Secondary Logo

Journal Logo


Hepatic injury associated with halogenated anaesthetics: cross-sensitization and its clinical implications

Mikatti, N. El.; Healy, T. E. J.

Author Information
European Journal of Anaesthesiology: January 1997 - Volume 14 - Issue 1 - p 7-14
  • Free


Acceptance of hepatic injury from halogenated anaesthetics

Hepatic injury was not associated with the use of the popular non-halogenated inhalational agents ether and cyclopropane. Their replacement by the halogenated agents was due in large part to the perceived unacceptability of the risks they presented of fire or explosion. Hepatic injury from halogenated inhalational agents was first recognized with chloroform, with more hesitation with halothane, and more hesitantly still with the newer agents as they have been introduced in succession. The early reluctance to accept the potential hepatotoxicity of the halogenated anaesthetic agents was partly because of their great effectiveness and safety in most other respects, partly because of the relative rarity of fulminant and fatal effects, and partly because acceptable alternatives were not yet available. The reluctance was relatively easy to justify because the diagnosis was clinical—primarily by exclusion—and it was always easy to find other plausible culprits to blame for post-operative hepatic dysfunction. Almost any type of surgical procedure can be followed by impairment of hepatic function to some degree, irrespective of which anaesthetic agents are used. It may be attributable to surgical trauma, hypoxia or shock [1], and may include jaundice amongst its manifestations. Jaundice itself may result simply from bilirubin overload associated with haemolysis, resorption of large haematomas and massive blood transfusions. Bacterial infection, adverse reactions to other drugs or hepatitis associated with blood transfusion may further obscure any link between hepatic injury and any halogenated anaesthetic agent that may have been used.

The first two cases of jaundice and death after chloroform occurred in 1847, the very year of its introduction into anaesthesia. In 1912, the Committee on Anaesthesia of the American Medical Association suggested that the use of chloroform should be stopped because of liver damage and cardiovascular collapse, but chloroform continued to be used until 1957. The renal and hepatic dysfunction caused by chloroform is now viewed as a classic example of toxicity caused by metabolic bioactivation, in this case by hepatic microsomal cytochrome P-450 to a highly reactive metabolite, phosgene (COCl2). The mechanisms of cellular toxicity of phosgene are likely to be complex and have not been defined in detail [2].

The first reports that halothane might be hepatotoxic appeared 2 years after its introduction in 1956. Increasing concern led to the National halothane study in 1966. This was a retrospective analysis of about 850 000 anaesthetics between 1959 and 1962. It neither proved nor disproved a relation between hepatic necrosis and halothane hepatitis, but did support its safety as an inhalational anaesthetic agent. It tooks some years before halothane hepatitis (HH) was generally accepted as an entity—that acceptance being assisted by the development of animal models and the demonstration of mechanisms by which it may come about. The perception of greater risk from repeated exposure was established relatively early. Liver damage may occur in two distinct forms after halothane anaesthesia. The more common is a mild form that occurs soon after anaesthesia and may result from a reductive biotransformation: this may sometimes be accentuated by prior induction of the hepatic enzymes that mediate the biotransformation. It is quite plausible that, with this form and mechanism of hepatotoxicity, there might well be a period (of relatively short but indeterminate length) after a previous exposure in which further exposure to a halogenated agent would be especially dangerous. The less common fulminant form is associated with repeated administration of halothane, appears to be immune-mediated [3], and can be marked by the detection of antibodies to adducts between trifluoroacetyl halothane metabolites and hepatic cellular components: its likelihood and severity may also be increased by induction of the appropriate enzymes for oxidative biotransformation. With this form and mechanism of hepatotoxicity, it is reasonable to expect the added risk from a repeat exposure to be lifelong.

The biotransformation of halothane is a central feature of both of these forms of halothane hepatitis. A number of other mechanisms have been suggested and investigated that may be adjuvants in recognized forms of the condition, or causes (in their own right) of separate forms of the disease. The acceptance and study of halothane hepatitis was vitally important in setting the stage for the introduction of the halogenated ethers enflurane and isoflurane. The perceived importance of biotransformation of halothane in the aetiology of halothane hepatitis provoked the expectation that agents that were less subject to biotransformation should be less likely to cause hepatitis.

