Secondary Logo

Journal Logo

Short Communications: Gastroenterology: Celiac Disease

The Effect of Gluten-free Diet on Clinical Symptoms and the Intestinal Mucosa of Patients With Potential Celiac Disease

Mandile, Roberta; Discepolo, Valentina; Scapaticci, Serena; Del Vecchio, Maria Rosaria; Maglio, Maria Antonia; Greco, Luigi; Troncone, Riccardo; Auricchio, Renata

Author Information
Journal of Pediatric Gastroenterology and Nutrition: April 2018 - Volume 66 - Issue 4 - p 654-656
doi: 10.1097/MPG.0000000000001745
  • Free


What Is Known

  • Potential celiac disease is a clinical condition characterized by the presence of anti-tissue tranglutaminase antibodies in the blood, but with a preserved intestinal villi architecture.
  • Patients with potential celiac disease can present or not clinical symptoms and/or signs.
  • Whether patients with potential celiac disease should or should not receive a gluten-free diet has not been well established yet.

What Is New

  • Symptoms after gluten-free diet improve in only about half of the cases.
  • The intestinal inflammation and infiltration does not always improve after gluten-free diet in potential celiac disease.

Celiac disease (CD) is defined as an immune-mediated systemic disorder elicited by gluten and related prolamines in genetically susceptible individuals, characterized by the presence of a variable combination of clinical manifestations, CD-specific antibodies, human leukocyte antigens (HLA) compatible haplotype and enteropathy (1).

The intestinal damage can vary from villous atrophy, typical of the overt disease, to minor or absent histological alterations. Indeed, the term potential celiac disease (PCD) has been coined to identify a subset of patients characterized by serum endomysium and tissue transglutaminase (TG2) antibodies, but a normal (Marsh 0) or infiltrated (Marsh 1) intestinal mucosa. These patients may present or not clinical symptoms of intestinal or extraintestinal origin (2). Despite it is well known that in overt CD a gluten-containing diet is related to a higher risk of nutritional, autoimmune, or even oncologic disorders, very few studies have analyzed up to now the risks of a gluten-containing diet in PCD. This is why, from a clinical point of view, the management of patients with PCD remains vividly debated.

The major clinical issue deals with the treatment of asymptomatic patients with PCD. Some studies have suggested that patients with mild enteropathy should all be addressed to a GFD since this condition can be considered as the first step of the overt disease (3). Recent data, however, highlight that only about one third of asymptomatic patients with PCD will develop intestinal damage over time. For this reason, a generalized prescription of GFD to all asymptomatic patients could often reveal to be an overtreatment (4).

On the contrary, as far as symptomatic patients with PCD are concerned, the scientific community is eager to suggest GFD. Nonetheless, even for these patients, the effects of GDF in PCD have never been specifically investigated.

In the present work, we evaluated the effect of GFD in patients with PCD, analyzing changes in clinical symptoms and intestinal mucosa status.


We enrolled 330 patients from 2007 to 2016, classified as Potential Celiac because of positive anti-TG2 antibodies in 2 different determinations, confirmed with endomysial antibodies, and a normal or slightly infiltrated mucosa (Marsh stages 0 or 1). All had HLA-DQ2 and/or -DQ8 haplotypes. Total serum IgA antibodies were within the normal range in all of them. We selected from this cohort 65 patients (21 boys and 44 girls, median age at diagnosis 7.27, range from 1 to 13 years) who received a gluten-free diet (GFD) because of the presence of symptoms (N = 47) or for parents’ choice (N = 18). Fifty-nine of them started the GFD at the time of first diagnosis, whereas 6 started a GFD during follow-up (in this case a biopsy was performed before starting the GFD to assess the status of the mucosa). These last cases were previously on a gluten containing diet for a median time of 36 months.

Every 6 months, symptoms were checked (mean follow-up time of 42.3 months) to evaluate clinical response to the GFD. We defined as positive clinical response the disappearance of all the symptoms within 12 months of GFD, partial clinical response the disappearance of only some of the symptoms presented (for a patient who presented at the same time more than 1 symptom) and no clinical response the persistence of symptoms upon at least 12 months of GFD. We also considered as nonresponders those patients who seemed to improve after a GFD, but in whom symptoms did not reappear when they were rechallenged with gluten for at least 12 months, because symptoms could not be defined as gluten dependent.

After at least 24 months of GFD (mean time 34 months, range 24–60 months), a duodenal biopsy was also proposed. Only 9 patients out of 65 accepted. The intestinal evaluation included both morphologic and immunohistochemical analysis. In particular, biopsies before and after GFD were compared for Mash grade, lamina propria CD25+ cells count, CD3+ and γδ+ intraepithelial cells count, and γδ+/CD3+ cells ratio as described elsewhere (5). The presence of intestinal IgA anti-TG2 deposits was also assessed by double immunofluorescence. For intestinal anti-TG2 deposits, a semiquantitative scoring was assigned by a pathologist (M.M.) as described elsewhere (6).

