Paracetamol poisoning

Paracetamol poisoning is the single commonest cause of acute liver failure in the developed world and one of the few toxin exposures with a specific, highly effective antidote that is time-critical. The Fellowship candidate must understand the mechanism (the CYP2E1 oxidation to NAPQI, the glutathione depletion, the zone-3 centrilobular necrosis), the four-phase clinical course, the Rumack-Matthew nomogram and its limitations, the three-bag 21-hour intravenous N-acetylcysteine regimen, and the King's College criteria that trigger a liver-transplant referral.[1][2] The dangerous feature is the silent first 24 hours — the patient who has taken a lethal dose feels well, has normal liver function tests, and may be discharged home unless the clinician reads the level against the nomogram and treats with N-acetylcysteine.[4]

Definition and classification

Paracetamol (acetaminophen, N-acetyl-p-aminophenol) is an analgesic and antipyretic available without prescription. Acute paracetamol poisoning is the ingestion of a potentially hepatotoxic dose, defined as 150 mg/kg or 7.5 g in an adult, whichever is less, taken over a period of one hour or less. Staggered overdose (also called repeated supratherapeutic ingestion, RSI) is the ingestion of doses above the therapeutic maximum over a period longer than one hour, often over days for pain; it is more dangerous than a single acute overdose because the nomogram cannot be applied, the patient often presents late, and the clinician may underestimate the risk.[1] The clinical course is divided into four phases that frame every management decision.

Pathophysiology — NAPQI, glutathione, and the centrilobular necrosis

Therapeutic paracetamol is metabolised in the liver by three pathways. About 90 per cent is conjugated — 40 to 60 per cent by glucuronidation and 20 to 40 per cent by sulfation — to non-toxic, water-soluble metabolites excreted in the urine. About 2 to 5 per cent is oxidised by CYP2E1 (and a small amount by CYP1A2 and CYP3A4) to the highly reactive electrophile N-acetyl-p-benzoquinone imine (NAPQI).[1]

[10]

Glutathione depletion is accelerated by anything that pre-depletes stores — fasting, malnutrition, chronic alcohol misuse, anorexia, HIV, cachexia — or that induces CYP2E1 — chronic ethanol, rifampicin, phenytoin, carbamazepine, isoniazid, St John's wort. These are the "high-risk" patients in whom some guidelines apply a lower treatment line (the UK 100 mg/L line at 4 hours, versus the standard 200 mg/L line for the lower-risk patient).[4]

Sources, epidemiology, and the rationale for pack-size restrictions

Paracetamol is the most commonly ingested drug in self-poisoning worldwide, the leading cause of acute liver failure in the United Kingdom, the United States, and Australasia, and accounts for the largest single call volume to poison information centres. The UK introduced pack-size restrictions (16 g per pack in non-pharmacy outlets, 32 g in pharmacies) in 1998, with a measurable fall in paracetamol-related death and transplant.[1] Despite this, paracetamol remains the leading toxin cause of acute liver failure, and most fatalities follow a staggered overdose or a delayed presentation rather than a large single acute ingestion.[4]

Clinical presentation — the four phases

The single acute overdose follows a stereotyped four-phase course. The Fellowship candidate must know the phases because the timing of presentation dictates what is found clinically and how the level is interpreted.[6]

[3]

In phase 1 the patient feels well or has only nausea, vomiting, pallor and diaphoresis; liver enzymes are normal and the patient can be entirely asymptomatic despite having taken a lethal dose — this is the trap. In phase 2 the right-upper-quadrant pain appears and the ALT and INR begin to rise; an oliguria may develop. In phase 3 the patient develops the full picture of fulminant hepatic failure — jaundice, encephalopathy, coagulopathy, hypoglycaemia, metabolic acidosis and acute kidney injury. In phase 4, surviving patients recover completely without chronic liver disease, while non-survivors progress to multi-organ failure and death typically between days 3 and 5.[3]

