Celiac disease (CD) is the most common chronic autoimmune enteropathy in Western Countries with an estimated prevalence ranging from 1:100 to 1:200. It is usually characterized by a benign course with clinical and histological remission, provided that a strict gluten-free diet (GFD) is followed by patients. Less frequently, CD can be characterized by a complicated course, when facing with a refractory disease (RCD) or with malignancies of the gastrointestinal (GI) tract, namely lymphoma and adenocarcinoma of the small bowel (SB). Different studies estimated a relative risk (RR) for neoplastic GI complications in CD ranging from 2 to 40 and from 10 to 60 for primary gut lymphoma and adenocarcinoma, respectively. Although uncommon, the discussed malignancies has a severe prognosis, reflecting the need for an early diagnosis. This project aims to establish an enteroscopic approach to improve the diagnostic timing and survival of CD patients at risk to develop SB tumors.

In the last years, the difficulty to explore SB represented a problem for an early diagnosis of the intestinal tumors. Small bowel videocapsule endoscopy (VCE) and double-balloon enteroscopy (DBE) currently represent the diagnostic gold standard for the SB mucosal lesions. The enteroscopic techniques have been applied for both the diagnosis and management of patients with complicated CD, even if the currently available data are derived from retrospective investigations. Moreover,different clinical and demographic risk factors for SB tumors in CD have been identified; thus, the following statements should be taken in consideration to plan a diagnostic strategy for SB malignancies in a subset of CD patients: i) the risk of GI malignancy in CD is not homogenously distributed; ii) attention should be paid to CD patients with particular demographic and clinical features; iii) in this subset of CD patients an early diagnostic strategy for SB tumors is yet to be evaluated.

Hyphotesis and Significance: In a cohort of CD patients with particular clinical and demographic characteristics is possible to establish a programme for early diagnosis of small bowel complications. Specific Aim:To identify CD patients at higher risk of developing small bowel neoplasia and to evaluate, in this subset, the usefulness of VCE and DBE. To evaluate the diagnostic yeld of DBE and VCE in this setting of patients. To establish a flowchart for early diagnosis of small bowel tumors in CD patients.

It is our intention to evaluate at least 180-240 CD patients satisfying the entry criteria, to be enrolled during the three years. At least 500 OGIB controls will be enrolled including the prospective and retrospective (VCE lombard registry) parts (rate of SB tumors in this group 0.6%, see image of preliminary data). The known rate of SB tumors in the general population is 0.009%. It is expected a rate of SB tumors in selected CD of 5% (data from the preliminary findings). Comparison between groups (5% level of significance), if these neoplastic rates and result expectation should be confirmed, has a statistical power greater than 80%. All the assumptions will be verified using SPSS ver. 18. A P value of <0.05 will be considered statistically significant (significance level of the tests 5%, two tails). As mentioned, the sample size has been calculated presuming a 5%prevalence of malignancies in the CD cohort. The normal distribution of the sample will be verified through the Kolmogorov-Smirnov test. Continuous variables will be analyzed with the Anova Oneway variance test or with the non parametric Kruskal-Wallis test. The significance level will be further verified by multiple comparison analysis (Tukey or Mann-Whitney' s test). Categorical variables will be compared wih X2 or Fisher' s Exact test.

Inclusion Criteria:

i) diagnosis of celiac disease when aged over 50 years, with persistence/recurrence of gastrointestinal symptoms after 6 months on gluten free diet ii) lack of compliance to gluten free diet (defined as conscious and regular gluten ingestion for at least two years) iii) presence of alarm symptoms/signs, either at diagnosis or during the follow up

Exclusion Criteria:

• age lower than 18 years
• absence of written informed consent to participate in the study