Does the patient history predict hepatotoxicity after acute paracetamol overdose?

Transcription

1 Q J Med 28; 11: doi:1.193/qjmed/hcm139 Advance Access published on 7 January 28 Does the patient history predict hepatotoxicity after acute paracetamol overdose? W.S. WARING, O.D.G. ROBINSON, A.F.L. STEPHEN, M.A. DOW and J.M. PETTIE From the Scottish Poisons Information Bureau, Royal Infirmary of Edinburgh, 51 Little France Crescent Edinburgh, Received 8 August 27 and in revised form 4 September 27 Summary Background: Initial management of patients who were presented to hospital after acute paracetamol overdose depends on the suspected amount ingested and more than 12 g is potentially fatal. However, the validity of this approach has received comparatively little attention. Methods: The present study is sought to establish whether the stated paracetamol dose might predict systemic exposure and risk of hepatotoxicity. A prospective observational study of consecutive patients presenting to the Emergency Department due to acute paracetamol overdose was performed. Serum paracetamol concentrations between 4 and 15 h post-ingestion were compared with the Rumack-Matthew 2-line nomogram, and Introduction Paracetamol (acetaminophen) is the most common means of deliberate self-poisoning in the United Kingdom and other European countries. 1 For example, in the United Kingdom alone, paracetamol overdose gives rise to more than 7 emergency hospital attendances and causes around 18 deaths per million population annually. 1 3 The need for N-acetylcysteine is normally determined from the suspected quantity of paracetamol ingested, interval between ingestion and presentation to hospital and factors that increase susceptibility to hepatotoxicity such as chronic heavy ethanol consumption, enzyme-inducing drugs and established liver disease. 4 Risk may also be determined by the comparison of serum paracetamol concentrations to hepatotoxicity was defined by prothrombin time ratio 41.3 or alanine transaminase 51 U/l. Results: There were 987 patients, and the stated quantity of paracetamol ingested was 12 g in 475 (48.1%),412 g in 349 (35.4%) and unknown in 163 (16.5%). Ingestion of 412 g was associated with paracetamol concentration above the 2-line in 31.8% (95% CI %) vs. 3.2% ( %), P <.1 by 2 proportional test, and associated with hepatotoxicity in 6.9% ( %) vs. 1.3% (.5 2.8%), P ¼.1. Conclusions: Therefore, ingestion of412 g predicted higher paracetamol exposure and increased risk of hepatotoxicity and supports the validity of patient history in this context. the Rumack-Matthew nomogram, which originally described an exponential decay between 2 mg/dl at 4 h and 3 mg/dl at 15 h. 5 Several nomograms are used worldwide, all with minor modifications but based on the same principles, for example the 1-line represents 5% of the concentrations in the Rumack-Matthew nomogram and may be used as a lower threshold for antidote treatment in selected high-risk patients, and a 15-line is widely accepted in the United States. 6 There are important limitations to the nomogram, and this method of risk stratification cannot be relied on if patients have taken a staggered overdose, co-ingested drugs capable of delaying gastrointestinal absorption, or presented late to hospital (415 h post-ingestion). 7 Address correspondence to W.S. Waring, Scottish Poisons Information Bureau, Royal Infirmary of Edinburgh, 51 Little France Crescent, Edinburgh, EH16 4TJ,. s.waring@ed.ac.uk! The Author 28. Published by Oxford University Press on behalf of the Association of Physicians. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 122 W.S. Waring et al. Table 1 Selected clinical studies addressing the validity of the patient-reported dose with respect to serum paracetamol concentrations or risk of hepatotoxicity Author n Key Findings Prospective studies Thomas SHL et al. 8 4 Correlation between dose and 4 h level; <12 g and 512 g led to concentrations above 2-line in 3% and 2% Brotodihardjo AE et al Reported dose 41 g predictive of increased risk of hepatotoxicity Daly FF et al Positive association between stated dose and risk of hepatotoxicity after staggered overdose Schiodt FV et al High stated dose confers increased risk of hepatic injury, independent of acute or chronic ethanol intake Gazzard BG et al Stated dose correlates with severity of hepatic insult but insufficient to assess risk in individual patients Hawton K et al Stated dose g associated with 4.5-fold higher risk of hepatotoxicity Shnaps Y et al Reported quantities <1 g and 51 g associated with similar paracetamol concentrations and outcome Retrospective studies Edwards DA et al Reported paracetamol dose allows prediction of 4 h serum concentration using a pharmacokinetic equation Bond GR et al Dosages similar between those with and without hepatotoxicity after supra-therapeutic ingestion Therefore, N-acetylcysteine is normally administered if the patient is suspected of having ingested 412 g (or 415 mg/kg) in the previous 24 h. 6,7 A common perception is that patient-reported dosages are unreliable in the context of acute overdose, but the validity of this approach has received comparatively little attention. A literature search using PubMed ( for keywords { paracetamol OR acetaminophen } AND { dose } AND { overdose OR ingestion OR toxicity OR poisoning } revealed only a small number of English language articles concerning the validity of patient reported dose (Table 1) Limited data suggest that patients tend to overestimate the quantity of paracetamol ingested, and that the stated amount ingested correlates poorly with the risk of hepatotoxicity. 