Acute Alcoholic Hepatitis

Acute alcoholic hepatitis characterised histologically by a florid inflammatory lesion and liver injury is part of a spectrum of liver damage resulting from alcoholic excess (Figure [1]). Although many cases of alcoholic cirrhosis develop insidiously, others will pass through one or more episodes of acute alcoholic hepatitis (AH). Clinically AH is manifest by acute onset of jaundice with fever and other signs of constitutional upset and with the rapid development of signs of acute liver failure, including renal impairment and encephalopathy. Very often it follows an acute alcoholic binge. It may be the first presentation of liver disease for that particular patient. Some episodes of AH can resolve without the development of cirrhosis if abstinence is maintained and acute alcoholic hepatitis is to be distinguished from acute episodes of decompensation in patients with an established underlying cirrhosis. AH carries a high mortality more than 50 % when the MADDREY index is > 32. The MADDREY index which has proved to be of considerable value as a prognostic tool and in estimating the need for particular methods of treatment such as corticosteroids, is based on the following formula which includes the height of the bilirubin and degree of prolongation of prothrombin time.

Figure 1 Liver Damage from Alcohol Excess.

Index = (Bilirubin in micromol/L divided by 17.1) + (4.6 X PT Patient - PT control).

Standard treatment measures for AH include corticosteroid therapy, particularly when encephalopathy is a prominent feature and with the important exceptions of infection and bleeding. However, as shown in Table [1] its value is unproven, at least as judged by the results of the various meta analyses of controlled trials carried out to date. The same applies to the other favoured clinical therapy, namely nutritional supplementation, the results of the most recent study being shown in Table [2]. The treatment, therefore, mainly consists of intensive care management based on the management of multi-organ failure as it develops, with the prevention of infection, maintenance of nutrition and other standard directions of intensive liver care.

Table 1 Value of corticosteroids in treatment of acute alcoholic hepatitis Two of the three published meta-analyses of placebo-controlled randomised trials showed a beneficial effect of steroids. Reynolds et al, Gastroenterology Int 1 989 Imperiale et al, Ann Intern Med 1 990 The third found no differences. Christenson et al, Gut 1 995 Nevertheless, continued impression that those with severe AH and encephalopathy can benefit.

Table 2 Short and long term outcome of severe alcohol induced hepatitis treated with steroids or enteral nutrition - 71 patients randomised 40 mg/d Prednisolone or enteral tube feeding (2 000 kcal/d) for 28 days and followed for 1 year or until death. - Side effects occurred in 5 patients on steroids and 10 on enteral nutrition - 8 patients of the latter group prematurely withdrawn from trial. - Mortality during treatment similar in both groups (9 of 36 vs. 11 of 37), but occureed earlier with enteral feeding (median 7 vs. 23 days) - During follow up higher with steroids, largerly infection 7 patients dying within first 1 œ months. From Cabre et al on behalf of Spanish Group for study of alcoholic hepatitis, Hepatology 2 000; 32 : 36

Despite the initiation of such optimal care and the withdrawal of alcohol, the patient often deteriorates further during the first 2-3 weeks after admission, with deepening jaundice and progression of the manifestations of multi-organ impairment and failure. This well known but ill-explained clinical observation may now be better understood in the light of new knowledge gained in experimental animal models of acute alcohol hepato-toxicity. The results of these are consistent in showing a dominant influence of TNF-α along with endotoxin in driving an early cytokine cascade (Table [3]). The latter is likely to be responsible for the acute inflammatory like injury. Correlations with the important stellate cell activation and the very characteristic release of neutrophil chemotaxic factors accounting for the polymorph infiltration, so characteristic on histological examination, are demonstrable. Activation of TNF system as measured by the gene expression of TNF-receptors in peripheral blood mononuclear cells has been shown to reflect disease severity and correlates with plasma endotoxin level. No significant up regulation of interleukin-10 known to inhibit secretion of TNF-α.

Table 3 Role of Kupffer cells and gut-derived endotoxins in alcoholic liver injury - Ethanol alters gut permeability and activation of Kupffer cells by gut derived endotoxin. - This primary event in initiation of alcoholic liver injury, followed by release of inflammatory mediators, including the cytokines TNF-α, interleukin 1 & 6, TGF-β as well as reactive oxygen intermediates. Also factors for neutrophil chemotaxis and expression of intracellular adhesion molecule-1 (ICAM-1) leading to microcirculatory disturbances. Liver damage in animal rat model diminished by antibody to TNF-α and does not occur in TNF-α receptor knockout mice. Inactivation of Kupffer cells with gadolinium chloride and intestinal sterilisation also prevents early liver damage including hypermetabolic state.

Further support for this view is the finding that the administration of antibody to TNF-α has been shown to ameliorate the course of experimental induced liver injury from alcohol. Our experience with this agent in a pilot clinical trial of 3 patients with acute alcoholic hepatitis treated at the Cromwell Hospital, London, is shown in Table [4]. The preparation used was Infliximab (Remocade 25 mouse, 75 human), given as an infusion over a 2 hour period 5 mg/Kg in 250 ml of 0.9 % saline. All three patients survived, although the falls in serum bilirubin observed were gradual. It is difficult to assess without a controlled clinical trial in which the agent is compared with placebo whether the clinical effects observed and changes in cytokine levels are related to the infused antibody. One such multicentre trial involving a large number of patients is, I understand, currently underway in Paris and the results of this will be awaited with interest.

