INTERNATIONAL JOURNAL OF IMMUNOPATHOLOGY AND PHARMACOLOGY Vol. 24, no.

2, 277-284 (2011)

EDITORIAL
PATHOGENESIS AND CLINICAL APPROACHES TO ANTICONVULSANT
HYPERSENSITIVITY SYNDROME: CURRENT STATE OF KNOWLEDGE

A. SCAPARROTTAI, A. VERROTTP, N.P. CONSILVI0 1, A. CINGOLANP, S. DI PILLOl,

M. DI GIOACCHIN02, M. VERINP and F. CHIARELLI I

'Department ofPaediatrics, "G. d'Annunzio" University, Chieti; 'Aliergy Related Disease Unit,
"G. d'Annunzio" University, Chieti, Italy

Received February 3, 2011 - Accepted April 13, 2011

Anticonvulsant hypersensitivity syndrome (AHS) is a rare, but severe and potentially fatal, adverse
reaction that occurs in patients who are treated with commonly used older anticonvulsant drugs
(phenytoin, carbamazepine and phenobarbital) and/or with some newer agents (lamotrigine). Paediatric
patients are at an increased risk for the development of AHS for the higher incidence of seizure disorder
in the first decade of life. Hypersensitivity reactions range from simple maculopapular skin eruptions
to a severe life-threatening disorder. AHS is typically associated with the development of skin rash,
fever and internal organ dysfunctions. Recent evidence suggests that AHS is the result of a chemotoxic
and immunologically-mediated injury, characterized by skin and mucosal bioactivation of antiepileptic
drugs and by major histocompatibility complex-dependent clonal expansion ofT cells. Early recognition
of AHS and withdrawal of anticonvulsant therapy are essential for a successful outcome. In vivo and
vitro tests can be helpful for the diagnosis that actually depends essentially on clinical recognition.

BACKGROUND (PB) as well as some newer agents, including

lamotrigine (LTG), oxcarbazepine (OXC), felbamate
An idiosyncratic drug reaction may be defined and zonisamide, have been implicated in eliciting
as "any adverse effect that cannot be explained a whole spectrum of hypersensitivity reactions,
on the basis of the known mechanisms of action ranging from simple maculopapular skin eruptions
of the drug. and occurs mostly unpredictably in to a severe life-threatening disorder (4).
susceptible individuals only, irrespective of dosage". Anticonvulsant hypersensitivity syndrome
Idiosyncratic reactions are relatively common in (AHS), that can also appear as drug rash with
patients treated with antiepileptic drugs (1-2). eosinophilia and systemic symptoms (DRESS), is
Antiepileptic hypersensitivity syndrome is a typically associated with the development of skin
severe, dose-independent, idiosyncratic cutaneous rash, fever and internal organ dysfunctions, that
reaction to aromatic anticonvulsants that may result may include blood dyscrasias, hepatitis, nephritis,
in-organ damage and death (3). myocarditis, thyroiditis, interstitial pneumonitis and
Aromatic anticonvulsant drugs such as phenytoin encephalitis (2, 5).
(PHT), carbamazepine (CBZ) and phenobarbital The initial reports ofAHS that primarily involved

Key words: anticonvulsant hypersensitivity syndrome, antiepileptic drugs, adverse drug reaction, phenytoin,
carbamazepine, lamotrigine, pathogenesis
Mailing address: Dr Marcello Verini,
Department of Paediatrics,
"G. d'Annunzio" University
0394-6320 (2011)
Via Dei Vestini 5 Copyright © by BlOLIFE, s.a.s.
66100 Chieti, Italy This publication and/or article is for individual use only and may not be further
Tel: ++39 0871358018 Fax ++39 0871584731 reproduced without written permission from the copyright holder.
e-mail: mverini@alice.it 277 Unauthorized reproduction may result in financial and other penalties
278 A. SCAPARROTTA ET AL.

