Adverse drug reactions affecting the liver represent an important challenge for physicians, health authorities and pharmaceutical companies. Drugs can reproduce practically the whole spectrum of liver diseases, but acute hepatitis is the most common syndrome [1–3]. Acute liver failure is the most severe clinical expression and represents the first cause of fatalities related to drugs. As a consequence, it is also the first cause of drug withdrawal from the pharmaceutical market [1–3]. This is particularly well illustrated by the recent example of troglitazone . This drug, representing a very promising prototype of a new class of anti-diabetic agents, was approved in the United States in 1997 . The occurrence of many cases of acute liver failure that required liver transplantation or had a fatal course, led to its withdrawal from the market in the United Kingdom by 1999 and in the United States in March 2000 .
Besides the large number of classical drugs (more than 1100) reported to exhibit potential hepatotoxicity [1–3], other agents should also be taken into consideration. These include the excipients present in formulations of drugs , herbal medicines which are increasingly consumed and often not disclosed [6,7], recreational and illegal compounds such as cocaine and synthetic amphetamine derivatives (such as ecstasy) .
Drug-induced acute liver failure raises several important issues: what is the epidemiology of drug-induced liver failure compared to other causes? What is the prognosis? Is there any mean to predict such an event?
Before addressing these questions, it is necessary to be reminded of what constitutes acute liver failure. Although there is no complete agreement on the definitions of acute liver failure and its potential subtypes, there is a very large majority for defining it as a condition characterized by sudden and massive liver cell dysfunction, occurring in a previously healthy person with no known underlying liver disease, and leading to a marked alteration of blood clotting tests and hepatic encephalopathy [8–10].
Epidemiological aspects of drug hepatotoxicity and drug-induced acute liver failure
The prevalence and incidence of drug hepatotoxicity are still very partially known . The majority of data are provided by retrospective studies of database from pharmacovigilance centres and/or pharmaceutical companies, aimed to describe what are the most frequently hepatotoxic drugs and their clinical characteristics . These studies have also tried to estimate the prevalence of hepatotoxicity of a given drug. By this imperfect process, it is clear that many events remain ignored and we become aware only of the ‘tip of the iceberg’ .
Prospective studies mostly consist of the assessment of hepatic adverse events during clinical trials required for new drugs at the pre-marketing period. Now, particular attention is being paid to the frequency of serum alanine aminotransferase (ALT) elevation over three times the upper limit of normal (ULN), the increase in bilirubin, and clinical events, in the treated group compared to the placebo group or another well known comparator. Unfortunately, these studies have been of limited value. The number of subjects involved in clinical trials generally ranges from 2000 to 5000 . As a result, only frequent events (>1% of subjects affected) can be detected. However, for the majority of drugs, the risk of hepatotoxicity ranges from 1/10 000 to 1/100 000 [1–3]. Such was the case for troglitazone, for which hepatotoxicity was estimated to be around 1/50 000 persons exposed . This explains why liver injury was not clearly detected among the 5000 patients included in clinical trials. For some drugs, such as anti-histamines, penicillins and minocycline, hepatotoxicity is exceedingly low (below 1/1 000 000). This makes it impossible to detect at an early stage, and also makes the demonstration of a particular susceptibility factor extremely difficult [1–3].
Finally, our knowledge of the incidence of drug-induced adverse effects in the liver, for a general population, is based on a single recent prospective study performed over 3 years . It was found to be 14/100 000 persons and corresponded to a number of events 16 times higher than that collected by pharmacovigilance centres. This study shows that the low reporting rate previously estimated to be around 10% [1–3,5] is probably worse and that one should be very careful in the interpretation of studies trying to compare the hepatotoxicity risks of different drugs, on the basis of spontaneously reported cases analysed retrospectively.
Epidemiological studies regarding drug-induced acute liver failure exhibit similar limitations. Retrospective studies suggested that drugs may have caused around 10–20% of all cases of fulminant and subfulminant hepatitis . Recently, the first prospective study focused on the causes of acute liver failure was performed on the basis of 308 consecutive patients admitted over a 41 month period in 17 tertiary care centres, in the US . Paracetamol overdose was the most common apparent cause of acute liver failure, accounting for 39% of cases . Idiosyncratic drug reactions were the presumptive cause in 13% of cases, in the same range as viral hepatitis A and B combined, indicated in 12% of cases . These data strongly support the suggestion that drugs have replaced viral hepatitis as the most frequent causes of acute liver failure.
Main drugs responsible for acute liver failure
The main drugs responsible for acute liver failure vary over time and with the area being considered. In UK and the US, paracetamol is clearly, and by far, the first causative drug [9,12]. In other Western countries, there is evidence that paracetamol intoxication represents an increasing contribution . Troglitazone has been an example limited to the US . Worldwide, the major drugs are anti-tuberculous agents, other antibiotics, anaesthetic drugs of halothane family, and non-steroidal anti-inflammatory drugs (NSAIDs) as recently illustrated by two compounds, bromfenac and benoxaprofen, which have been quickly withdrawn because of severe liver injury [12,13]. Compounds other than classical drugs can have a significant place. Ecstasy, an amphetamine derivative which is increasingly used by teenagers from Western countries, is one of the major agents causing drug-induced hepatic failure in Scotland as shown by Smith et al. in this issue of the Journal . Herbal medicines are playing an important role in Asian and African countries [6,7].
