Guidelines for performing skin tests with drugs in the investigation of cutaneous adverse drug reactions.
ABSTRACT Skin testing with a suspected drug has been reported to be helpful in determining the cause of cutaneous adverse drug reactions (CADR). Many isolated reports of positive drug skin tests are published, but without detailed information concerning the clinical features of the CADR and the method used in performing drug skin tests, such data are not very informative. A working party of the European Society of Contact Dermatitis (ESCD) for the study of skin testing in investigating cutaneous adverse drug reactions, has proposed the herein-reported guidelines for performing skin testing in CADR in order to standardize these procedures. In each reported case, the imputability of each drug taken at the onset of the CADR and a highly detailed description and characterization of the dermatitis need to be given. Drug skin tests are performed 6 weeks to 6 months after complete healing of the CADR. Drug patch tests are performed according to the methods used in patch testing in studying contact dermatitis. The commercialized form of the drug used by the patient is tested diluted at 30% pet. (pet.) and/or water (aq.). The pure drug is tested diluted at 10% in pet. or aq. In severe CADR, drug patch tests are performed at lower concentrations. It is also of value to test on the most affected site of the initial CADR. Drug prick tests are performed on the volar forearm skin with the commercialized form of the drug, but with sequential dilutions in cases of urticaria. Intradermal tests (IDT) are performed with sterile sequential dilutions (10-4, 10-3, 10-2, 10-1) of a pure sterile or an injectable form of the suspected drug with a small volume of 0.04 ml. Drug skin tests need to be read at 20 min and also later at D2 and D4 for patch tests, at D1 for prick tests and IDT. All these tests also need to be read at 1 week. The success of skin tests varies with the drug tested, with a high % of positive results, for example, with betalactam antibiotics, pristinamycin, carbamazepine and tetrazepam on patch testing, or with betalactam antibiotics and heparins on delayed readings of IDT. The results of drug skin tests also depend on the clinical features of the CADR. The use of appropriate control patients is necessary to avoid false-positive results.
- [show abstract] [hide abstract]
ABSTRACT: Skin testing with the suspected compound has been reported to be helpful in determining the cause of cutaneous adverse drug reactions (ADRs), but the value and specificity of these tests need to be determined. In this study, 72 patients with presumed drug eruptions (27 maculopapular, 18 urticarial, seven erythrodermic, nine eczematous, four photosensitivity, three fixed drug eruptions, three with pruritus and one with acute generalized exanthematous pustulosis) were assessed. All had drug patch tests; 46 also had prick tests and 30 had intradermal tests (performed on hospitalized patients using a sterile solution of the suspected drug, diluted sequentially) with immediate and delayed readings. Among these patients, 52 (72%) had a positive skin test reaction, 43%, 24% and 67% in patch, prick and intradermal skin tests, respectively. The results of skin tests varied with the drug tested and with the clinical type of cutaneous ADR, as a significantly higher number of positive patch tests was observed in maculopapular rashes than in urticarial reactions (P = 0.001). This study supports the value of careful sequential drug skin testing in establishing the cause of cutaneous ADR. Guidelines are proposed for performing these tests, and these include the use of appropriate negative control patients to avoid false-positive results.British Journal of Dermatology 08/1998; 139(1):49-58. · 3.76 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Skin test reactions were evaluated in 242 patients who appeared to develop delayed type drug eruptions from the clinical course. The patch testing was positive in 62 (31.5%) of 197 patients tested and the intradermal testing in 105 (89.7%) of 117 patients. The positive ratios of intradermal testing were higher in maculopapular (MP), erythema multiforme (EM), and erythrodermic (ED) types than in eczematous (Ecz) type drug eruptions, while those of patch testing were comparatively high in ED, Ecz type, and anticonvulsant-induced drug eruptions. It is considered that the combination of patch testing and intradermal testing is useful for determination of causative drugs in delayed type drug eruptions.The Journal of Dermatology 05/1990; 17(4):235-9. · 1.77 Impact Factor
- Clinics in Dermatology 01/1997; 15(4):479-84. · 2.34 Impact Factor
Contact Dermatitis, 2001, 45, 321–328
Printed in Denmark . All rights reserved
Copyright C Munksgaard 2001
Proposed by the Working party of the ESCD for the study of skin testing in investigating
cutaneous adverse drug reactions
A B1, M G ¸2, D B3 A B4
1Dermatology Department, Hopital Fournier, 54000 Nancy, France
2Dermatology Department, Hospital da Universidade, 3000-075 Coimbra, Portugal
3Department of Dermatology, Free University Hospital, De Boelelaan 1117, 1081 HV Amsterdam,
4Allergologische Poliklinik, Kantonsspital, Basel, Switzerland
Skin testing with a suspected drug has been reported to be helpful in determining the cause of
cutaneous adverse drug reactions (CADR). Many isolated reports of positive drug skin tests are
published, but without detailed information concerning the clinical features of the CADR and the
method used in performing drug skin tests, such data are not very informative. A working party
of the European Society of Contact Dermatitis (ESCD) for the study of skin testing in investigating
cutaneous adverse drug reactions, has proposed the herein-reported guidelines for performing skin
testing in CADR in order to standardize these procedures. In each reported case, the imputability
of each drug taken at the onset of the CADR and a highly detailed description and characteriza-
tion of the dermatitis need to be given. Drug skin tests are performed 6 weeks to 6 months after
complete healing of the CADR. Drug patch tests are performed according to the methods used in
patch testing in studying contact dermatitis. The commercialized form of the drug used by the
patient is tested diluted at 30% pet. (pet.) and/or water (aq.). The pure drug is tested diluted at
10% in pet. or aq. In severe CADR, drug patch tests are performed at lower concentrations. It is
also of value to test on the most affected site of the initial CADR. Drug prick tests are performed
on the volar forearm skin with the commercialized form of the drug, but with sequential dilutions
in cases of urticaria. Intradermal tests (IDT) are performed with sterile sequential dilutions (10–
4, 10–3, 10–2, 10–1) of a pure sterile or an injectable form of the suspected drug with a small
volume of 0.04 ml. Drug skin tests need to be read at 20 min and also later at D2 and D4 for
patch tests, at D1 for prick tests and IDT. All these tests also need to be read at 1 week. The
success of skin tests varies with the drug tested, with a high % of positive results, for example,
with betalactam antibiotics, pristinamycin, carbamazepine and tetrazepam on patch testing, or
with betalactam antibiotics and heparins on delayed readings of IDT. The results of drug skin tests
also depend on the clinical features of the CADR. The use of appropriate control patients is
necessary to avoid false-positive results.
