Location and extent of cavernous transformation of the portal vein dictates different visceral side revascularization in Meso-Rex bypass

Background: As an emerging standard of care for portal vein cavernous transformation (PVCT), Meso-Rex bypass (MRB) has been complicated and variated. The study aim was to propose a new classification of PVCT to guide MRB operations. Methods: Demographic data, the extent of extrahepatic PVCT, surgical methods for visceral side revascularization, intraoperative blood loss, operating time, changes in visceral venous pressure before and after MRB, postoperative complications and the condition of bypass vessels after MRB were extracted retrospectively from the medical records of 19 patients. Results: The median age of the patients (13 males and 6 females) was 32.5 years, while two patients were underage. Causes of PVCT can be summarized as follows: thrombophilia such as dysfunction of antithrombin III or proteins C; secondary to abdominal surgeries; secondary to abdominal infection or traumatic intestinal obstruction, and unknown causes. Intraoperatively, the median operation time was 9.5 h (7-13 h), and the intraoperative blood loss was 300 mL (100-1,600 mL). Ten cases used autologous blood vessels while 10 used allogeneic blood vessels. The vascular anastomosis was divided into the following types according to the site and approach: Type (T) 1-PV pedicel type, T2-confluence type, T3-major visceral vascular type; and T4-collateral visceral vascular type. Furthermore, the visceral venous pressure before and after MRB dropped significantly from 36 cmH(2)O (28-44) to 24.5 cmH(2)O (15-31) (P < 0.01). Postoperatively, one patient had delayed wound healing, two developed biochemical pancreatic fistulae, one experienced lymphatic leakage, the former caused by heat damage of the pancreatic tissues, the latter by cutting lymphatic vessels in the mesentery or removing the local lymph nodes during the process of separating the superior mesenteric vein, and one was re-operated on for an intervening intestinal fistulae. Postoperative enhanced CT scans revealed a significant improvement in abdominal varix in the patients with patent bypass, and at the 1-year postoperative follow-up, enhanced CT scans of six patients showed that the long axis of the spleen was reduced by >= 2 cm. Conclusions: MRB can effectively reduce visceral venous pressure in patients with PVCT. It is feasible to determine the PVCT type according to the extent of involvement and to choose individualized visceral side revascularization performances.