Planned versus unplanned portal vein resections during pancreaticoduodenectomy for adenocarcinoma

Abstract

Abstract Background The management of portal vein (PV) involvement by pancreatic adenocarcinoma during pancreaticoduodenectomy (PD) is controversial. The aim of this study was to compare the outcomes of unplanned and planned PV resections as part of PD. Methods An analysis of PD over 11 years was performed. Patients who had undergone PV resection (PV-PD) were identified, and categorized into those who had undergone planned or unplanned resection. Postoperative and oncological outcomes were compared. Results Of 249 patients who underwent PD for pancreatic adenocarcinoma, 66 (26·5 per cent) had PV-PD, including 27 (41 per cent) planned and 39 (59 per cent) unplanned PV resections. Twenty-five of 27 planned PV resections were circumferential PV-PD, whereas 25 of 39 unplanned PV resections were partial PV-PD. Planned PV resections were performed in slightly younger patients (mean(s.d.) 60(9) versus 65(10) years; P = 0·031), and associated with longer operating times (mean(s.d.) 602(131) versus 458(83) min; P < 0·001) and more major complications (26 versus 5 per cent; P = 0·026). Planned PV resections were associated with a lower rate of positive margins (4 versus 44 per cent; P < 0·001) despite being carried out for larger tumours (mean(s.d.) 3·9(1·4) versus 2·9(1·0) cm; P = 0·002). There was no difference in survival between the two groups (P = 0·998). On multivariable analysis, margin status was a significant predictor of survival. Conclusion Although planned PV resections for pancreatic adenocarcinoma were associated with higher rates of postoperative morbidity than unplanned resections, R0 resection rates were better.