PHT were in the 1930s, and in the following years erythematous or maculopapular rashes) are the most
numerous case reports described similar reactions common idiosyncratic reactions that affect 5-15% of
to other antiepileptic drugs, including PB, CBZ and patients started on treatment with CBZ, PHT, PB and
LTG. Since then a large number of adult patients and LTG. However, the same antiepileptic drugs are also
an increase of paediatric patients with AHS have associated with a risk of potentially life-threatening
been reported (3,6). Stevens-Johnson Syndrome (SJS), Toxic Epidermal
The pathogenetic mechanisms for AHS are not Necrolysis (TEN) and DRESS, with a frequency that
completely understood. Recent evidence suggests ranges between 1 and 10 per 10,000 new users of
that AHS is due to a sequence of chemotoxic these drugs (1-2).
and immunologically-mediated injury; AHS may The incidence ofAHS varies in the literature from
depend more on skin and mucosal bioactivation of 1per 1,500 exposures in new users ofanticonvulsants
antiepileptic drugs and on major histocompatibility to 1 per 10,000 exposures, with an approximate
complex (MHC)-dependent clonal expansion of T incidence of AHS of 1 per 3,000 exposures (6, 10,
cells (7). 12). The risk of AHS within the first 2 months of
This review discusses current understanding of new antiepileptic therapy is estimated to be 2.3 to
pathogenesis and clinical approaches to AHS in 4.5 for 10,000 PHT users and 1 to 4.1 for 10,000
adults and children. CBZ users (3,6).
A PubMed search indexed for MEDLINE was Paediatric patients may be at increased risk for
undertaken from 1962 to 2010 using the keywords: the development of this syndrome for the higher
"anticonvulsant hypersensitivity syndrome" and incidence of seizure disorder in the first decade of
"drug hypersensitivity anticonvulsant pathogenesis". life (3). No gender variation in the risk of developing
English language articles were reviewed. References AHS has been observed (6,9).
were reviewed for possible relevant articles. Clinical features ofAHS are also associated with
LTG administration in adult patients (13), with an
EPIDEMIOLOGY incidence of severe skin reaction of 19%, compared
with the 9% of patients treated with PB and CBZ
The exact incidence of AHS is not known for (14) and with the 9-13% ofpatients treated with PHT
under-reporting cases, variability in presentation and (14-15). On the contrary, AHS associated to valproic
lack of strict diagnostic criteria (8); probably AHS is acid is very rare, but valproic acid may affect
often unrecognised. Diagnosis can be difficultbecause the metabolism of LTG, so the risk of cutaneous
the syndrome may mimic infectious, neoplastic or reactions is significantly increased when LTG and
collagen vascular disorders. Another complicating valproic acid are used in the same patient (6).
factor is a lack of consensus of a generally accepted
term for the syndrome (9). For example, terms such PATHOGENESIS
as "Phenytoin/Dilantin syndrome", "Kawasaki-like
syndrome" or "Mononucleosis-like syndrome", Drug molecules are' generally considered to
have been used to describe a similar constellation of be too small (Molecular Weigh < 1,000 kDa) to
symptoms (9-10). be inherently immunogenic. It has therefore been
A variety of other drugs, such as sulfonamides, proposed that the anticonvulsant drug is oxidatively
dapsone, azathioprine, allopurinol have been metabolized to reactive,' potentially cytotoxic
associated with hypersensitivity syndrome while intermediate capable of interacting covalently
non-steroidal anti-inflammatory drugs, captopril, with cellular macromolecules. An imbalance in
calcium channel blockers, mexiletine, fluoxetine, bioactivating and detoxifying pathways results
terbinafine, metronidazole, minocycline and in increased covalent adduct (hapten) formation.
antiretroviral drugs have been particularly associated These drug-modified proteins could then function as
with DRESS (11). immunogens, with the potential to initiate immune
T-cell immune-mediated cutaneous response against drug-modified or native proteins.
hypersensitivity reactions (usually mild The generation of metabolites more reactive than
 Int. J. Immunopatbol. Pbarmacol. 279

