Document resource
Introduction The national endoscopy database has shown that gastric biopsies are often performed for endoscopically normal mucosa or uncomplicated inflammation. Most gastric biopsies involve four or less samples which are insufficient to accurately diagnose cancer or atrophic gastritis. Similarly, biopsies are often taken of the duodenum where the indication is not suggestive of coeliac disease. This project aimed to assess the utility of gastric and duodenal biopsies taken under low-value conditions.Methods A retrospective quality improvement study was performed using endoscopy and histology reports of low value biopsies taken at endoscopy at a UK hospital, between 2023 and 2025. A low value biopsy was defined as four or less gastric biopsies from endoscopically normal or inflammatory mucosa, and in the duodenum biopsy of normal or inflammatory mucosa in the absence of an indication suggestive of coeliac disease (weight loss, anaemia, diarrhoea and positive tissue transglutaminase).We analysed the indication of oesophagogastroduodenoscopy (OGD), the number of biopsies taken, and histology results.Results We identified 325 patients who had OGD from 2023 to 2025. Primary indications for OGD were dyspepsia (32.9%), dysphagia (30.5%), abdominal pain (12.9%), nausea/vomiting (7.1%), and reflux (6.7%). Others included haematemesis, melaena, weight loss and anaemia (for gastric biopsies) and abnormality on imaging.Of the 269 gastric biopsies, most showed inflammation on histology (198, 73.6%) (table 1). Premalignant findings were identified in 22 (8.2%) cases: gastric intestinal metaplasia in 14 (5.2%) and atrophic gastritis in 8 (3.0%). Dysplasia was found in 1 (0.4%) case with no cancers detected. Notably, 68.2% of premalignant biopsies were sampled from antrum or pre-pylorus only, preventing accurate risk stratification. One third of these patients subsequently required repeat OGD to obtain adequate corpus sampling for surveillance decisions.Of the 70 patients who had duodenal biopsies, 1 patient presenting with reflux was diagnosed with coeliac disease (table 1).Conclusions Despite guidance against performing biopsies for simple gastritis and duodenitis, this practice is still evident in our center, and only a small proportion of these biopsies result in clinically actionable findings with no new cases of malignancy. Many premalignant gastric biopsies were not sampled from the gastric body, therefore, clinicians cannot make accurate decisions on surveillance, which leads to unnecessary repeat OGD. Further investigation into cost analysis of low value biopsies is required.Abstract P31 Table 1Distribution of gastric and duodenal biopsy numbers and histology