- Email: [email protected]
A randomized trial of antioxidant therapy alone or with corticosteroids in acute alcoholic hepatitis
Journal of Hepatology 47 (2007) 277–283 www.elsevier.com/locate/jhep
A randomized trial of antioxidant therapy alone or with corticosteroids in acute alcoholic hepatitisq Stephen Stewart1, Martin Prince1, Margaret Bassendine1,2, Mark Hudson2, Oliver James1,2, David Jones1,2, Chris Record2, Christopher P. Day1,2,* 1
Liver Research Group, Institute of Cellular Medicine, Medical School, Newcastle University, Newcastle upon Tyne NE2 4HH, UK 2 The Liver Unit, Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK
Background/Aims: Oxidative stress is putatively involved in the pathogenesis of alcohol-induced liver injury. This trial was devised to determine whether antioxidant therapy, alone or as an adjunct to corticosteroids, improved survival in patients with acute alcoholic hepatitis. Methods: Patients with a severe alcoholic hepatitis were stratified by sex and steroid use, and then randomized. The active group received N-acetylcysteine for one week, and vitamins A–E, biotin, selenium, zinc, manganese, copper, magnesium, folic acid and Coenzyme Q daily for 6 months. The trial was double blinded and placebo controlled. The primary end-point was mortality within 6 months. Results: Thirty-six (20 male, 16 female; mean discriminant function (DF) 86.6) received active drug, and 34 (18 male, 16 female; mean DF 76.4) received placebo. 180-day survival was not significantly different between patients receiving drug and placebo (52.8% vs. 55.8%, p = 0.699). This was not affected by stratification for steroid use or sex. The only predictors of survival in multivariate analysis were initial bilirubin (p = 0.017), white cell count (p = 0.016) and age (p = 0.037). Treatment allocation did not affect survival in multivariate analysis (p = 0.830). Conclusions: Antioxidant therapy, alone or in combination with corticosteroids, does not improve 6-month survival in severe alcoholic hepatitis. 2007 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Alcoholic hepatitis; Clinical trial; Antioxidants
1. Introduction Alcoholic hepatitis is a disease with a wide clinical spectrum, ranging from sub-clinical forms to a severe acute disease. The mild forms have a good short-term prognosis, but the in-hospital mortality rate of the severe form is around 40% [1]. Features that identify Received 24 January 2007; received in revised form 22 February 2007; accepted 3 March 2007; available online 4 May 2007 q The authors who have taken part in this study declared that they have no relationship with the manufacturers of the drugs involved either in the past or present and did not receive funding from the manufacturers to carry out their research. * Corresponding author. Tel.: +44 191 222 7043; fax: +44 191 222 0723. E-mail address: [email protected] (C.P. Day).
patients with a high mortality include the presence of encephalopathy, a high serum bilirubin and a prolonged prothrombin time. These latter two variables have been combined to form the discriminant function (DF) [2], which has been shown to prospectively identify patients with a 40–50% risk of dying within 2 months [3]. Conventional management consists of abstinence from alcohol, the correction of dietary deficiencies and general support. Of the specific therapeutic agents evaluated prior to trial commencement, only glucocorticoids have been shown to have any benefit on survival. A recent analysis of individual patient data from the last three randomized controlled trials (RCTs) showed a higher 28 day survival (84.6 ± 3.4% vs. 65.1 ± 4.8%, p = 0.001) in corticosteroid compared to placebo treated patients with an initial DF of >32 [4]. Importantly,
0168-8278/$32.00 2007 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jhep.2007.03.027
278
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
active infection and gastrointestinal bleeding are still contraindications to steroid treatment in most centers, and both are relatively common in these patients. Furthermore, mortality is still high whether patients do or do not receive steroids, implying the need for alternative therapeutic options. Interest in the possible value of antioxidant therapy in the treatment of alcoholic hepatitis has arisen as a result of increasing evidence implicating oxidative stress as a key mechanism in alcohol-mediated hepatotoxicity [5]. This evidence falls into five categories. First, products of lipid peroxidation can be detected in the peripheral blood of heavy drinkers [6] and in the livers of patients with ALD [7], and the magnitude of lipid peroxidation correlates with the degree of liver injury [8]. Second, in patients and animal models of ALD, lipid peroxidation is most prominent in the perivenular region where the liver injury is typically most severe [7]. Third, a variety of sources of oxidative stress have been identified in patients with ALD and in animal models of disease. Considerable controversy remains over the most important source of the prooxidant reactive oxygen species (ROS) which are capable of initiating lipid peroxidation and damaging plasma and intracellular membranes [9], but the most likely appear to be microsomal cytochrome P4502E1 (CYP2E1; the only source of hydroxyethyl radicals [10,11]), the mitochondrial electron transport chain [12], inducible nitric oxide synthase [13] and Kupffer cells [14,15]. Fourth, ethanol consumption results in the depletion of endogenous antioxidant capabilities and patients with ALD have evidence of antioxidant deficiencies. Consumption of glutathione (GSH) during oxidative stress and inhibition of two enzymes involved in the synthesis of its precursor, S-adenosylmethionine (SAMe), methionine synthase (MS) and methionine adenosyltransferase (MAT), contribute to the decreased levels of hepatic SAMe and GSH observed in patients with ALD [16]. Heavy drinkers, including those with ALD, are also deficient in the antioxidant trace element selenium [17], which is required for the activity of the antioxidant enzyme GSH peroxidase, the antioxidant vitamins A, C and E [18,19] and Coenzyme Q [20]. This latter compound is present in plasma and mitochondrial matrix membranes and has emerged as one of the most important natural free radical scavengers. It is partly derived from the diet, but is also synthesized in the liver. Finally, in animal models of ALD, dietary [21,22] and genetic manipulations [23,24] that increase oxidative stress increase the severity of liver injury and reducing oxidative stress ameliorates injury [25,26]. Given the nature and extent of this evidence, we were keen to determine whether antioxidant supplementation conferred any benefit to patients with severe alcoholic hepatitis.
2. Methods After local Ethical Committee permission was obtained, consecutive heavy drinkers admitted to the Freeman hospital liver unit with alcoholic hepatitis with hepatic encephalopathy and/or a discriminant function (DF) of P32 were considered for inclusion in the study. Recruitment to the study started in 1998 and follow-up was completed in 2002. Inclusion criteria were; recent heavy (>40 g/day for women; >60 g/day for men) alcohol intake, age between 18 and 65 and either a diagnostic liver biopsy or two of the following: hepatomegaly, leukocytosis >11 · 109 cells/L and ‘‘white out’’ on liver and spleen isotope scanning. Exclusion criteria were; evidence of malignancy, positive HBV or HCV serology, pregnant or lactating women and cirrhotic patients admitted primarily for control of the complications arising from portal hypertension. Steroids were given unless there was evidence of active infection or gastrointestinal bleeding. Patients were then stratified by sex and steroid use, and randomized by an independent pharmacist using a computer-derived randomization schedule. The active group received loading dose N-acetylcysteine (NAC) of 150 mg/kg followed by 100 mg/kg/day for one week, and vitamins A–E, biotin, selenium, zinc, manganese, copper, magnesium, folic acid and Coenzyme Q daily for 6 months. The trial was blinded and control patients received identical placebo. Assessments were carried out weekly for one month and monthly thereafter. The primary end-point was 6-month mortality. We hoped to recruit 80 patients to the study, allowing 10 per group stratifying for sex, steroid use and randomization to antioxidants or placebo. Power calculations were based on 80% power to identify a 40% reduction in mortality with a 2-sided significance of 5%. Continuous variables were compared using the Mann–Whitney U test. Univariate survival analysis was performed using stratified Kaplan–Meier analysis and the log rank test. Multivariate survival analysis was performed using stratified Cox proportional hazards regression. All possible predictive variables were entered into a Cox model accounting for stratification and using a backwards stepwise method, removing the least predictive variable at each step. Results were verified using forwards stepwise regression with a cut off for inclusion of variables of p = 0.10. Multivariate analysis was performed to determine factors associating with survival. All analyses were performed using SPSS 11.0 (SPSS Corp., USA) on a per protocol basis.
