Successful venous repair and reconstruction for oncologic resections

Objective We report our institutional experience of various venous reconstruction methods during oncologic resections, especially examining the patency of venous reconstructions and the conduits used. Methods All patients undergoing venous repair or reconstruction for oncologic resections between 2008 and 2014 were identified by a retrospective search of a prospectively maintained database at a single university hospital. Extent and manner of venous reconstruction and conduit or patch material were recorded. Need for intraoperative venovenous bypass or cardiopulmonary bypass was also recorded. Whereas no prescribed follow-up protocol has been instituted, patency and survival data as available were analyzed. Results During the study period, 127 patients were identified. Five patients had primary ligations, without limb loss. Of the remaining 122 patients, 77 (63%) underwent primary repairs, 23 (19%) had patch repair, and 22 (18%) had bypasses. Of these, 27 (22%) were for portal vein reconstruction during a Whipple procedure, 47 (39%) were for caval repair during caval thrombectomy in the setting of renal cell cancer, and 28 (23%) were for caval repair during resection for other abdominal malignant neoplasms. Venovenous bypass was used in 16 repairs and cardiopulmonary bypass in 10. The 1-year patency rates were 100% for primary and patch repairs and 86% for bypass graft reconstructions. Occlusions were suffered only in the prosthetic grafts group. There was no limb loss or significant long-term morbidity in patients with occluded grafts. Rate of infection was 0%, and there was no evidence of an increased infection rate in prosthetic or bioprosthetic conduits or patches. Perioperative mortality was 5.5%. Conclusions Overall, venous reconstruction for oncologic resection can be done safely with very low complication rates and low perioperative mortality. Prosthetic grafts can be used for most reconstructions with no infections and good patency rates.