4.1. GI Symptoms
In accordance with other studies, we showed a high prevalence of GI symptoms in children with ASD that persisted across all age groups. Individuals with ASD were more severely affected by GI problems than the controls: almost half of the children and adolescents with ASD suffered from GI symptoms several times per week or daily, in addition, subjects with ASD tended to present more GI problems than the controls. The main complaint was constipation and/or hard stools, which is in agreement with findings of other studies [
14,
15,
33,
34].
We demonstrated that higher frequency of GI symptoms was associated with higher scores in all ADI-R domains, i.e., with more severe presentation of ASD symptomatology: social interaction, social communication, and repetitive and restricted behavior and interests. These results are similar to other studies that have shown more severe ASD core symptoms in children with GI problems than in children without GI problems [
11,
35]. Alterations of behavior, such as aggression, self-injurious activity, or sleep problems are often misleadingly attributed to be just “symptom of the autism” [
14], but it is possible that behavioral impairments in ASD are exacerbated or even partially due to the underlying gastrointestinal problems [
11]. Additionally, the evidence linking GI dysfunction with behavioral presentations of ASD may indicate that they share some common pathomechanisms [
36].
The existence of a GI pathology specific to ASD has not been established, and mechanisms of GI disorders in ASD are not fully explained. GI symptoms seem to have multifactorial basis [
37]. Recently, much attention is focused on the microbiota residing in the GI system [
38]. Evidence shows that gut microbiota and its metabolites influence the brain via the gut–brain axis, i.e., a bidirectional communication between the gut and the brain through neural, endocrine, and immune mechanisms. In ASD, differences in abundance and composition of gut microbiota have been observed [
39], and the altered signaling from gut to the brain has been suggested as a potential contributor to the development of ASD and its behavioral presentations. At the same time, altered gut microbiota may contribute to the GI problems associated with ASD [
36].
There is growing evidence of gender differences in ASD symptoms [
40]. In our study, girls with ASD exhibited GI problems more frequently than boys; 70.6% of girls but only 44.5% of boys had problems several times per week or daily. As found on bivariate analysis stratified by gender, among boys, those who scored by one unit more on the Likert scale of GI symptoms were, on average, at 1.39 times higher risk of being classified with ASD, compared to those in the reference GI category (free of GI problems). In girls, an increase by one unit on the Likert scale of GI symptoms was associated with a 2.55 times higher risk of having an ASD, compared to those in the reference GI category. Modifying effect of gender on this relationship was also confirmed in the multivariable analysis. This finding suggests that presence and severity of GI symptoms may be a supplementary biomarker in diagnosing ASD in girls. More investigation is needed for explanation of this observation. Kushak and Winter [
41] suggested that intestinal microbiota that is believed to interact with sex hormones is a factor associated with higher ASD prevalence in boys. Thus, we may hypothesize that gut microbiota might contribute also to gender differences in GI symptoms. Still, due to limited number of girls in our sample, this finding needs to be interpreted with caution.
4.2. Food Selectivity and Mealtime Problems
Food selectivity resulting in imbalances in the diet composition may be a factor contributing to GI problems, at least in a proportion of individuals with ASD [
12]. In our study, significant correlation between severity of GI symptoms with food selectivity and mealtime problems was found. Similarly, in our earlier survey, children with ASD who were food selective, had more GI symptoms compared to children free of food selectivity [
42].
About 70% of children with ASD are reported to have some feeding problems [
17]. Besides the “picky eating”, other undesirable mealtime behaviors observed in ASD, such as adherence to routines, resistance to new or non-preferred foods, or tantrums have been shown to adversely influence the diet composition of many children with ASD [
17]. In our sample with ASD, high rates of problem behavior related to food intake were demonstrated in boys and girls of all ages that correlated with severity of ASD symptoms in ADI-R domains. There is evidence that feeding problems in ASD are of multifactorial origin, and they appear to be associated with the core behavioral characteristics of ASD. They reflect repetitiveness and preference for sameness, ritualism, unusual interest in sensory properties of food, but also diminished responsiveness to social reward, and increased reactivity in response to frustration [
17,
43].
To some extent, food selectivity and feeding disorders are present in neurotypical children, but they are more severe in ASD [
44]. A study revealed that in ASD, the “picky eating” appears at early age and escalates more quickly than in typically developing children [
43]. The food aversion does not resolve over time as the child develops, therefore it is not recommended to wait for the difficulties to disappear or that the child will “grow out of the problem” [
45,
46].
