Duodenal Biopsy Audit: Relative Frequency of Diagnoses, Key Words on Request Forms Indicating Severe Pathology, and...

Background/Objectives: Understanding the diagnostic landscape is essential prior to developing artificial intelligence (AI)-based diagnostic strategies for automating the diagnosis of duodenal biopsies. This study aims to (1) determine the frequencies of different diagnoses seen in endoscopic duodenal biopsies in a large, tertiary referral centre; (2) identify key words on histopathology request forms that could indicate that a biopsy may contain a serious pathology and should not be diagnosed by an AI system; and (3) investigate the proportion of cases described as showing "intraepithelial lymphocytosis" that might be coeliac disease. Methods: To achieve this, we audited 18 months' worth of duodenal biopsy reports in our centre. Results: A total of 6245 duodenal biopsies were identified, of which 73.76% were normal and at least 8.84% fell within the spectrum of coeliac disease. Additionally, 6.47% were classified as showing non-specific inflammation, 1.86% were adenomas, 0.45% were carcinomas, 0.06% were neuroendocrine tumours, 0.10% were lymphomas, and 0.03% were cases of flat dysplasia, giving a total of 0.64% of dysplastic or malignant diagnoses. Rarer diagnoses included ulceration, Helicobacter pylori infection, giardiasis, lymphangiectasia, transplant rejection, and lymphoma. Furthermore, 227 biopsies (3.63%) showed isolated intraepithelial lymphocytosis, of which 33 cases (14.5%) gave an overall clinicopathological picture of coeliac disease. Conclusions: We present the first long-term audit of all endoscopic duodenal biopsies received by the histopathology department of a tertiary-care facility. The results indicate that a fully automated diagnostic histopathology reporting system able to identify normal duodenal biopsies and biopsies within the spectrum of coeliac disease-associated enteropathy could decrease pathologists' endoscopic duodenal biopsy workload by up to 80%.