the anticonvulsant drug is believed to be an essential is ingested with grapefruit juice. The major site of
component in the development of hypersensitive . drug bioactivation and injury with AHS appears
reactions. This process of bioactivation can be to occur in skin epidermis and other target tissues
demonstrated for both PHT and CBZ in two ways, (e.g., epidermal keratinocytes, mucosal cells, and
either by the generation of metabolites cytotoxic hepatocytes) (7).
to cells in culture or by the irreversible binding of However, accumulation of reactive drug
drug to microsomal proteins (referred to as covalent intermediates alone is unlikely to be sufficient to
adduct formation) (16). trigger AHS. Severe hypersensitivity to CBZ may
The cytochrome P-450 enzyme system be associated with a genetic polymorphism of the
is responsible for metabolizing the aromatic promoter region of the proinflammatory cytokine
anticonvulsant agents to arene oxide, and the tumour necrosis factor-a and with the heat shock
epoxide hydrolase enzyme system is responsible protein-70 gene cluster, which may be involved in
for detoxifying these metabolites. Susceptible reactions to stress (1).
individuals may have a diminished activity or a A study in Han Chinese has shown a strong
relative deficiency of epoxide hydrolase, probably association between HLA-B*1502 allele and CBZ-
related to an autosomal codominant inheritance at induced SJSITEN (17-20); HLA-B*1502 allele is
the cellular level. Since a sensitization period is very frequently observed also in the Thai population
required to develop delayed-type anticonvulsant (20). On the contrary, the association between HLA-
hypersensitivity, it usually occurs at least 7-10 days B*1502 and CBZ-induced SJSITEN has not been
after the first exposure; however, it may develop demonstrated in Caucasians (20-21) and Japanese
much sooner after a repeated exposure (3). patients, probably due to the low frequency of this
Although the exact mechanism of AHS remains allele in these ethnic groups (22), an observation
to be determined, it is thought to have at least three that may also explain the higher incidence of CBZ-
components:deficiencyor abnormalityofthe epoxide induced SJS in Chinese compared with Caucasians
hydroxylase enzyme that detoxifies the metabolites (1, 23). Another study on Japanese patients has
of aromatic amine anticonvulsants; herpes-type found that HLA-B*1518, HLA-B*5901, HLA-
virus reactivation; ethnic predisposition with certain C*0704 alleles and the haplotype HLA-A*2402-
specifichuman leukocyte antigen subtypes. The toxic B*5901-C*0102 are associated to a higher relative
intermediaries in the metabolism of anticonvulsant risk for the development ofsevere cutaneous adverse
drugs, arene oxides, can accumulate and directly drug reactions, suggesting that HLA-B*5901 can be
bind to macromolecules, causing subsequent cell one of the candidate markers for CBZ-induced SJS
death as well as acting as prohaptens that bind to in Japanese (24).
T cells, initiating a T-cell immune response and Most recent data also confirm that HLA-B*1502
systemic reactions (12). is not only a common risk allele for CBZ-induced
Loss of detoxification capacity, however, does SJSITEN, but also for SJS and TEN induced by
not explain individual susceptibility to AHS, other aromatic antiepileptic drugs (25).Sullivan and
because only a small subset of patients has defective Shear (26) have suggested that DRESS can be the
production of epoxide hydrolase (9). The rapid result of the combination ofa susceptible patient and
accumulation of reactive metabolites may trigger exposure to a drug capable of causing the reaction
cellular injury in susceptible patients, but does not and given at a sufficient dose and duration (8).
explain AHS susceptibility. Only recently it has Naisbitt et al. observed in patients with CBZ-
been appreciated that epidermal keratinocytes and induced AHS a clonal proliferation of CD4+ or
mucosal cells are the major sites of oxidative and CD4+/CD8+ cells, associated to the interferon-y
conjugative processing of drugs. A simple example secretion and to cytotoxic effects (27).
of this action is the ability of grapefruit juice to T-cellrecognition of CBZ is dependent on human
inhibit CYP3A4 in intestinal mucosa; consequently leukocyte antigen (HLA) class II-matched antigen-
patients being administered CBZ may have marked presenting cells. Posadas et al. showed in patients
increases in blood concentrations, when the drug with TEN the early activation of peripheral blood T
280 A. SCAPARROTTA ET AL.