Key words: cutaneous adverse drug reactions; patch testing; prick testing; intradermal testing;
guidelines. C Munksgaard, 2001.
Accepted for publication 29 August 2001
322 BARBAUD ET AL.
Cutaneous adverse drug reactions (CADRs) are a
frequent problem in dermatology, especially in
cases which concern drug imputability, because pa-
tients are often on multiple drug regimes. Besides
clinical and chronological parameters, there is no
widely usable complementary test to help estab-
lishing the definite cause of the CADR.
Skin testing (patch testing and also prick and
intradermal (IDT) testing), with the suspected
compound, has been reported to be helpful in de-
termining the cause of a CADR (1–5) and in
studying the pathophysiological mechanisms in-
volved in these reactions. According to previous
data, the results of drug skin tests mainly depend
on the drug tested and the clinical features of the
initial CADR (1, 2, 4–6), but there are, at present,
a few extensive studies that determine the sensi-
tivity and specificity of these drug skin tests as a
complementary tool for drug imputability in
CADR. Too many isolated reports of positive drug
skin tests in investigating CADR are published,
but without detailed information concerning the
clinical features of the CADR, the imputability of
the suspected drug, the methods used in perform-
ing drug skin tests, namely the concentrations and
vehicles used for testing the suspected drugs, this
data is not always very informative. Moreover,
there are undoubtedly many negative results that
are not published, which make it still more difficult
to ascertain the sensitivity and specificity of such
tests in the study of CADR.
Therefore, the aim of a working party of the
ESCD for the study of skin testing in investigating
cutaneous adverse drug reactions was to promote
a prospective multicentre study devoted to the use-
fulness of skin testing in CADRs and, also, to pro-
pose guidelines for performing skin testing in
CADRs, in order to help standardize this pro-
cedure in the investigation of CADR and to assess
their sensitivity and specificity, namely which con-
cern the suspected drug and the clinico-evolutive
features of the CADR.
Guidelines in Skin Testing
In order to validate the results collected individu-
ally in different centres, skin tests should be per-
formed according to common guidelines, such as
those proposed. Furthermore, it is highly recom-
mended that several aspects are considered in each
case, namely concerning data on the patient, on
the drugs taken, drug imputability score, clinico-
evolutive characterization of the CADR and,
eventually, additional data that can be useful in
establishing a more strictly defined diagnosis of the
Besides age, sex and accessory diseases, personal
or family history of adverse drug reactions, espe-
cially with the suspected drug or chemically related
drugs, have to be noted.
All the drugs taken during the onset of the CADR,
even if they have been prescribed for many
months, have to be listed. Imputability for each
drug taken before or during the onset of the skin
reaction has to be determined according to criteria
proposed by Moore et al. (7).
Clearly, it is of great value to note whether the
patient has had previous exposure to the suspected
drug and to know whether he has had at least 1
other episode of similar CADR after taking the
culprit drug (accidental rechallenge) or if a drug
provocation test has been performed. Other test
results, namely positive or negative in vitro tests,
such as RAST or a lymphocyte transformation
test, should also be reported, as they may modify
the imputability score.
Concerning the suspected drug, the history of
drug intake must be reported with the dates the
treatment was begun and stopped, the interval be-
tween the beginning of drug intake and the onset
of the CADR, the mode of administration, the
prescribed dosage, the common international no-
menclature of the drug and the disease that it was
prescribed for. Concurrently, the evolution of the
CADR must be given.
