Drug-Induced Liver Injury: Icelandic Lessons

Abstract

See “Incidence, presentation, and outcomes in patients with drug-induced liver injury in the general population of Iceland,” by Björnsson ES, Bergmann OM, Björnsson HK, et al, on page 1419. See “Incidence, presentation, and outcomes in patients with drug-induced liver injury in the general population of Iceland,” by Björnsson ES, Bergmann OM, Björnsson HK, et al, on page 1419. Drug-induced liver injury is among the more challenging forms of liver disease. It occurs without warning, has no specific distinguishing features, and is capable of mimicking almost any form of liver disease. The diagnosis is particularly challenging because there are no specific markers; drug-induced liver injury being a “diagnosis of exclusion” and often only considered after the fact.1National Institutes of HealthClinical and research information on drug-induced liver injury.www.LiverTox.nih.govGoogle Scholar, 2Fontana R.J. Seeff L.B. Andrade R.J. et al.Standardization of nomenclature and causality assessment in drug-induced liver injury: summary of a clinical research workshop.Hepatology. 2010; 52: 730-742Crossref PubMed Scopus (229) Google Scholar The literature on drug-induced liver injury is also challenging because it is immense and spread over hundreds of medical journals in different languages and representing medical disciplines from hepatology to all subspecialties of internal medicine as well as surgery, gynecology, pediatrics, psychiatry, pharmacology, toxicology, genetics, epidemiology, regulatory affairs, and even ethics. Despite this attention, drug-induced liver injury remains poorly understood and without specific means of diagnosis, treatment or prevention. How common is it? What proportions of cases of acute onset of jaundice or hospitalization for liver disease are caused by medications? What are the major causes of drug-induced liver injury and how frequently does it occur with each implicated medication, and herbal or dietary supplement? What is its mortality rate and what predicts poor outcome? In this issue of Gastroenterology, Björnsson et al3Björnsson E.S. Bergmann O.M. Björnsson H.K. et al.Incidence, presentation, and outcomes in patients with drug-induced liver injury in the general population of Iceland.Gastroenterology. 2013; 144: 1419-1425Abstract Full Text Full Text PDF PubMed Scopus (507) Google Scholar from Reykjavik, Iceland, attempt to answer some of these questions, at least in part. The authors took advantage of the limited but well characterized population of Iceland (about 250,000 adults, similar to the population of Fort Wayne, Indiana, or Orlando, Florida) and the excellent and centralized medical care system enjoyed by that population. This information provided the denominator for what they then systematically collected—all cases of acute drug-induced liver injury that occurred in that country during a 2-year period. By enlisting physicians and all gastroenterologists in Iceland, they then prospectively identified and carefully evaluated all patients presenting with the new onset of suspected drug-induced liver injury. The medical care system and their knowledge of the diagnostic challenges allowed them to obtain a complete battery of tests needed for reliable diagnosis. In this manner, they identified 96 cases over a 24-month period. Because they knew the population from which these cases arose, they could estimate the incidence of drug-induced liver disease as 19 cases per 100,000 persons per year. This incidence is somewhat higher than previous estimates. The higher rate was probably owing to the care and effort made in identifying all cases, some of which were subclinical, many without jaundice. Indeed, using jaundice as the criterion for identifying cases, the incidence would have been 5 per 100,000 population per year, which is not very different from estimates based on retrospective and cross-sectional studies.4Sgro C. Clinard F. Ouazir K. et al.Incidence of drug-induced hepatic injuries: a French population-based study.Hepatology. 2002; 36: 451-455Crossref PubMed Scopus (595) Google Scholar, 5Ibanez L. Perez E. Vidal X. et al.Prospective surveillance of acute serious liver disease unrelated to infectious, obstructive, or metabolic diseases: epidemiological and clinical features, and exposure to drugs.J Hepatol. 2002; 37: 592-600Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar, 6De Abajo F.J. Montero D. Madurga M. et al.Acute and clinically relevant drug-induced liver injury: a population based case-control study.Br J Clin Pharmacol. 2004; 58: 71-80Crossref PubMed Scopus (307) Google Scholar Several other important findings were that the incidence of drug-induced liver injury was similar in women and men, but increased markedly with age (from 9/100,000 among 15- to 29-year-olds to 41/100,000 among patients >70). The increase in rates with age, however, paralleled the increase use of medications (Figure 1), suggesting that the elderly were more likely to suffer liver injury from medications because they were more likely to be taking them. Other conventional “risk factors” for drug-induced liver injury2Fontana R.J. Seeff L.B. Andrade R.J. et al.Standardization of nomenclature and causality assessment in drug-induced liver injury: summary of a clinical research workshop.Hepatology. 2010; 52: 730-742Crossref PubMed Scopus (229) Google Scholar, 7Danan G. Benichou C. Causality assessment of adverse reactions to drugs-I A novel method based on the conclusions of International Consensus Meeting: application to drug-induced liver injuries.J Clin Epidemiol. 