Know the whole history
- Department of Histopathology, Torbay Hospital, Lawes Bridge, Torquay TQ2 7AA, UK; desmond.farrellnhs.net
As histopathologists, we rely heavily on the clinical information provided with request forms to inform us of the patient’s current complaint and relevant medical history. This varies enormously between clinicians. We also build up a relationship with our clinicians who regularly send biopsy material. This is particularly relevant in gastrointestinal pathology—for example, in assessing the endoscopic appearance of inflammatory bowel disease and the subsequent interpretation of the histological findings. With time, we develop an understanding with the clinicians who we deal with regularly and learn to judge the accuracy of the proposed diagnosis, particularly with the more experienced endoscopists.
A 57 year old woman underwent endoscopy by an experienced gastroenterologist who noted a deep gastric ulcer and infiltrated looking duodenal cap carcinoma. The pathology data base showed that seven months previously she had a right hemicolectomy for a poorly differentiated Dukes’s B adenocarcinoma of the hepatic flexure, which was infiltrating the omentum and involved the peritoneal surface of the specimen. Histological examination of the antral gastric biopsies showed abnormal glands with pronounced nuclear atypia (fig 1); the duodenal biopsies were mildly inflamed and oedematous. The gastric biopsies were considered suspicious of malignancy, particularly in view of the endoscopic appearances, and multiple repeat biopsies were suggested. Repeat endoscopy again showed an ulcer in the duodenal cap, but this time the essential information of radiotherapy for the colonic carcinoma was given. Further enquiries from the treating oncologist indicated that the treatment field included the duodenum and pancreas. At initial inspection the duodenal biopsies had a bizarre appearance with apparent underlying malignancy (fig 2); however, immunohistochemistry showed the underlying tissue to be pancreas with residual islets.
Crowded, atypical glands lined by pleomorphic nuclei, which are suspicious of malignancy. A normal gland is present at the top of the figure for comparison.
Ulcer slough overlying abnormal looking glandular tissue, which turned out to be pancreas in the base of the ulcer.
The pitfall of pancreatic tissue in the base of an ulcer is well known,1 although not often seen. However, our case was further complicated by the effects of radiotherapy. Radiation induced changes in the gastrointestinal tract are well described but histolopathologists need to be aware when radiotherapy has been given. As such, we rely on our clinical colleagues to provide this information to us but, on occasion, even the best clinicians may fail to provide a crucial piece of information, as was the case here, which can then trap the unwary.
