TOXICITY OF INHALATION ANAESTHETIC AGENTS

Br.J. Anaesth. (1978), 50, 665

TOXICITY OF INHALATION ANAESTHETIC AGENTS E. N. COHEN

0007-0912/78/0050-0665 §01.00

© Macmillan Journals Ltd 1978

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

Although hundreds of compounds on the chemist's For example, bromobenzene is metabolized to an shelf are known to produce anaesthesia, relatively few active epoxide (Reid et al., 1971; Mitchell, Jollow are used clinically. The requirements for a clinically and Gillette, 1973). A second example is seen with useful anaesthetic agent are necessarily stringent and the hepatic necrosis produced by acetaminophen include: a wide margin of safety, total reversibility of U.S.P. (paracetamol) (Brodie et al., 1971; Gillette, anaesthetic effect and an absence of long-term toxicity. 1974). The centrilobular necrosis produced by this In addition, a successful inhalation anaesthetic drug is enhanced by pretreating animals with agent must combine the capacity for depression of phenobarbitone which induces the mixed function the central nervous system with low levels of inter- oxidase system: in contrast, liver necrosis is prevented ference with respiration and the circulation. The by prior use of an enzyme inhibitor such as piperonyl central nervous system effects must be quickly butoxide (Mitchell, Jollow and Gillette, 1973) or reversible. cobaltous chloride (Tephly and Hibbeln, 1971). The acute depressant effects produced by the A similar relationship between drug metabolism inhalation anaesthetics contrast with the slower and toxicity would appear to apply to many of our development of long-term toxicity involving organs inhalation anaesthetics. Few anaesthetic agents, with such as the liver, the kidneys or the reproductive the exception of chloroform, are capable of producing system. Although such long-term toxicity associated direct cellular damage by themselves, but the with anaesthetics has been recognized for many hepatic and renal toxicity is related to metabolism in years, in the past it has been considered to occur a dose-dependent fashion. For example, Scholler infrequently, to be of little importance and of unde- (1970) found that the hepatotoxicity of chloroform termined aetiology. More recent findings suggest was associated with its metabolism. Chloroform that not only are these long-term effects serious, but anaesthesia in rats pretreated with phenobarbitone they occur more frequently than was previously which induces the liver enzymes led to extensive recognized. In addition the aetiological factors are hepatic necrosis; pretreatment with the enzyme now beginning to be understood. inhibitor disulfiram reduced or abolished the hepatic damage. RELATIONSHIP BETWEEN ANAESTHETIC An association between anaesthetic metabolism METABOLISM AND TOXICITY and toxicity also occurs with other agents. Cascorbi Several reports now suggest there is a significant and Singh-Amaranth (1972) observed that, whereas association between the metabolism of anaesthetic 20% of mice anaesthetized with fluroxene for 1 h agents and the development of toxicity. Whilst the died within 5-24 h, if animals were pretreated with majority of drugs are metabolized in the body to less phenobarbitone the mortality was 95%. However, toxic derivatives, this is not always so. Occasionally, pretreatment with carbon tetrachloride which dechemically inert compounds can be transformed into pressed the non-specific hepatic enzymes markedly reactive metabolites which can combine with tissue reduced the mortality following fluroxene anaesthesia. macromolecules. These combinations may be toxic. With halothane, the association between metabolism and toxicity has been more difficult to define. ELLIS N. COHEN, M.D., Department of Anesthesiology, Stanford University Medical Center, Stanford, California The lack of an adequate animal model was a serious problem. The important advance was the use of 94305, U.S.A. Portions of this material have been abstracted from animals pretreated with an enzyme-inducing agent. Metabolism of the Volatile Anesthetics: Implications for Stenger and Johnson (1972) and Reynolds and Toxicity (1977) (Eds. E. N. Cohen and R. A. Van Dyke). Menlo Park, California: Addison Wesley Publishing Moslen (1974) observed sub-capsular hepatic necrosis following the administration of halothane to rats Company. pretreated with phenobarbitone. More recently, Supported by NIH grant OH 00622-01.

666

BRITISH JOURNAL OF ANAESTHESIA

Trifluoroacetaldehyde is a suggested urinary metabolite of halothane which has yet to be isolated in man or experimental animals. When given i.v. or i.p. in mice its LD 50 is 600 mg kg" 1 , which places it between trifluoroacetic acid and trifluoroethanol (Airaksinen and Tammisto, 1968). Inorganicfluoride.This toxic metabolite is released in large amounts during and following methoxyflurane anaesthesia and in lesser amounts with enflurane, halothane and isoflurane. The amount released with halothane and isoflurane is normally very small. Fluoride concentrations can increase to nephrotoxic values (in excess of 40-50 (xmol litre"1) following methoxyflurane (Mazze, Shue and Jackson, 1971). The amount of renal damage is related to the fluoride concentration. Oxalic acid. There is an increased urinary excretion of oxalic acid following methoxyflurane anaesthesia MECHANISMS OF TOXICITY and crystalline oxalate deposits have been demonThree alternative pathways have been proposed to strated in the kidneys. Although renal toxicity as a link metabolism of anaesthetic agents and their result of tubular obstruction is known to follow high toxicity: the accumulation to toxic concentrations of urinary oxalate concentrations, toxicity only occurs metabolites normally readily excreted; the production at concentrations 10 times greater than those found of reactive intermediates that form irreversible bonds following the clinical use of methoxyflurane. The with tissue macromolecules and, finally, the formation renal lesion produced by methoxyflurane is histoof haptens that could produce hypersensitive or logically and functionally different from that seen immune responses. In each case metabolism of the with oxalate (Cousins, Mazze and Kosek, 1974). agent is implicated. Reactive intermediates Although the halogenated inhalation anaesthetic Excreted metabolites A number of metabolites of anaesthetic agents agents are normally considered to be unreactive, it which are excreted in the urine have been identified appears that under appropriate conditions they can be metabolized to reactive free radicals capable of and their potential toxicity determined. Trifluoroacetic acid is a major metabolite of combining with cellular constituents. In most stable halothane, fluroxene and isoflurane in several animals. compounds the electron orbitals of all the atoms are However, chronic administration shows it to have a filled with pairs of electrons of opposite spin. Howlow level of toxicity (Schimmasek et al., 1966; ever, some physical or chemical stresses, such as Stier et al., 1972). Its low toxicity following oral, those induced by metabolism, may disrupt this i.v. or i.p. administration is related to its poor balanced state, producing compounds containing an absorption. In the physiological range of pH values atom with a single impaired electron in an outer it is ionized and is thus unlikely to penetrate cell orbit. This type of compound is a reactive free radical. membranes from the outside. There are a number of ways in which such a free Trifluoroethanol is present in the urine following radical might be produced by metabolism of an fluroxene anaesthesia and is a suggested (but un- anaesthetic drug. For example, hydroxylation and proven) metabolite of halothane. In mice its LD 50 dehalogenation of halothane by its interaction with varies from 158 to 350 mg kg" 1 (Airaksinen and the cytochrome P-450 system could produce a Tammisto, 1968; Blake et al., 1969). Trifluoroethanol reactive acyl halide: is produced in only small amounts in humans F Br F Br F O anaesthetized with fluroxene and has not been II II I II demonstrated after administration of halothane: it HB F-C-C-H!j» F-C-C-OH /F-C-C-C1 seems that it poses little clinical risk (Gion et al., II II I 1974).