Enflurane, which is metabolized far less than halothane, was introduced in the early 1970s and seemed to produce little hepatic injury. Injury was difficult to demonstrate experimentally and the mechanism underlying the relatively few reported cases of hepatic injury was hotly debated [4]. Lewis et al.[5] reported enflurane hepato-toxicity in 24 patients, five of whom died. They blamed metabolic idiosyncrasy but also concluded that previous exposure to halogenated anaesthetics might increase the potential hepatotoxicity of enflurane because of cross-sensitization. Eger et al.[6] challenged this and other reports, concluding that the evidence did not support the existence of a syndrome of 'enflurane hepatitis'. The reports of hepatic dysfunction after repeated exposure to enflurane anaesthesia are controversial. Allen & Downing [7] failed to demonstrate changes in the plasma concentrations of hepatic enzymes in a group of 49 women who received enflurane on up to three occasions, but Dundee et al.[8] reported an increase in enzyme concentrations in almost 25% of those who received enflurane on between two and five occasions.

Paull and Fortune [9] reported a case of fatal hepatotoxicity after two enflurane anaesthetics and presented clinical and histological evidence implicating enflurane as the causative agent.

Hepatotoxicity after isoflurane anaesthesia remains a controversial issue, and many fewer case reports have been published than after enflurane despite the advent of reportedly more sensitive tests for hepatic damage, such as the plasma concentrations of glutathione S-transferase (GST) [10,11]. Isoflurane unlike halothane and enflurane was not followed by a significant increase in GST. A recent Food and Drug Administration study reviewed 45 suspected cases of isoflurane hepatotoxicity reported between 1981 and 1984: they found 36% of these to be 'possibly related' to isoflurane but concluded that the current evidence did not suggest a reasonable likelihood of an association between the use of isoflurane and the occurrence of post-operative hepatic dysfunction [12]. Retrospective, non-randomized, epidemiological studies are at best difficult to interpret. Analysis of a small number of cases, without a mechanistic hypothesis, minimizes the probability of drawing meaningful conclusions [13]. Other cases have been reported of hepatotoxicity after isoflurane, though without a definitive causal relation between isoflurane and hepatic dysfunction [14–16]. One patient has undergone successful orthotopic liver transplantation following fulminant hepatic failure after three consecutive, closely spaced, isoflurane anaesthetics for paranasal sinus surgery [17]. There were many similarities with halothane-induced hepatic necrosis, including the histological findings in the resected liver. These strongly supported a link between isoflurane and the patient's hepatic failure. Brown and Frink [18] reported that, though hepatotoxicity following isoflurane is rare, it may nonetheless occur. They also reported that serum trifluoroacetyl antibodies had been identified in one unreported case of hepatic necrosis after three exposures to isoflurane. Although the likelihood of antibody detection is generally small (in keeping with the reduced metabolism of isoflurane compared with halothane) the titre in the reported case was high enough to suggest that the reaction could have been as a result of isoflurane.

Goldfarb et al.[19] reported that halothane may induce hepatic ultrastructural changes in the form of significantly increased hepatocyte lysosomes. Similar changes have not been seen in liver biopsies taken 1 h after induction with isoflurane, though they might well be observable under different conditions, namely administration of isoflurane that is sufficiently prolonged to allow even reduced rates of metabolism to generate threshold amounts of biotransformation products.

The introduction of the newer agents, sevoflurane and desflurane (which are even more minimally biotransformed) has probably been too recent to allow judgement on their propensity to cause hepatitis, and present controversies still centre mainly on results of experimental studies in animals.

The mechanisms of hepatotoxicity

Halothane as a model

The mechanisms underlying halothane hepatitis are not fully understood in spite of having been intensively studied, but just as the acceptance of the entity (or entities) of halothane hepatitis was promoted by the experimental demonstration of mechanisms by which halothane might damage the liver, so the acceptance of the potential hepatotoxicity of the newer agents has been heavily based on demonstrations of damaging mechanisms analogous to those established for halothane hepatitis. The behaviour of the newer agents in experimental models has therefore usually been compared with the behaviour of halothane.