3. Results In total, 77 patients were recruited to the study and 70 were followed up until death or censor at 6 months. The shortfall of 10 patients arose due to difficulty in recruiting females that could not receive steroids. The trial had to be terminated when it was no longer reasonable not to treat these patients with pentoxifylline [27]. From the 70 patients, 36 (20 male, 16 female; mean DF 86.6) were randomized to receive active drug, and 34 (18 male, 16 female; mean DF 76.4) received placebo (Fig. 1). At trial entry there were no significant differences between the active and placebo groups with regard to age or the major laboratory indices that might affect mortality. Furthermore, the incidence of hepatic encephalopathy, gastrointestinal bleeding and sepsis was similar between the two groups (see Table 1). Liver histology was available for 43 patients (63%) and cirrhosis was present in 84% of these. For the 38 patients treated with steroids the mean period of treatment was 16 days. Treatment was discontinued prematurely in 14 (37%) patients because of gastrointestinal bleeding (5 patients) or clinical sepsis (9 patients). There were no
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
279
Assessed for eligibility (n=86)
Excluded Not AAH No consent
(n= 9) (n=8) (n= 1)
Stratified by sex and steroid use (n=77)
Males given steroids (n=21)
Males not given steroids (n=18)
Females given steroids (n=22)
Females not given steroids (n=16)
Randomization
AO (n=11) Placebo (n=10)
AO (n=9) Placebo (n=9)
Did not attend follow-up clinic (n=1)
AO (n=11) Placebo (n=11)
Follow-Up
AO (n=8) Placebo (n=8)
Did not attend follow-up clinic (n=4)
Did not attend follow-up clinic (n=2)
Analysis AO (n=11) Placebo (n=9)
AO (n=9) Placebo (n=9)
AO (n=9) Placebo (n=9)
AO (n=7) Placebo (n=7)
Fig. 1. Trial profile. AO, antioxidants.
Table 1 Comparison of active and placebo groups prior to treatment Active
Placebo
p value
Variable (median/interquartile range) Sex (male:female) Age (years) White cell count (·109/L) Prothrombin time (s) Creatinine (lmol/L) Bilirubin (lmol/L) Albumin (g/L) Child–Pugh score Discriminant function
36 (20:16) 44.00 (38.25-50.25) 13.20 (9.30–20.70) 27.00 (22.00–32.50) 117.00 (88.50–168.00) 473.5 (246.00–581.25) 27.00 (25.00–30.75) 12.00 (11.25–13.00) 75.50 (60.50–108.25)
34 (18:16) 44.50 (39.75-50.00) 14.40 (8.47–19.10) 24.50 (22.00–29.00) 110.50 (81.50–175.50) 378.00 (240.25–603.75) 28.50 (25.00–31.00) 12.00 (11.00–13.00) 67.00 (51.00–87.50)
0.984 0.773 0.677 0.179 0.601 0.659 0.637 0.532 0.110
Complication Portosystemic encephalopathy Gastrointestinal bleeding Sepsis
24/36 (66.7%) 4/36 (11.1%) 12/36 (33.3%)
26/34 (76.5%) 4/34 (11.8%) 13/34 (38.2%)
0.433 1.000 0.804
There was no significant differences between the two groups prior to treatment.
280
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
Kaplan Meier Survival
01.00
0.80
0.60
0.40
TREATMENT
0.20
active
00.00
placebo
0
60
120
180
Time (days) No at risk
70
Survival active 100% Placebo 100%
44
39
38
65% 62%
56% 56%
56% 53%
Fig. 2. Kaplan–Meier survival curves. After stratification for sex and steroid use, there is no significant difference (p = 0.710) between the Kaplan–Meier survival curves for patients receiving active drug or placebo.
adverse effects reported attributable to antioxidant or placebo. Patients were followed until death or censor at six months. In total, 7 patients were lost to follow-up; 3 from the active group and 4 from the placebo group. The 6 month survival was not significantly different between patients receiving active drug and placebo (19/36; 52.8% vs. 19/34; 55.8%, p = 0.699) (see Fig. 2). This result was unchanged in intention to treat analysis (with patients censored at point of last contact). As shown in Table 2, treatment allocation did not affect survival in univariate (p = 0.884) or multivariate (0.830) analysis adjusted for stratification by Cox proportional hazards regression. The only variables that associated with mortality in multivariate analysis were the white cell count (p = 0.016), bilirubin (p = 0.017) and age (p = 0.037) at trial entry (see Table 2).
In view of a small non-randomized trial in 12 patients (9 with alcoholic liver disease) with hepatorenal syndrome [28] showing a significant improvement in renal function on N-acetylcysteine (NAC), we conducted a post hoc analysis of the effect of our regime on patients with initial renal impairment. From 17 patients with significant renal impairment (creatinine > 130 lmol/L) that received active therapy, 15 survived the full weeks treatment and could be analyzed. The median reduction in creatinine in these patients was 40l mol/L (29% of baseline, p = 0.041). However this was not significantly different from the reduction observed in patients receiving placebo (n = 11, median reduction in creatinine 43lmol/L, 23% of baseline p = 0.019; p for comparison of active with control = 0.656) (see Fig. 3).
4. Discussion In spite of the evidence for oxidative stress in the pathogenesis of ALD, and the concomitant depletion of antioxidant defenses in heavy drinkers, only two trials have investigated the effect of antioxidant supplementation in acute severe alcoholic hepatitis. In the first study, 56 patients were randomized to receive Vitamin E, selenium and zinc, or placebo [29]. While treated patients had an in-hospital mortality of 6.5% compared to 40% in the control group, the entry criteria and patient details are not clearly discussed. The second trial [30] compared steroids with an antioxidant cocktail (vitamins A, C, E, selenium, allopurinol, desferrioxamine and N-acetylcysteine), and was stopped after an interim assessment found steroids to be of greater benefit. This latter trial included patients with moderate and severe disease (including patients with a DF < 32) and did not determine whether antioxidants conferred any benefit in patients that had a contraindication to steroids, or in addition to steroids. Our aim was to answer these questions using a panel of antioxidants in patients with severe acute alcoholic hepatitis.
