DERMATOLOGICA SINICA 31 (2013) 169e174

Contents lists available at ScienceDirect

Dermatologica Sinica
journal homepage: http://www.derm-sinica.com

REVIEW ARTICLE

Epidermal necrolysis (StevenseJohnson syndrome and toxic

epidermal necrolysis): Historical considerations
Jean-Claude Roujeau*
Department of Dermatology, Université Paris-Est Créteil (UPEC), Créteil, France

a r t i c l e i n f o a b s t r a c t

Article history: Objective: To describe the history of toxic epidermal necrolysis, before and after the original report by the
Received: Aug 22, 2013 British dermatologist Alan Lyell in 1956.
Revised: Sep 24, 2013 Methods: Subjective expert choice of key advances in the comprehension of the nosology, classification,
Accepted: Sep 25, 2013
causality, and mechanisms of epidermal necrolysis (EN) over more than a century.
Results: Epidermolysis had been reported long before Lyell’s paper, but most cases had likely been
Keywords:
staphylococcal scalded skin syndrome in children. Concerning non-Staphylococcus EN, confusion with
drug hypersensitivity
erythema multiforme dissipated and its relation to StevenseJohnson syndrome was clarified. Tremen-
epidermal necrolysis
erythema multiforme
dous advances were made in understanding the causes and mechanisms, with increased acceleration in
StevenseJohnson syndrome the last 10 years.
Conclusion: The next decade should be devoted to improve the prevention and management of a disease
that is the most terrible form of drug hypersensitivity.
Copyright Ó 2013, Taiwanese Dermatological Association.
Published by Elsevier Taiwan LLC. All rights reserved.

History of the disease: Nosology/classification, and diagnosis Present knowledge remains insufficient to define the boundaries of
GBFDE versus TEN clearly.
Original description of toxic epidermal necrolysis The description of TEN had already been provided by some
predecessors to Lyell under different denominations: “unusual
The name “toxic epidermal necrolysis (TEN)” was introduced in 1956 bullous eruption” (Lang and Walker in 195610), “erythroderma with
by Lyell1 to describe the clinical disorder characterized by extensive epidermolysis” (Debré et al 193911).
destruction of epidermis that resembled scalding. Since then, the
eponym Lyell’s syndrome became widely accepted. As acknowledged
Erythema multiforme, StevenseJohnson syndrome, and related
later by Lyell,2 his original paper mixed three different disorders that
diseases: History and confusion
all had been already described more or less clearly.
The first one was rapidly identified as staphylococcal scalded skin
The initial description of erythema exudativum multiforme is
syndrome,3 and Lyell himself4 (Figure 1) contributed to the discov-
attributed to von Hebra.12 I was not able to consult von Hebra’s
ery of the toxin responsible for a superficial, subcorneal detachment
original reports but from photographic pictures of Hebra’s draw-
that is histologically very distinct from the deep necrolysis charac-
ings it seemed to encompass many skin diseases with a circinate
terizing TEN.5 This had been previously recognized and reported as
pattern, of which some would probably be given a variety of other
“neonatal pemphigus”6,7 or von Rittershain disease.8
denominations today [e.g., erythema annulare, pyoderma gan-
The second is now labeled as generalized bullous fixed drug
grenosum, subacute lupus erythematosus (LE), Sweet’s syndrome].
eruption (GBFDE),9 a disease characterized by necrolysis of full-
Rendu in 1916,13 Fiessinger and Rendu in 191714 described a
thickness epidermis on large well-demarcated blisters, which is
mucocutaneous eruption without clear cause that was later on
similar to that of TEN; however, it is distinct from TEN in the
reported as FiessingereRendu syndrome or pluriorificial erosive
absence or, if present, mildness of mucous membranes erosions,
ectodermosis, and retrospectively seems very similar to the two
usual relapses, as well as with mild differences in histopathology.
cases described in 1922 by Stevens and Johnson.15 In Germany
Fuchs syndrome is used to describe a variant of erythema multi-
* Université Paris-Est Créteil (UPEC), 19 Avenue d’Alembert, 92160 Antony,
forme, and this syndrome mainly involves the mucosal surfaces. As
Créteil, France. the skin may be completely unaffected, it is an underrecognized
E-mail address: jean-claude.roujeau@wanadoo.fr. diagnosis and often difficult to confirm.16

