Abstract
According to the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association [APA], 2013), ADHD is defined as a persistent pattern of inattention and/hyperactivity–impulsivity that interferes with an individual’s development and everyday functioning. The symptoms and signs must present before the age of 12 years, be present in at least two situations (at school, home, work, etc.), and negatively influence social, academic, and occupational functioning (APA, 2013). The 10th revision of the International Classification of Diseases and Related Health Problems for Statistical Purposes (ICD-10), however, uses the term hyperkinetic disorder (HK), the symptoms of which must be present before the age of 7 years (World Health Organization, 1993). It is a matter of two designations, which are not totally interchangeable because of the different diagnostic criteria. According to the diagnostic criteria of ICD-10, which are used in Europe, for confirmation of the diagnosis of HK, criteria must be fulfilled in the areas of attention, and simultaneously also in the areas of hyperactivity and impulsivity, whereas for confirmation of the diagnosis of ADHD according to the American criteria of DSM-V, symptoms in the area of inattention and/or symptoms of hyperactivity and impulsivity suffice. Therefore, almost all cases of hyperkinetic disorder fulfill the criteria for ADHD, but the reverse is not true (Taylor et al., 2004). The prevalence of the disease depends on the diagnostic measures used, but the estimated worldwide-pooled prevalence of ADHD among people up to and including 18 years of age is 5.29% (Polanczyk, de Lima, Horta, Biederman, & Rohde, 2007).
Celiac disease (CD) is a chronic, systemic immune-mediated disease, which initially affects the small intestine. It occurs as a consequence of a hypersensitivity to gluten and related prolamines in genetically predisposed people. Characteristic of the disease is the presence of specific antibodies, the typical genetic code for human leukocyte antigen (HLA) HLA-DQ2 and/or HLA-DQ8 and various degrees of intestinal impairment. The disease can manifest in diverse ways: it can be asymptomatic, atypical, or the clinical symptoms and signs can be clearly expressed and characteristic of CD (Husby et al., 2012; Krzisnik & Brecelj, 2014; Dolinsek & Mičetić-Turk, 2014). Some authors also treat ADHD as an extra-intestinal, neurological symptom of CD (Diaconu, Burlea, Grigore, Anton, & Trandafir, 2013; Zelnik, Pacht, Obeid, & Lerner, 2004). Gujral, Freeman, and Thomson (2012) report that the prevalence of CD is approximately 0.5% to 1% in the general population in different parts of the world. The only effective treatment is a strict, lifelong gluten-free diet (Hiller, 2003).
There are few studies investigating neuropsychiatric symptoms in patients with CD, including ADHD (Lahat, Broide, Leshem, Evans, & Scapa, 2000), but clear data on prevalence of CD among ADHD patients are scarce and controversial (Niederhofer & Pittschieler, 2006; Zelnik et al., 2004).
Only recently a systematic review of the literature concerning a possible link between ADHD and CD was published in a present journal (Ertürk, Wouters, Imeraj, & Lampo, 2016), which found eight studies that discussed the possible association. In five of these studies, they looked at whether the prevalence of ADHD is higher in patients with CD than in the general population (Dazy, Rubenstein, Holevinski, & Kao, 2013; Niederhofer & Pittschieler, 2006; Pynnönen et al., 2004; Ruggieri et al., 2008; Zelnik et al., 2004). In three studies they tested whether the prevalence of CD is higher in patients with ADHD than in the general population. In two of the three studies looking at the prevalence of CD among patients with ADHD, they found no difference between the prevalence of CD among ADHD patients and the general population (Güngör, Celiloğlu, Ozcan, Raif, & Selimoğlu, 2013; Lahat et al., 2000). However, Niederhofer (2011) found the prevalence of CD to be significantly higher (14.9%) in children and adults with ADHD. In the latter study, after a 6-month gluten-free diet, patients or their parents even reported an improvement in the symptoms of ADHD (Niederhofer, 2011). A comparison between these studies is difficult because some studies included not only ADHD patients, but also patients with learning disabilities (Zelnik et al., 2004) and ADHD-like symptomatology (Dazy et al., 2013; Niederhofer & Pittschieler, 2006). Further, the results are difficult to compare, as none of the stated authors consistently followed the European Society for Paediatric Gastroenterology Hepatology and Nutrition (ESPGHAN) criteria for stepwise diagnostics of CD.