Table 2 Cox proportionate regression of factors affecting survival stratified by use of steroids and sex Variable tested
Treatment allocation White cell count (·109/L) Creatinine (lmol/L) Albumin (g/L) Bilirubin (lmol/L) Prothrombin time (s) Age (years) Sepsis Bleed PSE
Univariate analysis
Multivariate analysis
B
SE (B)
p value
B
SE (B)
p value
0.054 0.039 0.000 0.019 0.003 0.064 0.047 0.343 0.507 0.295
0.365 0.021 0.001 0.044 0.001 0.027 0.023 0.413 0.501 0.431
0.884 0.063 0.714 0.663 0.012 0.017 0.041 0.406 0.312 0.494
0.081 0.063 0.001 0.012 0.003 0.052 0.050 0.636 0.494 0.376
0.378 0.026 0.001 0.051 0.001 0.035 0.024 0.518 0.543 0.476
0.830 0.016 0.332 0.816 0.017 0.137 0.037 0.220 0.363 0.429
Pre-treatment white cell count, bilirubin and age were the only factors studied that affected survival.
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
Cr (micromol/L)
a
400
Cr (micromol/L)
*
300
200
100
0
b
*
week 0
week 1
300
** **
200
100
0
week 0
week 1
Fig. 3. The effect of antioxidants including N-acetylcysteine on renal function in alcoholic hepatitis. Among patients with renal impairment at trial entry, a significant improvement in mean serum creatinine was seen after 1 week in patients that received (a) antioxidant therapy (*p = 0.041), or (b) placebo (**p = 0.019). There was no significant difference between the fall in mean creatinine between these 2 groups (p = 0.656).
The 180-day mortality rate of severe alcoholic hepatitis in our cohort of 70 patients was 46%, which is similar to previous reports. Two-thirds of the patients that died did so in the first month, however death events occurred beyond three months in both the active and placebo treated groups. In a recent meta-analysis, age and creatinine on trial entry have been associated with mortality [4]. We found mortality to be associated with initial white cell count, bilirubin and age. There is now good evidence that corticosteroids improve 28 day mortality in patients with an initial DF of >32 [4], so all patients that did not have a contraindication received this treatment. This allowed us to assess whether there was an additional benefit to antioxidants; a relevant clinical question. Antioxidant therapy, however, alone or in combination with corticosteroids, did not improve 6month survival. With good evidence implicating oxidative stress in the pathogenesis of alcoholic liver disease, why do antioxidants given on admission to hospital with severe alcoholic hepatitis not result in improved survival? The first possibility is that the drug doses prescribed were insufficient. Although we did not monitor serum levels, we did provide a broad cocktail of antioxidants, which, in most cases, amounted to several times the daily recommended dose. Furthermore, even after a month of high dose administration, the survival curves for active and placebo treated patients were very similar. This
281
would imply that, at this stage in disease pathogenesis, even high dose, and prolonged antioxidant treatment cannot affect disease progression. The second possibility is that the cocktail we used did not perform an antioxidant function. It may have been possible to assess this function in a dynamic way by monitoring markers of oxidative stress, and there are several recognized techniques for doing this. These include measuring levels of F2-isoprostanes or the aldehyde end-products of lipid peroxidation, malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE). It is less clear which method is the best and, more importantly, whether levels of these markers in the serum or urine reflect levels in the liver. For example it has been demonstrated that the level of vitamin E in the serum does not correlate with hepatic levels [31]. It has also been recognized that components of standard antioxidant therapies may be ineffective or act as pro-oxidants under certain circumstances; namely vitamin E and vitamin C. One trial in healthy volunteers showed no benefit of vitamin E on lipid peroxidation [32]. This can be compared with other studies showing a beneficial effect [33–35]. Concern regarding the role of vitamin E in the development of lipid peroxidation in atherosclerosis is ameliorated by evidence that this effect is suppressed in the presence of other antioxidant agents [36]. While data suggesting that vitamin C has pro-oxidant properties [37] have been questioned [38], it must be accepted that the ideal combination and dose of antioxidant cocktail, and the effect it has on oxidative stress in the liver are not entirely clear and may be difficult to assess. Further study is clearly warranted in this area. The third possibility for the lack of survival benefit is that ongoing oxidative stress is not the primary mechanism underlying mortality in severe alcoholic hepatitis after presentation to hospital. Other mechanisms, that could be potential therapeutic targets, are implicated in disease progression in humans and rodents. These involve endotoxin-mediated cytokine release [39,40], neutrophil recruitment and activation [41] and adaptive immune responses to the products of ethanol metabolism and oxidative stress [42,43]. Furthermore, it is now clear that hepatocyte apoptosis is a pathological feature of alcoholic hepatitis, and the extent to which this occurs correlates with disease severity [44]. These mechanisms of disease will presumably interact during disease development, however, histologically, alcoholic hepatitis is a neutrophil predominant inflammatory condition. Future therapeutic strategies for acute severe alcoholic hepatitis may focus more on reducing recruitment or inhibiting activation of neutrophils, and suppressing Kupffer cell cytokine release and apoptosis in the short term. In summary, high dose antioxidant therapy confers no survival benefit in acute severe alcoholic hepatitis. In addition, it appears that the improvement in renal function seen in the first week after trial entry is due
282
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
to factors other than antioxidants or N-acetylcysteine. This does not exclude a role for this therapy in the delay or prevention of disease in heavy drinkers, or in reducing the risk of disease progression in persistent drinkers that have had an episode of hepatitis. It does, however, highlight that therapeutic strategies aimed at other putative mechanisms of disease should now undergo trials aimed at reducing the mortality rate of this devastating disease. References [1] Hardison WG, Lee FI. Prognosis in acute liver disease of the alcoholic patient. N Engl J Med 1966;275:61–66. [2] Maddrey WC, Boitnott JK, Bedine MS, Weber Jr FL, Mezey E, White Jr RI. Corticosteroid therapy of alcoholic hepatitis. Gastroenterology 1978;75:193–199. [3] Ramond MJ, Poynard T, Rueff B, Mathurin P, Theodore C, Chaput JC, et al. A randomized trial of prednisolone in patients with severe alcoholic hepatitis. N Engl J Med 1992;326:507–512. [4] Mathurin P, Mendenhall CL, Carithers RL, Ramond MJ, Maddrey WC, Garstide P, et al. Corticosteroids improve shortterm survival in patients with severe alcoholic hepatitis (AH): individual data analysis of the last three randomized placebo controlled double blind trials of corticosteroids in severe AH. J Hepatol 2002;36:480–487. [5] Dey A, Cederbaum AI. Alcohol and oxidative liver injury. Hepatology 2006;43:S63–S74. [6] Clot P, Tabone M, Arico S, Albano E. Monitoring oxidative damage in patients with liver cirrhosis and different daily alcohol intake. Gut 1994;35:1637–1643. [7] Paradis V, Kollinger M, Fabre M, Holstege A, Poynard T, Bedossa P. In situ detection of lipid peroxidation by-products in chronic liver diseases. Hepatology 1997;26:135–142. [8] Situnayake RD, Crump BJ, Thurnham DI, Davies JA, Gearty J, Davis M. Lipid peroxidation and hepatic antioxidants in alcoholic liver disease. Gut 1990;31:1311–1317. [9] Halliwell B. Free radicals, antioxidants, and human disease: curiosity, cause, or consequence?. Lancet 1994;344:721–724. [10] Ekstrom G, Ingelman-Sundberg M. Rat liver microsomal NADPH-supported oxidase activity and lipid peroxidation dependent on ethanol-inducible cytochrome P-450 (P-450IIE1). Biochem Pharmacol 1989;38:1313–1319. [11] Albano E, Tomasi A, Goria-Gatti L, Dianzani MU. Spin trapping of free radical species produced during the microsomal metabolism of ethanol. Chem Biol Interact 1988;65:223–234. [12] Slater TF, Sawyer BC, Strauli UD. Changes in liver nucleotide concentrations in experimental liver injury. 2. Acute ethanol poisoning. Biochem J 1964;93:267–270. [13] McKim SE, Gabele E, Isayama F, Lambert JC, Tucker LM, Wheeler MD, et al. Inducible nitric oxide synthase is required in alcohol-induced liver injury: studies with knockout mice. Gastroenterology 2003;125:1834–1844. [14] Niemela O, Parkkila S, Bradford B, Iimuro Y, Pasanen M, Thurman RG. Effect of Kupffer cell inactivation on ethanolinduced protein adducts in the liver. Free Radic Biol Med 2002;33:350–355. [15] Kono H, Rusyn I, Yin M, Gabele E, Yamashina S, Dikalova A, et al. NADPH oxidase-derived free radicals are key oxidants in alcohol-induced liver disease. J Clin Invest 2000;106:867–872. [16] Paredes SR, Kozicki PA, Fukuda H, Rossetti MV, Batlle AM. SAdenosyl-L-methionine: its effect on aminolevulinate dehydratase and glutathione in acute ethanol intoxication. Alcohol 1987;4:81–85.
