Title: The debate on paracetamol hepatotoxicity continues
Abstract: Björn Fischler Claudia Mandato Pietro Vajro Nearly 1300 probable cases of paediatric severe acute hepatitis of unknown origin were reported in 37 countries or regions between the first alerts in late 2021 and September 2022.1 The 2022 outbreak mainly affected previously healthy children under 6 years of age. Inexplicably, there were no consistent epidemiological associations between the cases and the incidence did not increase above the expected baseline in a number of countries. Most children recovered, but some developed liver failure, 55 required liver transplants and 29 died. Fortunately, cases began to stagnate after the 2022 outbreak and decreased to the levels seen before the COVID-19 pandemic.1 Several possible causes have been suggested for the outbreak and largely dismissed, including known major and minor hepatotropic hepatitis viruses and toxicity. The severe acute respiratory syndrome coronavirus 2 was discounted because most of the patients tested negative and COVID-19 vaccines were ruled out because most of the patients were too young to be eligible. Although some data seemed to point to an adenovirus, particularly adenovirus type 41, the evidence to support this was inconclusive, as only about 50% of patients tested positive in diverse series. This suggested that other factors may be involved. Most experts supported the immunity gap theory, which suggests that children were more susceptible to infectious agents due to social isolation and using masks during COVID-19 lockdowns and other restrictions.1 In March and July 2023, teams in the UK and USA independently reported that another virus, the adeno-associated virus type 2, was involved. However, their data showed that patients also appeared to be infected with another virus at the same time, including the adenovirus type 41, Epstein–Barr virus, human herpesvirus 6 or severe acute respiratory syndrome coronavirus 2. The authors also confirmed that the timing of the outbreak had probably been related to the global easing of COVID-19 pandemic restrictions after periods of relative isolation. This increased the risk of developing serious diseases, particularly if human leukocyte antigen class II status, associated with autoimmune disease, was present.2, 3 The remaining question is why this severe acute hepatitis outbreak only occurred in some children and some parts of the world. In this issue of Acta Paediatrica, an interesting Different View paper by Hoption Cann states that there is no strong evidence to date to unquestionably exclude the role of paracetamol in this outbreak of severe paediatric acute hepatitis of unknown origin.4 It is difficult to subscribe to this theory, considering the extensive global use of paracetamol in paediatrics, as it is regarded as a reliable and safe antifebrile. However, perhaps we should consider some other, possibly underestimated, issues related to paracetamol drug-induced liver injuries. In fact, the American Academy of Pediatrics (AAP) warns that many existing preparations contain paracetamol and the simultaneous use of more than one of these products can be dangerous. Because early symptoms of paracetamol toxicity are non-specific, the AAP advises healthcare providers to include paracetamol toxicity in the initial differential diagnosis for many diseases. This should particularly be the case with unexplained hepatic dysfunction, especially in the presence of certain situations that may increase the risk of paracetamol toxicity. These include prolonged fasting and concomitant viral infections, which are two possible conditions observed in children with paediatric severe acute hepatitis of unknown origin.5 This warning sits well with several studies of acute viral-like febrile illness in children who developed fulminant hepatic failure after exposure to multiple unintentional, miscalculated, supratherapeutic doses of paracetamol.6 Not being aware of this danger may result in delayed presentations to emergency departments. This can then happen after the window for the optimal administration of an N-acetyl-cysteine antidote has passed and symptoms of fulminant hepatic necrosis are already evident. Testing for paracetamol protein adducts has proven to be very valuable for identifying exposure to paracetamol toxicity in adults and children. Despite this, such testing was not referred to in any of the reports on outbreaks of paediatric severe acute hepatitis of unknown origin. With this in mind, we could speculate that adeno-associated virus type 2 infections, with or without other viruses such as adenovirus type 41, became more virulent due to previous COVID-19 restrictions and the related immunity gap. This could have paved the way for drug-induced liver injuries, due to young and otherwise healthy children with a possible specific human leukocyte antigen type receiving more than the recommended dose of paracetamol. This then predisposed them to hepatic immune-mediated damage. However, one large question remains. If paracetamol was the culprit, why did it result in clusters of severe drug-induced liver injury that only occurred in some infected children and in some parts of the world? We also need to question why cases slowed down, stagnated and probably disappeared. Could other paracetamol-related hypotheses not considered by Hoption Cann's paper4 come into play and possibly complete the puzzle? We think it would be helpful to discuss two plausible hypotheses about the relentless problem of falsified medicines, also known as counterfeit drugs. The scientific literature has been increasingly discussing the unpredictable public health risk of substandard