Can Autism be Cured via a Gluten Free Diet?

I’d like to share a story from a friend and her son–let’s call them Heidi and Sven.

Sven was always a sickly child, delicate and underweight. (Heidi did not seem neglectful.) Once Sven started school, Heidi started receiving concerned notes from his teachers. He wasn’t paying attention in class. He wasn’t doing his work. They reported repetitious behavior like walking slowly around the room and tapping all of the books. Conversation didn’t quite work with Sven. He was friendly, but rarely responded when spoken to and often completely ignored people. He moved slowly.

Sven’s teachers suggested autism. Several doctors later, he’d been diagnosed.

Heidi began researching everything she could about autism. Thankfully she didn’t fall down any of the weirder rabbit holes, but when Sven’s started complaining that his stomach hurt, she decided to try a gluten-free diet.

And it worked. Not only did Sven’s stomach stop hurting, but his school performance improved. He stopped laying his head down on his desk every afternoon. He started doing his work and responding to classmates.

Had a gluten free diet cured his autism?


A gluten free diet cured his celiac disease (aka coeliac disease). Sven’s troublesome behavior was most likely caused by anemia, caused by long-term inflammation, caused by gluten intolerance.

When we are sick, our bodies sequester iron to prevent whatever pathogen is infecting us from using it. This is a sensible response to short-term pathogens that we can easily defeat, but in long-term sicknesses, leads to anemia. Since Sven was sick with undiagnosed celiac disease for years, his intestines were inflamed for years–and his body responded by sequestering iron for years, leaving him continually tired, spacey, and unable to concentrate in school.

The removal of gluten from his diet allowed his intestines to heal and his body to finally start releasing iron.

Whether or not Sven had (or has) autism is a matter of debate. What is autism? It’s generally defined by a list of symptoms/behaviors, not a list of causes. So very different causes could nonetheless trigger similar symptoms in different people.

Saying that Sven’s autism was “cured” by this diet is somewhat misleading, since gluten-free diets clearly won’t work for the majority of people with autism–those folks don’t have celiac disease. But by the same token, Sven was diagnosed with autism and his diet certainly did work for him, just as it might for other people with similar symptoms. We just don’t have the ability right now to easily distinguish between the many potential causes for the symptoms lumped together under “autism,” so parents are left trying to figure out what might work for their kid.

Interestingly, the overlap between “autism” and feeding problems /gastrointestinal disorders is huge. Now, when I say things like this, I often notice that people are confused about the scale of problems. Nearly every parent swears, at some point, that their child is terribly picky. This is normal pickiness that goes away with time and isn’t a real problem. The problems autistic children face are not normal.

Parent of normal child: “My kid is so picky! She won’t eat peas!”

Parent of autistic child: “My kid only eats peas.”

See the difference?

Let’s cut to Wikipedia, which has a nice summary:

Gastrointestinal problems are one of the most commonly associated medical disorders in people with autism.[80] These are linked to greater social impairment, irritability, behavior and sleep problems, language impairments and mood changes, so the theory that they are an overlap syndrome has been postulated.[80][81] Studies indicate that gastrointestinalinflammation, immunoglobulin E-mediated or cell-mediated food allergies, gluten-related disorders (celiac diseasewheat allergynon-celiac gluten sensitivity), visceral hypersensitivity, dysautonomia and gastroesophageal reflux are the mechanisms that possibly link both.[81]

A 2016 review concludes that enteric nervous system abnormalities might play a role in several neurological disorders, including autism. Neural connections and the immune system are a pathway that may allow diseases originated in the intestine to spread to the brain.[82] A 2018 review suggests that the frequent association of gastrointestinal disorders and autism is due to abnormalities of the gut–brain axis.[80]

The “leaky gut” hypothesis is popular among parents of children with autism. It is based on the idea that defects in the intestinal barrier produce an excessive increase of the intestinal permeability, allowing substances present in the intestine, including bacteria, environmental toxins and food antigens, to pass into the blood. The data supporting this theory are limited and contradictory, since both increased intestinal permeability and normal permeability have been documented in people with autism. Studies with mice provide some support to this theory and suggest the importance of intestinal flora, demonstrating that the normalization of the intestinal barrier was associated with an improvement in some of the ASD-like behaviours.[82] Studies on subgroups of people with ASD showed the presence of high plasma levels of zonulin, a protein that regulates permeability opening the “pores” of the intestinal wall, as well as intestinal dysbiosis (reduced levels of Bifidobacteria and increased abundance of Akkermansia muciniphilaEscherichia coliClostridia and Candida fungi) that promotes the production of proinflammatory cytokines, all of which produces excessive intestinal permeability.[83] This allows passage of bacterial endotoxins from the gut into the bloodstream, stimulating liver cells to secrete tumor necrosis factor alpha (TNFα), which modulates blood–brain barrier permeability. Studies on ASD people showed that TNFα cascades produce proinflammatory cytokines, leading to peripheral inflammation and activation of microglia in the brain, which indicates neuroinflammation.[83] In addition, neuroactive opioid peptides from digested foods have been shown to leak into the bloodstream and permeate the blood–brain barrier, influencing neural cells and causing autistic symptoms.[83] (See Endogenous opiate precursor theory)

