- Email: [email protected]
Spontaneous reporting of drug-related hepatic reactions from two Italian regions (Lombardy and Veneto)
LIVER,PANCREAS,AND BlllARY TRACT
lJlGESTLIYER OIS 2000;32:716-23
Spontaneous reporting of drug-related hepatic reactions from two Italian Regions (Lombardy and Veneto) A. R. E. G. U. M. F.
Conforti Leone Ghiotto Velo Moretti Venegonil Bissoli2
&ion: Clinical Pharmacology Unit, “G.8. Rossi” University Hospital Verona; ’ Department of lf7ternal Medicine, Fatebenefratelii Hospita!, Man; p Liver Unit, Fomarvii Hospital, Megenta, Italy. lS‘lMp#&filW: Almw#k Dr. A. Conforti, lstituto di Farmacologia, Universicci di Verona, 37134 Vemna, &e/y. Fax: +39-045-581 I I 1. E-meil: [email protected] 1-b: The Authors are vary grateful Phermeceutice~ Departments Heaith Districts of Lombardy Veneto Regions for co/letting adverse reaction forms.
to the and local and the
Submitted May 18, 2000. Revised September 8, 21000. Accepted September 18, 2000,
716
Background and aims. Many currently used drugs are implicated in liver disorders ranging from asymptomatic abnormal liver function tests to fatal liver failure, and drug-induced hepatic reactions are the most frequent cause of the withdrawal of drugs from the market in the United States, United Kingdom and Spain. The aim of this study was to review the drug-induced hepatic reactions notified to the spontaneous surveillance system of two Italian regions [Lombardy and the Veneto] up to 31 December 1998. Patients and methods. The reports of hepatic adverse drug reactions were extracted from an inter-regional database of spontaneous adverse drug reaction reports and analysed by reporter category patient age and sex, the underlying disease, the adverse reaction and drug exposure. In the case of drugs involved in more than four reports, a search was made of the Medline, Hepatox, Micromedex and WHO databases. Results. Between 1988 and 1998, there were 3 IO reports of hepatic lesions associated with single drugs or fixed combinations, the most frequently involved patient categories being females and the elderly Sixty percent of the reactions were classified as serious, with three fatal cases. Thirteen drugs were associated with more than four reports and accounted for 35% of all of the hepatic reactions. Six drugs not known as hepatotoxic agents were associated with hepatic reactions. Conclusions. In this study spontaneous reports have suggested previously unknown hepatotoxicty of six drugs, and have improved drug safety profiles as shown in an Italian inter-regional health service.
Digest
Liver
Key words:
Ilk
2000;32:716-23
drug-induced
hepatic
reaction;
spontaneous
reporting
Introduction Many currently used drugs are implicated in liver disorders ranging from asymptomatic abnormal liver function tests to fatal liver failure. Although the risk of acute hepatic injuries due to individual drugs other than some well-known exceptions is generally considered to be low ‘, hepatic drug reactions are the most frequent cause of the withdrawal of drugs from the market in the United States, United Kingdom and Spain 2. Despite under-reporting, the impossibility of making reliable estimates of reaction rates, and its variability in terms of accuracy and precise terminology, spontaneous adverse drug reaction (ADR) reporting and publications of case
A. Conforti et al.
reports 3-5acts as an early warning system and provides data for the generation of research hypotheses. The spontaneous reports of drug-associated liver injuries received over a period of 21 years and reviewed by the New Zealand Centre of ADR Monitoring represented 4.2% of all reports and 7.4% of all fatal occurrences 6, and similar results were found in Denmark, where 6% of the ADRs reported during a 20-year period involved hepatotoxicity ’ 8. The aim of this study was to analyse the hepatic drug reactions notified to the spontaneous surveillance system of two Italian Regions (Lombardy and Veneto) up until 31 December 1998. The most frequently reported associations were correlated with published data, a specific hepatic drug toxicity bibliographical database 9, and the WHO database of spontaneous reports coming from 59 countries worldwide lo. Some previously unreported or poorly described hepatotoxic drug effects are also discussed.
Methods In Italy, spontaneous reporting of ADRs has been mandatory since 1987 and, in keeping with the law, doctors and pharmacists have to send all suspected ADRs to the Local Health authorities, which forward the reports to the Ministry of Health. The results of such a centralized organization are inadequate, the main problem being represented by the widespread under-reporting. In 1994 (the last year with available national data), the reporting rate by doctors was about 33/million inhabitants. However, some local initiatives were commenced over the last few years. The data presented in this paper were obtained from a database containing all of the spontaneous ADR reports coming from Lombardy and Veneto, two Italian Regions with a combined population of about 13.7 million and a 1998 reporting rate of 202.7 per million inhabitants. In Veneto, the reports have been examined regularly and entered into a database since 1988 by the Clinical Pharmacology Unit of the University of Verona; in Lombardy, a similar analysis has been carried out by the Regional Pharmacovigilance Group since 1993. All of the reports notifying one or more hepatic reactions and entered in the database as of 31 December 1998 were analysed by a panel of experts consisting of gastroenterologists, hepatologists, specialists in internal medicine, pharmacologists and pharmacists, who took into account the reporter category (hospital doctors, general practitioners and others), patient age and sex, the underlying disease, the characteristics of the adverse reaction (latency, duration, outcome, rechallenge, seriousness) and drug exposure (indication, duration of treatment and dosage).
The drugs were classified according to the Italian Codifa system and their Anatomical Therapeutic Chemical Classification (ATC), and the adverse reactions according to the terminology adopted by the WHO I’, while maintaining as far as possible the terms used by the reporter in defining the liver disorders. The seriousness of the reactions were classified on the basis of the WHO Critical Term List, in which a serious adverse reaction is one that is fatal, life-threatening, involved or prolonged hospitalisation, or led to persistent or significant disability or incapacity. Grouping the reports on the basis of the high-level terms of this list, the following types of serious liver disorders were found: hepatitis (including toxic and acute hepatitis), cholestatic hepatitis (including cholestatic jaundice), hepatocellular damage (including hepatocellular jaundice and hepatotoxicity), hepatic necrosis, hepatorenal syndrome, and hepatitis C. The type of hepatic damage was classified on the basis of the WHO criteria and, when laboratory test results were available, also using the criteria adopted by an international consensus meeting 12. Each report was graded according to the WHO documentation criteria: grade 1 when it included only the date of onset of the reaction and the dates of treatment; grade 2 when it also included the reason for the treatment and the outcome of the reaction; and grade 3 in the case of a report notifying a positive rechallenge and including all of the other data mentioned for grades 1 and 2. All of the drugs reported as causing liver adverse effects were used as keywords for a general search of the literature (Medline 1966- 1999 and Micromedex 19741999 “), with the Medline search being based on a full text strategy crossing each drug with all of the terminology compatible with a liver disorder; the selected articles were checked and accepted after reading the abstract. We also referred to the specialised Hepatox 9 and WHO Monitoring Centre Spontaneous Reporting databases lo. Any drug- associated hepatotoxicity that had not previously been clearly reported in the literature was flagged as a possible “signal”. Results
A total of 13,118 ADR reports were received by the Veneto and Lombardy regional centres during the period under consideration, 388 of which referred to a liver disorder. The annual percentage of hepatic ADR reports in Veneto and Lombardy is shown in Figure 1. The number of both types of report increased over time, however a comparison of data of the two regions in the common observation period shows a higher detection of hepatic ADR in Lombardy, the global percentage of reported hepatic disorders being 2.5 in 717
Spontaneous reporting of adverse hepatic drug reactions
Fip.1. Annual percentage of hepatic reactions in Veneto Im1 and Lomlardy (01 regions.
