Effect of portacaval shunts on lymph flow in the thoracic duct: Experiments with normal dogs and dogs with cirrhosis and ascites

Abstract

It has been shown that hepatic venous outflow obstruction is the primary hemodynamic lesion in cirrhosis. Furthermore, considerable evidence indicates that both the increased production of lymph and the formation of ascites in experimental and human cirrhosis are caused by the elevation of intrahepatic pressure which results from the outflow obstruction. Accordingly, the capacity of end to side and side to side portacaval shunts to reduce intrahepatic pressure was evaluated by measuring the effects of these procedures on pressure and lymph flow in the thoracic duct of fifty dogs. Portacaval shunts of both types were made in normal animals and in dogs with congestive cirrhosis, portal hypertension, and massive ascites produced by ligation of the hepatic veins. Normal dogs and dogs with cirrhosis but without shunts served as controls. Hepatic vein ligation resulted in massive ascites, which averaged 4.0 L., portal hypertension which averaged 208 mm. of saline solution, a mean thirteenfold increase in thoracic duct lymph flow, and a mean elevation of thoracic duct pressure which was almost three times the normal level. The end to side portacaval shunt reduced the high rate of lymph production and the elevated pressure significantly, but a fivefold increase in lymph flow and an approximately twofold elevation of thoracic duct pressure persisted. The side to side portacaval shunt lowered thoracic duct pressure to normal in every dog, and either abolished the excessive lymph formation or markedly reduced the rate of lymph production. The results of this study indicate that the side to side portacaval shunt is more effective than the end to side anastomosis in overcoming intrahepatic hypertension and decompressing the obstructed hepatic vascular bed. These findings represent an important consideration in the selection of surgical therapy for intractable cirrhotic ascites, and may have some bearing on the effects of portacaval shunts on nutrition of the hepatic parenchyma.