Indirect mechanisms. There seems to be a genetic susceptibility to halothane hepatitis though the underlying mechanisms are obscure. There is also known to be a relation between hepatic hypoxia and halothane hepatitis [3], though the hypoxia itself may be more directly relevant than any part played by reductive biotransformation of halothane [20]. Because all anaesthetic agents affect systemic and regional haemodynamics, all are potentially capable of damaging the liver indirectly by compromising its blood flow and causing ischaemia. Isoflurane and to a lesser extent enflurane facilitate oxygen delivery to the liver much better than halothane, mainly by more effective preservation of hepatic arterial blood flow [3]. Gelman and Van Dyke [21] have proposed that hepatic damage occurs as a consequence of perturbations in the mechanisms that maintain cellular calcium homeostasis. However, these perturbations occur in the final common pathways of many types of cellular damage, and could well be non-specific markers rather than the causes of the damage. The links, if any, have yet to be established between calcium balance and hepatic damage after any volatile anaesthetic agent.

More direct drug-related mechanisms. Extensive experimentation in animal models has shown that halothane has a direct hepatotoxic potential, but the relevance of this to human patients has not yet been confirmed [22]. It seems more likely that its clinically important, agent-specific toxicity is less direct, via at least two types of biotransformation [23]. There are reductive pathways that are favoured in hypoxic conditions, and which may be induced by the generally recognized inducers of hepatic enzyme systems (ethanol, isoniazid, phenobarbitone). These may yield free radicals of the type produced from carbon-tetrachloride, which cause zone three necrosis of only minor degree unless enhanced by an immunological response. There are also oxidative pathways via the cytochrome P450 system which are favoured when oxygen tensions are high: these may yield reactive trifluroacetyl (TFA) substances that combine covalently with cellular elements of the hepatocyte to create neoantigens that may excite autoimmune antibody responses. These are discussed in more detail later in relation to cross-sensitivity between halogenated agents.

Hepatotoxic mechanisms with the newer agents

Hepatic ischaemia. Enflurane and isoflurane both decrease portal blood flow and oxygen delivery, albeit less than does halothane [24]. However, neither enflurane nor isoflurane undergo reductive metabolism that might produce toxic free radicals, so that even if hepatic ischaemia were to occur, any resulting damage would have to be attributed to some other mechanism. The two most recent agents, sevoflurane and desflurane, have attracted attention as alternatives to those presently used. Sevoflurane, it is suggested, is not hepatotoxic [25,26], but studies using rats with hepatic microsomal induction did not support this view [27], though the directness of the hepatotoxicity remains in question. It is pertinent that, in the presence of hypoxia and hepatic enzyme induction, sevoflurane produced hepatic injury indistinguishable from that following isoflurane. In guinea pigs this is typical of ischaemic or low oxygen injury [28]. Recently it was shown that sevoflurane is biotransformed (3–5%) to a lesser degree than halothane. The organic metabolite of sevoflurane is not TFA, but a unique compound, hexafluoroisopropanol (HFIP). This is bound less to protein and does not accumulate, instead undergoing rapid phase II biotransformation to HFIP-glucuronide which is rapidly excreted in the urine [29]. In contrast to halothane, sevoflurane and desflurane did not show any increase in covalently bound fluorine over control (unaesthetized) values. These results suggest that the ability of free and conjugated HFIP to bind to liver macromolecules may be low enough to preclude the initiation of an immune-mediated response [30]. Current data therefore suggest that the hepatotoxic potential of sevoflurane is quite low [29].

Calcium changes. Iaizzo et al.[31] reported on the release of radio-labelled calcium from isolated rat hepatocytes pretreated with mitochondrial inhibitors to make it more likely that any observed release would be from endoplasmic reticulum. Though there were dose-related increases in calcium release with all three agents, halothane and enflurane gave similar increases in calcium release at the in vivo equivalent of one MAC, whereas isoflurane had little or no effect at all. This is at variance with the gradation in their clinical propensity to induce hepatic necrosis. Further studies are required to resolve the question of whether the elevated calcium is the cause or the effect of hepatic necrosis.

Biotransformation. Though neither enflurane nor isoflurane appear to undergo reductive metabolism, they do undergo oxidative metabolism (albeit less so than does halothane) by pathways mediated by cytochrome P450. About 2% of enflurane and 0.2% of isoflurane undergo metabolism compared with 20% of halothane. The oxidative metabolism of enflurane and isoflurane occurs predominantly in the liver [32] and might therefore lead to the formation of reactive acylating substances capable of forming covalently altered tissue adducts similar to those produced by halothane. A trifluracetyl metabolite is produced from isoflurane though much less than from halothane [13], although Davidkova et al.[33] recently reported that there might be a greater degree of biotransformation of isoflurane than had previously been accepted. The formation of TFA-adducts would certainly be possible if the types and degrees of biotransformation reported by Gil et al.[34] are accompanied by hepatic binding.