Veneto (in the period 1988- 1998) and 3.5 in Lombardy (in the period 1993-1998). As far as concerns the origin of the reports, 60% came from hospital doctors, 37% from general practitioners, 1% from others, and 2% from unspecified reporters. The age-related distribution of the reports indicating drug-induced liver disorders (Fig. 2) shows that the largest number referred to patients aged 65-74 years. This pattern is similar to that related to the number of
reports as a whole and, therefore, does not indicate any age-specific risk for liver reactions. The female/male ratio was 1.4, which is slightly lower than the 1.6 observed for reactions affecting other organ systems. The documentation grading of the reports was 7.5% grade 1, 78.4% grade 2, and 2.7% grade 3; 11.3% of the reports were graded 0 because they did not include the dates of onset or treatment. In about 16% of reports, laboratory tests were available. The outcome was complete recovery in 62.6% of the patients, and is unknown in 33.2% because of the incomplete nature of the report (although the need for hospitalisation was specified in 35 cases). Three patients died (two during treatment with flutamide and one during treatment with nevirapine), and two patients suffered permanent disability (one case of chronic HCV-related hepatitis attributed to immunoglobulins, and one of liver cirrhosis attributed to amiodarone) . The suspected drug was a single agent in 283 reports, a fixed combination in 27, and two or more drugs in 78. Our analysis was limited to the 310 reports concerning single drugs or fixed combinations, 187 (60%) of which involved a serious hepatic reaction (a percentage that is nearly double the 27% of serious reactions involving other systems or organs in our database): these are listed in Table I, but it must be emphasised that their seriousness is based on their potential negative consequences and not on their real outcome in the individual patients. The 13 drugs involved in more than four reports, including their documentation grading, patient age and sex, the type of liver damage, and its latency, defined as the time interval between the first drug administration and the first appearance of liver adverse reaction are listed in Table II. These drugs were responsible for 35% of all of the liver reactions reported in the database and caused both cytotoxic and cholestatic reac-
Tat& I. Serious hepatic adverse drug reactions reported in Lombardy and Veneto Regions, grouped according to the high-level terms listed in the WHO Critical Term List [see Methods). lli@li-leuol
-25
5-14
15-44 Age
45-64
65-74
=-74
Total
(ye=4
Fig. 2. Mean annual reporting rate of hepatic ADRs: distribution according to age.
718
term
Hepatitis Cholestetic hepatitis Hepatocellular damage Hepatic necrosis Hepatorenal syndrome Hepatitis C Total
Inaludod
teml
Toxic hepatitis, acute hepatitis Jaundice, chotestatic jaundice Liver disease, hepatotoxicity
No. t%POttS 107
59 16 3
1 1 187
R. Conforti et al.
Table II. Single agents or fixed combinations associated with more than four reports of hepatic reactions.
Amineptine
21 (6.81
1
Flutamide
10 (3.21
1
3
I4
3
A@ hlll#d
Selt F nil
23-78
17 4
Hepatotoxic effect (61, acute hepatitis 161, hepatocellular liver inj. 121, hepatic function abnormal, mixed liver inj., increased liver em., cholestatic hepatitis, hyperbilirubineemia, jaundice, jaundice and hepatotoxicity
7 days 3 months
22-83
1
Cholestatic hepatitis 141,fatal acute hepatitis (21, increased liver em., toxic hepatitis, iaundice end hepatic failure,
43 days 5 months
9
Type 0finjuqP
hepatocelluler
Fenofibrate
10 (3.21
10
42-70
5
5
htancy
liver inj. 2 weeks 3 years
h'epetocellular liver inj. (21, cholestatic liver inj. (2l, liver inj., increased liver
em., hepatic function abnormal, hepatitis and cholangitis, cholestatic hepatitis, hepatitis Tclopidine
Amioderone Simvastatin
9 (2.91
3
8 (2.61
6
8
8 (2.6)
8
42-81
54-82 49-72
3
4 5
6
4 3
Cholestatic hepatitis (31, cholestatic liver inj. PI, acute hepatitis f21, hepatitis, hepatocellular liver inj. Hepatitis (31, acute hepatitis (31, cholestatic hepatitis, cirrhosis Increased liver enz. f61, toxic hepatitis, hepatacelluler
Carbamazepine
8 (2.61
Aceteminophen
7 (2.21
Nimesulide
6 (I.91
1
5 2
liver inj.
days years
1 day years 2 months 16 months 3
IO-86
5
3
Acute hepatitis I31, hepatitis (21, toxic hepatitis, cholestetic liver inj., increased liver enz.
11 days 3 months
7
IO-86
1
6
I-IO days
6
39-65
4
2
Increased liver enz. (31, acute hepatitis (21, hepatocellular liver inj., liver inj. Increased liver enz., toxic hepatitis, hepatitis, cholestatic hepatitis, cholestatic jaundice, hepatocellular
16
2 days 2 months
liver inj.
lndinavir
6 (I.91
Interferon beta
6 [I .91
Rifampicin
5 (I.61
Diclofenec
5 (I.61
0
6
14
1
4
30-57
0
6
Hyperbilirubinaemia (41, increased liver enz., liver inj.
18-38
5
1
Increased liver enz. (61
25-85
3
2
Jaundice (21, hepatitis (21, increased liver enz.
l-9 days
25-85
2
3
Increased liver enz. (21, cholestatic hepatitis (21, hepatitis
few hours 2 months
italics indicate liver disorders codified on the basis of liver test abnormalities, *Number of reports is indicated in brackets ifmore than cne; "years.
tions; seven patients (treated with amineptine, ticlopidine, amiodarone, carbamazepine, acetaminophen and diclofenac) also experienced serious concomitant manifestations affecting the pancreas, skin and blood systern. In general, the reports were accurately completed and the temporal relationships with drug intake and possible alternative causes were indicated (83.5% of
as defined by the criteria
adopted
by the International
NR 1 month 10 months
Consensus
Meeting.
the reports have a documentation grading of 2 or 3). Most of the reports indicate a “probable” drug relationship, but a rechallenge in three patients treated with amineptine led to the reappearence of acute hepatitis. Of the 8 reports of liver injuries attributed to amiodarone, 8 refer to acute liver damage following acute drug administration (oral in two cases, iv in two,
719
Spontaneous reporting
of adverse hepatic drug reactions
and iv followed by oral in one); the other three refer to patients developing acute damage (rapidly evolving to cirrhosis in one case) after prolonged drug administration up to a maximum of 3.3 years. The acetaminophen-associated liver injuries occurred in patients taking routine therapeutic doses. Again, in relation to these 13 drugs, Table III shows the number of articles referring to hepatotoxicity extracted from Medline and Hepatox databases, and the number of spontaneous reports received by the WHO Collaborating Centre between 1996 and 1998: the hepatotoxicity of most of the drugs was well documented but, in a few cases, the published references were poor. As far as the WHO database is concerned, it needs to be stressed that the total number of reports related to each drug depends not only on toxicity, but also on the lev-
el of consumption and the number of countries in which the particular drug is marketed. Although this means that it is impossible to make any direct comparisons between the different drugs, it does allow a comparison to be made of the percentages of drug-induced liver and other organ toxicities: e.g. the WHO database clearly shows that liver toxicity is more frequently reported in relation to amineptine than in relation to diclofenac (39% vs 4% of the reports concerning liver disorders). The main characteristics of the reports concerning previously undescribed or poorly described associations between six drugs and serious liver disorders are summarised in Table IV. With the exception of glyceryl trinitrate, there are few reports in the WHO database and these “new signals” deserve monitoring over time.
