Surgery for advanced and metastatic pancreatic cancer--current state and perspectives.

Advanced disease, defined as vascular invasion or invasion into adjacent organs, in pancreatic ductal adenocarcinoma still remains a major diagnostic and therapeutic challenge. In most cases, only exploratory laparotomy will ultimately ensure surgical resectibility. A physician is ill-advised to make any decision regarding palliation relying on CT-scan, MRI, ultrasonography or angiography, since vascular invasion is difficult to diagnose because of peritumoral pancreatitis mimicking vascular invasion. Only in the case of complete vascular encasement of the mesenterico-portal axis or celiac trunk is a laparotomy unnecessary. If a T3 lesion is present, the patient will benefit greatly from R0 surgical resection, even if this includes en bloc resections of the transverse colon, or the portal vein, which can be reconstructed without vascular grafting in most cases. In the presence of distant metastases only palliative treatment is useful. If liver metastases are identified pre-operatively, palliation should include endoscopic common bile duct stenting in the presence of icterus, or endoscopic duodenal stenting in the case of percutaneous endoscopic gastrostomy. If metastases are found during exploratory laparotomy, surgical palliation should be considered (bilio-digestive anastomosis or gastro-enterostomy), since these procedures do not lead to a significantly longer hospital stay and are not associated with significant morbidity or mortality. Pain control can be ensured using morphine analogs, CT-guided sympathectomy or thoracoscopic sympathectomy. Currently, there is no answer as to which option offers the best pain control and quality of life. There is also an ongoing debate on the palliative Whipple's procedure, even in the event of single liver metastases, since this procedure is associated with limited mortality (well below 5% in high-volume centers) and ensures excellent pain control. This needs an individual assessment of risk and, furthermore, a detailed discussion with the patient. There are no studies in which resection has been performed as a standard procedure for palliation. This question should be answered in a multicenter randomized trial, otherwise the palliative Whipple's operation should still be considered experimental, since it is not likely to significantly prolong survival.