Here is an interesting case report of psychosis caused by gluten sensitivity:

 In May 2012, after a febrile episode, she became increasingly irritable and reported daily headache and concentration difficulties. One month after, her symptoms worsened presenting with severe headache, sleep problems, and behavior alterations, with several unmotivated crying spells and apathy. Her school performance deteriorated… The patient was referred to a local neuropsychiatric outpatient clinic, where a conversion somatic disorder was diagnosed and a benzodiazepine treatment (i.e., bromazepam) was started. In June 2012, during the final school examinations, psychiatric symptoms, occurring sporadically in the previous two months, worsened. Indeed, she began to have complex hallucinations. The types of these hallucinations varied and were reported as indistinguishable from reality. The hallucinations involved vivid scenes either with family members (she heard her sister and her boyfriend having bad discussions) or without (she saw people coming off the television to follow and scare her)… She also presented weight loss (about 5% of her weight) and gastrointestinal symptoms such as abdominal distension and severe constipation.

So she’s hospitalized and they do a bunch of tests. Eventually she’s put on steroids, which helps a little.

Her mother recalled that she did not return a “normal girl”. In September 2012, shortly after eating pasta, she presented crying spells, relevant confusion, ataxia, severe anxiety and paranoid delirium. Then she was again referred to the psychiatric unit. A relapse of autoimmune encephalitis was suspected and treatment with endovenous steroid and immunoglobulins was started. During the following months, several hospitalizations were done, for recurrence of psychotic symptoms.

Again, more testing.

In September 2013, she presented with severe abdominal pain, associated with asthenia, slowed speech, depression, distorted and paranoid thinking and suicidal ideation up to a state of pre-coma. The clinical suspicion was moving towards a fluctuating psychotic disorder. Treatment with a second-generation anti-psychotic (i.e., olanzapine) was started, but psychotic symptoms persisted. In November 2013, due to gastro-intestinal symptoms and further weight loss (about 15% of her weight in the last year), a nutritionist was consulted, and a gluten-free diet (GFD) was recommended for symptomatic treatment of the intestinal complaints; unexpectedly, within a week of gluten-free diet, the symptoms (both gastro-intestinal and psychiatric) dramatically improvedDespite her efforts, she occasionally experienced inadvertent gluten exposures, which triggered the recurrence of her psychotic symptoms within about four hours. Symptoms took two to three days to subside again.

Note: she has non-celiac gluten sensitivity.

One month after [beginning the gluten free diet] AGA IgG and calprotectin resulted negative, as well as the EEG, and ferritin levels improved.

Note: those are tests of inflammation and anemia–that means she no longer has inflammation and her iron levels are returning to normal.

She returned to the same neuro-psychiatric specialists that now reported a “normal behavior” and progressively stopped the olanzapine therapy without any problem. Her mother finally recalled that she was returned a “normal girl”. Nine months after definitely starting the GFD, she is still symptoms-free.

This case is absolutely crazy. That poor girl. Here she was in constant pain, had constant constipation, was losing weight (at an age when children should be growing,) and the idiot adults thought she had a psychiatric problem.

This is not the only case of gastro-intestinal disorder I have heard of that presented as psychosis.

Speaking of stomach pain, did you know Curt Cobain suffered frequent stomach pain that was so severe it made him vomit and want to commit suicide, and he started self-medicating with heroin just to stop the pain? And then he died.

Back to autism and gastrointestinal issues other than gluten, here is a fascinating new study on fecal transplants (h/t WrathofGnon):

Many studies have reported abnormal gut microbiota in individuals with Autism Spectrum Disorders (ASD), suggesting a link between gut microbiome and autism-like behaviors. Modifying the gut microbiome is a potential route to improve gastrointestinal (GI) and behavioral symptoms in children with ASD, and fecal microbiota transplant could transform the dysbiotic gut microbiome toward a healthy one by delivering a large number of commensal microbes from a healthy donor. We previously performed an open-label trial of Microbiota Transfer Therapy (MTT) that combined antibiotics, a bowel cleanse, a stomach-acid suppressant, and fecal microbiota transplant, and observed significant improvements in GI symptoms, autism-related symptoms, and gut microbiota. Here, we report on a follow-up with the same 18 participants two years after treatment was completed. Notably, most improvements in GI symptoms were maintained, and autism-related symptoms improved even more after the end of treatment.