Characterization of the CADR
Clinical and evolutive features of the cutaneous ad-
verse reaction need to be clearly described. Ad-
ditionally, when possible, the CADR should be
classified according to the following: pruritus, urti-
caria and/or angioedema, anaphylactoid or ana-
phylactic shock, maculopapular and/or purpuric
rash, erythroderma (exfoliative dermatitis), hyper-
sensitivity syndrome or drug rash with eosino-
philia and systemic syndrome (DRESS), pseudo-
lymphoma, generalized or localized eczema, ba-
boon syndrome, systemically-induced
dermatitis, flexural cutaneous adverse drug reac-
tion, acute generalized exanthematic pustulosis
(AGEP), purpura, leucocytoclastic vasculitis, li-
chenoid dermatosis, erythema multiforme, Stev-
ens-Johnson syndrome, toxic epidermal necrolysis
(Lyell’s syndrome), fixed drug eruptions (FDE) or
eczematous photosensitivity reactions.
323 SKIN TESTS IN CADR
Facultative additional data
To improve the diagnosis, the pathophysiological
mechanism or to determine whether there is as-
sociated organ dysfunction, the following labora-
tory investigations should be carried out during
the CADR: full blood count, platelet count, sedi-
mentation rate, serum creatinine, transaminases,
alkaline phosphatase, bilirubin, eosinophil cationic
protein and tryptase in case of anaphylactoid reac-
tion. Histological examination of a skin biopsy
performed at the onset of the cutaneous adverse
drug reaction can be of value in classifying the
clinical and pathophysiological type of CADR.
In particular types of CADR, serological tests
for viral infection that may interfere with the onset
of the drug eruptions are welcomed, particularly
tests for cytomegalovirus, Epstein-Barr virus and
parvovirus B19 for patients who have developed a
maculopapular rash, hepatitis B and C for patients
with urticaria, Herpes simplex virus hominis type
I and II for patients with erythema multiforme,
and tests for mycoplasma for patients with Stev-
Guidelines in Drug Skin Testing
After informed patient consent, drug skin tests
should be performed 6 weeks to 6 months after
complete healing of the CADR, at least 1 month
after discontinuation of systemic corticosteroid or
immunosuppressive therapy. It is preferable, if
possible, not to test during pregnancy.
Skin testing should be performed with the com-
mercialized drug and, whenever possible, also with
the pure active products and excipients. Testing
with drugs with a similar chemical structure, or
from the same pharmacological family, may also
be of importance in understanding cross-reactions
In order to avoid adverse effects, namely severe
immediate reactions, skin testing should be per-
formed according to the following sequence: patch
tests with an immediate reading at 20 min, then
prick tests and, if negative, delayed readings are
recommended in all such procedures.
Drug Patch Testing
Patch tests should be performed on the upper back
using Finn ChambersAon ScanporAtape (Epitest,
Tuusula, Finland), Van der Bend Square Chambers
(Chemotechnique), according to the methods used
in patch testing for contact dermatitis.
Patch test reactions need to be read at 20 min,
day (D)2 and D4, or, if this is not possible, on D3.
Whenever possible, if the patch tests are negative
on D4, a reading should be performed on D7. Re-
sults of patch testing should be reported according
to the International Contact Dermatitis Research
Group (ICDRG) criteria (9).
In investigating a photosensitivity reaction in-
duced by a drug, both drug patch tests and drug
photopatch tests with the responsible drug need to
be performed. Irradiation for drug photopatch
tests should be performed on D1, or for practical
reasons on D2, with 5 J/cm UVA irradiation (10).
In fixed drug eruptions (FDE), patch tests
should be performed both on the normal skin of
the back and on the residual pigmented site of the
Concentrations and vehicles
Patch testing with the commercialized drug. The
drug possibly responsible for the CADR can be
tested with the commercialized form used by the
patient. Pills should have their coating removed,
then be ground to a very fine powder. This powder
needs to be tested as is (facultative) and also incor-
porated at 30% in white petrolatum (pet.) and di-
luted at 30% in water (aq.), in highly controlled
The powder contained in capsules needs to be
tested as is, diluted at 30% in pet. and/or at 30%
in aq. The gel jacket portion of the capsules should
be moistened and tested as is. Liquid preparations
need to be tested both as is and diluted at 30% in
aq. With commercialized forms of the drugs, each
preparation is made for only 1 patient and kept no
more than 1 D. The name of the chemical form
of the molecule (salt, molecule base) needs to be
Patch testing with pure substances. Whenever
possible, the pure drug obtained from the manu-
facturer should be tested diluted at 10% in pet.
and, if possible also, at 10% in aq. or alcohol (alc.).
Concentrations and vehicles previously considered
as most adequate for certain drugs should also be
chosen. For instance, for patch testing with estro-
gens and progesterone, we recommend, besides
testing them diluted in aq. and pet., dilutions also
in alc. (14).
To avoid any relapse of a severe CADR, in pa-
tients who have developed DRESS, Stevens-John-
son syndrome, Lyell’s syndrome or in testing with
aciclovir, carbamazepine or pseudoephedrine, we
recommend that patch tests are performed, either
with the commercialized form of the drug or the
pure substance, first diluted at 0.1% and, if nega-
tive, at higher concentrations of 1% up to 10%.