1027-8117/$ e see front matter Copyright Ó 2013, Taiwanese Dermatological Association. Published by Elsevier Taiwan LLC. All rights reserved.
http://dx.doi.org/10.1016/j.dsi.2013.09.009
170 J.-C. Roujeau / Dermatologica Sinica 31 (2013) 169e174
Figure 1 Alan Lyell speaking at the first International Symposium on TEN, Créteil
1985.
Use of the terms “EM minor (erythema multiforme minor)” and
“EM major (erythema multiforme major, EMM)” was proposed in
the early 1950s for separating the classical mild cutaneous syn-
drome, as described by Hebra (erythema multiforme minor), from
the usually more severe syndrome, with marked mucosal damage,
as described by Stevens and Johnson, Fiessinger and Rendu, or
Fuchs (all grouped under erythema multiforme major or majus).17
Because many physicians had observed the progression of cases
from a phenotype of StevenseJohnson syndrome (SJS) to a
phenotype of TEN,18,19 TEN was included for years in an “EM
spectrum”19 or in an even larger “mucocutaneous syndrome.”20
Attempts to clarify the confusion
Several facts were challenging the concept of the EMeSJSeTEN
spectrum. First, the individual lesions that preceded detachment of
epidermis were not suspected of being EM by Stevens and Johnson.
In clinical reports of TEN, early lesions were clearly reported as “red
or purple macules, often brown, occasionally purpuric, sometimes
looking like ‘targets’ close to EM.”21 Second, it was difficult to
explain why “EM is often provoked by herpes simplex infection,
whereas I know of no evidence that this provokes TEN, if at all”
(Lyell19).
A first proposal to split the “EM spectrum” was done by Ruiz-
Maldonado, a Mexican dermatologist who proposed the classifi-
cation of acute disseminated epidermal necrosis (ADEN) types 1, 2,
and 3 for what we would call today SJS, overlap SJSeTEN, and TEN,
respectively. ADEN was ipso facto separated from EM.22
In 1987, as a preliminary step to a large case-control study of the
risk factors for EMM, SJS, and TEN, an international group of in-
vestigators had to agree on definitions and classification criteria.
After reviewing several hundred photographs of historical cases,
they proposed a classification based on the individual pattern and
topographic distribution of the skin lesions and maximum extent of
blisters/erosions (expressed as percentage of body surface area
involved).23
Implicit to this new classification were the hypotheses that: (1)
EMM is different from SJS, and (2) SJS and TEN are only severity
variants of a single entity.
Both hypotheses were strongly supported by the results of the
case-control analyses done on 552 patients and 1720 controls and
showing that EMM differed from both SJS and TEN by de-
mographics, associated diseases, causes, and severity, and that SJS,
Table 1 SCAR study classification of EMM and Epidermal necrolysis.
Classification Type of lesions Distribution % BSA
eroded
EMMa Typical targets Acral d
SJSa Spots
atypical targets Widespread <10
Overlap SJSeTEN Spots
atypical targets Widespread S10 to 30
TEN with spots Spots
atypical targets Widespread S30
TEN without spotsb Diffuse erythema, no spot Widespread S30
or target
BSA ¼ body surface area; SCAR study ¼ Severe Cutaneous Adverse Reactions study;
SJS ¼ StevenseJohnson syndrome; TEN ¼ toxic epidermal necrolysis.
a
Some cases with discordant criteria (widespread typical targets or acral spots)
were classified as atypical EMM or atypical SJS, respectively
b
Most cases of “TEN” without spots” had an alternative diagnosis of generalized
bullous fixed drug eruption.
TEN, and overlapping cases differed only by the extent of detach-
ment, and therefore severity.24 These results strongly suggested
that a “spectrum from EMM to TEN” does not exist, but rather that
there is on one hand a group of typical/atypical EMM and on the
other a real continuum from SJS to TEN. The simplest way to express
this continuum would be the denomination of epidermal necrolysis
(EN) that is used in this review.
Current classification
Most authors currently use the “Severe Cutaneous Adverse Re-
actions (SCAR) study” classification,23 presented in Table 1.
Diagnosis criteria are based on expert consensus and were never
submitted to formal validation. These criteria are erosions of mu-
cous membrane on at least two different sites, “spots” or “atypical
targets,” skin blisters or erosions, skin pain, Nikolsky’s sign, and
detachment of large epidermal sheets. The diagnosis is probable if
three or more of the aforementioned criteria are present, definite if
confirmed by unequivocal clinical photographs, and/or biopsy. The
typical histology pattern shows extensive apoptosis of keratino-
cytes on the full thickness of the epidermis and mild interface
dermatitis. Negative direct immunofluorescence findings are
important in case of ambiguous histology.
Differential diagnoses are listed in Table 2, and
Table 31,3,9,11e15,18,22,23,28,29,32,33,37,38,43e48,50,52,53,58e60,63e66 presents
the key dates and publications related to epidermal necrolysis.
Epidemiology
The best available evidence on the incidence of EN was provided by
studies done before the generalized use of current classification.
There was obviously some room for ambiguity. For EMM, SJS, TEN,
altogether the figure was four cases/million/year in the United
States.25 In France, Germany, Italy, and Singapore, the incidence of
TEN was reported to be one case/million/year26,27 using disease
definitions encompassing what we would now call TEN (more than
30% detachment) and SJSeTEN overlap (10e29% detachment).
Concerning SJS, more recent data suggest an incidence that is about
equal to the total of TEN and SJSeTEN overlap.28 Taking all these
data into account, a figure of two cases/million inhabitants/year can
be considered highly plausible for EN in Europe. The incidence may
be more elevated in other parts of the world because of risks related
to ethnicity29 or to more frequent use of “highly associated” med-
ications (Africa).
Severity
The high mortality associated with TEN was recognized very early.
Actually the qualification of “toxic” chosen by Lyell referred to the
 J.-C. Roujeau / Dermatologica Sinica 31 (2013) 169e174 171

Table 2 Main differential diagnoses of SJS/TEN.

