Mesenterico left portal bypass for variceal bleeding owing to extrahepatic portal hypertension caused by portal vein thrombosis

Abstract

Background/purpose: Portosystemic shunt operations are indicated in patients with extrahepatic portal hypertension owing to portal vein thrombosis (EPH-PVT) suffering from recurrent variceal bleeding despite endoscopic sclerotherapy. Mesenterico left portal bypass procedure (MLPB) is an alternative procedure to the portosystemic shunt operations in patients with EPH-PVT. MLPB operation reestablishes hepatopetal portal blood flow. We herein present our experience with MLPB in children with EPH-PVT.

Methods: Six patients were treated for EPH-PVT with recurrent bleeding despite endoscopic sclerotherapy (2 boys and 4 girls) in our unit. All patients were evaluated preoperatively with complete blood count, portal duplex system Doppler ultrasonography, magnetic resonance angiography, and upper gastrointestinal (GI) endoscopy. MLPB operation was performed as described by de Ville de Goyet. During the postoperative period, patients were evaluated with complete blood count, portal duplex system Doppler ultrasonography, upper GI endoscopy, and magnetic resonance angiography.

Results: Six patients were assessed to be candidates for MLPB procedure and were operated to perform the MLPB procedure. Left portal veins were found to be patent during the operation in 4 patients, and the MLPB procedure was performed. Internal jugular vein was used in 3 patients and enlarged inferior mesenteric vein in 1 patient. Left portal veins of the remaining 2 patients were found to be obliterated; therefore, mesocaval shunt was performed. The postoperative course of the patients was uneventful except for 1 patient. During the following period, the leukocyte and the platelet counts were significantly increased in 3 of the 4 patients after the MLPB procedure. Upper GI bleeding occurred in the early postoperative period in 1 patient with MLPB procedure because of prepyloric ulcer that was successfully treated by endoscopic sclerotherapy. Internal jugular vein graft thrombosis was detected on the 10th postoperative day. This patient underwent a second laparotomy, the distal half of the graft was found to be sclerosed and narrowed that the graft was revised with a synthetic allograft.

Conclusions: Based on a review of the literature, the MLPB functions well in patients with portal hypertension caused by portal vein thrombosis and appears to have a physiologic advance over shunts that decompress but do not return blood directly to the liver. Because intra-abdominal veins appear to function well as a conduit in this operation, it may be favored by eliminating additional incision and increased risk in such patients.