Experience with laparoscopic ultrasonography for defining tumour resectability in carcinoma of the pancreatic head and periampullary region

Background: Computed tomography (CT) is currently the most widely available staging investigation for pancreatic tumours. However, the accuracy of CT for determining tumour resectability is variable and can be poor. Laparoscopic ultrasonography (LUS) is potentially a more accurate method for disease staging. The authors' experience of LUS for staging carcinoma of the pancreatic head and periampullary region is described. Methods: Fifty-one patients with potentially resectable pancreatic tumours defined at CT underwent further investigation with LUS. Twenty-seven patients subsequently had an open laparotomy. The evaluations of tumour resectability at CT and LUS were compared with the operative findings. Results: At LUS, 24 patients were considered to have resectable tumours, 21 non-resectable tumours and six patients were shown to have no pancreatic tumour mass. Twenty-two patients deemed to have a resectable tumour at LUS underwent surgery. Twenty patients were confirmed to have resectable disease and two patients had non-resectable disease. A further five patients underwent surgery. In all five the ultrasonographic diagnosis was confirmed at surgery (four patients with non-resectable disease and no pancreatic tumour in one patient). LUS prevented unnecessary extensive surgery in 53 per cent of patients. For the 22 patients who underwent surgery for potentially resectable disease, the positive predictive value of LUS for defining tumour resectability was 91 per cent. Conclusion: LUS is an accurate additional investigation for defining tumour resectability and directing management in patients with potentially resectable carcinoma of the pancreatic head or periampullary region.