LIVER INJURY/REGENERATION

Categorization of Drugs Implicated in Causing Liver

Injury: Critical Assessment Based on Published
Case Reports
Einar S. Bjornsson1,2 and Jay H. Hoofnagle3
An important element in assessing causality in drug-induced liver injury is whether
the implicated agent is known to cause hepatotoxicity. We classified drugs into categories based on the number of published reports of convincingly documented, clinically
apparent, idiosyncratic liver injury. Drugs described in the website LiverTox (http://
livertox.nih.gov) were classified into five categories based on the number of published
cases (category A, 50; category B, 12-49; category C, 4-11; category D, 1-3; category
E, none). Case reports in categories C and D were individually reanalyzed using the
Roussel Uclaf Causality Assessment Method. Drugs with fatal cases or with rechallenge were noted. Among 671 individual drugs or closely related agents, 353 (53%)
were considered convincingly linked to liver injury in published case reports; 48
(13%) were assigned to category A, 76 (22%) were assigned to category B, 96 (27%)
were assigned to category C, and 126 (36%) were assigned to category D. Another 7
(2%) were direct hepatotoxins but only in high doses and placed in a separate category (T). The remaining 318 (47%) drugs had no convincing case report of hepatoxicity in the literature (category E). All except one in category A have been available
since 1999, 98% had at least one fatal case and 89% a positive rechallenge. In category B, 54% had a fatal case and 41% a rechallenge. Drugs in categories C and D
less frequently had instances of fatal (23% and 7%) or rechallenge cases (26% and
11%). Conclusion: Documentation of hepatoxicity in the medical literature is variable,
and many published instances do not stand up to critical review. A standardized system for categorizing drugs for hepatotoxicity potential will help develop objective and
reliable, computer-based instruments for assessing causality in drug-induced liver
injury. (HEPATOLOGY 2016;63:590-603)

rug-induced liver injury is an important and

not uncommon clinical problem in the practice
of hepatology. It is also perhaps the most diagnostically challenging of all liver-related conditions.
Critical elements in diagnosis include an appropriate
history of exposure, improvement upon stopping the
drug, and the exclusion of other causes of liver injury.
Also of critical importance is whether the agent is
known to be hepatotoxic and has been implicated previously in cases of liver injury. Information on the hepatotoxicity of drugs, however, is not always available or

easily found. Product labels (i.e., package inserts) provide some guidance as a part of the prescribing information, but they can be misleading or insufficient and
focus largely on the frequency of serum enzyme elevations during therapy.1-4 In contrast, clinically significant
liver injury from currently available medications is rare,
and its frequency and clinical features are not reliably
reported in product labels5 or in general reviews on the
efficacy and safety of medications. Most instances of
severe hepatotoxicity are first described after the medication is marketed and may be listed in the product label

Abbreviations: CNS, central nervous system; RUCAM, Roussel-Ulcaf Causality Assessment Method.
From the 1The National University Hospital of Iceland, Reykjavik, Iceland; 2Faculty of Medicine, University of Iceland, Reykjavik, Iceland; 3Liver Disease
Research Branch, Division of Digestive Diseases and Nutrition, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health,
Bethesda, MD
Received August 8, 2015; accepted October 29, 2015.
Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.28323/suppinfo.
Supported in part by the Intramural Research Program of the National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health.
590

HEPATOLOGY, Vol. 63, No. 2, 2016

as a post-marketing reported adverse event of unclear

significance or relationship to the drug.
Searching the medical literature for information on
the hepatotoxicity of drugs is also difficult and is not
always reliable. The most common causes of druginduced liver injury are well described in case reports
and case series and may have a known clinical signature
(i.e., phenotype) of liver injury, which is useful in making a diagnosis. The less common causes of druginduced liver injury, however, are more problematic;
they are documented only in isolated case reports or
small case series, which are often spread among multiple
journals from a variety of subspecialities from different
countries and in different languages. Furthermore, the
actual documentation of the hepatotoxicity and its relationship to the implicated drug may be poor, with inadequate descriptions of medical history, clinical features,
laboratory results, and suitable follow up, much less the
results of rechallenge. Although information about these
rare causes of hepatotoxicity is available in textbooks,
such books are often not widely available and rapidly
become outdated.
A newly developed open access and commercial-free
website established by the National Institutes of Health
called LiverTox (http://livertox.nih.gov) was an attempt
to address the need for more reliable and easily assessed
information on the hepatotoxicity of currently used
medications.6 The website includes information on virtually all commonly used medications, whether or not
they have been shown to cause liver injury. The website
is based on a rigorous literature search with the intent to
include all published case reports and case series from
the previous 50 years. Although the listing is thorough,
no attempt was made to judge the quality of the published reports or the degree of documentation of
causality.
We attempted to assess the accuracy of the categorization of drugs listed in LiverTox using a single, reasonably well-validated instrument for assessment of
causality in drug-induced liver injury: the Roussel Uclaf
Causality Assessment Method (RUCAM). This method
was developed by an international consensus group and
has been in use for 20 years.7,8 The current exercise was
meant also as an initial step in improving RUCAM,

BJORNSSON
AND HOOFNAGLE

591

which employs information on the hepatotoxicity

potential of medications but relies on product labels for
scoring of this potential. The exercise was also meant to
provide physicians in clinical practice with updated
information on the most common drugs associated with
hepatotoxicity and whether these agents have been
linked to cases of fatal liver injury or instances of positive rechallenge.

Materials and Methods

The current analysis of likelihood of hepatotoxicity
was based upon all drugs that were listed and described
in LiverTox.6 LiverTox is based upon drugs approved
for use in humans in the United States, but includes a
small number of agents with a well known hepatotoxicity profile that were recently withdrawn from use in the
United States or that are only marketed elsewhere. LiverTox was meant to include all medications that might be
considered in assessing drug-induced liver injury. Agents
that are not absorbed systemically (e.g., topical creams
and ointments, nasal sprays, eye and ear drops), normal
substances used in replacement amounts (e.g., immunoglobulin, plasma products, normal saline, food substances), and drugs with limited one-time usages or very
narrow applications (e.g., vaccines, local anesthetics,
recombinant proteins used as replacement for rare metabolic syndromes) were not included in LiverTox. For the
current analysis for likelihood of hepatotoxicity, herbal
and dietary supplements, metals, and illicit substances
that are currently described in LiverTox were excluded.
In the creation of LiverTox, drugs were divided into five
categories of likelihood for causing liver injury based on
the number of convincing reports in the published literature: category A, 50; category B, 12-49; category C,
4-11; category D, 1-3; and category E, none.
For the current analysis, the A-E category system was
used, and an additional category T was created for agents
that have not been implicated in causing liver injury at
standard doses, but are clearly hepatotoxic in high concentrations or with drug overdoses, examples being acetaminophen and niacin. In this analysis, the number of
cases published in the literature were counted. When
the total tally was less than 12 (categories C and D), the

Address reprint requests to: Einar S. Bj

ornsson, M.D., Ph.D., Professor of Gastroenterology and Hepatology, The National University Hospital of Iceland, 101
Reykjavik, Iceland. E-mail: einarsb@landspitali.is. Address reprint requests to: Jay H. Hoofnagle, M.D., Liver Disease Research Branch, National Institutes of
Health, Room 655, 6707 Democracy Boulevard, Bethesda, MD 20892-5450. E-mail: hoofnaglej@extra.niddk.nih.gov.
C 2015 by the American Association for the Study of Liver Diseases.
Copyright V
View this article online at wileyonlinelibrary.com.
DOI 10.1002/hep.28323
Potential conflict of interest: Nothing to report.