Table III. Drugs associated with more than four reports of liver reactions in this study compared with published data and the two-year analysis of the WHO hepatotoxicity database. . Mtelmablt Medlllla owl lm&l 11#111 ?Eli Amineptine Flutamide Fenofibrate Eclopidine Amiodarone Simvastatin Carbamazepina Acetaminophen Nimesulide indinavir Interferon beta Rifampicin Diclofenac
26 41 29 43 101 81 90 99 3 4 30 56 85
>20 z-20 15 >20 z-20 15 >20 >20 4 8’ 9 >20 >20
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
214 809 567 4339 3091 4930 5920 3242 294 5800 3078 1295 5864
78 222 122 432 287 377 510 421 37 435 93 275 181
(391 1271 [I51 [Ill (91 (81 (91 [I31 (131 C81 (3) [211 141
*aI/ articles published in 1999
T&la IV, New signals. Individual reports concerning serious unlabelled drug-induced liver disorders not mentioned in Micromedex and for which no or only one reference was found in Madline or Hepatox. The last column indicates the number of reports of any liver disorder contained in the WHO database.
Female Female Male Male Female Male Male Female
720
72 57 56 65 51 50 70 36
Hepatitis, jaundice Hepatic cytolysis Cholestatic hepatitis Cholestatic hepatitis Hepatotoxicity Hepatitis + urticaria Acute hepatitis Cholestatic hepatitis
Aufloxacin Rufloxacin Rufioxacin Cilazapril Glyceryl trinitrate lopamidol Cadralazine Azanidazole
200 mg 200 mg 200 mg 5w 15mgllS
5 days 8 days 4 days 2 months IOmonths
10mg 200 mg
9 days 1 week/4 months
No No No Positive No No No No
Sotalol No No Carbamazepine No Furosemide + nifadipine Furosemide + digoxin Mathylpradnisolone
4 4 4 11 89 10 0 1
A. Conforti et al.
Discussion It is well known that drugs are a major cause of acute liver failure I4 and may also be involved in chronic injuries I4. However, there are few data concerning the frequency of hepatotoxic ADRs, which seem to be rare in general practice but may be responsible for a larger proportion of cases of hepatic disease than previously thought Is. Drug-induced liver diseases have been associated with more than 800 drugs and have been the cause of one in 600-3500 hospital admissions I6 l’. Despite their well-known limitations, spontaneous reporting systems can provide useful information concerning the extent and severity of hepatic disorders 18. A New Zealand review of 22,455 spontaneous ADR reports made over a period of 20 years has shown that 943 concerned liver injuries associated with 205 drugs and indicated a mortality rate of 3.3% ‘j. A Danish review of 1100 reports of hepatic drug reactions, recorded between 1978 and 1987, found that hepatic injuries accounted for 5.9% of all ADRs: 47.2% of these were classified as being acute cytotoxic reactions, 16.2% as acute cholestatic reactions, and 26.9% as reflecting abnormal hepatic function; the hepatic injury was lethal in 52 cases “. The annual reporting rate in 1998 in the Veneto and Lombardy regions of Italy was similar to that of countries with well-developed national systems (202/million inhabitants vs 294 in the United Kingdom and 287 in France) lo, but the total percentage of reported hepatic drug reactions was less than that notified through the similar spontaneous reporting systems existing in New Zealand and Denmark 6-8. It is likely that this difference is due not only to the tendency of Italian physicians to prefer the spontaneous reporting of more typical and well-known ADRs (such as cutaneous reactions), but also to the difftculty of attributing liver disorders to a particular drug, as can be seen from the recent report of a mistaken diagnosis of non-A non-C acute hepatitis instead of drug-induced liver damage 19. In Lombardy, a larger percentage of hepatic ADRs was signalled compared to Veneto; one reason for this difference could be the existence, in that region, of centres specialized in liver diseases, where doctors are more accustomed to in recognizing druginduced liver disorders. A detailed analysis of the causes of regional differences in the reporting of hepatic ADRs was not within the scope of this investigation, attention being focused, instead, on the reports of hepatic ADRs independently of the Region from which they come. The reports shows a generally favourable outcome despite the fact that more than half of the reactions were considered serious, thus reflecting the cautious attitude of Italian doctors towards drug-induced liver damage.
In this study, the rate of the reporting of liver drug reactions in our surveillance system increased with age, as in the case of reactions involving other systems *O. This may simply reflect an age-related increase in drug consumption, but it has been suggested that age per se may be an independent risk factor for the development of ADRs *‘. It is comforting to note that 81% of the reports were given a documentation grading of 2 or 3, since this is a crucial aspect of drug causality assessments and it is often difficult to ensure accurate reporting **. The largest number of hepatic reaction reports involved amineptine, which is known to cause hepatocellular, cholestatic or mixed damage, the latency of which varies from one day to ten months and usually has a favourable outcome (no deaths have been reported); a rechallenge may be positive or negative, and associated manifestations have frequently been described 23 24. This drug was first marketed in Italy in 1993 and then withdrawn in February 1999 because of the risk of drug abuse; it has been reported that the consumption of amineptine in Italy between 1996 and 1998 accounted for 11% of the total consumption of antidepressants 25. Published data also show that flutamide-induced damage may be hepatocellular, cholestatic or mixed, with a described latency of between five days and ten months; a rechallenge leads to positive results. The outcome is generally favourable but some fatal cases have been reported 26*’ and our own database includes the deaths of two prostate carcinoma patients, aged 75 and 82 years. We have also received a report of acute hepatitis in a 22-year-old female taking flutamide for acne (an indication that is not registered in Italy), an event that raises the question of the safety of drugs used for unregistered indications. The possibility of acute liver damage (including some fatal cases) after the acute iv or po administration of amiodarone is well documented in the literature ** 29. Long-term treatment frequently causes minor asymptomatic liver damage (e.g. an increase in ALT levels to less than four times the upper normal limit) that may resolve during treatment, and symptomatic damage which is less frequent and rarely fatal. An asymptomatic increase in transaminases has frequently been described during fenofibrate, simvastatin and ticlopidine treatment, but clinically significant cytolytic or cholestatic damage is rare; chronic damage has been reported in the case of fenofibrate, but not in the case of simvastatin and ticlopidine. An increase in hepatic enzymes has been described during nimesulide treatment, but published data are very few since the drug is not marketed worldwide; six cases of acute cytolytic or cholestatic damage have recently been reported, none of which were fatal 30. The acute hepatotoxicity of high-dose acetaminophen
791
Spontaneous reporting
of adverse hepetic drug reactions
has been clearly documented in adults and children, but is much rarer after the administration of therapeutic doses. The usually non-fatal liver toxicity of carbamazepine, rifampicin and diclofenac has been widely documented, and it is well known that indinavir may lead to an increase in bilirubin levels (rarely associated with increased transaminases or alkaline phosphatases), which usually return to normal without the need for treatment withdrawal. An increase in transaminases alone is less frequent and usually resolves after drug withdrawal, but some cases of severe fatal hepatitis have also been described. The significant deterioration of liver function observed during interferon therapy may be related to various mechanisms: autoimmune hepatitis has been described in up to 5% of patients, and the transient increase in hepatic cytolytic enzymes that occurs in 20% of patients during the first few months of interferon beta treatment may reflect immune activation and autoimmune hepatic damage at least in some cases. Acute interferon-induced liver damage has also been associated with virus C clearance, the triggering of acute alcoholic hepatitis or direct liver tissue toxicity 3’. The principal aim of any spontaneous reporting system is the early recognition of adverse drug reactions not recognised during clinical trials because of the limited number and selection of the patients involved, and our analysis allowed us to identify some drug-related liver damage not hitherto described. Rufloxacin is a fluoroquinolone first marketed in Italy in 1992: we could not find any reports of drug-related liver disorders in the literature and the WHO database contains only four cases. There are some published reports of severe hepatitis associated with some fluoroquinolones (ciprofloxacin, enoxacin, ofloxacin), whereas others only seem to cause a transient increase in liver enzymes 13. Cilazapril was first marketed in Italy in 1992, and there are no published case reports of hepatotoxicity. The eleven reports included in the WHO database come from Austria, GFR, Switzerland, Sweden, the UK and Italy, and refer to biliary disorders (4), increased hepatic enzyme levels (3), hepatitis, cholestatic hepatitis, jaundice and porphyria. Our own findings suggest that a concomitant role of carbamazepine in the onset of cholestatic hepatitis cannot be excluded. Some individual cases of cholestatic jaundice, liver dysfunction and hepatitis have been described in patients receiving some angiotensin converting enzyme (ACE) inhibitors (captopril, lisinipril, fosinopril and ramipril), in these cases other ACE inhibitors should be avoided. No hepatic disorders were reported during treatment with glyceryl trinitrate, but the WHO database contains 89 reports of liver disorders, the most frequent being
abnormal hepatic function and hepatitis. In our patient, the drug was administered transdermically. Cadralazine is a hydralazine-related direct arterial vasodilator that has been marketed in Italy since 1988. Liver injuries associated with hydralazine and dihydralazine, and acute liver injury attributed to hydralazine, have been reported in more than 20 patients with different bioptic patterns: hepatocellular necrosis, intrahepatic cholestasis, granulomatous hepatitis r3. In the case reported in our database, the temporal relationship between the start of treatment and the temporal relationship between the start of treatment and the liver injury, the absence of alternative aetiological causes, and the improvement of biochemical tests after drug withdrawal strongly suggest the causal role of cadralazine. Azanidazole does not appear to induce hepatotoxicity: we found only one case report of hepatocellular damage in the WHO database, which came from Italy and was dated 1983. In this regard, it is worth pointing out that Italy stopped sending spontaneously reported data to the WHO Centre in 1994. In conclusion, spontaneous reports of ADR are a useful tool in the pharmacovigilance of an Italian inter-regional health service. They detect signals suggesting previously unreported drug hepatotoxicity, and improve the drug safety profiles in the reference general population.