Differential diagnosis — the cause of the transaminitis and the acute liver failure

When a patient presents with a markedly raised ALT or with acute liver failure, the differential is the cause of the hepatocyte injury; the history and the paracetamol level resolve most, but the others must be considered because the management diverges and several (the mushroom, the viral, the ischaemic) are missed by a narrow tox-focused history.[3]

[3]

Bedside assessment

The bedside assessment of the paracetamol-overdose patient has two goals: to estimate the time of ingestion (which determines whether the nomogram can be used) and to detect complications. The history must establish the time, the dose, the formulation (immediate-release vs modified-release), the intent, and any co-ingestants. A staggered history, an unknown time, or a mixed overdose each remove the option of using the nomogram and force the clinician toward empirical N-acetylcysteine.[1] The examination documents the conscious level, the vital signs, the hydration, the presence of right-upper-quadrant tenderness, the stigmata of chronic liver disease, and any sign of hepatic encephalopathy (asterixis is the bedside marker of the early encephalopathy). A psychiatric and self-harm risk assessment is begun once the patient is medically stable.

Investigations — the 4-hour paracetamol level and the nomogram

The cornerstone investigation is the serum paracetamol level drawn at 4 hours post-ingestion. Levels drawn before 4 hours are uninterpretable because absorption and distribution are incomplete; if the patient presents after 4 hours, the level is drawn immediately on arrival.[1]

[1]

The accompanying blood panel is the ALT (or AST), the INR, the creatinine, the venous blood gas (pH, lactate, bicarbonate), the glucose, the phosphate, the bilirubin, the lipase, and the beta-hCG in any woman of child-bearing age. The lactate is a marker of massive overdose: an early high lactate reflects mitochondrial NAPQI toxicity and predicts hepatotoxicity even before the transaminases rise.[5] A low phosphate (under 0.5 mmol/L) in phase 2 to 3 is a marker of depleted ATP and a poor prognosis. A salicylate level, an ethanol level, and a urine drug screen are sent if co-ingestion is suspected; the ECG rules out a co-ingested cardiotoxic agent (the widened-QRS tricyclic, the tachyarrhythmia of the salicylate).[1]

Immediate management — decontamination, the nomogram decision, and the first bag of N-acetylcysteine

The immediate management is built around three decisions: resuscitation, decontamination, and the antidote. Resuscitation follows ABCDE — most paracetamol overdoses are cardiovascularly stable, but the late-presenting patient may be in shock from hepatotoxicity, sepsis, or haemorrhage. Activated charcoal 50 g (1 g/kg in the child) orally or via nasogastric tube is given within 1 hour of ingestion (some centres extend to 2 to 4 hours for modified-release preparations) and reduces paracetamol absorption; it is most useful when the ingestion was large and within the time window.[1]

The nomogram decision follows. If the patient has taken a single acute ingestion and presents within 24 hours, a 4-hour (or later) level is plotted on the nomogram; if the level is on or above the treatment line, N-acetylcysteine is started. If the time of ingestion is unknown, the overdose was staggered, or the patient presents in or after phase 2 with hepatic injury, N-acetylcysteine is started empirically and the level is awaited.[4]

[2]

The antidote is N-acetylcysteine (NAC), which works by replenishing hepatic glutathione and by supplying cysteine, the rate-limiting precursor for glutathione synthesis; it also acts as a free-radical scavenger and a sulphur donor, and improves tissue oxygen delivery in established hepatotoxicity. NAC is remarkably effective when given within 8 hours of ingestion — it provides virtually complete protection against hepatotoxicity — and it remains beneficial when given later, even after hepatotoxicity has developed.[2]

The standard intravenous regimen across ANZ, the UK, and North America is the three-bag 21-hour regimen: 150 mg/kg over 1 hour, then 50 mg/kg over 4 hours, then 100 mg/kg over 16 hours — a total of 300 mg/kg over 21 hours.[2] The first bag was originally given over 15 minutes (the regimen was designed for oral NAC, where the load is one large dose), but the high plasma concentration produced a non-allergic, dose-related anaphylactoid reaction (flushing, urticaria, bronchospasm) in up to 20 per cent of patients; lengthening the loading bag to 1 hour reduced this rate substantially.[2]