F Cl

F Cl

F

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

pretreatment of rats with the polychlorinated biphenyl Aroclor 1254, was found to lead to a widespread hepatic necrosis following a halothane anaesthetic (Reynolds, Moslen and Szalo, 1976; Sipes and Brown, 1976). Within a few hours of anaesthesia the serum transaminase concentrations were increased and there were severe morphological changes in the centrilobular parenchymal cells. There are convincing reports that the nephrotoxicity seen following methoxyflurane is the result of the metabolic production of fluoride ions in both Fischer 344 rats and in man (Mazze, Cousins and Kosek, 1972; Cousins, Mazze and Kosek, 1974). In rats Mazze, Cousins and Kosek (1972) showed that the direct injection of sodium fluoride is capable of producing renal damage identical to that seen following methoxyflurane anaesthesia.

667

TOXICITY OF INHALATION ANAESTHETIC AGENTS It has also been suggested that, under some conditions, cytochrome P-450 is capable of removing a proton to form a reactive carbanion (Ullrich and Schnabel, 1973):

Malonaldchyde and other cleavage products

FIG. 1. Possible mechanism for lipoperoxidation.

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

associated with these lipoprotein membranes. The lipid constituents are particularly rich in unsaturated fatty acids and, thus, binding of the reactive intermediates would tend to occur within this milieu in the hepatic endoplasmic reticulum. F Br F Br The action of the reactive intermediate may commence with an attack on the alpha-methylene H F-C—C—H ~ + , F—C— Ccarbon atom of the unsaturated fatty acid. The presence of the double bond weakens the carbonI I F Cl F Cl hydrogen bond adjacent to the unsaturated bond. Free radical intermediates derived from the haloBy analogy with the work of Van Duuren and his genated anaesthetics are thus able to initiate peroxidacolleagues (1972) with alpha-haloethers it is possible tive decomposition of the lipids by abstracting allylic that the following reaction may occur: hydrogen from the alpha-methylene carbon atom, thereby rearranging the position of the double bonds. F Br F H A subsequent attack by oxygen produces cleavage of the radical (fig. 1) (Brown and Sagalyn, 1975). F—C—C—H F—C—CUnless terminated, this oxidative deterioration will be transferred sequentially to adjacent fatty acid F Cl F Cl complexes in a propagated autocatalytic chain The formation of free radicals and reactive (Recknagel and Ghoshal, 1966). intermediates is consistent with evidence showing Brown (1972) produced evidence of hepatic that the metabolites bind covalently to liver micro- microsomal lipoperoxidation in the rat following the somes and to tissue macromolecules (Uehleke, Hellmer administration of chloroform and halothane. Using and Tarbelli-Poplawski, 1973; Van Dyke and Wood, rats pretreated with phenobarbitone, he measured the 1975). increase in diene conjugation resulting from lipoperThe cellular damage produced as a result of the oxidation in hepatic microsomes. Brown, Sipes and binding of a reactive metabolite varies depending on Sagalyn (1974) found, in animals pretreated with the molecular aggregate formed. The molecules with phenobarbitone, that chloroform anaesthesia the highest susceptibility for damage include the produced a decrease in the liver content of glutaunsaturated fatty acids and nucleic acids. Fatty acid thione and an increase in hepatic necrosis. Pretreatcomplexes with phosphatidylcholine and phos- ment with diethyl malate also decreases glutathione phatidylethanolamine are the major components of concentrations and is associated with liver damage. It the membranes of liver endoplasmic reticulum and appears that, if the tissue concentrations of such mitochondria. Since inhalation anaesthetic agents are antioxidants reach critically low values, the reactive strongly lipophilic they can be expected to be closely metabolites formed from chloroform are no longer

668

Hypersensitivity and immune responses It is possible that the rare hepatic injury seen following the use of halogenated anaesthetic agents might have an allergic or hypersensitivity basis. The reactions seen are usually considered to represent a delayed or cell-mediated sensitivity produced by specialized mononuclear cells, thymus-derived lymphocytes, or T cells which react with membrane-bound antigens. Whilst it is generally accepted that simple non-reactive molecules such as the halogenated anaesthetics cannot be antigens, their metabolites are reactive and capable of forming stable bonds with proteins or other macromolecules. The presence in vivo of such metabolites which are covalently bound to macromolecules in the liver has been demonstrated by a number of workers. Rosenberg and Wahlstrom (1973) attempted to show whether trifluoroacetic acid and the other "presumed" halothane metabolites such as trifluoroethanol, trifluoroacetaldehyde and trifluoroacetic anhydride could act as haptens in rabbits immunized with a chicken serum globulin complex. Although their results suggested that the immune serum formed similar precipitins against the complexes, the findings are not totally convincing. Later studies by Mathieu and his colleagues (1974) using a trifluoroacetyl guineapig albumin complex showed that guineapigs display a typical delayed-type hypersensitivity. There is therefore evidence to show that anaesthetic metabolites, when complexed to proteins, can act as haptens. The fact that a delayed skin hypersensitivity can be induced does not mean that such hapten complexes are also capable of causing autoimmune liver destruction.

