Anticoagulation policy after venous resection with a pancreatectomy: a systematic review

Background Portal vein (PV) resection is used increasingly in pancreatic resections. There is no agreed policy regarding anticoagulation. Methods A systematic review was performed to compare studies with an anticoagulation policy (AC+) to no anticoagulation policy (AC-) after venous resection. Results There were eight AC+ studies (n = 266) and five AC- studies (n = 95). The AC+ studies included aspirin, clopidogrel, heparin or warfarin. Only 50% of patients in the AC+ group received anticoagulation. There were more prosthetic grafts in the AC+ group (30 versus 2, Fisher's exact P < 0.001). The overall morbidity and mortality was similar in both groups. Early PV thrombosis (EPVT) was similar in the AC+ group and the AC- group (7%, versus 3%, Fisher's exact P = 0.270) and was associated with a high mortality (8/20, 40%). When prosthetic grafts were excluded there was no difference in the incidence of EPVT between both groups (1% vs 2%, Fisher's exact test P = 0.621). Conclusion There is significant heterogeneity in the use of anticoagulation after PV resection. Overall morbidity, mortality and EPVT in both groups were similar. EPVT has a high associated mortality. While we have been unable to demonstrate a benefit for anticoagulation, the incidence of EPVT is low in the absence of prosthetic grafts.