The surgical treatment of cirrhosis of the liver

Abstract

Portacaval shunting for the relief of portal hypertension and portal stasis has proved to be an effective, rational, and relatively safe procedure to protect patients from untimely death due to hemorrhage, and to protect the liver from the deleterious effects of recurring hemorrhage and wasting ascites. A characteristic feature of cirrhosis is the architectural distortion which, with progressive disease, inevitably leads to obstruction to the inflow of portal blood. Obstruction to portal blood flow frequently results in the development of esophageal varices which are disposed to rupture, causing hemorrhage. In patients having portal block, portal venous pressure is elevated. The establishment of a portacaval shunt, employing either the portal vein anastomosed directly to the vena cava, or the splenic vein (following splenectomy) anastomosed end to side with the left renal vein, affords a large channel having a low resistance to blood flow, and capable of shifting a large volume of blood rapidly from the high-pressure (portal system) to the low-pressure (caval) system, and thereby effecting a reduction in portal pressure with relief of portal stasis. In 1943, we undertook a study of portacaval shunting. To date the experience has been expanded to embrace 262 shunted patients. The site of portal block was intrahepatic, due to cirrhosis of the liver, in 203 patients (77.4 per cent) and extrahepatic, due to a block in the portal vein outside of the liver, in 59 patients (22.6 per cent). The shunting procedure was accomplished in the first 117 patients having cirrhosis of the liver with a postoperative mortality rate of 23 per cent. The mortality rate in the remaining 86 patients was 15 per cent. The decrease in mortality rate is attributed primarily to improvement in the preoperative preparation of patients. The high incidence of liver failure in association with hemorrhage is discussed with special reference to emergency management. An efficient nasogastric balloon tube is presented with detailed instructions on its use for the emergency arrest of hemorrhage and in the preparation of patients for operation. Patients who show a tendency to immediate recurrence of bleeding after the customary four days of balloon tamponage and tube feeding are classified as refractory, and treatment is continued until their status can be ascertained on the basis of clinical behavior and dependable hepatic function tests. Many, after one week or ten days of tube feeding, are proved to have well-compensated livers, their tendency to hemorrhage being the result primarily of severe portal hypertension. In such patients the definitive operation of portacaval shunting can be carried out forthwith. More difficult patients are those in whom ten days or two weeks of treatment does not relieve impairment of liver function, and in whom the operation must be postponed or abandoned. It is with this group that suture of the esophageal varices may be carried out at a propitious time. Our present policy, however, is to continue with tamponage and tube feeding in those patients who are adapted to it, because after a few more weeks, many will improve to a point that shunting may be proceeded with at a reasonable risk. Early in our experience with portacaval shunting for portal hypertension the complication of ascites diagnosed clinically was encountered in 41.5 per cent of the cases. The presence of ascites may in itself be taken as a sign of decompensation of the liver, and, if unstabilized before operation, materially increases operative risk. Portal pressure was elevated to hypertensive levels in all patients with ascites; however, there was no quantitative relationship between the degree of portal hypertension and the severity of the ascites. An early appreciation of this lent encouragement to a view that portal stasis (the relief of which regularly follows the establishment of a portacaval shunt) and not the change in portal pressure was the responsible factor in the subsidence of ascites in a few early cases of long-standing ascites operated upon. This view is corroborated by the work of Hunt, who has shown that the speed of blood flow in the portal vein is increased from 5 cm. per second (about the average determined speed in patients with cirrhosis of the liver) to 12 cm. per second following a direct anastomosis of the portal vein to vena cava, end to side. The normal speed of flow of the blood in the portal vein was determined to be about 10 cm. per second. The indications for portacaval shunting and the evaluation of operative risk with respect to hepatic function tests are discussed.