Since the beginning of this century, investigators have faced the formidable task of developing a suitable experimental model of liver fibrosis. To date, no animal model reproduces the disease in man. The criteria for an ideal model may obviously vary, depending on the etiologic type of liver fibrosis but should include the following:'.'(1) duplication of morphologic features seen in human disease;(2) gradual and discrete progression of pathologic changes;(3) high reproducibility and low mortality;(4) reversibility and irreversibility of fibrotic changes;(5) development of pathophysiologic sequelae. Despite certain inherent limitations of animal models,'.'innovative experimental models of liver fibrosis have been developed that have made important contributions to the understanding of cellular and molecular mechanisms underlying excessive accumulation of extracellular matrix in the liver.

The purpose of this article is to review these experimental models, with the major emphasis on those that were described in the American Association for the Study of Liver Disease Spring Conference on" Connective Tissue Biology and the Liver" held in Asilomar, California, in April, 1989. A portion of this article is devoted to reviewing the Tsukamoto-French rat model of alcoholic liver fibrosis. This model was first described in 1985 and has been successfully used in two independent laboratories for studies of pathogenic mechanisms of alcoholic liver necrosis and fibrosis. Animal models of hepatic fibrosis can be categorized by the etiologic factor, as shown in Table 1:(1) toxic;(2) nutritional;(3) immunologic;(4) biliary;(5) alcoholic; and (6) genetic. This classification is not meant to be all inclusive, and some categories overlap. For instance, the models of alcoholic liver fibrosis can be considered toxic, nutritional, or immunologic, depending on one's view of the underlying mechanism.