References ’ Garcia Rodriguez LA, Ruigomez A, Jick H. A review of epidemiologic research on drug-induced acute liver injury using the general practice research data in the United Kingdom. Pharmacotherapy 1997;17:721-8. ’ Bakke OM, Manocchia M, de Abajo F, Kaitin KI, Lasagna L. Drug safety discontinuations in the United Kingdom, the United States and Spain from 1974 through 1993: a regulatory prospective. Clin Pharmacol Ther 1995;58:105-17. 3 Lee WM. The case of case reports. Ital J Gastroenterol Hepatol 1998;30:3 18-9. ’ Pontiroli L, Sartori M, Pittau S, Morelli S, Boldorini R, Albano E. Flutamide-induced acute hepatitis: investigation on the role of immunoallergic mechanisms. Ital J Gastroenterol Hepatol 1998;30:310-4. 5 Ceriani R, Borroni G, Bissoli F. Ritodrine-related liver injury. Case reports and reviews of the literature. Ital J Gastroenterol Hepat01 1998;30:315-7. 6 Pillans PI. Drug associated hepatic reactions in New Zealand. 21 years experience. NZ Med J 1996; 109:3 15-9. ’ Dossing M, Andreasen PB. Drug-induced liver disease in Denmark. Stand J Gastroenterol 1982;17:205-1 1. 8 Friis H, Andreasen PB. Drug-induced hepatic iniury: an analysis of 1100 cases reported to the-Danish Committee on Adverse Drug Reactions between 1978 and 1987. J Intern Med 1992;232:133-8. 9 Biour M, Poupon R, Grange JD, Chazouilleres 0, Jaillon P. Hepatotoxicitt des medicaments. 11’ Mise a iour du fichier biblibgraphique des atteintes hepatiques et des medicaments responsables. Gastroenterol Clin Biol 1998;22: 1004-44.
A. Conforti et al.
‘” Olsson S, editor. National Pharmacovigilance Systems. Country profiles and overview. Uppsala: WHO Monitoring Centre; 1999. I’ Olsson S. Role of WHO programme on international drug monitoring in co-ordinating world-wide drug safety efforts. Drug Saf 1998;19:1-10. I? Benichou C. Criteria of drug-induced liver disorders. Report of an International Consensus Meeting. J Hepatol 1990;11:272-6. I’ Drugdex, 1974-1999 Micromedex, Inc. Vol. 102. I4 Fontana RJ. Acute liver failure. Curr Opin Gastroenterol 1998;14;265-73. ‘z Zimmerman HJ. Drug-induced liver injury. In: Postgraduate course 1998. Clinical and Pathological Correlation in Liver Disease: Approaching the next Millennium. American Association for the Study of Liver Diseases 1998:252-68. Ih Jick H, Walker AM, Porter J. Drug-induced liver disease. J Clin Pharmacol 1981;21:359-64 ” Hallas H, Gram LF, Grodum E, Damsbo N, Brosen K, et al. Drug-related admission to medical wards. A population-based survey. Br J Clin Pharmacol 1992:33:61-S. Ix Bruppacher R. Epidemiological identification and evaluation of hepatic adverse drug reactions. Semin Liver Dis 1995;15:301-8. I’) Traversa G, Ciccozzi M, Mele A. Non-A non-C hepatitis and drug use. Pharmacoepidemiology and Drug Safety 1999;8(Suppl 2):6. *” Naldi L, Conforti A, Venegoni M, Troncon MG, Caputi A, Ghiotto E, et al. Cutaneous reactions to drugs. An analysis of spontaneous reports in four Italian regions. Br J Clin Pharmacol 1999;48:839-46. ?’ Carbonin P, Pahor M, Bemabei R, Sgadari A. Is age an indepen-
dent risk factor of adverse drug reactions in hospitalized medical patients? J Am Geriatr Sot 1991;39:1093-9. *2 Aithal GP, Rawlins MD, Day CP. Accuracy of hepatic adverse drug reactions reporting in one English health region. Br Med J 1999;319:1541. x Lazaros GA, Stavrinos C, Papatheodoridis GV, Delladetsina JK, Toliopoulos A, Tassopoulos NC. Amineptine-induced liver injury. Report of two cases and brief review of the literature. Hepatogastroenterology 1996;43:1015-9. 24 Ricca Rosellini S, Grilli F, Gaudio M, Saragoni A, Miglio F. Hepatic injury associated with amineptine therapy. Ital J Gastroenterol 1990;22:40-3. 25 Brazzale G, Font M, Lopatriello S. Meglio tardi the mai. Dialogo sui Farmaci 1999;2:76-7. 26Dourakis SP, Alexopoulou AA, Hadzivannis SJ. Fulminant hepatitis after flutamide treatment. J Hepatol 1994;20:350-3. ” Wysowski DK, Fourcroy JL. Flutamide hepatotoxicity. J Urol 1996;155:209-12. ‘* Morelli S, Guido V, De Marzio P, Aguglia F, Balsano F. Early hepatitis during intravenous amiodarone administration. Cardiology 1991;78:291-4. 29 Richer M, Robert S. Fatal hepatotoxicity following oral administration of amiodarone. Ann Pharmacother 1995;29:582-6. 3o Van Steenbergen W, Peeters P, De Bondt J, Staessen D, Buscher H, Laporta T, et al. Nimesulide-induced acute hepatitis: evidence from six cases. J Hepatol 1998;29:135-41. 3’ Durelli L, Ferrer0 B, Oggero A, Verdun E, Bongioanni MR, Gentile E, et al. Autoimmune events during interferon beta-lb treatment for multiple sclerosis. J Neurol Sci 1999; 162:74-83.
EASL SINGLE TOPIC CONFERENCE LIVER FIBROSIS: FROM BASIC SCIENCE TO CLINICAL TARGETS Florence, Italy, October 12-13,200l The aim of the conference is to provide the latest information on this key area of hepatology and to translate the current knowledge into clinical terms. The meeting is directed to both the expert in the field and to the general hepatologist. Preliminary programme and call for abstract: January 2001 Deadline for abstract: April 2001 Organizers: Massimo Pinzani, University of Florence (Italy) and Detlef Schuppan, University of Erlangen -Nuernberg (Germany).