14,16 In contrast, others have found the reported dose of paracetamol to be sufficiently accurate to allow development of a predictive pharmacokinetic model and to determine those at highest risk of hepatotoxicity. 15,17 However, these studies have included comparatively small patient numbers, some involving a retrospective design, and relate to acute overdose, staggered overdose and inadvertent supra-therapeutic administration. Therefore, the present study was designed to prospectively examine the significance of the patient history in a large group of patients who are managed in a consistent manner. The aim was to study whether the reported dose of ingested paracetamol might correspond to systemic exposure determined by serum paracetamol concentrations or the risk of subsequent hepatotoxicity. Methods Study design The protocol was reviewed and approved by the local research ethics committee. This was a prospective, observational study of consecutive patients presenting to the Emergency Department after acute paracetamol overdose between March 25 and June 26. Exclusion criteria were presentation 415 h after overdose, uncertainty about the timing of ingestion and ingestion staggered over 2 h or more. A standard operating procedure is used in management of paracetamol overdose patients to ensure consistency between the Emergency Department and Toxicology Unit, and is in accordance with TOXBASE Õ, the standard resource for poisoning management advice in the United Kingdom. 18 In brief, intravenous N-acetylcysteine is indicated after acute ingestion if serum paracetamol concentration is above the Rumack-Matthew 2-line nomogram, or above the 1-line and the patient is considered at high risk due to chronic ethanol excess, enzyme inducing drugs (carbamazepine, phenobarbital, phenytoin, rifampicin, St John s Wort), chronic liver disease, or malnutrition. Chronic ethanol excess is defined by regular consumption of 421 U (168 g) per week in men or

3 Does the patient history predict hepatotoxicity? U (112 g) in women. 19 If more than 8 h have elapsed after ingestion and the patient is thought to have ingested 412 g or 415 mg/kg, then N-acetylcysteine may be initiated before paracetamol concentrations become available. Intravenous N-acetylcysteine 3 mg/kg is administered over 2.25h and serum creatinine, prothrombin time ratio, and liver biochemistry are monitored. Measurements A standardized data collection sheet was used to record patient age, gender, date and time of overdose, stated quantity ingested, risk factors for hepatotoxicity and serum paracetamol concentration. If the patient was unwilling or unable to give the amount ingested, then this was recorded as unknown. The primary outcome variables were serum paracetamol concentration above the 1-line or 2-line at 4 15 h post ingestion, administration of N-acetylcysteine and incidence of hepatotoxicity predefined by prothrombin time ratio 41.3 or alanine transaminase 51 U/l. 2 Data analyses Where appropriate, data are presented as median, absolute values and proportions, with 95% confidence intervals constructed by the modified Wald method. 21 The relationship between stated paracetamol dose and estimated 4 h acetaminophen concentration was determined in men and women separately using the Spearman rank correlation coefficient (rho), and compared by z-test. Convenience subgroups were created as multiples of the maximum daily therapeutic dose (4 g) and betweengroup comparisons were made using two-tailed Yate s corrected 2 proportional tests (MedCalc statistical software v , Mariakerke, Belgium). P-values <.5 were accepted as statistically significant. Results There were 987 patients (672 women, 68.1%) aged 3 years (ranged years) and weight 65 kg (ranged kg). The stated quantity of paracetamol ingested was 12 g in 475 (48.1%), 412 g in 349 (35.4%) and unknown in 163 (16.5%). Patients who reported ingestion of 412 g had paracetamol concentrations above the 2-line in 31.8% ( %), compared with only 3.2% ( %) after ingestion of 12 g (P <.1). Patients who reported ingestion of 412 g were also more likely to require N-acetylcysteine, 62.5% ( %) vs. 13.5% ( %) (P <.1) and suffered a higher incidence of hepatotoxicity, 6.9% ( %) vs. 1.3% (.5 2.8%), P ¼.1. There was a positive correlation between stated dose ingested and estimated 4 h paracetamol concentration in men, rho ¼.654 ( ) and women, rho ¼.667 ( ) and no significant difference between men and women (z ¼.396, P ¼.7569). Across the subgroups 4 g, g, g, g, g and 424 g, there was