The oral 72-hour regimen (140 mg/kg loading, then 70 mg/kg every 4 hours for 17 doses) is an alternative where IV NAC is unavailable; it is effective but vomiting is the major practical problem and an antiemetic is routinely required. The oral route is now rarely used in the developed world.[2]

Definitive management — the escalation ladder, the modified regimen, and stopping NAC

The escalation ladder in paracetamol poisoning is short: most patients complete the 21-hour course and are discharged. The patient with a rising INR or ALT at the end of the 21-hour course, or with acute liver failure, continues NAC — either by extending the third bag at 100 mg/kg over 16 hours (the "modified" or "prolonged" regimen) or by switching to a continuous infusion at 6.7 mg/kg per hour until the INR is below 1.3, the ALT is falling, the paracetamol level is undetectable, and the encephalopathy has resolved.[3] The prognostic markers guide the duration: a falling ALT, a falling INR, an improving lactate, and a clearing paracetamol level signal recovery; a rising INR, a falling phosphate, a rising lactate, or a rising creatinine signal deterioration and the need for a transplant referral.

Subtypes and scenarios — the modified-release preparation and the staggered overdose

Two scenarios deserve special attention because the nomogram fails in each. The modified-release (sustained-release) paracetamol preparation (common in Australasia) produces a biphasic absorption with two peak serum concentrations; the recommendation is to draw two levels 4 hours apart (the first at 4 hours post-ingestion), and to start N-acetylcysteine if either level is above the treatment line.[1] The staggered overdose / repeated supratherapeutic ingestion is the more dangerous pattern: the nomogram cannot be applied because no single ingestion time exists, and these patients present later with established injury. The threshold to treat is lower — any detectable paracetamol level with a raised ALT or INR, or a substantial ingestion history alone (over 150 mg/kg or over 7.5 g in any 24-hour window) — and NAC is started empirically.[4]

Complications and pitfalls — hepatic failure, acidosis, and the anaphylactoid reaction

The complications of untreated or late-treated paracetamol overdose are the complications of fulminant hepatic failure: encephalopathy (graded I to IV), coagulopathy (the rising INR is the most sensitive marker of synthetic failure), hypoglycaemia (impaired gluconeogenesis), metabolic acidosis (the high lactate from mitochondrial failure), acute kidney injury (the same NAPQI mechanism in the proximal tubule, plus the hepatorenal physiology), and progression to multi-organ failure.[3][5]

The pitfalls are equally well described. The first is the missed 4-hour level — drawing the level too early (before absorption is complete) or too late (after 24 hours) and either over- or under-treating. The second is the misapplied nomogram — applying it to a staggered overdose, an unknown-time ingestion, or a co-ingestion that delays gastric emptying. The third is the delayed NAC — withholding NAC because the patient looks well in phase 1, when phase 1 is exactly the window in which NAC is most effective. The fourth is the anaphylactoid reaction to NAC, which is managed by slowing or pausing the infusion and giving an antihistamine (chlorphenamine 10 mg IV) — not by abandoning the antidote; the reaction is dose-rate-related and largely affects the first bag. The fifth is the false reassurance of a falling ALT in the recovery phase — the ALT falls as the hepatocytes die (no live cells to release enzyme), so a falling ALT with a rising INR is a deterioration, not an improvement.[3]

Prognosis and disposition — the King's College criteria for transplantation

The prognosis depends entirely on the timing of NAC. A patient treated within 8 hours of a single acute ingestion has a near-zero mortality; a patient presenting in phase 3 has a mortality that approaches 30 per cent without transplantation. The King's College Criteria (KCC) identify the paracetamol-induced acute liver failure patient who will not survive without a transplant; meeting KCC carries a mortality of 80 per cent or more with medical management alone and is the trigger for an urgent transplant-centre referral.[3]

