Monday, 5 December 2016

Potassium Bromide for Intractable Epilepsy and perhaps some Autism

Potassium Bromide has been on my to do list ever since I read a case study about Ida, a girl with epilepsy and non-verbal autism being treated at London’s Great Ormond Street Hospital 150 years ago.  Of course, the doctor not did not use the term autism, but it was obviously present.  

What I took away was not the resolution of her seizures but her behavioral change and most importantly the initiation of age-appropriate play.

My first toe in the water in treating my son’s autism was to use Bumetanide.  That trial was successful and ever since I have looked at ways of increasing this bumetanide effect.

Bumetanide partially blocks the flow of chloride (Cl-) into neurons and over time lowers the concentration towards where it should be, in typical mature neurons.  This allows the neurotransmitter GABA to function as it should and brings back neurons into a less excitatory state and hence gives better cognitive function.

Other ideas to further lower the level of chloride included using the AE3 ion exchanger and so I proposed the possible use of Diamox.

It might also be possible to increase the expression of KCC2, the transporter that takes chloride out of neurons; this might be achieved using intranasal insulin or indeed IGF-1.

Yet another theoretical method might be to introduce bromide and allow it to compete with chloride.  We know that Br ions cross cellular membranes more quickly than Cl. So by adding bromide we should automatically reduce chloride concentration within neurons.

Medical use of Potassium Bromide

It is surprising how medicine varies so much by country.  One example is the continued use of potassium bromide (KBr) to treat childhood epilepsy in Germany, Austria and Japan.

It is currently used to treat severe forms of generalized tonic-clonic seizures, early-childhood-related Grand-Mal-seizures, and also severe myoclonic seizures during childhood.

KBr was the world’s first epilepsy drug and its use was pioneered by Sir Charles Locock in 1857.  It is still the first-line treatment for treating epilepsy in dogs, but no longer in humans.

Due to a very long half-life, it takes a month of use to reach a stable level, so in the earlier years it is likely that un-necessarily high doses (up to 6g per day) were used.  This led to side effects.  The modern dosage is 50 to 70 mg/kg in infants and toddlers, 30 to 50 mg/kg in school children and 20 to 30 mg/kg in adults.  Tolerability of bromide treatment is much improved.

It is possible to start therapy with a loading dosage to overcome the problem of the long half life, but I expect this just increases the chance of side effects.

My thought was that at a lower concentration than prevents seizures, bromide might still be effective in some autism that responds to bumetanide.  At such a dosage the side effects that occur in German epilepsy therapy might become trivial.

The main side effects are usually drowsiness (19%) and acneiform skin eruption (13%) at the 50mg/kg dosage.  I was thinking that at a quarter of this dose you might get the good without the bad.

If you have one of the many kids with autism and intractable epilepsy then you might as well follow the standard dosage and just accept the risk of some spots.  After all, the standard anti-epileptic drugs (AEDs) all have side effects and we are not just taking about spots.

Interestingly, while KBr does not interact directly with other AEDs, it is found in Germany that previously ineffective AEDs can become effective when the person is given KBr.  There are various theories to explain this.  As a result KBr look doubly useful for intractable epilepsy.  

Dravet Syndrome

KBr seems to be particularly effective in people with SCN1A-mutations suffering from Dravet syndrome.  You may recall that Professor Catterall trialed his low dose clonazepam therapy in the mouse model of Dravet syndrome.  German and Austrian clinicians have shown that KBr is highly effective in treating seizures in the human form of Dravet, while a Japanese retrospective analysis of 99 patients which found complete prevention of status epilepticus in 41.7% of patients receiving bromide.

Mode of Action

Nobody knows exactly why KBr is effective in epilepsy, but that also applies to many other AEDs.