Fecal transplant is exactly what it sounds like. The doctors clear out a person’s intestines as best they can, then put in new feces, from a donor, via a tube (up the butt or through the stomach; either direction works.)

Unfortunately, it wasn’t a double-blind study, but the authors are hopeful that they can get funding for a double-blind placebo controlled study soon.

I’d like to quote a little more from this study:

Two years after the MTT was completed, we invited the 18 original subjects in our treatment group to participate in a follow-up study … Two years after treatment, most participants reported GI symptoms remaining improved compared to baseline … The improvement was on average 58% reduction in Gastrointestinal Symptom Rating Scale (GSRS) and 26% reduction in % days of abnormal stools… The improvement in GI symptoms was observed for all sub-categories of GSRS (abdominal pain, indigestion, diarrhea, and constipation, Supplementary Fig. S2a) as well as for all sub-categories of DSR (no stool, hard stool, and soft/liquid stool, Supplementary Fig. S2b), although the degree of improvement on indigestion symptom (a sub-category of GSRS) was reduced after 2 years compared with weeks 10 and 18. This achievement is notable, because all 18 participants reported that they had had chronic GI problems (chronic constipation and/or diarrhea) since infancy, without any period of normal GI health.

Note that these children were chosen because they had both autism and lifelong gastrointestinal problems. This treatment may do nothing at all for people who don’t have gastrointestinal problems.

The families generally reported that ASD-related symptoms had slowly, steadily improved since week 18 of the Phase 1 trial… Based on the Childhood Autism Rating Scale (CARS) rated by a professional evaluator, the severity of ASD at the two-year follow-up was 47% lower than baseline (Fig. 1b), compared to 23% lower at the end of week 10. At the beginning of the open-label trial, 83% of participants rated in the severe ASD diagnosis per the CARS (Fig. 2a). At the two-year follow-up, only 17% were rated as severe, 39% were in the mild to moderate range, and 44% of participants were below the ASD diagnostic cut-off scores (Fig. 2a). … The Vineland Adaptive Behavior Scale (VABS) equivalent age continued to improve (Fig. 1f), although not as quickly as during the treatment, resulting in an increase of 2.5 years over 2 years, which is much faster than typical for the ASD population, whose developmental age was only 49% of their physical age at the start of this study.

Important point: their behavior matured faster than it normally does in autistic children.

This is a really interesting study, and I hope the authors can follow it up with a solid double-blind.

Of course, not all autists suffer from gastrointestinal complaints. Many eat and digest without difficulty. But the connection between physical complaints and mental disruption across a variety of conditions is fascinating. How many conditions that we currently believe are psychological might actually be caused a by an untreated biological illness?


15 thoughts on “Can Autism be Cured via a Gluten Free Diet?

  1. On the one hand, I was diagnosed with autism as a child and have had lifelong gastrointestinal problems. On the other hand, I’m rather attached my mode of thinking. It’s a bit disconcerting, like telling a cat lover their behavior comes from a brain worm.

    Liked by 1 person

  2. Those are some fascinating links: Autism – Gluten – Iron. Gut health has strong links to our mood, and probably more so the behaviour in a child. Someone might have a particular sensitivity to something like gluten and not even be aware of it, or how it affects them.

    Liked by 1 person

  3. I’m working through a review copy of Superbugs: The Race to Stop an Epidemic, and there have been several passages in the book which make me think that the things we often categorize as chronic diseases or diseases of old age might be ultimately caused by an infection, either years or decades earlier.

    Like the 29-year-old with an aortic aneurysm, which was caused by a Salmonella infection. Sure, this could be an unusual case, or it could only be unusual in that the back pain caused an aneurysm caused by a weakening of the aortic wall showed up 20 or 30 years sooner than it otherwise might have.

    People can readily survive Salmonella, but it is less clear whether something like this could cause effects seen far later, like Strep leading to rheumatic heart disease.

    Liked by 1 person

  4. Allana Collen’s book 10 Percent Human (the title is in reference to the now falsified idea that 90 percent of our body was made up of microbes if I recall correctly) has a solid chapter on this matter. I’ll go in depth tomorrow.

    Liked by 1 person

  5. This actually happened to me. We started our dietary investigation on the autism root because my son’s mental symptoms were, frankly, much more serious than his physical ones.. On top of that, we were attributing his thinness and the bags under his eyes to a consequence of his point-blank refusal to eat almost anything except (you guessed it!) bread and cake. He also wasn’t really able to communicate the stomach aches he was having.

