We present two patients who underwent a portal stent placement for bleeding jejunal varices of the afferent loop caused by extrahepatic portal venous stenosis. Case 1 involved a 66-year-old woman who developed bleeding jejunal varices due to extrahepatic portal venous stenosis 1 year after a pancreaticoduodenectomy with intraoperative radiation therapy. Percutaneous transhepatic balloon dilatation and stent placement were performed. Since undergoing the procedure, no bleeding has occurred. Case 2 concerned a 44-year-old woman who had a rupture and bleeding of jejunal varices 16 years after a choledocojejunostomy. Stenosis was observed from the right and left branches of the portal vein to its intrahepatic branches. Both balloon dilatation and stent placement were attempted. However, the stent could not be fully inserted into the intrahepatic portal vein. Portal stent placement is less invasive and radical, and therefore should be attempted for the treatment of extrahepatic portal venous stenosis. However, there are limits to its application if the stenosis extends to the intrahepatic branches of the portal vein.

Cellular adhesion of sialyl-Lewis-a(SLea)-positive pancreas carcinoma to endothelial cells (EC) is augmented by activation of EC via up-regulated E-selectin expression on EC. Co-cultivation of pancreas-carcinoma cells, PCI-24, with human umbilical-vein endothelial cells (HUVEC) for 5 hr at the PCI-to-HUVEC ratio of 1:10 induced E-selectin expression on the endothelial-cell surface, augmenting SLea-positive pancreas-carcinoma cell attachment with HUVEC. Culture supernatants of 6 tested pancreas-carcinoma cell lines contained soluble, E-selectin-inducing factor(s). The E-selectin-inducing effect by the supernatants was blocked by the protein-kinase-C inhibitor, H7. Antibodies against SLea and E-selectin but not SLex or ICAM-1 blocked the increased pancreas-carcinoma-to-endothelial attachment. Paraformaldehyde(PFA)-fixed PCI-24 cells also induced E-selectin on vascular endothelial cells upon direct contact with endothelial cells, indicating the presence of a membrane-bound form. The 6 pancreas-carcinoma lines all produced IL-1 alpha mRNA and protein but not IL-1 beta or TNF-alpha protein and/or mRNA. Absorption of IL-1 alpha from the supernatants by IL-1 alpha-specific antibody almost completely abolished E-selectin-inducing activity. Anti-IL-1 alpha antibody also abolished the E-selectin-inducing activity of PFA-fixed PCI. IL-1 alpha production by PCI cells was up-regulated by TNF-alpha. These observations suggest that substance(s) produced by pancreas-carcinoma cells, in this case, IL-1 alpha, may contribute to pancreas-carcinoma-cell colonization in non-inflamed, distant locations in vivo, by activating vascular endothelial cells.

Radical esophagectomy is a highly invasive operation for esophageal cancer, and improved techniques are being sought to reduce the invasiveness of this procedure. We devised a method in which an assistant inserts their left hand into the thoracic cavity, and the operator inserts their left hand into the abdominal cavity through a small incision in the upper quadrant during an endoscopic procedure. Between 1996 and 1999, we performed endoscopic esophagectomy on 18 patients. The median number of mediastinal lymph nodes removed by thoracoscopic surgery was 20.1 +/- 9.4 and the median number of abdominal lymph nodes removed by laparoscopic surgery was 11.1 +/- 5.6. The number of nodes dissected by endoscopic surgery did not differ significantly from the number of nodes dissected by conventional thoracotomy with laparotomy. Our experience shows that endoscopic esophagectomy with reconstruction of the esophagus assisted by inserting the hand into the thoracic and abdominal cavity, for safety and certainty, is an effective technique that is much less invasive than radical esophagectomy performed by conventional thoracotomy with laparotomy.

We report the successful treatment of Budd-Chiari syndrome (BCS) due to an obstruction of the hepatic veins. A 19-year-old man developed bleeding esophageal and gastric varices, and a large amount of ascites was thus caused by portal hypertension. BCS complicated by an obstruction of the major hepatic veins was diagnosed after examinations. We performed percutaneous transluminal angioplasty (PTA). This resulted in a great improvement of BCS, a marked decrease in the pressure of the hepatic veins, and the disappearance of ascites. Restenosis occurred at 1 year and 5 years after the initial angioplasty, for which PTA was repeatedly performed. Nine years after the initial treatment, no stenosis has been observed and the patient has shown a favorable outcome.

No abstract