THE CONSTRUCTION of large shunts between the portal and caval systems of veins constitutes a logical method of treating patients with bleeding esophageal varices associated with portal hypertension. Both splenorenal and portacaval anastomoses have been used with success for this condition, some surgeons preferring the one and some the other type of operation. However, the procedure, which is most physiological as far as liver function is concerned, remains open to some question.
Whipple and his co-workers1 have shown that dogs with complete shunts of the portal blood away from the liver (Eck fistula) fail to regenerate protein and hemoglobin in a normal manner, and Mann and others2 demonstrated years ago that the livers of Eck fistula animals fail to regenerate after partial hepatectomy. On the basis of this evidence, as well as on theoretical grounds, it would seem wise to recommend the construction of partial rather than complete