Liver transplant associated with paracetamol overdose: results from the seven-country SALT study

Abstract

Aims

Acute drug overdose, especially with paracetamol, may cause acute liver failure leading to registration for transplantation (ALFT). Population statistics and between-country differences for ALFT related to overdose have been poorly described. The aim of the present study was to evaluate overdose ALFT in the multi-country Study of Acute Liver Transplantation (SALT).

Methods

All adult overdose-related ALFT, with or without suicidal intent, in France, Greece, Ireland, Italy, the Netherlands, Portugal and the UK between 2005 and 2007 were identified from liver transplant registries and hospital records. These were compared with whole-country and per capita use of paracetamol.

Results

Six hundred cases of ALFT were identified in 52 of 57 eligible transplant centres, of which 114 involved overdose (72 intentional, 10 non-intentional, 32 uncertain). Overdose represented 20% of all-cause ALFT: Ireland 52%, UK 28%, France 18%, the Netherlands 8%, and Italy 1%. Overdose ALFT were mostly females (61%), mean age 33.6 ± 10.9 years. A total of 111 (97%) of the overdoses involved paracetamol. Event rates ranged from one ALFT for 20.7 tons of paracetamol in Ireland, to one for 1074 tons in Italy and one case in 60 million inhabitants over 3 years in Italy to one case in 286 000 inhabitants per year in Ireland. Per-country event rates for non-overdose ALFT exposed to paracetamol were between 2.5 and 4.0 per million treatment-years sold.

Conclusions

Paracetamol overdose was found to represent one-sixth of all-cause ALFT. There was a 50-fold difference in Europe in the rates of paracetamol overdose ALFT, and a 200-fold difference per million inhabitants.

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT

• Paracetamol overdose-related hepatotoxicity is well known. Paracetamol overdose is a frequent, preventable cause of hepatotoxicity and liver unit admissions.

WHAT THIS STUDY ADDS

• Paracetamol overdose, even without suicidal intent, represents a large proportion of acute liver failure leading to registration for transplantation (ALFT). The frequency of paracetamol overdose leading to ALFT varies considerably between countries, whether per ton of paracetamol sold or by number of inhabitants.

Introduction

Acute liver failure (ALF) in its worst form leads to death or transplantation, although in some cases recovery may occur without transplantation. Less severe cases will usually resolve spontaneously and completely, making transplantation a major outcome in acute liver injury. Liver transplantation can only be done in selected and approved centres, so it is relatively easy to obtain complete country-wide data that can be compared with the exhaustive data on population exposure to drugs. In many countries, patients for whom transplantation is planned or who have been identified as requiring transplantation are inscribed in a registry, so that even patients who are not ultimately transplanted are identified. ALF leading to registration on transplantation lists (ALFT) can occur during drug treatment at normal doses, usually as an idiosyncratic (presumably, immuno-allergic) reaction, and we have previously reported on such cases of ALFT from the main Study of Acute Liver Transplantation (SALT), a seven-country study of non-steroidal anti-inflammatory drug (NSAID)-associated ALFT [1–3]. Another mechanism for ALFT is direct toxicity, such as is found with carbon tetrachloride or mushroom poisoning. Paracetamol is about the only marketed drug that is directly hepatotoxic and can lead to ALFT at doses that are close to the usual therapeutic doses [4–6]. Paracetamol overdose can be voluntary, with suicidal intent, or inadvertent [7].

Most studies of paracetamol overdose originate from emergency departments or hepatology units, and give relative rates of transplantation relative to all overdoses. Very few have attempted to identify all cases of overdose ALFT within one or several countries. We took advantage of SALT [1–3] to quantify paracetamol-related overdose leading to ALFT in different countries in Europe.

Methods

SALT is a multinational, multicentre, case-population study comprising France, Greece, Ireland, Italy, the Netherlands, Portugal and the UK, evaluating a 3-year period (1 January 2005 to 31 December 2007),performed at the request of the European Medicines Agency (EMA) in 2009–2012 [1,2,8].

Briefly, all cases of ALFT were identified from liver transplant registries. Demographic, clinical and drug use data within 30 days prior to initial symptoms of liver disease [index date (ID)] were collected for hospital charts. Cases were reviewed and validated by a national case selection committee hepatologist, who defined the ID, and confirmed by the case adjudication committee (CAC) [9].

ALF was classified as: (1) with an identified clinical cause (e.g. viral, immuno-allergic or vascular, not further considered for drug exposure); or (2) without an identified clinical nondrug cause. The latter cases were further classified into (i) acute drug overdose (with or without suicidal intent); (ii) exposed to drugs without demonstrated overdose (including herbal and homeopathic medicines) within 30 days prior to ID; or (iii) not exposed to drugs [2].