The Brazilian view is:-

“bromide may exert antiepileptic activity not only because of its reinforcement of the Cl hyperpolarizing Nernst potential, but also because of its low affinity for the NKCC enzyme in comparison with Cl . In summary, bromide's antiepileptic effect may be divided into three parts: (1) compensation of Cl accumulation by means of its hyperpolarizing effect on chloride channels; (2) antagonism of chloride flow through the channels because of its competition with chloride; (3) low affinity for the NKCC enzyme”

That paper is:-

The German view is:-

“While the exact mode of action of bromide is still unknown, the most acceptable hypothesis besides an inhibition of carbonic anhydrase is stabilization of excitable membranes through hyperpolarization of neurons. Bromide crosses cellular membranes more quickly than chloride, enhancing
GABA-activated inhibitory postsynaptic potentials and leading to hyperpolarization. Not only GABA-activated chloride channels are more permeable to bromide, but also voltage dependent channels. Studies using combined rat hippocampus-entorhinal cortex slices showed that bromide reduced or even blocked low calcium and low magnesium induced recurrent discharges, including the low magnesium induced late recurrent discharges which do not respond to most clinically used anticonvulsants. This mechanism might explain why our patients who previously did not
improve with various other antiepileptic drugs responded to treatment with bromide.

The above is from one of many good German papers on KBr :-

Intractable Epilepsy
About one-third of people with epilepsy will eventually develop intractable epilepsy. This means that standard anti-epileptic drugs (AEDs) do not work well, or at all, to control the seizures.
Intractable epilepsy can have a big effect on life. People with intractable epilepsy may have trouble at work or school. They may worry a lot about when their next seizure will come. They may also have injuries that result from their seizures.
In the case of the 30+% of people with strictly define autism (SDA) and epilepsy things can get particularly difficult and depend a great deal on where you live.
In the US some children with severe autism and recurring seizures can still be collected from home by the school bus and dropped back at the end of the day.  Not only do they have qualified nurses at school to deal with any seizures but even the bus has a nurse.
I was just reading about a teenage girl in the UK who no longer attends school at all because she may have a seizure.  The irony here is that the girl has been to the county’s top children’s hospital, Great Ormond Street.  Had she been there one hundred and fifty years ago she would have been prescribed KBr.  Had she attended a hospital in Innsbruck or Salzburg, Austria this year she would very likely also have been prescribed KBr.

The literature supporting the use of KBr is published in the English language and so there is no excuse for epilepsy experts not to be aware of it. Both the US and the UK have provisions in place where clinicians can apply to treat patients with non locally approved drugs.  So there is nothing to stop a neurologist or epileptologist in the US or UK from using KBr if he really wants to.  He just has some extra paperwork.  The simpler solution if you have intractable epilepsy might be to pay a visit to Germany, Austria or indeed Japan. Or you go see the vet.


This blog does not have many German/Austrian readers, in fact for a condition “invented” by Austrians (Kanner and Asperger) there is very little coming out of that part of the world nowadays.
German/Austrian parents would be the ones best placed to see the effect of KBr on intractable epilepsy and perhaps some autism.
Any readers that do try potassium bromide are very welcome to share their experiences.


  1. Peter, do you think that Phenibut could be an alternative to baclofen?

    1. Be very careful with Phenibut, it is very impactful but tolerance can build unless other compounds like fasoracetam used.

    2. Baclofen was originally suggested by a risk averse UK pediatrician, who had only positive things to say about its use in Asperger's. There are other drugs that affect GABAb, but all drugs are different.

    3. Thanks Peter and Em, if baclofen wasn't so difficult to get without prescription, I wouldn't doubt.

    4. Most strong GABAergics are addictive and will induce withdrawal symptoms upon termination of its use. The problem with Baclofen and Phenibut is also their short half-life (a problem the creation of Arbaclofen was supposed to resolve).

      Ironically, I think Phenibut is still over the counter and legal in the United States, but thanks to the big pharma's goto gal and super-corrupt Senator Claire McClaskill of Missouri, Picamilon has been banned by the FDA even though it is far safer than Phenibut which can be abused, but is not even remotely on the level of abuse from the kinds of drugs doctors and psychiatrists liberally prescribe for their patients just to make them happy with their service.

      Most of the research on Phenibut is also quite old and since it is a Soviet era drug, not much new research will likely go into it because of the stigma of anything Russian or Soviet has in the west (even to this day).

      From what I know about it, I would stay away from using it.

    5. This is an interesting survey on Phenibut & other Nootropics:

      From personal experience it can have a tremendous effect and from what we know of the action could be of great help, but must be used very carefully.

      Another compound on that list, tianeptine, has a much safer profile and could be of use for a certain subset of children with ASD..