723
lJlGESTLIYER OIS 2000;32:716-23
Spontaneous reporting of drug-related hepatic reactions from two Italian Regions (Lombardy and Veneto) A. R. E. G. U. M. F.
Conforti Leone Ghiotto Velo Moretti Venegonil Bissoli2
&ion: Clinical Pharmacology Unit, “G.8. Rossi” University Hospital Verona; ’ Department of lf7ternal Medicine, Fatebenefratelii Hospita!, Man; p Liver Unit, Fomarvii Hospital, Megenta, Italy. lS‘lMp#&filW: Almw#k Dr. A. Conforti, lstituto di Farmacologia, Universicci di Verona, 37134 Vemna, &e/y. Fax: +39-045-581 I I 1. E-meil: [email protected] 1-b: The Authors are vary grateful Phermeceutice~ Departments Heaith Districts of Lombardy Veneto Regions for co/letting adverse reaction forms.
to the and local and the
Submitted May 18, 2000. Revised September 8, 21000. Accepted September 18, 2000,
716
Background and aims. Many currently used drugs are implicated in liver disorders ranging from asymptomatic abnormal liver function tests to fatal liver failure, and drug-induced hepatic reactions are the most frequent cause of the withdrawal of drugs from the market in the United States, United Kingdom and Spain. The aim of this study was to review the drug-induced hepatic reactions notified to the spontaneous surveillance system of two Italian regions [Lombardy and the Veneto] up to 31 December 1998. Patients and methods. The reports of hepatic adverse drug reactions were extracted from an inter-regional database of spontaneous adverse drug reaction reports and analysed by reporter category patient age and sex, the underlying disease, the adverse reaction and drug exposure. In the case of drugs involved in more than four reports, a search was made of the Medline, Hepatox, Micromedex and WHO databases. Results. Between 1988 and 1998, there were 3 IO reports of hepatic lesions associated with single drugs or fixed combinations, the most frequently involved patient categories being females and the elderly Sixty percent of the reactions were classified as serious, with three fatal cases. Thirteen drugs were associated with more than four reports and accounted for 35% of all of the hepatic reactions. Six drugs not known as hepatotoxic agents were associated with hepatic reactions. Conclusions. In this study spontaneous reports have suggested previously unknown hepatotoxicty of six drugs, and have improved drug safety profiles as shown in an Italian inter-regional health service.
Digest
Liver
Key words:
Ilk
2000;32:716-23
drug-induced
hepatic
reaction;
spontaneous
reporting
Introduction Many currently used drugs are implicated in liver disorders ranging from asymptomatic abnormal liver function tests to fatal liver failure. Although the risk of acute hepatic injuries due to individual drugs other than some well-known exceptions is generally considered to be low ‘, hepatic drug reactions are the most frequent cause of the withdrawal of drugs from the market in the United States, United Kingdom and Spain 2. Despite under-reporting, the impossibility of making reliable estimates of reaction rates, and its variability in terms of accuracy and precise terminology, spontaneous adverse drug reaction (ADR) reporting and publications of case
A. Conforti et al.
reports 3-5acts as an early warning system and provides data for the generation of research hypotheses. The spontaneous reports of drug-associated liver injuries received over a period of 21 years and reviewed by the New Zealand Centre of ADR Monitoring represented 4.2% of all reports and 7.4% of all fatal occurrences 6, and similar results were found in Denmark, where 6% of the ADRs reported during a 20-year period involved hepatotoxicity ’ 8. The aim of this study was to analyse the hepatic drug reactions notified to the spontaneous surveillance system of two Italian Regions (Lombardy and Veneto) up until 31 December 1998. The most frequently reported associations were correlated with published data, a specific hepatic drug toxicity bibliographical database 9, and the WHO database of spontaneous reports coming from 59 countries worldwide lo. Some previously unreported or poorly described hepatotoxic drug effects are also discussed.
Methods In Italy, spontaneous reporting of ADRs has been mandatory since 1987 and, in keeping with the law, doctors and pharmacists have to send all suspected ADRs to the Local Health authorities, which forward the reports to the Ministry of Health. The results of such a centralized organization are inadequate, the main problem being represented by the widespread under-reporting. In 1994 (the last year with available national data), the reporting rate by doctors was about 33/million inhabitants. However, some local initiatives were commenced over the last few years. The data presented in this paper were obtained from a database containing all of the spontaneous ADR reports coming from Lombardy and Veneto, two Italian Regions with a combined population of about 13.7 million and a 1998 reporting rate of 202.7 per million inhabitants. In Veneto, the reports have been examined regularly and entered into a database since 1988 by the Clinical Pharmacology Unit of the University of Verona; in Lombardy, a similar analysis has been carried out by the Regional Pharmacovigilance Group since 1993. All of the reports notifying one or more hepatic reactions and entered in the database as of 31 December 1998 were analysed by a panel of experts consisting of gastroenterologists, hepatologists, specialists in internal medicine, pharmacologists and pharmacists, who took into account the reporter category (hospital doctors, general practitioners and others), patient age and sex, the underlying disease, the characteristics of the adverse reaction (latency, duration, outcome, rechallenge, seriousness) and drug exposure (indication, duration of treatment and dosage).
The drugs were classified according to the Italian Codifa system and their Anatomical Therapeutic Chemical Classification (ATC), and the adverse reactions according to the terminology adopted by the WHO I’, while maintaining as far as possible the terms used by the reporter in defining the liver disorders. The seriousness of the reactions were classified on the basis of the WHO Critical Term List, in which a serious adverse reaction is one that is fatal, life-threatening, involved or prolonged hospitalisation, or led to persistent or significant disability or incapacity. Grouping the reports on the basis of the high-level terms of this list, the following types of serious liver disorders were found: hepatitis (including toxic and acute hepatitis), cholestatic hepatitis (including cholestatic jaundice), hepatocellular damage (including hepatocellular jaundice and hepatotoxicity), hepatic necrosis, hepatorenal syndrome, and hepatitis C. The type of hepatic damage was classified on the basis of the WHO criteria and, when laboratory test results were available, also using the criteria adopted by an international consensus meeting 12. Each report was graded according to the WHO documentation criteria: grade 1 when it included only the date of onset of the reaction and the dates of treatment; grade 2 when it also included the reason for the treatment and the outcome of the reaction; and grade 3 in the case of a report notifying a positive rechallenge and including all of the other data mentioned for grades 1 and 2. All of the drugs reported as causing liver adverse effects were used as keywords for a general search of the literature (Medline 1966- 1999 and Micromedex 19741999 “), with the Medline search being based on a full text strategy crossing each drug with all of the terminology compatible with a liver disorder; the selected articles were checked and accepted after reading the abstract. We also referred to the specialised Hepatox 9 and WHO Monitoring Centre Spontaneous Reporting databases lo. Any drug- associated hepatotoxicity that had not previously been clearly reported in the literature was flagged as a possible “signal”. Results
A total of 13,118 ADR reports were received by the Veneto and Lombardy regional centres during the period under consideration, 388 of which referred to a liver disorder. The annual percentage of hepatic ADR reports in Veneto and Lombardy is shown in Figure 1. The number of both types of report increased over time, however a comparison of data of the two regions in the common observation period shows a higher detection of hepatic ADR in Lombardy, the global percentage of reported hepatic disorders being 2.5 in 717
Spontaneous reporting of adverse hepatic drug reactions
Fip.1. Annual percentage of hepatic reactions in Veneto Im1 and Lomlardy (01 regions.