Diagnosis Context Main clinical differences Pathology Causes

SSSS Infants, renal failure in adults No mucous membrane involvement Subcorneal detachment Staphylococcus toxins
AGEP Recent initiation of medicines Pustules Subcorneal pustules Medications a
Burns Unconsciousness Absence of “spot” or atypical targets Epidermis involved on variable Thermal or chemical
thickness
aGVHDb Bone marrow transplantation Slower progression, gut and liver May be very close to that of TEN Alloimmunity
involved
b
LE SLE or SCLE Slower progression, photo May be very close to that of TEN Autoimmunity
distribution
Paraneoplastic pemphigus Lymphoma Slower progression, severe mouth Acantholysis, positive DIF Ab-mediated
lesions autoimmunity
LABD, drug induced Recent initiation of medicines Some tense blisters Subepidermal blister, no necrosis, Medicationsc
pos. DIF
E(E)MM Children, young adults, frequent Typical targets, no confluence Less necrosis, more infiltrates Infections, idiopathic
recurrences

AGEP ¼ acute generalized exanthematous pustulosis; aGVHD ¼ acute graft-versus-host disease; E(E)MM ¼ erythema (exsudativum) multiforme majus; LE ¼ lupus ery-
thematosus; LABD ¼ linear IgA bullous disease; pos. DIF ¼ positive DIF; SCLE¼subacute cutaneous lupus erythematosus; SLE ¼ systemic lupus erythematosus;
SSSS ¼ staphylococcal scalded skin syndrome; SJS ¼ StevenseJohnson syndrome; TEN ¼ toxic epidermal necrolysis.
a
Frequently responsible drugs: amoxicillin, diltiazem, hydroxychloroquine, terbinafine
b
It is not yet clear whether aGVHD should be considered a cause of EN or a differential
c
Frequently responsible drugs: vancomycin, diclofenac, several antibiotics.

life-threatening severity of the disease and not to its causes or
mechanisms as it was too often believed later. For years, SJS was
considered much more benign, probably in large part because of
confusion with EMM. A recent cohort study that included 460 pa-
tients with EN observed an overall disease-related mortality of 30%
at 90 days after onset. The death rates were 20%, 30%, and 50% for
subgroups of SJS, overlap, and TEN, respectively.28 In addition to
very high mortality, EN is the cause of nearly constant sequelae,
occasionally progressing and often invalidating.
Evolution of concepts on causality
The concept that medications were a frequent, if not the only, cause
of TEN was rapidly accepted by the medical community . may be
too quickly in the opinion of Lyell.19 Strong evidence of such cau-
sality (and also of relationship between SJS and TEN) was provided
by several mass campaigns of prevention of infectious diseases by
long-acting sulfonamides, as these resulted in “epidemics” of SJS
and TEN.18,30

Table 3 Key dates and publications related to epidermal necrolysis.

Date Author(s) Concept Refs

1866 von Hebra Erythema exsudativum multiforme 12

1916, 1917 Rendu, Fiessinger and Rendu Multiorificial ectodermosis 13,14
1922 Stevens and Johnson StevenseJohnson syndrome 15
1939 Debré et al Bullous erythroderma with epidermolysis 11
1956 Lyell Toxic epidermal necrolysis 1
1968 Bergoend H et al High risk of long-acting sulfonamides 18
1968, 1970 Billingham and Streilein, Streilein and Billingham TEN and a graft-versus-host model in hamsters 37,38
1970 Mellish and Glasgow Staphylococcal scalded skin syndrome 3
1972 Peck et al Human graft-versus-host reaction as a cause of TEN 39
1972 Kauppinen Drug rechallenge often negative in SJS and TEN 9
1985 Ruiz-Maldonado ADEN types 1, 2, and 3 22
1985 Revuz et al First international meeting on TEN, Créteil, France d
1985 Lyell A (Créteil meeting) The Jackpot hypothesis d
1986 Roujeau et al Mild links between HLA and TEN 58
1987 SCAR study group Initiation of multinational case-control study on EN d
1990 Roujeau and Revuz Acute skin failure 64
1993 Correia et al Studying T cells in blister fluid of TEN 43
1993 Bastuji-Garin et al Consensus definition of EEMM, SJS, and TEN 23
1995e2002 Pichler et al Drug-specific T cell clones, p-i concept 44e47
1995 Roujeau et al Results from first case-control study of drug risks in EN 32
1998 Wolkenstein et al First RCT in TEN (thalidomide proved deleterious) 65
2000 Bastuji-Garin et al SCORTEN, a prognosis score allowing comparisons between series 66
2002 Nassif et al Drug-specific cytotoxic cells in blister fluid of TEN 48
2004 Chung et al Carbamazepine (CBZ)-related TEN and HLA-B*15:02 29
2005 Hung et al Allopurinol-related TEN and HLA-B*58:01 59
2008 Mockenhaupt et al EuroSCAR case-control study of SJS/TEN 33
2008 Chung et al Granulysin as the key cytokine in necrolysis 50
2011 Ko et al Restricted TCR needed for CBZ-related SJS/TEN 52
2011 Genin et al GWAS on a large European series of SJS/TEN cases 60
2012 Wei et al Direct noncovalent link between CBZ and HLA-B1502 53
2012 Chen et al Eradication of CBZ-induced TEN from Taiwan 63
2013 Sekula et al SJS/TEN even more severe than suspected 28