SPECIFIC ORGAN TOXICITY

A direct relationship between the metabolism of anaesthetic drugs and their long-term toxicity has been firmly established. A reduction in the metabolism of such drugs by pretreatment with enzyme inhibitors or by antioxidants prevents the toxicity of chloroform and fluroxene in several species. In addition, increased toxicity with chloroform, halothane, methoxyflurane or fluroxene follows pretreatment with enzyme induction agents such as phenobarbitone or 3-methylcholanthrene. The aetiology of such toxicity may relate to the formation of either reactive metabolites which bind to macromolecules or toxic metabolic end-products such as trifluoroethanol or fluoride ions. Three organ systems appear to be the major targets: the liver, the kidneys and the reproductive organs. Hepatotoxicity Liver damage produced by chlorinated hydrocarbons such as carbon tetrachloride and chloroform is well known. Within 3 years of its introduction in 1847, Casper reported the first case of delayed chloroform toxicity (cited by Techendorf, 1921). Chloroform can be considered a true hepatotoxin, since its effects on the liver increase with dose and duration of exposure. The necrosis is preceded by fatty changes in the liver cells nearest the central veins. As the damage increases the necrosis spreads to involve the whole lobule with the last area of destruction being adjacent to the portal tracts. An alternative explanation to a direct effect would be related to metabolism of the chloroform. Scholler (1970) found that pretreatment with the antimetabolite disulfiram prevented the hepatotoxicity of chloroform. Similarly, in the newborn chloroform does not damage the liver, presumably as a result of the lack of development of the drug-metabolizing enzyme system (Kato and Gillette, 1965). In contrast, phenobarbitone pretreatment significantly increases the toxicity of chloroform (Brown and Sagalyn, 1974). The exact mechanism has not yet been defined, but it has been suggested that the damage is produced as a result of the binding of free radicals to cellular constituents of the liver. Although one might suspect that most halogenated anaesthetic agents produce liver damage in an identical fashion, this does not appear to be the case. Admittedly, each is capable of causing hepatotoxicity, but the incidence and mechanisms are different. For example, halothane was introduced only after careful

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

quenched and become free to promote tissue damage by lipoperoxidation. Wood, Gandolfi and Van Dyke (1976) investigated the binding of halothane metabolites to rat liver microsomes incubated in nitrogen. Under such anaerobic conditions greater diene conjugation occurred although the formation of malonaldehyde and other "non-peroxide-conjugated-dienes" was less in the anaerobic than in the aerobic state. Thus it seems that the initial step in lipoperoxidation occurs in the absence of oxygen, but peroxidation does not occur. In aerobic conditions the smaller amount of binding of the products of halothane to lipids may reflect a preferential oxidation of halothane and its radicals with the ultimate formation of trifluoroacetic acid.

BRITISH JOURNAL OF ANAESTHESIA

TOXICITY OF INHALATION ANAESTHETIC AGENTS

bolism appears to play an essential role whether the presumed cause is hypersensitivity, the accumulation of toxic metabolites, or the conjugation of reactive intermediates. The clinical picture of fever, malaise, arthralgia eosinophilia or lymphocytopaenia supports the diagnosis of a hypersensitivity-type response. Unfortunately, these signs do not always appear and, in addition, similar symptoms accompany viral hepatitis. A number of in vitro tests which depend on the development of cell-mediated sensitivity have been designed to diagnose "halothane hepatitis". Such tests include lymphocyte transformation and the appearance of antimitochondrial antibodies (Rodriques et al., 1969; Paronetto and Popper, 1970). Although both tests were reported to be highly specific, subsequent investigators have failed to confirm these findings. The presence of a previous exposure to halothane in more than half the reported cases of hepatitis is in accord with the sensitization hypothesis. However, with the prevalent use of this anaesthetic, multiple exposures would be expected to occur with increasing frequency. A small number of individuals with previous episodes of hepatitis have actually undergone a challenge test with halothane (Belfrage, Ahlgren and Axelson, 1966; Klatskin and Kimberg, 1969). Two well-known cases of anaesthetists challenged with small doses of halothane attest to the existence of a hypersensitivity reaction, although the significance of these reports has recently been questioned (Simpson, Strunin and Walton, 1973). Finally, there are data which suggest a significant relationship between repeated exposures to halothane and the rapidity with which jaundice develops following subsequent exposures (Inman and Mushin, 1974). Although there appears to be a decreasing time interval for the development of jaundice following repeated exposures to halothane, these data have also come under question. Thus, while it is tempting to postulate a causal relationship between the hypersensitivity response and hepatotoxicity to halothane, the data are equivocal. The metabolism of halothane to toxic end-products has also been suggested as an explanation for the liver damage. Studies in the mouse and guineapig with parenterally administered trifluoroacetic acid, as well as with some presumed halothane metabolites, indicate varying degrees of damage (Airaksinen and Tammisto, 1968). However, the route of administration employed in these studies may not have allowed the metabolites to reach certain critical intracellular

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

animal experimentation had indicated it not to be metabolized and to lack hepatotoxicity. Millions of anaesthetics later, it would seem that halothane may be the cause of rare cases of hepatic necrosis. Although this relationship of halothane to liver damage is difficult to establish in the clinical situation, such a relationship has been found to occur in rats following pretreatment with the enzyme inducer, Aroclor 1254 (Reynolds, Moslen, and Szalo, 1976; Sipes and Brown, 1976). Similarly, there appears to be a relationship between trace-dose concentrations of halothane and hepatotoxicity. Stevens and his colleagues (1975) exposed rats in a chamber for 4 weeks to halothane concentrations equivalent to 1/30 MAC. The animals showed hepatotoxicity with vascular degeneration, lipidosis, focal and zonal necrosis. Similar animals exposed at 1/600 MAC showed no changes, and animals exposed to intermediate concentrations exhibited corresponding degrees of hepatic injury. These findings were interpreted as showing that halothane was capable of producing injury by a direct hepatotoxic action through the mechanism of anaesthetic metabolism. There is some evidence that personnel working in operating rooms who are exposed to trace concentrations of inhalation anaesthetic agents may develop hepatotoxicity. Measurable concentrations of these gases can be found in all operating rooms where they are used (Linde and Bruce, 1969; Whitcher, Cohen and Trudell, 1971). Although the concentrations are small and measured in parts per million, continuing exposure can result in the absorption of a significant amount of anaesthetic. Once these highly lipid-soluble anaesthetics are taken into the body they are eliminated relatively slowly. While most are excreted intact within several hours, significant amounts remain in the body and are slowly biodegraded over days and weeks. The incidence of hepatotoxicity in exposed operating room personnel has been examined in several large surveys; it appears to correlate with the presence of trace anaesthetic gas concentrations and the duration of the operating room exposure (Cohen et al., 1974). The increase in hepatic disease among exposed physician anaesthetists was found to be approximately twice that of the unexposed control group. A similar study amongst dentists exposed to higher waste anaesthetic gases in their practice showed an even greater incidence of hepatotoxicity (Cohen, Brown et al., 1975). Halothane. Although the precise cause of halothane hepatotoxicity remains to be established, its meta-

