A Prospective Study of the Incidence of Drug-Induced Liver Injury by the Modern Volatile Anaesthetics Sevoflurane and Desflurane

Bridget Bishop; Nicholas Hannah; Adam Doyle; Francesco Amico; Brad Hockey; David Moore; Siddharth Sood; Alexandra Gorelik; Danny Liew; Dolores Njoku; Amanda Nicoll


Aliment Pharmacol Ther. 2019;49(7):940-951. 

In This Article

Abstract and Introduction


Background: Volatile anaesthetics are known to cause drug-induced liver injury, a hepatotoxic reaction characterised by antibodies to trifluoroacetylated lipid and protein adducts and cytochrome p450 2E1. The incidence of volatile anaesthetic drug-induced liver injury from older agents has been described, but modern agents have not been prospectively studied.

Aim: To determine prospectively the incidence of volatile anaesthetic drug-induced liver injury from sevoflurane and desflurane.

Methods: Adult surgical patients with a predicted post-operative stay of at least 4 days were recruited. If volatile anaesthetic was administered, liver biochemistry was performed regularly. Medications, observations and other investigations were documented. Patients with abnormal liver biochemistry were classified as likely volatile anaesthetic drug-induced liver injury or not based on clinical assessment, Roussel Uclaf Causality Assessment Method score, and the absence of other likely pathology. Some patients were also tested for antibodies to both trifluoroacetylated lipid and protein adducts, and cytochrome p450 2E1.

Results: A total of 209 patients were recruited, of which 121 were included for analysis. Post-operative liver biochemistry was abnormal in 62 patients (51.2%); further classified as not volatile anaesthetic drug-induced liver injury in 47 cases (38.8%), and likely volatile anaesthetic-drug induced liver injury in 15 cases (12.4%). Of the likely volatile anaesthetic drug-induced liver injury patients, only one had severe disease with alanine transaminase greater than five times the upper limit of normal, while four cases had moderate disease with alanine transaminase greater than three times the upper limit of normal. Thus, the incidence of clinically significant volatile anaesthetic drug-induced liver injury was 4.1%. No risk factors were identified.

Conclusions: Volatile anaesthetic drug-induced liver injury from modern agents seems to be as common (4.1%) as previously reported with older agents (3%), and may identify patients at risk of severe acute liver injury with subsequent re-exposure.


Drug-induced liver injury, or drug-related hepatotoxicity, is injury to the liver due to exposure to a drug, as demonstrated by impaired liver function tests.[1] Volatile anaesthetics are commonly used inhaled halogenated agents that are a recognised cause of drug-induced liver injury.[2] Volatile anaesthetic drug-induced liver injury (VA-DILI) typically causes a rise in the alanine transaminase (ALT) 2 to 14 days post-operatively,[3] and can manifest as a spectrum of disease, from asymptomatic derangement in liver biochemistry,[4] to acute severe hepatitis,[4–16] and rarely, fatal hepatic necrosis.[17,18]

The most well-known type of VA-DILI is 'halothane hepatitis'.[19] Halothane was widely used until it was found to cause liver injury in up to 24.4% of patients,[20,21] and has since been replaced by newer volatile anaesthetics. Volatile anaesthetics are primarily metabolised by the cytochrome p450 enzyme, CYP2E1, which generates intermediates that can covalently bind to native cellular components, creating trifluoroacetylated lipid and protein adducts (TFA).[22] Immune-mediated drug-induced liver injury has been demonstrated in patients following modern volatile anaesthetics, but is less common than with halothane or enflurane.[2,22]

Modern volatile anaesthetics that are widely used include isoflurane (introduced in 1981), desflurane (1992) and sevoflurane (1995). These newer agents are thought to be much safer, with a lower risk of causing VA-DILI, since they have a lower rate of hepatic metabolism and thus generate less TFA.[20,23] In fact, sevoflurane is thought not to undergo hepatic oxidative metabolism at a clinically relevant level.[24] However, there are case reports,[4–18] as well as retrospective data,[2] that suggest that VA-DILI with modern volatile anaesthetics can still occur.

The diagnosis of VA-DILI is difficult and requires exclusion of other causes of deranged liver biochemistry, such as infection, ischaemia and other drug-induced or autoimmune phenomena. VA-DILI is characterised by a rise in ALT in the 2 to 14 days post-exposure, and is often accompanied by fever and eosinophilia in severe cases.[25] Liver injury from volatile anaesthetics is due to a combination of toxicity and autoimmunity, with the former occurring early in the post-exposure period, and the latter manifesting later.[26] Most cases resolve, but evolution into fulminant and sometimes fatal hepatic failure, as well as chronic liver disease, has been described.[4]

The aim of this study was to prospectively estimate the incidence of VA-DILI from the newer generation volatile anaesthetics, sevoflurane and desflurane, using accepted methods to diagnose drug-induced liver injury, including careful case review, exclusion of other possible aetiologies, and calculation of the Roussel Uclaf Causality Assessment Method (RUCAM) score.[27] Appreciation of VA-DILI with modern agents may aid hepatologists in recognising possible VA-DILI, and help anaesthetists identify those at risk and minimise the chance of severe VA-DILI occurring with rechallenge. This study was a pilot, proof of concept study, to determine if a prospective study of drug-induced liver injury was possible, and to guide future work in this field.