a positive dose-exposure relationship as determined by the proportion of patients with paracetamol concentrations higher than the 1-line and 2-line, and the need for N-acetylcysteine administration (Figure 1). There was also a positive association between the stated dose ingested and the incidence of hepatotoxicity. Patients in whom the dose was unknown had serum paracetamol concentrations similar to those measured in the groups who reported ingestion of g and g. Discussion In patients who were presented to hospital after acute paracetamol overdose, the reported quantities ingested are closely related to the extent of systemic exposure, as evidenced by serum paracetamol concentrations. The dosages reported by the patient informed both the extent of paracetamol exposure and also the likelihood of subsequent hepatotoxicity. This demonstrates the potential validity of the patient history in determining the quantity of drug ingested in poisoned patients who were presented to hospital. Furthermore, the data indicate a close correlation between the stated doses ingested and estimated 4 h paracetamol concentration for both men and women. It is rarely feasible to determine serum concentrations for a variety of drugs in the context of acute overdose and, therefore, the patient history might be an important surrogate for estimating overall drug exposure. Cautious interpretation is required for individual patients because of reporting variability, perhaps due to altered mental status, or the effects of co-ingested ethanol or sedative medications. Nonetheless, a correlation between stated dose and paracetamol concentrations higher than the 2-line has previously been found, irrespective of acute or chronic ethanol intake. 11 The present data are directly applicable to clinical research in poisoned patients in whom conventional methods are rarely appropriate. Validity of the patient history for estimating overall exposure might allow dose-dependent effects to be studied in groups of poisoned patients, for example evaluation of dose-dependent QT prolongation. 22

4 124 W.S. Waring et al. A Proportion of patients >'1-line' (%) B Proportion of patients >'2-line' (%) C Prevalence of N-acetylcysteine (%) D Incidence of hepatotoxicity (%) n=12 4 n=12 4 n=12 4 n=12 Figure 1. Relationships between stated dose ingested and (A) paracetamol concentration 41-line, (B) paracetamol concentration 42-line, (C) N-acetylcysteine Patients who reported ingestion of more than 12 g of paracetamol showed a substantially higher risk of developing hepatotoxicity compared with ingestion of smaller quantities (6.9% vs. 1.3%), despite administration of N-acetylcysteine. The incidence of hepatotoxicity in patients who were unwilling or unable to state the dose ingested was comparatively low (1.2%), indicating that this subgroup is not exposed to additional risk. Overall, the incidence of hepatotoxicity across the whole study population (3.2%) was consistent with other studies involving young adults who were presented to hospital within 15 h of acute paracetamol overdose. 7 9 Moreover, the higher incidence among those who ingested more than 12 g supports the use of this cut-off value when considering the need for N-acetylcysteine administration. A limitation is that only patients who were presented to hospital early were included, because the risk nomogram was originally validated between 4 and 15 h after ingestion. 5 Patient characteristics might conceivably differ between those who were presented to hospital early, late, or after a staggered overdose. Nonetheless, inclusion of these additional groups in an informal post hoc analysis did not alter the findings. Another potential concern is that data concerning paracetamol overdose might not be applicable to patients who have ingested other drugs. Nonetheless, a close correlation between the stated dose of quetiapine and serum concentrations has been reported elsewhere after deliberate overdose. 23 Activated charcoal was rarely administered in this series, despite being recommended within 1 h of paracetamol ingestion. 24 This might be due to specific contraindications, or limited opportunity to administer treatment within 1 h, or lack of awareness by Emergency Department staff. A further potential limitation is that reporting might have been enhanced if patients were aware that confirmatory paracetamol measurements would be made, but we do not believe that this would have had a significant effect here because the patient history was elicited before permission was sought for blood sampling. In conclusion, it is possible to make comparisons between groups of patients stratified on the basis of self-reported quantities ingested. The patient history is sufficiently reliable to indicate both the extent of drug exposure and also the risk of subsequent (NAC) treatment and (D) hepatotoxicity defined by prothrombin time ratio 41.3 or alanine transaminase 41 U/l. Data shown as percentage and 95% confidence interval, P <.5, P <.1, P <.5 by Yate s corrected 2 proportional tests compared with 4 g group. ¼ dose unknown.

