Complementary, Holistic, and Integrative Medicine: Autism Spectrum Disorder and Gluten- and Casein-Free Diet

After completing this article, readers should be able to:Many parents of children with autism want to try a gluten-free and casein-free diet (gfcf-d) for their child to see if it will help. Some studies have found positive results, but questions remain unanswered. Parents should consider potential benefits and potential harms of trying the diet. Potential benefits to the affected child include improved communication, social interaction, and behavioral flexibility and decreased inattention and hyperactivity. Potential harms of the diet include nutritional deficiencies and the effort and costs associated with maintaining it.Caution should be used when trying to decide whether to try a gfcf-d for a child who already has nutritional deficiency, growth problems, or restricted diet due to difficulty accepting new foods. Before starting a gluten-free diet, the child should be screened for celiac disease, especially if the child has had gastrointestinal symptoms, such as diarrhea, constipation, anorexia, vomiting, abdominal pain, or weight loss. If parents want to try a gfcf-d, they should consider seeking guidance from a therapist to help introduce the diet in a gradual way and a dietitian to monitor their child’s nutritional intake while on the diet and to recommend supplementation if necessary. Weight also needs to be monitored. Parents and health care professionals or teachers should document behavioral improvement, ideally using validated measures.Autism spectrum disorder (ASD) (prevalence of approximately 1 in 100) is a neurologically based, lifelong disability characterized by impairments in social interaction, communication, and behavioral flexibility, (1)(2) with its basis in genetic and nongenetic factors. (1) Current treatment focuses on having the child acquire skills and decrease comorbidities through educational services, provided through family and health or school systems, with consultation when necessary from speech-language, occupational, physical, and behavioral therapists, psychologists, and physicians. (1)(3) Studies suggest promising outcomes for some children from early intensive behavioral and developmental interventions, although more large-scale randomized controlled trials (RCTs) and data on potential harms are needed. (3)(4)(5)In absence of cure, many parents search for complementary and alternative therapies, (6) especially when there are comorbid gastrointestinal and behavioral issues. (7) One popular approach is the gluten-free and casein-free diet (gfcf-d), used by 15% to 38% of the ASD population (8)(9)(10)(11) and endorsed as effective by half of responding parents (n=479). (9) Gluten is a protein in wheat, rye, triticale, and barley. Casein is a protein in mammalian milk products (eg, milk, cheese, yogurt, butter, processed foods).Investigators have reported conflicting evidence regarding a leaky gut hypothesis (ie, increased gut permeability and uptake of inadequately digested gluten and casein peptides due to mucosal inflammation). (12)(13)(14)(15)(16)(17)(18)(19) In the 1980s and 1990s, Reichelt et al reported increased urine casein peptide levels in children with ASD, (20)(21) elevated serum IgA antibody levels to gluten and casein, and social and behavioral benefits from a total or partial gfcf-d. (12)(21)(22)(23)(24) This review explores current evidence for the potential benefits and risks of using gfcf-d for children with ASD.Two recent systematic reviews by Mulloy et al (25) and Rossignol et al (26) included case reports, observational studies, and controlled clinical trials, including 2 RCTs. (27)(28)Mulloy et al (25) identified 14 studies, with a total of 188 participants, between the ages of 2 and 17 years. Seven studies reported positive results, 4 negative results, 2 mixed results, and 1 inconclusive. Methodologic concerns were identified in all the studies. An important finding was that studies reporting negative results used the intervention for a much shorter duration than those with positive results, for a mean of 5 vs 18 months.Rossignol et al (26) identified 14 studies that included 930 participants (including one survey of 479 parents). (9) The children ranged in age from infants to 17 years. One study was only published as an abstract. (29) Of the remaining 13 studies, 8 reported positive results, 3 reported negative results, and 2 found no significant difference.Although both Mulloy et al and Rossignol et al included 2 RCTs, most of their included studies were of less rigorous designs, such as surveys, case reports, and observational