Vol. 13, No. 6, 1991

1257

HEPATOLOGY Elsewhere

The possibility that it may result in intrauterine growth retardation has been suggested (8,9), and that was seen in three patients in the Scantlebury et al. series and one patient reported by Cundy, O’Grady and Williams. The importance of this is unclear, given that intrauterine growth retardation can be seen in up to 10%of ostensibly normal pregnancies. The single case reported in the Scantlebury et al. article of a patient who stopped all medication in pregnancy is fascinating. Pregnancy is a state of altered immunity allowing retention of a foreign tissue mass (the fetus). The low incidence of allograft rejection reported by Scantlebury et al. and the successful withdrawal of immunosuppressive agents, albeit in one recipient only, imply that lower doses of immunosuppressive agents may be effective during pregnancy. It appears from these articles that (a) menstruation returns in concert with restored hepatic function, and (b) pregnancy is possible and relatively safe after liver transplantation. Indeed, liver transplantation can sustain the life of mother and fetus even in the extreme circumstances of decompensated liver failure during pregnancy (10). All in all, these articles contain reassuring news for young women facing the ordeal of liver transplantation.

URSODEOXYCHOLIC ACID AND TAURINE AS THERAPY FOR CHOLESTATIC LIVER DISEASE

Podda M , Ghezzi C, Battezzati PM, Crosignani A, Zuin M, Roda A. Effects of ursodeoxycholic acid and taurine on serum liver enzymes and bile acids in chronic hepatitis. Gastroenterology 1990;98:1044-1050. ABSTRACT

Relatively hydrophobic bile acids have been shown to produce some hepatotoxicity, whereas treatment with a more hydrophilic bile acid, ursodeoxycholic acid, has improved liver function indices in patients with certain chronic liver diseases. Taurineconjugated bile acids are more hydrophilic than glycine-conjugatedbile acids, and thus, taurine administration has also been suggested for the treatment of chronic hepatitis. To determine if taurine and ursodeoxycholic acid are beneficial and if their effects are additive, this double-blind, randomized trial was conducted to compare the effects of ursodeoxycholic acid, taurine, and a combination of the two on indices of liver injury in 24 patients with chronic hepatitis. The subjects were assigned at random to two of the four following treatments: ursodeoxycholic acid (600 mg/d), taurine (1.5 g/d), ursodeoxycholic acid plus taurine (600 mg and 1.5 g/d) or placebo, given in two successive cycles of 2 months each, according to a balanced incomplete block design. KIMBERLY A. BROWN, M.D. As expected, ursodeoxycholic acid became the preMICHAEL R. LUCEY, M.D. dominant bile acid in bile when administered alone or Division of Gastroenterology in combination with taurine, and taurine conjugate University of Michigan Medical Center concentrations increased during taurine adminisAnn Arbor, Michigan 481 09 tration. Ursodeoxycholic acid reduced serum aminotransaminase and y-glutamyl transpeptidase activities significantly, whereas taurine did not. The effects REFERENCES of taurine and ursodeoxycholic acid were not significantly different from those produced by ursodeoxy1. Van Thiel DH. Disorders of the hypothalamic-pituitary-gonadal and thyroidal axes in patients with liver disease. In: Zakim D, cholic acid alone. It is concluded that ursodeoxycholic Boyer TD, eds. Hepatology. 2nd ed. Philadelphia: W.B. Saunders acid, but not taurine, improves biochemical indices of liver injury in patients with chronic hepatitis. CO.,1990:513-529.

2. Van Thiel DH, Gavaler JS, Spero JA, Egler KM, Wright C, Sanghvi

A, Hasiba U, et al. Patterns of hypothalamic-pituitary-gonadal dysfunction in men with liver disease due to differing etiologies. HEPATOLOGY 1981;1:39-46. 3. Van Thiel D, Kumar S, Gavaler J , Tarter R. Effect of liver transplantation on the hypothalamic-pituitary-gonadal axis of chronic alcoholic men with advanced liver disease. Alcohol Clin EXPRe6 1990;14:478-481. 4. Penn I, Makowski E, Harris P. Parenthood following renal and hepatic transplantation. Transplantation 1980;30:397-400. 5. Van Dyke RW. The liver in pregnancy. In: Zakim D, Boyer TD, eds. Hepatology. 2nd ed. Philadelphia: W.B. Saunders Co., 1990:14381458. 6. Stafford RS.Alternative strategies for controlling rising cesarean section rates. JAMA 1990;263:683-687. 7. Alstead EM, Ritchie JK, LennardJones JE, Farthing MJG, Clark ML. Safety of azathioprine in pregnancy in inflammatory bowel disease. Gastroenteroloav 1990:99:443-446. 8. Pickrell M, Sawers R, Mshael J: Pregnancy after renal transplantation: severe intrauterine growth retardation during treatment with cyclosporin A. BMJ 1988;296:825. 9. Seeds JW.Impaired fetal growth: definition and clinical diagnosis. Obstet Gynecol 1984;64:303-310. 10. Fair J, Klein A, Feng T, Merritt W, Burdick J. Intrapartum orthotopic liver transplantation with successful outcome of pregnancy. Transplantation 1990;50:534-535.