5 Does the patient history predict hepatotoxicity? 125 toxicity. This approach merits further consideration in qualitative assessment of patients who were presented to hospital after deliberate overdose. Conflict of interest: None declared. References 1. Camidge DR, Wood RJ, Bateman DN. The epidemiology of self-poisoning in the. Br J Clin Pharmacol 23; 56: Atcha Z. Paracetamol related deaths in England and Wales, (Office for National Statistics). Health Stat 2; Q1: Sheen CL, Dillon JF, Bateman DN, Simpson KJ, MacDonald TM. Paracetamol-related deaths in Scotland, Br J Clin Pharmacol 22; 54: Brok J, Buckley N, Gluud C. Interventions for paracetamol (acetaminophen) overdose. Cochrane Database Syst Rev 26;CD Rumack BH, Peterson RC, Koch GG, Amara IA. Acetaminophen overdose. 662 cases with evaluation of oral acetylcysteine treatment. Arch Intern Med 1981; 141: Buckley N, Eddleston M. Paracetamol (acetaminophen) poisoning. Clin Evid 25; 14: Jones AL. Mechanism of action and value of N-acetylcysteine in the treatment of early and late acetaminophen poisoning: a critical review. J Toxicol Clin Toxicol 1998; 36: Thomas SH, Horner JE, Chew K, Connolly J, Dorani B, Bevan L, et al. Paracetamol poisoning in the north east of England: presentation, early management and outcome. Hum Exp Toxicol 1997; 16: Brotodihardjo AE, Batey RG, Farrell GC, Byth K. Hepatotoxicity from paracetamol self-poisoning in western Sydney: a continuing challenge. Med J Aust 1992; 157: Daly FF, O Malley GF, Heard K, Bogdan GM, Dart RC. Prospective evaluation of repeated supratherapeutic acetaminophen (paracetamol) ingestion. Ann Emerg Med 24; 44: Schiodt FV, Lee WM, Bondesen S, Ott P, Christensen E. Influence of acute and chronic alcohol intake on the clinical course and outcome in acetaminophen overdose. Aliment Pharmacol Ther 22; 16: Gazzard BG, Widdop B, Davis M, Hughes RD, Goulding R, Williams R. Early prediction of the outcome of a paracetamol overdose based on an analysis of 163 patients. Postgrad Med J 1977; 53: Hawton K, Ware C, Mistry H, Hewitt J, Kingsbury S, Roberts D, et al. Paracetamol self-poisoning. Characteristics, prevention and harm reduction. Br J Psychiatry 1996; 168: Shnaps Y, Halkin H, Dany S, Tirosh M. Inadequacy of reported intake in assessing the potential hepatotoxicity of acetaminophen overdose. Isr J Med Sci 198; 16: Edwards DA, Fish SF, Lamson MJ, Lovejoy FH Jr. Prediction of acetaminophen level from clinical history of overdose using a pharmacokinetic model. Ann Emerg Med 1986; 15: Bond GR, Wiegand CB, Hite LK. The difficulty of risk assessment for hepatic injury associated with supratherapeutic acetaminophen use. Vet Hum Toxicol 23; 45: Sivilotti ML, Yarema MC, Juurlink DN, Good AM, Johnson DW. A risk quantification instrument for acute acetaminophen overdose patients treated with N-acetylcysteine. Ann Emerg Med 25; 46: Bateman DN, Good AM. Five years of poisons information on the internet: the experience of TOXBASE. Emerg Med J 26; 23: Report of a Working Party of the Royal College of Physicians. Alcohol- can the NHS afford it? Recommendations for a coherent alcohol strategy for hospitals. Royal College of Physicians of London, 21. ISBN McClain CJ, Price S, Barve S, Devalarja R, Shedlofsky S. Acetaminophen hepatotoxicity: an update. Curr Gastroenterol Rep 1999; 1: Zou G, Klar N. A non-iterative confidence interval estimating procedure for the intraclass kappa statistic with multinomial outcomes. Biom J 25; 47: Isbister GK, Friberg LE, Duffull SB. Application of pharmacokinetic-pharmacodynamic modelling in management of QT abnormalities after citalopram overdose. Intensive Care Med 26; 32: Balit CR, Isbister GK, Hackett LP, Whyte IM. Quetiapine poisoning: a case series. Ann Emerg Med 23; 42: Chyka PA, Seger D, Krenzelok EP, Vale JA. American academy of clinical toxicology; European association of poisons centres and clinical toxicologists. Position paper: single-dose activated charcoal. Clin Toxicol 25; 43:61 87.