Portacaval Transposition in the Rat: A New Model for Investigation of Portal-Systemic Encephalopathy

The object of this work was to establish an experimental model in the rat to examine the relative roles of portal-systemic diversion and of impaired liver function in the pathogenesis of chronic portal-systemic encephalopathy (PSE). The problem with conventional end-to-side portacaval shunting, both clinically and in the experimental animal, is that in addition to producing total shunting of portal blood into the systemic circulation, the procedure itself leads to hepatocellular atrophy and impaired hepatic function. For this reason the new model of portacaval transposition (PCT) was developed in the rat. In this preparation total portal shunting takes place, but blood flow through the portal tracts of the liver is replaced by systemic blood from the inferior vena cava. Comparison of this model with the conventional portacaval shunt (PCS) should allow separation of the two phenomena of shunting and liver dysfunction. This is of some clinical importance, since it has been proposed that selection of patients or "tailoring" of portal diversion procedures on the basis of haemodynamic characteristics should lead to a lower incidence of chronic post-shunt encephalopathy, which remains the major clinical disadvantage of such operations. It was first established (Chapter III) that PCT is attended by retention of a normal body growth pattern, little reduction in relative liver weight, and maintenance of normal hepatic morphology. This contrasts with PCS, following which there is impaired body growth and specific atrophy of the liver with marked histological changes. Since PCS rats were shown to have reduced food intake, the specific nature of these findings was confirmed by using PCT and control animals which were pair-fed against PCS rats. It was also demonstrated that biochemical parameters of liver function were better conserved in the PCT preparation. It was further confirmed by use of radio-labelled Rose Bengal clearance that liver blood flow was in fact reduced in PCS rats and normal after PCT. Sections prepared from the rapidly perfused brains of these animals were examined microscopically (Chapter IV). In both groups (and in a small number of control animals) it was possible to identify changes in the astroglial cells similar to those described by Alzheimer and characteristic of the brain in PSE in man and in experimental animals. However, these changes were significantly greater in the PCS rats than in the PCT rats, and in one experiment the abnormal cell count in PCT rats did not differ from control values. These results were also validated in a small group of pair-fed animals. This finding confirms the importance of hepatic dysfunction in the pathogenesis of central nervous system damage in these models, and suggests that maintenance of total hepatic blood flow even in the absence of direct perfusion by portal blood might confer protection against the development of PSE. A characteristic pattern of plasma amino acids was seen following PCS, with reduction in the branched-chain amino acids valine, leucine and isoleucine, and elevation of the aromatic amino acids tyrosine and phenylalanine. This pattern, which characterizes chronic liver failure and PSE in man and experimental animals, was not seen after PCT (Chapter V). SUMMARY Similar results were found in the pair-feeding experiment. Plasma insulin and glucagon levels were elevated to a similar degree after PCS and PCT. These results suggest that the hormone elevation seen as a result of portal-systemic diversion cannot entirely explain the amino acid imbalance, which appears to depend largely upon hepatic dysfunction in the PCS rats. Changes in levels of glutamine, glutamate and tryptophan, substances known to be associated with central nervous system neurotransmission and possibly with glial function, were also examined (Chapter V). The differences observed did not achieve statistical significance, although the patterns seen amongst the groups were consistent with small but biologically important changes masked by metabolic compartmentation within the brain. We conclude that the PCT model has proved valuable in separating the effects of portal diversion from those of hepatic dysfunction. The results confirm the vital role of normal liver function in protecting the brain against damage resulting from portal-systemic diversion, and suggest that liver function may be maintained close to normal even in the absence of direct portal perfusion if total liver blood flow is maintained. These findings are in keeping with the observation in man that a substantial compensatory increase in arterial flow following total portal diversion results in a lesser degree of encephalopathy. The establishment of the PCS/PCT model permits a further degree of refinement in future studies of the fundamental biochemical and cellular mechanisms of hepatic coma