Desflurane is structurally very similar to isoflurane, but has a fluorine instead of the lone chlorine atom present in isoflurane. Eger et al.[35] and Holmes et al.[36] reported that desflurane does not cause hepatic injury in animals. This could be because of both its lower blood/gas partition coefficient (which reduces the body burden and the time it is retained in the body) and to its higher resistance to physical degradation. Jones et al.[37] found no significant changes in liver and renal function in 10 healthy volunteers exposed to an average inspired concentration of 3.6% desflurane.

Hypersensitivity and cross-sensitization

There is considerable clinical and experimental evidence to suggest that at least one form of halothane hepatitis (HH) is an immune-mediated toxicity [38,39] with the accompanying manifestations of a drug hypersensitivity reaction. The sera from many patients with clinically diagnosed HH have been found to contain specific antibodies that react with halothane-induced liver antigens (neoantigens), whereas the sera of patients with other forms of hepatitis contain no such antibodies. It has recently been shown that the neoantigens are formed by the covalent interaction of the reactive oxidative TFA metabolite of halothane with at least five distinct classes of liver microsomal proteins (100, 76, 59, 57 and 54 kilo Daltons (kDa)) [40–42]. There is insufficient information to determine the dose of halothane that is required to sensitize or to trigger HH in sensitized patients but the amounts that contaminate anaesthetic machines increase with the usage of halothane [43] and may well be sufficient.

Recent efforts to determine the role of the immune system in halothane-induced liver injury have tended to concentrate on the humoral response, but a cellular immune response must also be considered [44]. A weak cellular immune response has been reported though it lacks the intensity and specificity of the humoral reaction [45].

Detection of antibodies

Two general types of immunochemical assay have been reported for the detection of the antibodies [46].

  1. Immunoblotting assay: This is laborious, time-consuming and less sensitive than other methods (being positive in only 62% of patients with a clinical diagnosis of HH).
  2. Enzyme-linked immunosorbent assay (ELISA): Antibodies were detected in a sera of 67% [47] and 72% [48] of patients diagnosed clinically as having HH. Another ELISA assay with TFA-rabbit serum albumin as the hapten (the TFA-RSA) gave positive responses from between 30% [49] and 83% [50] of patients with clinically diagnosed HH. The TFA-RSA assay is reportedly a quick and convenient way of confirming the diagnosis of halothane-associated hepatitis in fulminant hepatic failure, but is less sensitive when the illness follows a milder course; its specificity also varied depending upon the laboratory [49].

The sensitivity of the ELISA can be expected to improve significantly once all the TFA neoantigens are covered because not all patients may be sensitized to the currently included neoantigens that have been raised in animals (namely TFA-57 kDA, TFA-76 kDa, or TFA-100 kDA). Eventually, TFA neoantigens will be purified from human liver and should prove best for detecting sensitized patients [46]. Hastings et al.[51] reported that a combination of the TFA-RSA ELISA and TFA-lysing blocking ELISA gave a sensitivity of 92% in cases of presumptive HH.

Evidence of cross-sensitivity

The known oxidative metabolism of both enflurane and isoflurane by cytochrome P450 can produce immunoreactive, covalently bound acylated protein adducts similar to those implicated in the genesis of halothane-induced hepatic necrosis [52]. Microsomal adducts were detected in rat liver after administration of enflurane, isoflurane and halothane. The detection used specific anti-TFA IgG hapten antibodies. There were decreases in the order halothane >enflurane >isoflurane in the relative amounts of immunoreactive protein adducts formed, which correlates directly with the relative degree of metabolism of these agents. Although anti-TFA immunoreactivity was most evident after multiple doses of enflurane or isoflurane, appreciable adduct formation was observed after single exposures when more rigorous analytical conditions were used. The results of this work support the view that acyl metabolites of the volatile anaesthetics may become covalently bound to hepatic proteins and serve as antigens, and this may account for the apparent cross-sensitization and idiosyncratic hepatotoxicity reported with these drugs.

