Objective: As it has been demonstrated that a careful duodenal inspection during upper gastrointestinal endoscopy may be useful in predicting coeliac disease, we tried to define the usefulness of endoscopy in detecting unsuspected coeliac patients.
Design and methods: We considered all the first diagnoses of coeliac disease from 1992 to 2001, i.e. 110 patients with a biopsy-proven diagnosis of coeliac disease. From 1992 to 1997, neither of the endoscopists paid careful attention to the endoscopic features of coeliac disease in the course of the examinations performed for indications other than coeliac disease. From 1998 to 2001, the same endoscopists looked very carefully at these endoscopic features, regardless of the indication for the procedure.
Results: Over the first period, 22/16,081 patients endoscoped for the first time had a histological diagnosis of coeliac disease, with a prevalence of 1/731. In all 22 patients the indication for the examination was the suspicion of coeliac disease. The endoscopic appearance of the duodenum was indicative in 16/22 (72.7%) patients. Over the second period, the diagnosis of coeliac disease was made in 88/10,410 patients endoscoped for the first time. The prevalence of the disease was 1/118 examinations performed. The endoscopic appearance of the duodenum was indicative in 70/88 (79.5%) patients. In 13/88 patients, the diagnosis of coeliac disease was presumed because of the macroscopic appearance of duodenum, lacking a past history suggestive of coeliac disease.
Conclusions: Despite a still open controversy on the accuracy of endoscopic markers in the diagnosis of coeliac disease, we have found that in subjects not suspected for coeliac disease and undergoing an upper gastrointestinal endoscopy for other reasons, attention to the endoscopic pattern could facilitate the identification of a relevant number of cases.