Chapter 13
Dietary Intervention for the Treatment of Autism:
Why Implement a Gluten and Casein Free Diet?
By Dr. Lisa S. Lewis


When my son was first diagnosed with autism (in 1991) my husband and I were both stunned, and in an odd way, relieved. Relief may seem like a bizarre reaction, but for almost two years we had been dealing with the unknown. It was terribly frightening and we never really knew if we were on the right track. Professionals were of little help. At Sam’s three-year checkup I asked our doctor point blank: “Could he be autistic?” At least he was honest. Shaking his head slowly, he replied, “I just don’t know.”

With a diagnosis, we had something to grab and run with. We both hold doctorate degrees in Anthropology, and as a result we were already trained to do research. Without discussing it, or formally dividing up what needed to be done, a division of labor seemed to occur naturally. As I look back on it now, I realize that doing research was our coping mechanism. We didn’t have time to cry and ask “Why us?” We sprang into action, devoting ourselves to doing something positive for our son.

My husband, Serge, immediately called a wonderful local organization called COSAC (New Jersey Center Outreach and Support for the Autistic Community) and set up an appointment. There we got valuable information on services available to us, our legal rights, whom we should call and where we should start, as well as a copy of Ivar Lovass’s Me Book. We signed up for a six-week parent training course due to start in a few weeks. Serge began looking into schools and other educational matters. (At that time Sam was in a half-day, “preschool handicapped” program that was not meeting his needs.) Armed with the information he had gathered, we were ready for our IEP (individualized educational plan) meeting, where we presented a case for moving Sam to a more specialized (and far more expensive!) school.

While Serge was looking into education and therapeutic interventions, I began an extensive search for anything I could find on autism—medical and other interventions, outcome and etiology. Fortunately for me, the World Wide Web was just then “taking off,” and I was able to do much of my research without leaving my desk. As a (then) employee of Princeton University, I was able to access anything on-line quickly, and at no expense. I had already learned to use the basic tools of electronic research for my work at the university, but now I really honed these skills to find specific information.  I found a lot, and I also found how many other people were out there “seeking.” I began extensive correspondence with many other parents, and with some notable professionals too, including (the late) Roland Ciarnello who was running the Stanford University Genetics Research program on Autism. Dr. Ciarnello was a big help, and sent us several his articles when we were looking for specific information.

We combined the results of our research to come up with a plan, which included where to have a good educational evaluation done on Sam, where to send him for speech and occupational therapy and what kind of school to put him in. Our school district was helpful and supportive, and gave us most of what we asked for. Still, we continued to search. At that time, there were few Internet news groups on the subject, and locally, we only found “support” groups that stressed coping with the diagnosis rather than research or treatment. After reading much of the literature on the subject we weren’t surprised by the focus of these groups. Autism was, after all, described a lifelong developmental disability, intractable to most medical treatment.

When Sam was three, he began having violent tantrums. I had read that many children who were allergic to milk behaved that way. He loved milk and any milk products and consumed lots of milk, yogurt, cottage cheese and ice cream. And why not? After all, I was raised to believe that milk was the perfect food! I never loved it but my mother insisted that a certain amount be consumed every day. But as Sam’s tantrums intensified I started reading about allergies, and found that children with food allergies generally craved most, those foods that they should not have.

I removed dairy from Sam’s diet, and while he did seem to settle down a bit, there wasn’t really a profound difference. I still suspected that something he ate was affecting him, and his pattern of recurrent ear and upper respiratory infections which eventually led to asthma, seemed to indicate an immune system gone awry. We took him to a highly recommended pediatric allergist in Philadelphia. Her tests confirmed that Sam was very allergic to pollens and molds, but she found no evidence of food allergy. That was the end of dietary intervention—for a while.

Two years later, when Sam was five, I saw Dr. Doris Rapp on a talk show. She spoke about environmental allergies and food allergies, which I found very interesting. But then she showed videotapes of children who were given concentrations of foods to which they were allergic. These nice children suddenly became wild animals! Screaming, attempting to hit or scratch, throwing tantrums and worse. Milk was said to be one of the main allergens that produced this response. Wheat was also named as a common cause culprit.

At this, a bell went off for me. In addition to his love of dairy, Sam loved crackers, bread, rolls, pretzels—anything starchy and most foods made of wheat. We had been calling him “Carbo-man” because he so loved these (mostly) wheat-based foods. I could not help noticing the resemblance between those videotapes, and the behavior I had been witnessing and dealing with in my own son. Though he had tested negative for a wheat allergy, I decided then and there to remove it from his diet.

Within three days, I began getting notes home from school, saying that Sam was doing beautifully and that his behavior was enormously improved. What had we done? I decided not to reveal the removal of wheat at this point. But two weeks later I was eating a slice of pizza. Sam came by and asked for some. I was watching something on television, and was preoccupied. I mindlessly handed him a slice and only after he was halfway through did I realize what I had done. Oh well, I thought, the damage (if any) was already done.

The next morning when Sam asked for a Pop-Tart® I went ahead and gave it to him. Then I waited. At 4:00 his bus pulled up and even before he was out his aide said, “I hear he had a pretty rough day.” I barely looked at Sam, instead grabbing his backpack to find his notebook. As I had suspected, the note was not good. During the course of the day, Sam had numerous tantrums, had been extremely aggressive and very unfocused. Only late in the day did he start to come around, telling his teacher “don’t write in the book” realizing that I would read of his many transgressions!

From that point on, I explained to his teachers and therapists what we were doing (no more pretzel reinforcers during speech therapy!) I began sending wheat free lunches and snacks, and wrote up a long list of what he should not be given at school.