Acute drug overdoses were confirmed by the CAC, based on the case description of documented overdosing (number of tablets taken reported in the hospital medical files, empty packs found with the patient, or family report of overdose, etc.) and/or toxic plasma paracetamol concentrations. These cases were also analysed for indications of the intent of overdose (direct evidence such a suicide note or patient admission of suicidal intent) and classified as definitely suicidal, definitely accidental or uncertain intent.

Statistical analysis

Intercontinental Medical Services Ltd (IMS; London, UK) provided per-country population exposure to NSAIDs and paracetamol. Country population sizes were retrieved from publicly available information. Event rates for overdoses are reported as the number of cases per ton of paracetamol sold in the countries of interest and as the number of cases per million inhabitants per year in the country, with 95% Poisson confidence intervals (95% CI).

All statistical analyses were performed using SAS software (SAS Institute, Cary, NC, USA), version 9.3.

Results

The 52 participating transplant centres out of 57 eligible centres (91.2%) contributed 9479 cases registered for liver transplantation, of which 600 were ALFT and 114 were overdose (Figure1). In 111 of these (97.3%), the toxic drug was paracetamol. Of the 301 ALFT cases without clinical aetiology, 81 had been found exposed to paracetamol but without overdose within 30 days before the ID. In the three non-paracetamol overdose cases, the suspect drugs were benzodiazepine derivatives + opioids, ecstasy, and diclofenac + iron. Thirty-one (27.2%) overdose cases had also been exposed to antidepressants, and 30 (26.3%) to psycholeptic agents [1].

Figure 1.

Final case inclusion in seven participating countries over a 3-year period (adapted from Gulmez et al., Drug Saf 2013; 36: 757–64)

Overdose patients were mostly female (61.3%); mean age 33.6 (standard deviation 10.9) years. Most (72.8%) were eventually transplanted (Table 1).

Table 1.

Demographic characteristics of ALFT cases with or without acute overdose

	With drug overdose* (n = 114)	Without overdose (n = 187)
Male [n (%)]	44 (38.7)	88 (47.1)
Mean age at registration [years (SD)]	33.6 (10.9)	40.9 (13.5)
Transplanted [n (%)]	82 (73.9)	161 (86.1)

^*Includes 111 paracetamol overdoses. SD, standard deviation.

Most overdoses (63.0%) were intentional (suicide attempts). Intentionality was uncertain in 28%. Overdose was responsible for 19.0% of all-cause ALFT in the seven participating countries. This was highest in Ireland (52.0%), followed by the UK (28.0%), France (18.0%), the Netherlands (8.0%) and Italy (1.0%). No overdose ALFTs were identified in Greece (122 million Defined Daily Dose [DDD], 212 017 treatment-years, 10.8 million inhabitants) or Portugal (106 million DDD, 154 962 treatment-years, 10 million inhabitants).

Over the same period, 18,530 tons of paracetamol were sold in the seven SALT countries (17 800 in the five countries where overdose ALFTs were observed), so that, on average, one case of paracetamol overdose resulting in ALFT occurred for each 167 tons of paracetamol sold. The risk varied from no events in 118 and 161 tons in Portugal and Greece, respectively, to one event for every 1074 tons of paracetamol sold in Italy, to one for every 20.7 tons in Ireland (Table 2). Per-population rates ranged from one case of overdose ALFT for 60 million adult inhabitants over 3 years in Italy to one case per 280 000 in Ireland (or 0.006 cases per million inhabitants per year in Italy to 1.167 in Ireland) (Table 2). Per capita use of paracetamol estimated from tonnage sold per country and number of inhabitants ranged from 6 tons per million inhabitants per year in Italy to 51.5 tons in France (Table 2). In Greece and Italy, per capita use of paracetamol was 5.4 and 3.6 tons per million inhabitants per year, respectively.

Table 2.

Acute paracetamol overdoses leading to liver transplantation in seven countries

Country	APO	% of all ALFT cases	Tons of paracetamol sold	Population (million)	Tons of paracetamol per million inhabitants per year	APO per 1000 tons of paracetamol (95% CI)	APO per million inhabitants per year (95% CI)
France	31	18	10043.2	65	51.5	3.09 (2.10–4.38)	0.159 (0.108–0.267)
Ireland	14	52	289.6	4	24.1	48.8 (26.4–81.2)	1.167 (0.638–1.95)
Italy	1	1	1074.4	60	5.97	0.94 (0.03–5.18)	0.006 (0.0002–0.031)
Netherlands	2	8	557.6	16	11.6	3.59 (0.43– 13.0)	0.042 (0.005– 0.151)
United Kingdom	63	28	6286.5	60	34.9	10.0 (7.70–12.8)	0.350 (0.269–0.448)
Greece	0	0	161.5	10	5.38	–	–
Portugal	0	0	118.1	11	3.58	–	–
All	111	19	18531	226	27.3	5.99 (4.93–7.22)	0.164 (0.135–0.197)

AFLT, acute liver failure leading to registration for transplantation; APO, acute paracetamol overdose leading to registration for liver transplantation. Population, country population in million (from Wikipedia); Tons of paracetamol sold provided by IMS international.

