Background: The value of mesentericoportal vein resection (VR) associated with pancreaticoduodenectomy (PD) for pancreatic-head adenocarcinoma still remains controversial.
Methods: From 1989 to 2003, 45 consecutive patients with pancreatic-head adenocarcinoma underwent PD with mesentericoportal VR due to intraoperative macroscopic involvement of the superior mesenteric or portal vein (VR+ group). They were compared with 88 patients who underwent PD for adenocarcinoma without VR over the same time period (VR- group) and matched for age, American Society of Anesthesiologists (ASA) score, pathological diagnosis, and nodal involvement. Preoperative features, intraoperative parameters, postoperative course, and survival were compared between the VR+ group and VR- group. Factors that may influence survival were determined by univariate and multivariate analyses.
Results: Mortality, morbidity, and mean hospital stay did not differ between the two groups (VR+ 4.4%, 56%, and 22.6 days, respectively; VR- 5.7%, 64%, 24.6 days, respectively). In the group VR+, vein invasion was histologically proven in 29 (64%) patients. Three-year global survival and 3-year disease-free survival did not differ between the two groups: VR+ 22% and 14%, respectively; VR- 25% and 21%, respectively (log-rank: P=0.69 and P=0.39, respectively). Neither VR nor histologically proven vein involvement significantly impacted survival duration. On multivariate analysis, tumor-free margin was the most important prognostic factor.
Conclusions: Vein resection during PD can be performed safely. Patients with adenocarcinoma who require VR during PD have a survival not different from that of patients who undergo standard PD. Macroscopic isolated mesentericoportal vein involvement is not a contraindication for PD in patients with adenocarcinoma provided disease-free margins can be obtained.