Challenges in endoscopy and histological diagnosis of celiac disease

Narrow-band imaging (NBI) is an interesting endoscopic imaging technique that helps in visualizing some mucosal changes consistent with celiac disease. The technique seems to be more effective than conventional light endoscopy and the paper of Singh et al. opens up a new prospect in the endoscopic aspects of celiac disease [ 1 ]. Unfortunately, even with this new technique the value of endoscopy in the diagnosis of celiac disease is limited to the particular grade of atrophy (Marsh 3) that is a constant finding in a nonchallenging subgroup. Celiac patients with Marsh 3 lesions are usually easily diagnosed, and most of them present with positive serology. The histology in this group is clearly defined and displays characteristic lesions; although relatively rare, this form cannot be mistaken for other disorders. The main challenge lies in dealing with those patients presenting with milder enteropathy (Marsh 1 or 2) and other atypical features: the most prevalent form of disease is now represented by this group, and there is no gold standard for identifying these patients in whom the villi are normal and there is only a pathological increase of T lymphocytes (> 25 per 100 epithelial cells) [ 2 ] [ 3 ]. This common feature of gluten sensitivity falls under the definition of microscopic enteritis [ 4 ], yet different pathological entities might present with the same feature and characteristics described in this entity. An accurate diagnosis requires accurate diagnostic work-up. Unfortunately, there is a massive gap between evidence-based knowledge and our practice. A survey in Birmingham on the numbers of biopsies taken to diagnose celiac disease showed that 63 % of cases had fewer than four duodenal biopsies. Many endoscopists seem very reluctant to take an adequate number of biopsies as recommended in the major guidelines. From the practical point of view, it is also important to stress that the biopsy samples are usually not correctly oriented, that histological evaluation is poor and insufficient, or that the endoscopist often provides no clear clinical indication. Milder abnormalities such as Marsh 1 and 2 are ignored and under-reported by some European pathologists. Some general practitioners and other clinicians might not be very interested in screening the high risk groups such as symptomatic first-degree relatives, those with type 1 diabetes, and those labeled as having “irritable bowel syndrome.” As the histological features of celiac disease present patchily in some cases, NBI could be very useful for selecting the biopsy sites in the small bowel [ 5 ]. In recent years, new endoscopic approaches to duodenal villous abnormalities have been proposed, with the aim of facilitating the identification of endoscopic abnormalities typical of celiac disease (such as a mosaic or a nodular mucosal pattern, or scalloped folds), which have been associated with villous atrophy. In particular, endomicroscopy is another technique that has been compared with histological evaluation [ 6 ] [ 7 ] [ 8 ] [ 9 ]. Concerning total atrophy in celiac disease, the cerebroid appearance or mosaic pattern was evident with endomicroscopy and an analysis of villi architecture may suffice for a correct diagnosis, whereas the identification of T lymphocytes, a marker of initial celiac disease, is more difficult. Thus these techniques enable more accurate targeting of biopsies in selected patients with patchy findings or early celiac disease, and consequently might yield greater accuracy in pathological examinations. Further controlled, prospectively collected data are needed to confirm the effectiveness of NBI and CLE in diagnosing celiac disease at earlier stages.