The statistical analysis was performed, according to the type of parameter, with Fisher exact test or Wilcoxon rank sum test.

This study was approved by the University Federico II ethical committee.


Of the 65 patients with PCD considered for this study, 47 received a GFD because of the presence of symptoms/signs, whereas 18 were asymptomatic. Symptoms/signs at diagnosis were low body mass index (BMI) (36%), recurrent abdominal pain (34%), followed by diarrhea (19.15%), osteoporosis (10.6%), short stature (6.4%), vomiting (4.25%), hypertransaminasemia (4.25%), low blood ferritin (4.25%), pervasive developmental disorder (2.1%), dermatitis herpetiformis (2.1%). On note, some patients presented more than 1 symptom at the same time.

Out of 47 symptomatic subjects we excluded from the analysis all those who were lost at the follow-up (7/47) and those who received other therapies in addition to GFD since we could not evaluate for them the specific contribution of each therapeutic strategy to the clinical resolution. Of the remaining 35 patients, 54% (19/35) showed a positive clinical response in the first 12 months (71% of patients with abdominal pain, 55.6% of patients with diarrhea, and 50% of patients with vomiting), 6% (2/35) showed a partial clinical response, and 40% (14/35) no clinical response (Fig. 1). As expected, in all patients GFD induced a fall of anti-TG2 blood levels in the first 6 months, and thus confirming the adherence to the diet.

Symptoms before and after GFD. Number of patients with a certain symptom at the diagnosis (black column) and after at least 12 months of GFD (grey column). BMI = body mass index; GFD = gluten-free diet.

Nine patients out of 65 received a second biopsy while on GFD (mean GFD duration 34 months, range 24–60 months). Two of 9 (22%) patients were Marsh 0 and 7 of 9 (78%) were Marsh 1 at the time of diagnosis. A variable outcome in terms of Marsh grade was observed after at least 12 months of GFD, indeed 2 of 9 (22%) remained Marsh 0, 4 of 9 (44.5%) remained Marsh 1, 3 of 9 (33.5%) improved from M1 to M0, and, as expected, none passed from an M0 to M1. Not significant difference was found in Marsh grade (exact Fisher P = 0.33) nor in the number of CD3+ cells (mean value at diagnosis 43 ± 18 cells/mm epithelium, mean value after GFD 32 ± 19, P = 0.9) (Fig. 2A), the number of CD25+ cells (mean value at diagnosis 11 ± 12 cells/mm2 lamina propria, mean value after GFD 9 ± 9, P = 0.8) (Fig. 2B) and the number of γδ+ cells (mean value at diagnosis 10 ± 9 cells/mm epithelium, mean value after GFD 12 ± 16, P = 0.59) (Fig. 2C). Equally, no statistically significant difference was observed for the presence of intestinal deposits of anti-TG2 antibodies after GFD (P = 0.60) (Fig. 2D).

Intestinal CD3+ (A), CD25+ (B), γδ+ cells (C) and intestinal anti TG2 deposits (D) in potential celiac disease (PCD) patients on gluten-free diet (GFD). We evaluated the counts of CD3 positive (A) and γδ positive (C) cells per mm of epithelium and the number of CD25 positive (B) cells per mm2 of lamina propria by IHC. The density of the intestinal anti-TG2 deposits (D) was evaluated by double immunofluorescence (the black part of the column represents the number of patients with negative anti-TG2 deposits, the grey part of the column represents the number of patients with positive anti TG2 deposits). No statistically significant difference was observed in these parameters after 34 months (on average) of GFD (P > 0.05). N = 8 (9 biopsies were analyzed before and after GFD, but for 1 biopsies only the Marsh grade was evaluated).


PCD is an emerging condition, as it is now estimated to represent approximately 11% of the total diagnosis of CD (4). Very few studies have, however, analyzed long- and short-term risks and benefits related to a gluten-containing diet in these patients. The lack of information regarding the clinical management of PCD in children prompted us to collect and analyze our data regarding patients with PCD who received a GFD. We evaluated both symptomatic and asymptomatic patients on GFD and we followed them up over time. Asymptomatic patients received a GFD based on parents’ choice.