Though I was completely convinced that Sam was greatly affected by what he ate, I was still puzzled over why he tested negative to wheat and milk allergies. I mentioned this to an “electronic friend” (someone I’d met on the Internet, and subsequently in person) named Jean Jasinski. She recalled reading an article about autism and gluten intolerance. I vaguely remembered something about it too, but couldn’t remember the authors or where I had seen them. Bless Jean—she searched until she came up with one article, which she mailed to me. I then was able to find other articles by the author, Paul Shattock of the University of Sunderland in Sunderland, England. Shattock’s articles in turn led me to the work of Dr. Karl Reichelt in Norway.

After reading about the research going on in England and Norway, I came to realize that I should have removed all gluten grains from Sam’s diet, rather than just wheat. I did this, as well as I could. It was very hard to manage since it was new to me, but I found information about celiac sprue and went from there (more on this below.) I did not see the huge change in Sam that I’d seen after first removing wheat, but this seemed natural to me. I had already removed the grain that had the highest concentration of gluten proteins; perhaps I had succeeding in weaning him from the offensive proteins.

Two years later I found Mr. Shattock on-line. I sent him e-mail immediately, and so began a correspondence which continues to this day. He and his colleagues have proved tremendously helpful to me.

So many people began asking me for information, that I wrote a twenty-page document on how and why to try this intervention. Before long, I was spending a lot of time and money duplicating the article and mailing it to people. There had to be a better way.  At that point, many more people had Internet access from their offices or homes, and the World Wide Web was becoming widely available. Since nearly all requests for the article were coming to me via e-mail, I decided to put this document up on the web. I created a home page, and announced it to the autism world, and what a reaction it has had.

I began to receive phone calls, letters and e-mail from all over the world. From these contacts I met many more parents with whom I began exchanging information. It was from this and references from friends who had read my article, that I made contact with many of the “seeker parents” out in the world.

At first, when I began talking and writing on this topic, I was thought to be something of a “nut.” For the last year or so, however, the subject of diet is on “the net” constantly, and even doctors have begun taking it seriously. It may not cure, and it may not even help all that try it, but it will help many. I hope the information provided in this book and in this chapter will help you to decide whether or not to try new interventions for your child. I hope that I can provide you with information on why it might be a useful experiment. And if you do want to try it, some information is included that will help you get started.

 What is Gluten? Why Eliminate it from the Diet?

Gluten is a protein found in the Plant Kingdom Subclass of Monocotyledonae (monocots.) These plants are members of the grass family of wheat, oats, barley, rye and their derivatives. Derivatives include: malt, grain starches, hydrolyzed vegetable/plant proteins, textured vegetable proteins, grain vinegar, soy sauce, grain alcohol, flavorings and the binders and fillers found in vitamins and medications. Casein is a phospho-protein of milk, which has a molecular structure that is extremely similar to that of gluten.
In the early 1980's, two scientists noted that the behavior of animals, under the influence of opioid drugs such as morphine, was very similar to that of some people with autism. Dr. Jaak Panksepp proposed that people with autism might have elevated levels of naturally occurring opioids in their Central Nervous System.  There are several such naturally occurring compounds. The best known of these are the beta-endorphins which produce the so-called “runner’s high.”

At about the same time, work by Swedish autism expert Christopher Gillberg showed elevated levels of "endorphin-like substances" in the cerebrospinal fluid of some people with autism. It is particularly interesting that levels are high in those children with autism who are insensitive to pain and those who engage in self-injurious behaviors. Dr. Karl Reichelt found abnormal peptides in the urine of people with autism; these peptides are apparently similar to those found by Gillberg. The Autism Research Unit later replicated Reichelt’s findings at the University of Sunderland, under the direction of Paul Shattock.

According to Shattock, “In the urine of about 50% of people with autism there appear to be elevated levels of substances with properties similar to those expected from opioid peptides.”  Because the urinary compounds greatly exceed what could be possibly be of CNS origin, it is presumed that they result from the incomplete breakdown of certain foods. 

Proteins consist of long chains of amino acids. Normally intestinal enzymes digest them, breaking the bonds that connect the protein’s amino acids. Genetic mutations, caused by changes in the DNA, can mean that specific enzymes cannot do their work.

Enzymes are also proteins; they are long chains of amino acids that fold into specific three dimensional shapes.  Each enzyme has an active site into which the protein it is designed to digest can fit. An alteration in the gene that codes for a particular enzyme can mean that it folds in new way, and the protein to be modified no longer fits into the active site. “Mutations...can change the chemistry of the body by preventing or altering the way certain enzymes and chemical reactions work” (Comings, 1990). 

In this case, an incomplete digestive process would leave amino acids bound into short chains called peptides. If the peptides still have biological activity—that is, if they still function as opioids—they could result in symptoms we see in autism. Most of the peptides would be dumped harmlessly into the urine, but if a portion escapes the gut and enters the bloodstream, they could cross the blood-brain barrier and cause serious neurological problems.

Two commonly ingested proteins are known to break down into peptides that have opioid activity. Casein, a protein in cow’s and goat’s milk, breaks down to produce a peptide called casomorphine, and gluten from wheat breaks down to form gliadinomorphins. The amino acid sequences of these two molecules are extremely similar, which is why the elimination of both gluten and casein is usually recommended.           

If one lacks the ability to break down these proteins appropriately, there must be a strategy for reducing the effects of the resulting opioids to minimize neurological effects. One approach would be the anti-opioid drug "naltrexone." Naltrexone has shown very mixed results, however, and there are some difficulties associated with its administration. (Dr. Shaw’s note: See Dr. Semon’s chapter for effective dosing of Naltrexone to supplement dietary changes.) Finding the optimal dose has proved difficult, and it is a very bitter pill, which most children will resist taking.  A second approach is excluding casein and gluten from the diet.

Many parents have had traditional allergy tests run, and most report that their children are not allergic to wheat or milk. This is probably true.  Children who are helped by this diet are generally not allergic in the traditional sense; they are gluten or casein sensitive. According to Shattock, “The results are akin to poisoning rather than an extreme sensitivity such as occurs in coeliac disease or sensitivity to certain food colourings.”

