904 HEPATOLOGY ELSEWHERE HEPATOLOGY

plausible, but the alternative, that they are derived from smooth muscle, is also possible. Smooth muscle cells are heterogeneous with respect to intermediate filament con- tent. Certain vascular, smooth muscle cells, in particular, contain vimentin but not desmin. In the context of liver, thus, it seems logical to postulate that myofibroblasts originate from smooth muscle cells of the terminal he- patic venule.

Based solely on intermediate filament content, FSC

ANIMAL MODELS FOR ALCOHOLIC LIVER DISEASE

Ainley CC, Senapati A, Brown IMH, Iles CA, Slavin BM, Mitchell WD, Davies DR, Keeling P W N , Thompson RPH. Is alcohol hepatotoxic in the baboon? J Hepatol 1988; 7935-92.

ABSTRACT and myofibroblasts appear to be independent nonpar- enchymal cells within the liver. That both cell types are present in normal liver also argues in favor of a dual population. Furthermore, a clear immunohistochemical distinction between the two is maintained during pro- longed culture, a state which may be analogous to acti- vation in uivo. Does this study, then, refute the notion that FSC transform into myofibroblasts? Viewing the entire population of FSC as a whole, it does not seem that they are predisposed to this particular alteration. Some cells are capable of modifying their intermediate filament profile, however, as in the case of aortic smooth muscle in the present study. Thus, in the presence of an appropriate stimulus, transformation of at least a sub- population of FSC into myofibroblasts cannot be ex- cluded.

At present, there is no specific histological marker to designate myofibroblasts. Classifying them as separate from FSC, in essence by default, as is the case with desmin, is less than optimal. Comparing the two cell types in culture also has its disadvantages, because of phenotypic modulation which may obscure any differ-

The baboon is the only animal in which alcoholic fibrosis and cirrhosis of the liver has been pro- duced with a nutritionally adequate diet. Zinc de- ficiency is associated with alcoholic liver disease and may contribute to liver damage. We have therefore investigated whether zinc supplementa- tion would reduce liver damage in ten baboons receiving ethanol and an adequate diet. Eight re- ceived ethanol at up to 25 g/kg/day (70% of calo- ries) for up to 60 months (four were supplemented with 50 mg zinc/day). All animals gained weight, and blood concentrations of ethanol were 63-342 mg/dl. Changes in liver blood tests were slight. Liver histology only showed fatty change in six animals, severe in two, and minor inflammatory changes but no significant fibrosis or cirrhosis. In one of the animals with severe fatty change there were also degenerative changes in parenchymal cells. There was thus no significant hepatic fibro- sis or cirrhosis in baboons given large amounts of ethanol and an adequate diet for up to 5 years.

COMMENTS ences between them. Until better markers are identified which allow a positive distinction between FSC and myofibroblasts, it is best to avoid a dogmatic posture regarding the two cell types and their individual contri- butions to heDatic fibrosis.

A major difficulty in the investigation of the pathogen- esis of alcoholic liver disease has been the lack of a consistently reliable and inexpensive animal model for

1.

2.

3.

4.

5.

6.

7.

8.

JACQUELYN J. MAHER, M.D. University of California San Francisco General Hospital San Francisco, California 941 10

REFERENCES Friedman SL, Roll FJ, Boyles J, et al. Hepatic lipocytes: the principal collagen-producing cells of normal rat liver. Proc Natl Acad Sci USA 1985; 82:8681-8685. Leo MA, Mark KM, Savolainen E, et al. Isolation and culture of myofibroblasts from rat liver. Proc SOC Exp Biol Med 1985;