[17] Dworkin B, Rosenthal WS, Jankowski RH, Gordon GG, Haldea D. Low blood selenium levels in alcoholics with and without advanced liver disease. Correlations with clinical and nutritional status. Dig Dis Sci 1985;30:838–844. [18] Thurnham DI, Davies JA, Crump BJ, Situnayake RD, Davis M. The use of different lipids to express serum tocopherol: lipid ratios for the measurement of vitamin E status. Ann Clin Biochem 1986;23:514–520. [19] Leo MA, Lieber CS. Hepatic vitamin A depletion in alcoholic liver injury. N Engl J Med 1982;307:597–601. [20] Bianchi GP, Fiorella PL, Bargossi AM, Grossi G, Marchesini G. Reduced ubiquinone plasma levels in patients with liver cirrhosis and in chronic alcoholics. Liver 1994;14:138–140. [21] Nanji AA, French SW. Dietary linoleic acid is required for development of experimentally induced alcoholic liver injury. Life Sci 1989;44:223–227. [22] Nanji AA, Mendenhall CL, French SW. Beef fat prevents alcoholic liver disease in the rat. Alcohol Clin Exp Res 1989;13:15–19. [23] Kessova IG, Ho YS, Thung S, Cederbaum AI. Alcohol-induced liver injury in mice lacking Cu,Zn-superoxide dismutase. Hepatology 2003;38:1136–1145. [24] Lu SC, Alvarez L, Huang ZZ, Chen L, An W, Corrales FJ, et al. Methionine adenosyltransferase 1A knockout mice are predisposed to liver injury and exhibit increased expression of genes involved in proliferation. Proc Natl Acad Sci USA 2001;98:5560–5565. [25] Lieber CS. S-Adenosyl-L-methionine and alcoholic liver disease in animal models: implications for early intervention in human beings. Alcohol 2002;27:173–177. [26] Wheeler MD, Nakagami M, Bradford BU, Uesugi T, Mason RP, Connor HD, et al. Overexpression of manganese superoxide dismutase prevents alcohol-induced liver injury in the rat. J Biol Chem 2001;276:36664–36672. [27] Akriviadis E, Botla R, Briggs W, Han S, Reynolds T, Shakil O. Pentoxifylline improves short-term survival in severe acute alcoholic hepatitis: a double-blind, placebo-controlled trial. Gastroenterology 2000;119:1637–1648. [28] Holt S, Goodier D, Marley R, Patch D, Burroughs A, Fernando B, et al. Improvement in renal function in hepatorenal syndrome with N-acetylcysteine. Lancet 1999;23:294–295, 353. [29] Wenzel G, Kuklinski B, Ruhlmann C, Ehrhardt D. [Alcoholinduced toxic hepatitis-A ‘‘free radical’’ associated disease. Lowering fatality by adjuvant antioxidant therapy]. Z Gesamte Inn Med 1993;48:490–496. [30] Phillips M, Curtis H, Portmann B, Donaldson N, Bomford A, O’Grady J. Antioxidants versus corticosteroids in the treatment of severe alcoholic hepatitis-A randomised clinical trial. J Hepatol 2006;44:784–790. [31] Nagita A, Ando M. Assessment of hepatic vitamin E status in adult patients with liver disease. Hepatology 1997;26:392–397. [32] Meagher EA, Barry OP, Lawson JA, Rokach J, FitzGerald GA. Effects of vitamin E on lipid peroxidation in healthy persons. JAMA 2001;285:1178–1182. [33] Kaikkonen J, Porkkala-Sarataho E, Morrow JD, Roberts 2nd LJ, Nyyssonen K, Salonen R, et al. Supplementation with vitamin E but not with vitamin C lowers lipid peroxidation in vivo in mildly hypercholesterolemic men. Free Radic Res 2001;35:967–978. [34] Sacheck JM, Milbury PE, Cannon JG, Roubenoff R, Blumberg JB. Effect of vitamin E and eccentric exercise on selected biomarkers of oxidative stress in young and elderly men. Free Radic Biol Med 2003;34:1575–1588. [35] Upritchard JE, Schuurman CR, Wiersma A, Tijburg LB, Coolen SA, Rijken PJ, et al. Spread supplemented with moderate doses of vitamin E and carotenoids reduces lipid peroxidation in healthy, nonsmoking adults. Am J Clin Nutr 2003;78:985–992.
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283 [36] Thomas SR, Stocker R. Molecular action of vitamin E in lipoprotein oxidation: implications for atherosclerosis. Free Radic Biol Med 2000;28:1795–1805. [37] Podmore ID, Griffiths HR, Herbert KE, Mistry N, Mistry P, Lunec J. Vitamin C exhibits pro-oxidant properties. Nature 1998;392:559. [38] Poulsen HE, Weimann A, Salonen JT, Nyyssonen K, Loft S, Cadet J, et al. Does vitamin C have a pro-oxidant effect? Nature 1998;395:231–232. [39] Adachi Y, Moore LE, Bradford BU, Gao W, Thurman RG. Antibiotics prevent liver injury in rats following long-term exposure to ethanol. Gastroenterology 1995;108:218–224. [40] Yin M, Wheeler MD, Kono H, Bradford BU, Gallucci RM, Luster MI, et al. Essential role of tumor necrosis factor alpha in alcoholinduced liver injury in mice. Gastroenterology 1999;117:942–952.
283
[41] Taieb J, Mathurin P, Elbim C, Cluzel P, Arce-Vicioso M, Bernard B, et al. Blood neutrophil functions and cytokine release in severe alcoholic hepatitis: effect of corticosteroids. J Hepatol 2000;32:579–586. [42] Viitala K, Israel Y, Blake JE, Niemela O. Serum IgA, IgG, and IgM antibodies directed against acetaldehyde-derived epitopes: relationship to liver disease severity and alcohol consumption. Hepatology 1997;25:1418–1424. [43] Mottaran E, Stewart SF, Rolla R, Vay D, Cipriani V, Moretti M, et al. Lipid peroxidation contributes to immune reactions associated with alcoholic liver disease. Free Radic Biol Med 2002;32:38–45. [44] Natori S, Rust C, Stadheim LM, Srinivasan A, Burgart LJ, Gores GJ. Hepatocyte apoptosis is a pathologic feature of human alcoholic hepatitis. J Hepatol 2001;34:248–253.