1993; 46: 1323-1330Abstract Full Text PDF PubMed Scopus (1109) Google Scholar were not common in this population. Thus, apparently none of the cases occurred in pregnant women, and alcohol abuse was not frequent. The truth, however, is that drug-induced liver injury is not a single, uncommon disease of the general population, but rather a series of rare diseases that occur only in persons who take specific medications. For these reasons, the incidence of hepatotoxicity in the general population is not particularly meaningful. More important is the risk of developing liver injury with a medication. A particular strength of the study from Björnsson et al was the availability of a denominator of the number of persons in the population who were prescribed each medication of interest. Using this information, the authors were able to estimate the risk of liver injury from the implicated drugs. The accuracy of the estimates was limited by the numbers of cases; only 8 medications were implicated in >1 case. But from these 8 the estimated risk of liver injury from amoxicillin/clavulanate (Augmentin) was 1 per 2350 users, diclofenac, 1 per 9148; azathioprine, 1 per 133 (largely anicteric disease); infliximab, 1 per 148; nitrofurantoin, 1 per 1369; isotretinoin, 1 per 732; atorvastatin, 1 per 3693; and doxycycline, 1 per 16,339. Only single instances were identified of other common causes of drug-induced liver injury, such as trimethoprim/sulfomethoxazole, isoniazid, simvastatin, methotrexate, interferon beta, phenytoin, and amiodarone. Interestingly, however, some commonly implicated agents8Chalasani N. Fontana R.J. Bonkovsky H.L. et al.Causes, clinical features, and outcomes from a prospective study of drug-induced liver injury in the United States.Gastroenterology. 2008; 135: 1924-1934Abstract Full Text Full Text PDF PubMed Scopus (649) Google Scholar, 9Andrade R.J. Lucena M.I. Fernandez M.C. et al.Drug-induced liver injury: an analysis of 461 incidences submitted to the Spanish Registry over a 10-year period.Gastroenterology. 2005; 129: 512-521PubMed Google Scholar, 10Reuben A. Koch D.G. Lee W.M. Acute Liver Failure Study GroupDrug-induced acute liver failure: results of a U.S. multicenter, prospective study.Hepatology. 2010; 52: 2065-2076Crossref PubMed Scopus (487) Google Scholar did not appear at all in this cohort from Iceland, examples being ciprofloxacin and the fluoroquinolones, azithromycin and other macrolide antibiotics, minocycline, anabolic steroids, valproic acid, and cancer chemotherapeutic agents (except for 1 case attributed to imatinib). A possible reason for this is suggested by the data from Björnsson et al shown in Supplementary Table 5; these agents are not commonly used in Iceland. Special notice should be made that 16% of cases of liver injury were attributed to use of herbals and dietary supplements (HDS). The clinical phenotype was typically mild-to-moderate acute hepatocellular or “mixed” injury. The specific risk of taking these products could not be calculated because the number of Icelanders who used them was not available. However, the message from these findings is clear and agrees with results from other areas of the world, that herbal products and nutritional supplements are being used more and more frequently and are increasingly implicated in cases of liver injury.8Chalasani N. Fontana R.J. Bonkovsky H.L. et al.Causes, clinical features, and outcomes from a prospective study of drug-induced liver injury in the United States.Gastroenterology. 2008; 135: 1924-1934Abstract Full Text Full Text PDF PubMed Scopus (649) Google Scholar, 9Andrade R.J. Lucena M.I. Fernandez M.C. et al.Drug-induced liver injury: an analysis of 461 incidences submitted to the Spanish Registry over a 10-year period.Gastroenterology. 2005; 129: 512-521PubMed Google Scholar, 10Reuben A. Koch D.G. Lee W.M. Acute Liver Failure Study GroupDrug-induced acute liver failure: results of a U.S. multicenter, prospective study.Hepatology. 2010; 52: 2065-2076Crossref PubMed Scopus (487) Google Scholar, 11Navarro V.J. Herbal and dietary supplement hepatotoxicity.Semin Liver Dis. 2009; 29: 373-382Crossref PubMed Scopus (79) Google Scholar In this series, HDS accounted for 15 cases, 14 of which were symptomatic, 4 jaundiced, and 3 hospitalized. Various HDS products were implicated and the responsible specific herbal or compound was often not obvious, possibly being Camellia sinensis (green tea) in 4 instances. Herbalife products were implicated in 5 cases, thus ranking second only to amoxicillin/clavulanate as a specific cause of drug-induced liver injury in this population. Finally, the results must be viewed in the context of the uniquely well characterized genealogy of Icelanders. Genetic susceptibility has been implicated in injury owing to several drugs.12Daly A.K. Donaldson P.T. Bhatnagar P. et al.HLA-B*5701 genotype is a major determinant of drug-induced liver injury due to flucloxacillin.Nat Genet. 2009; 41: 816-819Crossref PubMed Scopus (830) Google Scholar Whether the homogeneity of the Icelandic population relative to many other countries impacts the frequency of drug-related injury or the types of drugs implicated was not addressed. Therefore, extending the estimated risk for injury owing to specific drugs to populations with more diverse genetic backgrounds must be done with caution. This careful and expert prospective analysis of the incidence of drug-induced liver injury in a well-defined population with an excellent medical care system and thorough records of pharmacy use has provided valuable information on the incidence, spectrum, risk and significance of drug-induced liver injury. Incidence, Presentation, and Outcomes in Patients With Drug-Induced Liver Injury in the General Population of IcelandGastroenterologyVol. 144Issue 7PreviewLittle is known about the incidence of drug-induced liver injury (DILI) in the general population. We investigated the incidence and the quantitative risk of DILI in a population-based cohort. Full-Text PDF Covering the CoverGastroenterologyVol. 144Issue 7PreviewDespite a decreasing burden of peptic ulcer disease and increasing use of proton pump inhibitors, the incidence of nonvariceal upper gastrointestinal bleeding is largely unchanged, suggesting that other undefined risk factors may have an important etiologic role. In this issue of Gastroenterology, Crooks et al, using a well-established database representative of the general English population, conduct a case-control study and develop a model corrected for known risk factors to determine the role of nongastrointestinal comorbidity in nonvariceal upper gastrointestinal bleeding. Full-Text PDF