The first evidence emerging from this study was that only 47 of 330 (14.25%) patients with PCD were prescribed a GFD for the presence of symptoms. Indeed, in our setting the majority of pediatric patients with a diagnosis of PCD is asymptomatic. Among symptomatic patients, symptoms of intestinal origin, in particular low BMI and abdominal pain, were more frequent, representing overall the 76% of the recruited symptomatic population. Notably, this situation is different from what is found in adults with PCD. A recent study of Volta et al (7) has highlighted that the majority of adult patients with PCD are symptomatic, and most common symptoms are of extraintestinal origin. Two hypotheses can explain these discrepancies. Extensive screening for CD is more common in children than in adults, thus asymptomatic patients could just be more frequently detected among children. It has, however, been demonstrated that the mean age of adults with asymptomatic PCD is significantly lower compared to adult with symptomatic PCD (7). This is consistent with our results, and it could suggest that symptoms require time to manifest.

When we followed up our cohort of symptomatic patients on GFD, we noted that the percentage of responding and nonresponding patients was similar, even though anti-TG2 serum level declined upon antigen elimination in all of them, confirming the adherence to the diet. In particular, about half of the patients showed a positive clinical response in the first 12 months while the other half showed a partial or no clinical response. Moreover, not all symptoms showed the same tendency to disappear after GFD. For instance, abdominal pain and diarrhea were more prone to improve than low BMI. Anyway, it is clear that not all symptoms are gluten dependent in PCD. We can speculate that, in some cases, the association between CD and irritable bowel syndrome may be a significant confounding factor (8).

Also the intestinal inflammation and infiltration is not significantly affected by GFD. All the histological parameters analyzed, surprisingly, did not change significantly after GFD, although we acknowledge the small size of the group. In 1 patient with a codiagnosis of type 1 diabetes and a positive family history of CD, CD25+ cells, γδ+ cells, and γδ+/CD3+ cells ratio were even increased after GFD. This is consistent with previous observations that have been shown in literature: first-degree family relatives of patients with CD, particularly if homozygous for DQ2, have a higher count of γδ+ intraepithelial lymphocytes, even when healthy and seronegative (9). In addition, the increase of the γδ+ T cells fraction has been found persisting after years of GFD (10). On the contrary, the persistence of intestinal anti-TG2 deposits could just be related to a too short follow-up time.

It is important to highlight that in the present study we did not analyze long-term effects of GFD in PCD, such as the development of other autoimmune diseases or oncologic disorders. This issue remains a major point that needs to be evaluated in the decision whether patients with PCD should receive a GFD. Unfortunately, up to now, a clinical study with an adequate follow-up time and control group has not been performed yet.

In conclusion, caution is necessary in PCD before attributing all abnormalities to gluten, as we demonstrated that often symptoms and inflammatory markers of intestinal mucosa do not improve after a prolonged period of GFD. Prospective randomized clinical trials on the effect of GFD in PCD are necessary to offer evidence-based guidelines on the management of patients with PCD.


1. Husby S, Koletzko S, Korponay-Szabó IR, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition Guidelines for the diagnosis of coeliac disease. J Pediatr Gastroenterol Nutr 2012; 54:136–160.
2. Biagi F, Trotta L, Alfano C, et al. Prevalence and natural history of potential celiac disease in adult patients. Scand J Gastroenterol 2013; 48:537–542.
3. Kurppa K, Collin P, Viljamaa M, et al. Diagnosing mild enteropathy celiac disease: a randomized, controlled clinical study. Gastroenterology 2009; 136:816–823.
4. Auricchio R, Tosco A, Piccolo E, et al. Potential celiac children: 9-year follow-up on a gluten-containing diet. Am J Gastroenterol 2014; 109:913–921.
5. Paparo F, Petrone E, Tosco A, et al. Clinical, HLA, and small bowel immunohistochemical features of children with positive serum antiendomysium antibodies and architecturally normal small intestinal ucosa. Am J Gastroenterol 2005; 100:2294–2298.
6. Tosco A, Auricchio R, Aitoro R, et al. Intestinal titres of anti-tissue transglutaminase 2 antibodies correlate positively with mucosal damage degree and inversely with gluten-free diet duration in coeliac disease. Clin Exp Immunol 2014; 177:611–617.
7. Volta U, Caio G, Giancola F, et al. Features and progression of potential celiac disease in adults. Clin Gastroenterol Hepatol 2016; 14:686.e1–693.e1.
8. Saps M, Sansotta N, Bingham S, et al. Abdominal pain-associated functional gastrointestinal disorder prevalence in children and adolescents with celiac disease on gluten-free diet: a multinational study. J Pediatr 2017; 182:150–154.
9. Holm K, Mäki M, Savilahti E, et al. Intraepithelial gamma delta T-cell-receptor lymphocytes and genetic susceptibility to coeliac disease. Lancet (London, England) 1992; 339:1500–1503.
10. Jarvinen TT, Kaukinen K, Laurila K, et al. Intraepithelial lymphocytes in celiac disease. Am J Gastroenterol 2003; 98:1332–1337.

gluten-free diet; intestinal biopsies; intraepithelial lymphocytes; potential celiac disease

© 2018 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,