This view is supported by a study [53] in which immunoblotting techniques revealed that antibodies in the sera of six HH patients recognized liver microsomal antigens formed in rats treated with enflurane or halothane. These antigens were not detected in microsomes from rats treated with isoflurane or sesame oil, although isoflurane metabolism does produce very small amounts of covalently bound liver adducts in rats [52]. These data can support cautious speculation that the immunogenic potential for isoflurane should be relatively low, though not necessarily negligible, because the degree of haptenic alteration necessary to provoke an immune response in a sensitized individual is unknown. This anxiety is reinforced by a recent report of hepatic necrosis after isoflurane in a patient who had on previous occasions been anaesthetized with halothane and enflurane [16].

Hepatic enzyme induction. The quantities and types of hepatic neo-antigens produced by halogenated agents may be increased in patients exposed to known inducers of cytochrome P450. Pretreatment of rats with isoniazid or ethanol results in the expression of a novel isozyme of cytochrome P450 which has been shown to induce oxidative metabolism of the three halogenated volatile agents presently in common clinical use [54–56]. Experimentally produced diabetes is also associated with an increased rate of oxidative defluorination of enflurane. It induces the same cytochrome P450 isozyme as do ethanol and isoniazid, possibly by way of the ketone body acetone. Whether this isozyme is found in human liver and is susceptible to the same inducers remains to be demonstrated [57].

In summary, the most dangerous known form of hepatotoxicity from halogenated anaesthetics is related to the production of acyl (TFA) metabolites that may become covalently bound to hepatic proteins and become neoantigens. These TFA-adducts induce antibodies in the blood that can attack the liver. This accounts for the apparent cross-sensitization and idiosyncratic hepatotoxicity reported after these drugs.

Further development of assays for the detection of the antibodies against the TFA neoantigens will be clinically important for several reasons [46]:

  1. The presence of the antibodies in the sera of patients previously exposed to halothane will indicate prior sensitisation and will thus identify these patients as being at increased risk of developing a hypersensitivity reaction on further exposure to halothane.
  2. Because these same antibodies can react with liver microsomal neoantigens produced by the structurally related oxidative acyl-halide metabolite of enflurane (and possibly also isoflurane), their presence will warn of the potential cross-sensitization reaction after exposure to other halogenated agents [52,53].
  3. The assay can be used to monitor sensitization to halothane in animal studies of the immunological mechanism underlying halothane hepatitis.

Clinical implications

Repeated anaesthesia in susceptible patients. Because of anxieties over cross-sensitization and cross-reactivity, it was recommended in 1983 that neither halothane, methoxyflurane, nor enflurane be used in any patients with documented hepatic dysfunction after previous exposure to any one of them [5]. This view was supported in 1990 by Brown and Frink [18] who recommended that the anaesthetic regimen should probably exclude any halogenated anaesthetic. This seems eminently sensible and acceptable. Alternating halogenated anaesthetics will not necessarily reduce the risk in susceptible individuals, because the antibodies generated by one anaesthetic can apparently cross-react with antigens generated by a different one. Changing anaesthetics may perhaps reduce the likelihood of a hepatotoxic reaction, but not necessarily to zero. The relative commoness of HH compared with enflurane or isoflurane hepatitis raises the possibility that the hepatotoxic potential of the halogenated anaesthetics is related to the extents to which they undergo biotransformation to reactive metabolites that bind covalently to hepatic proteins [2]. It may therefore be expected that the risk of hepatitis by cross-sensitization may be smaller if a halothane exposure follows, rather than precedes, an exposure to enflurane or isoflurane. However, this appears not to be the case [44].

Though the TFA-antibody is expensive and available in only a few laboratories, it would seem reasonable for patients with a documented clinical history to be referred to these for antibody testing. This is partly to strengthen the documentation on the patients themselves, partly to improve the documentation on the reliability of the tests, and partly to create a more centralized knowledge-base of the natural history of the condition.

Repeated anaesthesia in patients with no documented susceptibility. Recommending is much more problematical for patients who have no documented evidence of previous reactions but who nonetheless require repeat general anaesthesia after previous exposure to a halogenated anaesthetic agent. Is it safe to repeat the same agent, or, if not, is it any less dangerous to use a different one? Should the current guidelines on repeating halothane anaesthesia (i.e. avoidance especially for lengthy surgical procedures) be applicable to any other halogenated agent irrespective of whether it has previously been used? Would it be cost-effective to use antibody tests to screen all such patients before any re-exposure to a halogenated anaesthetic?

