Sunday, 17 November 2013

Magnesium in Autism and other Neurological/Psychiatric Diseases

You may have my read earlier posts about the surprising role of potassium in autism; in those posts I also noted the importance of magnesium for the body to maintain a sufficient level of potassium.  I had thought I had really finished this subject once and for all.

Last week I was discussing my findings with the Endocrinologist.  She was asking how I could possibly tell whether a new therapy was working, given that I already have others in place.  I thought this was a very good question; I replied that if you only change one thing then you can determine whether a therapy is good, bad or has no effect.  If you are new to autism, you are not aware that the condition has many separate dimensions; it is not just a linear scale from 1 to 10.
A few days ago there was an excellent example.  Monty, aged 10 with autism, has an assistant, Nela, who goes to school with him.  When I asked how he was that day, Nela said that he was not as good as recently; he was not making good eye contact and not answering the teacher’s questions.  I asked more details and then Nela mentioned he had been covering his ears.  Then I had to think what had changed.  No potassium/magnesium supplement at breakfast.  Could it really make such a difference, and so quickly?  The only way to tell was to give K/Mg straight away.  It was like “a curtain had lifted”; Nela’s words not mine.
Rather shocked by this further proof, and since almost nothing has been written about potassium and autism; I thought I would do some digging about the other mineral, magnesium.  I was aware that in autism, people do give magnesium and vitamin B6, but I was unaware about its broader role in other neurological/psychiatric Diseases.
There is a big question about what controls the flow of magnesium across the blood brain barrier (BBB).  It clearly must cross somehow, but it is not a simple process.  Because of this, researchers at MIT tried to find a form of magnesium that would easily cross the BBB, they succeeded in mice; but it is far from clear that their new compound magnesium l-threonate has the same effect in humans. 
From the research, it is clear that most people do not have enough magnesium in their diet and anybody with any kind of neurological or psychiatric disorder should make sure their diet is rich in this mineral.  The rest of this post is really for those who want to know why supplementing Potassium and Magnesium should be good for anybody with ASD.  If you do not feel the need to know why, just go buy your supplements.
All you could ever want to know about the neuroscience of magnesium is available in one place, and for free:-

We have to thank Robert Vink, from Adelaide, Australia and Mihai Nechifor from Iaşi, Romania for this 355 page collection of research papers; if only there was one for potassium.
I made a summary of the parts I found interesting that relate to what I am interested in.  Many of the papers are not too science-heavy and you can skip through them.  
  • Magnesium levels are reduced in acute and chronic brain diseases
  • Extracellular magnesium deficiency induces apoptosis, mainly through increased oxidative stress  

Neuronal apoptosis can be triggered by three main mechanisms:

1)    Lack of growth factors;

2)    Overstimulation of glutamate receptors; and

3)     Oxidative stress.

Magnesium could play a (different) role in each of these signalling pathways.

Brain magnesium decline is a ubiquitous feature of traumatic brain injury and is associated with the development of motor and cognitive deficits.
Experimentally in TBI, parenteral administration of magnesium up to 12 h post-trauma restores brain magnesium homeostasis and profoundly improves both motor and cognitive outcome.

Magnesium has been shown to attenuate a variety of secondary injury factors, including brain edema, cerebral vasospasms, glutamate excitotoxicity, calcium-mediated events, lipid peeoxidation, mitochondrial permeability transition, and apoptosis.

Magnesium therapy has failed in clinical trials. Increase in brain free magnesium concentration seems to be essential to confer neuroprotection, and intravenous magnesium administration only marginally increases CSF magnesium concentration, which suggests that the integrity of the blood—brain barrier and the regulation of magnesium in the cerebrospinal fluid are largely maintained following acute brain injury and limit magnesium bioavailability in the brain.

Calcium and Mg cellular contents classically follow the same pathway – when Mg increased, calcium also increased. This May explain the significant correlation between Erc--Mg and intracellular calcium values as well as the fact that in children who have low intracellular calcium values, Mg therapy increased intracellular calcium levels. It can be hypothesized that a genetic factor, which modulates Na+/Mg2+ exchanger activity, may be important in the regulation of Mg