studies. Limitations included lack of control groups and heterogeneous study conditions and interventions.A recent Cochrane systematic review evaluated the 2 RCTs, which report conflicting results. (27)(28) The trial by Knivsberg et al (27) was a single-phase single-blind trial comparing gfcf-d to standard diet for 12 months (n=10 on each diet). It found improvements in communication, social interaction, and behavioral flexibility (DIPAB or Diagnose of Psykotisk Adfærd hosBørn [Diagnosis of Psychotic Behavior in Children]) in the intervention group compared with the control group; all were statistically significant (p<.05). Differences in nonverbal cognitive (Leiter International Performance Scale) and gross and fine motor skills (Movement Assessment Battery for Children) were nonsignificant.The trial by Elder et al (28) was a double-blind crossover trial comparing gfcf-d to standard diet for 6 weeks (n=15 on each diet). Although the trial reported no significant differences between the 2 treatment groups for communication, social relationships, behavioral flexibility, activity level, or intellectual ability, parents of 7 children reported improved language, hyperactivity, and tantrums, possibly reflecting subtle changes not detected by the outcome measure (Childhood Autism Rating Scale). Some parents decided to continue gfcf-d despite lack of evidence of effect.In a more recent single-blind RCT by Whiteley et al, (30) children were randomized to 12 months of gfcf-d or standard diet; both groups had essential fatty acid supplementation. The authors report benefits for the gfcf-d group in communication (Autism Diagnostic Observation Schedule), social interaction (Gilliam Autism Rating Scale), inattention, and hyperactivity (attention deficit hyperactivity disorder-IV rating scale). In the gfcf-d group, 11 of 37 participants withdrew from the study despite extensive nutritional support because of difficulty in accepting the diet, and lack of time, resources, and perceived beneficial effect. Gastrointestinal disease or concomitant treatments were not reported.Another single-blinded RCT (31) compared gfcf-d (n=8) to a low-sugar diet (n=14) for 3 months in patients with ASD. Although there was no significant difference on the receptive language domain (Mullen Scales of Early Learning AGS Edition) in the gfcf-d group in comparison to the control group (p=.06), the control group did significantly better on the Mullen visual reception domain (p=.04). The gfcf-d group scored significantly better on aggression and attention deficit hyperactivity disorder (Child Behavior Checklist), whereas the control group scored as significantly less withdrawn. The authors suggest that more time may be needed before effects can be seen.Although not yet completed, it is worth noting that an ongoing double-blind RCT sponsored by the National Institutes of Mental Health is examining 6 weeks of gfcf-d in preschoolers followed by placebo for 12 weeks. This study is currently being prepared for publication; therefore, the results are not yet available (Susan Hyman, personal communication).Adequate short- and long-term safety data from gfcf-d are not yet available. (30) The RCTs reviewed did not state whether any adverse effects were seen. (27)(28) Current evidence comparing macronutrient and micronutrient deficiencies with and without a gfcf-d in children with ASD is limited and conflicting. In one report, minimal effect on energy, protein, and micronutrient intake is attributed to gfcf-d (n=4) compared with those not on a gfcf-d (n=12). (8) However, when comparing the effects of a gfcf-d on children with autism to healthy controls, some reports associate a gfcf-d with exacerbation of being underweight in children with autism (n=252) compared with healthy controls, (32) less calcium intake (n=14) compared with healthy controls without gfcf-d (n=31), (10) decreased bone development (n=9), (33) and plasma essential amino acid deficiency in 6 of 10 children. (34) A study involving a short audit of dietetic records (11 of 26 on or planning to begin a gfcf-d) suggested the complexity of managing diet within the context of parent-child relationships/family dynamics, dual diagnosis, and the behavioral basis of both the disorder and normal feeding. (35) A gfcf-d can be expensive and might restrict lifestyle. (36)Adequate evidence for nutritional adequacy in children with ASD compared with healthy children is also lacking and conflicting, (37)(38) although findings indicate a potential increased risk for nutritional deficiencies. (37)(39)(40)(41) Small studies confirm that many children with ASD often have food