    It was only a few months after we took him off gluten (having seen a miraculous change in behaviour in two weeks) that we started also seeing him gain more weight and looking physically healthier and then we went to a gastroentorologist who gave a probable celiac diagnosis. However, he was never really *tested* because testing involves weeks of gluten consumption and that’s something we can never do. He had a positive result on the celiac screening test, but that isn’t considered enough for a diagnosis. Because he subsequently got a negative result on the screen when we went super-ultra-double-plus strict about his diet, we are 99% sure. Our current pediatrician does say that she would have put him down as possible celiac based on his low iron levels, and low weight gain from age 1 and a half.

    That was bit rambling, but one more thing I would say about this is that people who say ‘autistic children are a gift’ or equivalent need to STFU. Considering what our life would be like if our uncontrollable, violent, insane 3 year old stayed that way into his teens literally gave me nightmares. One day, a classroom assistant gave him a piece of the wrong cake and the next month was a living hell.

    Liked by 1 person

    • Thank you for sharing your story. I am so glad you were able to figure out your son’s problem and fix it. Just imagine being three years old and feeling so sick. Poor little guy. I definitely understand what you mean about the diagnostic process.Really, once you know yourself that eating X is bad and avoiding X is good, more tests aren’t going to tell you anything new.

      Good luck avoiding all of the cake and congratulations on a healthy child.


    • Except how do you put an autistic child on any diet? What EvoX said about extreme pickiness is true. If you don’t give them the exact thing they want to eat, they will not eat. And then hunger makes them behave crazy and tear your nerves apart by hysterically bawling about 5 different things at the same time and it will not get better until the hunger is gone by giving them what they want to eat.

      If diet fixed it, I think it was not the real thing, I mean, autistic looking behavior sure but different causation.

      Because you see, wanting to eat only bread and cake is pretty normal. Carbs are fun. Ours only wants to eat raw vegetables. Yes. Cucumbers, bell peppers, tomatoes. Which is what most kids hate. So this is how we knew something is odd here. Refuses bread. Given a cake, maybe licks the glazing a bit but does not eat it.

      I think the logic is my mother wanted to be a good mom, good moms make kids eat vegetables so when it was time to wean her off pureed stuff and into real food she started with raw vegetables. So she ate it and then plain simply does not like change so does not want to eat anything else. Very repetitive behavior. Typical autistic trait.


      • Apart from bread and cake, he also ate porridge and raw vegetables. So for a few months he ate porridge and raw vegetables. After that he calmed down sufficiently that we could start introducing him to new foods. It’s been a slow process, but at least it’s a process).

        (I remember before that trying all sorts of methods to coax him into eating. We used to cut up some chicken and put it on a small plate next to him. We told him he didn’t have to try it, we just wanted him to get used to it being there. Read that one in a book. He picked up the plate and threw it against the wall.)

        Liked by 1 person

  6. >Sven’s teachers suggested autism. Several doctors later, he’d been diagnosed.

    WTF, WTF! The parents should have never believed that. Autism, even the mildest kind is something you absolutely notice waaay before the school age! We have noticed like 5 of such symptoms as arranging toys but not playing with them of our daughter way before 5 years old and the docs are still unsure, because if it exists, it is very mild. Mine is also very mild, and autism was not something talked about in my childhood, my parents had no idea and were not particularly well versed in things like this but I got still taken to a child neurologist at 3 years old because “he is odd”.

    Same thing with ADHD seen as a learning disability diagnosed at school. Are parents not paying attention anymore? A 4 or 5 years old kid should be able to dress themselves. If you find that between every item of clothing they completely forget they are supposed to be dressing themselves and start doing something else half naked, that is ADHD. Maybe – even that is not sure. If they ask you 10 times a day where is my puppet with which they were playing 10 minutes ago, and it is right in front of them on their desk, that may be ADHD. But if you have a perfectly normal kid at 4 or 5 and then at 6 the teachers tell you she or he has ADHD because behaving a bit less obediently than they want them to, tell them to go fuck themselves.


  7. So as a summary of both of my comments:

    – Docs massive overdiagnose all kinds of behavior as autism or ADHD which is not in fact so
    – You can fix it with diet
    – Because it wasn’t autism
    – Actual, even very mild autism you can’t

    Not only that extremely repetitive behavior (like peas only) makes it highly unlike you will change their diets, my n=2 sample, I know, not big, also shows really if mildly autistic children not have the usual, crab, gluten, sweets oriented preferences and their natural pickiness often excludes gluten. My daughter is basically a raw vegan, for me as a child the meal for lunch and dinner was for years on roasted chicken breast and canned corn/maize. Because I just didn’t want to deal with change and surprise and it was easy to prepare.

    Liked by 1 person

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 )

Twitter picture

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

Facebook photo

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

Connecting to %s