Acute liver failure: early supportive medical management for the liver is better than delayed intensive care for the brain

Abstract

Acute liver failure (ALF) is a complex syndrome reflecting a rapidly progressing, life-threatening but potentially reversible, deterioration of liver function in patients recently free of any overt chronic liver disease. Since the early 1990s, a changing epidemiology of ALF has emerged. In Western countries, the incidence of viral ALF decreased [[1]Ostapowicz G. Fontana R.J. Schiodt F.V. Larson A. Davern T.J. Han S.H.B. et al.Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States.Ann Int Med. 2002; 137: 947-954Crossref PubMed Scopus (1623) Google Scholar] whereas that of drug-induced, especially acetaminophen-induced, ALF increased [1Ostapowicz G. Fontana R.J. Schiodt F.V. Larson A. Davern T.J. Han S.H.B. et al.Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States.Ann Int Med. 2002; 137: 947-954Crossref PubMed Scopus (1623) Google Scholar, 2Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar]. In Asian countries, although widespread use of vaccination against hepatitis B virus (HBV) was associated with a highly significant decrease in HBV-induced ALF in children [[3]Kao J.H. Hsu H.M. Shau W.Y. Chang M.H. Chen D.S. Universal hepatitis B vaccination and the decreased mortality from fulminant hepatitis in infants in Taiwan.J Pediatr. 2001; 139: 349-352Abstract Full Text Full Text PDF PubMed Scopus (93) Google Scholar], ‘flares’ in adults with chronic HBV infection [[4]Cheng V.C.C. Lo C.-. Lau G.K.K. Current issues and treatment of fulminant hepatic failure including transplantation in Hong-Kong and the far east.Sem Liv Dis. 2003; 23: 239-250Crossref PubMed Scopus (22) Google Scholar] and acute hepatitis E, especially in India [[5]Acharya S.K. Dasarathy S. Kumer T.L. Sushma S. Prasanna K.S. Tandon A. et al.Fulminant hepatitis in a tropical population: clinical course, cause, and early predictors of outcome.Hepatology. 1996; 23: 1448-1455Crossref PubMed Google Scholar], are the main causes of ALF. According to the cause, the overall mortality rate ranges from 40 to 100% [[1]Ostapowicz G. Fontana R.J. Schiodt F.V. Larson A. Davern T.J. Han S.H.B. et al.Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States.Ann Int Med. 2002; 137: 947-954Crossref PubMed Scopus (1623) Google Scholar] and, thus, prevention of ALF [2Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 3Kao J.H. Hsu H.M. Shau W.Y. Chang M.H. Chen D.S. Universal hepatitis B vaccination and the decreased mortality from fulminant hepatitis in infants in Taiwan.J Pediatr. 2001; 139: 349-352Abstract Full Text Full Text PDF PubMed Scopus (93) Google Scholar, 6Bernuau J. Durand F. L'insuffisance hépatique fulminante et subfulminante, urgence de prévention.Gastroenterol Clin Biol. 1997; 21: 387-390PubMed Google Scholar], so often an iatrogenic complication [1Ostapowicz G. Fontana R.J. Schiodt F.V. Larson A. Davern T.J. Han S.H.B. et al.Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States.Ann Int Med. 2002; 137: 947-954Crossref PubMed Scopus (1623) Google Scholar, 2Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar], should become a key preoccupation of the medical community. In the last 20 years, emergency liver transplantation (ELT) emerged as the ‘only therapeutic intervention of proven benefit for patients with advanced ALF’ [[7]Bernal W. Wendon J. Liver transplantation in adults with acute liver failure.J Hepatol. 2004; 40: 192-197Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar]. However, without any doubt, some patients underwent unnecessary transplantation [6Bernuau J. Durand F. L'insuffisance hépatique fulminante et subfulminante, urgence de prévention.Gastroenterol Clin Biol. 1997; 21: 387-390PubMed Google Scholar, 7Bernal W. Wendon J. Liver transplantation in adults with acute liver failure.J Hepatol. 2004; 40: 192-197Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar] and survival after ELT—55% after 10 years in Europe [[8]Adam R. McMaster P. O'Grady J.G. Castaing D. Klempnauer J.L. Jamieson N. et al.Evolution of liver transplantation in Europe: report of the European liver transplant registry.Liver Transpl. 2003; 9: 1231-1243Crossref PubMed Scopus (484) Google Scholar]—is much more pejorative than spontaneous survival. So, optimized efficacy of the medical management, both cause specific and not specific, of patients with, or—even better—threatened by, ALF is clearly beneficial when it allows to save the patient's liver. Supportive, non-specific, medical management of patients with established ALF (i.e. with clinical encephalopathy) was frequently reviewed [2Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 9Carceni P. Van Thiel D.H. Acute liver failure.Lancet. 