COMMENTS

Ursodeoxycholic acid (UDCA) has been reported to improve pruritus and biochemical profiles of patients with cholestatic liver diseases such as PBC. Several hypotheses have been proposed to explain these results, with reports based on what is known about bile acid (BA) metabolism, physiology and the enterohepatic circulation, but the exact mechanism(s) of these beneficial effects is not known. Chenodeoxycholic acid (CDCA) and lithocholic acid appear to be more toxic than more hydrophilic BAS. Hydrophilicity in viuo is achieved by hydroxylation and/or conjugation, including glucuronidation, sulfation and amidation with glycine or taurine. Taurine-conjugated BAS are more hydrophilic than glycine conjugates, but glycine conjugates predominate under normal circumstances. Previous studies have shown that feeding taurine to normal humans causes the reversal of the glycinehaurine conjugation ratio (1). UDCA is the 7p-epimer of CDCA. The 7P-hydroxyl

1258

HEPATOLOGY Elsewhere

group renders this molecule more hydrophilic than other BAS with two hydroxyl groups. It has been proposed that the hydrophilicity of UDCA accounts for its lack of toxicity and improvement in biochemical profile in cholestatic liver disease. To test the hypothesis that hydrophilicity accounts for the beneficial effects of UDCA, Podda and colleagues designed a study to see whether taurine had added efficacy when given with UDCA to persons with chronic hepatitis. Their results suggest that taurine has no benefit alone or when given with UDCA. This suggests that focusing on hydrophilicity alone may be an oversimplification of the effects of UDCA in the liver. What other factors might contribute to the therapeutic efficacy of UDCA in chronic liver disease? First, UDCA given orally inhibits the ileal uptake of other potentially toxic BAS. For patients with ileostomies, UDCA administration was followed by an increased excretion of cholic acid and CDCA, indicating decreased absorption of these BAS. This decreased intestinal absorption is likely responsible for the decrease in cholic acid and CDCA pool sizes and their increased fractional turnover rates (2). Second, canalicular bile flow is driven in part by the secretion of BA anions across the canalicular membranes of hepatocytes that leads to the osmotic flow of water and accompanying solutes. Kitani and Kanai (3) have hypothesized that part of the UDCA-stimulated bile flow could arise not only by the excretion of BA into the bile canaliculus (osmotic choleresis) but also by its presence in the hepatocyte, in some way increasing the movement of water and electrolytes into the canaliculus. In addition, Kitani and Kanai (4) have shown that UDCA af'fects excretion of other BAS by hepatocytes. When taurocholate (TCA) was infused simultaneously with TUDCA in rats, not only was the cholestasis caused by excess TCA prevented but TCA and total BA excretion were higher. Galan et al. (5) have reported that the maximal biliary secretion of bilirubin in the rat also increased with the infusion of low-dose UDCA. Presumably the same holds true for other organic anions. The mechanism that underlies these findings may also account for the clinical improvement seen in patients with PBC and other cholestatic diseases treated with UDCA because excretion of the yet unidentified pruritogen, possibly an organic anion, may be enhanced. Third, UDCA appears to lack the membrane toxicity seen with other BAS and may also decrease the membrane damage caused by more toxic BAS. In rat hepatocytes, TCA and UDCA caused no significant toxicity in concentrations up to 1,500 pmoVL. However, CDCA and deoxycholate (DCA) caused enzyme release, membrane lysis and urea-synthesis inhibition above 500 pmoVL. The monohydroxylated BA, taurolithocholate, caused membrane damage with enzyme loss at low concentrations (30 to 100 pmoVL). Amidation did not have a significant effect, with the exception of DCA conjugates that produced enzyme leakage more rapidly. Studies using artificial lipid membranes produced similar alterations (6).

