ARTICLE
Auteur(s) :, Maya Krasteva1,*,
Martin Cottin1, Antonio Cristaudo2, Gérard
Lainé3, Gerhard Nohynek1, David
Orton4, Hervé Toutain1, Vincenzo
Severino2, John Wilkinson4
1L’Oréal Research and Development, Worldwide Safety
Department, 25-29, quai Aulagnier, 92600 Asnières, FranceFax: (+33)
1 47 56 72 68.
2Department of Dermatology, IFO, Polo Dermatologico San
Gallicano, 53, Via E. Chianesi, 00144 Roma, Italy
3Department of Dermatology, Hôpital Saint Louis, 1,
Avenue Claude Vellefaux, 75010 Paris, France
4Department of Dermatology, Amersham Hospital, Whielden
street, Amersham, Buckinghamshire HP7 OJD, England
accepté le 10 Septembre 2004
In recent decades, the prevalence of contact allergy to the
oxidative hair dye ingredient para-phenylenediamine (PPD) has
remained stable in the populations of the industrialised world
against a background of increasing use of hair dyes. Some authors
have even reported a decreasing prevalence [1, 2]. Nevertheless,
contact dermatitis to hair colouring products remains an important
health concern for consumers and the hair dye industry.In order to
protect the consumer against potential adverse reactions to hair
colouring products, the hair dye industry recommends that consumers
perform a skin allergy test (SAT) prior to hair dyeing [3, 4]. The
SAT consists of a 48-hour open application of the neat colourant
base of the commercial hair colouring product to the retroauricular
area. Labelling precautions advise that consumers should refrain
from hair dyeing if the SAT produces itching, reddening or swelling
at the application site.In the present study, we investigated the
sensitivity and specificity of the SAT to detect contact allergic
reactions to hair colouring products. We selected PPD as a model
allergen, since a) PPD is a major contact allergen in hair
colouring products [5, 6], b) the sensitivity of the test procedure
could be best appreciated by using as a model allergen an
ingredient which is found at different concentrations in consumer
products, thus permitting establishment of a dose-response and c)
recruitment of volunteers was facilitated by the presence of PPD in
the European standard series and the fact that all patients with
suspected contact dermatitis are routinely tested with it [7, 8].A
previous multicenter study [4] confirmed the effectiveness of the
SAT using a black hair dye with a high (1.8%) content of PPD. All
30 subjects with a history of contact dermatitis to hair colouring
products included in the study developed type IV allergic reactions
at the test site. However, a suitable preventive test should be
able to identify an allergic reaction by testing with an allergen
at an equal or lower concentration than that which elicited the
adverse reaction under use conditions. In the present study, we
investigated the sensitivity and specificity of the SAT to detect
contact dermatitis to hair colouring products over a range of
concentrations of PPD and typical shades (light, medium and dark)
in subjects with confirmed PPD-related contact allergy. We
attempted to establish the lowest eliciting concentration of PPD
and investigated whether the test concentration was within or below
the range of PPD concentrations of the colourant base in the
commercial product that had elicited the previous reaction in
PPD-positive subjects. Our aim was to confirm the suitability of
the SAT for the detection and the secondary prevention of contact
sensitivity to hair dyes.
Materials and methods
Three dermatological departments in three different countries
(England, Italy and France) participated in the study. The study
protocol was approved by the respective Ethics Committees. Informed
consent was obtained from all subjects.
Test materials
The PPD concentrations of more than two thousand PPD-containing
commercial products marketed worldwide were reviewed and 3 groups
of shades were defined based on the results: a) light shades
(typical range: 0.02-0.39% PPD in colourant base before mixing with
the developer), b) medium shades (0.14-1.34% PPD) and c) dark
shades (0.74-2.0% PPD). Accordingly, three experimental test
product prototypes from these groups were formulated: Product A
(containing 0.1% PPD, representative of light shades), Product B
(containing 0.5% PPD, representative of medium shades) and Product
D (containing 1.5% PPD, representative of dark shades). An
additional experimental test product, Product C, containing 1.0%
PPD, corresponded to the test material in the European standard
series.
The 4 experimental test colourant formulations corresponded to
modern hair dyes and contained typical ingredients, such as
resorcinol, meta-Aminophenol and 2,4-Diaminophenoxyethanol, ranging
from 0.1% (product A) to 1.64% (product D). The matrices consisted
of water, surfactants (max. 30%), conditioning agents (max. 5%),
alkalising agents (max. 3%), antioxidants and stabilisers (max. 1%)
and perfume (max. 0.5%).
