Current dietary approaches in autism spectrum disorder

Introduction

Autism spectrum disorder (ASD) is defined as a biologically based neurodevelopmental disorder characterized by persistent deficits in social communication and interaction, along with restricted, repetitive patterns of behavior, interests, or activities. This condition is typically diagnosable within the first three years of life (American Psychiatric Association 2013). The Centers for Disease Control and Prevention (CDC) estimate that one in every 36 children will have ASD, while the World Health Organization (WHO) estimates that in 2023, one in every 100 children will have ASD (CDC 2023; WHO 2024). There has been a significant increase in the number of diagnosed ASD cases in the last two years (CDC 2023). Although the etiology of ASD is not fully understood, genetic factors are thought to play a critical role. Besides genetic factors, environmental factors, including nutrition, are also believed to contribute to this condition (Alamri 2020; Yılmaz 2022).
Although there is no consensus on the biomarkers of ASD, disruptions in amino acid metabolism, elevated homocysteine levels, altered folate concentrations, and imbalances in serum B and D vitamin levels are used as biomarkers for early diagnosis (Salomone et al. 2015). Studies have indicated that while individuals with ASD generally have adequate macronutrient intake, they face issues with micronutrient intake (APA 2023; Arentz-Hansen et al. 2002; Bavykina et al. 2019). Foods produced from gluten- and casein-containing products metabolize into gluteomorphin and casomorphin in these individuals. These peptides bind to opiate receptors in the central nervous system, mimicking the effects of opioid drugs. Additionally, due to increased intestinal permeability, these peptides can enter the bloodstream and cross the blood-brain barrier, potentially affecting the brain (Piwowarczyk et al. 2018). The gastrointestinal symptoms commonly seen in ASD are associated with gluten sensitivity (Whiteley et al. 2013). The increase in anti-gliadin antibody levels and gastrointestinal symptoms supports immunological and intestinal permeability disorders (Piwowarczyk et al. 2018). This study aimed to investigate the potential effects of a gluten- and casein-free diet on ASD.

Method

Articles included in the study were thoroughly searched in electronic databases such as PubMed, Scopus, and Google Scholar. The search terms were defined as “gluten”, “autism”, “casein”, “casein free”, “gluten free”, and combinations of these terms. Filters were applied to include studies published from the inception of the databases to the present. Studies examining the effects of propionic acid on rats in clinical settings were deemed suitable for inclusion. Descriptive studies, letters to the editor, book chapters, conference proceedings, literature reviews, narrative reviews, systematic reviews with or without meta-analyses, and nutritional adjunctive intervention studies were excluded (Fig. 1).

Gluten structure

Gluten is a very intricate substance known for its extensive genetic variation that determines its unique proteins, glutenin and gliadin. Different varieties of wheat yield diverse quantities and types of these elements. The gluten levels in wheat may fluctuate due to farming practices and industrial procedures. For example, -5 gliadin levels rise with increased fertilization and higher temperatures during the ripening stage. Certain α-gliadins found in the subaleurone layer of wheat grains can be partially eliminated through roller milling (Kucek et al. 2015).

Toxic components

Gliadin contains peptide sequences that resist breakdown by stomach, pancreas, and intestinal enzymes due to its high content of proline and glutamine amino acids, which are not easily cleaved by many proteases. These proline-rich residues create dense, compact structures capable of triggering immune reactions in individuals with celiac disease (Arentz-Hansen et al. 2002).
Various sequences from α, γ, and ω gliadins, as well as glutenins, have been identified as triggers for celiac disease. It is estimated that hundreds of gluten peptides are immunogenic and initiate an immune response mediated by T-cells (Hausch et al. 2002). The predominant T-cell epitope originates from α-gliadin, although there is also T-cell reactivity to peptides derived from gluten, secalin, and hordein. Furthermore, there exists a specific hierarchy of immunostimulatory gluten peptides within each type of grain (Janatuinen 2002). Despite numerous peptides capable of stimulating T-cell responses in celiac disease, individual patients typically respond to only a subset (Biesiekierski 2017; Hausch et al. 2002).

