Clinical implications of combined portal vein resection as a palliative procedure in patients undergoing pancreaticoduodenectomy for pancreatic head carcinoma

Background: The clinical implications of combined portal vein resections are controversial. Methods: One-hundred and forty-nine consecutive patients underwent macroscopically curative pancreatectomies for pancreatic head carcinoma between January 1, 1996 and December 31, 2004. Portal vein resection was performed in 86 patients (58%). Data on surgical mortality, morbidity, perioperative outcome, pathological factors, initial recurrence site, and survival were retrospectively compared between the patients with and without portal vein resection. Results: The incidence of postoperative pancreatic fistula was lower among patients who underwent portal vein resection. The median survival period was 14 months for the portal vein resection group and 35 months for the non-portal vein resection group, respectively. Combined portal vein resection was a significant predictor of poor survival using a multivariate analysis. Portal vein resection was strongly associated with larger tumor size, the degree of retropancreatic tissue invasion, the presence of extrapancreatic nerve plexus invasion, lymph node metastases, and positive cancer infiltration at the surgical margins. Conclusions: Portal vein resection at the time of pancreaticoduodenectomy can be safely performed. However, most of patients requiring portal vein resection do not achieve a potentially curative resection or a favorable survival term. As a result, the aggressive application and the strict selection of portal vein resection might reduce the incidence of positive surgical margins, enabling long-term survival in patients who do not require portal vein resection.