The control test materials coded Y (control product for test
products A and B) or Z (control for C and D) corresponded to the
respective test substances, but contained neither PPD, nor any
other colourant molecule.
Subjects
PPD-positive subjects
The experimental group comprised 34 subjects, 1 male and 33
females, aged 23-68 years (mean: 48 ± 11 years). Four subjects
belonged to the skin phototypes V and VI (two were of Asian and two
of African origin), while the remaining 30 were Caucasians.
Thirty seven subjects with positive patch test reactions to PPD
(+ to +++) during routine investigation in the last five years were
initially included in the experimental group. All had self-declared
clinical manifestations compatible with non-occupational contact
dermatitis to oxidative hair colouring products. Exclusion criteria
were current acute or widespread eczema at any site, any eczema at
the test sites within the last 3 months and participation in
diagnostic patch testing during 6 weeks prior to the study. One
subject was excluded from the study because the clinical diagnosis
and the eliciting agent could not be confirmed, whereas another
subject dropped out from the study after the first product
application. An additional subject was excluded from the
statistical analysis because she was patch test-negative to PPD
when re-tested at the end of the study (see results). Therefore the
statistical analysis was carried out on 34 subjects.
Subjects were requested to report the shade and brand of
commercial hair colouring products that had elicited previous
contact dermatitis reactions. This was achieved with the help of
colour charts and presentation of packages of commercial hair
colouring products. Nine subjects reported reactions to light
shade, 15 to medium shade and 10 to dark shade oxidative hair dyes.
Seventeen study subjects reported reactions to professional, and 17
to home-use products. In 19 subjects, the reported eliciting
products contained PPD, whereas in 4 they contained
para-toluenediamine (PTD). In the remaining 11 subjects the
eliciting commercial product could not be identified. Clinical
manifestations following product use prior to the inclusion in the
study are shown in table 1( Table 1 ).
Medical treatment was confirmed in 20 subjects: 8/20 were treated
with antihistamines, 15 with topical corticosteroids, 3 with
systemic corticosteroids, whereas 3 had no medical treatment.
The results from diagnostic patch testing of the study subjects
are shown in table 2( Table 2 ).
Table 1 Clinical manifestations in PPD patch
test-positive subjects (declarative data by study subjects,
verified by the investigators)
|
Eczematous contact dermatitis of the scalp ± neighbouring skin
(forehead, ears, etc)
|
Contact dermatitis on remote areas (arms, forearms, body)
|
Oedema of eyelids ± forehead or neck
|
Whole facial oedema
|
|
33/34
|
4/34
|
10/34
|
4/34
|
Table 2 Positive results from diagnostic patch testing
prior to the inclusion of the subjects in the study. All subjects
were tested with the European standard series and some were tested
with additional series
|
PPD
|
Other contact allergens
|
|
(Toluene-2,5-diamine) (PTD) : 14/22
|
|
p-Aminophenol : 3/6
|
|
p-Aminoazobenzene : 5/6
|
|
+ : 2/34
|
Disperse orange 3 : 16/20
|
|
++ : 24/34
|
Disperse yellow 3 : 2/19
|
|
+++ : 8/34
|
o-Nitro-p-phenylenediamine (ONPPD) : 1/16
|
|
N-Isopropyl-N-phenyl-4-phenylenediamine (IPPD) : 2/34
|
|
Nickel : 10/34
|
|
Fragrance mix : 2/34
|
PPD-negative control subjects
Fifty control subjects with a negative patch test to PPD and
related allergens and no history of adverse reactions to hair dyes
entered the study. Four control groups of 12 to 13 control subjects
were sex- and age-matched to the groups of PPD-positive subjects
that had reacted respectively to products A, B, C and D. One
subject was found to be positive to an ingredient of the test
products and was excluded from the control group (see results);
therefore the statistical analysis was carried out on 49 control
subjects.
Study design
PPD-positive subjects
PPD-positive subjects were tested consecutively with experimental
products containing rising concentrations of PPD (products A
through D); the corresponding control product was applied
simultaneously. Subjects were allowed a rest period of 3 to 6 weeks
between two consecutive applications of test products. When test
products produced definite allergic reactions (at least erythema
with papules or homogeneous infiltration/œdema), the subject was
classified as SAT-positive and testing with products containing
higher concentrations of PPD was discontinued for ethical reasons.
PPD-negative control subjects
PPD-negative subjects were tested with one experimental product
only and with the corresponding control product. The experimental
product was the one that had elicited positive reactions in the
group of experimental subjects to which the control group was sex-
and age-matched.