Gluten-free diet

The definition of “gluten-free” lacks uniformity across different regulations. The Australia New Zealand Food Standards stipulate that gluten-free products must have undetectable gluten levels (less than 5 ppm) and must not include oats or malt (Biesiekierski 2017). In contrast, the United States Food and Drug Administration (FDA) and the international Codex Alimentarius Commission define gluten-free as containing less than 20 ppm of detectable gluten. Aligning with Codex standards, the Australian Food and Grocery Council specifies that gluten levels should not exceed 20 mg/kg in Australia and New Zealand (Biesiekierski 2017; TGKY 2012). Additionally, foods labeled as gluten-reduced must contain less than 100 mg/kg of gluten on a dry matter basis (TGKY 2012).
Previously, there has been debate surrounding the inclusion of oats in gluten-free diets. While some individuals with celiac disease report minimal T-cell responses specific to avenin and generally tolerate oats, others experience sensitivity to oats and may develop histological damage (Lundin 2003). Moreover, the safe threshold for gluten consumption varies significantly among individuals with celiac disease, ranging from 10 to 100 mg per day (Janatuinen 2002; Lundin 2003). Table 1 illustrates foods that contain gluten and those that do not.

Gluten-free diet mechanism

Microbiome-gut-brain axis

It has been hypothesized that the homeostasis of the gut-brain axis can be disrupted due to factors such as aging, obesity, diet, and drug use (Daulatzai 2015). Chronic intestinal inflammation, triggered in predisposed individuals (such as those with CD/NCGS) by consuming a gluten-containing diet, is closely associated with intestinal dysbiosis and a leaky gut (Chandra et al. 2015; Mohan et al. 2016). Both our research and studies conducted by others have shown that inflammation and dysbiosis induced by dietary gluten are associated with disruptions in genetic regulatory factors involved in neuroinflammation, cognition, and neurodegeneration (De Palma et al. 2009; Scheperjans et al. 2015; Sestak et al. 2011).
The enteric nervous system, a sophisticated network of neurons located within the gastrointestinal (GI) tract, is often called the “second brain”. This system functions independently of the central nervous system, hence its nickname “gut brain”, and has its own reflexes and sensory functions (Rao and Gershon 2016). Consisting of millions of neurons, the enteric nervous system communicates internally and with the central nervous system through the vagus nerve and other neural pathways. It primarily regulates digestive processes such as the movement of food, secretion of digestive enzymes, and nutrient absorption (Furness et al. 2014). The gut-brain axis refers to the complex bidirectional communication between the brain and the GI tract. This axis includes components such as the gut microbiota, the immune system, and the autonomic nervous system within the enteric nervous system (Al-Beltagi et al. 2023). It manages physiological functions such as digestion and metabolism and influences cognitive functions, behavior, and other brain activities. Communication between the gut and brain occurs via various pathways, including neurocrine and endocrine signaling (Al-Beltagi et al. 2023). At the molecular level, the expression of peroxisome proliferator-activated receptor γ (PPAR-γ), a key gene with anti-inflammatory effects (affecting peripheral, intestinal, and neuroinflammation) (Villapol 2018) and anti-dysbiotic properties (Byndloss et al. 2017), is significantly reduced in patients with ulcerative colitis and celiac disease (Kang et al. 2013; Soares et al. 2013; Sziksz et al. 2014). Research by Byndloss et al. (2017) demonstrated that downregulation of PPAR-γ is associated with a dysbiotic expansion of bacteria from the family Enterobacteriaceae (phylum Proteobacteria) and a decrease in the relative abundance of obligate anaerobic bacteria.