669

670

The appearance of a cysteine conjugate of halothane in the urine is of considerable importance. The conjugate found in human urine was 2-bromo-2-

chloro-1-l-difluoroethylene. It was suggested that this occurred following the reaction of glutathione with a reductive dehydrofluorination product of halothane. The highly reactive bromochlorodifluorethylene intermediate could be formed as a result of proton abstraction from halothane by cytochrome P-450 (Ullrich and Schnabel, 1973). It has been suggested that gluthathione plays an important part in the body by its preferential conjugation of the reactive intermediates formed during metabolism of foreign compounds. In vivo studies in rats pretreated with phenobarbitone indicate that anaesthesia with chloroform, but not with halothane, decreases the concentration of glutathione in the liver (Brown, Sipes and Sagalyn, 1974). This latter finding may represent a species difference. Although the protective role of glutathione conjugation in man remains to be established, the administration of nucleophiles such as cysteine, cysteamine and dimercaprol prevents the liver damage produced in mice by paracetamol (acetaminophen) (Boyland and Chasseaud, 1967; Judah, McLean and McLean, 1970). Theoretically, the use of these agents offers a possibility for the prevention and treatment of halothane-induced hepatitis. The pathways described in the preceding discussion indicate that the metabolism of halothane may proceed along either an oxidative or a reductive route. Under aerobic conditions, metabolism preferentially leads to fluoride, bromide and trifluoroacetic acid. Widger, Gandolfi and Van Dyke (1976) investigated the effects of a reduced oxygen tension on the metabolism of halothane. Rats exposed to a 7% inhaled oxygen concentration responded with significantly increased concentrations of fluoride ions in the plasma. This increase in dehalogenation under hypoxic conditions was also accompanied by a more than three-fold increase in the covalent binding of metabolites to the microsomal lipids. This greater binding during hypoxia supports the hypothesis that reductive metabolism will produce a more reactive chemical species. Methoxyfturane. Hepatitis associated with methoxyflurane has also been reported. Joshi and Conn (1974) reviewed 24 cases. In most respects the clinical aspects of this syndrome were indistinguishable from those found following halothane. Histologically, the lesion is identical with that of viral hepatitis. The majority of cases were in women. More than half had a previous exposure to either halothane or methoxyflurane, and a cross sensitization between these two anaesthetic agents seems Likely. Although most

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

areas, which would be directly accessible to metabolites produced in vivo. There also appears to be only limited toxicity to the final metabolites of halothane. These metabolites, although weakly toxic, may conceivably accumulate to hazardous concentrations as a result of repeated administrations of the anaesthetic, depression of excretory pathways, or increased production of the metabolites following induction of the drug-metabolizing enzymes. Of more serious concern are the studies which indicate covalent binding of metabolites of halothane to macromolecules within the liver. The reactive intermediates are able to bind to both lipid and protein. Gandolfi and Van Dyke (1978) showed that the binding to phospholipids is through the fatty acid moiety. This binding is enhanced at lower oxygen tensions, following diethyl maleate-induced glutathione depletion or after pretreatment of the animal with a diet rich in polyunsaturated fatty acids. Enzyme induction increased the production of reactive intermediates and tended to promote increased binding. Hypoxia and a reduced antioxidant concentration also act to prolong the half-life of the intermediates and thus produce increasing binding. Recent studies have described the final urinary metabolites of halothane in man and also indicate the presumed toxicity of the reactive intermediates (Cohen, Trudell et al., 1975). Although trifluoroacetic acid has been recognized as the final metabolite with limited toxicity, it may be that the oxidative formation of the trifluoroacetyl radical represents the key step in the metabolism of halothane. There is evidence that this radical may bind covalently with phosphatidylethanolamine, for N-trifluoroacetylethanolamide is present as a urinary metabolite. The most likely source of the ethanolamide would be the phosphatidylethanolamine, which is a normal lipid constituent of cell membranes. Presence of the urinary metabolite suggests conjugation of the radical with the phosphatidylethanolamine in the cell membrane. The urinary metabolite would be produced as a result of enzymatic cleavage of this molecule. The accumulation of this metabolite within the cell could result from either the increased production of the radical following increased metabolism of halothane or the reduction in the amount of the phosphatase enzyme available for cleavage of the conjugated molecule.

BRITISH JOURNAL OF ANAESTHESIA

TOXICITY OF INHALATION ANAESTHETIC AGENTS

dose methoxyflurane anaesthesia and that the syndrome of high output renal failure follows metabolism of the drug to inorganic fluoride ions. Evidence that it is the inorganic fluoride ion which produces the nephrotoxicity in man may be seen with the production of similar abnormalities in the rat following the i.v. injection of sodium fluoride in amounts calculated to produce concentrations similar to those seen with methoxyflurane anaesthesia (Cousins, Mazze and Kosek, 1974). Although the mechanism by which the fluroide ion causes renal toxicity is unknown it is likely to be related to an interference in the transport of sodium in the proximal convoluted tubules (Kosek, Mazze and Cousins, 1972). The toxicity may also be produced from the potent inhibitory effects of fluoride ions on enzyme systems involved with cellular energy transfer (Wiseman, 1970). An additional possibility is that inhibition of adenyl cyclase leads to interference with the action of antidiuretic hormone (Orloff and Handler, 1964). A fourth possibility is that inorganic fluoride, which is a potent vasodilator, interferes with the counter-current system of the kidney, leading to an increase in medullary blood flow, washout of sodium from the interstitium and loss of hypertonicity (Caruso, Maynard and DiStefano, 1970). Although ethical considerations preclude studies in humans, the induction of the drug metabolizing systems in the rat with phenobarbitone has been shown to result in an increased production of inorganic fluoride and an increase in toxicity (Cousins, Mazze and Kosek, 1974). Pretreatment of the rats with the antimetabolite, SKF 525A-A, was shown to decrease the metabolism of methoxyflurane and to diminish its nephrotoxicity. There can be little question that the renal damage produced by methoxyflurane is associated with its metabolism and the Nephrotoxicity In the broadest sense all anaesthetic agents are production of inorganic fluoride ions. Enflurane, halothane, isoflurane andfluroxene.Since nephrotoxic, since they characteristically produce a generalized depression of renal functions. These each of these anaesthetic agents liberates fluroide ions changes are largely secondary to effects of anaesthesia during its metabolism, concern has arisen as to on the cardiovascular, sympathetic and endocrine whether there is an associated nephrotoxicity. The systems. Renal blood flow and glomerular filtration metabolism of enflurane to inorganic fluoride is rates usually return to normal within a few hours considerably less than that for methoxyflurane, with after the termination of the anaesthetic, although the a peak serum inorganic fluoride concentration ability to excrete a water load may be impaired averaging 22 jxmol litre" 1 following administration of somewhat longer. Persistent abnormalities of renal a clinically anaesthetic concentration (Cousins et al., function, however, have been associated with the 1974). This peak concentration is significantly below that known to be associated with nephrotoxicity. prolonged use of fluorinated anaesthetics. Methoxyflurane. There is abundant evidence in man Nonetheless, at least two case reports have appeared to show that renal toxicity is associated with high- indicating nephrotoxicity following enflurane. In one,