Table 4 Serum bilirubin levels (umol/L) in acute alcoholic hepatitis given TNF-α antibody (fleximab) * Case 1 Aged 55 2 61 3 39 Pre INF 271 289 625 Post 1 wk 348 235 554 2 wk 318 260 438 3 wk 274 235 344 4 wk 190 184 (Personal data of Williams et al. 2000)

Experimental studies by Lieber and colleagues has shown the importance of free radical and super oxide mediated liver cell damage in various animal models of alcohol induced liver injury, consequent on the peroxidation of lipids including Phosphatidylcholine (PC) the backbone of cell membrane structure (Table [5]). Replenishment of PC by administration of polyethyline phosphatidylcholine can prevent lesions in the baboon animal model of alcoholic liver injury (Table [5]). There is also evidence in a pilot study of this agent in subjects with alcoholic liver injury that the plasma markers of oxidative stress were reduced (Table [6]). We will have to await the results of a large multi-centre control trial involving 800 subjects which has been carried out and the analysis of which is currently ongoing. The place of such anti-oxidant therapy, particularly in the early treatment phase also needs to be assessed in relation to the concomitant or sequential use of anti TNF -α.

Table 5 Correction of lipid peroxidation and prevention of fibrosis in experimental liver injury Background: Peroxidation of lipids, including phosphatidylcholine (PC) - the backbone of cell membrane structure - is an important consequence of oxidative stress produced by acetaldehyde induced generation of superoxide as well as hydroxy and other free radicals together with impairment of antioxidant systems. Finding: Replenishment of PC by administration of PPC (polyenylphosphatidylcholine) prevented septal fibrosis and cirrhosis in baboon animal model of alcoholic liver disease with associated reduction in number of activated stellate cells (from work of Lieber et al.)

Table 6 Polyenylphosphatidylcholine (PPC) opposes alcohol-induced oxidative stress and hepatic fibrosis in man: Preliminiary Results - 18 subjects with significant liver injury including some fibrosis but not complete cirrhosis. - PPC given in dose of 3 tablets daily (1.5 grams) or placebo-random allocation - for 2 years. - Liver fibrosis - no change or regression in 9 given PPC compared with progression in 5 of 9 placebo treated. - Levels of 3 plasma markers of oxidative stress - malondialdehyde, free F2-isoprostanes, Total Peroxyl Radical-Trapping Potential - significantly reduced by PPC. (from Lieber et al, 2 000)

In those patients in whom efforts to reverse the acute phase of cell injury have been initiated, the use of MARS may have a further important place in providing additional time for the processes of hepatic regeneration and recovery (Table [7]). In our initial clinical experience, as will be reported later in the programme, substantial reductions of serum bilirubin level were observed with improvements in encephalopathy and general clinical state. However, the value of such a two pronged approach can only be assessed in properly conducted multi-centre controlled clinical trials.

Table 7 Treatment of acute alcoholic hepatitis a two-prong approach 1. Control of initial inflammatory/cytokine dependent processes (? by anti-TNF-α) to limit damage to hepatocytes and supporting framework and thereby prevent clinical progression. 2. Use of liver support device (MARS) for correction of biochemical disturbances and consequences of hepatic decompensation, to gain extra time for regeneration/recovery processes.

Selected References

• 1 Imperiale T F, O’Connor J B, McCullough A J. Corticosteroids are effective in patients with severe alcoholic hepatitis.  Am J Gastroenterol. 1999 Oct;  94 3066-3068 (10)
• 2 Christensen E, Gluud C. Glucocorticosteroids are not effective in alcoholic hepatitis.  Am J Gastroenterol. 1999 Oct;  94 3065-3066 (10)
• 3 Cabré E, Rodriguez-Iglesias P, Caballéria J, Quer J C, Sanchez-Lombraña J L, Parés A, Papo M, Planas R, Gassull M A. on behalf of The Spanish Group for the Study of Alcoholic Hepatitis . Short and Long-Term Outcome of Severe Alcohol-Induced Hepatitis Treated with Steroids or Enteral Nutrition: A Multicenter Randomized Trial.  Hepatology. 2000;  Vol 32 (No 1)
• 4 Enomoto N, Ikejima K, Bradford B U, Rivera C A, Kono H, Goto M, Yamashina S, Schemmer P, Kitamura T, Oide H, Takei Y, Hirose M, Shimizu H, Miyazaki A, Brenner D, Sato N, Thiurman R G. Role of Kupffer cells and gut-derived endotoxins in alcoholic liver injury.  Journal of Gastroenterology and Hepatology. (2000);  15 (Suppl.) D20-D25
• 5 Hanck C, Glatzel M, Singer M V, Rossol S. Gene expression of TNF-receptors in peripheral blood mononuclear cells of patients with alcoholic cirrhosis.  Journal of Hepatology. 2000;  32 51-57
• 6 Lieber S C. Prevention and Treatment of Liver Fibrosis Based on Pathogenesis. Alcoholism: Clinical and Experimental Research May 1999 23
• 7 Iimuro Y, Gallucci R M, Luster M I, Kono H, Thurman R G. Antibodies to Tumor Necrosis Factor Alfa Attenuate Hepatic Necrosis and Inflammation Caused by Chronic Exposure to Ethanol in the Rat.  Hepatology. 1997;  Vol. 26 (No. 6)
• 8 Yin M, Wheeler M D, Kono H, Bradford B U, Gallucci R M, Luster M I, Thurman R G. Essential Role of Tumor Necrosis Factor α in Alcohol-Induced Liver Injury in Mice.  Gastroentrology. 1999;  117 942-952