Gut-brain axis and neurodevelopment

The establishment of the microbiome-gut-brain axis begins early in life, with bacterial colonization occurring soon after birth. By the ages of one to three, a complex and stable microbiome is formed and remains fairly consistent (Koch and Demontis 2022; Vellingiri et al. 2022). There are parallels between the development of the microbiome and the central nervous system, implying that critical periods in neurodevelopment might align with essential periods of microbiome colonization (Maslen et al. 2022). This has led to the hypothesis that the microbiome-gut-brain axis plays a crucial role in neurodevelopmental disorders (Marler et al. 2017; Maslen et al. 2022). However, the exact elements of microbial composition that might disrupt neurodevelopmental processes are not yet fully understood.
Supporting this theory, studies have demonstrated that germ-free mice show an exaggerated HPA axis response to external stressors, a reaction that can be normalized only by introducing normal gut microbiota from specific pathogen-free mice during early development, but not later (Mazefsky et al. 2014). This highlights the importance of early exposure to endogenous microbiota for the HPA system to respond appropriately to regulatory neural inputs. More broadly, these findings suggest that the microbiome could impact critical windows in neurodevelopment, emphasizing its potential role in neurodevelopmental disorders such as ASD and attention deficit hyperactivity disorder (ADHD) (Gkougka et al. 2022; González-Domenech et al. 2020; Hausch et al. 2002).

Gut-brain axis and ASD

In individuals diagnosed with ASD, studies have consistently shown changes in the composition of gut microbiota (Berry et al. 2015; Bresciani et al. 2023; Cryan et al. 2019; Parmar et al. 2021; Zhao et al. 2020). For instance, research indicates that certain bacterial clusters such as Clostridium or Desulfovibrio are more prevalent in children with ASD compared to peers with gastrointestinal issues and typical neurobehavioral development (Saurman et al. 2020; Zahra et al. 2022; Zhao et al. 2020). While findings regarding specific microbial species in ASD vary among studies, alterations in Clostridium species have been frequently noted in research involving ASD individuals (Bresciani et al. 2023; Quigley 2017; Zhao et al. 2020). Some anecdotal evidence suggests that children with ASD may experience clinical improvements following episodes of fever, use of oral antibiotics, or intake of probiotics, which could potentially influence gut microbiome dynamics (Al-Beltagi and Saeed 2022; Gkougka et al. 2022; Pan et al. 2023). In animal models simulating ASD symptoms, administering the gut commensal Bacteroides fragilis has demonstrated benefits such as alleviating gastrointestinal symptoms, enhancing gut barrier function, and mitigating behavioral abnormalities associated with ASD (Al-Beltagi and Saeed 2022).
One hypothesis posits that autism may result from a metabolic dysfunction where opioid peptides, produced during the digestion of gluten and casein, traverse an unusually permeable gut membrane (“opioid excess theory” or “leaky gut theory”) (Darch and McCafferty 2022; Rubenstein et al. 2018). These neuroactive peptides could then enter the bloodstream and attach to opioid receptors, affecting neurotransmission. It has been observed that children with ASD often exhibit higher gut permeability compared to healthy controls (Mukhtar et al. 2019). However, research has not consistently found elevated levels of opioid peptides in children with ASD (Kerr et al. 2009). Another theory focuses on oxidative stress and deficiencies in sulfur metabolism. Here, sulfur metabolic deficiencies can modify bacterial composition and their products, resulting in increased oxidative stress in individuals. The combined impact of bacterial byproducts, oxidative stress, and dietary allergens could enhance gut permeability (Jiang et al. 2022). This increased gastrointestinal permeability might lead to neuroinflammation, affecting the protective function of the blood-brain barrier and potentially contributing to ASD. As a result, dietary treatments aimed at restoring gut permeability could be promising therapeutic approaches for children with ASD (Jiang et al. 2022; Mallory and Keehn 2021; Nikolov et al. 2009).

Gut-brain axis and ADHD

Currently, there is limited direct evidence regarding the impact of altered interactions between the microbiome, gut, and brain on ADHD. However, accumulating evidence suggests that similar alterations could potentially contribute to ADHD symptoms (Neuhaus et al. 2018; Santhanam 2023). Recent unpublished data from our research group indicate that the gut microbiome shows differences in individuals with ADHD. Specifically, there is an increase in the prevalence of the genus Bifidobacterium in ADHD, which correlates with heightened metabolic activity of cyclohexadienyl dehydratase – an enzyme involved in synthesizing essential amino acids and the dopamine precursor phenylalanine. Furthermore, this study highlights that these metabolic shifts are linked to reduced signaling in the ventral striatum during anticipation of rewards, which serves as a neural marker for ADHD (Santhanam 2023). These findings mark an important initial stride toward comprehending the connection between microbiome alterations and ADHD symptoms.

