Aims Duodenal polyps (DP) are found incidentally during diagnostic upper GI endoscopy.
The proportion of DPs that are duodenal adenomas and have malignant potential is not
clearly known due to their low incidence. We aimed to determine the endoscopic-histological
correlation for DPs.
Methods Retrospective study at a tertiary London-based hospital Trust. Endoscopy software
(Unisoft GI reporting tool) used to identify the last 200 patients to be diagnosed
with a duodenal polyp in reverse chronological order from December 2018.
Results 200 patients had duodenal polyps diagnosed between February 2016 and December 2018
(median age 70 (IQR 59–77), Female 94 (47%)). The size of the polyp was not described
in 88 patients (44%), the median size in the remain 112 patients was 6 mm (IQR 4–10).
13 (6.5%) polyps were > 20 mm.
Polyp morphology was described as sessile in 30 (15%), pedunculated in 11 (5.5%) and
not described in 159 (79.5%). Pit pattern was described as hyperplastic in 6 (3%)
and adenomatous in 20 (10%).
Biopsies of the polyp were taken in 189 patients (94.5%) and polypectomy was performed
in 15 (7.5%). Of those resected, polyps were retrieved in 13 (86.7%). Only 7 of 20
polyps thought to be adenomas at endoscopy were confirmed on histology (35%).
Conclusions 3 out of 4 patients diagnosed with DPs do not have a description of the morphology
or pit pattern in the report and less than half describe the size. Less than 10% of
DPs undergo polypectomy. One third of patients have normal duodenal mucosa on histology.
There is significant variability of practice with regards to management of DPs. Better
endoscopic descriptions are required for DPs which may in turn reduce the number of
unnecessary histological samples being taken. Automated duodenal polyp characterisation
on the endoscopy reporting tool may help in better documentation of DPs.