1995; 345: 163-169Crossref PubMed Scopus (106) Google Scholar, 10Ellis A. Wendon J. Circulatory, respiratory, cerebral and renal derangements in acute liver failure: pathophysiology and management.Sem Liv Dis. 1996; 16: 379-388Crossref PubMed Scopus (134) Google Scholar, 11Mas A. Rodés J. Fulminant hepatic failure.Lancet. 1997; 349: 1081-1085Abstract Full Text Full Text PDF PubMed Scopus (113) Google Scholar]. This highly complex management includes treatment and, when possible, prevention of multiple organ failure and bacterial and fungal superinfections, but its most critical target at late stages of encephalopathy is the life-threatening intracranial hypertension due to brain edema [12Blei A. Medical therapy of brain edema in fulminant hepatic failure.Hepatology. 2000; 32: 666-669Crossref PubMed Scopus (34) Google Scholar, 13Larsen F.S. Wendon J. Brain edema in liver failure: basic physiologic principles and management.Liver Transpl. 2002; 8: 983-989Crossref PubMed Scopus (70) Google Scholar]. A recent controlled trial suggested that phenytoin may be effective as a prophylaxis of subclinical seizure activity [[14]Ellis A.J. Wendon J. Williams R. Subclinical seizure activity and prophylactic phenytoin infusion in acute liver failure: a controlled trial.Hepatology. 2000; 32: 536-541Crossref PubMed Scopus (124) Google Scholar]. In this issue of the Journal, Bhatia et al. report the results of a controlled trial of phenytoin as prophylaxis of seizures in patients with ALF and stage III–IV encephalopathy [[15]Bhatia V. Batra Y. Acharya S.K. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure—a controlled clinical trial.J Hepatol. 2004; 41: 89-96Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar]. A controlled trial in the field of ALF, where so few are available [2Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 9Carceni P. Van Thiel D.H. Acute liver failure.Lancet. 1995; 345: 163-169Crossref PubMed Scopus (106) Google Scholar], is always a noteworthy achievement. Experienced hepatologists from New Delhi, after having stratified patients with established ALF according to their clinical severity at inclusion, demonstrated that IV phenytoin used at recommended dosages for preventing clinical seizures was ineffective: rates of seizures and severe cerebral edema as well as fatality rates were similar in patients receiving the antiepileptic drug and in those who did not [[15]Bhatia V. Batra Y. Acharya S.K. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure—a controlled clinical trial.J Hepatol. 2004; 41: 89-96Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar]. Since liver transplantation is not yet available in India, Bhatia et al.'s study offers the opportunity to reconsider many aspects, and pitfalls, of medical supportive care of patients with established ALF in order to optimize their survival. Several features of special interest of the Bhatia et al.'s study deserve further comments. First, the question could be raised whether the results of the Indian trial may be applied worldwide. Although etiology of ALF in the 42 patients included in the study was dominated by hepatitis E, a major cause of ALF in India [[5]Acharya S.K. Dasarathy S. Kumer T.L. Sushma S. Prasanna K.S. Tandon A. et al.Fulminant hepatitis in a tropical population: clinical course, cause, and early predictors of outcome.Hepatology. 1996; 23: 1448-1455Crossref PubMed Google Scholar], I confidently hypothesize that results would have been similar had the study been performed in a Western country where drug-induced hepatotoxicity is currently the leading cause of ALF [[1]Ostapowicz G. Fontana R.J. Schiodt F.V. Larson A. Davern T.J. Han S.H.B. et al.Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States.Ann Int Med. 2002; 137: 947-954Crossref PubMed Scopus (1623) Google Scholar]. Indeed, in patients with ALF and late stages of encephalopathy, the latter more than the cause of the acute liver disease is the main prognostic factor of short term outcome [[1]Ostapowicz G. Fontana R.J. Schiodt F.V. Larson A. Davern T.J. Han S.H.B. et al.Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States.Ann Int Med. 2002; 137: 947-954Crossref PubMed Scopus (1623) Google Scholar]. So, in my view, results of the Indian study may be applied worldwide in patients with ALF and stage III–IV encephalopathy. Second, among the 42 patients with a 74% rate of clinically significant cerebral edema, the overall rate of clinical seizures was 24% [[15]Bhatia V. Batra Y. Acharya S.K. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure—a controlled clinical trial.J Hepatol. 2004; 41: 89-96Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar]. The uncommon occurrence of seizures in patients with ALF, previously pointed out [[12]Blei A. Medical therapy of brain edema in fulminant hepatic failure.Hepatology. 2000; 32: 666-669Crossref PubMed Scopus (34) Google Scholar], fits well our own clinical experience. The 24% rate in the Bhatia et al.'s study shows that severe cerebral edema by itself is not the only requirement for the occurrence of seizures. The latter could simply reflect the intrinsic aggravation of cerebral edema due to the continuing downhill course of ALF. However, since specificity of symptoms is always very low in patients with ALF, it is not unjustified to question whether seizures could be, at least in part, related to additional factors concomitant to ALF, namely acute renal failure and the use of potentially neurotoxic drugs. Bhatia et al. adequately defined number of expressions and variables used in their study but, on the only basis of a normal result, 0.7 mg/dl, of serum creatinine, concluded that ‘no patient developed renal failure’ [[15]Bhatia V. Batra Y. Acharya S.K. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure—a controlled clinical trial.J Hepatol. 2004; 41: 89-96Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar]. Several arguments make me reluctant to support this statement. Normal serum creatinine may coexist with acute renal failure. In patients with well defined ALF, the incidence of acute renal failure (a) is higher in those with, than in those without, severe cerebral edema, (b) was 30% in cases not due to acetaminophen hepatotoxicity in United Kingdom [[2]Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar] and (c) 38% among 50 consecutive Indian patients with an increased frequency in those with jaundice [[16]Jain S. Pendyala P. Varma S. Sharma N. Joshi K. Chawla Y. Effect of renal dysfunction in fulminant hepatic failure.Trop Gastroenterol. 2000; 21: 118-120PubMed Google Scholar]. Finally, in patients managed by Bhatia et al., mean serum bilirubin level was 16.5 mg/dl (i.e. 280 μmol/l) [[15]Bhatia V. Batra Y. Acharya S.K. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure—a controlled clinical trial.J Hepatol. 2004; 41: 89-96Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar]. This key result must lead to be extremely cautious regarding a normal value of serum creatinine because of the negative interference between bilirubin-derived biliverdin and serum creatinine when the latter is assayed, at high pH, by a Jaffé reaction derived technic [17Hawker F. The liver. WB Saunders, London1993Google Scholar, 18Bernuau J. Benhamou J.P. Fulminant and subfulminant hepatic failure.in: Bircher J. Benhamou J.P. McIntyre N. Rizzetto M. Rodés J. Oxford textbook of clinical hepatology. Oxford University Press, Oxford1999: 1341-1372Google Scholar]. This methodological drawback was confirmed by a recent study using enzymatic creatinine [[19]Bzowej N.H. Huilgol V.R. Sebane H. Bonacini M. Rabkin J. Doo E. et al.MELD underestimated unless enzymatic creatinine used.J Hepatol. 2004; 40 ([abstract]): 40Abstract Full Text PDF PubMed Google Scholar]. Therefore, I hypothesize that a significant number of patients included in the Bhatia et al.'s study did actually develop acute renal failure. Thus, according to this hypothesis, it results that using ranitidine, a potentially neurotoxic anti-H2 receptor antagonist with delayed clearance in patients with severe cirrhosis [[20]Gonzalez-Martin G. Paulos C. Veloso B. Chesta J. Novoa X. Arancibia A. Ranitidine disposition in severe hepatic cirrhosis.Int J Clin Pharmacol Ther Toxicol. 1987; 25: 139-142PubMed Google Scholar], exposes patients with ALF complicated by severe cerebral edema and acute renal failure—a condition comparable to severe cirrhosis—to an unpredictable risk of neurotoxicity with seizures. This, admittedly controversial, hypothesis could have been confirmed, or denied, by plasma drug assays but can hardly be denied only on clinical grounds in so critically ill patients. Pharmacokinetics of drugs is profoundly altered in patients with both severe hepatic and renal dysfunction [[21]Power B.M. Forbes A.M. van Heerden P.V. Ilett K.F. Pharmacokinetics of drugs in critically ill adults.Clin Pharmacokinet. 