Family relationships and structure

The implementation of an elimination diet necessitates stringent parental oversight, meaning that the effects of such diets might also be shaped by concurrent changes in parental caregiving strategies and involvement. It is well documented that parents of children with ADHD and ASD adopt distinct parenting styles and maintain different relationships with their children (Cermak et al. 2010; Ferguson et al. 2019; Mitrea et al. 2022). Consistent parenting practices and positive parent-child interactions are linked to improvements in the child’s behavior (Bowman et al. 2021; Mitrea et al. 2022). However, it is important to note that behavioral interventions for ADHD typically lead to positive outcomes related to the disorder’s associated symptoms rather than its core symptoms (Carlier et al. 2023).
Therefore, it is plausible that the behavioral improvements seen with an elimination diet are due to changes in parental caregiving practices necessary for adhering to the diet, rather than solely from the dietary changes themselves. However, this hypothesis lacks empirical support to date. A previous study indicated that families willing to adopt an elimination diet for their children with ADHD had a favorable family environment compared to families with healthy controls, suggesting that the elimination diet did not negatively affect family dynamics and structure (Dineen-Griffin et al. 2019).
The exact mechanism behind the effectiveness of elimination diets remains unclear. The microbiome-gut-brain axis is a leading candidate, likely involving a complex interplay of allergic reactions, gut permeability, oxidative stress, and changes in microbiome composition and function. Understanding the direct and indirect pathways of this connection is crucial, as it could provide clinicians with markers and targets for preventive and therapeutic treatments. Additionally, this knowledge is a vital step toward developing personalized nutritional and therapeutic interventions for individual patients. More research is needed to clarify how elimination diets impact ADHD and ASD (Dineen-Griffin et al. 2019).

Effects of diets

Table 2 shows the potential effects of gluten-free and/or casein-free diets on ASD. Additionally, Table 2 also highlights the potential side effects that may result from following these dietary practices.

