Analysis of Biopsies From Duodenal Bulbs of All Endoscopy Patients Increases Detection of Abnormalities but has a Minimal Effect on Diagnosis of Celiac Disease

Abstract

Background & aims: In patients with positive results from serologic tests for celiac disease, analysis of tissues samples from the duodenal bulb, in addition to those from other parts of the small bowel, might increase the diagnostic yield. However, biopsies are not routinely collected from the duodenal bulb because of concerns that villous atrophy detected there could be caused by other disorders (Brunner glands or peptic duodenitis, gastric metaplasia, shorter villi, or lymphoid follicles). We investigated whether analysis of biopsies from duodenal bulbs of all patients undergoing endoscopy (a population with a low probability for celiac disease) increases diagnoses of celiac disease.

Methods: We performed a retrospective analysis of data from 679 patients (63% female; mean age, 50 years) from whom duodenal bulb and small bowel biopsies were collected during endoscopy at 3 Mayo Clinic sites, from January 1, 2011 through December 31, 2011. Records were reviewed for age, sex, pathology findings, serology test results (HLA DQ2 or DQ), indications for biopsy analyses, and adherence to a gluten-free diet. Patients with celiac disease were identified on the basis of increased intraepithelial lymphocytosis, with or without villous atrophy and crypt hyperplasia, and results from serology tests. Findings from duodenal bulbs were compared with diagnoses using the Fisher exact test.

Results: Of all patients undergoing endoscopy, 16 patients (2%) were found to have celiac disease. Analysis of the duodenal bulb biopsies identified 1 patient (0.1%) with celiac disease limited to this region. Of 399 patients whose celiac serology was not known before endoscopic examination, only 2 patients had histologic changes consistent with celiac disease but not limited to duodenal bulb. Abnormal duodenal histology was detected in 265 patients (39%), most commonly in the bulb (n = 241; P < .0001). Of abnormal bulb histologies, chronic peptic duodenitis was most common (observed in 114 patients, 47%). In patients with a normal distal duodenum (n = 576), the duodenal bulb had abnormal histology in 162 (28%).

Conclusions: In a low pretest probability cohort, separate sampling of the duodenal bulb had minimal effect on celiac disease detection. Abnormal histologic findings are more commonly detected in the duodenal bulb; although they do not seem to impair identification of celiac disease, their clinical implications are unclear.

Keywords: Diagnostic Accuracy; Screening; Undiagnosed; Wheat.