Background Endoscopic confocal laser endomicroscopy (eCLE) is an established technique that allows clinical evaluation of mucosal integrity by fluorescein leaking through the mucosa upon duodenal food challenge (DFC). Analysis of eCLE with DFC in eosinophilic esophagitis (EoE) would be interesting to evaluate epithelial barrier dysfunction also in other regions of the gastrointestinal tract and to characterize potential individual food allergens that trigger the esophageal inflammation.
Methods In an observational and proof of concept study we evaluated 9 patients with histologically proven EoE by eCLE and DFC. Severity of symptoms were graduated according to the validated symptom-based EoE activity index. The endoscopic appearance of the esophagus was described according to the Endoscopic Reference Score System (ERERS). Spontaneous and food induced transfer of fluorescein into duodenal lumen were detected 10 minutes following intravenously application of fluorescein and 10 minutes after DFC. Food allergens were yeast, egg, soy, milk, and wheat, respectively. Local application of sodium chloride solution 10 % to the duodenal mucosa before DFC served as a control. Patients responding to DFC received a dietary exclusion therapy according to the results of DFC.
Results We investigated 9 patients with EoE (8 men, 49.7±13.8, 36–76 years). Symptom-based EoE activity index was 79±27.4, 33–100. In all patients EoE was confirmed by histology with number of esophageal mucosal eosinophilic granulocytes > 15/HPF, (91.4±77.4, 42–263). Mean ERERS score was 4.5±1.3, 3–7. None of the patients was aware of any food intolerance. eCLE revealed one patient with spontaneous transfer of i. v. fluorescein into duodenal lumen before DFC (“leaky gut”). 40 DFC were performed in the remaining 8 patients of whom 5 patients (61 %) responded to DFC. Rank order of fluorescein leakage upon DFC was wheat and milk in 37.5 % each, soy in 25 %, and egg in 12.5 %. The patients were treated by PPI (n=9), esophageal bouginage (n=5) and/or local corticoid therapy (n=3). The 5 patients responding to DFC received an additional food exclusion dietary advice focussed on the results of DFC. All patients reported a reduction of their symptoms. EoE activity indexes of patients with positive DFC were 73.7+28.6, 33–100 before and 22.7+37.9, 0–79 four weeks after food exclusion.
Conclusion The findings of our proof of concept study suggest that eCLE with DFC may be an interesting tool to further evaluate patients with EoE. This technique has the potential to identify patients who may benefit from an additional individual dietary therapy.