Many children suffer an initial bad reaction to the removal of their favorite foods. Often, these children seem nearly addicted to a specific type of food--often consuming large quantities of dairy or wheat products.  Some children do very well for a few days, then suffer a regression. According to Reichelt, this bodes well for the success of the intervention. Once this period passes, it is generally followed by a good response. Younger children are more likely to benefit dramatically from this intervention, but adults have also been noted to have improved concentration and communication, as well as lessened sensory scrambling.
When I first learned of this research, Dr. Reichelt was working with many families in his native Norway. This work convinced him of the diet’s efficacy.  He was the only person that I knew of who had data from tests of these theories, and I wanted to know more. I tracked down a fax number for him and sent him a letter requesting more information. I included my e-mail address in my fax, and was delighted to receive a response via e-mail, in just two days. In this mail Dr. Reichelt said:

“In general we recommend a diet free of gluten and casein for autistic...patients. The reason for this is that opioid peptides from gliadin are almost of the same structure as casomorphins from casein. We also recommend addition of multivitamin with trace minerals and magnesium, cod liver oil and calcium. We usually remove both casein and gluten. Opioids from these proteins are very similar."

He ended with: "Effects of diet if useful, tends to be cumulative. Must be tried for 1 year."

 Further Research

Mr. Shattock, along with colleague Dawn Savery, has recently written a paper that brings their work on this topic up to date.  At the time of writing, Shattock and Savery had examined urine samples from nearly 1,000 subjects. While little other information about the subjects was collected initially, the study is now more formal and involves the collection of much more detailed behavioral and other information.

The theoretical model on which their study is based remains the same, relying heavily on work by Reichelt and colleagues (Knvisberg and Waring.) To summarize:

...autism could be the consequence of the action of peptides of exogenous origin effecting neurotransmission within the Central Nervous System (CNS.) We believe that these peptides result in effects which are basically opioid in nature....The CNS neuroregulatory role which is normally performed by the natural opioid peptides...would be intensified to such an extent that normal processes within the CNS would be severely disrupted.
The presence of this intense opioid activity would result in a large number of the systems of the CNS being disrupted....Perception; cognition; emotions; mood and behaviour would all be affected. ...Many and diverse symptoms by which autism is...defined would result. We believe that these peptides are derived from an incomplete breakdown of certain foods, and in particular, gluten....and from casein.

--"Autism as a Metabolic Disorder,"
Paul Shattock and Dawn Savery (1997)

Any time proteins are broken down in the gut, peptides result; they are intermediate compounds which should then be broken down further into their amino acid components. In all individuals, a proportion of these may cross from the intestines into the bloodstream, and hence, cross the blood-brain barrier. However, if the gut is "leaky" then the proportion of improperly broken down peptides that cross this protective barrier will be far larger, with potentially devastating consequences. [See discussion on Phenol sulfur Transferase below.]

 Sam's Story

I gave an overview of what we did when our son was diagnosed, but here is a little more background about him, and his family.

Sam is now nine years old, and was diagnosed as PDDNOS, pervasive developmental disorder-not otherwise specified, a form of autism at age three and a half. We believe that his development was normal for approximately the first 18 months of his life, but by the age of two and a half he was in an early intervention program. Country officials who ran this program never gave us a specific diagnosis, saying only that he had "sensory integration difficulties." This is certainly true—he was significantly delayed in motor planning, he had a poor sense of where he was in space he had “tactile defensiveness”—refusing to touch textures such as shaving cream, finger paint or sand. But as I read the little that was available about the subject, it seemed clear to me that sensory integration problems were likely a symptom of something else, rather than the cause of his many problems.

By age three Sam was in a multiply-handicapped half-day preschool, and was receiving private speech therapy. Though he had language from an appropriate age (13 months), by two it was far behind that of peers and was characterized by (appropriately placed) echolalic utterances. When Sam was three and a half, a neurologist at the University of Medicine & Dentistry of New Jersey (at Rutgers University) finally confirmed our suspicions that he might be autistic.

At this time we began doing the research outlined previously, and we sought an independent educational evaluation at the Eden Institute in Princeton, New Jersey. A placement more specific to autism was recommended, and Sam was accepted at the Douglass Developmental Disabilities Center (DDDC) for the next year.

Because DDDC is a data based program, we have hard data on what happened when wheat was removed from Sam’s diet. After five days, Sam's aggressions had dropped dramatically, from double digits over the course of a five-hour day to an average of 6.1/day. During the month he was on vacation, Sam did not aggress at all. When he returned to school in September, his aggressions dropped further, averaging 2.47/day over the next seven months.

Reduced aggression was not the only change we noted when dairy, wheat and gluten were removed from Sam’s diet. For two years we had struggled with Sam's reversed pronouns. We did drills at home; he worked on it at school and with his private speech therapist. Within one week of his dietary change, his use of pronouns was suddenly and completely correct. His attention span increased and he responded more quickly to lessons at school, home and in private speech and occupational therapies. His speech therapist referred to him as her "one-trial learner."

In the spring of 1994, we visited Dr. Sidney Baker of Westport, Connecticut. He placed Sam on a strict with anti-yeast diet coupled with high doses of the anti-fungal drug Nystatin. We were told that before seeing any improvement we might first see a regression. Sam did have a regression that lasted for three weeks, with terrible behavior but no loss of previously attained skills. After a few weeks, Sam's behavior normalized and his aggressions gradually decreased to a level of 2.47 day during the remaining school year. I did not see a marked improvement as a result of this treatment, however, which was a disappointment. I later had the organic acids test performed by Dr. Shaw, which indicated that a new trial of anti-fungal medication might be in order. Since the test has been expanded and refined, we will likely do it again and try, if necessary, an antifungal other than Nystatin.

Dr. Baker also ordered extensive testing of blood, urine, saliva and stool. While most were normal, Sam was deficient in eight amino acids, and low in zinc. In order to put Sam's system into better balance, we added a vitamin compound, additional calcium and zinc to Sam's daily regimen.