The most frequently omitted food group is vegetables, followed by fruits [
16,
42]. Selective children with ASD often prefer foods with low nutritional value and high in fat, salt, and sugar [
19]. Children with ASD were shown to have a lower protein intake, and their diets may be low in micronutrients [
47,
48]. Food variety has been shown to be a predictor of nutritional status of children with ASD [
49]. Mealtime problems and unusual dietary patterns are one of the factors contributing not only to undernutrition, but also to higher risk of obesity in children with ASD [
50].
Feeding problems in ASD may remain unattended by the healthcare providers. This is because selective eating patterns are not necessarily associated with higher risk for growth retardation that is a marker of nutritional deficits and triggers clinical attention of pediatricians [
18].
4.3. Dietary Interventions
There is evidence that individuals with ASD are more frequently affected by food allergies and intolerances, requiring diet as a treatment method of their medical condition [
22]. In our study, more than 20% of children and adolescents were on a diet, of that about one third reported that they followed a diet on recommendation of their pediatrician. However, more than 50% of children and adolescents were put on a diet, although they did not suffer from a medical condition that is typically treated with a diet. We have not found a correlation between GI problems and diets, thus it does not seem that GI problems were a significant cause for putting the child on a diet. Evidence shows that in a sizeable proportion of individuals with ASD, the dietary intervention is a form of complementary and alternative medicine, i.e., practices and products that are not part of conventional Western medical care [
51]. Implementation of a GFCF (gluten-free/casein-free) diet is the most used dietary intervention in ASD [
23,
52]. It is based on the “opioid excess hypothesis”. It postulates that ASD results from a metabolic disorder in which opioid peptides produced by metabolism of gluten and casein pass through an abnormally permeable intestinal wall into blood, and exert harmful effects on processes in the brain [
52]. Although scientific evidence shows some improvements in behavior on GFCF diet, there is insufficient data to support its broader use [
23,
53]. Several caregivers of our ASD subjects reported that they rather reduced than fully eliminated either gluten or casein, which raises questions about efficacy of such “partial” dietary treatment.
GAPS (Gut and Psychology Syndrome) diet is a strict elimination diet that recommends to cut out foods such as grains, pasteurized dairy, starchy vegetables, or refined carbohydrates. Although it is broadly suggested in non-scientific resources, it was very scarcely studied, thus there is lack of evidence about efficacy of this diet and doubts about its safety. Better studied is the ketogenic diet (KD) that is also based on major restriction of carbohydrates. Although the evidence is promising, data are still limited, and insufficient to encourage KD as a treatment for ASD [
54].
A review of existing data shows that gluten-free diet is not an appropriate choice in individuals without a medical condition due to their lower content of energy, several vitamins and minerals and the dietary fiber, and potential heavy metal exposure [
55]. Gluten free diets reduce consumption of starchy foods that are often preferred by children with ASD [
52]. Children with reduced milk intake or milk-free diet face an increased risk of growth deceleration and suboptimal intake of several micronutrients, mainly calcium [
52,
56].
We demonstrated in our study that elimination diets were introduced also to individuals who exhibited food selectivity. There is evidence that children with ASD who present mealtime problems are at higher risk of nutrient deficiencies than their nonselective peers [
16,
17]. The combination of food selectivity and nutritional limitations raises important questions regarding the use and elevated risk of adverse effects of dietary interventions. As the elimination diets are associated with higher risk of nutrient deficiencies, the caregivers should employ utmost caution when deciding to pursue this form of treatment.
A meta-analysis of dietary approaches to ASD does not support non-targeted dietary interventions as treatment of ASD. Due to heterogeneity in ASD presentation, it appears that there is a potential role for some specific dietary interventions in the management of some symptoms in some individuals. There is a need for better-designed clinical trials assessing efficacy and safety of dietary interventions in ASD [
21].
4.4. Food Supplements
High rates of food supplement use were found in our study in the ASD group, but also in the control sample. Generally, an increasing trend in food supplement use among children and adolescents has been observed [
57], that are given mainly to prevent or treat illnesses [
58]. A study that involved children with ASD reported that the top reasons of parents for giving supplements were also to enhance the child’s diet, and to promote immune system function [
51]. Evidence shows that the most frequently used forms of supplements are vitamins and minerals in a variety of combinations; also in the ASD population [
47,
51,
57]. Interestingly, there are reports that supplements were given to prevent or treat conditions for which there is only limited evidence of their efficacy [
58].
Our results show that supplement use correlated with “being on a diet”. One explanation may be that the parents attempted to substitute potential nutritional deficits resulting from lower variety of diet. On the other hand, despite high rates of food selectivity and mealtime problems found in our study, we did not observe their correlation with supplement use. The majority of our sample, both with and without ASD, were in preschool age, which is associated with increased occurrence of common childhood illnesses, thus, we hypothesize that part of the individuals were receiving supplements to boost immunity.