HEPATOLOGY

Leuschner and Kurtz (7) found UDCA was less toxic to cell membranes than CDCA in rats at doses of 20 to 1,000 mg/kg/day. Severe alterations of outer liver cell membranes were seen at doses of CDCA between 20 and 250 mg/kg/day and damage to organelle membranes was seen at doses of 500 to 1,000 mg/kg/day. UDCA had no such effects. Electron spin resonance (which reflects membrane polarity and therefore fluidity) of two types of cell membranes showed a toxic influence of TCDCA less than GCDCA less than CDCA and of DCA and its conjugates. UDCA when added simultaneously with the toxic bile salts prevented changes in fluidity. These data suggest that UDCA appears nontoxic to liver cell membranes and, moreover, may prevent CDCA toxicity. Using a different model, Miyazaki, Nakayama and Koga (8)found similar results. They assessed the direct effect of CDCA or UDCA and their glycine and taurine conjugates on isolated human hepatocytes. CDCA reduced the number of microvilli and disrupted cell membranes. A pronounced release of AST was observed. In contrast, UDCA produced only slight structural changes and enzyme release. Conjugation of each BA had only moderating effects. Finally, UDCA may affect major histocompatibility complex (MHC) antigen expression in the liver. The expression of MHC class I molecules in the liver is normally restricted to sinusoidal endothelial cells and biliary epithelial cells, whereas that of class I1 antigens is restricted to sinusoidal endothelial cells and antigenpresenting cells. Aberrant hepatic expression of MHC class I and class I1 molecules has been documented in PBC (9). In PBC, bile ductular epithelial cells exhibit abnormal expression of MHC class I1 molecules that may lead to the induction of autoreactive T cells and thus play a role in the pathogenesis of PBC. MHC class I molecules have been shown to be abnormally expressed in the hepatocyte in PBC, particularly in areas of piecemeal necrosis. Because the presence of MHC class I molecules is thought to be required for efficient lysis of target cells by cytotoxic T-lymphocytes, the presence of MHC class I antigens on hepatocytes might help explain the periportal and lobular cell necrosis observed in advanced stages of PBC. In a nonrandomized trial of UDCA for PBC, Calmus et al. (10) showed that long-term UDCA administration led to clinical, biochemical and histological improvements. They have since shown immunohistochemically that MHC class I antigen expression on hepatocytes was much weaker than in untreated patients but still significantly higher than in controls. The expression of MHC class I1 molecules was similar in both the treated and the untreated groups. These findings suggest that the beneficial effect of UDCA therapy in PBC could be mediated in part by altering immune mechanisms of liver injury. The reduction in hepatocyte MHC class I expression, by suppressing the cytotoxic T-cell target, could block periportal and lobular cell necrosis and thus slow the progression of PBC. The effects of UDCA in patients with chronic liver disease appear to be mediated through several mecha-

Vol. 13, No. 6, 1991

1259

HEPATOLOGY Elsewhere

nisms that may reflect in part its hydrophilicity. The issue, however, would seem more complex, in that increasing hydrophilicity alone (i.e., with taurine conjugation) does not enhance the clinical response to UDCA. Additional answers may come from understanding how UDCA affects BA absorption in the intestine, BA secretion by the hepatocyte, membrane toxicity and MHC-antigen expression. We are only beginning to understand the complex intricacies of UDCA and the liver, and it is hoped this line of research will provide us with more nontoxic, efficacious therapies for cholestatic liver disease.

MARY F. CHAN,M.D. GI Fellow JOHN R. LAKE,M.D. Department of Medicine Liver Center University of California, San Francisco San Francisco, California 94143 REFERENCES 1. Hardison WGM, Grundy SM. Effect of bile acid conjugation pattern on bile acid metabolism in normal humans. Gastroenter010a 1983;84:617-620. 2. Stiehl A, Raedsch R, Rudolph G. Acute effects of ursodeoxycholic and chenodeoxycholic acid on the small intestinal absorption of bile acids. Gastroenterology 1990;98:424-428. 3. Kitani K, Kanai S. Tauroursodeoxycholate prevents taurocholate induced cholestasis. Life Sci 1982;30:515-523. 4. Kitani K, Kanai S. Interactions between different bile salts in the biliary excretion of the rat. Res Commun Chem Pathol Pharmacol

