Distal splenorenal shunt vs. portal-systemic shunt: current status of a controlled trial.

A prospective controlled comparison of portal-systemic (PSS) and distal splenorenal shunts (DSRS) in cirrhotic patients who had survived hemorrhage from esophagogastric varices was undertaken 5 yr ago at five hospitals by the Boston-New Haven Collaborative Liver Group. The clinical and endoscopic criteria for massive hemorrhage were satisfied in 155 patients. Thirty-four patients were excluded, primarily because of uncontrolled hemorrhage. Thirty-four were rejected because the were poor operative risks and 21 because they did not satisfy criteria. Thirteen patients refused to participate; the remaining 53 were randomized; 29 to receive PSS and 24, DSRS. The two groups were similar in clinical, laboratory, and manometric characteristics. The DSRS group was older and tended to have had more previous hemorrhages. Followup ranged from 1 to 56 months (mean 21). After PSS, which was performed by 10 different surgeons, 6 patients died during the hospital admission (21%) compared to 2 after DSRS (12%). There were 6 late deaths in the PSS group and 4 in the DSRS group. Portal-systemic encephalopathy occurred in 5 of the 23 survivors of PSS (23%), and in 6 of the 19 who survived DSRS (32%. Two patients in the PSS group bled (9%), 1 after thrombosis and 1 after stenosis of the shunt. Three patients in the DSRS group bled (16%) and all had thrombosis of the shunt. PSS was associated with an unexplained but inordinately high operative mortality. Although the DSRS was accomplished with an acceptably low operative mortality, it was associated with frequent portal-systemic encephalopathy, shunt occlusion, and recurrent hemorrhage. Similar incidences of portal-systemic encephalopathy, shunt occlusion, and recurrent hemorrhage were observed in the PSS group. More patients and longer followup are necessary to determine which of these portal decompressive procedures is superior.