Schizophrenia and bipolar disorders are two of the most severe CNS conditions. Changes in plasma and intracellular magnesium concentration, as well as in other bivalent cations, have been found in both psychoses. Our data, as well as that of other authors, has shown that schizophrenic, paranoid patients admitted in the acute state and without previous treatment, have significantly decreased intracellular magnesium levels compared to healthy subjects. Therapy with haloperidol (a typical antipsychotic) or with risperidone (an atypical antipsychotic) both significantly raised the intracellular magnesium concentration without causing significant changes in plasma magnesium concentration. The increase in intracellular magnesium concentration was positively correlated with the improvement in clinical  symptomatology.
We consider that magnesium acts foremost by reducing glutamate release and by its Action upon NMDA receptors, and results in an augmentation in the activity of the GABAergic systems. Unlike the hypothesis that only implicates zinc deficits in the Pathogeny of schizophrenia, we consider that both intracellular magnesium and extracellular zinc deficits are equally involved in schizophrenia pathogeny.

In patients with untreated bipolar disorder, our data showed a significant decrease In intracellular magnesium concentration and plasma zinc concentration during the manic episode. 

Therapy with mood modulators (carbamazepine and valproic acid) increased total intracellular magnesium and plasma zinc concentrations without having a significant effect on total plasma magnesium concentration. Other data showed that lithium also increases intracellular magnesium concentration. The fact that mood modulators with different mechanisms of action have in common the increase of intracellular magnesium concentration is an argument to consider this augmentation as an important element of their mechanism of action.

 Magnesium in Depression

One 2008 randomized clinical trial showed that Mg was as effective as the tricyclic Antidepressant imipramine in treating Major Depression (MD). Intravenous and oral Mg protocols have been reported to rapidly terminate MD safely and without side effects. Brain Mg deficiency reduces  serotonin levels, and antidepressant drugs have been shown to have the action of raising brain Mg.

Excessive calcium, glutamate and aspartate intake can greatly worsen MD.

We believe that, when taken together, there is more than sufficient evidence to Implicate inadequate dietary Mg as contributing to the cause of MD, and we suggest that physicians prescribe Mg for its prevention and treatment.
Magnesium in autism

In this chapter (21) , a brief overview of pharmacology and genetics of magnesium
transport will be followed by a review of clinical and biological studies of Mg vitamin B6 supplementation in attention deficit/hyperactivity disorder (ADHD) and autism (autistic spectrum disorders family, ASD) in children.

Although no study carried out on a rational basis has been published to date, some experimental and/or clinical works support a positive effect of such therapy in these pathologies.

All the individual observations report a decrease in hyperactivity and a stabilisation of scholarly behaviour with treatment. These data strongly support the need for a controlled study to confirm or invalidate these assumptions.

Magnesium is known to be crucial for brain activity and its involvement in the prevention of neurobehavioural  diseases seems to be established. A  clinical double-blind study with Mg-B6 treatment over placebo cannot be accepted for regulatory and ethical reasons. 

This review brings additional information about the therapeutic role of a Mg-B6 regimen In children with ADHD or ASD/autism syndrome. This effect seems to be associated, At least in part, to a cellular Mg depletion as evidenced by intraeythrocyte Mg measurements.

Children with ADHD or PDD/ASD (pervasive developmental disorders/autistic spectrum disorders), including autism, exhibit low Erc-Mg levels.

Parents frequently showed similar low Erc-Mg values suggesting a genetic defect in Mg transport. Installing a Mg-B6 supplementation for some weeks restored higher intraerythrocyte Mg values and significantly reduced the clinical symptoms of these diseases.


Magnesium turned out to be a surprisingly interesting subject for me.  While it is clear that the science is only partially understood, at least we know that magnesium levels in the diet are important.  In the ideal world you would be able to take a special magnesium molecule that better penetrates the BBB; it does not yet exist for humans.  

Perhaps, in some types of autism, the BBB is compromised enough to allow magnesium to enter more freely. Perhaps this is why some people with ASD respond to Mg + B6 treatment, while others do not. 

Again we learnt that in human biology everything is interconnected.  Low brain Mg lowers serotonin, which is the opposite of what we want.  The thyroid axis is known to play a role in regulation of the Mg metabolism.  When Mg levels increase, so do Ca levels.  Intra/extra cellular levels of all electrolytes in the brain are very important; it is part of the brain's control system. 

The so-called ion channels are how the brain controls itself, when one malfunctions there is likely to be a cascade affecting them all.  We know from Dr Ben-Ari that the NKCC1 transporter is the location of one much malfunction, I suspect there are many others.


  1. Which form of magnesium do you find most helpful for your son?

  2. For us it is potassium that helps. We give 250mg potassium twice a day, it does also include 75mg of magnesium. Epsom baths did nothing good. Magnesium may indeed help other people.