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

authors support an immunological basis for this hepatitis, no patients have been purposely challenged with methoxyflurane to induce a recurrence. Several case reports, however, discuss patients with previous hepatic injury from methoxyflurane who have received a subsequent anaesthetic with a recurrence of the hepatic injury (Rosander, 1970; Brenner and Kaplan, 1971). There is also one report of hepatic injury following methoxyflurane administered in subanaesthetic concentrations for obstetric analgesia on each of two occasions (Rubinger, Davidson and Melmed, 1975). At present there are no data which define the specific reactive metabolites of the methoxyflurane that might act as possible haptens, although their existence seems possible. Fluroxene. This anaesthetic drug has enjoyed a remarkable history of clinical safety in over 500 000 administrations. However, recent reports suggest the occasional occurrence of liver damage in man. Two fatal cases have been described in which the patients were on enzyme-inducing drugs, and in one individual this included both phenobaritone and diphenylhydantoin (Reynolds, Brown and Vandam, 1972; Tucker et al., 1973). The overall toxicity of fluoroxene noted in several animal species appears to correlate with its metabolism to trifluoroethanol. However, humans metabolize fluroxene to trifluoroacetic acid, and only a very small amount is present in the urine as trifluoroethanol (Gion et al., 1974). There is no evidence which defines the cause of the hepatic injury in man following fluroxene anaesthesia; this syndrome does not correspond with the toxicity seen in the experimental animals. However, it does seem likely that fluroxene toxicity is associated with its metabolism, although perhaps along different pathways.

671

672

Toxicity to the reproductive organs The clinical observations of an increase in the spontaneous miscarriage rate, of a decreased fertility, and an increase in foetal abnormalities amongst children of operating room personnel exposed to waste anaesthetic gases suggest an effect of inhalation anaesthetics on the reproductive system (KnillJones et al., 1972; Cohen et al., 1974). Epidemiological surveys show that exposed female anaesthetists and their progeny may show a doubling of spontaneous miscarriage and foetal abnormality rates when compared with control groups of unexposed female paediatricians. Whilst there are no data which confirm a cause-effect relationship, there remains the possibility that these effects result from chronic exposure to waste anaesthetic gases. Furthermore, it is likely that the metabolism of such anaesthetics will be causally implicated. Cascorbi, Blake and Helrich (1970) suggested, in a small-scale study, that anaesthetists metabolized halothane more efficiently than did a control group of pharmacists, although the differences were not statistically significant. It is possible that the anaesthetists underwent an induction of their microsomal enzymes as a result of the chronic exposure to anaesthetic gases in the operating room. Ghoneim and his colleagues (1975) found evidence

to show that operating room exposure may also exert an inhibitory effect on other enzyme systems. Their study of the kinetics of Warfarin in anaesthesia residents showed a prolongation of the primary half-life by 53% following 4 months of operating room exposure. Recent studies with radioactive halothane indicate an unusually high concentration of the non-volatile metabolites in the ovaries and testes of heart transplant donors (Cohen and Van Dyke, 1977). The role of these metabolites in the previously mentioned toxicity in anaesthetists has not been defined. Despite the possibly serious implication of these studies, much further information is needed to identify the precise role of anaesthetic agents and their metabolites in the toxicity associated with the reproductive system. ANAESTHETICS AND CARCINOGENICITY

Although there is legitimate concern regarding the relationship between cancer and anaesthesia, the evidence is fragmentary and only limited studies are at present available. Eschenbrenner (1945) reported the prodution of hepatomata in mice following the repeated oral administration of chloroform; however, the hepatic lesions appeared only if the anaesthetic dosage was large enough to produce liver necrosis. Kunz, Schaude and Thomas (1969) demonstrated that a different histological picture of rat liver carcinoma resulted from the administratidn of nitrosamines, depending on whether the animals were anaesthetized with phenobarbitone, halothane or methoxyflurane. On the other hand, Peraino, Fry and Staffeldt (1971) showed a protective influence of phenobarbitone on the hepatocarcinogenesis induced in the rat by 2-acetyl-aminofluorene. Using an in vitro microbial system, Baden and his group (1976) were unable to produce mutagenicity in two histidinedependent strains incubated with halothane in concentrations ranging from 0.1 to 30%. In an extension of these studies, mutagenicity was also absent following incubation with methoxyflurane, enflurane, trichloroethylene and isoflurane; in contrast, exposure to fluroxene in concentrations of 0.1 and 30% produced a six-fold increase in the numbers of mutant colonies (Baden et al., 1977). In another approach, Saffiotti (1978) found that trichloroethylene, when administered in large intragastric doses, was capable of producing hepatic cancer in mice in a dose-dependent fashion. Concern over the effects of chronic exposure to inhalation anaesthetic agents led Corbett (1976) to investigate the offspring of mice exposed transplacentally and

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

histological evidence of renal damage appeared with only a modest increase in the serum inorganic fluoride concentration (Loehning and Mazze, 1974). Since this patient had severe pre-existing renal disease with a failing kidney transplant, the authors suggested that the toxicity threshold may have been lowered in the diseased kidney. In the second case it was reported that, following 6 h of uneventful enflurane anaesthesia, the serum inorganic fluoride ion concentration reached 93 (xmol litre" 1 coinciding with evidence of renal failure (Eichhorn et al., 1976). Since this patient had received an enflurane anaesthetic 6 weeks earlier, the possibility of enzyme induction was suggested; no alternative causes for the renal failure were found. Isolated cases of renal damage have also been reported in association with fluroxene anaesthesia (Reynolds, Brown and Vandam, 1972; Tucker et al., 1973). These patients suffered from hepatic involvement and in addition were receiving enzyme-inducing drugs. Under usual clinical conditions, one would not anticipate renal symptoms in conjunction with the limited release of fluoride seen with this anaesthetic agent.