cells with skin homing receptors (CLA-T cells), as a been reported (8).
result ofaromatic anticonvulsant drug administration. The current understanding of the chemical basis
They linked epidermal injury to accumulation of T ofimmune-mediated reactions is based on the hapten
cells and cytokines in skin, including tumor necrosis hypothesis, which requires drug bioactivation,
factor-a (7, 28). covalent binding to proteins, followed by uptake,
Drug-specific T-cell clones were demonstrated antigen processing and polyclonal T-cell immune
also in PB-sensitive children (29) and in LTG- response. The recently proposed "danger hypothesis"
sensitive patients (30), in which LTG binding of T can be considered an essential addition to the hapten
cells subsequently triggered clonal production of hypothesis. According to the danger hypothesis, the
CD4+ cells and some CD8+ cells in culture. immune response to a drug-derived antigen requires
The mechanisms producing AHS can be seen the presence ofco-stimulatory signals and cytokines,
as a defect in the human ecological barrier system. which propagate and determine the type of immune
Cytochrome enzymes are present in skin, mucosal response. The "danger signal" might result from
(oropharyngeal, intestine, vagina, conjunctiva), chemical, physical, or viral stress (32).
lung and hepatic tissue. These tissues also are sites
of MHC-dependent T-cell-mediated immunity and CLINICAL FEATURES
together the oxidative and immunologic systems are
adaptable to removing xenobiotics. MHC proteins AHS occurs most frequently 1 to 8 weeks after
coordinate T-cell responses for foreign molecules, the first exposure to anticonvulsant drugs, with
including some antiepileptic drugs that are processed fever (ranging from 38° to 40°C) in 90-100% (3,
as antigens by keratinocytes. It remains unclear 10, 12, 33-36) of patients, often associated with
whether small molecules, such as antiepileptic malaise and pharyngitis, followed, 1 or 2 days later,
drugs, may bind directly to specific T-cell receptors by skin rash and lymphadenopathy. In previously
without prior T-cell sensitization or drug-antigen sensitised individuals, AHS may occur within 1
presentation by MHC (7, 31). day on rechallenge. The syndrome is not related
The variability in the clinical presentation ofAHS, to anticonvulsant dosage or serum concentrations
such as epidermal eruptions and mucosal blistering and the symptoms often resolve within a few days
in SJS, epidermal necrosis and delamination in of discontinuing the drug, usually with complete
TEN, or eosinophilia in DRESS, is likely to reflect recovery; however, there is not a rapid clearing of
host-specific metabolic and immunologic reactions symptoms, but patients may continue to deteriorate
associated with T-cell-mediated responses in these weeks after the drug has been stopped (6). Diagnostic
tissues (27-28, 30-31). These reactions include criteria for AHS in both adults and children are
varying levels of cytochrome induction: MHC/non- reported by Kennebeck (33).
MHC-dependent T-cell activation; involvement of Drug eruption occurs in approximately 90-100%
cytotoxic CD8+ and CD4+ mediated responses; (3, 10, 12, 33-36) of patients with AHS and can
expression of adhesion and homing molecules on range from an exanthematous eruption to more
keratinocytes; chemokine release with infiltration serious reactions like SJS or TEN (9). Desquamation
by monocytes/macrophages and eosinophils. occurs with resolution (8).
Keratinocytes are apt to be directly involved in drug Frequent symptoms are also the facial and
bioactivation and T-cell-mediated injury, since they periorbital oedema, and mucosal involvement, with
express monocyte adhesion molecules, form adducts conjunctivitis and ulceration of the vaginal and
with antiepileptic drugs, and secrete cytokines (such buccal mucosae (6). Internal organ involvement
as TNF-a) and chemokines involved in immune cell may not develop for 1 or 2 weeks into the reaction
migration and cell injury (7). and may even appear 1 month later (9), with a great
Active viral infections can be another cofactor; in variability regarding the target organs involved as
patients with DRESS, reactivation of a latent human the spectrum of presentation (mild to severe) of the
herpesvirus-6 (HHV-6), cytomegalovirus (CMV), involvement (6).
Epstein-Barr virus (EBV), and HHV-7 infections has The most frequently involved internal organs are
 Int. J. Immonopathol. Pharmacol. 281