  3. I had read, based on the E-book and links, that magnesium threonate was the preferred form. We are going to give it a try at 6mg/kg/d and pyridoxine at 0.6mg/kg/d, which is what most of the case studies used, and see what happens. Thanks so much for the link!

  4. Marc, Any results, good or bad from the magnesium threonate? Sorry for the late comment,but I am curious whether it workked out? -J

  5. Finding a good Magnesium dosage for my me and my daughter, both Aspies, has been difficult. My initial assumption was that as long as I stayed at safe levels, negative side effects were very unlikely. I was wrong as I discovered last year after reducing her dosage... she was better. For her, too much Magnesium, more than about 120mg a day (from magnesium lactate dihydrate) result in deterioration, the most obvious being anxiety related: difficulties falling asleep at night because of various fears, she also starts biting her nails and being more fidgety.

    For me, more magnesium clears the food related skin problems I have dealt with since my teenage years, probably rosacea. On the other hand, too much of it results in chronic anxiety and panic attacks that had been under control for many years. To control anxiety, I use among other supplements 5-HTP and L-tryptophan supplements to increase my serotonin. Taking too much magnesium however makes my low serotonin issues resurface: anxiety, reflux, difficulty swallowing, mid-day diarrhea, old nerve pains come back in my shoulder and other places, etc. Increasing 5-HTP and zinc helps but doesn't completely eliminate the negative side effects of higher magnesium so in the end I try to keep my magnesium at a level that allow me to eat more problematic food without rosacea issues while not having too much anxiety. On a related note, too much B12 also increases my skin issues.

    Magnesium from supplements or food seems to be mostly equivalent when it comes to anxiety, if we take more magnesium from food, supplements needs to be reduced accordingly. This is actually how I found out that magnesium was so important for us. I noticed that binging on chocolate was making my rosacea better but that I was also starting to have panic attacks. It took me a while, over a year actually, to discover it was all related to magnesium.

    This long delay in finding the best dosage for my daughter has been quite frustrating since starting supplements 4 years ago, she is now 8. Every time I think I found the perfect mix because her behavior is just how I think it should be, she will inevitably regress after a while, be it less than a day or in the best of cases maybe 2 weeks. The reasons are many, she is very sensitive to food (probably less so with magnesium), she gets sick, new teeth are coming out, her system changes with age, her body feedback mechanisms are working against me every step of the way, etc.

    In the last few months I have been mostly concentrating on the dosage of bumetanide, atorvastatin and magnesium as they seem to be inter-dependent and sensitive. We take other supplements but their dosage don't seem to be quite as sensitive. My latest discovery came about after struggling with adjusting atorvastatin for a few months. It use to work magic as long as I kept it under a certain level, even at the very low dosage I am using. But lately the effect seemed to be waning and increasing dosage just resulted in bad behavior. I was quite surprised when I reduced bumetanide by only 8%... she was back to her top self (I reduced it more than that initially but that wasn't good) as if atorvastatin was working well again. I don't know why she needs less bumetanide now, there are many explanations...

    My point however is that keeping her in the zone has been challenging. I use to establish most supplements dosage withing 10% or 20% in the first three years and her behavior was much improved overall but rarely extremely good and often very bad, sometimes I would have fleeting glimpses of "perfection". In the last little while however, I can clearly see that changing some supplements (atorvastatin, bumetanide, magnesium) by as little as a few percentage points has a visible impact on her and getting the values right keep her closer to the top and for longer periods of time, the regressions also being less important.

    Trying to hit a moving slowly moving target in a multidimensional space within 5% is not an easy task...

    1. christi999, it's interesting that you experience bad effect from magnesium, it is usually one of those magical compounds that benefits the majority of people (possibly because most of us are deficient).
      From what I know, the only group I would discourage from trying magnesium is those with such mutations in the NMDA receptor subunit NR2B that enhance Ca2+ influx ("gain of function"). In these people Mg can have both positive and negative effects since Mg closes the NMDA receptor for Ca2+ influx, but also upregulates the faulty subunit gene so that in the end there might be more calcium flowing in instead of less.
      This is just a thought, you seem to have a different explanation for this effect which might be more correct. :-)


    2. Hi Ling,

      Thank you very much for your suggestion. I don't think I have a clear explanation yet so I will try to read more on what you are suggesting.



Post a comment