Liver Disease Caused by Anesthetics, Chemicals, Toxins, and Herbal and Dietary Supplements

Halothane

The retrospective National Halothane Study, cited in the past as the basis for exonerating halothane as a cause of hepatotoxicity, is now considered seriously flawed. Nearly 1000 cases of halothane hepatotoxicity were reported worldwide during the 1960s and 1970s. ^, A fairly uniform clinical picture of postoperative fever, eosinophilia, jaundice, and hepatic necrosis occurred a few days to weeks after administration of anesthesia, usually after repeat exposure to halothane, and the case-fatality rate was high ( Box 89.1 ). Rare cases of halothane-induced liver injury have been reported after workplace exposure among anesthesiologists, surgeons, nurses, and laboratory staff and after halothane sniffing for recreational use ; in affected persons, antibodies to trifluoroacetylated (TFA) proteins were demonstrated, indicating previous exposure. More recently, elevated urinary bromide levels have been reported to indicate exposure to halothane in anesthesiology personnel.

Two types of postoperative liver injury have been associated with halothane. A minor form (type 1) is seen in 10% to 30% of patients in whom mild, asymptomatic, self-limited elevations in serum ALT levels develop between the first and 10th postoperative days; the risk of hepatotoxicity after 2 or more exposures to halothane is higher than that for repeated use of alternative agents such as enflurane, isoflurane, and desflurane. Evidence of immune activation is lacking in these patients, in whom the serum ALT elevations generally reverse rapidly.

The major form of halothane-induced hepatotoxicity (type 2) is a rare, dose-independent, severe hepatic drug reaction with elements of immunoallergy and metabolic idiosyncrasy (see Box 89.1 ). After an initial exposure to halothane, the frequency of this form of toxicity is only about 1 per 10,000, but the rate increases to approximately 1 per 1000 after 2 or more exposures, especially when the anesthetic agent is readministered within a few weeks. Typically, zone 3 (centrilobular) hepatic necrosis is seen histologically. The case-fatality rate ranged from 14% to 71% in the pre-LT era and remains high in developing countries where halothane is still used. ^,

Others

The likelihood that individual haloalkane anesthetics will cause liver injury appears to be related to the extent to which they are metabolized by hepatic CYP enzymes: 20% to 30% for halothane, greater than 30% for methoxyflurane, 2% for enflurane, less than 1% for sevoflurane, and 0.2% or less for isoflurane and desflurane. Accordingly, the estimated frequency of hepatitis from the newer agents is much less than that for halothane ( Table 89.1 ).

Methoxyflurane caused hepatotoxicity and a high frequency of nephrotoxicity that led to its withdrawal. Enflurane caused a clinical syndrome similar to that for halothane, with the onset of fever within 3 days and jaundice in 3 to 19 days after anesthesia ^, and with an estimated incidence of enflurane-induced liver injury of about 1 in 800,000 exposed patients.

Despite its low rate of metabolism, several instances of isoflurane-associated liver injury have been reported. In one case, cross-sensitivity was suspected 22 years after initial exposure to enflurane. TFA liver proteins have been detected in patients with suspected isoflurane hepatotoxicity. In rats, the number of apoptotic hepatocytes seen after multiple exposures to isoflurane was only about 3% in a periacinar distribution. They were seen in only a small number of lobules, indicating a low propensity to cause hepatotoxicity.

The newer haloalkane anesthetics, desflurane and sevoflurane, appear to be nearly free of adverse hepatic effects. Desflurane undergoes minimal biotransformation and is not associated with the development of TFA antibodies in exposed rats. Only isolated reports of liver injury in patients receiving desflurane anesthesia have been published. The biotransformation of sevoflurane is also minimal, and only rare reports have implicated this agent in postoperative hepatic dysfunction. ^, In rats, sevoflurane produced fewer than 1% apoptotic hepatocytes after repeated exposures.

Ether, nitrous oxide, and cyclopropane apparently are devoid of significant hepatotoxic potential because of their lack of halogen moieties, and ketamine has only rarely been reported to cause hepatic injury. Propofol is considered largely free of hepatotoxic effects, even in patients with cirrhosis. Although propofol has a high affinity for mitochondrial membranes, no significant impairment in mitochondrial function has been shown in animal models.

Jaundice in the Postoperative Period

Between 25% and 75% of patients undergoing surgery experience postoperative hepatic dysfunction, ranging from mild elevations in liver biochemical test levels to hepatic failure, with postoperative jaundice reported in nearly 50% of patients with underlying cirrhosis. Patients undergoing upper abdominal surgical procedures are at highest risk of postoperative liver dysfunction, as well as pancreatitis, cholecystitis, and bile duct injury, because of impaired blood flow to the liver. Box 89.2 lists many causes of postoperative jaundice and hepatic dysfunction, broadly divided into hepatocellular injury, cholestasis, and indirect hyperbilirubinemia. Drugs that may cause hepatotoxicity in this setting include antibiotics (e.g., erythromycin, amoxicillin–clavulanic acid, trimethoprim/sulfamethoxazole, fluoroquinolones) (see Chapter 88 ) as well as the halogenated anesthetics discussed earlier.

Table 89.2 contrasts the features of halogenated anesthetic-induced hepatitis, ischemic hepatitis (shock liver) (see Chapter 85 ), and cholestatic injury in the early postoperative period. Bile cast nephropathy is a relatively newly recognized clinical entity that can contribute to hyperbilirubinemia and the development of hepatorenal syndrome (see Chapter 94 ) in patients with acute-on-chronic liver injury (see Chapter 74 ), including postoperatively.

Commercial and Industrial Agents

Among the tens of thousands of chemical compounds in commercial and industrial use, several hundred are listed as causing liver injury by the National Institute for Occupational Safety and Health, as published in their Pocket Guide to Chemical Hazards . The National Library of Medicine maintains a database of chemical toxins in its Toxicology and Environmental Health Information Program, as do other sources. ^, Table 89.3 lists the various chemical classes associated with hepatotoxicity as a primary toxic effect.

Toxic exposure to chemical agents occurs most often from inhalation or absorption by the skin and less often from absorption by the GI tract after oral ingestion or through a parenteral route. Because most chemical toxins are lipid soluble, when absorbed they can easily cross biological membranes to reach their target organ(s), including the liver. ^{, ,} Hepatotoxic chemical exposure (as with carbon tetrachloride [CCl ₄ ] and phosphorus) usually results in an acute cytotoxic injury that typically consists of 3 distinct phases, similar to those observed after an acetaminophen overdose (see Chapter 88 ) or ingestion of toxic mushrooms (see later) ( Table 89.4 ). ^, Less commonly, acute cholestatic injury may occur. Many chemicals (e.g., vinyl chloride) are also carcinogenic, and hepatic malignancies have been part of the clinicopathologic spectrum of chemical injury (see Chapter 96 ) ( Box 89.3 ). ^, Although liver injury is the dominant toxicity for some agents (see Table 89.3 ), hepatic damage may be only one facet of more generalized toxicity for other agents.