ADEN ¼ acute disseminated epidermal necrosis; EN ¼ epidermal necrolysis; GWAS ¼ genome-wide association studies; HLA ¼ human leukocyte antigen; p-i
concept ¼ pharmoco-immune concept; RCT ¼ randomized controlled trial; SCAR study ¼ Severe Cutaneous Adverse Reactions study; SJS ¼ StevenseJohnson syndrome;
TCR ¼ T-cell receptor; TEN ¼ toxic epidermal necrolysis.
172 J.-C. Roujeau / Dermatologica Sinica 31 (2013) 169e174
It is anyhow important to keep in mind that no drug cause was
suspected in the two original cases published by Stevens and
Johnson.15 However, only one of the four original cases was re-
ported by Lyell.1
In 1990, I contributed a review on SJS and TEN stating that
“Drugs are the most important, if not the only, cause of TEN .
Infections are well-recognized causes of SJS but not of TEN.”31 Both
statements were erroneous: the hypothesis of different causes for
SJS and TEN was wrong and so is the suggestion that TEN had no
other cause than drugs.
The results of the SCAR case-control study clearly indicated that
the “etiologic fraction” for “associated medications” was very
similar for SJS, overlap SJSeTEN, and TEN and around 0.65 in all
three groups.24,32 Because the etiologic fraction was calculated only
for drugs with a statistically significant association, it was under-
estimated by lack of statistical power. Further experience on larger
numbers of cases confirmed the existence of low percentage of cases
without any exposure to medications (2e3%) and a larger number
(10e15%) of cases exposed to several drugs that all were unlikely the
cause. In 20e25% of cases, one or several medications were possibly
the cause without any definite certainty.33 In conclusion, no more
than 70e80% of cases are drug induced and the percentages are
similar for SJS, TEN, and overlaps.34 The percentage of “idiopathic
cases” is probably higher in children. Up to now only a very low
proportion of cases of EN (z1%) had an established nondrug cause.
These causes include acute graft-versus-host-disease (GVHD), some
variants of acute LE,35 and infections. Infection-related EN was
especially demonstrated with Mycoplasma pneumoniae36 but other
agents were also suspected (Klebsiella pneumoniae, viruses). Over
dosages of methotrexate or colchicine may also induce damages of
skin and mucous membrane that closely resemble EN.
Progress on comprehension of mechanisms
In early reports, the mechanisms leading to EN appeared very
mysterious. Histological slides of skin lesions actually evidenced
the total destruction of the epidermis upon a “silent dermis” with a
very mild cell infiltrate, no lesion of vessels, no deposits of immu-
noglobulins (Ig) or complement.
Some important steps in the advancement of knowledge are
summarized below.
Resemblance to GVHD
The first animal model resulting in necrolysis of epithelial cell was
established by Billingham and Streilein, two famous immunologists,
in 1968 and 1970.37,38 The EN developed around a site of local GVHD
in a complex model of cutaneous GVHD, at a time when the key role
of dendritic cells was still unknown. In this model, an unidentified
substance present in the plasma was able to cause epidermolysis.
In 1972, TEN was reported as a possible expression of acute
GVHD in humans.39
More recently, a Japanese team developed a mouse model of
GVHD in transgenic mice expressing OVA in keratinocytes and
developing epidermolysis after the transfer of OVA-specific cyto-
toxic T cells.40 The development of epidermolysis needed some
immunosuppression and implied inactivation of regulatory cells.41
Focus on blister fluid
The first attempts to study T cells at the site of lesions used immu-
nolabeling techniques and extraction of T cells from trypsinized
sheets of necrotic epidermis. Very few cells were obtained and
viability was poor.42 Tremendous progresses were made possible by
studying cells present in the blister as first done by Osvaldo Correia
et al in Portugal.43 Many advances were made by studying this
blister fluid and characterizing the cells present in this fluid.
Drug-specific T cells
In the last 20 years, the key role of drug-specific T cells in drug
allergy was definitely demonstrated by the establishment of human
T-cell clones, derived from the blood lymphocytes or from skin
lesions of patients with a variety of reactions. T-cell clones
demonstrated that drugs can be recognized by human T cells and
suggested original pathways of activation. Clones have been ob-
tained with most medications that induce allergic reactions in
humans, including penicillin G, amoxicillin, sulfamethoxazole,
phenobarbital, carbamazepine (CBZ), and lamotrigine. They were