There was no apparent relationship between per capita use of paracetamol and the risk of ALFT by overdose per capita or per ton of paracetamol. However, the event rate per ton of paracetamol sold, or per inhabitant in the country, varied compared with all countries pooled, with a significantly greater risk in Ireland than in other countries (Table 3) compared with all countries pooled. Ireland and the UK had a significantly greater risk of ALFT per ton of paracetamol sold, or per inhabitant in the country, with 7.75 and 1.6 times more cases in Ireland and the UK, respectively, than in the rest of Europe (Figure2).

Table 3.

Relative risk of acute liver failure leading to liver transplantation, in different European countries, compared with all countries together

Country	ALFT by paracetamol overdose	Tons of paracetamol sold	Population (million)	RR of overdose ALFT per ton of paracetamol (95% CI)	RR of overdose ALFT per million inhabitants per year (95% CI)
France	31	10043.2	65	0.52 (0.39–0.68)	0.97 (0.65–1.45)
Ireland	14	289.6	4	8.07 (3.00–21.7)	7.13 (4.09–12.4)
Italy	1	1074.4	60	0.16 (0.12–0.21)	0.03 (0.005–0.24)
Netherlands	2	557.6	16	0.60 (0.36–0.99)	0.25 (0.06–1.03)
United Kingdom	63	6286.5	60	1.67 (1.26–2.22)	2.14 (1.57–2.91)
All (reference)*	111	18531	226	1.00	1.00

^*Includes Greece and Portugal with no ALFT. ALFT, acute liver failure leading to liver transplantation; CI, confidence interval; RR, relative risk.

Figure 2.

Number of cases of overdose-related acute liver failure leading to transplantation per ton of paracetamol sold in the different countries and overall

By contrast, when non-overdose paracetamol-associated ALFTs are considered, the event rates were very similar, between 2.6 and 4.1 (mean 3.5) cases per million treatment-years of paracetamol sold (Table 4), with no difference between countries (Figure3).

Table 4.

Non-overdose ALFT with exposure to paracetamol within 30 days before first symptoms, per country (95% CI)

Country	Non-overdose ALFT	MTTY	ALFT/MTTY
France	49	13.2	3.72 (2.75–4.91)
Ireland	1	0.38	2.63 (0.079–14.7)
Italy	4	1.41	2.84 (0.77–7.26)
Netherlands	3	0.73	4.10 (0.62–8.77)
United Kingdom	24	8.25	2.91 (1.86–4.33)
All*	81	24.3	3.33 (2.65–4.14)

^*Includes Portugal and Greece with no cases. ALFT, acute liver failure leading to registration for transplantation; CI, confidence interval; MTTY, million treatment-years sold.

Figure 3.

Non-overdose acute liver failure leading to transplantation in patients exposed to paracetamol within 30 days before first symptoms, per million patient-years of paracetamol sold in each country

Discussion

In the present study, which compiled exhaustively all cases of liver transplantation over 3 years in seven countries [1,2], paracetamol was involved in 111/114 (97.3%) ALFTs. The three other cases involved opiates (possibly hepatotoxic due to biliary spasm), ecstasy (where the use of other drugs of abuse cannot be eliminated), and diclofenac + iron (where the role of acute iron overdose may be suspected). Like most NSAIDs, diclofenac is associated with liver toxicity, but this is idiosyncratic and usually not related to overdose [1]. We cannot eliminate that these three cases might also have ingested paracetamol without it being recorded.

The method we used within the liver transplant units, coupled with the transplant registries, ensured that we missed no cases of ALFT, whatever the cause, over the period and area of the study. Because paracetamol overdose is one of the main causes of ALF, it is highly unlikely that paracetamol overdose would not have been sought as an explanation in a case of otherwise unexplained ALF, from patient history or family, or biological indicators of paracetamol poisoning. The course of ALF caused by paracetamol overdose is well known, and usually progressive over a few days, so that the hypothesis of acute overdoses leading to fatal liver failure that would not reach the hospital is unlikely unless the patient specifically refused admission. Overdose represented 10–50% of ALFT according to the country.

