Fatal hepatotoxicity due to sevoflurane use in a paediatric patient after aortic repair: reality or fiction? A case of pharmacovigilance

Abstract

Hepatotoxicity secondary to exposure of volatile anaesthetics is an exceptional finding, but its clinical interest depends on their frequent use, unpredictable appearance and potential severity. Halothane is the volatile anaesthetic most frequently involved in the development of liver dysfunction, especially after re-exposure. Sevoflurane has rarely been related to this life-threatening complication. We present the case of a 1-year-old girl who had undergone previous surgery for closure of a patent ductus arteriosus, and who developed severe acute hepatitis and died after undergoing surgical repair of an aortic isthmus narrowing by means of general anaesthesia with sevoflurane. Other possible causes of liver failure were excluded. This adverse event was classified as serious and was included in the national and European pharmacovigilance network, with the aim of preventing dangerous effects on patient health in clinical practice, by contributing to the enrichment of the literature.

Background

Sevoflurane is the most worldwide-used volatile anaesthetic agent, with a low toxicity profile. It is an inhalation anaesthetic based on halogenated methyl isopropyl ether which produces a rapid induction and awakening phase. The minimum alveolar concentration (MAC) of sevoflurane decreases with age. Sevoflurane produces loss of consciousness, reversible abolition of pain sensitivity and motor activity, a decrease in autonomic reflexes, and respiratory and cardiovascular depression. These effects are dose dependent. Sevoflurane has a low blood/gas partition coefficient (0.65) which results in a rapid awakening from anaesthesia. While most halogenated volatile anaesthetics (eg, halothane, enflurane, isoflurane, desflurane) undergo cytochrome P450 metabolism and produce trifluoroacetate (TFA) proteins, which can act as haptens and induce immune-mediated hepatotoxicity,¹ in contrast sevoflurane does not develop TFAs and therefore is considered safer than other halogenated agents regarding hepatotoxicity. Acute hepatotoxicity secondary to sevoflurane exposure has been rarely reported, especially after cardiac surgery and in children.^2–5 In sevoflurane’s summary of product characteristics (SCP), acute hepatotoxicity is classified as a not known adverse reaction. This case report aims to analyse and to consider acute hepatotoxicity by sevoflurane as an exceptional adverse reaction that can occur after sedation in cardiac surgery in children. Indeed, the report by the hospital pharmacy of an adverse reaction in this type of patient population could help to preserve their health by preventing possible adverse events resulting from the administration of sevoflurane.

Case presentation

We describe the case of a 1-year-old female patient (weight 8.8 kg) with a known diagnosis of aortic isthmic narrowing and who was scheduled for aortic repair. She was under treatment with atenolol 6.25 mg twice daily by mouth. Her past surgical history included a previous patent ductus arteriosus closure through a left thoracotomy at 6 months of age, using general anaesthesia with sevoflurane. Preoperative tests showed normal liver function (with international normalised ratio (INR), albumin, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) within normal limits) with negative serology for hepatitis B and C virus as well as for HIV infection. Anaesthetic premedication included glycopyrrolate 0.01 mg/kg + midazolam 0.1 mg/kg intravenously (IV). General anaesthesia was induced with fentanyl 2 μg/kg + midazolam 0.1 mg/kg + vecuronium 0.1 mg/kg IV, and anaesthetic maintenance was achieved with 2% sevoflurane (~0.9 MAC) and a fresh gas flow of 3 L/min. The aortic isthmus was repaired in euthermia, without cardiopulmonary bypass (CPB); the proximal and distal aorta was temporarily clamped (35 min). The surgery was uneventful (total surgical time 4 hours; immediate postoperative lactacidaemia 1.8 mmol/L) and the patient was transferred to the intensive care unit in a stable haemodynamic condition with low inotropic support (milrinone 0.75 μg/kg/min + glyceryl trinitrate (nitroglycerin) 6 μg/kg/min), where she was extubated 8 hours after the surgical procedure. On postoperative day (POD) 3 the patient developed a sudden generalised seizure which required reintubation. Gastrointestinal bleeding was found when positioning the nasogastric tube. On POD 4 the patient’s general condition deteriorated, with systemic hypotension, appearance of jaundice, and bleeding.

Investigations

The emergent laboratory tests revealed severe hypoglycaemia (<50 mg/dL), eosinophilia (1980/µL) and thrombocytopenia (<50×10⁹/L). A brain tomography performed on POD 3 did not show any acute abnormalities. Hepatic tests on POD 4 showed ALT >1000 U/L, AST >750 U/L, lactate dehydrogenase (LDH) >2150 U/L, alkaline phosphatase (ALP) 332 U/L and haematic ammonium 240 μmol/L. Hepatic biopsy was not performed because of the absence of a reliable pathological unit in the hospital.

Treatment

The treatment was empiric and supportive; the generalised seizure was managed by means of an initial dose of benzodiazepine (diazepam 0.2 mg/kg IV) and, after reintubation, the patient required sedation and myorelaxation according to the paediatric intensive care unit protocol.