324 BARBAUD ET AL.
agents and excipients should also be tested, as is
or diluted at 10% in pet., or in the vehicles and
concentrations usually proposed for testing in al-
lergic contact dermatitis.
Drug prick tests should be performed on the volar
forearm skin, with the commercialized form of the
drug thought to be responsible for inducing the
CADR. Whenever possible, both the pure drug
and excipients should be tested as is (in the form
used for patch testing).
When testing patients who have developed urti-
caria, the drug needs to be tested with the same
sequential solutions as those used for IDR at 10ª3,
10ª2, 10ª1, then pure. Reactions are considered
positive when a weal with a diameter larger than 3
mm and than that of the negative control (0.9%
saline) is present 20 min later. Positive controls are
performed with codeine phosphate at 9% and/or
with histamine (10 mg/ml). In all cases, a delayed
reading needs to be done 1 D after performing
Intradermal Tests (IDT)
Intradermal tests are performed only when prick
tests give negative results 20 min after testing with
the suspected drug. Intradermal tests are contra-
indicated in patients who have developed erythema
multiforme, Stevens-Johnson syndrome, toxic epi-
dermal necrolysis or leucocytoclastic vasculitis on
The following guidelines should be applied for
(i) Dilutions are prepared, under laminar flow,
no longer than 2 h preceding administration.
(ii) All necessary precautions have to be taken
in case emergency treatment should be needed. It
is highly recommended, especially in investigating
anaphylactoid reactions or urticaria, that patients
should be tested under hospital surveillance for a
minimum of 6 h following IDT. Intravenous glu-
cose solution should be given for at least 2 h fol-
lowing the last IDT, when the investigation results
remain negative, and for at least 6 h if the IDT is
positive. Blood pressure, pulse and peak-flow
should be regularly monitored.
(iii) Intradermal tests are performed using a
sterile solution of the suspected drug, diluted se-
quentially (10ª4, 10ª3, 10ª2and 10ª1) in phenol-
ated saline (0.5% phenol in 0.9% saline) or in 0.9%
(iv) Intradermal tests are performed on the ex-
tensor surface of the arm, with a small volume
(0.04 ml) that produces a weal of 4 to 6 mm in
diameter. A negative control is performed with
phenolated saline or saline. When read after 30
min, IDT results are considered positive if a weal
of more than 10 mm in diameter is observed. The
first IDT is performed with a 10ª4dilution. If re-
sults are negative after 30 min, subsequent IDT are
performed with increasingly higher serial dilutions
up to the pure solution, at 30-min intervals, as long
as the test results of the previous dilution remain
(v) Intradermal test result readings are per-
formed at 30 min, 6 h and 1 D. If a delayed positive
reaction is observed, the diameter of the papule is
measured. When the IDT results are negative, a
telephone call to the patient, 1 week later, con-
firming whether the results remain negative or not,
or a delayed reading at 1 week, should be per-
Negative Control Patients for Drug Skin Testing
The most informative negative controls are pa-
tients who have presented with a dermatitis, 6
weeks to 6 months previously, after taking sev-
eral drugs, including 1 of those implicated in the
study. They are considered negative controls if
they fail to display reactivity to 1 of these drugs,
while another drug is confirmed to be respon-
sible for the CADR with which they have pre-
sented; e.g., if a patient having taken amoxicillin
CADR due to amoxicillin but not to paracet-
amol, if he has had skin tests with paracetamol
and if he has taken paracetamol orally again
without any CADR, he/she is enrolled as a nega-
tive control for paracetamol (1).
Healthy volunteers with or without previous ex-
posure to the drug can be used as negative con-
trols, but this requires signed informed consent of
such control patients, with approval of an ethical
Following the guidelines reported herein, drug skin
tests are of value in investigating CADR as posi-
tive relevant results were obtained in 43%, 24% and
67% of (photo)patch, prick and IDT drug skin
tests, respectively, among 72 patients (1). Among
these 3 different skin test procedures, patch testing
is the one most extensively studied in CADRs. In
patients with a high imputability of one drug in the
onset of their CADR, drug patch tests had positive
results in 43% of 72 patients (1), 50% of 108 pa-
tients (2), 43.9% of 66 patients (5), and 31.7% of
197 patients (3). Positive relevant results depend on
the clinico-evolutive type of CADR, on the respon-
325 SKIN TESTS IN CADR
sible drug, on the drug concentration and vehicle
used (1, 2, 5), and even on the skin sites where tests
are performed (8, 11, 12, 15–17).
Due to the possibility that a low concentration
might yield false-negative results, drug patch tests
need to performed with high concentrations of the
commercialized form of drug. After pulverizing a
tablet, the highest concentration to obtain homo-
geneous dilution of the powder is 30% in pet., in
aq. or in alc. This threshold was determined in
Nancy (France) by V. Noirez (unpublished data).
When testing with pure substances, as the thresh-
old of sensitivity for each drug is not yet deter-
mined, we advise a 10% concentration in pet., and
if necessary in other vehicles, although for some
drugs lower concentrations may be sufficient.