The experimental product and the corresponding control product
were applied after placing circular adhesive devices (3M) behind
each ear. 0.1 ml of products was applied to the centre of the
device by means of a micropipette (Multipette 4780, Eppendorf) and
combitips (Eppendorf). The products were spread over the skin
surface enclosed by the adhesive device (1.75 cm2). The
actual volume of the test material applied was approximately 0.08
ml. Given that discolouration at the test site revealed the
presence of colourants in the experimental product, the
applications of experimental and control products were not
blinded.
Adhesive devices were removed 1 hour after application. The test
products were left for 48 hours without washing. Reactions were
recorded on Day 0 (1 hour post-application), Day 2 (48 ± 2 hours
post-application) and Day 4 (96 ± 2 hours post-application).
In the absence of a universally accepted method for grading
results of open tests, we used a scoring method based on the
proposal of Johansen et al. [9]. The method included: a) a global
evaluation of the severity of the reaction, corresponding to the
overall clinical impression of the investigator using a five-point
grading scale for the allergic reactions. To enable statistical
calculations, values were attributed as follows: negative = 0;
doubtful = 1; weakly positive = 2; moderately positive = 3 and
strongly positive = 4; b) an objective evaluation using a set of
objective parameters including: i) involved area of application
(scored 0-5), ii) erythema: pattern of involvement (0-2) and
strength (1-3), iii) papules, infiltration and/or oedema (0-4), iv)
vesicles and/or erosions (0-4) and v) sensory manifestations
reported by the patients (0-1). The maximal total score for each
reading time point was 19. On Day 2 after treatment, the subjects
were requested to report the earliest manifestations they had noted
and the time of onset.
Follow-up
PPD-positive subjects with no reaction or doubtful reactions to
product D underwent a diagnostic patch test to PPD at least 3 weeks
after the SAT. At least 3 weeks after any positive reaction to the
SAT to either the test or control products in control subjects or
to the control product in PPD-positive subjects, a diagnostic patch
test including all the ingredients of the incriminated formulation
was performed. The ingredients were tested at their usual
diagnostic concentrations or at typical product concentrations. The
diagnostic patch tests were graded according to the ICDRG grading
scale [8, 10].
A confirming challenge test (use test) was carried out in any
PPD-positive subject who reacted to test product D containing a PPD
concentration above the concentration range of the shade group to
which they had reported previous reactions. For this purpose
product B, representative of the group of shades to which the
eliciting hair colouring product belonged, was mixed with a
developer (6% hydrogen peroxide) and applied to a large strand of
hair. The corresponding control product Y was applied on the
contralateral side under the same conditions. Each contact area on
the scalp had a surface area of 7 cm2. The hair
colouring products were rinsed off after 30 minutes and skin
reactions were scored 20 minutes, 1, 3 and 6 days later. The use
test reactions were graded using the same method as the skin
sensitivity test.
Statistical analysis
Student, Levene and Mantel-Haenzel tests were used to assure that
the control subjects were sex- and age-matched to the groups of
PPD-positive subjects. Probability values of p < 0.05
were considered to be significant.
Mantel-Haenzel’s tests were used to compare the evaluation of
the reactions scored as “overall clinical impression” to each of
the test products and to the corresponding control products in
PPD-positive and PPD negative subjects at all reading times. To
assure a probability value of p < 0.05 for the
totality of the comparisons, p < 0.008 was considered
to be significant for each of the paired comparisons.
Analysis of variances followed by Tukey’s tests was used to
compare the objective numerical scores of the reactions between
PPD-positive and PPD-negative subjects, and between test products
and corresponding control products. Probability values of
p < 0.05 were considered to be significant.
Kendall’s Tau-b was used to evaluate the correlation between the
overall clinical evaluation and objective numerical scores of the
reactions. Correlation was considered to be significant when the
absolute value of Tau-b was greater than 0.6.
Results
Reactions to Product A (0.1% PPD) and/or Control Y
Twenty seven out of 34 PPD patch test-positive subjects reacted to
the SAT with product A (light shade prototype) with a definite
allergic reaction. Reactions included at least erythema with either
papules or homogeneous infiltration on Day 2. In 22/27 subjects,
vesicles were present. Reactions were evaluated as weakly positive
(2/27), moderately positive (12/27) or strongly positive (13/27).