I later added molybdenum and magnesium, as well as essential fatty acids (evening primrose oil.) At various times I have also experimented with DMG, L-Carnitine, a (milk-free) acidophilus powder, extra inositol, pycnogenol and octocosanol. Since I never saw any benefit from these additions, I have not continued to use them on a regular basis. I do know of many children who have responded very well to some of these compounds, and I retry them on a period basis just in case something has changed in his system that might make them helpful.

Sam's diet and nutritional supplements are certainly not the only things that have helped him. He spent four years at an excellent special school, and attended weekly speech and sensory integration therapy for five years. For two years he wore yoke prism glasses and did visual therapy prescribed by Dr. Melvin Kaplan of Tarrytown, New York. The prism glasses helped Sam to focus his visual attention for longer periods (though this remains a problem) and it also put an end to his habit of looking at the world using primarily peripheral vision (i.e. squinting his eyes and turning his head and looking from the side.)

However, the change after removing wheat was both remarkable and undeniable, as is what happened on the occasions that he accidentally ingested gluten. On several occasions Sam has eaten gluten without our knowledge, and the changes in him were fast and quite marked. In each case we were able to determine what had caused the sudden, and thankfully short-lived, regression.

While I cannot be sure what has helped Sam the most, I do have a daily record of his behavior and many of his utterances dating back to when he was three. I can therefore correlate changes with particular interventions. Because autism is likely a disorder with multiple etiologies, it is doubtful that every person with autism would benefit from this diet. I believe strongly, however, that the approach would help many children. Indeed, in the three years since I first began to write on this subject, I know the diet has helped thousands that I am aware of and no doubt countless others who have never contacted me. It is certainly worth trying.

For a child with a limited diet, I would start with lab tests to determine if he is likely to benefit from the diet (see below). All parents of children with very limited diets want to broaden the food choices the child accepts. However, if positive test results show that gluten and or casein could be causing damage to the CNS, changing the diet is critically important. NOTE: tests will not be valid once the child has had gluten and/or casein removed from his diet for any length of time.

Because Sam responded so well to a GF (and greatly reduced casein) diet, I feel frustration that more parents have not been willing to try this diet. However, I also know that I am lucky.  My son is not a fussy eater, and accepts the various substitutes I provide for him. He can now monitor his own diet to a certain extent, refusing "regular" bread or cookies. He also eats a wide variety of foods, much of it healthful. He is now able to swallow pills, and can take the vitamin supplements I give him without trouble.

 Jake's Story

Sam's brother Jacob was born when Sam was three, and we watched his development very carefully. He was only three months old when we had a definitive diagnosis for Sam. I tried to remain calm about his development, but when Jake was nine months old I began to worry in earnest.

Jake showed little pre-verbal development; he did not babble and he made very few sounds. While his pediatrician understood the source of my concern, he maintained that it was far too early to see anything, and that I should just keep a close eye on him. At this same visit, the pediatrician told me that Jake was ready for cow’s milk. I was working full time and under a great deal of stress. My own milk supply had waned and to keep him happy I had added formula to his diet. I was thrilled to throw away the breast pump and the nasty smelling formula supplements. I bought some whole milk and Jake drank it with gusto.

Cow's milk, however, seemed to cause an immediate change in Jake.  He got fussier and had more stomach upsets. Within a day I knew that this child was not ready for cow's milk. I immediately went back to the store to buy formula. Then, in a moment that in retrospect seems like an epiphany, I bought soy formula instead. Within two days Jake was happier than he had ever been. Within three days he was saying "mamamamam" and "dadadada".

On his first birthday Jake had about ten words; by 15 months he had 200; by 18 months he spoke in sentences. He has continued to develop into an incredibly imaginative, verbally precocious little boy. At five, Jake calls himself a scientist and is fascinated by all sea creatures (especially sharks.) He wants to be an underwater photographer and to make nature films, but agrees that perhaps kindergarten should come first. He was four before he had cow’s milk products again, and by that time they had no effect on him so they are back in his diet to a limited extent. In addition to the joy he has brought to his parents, Jake is the best "therapist" Sam ever had!

Did I "save" Jacob from autism by removing a potentially harmful protein at a vulnerable age? Did I save him from some other developmental disability? Of course, I will never know.

 Testing for Urinary Peptides

Because modification of the diet is far less invasive or harmful than most interventions, it would seem logical to try this method. Many children with autism, however, have such finicky eating habits that the idea of cutting anything they will actually eat out of their dietary repertoire, strikes fear the hearts of their parents. For this reason, some might prefer to test their child's urine for the presence of the urinary peptides found by Reichelt and others. If there are no peptides found, it is unlikely that the diet would help the child. However, if the peptides are present and are escaping from the gut into the bloodstream, it is believed that they can "mimic" neurotransmitters and thus result in the scrambling of sensory input.

If you have already tried the diet you will not learn anything meaningful from the urine test. By eliminating gluten and casein from the child's diet, you have removed the source of the peptides. It can take a long time to build them back up to pre-diet (baseline) levels, and this is not advisable, especially if the diet has proven helpful.

Dr. Cade, an autism researcher now deceased, has had some success with adding prescription strength digestive enzymes to the regimen of children with autism. These enzymes apparently help break down proteins in individuals whose digestive system is not functioning properly. While it may not mean that the child can start eating whatever he or she would like to, it can be one more weapon in our treatment arsenal.

 Testing for Celiac Disease

What exactly IS Celiac Disease (CD)?

“Celiac disease (also known as Celiac Sprue or gluten-sensitive enteropathy) is a chronic disease in which malabsorption of nutrients is caused by a characteristic...lesion of the small intestine mucosa. The lesion is produced, through unclear mechanisms, by protein constituents of some cereal grains"(J.S. Trier, 1993). Traditionally, doctors have suspected celiac disease only when patients show poor growth, extreme gastrointestinal problems and fatty stools. It is now known, however, that many patients with sensitivity to gluten serious enough to damage the gut wall show no such symptoms!