592

BJORNSSON
AND HOOFNAGLE

individual cases were reanalyzed by one author (E.S.B.)

using RUCAM.7,8 Cases classified as unlikely by this
method were excluded from the total tally of published
cases.
RUCAM is based on seven areas of clinical information and yields total scores that can range from 29 to
114, although the majority are between 0 and 10. The
score is then converted to one of the following likelihood categories based on an accepted convention: 0-2,
unlikely; 3-5, possible; 6-8, probable; and 9-14, highly
probable. The clinical information to assign a RUCAM
score was not always available in published reports, and
scores of highly probable were uncommon, for which
reason this classification was collapsed into the probable
category. Only case reports with RUCAM scores in the
range of possible or probable (3) were used to categorize drugs. Common reasons for why published cases
were scored as less than possible included (1) liver test
results that were not provided or never more than twice
the upper limit of normal, (2) lack of information on
the course of liver tests after stopping the implicated
drug, (3) lack of documentation of exclusion of other
important causes of liver injury, or (4) presence of a concomitant condition or medication that might explain
the liver injury.7,8 Important conditions that were often
not excluded and were possible causes included ischemic
liver injury, autoimmune hepatitis, gallstone disease,
reactivation of hepatitis B, and exacerbation of chronic
hepatitis C. In other cases, multiple potential hepatotoxic drugs were taken, and it was impossible to assign
causality to a single agent. These are common problems
in assessing drug-induced liver injury.
Well-documented case reports were the basis of the
majority of the analyses, but case series were also used if
the individual cases were subjected to formal causality
assessment. Thus, case series with defined causality
assessment were included in the analyses, such as from
the French,9 Spanish,10,11 Swedish,12-14 Icelandic,15
United States,16,17 Indian,18 and Latin American19 hepatotoxicity registries or prospective studies. Information
on the frequency of elevated liver tests in clinical trials
was not used, because individual cases were rarely presented in detail or subjected to specific causality assessment. Most information on clinically apparent
hepatotoxicity appears in the post-marketing phase,
once the drug is more broadly available.20
During the analysis and review, specific information
was collected on each drug, including its classification
(e.g., antimicrobial, cardiovascular, gastrointestinal), the
year of its initial approval in the United States, whether
the case reports included an instance with a fatal outcome, and whether a convincing instance of rechallenge

HEPATOLOGY, February 2016

with recurrence of injury had been published. A positive

rechallenge was defined as elevation of liver tests more
than twice the upper limit of normal or twice baseline
values following readminstration of the same drug and
after normalization or improvement of the liver tests
from the initial liver injury.7,8,21,22 The occurrence of a
fatal case of liver injury defines the drug as a serious
potential hepatoxin, whereas a positive rechallenge represents the most convincing evidence for causality in
drug-induced liver injury.

Results
As of May 1, 2015, a total of 863 agents were listed
and discussed separately in LiverTox. Of this total, 116
were excluded in the current analysis: 49 were herbal or
dietary supplements, 16 were metals, 11 were agents
that were uncommonly used and are no longer available
in the United States, and three were illicit substances of
abuse not given by prescription (cocaine, heroin,
opium) (Fig. 1). Another 37 agents had been approved
within the previous 5 years and were considered to have
too little clinical use for an accurate assessment of hepatotoxic potential. The remaining 747 agents comprised
the agents eligible for this analysis, but a number of
these were grouped because of similarity of structure,
activity, and adverse event profile. The most important
of these groups were the oral contraceptives consisting
of estrogens (n 5 9) and progestins (n 5 12), the
androgens or anabolic steroids (n 5 10), corticosteroids
(n 5 8), sulfonamides (n 5 5), amphetamines (n 5 4),
salicylates (n 5 3), gold (n 5 3), and mesalamine derivatives (n 5 3). Another 54 drugs that were isomers or
had similar structure were also combined for analysis
(e.g., esomeprazole and omeprazole, interferon-a and
peginterferon alfa). For these reasons, 76 different marketed drugs were not counted separately. Thus, a total
number of 671 distinct agents were available for the
hepatotoxicity likelihood analysis (Fig. 1).
Of the 671 distinct agents assessed, 353 drugs (53%)
were classified as having been linked to liver injury: 48
(13%) were assigned to category A, 76 (22%) were
assigned to category B, 96 (27%) were assigned to category C, and 126 (36%) were assigned to category D.
Another 7 (2%) were placed in a separate category T
because they were predominantly direct toxins and had
little or no hepatotoxic potential at standard doses. The
remaining 318 drugs (47%) had no convincing published report of clinical apparent liver injury and were
thus placed in category E; 281 of these agents had no
published report of hepatotoxicity, and 37 had a few


BJORNSSON
AND HOOFNAGLE

HEPATOLOGY, Vol. 63, No. 2, 2016

Fig. 1. Flow chart of analysis cohort from the LiverTox website.

reports, but all of them were scored as unlikely using

RUCAM.
The numbers of distinct agents (individual drugs or
related classes) for each category, the medication classes,
the year of initial approval or marketing, and the proportion of drugs with a fatal case of liver injury or an
instance of a positive rechallenge are given in Table 1.
Drugs with higher hepatotoxicity scores tended to have
been approved for longer and were more likely to have a
fatal case and a positive rechallenge case than drugs with
lower scores. Among major classes of drugs, antimicrobials, central nervous system agents, cardiovascular drugs,
antineoplastic and rheumatologic-immunomodulatory
agents were more likely and gastrointestinal, respiratory
and urologic agents less likely to be associated with liver
injury (Fig. 2).
Category A. A total of 48 distinct drugs were
assigned to category A having been implicated in more
than 50 cases of liver injury in the published literature
(Table 2). This category actually included 86 named
agents, but 39 were were contained in four classes of