A PPD-negative control subject reacted strongly positive
(erythema, infiltration and vesicles) to test product A and to the
corresponding control product Y. The subject was subsequently patch
tested with the ingredients of the two formulations and was found
positive to sodium metabisulfite present in both formulations. The
subject was therefore excluded from the control group. All
remaining 11 PPD-negative subjects showed negative reactions to
products A and Y.
The time-course of the reactions to products A and Y in the 27
PPD-positive subjects who reacted to product A and in the 11
PPD-negative controls is shown on ( figure 1 ). The most severe
reactions to product A in PPD-positive subjects were observed on
Day 2 and continued to be present, albeit with lower severity, on
Day 4.
No reactions were graded as positive in PPD-positive or
PPD-negative subjects to the control product Y, or in PPD-negative
control subjects to the test product A, although several test sites
showed low grade erythema, which remained unnoticed by the
subjects.
The severity of the reactions to product A in PPD-positive
subjects was statistically higher than the severity of the
reactions to the control product in PPD-positive subjects and
higher than the severity of the reactions to the experimental and
to the control products in PPD-negative subjects, both on Day 2 and
Day 4 (p < 0.001). There was no statistically
significant difference on Day 0. The same conclusions were obtained
when analysing the “overall clinical impression” or the objective
numerical scores of the reactions.
The remaining 7 PPD-positive subjects that failed to react to
test product A were subsequently tested with products B and Y.
Reactions to Products B (0.5% PPD), C (1% PPD) or D (1.5%PPD)
and/or corresponding Controls Y or Z
Three PPD-positive subjects responded on Day 2 to product B with a
definite allergic reaction (erythema, infiltration and vesicles)
and completed the study.
The 4 remaining PPD-positive test subjects were treated with
product C and the corresponding control product Z. Three subjects
developed on Day 2 definite allergic reactions to product C.
The last PPD-positive subject that had failed to respond with an
allergic reaction to products A, B or C showed a clear allergic
reaction to product D (erythema, infiltration and vesicles) on Day
2.
One subject with an initial positive patch test to PPD included
in the study was negative to all test products at all readings. The
subject was patch tested with PPD 1% in petrolatum on two different
occasions two months apart and was found repeatedly negative.
Therefore, the subject was classified as no longer being sensitised
to PPD and was excluded from the statistical evaluation.
Three groups of sex and age-matched PPD-negative control
subjects were tested respectively with products B
(n = 13), C (n = 12) or D (n = 13)
and corresponding control products. All reactions in the control
subjects were graded negative.
Given the low number of PPD-positive subjects reacting to
product B (n = 3), C (n = 3) and D
(n = 1), no statistical comparison between these groups
and corresponding control groups was performed. However, in these
groups the reactions to the test products and the corresponding
control products both in PPD-positive and in PPD-negative subjects
followed the same pattern as the reactions of observed to products
A and Y, i.e. PPD-positive subjects developed definite allergic
reactions on Days 2 or 4 to the experimental products containing
PPD, whereas PPD-negative subjects did not (table 3( Table 3 )).
A statistical analysis carried out in all 34 PPD-positive
subjects and 49 PPD-negative controls showed that, on Days 2 and 4,
PPD-positive subjects had a statistically significant greater
severity of reactions to PPD-containing products when compared with
that to the control products (p < 0.001). On Day 0,
there were no statistically significant differences between the
severities of the reactions in PPD-positive or negative subjects to
the experimental or control products. The highest values of
objective scores and clinical gradings of the reactions to
PPD-containing experimental products in PPD-positive subjects were
observed on Day 2. The results obtained by the two methods of
evaluation on Day 2 in all the subjects were significantly
correlated (Kendal’s tau b = 0.78).
Table 4( Table 4 ) shows the number
of PPD-positive subjects that developed a definite allergic
reaction to one of the test products on Day 2. Keeping in mind that
PPD-positive subjects were progressively withdrawn from the study
as soon as they developed an allergic reaction and assuming that
the application of a product containing a higher concentration of
allergen would be expected to elicit the same or a stronger
allergic reaction, cumulative percentages were calculated for
products B, C and D. Accordingly, 88% of the subjects would have
reacted to Product B (0.5% PPD), 97% to product C (1% PPD) and 100%
to product D (1.5% PPD). Overall, 15/34 (44%) subjects showed
strongly positive reactions, 13/34 (38%) had moderately positive
reactions and 6/34 (18%) had weakly positive reactions to
PPD-containing test products. Twenty seven from 34 reactions (79%)
were vesicular.
All PPD-positive subjects reported that they had noticed a
reaction to the SAT with the PPD-containing product. Pruritus,
appearing 5-28 hours post-application (mean time of onset 15.4
hours) was the first symptom reported by 33/34 subjects.