In people who have celiac disease but continue to eat gluten, the intestinal wall is excessively porous; not only are nutrients improperly absorbed, but large molecules which should be contained by the gut wall are not. This could be the way in which improperly digested peptides pass into the bloodstream and then cross the blood-brain barrier. Thus, the speculation that celiac disease is present in some children with autism who would benefit from a gluten free diet is not inconsistent with the opioid excess theory of Reichelt and Shattock.

Various experts on autism long ago dismissed the idea that gluten could be a significant causal factor. However, gluten exists as a "hidden ingredient" in many foods, medicines and even in the envelope glue we lick. It is possible that children with autism children put on a so-called gluten free diet were inadvertently ingesting gluten in small amounts.

For those with full blown celiac disease, tiny amounts of gluten can be toxic; it is not so far fetched to imagine that in less severe forms of gluten intolerance, minute amounts could also cause harm. When full blown celiac disease is diagnosed, it can take more than a month on a gluten-free diet to see changes; again, it is not far fetched to assume that the same is true for people with gluten intolerance that have different outward symptoms.

It may be then, that early researchers and parents who tried this intervention in the past simply gave it up too soon. Patients with full-blown celiac disease often have terrible symptoms of gastrointestinal distress, fatigue, and failure to grow or gain weight. These kinds of symptoms cannot be ignored, and the diet is changed when the child is relatively young. But it is possible that far less severe forms of CD exist and are, in fact, quite common. If so, they could go undiagnosed for years. Undiagnosed, the toxic effects of the ingested gluten could prove extremely damaging and could cause what is likely to be permanent damage to the central nervous system.

According to an article by Dr. Allessio Fasano in a 1994 newsletter of the American Celiac Society:
In recent years there has been a noticeable change in the age of onset of symptoms and the clinical presentation of celiac disease. Because the typical symptoms of gastrointestinal dysfunction are frequently absent in older children, the diagnosis beyond the first two years of life is more difficult and often delayed. These cases are now regarded as having atypical or late onset forms of celiac disease.
Rimland and Meyer noted as long ago as 1967 that children with the highest score on Rimland’s E-2 Diagnostic Checklist also showed many gastrointestinal symptoms. It has also been suggested that celiac disease is an autoimmune disorder with gluten stimulating increased synthesis of some antibodies in celiac disease patients. Ruth Sullivan noted that "though few children with celiac disease have autism, it seems a disproportionate number of children with autism have celiac. Why?

Does malabsorption of the small intestine prohibit vital substances (like serotonin...) from reaching the brain? If so, why do not all classic cases have celiac? Or do they? (1975)

A disorder very closely related to celiac disease, and necessitating the same dietary intervention, is a skin disease known as dermatitis herpetiformes (DH). According to the newsletter of the American Celiac Society, "Dermatitis herpetiformes is the skin manifestation of gluten sensitivity and 70-80% of DH patients have coexisting damage in the intestine." In many cases DH sufferers have no outward signs of intestinal difficulty, and yet at least 70% actually do suffer from celiac disease! DH appears as a bumpy rash, usually on the arms, legs or buttocks. It is extremely itchy and may also burn.

Interestingly, Sam had such a rash on his arm and inner thigh. This rash first appeared at approximately age 2 (around the age his autistic symptoms also appeared) and was diagnosed by our pediatrician and two dermatologists as severe eczema. All prescribed cortisone creams but the rash did not improve. It was so itchy that my son would frequently scratch until he bled. We removed all synthetic fibers, dressing him in only 100% cotton washed in soap that had no colors or dyes. Nothing helped.

Then, as mysteriously as it appeared, the rash went away. Around the time that I changed Sam’s diet I also began giving him evening primrose oil, which was said to help eczema. I credited the oil and bought several bottles. Then I stopped using it and the rash did not reappear. I now realize that the cause of the improvement was probably not the oil, but rather the removal of gluten from Sam's diet! Though I cannot have the tests run because he is been off gluten too long) I am convinced that he was likely showing signs of DH, which were unrecognized by the doctors who saw it. [Note: I later realized the importance of evening primrose oil for the essential fatty acids it provides, after reading Leo Galland’s excellent book, SuperImmunity For Kids. I then reinstated the oil into Sam’s daily regimen.]

New blood tests show latent and sub-clinical cases of celiac disease. Because even latent celiac disease will cause damage to the intestinal wall, it makes sense to have these tests run. The relevant tests involve screening the blood for celiac antibodies. The tests are called endomysial IgA, gliadin IgA and reticulin IgA. The blood test can rule out or suggest Celiac Disease.  If celiac disease is not ruled out, a diagnosis still cannot be made. Celiac disease can only be positively identified via intestinal biopsy. If a gluten free diet has already been implemented, these tests will not be valid. While these tests will not reveal a possible sensitivity to casein, they should certainly be done on children who developed normally for up to two years (and who are thus more likely sensitive to gluten). Not all labs are equipped to run these tests. If a local lab cannot do it, you might want to contact Specialty Laboratories, Inc., in Santa Monica, CA at 310-828-6543.

Although no child will willingly donate blood, all four tests can be performed following a single draw. While it is doubtful that all people with autism will turn out to have celiac disease, these tests should be performed to rule it out.

Certainly celiac disease causes a leaky gut; if various proteins are being improperly metabolized, such a gut would provide a pathway into the bloodstream for these peptides. Clearly these tests should be added to the battery that children undergo when a diagnosis of autism, PDDNOS or atypical autism is made.

Intestinal Permeability tests also exist, and should be performed, if possible [see section on the DAN! protocol below.] This test requires a patient to ingest a sweet drink provided by the lab performing the test, then eat nothing for several hours. This is followed by a collection of all urine for the next 24 hours. This test must be ordered by a doctor, and will show whether or not the patient has a "leaky gut." If the child is not toilet trained, a bag (obtainable from your doctor) can be taped used to collect urine at each diaper change.