BRITISH JOURNAL OF ANAESTHESIA

TOXICITY OF INHALATION ANAESTHETIC AGENTS

compounds to alkylnitrosamines and diazo alkanes have already been demonstrated (Wolfe and Wasserman, 1972). CONCLUSION

It would appear that a number of mechanisms exist for the potential toxicity of volatile anaesthetic agents. Although the pathways are as yet incompletely defined there can be little doubt that metabolism plays a central role. It is thus essential that we continue our search for new and non-metabolizable anaesthetic agents. Investigations into the possible therapeutic use of antimetabolite and antioxidant drugs would seem to be of interest. With persistent efforts we shall move closer towards the realization of our goal: to provide the safest possible anaesthetic to each of our patients. REFERENCES

Airaksinen, M. M., and Tammisto, T. (1968). Toxic actions of the metabolites of halothane: LD 50 and some metabolic effects of trifluoroethanol and trifluoroacetic acid in mice and guinea pigs. Ann. Med. Exp. Fenn., 46, 242. Baden, J. M., Brinckenhoff, M., Wharton, R. S., Hitt, B. A., Simmon, V. F., and Mazze, R. (1976). Mutagenicity of volatile anesthetics: halothane. Anesthesiology, 45, 311. Kelley, M., Wharton, R. F., Hitt, B., Simmon, V. F., and Mazze, R. (1977). Mutagenicity of halogenated ether anesthetics. Anesthesiology, 46, 346. Belfrage, S., Ahlgren, I., and Axelson, I. (1966). Halothane hepatitis in an anaesthetist. Lancet, 2, 1466. Blake, D. A., Cascorbi, H. F., Rozman, R. S., and Meyer, F. J. (1969). Animal toxicity of 2,2,2,-trifluroethanol. Toxicol. Appl. Pharmacol., 15, 83. Block, J. B. (1974). Angiosarcoma of the liver following vinyl chloride exposure. J.A.M.A., 299, 53. Boyland, E., and Chasseaud, L. F. (1967). Enzymecatalysed conjugations of glutathione with unsaturated compounds. Biochem. J., 104, 95. Brenner, A. I., and Kaplan, M. M. (1971). Recurrent hepatitis due to methoxyflurane anesthesia. N. Engl. J. Med., 284, 961. Brodie, B. B., Reid, W. D., Cho, A. K., Sipes, G., Krishna, G., and Gillette, J. R. (1971). Possible mechanism of liver necrosis caused by aromatic organic compounds. Proc. Natl Acad. Set., U.S.A., 68, 160. Nitrous oxide is an intermediate ( + ) 1 oxidation state Brown, B. R. (1972). Hepatic microsomal lipoperoxidation and inhalation anesthetics: a biochemical and morphologic and could conceivably be metabolized by either an study in the rat. Anesthesiology, 36, 458. oxidative or a reductive mechanism. There is ample Sagalyn, A. M. (1974). Hepatic microsomal enzyme evidence to indicate the hepatic oxidation of amines induction by inhalation anesthetics. Anesthesiology, 40, 152. and the reduction of nitro compounds (Mitchard, (1975). Reactive Intermediates of Anesthetic 1971; Weisburger and Weisburger, 1971). If the Biotransformation and Hepatotoxicity (ed. R. B. Fink). human liver can be shown to be capable of metabolizNew York: Raven Press. ing nitrous oxide, particularly to nitric oxide or to the Sipes, I. G., and Sagalyn, A. M. (1974). Mechanisms nitrite ion, the additional steps converting the latter of acute hepatic toxicity: chloroform, halothane and glutathione. Anesthesiology, 41, 554.

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

after gestation to sub-anaesthetic concentrations of isoflurane. The results obtained after 15 months indicated a significant increase in hepatic neoplasms, with no increase in the control groups. Evidence for the direct induction of human cancer by anaesthetic drugs is largely epidemiologic. At least two current studies suggest a relationship between female cancer and operating exposure, presumably causally related to the waste anaesthetic gases. A limited study by Corbett and his colleagues (1973) showed the incidence of cancer in nurse anaesthetists in Michigan to be three times that of the control group. Evidence from a large-scale national survey (Cohen et al., 1974) showed the incidence of cancer in exposed female physician anaesthetists to be twice that of unexposed female paediatricians. Somewhat smaller increases in the cancer rates were seen in operating room nurses. Of considerable interest is the as yet unexplained finding that this increase in cancer does not apply to the exposed male. In the exposed women, there was little differentiation in the type or location of cancer that developed, with the exception that the frequencies of leukaemia and lymphoma were increased three-fold. The strong alkylating property of certain haloethers has been shown to be closely related to their carcinogenicity (Van Duuren et al., 1969, 1972). Of these compounds, bis(chloromethyl)ether was shown to be a strongly active carcinogen. An even higher degree of carcinogenicity has been assigned to vinyl chloride and its reactive by-products (Block, 1974). The ability of both these groups of hydrocarbons to form reactive intermediates is a property shared with many halogenated anaesthetic agents. Although only limited data are available with regard to nitrous oxide there is some evidence that it is metabolized (Matsubara and Mori, 1968). It is known to be biologically active and an enzyme inducer (Remer, 1961; Eastwood et al., 1963). The following equilibrium serves to illustrate the oxidation states of nitrogen in aqueous solution:

673

674

Inman, W. H. M., and Mushin, W. W. (1974). Jaundice after repeated exposure to halothane: an analysis of reports to the Committee on Safety of Medicines. Br. Med. J; 1, 5.