1998; 34: 25-56Crossref PubMed Scopus (164) Google Scholar] and cerebral aggravation in patients with ALF may be due to drugs associated with fitting [[10]Ellis A. Wendon J. Circulatory, respiratory, cerebral and renal derangements in acute liver failure: pathophysiology and management.Sem Liv Dis. 1996; 16: 379-388Crossref PubMed Scopus (134) Google Scholar] and also to terlipressin [[22]Shawcross D.I. Davies N.A. Mookerjee R.P. Hayes P.C. Williams R. Lee A. et al.Worseninng of cerebral hyperemia by the administration of terlipressin in acute liver failure with severe encephalopathy.Hepatology. 2004; 39: 471-475Crossref PubMed Scopus (91) Google Scholar]. Third, Bhatia et al. provided three other critical data which may share some relationship with the occurrence of cerebral edema and seizures among their 42 patients with ALF. Mean arterial pressure (MAP) was low (<60 mmHg). Since, in patients with stage III–IV encephalopathy, cerebral blood flow closely depends on systemic arterial pressure, this low level of MAP suggests that, at least in those of the Indian patients with the lowest values, some degree of cerebral hypoperfusion might have been a cofactor of seizures. Half the patients of the trial had significant hyponatremia, ranging between 115 and 135 μmol/l. Although natremia levels of patients who developed seizures and those of patients who did not were not compared, hyponatremia could have been another cofactor of seizures in these patients. Hyponatremia may contribute to the development, or the aggravation, of cerebral edema and a recent, noteworthy, controlled trial at King's (London) showed that hypertonic sodium chloride administration allows a significant decrease in intracranial pressure [[23]Murphy N. Auzinger G. Bernal W. Wendon J. The effect of hypertonic sodium chloride on intracranial pressure in patients with acute liver failure.Hepatology. 2004; 39: 464-470Crossref PubMed Scopus (267) Google Scholar]. Finally, the high mean hyperglycemia, 243 mg/dl, in addition to its unfavourable prognostic value in critically ill patients [[24]Van den Berghe G. Wouters P. Weekers F. Verwaest C. Bruyninckx F. Schetz M. et al.Intensive insulin therapy in critically ill patients.N Engl J Med. 2001; 345: 1359-1367Crossref PubMed Scopus (7939) Google Scholar], may also result in increased intracranial pressure [[25]Kodakar S.K. Gopal P.B. Wendon J.A. Hyperglycaemia is associated with intracranial hypertension in patients with acute liver failure.Liver Transpl. 2001; 7 ([abstract]): C21Google Scholar]. Despite so many possible cofactors of seizures, the overall 31% survival rate reported by Bhatia et al. is within the expected range [[18]Bernuau J. Benhamou J.P. Fulminant and subfulminant hepatic failure.in: Bircher J. Benhamou J.P. McIntyre N. Rizzetto M. Rodés J. Oxford textbook of clinical hepatology. Oxford University Press, Oxford1999: 1341-1372Google Scholar]. The 42 patients are mainly younger than 30, but none of them had ALF due to acetaminophen, a cause frequently associated with increased survival rates [1Ostapowicz G. Fontana R.J. Schiodt F.V. Larson A. Davern T.J. Han S.H.B. et al.Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States.Ann Int Med. 2002; 137: 947-954Crossref PubMed Scopus (1623) Google Scholar, 2Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar]. Finally, what is the most important message delivered by Bhatia et al. regarding patients with ALF? In my view, it is not the recommendation for not using phenytoin in patients with stage III–IV encephalopathy, but the vigorous emphasis that their patients ‘presented too late to the hospital’ [[15]Bhatia V. Batra Y. Acharya S.K. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure—a controlled clinical trial.J Hepatol. 2004; 41: 89-96Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar]. Although repeatedly denounced for decades, the ‘too late’ referal to liver units, or at least to a hepatogastroenterologist (even by a phone call), of patients affected by acute liver disease and severe coagulopathy (prothrombin ratio <50% of normal) remains a key factor of the development of established ALF [2Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 18Bernuau J. Benhamou J.P. Fulminant and subfulminant hepatic failure.in: Bircher J. Benhamou J.P. McIntyre N. Rizzetto M. Rodés J. Oxford textbook of clinical hepatology. Oxford University Press, Oxford1999: 1341-1372Google Scholar]. This universal message, unfortunately, is still true all around the world including countries where ELT is available [[1]Ostapowicz G. Fontana R.J. Schiodt F.V. Larson A. Davern T.J. Han S.H.B. et al.Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States.Ann Int Med. 2002; 137: 947-954Crossref PubMed Scopus (1623) Google Scholar]. It is crucial to understand why delayed hepatologic evaluation of these patients may be so deleterious. At least three processes may be involved. First, the etiological diagnosis of the acute liver disease, especially that of the cases with an unusual clinical profile (such as non-intentional acetaminophen-induced hepatotoxicity) or due to a relatively uncommon cause, is too often delayed. Second, and consequently, specific treatments targeting the liver—the key organ to be preserved and supported in patients with acute liver disease still without clinical encephalopathy but with severe coagulopathy and thus threatened by ALF—may also be delayed. Whereas these specific treatments have their full efficacy when liver tissue lesions are still (relatively) limited, their delayed application allows the spread of liver destruction. Classical examples of this issue include all the treatable acute liver diseases [2Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 18Bernuau J. Benhamou J.P. Fulminant and subfulminant hepatic failure.in: Bircher J. Benhamou J.P. McIntyre N. Rizzetto M. Rodés J. Oxford textbook of clinical hepatology. Oxford University Press, Oxford1999: 1341-1372Google Scholar] such as acetaminophen-induced hepatotoxicity [1Ostapowicz G. Fontana R.J. Schiodt F.V. Larson A. Davern T.J. Han S.H.B. et al.Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States.Ann Int Med. 2002; 137: 947-954Crossref PubMed Scopus (1623) Google Scholar, 2Riordan S.M. Williams R. Fulminant hepatic failure.Clin Liv Dis. 2000; 4: 25-45Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar], herpes simplex hepatitis, acute fatty liver of pregnancy, auto-immune hepatitis, hypoxic hepatitis, Budd-Chiari syndrome and the acute form of Wilson's disease [[26]Durand F. Bernuau J. Giostra E. Mentha G. Shouval D. Degott C. et al.Wilson's disease with severe hepatic insufficiency: beneficial effects of early administration of d-penicillamine.Gut. 2001; 48: 849-852Crossref PubMed Scopus (80) Google Scholar]. Third, and most important, there is a risk of unappreciated aggravation of either the destruction of liver tissue or the patient's condition. A key factor of such an aggravation, often associated with both delayed diagnosis and the lack of specific treatment, is the undiscriminating use of various hepatotoxic, nephrotoxic or neurotoxic substances when the first therapeutic law in patients with acute liver diseases, ‘stop any drug’, is not followed [[18]Bernuau J. Benhamou J.P. Fulminant and subfulminant hepatic failure.in: Bircher J. Benhamou J.P. McIntyre N. Rizzetto M. Rodés J. Oxford textbook of clinical hepatology. Oxford University Press, Oxford1999: 1341-1372Google Scholar]. For many years, this concept of deleterious ‘aggravating cofactors’ is consistent with the clinical experience of the author in many patients with ALF associated with acute viral hepatitis due to hepatitis A, or B (or B and D), or C, and also with dengue fever, herpes simplex hepatitis, varicella and isoniazid-induced hepatitis. The hypothesis is now supported by recent reports of recent (prior to encephalopathy) ingestion of acetaminophen in patients with ALF complicating acute viral hepatitis type B in the United States [27Garfein R.S. Bower W.A. Hutin Y.J.F. Jawanda J. Groom A.V. Murphy J.S. et al.Factors associated with fulminant liver failure among injection drug users with acute hepatitis B.Hepatology. 1999; 30: 478AGoogle Scholar, 28Polson J. Ocama P. Larson A.M. Hynan L. Lalani E.K. Harrison M.E. et al.Role of acetaminophen in acute liver failure due to viral hepatitis.Hepatology. 2003; 38: 544ACrossref Google Scholar] or type A in France [[29]Rezende G. Roque-Alfonso A.M. Samuel D. Gigou M. Nicand E. Ferre V. et al.Viral and clinical factors associated with the fulminant course of hepatitis A infection.Hepatology. 2003; 38: 613-618Crossref PubMed Scopus (143) Google Scholar]. Inadequate consumption of acetaminophen could also explain life-threatening cases of ALF having complicated parvovirus B19 infection, a febrile illness, in Belgian children [[30]Bernuau J. Durand F. Valla D. Parvovirus B19 infection and fulminant hepatitis.Lancet. 