In the literature, we find three possible answers to the question, how events in the intestine could influence the operation of the nervous system: the malnutrition theory, the inflammation theory, and the so-called “leaky gut syndrome.” According to the first theory, because of inflammation of the intestinal mucosa, there is a deficiency in absorption and malnutrition. Some studies have confirmed the lower concentrations of certain substances in children with ADHD, especially the minerals zinc, magnesium, and iron; B-vitamins; and essential fatty acids (Loscalzo, 2004), which are important in the synthesis of neurotransmitters. Studies that measured the effect of replacement of the deficient substances did not give unambiguous results (Curtis & Patel, 2008). According to the second theory, inflammatory cells, antibodies, and stress hormones, which arise with the inflammation, would negatively affect the nervous system. In this way, they explain the occurrence of neurological symptoms of CD. In patients with gluten ataxia, they found infiltrates of T lymphocytes in both the white substance of the cerebellum and in the peripheral nervous system. T lymphocytes and antibodies, which arise with intestinal inflammation, cross-react with some nerve cells, which have similar epitopes to their correct targets, and cause neurological symptoms (Hadjivassiliou et al., 2010). According to the third theory, the inflamed intestine leaks toxins, which disturb the functioning of the nervous system. The so-called “leaky gut syndrome” refers to the intestinal-brain model, which has still not been completely validated. Because of the abnormal metabolism of proteins (deficiency of enzymes for protein decomposition), aggregates of proteins and peptides are formed, which because of the increased permeability of the intestinal wall, cross the blood-brain barrier and cause problems in the brain (Whiteley et al., 2013). The pathological composition of the intestinal flora, which produce toxins that are thought to interact with the neuroendocrine system and disturb its functioning, is also thought to play an important role (Esparham, Evans, Wagner, & Drisko, 2014).
The primary aim of this study was to examine the prevalence of CD in children and teenagers with ADHD and to find a possible association between the two disorders to determine the reasonableness of screening tests for CD and the introduction of a gluten-free diet in children with ADHD. As mentioned already, only three studies examined ADHD patients for CD, and only one found a strong association. In light of these studies, we hypothesized that the prevalence of CD among young ADHD patients would not be higher than in the general population, and that there are no data supporting performing routine diagnostic tests for CD in ADHD patients (and vice versa) or for suggesting a gluten-free diet as a treatment option for ADHD unless there are additional indications.
Method
Participants
We invited 214 children and adolescents, who had been diagnosed with ADHD according to the valid diagnostic criteria of DSM-V, to participate in the study. A total of 102 children and adolescents (47.66 %) responded to the invitation. There were 18 (17.6%) girls and 84 (82.4%) boys aged from 4 to 18 years (mean age 12.8 years). Comorbidity was not an exclusion criterion. All patients were prospectively included in the study.
Instruments
ADHD was diagnosed according to DSM-V. All participants also fulfilled the diagnostic criteria for HK according to ICD-10. So, patients included in this study constitute the most severe presentation of ADHD symptoms.