The arterial pH criterion alone (below 7.3 after fluid resuscitation) meets KCC regardless of the encephalopathy grade; the triad of INR over 6.5, creatinine over 300 micromol/L and grade III to IV encephalopathy meets KCC only when all three are present. The Bernal dynamic model (Lancet Gastroenterology & Hepatology 2016) refined the static KCC by serially tracking the lactate, the INR and the encephalopathy and predicts survival with a higher sensitivity, especially in the borderline patient; the modern transplant-centre referral uses both.[3]

The disposition is determined by the clinical phase. The phase 1 patient with a level below the treatment line is discharged (with a psychiatric assessment if the overdose was deliberate). The patient given NAC within 8 hours and recovering is discharged when the INR is normal and the ALT is falling. The patient with a rising INR, encephalopathy, or any KCC component is admitted to a high-dependency or intensive-care unit and referred to the liver transplant service.[3]

Special populations

Pregnancy is not a contraindication to NAC, which crosses the placenta and protects the fetus; the same mg/kg regimen is used, and the threshold to treat is unchanged because the maternal hepatocyte is the target. The malnourished, alcoholic, anorexic, HIV-positive, or CYP-inducing-drug patient is the "high-risk" patient in whom the lower (100 mg/L at 4 hours) treatment line is applied, because glutathione stores are depleted and NAPQI formation is induced. The paediatric patient is dosed identically by weight; liquid formulations are absorbed faster and the 4-hour level remains the reference. The late presenter (over 24 hours) with any sign of hepatic injury is treated with NAC regardless of the nomogram — the antidote works in established hepatotoxicity.[4][3]

The worked algorithm — when to give N-acetylcysteine

The single most common Fellowship error is mis-applying the nomogram to a presentation it was never designed for. The decision to treat branches on the pattern of ingestion, not on the level alone. The algorithm below collapses the UK, US, and ANZ guidance into one pathway.[1][4]

[7] [7]

The treatment lines compared — UK, US, and ANZ

The three English-speaking guideline bodies use slightly different thresholds, and the Fellowship candidate is expected to know that the principle is identical (a 4-hour level on a log-linear treatment line) while the cut-off differs. The discrepancy reflects different philosophies of sensitivity versus NAC-anaphylactoid-reaction burden.[4]

[11] [4]

Intravenous versus oral N-acetylcysteine

Both routes replenish glutathione and detoxify NAPQI; both are effective when given early. The intravenous route is the modern standard because it is reliable, faster to peak, and does not depend on the vomiting patient retaining the dose.[2][8]

[2]

Managing the anaphylactoid (non-allergic) reaction to NAC

The reaction to NAC is an anaphylactoid (non-IgE, dose-rate-related) reaction, not a true allergy — it is driven by the high peak plasma histamine concentration produced by the rapid loading bag, and it is almost always confined to the first hour of the infusion. The rate is 8 to 20 per cent with the old 15-minute load and under 5 per cent with the modern 1-hour load. The cardinal rule is: do not abandon the antidote — the reaction is manageable, the untreated hepatotoxicity is lethal.[2][11]

[11]

Single acute versus staggered overdose — the key comparison

The single acute overdose is the textbook case; the staggered overdose is the one that kills. The Fellowship candidate must contrast the two because the management, the nomogram validity, and the prognosis all diverge.[4]

[3]

The landmark trials

The evidence base for paracetamol-poisoning management spans five decades and is unusually complete for a toxicological topic. The Fellowship candidate should be able to name the pivotal studies behind the nomogram, the NAC regimen, and the transplant criteria.[1]

[6] [7] [11] [10]

Memory aids — the mechanism and the transplant criteria

[9] [3] [1] [10]

