Screening for celiac disease among children with overweight and obesity: toward exploring celiac iceberg

Introduction

Celiac disease (CD) is a systemic, lifelong immune mediated disease characterized by small intestinal enteropathy triggered by exposure to dietary gluten, that may occur in individuals with a genetic susceptibility [1], [2], [3], [4], [5]. Its global prevalence is estimated at approximately 1%, with regional differences and studies suggest it is increasing [3], [4]. In Italian school children, the most recent literature suggests that prevalence has nearly doubled in the past 25 years from 0.88 to 1.58% [5].

The clinical presentation of CD has changed. In the last two decades, diarrhea, poor growth, and malabsorption have progressively decreased as presenting symptoms of CD among adults and children, whereas symptoms such as fatigue, constipation, headache, nausea, iron deficiency anemia, and joint pain have increased [3], [5], [6]. As reported by Nenna R et al. [7], a modified clinical spectrum was also noted and referred to a large study on school-age children screened for CD, showing a silent, typical, atypical, and potential CD respectively in 64, 28, 7, and 1% of cases. In addition to these changes in presentation, several studies have shown that CD is common in patients who are normal weight, overweight, or affected by obesity and not limited to those who have developed under-nutrition; previous work analyzing weight at CD presentation found that 8.8–20.8% of patients might be overweight and 0–6% of patients might be classified as obese at CD onset [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17]. The link between CD and excessive weight gain is unknown; however, this co-presence suggests that CD may be more widespread than previously suspected, generating a need for further research to define an appropriate assessment and monitoring according to an overall healthcare plan.

The aim of this study was to investigate the prevalence and clinical presentation of CD, detected by screening children with excessive weight gain in order to explore a new potential part of the celiac iceberg.

Patients and methods

Results

CD was detected in 8 patients (4%), mean age 9.43 ± 4.68 years, five female and three males. Both tTG-IgA and EMA-IgA autoantibodies were positive in all subjects with CD; no patient showed tTG-IgA 10 times higher than the upper cutoff. CD diagnosis was confirmed by EGDS (Marsh stage 3b in seven patients and stage 3c in one subject). HLA-DQ2 and/or DQ8 haplotypes positivity was reported in all subjects with CD.

Clinical features and biochemical parameters of the patients are reported in Tables 1 and 2. Personal history and lifestyle variables are showed in Tables 3 and 4.

Discussion

We found the prevalence of CD in overweight and patients with obesity referred to our outpatient clinic of the Pediatric Endocrinology Unit to be 4% which is higher than previously reported in similar cohorts. In addition we found that these patients found to have CD upon screening were more likely to consume less fruits and vegetables weekly. Additionally, these patients were more likely to report headache as a symptom which is consistent with work by Hom et al. which found that one-third of children and adolescents with CD have recurrent headaches at the time of diagnosis [29]. We reported that the familiar positivity for other autoimmune diseases was revealed in six of eight patients who screened positive in our study for CD. These diseases included one rheumatoid arthritis and five autoimmune thyroiditis (AT) confirming the close relationship between CD and AT disease, as already reported by other authors partly and due to a common genetic predisposition [30]. We did not find significant associations between sex, ponderal status, or other lifestyle variables and CD in our sample.

Celiac disease prevalence has increased over recent decades [4], [5] and our findings support previous work suggesting that CD is prevalent in patients with overweigh and obesity. While traditionally CD is thought of as causing malabsorption, and thus researchers and clinicians have focused on weight loss and poor growth, recent studies have found that patients who were normal weight, overweight, and affected by obesity are common [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17]. Nenna et al. [31] evaluated retrospectively CD prevalence in a large series of overweight/obese children and adolescents and reported that 1.11% of patients were positive for serology and showed villous atrophy; in this population some patients with CD showed gastrointestinal symptoms, though they had not been complaining about them, and attended the outpatient visit for the nutritional disorder. Here, we reported a remarkable CD prevalence (4%) in an Italian pediatric population referred to an auxological outpatients’ clinic for overweight/obesity. Our overall prevalence is likely even higher than reported here, given that we excluded patients with gastrointestinal symptoms, with previous diagnosis of CD and or family history of CD. Our observation confirms the importance of screening also when excessive weight gain exists [6].

