Background In children, paracetamol overdose due to deliberate self-poisoning, accidental exposure or medication errors can lead to paediatric acute liver failure and death. In Australia and New Zealand, the nature of ingestion and outcomes of paracetamol-associated paediatric acute liver failure have not been described.
Objective To describe the nature and outcomes of paracetamol-associated paediatric acute liver failure.
Design Retrospective analysis of paracetamol-associated paediatric acute liver failure cases presenting 2002–2012.
Setting New Zealand and Queensland Paediatric Liver Transplant Services.
Results 14 of 54 cases of paediatric acute liver failure were attributed to paracetamol, the majority were secondary to medication errors. 12 of the 14 children were under the age of 5 years. Seven children received doses in excess of 120 mg/kg/day. Many of the other children received either a double dose, too frequent administration, coadministration of other medicines containing paracetamol or regular paracetamol for up to 24 days. Three children underwent transplant. One of these and one other child died.
Conclusions In Australia and New Zealand, paracetamol overdose secondary to medication errors is the leading cause of paediatric acute liver failure. A review of regional safety practices surrounding paracetamol use in children is indicated.
- Drug Abuse
Statistics from Altmetric.com
What this study adds?
Medication errors in relation to paracetamol are a major problem for parents and caregivers in Australia and New Zealand.
Paracetamol overdose due to medication errors is one of the main causes for paediatric acute liver failure in Australia and New Zealand.
What is already known?
Inappropriate paracetamol ingestion is the leading cause for contacting Poisons Information Centres in Australia and New Zealand.1
In children, paracetamol overdose can lead to paediatric acute liver failure and death.
In the UK and North America, intentional paracetamol overdose is the leading known cause of paediatric acute liver failure.
Paracetamol (acetaminophen) is an analgesic and antipyretic agent commonly used during childhood illness by parents and physicians worldwide. In Australia and New Zealand, paracetamol is a common pharmaceutical agent leading to contact with Poisons Information Centres and the most common cause for all age drug overdose-related presentation and admission to hospital.1
Children are typically brought to medical attention following one of three methods of paracetamol ingestion: deliberate self-poisoning; accidental exposure or medication error. Despite a perceived good safety profile, paracetamol overdose under any of these scenarios may cause hepatotoxicity, acute liver failure and occasional death.1–6
The true incidence of hepatotoxicity remains unknown.4 North America and the UK report paracetamol as the leading cause for paediatric acute liver failure, accounting for 14% of cases with 4% mortality.6 ,7 While most of these presentations relate to acute intentional overdose in older children, there has also been an increase in reports worldwide describing hepatotoxicity associated with inadvertent overdoses or prolonged dosing within the expected therapeutic range.2–4 ,7–10
The purpose of this retrospective study is to report the nature of ingestion, risk factors and outcomes of 14 children with paracetamol-associated acute liver failure, admitted at two paediatric liver units in Australia and New Zealand. A secondary aim is to remind prescribers, carers and health practitioners of the risks of significant hepatotoxicity with paracetamol use.
The New Zealand Liver Transplant Unit (NZLTU) at Starship Children's Hospital, Auckland, and the Queensland Liver Transplant Service (QTLS) at the Royal Children's Hospital, Brisbane, are two of four paediatric liver transplant centres in Australia and New Zealand.
Eligible patients admitted between January 2002 and December 2012 were extracted from hospital-wide databases: NZLTU Registry, Starship Clinical Coding Database, Starship Paediatric Intensive Care database, QLTS referral Registry; Queensland Metabolic Unit Database (MMS 1.2) and Health Information Management Database (DRG). Respective medical records were reviewed for demographic, laboratory and outcome information.
Patients from 1 month to 16 years of age who satisfied paediatric acute liver failure criteria were eligible for the study6:
children with no evidence of chronic liver disease
biochemical evidence of acute liver injury
hepatic-based coagulopathy (defined as prothrombin time (PT) ≥15 s or international normalised ratio (INR) ≥1.5 not corrected by vitamin K in the presence of hepatic encephalopathy (HE); or PT ≥20 s or INR ≥2.0 regardless of presence or absence of clinical encephalopathy).