The total amount of paracetamol sold per country can be ascertained from wholesalers and manufacturers. IMS has a strong reputation for providing reliable sales figures; these include pharmacy and non-pharmacy sales but do not cover ‘grey market’ sales – i.e. parallel imports, sales over the internet or products bought abroad, or the export of the drug to other countries [10–12]. Other studies have not used exhaustive nationwide data from transplantation registries and therefore have not provided the information per million inhabitants or per ton of paracetamol sold nationwide [13].

The event rate for paracetamol overdose-related ALFT was not found to be homogeneous throughout Europe. It ranged from no cases in Greece or Portugal, out of about 10 million persons each, over 3 years, to one case in 60 million inhabitants over 3 years in Italy, to one per 0.3 million per year in Ireland – a 200-fold difference. The event rate per ton of paracetamol sold was also very variable, ranging from one overdose ALFT for 1074 tons of paracetamol sold in Italy, to one for 20.7 tons sold in Ireland. Per-person use of paracetamol ranged from 3.5 tons per million inhabitants per year in Portugal, to 51.5 tons in France. The highest rates of overdose ALFT per ton of paracetamol sold or per inhabitant were found in the two English-speaking countries (Ireland and the UK).

There are two hypotheses to explain the latter result: that there is a higher overall rate of paracetamol overdose in these two countries, with the same proportion of the more severe ones leading to transplantation; or that, for some reason, overdoses may be more severe in these countries, resulting in more severe liver failure, suggesting a possible genetic or a societal background for this higher incidence of ALFT. However, the fact that non-overdose ALFT occurs with about the same incidence in all European countries is an argument against a genetic predisposition to more severe hepatic injuries, and probably also against an environmental or societal explanation such as per capita alcohol consumption, which is highest in Portugal and France according to the World Health Organization [14]. As we do not have event rates for overdoses not leading to liver failure, we cannot conclude on rates of non-ALFT overdoses in the different countries, but indicators point to more common use of paracetamol for self-poisoning in these countries.

There was no relationship between the amount of paracetamol used yearly per million inhabitants and the number of overdoses per inhabitant or per ton of paracetamol sold. Reducing total or per capita use of paracetamol in a country would not necessarily reduce the number of acute drug overdoses leading to transplantation. France has the highest per capita use of paracetamol but the third lowest ALFT rate.

Changing the amount per preparation or per box might change the risk of liver failure. In Ireland, legislation was introduced in October 2001 to control the sale of paracetamol in non-pharmacy outlets. Donohoe et al. assessed the impact of this legislation on acute deliberate paracetamol overdoses and concluded that the legislation controlling the sales and packaging of paracetamol preparations appeared to be associated with a significant fall in the number of tablets taken in acute deliberate paracetamol overdoses [15]. Despite this, paracetamol overdose was still involved in over 50% of all ALFTs in Ireland [16]. In the UK, paracetamol was found to be the principal cause of poison-related deaths and liver transplants for ALF [17]. Paracetamol overdose accounted for about 40% of cases of ALF in the UK [18], and also accounted for over 200 admissions to a liver unit and 20 or so liver transplants per year [19]; our results were similar (i.e. 60 overdose ALFTs over 3 years in SALT in the UK), which confirms the validity of both studies. In a recent study by Teo et al., in adults (>13 years) admitted to the emergency department short-stay ward of Aberdeen Royal Infirmary in 2009 because of poisoning, almost half of 1062 cases were polypharmacy; alcohol was involved in 40% of cases and overdoses most commonly involved paracetamol (43%) [20]. As paracetamol hepatotoxicity is the leading cause of overdose ALFT and yet is completely preventable, regulatory changes regarding the labelling and dispensing of paracetamol-containing products have been implemented [21–23]. In the UK, blister packaging and restrictions on the dispensing of paracetamol tablets in 1998 led to a reduction in the number of patients intentionally overdosing and of those referred for liver transplantation during the 11 years after the introduction of the legislation [24–26]. Despite this, after Ireland, the UK remains the country with the highest rate of ALFT due to paracetamol overdose per million inhabitants or per ton of paracetamol sold. The number of cases we found in SALT was too small to derive temporal trends over only 3 years. Paracetamol overdose was responsible for only 1.0% of all-cause ALFT in Italy, and no cases of overdose ALFT were identified in Greece or Portugal. In a recent retrospective study [27] analysing the results of liver transplantation for ALF in the region of Castilla y Leon in Spain, the most common aetiology was toxic exposure to agricultural products. No cases were related to paracetamol overdose, which is reported to be a rare cause of ALF in Spain [28].

