Multivisceral Resections in Pancreatic Cancer: Identification of Risk Factors

There is an assumption that multivisceral resections (MVRs) in patients with a pancreatic malignancy are associated with higher morbidity. The oncologic benefit, however, remains controversial. The aim was to identify risk factors for complications in cases of MVR in patients with pancreatic cancer. Of 1099 patients who underwent major pancreatic resection at our institution between January 1992 and October 2008, a total of 55 were treated with an MVR involving resection of one or more additional organs. This group was compared with 154 patients who had palliative bypass surgery and 303 patients who underwent standard pancreatic head resection. Multivisceral resection patients had an overall higher incidence of major surgical complications (p < 0.001). In-hospital mortality was comparable in all groups. Median survival after MVR was inferior to that after standard resection but was significantly better than that after palliative bypass. Univariate logistic regression analysis identified concomitant colon, kidney, and liver resections and any intraoperative transfusion as predictors of complications; in the multivariate analysis, only kidney resections and any intraoperative transfusion were confirmed predictors. In contrast, T status, kidney resection, resection of four or more organs, any postoperative transfusion, and intensive care unit stay of > 2 days were identified as predictors of survival in the univariate Cox regression analysis; in the multivariate analysis, only the T status was confirmed. Median survival after MVR was 16 months, after palliative bypass 6 months, and after standard resection 18 months (p < 0.001). Multivisceral resections are technically feasible procedures with increased survival when compared to palliative bypass procedures. The incidence of postoperative complications was increased with kidney resection and when intraoperative transfusion was required.