A reasonable and conservative recommendation might be to use the least metabolized of the available halogenated volatile anaesthetic agents for any patient requiring a repeat anaesthetic. Although isoflurane seems to be the most suitable inhalation agent that is generally available just now, any such recommendation ought to take into consideration the remote possibility of an underestimation so far of the extent to which isoflurane is biotransformed [33], either generally, or in particular cases with a genetically determined metabolic idiosyncracy, or in patients whose cytochrome P450 has been induced by agents such as alcohol or phenobarbitone, or by diseases such as diabetes or hyperthyroidism [4]. A more demanding approach might be to encourage more deliberate forethought about which patients might be latently more susceptible for the reasons listed, and to be more restrictive in the use of halogenated agents, and more inclined to screen for antibodies in such patients.

Because availability of suitable agents and delivery devices is making total intravenous (i.v.) anaesthesia an increasingly feasible proposition, more radical thoughts might yet become thinkable, such as extending the recommendations of Brown and Frink [18] to include any patient requiring multiple anaesthetics. This might well sound the death knell for inhalational anaesthesia, but will probably have to await a more complete appreciation of the potential hazards of total i.v. anaesthesia.


1 Koff RG. Postoperative jaundice. Surg Clin North Am 1975; 59: 823–837.
2 Kenna JG, Jones RM. The organ toxicity of inhaled anaesthetics. Anesth Analg 1995; 81: S51–66.
3 Ray DC, Drummond GB. Halothane hepatitis. Br Anaesth 1991; 67: 84–99.
4 Zimmerman H. Even isoflurane. Editorial. Hepatology 1991; 13,6: 1251–1253.
5 Lewis JH, Zimmerman HJ, Ishak KG, Mullick FG. Enflurane hepatotoxicity. A clinicopathologic study of 24 cases. Ann Int Med 1983; 98: 984–992.
6 Eger Ei II, Smuckler EA, Ferrell LD, Goldsmith CH, Johnson BH. Is enflurane hepatotoxic? Anesth Analg 1986; 65: 21–30.
7 Allen PJ, Downing JW. A prospective study of hepatocellular function after repeated exposures to halothane or enflurane in women undergoing radium therapy for cervical cancer. Br J Anaesth 1977; 49: 1035–1039.
8 Dundee JW, Fee JPH, Mcllory PDA, Black GW. Prospective study of liver function following repeat halothane and enflurane. J Roy Soc Med 1981; 74: 286–291.
9 Paull JD, Fortune DW. Hepatotoxicity and death following two enflurane anaesthetics. Anaesthesia 1987; 42: 1191–1196.
10 Allan LG, Hussey AJ, Howie J et al. Hepatic glutathione S-transferase release after halothane anaesthesia: open randomised comparison with isoflurane. Lancet 1987; 1: 771–774.
11 Hussey AJ, Aldridge LM, Paul D, Ray DC, Beckett GJ, Allan LG. Plasma glutathione S-transferase concentration as a measure of hepatocollular integrity following a single general anaesthetic with halothane, enflurane or isoflurane. Br J Anaesth 1988; 60: 130–135.
12 Stoelting RK, Blitt CD, Cohen PJ, Merin RG. Hepatic dysfunction after isoflurane anaesthesia. Anesth Analg 1987; 66: 147–153.
13 Brown BR. Hepatotoxicity of inhalational anaesthetics. Current opinion in Anaesthesiology 1989; 2: 414–417.
14 McLaughlin DF, Eger II Ei. Repeated isoflurane anaesthesia in a patient with hepatic dysfunction. Anesth Analg 1984; 63: 775–778.
15 Gregoire S, Smiley RK. Acute hepatitis in a patient with mild factor IX deficiency after anaesthesia with isoflurane. Can Med Assoc J 1986; 135: 645–646.