Progressive cardiocirculatory shock required increasing inotropic support, which was ineffective. The hypoglycaemia was treated with 50% dextrose infusion, whereas bleeding due to thrombocytopenia was managed by means of platelet infusion.

Outcome and follow-up

Despite maximal supportive treatment (described above), the patient underwent a rapid irreversible cardio-circulatory shock and died on POD 6. Consequently, the patient’s death was classified as a serious adverse event and, in accordance with internal procedure rules, it was reported by the company hospital pharmacist within the AIFA (Italian Medicines Agency) pharmacovigilance network,⁶ the system used by healthcare professionals to report adverse events that occur as a result of the use of medicines in clinical practice.

Discussion

Halogenated anaesthetic agents have been associated with perioperative liver dysfunction and injury. Most cases of hepatotoxicity reported involve halothane exposure. In the literature, there have been no reports of known fatal cases of acute hepatotoxicity secondary to sevoflurane exposure, especially in children after cardiac surgery.⁷ For example, a relevant non-fatal case of acute hepatotoxicity caused by sevoflurane was reported during neurosurgical treatment in an 11-year-old child.³ Another case of non-fatal acute hepatotoxicity associated with sevoflurane was reported during a cardiosurgical intervention in a 67-year-old woman.⁸ Cross-sensitisation between halothane and other volatile agents has been previously reported with isoflurane, enflurane and desflurane.^9–11 Acute liver dysfunction in a postoperative patient can be due to several causes: ischaemic (especially after surgery, acute shock or cardiopulmonary arrest), toxic (eg, drugs up to 25% of fulminant hepatitis), infectious (viral or bacterial) or traumatic. CPB is also a recognised cause of hepatic dysfunction. However, in this case the patient did not undergo CPB. In the paediatric population, the youngest patients (especially <6 months of age) showed hepatic damage more frequently (probably due to liver immaturity or a lower adaptation threshold to metabolic and haemodynamic changes after cardiac surgery). Among different types of cardiac disease, the Fontan operation and operations for transposition, coarctation or complete atrioventricular canal repair are more frequently prone to hepatic ischaemia.² Clinically, systemic hypoperfusion after major cardiac surgery can be easily detected by means of plasmatic lactate and venous saturation; all of these parameters were negative so we excluded this cause. Thus, the temporal relation between exposure and liver injury is consistent with sevoflurane hepatotoxicity as a diagnosis.

As previously stated, other possible perioperative aetiologies were excluded, such as pre-existing liver disease, new onset of biliary obstruction and cholangitis, systemic viral infection, septicaemia, adverse reactions to other medications given during the perioperative period, and various metabolic and immunogenic diseases. Clinical signs of hepatic failure secondary to volatile halogenated anaesthetics include malaise, fever, encephalopathy, hypoglycaemia, eosinophilia, jaundice, and increases in liver enzymes (all found in our patient). Hepatic damage due to these anaesthetics is primarily related to re-exposition. Even if sevoflurane does not produce TFAs, the mechanism of its toxicity might be related to compound A (fluoromethyl-2,2-difluoro-1-(trifluoromethyl) vinyl ether), which is a consequence of the metabolism between sevoflurane and carbon dioxide. This product increases during anaesthesia with low flows of gases, using closed circuits and high concentrations of sevoflurane. Compound A is involved in the production of antigens which can lead to immunogenicity after re-exposition; this has been shown in animal experiments.¹² According to recent evidence,⁸ an immune response to compound A takes the same time as the response to halothane, so that compound A can be involved in hepatic damage through an immunologic mechanism similar to other volatile halogenated anaesthetics. We know that a definitive diagnosis would have been reached by means of a liver biopsy, but it was not possible to perform such a biopsy due to the available hospital resources.

To the best of our knowledge, this is the first case that shows acute hepatotoxicity secondary to sevoflurane exposure in a child who had undergone cardiac surgery. The mechanism might be related to the immunogenic effect of sevoflurane metabolites. Postoperative laboratory tests have excluded viral, metabolic and organic aetiologies. This report feeds the flow of information that constitutes the pharmacovigilance network; in this way healthcare professionals can be made aware of the importance to carefully monitor for any adverse event related to hepatic toxicity which can manifest during treatment with sevoflurane in paediatric cardiac surgery, with the aim of preventing dangerous effects on patient health and also to contribute and enrich the literature surrounding the use of sevoflurane. Anaesthetists must therefore carefully evaluate the past perioperative clinical history and the technology of anaesthetic gas release of paediatric patients undergoing redo cardiac surgery, to avoid unexpected drug-related adverse events.

Learning points.

• The importance of the pharmacovigilance network for enriching the literature on the use of medicines.

• The primary role of the hospital pharmacist as a reporter of adverse events that may occur in clinical practice.

• The role and expertise of the hospital pharmacist in supporting physicians in their clinical activity.

Data availability statement

All data relevant to the study are included in the article.

Ethics statements

Patient consent for publication

Not required.