Paediatric acute hepatitis and therapeutic doses of paracetamol

While the majority of acute hepatitis cases of unknown origin have been reported in the United Kingdom and United States, elevated case numbers have been observed internationally with reports in nearly 40 countries.1 The UK Health Security Agency's (UKHSA) series of reports2 on the outbreak have extensively reviewed possible links to adenovirus infection. They also noted that ‘approximately three-quarters of respondents in data for England mentioned paracetamol use’, yet surmised that ‘detection of paracetamol is likely to be related to appropriate therapeutic use…which would not be a concern’.2 Even though this cluster has been characterised as an isolated event, there is evidence to suggest that this may not be the case. A recent study by Jagadisan et al.3 examining the association between adenovirus infection in children and indeterminate acute liver failure over the past 6 years provides evidence that this may be an ongoing phenomenon. They observed that the incidence of this condition correlates with the varying incidence of adenovirus infection. From their findings, they further contend that an ‘exclusive role of human adenovirus infection in the causation of this outbreak of acute hepatitis seems unlikely’, suggesting that other co-factors should be explored. Thus, the suggestion by the UKHSA to establish an ongoing surveillance scheme for non-A–non-E hepatitis2 appears warranted. Although liver injury could be expected following an overdose, hepatitis following doses within the therapeutic range could also occur. Of note is a previous outbreak of fulminant hepatic failure in children (ages 3 months to 5 years) that occurred over 25 years ago.4 The hepatitis outbreak followed acute viral infections in all children, which ‘represented at least a threefold increase in the yearly incidence’.4 All cases had received paracetamol, but none were above the therapeutic range—yet the authors suspected that paracetamol may have been causative. In an earlier report on a measles outbreak in Israel, investigators followed 118 military personnel (18–21 years of age) of which 56% were noted to have elevated liver function tests.5 All personnel had been treated with either paracetamol or dipyrone therapeutically for symptoms of the infection. While all individuals received medication within the therapeutic range, only 15% of those treated with dipyrone had elevated AST and ALT. In contrast, in those treated with paracetamol, 65% had AST and 58% ALT elevation. The relative risk for abnormally raised AST was 3.8 (95% CI 1.03–13.86), and for ALT elevation 4.2 (95% CI 1.16–15.43). Examining paracetamol protein adduct levels has also been shown to be useful marker of liver toxicity. Jiang et al.6 conducted a prospective study examining adduct levels in 181 children under 19 years who were expected to receive paracetamol at therapeutic doses during hospitalisation. It was observed that compared to other admission diagnoses, patients with suspected infections had higher adduct concentrations. Although only 2% had adduct concentrations above 1.0 nmol/mL (the toxicity cut point for adducts), this is not a trivial number since all subjects received doses within the therapeutic range. Finally, it was observed that there was a dose-dependent association for paracetamol protein adducts, with the infant adduct profile showing a more rapid adduct elevation relative to other age groups. Thus, paracetamol use during infections may put patients at a higher risk and the accumulated dose should be considered in addition to whether paracetamol was used or not. With respect to the current outbreak, could in rare cases some synergism occur between paracetamol and adenovirus infection? Unfortunately, it is rarely questioned whether a recommended dose during an infection could precipitate hepatitis. To illustrate, a recent case report described a 35-year-old patient who developed acute hepatitis following a febrile adenovirus infection.7 She had taken ibuprofen and then a combination of paracetamol, dextromethorphan and phenylephrine in the daytime and paracetamol, dextromethorphan and doxylamine succinate at night for 4 days. Yet in the discussion, the event was solely attributed to the adenovirus infection without any comment on a possible role for paracetamol or the paracetamol/drug combination. The challenge is that if a patient were to receive an excessive dose of paracetamol and develop hepatitis, this would be attributed to a paracetamol overdose; however, if a patient was administered a dose within the therapeutic range and developed hepatitis, the cause would be labelled as undetermined. Not so long ago there were outbreaks of encephalopathy with fatty liver in children, usually following flu-like or gastrointestinal symptoms. When Reye and colleagues first published their case series on this mysterious condition, they were as lost then as we are now on possible causes; yet shortly after, Giles8 published a compelling argument that salicylates may be the trigger. Among his patients and those he obtained from colleagues, he found 15 of 31 cases had used aspirin, while a further six were inconclusive. He asserted ‘the true incidence of aspirin administration may well be higher, since in three of my own cases only close and direct questioning elicited the fact that it had been given’. Underreporting is not surprising as a US survey concerning the most popular paracetamol-containing analgesic found only 31% knew that paracetamol was the active ingredient.9 Therefore, paracetamol use in cases of acute paediatric hepatitis may be much higher than reported. Biological samples were also analysed in the UKHSA study; however, they concluded it ‘appears highly unlikely that paracetamol is causative’ due to samples that were most strongly positive being possibly due to hospital use, implying again that only excessive doses could cause hepatitis.2 In contrast, Pinsky et al.10 examined the relationship between Reye syndrome and aspirin dose administered during the antecedent illness. They found that ‘larger doses of aspirin (though generally, as in previous studies, well below the recommended upper limit for antipyretic therapy of 80 mg/kg per day) administered during the antecedent illness are related to an increased risk of disease’. Before its link to Reye syndrome, aspirin was not known to cause the adverse effects associated with this condition. In contrast, paracetamol remains the most frequent cause of paediatric acute liver failure11 and thus should remain a key suspect in this outbreak until conclusive evidence demonstrates otherwise. In addition to adenovirus being a possible factor, reports have also implicated adeno-associated virus 2 (AAV2), found in 81–96% of children investigated in several smaller but more recent studies conducted in the United Kingdom and United States.12 This is a replication-deficient virus, and to generate new AAV2 particles, requires ‘helper viruses’, such as adenovirus, human herpes virus 6 (HHV-6) or Epstein–Barr virus (EBV). The latter two viruses were also present in some patients.12 In the UKHSA investigations, 65.9% of UK cases tested positive for adenovirus and 71.1% of cases in England.2 Nevertheless, at least three-quarters of cases in England had taken paracetamol, and regrettably, these series of reports have directed investigations away from one of the most common factors linking these cases together. During the previous outbreak of fulminant hepatitis,4 an author who reported a similar case of severe hepatitis following an acute viral infection and ‘appropriate dosages’ of paracetamol, asserted that the role for paracetamol ‘warrants further investigation’.13 If we had heeded this warning over a quarter of a century ago, could we have avoided the current outbreak? In this light, it seems premature to rule out paracetamol when present in the majority of cases. Furthermore, histological studies, when possible, should be considered to explore the presence of signs characteristic of paracetamol toxicity in the liver (i.e. centrilobular necrosis) and kidneys (i.e. acute tubular necrosis).14-16 Alternatively, one could investigate whether there is evidence of mitochondrial dysfunction,17 although acute hepatitis due to multifactorial causes may not resemble the classic histology seen in paracetamol overdose. Finally, investigation of genetic variants may provide further insights into causes of drug-induced liver injury at therapeutic doses of paracetamol18; however, more immediately, rigorous questioning is required on whether this drug was used in each case, and case–control studies should be conducted to explore the association between paracetamol and acute hepatitis—as was done to confirm the link between aspirin and Reye syndrome. The author has no conflicts of interest to declare.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.