Prognosis and treatment of drug-induced acute liver failure
The natural prognosis varies between drugs. In the case of paracetamol intoxication, the spontaneous mortality rate is estimated to be around 32–50% [1–3,12]. For other drugs, it ranges from 75% to 90% [1–3,10,12].
Once acute liver failure has occurred, there is no specific treatment and emergency liver transplantation represents the best chance of survival. Acute liver failure represents 9% of indications for liver transplantation . Taking all causes of liver failure into account, the 5 year survival rate of transplanted patients is 59% . In patients specifically transplanted for drug-induced liver failure, data are scant. The prospective survival rate has only been described as short term (3 weeks) and was 94% for paracetamol and 53% for other drugs . The measures to prevent acute liver failure are quite limited. In paracetamol intoxication, the rapid administration of N-acetylcysteine is a classical antidote. Once overt hepatitis is present or when ALT level increases over 3–5 times the upper limit of normal (ULN), the main therapeutic measure consists of stopping the offending drug. At the stage of liver failure, treatment is mostly supportive. Since irreversible damage is unpredictable, early transfer to a transplantation centre should be considered.
Predictive factors of drug-induced acute liver failure
The course of acute liver injury to liver failure may be modulated by the following parameters: the continuation of causative drug administration despite the onset of liver injury, the age with a higher risk for older people, a pre-existing cirrhosis, fasting, denutrition, chronic alcohol abuse, and sometimes the amount of ingested drug (paracetamol) [1–5].
In contrast, the frequency of ALT elevation does not appear to have a predictive value on the risk of acute liver failure for drugs with rare hepatotoxicity as illustrated by troglitazone and some other instances [4,16]. Although the combination of ALT>3×ULN and total bilirubin >2×ULN, called Hy's rule (from Hyman Zimmerman's clinical experience) , has recently been proposed by the US Food & Drug Administration to predict a 10% risk of liver failure , this point has not been really validated and requires further studies.
Acute liver failure is the most severe complication of drug-induced hepatotoxicity with a high spontaneous mortality rate and represents the first cause of drug withdrawal from the market. Estimates of overall incidence in a general population remain unknown. However, drugs appear to be the first of all causes of acute liver failure. In Western countries, paracetamol represents the first cause of all liver failure. Non-paracetamol drugs given at normal dosage contribute in an equivalent manner to combined viral hepatitis A and B. As the rapid deterioration is unpredictable, early liver transplantation should be considered whenever it is possible because it represents the best chance of survival.
1. Pessayre O, Larrey D, Biour M. Drug-induced liver injury. In: Bircher J, Benhamou JP, McIntyre N, Rizzetto M, Rodés J (editors): Oxford Textbook of Clinical Hepatology, 2nd edition, Vol. 2. Oxford: Oxford University Press; 1999, pp. 1261–1315.
2. Zimmerman HJ. The Adverse Effects of Drugs and Other Chemicals on the Liver. 2nd edition. Lippincott Williams & Wilkins; 1999.
3. Larrey D. Drug-induced liver diseases. J Hepatol 2000; 32:77–88.
4. Watkins PB, Whitcomb RW. Hepatic dysfunction associated with troglitazone. N Engl J Med 1998; 13:916–917.
5. Larrey D. Epidemiology
and individual susceptibility to adverse drug reactions affecting the liver. Semin Liver Dis 2002; 22:145–155.
6. Larrey D. Hepatotoxicity of herbal medicines. J Hepatol 1997; 26 (suppl 1):47–51.
7. Seeff LB, Lindsay KL, Bacon BR, Kresina TF, Hoofnagle JH. Complementary and alternative medicine in chronic liver disease. Hepatology 2001; 34:595–603.
8. Williams R. Classification, etiology, and considerations of outcome in acute liver failure
. Semin Liver Dis 1996; 16:343–348.
9. Williams R. Fulminant hepatic failure. Semin Liver Dis 2003; 23:201–202.
10. Bernuau J, Benhamou JP. Fulminant and subfulminant liver failure
. In: Bircher J, Benhamou JR, McIntyre N, Rizzetto M, Rodés J Oxford Textbook of Clinical Hepatology, 2nd edition, Vol, 2. Oxford: Oxford University Press; 1999, pp. 1341–1372.
11. Sgro C, Clinard F, Ouazir K, Chanay H, Allard C, Guilleminet C, et al. Incidence of drug-induced hepatic injuries. A French population-based study. Hepatology 2002; 36:451–455.
12. Ostapowicz G, Fontana RJ, Schiodt SV, Larson A, Davern TJ, Han SH, et al. Results of a prospective study of acute liver failure
at 17 tertiary care centers in the United States. Ann Intern Med 2002; 137:947–954.
13. World Health Organization. Bromfenac withdrawal. WHO Drug Information 1998; 12:146.
14. Smith IDM, Simpson KJ, Garden OJ, Wigmore SJ. Non-paracetamol drug-induced fulminant hepatic failure among adults in Scotland. Eur J Hepatol Gastroenterol 2005; 17:161–167.
16. Russo M, Watkins P. Are patients with elevated liver tests at increased risk of drug-induced liver injury? Gastroenterology 2004; 126:1477–1479.