16 Carrigan TW, Straughen WJ. A report of hepatic necrosis and death following isoflurane anaesthesia. Anesthesiology 1987; 67: 581–583.
17 Brunt EM, White H, Marsh JW, Holtmann B, Peters MG. Fulminant hepatic failure after repeated exposure to isoflurane anaesthesia. A case report. Hepatology 1991; 13: 1017–1021.
18 Brown BR, Frink EJ. Status of inhalational anesthetic hepatotoxicity in the 1990s. Anesthesiology Review 1990; 17,6: 15–21.
19 Goldfarb G, Rogier E, Gebauer C et al. Comparative effects of halothane and isoflurane anaesthesia on the ultrastructure of human hepatic cells. Anesth Analg 1989; 69: 491–495.
20 Gelman S. Halothane hepatotoxicity again? Anaesth Analg 1986; 65: 831–834.
21 Gelman S, Van Dyke R. Mechanism of halothane-induced hepatotoxicity: another step on a long path. Anesthesiology 1988; 68: 479–482.
22 Neuberger JM. Halothane and hepatitis. Incidence, predisposing factors and exposure guidelines. Drug Saf 1990 5(1): 28–38.
23 Neuberger JM, Kenna JG. Halothane hepatitis: a model of immune-mediated drug hepatotoxicity. Clin Sci 1987; 72: 263–271.
24 Kanaya N, Nakayama M, Fujita S, Watanabe H, Namiki A. The effects of halothane and isoflurane on hepatic blood flow as measured by the finger piece indocyanine green test. Anaesth Analg 1993; 76: S1–S476.
25 Cook TL, Beppu WJ, Hitt BA, Kosek JC, Mazze RI. Renal effects and metabolism of sevoflurane in Fischer 344 rats: an in-vivo and in-vitro comparison with methoxyflurane. Anesthesiology 1975; 43: 70–77.
26 Takeda J, Kobayashi Y, Ochiai R, Sekiguchi H. Clinical survey of sevoflurane anaesthesia on hepato-renal function. Anaesth Analg 1993; 76: S1–S476.
27 Strum DP, Eger II Ei, Johnson BH, Steffey EP, Ferrell LD. Toxicity of sevoflurane in rats. Anesth Analg 1987; 66: 769–773.
28 Lind RC, Ghantous HN, Gandolfi AJ, Brown BR. Sevoflurane biotransformation and hepatotoxicity in the guinea pig. Anesthesiology 1989; 71: A310.
29 Frink EJ. The hepatic effects of sevoflurane. Anesth Analg 1995; 81: S46–50.
30 Green Jr WB, Eckerson ML, Depa R, Brown Jr BR. Covalent binding of oxidative metabolites to hepatic protein not detectable after exposure to sevoflurane or desflurane. Anesthesiology 1994; 81: A437.
31 Iaizzo PA, Seeward MJ, Powis G, Van Dyke RA. The effects of volatile anaesthetics on Ca2+ mobilization in rat hepatocytes. Anesthesiology 1990; 72: 504–509.
32 Blitt CD, Gandolfi AJ, Soltis JJ, Brown Jr BB. Extra-hepatic biotransformation of halothane and enflurane. Anesth Analg 1981; 60: 129–232.
33 Davidkova T, Kikuchi H, Fuji K et al. Biotransformation of isoflurane: urinary and serum fluoride ion and organic fluorine. Anesthesiology 1988; 69: 218–222.
34 Gil F, Fiserova-Bergerova V, Altman NH. Hepatic protection from chemical injury by isoflurane. Anesth Analg 1988; 67: 860–867.
35 Eger II Ei, Johnson BH, Strum DP, Ferrell LD. Studies of the toxicity of I-653, halothane and isoflurane in enzyme-induced hypoxic rats. Anesth Analg 1987; 66: 1227–1229.
36 Holmes AA, Weiskopf RB, Eger II Ei, Johnson BH, Rampil IJ. Hepatocellular integrity in swine after prolonged desflurane (I-653) and isoflurane anaesthesia: Evaluation of plasma alanine aminotransferase activity. Anesth Analg 1990; 71: 249–253.
37 Jones RM, Koblin JN, Cashman JN, Eger II Ei, Johnson BH, Damask MC. Biotransformation and hepato-renal function in volunteers after exposure to desflurane (1–653). Br J Anaesth 1990; 64: 482–487.
38 Pohl LR, Kenna JG, Satoh H, Christ DD, Martin JL. Neoantigens associated with halothane hepatitis. Drug Metab Rev 1989; 20: 203–217.
