Initial failure of angiography to demonstrate a bleeding pancreatic cancer: a case for provocative agents

Mesenteric angiography is commonly employed in the modern-day investigation of gastro-intestinal bleeding if the bleeding sites cannot be identified by endoscopic means. Angiography is optimally sensitive in the presence of active bleeding. However, vasospasm may occasionally account for a negative study shortly after bleeding.

A 70-year-old lady with inoperable carcinoma of the pancreas presented with gastro-intestinal bleeding. Although upper endoscopy visualised active bleeding from the tumour, which had invaded into the duodenum, haemostasis could not be achieved endoscopically. Therefore, mesenteric angiography was arranged.

The initial angiography failed to demonstrate the bleeding site, which only became obvious on a repeat study, when embolisation was performed to achieve haemostasis.

Vasospasm probably accounted for the initial negative study, as the second angiography was able to demonstrate contrast extravasation without the use of any anticoagulant or thrombolytic agent. It is not our routine to give pharmacological agents to provoke bleeding after a negative angiography, but for selected patients this manoeuvre may turn out to be more cost-effective.