often of both CD4 and CD8 phenotypes, whatever the original type
of eruption had been.44e47
TEN drug-specific T cells were found in the blister fluid of pa-
tients with TEN, which were reported to kill autologous cells
(lymphocytes and keratinocytes) in a major histocompatibility
complex (MHC) class I-restricted pathway.48 In addition to the
drug-specific CD8 memory cells, natural killer (NK) cells and NK T
cells also contribute to apoptosis of epithelial cells.49e51
Pharmoco-immune concept
With many drugs, a very original observation was that the drug
could be presented to the T-cell receptor (TCR) and activate specific
clone without prior processing by the antigen-presenting cell and
through a noncovalent binding to the MHC or its embedded pep-
tide.46 Because the noncovalent binding is reminiscent of the
pharmacological interaction between a drug and its receptor, the
denomination of pharmoco-immune concept has been proposed.47
This concept is supported by recent findings about direct links of
medications (e.g., CBZ) with both human leukocyte antigen (HLA)
and TCR.52,53
Fas ligand
It was generally agreed that there were too few cytotoxic cells in the
lesions of EN to explain the extent of epidermolysis and that soluble
death mediators were also involved. A succession of cytokines had
been suspected, such as tumor necrosis factor-a (TNF-a), perforin,
granzyme B, but the most popular for more than 10 years has been
Fas ligand.54 Activated keratinocytes express Fas-L and epidermal
sheets of patients with EN-killed Jurkat cells (a lymphocyte cell line
that is exquisitely sensitive to Fas-L-induced apoptosis). Human Ig
inhibited the apoptosis of Jurkat cells. These findings were exag-
gerated as: (1) “collective suicide” of keratinocytes through Fas-Le
Fas interactions (even though no evidence was provided that Fas L
killed keratinocytes), and (2) possible inhibition of apoptosis of
keratinocytes by human Ig (even though evidence points that the
effect was only on Jurkat cells). Further studies have demonstrated
the absence or very low impact of Fas-L on keratinocytes and no
benefit from using high-dose human Ig (intravenous Ig) to treat
patients with EN.55
Regulatory cells
Clinicians noticed that patients developing EN suffered more often
than controls from conditions that impaired immune response,
advanced malignancy, autoimmune diseases, and long-term ther-
apy with corticosteroids.33,56 In Azukizawa’s mouse model, devel-
opment of EN needed prior inactivation of regulatory cells.41
Studies in humans strongly suggested that a deep depression in
 J.-C. Roujeau / Dermatologica Sinica 31 (2013) 169e174
number and functions of regulatory T cells was present at the acute
stage of EN.57 Further studies are definitely needed on this topic.
Granulysin
In 2008, a key paper50 reported the work of the team directed by
W.H. Chung and S.I. Hung demonstrating that granulysin was
actually the main cytokine responsible for the apoptosis of
epithelial cells in EN. Not only was granulysin present in the blisters
at concentrations 100 times more elevated than perforin or gran-
zyme B, but it also killed target cells in vitro and induced blisters
with a necrotic roof when injected in the skin of normal mice at the
concentration observed in blisters. Fas-L was detected in very low
concentrations without any detectable effect. Granulysin was
released by drug-specific cytotoxic CD8 T cells and NK cells. Mac-
rophages often recruited in the lesions are also able to produce
granulysin. These findings have several tremendous consequences:
(1) clearly they demonstrate that the many other cytokines previ-
ously considered as contributing to EN (e.g., TNF-a, Fas-L, TRAIL,
Tweak) have a mild and likely accessory role; and (2) they point to
inhibiting the release and/or the effects of granulysin as the main
goal in an emergency treatment of EN.
HLA associations
A mild association of EN with HLA had been reported more than 25
years ago58 but since 2004 several studies have shown strong and
complete associations of HLA-B*15:02 allele with CBZ-induced
EN29 and of HLA-B*58:01 with allopurinol-induced EN.59 Associ-
ated alleles show very different frequency distribution around the
world being frequent in some parts of Asia and less frequent or rare
in Europe.60 That association appeared both drug specific and
phenotype specific and related to ethnicity. HLA associations are a
field of intense and productive research. A direct link between the
medication and the relevant HLA molecule was demonstrated for
CBZ53 and abacavir61 as suggested years ago for penicillin.62 The
demonstration that patients suffering from EN also have a
restricted use of specific clonotypes of the TCR52 is a major progress