(3, 10, 12, 33-36): liver (50-90% of cases) involved In vivo systemic rechallenge (drug provocative test
with mild elevations in serum transaminase levels or or controlled re-exposure) is considered the gold
granulomatous hepatitis or fulminant hepatic necrosis standard, but ethically this is highly contentious,
(9,33); kidney, with a spectrum of presentation from because the rechallenge with the implicated drug
nephritis (7-55%) to acute renal failure; heart, with may result in severe morbidity or even death (4).
myocarditis, pericarditis or congestive heart failure; The tests available for diagnosis of AHS are in
lung involvement (9-45%) from pneumonitis to vivo tests (patch test) and in vitro tests, which include
respiratory distress syndrome. the Lymphocyte Transformation Test (LTT) and the
Arthritis (37), pancreatitis, thyroiditis, colitis, Lymphocyte Toxicity Assay (LTA). The usefulness
myositis and meningitis have also been described of patch test has been evaluated by several studies. It
(8, 26). Lymphadenopathy has been reported in should be performed 2-6 months after recovery from
57%-84% of patients (3, 33-36) and there are the date of the adverse drug reaction for best results.
numerous case reports describing PHT/CBZ-induced Limits of this test are various: the benefit of patch
pseudolymphoma and some cases of mycosis test appears to be maximal with certain drugs (CBZ
fungoides-like lesions (9). and PHT) and for specific clinical manifestations
The main haematological abnormalities in patients (strong reactions); it can detect only a small portion
with AHS consist of: leukocytosis (28-75%), with of the immunological reactions that underlie AHS;
atypical lymphocytes (50-71%), eosinophilia (42- the test procedure must be standardized (4-5).
73%) and lymphocytosis (11-71%) (3, 12, 34-36); The LTAmay predict the susceptibility ofpatients
less frequently leukopenia (35), anemia, positive to develop AHS, based on a genetic deficiency in
rheumatoid factor, antinuclear antibodies, anti- their cellular defence systems against toxic reactive
double-stranded DNA smooth muscle antibodies, cold drug metabolites, whereas the LTT may confirm
agglutinin, hypo/hypergammaglobulinaemia (6). the development of AHS by detecting peripherally
circulating drug-specific T cells. It is evident that
DIAGNOSIS these tests lack standardization and large-scale
validation to determine their appropriateness in
Diagnosis of AHS is very difficult, because it terms of sensitivity and specificity in addition to
may be confused with many diseases that must be positive and negative predictive values. In the
considered in the differential diagnosis: collagen diagnosis of AHS due to aromatic anticonvulsant,
vascular diseases, lymphoma, autoimmune diseases, the LTT has frequently shown a sensitivity of 71%-
infections and/or viral diseases (CMV, HIV, EBV), 100% (27, 38), but this also ranges as low as 19-40%
Kawasaki and hypereosinophilic syndromes. Also, (39). Estimate of specificity, however, seems to be
if a patient is being treated with more than one quite good (close to 100%) (5, 27-28). The use of
anticonvulsant, it is difficult to differentiate which the LTAhas revealed a range of sensitivity between
drug may be responsible for the reaction, and 85% and 100% with well documented AHS cases,
diagnostic rechallenge may result in severe reactions with satisfactory negative and positive predictive
in the truly drug-sensitive patient (6). values (5).
Currently the diagnosis depends essentially on: Recently, a new test allows the simultaneous
clinical recognition (9), comprehensive physical assessment of the frequencies and the cytokine-
examination and appropriate laboratory tests (4), producing phenotypes of allergen-specific T cells:
which may need to be repeated, if initially negative, the carboxyfluorescein succinimidyl ester (CFSE)
since reactivation may take 2-4 weeks to manifest assay has a greater sensitivity in the detection of
(12). phenytoin-specific proliferation than the LTT (38).
A validated gold standard in vitro test for
diagnosis or prediction of AHS is not yet available. TREATMENT AND PREVENTION
The value of all currently used in vitro and in vivo
tests is widely controversial and their sensitivity, Supportive care is the main therapy of AHS; the
specificities andvariability are yet to be determined. discontinuation of any aromatic anticonvulsant drug
282 A. SCAPARROTTA ET AL.

