Nonceliac wheat sensitivity in the context of multiple food hypersensitivity: new data from confocal endomicroscopy.

Dear Editor, We enjoyed reading the article by Fritscher-Ravens et al who showed, by confocal endomicroscopy, that candidate food antigens caused immediate duodenal mucosa damage in irritable bowel syndrome (IBS) patients with a prolonged clinical history of symptoms after meals. Their in vivo data add evidence to the relationship between IBS and food allergy and seem to reinforce our hypothesis that a percentage of “nonceliac wheat sensitive” (NCWS) -patients with an IBS-like clinical presentation could suffer from non-immunoglobulin E-mediated wheat allergy. However, we would suggest that the very high percentage of positive confocal laser endomicroscopy patients (CLE) -22 out of 36- found in the study of Fritscher-Ravens et al could depend on their inclusion criteria (refractory daily symptoms >1 year, daily shortly after meal symptoms); in our experience, the frequency of food hypersensitivity diagnosed by double-blind, placebo-controlled (DBPC) food challenges in IBS is slightly <30% (276 patients out of 920). Apart from the epidemiologic data, which were not the objectives of this pilot study, we would like to underline some aspects of the study and make some suggestions for future research. It is interesting that a total of 32 reactions were analyzed, with different food antigens, in 22 CLE-positive patients and that the second most frequently offending food, after wheat, was cow’s milk. This is in keeping with our data about the high frequency of multiple food hypersensitivities in patients with NCWS. We showed that 206 of 276 NCWS subjects also became symptomatic after DBPC cow’s milk proteins challenge. These observations should induce the physicians who suspect a relationship between NCWS or food hypersensitivity and IBS to suggest an elimination diet with the exclusion of more food rather than just wheat, and that the reintroduction should be performed singly and with great caution, as described. In fact, a lack of response to a wheat-free diet could depend on hypersensitivity to other food antigens which are still included in the patients’ diet. We found also of great interest that CLE showed significantly higher intraepithelial lymphocyte (IEL) count in CLE positive than CLE-negative patients and in controls. Furthermore, histology showed that the mean values of IEL in CLE-positive patients were 26.4 ± 2.7 per 100 cells. Overall, this could indicate a state of mucosal inflammation owing to food hypersensitivity. A previous NCWS study which excluded patients with >25 IEL per 100 EC in the duodenal mucosa, very probably missed the group of NCWS patients who had an immunologic pathogenesis at the basis of their troubles. However, in our opinion, NCWS is a heterogeneous condition, which includes different subgroups of patients and the “allergic hypothesis” does not exclude that, in other NCWS patients, wheat amylase trypsin inhibitors or fermentable sugars4 could be the main pathogenetic triggers. Finally, we think that the authors showed that CLE is an excellent instrument to demonstrate food-related reactions in IBS and to separate a subgroup of the NCWS -those with non-immunoglobulin E-mediated wheat hypersensitivity- from the confuse melting pot that NCWS still is. However, awaiting a wider diffusion of this endoscopic means, and taking into account that the economic resources are decreasing in developed countries, it would be very important to correlate the CLE finding with simpler, noninvasive biomarkers. In this respect, it would be interesting to know whether CLE findings correlate with the eosinophil cationic protein concentrations in the stools or with the flow cytometric allergen stimulation assay results, biomarkers that showed a good concordance with the DBPC challenge results in IBS patients.