Introduction British Society of Gastroenterology guidelines on the management of gastric polyps1 recommend increased intensity of evaluation compared with traditional practice, including biopsy of ALL polyps, biopsy of intervening gastric mucosa (hyperplastic/adenomatous polyps) and 1 year follow-up of dysplasia; a considerable increase in endoscopic and histological workload. The purpose is to reduce risk, and increase early detection of gastric carcinoma. It would therefore be expected that in clinical practice there would be an association between gastric polyp detection and subsequent carcinoma, and that application of the guidelines would improve patient outcomes. The purpose of this study was to evaluate whether this association exists in a DGH setting where traditional polyp assessment and surveillance was followed.
Methods Our database was searched for gastroscopies where polyps were found over 5 years. These records were cross-referenced against the local cancer database for those 5 years plus 1 year of follow-up. Any cases where polyps were found along with subsequent cancers were audited.
Results Details of 15 489 gastroscopies in 11 938 patients over 5 years (2006–2010) were analysed. 670 patients (756 gastroscopies) were found to have gastric polyps (5.6%—in line with larger studies).2 In 2006–2011, 1328 upper GI cancers were recorded on the local register. Cross reference of these revealed 57 patients with polyps at gastroscopy who had a co-existent record on the local cancer register. Cases were excluded where cancer was diagnosed at the index endoscopy [45 patients with polypoid tumours or co-existent polyps where the guidelines would not have influenced outcome]. Seven of the remainder had non-gastric neoplasia. Of the remaining 5, 3 had a small neuroendocrine tumour on follow-up endoscopy, with no specific treatment. The other two were an 80 year old with a dysplastic polyp, followed up at 3 and 6 months when carcinoma was identified (the patient was not fit for radical treatment); and a patient with a large suspicious polyp—inflammatory on initial biopsies with dysplasia identified following intensive follow-up (leading to resection). Neither would have benefited from application of the guidance.
Conclusion This retrospective analysis reveals no patients where an initial suspicion of simple fundic polyps was followed by a subsequent diagnosis of significant neoplasia, or where follow-up of hyperplastic polyps would prevent progression. On the basis of these results, the upper 95% confidence limit for patient benefit from the proposed algorithm is ≈0.5%. These guidelines should therefore be more thoroughly assessed for cost effectiveness.
Competing interests None declared.
References 1. Goddard AF, Badreldin R, Pritchard DM, et al. Gut 2010;59:1270–6.
2. Carmack SW, Genta RM, Schuler CM, et al. Am J Gastroenterol 2009;104:1524–32.