Yokoi Y, Namihisa T, Matsuzaki K, et al. Distribution of Ito cells in experimental hepatic fibrosis. Liver 1988,848-52. Nakano M, Lieber CS. Ultrastructure of initial states of perivenular fibrosis in alcohol-fed baboons. Am J Pathol 1982; 106:145-155. Friedman SL, Blaner WS. Activation of cultured lipocytes by Kupffer cell medium (KCM) is accompanied by release of retinol (Abstract). Gastroenterology 1989 (in press). Mak KM, Lieber CS. Lipocytes and transitional cells in alcoholic liver disease: a morphometric study. Hepatology 1988; 81027-1033. Yokoi Y, Namihisa T, Kuroda H, et al. Immunocytochemical detection of desmin in fat-storing cells (Ito cells). Hepatology 1984;

Schurch W, Seemayer TA, Lagace R, et al. The intermediate filament cytoskeleton of myofibroblasts: an immunofluorescence and ultrastructural study. Virchows Arch A 1984; 403:323-336.

180:382-391.

4:709-714.

the disease: A first major obstacle in the development of a model was the natural aversion of animals to alcohol ingestion. This was overcome by DeCarli and Lieber (1) by incorporation of ethanol into a well-balanced liquid diet. A second obstacle was that chronic feeding of this diet to rats resulted in fatty liver, but no histologic evidence of hepatocellular necrosis or fibrosis (2). A successful approach to this problem was the administra- tion of the ethanol-containing diet to baboons resulting in the development of cirrhosis in one-third of them (3). A possible reason for this success is that chronically higher blood ethanol concentrations can be maintained in larger animals due to slower rates of ethanol elimina- tion in larger animals as compared to smaller animals such as rats. The hypothesis is that alcoholic liver disease only occurs when the amount of ethanol ingested mark- edly exceeds the metabolic capacity to dispose of the ethanol, resulting in elevated blood levels for prolonged periods of time. The approach of other investigators was to maintain high blood ethanol levels in rats either by continuous intragastric infusion of ethanol with a diet (4), or addition of 4-methylpyrazole, an inhibitor of al- cohol dehydrogenase, to the ethanol-containing diet (5). In both of the above experiments, there was success in the development of hepatocellular necrosis and inflam-

mation, but no fibrosis was observed. In addition, intra- gastric administration of ethanol is cumbersome and 4- methylpyrazole may itself produce liver injury.

Problems with the use of baboons as models are that they are difficult to handle and very expensive to keep and feed for prolonged periods of' time, and that cirrhosis develops only in a few of them. Furthermore. the baboon model has only been successful in the hands of Lieber and associates. In the study by Ainley et al. chronic feeding of ethanol did not result in any hepatic fibrosis or cirrhosis, even though s ix of the baboons were fed ethanol for periods ranging from 36 to 60 months. The ethanol intake of 25 gm per kg body weight per day was higher than the 4.5 to 8.3 gm per kg body weight per da.v ingested by the baboons in the study of' Rubin and Lieber (6), resulting in blood levels of 63 t o 143 mg per dl, which are similar to the levels of 92 to 184 mg per dl achieved by the baboons of Rubin and Lieber (6). In other studies using monkeys, feeding of ethanol to four Macaca mulata by Rogers et al. ( 7 ) for 52 months and to four Macaca radiata by Mezey et al. (2 ) for 40 to 48 months also did not result in any hepatocellular injury or fibrosis. The monkeys in these studies had been provided with addi- tional choline in their diet, and it was suggested that this had a protective effect. However, suhsequently. Lieber et al. (8) showed that hepatic fibrosis developed in one baboon and cirrhosis in another one of four baboons fed the ethanol-containing diet with choline supplementa- tion.