A randomized trial of antioxidant therapy alone or with corticosteroids in acute alcoholic hepatitisq Stephen Stewart1, Martin Prince1, Margaret Bassendine1,2, Mark Hudson2, Oliver James1,2, David Jones1,2, Chris Record2, Christopher P. Day1,2,* 1
Liver Research Group, Institute of Cellular Medicine, Medical School, Newcastle University, Newcastle upon Tyne NE2 4HH, UK 2 The Liver Unit, Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK
Background/Aims: Oxidative stress is putatively involved in the pathogenesis of alcohol-induced liver injury. This trial was devised to determine whether antioxidant therapy, alone or as an adjunct to corticosteroids, improved survival in patients with acute alcoholic hepatitis. Methods: Patients with a severe alcoholic hepatitis were stratified by sex and steroid use, and then randomized. The active group received N-acetylcysteine for one week, and vitamins A–E, biotin, selenium, zinc, manganese, copper, magnesium, folic acid and Coenzyme Q daily for 6 months. The trial was double blinded and placebo controlled. The primary end-point was mortality within 6 months. Results: Thirty-six (20 male, 16 female; mean discriminant function (DF) 86.6) received active drug, and 34 (18 male, 16 female; mean DF 76.4) received placebo. 180-day survival was not significantly different between patients receiving drug and placebo (52.8% vs. 55.8%, p = 0.699). This was not affected by stratification for steroid use or sex. The only predictors of survival in multivariate analysis were initial bilirubin (p = 0.017), white cell count (p = 0.016) and age (p = 0.037). Treatment allocation did not affect survival in multivariate analysis (p = 0.830). Conclusions: Antioxidant therapy, alone or in combination with corticosteroids, does not improve 6-month survival in severe alcoholic hepatitis. 2007 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Alcoholic hepatitis; Clinical trial; Antioxidants
1. Introduction Alcoholic hepatitis is a disease with a wide clinical spectrum, ranging from sub-clinical forms to a severe acute disease. The mild forms have a good short-term prognosis, but the in-hospital mortality rate of the severe form is around 40% [1]. Features that identify Received 24 January 2007; received in revised form 22 February 2007; accepted 3 March 2007; available online 4 May 2007 q The authors who have taken part in this study declared that they have no relationship with the manufacturers of the drugs involved either in the past or present and did not receive funding from the manufacturers to carry out their research. * Corresponding author. Tel.: +44 191 222 7043; fax: +44 191 222 0723. E-mail address: [email protected] (C.P. Day).
patients with a high mortality include the presence of encephalopathy, a high serum bilirubin and a prolonged prothrombin time. These latter two variables have been combined to form the discriminant function (DF) [2], which has been shown to prospectively identify patients with a 40–50% risk of dying within 2 months [3]. Conventional management consists of abstinence from alcohol, the correction of dietary deficiencies and general support. Of the specific therapeutic agents evaluated prior to trial commencement, only glucocorticoids have been shown to have any benefit on survival. A recent analysis of individual patient data from the last three randomized controlled trials (RCTs) showed a higher 28 day survival (84.6 ± 3.4% vs. 65.1 ± 4.8%, p = 0.001) in corticosteroid compared to placebo treated patients with an initial DF of >32 [4]. Importantly,
0168-8278/$32.00 2007 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jhep.2007.03.027
278
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
active infection and gastrointestinal bleeding are still contraindications to steroid treatment in most centers, and both are relatively common in these patients. Furthermore, mortality is still high whether patients do or do not receive steroids, implying the need for alternative therapeutic options. Interest in the possible value of antioxidant therapy in the treatment of alcoholic hepatitis has arisen as a result of increasing evidence implicating oxidative stress as a key mechanism in alcohol-mediated hepatotoxicity [5]. This evidence falls into five categories. First, products of lipid peroxidation can be detected in the peripheral blood of heavy drinkers [6] and in the livers of patients with ALD [7], and the magnitude of lipid peroxidation correlates with the degree of liver injury [8]. Second, in patients and animal models of ALD, lipid peroxidation is most prominent in the perivenular region where the liver injury is typically most severe [7]. Third, a variety of sources of oxidative stress have been identified in patients with ALD and in animal models of disease. Considerable controversy remains over the most important source of the prooxidant reactive oxygen species (ROS) which are capable of initiating lipid peroxidation and damaging plasma and intracellular membranes [9], but the most likely appear to be microsomal cytochrome P4502E1 (CYP2E1; the only source of hydroxyethyl radicals [10,11]), the mitochondrial electron transport chain [12], inducible nitric oxide synthase [13] and Kupffer cells [14,15]. Fourth, ethanol consumption results in the depletion of endogenous antioxidant capabilities and patients with ALD have evidence of antioxidant deficiencies. Consumption of glutathione (GSH) during oxidative stress and inhibition of two enzymes involved in the synthesis of its precursor, S-adenosylmethionine (SAMe), methionine synthase (MS) and methionine adenosyltransferase (MAT), contribute to the decreased levels of hepatic SAMe and GSH observed in patients with ALD [16]. Heavy drinkers, including those with ALD, are also deficient in the antioxidant trace element selenium [17], which is required for the activity of the antioxidant enzyme GSH peroxidase, the antioxidant vitamins A, C and E [18,19] and Coenzyme Q [20]. This latter compound is present in plasma and mitochondrial matrix membranes and has emerged as one of the most important natural free radical scavengers. It is partly derived from the diet, but is also synthesized in the liver. Finally, in animal models of ALD, dietary [21,22] and genetic manipulations [23,24] that increase oxidative stress increase the severity of liver injury and reducing oxidative stress ameliorates injury [25,26]. Given the nature and extent of this evidence, we were keen to determine whether antioxidant supplementation conferred any benefit to patients with severe alcoholic hepatitis.
2. Methods After local Ethical Committee permission was obtained, consecutive heavy drinkers admitted to the Freeman hospital liver unit with alcoholic hepatitis with hepatic encephalopathy and/or a discriminant function (DF) of P32 were considered for inclusion in the study. Recruitment to the study started in 1998 and follow-up was completed in 2002. Inclusion criteria were; recent heavy (>40 g/day for women; >60 g/day for men) alcohol intake, age between 18 and 65 and either a diagnostic liver biopsy or two of the following: hepatomegaly, leukocytosis >11 · 109 cells/L and ‘‘white out’’ on liver and spleen isotope scanning. Exclusion criteria were; evidence of malignancy, positive HBV or HCV serology, pregnant or lactating women and cirrhotic patients admitted primarily for control of the complications arising from portal hypertension. Steroids were given unless there was evidence of active infection or gastrointestinal bleeding. Patients were then stratified by sex and steroid use, and randomized by an independent pharmacist using a computer-derived randomization schedule. The active group received loading dose N-acetylcysteine (NAC) of 150 mg/kg followed by 100 mg/kg/day for one week, and vitamins A–E, biotin, selenium, zinc, manganese, copper, magnesium, folic acid and Coenzyme Q daily for 6 months. The trial was blinded and control patients received identical placebo. Assessments were carried out weekly for one month and monthly thereafter. The primary end-point was 6-month mortality. We hoped to recruit 80 patients to the study, allowing 10 per group stratifying for sex, steroid use and randomization to antioxidants or placebo. Power calculations were based on 80% power to identify a 40% reduction in mortality with a 2-sided significance of 5%. Continuous variables were compared using the Mann–Whitney U test. Univariate survival analysis was performed using stratified Kaplan–Meier analysis and the log rank test. Multivariate survival analysis was performed using stratified Cox proportional hazards regression. All possible predictive variables were entered into a Cox model accounting for stratification and using a backwards stepwise method, removing the least predictive variable at each step. Results were verified using forwards stepwise regression with a cut off for inclusion of variables of p = 0.10. Multivariate analysis was performed to determine factors associating with survival. All analyses were performed using SPSS 11.0 (SPSS Corp., USA) on a per protocol basis.