593

drugs (21 estrogens-progestins, 10 anabolic steroids, five

miscellaneous sulfonamides and three gold products)
and six were drugs of similar structure and grouped
together as three entities (azathioprine with mercaptopurine; interferon-a with peginterferon alfa; phenytoin
with fosphenytoin). Most of the drugs or entities in category A (81%) were implicated in more than 100 cases
in the literature, and all except one (estrogens-progestins) had at least one convincing case that resulted in a
fatality. Positive rechallenges were described for 44 of
the category A agents (92%). This Category included
two well-known hepatoxic agents that were included in
LiverTox but were not marketed in the United States
despite wide-scale use abroad (flucloxacillin and nimesulide). All except one of these agents (telithromycin) have
been available clinically for more than 15 years (prior to
2000) and many (63%) have been available for more
than 35 years. (References for category A agents are
given in the Supporting Information).
Direct Toxins, Category T. Seven other drugs or
classes of drugs that might be qualified to be placed in
category A were listed separately, as they were clearly
hepatotoxic, but only when given in high concentrations
either orally (acetaminophen, aspirin and salicylates,
niacin and vitamin A) or intravenously (buprenorphine,
methylprednisone and tetracycline). These agents were
distinctive in being direct toxins (T) but were nonhepatotoxic and usually well tolerated at conventional doses.
All seven agents were implicated in fatal cases of liver
injury, and all had at least one instance of positive
rechallenge, but only with repeat high doses; in many
cases, conventional doses were tolerated without
Table 1. Comparison of Drug Classes
Between Likelihood Categories
Feature

No. of drugs
Drug class
Antimicrobial
CNS
Cardiovascular
Antineoplastic
Endocrine
Immunomodatory
Rheumatologic
Gastrointestinal
Urologic
Other
Year approved
1980 or earlier
1981-1999
2000-2010
Fatal cases
Rechallenge

Total

48

76

96

126

346

318

33%
12.5%
12.5%
10%
6%
12.5%
0%
0%
0%
13%

30%
20%
16%
13%
7%
1%
0%
4%
0%
9%

19%
17%
20%
16%
4%
3%
3%
5%
1%
12%

27%
19%
13%
17%
6%
6%
3%
2%
2%
6%

27%
17%
15%
14%
6%
6%
1.5%
3%
1%
10%

20%
22%
14%
7.5%
4.5%
3%
8%
7%
4%
10%

63%
35%
2%
98%
92%

37%
51%
12%
53%
38%

27%
44%
19%
23%
26%

29%
44%
27%
7%
11%

39%
44%
15%
35%
32%

35%
38%
27%

Abbreviation: CNS, central nervous system.

594

BJORNSSON
AND HOOFNAGLE

HEPATOLOGY, February 2016

Fig. 2. Proportion of agents

linked to liver injury by drug class.

recurrence of injury. These agents were clearly and predominantly direct toxins (T) as opposed to the other
agents that generally caused idiosyncratic injury. (References for category T agents are given in the Supporting
Information).
Category B. Seventy-six distinct agents were
assigned to category B being implicated in 12-49 cases
of liver injury in the literature (Table 3). All except one
(cyproterone) were marketed in the United States. Overall, 41 (54%) of the category B agents were associated
with at least one fatal case and 30 (39%) with a positive
rechallenge. Most of these agents (91%) have been in
clinical use for more than 15 years (since before 2000),
the eight exceptions being duloxetine, imatinib, lamotrigine, oxaliplatin, rosuvastatin, rivaroxaban, sertaline,
and voriconazole. (References for category B agents are
given in the Supporting Information).
Category C. Ninety-six distinct agents were implicated in four to 11 published case reports of hepatotoxicity (Table 4). Category C included 22 (23%) agents
that were implicated in at least one case with a fatal outcome and 25 (26%) cases that were shown to have
recurrent injury after re-exposure. Most of the drugs had
case reports with both possible and probable relationship as judged by RUCAM, but 16 (17%) had cases
with a possible likelihood score only (amlodipine, citalopram, clomiphen, cytarabine, dactinomycin, diltiazem,
disopyramide, etoposide, fluorouracil, glipizide, mitomycin, pantoprazole, phenelzine, risperidone, ritonavir,
and vancomycin). Overall, 91 case reports among drugs
in category C did not fulfill causality according to
RUCAM. (References for category C agents are given in
the Supporting Information).
Category D. One hundred twenty-six distinct
agents were implicated in causing liver injury in one to

three cases reports in the literature (Table 5). Fatal

instances were rare in this group (n 5 8 [6%]), and only
14 (11%) cases had an instance of a convincing positive
rechallenge. Three agents in this category were never
marketed in the United States (gliclazide, plicamycin
[mithramycin], and tibolone). Overall, 70 case reports
among drugs in category C did not fulfill criteria for
causality using RUCAM. (References for category D
agents, in addition to case reports of drugs in category C
and D that were unlikely according to RUCAM, are
given in the Supporting Information).
Category E. A remaining 318 drugs were placed
into category E, having no convincing reports of hepatotoxicity (Table 1). These 318 drugs had either no case
reports published or no case reports fulfilling the
RUCAM causality assessment criteria (n 5 37; Table 6).
Among the 318 drugs listed as not hepatotoxic, 78
(25%) did not have case reports in the literature that
qualified as being possibly related using RUCAM but
nevertheless were considered suspicious by the authors
based on liver test abnormalities in clinical trials or
unpublished or unsubstantiated reports to authorities,
marked with an asterisk (*) in Supporting Table 1.
(References for category E agents are given in the Supporting Information).

Discussion
In the current study, drugs implicated in liver injury
were categorized based on published and reasonably
documented cases of hepatoxicity. The categorization
revealed that most drugs with well-known potential for
hepatotoxicity (categories A and B) have been associated
with at least one instance with a fatal outcome and a
positive rechallenge. Most of these agents are well


BJORNSSON
AND HOOFNAGLE

HEPATOLOGY, Vol. 63, No. 2, 2016

595

Table 2. Category A (50 Published Cases)

Drug No.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48

Ingredient

No. of Cases

Fatalities

Rechallenge

Year

Classification

Allopurinol
Amiodarone
Androgenic steroids*
Atorvastatin
Auranofin/Gold products*
Azathioprine/Mercaptopurine
Busulfan
Carbamazepine
Chlorpromazine
Clavulanate with Amoxicillin
Dantrolene
Diclofenac
Didanosine
Disulfiram
Efavirenz
Erythromycin
Estrogens/Progestins*
Floxuridine
Flucloxacillin
Flutamide
Halothane
Hydralazine
Ibuprofen
Infliximab
Interferon-a/Peginterferon
Interferon-b
Isoniazid
Ketoconazole
Methotrexate
Methyldopa
Minocycline
Nevirapine
Nimesulide
Nitrofurantoin
Phenytoin/Fosphenytoin
Propylthiouracil
Pyrazinamide
Quinidine
Rifampin
Simvastatin
Sulfamethoxazole with TMP
Sulfasalazine
Sulfonamides*
Sulindac
Telithromycin
Thioguanine
Ticlopidine
Valproate