 Phenol-Sulfur Transferase (PST) Deficiency

Preliminary studies by Rosemary Waring, of the University of Birmingham, UK, suggest a particular enzyme deficiency in many children with autism. This abnormality affects the sulfur-transferase system, which is one of the body’s major means of detoxification. In a recently published book, Dr. Sidney Baker describes this system very succinctly:

This system helps us get rid of leftover hormones, neurotransmitters and a wide variety of other toxic molecules. Some such molecules come from our own metabolism, like leftover hormones and neurotransmitters, and some come into us with our food or are made by the germs that live in our intestines (--Detoxification & Healing; The Key to Optimal Health, 1997).

With insufficient PST, individuals have an extremely low capacity to oxidize sulfur compounds. Children with this enzyme deficiency are unable to fully metabolize certain foods and chemicals that contain phenols and amines.

As stated by Dr. Baker, PST is necessary to break down hormones, some food components and toxic chemicals. If the enzyme is deficient, the body cannot detoxify the system—that is, it will be unable to render these substances harmless. Harmful substances that should be metabolized would build up to abnormal levels, substances which include serotonin, dopamine and noradrenaline. Many metabolic processes can be disturbed by phenolic compounds and cause many physical problems that may not have been previously thought connected to autism (excessive thirst, night sweating, facial flushing, reddened ears etc.) 

The children most likely to show this deficiency (based on Waring’s small sample size) showed normal development for the first 18 months to two years of life, and also show family histories of asthma, skin problems and migraine, as well as sensitivity to foods (especially wheat, milk and salicylates.)

There are some tests that can identify whether an individual has a weak detoxification pathway, however, normal levels have not been established for children under the age of twelve. Dr. Waring has a working test for children, which uses acetaminophen (Tylenol®) as a "probe" for finding weakness in the PST system. Testing does require a 24-hour collection of urine, a nearly insurmountable difficulty if a child is not reliably toilet trained. For more information, Dr. Waring can be contacted at: The School of Biochemistry, University of Birmingham, Edgbaston, Birmingham, B15 2TT England. Dr. Waring does not currently have Internet access. Dr. Robert Sinaiko, a San Francisco specialist in Allergy and Immunology, is working on perfecting a test in this country. Hopefully, such a test will be available soon.

Unfortunately, there is no standardized, recommended treatment for PST deficiency. Two approaches may be taken—you can try to increase the body's ability to detoxify itself, or you can try to decrease the toxic load you subject it too. Neither approach is particularly easy or 100% effective. To quote Developmental Delay Registry Founder and nutritionist Kelly Dorfman: 

Some parents have used diets that remove all known phenol compounds (such as Sara's Diet) to take pressure off the PST…system. While sometimes helpful, these diets are extraordinarily difficult to implement long-term as naturally occurring phenols are in every food with color. Except in extreme cases, a diet reducing toxic load form the most concentrated sources...appears to be the best.

That is, reduce juices (or limit to pear juice) and eliminate all artificial colors and flavors.

Unfortunately, no amount of intervention...can totally unburden PST…enzymes....That is why it is critically important to improve the efficiency of the faulty enzyme system while attempting to lessen the load. Several nutrients may help. They include vitamin C, vitamin E, reduced L-glutathione and N-acetylcysteine. All of the antioxidants (including selenium and bioflavonoids) are valuable for detoxification in general.
            -From New Developments, Winter 96-97, a DDR  publication.

Autism researchers have been intrigued by the fact that the PST deficiency can cause the improper metabolism of some neurotransmitters (serotonin, dopamine and noradrenaline.) It has been known for years that people with autism often have abnormal levels of serotonin, as least as is measured in the blood. But the buildup of serotonin may be less significant than another outcome of a PST deficiency—namely, the effect this deficiency would have on the permeability of the intestinal lining.

One outcome of an improperly operating sulfur-transferase system is insufficient connective tissue in the gut wall. Thus, a PST deficiency could be yet another reason (besides Celiac Disease and other gastrointestinal ailments) that the gut wall would be "leaky." As stated above, when improperly metabolized proteins (such as gluten or casein) are able to escape the gut lining into the bloodstream, they can cross the protective blood-brain barrier.

I noted above that my son’s urinary amino acids tests revealed a deficiency in eight amino acids. Five of these are sulfur-carrying amino acids. Dr. Baker informs me, that this is a pattern he sees very frequently in autistic patients. It will be interesting to follow Dr. Waring 's research to determine whether or not there is a relationship between her theories and the deficiencies he finds. Because the sulfur-carrying amino acids are involved in the detoxification of the body of both exogenous and endogenous pollutants, disturbances in these systems indicate disturbed immune systems.

Considering how frequently these children suffer from numerous infections and allergies, this is not an unlikely assumption. In some parts of the country immunological approaches are being taken with some benefits to children with autism, and it is possible that for some the cause of autism may be an autoimmune disorder.

Though it cannot yet be proven, there is good evidence that a diet that eliminates gluten and or casein may indeed be beneficial. In an unpublished (1993) manuscript, Waring and Reichelt state "We think that the demonstrated peptides may be central to the aetiology of the disease. Exorphins not only increase social isolation in animal models, but may cause CNS inhibition of maturation." Another observation is equally intriguing: "...because most bioactive peptides are found in different chain lengths, but with very similar activity, different peptidase defects would cause similar but not identical symptom profiles and peptide profiles." They believe that this indicates that such "effector peptides" would be the "final common path of several clinical subtypes involving different lengths of peptides. It would also suggest that other diseases may show autistic symptoms if peptides are involved, as is seen for coeliac disease."

If you are convinced (or become so after having tests run) that this dietary intervention is worth trying with your child, you have your work cut out for you. Even a very good cook must relearn how to shop, how to plan menus and how to cook.