39 Satoh H, Davies HW, Takemura T, Gillette JR, Maeda K, Phol LR. An immunochemical approach to investigating the mechanism of halothane-induced hepatotoxicity. In: Bridges JW, Chasseaud LF, Gibson CG, eds. Progress in Drug Metabolism. Philadelphia. Taylor and Francis, 1987: 187–206.
40 Kenna JG, Satoh H, Christ DD, Pohl LR. Metabolic basis for a drug hypersensitivity: antibodies in sera from patients with halothane hepatitis recognize liver neoantigens that contain the trifluoroacetyl group derived from halothane. J Pharmacol Exp Ther 1988; 245: 1103–1109.
41 Satoh H, Martin BM, Schulick AH, Christ DD, Kenna JG, Pohl LR. Human anti-endoplasmic reticulum antibodies in sera of halothane hepatitis patients are directed against a trifluoroacetylated carboxylesterase. Proc Nat Acad Sci, USA 1989; 86: 322–326.
42 Kenna JG, Neuberger J, Williams R. Identification by immunoblotting of three halothane-induced liver microsomal polypeptide antigens recognized by antibodies in sera from patients with halothane-associated hepatitis. J Pharmacol Exp Ther 1987; 242: 733–740.
43 Ritchie PA, Cheshire MA, Pearce NH. Decontamination of halothane from anaesthetic machines achieved by continuous flushing with oxygen. Br J Anaesth 1988; 60: 859–863.
44 Hubbard AK, Gandolfi AJ, Brown BR. Immunological basis of anaesthetic-induced hepatotoxicity. Anesthesiology 1988; 69: 814–817.
45 Hubbard AK, Roth TP, Gandolfi AJ. Elicitation of a cell mediated immune response to a reactive intermediate of halothane. Toxicologist 1988; 8: 12.
46 Martin JL, Kenna JG, Pohl LR. Antibody assays for the detection of patients sensitized to halothane. Anesth Analg 1990; 70: 154–159.
47 Kenna JG, Neuberger J, Williams R. An enzyme-linked immunosorbent assay for detection of antibodies against halothane-altered hepatocyte antigens. J Immunol Methods 1984; 75: 3–14.
48 Kenna JG, Neuberger J, Williams R. Specific antibodies to halothane-induced liver antigens in halothane-associated hepatitis. Br J Anaesth 1987; 59: 1286–1290.
49 Bird GLA, Williams R. Detection of antibodies to a halothane metabolite hapten in sera from patients with halothane-associated hepatitis. J Hepatol 1989; 9: 366–373.
50 Hubbard AK, Roth TP, Gandolfi AJ, Brown BR, Webster NR, Nunn JF. Halothane hepatitis patients generate an antibody response toward a covalently bound metabolite of halothane. Anesthesiology 1988; 68: 791–796.
51 Hastings KL, Thomas C, Hubbard AK, Gandolfi AJ. Screening for antibodies associated with halothane hepatitis. Br J Anaesth 1991; 67: 722–728.
52 Christ DD, Satoh H, Kenna JG, Pohl LR. Potential metabolic basis for enflurane hepatitis and the apparent cross-sensitization between enflurane and halothane. Drug Metab Dispos 1988; 16: 135–140.
53 Christ DD, Kenna JG, Kammerer W, Satoh H, Pohl LR. Enflurane metabolism produces covalently bound liver adducts recognized by antibodies from patients with halothane hepatitis. Anesthesiology 1988; 69: 833–838.
54 Rice SA, Maze M, Smith CM, Kosek JC, Mazze RI. Halothane hepatotoxicity in Fischer 344 rats pretreated with isoniazid. Toxicol Appl Pharmacol 1987; 87: 411–419.
55 Rice SA, Talcott TE. Effects of isoniazid treatment on selected hepatic mixed-function oxidase. Drug Metab Dispos 1979; 7: 260–262.
56 Rice SA, Dooley JR, Mazze RI. Metabolism by rat hepatic microsomes of fluorinated ether anesthetics following ethanol consumption. Anesthesiology 1980; 58: 237–241.
57 Favreau LV, Malchoff DM, Mole JE, Schenkman JB. Response to insulin by two forms of rat heptic microsomal cytochrome P450 that undergo major (RLM6) and minor (RLM5b) elevations in diabetes. J Biol Chem 1987; 262: 14319–14326.
© 1997 European Academy of Anaesthesiology