Vascular resection in pancreaticoduodenectomy for periampullary cancers. A 10 year retrospective cohort study

Background Pancreaticoduodenectomy (PD) is the only chance of cure for periampullary cancers. This study aims to evaluate survival and complication rates for PD with additional vascular resection performed for local vascular involvement and compare to standard PD. Materials and methods A retrospective cohort analysis of a departmental hepato-pancreatobiliary database from 2004 to 2014 was performed. All patients (n = 92) who underwent PD without vascular resection (n = 72), with venous resection (n = 16), with both arterial and venous resection (n = 4) were included in the study. Patients who received palliative double bypass (n = 6) were also included for survival analysis. Survival and post-operative complications were assessed. Results Median survival for standard PD and PD with venous resection was 21 months and 18 months respectively (P = 0.588). Patients who received PD with venous and arterial resection had a median survival of 7 months, significantly less than standard PD (P = 0.044). Median survival in the palliative bypass group was 4 months, comparable to PD with venous and arterial resection (P = 0.191). There was a significant survival advantage in patients who received an R0 resection (median survival 24 months) compared to those who received an R1 resection (median survival 18 months) (P < 0.02). Patients with a lymph node ratio <0.2 had a median survival of 25 months, which was significantly higher than that of patients who had a lymph node ratio ≥0.2 (9 months) (P < 0.005). Conclusion PD with venous resection has similar survival to standard PD with no increased risk of procedure specific post-operative complications. On the other hand, PD with venous resection and additional arterial resection has no survival benefit and may be a step too far in our experience.