Death after psychiatric contraindications to urgent liver transplant for paracetamol overdose

Abstract

BACKGROUND

Paracetamol overdose (POD) is the most common cause of acute hepatic failure (AHF) in the United Kingdom. Without urgent orthotopic liver transplant (OLT), mortality is high. Psychiatric assessment for transplant is time-pressured and often undertaken by psychiatrists without transplant experience. Assessors may identify absolute psychiatric contraindications (APCIs) precluding transplant in otherwise medically suitable patients. It is unknown how often this occurs. The combination of high but unknown mortality, time pressure, and relative inexperience is likely to provoke anxiety in assessors. This study hypothesised that the proportion of POD patients assessed for OLT who die because psychiatric contraindications preclude transplant would be small but not negligible.

AIM

To determine the proportion of patients with paracetamol-induced AHF, for whom psychiatric contraindications preclude transplantation, and the consequent mortality.

METHODS

This is an 18-year single-centre retrospective cohort study based in a national liver transplant centre. 524 participants were identified from a departmental database and included if they had AHF from suspected POD and received a psychiatric assessment for OLT. For those who died before discharge, records were reviewed for medical and psychiatric contraindications to transplant, alongside age, sex, and primary psychiatric diagnosis. We calculated the proportion of patients assessed for whom APCIs precluded transplant, resulting in death.

RESULTS

Among 524 patients undergoing psychiatric assessment for OLT, there were 102 in-episode deaths (19.5%). APCIs were identified in 46 patients who were otherwise medically suitable for transplant and went on to die. This statistic represents 8.8% of the number of persons evaluated and 45% of the number of deaths. Within this subgroup, 27 (59%) were female, with a mean age of 44.6 years (ranging from 19-72 years). The most common primary psychiatric diagnosis was alcohol dependence syndrome, which accounted for 67% (n = 31).

CONCLUSION

8.8% of medically suitable patients with AHF following POD died with APCIs to transplant. This indicates a need for ongoing assessor training and support, and (inter) national comparisons of practice.

Key Words: Paracetamol overdose; Acute hepatic failure; Orthotopic liver transplant; Psychiatric contraindications; Mortality

Core Tip: This is the first United Kingdom study to report the frequency of death in patients presenting with paracetamol overdose-related acute hepatic failure, who are declined urgent liver transplant because of psychiatric contraindications, despite their being otherwise medically suitable for transplant. We found that this occurs in almost 9% of patients assessed. Our results highlight the impact of psychiatric decision-making in this context, as well as the potential for inter-unit variability of patient outcomes that arises from the lack of standardised guidelines in this area.

INTRODUCTION

Paracetamol overdose (POD), whether deliberate or accidental, is common and can cause significant hepatotoxicity. In the United Kingdom, approximately 100000 people present to hospital following POD annually, half of whom require admission for emergency treatment[1]. When paracetamol is taken in overdose quantities, stores of glutathione, which is required for the conjugation and detoxification of paracetamol’s toxic metabolite, N-acetyl-p-benzoquinone imine, are depleted. This causes a build-up of N-acetyl-p-benzoquinone imine which directly damages hepatocytes. The first-line treatment for paracetamol poisoning is intravenous N-acetylcysteine, which works by restoring the body’s supply of glutathione. While this is effective if given in time, delayed administration, or very large overdoses taken over several hours, may render N-acetylcysteine insufficient to prevent the progression of hepatotoxicity to acute hepatic failure (AHF)[2].

POD is the leading cause of AHF in the United Kingdom, accounting for over 65% of cases[3]. AHF is associated with high mortality, and, for patients who continue to deteriorate despite maximum medical treatment, emergency orthotopic liver transplant (OLT) is the only intervention that may enable survival[4]. Transplant recipients who survive the critical peri- and post-operative period face significant longer-term challenges. Those who receive a liver transplant due to paracetamol-related AHF are at a tenfold higher risk of death secondary to psychosocial issues, such as suicide, trauma, and poor adherence to immunosuppressive treatment, compared to their non-paracetamol counterparts[5]. Given the scarcity of donor livers, the complexities of liver transplantation, and the lifelong morbidity and mortality burden for transplant recipients, careful consideration around listing is necessary. It is therefore required that potential transplant recipients undergo rigorous medical, psychiatric, and social assessment to determine suitability for OLT.