The pathogenetic basis of CD coexistence in subjects with excessive weight gain remains unclear; according to the “compensatory” hypothesis, the duodenum—jejunum atrophy in CD patients could be balanced by an increased distal intestinal absorption [32]. This adaptation is due to structural modifications that cause an increased enteral absorption; so unbalanced diets may lead children to gain weight [33]. Even though no significant difference in age was noted between subjects with and without CD, the age of those affected is young (9.43 ± 4.68) suggesting that intestinal adjustments occur overtime [34], [35], [36]. Moreover it can be expected that the youngest children diagnosed with CD show acute and classic symptomatology, while older children present a more nuanced or atypical symptomatology [34], [35], [36].

While 75% of patients identified as having CD had a family history of autoimmune disease, we did not find any significant differences in medical history or lifestyle variables in patients with or without CD. These results reinforce the knowledge that an individual’s genetic predisposition plays a crucial role in the development of CD. As reported, autoimmune diseases, including CD, result from an interaction between environmental factors and genetic predisposition. More than one autoimmune disease frequently develops and clusters in the same individual and family members, supporting the hypothesis that there are likely to be common chromosomal gene regions conferring susceptibility to more than one disease [37].

Regarding symptoms, our data confirmed that prevalence of idiopathic headache at diagnosis was higher in children with excessive weight gain [38], [39], [40]. The pathogenetic mechanism underlying the relationship between CD and headache remains unclear; however, this symptom should be considered as a neurological manifestation among children with CD [38], [39], [40]. Headaches associated to CD are predominantly migraines [41] and multiple nutritional, immunological and inflammatory factors could be involved in triggering migraine attacks [41].

Lifestyle including diet plays an important role in health improvement in CD in which the only treatment includes nutritional therapy [1], [2], [42]. Studies to date have focused on timing of gluten introduction and amount of gluten ingested as it relates to celiac disease onset [43]. However, we are unaware of studies evaluating other dietary habits in children prior to CD diagnosis.

The limited weekly consumption of fruits and vegetables noted in our sample with CD could hide an early onset of gastrointestinal symptoms that have driven to such poor plant food intake; however, evidence suggests that western diet, high in sugar and fat and low in vegetables and fruit, a can cause dysbiosis, an alteration in the composition of the microbiota, which could lead to aberrant immune responses. Low intake of fiber alters intestinal microbiota in humans resulting in decreased short chain fatty acid production by microbes which increase the intestinal pH [44]. Thus, it is plausible that low fruit and vegetable consumption promote the growth of microbes that could have detrimental effects on their host [44] besides being a risk factor for chronic diseases [45].

The emerging role of epigenetic mechanisms and the potential mechanisms of action of the intestinal microbiota and specific components that impact CD pathogenesis [46] may play a key role in CD and obesity coexistence. Additionally, the increase prevalence rate of overweight and obesity in children may also explain excessive weight coexistence in children affected by CD [6]. Finally, the unconfirmed but nonetheless interesting hypothesis that overnutrition is due to a compensatory high energetic yield secondary to slow functional adaptation of the atrophic mucosa could not be excluded [40].

Study limitations should be acknowledged starting from the choice the authors made to exclude patients with family history of CD and intestinal symptoms since in these patients screening is routinely performed and the authors instead wanted to investigate the asymptomatic ones. There is no control group since the prevalence in the general population is already well known also in pediatrics. No weight-loss follow-ups are reported, since this was not the aim of this study, although BMI improvements were detected. Further studies focusing on the relationship between CD and obesity are recommended.

Clinical observational studies with large number of pediatric patients with excessive weight gain are mandatory to confirm the observed results. Further research in this field is needed to achieve a better understanding of the networks involved in CD pathogenesis.

In conclusion, CD should be considered in pediatric patients with excessive weight gain, besides familial predisposition to other autoimmune diseases representing a well-known risk factor for CD development. Additionally, even though the relationship between headache and CD is not well defined, patients with idiopathic headache should be screened for CD. Professionals should be aware of the broad spectrum of this disease, in order to perform an early screening and to prevent the organic complications of untreated CD.