Acute liver failure was attributed to paracetamol overdose in instances where there was a documented history of paracetamol ingestion that fulfilled either of the following criteria:
cumulative dose greater than 150 mg/kg ingested within 8 h1
≥2 doses ingested over a period greater than 8 h resulting in a toxic cumulative dose with no other cause identified.1
Paracetamol-associated paediatric acute liver failure was further classified as either child exploratory in instances where a young child had found medication leading to ingestion; intentional overdose if there was an objective of harm; or medication error, where medication was administered with therapeutic intent and non-accidental intent was excluded based on consensus opinion involving medical staff, social work and child protection.
The decision to waitlist children for liver transplant was based on standard guidelines including (1) progressive coagulopathy and/or (2) development of encephalopathy, refractory to medical management. Survival post-transplant was defined as alive at 3 months following surgery. Poor outcome was defined as death or liver transplant.
The Ethics and Institutional Review Committee of the Royal Children's Hospital, Brisbane, approved this study. The Ethics and Institutional Review Committee of Auckland Hospital granted institutional approval with exemption from requiring ethical approval for this study.
Fifty-four children presented with acute liver failure, of whom 14 (9F:5M) had paracetamol attributed as the underlying cause. The median age and weight at presentation of 2.6 years (IQR 1.5–3.8 years) and 13.1 kg (IQR 10–17.1 kg), respectively. Twelve of the fourteen children were under 5 years of age at presentation.
The median paracetamol dosage, which could be estimated in 13 cases, was 135 mg/kg/day (range 62–250 mg/kg/day). The duration over which paracetamol was administered prior to presentation ranged from 2 to 24 days, with a median duration of 5 days. Dosing frequency varied with patients receiving paracetamol at half hourly to six hourly intervals with a median interval of 4 h. Seven children received dosages in excess of 120 mg/kg/day (see table 1). Three other children received double doses, paracetamol 2–3 hourly or coadministration of other medicines containing paracetamol. The four remaining children all received paracetamol for more than 48 h, including one who received it for a total of 24 days and two who were given paracetamol 4 hourly. Ten of the fourteen children had definite medication errors in relations to dosage or frequency of administration. The four remaining children received paracetamol inappropriately in relation to the duration of treatment or the frequency of administration.
More than half (8/14) of the patients had a clear clinical history of a viral prodrome and all children experienced one or more symptoms of vomiting, diarrhoea or poor oral intake prior to admission. Comorbid viral infection was subsequently proven via nasopharyngeal aspirate or serology in three cases. Five children had been recently started on penicillin-based antibiotics. Other factors likely to exacerbate liver injury, including malnutrition and coingestion of hepatotoxic medication, were not observed.
The admission INR was high in paracetamol-associated paediatric acute liver failure with median value of 5.4, whereas serum bilirubin at admission was low with median value of 36 μmol/L. The serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were high with a median ALT 6600 IU/L and AST 8800 IU/L.
HE was present in nine children presenting with paracetamol toxicity. HE could not be staged in three due to intubation prior to transfer or incomplete data. HE did not predict poor outcome in children with paracetamol-associated paediatric acute liver failure.
Overall, 10 of the 14 children survived without requiring transplantation, 1 died without transplant and 3 underwent liver transplantation with 2 surviving post-transplant.
The number of paracetamol-containing oral preparations available in New Zealand totalled 210 and in Queensland 121 products, of which liquid cough/cold formulations accounted for 28 and 26 preparations, respectively. Between 2009 and 2012, the New Zealand Poisons Information Centre received an average of 804 phone calls per year specifically relating to child paracetamol ingestion. Of these, medication errors accounted for 28% of phone calls and related to children of median age 23 months, who had reportedly ingested liquid preparation paracetamol in 77% of instances (see figure 1). Over the same period, Queensland's Poisons Information Centre averaged an annual rate of 998 phone calls. Medication errors accounted for 24% of phone calls and concerned children of a median age 41 months, who had ingested a liquid formulation in 78% of instances (see figure 1).