Whilst detailed prognostic criteria, such as the King’s College[6] (Table 1) or Clichy[7] (Table 2) criteria, are widely used to determine medical eligibility for transplant, no such evidence-based guidelines exist at a national level to aid psychiatric assessment. In some cases, psychiatric assessment may identify absolute psychiatric contraindications (APCIs) to transplantation, excluding some patients who are otherwise medically suitable for listing. If not listed for transplant due to APCIs, the high mortality of AHF implies that these patients may likely go on to die. The mortality rate for patients for whom psychiatric contraindications alone preclude OLT is not known.

Table 1 King’s College criteria for selection of acute hepatic failure patients suitable for liver transplantation[6].

Classifications	Characteristics
Paracetamol-induced acute hepatic failure	Arterial blood pH < 7.30 (irrespective of grade of encephalopathy)
	Or all of the following	Prothrombin time > 100 seconds (INR > 6.5)
		Serum creatinine > 300 μmol/L
		Grade III or IV hepatic encephalopathy
Non-paracetamol induced acute hepatic failure	Prothrombin time > 100 seconds (INR > 6.5) (irrespective of grade of encephalopathy)
	Or any 3 of the following (irrespective of grade of encephalopathy)	Age < 10 years or > 40 years
		Aetiology: Non-A/non-B hepatitis, drug-induced
		Duration of jaundice to encephalopathy > 7 days
		Prothrombin time > 50 (INR > 3.5)
		Serum bilirubin > 300 μmol/L

Citation: O'Grady JG, Alexander GJ, Hayllar KM, Williams R. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989; 97: 439-445. Copyright © 2025 Elsevier Inc. *INR: International normalized ratio.

Table 2 Clichy criteria for selection of acute hepatic failure patients for liver transplantation[7].

Number	Criteria
1	Presence of hepatic encephalopathy
2	Factor V level of < 20% (if patient’s age < 30 years) or < 30% (if patient’s age ≥ 30 years)

Citation: Bismuth H, Samuel D, Castaing D, Adam R, Saliba F, Johann M, Azoulay D, Ducot B, Chiche L. Orthotopic liver transplantation in fulminant and subfulminant hepatitis. The Paul Brousse experience. Ann Surg 1995; 222: 109-119 2025. Copyright © 2025 Wolters Kluwer Health, Inc. and/or its subsidiaries.

Psychiatric assessment in emergency circumstances is often time-pressured, performed by psychiatrists without transplant experience, and undertaken on patients from distant centres, for whom there is limited timely access to relevant medical records. Interview of patients with AHF is also often limited - or rendered impossible - by reduced consciousness, encephalopathy, or intubation, making accurate psychiatric assessment all the more challenging. This combination of high mortality, time pressure, limited background information and reduced access to interview may be daunting to psychiatric assessors, especially those with limited experience. This study aims to assist assessors by providing the first measure of mortality associated with these decisions.

The study examines in-episode deaths of patients with AHF secondary to POD, who were referred to the Scottish Liver Transplant Unit (SLTU) for assessment for emergency OLT between 2006-2024. It aims to determine the number of patients who die after APCIs to liver transplant are identified, and who were otherwise medically suitable for listing. It additionally explores demographic factors for this cohort of patients, such as their age, sex, and primary psychiatric diagnosis.

MATERIALS AND METHODS

This study is a single-centre, retrospective cohort analysis, conducted at the SLTU, located at the Royal Infirmary of Edinburgh, United Kingdom. The study was conducted by members of the Royal Infirmary of Edinburgh Department of Psychological Medicine. A total of 524 participants were identified using the Department of Psychological Medicine’s electronic database, which contains records of all cases referred for psychiatric assessment for urgent OLT suitability. The study covered an 18-year period, spanning referral records from 2006 to 2024. Participants were selected by searching the database for International Classification of Diseases codes related to AHF, which prompted psychiatric assessment for transplant suitability.