>100
>100
>100
65
>100
>100
>100
>100
>100
>100
51
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
52
>100
56
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
53
>100
68
>100
>100
>100
>100
79
56
>100
>100

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1
0
1
1
1
1
1
1
1
0
1
1
0
1
1
1
1
1
1
1
1
1
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
0
1
1

1965
1985
1981
1996
1985
1968
1954
1968
1957
1984
1974
1988
1991
1951
1998
1967
Pre-1980
1970
N
1989
1956
1984
1974
1998
1986
1993
1974
1981
1971
1962
1971
1996
N
1953
1946
1948
1971
1950
1971
1991
1980
1950
1973
1978
2004
1966
1985
1978

Rheumatologic
Cardiovascular
Endocrine
Cardiovascular
Rheumatologic
Immunomodulatory
Antineoplastic
CNS
CNS
Antimicrobial
CNS
Analgesic
Antimicrobial
Substance abuse agent
Antimicrobial
Antimicrobial
Endocrine
Antineoplastic
Antimicrobial
Antineoplastic
CNS
Cardiovascular
Analgesic
Immunomodulatory
Antimicrobial
Immunomodulatory
Antimicrobial
Antimicrobial
Antineoplastic
Cardiovascular
Antimicrobial
Antimicrobial
Analgesic
Antimicrobial
CNS
Endocrine
Antimicrobial
Cardiovascular
Antimicrobial
Cardiovascular
Antimicrobial
Rheumatologic
Antimicrobial
Analgesic
Antimicrobial
Antineoplastic
Hematologic
CNS

*Groups of agents (see text for details).

Abbreviations: CNS, central nervous system; N, never approved in the United States; TMP, trimethoprim.

represented in recent surveys of drug-induced liver

injury and have been available for several decades. The
analysis also demonstrated that many drugs considered
or labeled as occasionally hepatotoxic (initially placed in
category C or category D) have not been adequately
documented to cause liver injury in the literature when
causality was analyzed critically in a standardized, wellverified fashion.

While the lists of drugs in categories may be quite

useful clinically in assessing cases of suspected druginduced liver injury, the shortcomings and limitations of
this listing must be kept in mind. Most importantly, this
listing is based upon the published literature and not
upon spontaneous reports to regulatory agencies or the
drug manufacturers. Serious adverse events of medications are greatly underreported, and those that are

596

BJORNSSON
AND HOOFNAGLE

HEPATOLOGY, February 2016

Table 3. Category B (1249 Cases)

Drug No.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63

Ingredient*

No. of Cases

Fatalities

Rechallenge

Year

Classification

Acarbose
Amitriptyline
Amodiaquine
Amoxicillin
Asparaginase/Pegaspargase
Azithromycin
Captopril
Cefazolin
Ceftriaxone
Celecoxib
Chlorpropamide
Chlorzoxazone
Cimetidine
Ciprofloxacin
Clarithromycin
Clindamycin
Clopidogrel
Cloxacillin
Clozapine
Cyclophosphamide/Ifosphamide
Cyproterone
Dacarbazine
Dicloxacillin
Doxorubicin
Duloxetine
Enalapril
Enflurane
Etanercept
Ethionamide
Felbamate
Fenofibrate
Fluconazole
Fluvastatin
Glyburide (Glibenclamide)
Haloperidol
Heparin
Imatinib
Imipramine
Irinotecan
Isoflurane
Itraconazole
Ketamine
Lamotrigine
Leflunomide
Levofloxacin/Ofloxacin
Lisinopril
Lovastatin
Melphalan
Metformin
Methimazole (Thiamazole)
Moxifloxacin
Naproxen
Nifedipine
Olanzapine
Omeprazole/Esomeprazole
Oxacillin
Oxaliplatin
Paroxetine
Penicillamine
Phenobarbital
Piroxicam
Propafenone
Quinine

13
18
37
17
12
30
16
25
15
14
13
41
22
28
14
12
14
16
27
25
49
24
20
16
13
25
29
16
12
15
24
25
28
13
25
42
39
15
31
25
18
15
33
12
40
14
12
21
18
27
13
16
12
28
16
36
12
23
22
30
12
13
12

1
1
1
1
1
1
1
0
1
0
0
1
0
1
0
1
1
0
1
1
1
1
0
1
0
1
0
1
1
1
0
0
0
1
0
0
1
1
0
1
1
0
1
1
1
1
0
1
0
0
1
1
0
0
1
0
0
0
1
0
0
0
0

0
1
1
0
0
0
0
0
0
1
0
0
1
0
1
0
1
1
0
1
1
0
1
0
0
0
0
1
1
0
1
0
1
0
0
0
1
0
0
1
1
1
0
0
0
0
0
0
0
0
0
0
1
1
1
0
0
0
1
0
0
1
1

1995
1961
1977
1980
1972
1991
1981
1973
1984
1998
1958
1958
1977
1987
1991
1970
1997
1974
1989
1959
N
1975
1968
1974
2004
1985
1972
1998
1965
1993
1993
1990
1993
1964
1967
Pre-1980
2001
1959
1996
1979
1992
1970
2007
1998
1996
1987
1987
1964
1995
1959
1999
1976
1981
1996
1989
1989
2002
1992
1970
1911
1982
1989
2005

Endocrine
CNS
Antimicrobial
Antimicrobial
Antineoplastic
Antimicrobial
Cardiovascular
Antimicrobial
Antimicrobial
Analgesic
Endocrine
CNS
Gastrointestinal
Antimicrobial
Antimicrobial
Antimicrobial
Hematologic
Antimicrobial
CNS
Antineoplastic
Antineoplastic
Antineoplastic
Antimicrobial
Antineoplastic
CNS
Cardiovascular
CNS
Immunomodulatory
Antimicrobial
CNS
Cardiovascular
Antimicrobial
Cardiovascular
Endocrine
CNS
Hematologic
Antineoplastic
CNS
Antineoplastic
CNS
Antimicrobial
CNS
CNS
Rheumatologic
Antimicrobial
Cardiovascular
Cardiovascular
Antineoplastic
Endocrine
Endocrine
Antimicrobial
Analgesic
Cardiovascular
CNS
Gastrointestinal
Antimicrobial
Antineoplastic
CNS
Toxicology
CNS
Analgesic
Cardiovascular
Antimicrobial


BJORNSSON
AND HOOFNAGLE

HEPATOLOGY, Vol. 63, No. 2, 2016

597

Table 3. Continued
Drug No.

