective studies.

        Here’s hoping your clocks stay round and symmetrically distributed,


  4. Forgot to query about this: ” The blood-brain-barrier or BBB should prevent big and bulky antibodies from passing into the brain. But for reasons that are just beginning to be explored and understood, the BBB fails at its job.”
    Is the ‘upshot’ of this that ‘we’ don’t really know how the BBB does what it does? And are you going to post about the ‘reasons’ for the ‘failure’?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s