For many of the children who respond positively to dietary intervention, gluten and casein are not the only problematic foods.  Some children react to corn, soy and eggs, in addition to gluten and casein. Some can eat dairy but not gluten. Some can eat gluten, but no dairy or eggs. Dietary intervention is time-consuming and tedious because you must be systematic in determining which foods cause problems for a given child. But if your child responds, it is worth the trouble.

 Where to Go for Help with a Gluten and Casein Free Diet

There is a large population of celiac sufferers in this country; they are experienced in food substitutions and can be a great source of information for people trying to avoid gluten. Five organizations that have newsletters containing lots of valuable information are:

American Celiac Society
201-325-8837 (New Jersey)
Celiac Sprue Association, USA
402-558-0600 (Omaha, Nebraska)
Gluten Intolerance Group of Florida
Orlando, Florida 32837 
Gluten Intolerance Group
P.O. Box 23053
Seattle, Washington
Celiac Disease Foundation
13251 Ventura Blvd. Suite #3
Studio City, CA 91604

The Gluten-Free Baker Newsletter is published quarterly, and gives recipes for sweet and savory baked goods. 361 Cherrywood Drive, Fairborn, Ohio, 45324-4012
If you have Internet Access, there is a Celiac List, which has a great deal of useful information. For information on subscribing to this list, send e-mail to You will also receive instructions on how to access the archived recipes posted over the last few years.

The Autism Network for Dietary Intervention (A.N.D.I.) publishes a quarter newsletter. For more information write to The ANDI News, PO Box 77111, Rochester, NY  14617-0711 or send e-mail to

 Some Good Cookbooks 

Because so many people have asked for more details on the information contained in this chapter, I have recently completed a book on the topic. Over half of the book is devoted to actually implementing the diet, and it contains over 100 gluten and casein recipes. See the end of this chapter for more information.
The Gluten Free Gourmet, More From the Gluten Free Gourmet and The Gluten Free Gourmet Cooks Fast and Healthy, by undisputed GF Guru Bette Hagman, are all published by Holt. All are excellent. Each has over 200 gluten free recipes for bread, cookies, pizza, chicken pot pie, cakes etc. It's also full of advice about adapting regular recipes and what to use as substitutions. If you can buy only one cookbook, make it one of Bette Hagman’s.

Other useful and excellent cookbooks include:  Allergy Cooking With Ease by Nicolette Dumke and The Allergy Self-Help Cookbook by Marjorie Hurt Jones. For those who are limiting yeast, The Candida Control Cookbook by Gail Burton is a very good source of recipes.

Full of Beans by Kay Spicer and Violet Currie has recipes using beans and bean flour. These "odd" ingredients make wonderfully moist and delicious baked goods. No-Gluten Children's Cookbook by Pat Cassidy is available for $25.50 from RAE Publications, PO Box 731, Brush Prairie, WA 98606. The Practical Gluten-Free Cookbook by Arlene Stetzer is available from Main Street Systems (608) 534-6730.

There are MANY others-check bookstores and libraries! For Web surfers, visit

 Where Do We Go From Here?

Everyone agrees that autism is a puzzle. It seems at times, that all the pieces are black and we are trying to put it together in the dark!  Everyone has to do what he or she thinks is best for his or her children, and for their families. For forty years parents have been given false hopes and empty promises. There are still few definitive answers, and there are still lots of promises being made.

I hope that the reader of this book does not believe that special diets fall into this category. I do not believe that dietary intervention will "cure" children with autism (or at least, not very many.) But it can help, and it can help a lot. For those children who respond to dietary changes, the ones who respond the most will likely be the very young ones. The few children who were cured by dietary interventions (i.e. they have been reclassified after an initial diagnosis of autism) were generally started prior to the age of three, and usually prior to the age of two. My son was five, and while he has made tremendous progress, he is still autistic. I have even heard from autistic adults, who started the diet in their twenties and beyond. They are still disabled, but it has alleviated symptoms that were always present and thus cleared up some of the “fog” these folks felt they were in.

This diet probably won't help everyone who tries it, but should you decide to try, you must be both serious and scrupulous about it the trial. You must also make sure that school staff, sitters and (especially) grandparents gets with the program. Many an effort has been scuttled by a grandma saying “but just a little couldn’t hurt, could it?” Or by a disbelieving outsider betting you won’t notice if they slip a cookie or a pretzel to your child. This cannot work unless everyone who comes near your child follows the strict guidelines that you provide.

But what else should you do? If we all now know (no thanks to the late Bruno Bettleheim) that this disorder is an organic rather than an emotional problem, why has so little been done to find a medical answer? Fortunately, things are changing somewhat. In January of 1995, Dr. Bernard Rimland convened the first Defeat Autism Now!  (DAN!) Conference in Dallas, Texas. It was a gathering of serious researchers who want to find a way to help children with autism, now, not in twenty years. A major undertaking of the group was the writing of a medical protocol to be used by physicians who treat autistic patients.
Called "Clinical Assessment Options for Children with Autism and Related Disorders: A Biomedical Approach," it represents a consensus report of the participants. The protocol was written by Drs. Sidney Baker and Jon Pangborn, and approved by all but one of the practitioner participants. It has recently been revised, and includes a list of doctors who are willing to use the protocol.

The Dan! Protocol, as it is called, is available from The Autism Research Institute.  It is a somewhat daunting document, but it is very well worth the effort to buy the protocol and go through it with your child's doctor.

No one expects that every parent will do every test in the protocol. But when you study it, you will find those tests that seem most likely to give you important information for your child. It is a valuable document, and I highly recommend that you contact the Autism Research Institute at 4182 Adams Avenue, San Diego, CA 92116. The Autism Research Institute’s phone number is 619 563-6840 and its website is In addition to the DAN! Protocol, The Autism Research Institute’s newsletter will keep you up to date on everything going on in the autism world.