64
65
66
67
68
69
70
71
72
73
74
75
76

Ingredient*

Ranitidine
Rivaroxaban
Rosuvastatin
Sertraline
Sevoflurane
Stavudine
Tamoxifen
Terbinafine
Thiabendazole
Thioridazine
Venlafaxine/Dexvenlafaxine
Voriconazole
Zidovudine

No. of Cases

Fatalities

Rechallenge

Year

Classification

14
17
13
17
21
47
30
30
13
12
12
23
13

1
0
0
1
0
1
1
0
1
0
0
0
1

1
0
0
1
1
0
0
0
0
1
1
0
1

1983
2011
2003
2005
1995
1994
1997
1998
1967
1978
1965
2002
1987

Gastrointestinal
Hematologic
Cardiovascular
CNS
CNS
Antimicrobial
Antineoplastic
Antimicrobial
Antimicrobial
CNS
CNS
Antimicrobial
Antimicrobial

*Alternative names are given in parentheses.

Abbreviation: CNS, central nervous system; N, never approved in the United States.

Table 4. Category C (411 Cases)

Drug. No.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

Ingredient

Abacavir
Acebutolol
Acitretin
Adalimumab
Albendazole
Alfuzosin
Amlodipine
Amphetamines*
Amphotericin
Ampicillin
Anakinra
Atomoxetine
Bosentan
Bupropion
Candesartan
Cephalexin
Citalopram/Escitalopram
Clomiphene
Cyproheptadine
Cytarabine
Dactinomycin
Daptomycin
Desflurane
Diflunisal
Diltiazem
Disopyramide
Doxycycline
Dronedarone
Erlotinib
Etodolac
Etoposide
Ezetimibe
Famotidine
Flavocoxid
Fluorouracil
Fluoxetine
Gabapentin
Gefitinib
Gemcitabine

No. of Cases

Fatalities

Rechallenge

Year

Classification

6
4
5
10
10
4
6
4
5
4
6
8
6
9
5
7
4
6
4
4
7
4
7
6
5
7
10
4
11
5
4
8
6
4
4
6
4
5
6

1
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
1
0
0
0
0
0
0
0
1
1
1
0
1
0
0
0
0
0
0
0

0
1
0
1
0
1
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0

1998
1984
1996
2002
1996
1995
1992
Pre-1980
1971
1965
2001
2002
2001
1985
1998
1971
1998
1967
1961
1969
1964
2003
1992
1982
1982
1977
1967
2009
2004
1991
1983
2002
1986
2004
1962
1987
1993
2003
1996

Antimicrobial
Cardiovascular
Dermatologic
Immunomodulatory
Antimicrobial
Urology
Cardiovascular
CNS
Antimicrobial
Antimicrobial
Immunomodulatory
CNS
Respiratory
CNS
Cardiovascular
Antimicrobial
CNS
Endocrine
Respiratory
Antineoplastic
Antineoplastic
Antimicrobial
CNS
Analgesic
Cardiovascular
Cardiovascular
Antimicrobial
Cardiovascular
Antineoplastic
Analgesic
Antineoplastic
Cardiovascular
Gastrointestinal
Rheumatologic
Antineoplastic
CNS
CNS
Antineoplastic
Antineoplastic

598

BJORNSSON
AND HOOFNAGLE

HEPATOLOGY, February 2016

Table 4. Continued
Drug. No.

40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96

Ingredient

No. of Cases

Fatalities

Rechallenge

Year

Classification

Gemfibrozil
Glimepiride
Glipizide
Hydroxyurea
Indomethacin
Irbesartan
Isotretinoin
Ketoprofen
Labetalol
Lansoprazole/Dexlansoprazole
Lenalidomide/Thalidomide
Levetiracetam
Levocetirizine/Cetirizine
Linezolid
Losartan
Meloxicam
Mesalamines
Methoxsalen
Methylphenidate/Dexmethylphenidate
Metronidazole
Mexiletine
Mirtazapine
Mitomycin
Montelukast
Nafcillin
Natalizumab
Nefazodone
Nilutamide
Norfloxacin
Orlistat
Oxaprozin
Pantoprazole
Penicillin G
Phenelzine
Pioglitazone
Pravastatin
Pregabalin
Procainamide
Prochlorperazine
Pyrimethamine
Quetiapine
Ramipril
Risperidone
Ritonavir
Rosiglitazone
Sorafenib
Temozolomide
Tolcapone
Topiramate
Trastuzumab
Trazodone
Trimethoprim
Vancomycin
Verapamil
Vincristine
Warfarin
Zafirlukast

6
4
5
8
7
6
5
6
10
5
10
7
10
4
5
6
8
6
6
8
5
5
9
11
6
8
11
5
5
4
4
4
9
6
8
11
6
6
11
6
11
4
6
8
8
9
9
4
4
4
9
10
6
9
6
5
9

0
0
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
1
0
1
1
0
0
1
1
0
0
0
0
0
1
0
0
0
0
1
0
1
0
0
0
0
0
0
1
0
1

0
0
0
0
0
0
0
0
0
1
0
1
0
0
1
0
1
1
1
1
0
0
0
0
1
1
0
0
1
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
0
0
0
0
1
0
1
0
0
1
1
1
0

1981
1995
1984
1967
1965
1997
1982
1986
1984
1995
2006
1999
2007
2000
1995
2000
1987
1954
1955
1963
1985
1996
2002
1998
1970
1998
1994
1996
1986
1999
1992
2000
1958
1961
1999
1991
2004
1950
1956
1981
1997
1995
1993
1996
1999
2005
1999
1998
1996
1998
1981
1980
1986
1981
1963
1967
1996

Cardiovascular
Endocrine
Endocrine
Hematologic
Analgesic
Cardiovascular
Dermatologic
Analgesic
Cardiovascular
Gastrointestinal
Antineoplastic
CNS
Respiratory
Antimicrobial
Cardiovascular
Analgesic
Gastrointestinal
Dermatologic
CNS
Antimicrobial
Cardiovascular
CNS
Antineoplastic
Respiratory
Antimicrobial
Immunomodulatory
CNS
Antineoplastic
Antimicrobial
Weight loss agent
Analgesic
Gastrointestinal
Antimicrobial
CNS
Endocrine
Cardiovascular
CNS
Cardiovascular
Gastrointestinal
Antimicrobial
CNS
Cardiovascular
CNS
Antimicrobial
Endocrine
Antineoplastic
Antineoplastic
CNS
CNS
Antineoplastic
CNS
Antimicrobial
Antimicrobial
Cardiovascular
Antineoplastic
Hematologic
Respiratory

*Includes amphetamine, dextroamphetamine, lisdexamfetamine, and methamphetamine.

Includes mesalamine, balsalazide, and olsalazine.
Abbrviation: CNS, central nervous system.