1999; 353 ([letter]): 754-755Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar]. Aggravating cofactors may be extra-hepatic as well. As an example, admission hyponatremia, a frequent finding in patients with ALF in New Delhi [[15]Bhatia V. Batra Y. Acharya S.K. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure—a controlled clinical trial.J Hepatol. 2004; 41: 89-96Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar], but also in London [[23]Murphy N. Auzinger G. Bernal W. Wendon J. The effect of hypertonic sodium chloride on intracranial pressure in patients with acute liver failure.Hepatology. 2004; 39: 464-470Crossref PubMed Scopus (267) Google Scholar] and Clichy (personal experience), usually attests the previously unappreciated development of acute renal failure associated with inadequate low sodium rehydratation. So it is quite likely that early referal to hepatologists of patients with acute liver disease and severe coagulopathy (prothrombin ratio <50%) could actually reduce, even eliminate, the sometimes multiple aggravating cofactors that precipitate the patient in the life-threatening condition of established ALF. So the final question is: why referal to liver units (or at least an early phone call to a hepatologist) is delayed in so many patients with acute liver disease, severe coagulopathy and no clinical encephalopathy? This issue is fully pertinent in this editorial since, as mentioned above, Bhatia et al. clearly stressed that their patients ‘presented too late to the hospital’ [[15]Bhatia V. Batra Y. Acharya S.K. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure—a controlled clinical trial.J Hepatol. 2004; 41: 89-96Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar]. My view is that hepatologists, and still more non-hepatologists, are confined within the classical, but too rigid, Trey and Davidson's definition of ALF in which clinical encephalopathy is both an absolute requirement for the diagnosis of ALF and a marker of the catastrophic prognosis most often associated to it [[31]Trey C. Davidson C. The management of fulminant hepatic failure.in: Popper H. Schaffner F. Progress in liver diseases. Grune and Stratton, New York1970: 282-298Google Scholar]. For many years, Benhamou and the Beaujon's group do think that ‘the liver fails before encephalopathy’ [[32]Bernuau J. Benhamou J.P. Classifying acute liver failure.Lancet. 1993; 342: 252-253Abstract PubMed Scopus (48) Google Scholar] as attested by the early deterioration in coagulation factors activities and, often but inconstantly, by early jaundice. Encephalopathy should be considered as a late symptom of ALF (J. Wendon, personal communication) and the concept of ALF without encephalopathy was used by several authors [26Durand F. Bernuau J. Giostra E. Mentha G. Shouval D. Degott C. et al.Wilson's disease with severe hepatic insufficiency: beneficial effects of early administration of d-penicillamine.Gut. 2001; 48: 849-852Crossref PubMed Scopus (80) Google Scholar, 33Case Record of the Massachusetts General HospitalCase 6-2001.N Engl J Med. 2001; 344: 591-599Crossref PubMed Scopus (12) Google Scholar, 34Bernal W. Donaldson N. Wyncoll D. Wendon J. Blood lactate as an early predictor of outcome in paracetamol-induced acute liver failure: a cohort study.Lancet. 2002; 359: 558-563Abstract Full Text Full Text PDF PubMed Scopus (368) Google Scholar, 35Brandsoeter B. Höckersdtedt K. Friman S. Ericzon B.-G. Kirkegaard P. Isoniemi H. et al.Fulminant hepatic failure: outcome after listing for highly urgent liver transplantation-12 years experience in the Nordic countries.Liver Transpl. 2002; 8: 1005-1062Google Scholar, 36Bernuau J. Durand F. Belghiti J. Acute hepatic failure: a French pespective.in: Arroyo V. Bosch J. Bruix J. Ginés P. Navasa M. Rodés J. Therapy in hepatology. Ars Medica, Barcelona2001: 149-157Google Scholar]. So, practically speaking, hepatologists should teach repeatedly to non-hepatologists not to wait for even ‘early symptoms’ of encephalopathy prior to refer patients with acute liver disease and severe coagulopathy to a hepatologist: waiting for encephalopathy would lead to unacceptably delayed referal, as prospectively observed by Bhatia et al. [[15]Bhatia V. Batra Y. Acharya S.K. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure—a controlled clinical trial.J Hepatol. 2004; 41: 89-96Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar]! Any physician will agree that, in these patients who are threatened by established ALF, both avoidance of deleterious cofactors and early specific medical therapy for the liver must be prefered to late intensive care for the brain.