I was fortunate to have been one of four parents invited to join in the first meeting of Dan! Since that time, Dr. Rimland has convened another conference aimed at training doctors and informing parents (and more are planned.) One thing that came out of these meetings is the shameful lack of research money dedicated to autism. Another parent who attended the DAN! meetings decided to do something about this situation and along with her husband, started the Cure Autism Now Foundation. CAN is dedicated to raising money for research. Please contact CAN at 1-213-549-0500 or electronically at for information on what you can do to help. If you have access to the Internet, visit the CAN web page at It is absolutely vital that we all join in this effort!

 Sam’s Story…Today

Many people write and call me to discuss the issues covered in my original paper on this topic. Without a doubt, the question I am asked most often is "How is Sam doing NOW?" There is no simple answer to this question. In general, we are very gratified at the progress Sam has made and continues to make. But there are always "glitches" along the way.

As of this writing, Sam is nine, and is attending third grade at a "regular" district school. He was transitioned into the district gradually when he was almost seven, with a great deal of support and instruction from the staff of his former school. He has a personal aide who was also trained by DDDC staff.   Sam does his academic work in a self-contained classroom, and most of it is one-on-one with his teacher or aide. He attends morning meeting, art, music, gym, lunch, recess and other "specials" with his third grade peers. 

Sam's attendance at this school requires a lot of patience on the part of school personnel, and a lot of adjustment on our parts. At times we miss the extra attention and support we got as a family from his former school. But Sam is so happy to have peers who respond to him! Though his behavior has been difficult at times (we still struggle with bouts of non-compliance and aggression) his attendance at this school is a great experience for everyone. He is well liked by staff and students. His peers try hard to help and understand him, and my husband worked with them at the start of the year so that they could understand Sam's struggles better. They are proud to help him when they can, and he greatly enjoys his social times. They have benefited from learning that not everyone faces the same challenges in life.

Sam continues to have difficulty with visual processing, and reading remains a real challenge. It is a challenge he is meeting however, and I am encouraged that Sam will read fluently in the not too distant future.

Sam’s language is excellent, and he shows that he is able to generalize ideas and skills quite well. He clearly has a "theory of mind" and is improving at abstract thinking. His school provides two sessions of speech therapy per week, and two of occupational therapy. Noting that Sam really needed and benefited from input to his vestibular system (the system controlling balance), his occupational therapist obtained a weighted vest for Sam. He wears it for much of each day, and teachers report it calms him down a great deal.

Now, about those “glitches”...Sam has shown, over the last four years, a troubling and completely predictable seasonal cycle. As late autumn approaches, life seems to become increasingly difficult. By mid-winter we are deep into a funk, which includes an increase in the number of tantrums and extremely difficult behavior. At this time of year, if he is not under medication, he will regress into terrible aggression and to a lesser degree, self-aggression. We have found that a small dose of Risperdal (Risperidone) is necessary to get through this portion of the year. As spring approaches, behavior improves as does Sam's attitude and mood. Last year, I removed the medication during his "up" period, but he really seemed to need it even then. As of now, he is taking this medication and doing very well on it. I doubt very much that we could get through a winter without this chemical help.

Sometimes being Sam’s advocate is very hard work, but we are extremely proud of him. Getting along is the world is more challenging for him than I can even imagine. Despite the rough spots, however, he is in general sweet and friendly. He can (and does) charm the socks off most adults who meet him and work with him. To sum up: life goes on, and we all work hard to ensure that he will be the very best that he can be.

A final note: if you have found this chapter interesting and would like to pursue a dietary intervention, please contact The Great Plains Laboratory, Inc. to obtain a copy of my book on this subject, Special Diets for Special Kids: Understanding and Implementing Special Diets to Aid in the Treatment of Autism and Related Developmental Disorders.  In addition to discussing the topics covered here, it contains many recipes and tricks to get you started.


  1. Baker, S.M. (1997) Detoxification & Healing: The Key to Optimal Health. Keats Publishing: New Canaan, CT.
  2. D'Eufemia, P. Celli, M., Finocchiaro, R., Pacifico, L., Viozzi, L., Zaccagnini, M., Cardi, E., Giardini, O. (1996) "Abnormal Intestinal Permeability in children with Autism". Acta Paediatrica 85: 1076-1079.
  3. Gillberg, C. (1988) "The role of endogenous opioids in autism and possible relationships to clinical features" in Wing, L. (ed.) Aspects of Autism: Biological Research. Gaskell:London, pp. 31-37.
  4. Knivsberg A-M et al. (1990) "Dietary intervention in autistic syndromes." Brain Dysfunction, 3:315-27. Brain Dysfunction, 4 (6) 308-319.
  5. Reichelt, K.L., Lind, G., Nodland, M. (1991) Probable Etiology and Possible Treatment of Childhood Autism; Brain Dysfunction, 4 (6) 308-319.
  6. O'Reilly, B. A. and R.H. Waring (1990) "Enzyme and Sulfur Oxidation Deficiencies in Autistic Children with Known Food and Chemical Intolerances. Xenobiotica, 20:117-122.
  7. Panksepp, J. (1979) "A neurochemical theory of autism." Trends in Neuroscience, 2: 174-177.
  8. Reichelt, K.L., et. al. (1981) "Biologically Active Peptide-Containing Fractions in Schizophrenia and Childhood Autism." Adv. Biochem. Psychopharmacol. 28:627-643.
  9. Shattock, P., Kennedy, A., Rowell, F., Berney, T.P. (1990) "Proteins, Peptides and Autism. Part 2: Implications for the Education and Care of People with Autism." Brain Dysfunction, 3 (5) 328-345.
  10. G. Lowdon (1991) "Proteins, Peptides and Autism; Part 2: Implications for the Education and Care of People With Autism." Brain Dysfunction, 4:323-334.
  11. D. Savery (1997). “Autism as a Metabolic Disorder,” Paper presented to the Annual Geneva Centre Conference on Autism. Toronto, Canada.