Veneto (in the period 1988- 1998) and 3.5 in Lombardy (in the period 1993-1998). As far as concerns the origin of the reports, 60% came from hospital doctors, 37% from general practitioners, 1% from others, and 2% from unspecified reporters. The age-related distribution of the reports indicating drug-induced liver disorders (Fig. 2) shows that the largest number referred to patients aged 65-74 years. This pattern is similar to that related to the number of
reports as a whole and, therefore, does not indicate any age-specific risk for liver reactions. The female/male ratio was 1.4, which is slightly lower than the 1.6 observed for reactions affecting other organ systems. The documentation grading of the reports was 7.5% grade 1, 78.4% grade 2, and 2.7% grade 3; 11.3% of the reports were graded 0 because they did not include the dates of onset or treatment. In about 16% of reports, laboratory tests were available. The outcome was complete recovery in 62.6% of the patients, and is unknown in 33.2% because of the incomplete nature of the report (although the need for hospitalisation was specified in 35 cases). Three patients died (two during treatment with flutamide and one during treatment with nevirapine), and two patients suffered permanent disability (one case of chronic HCV-related hepatitis attributed to immunoglobulins, and one of liver cirrhosis attributed to amiodarone) . The suspected drug was a single agent in 283 reports, a fixed combination in 27, and two or more drugs in 78. Our analysis was limited to the 310 reports concerning single drugs or fixed combinations, 187 (60%) of which involved a serious hepatic reaction (a percentage that is nearly double the 27% of serious reactions involving other systems or organs in our database): these are listed in Table I, but it must be emphasised that their seriousness is based on their potential negative consequences and not on their real outcome in the individual patients. The 13 drugs involved in more than four reports, including their documentation grading, patient age and sex, the type of liver damage, and its latency, defined as the time interval between the first drug administration and the first appearance of liver adverse reaction are listed in Table II. These drugs were responsible for 35% of all of the liver reactions reported in the database and caused both cytotoxic and cholestatic reac-
Tat& I. Serious hepatic adverse drug reactions reported in Lombardy and Veneto Regions, grouped according to the high-level terms listed in the WHO Critical Term List [see Methods). lli@li-leuol
-25
5-14
15-44 Age
45-64
65-74
=-74
Total
(ye=4
Fig. 2. Mean annual reporting rate of hepatic ADRs: distribution according to age.
718
term
Hepatitis Cholestetic hepatitis Hepatocellular damage Hepatic necrosis Hepatorenal syndrome Hepatitis C Total
Inaludod
teml
Toxic hepatitis, acute hepatitis Jaundice, chotestatic jaundice Liver disease, hepatotoxicity
No. t%POttS 107
59 16 3
1 1 187
R. Conforti et al.
Table II. Single agents or fixed combinations associated with more than four reports of hepatic reactions.
Amineptine
21 (6.81
1
Flutamide
10 (3.21
1
3
I4
3
A@ hlll#d
Selt F nil
23-78
17 4
Hepatotoxic effect (61, acute hepatitis 161, hepatocellular liver inj. 121, hepatic function abnormal, mixed liver inj., increased liver em., cholestatic hepatitis, hyperbilirubineemia, jaundice, jaundice and hepatotoxicity
7 days 3 months
22-83
1
Cholestatic hepatitis 141,fatal acute hepatitis (21, increased liver em., toxic hepatitis, iaundice end hepatic failure,
43 days 5 months
9
Type 0finjuqP
hepatocelluler
Fenofibrate
10 (3.21
10
42-70
5
5
htancy
liver inj. 2 weeks 3 years
h'epetocellular liver inj. (21, cholestatic liver inj. (2l, liver inj., increased liver
em., hepatic function abnormal, hepatitis and cholangitis, cholestatic hepatitis, hepatitis Tclopidine
Amioderone Simvastatin
9 (2.91
3
8 (2.61
6
8
8 (2.6)
8
42-81
54-82 49-72
3
4 5
6
4 3
Cholestatic hepatitis (31, cholestatic liver inj. PI, acute hepatitis f21, hepatitis, hepatocellular liver inj. Hepatitis (31, acute hepatitis (31, cholestatic hepatitis, cirrhosis Increased liver enz. f61, toxic hepatitis, hepatacelluler
Carbamazepine
8 (2.61
Aceteminophen
7 (2.21
Nimesulide
6 (I.91
1
5 2
liver inj.
days years
1 day years 2 months 16 months 3
IO-86
5
3
Acute hepatitis I31, hepatitis (21, toxic hepatitis, cholestetic liver inj., increased liver enz.
11 days 3 months
7
IO-86
1
6
I-IO days
6
39-65
4
2
Increased liver enz. (31, acute hepatitis (21, hepatocellular liver inj., liver inj. Increased liver enz., toxic hepatitis, hepatitis, cholestatic hepatitis, cholestatic jaundice, hepatocellular
16
2 days 2 months
liver inj.
lndinavir
6 (I.91
Interferon beta
6 [I .91
Rifampicin
5 (I.61
Diclofenec
5 (I.61
0
6
14
1
4
30-57
0
6
Hyperbilirubinaemia (41, increased liver enz., liver inj.
18-38
5
1
Increased liver enz. (61
25-85
3
2
Jaundice (21, hepatitis (21, increased liver enz.
l-9 days
25-85
2
3
Increased liver enz. (21, cholestatic hepatitis (21, hepatitis
few hours 2 months
italics indicate liver disorders codified on the basis of liver test abnormalities, *Number of reports is indicated in brackets ifmore than cne; "years.
tions; seven patients (treated with amineptine, ticlopidine, amiodarone, carbamazepine, acetaminophen and diclofenac) also experienced serious concomitant manifestations affecting the pancreas, skin and blood systern. In general, the reports were accurately completed and the temporal relationships with drug intake and possible alternative causes were indicated (83.5% of
as defined by the criteria
adopted
by the International
NR 1 month 10 months
Consensus
Meeting.
the reports have a documentation grading of 2 or 3). Most of the reports indicate a “probable” drug relationship, but a rechallenge in three patients treated with amineptine led to the reappearence of acute hepatitis. Of the 8 reports of liver injuries attributed to amiodarone, 8 refer to acute liver damage following acute drug administration (oral in two cases, iv in two,
719
Spontaneous reporting
of adverse hepatic drug reactions
and iv followed by oral in one); the other three refer to patients developing acute damage (rapidly evolving to cirrhosis in one case) after prolonged drug administration up to a maximum of 3.3 years. The acetaminophen-associated liver injuries occurred in patients taking routine therapeutic doses. Again, in relation to these 13 drugs, Table III shows the number of articles referring to hepatotoxicity extracted from Medline and Hepatox databases, and the number of spontaneous reports received by the WHO Collaborating Centre between 1996 and 1998: the hepatotoxicity of most of the drugs was well documented but, in a few cases, the published references were poor. As far as the WHO database is concerned, it needs to be stressed that the total number of reports related to each drug depends not only on toxicity, but also on the lev-
el of consumption and the number of countries in which the particular drug is marketed. Although this means that it is impossible to make any direct comparisons between the different drugs, it does allow a comparison to be made of the percentages of drug-induced liver and other organ toxicities: e.g. the WHO database clearly shows that liver toxicity is more frequently reported in relation to amineptine than in relation to diclofenac (39% vs 4% of the reports concerning liver disorders). The main characteristics of the reports concerning previously undescribed or poorly described associations between six drugs and serious liver disorders are summarised in Table IV. With the exception of glyceryl trinitrate, there are few reports in the WHO database and these “new signals” deserve monitoring over time.