is the first step in the management of AHS. Most 40). Cross-sensitivity is significant among PHT,
patients are managed with topical corticosteroids CBZ, and PB, which obviously contain the aromatic
and antihistamines, mainly for pruritus, but the structure (12). CBZ, OXC, PHT, LTG and PB have a
evidence supporting antihistamine benefit is lacking 20-30% chance of cross-reacting. About 25-33% of
(12). The use of H2-receptor blockers or proton CBZ hypersensitive patients show cross-sensitivity
pump inhibitors is especially pertinent in patients to OXC and 25-70% of OXC-hypersensitive ones
with evidence of mucosal sloughing and suspected react to CBZ (25).
SJSITEN to prevent the risk of gastrointestinal Because a hereditary component is involved in
bleeding or ulcerations (3, 36). the development of AHS, the family members of
The use of systemic corticosteroids has not been the patient with AHS may also be at risk for this
substantiated in randomised blinded studies (34, 36), syndrome, if aromatic anticonvulsant drugs are
but is considered the standard of care for treatment used.
in severe cases of AHS and/or cases with internal
organ involvement (6,9,33); they have been used at CONCLUSIONS
initial dose of 1-2 mg/kg/day of prednisolone, with
favourable results (14, 37). The duration of therapy AHS is a rare, but serious, disorder associated
is variable, ranging from several weeks to months, with commonly used older anticonvulsants as well as
according to the clinical manifestation severity (3, with some of the newer ones; it should be suspected
34). Systemic corticosteroids should not be used in in patients with fever, rash and lymphadenopathy
patients with infectious processes or disrupted skin 2-3 months after starting anticonvulsant therapy.
barrier (SJS or TEN) for the increased mortality The mechanisms for AHS development are not
from sepsis (3, 36). In these cases intravenous completely understood, but involve genetic
immunoglobulin can be used with successful susceptibility with accumulation of anticonvulsant
outcomes, binding the Fas cellular surface receptor drugs and oxidized metabolites, eausing MHC/
on keratinocytes in patients with SJSlfEN, thus non-MHC-dependent clonal activation of T cells,
preventing cell death (6,34). and subsequent production of cytokine/chemokine
Benzodiazepines, valproic acid or gabapentin in T cells, keratinocytes and other target cells
have been used safely for seizure control. However, (7). Early recognition of AHS and withdrawal of
due to its inherent risk of hepatotoxicity, especially anticonvulsant drug therapy are essential for a
in children less than 2 years of age, valproic acid successful outcome. In vivo and in vitro tests can
therapy may not be a viable option during the be helpful for guiding future therapy, but these tests
first several weeks of AHS, if hepatic enzymes lack standardization and large-scale validation.
are elevated. Tiagabine, topiramate, levetiracetam Subsequently, the diagnosis depends essentially on
and zonisamide (the newest anticonvulsant non- clinical recognition.
aromatic agents) may represent another alternative;
REFERENCES
however, the relatively slow speed of dose titration
recommended for their administration may limit 1. Franciotta D, Kwan P, Perucca E. Genetic basis for
their usefulness in an actively seizing patient (3). idiosyncratic reactions to antiepileptic drugs. Curr
Anticonvulsant hypersensitivity. syndrome
Opin Neuro12009; 22:144-49.
can be prevented primarily through the increased
2. Zaccara G, Franciotta D, Perucca E. Idiosyncratic
education and awareness of patients, prescribers
adverse reactions to antiepileptic drugs. Epilepsia
and other health care providers. Patients need to
be informed that they may have the same type of 2007; 48:1223-44.
reaction to other anticonvulsant drugs (not in the 3. Bessmertny0, PhamT.Antiepileptichypersensitivity
same drug class as the original drug) and about syndrome: clinicians beware and be aware. Curr
the dangers of rechallenging (3), considering that AllergyAsthma Rep 2002; 2:34-39.
there is a 40%-80% incidence of cross-sensitivity 4. Elzagallaai AA, Knowles SR, Rieder MJ, et al.
between the different aromatic anticonvulsants (14, Patch testing for the diagnosis of anticonvulsant
 Int. J. Immunopathol. Pharntacol. 283

