Early detection of paracetamol toxicity using circulating liver microRNA and markers of cell necrosis

Paracetamol (acetaminophen) overdose is a common medical emergency. The decision to treat with acetylcysteine is predominately based on the dose of drug ingested and a timed blood paracetamol concentration. These risk markers are used because most patients present to hospital soon after overdose, before hepatotoxicity can be excluded using current biomarkers such as serum alanine aminotransaminase activity (ALT). To improve patient stratification, sensitive liver-specific and mechanism-based biomarkers are in development [1–3]. In a recent study we demonstrated that the liver-enriched microRNA, miR-122, and the necrosis marker proteins high-mobility group box 1 (HMGB1) and keratin-18 (K18) identified hepatotoxicity on hospital admission with areas under their respective receiver operator curves >90% and negative predictive values >95% [1]. Here we present a case that demonstrates their potential utility.

A 25-year-old man attended the Royal Infirmary of Edinburgh 4.5 h after a single Co-codamol overdose (60–70 tablets, total 30–35 g paracetamol/480–560 mg codeine). Blood was drawn 4.5 h after overdose and the paracetamol concentration was 107 mg l⁻¹ (Table 1). The patient was deemed at low risk of hepatotoxicity applying the criteria that were standard UK practice at that time [4] and, therefore, the paracetamol concentration was interpreted using the ‘200-line’ on the UK paracetamol treatment nomogram (line starting at 200 mg l⁻¹ at 4 h post-ingestion). According to the UK guidance used at the time of presentation, acetylcysteine treatment was not indicated and the patient was discharged after psychiatric review. The patient returned to hospital 43 h after overdose because he had been vomiting and felt lethargic. Repeat blood tests (43 h post-overdose) demonstrated hepatotoxicity with coagulopathy (Table 1). The patient was immediately treated with intravenous acetylcysteine and recovered fully.

Table 1.

Blood results from first presentation to hospital (4.5 h after overdose) and second presentation (43 h after overdose). For miR-122, HMGB1 and K18 the upper limit of normal (ULN) is the upper limit of 95% prediction interval based on the hospital presentation blood sample from paracetamol overdose patients without subsequent hepatotoxicity (peak ALT < 50 U l⁻¹, n = 82) [1]

	First presentation blood sample (4.5 h post-overdose)	Second presentation blood sample (43 h post-overdose)
Paracetamol (mg l−1)	107	9
ALT (U l−1) (ULN 50)	34	11314
INR	1.0	2.1
miR-122 (let-7d normalized) (ULN 5.2)	261
HMGB1 (ng ml−1) (ULN 0.9)	7.2
Necrosis K18 (U l−1) (ULN 480)	4018

Retrospectively, biomarkers were measured (by a researcher blind to the patient's data) in the patient's first presentation sample (taken 4.5 h after overdose) and hepatotoxicity was unambiguously reported (Table 1). MiR-122 is a liver-enriched microRNA that is released into the circulation following paracetamol-induced hepatotoxicity. Both rodent and human studies demonstrate it is elevated early, when ALT is still normal [1, 3, 5]. At 4.5 h after overdose in this patient, miR-122 was elevated 50-fold above the upper limit of normal (ULN). HMGB1 and K18 are proteins released by necrotic cells that are elevated in the circulation of rodents and humans with paracetamol toxicity [2, 6]. In this patient they both reported toxicity on first hospital presentation (approximately 8-fold above ULN).

At the time of first presentation to hospital, the UK guidelines were correctly applied and the patient was not treated. This misclassification may reflect the patient having increased susceptibility to hepatotoxicity, the time of overdose being earlier than reported, drug absorption being delayed by codeine or a second overdose after first discharge from hospital. However, had the new biomarkers been available toxicity would have been identified at first presentation. Since this case, the UK guidelines have changed and a single ‘100-line’ (line starting at 100 mg l⁻¹ at 4 h post-ingestion) is used for treatment decisions in all patients, meaning this case would now be correctly classified as needing treatment. However, this lowering of the treatment threshold is likely to result in increased hospital bed occupancy and overall financial cost, for only marginal benefits [7]. Adoption of these biomarkers into clinical practice could target treatment to patients at elevated risk of hepatotoxicity. In North America, Australia and New Zealand, where a ‘150-line’ is used (line starting at 150 mg l⁻¹ at 4 h post-ingestion), this patient would still not be treated. This case report demonstrates the potential of new blood biomarkers for hospital front door stratification. This may improve patient identification for trials of new management pathways, facilitate earlier, safe, patient discharge and/or identify the rare patient with hepatotoxicity missed by current biomarkers.

Competing Interests

All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare no support from any organization for the submitted work, no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years and no other relationships or activities that could appear to have influenced the submitted work.