Table III. Drugs associated with more than four reports of liver reactions in this study compared with published data and the two-year analysis of the WHO hepatotoxicity database. . Mtelmablt Medlllla owl lm&l 11#111 ?Eli Amineptine Flutamide Fenofibrate Eclopidine Amiodarone Simvastatin Carbamazepina Acetaminophen Nimesulide indinavir Interferon beta Rifampicin Diclofenac
26 41 29 43 101 81 90 99 3 4 30 56 85
>20 z-20 15 >20 z-20 15 >20 >20 4 8’ 9 >20 >20
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
214 809 567 4339 3091 4930 5920 3242 294 5800 3078 1295 5864
78 222 122 432 287 377 510 421 37 435 93 275 181
(391 1271 [I51 [Ill (91 (81 (91 [I31 (131 C81 (3) [211 141
*aI/ articles published in 1999
T&la IV, New signals. Individual reports concerning serious unlabelled drug-induced liver disorders not mentioned in Micromedex and for which no or only one reference was found in Madline or Hepatox. The last column indicates the number of reports of any liver disorder contained in the WHO database.
Female Female Male Male Female Male Male Female
720
72 57 56 65 51 50 70 36
Hepatitis, jaundice Hepatic cytolysis Cholestatic hepatitis Cholestatic hepatitis Hepatotoxicity Hepatitis + urticaria Acute hepatitis Cholestatic hepatitis
Aufloxacin Rufloxacin Rufioxacin Cilazapril Glyceryl trinitrate lopamidol Cadralazine Azanidazole
200 mg 200 mg 200 mg 5w 15mgllS
5 days 8 days 4 days 2 months IOmonths
10mg 200 mg
9 days 1 week/4 months
No No No Positive No No No No
Sotalol No No Carbamazepine No Furosemide + nifadipine Furosemide + digoxin Mathylpradnisolone
4 4 4 11 89 10 0 1
A. Conforti et al.
Discussion It is well known that drugs are a major cause of acute liver failure I4 and may also be involved in chronic injuries I4. However, there are few data concerning the frequency of hepatotoxic ADRs, which seem to be rare in general practice but may be responsible for a larger proportion of cases of hepatic disease than previously thought Is. Drug-induced liver diseases have been associated with more than 800 drugs and have been the cause of one in 600-3500 hospital admissions I6 l’. Despite their well-known limitations, spontaneous reporting systems can provide useful information concerning the extent and severity of hepatic disorders 18. A New Zealand review of 22,455 spontaneous ADR reports made over a period of 20 years has shown that 943 concerned liver injuries associated with 205 drugs and indicated a mortality rate of 3.3% ‘j. A Danish review of 1100 reports of hepatic drug reactions, recorded between 1978 and 1987, found that hepatic injuries accounted for 5.9% of all ADRs: 47.2% of these were classified as being acute cytotoxic reactions, 16.2% as acute cholestatic reactions, and 26.9% as reflecting abnormal hepatic function; the hepatic injury was lethal in 52 cases “. The annual reporting rate in 1998 in the Veneto and Lombardy regions of Italy was similar to that of countries with well-developed national systems (202/million inhabitants vs 294 in the United Kingdom and 287 in France) lo, but the total percentage of reported hepatic drug reactions was less than that notified through the similar spontaneous reporting systems existing in New Zealand and Denmark 6-8. It is likely that this difference is due not only to the tendency of Italian physicians to prefer the spontaneous reporting of more typical and well-known ADRs (such as cutaneous reactions), but also to the difftculty of attributing liver disorders to a particular drug, as can be seen from the recent report of a mistaken diagnosis of non-A non-C acute hepatitis instead of drug-induced liver damage 19. In Lombardy, a larger percentage of hepatic ADRs was signalled compared to Veneto; one reason for this difference could be the existence, in that region, of centres specialized in liver diseases, where doctors are more accustomed to in recognizing druginduced liver disorders. A detailed analysis of the causes of regional differences in the reporting of hepatic ADRs was not within the scope of this investigation, attention being focused, instead, on the reports of hepatic ADRs independently of the Region from which they come. The reports shows a generally favourable outcome despite the fact that more than half of the reactions were considered serious, thus reflecting the cautious attitude of Italian doctors towards drug-induced liver damage.
In this study, the rate of the reporting of liver drug reactions in our surveillance system increased with age, as in the case of reactions involving other systems *O. This may simply reflect an age-related increase in drug consumption, but it has been suggested that age per se may be an independent risk factor for the development of ADRs *‘. It is comforting to note that 81% of the reports were given a documentation grading of 2 or 3, since this is a crucial aspect of drug causality assessments and it is often difficult to ensure accurate reporting **. The largest number of hepatic reaction reports involved amineptine, which is known to cause hepatocellular, cholestatic or mixed damage, the latency of which varies from one day to ten months and usually has a favourable outcome (no deaths have been reported); a rechallenge may be positive or negative, and associated manifestations have frequently been described 23 24. This drug was first marketed in Italy in 1993 and then withdrawn in February 1999 because of the risk of drug abuse; it has been reported that the consumption of amineptine in Italy between 1996 and 1998 accounted for 11% of the total consumption of antidepressants 25. Published data also show that flutamide-induced damage may be hepatocellular, cholestatic or mixed, with a described latency of between five days and ten months; a rechallenge leads to positive results. The outcome is generally favourable but some fatal cases have been reported 26*’ and our own database includes the deaths of two prostate carcinoma patients, aged 75 and 82 years. We have also received a report of acute hepatitis in a 22-year-old female taking flutamide for acne (an indication that is not registered in Italy), an event that raises the question of the safety of drugs used for unregistered indications. The possibility of acute liver damage (including some fatal cases) after the acute iv or po administration of amiodarone is well documented in the literature ** 29. Long-term treatment frequently causes minor asymptomatic liver damage (e.g. an increase in ALT levels to less than four times the upper normal limit) that may resolve during treatment, and symptomatic damage which is less frequent and rarely fatal. An asymptomatic increase in transaminases has frequently been described during fenofibrate, simvastatin and ticlopidine treatment, but clinically significant cytolytic or cholestatic damage is rare; chronic damage has been reported in the case of fenofibrate, but not in the case of simvastatin and ticlopidine. An increase in hepatic enzymes has been described during nimesulide treatment, but published data are very few since the drug is not marketed worldwide; six cases of acute cytolytic or cholestatic damage have recently been reported, none of which were fatal 30. The acute hepatotoxicity of high-dose acetaminophen
791
Spontaneous reporting
of adverse hepetic drug reactions
has been clearly documented in adults and children, but is much rarer after the administration of therapeutic doses. The usually non-fatal liver toxicity of carbamazepine, rifampicin and diclofenac has been widely documented, and it is well known that indinavir may lead to an increase in bilirubin levels (rarely associated with increased transaminases or alkaline phosphatases), which usually return to normal without the need for treatment withdrawal. An increase in transaminases alone is less frequent and usually resolves after drug withdrawal, but some cases of severe fatal hepatitis have also been described. The significant deterioration of liver function observed during interferon therapy may be related to various mechanisms: autoimmune hepatitis has been described in up to 5% of patients, and the transient increase in hepatic cytolytic enzymes that occurs in 20% of patients during the first few months of interferon beta treatment may reflect immune activation and autoimmune hepatic damage at least in some cases. Acute interferon-induced liver damage has also been associated with virus C clearance, the triggering of acute alcoholic hepatitis or direct liver tissue toxicity 3’. The principal aim of any spontaneous reporting system is the early recognition of adverse drug reactions not recognised during clinical trials because of the limited number and selection of the patients involved, and our analysis allowed us to identify some drug-related liver damage not hitherto described. Rufloxacin is a fluoroquinolone first marketed in Italy in 1992: we could not find any reports of drug-related liver disorders in the literature and the WHO database contains only four cases. There are some published reports of severe hepatitis associated with some fluoroquinolones (ciprofloxacin, enoxacin, ofloxacin), whereas others only seem to cause a transient increase in liver enzymes 13. Cilazapril was first marketed in Italy in 1992, and there are no published case reports of hepatotoxicity. The eleven reports included in the WHO database come from Austria, GFR, Switzerland, Sweden, the UK and Italy, and refer to biliary disorders (4), increased hepatic enzyme levels (3), hepatitis, cholestatic hepatitis, jaundice and porphyria. Our own findings suggest that a concomitant role of carbamazepine in the onset of cholestatic hepatitis cannot be excluded. Some individual cases of cholestatic jaundice, liver dysfunction and hepatitis have been described in patients receiving some angiotensin converting enzyme (ACE) inhibitors (captopril, lisinipril, fosinopril and ramipril), in these cases other ACE inhibitors should be avoided. No hepatic disorders were reported during treatment with glyceryl trinitrate, but the WHO database contains 89 reports of liver disorders, the most frequent being
abnormal hepatic function and hepatitis. In our patient, the drug was administered transdermically. Cadralazine is a hydralazine-related direct arterial vasodilator that has been marketed in Italy since 1988. Liver injuries associated with hydralazine and dihydralazine, and acute liver injury attributed to hydralazine, have been reported in more than 20 patients with different bioptic patterns: hepatocellular necrosis, intrahepatic cholestasis, granulomatous hepatitis r3. In the case reported in our database, the temporal relationship between the start of treatment and the temporal relationship between the start of treatment and the liver injury, the absence of alternative aetiological causes, and the improvement of biochemical tests after drug withdrawal strongly suggest the causal role of cadralazine. Azanidazole does not appear to induce hepatotoxicity: we found only one case report of hepatocellular damage in the WHO database, which came from Italy and was dated 1983. In this regard, it is worth pointing out that Italy stopped sending spontaneously reported data to the WHO Centre in 1994. In conclusion, spontaneous reports of ADR are a useful tool in the pharmacovigilance of an Italian inter-regional health service. They detect signals suggesting previously unreported drug hepatotoxicity, and improve the drug safety profiles in the reference general population.