Joshi, P. H., and Conn, H. V. (1974). The syndrome of methoxyflurane associated hepatitis. Ann. Intern. Med., 80, 395. Judah, J. D., McLean, A. E. M., and McLean, E. K. (1970). Biochemical mechanisms of liver injury. Am.J. Med., 49, 609. Kato, R., and Gillette, J. R. (1965). Effects of starvation on NADPH-dependent enzymes in liver microsomes of male and female rats. J. Pharmacol. Exp. Ther., 150, 279. Klatskin, G., and Kimberg, D. V. (1969). Recurrent hepatitis attributable to halothane sensitization in anesthetists. N. Engl.J. Med., 280, 515. Knill-Jones, R., Moir, D. D., Rodriques, L. V., and Spence, A. A. (1972). Anaesthetic practice and pregnancy: a controlled survey of women anesthetists in the United Kingdom. Lancet, 2, 1326. Kosek, J. C , Mazze, R. I., and Cousins, M. J. (1972). The morphology and pathogenesis of nephrotoxicity following methoxyflurane (Penthrane) anesthesia: an experimental model in rats. Lab. Invest., 27, 575. Kunz, W., Schaude, G., and Thomas, C. (1969). Die Beeinflussung der Nitrosamincarcinogenese durch Phenobarbital und Halogenkohlen Wasserstoffe. Z. Krebsforsch., 72, 291. Linde, H. W., and Bruce, D. L. (1969). Occupational exposure of anesthetists to halothane, nitrous oxide and radiation. Anesthesiology, 30, 363. Loehning, R. W., and Mazze, R. I. (1974). Possible nephrotoxicity from enflurane in a patient with severe renal disease. Anesthesiology, 40, 203. Mathieu, A., Di Padua, D., Mills, J., and Kahan, B. (1974). Experimental immunity to a metabolite of halothane and fluroxene: cutaneous delayed-type hypersensitivity. Anesthesiology, 40, 385. Matsubara, T., and Mori, T. (1968). Studies on denitrification. IX: Nitrous oxide, its production and reduction of nitrogen. J. Biochem. {Tokyo), 64, 871. Mazze, R. I., Cousins, M. J., and Kosek, J. C. (1972). Dose-related methoxyflurane nephrotoxicity in rats: a biochemical and pathologic correlation. Anesthesiology, 36, 571. Shue, G. L., and Jackson, S. H. (1971). Renal dysfunction associated with methoxyflurane anesthesia: a randomized prospective clinical evaluation. J.A.M.A., 216, 278. Mitchard, M. (1971). Bioreduction of organic nitrogen. Xenobiotica, 1, 469. Mitchell, J. R., Jollow, D. J., and Gillette, J. R. (1973). Drug metabolism as a cause of drug toxicity. Drug Metdb. Dispos., 1, 418. Orloff, J., and Handler, J. S. (1964). The cellular mode of action of antidiuretic hormone. Am.J. Med., 36, 686. Paronetto, F., and Popper, H. (1970). Lymphocyte stimulation induced in patients with hepatitis following exposure to halothane. N. Engl. J. Med., 283, 277. Peraino, C , Fry, R. J., and Staffeldt, E. (1971). Reduction and enhancement by phenobarbital of hepatocarcinogenesis induced in the rat by 2-acetylaminofluroxene. Cancer Res., 31, 1506.

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

Caruso, F. S., Maynard, E. A., and DiStefano, V. (1970). Pharmacology of sodium fluoride; in Handbook of Experimental Pharmacology, Pharmacology of Fluorides (ed. F. A. Smith), p. 20. New York: Springer-Verlag. Cascorbi, H. F., Blake, D. A., and Helrich, M. (1970). Differences in biotransformation of halothane in man. Anesthesiology, 32, 119. Singh-Amaranth, A. V. (1972). Fluroxene toxicity in mice. Anesthesiology, 37, 480. Cohen, E. N., Brown, B. W., Bruce, D. L., Cascorbi, H. F., Corbett, T. H., Jones, T. W., and Whitcher, C. E. (1974). Occupational disease among operating room personnel: a national study. Anesthesiology, 41, 321. (1975). A survey of anesthetic health hazards among dentists. J. Am. Dent. Assoc, 90, 1291. Trudell, J. R., Edmunds, H. N., and Watson, E. (1975). Urinary metabolites of halothane in man. Anesthesiology, 43, 391. Van Dyke, R. A. (1977). Metabolism of Volatile Anesthetics: Implications for Toxicity. Menlo Park, California: Addison Wesley Co. Corbett, T. H. (1976). Cancer and congenital anomalies associated with anesthetics. Ann. N.Y. Acad. Sci., 271, 58. Cornell, R. G., Leiding, K., and Endres, J. L. (1973). Incidence of cancer among Michigan nurse anesthetists. Anesthesiology, 38, 260. Cousins, M. J., Greenstein, L. R., Hitt, B. A., and Mazze, R. I. (1974). Metabolism and renal effects of enflurane in man. Anesthesiology, 44, 44. Mazze, R. I., and Kosek, J. C. (1974). The etiology of methoxyflurane nephrotoxicity. J. Pharmacol. Exp. Ther., 190, 530. Eastwood, D. W., Green, C. D., Lambdin. M. A., and Gardner, R. (1963). Effect of nitrous oxide on the white cell count in leukemia. N. Engl.J. Med., 268, 297. Eichhorn, J. H., Hedley-Whyte, J., Steinman, T. I., Kaufman, J. M., and Laasberg, L. H. (1976). Renal failure following enflurane anesthesia. Anesthesiology, 45, 557. Eschenbrenner, A. B. (1945). Induction of hepatomas in mice by repeated oral administration of chloroform with observations on sex differences. J. Natl Can. Inst., 5, 251. Gandolfi, A. J., and Van Dyke, R. A. (1978). Effects of animal pretreatment on the in vivo covalent binding of halothane metabolites to hepatic lipids and proteins. Chem. Biol. Interaction (in press). Ghoneim, M. M., Delle, M., Wilson, W. R., and Ambre, J. J. (1975). Alteration of Warfarin kinetics in man associated with exposure to an operating-room environment. Anesthesiology, 43, 333. Gillette, J. R. (1974). A perspective on the role of chemically reactive metabolites of foreign compounds in toxicity. I: Correlation of changes in covalent binding of reactive metabolites with changes in the incidence and severity of toxicity. Pharmacology, 23, 2785. Gion, H., Yoshimura, N., Holaday, D. A., FigerovaBergerova, V., and Chase, R. E. (1974). Biotransformation of fluroxene in man. Anesthesiology, 40, 553.

