Portacaval shunting causes liver atrophy, hyperammonemia, and hepatic encephalopathy. A fundamental question is whether the changes, especially those to brain, are permanent. To answer this, it is necessary to have a model whereby a portacaval shunt can be created for a period of time and then the normal pattern of circulation to the liver restored at will. An end-to-side shunt, the most extensively studied model of liver dysfunction, is permanent. However, a side-to-side shunt can be constructed that results in a somewhat different pattern of circulation but with the potential to be reversed. The purpose of the present study was to compare the severity of the metabolic disturbances caused by the two models. Rats with an end-to-side shunt, a side-to-side shunt, or sham operation were prepared and studied after 14-19 days. Both models of shunting caused the same degree of liver atrophy, hyperammonemia, and indistinguishable disturbances in the amino acid content of plasma and brain. Furthermore, both models produced the same degree of cerebral depression as measured by glucose consumption.