A stepwise approach, as described in the recent ESPGHAN Guidelines (Husby et al., 2012) for the diagnosis of CD, was used in the study, with the detection of CD-specific antibodies against tissue transglutaminase 2 (t-TG2) as a first-line diagnostic method. In 53 children and adolescents, IgA class-tissue transglutaminase antibodies were determined in the laboratory by the ELISA method after exclusion of IgA immunodeficiency. In the remaining 49 children and adolescents, we performed the BiocardTM Celiac Test, a quick test for CD which detects antibodies against tissue transglutaminase in capillary blood. Testing was carried out in accordance with the manufacturer’s instructions, which ensured 97.8% sensitivity and 96.3% specificity. Rapid anti t-TG2 antibody detection at the point of contact can be performed with high accuracy that is similar to anti t-TG2 antibody detection by laboratory measurements. The evaluation of rapid tests is less reliable if done by untrained or lay people. Quantification, as in serum immunoassays, is not currently possible. In a participant with two invalid quick tests for CD, we performed laboratory measurements of the serum concentration of IgA antibodies because of the suspicion of IgA immunodeficiency. In proven IgA immunodeficiency, we excluded the possibility of CD by testing the serum concentration of AGA antibodies of IgG class (using the ELISA method, alpha-GliaPep IgG; Eurospital, Trieste, Italy). In addition, we also carried out genetic testing in this patient. None of the patients stated that they were on a gluten-free diet during or prior to the study.
Results
Among the 53 children and adolescents with a diagnosis of ADHD, in whom laboratory measurement of the concentration of serum IgA antibodies was performed, there were 42 boys and 11 girls. Their mean age was 13.2 years. With the determination of total serum IgA antibodies, IgA immunodeficiency was excluded in all children. The values of total serum class IgA ranged from 0.15 to 2.90 g/L, with an average of 1.32 g/L. The diagnosis of IgA immunodeficiency is made if the value of total serum IgA in a child, older than 4 years, is less than 0.05 g/L. The values of IgA antibodies against t-TG were between 0 in 4 pE, on average 1.19 pE. For a positive test, the value must be above 16 pE.
Among the 49 patients with ADHD, in whom we performed the quick test for CD, there were 42 boys and seven girls. Their mean age was 12.5 years. In 48 participants, the quick test was negative. In one participant (No. 13), the quick test for CD was performed twice, and both times it was invalid, so we performed laboratory measurement of the serum concentration of IgA antibodies, which confirmed our suspicion of IgA immunodeficiency. We excluded CD by measuring the serum concentration of IgG class antibodies. His result was 0 U/ml. We performed genetic testing, which proved that Participant No. 13 is heterozygous for HLA-DQ2/DQ5, thus, further follow-up is required.
Within the study, we excluded CD in all 102 patients with a diagnosis of ADHD and refuted the hypothesis that the prevalence of CD in children and adolescents with ADHD is higher than in the general population in Europe. In our study, the prevalence of CD in children and adolescents with ADHD was, therefore, less than 1%.
Discussion
The purpose of the study was to establish the prevalence of CD in a group of children and adolescents with ADHD and to discover a possible link between the two diseases. We wanted to establish whether it is reasonable to introduce routine screening tests for CD and a gluten-free diet in children and adolescents with ADHD. The question is not only academic as it is proven that a gluten-free diet is not only financially expensive but also associated with a poorer quality of life (Biagetti, Gesuita, Gatti, & Catassi, 2015), which can be particularly problematic in children and adolescents who already have a type of deprivation in their lives because of ADHD. Until now, there has been very little research on the possible link between CD and ADHD. In a recent systematic review published in a present journal, only eight studies were found on this topic and only three of these reported a positive association (Ertürk et al., 2016). In light of these studies, we hypothesized that the prevalence of CD among young ADHD patients would not be higher than in the general population.
In our sample of 102 children and adolescents with a diagnosis of ADHD, we did not find any patient whose results raised the suspicion of CD and warranted further diagnostic investigations. The suspicion was refuted on the basis of the determination of IgA class-t-TG antibodies: in 53 patients, by laboratory determination of IgA class-t-TG antibodies and in 48 patients by the quick test, Biocard™ Celiac Test. In one patient, in whom the quick test for CD was twice invalid because of IgA immunodeficiency, we performed laboratory measurement of IgG class antibodies and genetic testing. We confirmed the hypothesis that the prevalence of CD among children and adolescents with ADHD is not higher than in the general population.