References ’ Garcia Rodriguez LA, Ruigomez A, Jick H. A review of epidemiologic research on drug-induced acute liver injury using the general practice research data in the United Kingdom. Pharmacotherapy 1997;17:721-8. ’ Bakke OM, Manocchia M, de Abajo F, Kaitin KI, Lasagna L. Drug safety discontinuations in the United Kingdom, the United States and Spain from 1974 through 1993: a regulatory prospective. Clin Pharmacol Ther 1995;58:105-17. 3 Lee WM. The case of case reports. Ital J Gastroenterol Hepatol 1998;30:3 18-9. ’ Pontiroli L, Sartori M, Pittau S, Morelli S, Boldorini R, Albano E. Flutamide-induced acute hepatitis: investigation on the role of immunoallergic mechanisms. Ital J Gastroenterol Hepatol 1998;30:310-4. 5 Ceriani R, Borroni G, Bissoli F. Ritodrine-related liver injury. Case reports and reviews of the literature. Ital J Gastroenterol Hepat01 1998;30:315-7. 6 Pillans PI. Drug associated hepatic reactions in New Zealand. 21 years experience. NZ Med J 1996; 109:3 15-9. ’ Dossing M, Andreasen PB. Drug-induced liver disease in Denmark. Stand J Gastroenterol 1982;17:205-1 1. 8 Friis H, Andreasen PB. Drug-induced hepatic iniury: an analysis of 1100 cases reported to the-Danish Committee on Adverse Drug Reactions between 1978 and 1987. J Intern Med 1992;232:133-8. 9 Biour M, Poupon R, Grange JD, Chazouilleres 0, Jaillon P. Hepatotoxicitt des medicaments. 11’ Mise a iour du fichier biblibgraphique des atteintes hepatiques et des medicaments responsables. Gastroenterol Clin Biol 1998;22: 1004-44.
A. Conforti et al.
‘” Olsson S, editor. National Pharmacovigilance Systems. Country profiles and overview. Uppsala: WHO Monitoring Centre; 1999. I’ Olsson S. Role of WHO programme on international drug monitoring in co-ordinating world-wide drug safety efforts. Drug Saf 1998;19:1-10. I? Benichou C. Criteria of drug-induced liver disorders. Report of an International Consensus Meeting. J Hepatol 1990;11:272-6. I’ Drugdex, 1974-1999 Micromedex, Inc. Vol. 102. I4 Fontana RJ. Acute liver failure. Curr Opin Gastroenterol 1998;14;265-73. ‘z Zimmerman HJ. Drug-induced liver injury. In: Postgraduate course 1998. Clinical and Pathological Correlation in Liver Disease: Approaching the next Millennium. American Association for the Study of Liver Diseases 1998:252-68. Ih Jick H, Walker AM, Porter J. Drug-induced liver disease. J Clin Pharmacol 1981;21:359-64 ” Hallas H, Gram LF, Grodum E, Damsbo N, Brosen K, et al. Drug-related admission to medical wards. A population-based survey. Br J Clin Pharmacol 1992:33:61-S. Ix Bruppacher R. Epidemiological identification and evaluation of hepatic adverse drug reactions. Semin Liver Dis 1995;15:301-8. I’) Traversa G, Ciccozzi M, Mele A. Non-A non-C hepatitis and drug use. Pharmacoepidemiology and Drug Safety 1999;8(Suppl 2):6. *” Naldi L, Conforti A, Venegoni M, Troncon MG, Caputi A, Ghiotto E, et al. Cutaneous reactions to drugs. An analysis of spontaneous reports in four Italian regions. Br J Clin Pharmacol 1999;48:839-46. ?’ Carbonin P, Pahor M, Bemabei R, Sgadari A. Is age an indepen-
dent risk factor of adverse drug reactions in hospitalized medical patients? J Am Geriatr Sot 1991;39:1093-9. *2 Aithal GP, Rawlins MD, Day CP. Accuracy of hepatic adverse drug reactions reporting in one English health region. Br Med J 1999;319:1541. x Lazaros GA, Stavrinos C, Papatheodoridis GV, Delladetsina JK, Toliopoulos A, Tassopoulos NC. Amineptine-induced liver injury. Report of two cases and brief review of the literature. Hepatogastroenterology 1996;43:1015-9. 24 Ricca Rosellini S, Grilli F, Gaudio M, Saragoni A, Miglio F. Hepatic injury associated with amineptine therapy. Ital J Gastroenterol 1990;22:40-3. 25 Brazzale G, Font M, Lopatriello S. Meglio tardi the mai. Dialogo sui Farmaci 1999;2:76-7. 26Dourakis SP, Alexopoulou AA, Hadzivannis SJ. Fulminant hepatitis after flutamide treatment. J Hepatol 1994;20:350-3. ” Wysowski DK, Fourcroy JL. Flutamide hepatotoxicity. J Urol 1996;155:209-12. ‘* Morelli S, Guido V, De Marzio P, Aguglia F, Balsano F. Early hepatitis during intravenous amiodarone administration. Cardiology 1991;78:291-4. 29 Richer M, Robert S. Fatal hepatotoxicity following oral administration of amiodarone. Ann Pharmacother 1995;29:582-6. 3o Van Steenbergen W, Peeters P, De Bondt J, Staessen D, Buscher H, Laporta T, et al. Nimesulide-induced acute hepatitis: evidence from six cases. J Hepatol 1998;29:135-41. 3’ Durelli L, Ferrer0 B, Oggero A, Verdun E, Bongioanni MR, Gentile E, et al. Autoimmune events during interferon beta-lb treatment for multiple sclerosis. J Neurol Sci 1999; 162:74-83.
EASL SINGLE TOPIC CONFERENCE LIVER FIBROSIS: FROM BASIC SCIENCE TO CLINICAL TARGETS Florence, Italy, October 12-13,200l The aim of the conference is to provide the latest information on this key area of hepatology and to translate the current knowledge into clinical terms. The meeting is directed to both the expert in the field and to the general hepatologist. Preliminary programme and call for abstract: January 2001 Deadline for abstract: April 2001 Organizers: Massimo Pinzani, University of Florence (Italy) and Detlef Schuppan, University of Erlangen -Nuernberg (Germany).
723