3. Results In total, 77 patients were recruited to the study and 70 were followed up until death or censor at 6 months. The shortfall of 10 patients arose due to difficulty in recruiting females that could not receive steroids. The trial had to be terminated when it was no longer reasonable not to treat these patients with pentoxifylline [27]. From the 70 patients, 36 (20 male, 16 female; mean DF 86.6) were randomized to receive active drug, and 34 (18 male, 16 female; mean DF 76.4) received placebo (Fig. 1). At trial entry there were no significant differences between the active and placebo groups with regard to age or the major laboratory indices that might affect mortality. Furthermore, the incidence of hepatic encephalopathy, gastrointestinal bleeding and sepsis was similar between the two groups (see Table 1). Liver histology was available for 43 patients (63%) and cirrhosis was present in 84% of these. For the 38 patients treated with steroids the mean period of treatment was 16 days. Treatment was discontinued prematurely in 14 (37%) patients because of gastrointestinal bleeding (5 patients) or clinical sepsis (9 patients). There were no
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
279
Assessed for eligibility (n=86)
Excluded Not AAH No consent
(n= 9) (n=8) (n= 1)
Stratified by sex and steroid use (n=77)
Males given steroids (n=21)
Males not given steroids (n=18)
Females given steroids (n=22)
Females not given steroids (n=16)
Randomization
AO (n=11) Placebo (n=10)
AO (n=9) Placebo (n=9)
Did not attend follow-up clinic (n=1)
AO (n=11) Placebo (n=11)
Follow-Up
AO (n=8) Placebo (n=8)
Did not attend follow-up clinic (n=4)
Did not attend follow-up clinic (n=2)
Analysis AO (n=11) Placebo (n=9)
AO (n=9) Placebo (n=9)
AO (n=9) Placebo (n=9)
AO (n=7) Placebo (n=7)
Fig. 1. Trial profile. AO, antioxidants.
Table 1 Comparison of active and placebo groups prior to treatment Active
Placebo
p value
Variable (median/interquartile range) Sex (male:female) Age (years) White cell count (·109/L) Prothrombin time (s) Creatinine (lmol/L) Bilirubin (lmol/L) Albumin (g/L) Child–Pugh score Discriminant function
36 (20:16) 44.00 (38.25-50.25) 13.20 (9.30–20.70) 27.00 (22.00–32.50) 117.00 (88.50–168.00) 473.5 (246.00–581.25) 27.00 (25.00–30.75) 12.00 (11.25–13.00) 75.50 (60.50–108.25)
34 (18:16) 44.50 (39.75-50.00) 14.40 (8.47–19.10) 24.50 (22.00–29.00) 110.50 (81.50–175.50) 378.00 (240.25–603.75) 28.50 (25.00–31.00) 12.00 (11.00–13.00) 67.00 (51.00–87.50)
0.984 0.773 0.677 0.179 0.601 0.659 0.637 0.532 0.110
Complication Portosystemic encephalopathy Gastrointestinal bleeding Sepsis
24/36 (66.7%) 4/36 (11.1%) 12/36 (33.3%)
26/34 (76.5%) 4/34 (11.8%) 13/34 (38.2%)
0.433 1.000 0.804
There was no significant differences between the two groups prior to treatment.
280
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
Kaplan Meier Survival
01.00
0.80
0.60
0.40
TREATMENT
0.20
active
00.00
placebo
0
60
120
180
Time (days) No at risk
70
Survival active 100% Placebo 100%
44
39
38
65% 62%
56% 56%
56% 53%
Fig. 2. Kaplan–Meier survival curves. After stratification for sex and steroid use, there is no significant difference (p = 0.710) between the Kaplan–Meier survival curves for patients receiving active drug or placebo.
adverse effects reported attributable to antioxidant or placebo. Patients were followed until death or censor at six months. In total, 7 patients were lost to follow-up; 3 from the active group and 4 from the placebo group. The 6 month survival was not significantly different between patients receiving active drug and placebo (19/36; 52.8% vs. 19/34; 55.8%, p = 0.699) (see Fig. 2). This result was unchanged in intention to treat analysis (with patients censored at point of last contact). As shown in Table 2, treatment allocation did not affect survival in univariate (p = 0.884) or multivariate (0.830) analysis adjusted for stratification by Cox proportional hazards regression. The only variables that associated with mortality in multivariate analysis were the white cell count (p = 0.016), bilirubin (p = 0.017) and age (p = 0.037) at trial entry (see Table 2).
In view of a small non-randomized trial in 12 patients (9 with alcoholic liver disease) with hepatorenal syndrome [28] showing a significant improvement in renal function on N-acetylcysteine (NAC), we conducted a post hoc analysis of the effect of our regime on patients with initial renal impairment. From 17 patients with significant renal impairment (creatinine > 130 lmol/L) that received active therapy, 15 survived the full weeks treatment and could be analyzed. The median reduction in creatinine in these patients was 40l mol/L (29% of baseline, p = 0.041). However this was not significantly different from the reduction observed in patients receiving placebo (n = 11, median reduction in creatinine 43lmol/L, 23% of baseline p = 0.019; p for comparison of active with control = 0.656) (see Fig. 3).
4. Discussion In spite of the evidence for oxidative stress in the pathogenesis of ALD, and the concomitant depletion of antioxidant defenses in heavy drinkers, only two trials have investigated the effect of antioxidant supplementation in acute severe alcoholic hepatitis. In the first study, 56 patients were randomized to receive Vitamin E, selenium and zinc, or placebo [29]. While treated patients had an in-hospital mortality of 6.5% compared to 40% in the control group, the entry criteria and patient details are not clearly discussed. The second trial [30] compared steroids with an antioxidant cocktail (vitamins A, C, E, selenium, allopurinol, desferrioxamine and N-acetylcysteine), and was stopped after an interim assessment found steroids to be of greater benefit. This latter trial included patients with moderate and severe disease (including patients with a DF < 32) and did not determine whether antioxidants conferred any benefit in patients that had a contraindication to steroids, or in addition to steroids. Our aim was to answer these questions using a panel of antioxidants in patients with severe acute alcoholic hepatitis.
Table 2 Cox proportionate regression of factors affecting survival stratified by use of steroids and sex Variable tested
Treatment allocation White cell count (·109/L) Creatinine (lmol/L) Albumin (g/L) Bilirubin (lmol/L) Prothrombin time (s) Age (years) Sepsis Bleed PSE
Univariate analysis
Multivariate analysis
B
SE (B)
p value
B
SE (B)
p value
0.054 0.039 0.000 0.019 0.003 0.064 0.047 0.343 0.507 0.295
0.365 0.021 0.001 0.044 0.001 0.027 0.023 0.413 0.501 0.431
0.884 0.063 0.714 0.663 0.012 0.017 0.041 0.406 0.312 0.494
0.081 0.063 0.001 0.012 0.003 0.052 0.050 0.636 0.494 0.376
0.378 0.026 0.001 0.051 0.001 0.035 0.024 0.518 0.543 0.476
0.830 0.016 0.332 0.816 0.017 0.137 0.037 0.220 0.363 0.429
Pre-treatment white cell count, bilirubin and age were the only factors studied that affected survival.