Nevertheless, testing with high concentrations may
elicit a relapse of the initial CADR. This is a rea-
son why patch tests need to be performed first with
low concentrations in investigating DRESS, Stev-
ens-Johnson syndrome or Lyell’s syndrome (1). As
we have observed (2) some differences in testing
with amoxicillin base or trihydrate amoxicillin salt,
the name of the chemical form of the tested drug
needs also to be given (2).
We propose the use of various vehicles, at least
pet. and aq., for preparing drug patch tests, so as
to avoid false-negative results due to poor penetra-
tion of the drug into the epidermis. In 5 cases, we
have observed false-negative results when anti-
biotics were diluted in aq., whereas patch tests with
the same drug diluted in pet. yielded positive re-
sults (2). Gonc ¸alo et al. (14) observed false-positive
results when testing estrogens diluted in aq. or pet.,
but obtained positive results when steroid hormons
were diluted in alc. Alcohol has been reported as a
good vehicle for patch testing in investigating FDE
(19). Some drugs frequently induce a relapse of the
CADR during patch testing, such as aciclovir (1),
carbamazepine (20) or pseudoephedrin (1, 21, 22).
These drugs need to be tested at low concen-
trations, and if they give negative results, at higher
concentrations up to 30% if necessary.
Whenever possible, it may also be of value to
test on the most-affected site of the initial CADR.
In fixed drug eruptions, patch tests (11) or re-
peated application tests (12) with the suspected
drug are positive only when performed on residual
pigmented skin sites of the CADR, rather than
when applied on non-previously affected skin of
the back. Also, in 1 case of toxic necrolysis, Klein
et al. (18) obtained positive patch tests when co-
trimoxazole was tested on the cutaneous sites pre-
viously affected by necrolysis, while drug patch
tests performed on other less-affected skin sites re-
mained negative. It may also be of value to test on
the most highly affected skin sites in maculopapul-
ar rashes, as reported in 1 case with tetrazepam
(15). As a CADR, especially urticaria, may be due
to coloring agents and excipients, these compo-
nents also need to be tested (23).
As drug patch tests can elicit immediate positive
reactions, especially with betalactam antibiotics,
these tests need to be read at 20 min. Because most
CADR are related to delayed cellular hypersensit-
ivity, it is absolutely essential to carry out delayed
readings at D2, also at D4 and, if negative, on D7.
We have observed negative drug patch tests when
read on D2, but positive on D4, in 5 patients (2).
The usefulness of drug patch tests depends on
the clinical features of the CADR. From our re-
sults (1, 2, 5, 6) and those obtained from the litera-
ture, we consider that patch tests are of value in
determining the responsible drug in generalized ec-
zema, systemic contact dermatitis, baboon syn-
drome, maculopapular rash (1, 2, 5, 6, 24), AGEP
(25), lichenoid rash and fixed drug eruption (11),
and that photopatch tests may be useful in study-
ing drug photosensitivity (10, 26). On the other
hand, they are of less value in investigating urti-
caria (1), Stevens-Johnson or Lyell’s syndrome
(25), pruritus or vasculitis (2).
The usefulness of drug patch tests also depends
on the drug tested. The most frequent reports of
positive drug patch tests are related to the follow-
ing drugs: betalactam antibiotics, especially amoxi-
cillin (1, 2, 4, 6, 18, 24–30), cotrimoxazole (15),
corticosteroids (1, 31, 32), heparin derivatives (1,
33–36), pristinamycin (1, 36), carbamazepine (1, 2,
20, 37–44), diltiazem (1, 45, 46), diazepam (47),
hydroxyzine (1, 48), pseudoephedrine (1, 21, 22)
and tetrazepam (1, 16, 49–51).
Prick tests and IDT
The usefulness of prick tests and IDT in determin-
ing the cause of CADR has not been evaluated, as
much as the usefulness of drug patch tests. Osawa
et al. (3) obtained positive results in 89.7% of their
patients, but in the absence of negative controls,
the specificity of their IDT cannot be validated. We
obtained positive results on prick testing in 24% of
46 cases and in 64% of the 30 patients undergoing
Such drug skin tests are read at 20 or 30 min,
but delayed readings are also worthwhile doing
after 1 D (24) or later. Immediate positive results
are observed most frequently in patients who have
previously developed urticaria or angioedema.
Delayed positive results can be observed even with
prick tests, as reported with amoxicillin (1, 24) and
pseudoephedrine (1). Delayed positive results are
frequent with IDT, especially in investigating mac-
ulopapular rashes due to amoxicillin (1, 24, 29). In
326 BARBAUD ET AL.
testing heparin derivatives, it is of value to perform
more delayed readings as results can be positive
only at D3 (1, 35, 36). With IDT, delayed positive
results can be obtained in maculopapular rash, ec-
zema, erythroderma or fixed drug eruption.
Most of the papers published on the usefulness
of prick tests and IDT concern CADR due to be-
talactam antibiotics (1, 24, 29, 52–54). Prick tests
and IDT have also been performed with other
drugs such as diltiazem (45), heparin derivatives
(33–35), insulin (45), oestradiol (1) and cortico-
Drug skin tests can re-elicit the initial CADR.