Evidence and regional guidelines

The evidence base and the regional practice are well aligned across the Anglosphere. The UK national guideline (the 2012 revision published by the Commission on Human Medicines, audited by Bateman in 2014) introduced the single 100 mg/L "treatment line" for all patients (with a separate 50 mg/L line rarely used for the highest-risk) and lowered the threshold for NAC, which increased NAC use without increasing jaundice or transplant admissions.[4] The US guideline (the American College of Emergency Physicians and the FDA-approved labelling) uses the Rumack-Matthew 150 mg/L line. The Australian and New Zealand guideline (the ANZ position statements from the Australasian Poisons Information Centre network) endorses the 21-hour three-bag regimen, applies the lower treatment line in the high-risk patient, and recommends two levels 4 hours apart for the modified-release preparation.[1][2] The three-bag 21-hour regimen itself (150 mg/kg over 1 hour, 50 mg/kg over 4 hours, 100 mg/kg over 16 hours) is the modern global standard, having replaced the older 20.25-hour regimen that delivered the loading bag over 15 minutes; the change to the 1-hour load reduced the anaphylactoid rate without compromising efficacy.[2]

Exam practice

[10] [10]

Exam pearls

• The mechanism: paracetamol → CYP2E1 → NAPQI → glutathione depletion → centrilobular (zone 3) necrosis. Glutathione under 30 per cent is the tipping point.
• The 21-hour three-bag regimen: 150 mg/kg over 1 hour, then 50 mg/kg over 4 hours, then 100 mg/kg over 16 hours — total 300 mg/kg over 21 hours.
• The level is drawn at 4 hours post-ingestion (or as soon as possible if the presentation is later); never before 4 hours.
• The indications for NAC: any level above the treatment line, OR the unknown-time ingestion with a detectable level, OR the staggered overdose, OR the late presentation with hepatic injury.
• NAC works at any time — give it even in established hepatotoxicity; it improves survival in paracetamol-induced acute liver failure.
• The anaphylactoid reaction to the first bag is dose-rate-related: slow the infusion and give chlorphenamine; do not abandon the antidote.
• The King's College criteria: arterial pH below 7.3 alone, OR all three of INR over 6.5, creatinine over 300 micromol/L, grade 3 to 4 encephalopathy — refer for transplant.
• The high lactate and the low phosphate are markers of poor prognosis.
• A falling ALT with a rising INR is deterioration (the hepatocytes have all died) — not recovery.
• The staggered or unknown-time overdose: treat if the level is detectable, the ALT is over 50, the INR is over 1.3, or the dose exceeds 150 mg/kg or 7.5 g — the nomogram is never used.
• The five pathways to NAC: single-acute-above-line; unknown-time-detectable; staggered; late-with-injury; modified-release-two-levels.
• NAC works at any time — it improves survival in established paracetamol ALF by glutathione repletion, sulphur donation, and free-radical scavenging, not only NAPQI neutralisation.
• The four prognostic bloods: rising INR, rising lactate (over 3.5), falling phosphate (under 0.5), arterial pH under 7.3 — track the trend, not the single value.
• The anaphylactoid reaction is dose-rate-related and confined to bag 1: slow/stop the infusion, give chlorphenamine, restart at half rate — never abandon the antidote.
• The lactate is the earliest prognostic marker in massive overdose — it rises from mitochondrial NAPQI toxicity before the transaminases move.
• Co-ingest an opioid/anticholinergic and the nomogram can mislead — delayed gastric emptying keeps absorption going past 4 hours; send salicylate, ethanol, and a urine drug screen.
• NAC in obesity: dose per actual body weight but cap the bag weight (commonly 100–110 kg) per local policy — never under-dose the obese patient.[10]

Model answer — A 24-year-old woman, 20 g of paracetamol 6 hours ago, ALT 45 U/L, INR 1.1, paracetamol level 180 mg/L. Outline the immediate management.