Although vitamins and minerals were frequently used both in ASD and control sample, concerning other types of supplements, we found significant differences between the samples. This may result from the fact that supplements are in ASD commonly given as a form of complementary and alternative therapy with the aim to reduce the behavioral symptoms, such as repetitive behavior, to increase social skills, or to improve sleep [
51]. In our study, we observed weak negative correlations of supplement use with age and verbal skills indicating that younger and non-verbal children were more likely to use food supplements. In fact, some parents reported that they give the supplements in order to enhance verbal skills in their children presenting delay in language development.
Recently, a meta-analysis revealed that supplementation with vitamins and omega-3 PUFAs had positive effects on some behavioral presentations of subjects with ASD with small effects size [
21]. As shown by our results and supported by a study of Trudeau at al. [
51], besides vitamins and minerals, the most commonly used supplements included probiotics and omega-3 PUFAs. The administration of probiotics, especially in the context of the “gut–brain-axis”, is a promising intervention for behavioral symptoms, at the same time for GI dysfunction, but clinical studies are still limited and inconclusive [
38]. There is evidence that omega-3 PUFAs play role in neurodevelopment, and supplementation with omega-3 PUFAs exhibited positive effects on some core symptoms of ASD. Two meta-analyses gained discordant results, thus more studies are needed to elucidate their efficacy [
59]. In our sample with ASD, we also found use of carnosine. Neuroprotective action of carnosine has been reported, and beneficial effects on behavior of subjects with ASD have been observed in some studies, however, they were not confirmed by other studies. At present, there is insufficient support for efficacy of carnosine in ASD [
60]. Frequent vitamin D was reported in individuals with ASD [
51]. In our ASD sample, 15.0% of supplement users took vitamin D, but we did not include this vitamin into analysis of food supplement use because it is broadly recommended by pediatricians to due to high rates of low plasma vitamin D levels in the population [
32].
The effect of food supplements in individuals with ASD appears to depend on the nutritional status, showing that subjects with lower intakes benefit more from the intervention [
61]. Still, food supplements are commonly given without previous assessment of nutritional status. The use of food supplements by children and adolescents can improve micronutrient status, but it can also increase the risk of excessive intake of certain vitamins and minerals [
57]. According to a study of Stewart et al. [
47], despite supplementation, the most common micronutrient deficits in ASD (vitamin D, calcium, potassium, panthothenic acid) were found to have not been corrected. Additionally, the usage of supplement lead to excess intake of vitamin A, folate, and zinc.
Due to a variety of problems associated with ASD, families and caregivers of individuals with ASD face many challenges. GI and nutrition-related problems are often getting less attention in comparison to behavioral or other clinical concerns [
3]. There is evidence that GI dysfunction, feeding problems, or imbalanced diet may result in inadequate nutrient intakes in children with ASD, and subsequent nutrition related health consequences [
15,
44]. At the same time, symptoms of ASD may be exacerbated due to the underlying GI problems [
11]. This indicates that multi-professional approach to individuals with ASD should include early identification and management of GI symptoms, food selectivity, and mealtime problems. Currently, no special treatment methods to GI problems specific for subjects with ASD exist and clinical practice guidelines available for the general population should be used [
3,
62,
63]. Concerning food selectivity and mealtime problems, the only empirically supported treatment for children with ASD is behavioral intervention based on applied behavioral analysis (ABA) [
45,
64,
65].
At present time, no causal therapy for ASD core symptoms is known, however, there is scientific evidence about efficiency of behavioral interventions, such as ABA [
3]. Despite limited evidence and poor understanding of the potential mechanisms of dietary interventions that might affect behavior, special or elimination diets and nutritional supplementation are frequently used treatments in children and adolescents with ASD [
24]. Parents who are considering dietary intervention should contact a health professional (e.g., a pediatrician, or a dietitian), who can provide guidance in planning of the diet to ensure the nutritional needs and monitor the child’s nutritional status.
This is the first study that evaluated GI symptoms, feeding problems in Slovak children and adolescents. The strength of the study is the well-defined sample of children carefully diagnosed for the presence of ASD with golden standard methods, the number of participants, and inclusion of multiple markers into the analysis. The study has its limitations. The sample was not randomly selected, and it included all volunteers who met the criteria, thus, the results may be biased by the sample selection. Another limitation is the fact that data on GI problems were collected by a questionnaire filled in by parents without verification of the diagnosis by a medical doctor. However, according to a study [
12], parental report of a GI disorder displayed a high concordance with the clinical diagnosis by a medical professional. Additionally, the subjects in the control sample were not screened for symptoms of ASD, and absence of the disorder was based only on a parental report.