The reasons for the inability of various investigators to find development of hepatic fibrosis and cirrhosis in baboons or monkeys after chronic ethanol feeding remain unknown. The small number of animals used in the various studies may by chance have resulted in missing the injury, since Popper and Lieher (?I) showed the development of cirrhosis in only six of 18 baboons. How- ever, fibrosis developed in the majority of the baboons. Certainly, the findings by Lieber in the baboons are compatible with the observations that at most, 20% of chronic alcoholic patients develop cirrhosis. These ob- servations imply that factors in addition to chronic ethanol ingestion contribute to the pathogenesis of al- coholic liver disease. Deficiency in a specific nutrient o r a nutritional imbalance may be important,. Of note is that the ethanol-fed baboons in the study by I .ie . b er et al. (8 ) failed to gain weight. whereas in the study of' Ainley et al. the baboons fed ethanol without a zinc supplement gained the same weight as the controls. Unfortunately, the diet used by Ainlev et al. was different from the Lieber-DeCarli diet, which makes the interpre- tation of the different results difficult. Also. it is likely that there exist differences in intrinsic genetic or ac- quired susceptibility to alcohol-induced liver injury. Liver biopsies were performed in only three of 18 ba- boons, according to Figure 1 of the article by Popper and Lieber (3) . One of these baboons developed cirrhosis after 12 months of ethanol feeding and the other two developed steatosis with fibrosis. In the remaining baboons, it is unknown whether or not there was some liver abnor- mality prior to starting the ethanol feeding, which may have increased the susceptibility to alcohol-induced liver

injury. None of the other investigators obtained liver hiopsies before starting the ethanol feeding.

Further studies to investigate the usefulness of the baboon model of alcohol liver disease are needed, but it is recommended that investigators initiate such studies only if they have the facilit.ies and funds to investigate a large number of animals to avoid the chance of missing the development of cirrhosis in a few. Also, initially it makes sense to follow the ethanol-containing diet and protocol employed successfully by Lieber and collabora- tors, rather than utilizing modificat.ions. Once the find- ings are reproduced, then specific variation in nutritional factors can be tested for their beneficial or detrimental effect s .

ESTEBAN MEZEY, M.D. The Johns Hopkins Uniuersity School of Medicine Baltimore, Mapland 21205

REFERENCES I M ' a r l i LM, l ieher (IS. Fatty liver in the rat after prolonged intake 0 1 ethanol with a nutritionallv adequate new liquid diet. .I S u t r 1967: 9 1 ::136. Mezey E, Potter J . J . Slusser ItJ, et al. (:hanges in hepatic collagen metaholism in rats produced hv chronic ethanol feeding. Lab Invest 1977: :$6:206-214.

I'opper H. Lieher CS. Histogenesis of alcoholic fibrosis and cirrhosis i n the hahoon. .4m d Pathol 1980: 98:695-716. I.indros KO. Stowell I,, Vaananen H. et al. Cininterrupted pro- longed ethanol oxidation as a main pathogenetic factor of alcoholic liver damage: evidence from a new liquid diet animal model. Liver

Tsukamutt~ H. French SW, Henson N. Severe and progressive steatosis and focal necrosis in rat liver induced hy continuous intragastric infusion of ethanol and low fat diet. Hepatology 1985: ;;:224. 'L:3'L. Huhin E. I ieher (3. Fatty liver. alcoholic hepatitis and cirrhosis produced hy alcohol in primates. N Engl ,J Med 1974; 290:128-135. Iiiigerh AE. Fox J E . Murphy ,J('. Ethanol and diet interactions in male Hhesus monkeys. Drug-Nutr Interact 1981: 1:3-14. 1,ieher CS. DeCarli LM. Ruhin E. Sequential production of fatty liver. hepatitis. and cirrhosis in s u b h u m a n primates fed ethanol w i t h adequate diets. Proc Natl Acad Sci I'SA 1975: 72:437-441.

THE INITIAL HEMORRHAGE FROM ESOPHAGEAL VARICES

Ttici North Italian Endoscopic C'lub for the Study and Treatment of Esophageal Varices. Prediction of the first variceal hemorrhage in patients with cirrhosis of the liver and esophageal varices. N Engl ,J Med 1988; 319:983-989.

ABSTRACT

We conducted a prospective study of 321 pa- tients with cirrhosis of the liver and esophageal varices with no history of bleeding to see whether a comprehensive analysis of their clinical features and of the endoscopic appearances of their varices could help to identify those at highest risk for bleeding. Varices were classified endoscopically as suggested by the Japanese Research Society for