to explain why a minority of persons harboring the HLA-B*15:02
phenotype developed EN when taking CBZ. Furthermore, the
prevalence of CBZ-related EN was largely decreased from Taiwan,
where it had been the primary cause, by testing for HLA-B*15:02
before prescription in recent years.63
The Jackpot hypothesis
The “Jackpot” hypothesis was proposed by Lyell in 1985 during a
discussion about the mechanisms of EN in the first international
symposium on TEN. The disease is so rare that its occurrence needs
the conjunction of several rare risk factors, including intake of
“strongly associated” medication, adequate HLA, and certainly a
few others such as adequate TCR variant and perturbation of reg-
ulatory cells.
The future
Looking back to the history of EN, it appears that in the past decades
the knowledge about EN had improved tremendously, slowly
initially but more rapidly in the last few years. Most of these more
recent advances were provided by a leading Taiwanese team.
With the important exception of HLA testing for CBZ in Asia,
progresses in knowledge have not yet resulted in major benefits for
the victims of what is the most severe form of adverse reaction to
drugs.
173
The existence and development of international networks of
researchers (e.g., RegiSCAR, J-SCAR, Asian-SCAR), active patient
associations, and reference centers for treatment, provide a basis
for new progresses and especially for translating progresses from
laboratories to patient care.
Decreasing the incidence (detection of patients “at risk,” re-
striction of use of “strongly associated” medicines), earlier diag-
nosis and better care (reference centers, specific treatments),
prevention and treatment of sequelae should be priorities for re-
searchers and clinicians. No specific treatment has been proven to
be effective until now. Therefore, increased focus is needed on
agents inhibiting the release and/or blocking the effects of
granulysin.
More implication of pharmaceutical companies and regulatory
agencies in promoting research and helping the victims is definitely
needed to decrease the burden of EN on public health.
References
1. Lyell A. Toxic epidermal necrolysis: an eruption resembling scalding of the skin.
Br J Dermatol 1956;68:355e61.
2. Lyell A. Requiem for toxic epidermal necrolysis. Br J Dermatol 1990;122:837e8.
3. Mellish ME, Glasgow LA. The staphylococcal-scalded skin syndrome: devel-
opment of an experimental model. N Engl J Med 1970;282:1114e9.
4. Arbuthnott JP, Kent J, Lyell A, Gemmell CG. Toxic epidermal necrolysis pro-
duced by an extracellular product of Staphylococcus aureus. Br J Dermatol
1971;85:145e9.
5. Amon RB, Dimond RL. Toxic epidermal necrolysis. Rapid differentiation be-
tween staphylococcal- and drug-induced disease. Arch Dermatol 1975;111:
1433e7.
6. Hervieux E. Pemphigus épidémique des nouveaux-nés. Arch Union Med Balk
Bull nion Med Balk 1868;5:374 [Article in French].
7. von Rittershain GR. Jahresbericht fûr des kôniglichen bömischen Landes-Fin-
delanstalt. Oesteerreiches Jahrb Paediatr 1870;1:1 [Article in German].
8. Maguire GJ. Acute contagious pemphigus in the newly born: a clinical study. Br
J Dermatol 1903;15:427.
9. Kauppinen K. Cutaneous reactions to drugs with special reference to severe
bullous mucocutaneous eruptions and sulphonamides. Acta Derm Venereol
Suppl (Stockh) 1972;68:1e89.
10. Lang R, Walker J. An unusual bullous eruption. S Afr Med J 1956;30:97e8.
11. Debré R, Lamy M, Lamotte R. Un cas d’érythrodermie avec épidermolyse chez
un enfant de douze ans. Bull Soc Pediatr 1939;37:231e8 [Article in French].
12. von Hebra F. Atlas der Hautkrankheiten. Wien: Kaiserliche Akademie der Wis-
senchaften; 1866 [Article in German].
13. Rendu R. Sur un syndrome caractérisé par une l’inflammation simultanée de
toutes les muqueuses externes (conjonctivale, nasale, linguale, buccophar-
yngée, anale et balano-prépuciale) coexistant avec une éruption varicelliforme
puis purpurique des quatre membres. J Prat 1916;30:351e2 [Article in French].
14. Fiessinger N, Rendu R. Sur un syndrome caractérisé par l’inflammation
simultanée de toutes les muqueuses externes coexistant avec une éruption
vésiculeuse des quatre membres, non douloureuse et non récidivante. Paris
Méd 1917;25:54e8 [Article in French].
15. Stevens AM, Johnson FC. A new eruptive fever associated with stomatitis and
ophthalmia. Report of two cases in children. Am J Dis Child 1922;24:526e33.
16. Gädeke R. Erythema exsudativum multiforme major. Pluriorificial erosive
ectodermosis; dermatostomatitis; Fuchs’ acute cutaneo-ocular syndrome;
Stevens-Johnson syndrome. Hippokrates 1970;41:323e36.