There is a risk of eliciting a relapse of the initial
CADR especially when performing IDT. Among
30 IDT performed, 3 minor incidents were ob-
served (1). Because of this risk, we suggest hospital
surveillance while performing IDT. The volume of
drug administered for IDT should be as low as
possible, 0.04 ml being the safest quantity and the
lowest volume that gives reproducible responses
Interpretation of Results
Relevance and specificity of drug skin tests
A crucial point is the interpretation of results of
drug skin tests, whether negative or positive. Even
when performed according to reviewed guidelines,
skin testing is negative in 30–50% of patients (1–3,
5). Negative results may have several explanations:
the final responsible agent for the CADR is a drug
metabolite that is not formed in the skin when the
native drug is applied there; there is no immune
mechanism involved in the CADR; or concomitant
factors that are responsible in inducing a transient
oral drug intolerance, such as a viral infection, are
not present at the time of testing. Thus, if negative,
a drug skin test does not exclude the responsibility
of a drug.
On the other hand, if the test is positive, we need
to determine its specificity and its relevance. We
have observed many false-positive reactions when
performing IDT (57). We have already determined
the thresholds of specificity for many drugs, for
example betalactam antibiotics, erythromycin and
spiramycin (1, 57), but the thresholds of specificity
still need to be determined for many drugs in IDT
False-positive results have also been observed on
drug patch testing, namely with commercialized
forms of drugs containing sodium lauryl sulfate in
their formulations, with colchicine diluted at 10%
in pet., with pills containing misoprostol when di-
luted at 30% in pet. (59).
This emphasizes that it is necessary to compare
all skin test results with those obtained in negative
controls (58), namely patients who have taken the
drug within the last 6 months without adverse ef-
fects, or even patients who have never had contact
with the culprit drug. To collect such negative con-
trols and to enlarge studies that may help to deter-
mine the sensitivity and specificity of drug skin
tests in investigating CADR, it is necessary to or-
ganize multicentre studies using the same guide-
Another point to be kept in mind, when as-
sessing the relevance of drug patch tests, is that a
positive drug patch test may have past relevance to
contact dermatitis due to a drug, or to an excipi-
ent, without any relevance to the present CADR
1. Barbaud A, Reichert-Penetrat S, Trechot P, Jacquin-Petit
M A, Ehlinger A, Noirez V, Faure G C, Schmutz J L, Bene
M C. The use of kin testing in the investigation of cu-
taneous adverse drug reactions. Br J Dermatol 1998: 139:
2. Barbaud A, Bene M C, Faure G, Schmutz J L. Tests cutan-
e ´s dans l’exploration des toxidermies suppose ´es de me ´can-
isme immuno-allergique. Bull Acad Natle Med 2000: 184:
3. Osawa J, Naito S, Aihara M, Kitamura K, Ikezawa Z,
Nakajima H. Evaluation of skin test reactions in patients
with non-immediate type drug eruptions. J Dermatol 1990:
4. Bruynzeel D P, Van Ketel W G. Patch testing in drug erup-
tions. Sem Dermatol 1989: 8: 196–203.
5. Gonc ¸alo M, Fernandes B, Oliveira H S, Figueiredo A. Ep-
icutaneous patch testing in drug eruptions. Contact Derma-
titis 2000: 42: s22.
6. Bruynzeel D P, Van Ketel W G. Skin tests in the diagnosis
of maculopapular drug eruptions. Sem Dermatol 1987: 6:
7. Moore N, Paux G, Begaud B, Biour M, Loupi E, Boismare
F, Royer R J. Adverse drug reaction monitoring: doing it
the french way. Lancet 1985: ii: 1056–1058.
8. Oliveira H S, Gonc ¸alo M, Reis J P, Figueiredo A. Fixed
drug eruption to piroxicam. Positive patch tests with cross-
sensitivity to tenoxicam. J Dermatol Treatment 1999: 10:
9. Wilkinson D S, Fregert S, Magnusson B, Bandmann H J,
Calnan C D, Cronin E, Hjorth N, Maibach H J, Malten K
E, Meneghini C L, Pirilä V. Terminology of contact derma-
titis. Acta Dermato-venereologica 1970: 50: 287–292.
10. Gonc ¸alo M. Exploration dans les photo-allergies me ´dica-
menteuses. In: Progre `s en dermato-allergologie. Nancy:
John Libbey Eurotext (eds.) 1998: 67–74.
11. Alanko K, Stubb S, Reitamo S. Topical provocation of fix-
ed drug eruption. Br J Dermatol 1987: 116: 561–567.
12. Alanko K. Topical provocation of fixed drug eruption: a
study of 30 patients. Contact Dermatitis 1994: 31: 25–27.
13. Oliveira H S, Robalo-Cordeiro M, Fernandes B, Gonc ¸alo
M, Figueiredo A. Topical provocation in 9 cases of fixed
drug eruption due to non-steroidal anti-inflammatory
drugs. Contact Dermatitis 2000: 42: s30.
14. Gonc ¸alo M, Oliveira H S, Monteiro C, Clerins I, Figueire-
do A. Allergic and systemic contact dermatitis from estra-
diol. Contact Dermatitis 1999: 40: 58–59.