Once selected, the electronic medical records of each patient with AHF were reviewed. We examined these records to determine whether AHF was induced by suspected POD, whether the patient had died in-episode, and the outcomes of both medical and psychiatric assessments regarding eligibility for emergency liver transplantation. For patients who had APCIs but were deemed medically suitable for transplant, we collected additional demographic information such as age, sex, and primary psychiatric diagnosis. Finally, we calculated the proportion of patients with paracetamol-induced AHF who were assessed for transplantation, but for whom APCIs alone precluded transplant, leading to death.

RESULTS

524 patients underwent psychiatric assessment for potential OLT after presenting to the SLTU with POD-induced AHF between 2006 and 2024. 102 (19.5%) of these patients went on to die within their index admission episode without undergoing OLT. Amongst these 102 in-episode deaths, 46 patients were identified for whom APCIs precluded transplant, despite their otherwise being medically suitable. This statistic represents 8.8% of all patients undergoing psychiatric assessment for urgent OLT and 45% of all patients who died in our cohort. Of these 46 patients who died with APCIs to OLT, 27 (59%) were female and the average age was 44.6 years (range 19-72). The most common primary psychiatric diagnosis was Alcohol Dependence Syndrome, which was identified in 31 patients (67%).

DISCUSSION

The scarcity of organs for transplantation requires that recipients are carefully selected to ensure that they are used to maximum benefit. Candidates must undergo assessment to identify inter alia, comorbidities and other features that might predict a poor prognosis post-transplant. For non-urgent cases, time allows for careful multi-disciplinary discussions and an opportunity for patients to act upon potentially modifiable contraindications (for example, a patient with active alcohol dependence syndrome may be able to achieve six-months of sobriety and therefore be reconsidered for transplant). When a patient presents with AHF severe enough to require urgent transplantation after POD, selection assessment is challenging due to time pressure, limited collateral history information, and constraints on interview in patients who are unconscious, encephalopathic, or delirious.

Whilst detailed medical guidance exists to help physicians identify patients who are physiologically suitable for urgent OLT, there are no comparable national or international guidelines to aid decision-making when it comes to psychiatric assessment. In the SLTU, a 1997 Fatal Accident Inquiry recommended that the psychiatric contraindications to urgent OLT be codified, which led to the development of unit-specific guidance which remains in place today[8] (Table 3). While it is likely that a similar but less explicit approach is taken by other transplant centres in the United Kingdom, the lack of standardised guidelines leaves room for variation in practice between centres and potential regional inequality. An international study surveying mental health professionals involved in conducting urgent transplant psychiatric assessments found there was significant variability in decision-making when it came to making case-based transplant decisions[9], highlighting the potential for inconsistency between clinicians. There is therefore a clear need for further research comparing decision-making practices in psychiatric assessments for urgent OLT across transplant units in the United Kingdom.

Table 3 Absolute and relative psychiatric contraindications to urgent liver transplant in the Scottish Liver Transplant Unit, Royal Infirmary of Edinburgh[14].

Absolute contraindications	Relative contraindications
5 or more episodes of deliberate self-harm, unless previous events occurred many years ago	2-4 Lifetime episodes of self-harm
Current substance dependence of active misuse in a severe, chaotic fashion	Current/recent substance misuse
Chronic, severe, poor-prognosis mental disorders, especially if refractory to treatment or where leading to patient incapacity to express wishes or understand circumstances	Other significant or serious pervasive mental disorder
Repeated non-compliance with medical or psychiatric care	-
Patient refusal of transplant with capacity

Citation: Potts SG. Transplant psychiatry. J R Coll Physicians Edinb 2009; 39: 331-336. Copyright © 2025 Royal College of Physicians of Edinburgh. Copyright is held by the Scottish Liver Transplant Unit who have given permission for publication.

Given the nature of presentation with overdose, it is unsurprising that patients with POD-induced AHF assessed for OLT commonly have comorbid psychiatric illness. The primary diagnosis in our cohort, found in 67% of patients, was active alcohol dependence syndrome, meaning this was the commonest psychiatric contraindication to listing for transplant. This is in keeping with a previous study of patients presented to the SLTU between 1992 and 2014, where the commonest psychiatric contraindication to transplant was also alcohol use[10]. Active alcohol dependence syndrome is known to be associated with poor outcomes following liver transplant, due to the direct effects of alcohol on the allograft itself, as well as its commonly associated psychosocial factors (e.g. poor medication adherence)[11,12]. It is therefore widely agreed to be an absolute contraindication to transplant, whether urgent or not[13].