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
Cr (micromol/L)
a
400
Cr (micromol/L)
*
300
200
100
0
b
*
week 0
week 1
300
** **
200
100
0
week 0
week 1
Fig. 3. The effect of antioxidants including N-acetylcysteine on renal function in alcoholic hepatitis. Among patients with renal impairment at trial entry, a significant improvement in mean serum creatinine was seen after 1 week in patients that received (a) antioxidant therapy (*p = 0.041), or (b) placebo (**p = 0.019). There was no significant difference between the fall in mean creatinine between these 2 groups (p = 0.656).
The 180-day mortality rate of severe alcoholic hepatitis in our cohort of 70 patients was 46%, which is similar to previous reports. Two-thirds of the patients that died did so in the first month, however death events occurred beyond three months in both the active and placebo treated groups. In a recent meta-analysis, age and creatinine on trial entry have been associated with mortality [4]. We found mortality to be associated with initial white cell count, bilirubin and age. There is now good evidence that corticosteroids improve 28 day mortality in patients with an initial DF of >32 [4], so all patients that did not have a contraindication received this treatment. This allowed us to assess whether there was an additional benefit to antioxidants; a relevant clinical question. Antioxidant therapy, however, alone or in combination with corticosteroids, did not improve 6month survival. With good evidence implicating oxidative stress in the pathogenesis of alcoholic liver disease, why do antioxidants given on admission to hospital with severe alcoholic hepatitis not result in improved survival? The first possibility is that the drug doses prescribed were insufficient. Although we did not monitor serum levels, we did provide a broad cocktail of antioxidants, which, in most cases, amounted to several times the daily recommended dose. Furthermore, even after a month of high dose administration, the survival curves for active and placebo treated patients were very similar. This
281
would imply that, at this stage in disease pathogenesis, even high dose, and prolonged antioxidant treatment cannot affect disease progression. The second possibility is that the cocktail we used did not perform an antioxidant function. It may have been possible to assess this function in a dynamic way by monitoring markers of oxidative stress, and there are several recognized techniques for doing this. These include measuring levels of F2-isoprostanes or the aldehyde end-products of lipid peroxidation, malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE). It is less clear which method is the best and, more importantly, whether levels of these markers in the serum or urine reflect levels in the liver. For example it has been demonstrated that the level of vitamin E in the serum does not correlate with hepatic levels [31]. It has also been recognized that components of standard antioxidant therapies may be ineffective or act as pro-oxidants under certain circumstances; namely vitamin E and vitamin C. One trial in healthy volunteers showed no benefit of vitamin E on lipid peroxidation [32]. This can be compared with other studies showing a beneficial effect [33–35]. Concern regarding the role of vitamin E in the development of lipid peroxidation in atherosclerosis is ameliorated by evidence that this effect is suppressed in the presence of other antioxidant agents [36]. While data suggesting that vitamin C has pro-oxidant properties [37] have been questioned [38], it must be accepted that the ideal combination and dose of antioxidant cocktail, and the effect it has on oxidative stress in the liver are not entirely clear and may be difficult to assess. Further study is clearly warranted in this area. The third possibility for the lack of survival benefit is that ongoing oxidative stress is not the primary mechanism underlying mortality in severe alcoholic hepatitis after presentation to hospital. Other mechanisms, that could be potential therapeutic targets, are implicated in disease progression in humans and rodents. These involve endotoxin-mediated cytokine release [39,40], neutrophil recruitment and activation [41] and adaptive immune responses to the products of ethanol metabolism and oxidative stress [42,43]. Furthermore, it is now clear that hepatocyte apoptosis is a pathological feature of alcoholic hepatitis, and the extent to which this occurs correlates with disease severity [44]. These mechanisms of disease will presumably interact during disease development, however, histologically, alcoholic hepatitis is a neutrophil predominant inflammatory condition. Future therapeutic strategies for acute severe alcoholic hepatitis may focus more on reducing recruitment or inhibiting activation of neutrophils, and suppressing Kupffer cell cytokine release and apoptosis in the short term. In summary, high dose antioxidant therapy confers no survival benefit in acute severe alcoholic hepatitis. In addition, it appears that the improvement in renal function seen in the first week after trial entry is due
282
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283
to factors other than antioxidants or N-acetylcysteine. This does not exclude a role for this therapy in the delay or prevention of disease in heavy drinkers, or in reducing the risk of disease progression in persistent drinkers that have had an episode of hepatitis. It does, however, highlight that therapeutic strategies aimed at other putative mechanisms of disease should now undergo trials aimed at reducing the mortality rate of this devastating disease. References [1] Hardison WG, Lee FI. Prognosis in acute liver disease of the alcoholic patient. N Engl J Med 1966;275:61–66. [2] Maddrey WC, Boitnott JK, Bedine MS, Weber Jr FL, Mezey E, White Jr RI. Corticosteroid therapy of alcoholic hepatitis. Gastroenterology 1978;75:193–199. [3] Ramond MJ, Poynard T, Rueff B, Mathurin P, Theodore C, Chaput JC, et al. A randomized trial of prednisolone in patients with severe alcoholic hepatitis. N Engl J Med 1992;326:507–512. [4] Mathurin P, Mendenhall CL, Carithers RL, Ramond MJ, Maddrey WC, Garstide P, et al. Corticosteroids improve shortterm survival in patients with severe alcoholic hepatitis (AH): individual data analysis of the last three randomized placebo controlled double blind trials of corticosteroids in severe AH. J Hepatol 2002;36:480–487. [5] Dey A, Cederbaum AI. Alcohol and oxidative liver injury. Hepatology 2006;43:S63–S74. [6] Clot P, Tabone M, Arico S, Albano E. Monitoring oxidative damage in patients with liver cirrhosis and different daily alcohol intake. Gut 1994;35:1637–1643. [7] Paradis V, Kollinger M, Fabre M, Holstege A, Poynard T, Bedossa P. In situ detection of lipid peroxidation by-products in chronic liver diseases. Hepatology 1997;26:135–142. [8] Situnayake RD, Crump BJ, Thurnham DI, Davies JA, Gearty J, Davis M. Lipid peroxidation and hepatic antioxidants in alcoholic liver disease. Gut 1990;31:1311–1317. [9] Halliwell B. Free radicals, antioxidants, and human disease: curiosity, cause, or consequence?. Lancet 1994;344:721–724. [10] Ekstrom G, Ingelman-Sundberg M. Rat liver microsomal NADPH-supported oxidase activity and lipid peroxidation dependent on ethanol-inducible cytochrome P-450 (P-450IIE1). Biochem Pharmacol 1989;38:1313–1319. [11] Albano E, Tomasi A, Goria-Gatti L, Dianzani MU. Spin trapping of free radical species produced during the microsomal metabolism of ethanol. Chem Biol Interact 1988;65:223–234. [12] Slater TF, Sawyer BC, Strauli UD. Changes in liver nucleotide concentrations in experimental liver injury. 2. Acute ethanol poisoning. Biochem J 1964;93:267–270. [13] McKim SE, Gabele E, Isayama F, Lambert JC, Tucker LM, Wheeler MD, et al. Inducible nitric oxide synthase is required in alcohol-induced liver injury: studies with knockout mice. Gastroenterology 2003;125:1834–1844. [14] Niemela O, Parkkila S, Bradford B, Iimuro Y, Pasanen M, Thurman RG. Effect of Kupffer cell inactivation on ethanolinduced protein adducts in the liver. Free Radic Biol Med 2002;33:350–355. [15] Kono H, Rusyn I, Yin M, Gabele E, Yamashina S, Dikalova A, et al. NADPH oxidase-derived free radicals are key oxidants in alcohol-induced liver disease. J Clin Invest 2000;106:867–872. [16] Paredes SR, Kozicki PA, Fukuda H, Rossetti MV, Batlle AM. SAdenosyl-L-methionine: its effect on aminolevulinate dehydratase and glutathione in acute ethanol intoxication. Alcohol 1987;4:81–85.