15. Barbaud A, Trechot P, Reichert-Penetrat S, Granel F,
Schmutz J L. Drug patch testing: the usefulness in testing
327 SKIN TESTS IN CADR
on the most previously affected site in a systemic cutaneous
adverse drug reaction to tetrazepam. Contact Dermatitis
2001: 44: 259–260.
16. Robalo-Cordeiro M, Gonc ¸alo M, Fernandes B, Oliveira H
S, Figueiredo A. Positive lesional patch tests in fixed drug
eruptions from nimesulide. Contact Dermatitis 2000: 43:
17. Barbaud A, Tre ´chot P, Reichert-Pe ´ne ´trat S, Granel F,
Schmutz J L. The usefulness of patch testing on the pre-
viously most severely affected site in a cutaneous adverse
drug reaction to tetrazepam. Contact Dermatitis 2001: 44:
18. Klein C E, Trautmann A, Zillikens D, Brocker E B. Patch
testing in an unusual case of toxic epidermal necrolysis.
Contact Dermatitis 1995: 33: 448–449.
19. Kaupinnen K, Alanko K, Hannuksela M, Maibach H.
Skin reactions to drugs. Boca Raton: CRC Press, 1998.
20. Vaillant L, Camenen I, Lorette G. Patch testing with carba-
mazepine: reinduction of an exfoliative dermatitis. Arch
Dermatol 1989: 125: 299.
21. Tomb R R, Lepoittevin J P, Espinassouze F, Heid E, Fous-
sereau J. Systemic contact dermatitis from pseudoeph-
edrine. Contact Dermatitis 1991: 24: 86–88.
22. Sanchez T S, Sanchez-Perez J, Aragues M, Garcia-Diaz A.
Flare-up reaction of pseudoephedrine baboon syndrome
after positive patch test. Contact Dermatitis 2000: 42: 312–
23. Barbaud A. Place of excipients in drug related allergy. Clin
Rev Allergol Immunol 1995: 13: 253–263.
24. Barbaud A, Bene M-C, Schmutz J L, Ehlinger A, Weber
M, Faure G C. Ro ˆle of delayed cellular hypersensitivity
and adhesion molecules in maculopapular rashes induced
by amoxicillin. Arch Dermatol 1997: 133: 481–486.
25. Wolkenstein P, Chosidow O, Flechet M-L, Robbiola O,
Paul M, Dume L, Revuz J, Roujeau J C. Patch testing in
severe cutaneous adverse drug reactions, including Stevens-
Johnson syndrome and toxic epidermal necrolysis. Contact
Dermatitis 1996: 35: 234–236.
26. Gonc ¸alo M, Figueiredo A, Tavares P, Fontes-Ribeiro C A,
Teixeira F, Poiares-Baptista A. Photosensitivity to pirox-
icam: absence of cross reaction with tenoxicam. Contact
Dermatitis 1992: 27: 287–290.
27. Bruynzeel D P, Von Blomberg B M, Van der Flier M,
Scheper R J, Van Ketel W G, De Haan P. Penicillin allergy
and the relevance of epicutaneous tests. Dermatologica
1985: 171: 429–434.
28. Romano A, Quarantino D, Di Fonso M, Papa G, Venuti
A, Gasbarrini G. A diagnostic protocol for evaluating non-
immediate reactions to aminopenicillins. J Allergy Clin Im-
munol 1999: 103: 1186–1190.
29. Romano A, Di Fonso M, Papa G, Pietrantonio F, Federico
F, Fabrizi G, Venuti A. Evaluation of adverse cutaneous
reactions to aminopenicillins with emphasis on those mani-
fested by maculopapular rashes. Allergy 1995: 50: 113–118.
30. Vega J M, Blanca M M, Carmona M J et al. Delayed al-
lergic reactions to beta-lactams. Allergy 1991: 46: 154–157.
31. Dooms-Goossens A. Allergy to inhaled corticosteroids: a
review. Am J Contact Dermatitis 1995: 6: 1–3.
32. Bircher A J, Pelloni F, Langauer-Messmer S, Müller D.
Delayed hypersensitivity reactions to corticosteroids ap-
plied to mucous membranes. Br J Dermatol 1996: 135:
33. Bircher A J, Flückiger R, Buchner S A. Eczematous infil-
trate plaque to subcutaneous heparin: a type IV allergic
reaction. Br J Dermatol 1990: 123: 507–514.
34. Moreau A, Dompmartin A, Esnault P, Michel M, Leroy
D. Delayed hypersensitivity at the injection sites of a low-
molecular-weight heparin. Contact Dermatitis 1996: 34:
35. Figarella I, Barbaud A, Lecompte T, De Maistre E, Reich-
ert-Pe ´ne ´trat S, Schmutz J L. Re ´action d’hypersensibilite ´ re-
tarde ´e aux he ´parines et he ´parinoı ¨des. Ann Dermato-ve ´ne ´re ´-
ologica 2001: 128: 25–30.
36. Michel M, Dompmartin A, Szczurko C, Castel B, Moreau
A, Leroy D. Eczematous-like drug eruption induced by sy-
nergistins. Contact Dermatitis 1996: 34: 86–87.