    6. I found it interesting that tizanidine was one of the drugs recommended in autism long ago by dr Chez in his book:

      It seems like he used it quite often. In another paper by him tizanidine safety and pediatric dosing is discussed.:

      “tizanidine should be started at a dose of 1 and 2 mg at bedtime in children aged <10 and >10 years of age, respectively, with a maintenance dosage of 0.3-0.5 mg/kg/day divided four times daily”

      Tizanidine has been suggested to “inhibit glutamate release”, so do you think that tizanidine response may indicate that other drugs targeting E/I imbalance could be helpful: bumetanide, low dose clonazepam, Diamox (or KBr)?

      Valentina, do you still give tizanidine to your son with good effect?

    7. Clonidine is another drug that inhibits glutamate release and is used in ADHD and sometimes autism.

      Guanfacine is yet another drug known to inhibit glutamate release and yet again is an ADHD drug.

      Guanfacine shown safe and effective in autism treatment

      These two drugs and tizanidine are all α2 adrenergic agonists. There are interactions with other drugs like verapamil.

      Definitely an interesting line of thought.

    8. Agnieszka and Peter, the first thing I thought when discovered tizanidine by chance,was "this was made for my son", he responded incredible well with only a single dose of 2 mg at bedtime. Today I started to add 1 mg at midday and I plan to give him 1mg, 1mg, 2 mg, divided during the day.Great result for now. My son's greatgrandfather had MS and this is an antecedent that I don't forget. He shows both hypotonia and movements with hypertonia.

    9. This comment has been removed by the author.

  2. Here is some huge research on Alzheimer's disease that I believe may potentially explain a few big topics in autism:

    What these researchers did was show through a variety of experiments that 40hz stimulation of neurons provided a protective effect in clearing amyloid-beta proteins from Alzheimer's Disease model mice. They first optogenetically stimulated neurons in the hippocampus of mice and found that amyloid-beta was reduced by at leas 50%. Then they looked for possible biological reasons for this effect and found that microglia were activated to a phagocytotic state (i.e. ready to go eat stuff) and that in effect the microglia were recruited by the neurons when the neurons were stimulated at this frequency to create this neuroprotective effect.

    Now to pause for a second on the rest of the paper, it is important to point out in autism that microglia seem to be hyperactivated, while at the same time it is thought that they are not doing their jobs properly in terms of munching on excess synapses leading to one of the many hyperconnected theories of the autism brain. It is also worth pointing out that several post-mortem studies on adult autism brains show amyloid-beta buildup as well so obviously with respect to autism this finding is a little ambiguous.

    Anyways, the researchers then found that applying a GABAA receptor antagonist (i.e. blocks inhibition) blocks this 40hz optogenetic effect of inducing microglia to take out the synaptic cellular trash. GABAA receptor dysfunction and its effect on parvalbumin fast spiking interneurons (these are important because they inhibit other neurons strongly and are necessary for the generation of gamma rhythms in the brain) seems to be one of the core features of autism and if you can imagine that if GABAA function is compromised, then microglia function will as well leading to neurodegenerative conditions on par with what you find in Alzheimer's disease, but also potentially in autism.

  3. But wait there is more! The researchers then used photic driving at 40hz (i.e. they pulsed LED light at 40hz) and then looked at the visual cortex of mice to see if the same effect that occurred via optogenetic stimulation of the hippocampus was occurring as well in the visual cortex 40hz LED stimulation and it turns out it was. Then they tested this stimulation against one or more (can't remember how many) other Alzheimer's disease model mice and found that amyloid-beta was reduced in their visual cortex by a whole bunch in those mice as well.

    So with respect to humans, you can easily stimulate yourself with LED stimulation at 40hz, though getting your child to do this, especially if they are low-functioning, might be a bit of a challenge. Back in my late 20's to early 30's I used to toy around with a product called Neuroprogrammer II which can be found at I also bought some accessories to use with the program called an AudioStrobe decoder and some cheap LED glasses you wear. I experimented a whole lot on my own creating my own sessions and 40hz stimulation definitely was special. Once I had a family I didn't have the time or privacy to mess around with this stuff anymore so I shelved it more or less. The developer of the software now has a site along the same lines called (I have no financial interest in either of these companies and have not actually used myself). There are other programs and hardware as well that do the same thing (BWGen if I remember correctly is one of them), but NPII was what I used. I used to do this stuff all the time and unless you have epilepsy, it is completely safe.