The impact of other psychiatric illnesses on post-transplant morbidity and mortality is less clear[12]. It is known that patients who have previously self-harmed are more likely to engage in further self-harming or suicidal behaviour in the future[9]. If patients who have received a liver transplant for POD-induced AHF go on to self-harm again post-transplant, this could risk loss of the allograft, either due to direct effects of a toxin or self-injurious behaviour, or through patient death. A previous study of spontaneous survivors of AHF in the SLTU showed that the most common cause of patient death after initial survival without transplantation was by self-harm[10]. It therefore follows that a history of multiple previous self-harm or suicide attempts might contraindicate transplant. The number of previous episodes at which the threshold should be set is, however, somewhat arbitrary, and should be considered together with other factors in each individual case. In this previous study, the author also highlights that patients who spontaneously survive POD-induced AHF without transplant have increased mortality compared to matched controls for up to 20 years after discharge, which is possibly due to ongoing psychiatric illness[10]. This highlights a need for rigorous psychiatric follow-up for survivors who are denied OLT because of APCIs, as this may predict increased future mortality risk.

There is mixed evidence of the impact of other significant mental illnesses on short- and long-term prognosis following liver transplant. Undergoing transplantation is a major event that necessitates subsequent life-long commitment to medication compliance, long-term follow-up, and lifestyle modifications, such as abstinence from alcohol[14]. The presence of some severe, chronic mental disorders may limit a person’s ability to adhere to immunosuppression, engage with follow-up, and maintain a ‘transplant-friendly’ lifestyle, and this has been shown to increase the risk of transplant failure by up to ten times[5]. A 2015 meta-analysis found that the presence of depression pre-transplant significantly increased the risk of post-transplant mortality[15]. Other recent studies, however, found no impact on long-term survival in patients with significant mental illness who underwent transplantation[16,17]. Furthermore, there is also evidence that five-year mortality post-transplant is not predicted by the number of previous psychiatric presentations[10]. This variability in evidence means that it is important to consider the presence or absence of poor prognostic indicators (e.g. pre-existing poor compliance) when significant mental illness is being considered as a potential transplant contraindication.

Our finding that almost 9% of assessments for urgent OLT due to POD-induced AHF end with in-episode patient death due to the psychiatric decision not to transplant, highlights the significance and gravity of this assessment and decision. When this is considered in the context of the challenges discussed above, it is likely that these assessments are daunting for the less experienced psychiatrist. Though it was beyond the scope of this study to examine assessors’ experiences, this could be an area of interest to establish any specific support needs and explore how further support may be provided.

The primary limitation of this study lies in its retrospective nature and inability to control for potential confounding factors. We limited this study to include only those patients who died within the same hospital admission episode in order to identify immediate mortality outcomes. Expansion of the patient cohort to include those who died within a specified period post-discharge may help better predict short-term mortality outcomes.

CONCLUSION

This is the first United Kingdom study to report the frequency of death after patients with POD-induced AHF are declined OLT because of psychiatric contraindications, whilst being otherwise medically suitable for transplant. Our finding that this occurs in almost 9% of patients assessed, making up almost 45% of all in-episode deaths, highlights the significance of the psychiatric assessment for urgent liver transplant. It raises the question of whether the development of carefully considered standardised guidelines might help support clinicians to make these difficult decisions and help reduce inter-unit variability in decision-making. Due to the intrinsic nuances of mental illness, consideration would need to be given to the need to maintain an individual case-based approach to decision-making whilst applying any potential guidelines. There is a need for further research exploring quantitative and qualitative aspects of psychiatric assessment for urgent OLT after POD at other clinical centres, both intra- and inter-nationally.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the assistance of Dr Mhairi Donnelly for sharing her data and its analysis, as well as for her help in identifying relevant references.