BRITISH JOURNAL OF ANAESTHESIA

TOXICITY OF INHALATION ANAESTHETIC AGENTS

675

Downloaded from http://bja.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on April 19, 2015

Recknagel, R. O., and Ghoshal, A. K. (1966). Lipoperoxida- Stier, A., Kunz, H. W., Walli, A. K., and Schimassek, H. tion as a vector in carbon tetrachloride hepatotoxicity. (1972). Effects on growth and metabolism of rat liver by halothane and its metabolite trifluoroacetate. Biochem. Lab. Invest., 15, 132. Pharmacol., 21, 218. Reid, W. D., Christie, B., Krishna, G., Mitchell, J. R., Moskowitz, J., and Brodie, B. B. (1971). Bromobenzene Techendorf, W. (1921). Zur Chloroform Nachwirkung in metabolism and hepatic necrosis. Pharmacology;, 6, 41. Tieruersuch. Arch. Exp. Path. Pharmakol., 90, 288. Remer, H. (1961). Drugs as activators of drug enzymes, Tephly, T. R., and Hibbeln, P. (1971). The effect of cobalt metabolic factors controlling duration of drug action; chloride administration on the synthesis of hepatic in Proceedings of the First International Pharmacological microsomal cytochrome P-450. Biochem. Biophys. Res. Commun., 42, 589. Meeting (eds. B. B. Brodie and E. G. Erdos). New York: Pergamon Press. Tucker, W. K., Munson, E. S., Holaday, D. A., FigerovaReynolds, E. S., Brown, B. R., and Vandam, L. D. (1972). Bergerova, V., and Turner, B. M. (1973). Hepatorenal Massive hepatic necrosis after fluroxene anesthesia—a toxicity following fluroxene anesthesia. Anesthesiology, 39, 104. case of drug interaction ? N. Engl.J. Med., 286, 530. Moslen, M. T. (1974). Liver injury following halo- Uehleke, H., Hellmer, K. H., and Tarbelli-Poplawski, C. thane anesthesia in phenobarbital-pretreated rats. (1973). Metabolic activation of halothane and its covalent binding to liver endoplasmic proteins in vitro. Naunyn Biochem. Pharmacol., 23, 189. Schmeidebergs Arch. Pharmacol., 279, 39. Szalo, S. (1976). Halothane hepatotoxicity; enhancement by polychlorinated biophenyl pre-treatment. Ullrich, V., and Schnabel, K. H. (1973). Formation and Fed. Proc, 35, 376. binding of carbanions by cytochrome P-450 of liver Rodriquez, M., Paronetto, F., Schaffner, F., and Popper, H. microsomes. Drug Metab. Dispos., 1, 176. (1969). Antimitochondrial antibodies in jaundice Van Duuren, B. L., Katz, C , Goldschmidt, B. M., Frenkel, following drug administration. J.A.M.A., 208, 148. K., and Sivak, A. (1972). Carcinogenicity of halo-ethers. Rosander, G. (1970). Hepatic necrosis after methoxyflurane II: Structure-activity relationships of analogs of bis(chloro-methyl)ether. J. Natl Cancer Inst., 48,1431. anesthesia. Opusc. Med., 15, 305. Rosenberg, P., and Wahlstrom, T. (1973). Trifluoroacetic Sivak, A., Goldschmidt, B. M., Katz, C , and acid and some possible intermediate metabolites of Melchionne, S. (1969). Carcinogenicity of haloethers. halothane as haptens. Anesthesiology, 38, 224. J. Natl Cancer Inst., 43, 481. Rubinger, D., Davidson, J. T., and Melmed, R. N. (1975). Van Dyke, R. A., and Wood, C. L. (1975). In vitro studies on irreversible binding of halothane metabolite to microHepatitis following the use of methoxyflurane in obstetric somes. Drug Metab. Dispos., 3, 51. analgesia. Anesthesiology, 43, 593. Saffiotti, U. (1975). Memorandum Alert. March 21. Weisburger, J. H., and Weisburger, E. K. (1971). NJ. Nat! Cancer Inst. oxidation enzymes; in Handbook of Experimental PharmaSchimmasek, H., Helms, J., Kunz, W., and Stier, A. (1966). cology (eds. B. B. Brodie and J. R. Gillette). New York: Springer-Verlag New York, Inc. Lebervergrossung unter Trifluoressigsaure, Spalt Produkt des Halothan. Naunyn Schmiedebergs Arch. Exp. Path. Whitcher, C. E., Cohen, E. N., and Trudell, J. R. (1971). Pharmakol, 255, 67. Chronic exposure to anesthetic gases in the operating Scholler, K. L. (1970). Modification of the effects of room. Anesthesiology, 35,348. , chloroform on the rat liver. Br. J. Anaesth., 42, 603. Widger, L. A., Gandolfi, A. J., and Van Dyke, R. A. (1976). Simpson, B. R., Strunin, L., and Walton, B. (1973). Hypoxia and halothane metabolism in vivo: release of inorganic fluoride and halothane metabolite binding to Halothane hepatitis—fact or fallacy? Ada Anaesthesiol. cellular constituents. Anesthesiology, 44, 197. Belg., 2, 133. Sipes, I. G., and Brown, B. R. (1976). An animal model of Wiseman, A. (1970). Effect of inorganic fluoride on enzymes; in Handbook of Experimental Pharmacology, hepatotoxicity associated with halothane anesthesia. Pharmacology of Fluoride (ed. F. A. Smith), p. 20. New Anesthesiology, 45, 622. York: Springer-Verlag New York, Inc. Stenger, R. J., and Johnson, E. A. (1972). Effects of phenobarbital pretreatment on the responses of rat liver Wolfe, I. A., and Wasserman, A. E. (1972). Nitrates, nitrites and nitrosamines. Science, 177, 15. to halothane administration. Proc. Soc. Exp. Biol. Med., 140, 1319. Wood, C. L., Gandolfi, A. J., and Van Dyke, R. A. (1976). Stevens, W. C , Eger, E. I., White, A., Halsey, M. J., Lipid binding of a halothane metabolite: in vitro comMunger, W., Gibbons, R. D., Dolan, W., and Shargel, R. parison with lipid peroxidation. Drug Metab. Dispos., 4, 305. (1975). Comparative toxicity of halothane, isoflurane, and diethyl ether at subanesthetic concentrations in laboratory animals. Anesthesiology, 42, 408.