    But the main takeaway from this research is it potentially shows how a cascade of problems could potentially occur if GABAA receptor function is compromised not only in the visual cortex and hippocampus, but potentially the entire brain. If GABAA receptor function on parvalbumin interneurons is not functioning, then you won't get properly functioning gamma rhythms and this 40hz neuroprotective effect being induced to instruct microglia to take out the trash which in neurodevelopment also means pruning excess synapses.

    Also, if anyone is considering any sort of therapy related to this topic, the research was pretty clear that this neuroprotective effect from 40hz stimulation won't happen if GABAA receptor function is dysfunctional. I would also stress that since parvalbumin interneurons are driven from mglur5 receptor signaling from pyramidal cells (which a lot of autism research suggests is dysfunctional as well), you might get neutral or even negative results from this type of therapy UNLESS those two issues are addressed pharmacologically (my opinion only of course).

    Nevertheless, this type of therapy, though specific to Alzheimer's Disease, is non-invasive and I feel intersects strongly with the most current autism research so it is worth a strong look by serious autism researchers as well as parents looking for serious therapies to help their child now.

  4. Tyler,
    Do you know why Baclofen is not recommended for Parkinson's disease?

    1. I would assume that is because Parkinson's disease symptoms are typically treated with L-Dopa and/or Carbidopa to restore the lost dopamine production from the ventral tegmental area which is one of the first areas of the brain to essentially die, thereby leading to further downstream dopamine dysfunctions including dystonia.

      As far as preventing Parkinson's from fully manifesting, the most recent research I have come across has been on identification and prevention. A paper I read a few days ago suggested that oxidative stress (everyone's favorite neurodegenerative buzzwords) seemed to be the core problem that led to cell death (sound familiar?). Other hypotheses ranged from an excess of glutamate or an excitation/inhibition imbalance in the infected areas of the brain (infection in the sense of the abberrant PD protein a-synuclein spreading from neuron to neuron).

      Off the top of my head I don't know why Baclofen is "not recommended" for Parkinson's disease, though I would suspect some have investigated its use for decreasing neuron hyperexcitability and hopefully slowing the disease's progression that way.

      But to answer your question I do not know at this time why Baclofen is "not recommended" for Parkinson's disease.

    2. Tyler, thank you very much for your attempt to answer my question.
      I found the "not recommended for Parkinson's" statement on wikipedia.

      I needed this information as I am trialling Baclofen for the second time with mixed results. He sometimes acts as if his dopamine is blocked and thought I should investigate it a little bit further.

      I also found this article which might explain some things:

      The gamma-aminobutyric acid type B (GABAB) receptor agonist baclofen inhibits morphine sensitization by decreasing the dopamine level in rat nucleus accumbens
      Zhenyu Fu†, Hongfa Yang†, Yuqiang Xiao, Gang Zhao and Haiyan HuangEmail author
      †Contributed equally
      Behavioral and Brain Functions20128:20
      DOI: 10.1186/1744-9081-8-20© Fu et al.; licensee BioMed Central Ltd. 2012
      Received: 24 August 2011Accepted: 6 May 2012Published: 10 July 2012

  5. Tyler, Picamilon is still sold in Europe, but it is also known as nicotinoil gaba, perhaps with a combination of niacin and gaba could be obtained something similar to Picamilon?

    1. Well the strange thing about drug policy here in the United States is that you cannot even import Picamilon now as it is classified as only for research use now.

      Also, Niacin+GABA is not the same thing. The use of Picmilon according to its reported method of action is that because GABA itself crosses the blood brain barrier very poorly, the niacin bound to the GABA helps the GABA get an escort into the brain. Once in the brain it is thought that Picamilon then hydrolyzes back into nicacin and GABA. Nevertheless, there is some relatively recent research now that suggests that oral GABA supplementation might increase brain GABA levels directly which contradicts past research suggesting GABA does not easily cross the blood brain barrier.

  6. Peter, I use classic nescafe instant coffee by nestle with Bumetanide. I read on wikipedia that some recipies include potassium bromide as a fog-reducing agent. Do you know if there is any potassium bromide in nescafe nestle? I though, if actually there is any, this may add the extra benefit I see with this combination.

  7. Sorry Peter, my comment is not correct, I didn't realize that this formula is used by film developers.
    Instant coffee is supposed to be 100% natural.
    Some chips, flour, bread products may contain potassium bromate as a food additive, which then theoretically is conveted into potassium bromide.
    I don't know what is going on with coffee.


Post a comment