The present study was the first population-based study of paracetamol overdose leading to registration for liver transplantation. Most studies of drug-induced liver injury often have as the point of capture admittance to emergency or hepatology departments, and have only provided the relative weight of ALF sent for transplantation out of all cases of ALF or all acute overdoses, rather than absolute numbers. As in the present study, they found paracetamol overdose to be the leading cause of ALFT, representing up to 50% of all such cases. Overdose ALFT represents only a small percentage of all liver transplants (1%) but a non-negligible proportion (1/6) of cases of unexplained or drug-related ALF [1,2]. This avoidable cause deprives other potential recipients of a much-needed organ. There is a need to continue working on avoiding overdose ALFT. Reducing pack size is a likely answer, and seems to have worked in the UK. In France, pack sizes were limited to 8 g per pack some time ago.

It is unlikely that limiting individual tablet strength for prescription paracetamol in the US would be efficient at preventing overdose ALFT when paracetamol can still be bought over the counter at doses of 500 mg or 1 g per tablet in large pack sizes. These large pack sizes make large amounts of a dangerous product easily available for an impulse overdose. This is not the case for limited pack sizes in blister packs, as found in other countries for over-the-counter usage. However, this is certainly not the only reason for the differences in event rates between countries, and more work needs to be done to understand why people in Ireland and the UK commit suicide more often with paracetamol than in other countries, or at least have more cases of ALFT.

Conclusion

Paracetamol is still the almost exclusive cause for liver transplantation related to acute drug overdose, with an average of one case per 6 million inhabitants per year over seven countries in Europe. This seems to be about eight times less than the number reported for the US.

Large differences were found between countries in event rates, with a six-times higher risk in Ireland and a two-fold higher risk in the UK compared with the average of the countries participating in the study. The reasons for these differences are uncertain but could give indications for their prevention.

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 [SEG, MT, SLM, SM, FH, RL, JJ, PB, NM], and travel support or consulting honoraria from the University of Bordeaux [DL, GPP, JB, FB, YH, DT, AMC, BS] for the submitted work; no other relationships or activities that could appear to have influenced the submitted work.

The authors wish to thank all the physicians and staff at the transplant centres and all the patients whose acceptance of this study made it possible, and the various persons in administrative positions, data protection, ethics and the British Research & Development (R&D) committees.

The corresponding authors had full access to all of the data in the study and take responsibility for the integrity of the data and accuracy of the data analysis, which was conducted independently from the funder.

Contributors

Nicholas Moore and Dominique Larrey had the original idea for the study several years ago, and proposed it to the sponsor and to The Committee for Medicinal Products for Human Use (CHMP). NM was the overall study supervisor, intervening when necessary for the smooth operation of the study. DL was chairman of the scientific committee and of the CAC, and was the everyday hepatology adviser for the study. George-Philippe Pageaux was the vice-chairman of the CAC, and contributed enormously to case understanding and adjudication. Sinem Ezgi Gulmez was the scientific study coordinator, devising the initial document generation under the control of the scientific committee, organizing negotiations with transplant centres and study data retrieval, writing the study reports, drafting articles and verifying all contents. Séverine Lignot was the operational study coordinator. Sophie Micon and Fatima Hamoud coordinated the scientific and CAC activities. Régis Lassalle and Jérémy Jove provided data management and statistical analyses. Patrick Blin was the study epidemiological overseer, contributing to study design, operations, analysis and understanding. Massoud Toussi provided drug utilization data and helped considerably in their understanding and interpretation. Dominique Larrey, Georges-Philippe Pageaux, Jacques Bernuau, Franco Bissoli, Yves Horsmans, Bruno Stricker and Douglas Thorburn were members of the CAC and performed causality assessments. Aiden McCormick was the national coordination in Ireland.

All authors contributed comments and approved the final version of this paper.

This study was funded by Helsinn Healthcare SA (HHC), at the request of the EMA's CHMP, but was conducted independently. The sponsor had no role in the design of the study (which was submitted to and approved by the CHMP); the collection, management, analysis or interpretation of the data; or the preparation or review of this manuscript. They were given the opportunity to read and comment on this paper before submission, but the content of the paper was entirely under the control and responsibility of the authors. HHC contracted with the University of Bordeaux for the conduct of the study. All contracts with experts or transplant centres were with the University of Bordeaux. All study-related expenses were paid for by the University of Bordeaux from contractual financing by Helsinn. HHC had no direct contact with any of the participants in the study.

The report of this study was sent to the regulatory authorities and was presented to the CHMP on 17 May 2011.