Güngör et al. (2013) reported similar results to those in our study. The prevalence of CD in the group of children with a diagnosis of ADHD was the same as in the control group. They found a raised value of IgA-t-TG antibodies in 1.1% of children with a diagnosis of ADHD, changes typical of CD in a biopsy of the duodenal mucosa in 0.27% of children with a diagnosis of ADHD; raised values of IgA-t-TG antibodies were found in 0.8% children in the control group, while biopsy of the duodenal mucosa was normal in all children in the control group (Güngör et al., 2013).
An Israeli study also failed to find a link between CD and ADHD. In a group of 167 children and adolescents with neurological or psychiatric disorders (migraine headaches, epilepsy, hypotonia, motor dysfunction, ADHD [39 children]) and a control group without neurological or psychiatric disorders, they performed a stepwise diagnostic work-up of CD. In the group of children with neurological or psychiatric disorders, they did not find an increased prevalence of CD compared with the control group (Lahat et al., 2000).
Important different results were found in the study by Niederhofer (2011), where they found 14.9% of a group of 67 children and adolescents with ADHD had CD. The results presented by these authors can be questioned to a certain extent since the diagnosis of CD was not always confirmed according to the ESPGHAN criteria for the stepwise diagnostic investigation of CD, which requires intestinal biopsy and the determination of mucosal atrophy as the necessary diagnostic method for the diagnosis of the disease. Recent guidelines, however, warrant the stepwise approach used in our study, in which initial non-invasive tests can be used to select a subgroup of patients who need to undergo confirmatory intestinal biopsy.
The parents of some of the children with ADHD tried the introduction of a gluten-free diet in an attempt to reduce the symptoms of ADHD, even though the diagnostic work-up of CD had not been done. Some of them observed improvement in the symptoms of ADHD after the introduction of a gluten-free diet (Niederhofer, 2011). Considering the results of our study, which are in line with those of Lahat in Güngor, who did not find a link between CD and ADHD, the observed improvement cannot be attributed to the direct effect of the gluten-free diet. The improvement in symptoms after the introduction of a gluten-free diet can be explained in several other ways. Through the change in diet, there can also be a change in the parents–child relationship, which gives the child more positive attention. It is also possible that the children have subtle allergies to substances, which are wheat and other foods that contain gluten added during the manufacturing process. Avoiding these foods leads to improvement in the digestive symptoms and the children also feel better psychologically. Third, because of the smaller intake of carbohydrates, the children can simply have less energy, which has a beneficial effect on some symptoms of ADHD (Soreff, 2016).
The strengths of the present study are the relatively large sample size compared with some other studies (Lahat et al., 2000; Niederhofer, 2011; Pynnönen et al., 2004) and the precise diagnostic work-up of CD. There was a very good response from the invited patients or their parents or guardians as there was a response rate of 47.66%. We attribute the good response rate to the fact that parents or guardians are very aware of CD and that they have a good relationship with the child and adolescent psychiatrists who referred them for investigations.
The limitations of the study are the absence of a control group and the fact that the majority of participants came from the northeastern part of Slovenia. However, another study found the prevalence of CD among students in northeastern Slovenia to be 1.49% (Cencic & Beric, 2013) and the prevalence of CD in Slovenia to be 0.22% (Mičetić-Turk et al., 2011).
Conclusion
In our sample of children and teenagers with ADHD, the prevalence of CD was not higher than in the general population. Among the 102 participants, we did not find anyone with an increased concentration of CD-specific antibodies. On the basis of the results obtained and the results of similar studies, we can conclude that there are not enough data to support screening for CD and the introduction of a gluten-free diet in children with ADHD without additional indications. However, these conclusions were based on relatively small number of studies and participants, so there is still need for studies on larger sample size, which may bring different results.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.