hypersensitivity syndrome: a systematic review. adverse drug reactions. Pharmacogenet Genomics

Drug Saf2009; 32:391-408. 2006; 16:297-306.
5. Elzagallaai AA, Knowles SR, Rieder MJ, et al. In 19. Man CB, Kwan P,Baum L, et al. Association between
vitro testing for the diagnosis of anticonvulsant HLA-B*1502 allele and antiepileptic drug-induced
hypersensitivity syndrome: a systematic review. Mol cutaneous reactions in Han Chinese. Epilepsia 2007;
Diagn Ther 2009; 13:313-30. 48: 1015-18.
6. Verrotti A, Trotta D, Salladini C, Chiarelli F. 20. Tassaneeyakul W, Tiamkao S, Jantararoungtong
Anticonvulsant hypersensitivity syndrome In T, et al. Association between HLA-B*1502 and
children. Incidence, prevention and management. carbamazepine-induced severe cutaneous adverse
CNS Drugs 2002; 16:197-205. drug reactions in a Thai population. Epilepsia 2010;
7. Krauss G. Current understanding of delayed 51:926-30.
anticonvulsant hypersensitivity reactions. Epilepsy 21. Lonjou C, Thomas L, Borot N, et al. A marker
Curr 2006; 6:33-37. for Stevens-Johnson syndrome: ethnicity matters.
8. Ganeva M, Gancheva T, Lazarova R, et al. Pharmacogenomics J 2006; 6:265-68.
Carbamazepine-induced drug reaction with 22. Kaniwa N, Saito Y, Aihara M, et al. HLA-B locus
eosinophilia and systemic symptoms (DRESS) in Japanese patients with anti-epileptics and
syndrome: report of four cases and brief review. Int J allopurinol-related Stevens-Johnson syndrome and
Dermatol 2008; 47:853-60. toxic epidermal necrolysis. Pharmacogenomics
9. Knowles SR, Shapiro LE, Shear NH. Anticonvulsant 2008; 9:1617-22.
hypersensitivity syndrome: incidence, prevention 23. Locharernkul C, Loplumlert J, Limotai C, et al.
and management. Drug Saf 1999; 21:489-501. Carbamazepine and phenytoin-induced Stevens-
10. Schlienger RG, Shear NH. Antiepileptic drug Johnson syndrome is associated with HLA-B*1502
hypersensitivity syndrome. Epilepsia 1998; 39:S3-7. allele in Thai population. Epilepsia 2008; 49:2087-91.
11. Tas S, Simonart T. Management of drug rash with 24. Ikeda H, Takahashi Y, Yamazaki E, et al. HLA class
eosinophilia and systemic symptoms (DRESS I markers in Japanese patients with carbamazepine-
syndrome): an update. Dermatology 2003; 206:353-56. induced cutaneous adverse reactions. Epilepsia
12. Bohan KH, Mansuri TF, Wilson NM. Anticonvulsant 2010; 51:297-300.
hypersensitivity syndrome: implications for 25. Chung WH, Hung SI, Chen YT. Genetic
pharmaceutical care. Pharmacotherapy 2007; 27: predisposition of life-threatening antiepileptic-
1425-39. induced skin reactions. Expert Opin Drug Saf2010;
13. Schlienger RG, Knowles SR, Shear NH. Lamotrigine- 9:15-21.
associated anticonvulsant hypersensitivity syndrome. 26. Sullivan JR, Shear NH. The drug hypersensitivity
Neurology 1998; 51:1172-75. syndrome: what is the pathogenesis? Arch Dermatol
14. Shear NH, Spielberg SP. Anticonvulsant 2001; 137:357-64.
hypersensitivity syndrome. In vitro assessment of 27. Naisbitt DJ, Britschgi M, Wong G, et al.
risk. J Clin Invest 1988; 82: 1826-32. Hypersensitivity reactions to carbamazepine:
15. Haruda F. Phenytoin hypersensitivity: 38 cases. characterization of the specificity, phenotype, and
Neurology 1979; 29:1480-85. cytokine profile of drug-specific T cell clone. Mol
16. Leeder JS. Mechanisms of idiosyncratic Pharmacol2003; 63:732-41.
hypersensitivity reactions to antiepileptic drugs. 28. Posadas SJ, Padial A, Torres MJ, et al. Delayed
Epilepsia 1998; 39:S8-S16. reactions to drugs show levels of perforin, granzyme
17. Chung WH, Hung SI, Hong HS, et al. Medical B, and Fas-L to be related to disease severity. J
genetics: a marker for Stevens-Johnson syndrome. Allergy Clin Immuno12002; 109:155-61.
Nature 2004; 428:86. 29. Chiappini E, Peruzzi M, Galli L, et al. Evidence for
18. Hung SI, Chung WH, Jee SH, et al. Genetic a T-lymphocyte dependent mechanism involving
susceptibility to carbamazepine-induced cutaneous proinflammatory cytokines In anticonvulsant
284 A. SCAPARROTTA ET AL.

