Prospective evaluation of ultrasound and colour duplex imaging for the assessment of surgical resectability of pancreatic tumours

Background This study was performed to evaluate colour duplex imaging (CDI) for the assessment of resectability of pancreatic tumours (PTs).Method From October 1998 to December 2001, 182 patients consecutively having surgery for a PT were enrolled in this prospective study. Extension of the tumour to large blood vessels and retroperitoneum, the detection of liver metastases, enlarged lymph nodes and peritoneal carcinomatosis were defined as criteria for assessment. The patients were grouped into three classes of resectability: US-TU 1 = resectable/potentially curable, US-TU 2 = questionably resectable/curable, and US-TU 3 = non-resectable/not curable. CDI was performed by five different examiners. Results were compared with intra-operative findings.Results Using CDI, we classified 85 (46.7%) tumours as resectable, 64 (35.2%) as non-resectable, and 33 (18.1%) as questionably resectable. Overall, 46.2% (n=84) were found to be resectable, and 53.8% (n=98) to be non-resectable, intra-operatively. A correlation between CDI classification and intra-operative findings was found in 138 of 149 cases (92.6%) (sensitivity 88.4%, specificity 96.3%). With regard to the complete oncological status (local extension, metastases, lymph-node staging and peritoneal carcinomatosis), a sensitivity of 77.2% and specificity of 95.7% were found. Non-correlated findings were likely attributed to missing small liver metastases, peritoneal carcinomatosis without ascites, and on difficulties in the assessment of enlarged lymph nodes concerning tumour infiltration.Conclusion The use of CDI in evaluation of PTs may provide valuable pre-operative assessment of surgical resectability and may be performed in the clinical setting.