Article Text


PMO-111 Should all patients with locally advanced pancreatic cancer be offered intraoperative assessment?
  1. V Rao1,
  2. K Chantladze1,
  3. S Kugathasan1,
  4. O Byass2,
  5. A Razack2,
  6. K Wedgwood1
  1. 1Department of Surgery, Castle Hill Hospital, Cottingham, UK
  2. 2Castle Hill Hospital, Cottingham, UK


Introduction Surgery is the only potentially curative option in patients with pancreatic cancer. Hence it is extremely important that the diagnostic tests used to ascertain resectability is very reliable before this curative option is denied to this unfortunate group of patients. CT and Endoscopic Ultrasound (EUS) which are commonly used as part of pre operative staging was compared with intraoperative findings to assess diagnostic reliability in determining resectability in patients with pancreatic cancer.

Methods All patients diagnosed with pancreatic adenocarcinoma over a 2-year period who underwent trial dissection or resection after routine staging with CT and EUS were included in the study. CT and EUS images were retrospectively reviewed by two radiologists in a double blinded manner and the findings were compared with operative findings and final histology in those patients who underwent radical resection. Sensitivity, Specificity, Positive Predictive value (PPV), Negative predictive value and Accuracy were determined for assessing major vessel involvement which in most cases preclude radical resection.

Results 23 patients (M:F=13:10; mean age=68; range=56–78) underwent trial dissection or radical resection over a 2-year period. 13 were inoperable (nine inoperable due to locally advanced tumour, 1 inoperable due to liver mets, three both locally advanced and liver mets) and 10 underwent radical resection (three resected with cuff of portal vein (all R1), seven resected with six of them R1). Predictably EUS had superior sensitivity and accuracy over CT for both major vessel involvement (88% vs 53% & 87% vs 65%) and nodal involvement (43% vs 10% & 56% vs 30%). However CT was superior to EUS in excluding major vessel involvement (specificity = 100% vs 86%) and comparable to EUS in ruling out nodal disease (specificity = 100%). Importantly, three patients declared as having major vessel involvement by either of the modality underwent radical resection, two of them with PV resection. One patient who was staged as resectable with no vascular involvement was found to have major vessel involvement and underwent resection (R1).

Conclusion Though CT and EUS have important role in staging of patients with pancreatic cancer, a significant minority of patients will still be amenable to radical surgery and should be offered trial dissection with a view to radical surgery as surgery is the only realistic curative therapeutic option.

Competing interests None declared.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.