Evidence in specific syndromes

Elder et al. conducted a study on 15 participants aged 2-16 years to examine the effects of a gluten- and casein-restricted diet. They reported no significant differences in the Childhood Autism Rating Scale (CARS) and Ecological Communication Orientation Scale (ECOS), and behavioral frequencies. There was no significant difference in urinary peptide levels for gluten or casein in grouped data. Additionally, there was no statistically significant difference in observed parental behaviors (Elder et al. 2006). Another study reported improvements in Autism Diagnostic Observation Schedule (ADOS), Gilliam Autism Rating Scale (GARS), and ADHD-IV scores, as well as improved social interaction starting from the 12th month of diet implementation (Whiteley et al. 2010). Knivsberg et al. reported that a casein-free diet over one year resulted in reductions in resistance to learning, ability to close eyes, and the habit of covering the ears. They also noted significant improvements in distancing from teachers or showing direct indifference, peer relationships, anxiety, empathy, physical contact, nonverbal communication, eye contact, responsiveness when spoken to, and language features (Knivsberg et al. 2002). Johnson et al. conducted a study on 22 children aged 3-5 years, implementing a gluten- or casein-restricted diet with omega-3 supplementation for three months. The study found improvements in behavior, increased language skills, decreased aggression scale scores, and statistically significant improvement in the visual reception subscale. They also noted no significant difference between the diet groups (Johnson et al. 2011). A study conducted on 80 individuals over six weeks reported a significant reduction in the prevalence of gastrointestinal symptoms and behavioral disorders due to a gluten-free diet (Ghalichi et al. 2016). A similar study by González-Domenech et al. found a reduction in ATEC scale scores, ERC-III, and ABC scale scores after 12 months on a gluten-free diet, though no changes in behavior or urinary beta-casomorphin concentrations were observed (González-Domenech et al. 2020). Another study reported reductions in hyperactivity, irritability, and inattentiveness scores following a gluten-free and dairy-free diet. However, constipation complaints were observed in the gluten-free diet group by the second week (Navarro et al. 2015). A different study involving various diet groups found significant improvement in non-verbal intelligence and autism symptoms in autistic individuals. The study reported significant increases in serum EPA, DHA, carnitine, and vitamins A, B2, B5, B6, B12, folic acid, and coenzyme Q10. Additionally, vitamin/mineral supplements, essential fatty acids, and dietary therapy were identified as the most effective methods (Adams et al. 2018). Hyman et al. reported reductions in attention deficit, hyperactivity, and behavioral disorder rates after 30 weeks on a casein- and gluten-free diet. However, side effects such as abdominal pain and diarrhea were also reported (Hyman et al. 2016). Nazni et al. conducted a study on 50 children aged 3 to 11 years and reported improvements in attention scores, hyperactivity scores, and anxiety/compulsion scores following a gluten- and/or casein-restricted diet (Nazni et al. 2008). Another study involving 45 individuals over six months examined the effectiveness of a modified Atkins diet and a gluten-free diet. The results showed significant improvements in CARS and ATEC scales, with the modified Atkins diet group achieving better results in cognition and sociability compared to the gluten-restricted diet group (El-Rashidy et al. 2017). Piwowarczyk et al. reported in a separate study that a gluten-free diet did not show differences in autism symptoms, maladaptive behaviors, or cognitive abilities between the control group and the diet group (Piwowarczyk et al. 2020). In various studies excluding gluten and casein, it was noted that there were high levels of specific and anti-gliadin IgG antibodies, along with an increase in composite scores for average approach withdrawal problems. Additionally, significant differences were not observed in gastrointestinal symptoms or urinary intestinal fatty acid-binding protein (IFABP) excretion between the diet intervention groups and control groups (Bavykina et al. 2019; Pusponegoro et al. 2015).

Side effects of diet implementation

In studies where diet interventions were applied, various side effects have been reported. These include minor behavioral issues, irritable bowel syndrome, nausea, vomiting, night awakenings, appetite loss, gastrointestinal abnormalities, increased behavioral disorders, soft stools, constipation, abdominal pain, diarrhea, and cerebral ataxia (Adams et al. 2018; Ghalichi et al. 2016; Hyman et al. 2016; Johnson et al. 2011; Navarro et al. 2015; Piwowarczyk et al. 2020; Roberts et al. 2020; Whiteley et al. 2010).

Conclusions

This review examined various medical nutrition approaches and discussed relevant research findings. The study included 14 different RCT studies with more than 561 subjects in the meta-analysis. According to the results of the meta-analysis, gluten- and casein-restricted dietary intervention was found to have benefits in terms of behavior, cognition, gastrointestinal and biochemical values in more than 461 subjects. However, no significant change was observed in 102 subjects. Gluten-free, casein-free and ketogenic diets, camel milk, turmeric, probiotics and fermented foods are suggested to alleviate symptoms of autism spectrum disorder, while sugars, additives, pesticides, genetically modified organisms, processed inorganic foods and hard-to-digest starches are reported to worsen symptoms. Large-scale prospective controlled studies are needed before recommending an optimal diet for ASD. In the field of medical nutrition therapy, it would be advantageous to investigate the pathophysiology, nutritional intake, food allergies/intolerances, and dietary patterns of individuals with ASD, address energy and nutrient deficiencies through dietary modifications, provide adequate and balanced nutrition, and use therapeutic dietary approaches to alleviate symptoms. This review highlights the importance of identifying specific dietary strategies for individuals with ASD and evaluating their effects on symptoms. It also aims to instill optimism by identifying new therapeutic targets and interventions for individuals affected by this common developmental disorder.

Disclosures

This research received no external funding.
Institutional review board statement: Not applicable.
The authors declare no conflict of interest.

References