120 recipients of primary liver grafts and 20 more patients undergoing liver retransplantation. The patient survival rate after 2 to 8 months in the primary liver transplantationseriesis 93.3%, with original graft survival of 87.5%. Of the 20 patients in the hepatic retransplant series, 17 (85%) are living. Almost all of the surviving patients have good liver function. In addition 11 hearts, 2 double lungs, and a heart-lung have been transplanted under FK 506, with survival of all 14 patients. With all of the organ systems so far tested, including the kidney (which has been reported elsewhere), rejection usually has beem controlled without additional drugs and with lower average steroid doses than in the past. Nephrotoxicity has been observed, but not to an alarming degree, and there has been a notable absence of hypertension. There is a suggestion that serum cholesterol may be lowered by FK 506, but this is unproved. Although the adverse reactions of FK 506 and the immunosuppressivemechanisms resemble those of cyclosporine, our preliminary observations suggest that FK 506 may have a more advantageous therapeutic index. COMMENTS

Conventional maintenance immunosuppression for liver allograft recipients has included combination therapy with corticosteroids, azathioprine and cyclosporin A. Cyclosporin A, introduced in the early 1980s, rapidly achieved widespread use. Problems with dosage, therapeutic monitoring, drug interactions and toxicity (especially nephrotoxicity) have remained areas of concern. FK 506 is a new immunosuppressive agent with a structure similar to the microlide antibiotic erythro1983;39: 139-152. mycin. The mode of action of FK 506 appears broadly 5. Galan AI, Jimenez R, Nunoz ME, Gonzalez J. Effects of ursodeoxycholate on maximal biliary secretion of bilirubin in the rat. similar to Cyclosporin A. Both drugs block the proliferation of T-lymphocytes at an early stage and prevent Biochem Pharmacol 1990;39:1175-1180. 6 . Scholmerich J, Becher MS, Schmidt K, Schubert R, Kremer B, synthesis of interleukin-2 and other cytokines impliFeldhaus S, Gerok W. Influence of hydroxylation and conjugation cated in lymphocyte differentiation and function. Both of bile salts on their membranedamaging properties: studies on drugs bind to similar proteins, the peptidyl-prolyl isolated hepatocybs and lipid membrane vesicles. HEPATOLOGY isomerases, which are enzymes involved with the folding 1984;4:661-666. 7. Leuschner U, Kurtz W. Pharmacological aspects and therapeutic of their substrates (1, 2). FK 506 was introduced into effects of ursodeoxycholic acid. Dig Dis 1990;8:12-22. clinical practice in February 1989 (3) and even the 8. Miyazaki K, Nakayama F, Koga A. Effect of chenodeoxycholicand popular press ran articles on the drug (4). ursodeoxycholic acids on isolated adult human hepatocytes. Dig The recent publication by Todo and colleagues outDis Sci 1984;12:1123-1130. lining the Pittsburgh experience is therefore welcome. 9. Ballardini G, Gianchi FB, Doniach D, Mirakian R, Pisi E, Bottazzo GF. Aberrant expression of HLA-DR antigens on bile duct Of the 140 patients who received a liver graft (120 epithelium in primary biliary cirrhosis: relevance to pathogenesis. primary), all but 20 were at least 18 yr old. ImmunoLancet 1984;2:1009-1013. suppression was achieved with FK 506 and corticos10. Calmus Y, Gane P, Rouger P, Poupon R. Hepatic expression of teroids. FK 506 (0.075 mg/kg infused over 4 hr) was class I and class I1 major histocompatibility complex molecules in primary biliary cirrhosis: effect of ursodeoxycholic acid. started in the operating theater and continued every 12 HEPATOLQGY 1990;11:12-15. hr until oral dosing was established at a dose of 0.3 mg/kg/day in two doses. Although trough plasma levels FK 506-A PROMISE OF GOOD THINGS TO COME? were used to adjust dosage, other factors such as neurotoxicity, nephrotoxicity and liver dysfunction were Todo S, Fung JJ, Starzl TE, Tzakis A, Demetris AJ, indications for reducing the dose, whereas evidence of Kormos R, Jain A, et al. Liver, kidney, and thoracic liver allograft rejection was an indication for increased organ transplantation under FK 506. Transplantation dosage. Corticosteroids were given and the dose reduced 1990;212:295-307. during the second to sixth postoperative week. Survival results were impressive: Of 120 patients given primary ABSTRACT grafts, 112 (93.3%) were alive 2 to 7.5 mo after The new immunosuppressive drug FK 506 was used transplantation. Nine were regrafted, including seven from the outset with low doses of prednisone to treat within 60 days. Compared with 400 patients receiving