Chinese Medical Journal

Ag THE CHINESE MEDICAL JOURNAL

of the hepatic and splenic arteries was done. The patient vomited a small amount of blood following the operation and finally died of hepatic coma seven weeks later. The second case was a man of 44, who underwent splenectomy and hepatic artery ligation, and apart from occasional spitting of “bloody sputum” he has been enjoying fairly good general health and has resumed full duty. Now, over two years after the operation, there is striking improvement in his general condition. Examination shows no evidence of ascites and the liver function tests are all within normal limits. While opinions on the value of this operation are divergent and our limited experience can contribute little of value to the controversy, we feel that in relatively late cases of liver cirrhosis with a history of hematemesis, presence of ascites and marked impairment of liver function, where a shunt procedure is infeasible, this operation may lessen the occurrence of bleeding from the esophageal varices and also eliminate ascites. The point of ligation should be proximal to where it gives out the gastroduodenal branch so that some degree of arterial supply to the liver may be maintained.

Splenorenal anastomosis. The results following this type of operation have been reported by one of us(3). In a case in which the splenic vein was anastomosed to a large retroperitoneal collateral, profuse bleeding from the esophageal varices occurred about six months after the operation. The patient was readmitted and trans-esophageal transfixion of the esophageal veins was done. For over eight months since the second operation the patient has been well apart from an attack of.a small amount of bleeding. This case shows that the retroperitoneal collaterals may communicate with the esophageal veins, and therefore anastomosis may cause further engorgement of these veins, and should not be attempted.

Splenectomy. Splenectomy was done in 52 cases. Since the majority of our cases fall into this group, we shall deal with the operative results under this heading.

a. Mortality. There were two operative deaths within twenty-four hours after operation among the 52 cases of splenectomy. One patient died of bleeding because of non-coagulation of the blood; necropsy failed to show any apparent cause. The exact cause of the other death was not clear. In another case of splenectomy the postoperative course was smooth but the patient suddenly developed massive hematemesis on the twenty-fourth postoperative day and immediately died. Including the fourth death after ligation of the splenic and hepatic arteries, the hospital mortality of our 63 cases (2 cases of mediastinotomy excluded) was 6.3 per cent and the operative mortality was 3.2 per cent.

b. Change in portal pressure. In some of these cases, the portal pressure was determined through a gastroepiploic branch after ligation of the splenic artery, splenectomy and completion of the splenorenal shunt. A reduction of the portal pressure was observed in most of the