or falsified medicines containing an incorrect amount of the active pharmaceutical ingredients or solvents. All types of medications appear to have been targeted over the years, particularly during the COVID-19 pandemic. This has had a serious impact on the global health of both adults and children. No deaths have been directly attributed to substandard and falsified medicines in Europe and the UK, but there have been deaths attributed to these in other regions of the world. However, reliable statistics about the danger of these drugs are not available, due to their illegal nature and the relentless globalisation of the pharmaceutical market and online sale of all types of medical products. Some researchers therefore believe deaths caused by counterfeiting have gone unnoticed or biased.7, 8 The most frequently reported fatalities from falsified medicines relate to the inadvertent or deliberate use of cheap diethylene glycol as a paracetamol solvent in liquid formulations for children, instead of the more expensive propylene glycol. A review found that 17 liquid medicines were probably contaminated with diethylene glycol, but only paracetamol syrup contaminated with this was epidemiologically associated with disease. There could be three possible reasons for this finding. The paracetamol could have intensified the diethylene glycol toxicity, since both are toxic and can modify diethylene glycol metabolism The diethylene glycol concentration in the paracetamol syrup could have been higher than that of the other products. Lastly, clinical toxicity could have been more likely, because associated gastroenteritis contributed to the increased absorption or otherwise enhanced toxic effects of the diethylene glycol.9 In August 2023, the World Health Organization identified liquid medication contaminated with substandard paracetamol and chlorpheniramine maleate in its Eastern Mediterranean region. The syrups, which were manufactured in India, exceeded the safety limit for both chemicals, which is ≤0.10%, and were linked to the deaths of at least 89 children in 2022.10 It has been reported that glycol and diethylene glycol were considered responsible for these deaths, primarily due to their nephrotoxicity. The question is whether there is any evidence that falsified medicines could have caused cases of paediatric severe acute hepatitis of unknown origin. Data on diethylene glycol and the liver are very limited, but one study showed that accidental intravenous infusions of diethylene glycol worsened liver damage in patients with preexisting liver disease. Although most of the alterations in the patients' baseline liver functions were mild and transient, some cases of severe liver damage, secondary to diethylene glycol, were seen in those with concomitant renal failure.11 On the other hand, jaundice was common in patients poisoned with oral diethylene glycol in the USA in 1937. In a Nigerian study hepatomegaly was found more commonly in renal failure that was related to diethylene glycol (53%), than was not (33%).12 These findings may suggest that oral diethylene glycol poisoning also results in some liver damage in patients without underlying liver disease.13 As recently reviewed by Singh et al. glycol ethers hepatically metabolised to their respective toxic metabolites (e.g. diglycolic acid) may cause hepatic injuries such as centrilobular degeneration, necrosis, and ballooned hepatic cells. Hepatotoxicity is primarily through mitochondrial damage, production of reactive oxygen species, and decreased cellular metabolism with TNF-α mediation.14 Reaching a final conclusion on the hepatotoxicity of paracetamol contaminated with diethylene glycol is difficult, as ethylene glycol poisoning has not been considered to be a contraindication for donor livers used in transplants.15 These data on diethylene glycol liver injury prompts two questions. Firstly, did the adeno-associated virus type 2, plus the adenovirus and/or other viruses cause some minor hepatitis that was then aggravated by inadvertent paracetamol overdosing and responsible for a drug-induced liver injury? Secondly, were local clusters of paediatric severe acute hepatitis of unknown origin caused by substandard or falsified paracetamol containing diethylene glycol and/or other contaminants and/or uncontrolled higher concentrations of the active drug in children predisposed by particular human leukocyte antigens? (Figure 1). In conclusion, the Different View paper by Hoption Cann appearing in this issue advises that not enough attention was given to the possibility that severe acute hepatitis of unknown origin in young children was associated with the use of paracetamol in regular doses.4 However, given the peculiar epidemiological development of the 2021/22 outbreak and the quite convincing available data on virology and predisposing factors, we would certainly not suggest that paracetamol should be avoided if indications and dosages were correct. On the other hand, Dr. Cann's view is commendable for giving the opportunity to focus on the often overlooked problem of possible hepatotoxicity of drugs generally believed to be safe. More detailed attention to the type of paracetamol formulations, and factors related to its toxicity, are needed in children with severe acute hepatitis of unknown origin and similar future circumstances. Björn Fischler: Conceptualization; data curation; writing – original draft; writing – review and editing. Claudia Mandato: Data curation; writing – original draft; writing – review and editing. Pietro Vajro: Conceptualization; supervision; writing – original draft; writing – review and editing. The authors declare no conflicts of interest.