17. Huff JC, Weston WL, Tonnesen MG. Erythema multiforme: a critical review
of characteristics, diagnostic criteria and causes. J Am Acad Dermatol 1983;8:
763e75.
18. Bergoend H, Loeffler A, Amar R, Maleville J. Réactions cutanées survenuies au
cours de la prophylaxioe de masse de la méningite cérébrospinale par un sul-
famide long-retard. Ann Dermatol Syphiligr 1968;95:481e90 [Article in French].
19. Lyell A. Toxic epidermal necrolysis (the scalded skin syndrome): a reappraisal.
Br J Dermatol 1979;100:69e86.
20. Böttiger LE, Standberg I, Westerholm B. Drug-induced febrile mucocutaneous
syndrome. Acta Med Scand 1975;198:229e33.
21. Massol R, Mojon M, Bertrand JL, Bertoye A. Le syndrome de Lyell. J Med Lyon
1972;53:1109e22.
22. Ruiz-Maldonado R. Acute disseminated epidermal necrolysis types 1, 2 and 3:
study of sixty cases. J Am Acad Dermatol 1985;13:623e35.
23. Bastuji-Garin S, Rzany B, Stern RS, Shear NH, Naldi L, Roujeau J-C. A clinical
classification of cases of toxic epidermal necrolysis, Stevens-Johnson syndrome
and erythema multiforme. Arch Dermatol 1993;129:92e6.
24. Auquier-Dunant A, Mockenhaupt M, Naldi L, et al. Severe cutaneous adverse
reactions. Correlations between clinical patterns and causes of erythema
multiforme majus, StevenseJohnson syndrome, and toxic epidermal necrol-
ysis: results of an international prospective study. Arch Dermatol 2002;138:
1019e24.
174 J.-C. Roujeau / Dermatologica Sinica 31 (2013) 169e174
25. Chan HL, Stern RS, Arndt KA, et al. The incidence of erythema multiforme,
StevenseJohnson syndrome, and toxic epidermal necrolysis. A population-
based study with particular reference to reactions caused by drugs among
outpatients. Arch Dermatol 1990;126:43e7.
26. Roujeau JC, Guillaume JC, Fabre JP, Penso D, Fléchet ML, Girre JP. Toxic
epidermal necrolysis (Lyell syndrome). Incidence and drug etiology in France,
1981e1985. Arch Dermatol 1990;126:37e42.
27. Schöpf E, Stühmer A, Rzany B, Victor N, Zentgraf R, Kapp JF. Toxic epidermal
necrolysis and Stevens-Johnson syndrome. An epidemiologic study from West
Germany. Arch Dermatol 1991;127:839e42.
28. Sekula P, Dunant A, Mockenhaupt M, et al. RegiSCAR study group. Compre-
hensive survival analysis of a cohort of patients with Stevens-Johnson syn-
drome and toxic epidermal necrolysis. J Invest Dermatol 2013;133:1197e204.
29. Chung WH, Hung SI, Hong HS, Hsih MS, Yang LC, Ho HC, et al. A marker for
Stevens-Johnson syndrome. Nature 2004;248:486.
30. Hernborg H. StevenseJohnson syndrome after mass prophylaxis with sulfa-
doxine for cholera in Mozambique. Lancet 1985;2:1072e3.
31. Roujeau JC, Chosidow O, Saiag P, Guillaume JC. Toxic epidermal necrolysis
(Lyell syndrome). J Am Acad Dermatol 1990;23:1039e58.
32. Roujeau JC, Kelly JP, Naldi L, et al. Medication use and the risk of StevenseJohnson
syndrome or toxic epidermal necrolysis. N Engl J Med 1995;333:1600e7.
33. Mockenhaupt M, Viboud C, Dunant A, et al. StevenseJohnson syndrome and toxic
epidermal necrolysis: assessment of medication risks with emphasis on recently
marketed drugs. The EuroSCAR-study. J Invest Dermatol 2008;128:35e44.
34. Sassolas B, Haddad C, Mockenhaupt M, et al. ALDEN, an algorithm for assess-
ment of drug causality in Stevens-Johnson syndrome and toxic epidermal
necrolysis: comparison with case-control analysis. Clin Pharmacol Ther
2010;88:60e8.
35. Ting W, Stone MS, Racila D, Scofield RH, Sontheimer RD. Toxic epidermal
necrolysis-like acute cutaneous lupus erythematosus and the spectrum of the
acute syndrome of apoptotic pan-epidermolysis (ASAP): a case report, concept
review and proposal for new classification of lupus erythematosus ves-
iculobullous skin lesions. Lupus 2004;13:941e50.
36. Finkelstein Y, Soon GS, Acuna P, et al. Recurrence and outcomes of Stevens-
Johnson syndrome and toxic epidermal necrolysis in children. Pediatrics
2011;128:723e8.
37. Billingham RE, Streilein JW. Toxic epidermal necrolysis and homologous dis-
ease in hamsters. Arch Dermatol 1968;98:528e39.
38. Streilein JW, Billingham RE. An analysis of graft-versus-host disease in Syrian
hamsters. I. The epidermolytic syndrome: description and studies on its pro-
curement. J Exp Med 1970;132:163e80.
39. Peck GL, Elias PM, Graw RG. Graft-vs-host reaction and toxic epidermal nec-
rolysis. Lancet 1972;2:1151e3.
40. Azukizawa H, Kosaka H, Sano S, et al. Induction of T-cell-mediated skin disease
specific for antigen transgenically expressed in keratinocytes. Eur J Immunol
2003;33:1879e88.
41. Azukizawa H, Sano S, Kosaka H, Sumikawa Y, Itami S. Prevention of toxic
epidermal necrolysis by regulatory T cells. Eur J Immunol 2005;35:1722e30.
42. Roujeau JC, Dubertret L, Moritz S, et al. Involvement of macrophages in the
