Continuous intravenous ammonia infusion as a model for the study of hepatic encephalopathy in rabbits.

To obtain a more precise pathophysiological evaluation of the role of ammonia in acute hepatic encephalopathy, we compared the plasma ammonia concentrations and electroencephalographic recordings (EEGs) of rabbits with surgically induced acute hepatic failure (AHF, n = 10) and normal rabbits administered an infusion of ammonium acetate (NH4-Ac, n = 7) over a 10-hr period. AHF was surgically induced by portocaval shunting followed by hepatic artery ligation 48 hr later. In the infusion group the dose of NH4-Ac, initially 0.78 mmol/kg/hr, was increased every 2 hr by 0.13 mmole/kg/hr during a 10-hour period to simulate the arterial NH3 concentrations observed in AHF. Ammonia levels in rabbits administered the NH4-Ac infusion were identical to those observed in AHF, with the exception of the higher initial value in the AHF group. Moreover, the mean rates of increase in grade of encephalopathy in the two groups were similar, although the EEG grades in the infusion group were significantly less at all time points. In conclusion, this study demonstrates that a more pathophysiological approach to identification of the putative toxins in hepatic encephalopathy is feasible. Some of the EEG abnormalities of acute hepatic encephalopathy in rabbits are presumably due to hyperammonemia; the encephalopathy observed in AHF at zero time is probably caused by the previously constructed portocaval shunt via an undefined, but possibly also ammonia-related, mechanism.