hypersensitivity syndrome: a case report. Int J Antiepileptic hypersensitivity syndrome in children.

Immunopathol Pharmacol2002; 15:239-42. Ann Pharmacoth 2001; 35:533-38.
30. Naisbitt DJ, Farrell J, Wong G, et al. Characterization 37. Verrotti A, Feliciani C, Morresi S, et al.
of drug-specific T-cells in lamotrigine Carbamazepine-induced hypersensitivity syndrome
hypersensitivity. J Allergy Clin Immunol 2003; Ill: in a child with epilepsy. Int J Immunopathol
1393-403. Pharmacol 2000; 13:49-53.
31. Pichler WJ. Delayed drug hypersensitivity reactions. 38. Tsuge I, Okumura A, Kondo Y, et al. Allergen-
Ann Intern Med 2003; 139:683-93. specific T-cell response in patients with phenytoin
32. Naisbitt DJ, Pirmohamed M, Park BK. hypersensitivity: simultaneous analysis of
Immunopharmacology of hypersensitivity reactions proliferation and cytokine production by
to drugs. Curr Allergy Asthma Rep 2003; 3:22-29. carboxyfluorescein succinimidyl ester (CFSE)
33. Kennebeck GA. Anticonvulsant hypersensitivity dilution assay. Allergol Int 2007; 56: 149-55.
syndrome. J Am Board Fam Pract 2000; 13:364-70. 39. Romano A, Pettinato R, Andriolo M, et al.
34. Newell BD, Moinfar M, Mancini AJ, Nopper AJ. Hypersensitivity to aromatic anticonvulsants: in vivo
Retrospective analysis of 32 pediatric patients with and in vitro cross-reactivity studies. Curr Pharm Des
anticonvulsant hypersensitivity syndrome (ACHSS). 2006; 12:3373-81.
Pediatric Dermatol 2009; 26:536-46. 40. Aouam K, Ben Romdhane F, Loussaief C,
35. Carroll MC, Yueng-Yue KA, Esterly NB, Drolet BA. et al. Hypersensitivity syndrome induced by
Drug-induced hypersensitivity syndrome in pediatric anticonvulsants: possible cross-reactivity between
patients. Pediatrics 2001; 108:485-92. carbamazepine and lamotrigine. J Clin Pharmacol
36. Bessmertny 0, Hatton RC, Gonzalez-Peralta RP. 2009; 49: 1488-91.