[17] Dworkin B, Rosenthal WS, Jankowski RH, Gordon GG, Haldea D. Low blood selenium levels in alcoholics with and without advanced liver disease. Correlations with clinical and nutritional status. Dig Dis Sci 1985;30:838–844. [18] Thurnham DI, Davies JA, Crump BJ, Situnayake RD, Davis M. The use of different lipids to express serum tocopherol: lipid ratios for the measurement of vitamin E status. Ann Clin Biochem 1986;23:514–520. [19] Leo MA, Lieber CS. Hepatic vitamin A depletion in alcoholic liver injury. N Engl J Med 1982;307:597–601. [20] Bianchi GP, Fiorella PL, Bargossi AM, Grossi G, Marchesini G. Reduced ubiquinone plasma levels in patients with liver cirrhosis and in chronic alcoholics. Liver 1994;14:138–140. [21] Nanji AA, French SW. Dietary linoleic acid is required for development of experimentally induced alcoholic liver injury. Life Sci 1989;44:223–227. [22] Nanji AA, Mendenhall CL, French SW. Beef fat prevents alcoholic liver disease in the rat. Alcohol Clin Exp Res 1989;13:15–19. [23] Kessova IG, Ho YS, Thung S, Cederbaum AI. Alcohol-induced liver injury in mice lacking Cu,Zn-superoxide dismutase. Hepatology 2003;38:1136–1145. [24] Lu SC, Alvarez L, Huang ZZ, Chen L, An W, Corrales FJ, et al. Methionine adenosyltransferase 1A knockout mice are predisposed to liver injury and exhibit increased expression of genes involved in proliferation. Proc Natl Acad Sci USA 2001;98:5560–5565. [25] Lieber CS. S-Adenosyl-L-methionine and alcoholic liver disease in animal models: implications for early intervention in human beings. Alcohol 2002;27:173–177. [26] Wheeler MD, Nakagami M, Bradford BU, Uesugi T, Mason RP, Connor HD, et al. Overexpression of manganese superoxide dismutase prevents alcohol-induced liver injury in the rat. J Biol Chem 2001;276:36664–36672. [27] Akriviadis E, Botla R, Briggs W, Han S, Reynolds T, Shakil O. Pentoxifylline improves short-term survival in severe acute alcoholic hepatitis: a double-blind, placebo-controlled trial. Gastroenterology 2000;119:1637–1648. [28] Holt S, Goodier D, Marley R, Patch D, Burroughs A, Fernando B, et al. Improvement in renal function in hepatorenal syndrome with N-acetylcysteine. Lancet 1999;23:294–295, 353. [29] Wenzel G, Kuklinski B, Ruhlmann C, Ehrhardt D. [Alcoholinduced toxic hepatitis-A ‘‘free radical’’ associated disease. Lowering fatality by adjuvant antioxidant therapy]. Z Gesamte Inn Med 1993;48:490–496. [30] Phillips M, Curtis H, Portmann B, Donaldson N, Bomford A, O’Grady J. Antioxidants versus corticosteroids in the treatment of severe alcoholic hepatitis-A randomised clinical trial. J Hepatol 2006;44:784–790. [31] Nagita A, Ando M. Assessment of hepatic vitamin E status in adult patients with liver disease. Hepatology 1997;26:392–397. [32] Meagher EA, Barry OP, Lawson JA, Rokach J, FitzGerald GA. Effects of vitamin E on lipid peroxidation in healthy persons. JAMA 2001;285:1178–1182. [33] Kaikkonen J, Porkkala-Sarataho E, Morrow JD, Roberts 2nd LJ, Nyyssonen K, Salonen R, et al. Supplementation with vitamin E but not with vitamin C lowers lipid peroxidation in vivo in mildly hypercholesterolemic men. Free Radic Res 2001;35:967–978. [34] Sacheck JM, Milbury PE, Cannon JG, Roubenoff R, Blumberg JB. Effect of vitamin E and eccentric exercise on selected biomarkers of oxidative stress in young and elderly men. Free Radic Biol Med 2003;34:1575–1588. [35] Upritchard JE, Schuurman CR, Wiersma A, Tijburg LB, Coolen SA, Rijken PJ, et al. Spread supplemented with moderate doses of vitamin E and carotenoids reduces lipid peroxidation in healthy, nonsmoking adults. Am J Clin Nutr 2003;78:985–992.
S. Stewart et al. / Journal of Hepatology 47 (2007) 277–283 [36] Thomas SR, Stocker R. Molecular action of vitamin E in lipoprotein oxidation: implications for atherosclerosis. Free Radic Biol Med 2000;28:1795–1805. [37] Podmore ID, Griffiths HR, Herbert KE, Mistry N, Mistry P, Lunec J. Vitamin C exhibits pro-oxidant properties. Nature 1998;392:559. [38] Poulsen HE, Weimann A, Salonen JT, Nyyssonen K, Loft S, Cadet J, et al. Does vitamin C have a pro-oxidant effect? Nature 1998;395:231–232. [39] Adachi Y, Moore LE, Bradford BU, Gao W, Thurman RG. Antibiotics prevent liver injury in rats following long-term exposure to ethanol. Gastroenterology 1995;108:218–224. [40] Yin M, Wheeler MD, Kono H, Bradford BU, Gallucci RM, Luster MI, et al. Essential role of tumor necrosis factor alpha in alcoholinduced liver injury in mice. Gastroenterology 1999;117:942–952.
283
[41] Taieb J, Mathurin P, Elbim C, Cluzel P, Arce-Vicioso M, Bernard B, et al. Blood neutrophil functions and cytokine release in severe alcoholic hepatitis: effect of corticosteroids. J Hepatol 2000;32:579–586. [42] Viitala K, Israel Y, Blake JE, Niemela O. Serum IgA, IgG, and IgM antibodies directed against acetaldehyde-derived epitopes: relationship to liver disease severity and alcohol consumption. Hepatology 1997;25:1418–1424. [43] Mottaran E, Stewart SF, Rolla R, Vay D, Cipriani V, Moretti M, et al. Lipid peroxidation contributes to immune reactions associated with alcoholic liver disease. Free Radic Biol Med 2002;32:38–45. [44] Natori S, Rust C, Stadheim LM, Srinivasan A, Burgart LJ, Gores GJ. Hepatocyte apoptosis is a pathologic feature of human alcoholic hepatitis. J Hepatol 2001;34:248–253.