BJORNSSON
AND HOOFNAGLE

HEPATOLOGY, Vol. 63, No. 2, 2016

Table 5. Category D (One to Three Cases)

Drug No.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63

Ingredient

No. of Cases

Fatalities

Rechallenge

Year

Acetazolamide
Acetohexamide
Aliskiren
Alosetron
Alprazolam
Anastrozole
Atazanavir
Atenolol
Baclofen
Benazepril
Bicalutamide
Bortezomib
Bromocriptine
Canakinumab
Carbenicillin
Carboplatin
Carvedilol
Cefaclor
Cefadroxil
Cefdinir
Cefepime
Cefoperazone
Cefotaxime
Cefprozil
Ceftazidime
Cefuroxime
Chlorambucil
Chlordiazepoxide
Cisplatin
Clofibrate
Clomipramine
Clonazepam
Clotrimazole
Cromolyn
Cyclizine/Chlorcyclizine
Cyclosporine
Dabigatran
Dalteparin
Darbepoetin alfa
Dasatinib
Deferasirox
Deferoxamine
Docetaxel
Donepezil
Doxepin
Enoxaparin
Entacapone
Ethambutol
Etravirine
Exemestane
Febuxostat
Flecainide
Flurazepam
Fondaparinux
Fosfomycin
Fosinopril
Gemifloxacin
Gliclazide
Hydrochlorothiazide/Chlorothiazide
Hydroxychloroquine/Chloroquine
Imipenem
Ivermectin
Lapatinib

1
3
2
1
1
3
1
3
1
2
3
3
1
1
1
1
1
2
2
2
1
1
1
1
1
2
3
2
2
3
1
1
1
1
1
2
2
2
1
1
2
1
1
1
1
3
1
2
1
3
1
3
2
1
2
3
1
2
2
2
2
2
3

0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0

1986
1964
2007
2000
1981
1995
2003
1981
1977
1991
1995
2003
1978
2013
1972
1989
1995
1979
1978
1997
1996
1978
1992
1983
1983
1983
1957
1960
1978
1967
1989
1997
1975
1984
1972
2003
2010
1994
2001
2006
2005
1969
1996
1996
1969
2003
1993
1967
2008
1999
2009
1985
1970
2001
1996
1991
2003
N
1959
1994
1985
1996
2007

Classification

Cardiovascular
Endocrine
Cardiovascular
Gastrointestinal
CNS
Antineoplastic
Antimicrobial
Cardiovascular
CNS
Cardiovascular
Antineoplastic
Antineoplastic
CNS
Immunomodulatory
Antimicrobial
Antineoplastic
Cardiovascular
Antimicrobial
Antimicrobial
Antimicrobial
Antimicrobial
Antimicrobial
Antimicrobial
Antimicrobial
Antimicrobial
Antimicrobial
Antineoplastic
CNS
Antineoplastic
Cardiovascular
CNS
CNS
Antimicrobial
Respiratory
Respiratory
Immunomodulatory
Hematologic
Hematologic
Hematologic
Antineoplastic
Toxicology
Toxicology
Antineoplastic
CNS
CNS
Hematologic
CNS
Antimicrobial
Antimicrobial
Antineoplastic
Rheumatologic
Cardiovascular
CNS
Hematologic
Antimicrobial
Cardiovascular
Antimicrobial
Endocrine
Cardiovascular
Rheumatologic
Antimicrobial
Antimicrobial
Antineoplastic

599

600

BJORNSSON
AND HOOFNAGLE

HEPATOLOGY, February 2016

Table 5. Continued
Drug No.

64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126

Ingredient

Letrozole
Lomustine
Lopinavir
Mebendazole
Mefenamic acid
Mefloquine
Memantine
Meropenem
Methazolamide
Metoprolol
Micafungin
Milnacipran
Mitoxantrone
Molindone
Mycophenolate mofetil
Nelarabine
Nelfinavir
Nizatidine
Nortriptyline
Ondansetron
Oxcarbazepine
Pentamidine
Pentostatin
Plicamycin
Prasagrel
Procarbazine
Propofol/Fospropofol
Rabeprazole
Raltegravir
Repaglinide
Rifabutin
Rilpivirine
Riluzole
Rivastigmine
Ropinirole
Saquinavir
Sibutramine
Sildenafil
Sirolimus
Sitagliptin
Spironolactone
Sunitinib
Tacrolimus
Tamsulosin
Terbutaline
Thiotepa
Thyroxine
Tibolone
Tizanidine
Tobramycin
Tocilizumab
Tolazamide
Tolmetin
Tolvaptan
Tranylcypromine
Triamterene
Valacyclovir/Acyclovir
Valsartan
Varenicline
Vismodegib
Zileuton
Ziprasidone
Zonisamide

No. of Cases

Fatalities

Rechallenge

Year

Classification

1
1
2
3
2
1
1
2
1
2
2
1
1
1
2
1
1
1
2
2
2
1
1
3
2
1
3
1
1
3
2
1
3
1
1
1
1
2
1
2
3
1
3
1
2
1
3
3
2
1
2
3
1
3
1
1
2
2
3
1
1
3
1

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
0
0
0
0
0
1
1
0
0
0
0
0
0
0

1997
1978
2000
1974
1967
1989
2003
1996
1959
1978
1985
2009
1987
1974
1995
2005
1997
1988
1964
1999
2000
1984
1991
N
2009
1969
1989
1999
2007
1997
1992
2011
1995
2011
1997
1997
1997
1998
1999
2006
1960
2006
1994
2007
1974
1959
Pre-1980
N
1996
1980
2010
1966
1976
2009
1961
1982
1995
1996
2006
2012
1996
2001
2000

Antineoplastic
Antineoplastic
Antimicrobial
Antimicrobial
Analgesic
Antimicrobial
CNS
Antimicrobial
Ophthalmologic
Cardiovascular
Antimicrobial
CNS
Antineoplastic
CNS
Immunomodulatory
Antineoplastic
Antimicrobial
Gastrointestinal
CNS
Gastrointestinal
CNS
Antimicrobial
Antineoplastic
Antineoplastic
Hematologic
Antineoplastic
CNS
Gastrointestinal
Antimicrobial
Endocrine
Antimicrobial
Antimicrobial
CNS
CNS
CNS
Antimicrobial
Weight loss agent
Urologic
Immunomodulatory
Endocrine
Cardiovascular
Antineoplastic
Immunomodulatory
Urologic
Respiratory
Antineoplastic
Endocrine
Endocrine
CNS
Antimicrobial
Immunomodulatory
Endocrine
Analgesic
Endocrine
CNS
Cardiovascular
Antimicrobial
Cardiovascular
Substance abuse agent
Antineoplastic
Respiratory
CNS
CNS

Abbreviation: CNS, central nervous system; N, never approved in the United States.


