RESUMO
BACKGROUND: Portal vein resection (PVR) with pancreatectomy is now accepted practice in cases with involvement by tumour. We present our experience of this procedure with particular emphasis on morbidity and survival. METHODS: A retrospective case-control analysis of a prospectively maintained database between 2004 and 2012 was undertaken. A total of 17 patients had pancreatic resections with PVR for cancer and were compared with 17 patients with identical tumour type and stage who underwent pancreatic resection without PVR next in chronological order. Information obtained included patient demographics, radiological and histological evidence of major vein involvement and post-operative morbidity. Disease- and recurrence-free survival were calculated using Kaplan-Meier curves. RESULTS: Procedures associated with PVR included pancreatico-duodenectomy in 11 and total pancreatectomy in six. Three patients underwent pancreatic resection as a re-operation. Pathological staging showed 2× T2N0, 5× T3N0, 1× T1N1, 2× T2N1 and 7× T3N1 tumours. Seven PVR patients (41%) had post-operative morbidity Clavien 3 and 4, compared with none in no-PVR group, but rates of Clavien 1 and 2 complications were similar. Six PVR patients developed PV thrombosis (35%), all with significant clinical consequences. Comparing the PVR group with the no-PVR group, there was significantly reduced median overall survival in (13.8 versus 43.1 months; P = 0.028) and recurrence-free survival (7.5 months versus 39.7; P = 0.004). CONCLUSIONS: Survival of patients after pancreatectomy with PVR was reduced and morbidity was high compared with no-PVR. Delayed portal vein thrombosis due to recurrence was common. Routine post-operative anticoagulation may be indicated in this group.