Paracetamol toxicity comprises the single most commonly identified cause of paediatric acute liver failure presenting to two liver transplant units in Australia and New Zealand. In this study, children under 5 years of age were at greatest risk and the mode of toxicity was medication error in at least 10 cases. Common errors in administration that resulted in paediatric acute liver failure included (1) exceeding recommended dosages, with seven children receiving doses greater than 120 mg/kg/day; (2) increased dosing frequency, with many carers not adhering to recommended 4–6 hourly frequencies; and (3) prolonged duration of therapy, with 13 children administered regular paracetamol in excess of 48 h, despite product recommendations for medical review at 48 h prior to continuing.11 Factors reported by carers as reasons leading to dosing errors included (1) failure to read and understand labelled instructions, (2) using incorrect measuring devices, (3) using an alternative concentration of paracetamol than the one prescribed, (4) using adult regular strength tablets and (5) not recognising paracetamol to be contained in other cough/cold preparation. Overall, these findings highlight the need to address regional practice with regards to dosing instructions, address packaging information and education on the potential risk of toxicity in sick children.
To fully understand the cause of paediatric paracetamol-associated acute liver failure, consideration should also be given to host factors that modulate the balance between paracetamol bioactivation and detoxification.1–5 In this study, more than half of the patients in the paracetamol group presented with a history of viral illness. Although fever is a stated indication for paracetamol use, animal studies have shown that fever may increase the metabolism of paracetamol via the induction of mixed function oxidases towards the toxic metabolite N-acetylparabenzoquinoneimine (NAPQ1).2 ,11 ,12 Furthermore, associated vomiting, diarrhoea and/or poor oral intake, as observed in our entire paracetamol cohort, are hypothesised to cause acute depletion in hepatocellular sulfate and glucuronide precursors and glutathione stores, thereby increasing the risk of hepatic injury. Thus, medication errors may result in poorer outcomes compared with acute intentional overdose from a multifactorial liver injury due to (a) concurrent viral infection, (b) poor oral intake and (c) staggered paracetamol over length of time. Practitioners should therefore remain cautious in making therapeutic recommendations for using paracetamol in a viraemic child, particularly as no conclusive evidence exists of paracetamol's efficacy over placebo as an antipyretic agent.11 ,13
While nomograms provide physicians with a dosing threshold at which hepatic injury occurs in single paracetamol overdose, establishing similar thresholds in the setting of chronic or recurrent overdose due to medication errors has proved difficult and paracetamol levels may be falsely reassuring, as observed in this study (see table 1). Pharmacokinetic properties are altered in staggered overdoses, which can be further compounded by the coadministration of over-the-counter preparations containing paracetamol or drugs capable of inducing hepatic microsomal enzymes.1 In addition to the uncertainty of drug dosages and timing provided by historians, the kinetics of tablet, capsule and liquid preparations also adds to this complexity, with peak concentrations within 1–2 h for standard tablets or capsule formulations and within 30 min for liquid preparations.1 ,11 Another consideration and limitation of this study was the possibility of having attributed events to adverse drug reactions when they are in fact a consequence of substandard preparations.14 ,15 This not only highlights the importance of a thorough medication history but the need to ascertain data regarding specific brands administered to patients and relevant pharmacovigilance reports denoting substandard products.
A limitation of this study is that it was a retrospective review. It is therefore possible that several patients also received other paracetamol-containing medicines and that this was not documented in the notes. A review of regional safety practices surrounding paracetamol use in children is indicated, addressing (1) dosing guidelines in the setting of an unwell, young child; (2) quality and labelling of medicines and (3) information provided to carers in relation to paracetamol use in children.
The authors acknowledge the data contribution of the New Zealand National Poisons Centre and Queensland Poisons Information Centre. We thank Peter Reed and the Children's Research Centre, Starship Children's Hospital, New Zealand, for their assistance in data analysis. We acknowledge Dr J Khoosal for assistance with graphical art.
Contributors JRB and HME jointly conceived the study; JR designed the study, collected data, analysed data and wrote the paper; JRB, HME and PJL discussed the results and implications, commented and edited the manuscript at all stages.
Competing interests None.
Ethics approval Queensland Children's Health Services Human Research Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.