BJORNSSON
AND HOOFNAGLE

HEPATOLOGY, Vol. 63, No. 2, 2016

Table 6. Drugs With Published Case Reports but That

Did not Fulfill at Least a Possible Relationship, According
to RUCAM
Drug

Year Marketed

Amoxapine
Amprenavir/Fosamprenavir
Anidualfungin
Aripiprazole
Atovaquone
Bleomycin
Capecitabine
Carmustine
Ceftibuten
Clofarabine
Clorazepate
Desipramine
Diazepam
Dydrogesterone
Ethosuximide
Flucytosine
Fluphenazine
Fluvoxamine
Indinavir
Lithium
Maraviroc
Minoxidil
Piperacillin
Probenezide
Proguanil
Propranolol
Risedronate
Sumatriptan
Tolbutamide
Tipranavir
Tramadol
Triazolam
Trifluoperazine
Valganciclovir
Vardenafil
Zolmitriptan
Zolpidem

1992
1999/2003
2006
2002
1992
1973
1998
1977
1998
2004
1972
1964
1963
1961
1960
1991
1972
1994
1996
1970
2007
1979
1981
1951
2000
1967
1998
1997
1998
2005
1995
1982
1959
2001
2003
1997
1992

reported are often inadequately documented.20 This

limitation is confounded by a resistance among journal
editors to publish case reports, the poor regard with
which they are viewed, and the general impression that
clinical case descriptions are less worthy or less scientific
than experimental work or statistical analyses of large
datasets. Yet careful clinical observation is the basis for
much medical understanding and can lead to new
insights and identification of new medical conditions
and complications. This limitation is also exacerbated
by the resistance to publish any case history unless it is
the first instance of this form of adverse reaction, even
when an earlier report was incomplete or actually incorrect. The current analysis demonstrates the importance
and need to report serious liver-related adverse events in
the literature with adequate documentation, particularly
for rare events such as drug-induced liver injury. It

601

would have been helpful to be able to assess the frequency of reports of drug-induced liver injury as a function of prescriptions in use. However, it is not possible
to obtain this data for the United States, much less the
worldwide use of these medications for the entire 50
years or more that are covered in this literature search.
Another shortcoming of this categorization is that it
is based largely upon idiosyncratic drug-induced liver
injury and is less reliable in capturing cases of direct
toxic injury. Idiosyncratic liver injury is generally rare, is
not dose related, is not reproducible in animal models,
and is unexpected. Direct toxic injury, in contrast, is relatively common, is dose related, is reproducible in animal models, and is expected. For these reasons, a
separate category (category T) was created for agents
that are direct toxins and reliably cause liver injury but
only when given in high doses, either in an overdose
(such as acetaminophen) or with a narrow therapeutic/
toxic ratio (such as niacin).6 Actually, many other drugs
also have direct toxicity at high doses, particularly the
antineoplastic agents (cyclophosphamide, busulfan, fluorouracil, plicamycin), which when given intravenously
in high doses can cause an acute hepatocellular injury or
sinusoidal obstruction syndrome.6,23 The major categorization of agents in the current analysis was based
largely on their potential to also cause idiosyncratic
injury (azathioprine, chlorambucil, temozolomide) and
injury at conventional doses. Many antineoplastic agents
can cause both types of injury, which in an individual
case may be difficult to distinguish.
A particular problem in classifying the hepatotoxic
potential of drugs arises in assessing the antiretroviral
agents, which are often considered hepatotoxic; some
having boxed warnings of liver injury in their product
labels. However, only a small proportion of antiretroviral
agents have been well documented to cause clinically
apparent liver injury. Examples of agents with hepatotoxicity warnings with little evidence of hepatotoxicity are
lamivudine, tenofovir dipivoxil, and lopinavir. Competing causes of liver damage (hepatitis B and C, opportunistic infections, ischemia, other medications) often
confound causality assessment in the few detailed clinical
reports on these agents in the literature.24-26 Their hepatotoxicity potential is often based on retrospective analyses of large cohort studies in which detailed clinical
information is lacking.24-26 On the other hand, there are
several antiretroviral agents with well-documented hepatotoxicity, such as the older nucleoside analogues ziduvodine, didanosine, and stavudine. These agents were
assigned to categories A and B, largely because they cause
both idiosyncratic as well as direct hepatotoxicity (even
when given in standard doses) and have been implicated

602

BJORNSSON
AND HOOFNAGLE

in many cases of lactic acidosis with steatosis and hepatic

dysfunction and nodular regenerative hyperplasia. Other
antiretroviral agents such as abacavir, nevirapine, and efavirenz also have well-characterized idiosyncratic (largely
immunologically mediated) hepatotoxicity and were
assigned to categories A and B as well.
The current analysis was also limited largely to prescription medications and excluded herbal and dietary
supplements, which are being used increasingly by the
United States population and are increasingly implicated
in causing liver injury.15-17 However, there are many difficulties in categorizing these agents accurately. Most
commercial herbal and dietary supplements include
multiple components and provide little information on
their nature, source, concentrations, and purity. The
hepatotoxic component is often unknown or can only
be suspected. In addition, herbal and dietary supplement product labels may be deceptive, and the products
may not contain what is described in the labels or contain contaminants, undeclared medications, or synthetic
simulations of the active herbal motieties.
Because drug-induced liver injury is a frequent differential diagnosis in patients with acute and sometimes
chronic liver injury, information about the hepatotoxicity potential of the implicated agent is important.27 The
current classification of agents into five categories is
based on the published literature as of May 2015.
Clearly, this listing requires regular updating. For this
reason, these categorizations and tables will also be published on the LiverTox website in a searchable format
and with updating at regular intervals (http://livertox.
nih.gov/DrugCategory.html). Many agents are likely to
advance in categorization. Newly approved medications
will be added to LiverTox and will ultimately require
similar categorization.
Finally, this categorization of drugs for their hepatotoxic
potential should serve as a means for scoring causality
using clinical instruments such as RUCAM.7,8 Indeed,
this exercise was meant partially as a first step in improving RUCAM or developing a new causality assessment
system, particularly one that is more fully objective than
RUCAM and can be computer-based. In the current
RUCAM causality assessment tool, the information on
hepatoxicity (item 6) is not evidence-based, and a drug
with more than 100 case reports receives the same number
of points as a drug that has been labeled as hepatotoxic
due to a few cases.7,8 Our work could serve as a basis for a
new evaluation of hepatotoxicity. Because the current analysis also includes the specific number of cases in the literature, development of this computerized system could also
test and validate the reliability of the cut-points (50,

HEPATOLOGY, February 2016

12, 4 published cases) used for assigning different levels of likelihood.