[1]

Immediate management. The presentation is a single acute ingestion at a known time, so the paracetamol level can be plotted on the Rumack-Matthew nomogram. At 6 hours post-ingestion the treatment line crosses approximately 80 mg/L (it halves every 4 hours from 200 mg/L at 4 hours); a level of 180 mg/L is well above the line. The indication for N-acetylcysteine is met. The first step is the first bag of the 21-hour three-bag regimen — 150 mg/kg over 1 hour — commenced immediately, followed by 50 mg/kg over 4 hours and 100 mg/kg over 16 hours, total 300 mg/kg over 21 hours. A baseline set of bloods is sent: ALT, INR, creatinine, venous gas (pH, lactate), glucose, phosphate, beta-hCG, and a salicylate and ethanol level for any co-ingestion. Activated charcoal 50 g is offered (within the 1- to 2-hour window it is well past, so it is not given here). The patient is observed for the anaphylactoid reaction in the first hour; if it occurs, the infusion is slowed and chlorphenamine 10 mg IV given, the antidote is not stopped. The INR and ALT are checked at the end of the 21-hour course: if normal or falling, NAC is stopped; if the INR is rising or above 2.0, the third bag is continued. A psychiatric and self-harm risk assessment is performed once the patient is medically stable. The expected outcome with NAC within 8 hours is complete recovery.[10]

SAQs

[10] [10]

Red flags

[10] [7]

References

1. [1]Bateman DN, Dart RC, Dear JW, et al. Fifty years of paracetamol (acetaminophen) poisoning: the development of risk assessment and treatment 1973-2023 with particular focus on contributions published from Edinburgh and Denver. Clinical toxicology (Philadelphia, Pa.), 2023.PMID 38197864
2. [2]Bateman DN, Dear JW, Carroll R, et al. New regimens for intravenous acetylcysteine, where are we now? Clinical Toxicology (Philadelphia), 2016.PMID 26666290
3. [3]Bernal W, Lee WM, Wendon J, Larsen FS, Williams R. Development and validation of a dynamic outcome prediction model for paracetamol-induced acute liver failure: a cohort study. Lancet Gastroenterology & Hepatology, 2016.PMID 28404094
4. [4]Bateman DN, Carroll R, Pettie J, et al. Effect of the UK's revised paracetamol poisoning management guidelines on admissions, adverse reactions and costs of treatment. British journal of clinical pharmacology, 2014.PMID 24666324
5. [5]Shah AD, Wood DM, Dargan PI. Understanding lactic acidosis in paracetamol (acetaminophen) poisoning. British Journal of Clinical Pharmacology, 2011.PMID 21143497
6. [6]Smilkstein MJ, Knapp GL, Kulig KW, Rumack BH. Efficacy of oral N-acetylcysteine in the treatment of acetaminophen overdose. Analysis of the national multicenter study (1976 to 1985). New England Journal of Medicine, 1988.PMID 3059186
7. [7]Smilkstein MJ, Bronstein AC, Linden C, Augenstein WL, Kulig KW, Rumack BH. Acetaminophen overdose: a 48-hour intravenous N-acetylcysteine treatment protocol. Annals of Emergency Medicine, 1991.PMID 1928874
8. [8]Prescott LF, Illingworth RN, Critchley JA, Stewart MJ, Adam RD, Proudfoot AT. Intravenous N-acetylcystine: the treatment of choice for paracetamol poisoning. British Medical Journal, 1979.PMID 519312
9. [9]O'Grady JG, Alexander GJ, Hayllar KM, Williams R. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology, 1989.PMID 2490426
10. [10]Bernal W, Donaldson N, Wyncoll D, Wendon J. Blood lactate as an early predictor of outcome in paracetamol-induced acute liver failure: a cohort study. Lancet, 2002.PMID 11867109
11. [11]Schmidt LE, Rasmussen DN, Petersen TS, et al. Fewer adverse effects associated with a modified two-bag intravenous acetylcysteine protocol. Clinical Toxicology (Philadelphia), 2018.PMID 29792347