preferences restricted (8)(42) to certain textures, colors, or packaging due to sensory hypersensitivity and insistence on sameness; (8)(10) as food variety decreases, the number of nutrient intakes falling below reference nutrient intakes increases (n=17). (41) There is a reported lower intake of energy, vitamin A, vitamin C, zinc, and phosphorus in children with ASD (n=252) compared with those without ASD. (32) Small studies in children with ASD found they (n=32) consume fewer servings of dairy compared with healthy controls (n=23) (10) and have below recommended intake of vitamins A, D, and K, pantothenic acid, biotin, and choline (n=24). (43) Some children with ASD (10- to 18-year-olds) have low bone mineral density (BMD) (4 of 26) correlating with insufficient calcium and calorie intake; (44) although asymptomatic in these children, low BMD increases risk of fractures and future osteoporosis. Limited studies suggest that blood nutrients frequently deficient in children with ASD include essential amino acids (58% of n=26); (34) vitamin D (61% of a cohort of 89 children, which included controls and ASD patients with and without casein-free diet, with no group effect); (45) and ferritin (16% of n=96). (46)(47) Thus, further restrictions by gfcf-d could potentially exacerbate risk of nutritional deficiencies, (48) especially when introducing new foods may be challenging. Supplementation with multivitamins can significantly decrease these risks except in the rare child who refuses any multivitamin supplementation.Another potential harm of adopting gfcf-d is overlooking possible underlying food allergy, celiac disease, or lactose intolerance. (13) Food allergy can be IgE- or non–IgE-mediated wheat allergy; however, both are rare. (49) Celiac disease is the most common autoimmune gastrointestinal disorder (approximately 1% of North Americans). (50) The only treatment for it is a life-long gluten-free diet. (51)(52) Celiac disease may present with typical gastrointestinal symptoms of diarrhea, constipation, anorexia, vomiting, abdominal pain, or failure to thrive; it may also present with atypical symptoms, (53) such as headache, learning difficulties, and peripheral neuropathy. (54)(55) Therefore, celiac disease would be difficult to exclude in this population, without a screening test before commencing a gfcf-d, especially when behavioral changes (irritability, aggression, repetitive movements) are the only manifestation of gastrointestinal conditions. (13)(56) Current evidence about celiac disease risk in autism is conflicting, (57)(58) but the availability of IgA-antitissue-transglutaminase screening test for celiac disease has improved detection rates. No current consensus exists for recommending screening for celiac disease in individuals considering a gfcf-d. (13) However, these diagnoses can (at minimum) coexist at least at population rates. Lactose intolerance, although uncommon in young children, presents usually with flatulence and diarrhea. It is possible that for children with primary or secondary lactose intolerance a casein-free diet would reduce irritability from gaseous distention of the intestine.Evidence to date on the effectiveness of gfcf-d for children with ASD has been inconclusive due to methodologic limitations. Preliminary data suggest there may be a subgroup of children with ASD who respond to a gfcf-d. However, further research is necessary before health care professionals can recommend gfcf-d dietary modifications for ASD symptoms. (13) If parents wish to try a gfcf-d, in our opinion it is appropriate to educate them on a safe approach (Table).Well-conducted and adequately powered double-blind RCTs of sufficient duration are needed (36) to determine the outcome of a gfcf-d for children with ASD and to evaluate for possible markers for those children likely to benefit, (13) such as underlying gastrointestinal disease. Such studies should include assessment of long-term nutritional adequacy.The authors gratefully acknowledge Laura Rogers, Occupational Therapist, Faculty of Nursing, University of Alberta, Susan Jardine, Registered Dietitian, Autism Clinic, Glenrose Rehabilitation Hospital, Edmonton, Alberta, and Dr Lonnie Zwaigenbaum, Department of Pediatrics, University of Alberta, for their comments during the manuscript preparation, Soleil Surette for help with the search strategy, and Amy Moen for coordinating the Pediatrics in Review series for the American Academy of Pediatrics Section on Integrative Medicine. Dr Sunita Vohra receives salary support from Alberta Innovates Health Solutions as a Health Scholar.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.