At present, the only means of validation of causality
is against expert opinion and prospective analyses of
cases of drug-induced liver injury. In this regard, the
recent summary of the initial 8 years of the U.S. DrugInduced Liver Injury Network17 provides some validation of the current categorization schema. Among the
67 agents identified as causing three or more cases of
drug-induced liver injury between 2004 and 2012, all
except two (piperacillin and darunavir) were in categories A (n 5 25), B (n 5 24), C (n 5 15), or T (n 5 1).
In conclusion, the current categorization is a step
toward creating an internationally accepted instrument
for assessing causality in drug-induced liver injury.
These findings also call for a wider acceptance for publication of well documented liver-related adverse events,
not just for new or never-before-described agents, but
also for those that have been the subject of rare and not
always convincing reports.

References
1. Physician Desk Reference. http://www.pdr.net. Accessed May 1, 2015.
2. European Medicines Agency. www.ema.europa.eu. Accessed May 1, 2015.
3. Electronic Medicines Compendium. www.medicines.org.uk. Accessed
May 1, 2015.
4. The Swedish Association of the Pharmaceutical Industry. FASS: the
Swedish medicines information engine. http://service.lif.se/default.
aspx?id531983. Accessed May 1, 2015.
5. Bjornsson ES, Jacobsen EI, Einarsdottir R, Chalasani N. Discrepancies
in liver disease labeling in the package inserts of commonly prescribed
medications. Gastroenterology 2015;148:269-273.
6. National Institute of Diabetes and Digestive and Kidney Diseases. LiverTox: clinical and research information on drug-induced liver injury.
http://livertox.nih.gov. Accessed May 1, 2015.
7. Danan G, Benichou C. Causality assessment of adverse reactions to
drugsI. A novel method based on the conclusions of international
consensus meetings: application to drug-induced liver injuries. J Clin
Epidemiol 1993;46:1323-1330.
8. Benichou C, Danan G, Flahault A. Causality assessment of adverse
reactions to drugs-II. An original model for validation of drug causality
assessment methods: case reports with positive rechallenge. J Clin Epidemiol 1993;46:1331-1336.
9. Sgro C, Clinard F, Ouazir K, Chanay H, Allard C, Guilleminet C,
et al. Incidence of drug-induced hepatic injuries: a French populationbased study. HEPATOLOGY 2002;36:451-455.
10. Andrade RJ, Lucena MI, Fernandez MC, Pelaez G, Pachkoria K,
Garca-Ruiz E, et al.; Spanish Group for the Study of Drug-Induced
Liver Disease. Drug-induced liver injury: an analysis of 461 incidences
submitted to the Spanish registry over a 10-year period. Gastroenterology 2005;129:512-521.
11. Lucena MI, Andrade RJ, Kaplowitz N, Garca-Cortes M, Fernandez
MC, Romero-Gomez, et al. Phenotypic characterization of idiosyncratic
drug-induced liver injury: the influence of age and gender. HEPATOLOGY
2009;49:2001-2009.
12. Bj
ornsson E, Olsson R. Outcome and prognostic markers in severe
drug-induced liver disease. HEPATOLOGY 2005;42:481-489.
13. Bj
ornsson B, Bergqvist A, Jerlstad P, Olsson R. Fulminant druginduced hepatic failure leading to death or liver transplantation in Sweden. Scand J Gastroenterol 2005;40:1095-1101.

HEPATOLOGY, Vol. 63, No. 2, 2016

14. De Valle MB, Av Klinteberg V, Alem N, Olsson R, Bjornsson E.

Drug-induced liver injury in a Swedish University hospital out-patient
hepatology clinic. Aliment Pharmacol Ther 2006;24:1187-1195.
15. Bjornsson ES, Bergmann OM, Bjornsson HK, Kvaran RB, Olafsson S.
Incidence, presentation and outcomes in patients with drug-induced
liver injury in the general population of iceland. Gastroenterology
2013;144:1419-1425.
16. Chalasani N, Fontana RJ, Bonkovsky HL, Watkins PB, Davern T,
Serrano J, et al.; Drug Induced Liver Injury Network (DILIN). Causes,
clinical features, and outcomes from a prospective study of druginduced liver injury in the United States. Gastroenterology 2008;135:
1924-1934.
17. Chalasani N, Bonkovsky HL, Fontana R, Lee W, Stolz A, Talwalkar J, et al.
Features and outcomes of 889 patients with drug-induced liver injury: the
DILIN Prospective Study. Gastroenterology 2015;48:1340-1352.
18. Devarbhavi H, Dierkhising R, Kremers WK, Sandeep MS, Karanth D,
Adarsh CK. Single-center experience with drug-induced liver injury
from India: causes, outcome, prognosis, and predictors of mortality.
Am J Gastroenterol 2010;105:2396-2404.
19. Hernandez N, Bessone F, Sanchez A, di Pace M, Brahm J, Zapata R,
et al. Profile of idiosyncratic drug induced liver injury in Latin America: an analysis of published reports. Ann Hepatol 2014;13:231-239.
20. Senior JR. Evolution of the Food and Drug Administration approach
to liver safety assessment for new drugs: current status and challenges.
Drug Saf 2014;37(Suppl. 1):S9-S17.

BJORNSSON
AND HOOFNAGLE

603

21. Papay JI, Clines D, Rafi R, Yuen N, Britt SD, Walsh JS, et al. Druginduced liver injury following positive drug rechallenge. Regul Toxicol
Pharmacol 2009;54:84-90.
22. Andrade RJ, Robles M, Lucena MI. Rechallenge in drug-induced liver
injury: the attractive hazard. Expert Opin Drug Saf 2009;8:709-714.
23. Zimmerman HJ. Hepatotoxicity: The Adverse Effects of Drugs and
Other Chemicals on the Liver. 2nd ed. Philadelphia, PA: Lippincott
Williams & Wilkins; 1999:425-456.
24. Pavel S, Burty C, Alcaraz I, de la Tribonnie`re X, Baclet V, Ajana F,
et al. Severe liver toxicity in postexposure prophylaxis for HIV infection with a zidovudine, lamivudine and fosamprenavir/ritonavir regimen. AIDS 2007;21:268-269.
25. Brau N, Leaf HL, Wieczorek RL, Margolis DM. Severe hepatitis in
three AIDS patients treated with indinavir. Lancet 1997;349:924-925.
26. Vergis E, Paterson DL, Singh N. Indinavir-associated hepatitis in
patients with advanced HIV infection. Int J STD AIDS 1998;9:53.
27. Garca-Cortes M, Stephens C, Lucena MI, Fernandez-Casta~
ner A,
Andrade RJ. Causality assessment methods in drug induced liver injury:
strengths and weaknesses. J Hepatol 2011;55:683-691.

Supporting Information
Additional Supporting Information may be found at
onlinelibrary.wiley.com/doi/10.1002/hep.28323/suppinfo.