37. Silva R, Machado A, Brandao M, Goncalo S. Patch test
diagnosis in carbamazepine erythroderma. Contact Derma-
titis 1982: 8: 283–284.
38. Camarasa J G. Patch test diagnosis of exfoliative derma-
titis due to carbamazepine. Contact Dermatitis 1985: 12:
39. Romaguera C, Grimalt F, Vilaplana J, Azon A. Erythrod-
erma from carbamazepine. Contact Dermatitis 1989: 20:
40. Jones M, Fernandez-Herrera J, Dorado J M, Sols M, Ruiz
M, Garcia-Diez A. Epicutaneous test in carbamazepine cu-
taneous reactions. Dermatology 1994: 188: 18–20.
41. Rodriguez-Mosquera M, Iglesias A, Saez A, Vidal C.
Patch test diagnosis of carbamazepine sensitivity? Contact
Dermatitis 1991: 25: 137–138.
42. Alanko K. Patch testing in cutaneous reactions caused by
carbamazepine. Contact Dermatitis 1993: 29: 254–257.
43. Corazza M, Mantovani L, Casetta I, Virgili A. Exfoliative
dermatitis caused by carbamazepine in a patient with iso-
lated IgA deficiency. Contact Dermatitis 1995: 33: 447.
44. Puig L, Nadal C, Fernandez-Figueras M T, Alomar A.
Carbamazepine-induced drug rashes: diagnostic value of
patch tests depends on clinico-pathologic presentation.
Contact Dermatitis 1996: 34: 435–437.
45. Romano A, Pietrantonio F, Garcovich A, Rumi C, Bellocci
F, Caradonna P, Barone C. Delayed hypersensitivity to
diltiazem in two patients. Ann Allergy 1992: 69: 31–32.
46. Barbaud A, Trechot Ph, Gillet-Terver M N, Zannad F,
Schmutz J L. Investigations allergologiques dans une toxid-
ermie au diltiazem (Tildiem 300 LPR). The ´rapie 1993: 48:
47. Felix R, Comaish J. The value of the patch tests and other
skin tests in drug eruptions. Lancet 1974: i: 1017–1019.
48. Michel M, Dompmartin A, Louvet S, Szczurko C, Castel
B, Leroy D. Skin reactions to hydroxyzine. Contact Derma-
titis 1997: 36: 147–149.
49. Camarasa J G, Serra-Baldrich E. Tetrazepam allergy de-
tected by patch test. Contact Dermatitis 1990: 22: 246.
50. Collet E, Dalac S, Morvan C, Sgro C, Lambert D. Tetra-
zepam allergy once more detected by patch test. Contact
Dermatitis 1992: 26: 281.
51. Tomb R, Grosshans E, Defour E, Heid E. Allergic skin
reaction to tetrazepam detected by patch testing. Eur J
Dermatol 1993: 3: 116–118.
52. Blanca M, Vega J M, Garcia J, Carmona M J, Terados S,
Avila M J, Miranda A, Juarez C. Allergy to penicillin with
good tolerance to other penicillins; study of the incidence
in subjects allergic to betalactams. Clin Exp Allergy 1990:
53. Blanca M, Torres M J, Garcia J J, Romano A, Mayorga
C, De Ramon E, Vega J M, Miranda A, Juarez C. Natural
evolution of skin test sensitivity in patients allergic to b-
lactam antibiotics. J Allergy Clin Immunol 1999: 103: 918–
54. Audicana M, Bernaola G, Urrutia I, Echechipia S, Gasta-
minza G, Munoz D, Fernandez E, Fernandez De Corre `s
L. Allergic reactions to betalactams: studies in a group of
patients allergic to penicillin and evaluation of cross-reac-
tivity with cephalosporin. Allergy 1994: 49: 108–113.
55. Barbaud A, Got I, Trechot Ph, Noirez V, Schmutz J L.
Allergies cutane ´es et insulinothe ´rapie. Aspects re ´cents, con-
duite a ` tenir. Ann Dermatol Venereol 1996: 123: 214–218.
328BARBAUD ET AL.
56. Vidal C, Tome S, Fernandex-Redondo V, Tato F. Systemic
allergic reaction to corticosteroids. Contact Dermatitis
1994: 31: 273–274.
57. Barbaud A. Prick-tests et IDR dans les toxidermies: leurs
modalite ´s, leur inte ´re ˆt compare ´ a ` celui des tests e ´picutane ´s
me ´dicamenteux. In: Libbey J (ed): Progre `s en dermato-al-
lergologie. Montrouge, France: 1998: 4: 75–80.
58. Calkin J M, Maibach H. Delayed hypersensitivity drug re-
actions diagnosed by patch testing. Contact Dermatitis
1993: 29: 223–233.
59. Barbaud A, Trechot P, Reichert-Penetrat S, Commun N,
Schmutz J L. Relevance of skin tests with drugs in investi-
gating cutaneous adverse drug reactions. Contact Derma-
titis 2001: 45: 265–268.
Department of Dermatology, Fournier Hospital
36 Quai de la Bataille