pathology of toxic epidermal necrolysis. Br J Dermatol 1985;113:425e30.
43. Correia O, Delgado L, Ramos JP, Resende C, Torrinha JA. Cutaneous T-cell
recruitment in toxic epidermal necrolysis. Further evidence of CD8þ lympho-
cyte involvement. Arch Dermatol 1993;129:466e8.
44. Mauri-Hellweg D, Bettens F, Mauri D, Brander C, Hunziker T, Pichler WJ.
Activation of drug-specific CD4þ and CD8þ T cells in individuals allergic to
sulfonamides, phenytoin, and carbamazepine. J Immunol 1995;155:462e72.
45. Brander C, Mauri-Hellweg D, Bettens F, Rolli H, Goldman M, Pichler WJ. Het-
erogeneous T cell responses to beta-lactam-modified self-structures are
observed in penicillin-allergic individuals. J Immunol 1995;155:2670.
46. Schnyder B, Mauri-Hellweg D, Zanni M, Bettens F, Pichler WJ. Direct, MHC-
dependent presentation of the drug sulfamethoxazole to human alphabeta T
cell clones. J Clin Invest 1997;100:136e41.
47. Pichler WJ. Pharmacological interaction of drugs with antigen-specific immune
receptors: the p-i concept. Curr Opin Allergy Clin Immunol 2002;2:301e5.
48. Nassif A, Bensussan A, Dorothée G, et al. Drug specific cytotoxic T-cells in the
skin lesions of a patient with toxic epidermal necrolysis. J Invest Dermatol
2002;118:728e33.
49. Le Cleach L, Delaire S, Boumsell L, et al. Blister fluid T lymphocytes during
toxic epidermal necrolysis are functional cytotoxic cells which express
human natural killer (NK) inhibitory receptors. Clin Exp Immunol 2000;119:
225e30.
50. Chung WH, Hung SI, Yang JY, et al. Granulysin is a key mediator for dissemi-
nated keratinocyte death in Stevens-Johnson syndrome and toxic epidermal
necrolysis. Nat Med 2008;14:1343e50.
51. Morel E, Escamochero S, Cabañas R, Díaz R, Fiandor A, Bellón T. CD94/NKG2C is
a killer effector molecule in patients with StevenseJohnson syndrome and
toxic epidermal necrolysis. J Allergy Clin Immunol 2010;12:703e10.
52. Ko TM, Chung WH, Wei CY, et al. Shared and restricted T-cell receptor use is
crucial for carbamazepine-induced StevenseJohnson syndrome. J Allergy Clin
Immunol 2011;128:1266e76.
53. Wei CY, Chung WH, Huang HW, Chen YT, Hung SI. Direct interaction between
HLA-B and carbamazepine activates T cells in patients with Stevens-Johnson
syndrome. J Allergy Clin Immunol 2012;129:1562e9.
54. Viard I, Wehrli P, Bullani R, et al. Inhibition of toxic epidermal necrolysis by
blockade of CD95 with human intravenous immunoglobulin. Science 1998;282:
490e3.
55. Huang YC, Li YC, Chen TJ. The efficacy of intravenous immunoglobulin for the
treatment of toxic epidermal necrolysis: a systematic review and meta-anal-
ysis. Br J Dermatol 2012;167:424e32.
56. Lee HY, Dunant A, Sekula P, et al. The role of prior corticosteroid use on the
clinical course of StevenseJohnson syndrome and toxic epidermal necrolysis: a
case-control analysis of patients selected from the multinational EuroSCAR and
RegiSCAR studies. Br J Dermatol 2012;167:555e62.
57. Takahashi R, Kano Y, Yamazaki Y, Kimishima M, Mizukawa Y, Shiohara T.
Defective regulatory T cells in patients with severe drug eruptions: timing of
the dysfunction is associated with the pathological phenotype and outcome.
J Immunol 2009;182:8071e9.
58. Roujeau JC, Bracq C, Huyn NT, Chaussalet E, Raffin C, Duédari N. HLA
phenotypes and bullous cutaneous reactions to drugs. Tissue Antigens 1986;28:
251e4.
59. Hung SI, Chung WH, Liou LB, et al. HLA-B*5801 allele as a genetic marker for
severe cutaneous adverse reactions caused by allopurinol. Proc Natl Acad Sci U S
A 2005;102:4134e9.
60. Genin E, Schumacher M, Roujeau JC, et al. Genome-wide association study of
Stevens-Johnson syndrome and toxic epidermal necrolysis in Europe. Orphanet
J Rare Dis 2011;29:52.
61. Ostrov DA, Grant BJ, Pompeu YA, et al. Drug hypersensitivity caused by alter-
ation of the MHC-presented self-peptide repertoire. Proc Natl Acad Sci U S A
2012;109:9959e64.
62. Claas FH, Runia-van Nieuwkoop R, van den Berge W, van Rood JJ. Interaction of
penicillin with HLA-A and -B antigens. Hum Immunol 1982;5:83e90.
63. Chen P, Lin JJ, Lu CS, et al. Taiwan SJS Consortium. Carbamazepine-induced
toxic effects and HLA-B*1502 screening in Taiwan. N Engl J Med 2011;364:
1126e33.
64. Roujeau JC, Revuz J. Intensive care in dermatology. In: Champion RH, Pye RJ,
editors. Recent advances in dermatology. Edinburgh: Churchill Livingstone;
1990. p. 85e99.
65. Wolkenstein P, Latarjet J, Roujeau JC, et al. Randomised comparison of
thalidomide versus placebo in toxic epidermal necrolysis. Lancet 1998;352:
1586e9.
66. Bastuji-Garin S, Fouchard N, Bertocchi M, Roujeau JC, Revuz J, Wolkenstein P.
SCORTEN: a severity-of-illness score for toxic epidermal necrolysis. J Invest
Dermatol 2000;115:149e53.