Interestingly, the 4 subjects that had designated PTD-containing
hair dyes as eliciting reacted positively to PPD-containing test
products. Two of them reacted to product A (0.1% PPD), the
remaining 2 to product C (1% PPD). Subjects with skin phototypes V
and VI displayed the same type of reactivity as the lighter
phototypes: definite allergic reactions were recorded on Day 2 to
product A in 3/4 and to product B in 1/4 subjects.
Table 3 Mean severity of reactions (overall clinical
impression) to test products A (0.1% PPD), B (0.5% PPD), C (1.0%
PPD) and D (1.5% PPD) and to the corresponding control products (Y
and Z) in PPD-positive and PPD-negative subjects. 0 = negative
reaction; 1 = doubtful; 2 = weakly positive; 3 = moderately
positive; 4 = strongly positive
|
Subject group
|
Test product
|
Content PPD (%)
|
Day 0
|
Day 2
|
Day 4
|
|
PPD-positive
|
A
|
0.1
|
0.07
|
3.41
|
2.85
|
|
PPD-positive
|
B
|
0.5
|
0.3
|
3.67
|
3
|
|
PPD-positive
|
C
|
1.0
|
0
|
2
|
1.3
|
|
PPD-positive
|
D
|
1.5
|
0
|
2
|
2
|
|
PPD-positive
|
Y tested with A
|
0.0 (Control)
|
0.04
|
0
|
0
|
|
PPD-positive
|
Y tested with B
|
0.0 (Control)
|
0.3
|
0
|
0
|
|
PPD-positive
|
Z tested with C
|
0.0 (Control)
|
0
|
0
|
0
|
|
PPD-positive
|
Z tested with D
|
0.0 (Control)
|
0
|
0
|
0
|
|
PPD-negative
|
A
|
0.1
|
0
|
0
|
0
|
|
PPD-negative
|
B
|
0.5
|
0
|
0
|
0
|
|
PPD-negative
|
C
|
1.0
|
0
|
0
|
0
|
|
PPD-negative
|
D
|
1.5
|
0
|
0
|
0
|
|
PPD-negative
|
Y tested with A
|
0.0 (Control)
|
0
|
0
|
0
|
|
PPD-negative
|
Y tested with B
|
0.0 (Control)
|
0
|
0.1
|
0
|
|
PPD-negative
|
Z tested with C
|
0.0 (Control)
|
0
|
0
|
0
|
|
PPD-negative
|
Z tested with D
|
0.0 (Control)
|
0
|
0
|
0
|
Table 4 Number of PPD-positive subjects developing
positive reactions to the test products on Day 2 after
application
|
|
|
|
|
|
Number of subjects reacting
|
27/34
|
3/34
|
3/34
|
1/34
|
|
% of subjects reacting
|
79.4
|
8.8
|
8.8
|
2.9
|
|
Cumulative percentage of subjects reacting to the respective test
product
|
79.4
|
88.2
|
97
|
100
|
Comparison of the eliciting PPD concentrations in the study
(experimental data) with the PPD concentrations present in the
colourant base of the eliciting commercial product (declarative
data)
In order to evaluate the sensitivity of the SAT and its predictive
value for each individual test subject, the eliciting
concentrations of PPD under the study conditions were compared with
the PPD concentrations in the colourant base of the reported,
eliciting commercial products, i.e. the PPD concentration range in
the group of hair dye shades claimed as eliciting by the consumer.
( Figure 2 )
shows the relation between experimental and declarative data.
With a single exception, the SAT was predictive of contact
dermatitis to the corresponding shade of hair dyes reported as
eliciting by the subject. Indeed, most subjects reacted to test
products containing PPD concentrations lower than the concentration
ranges of the respective commercial group of shades (20/34).
However, a single subject reported a history of contact
dermatitis to medium shades, but reacted only to product D (1.5%
PPD, dark shade prototype). In this subject, the SAT seemed to be
less predictive of contact dermatitis produced by the respective
commercial hair dye. In order to verify the presence of allergy to
medium shades in this subject, we carried out a use test with
product B (medium shade prototype) and the control product Y.
Neither product elicited an allergic reaction on the scalp
(readings on Days 0, 1, 3 and 6). It was therefore concluded that
there was no evidence of contact dermatitis to medium shades at the
time of the study.
Discussion
We investigated the sensitivity and specificity of the SAT in 34
PPD-positive subjects with a history of contact dermatitis to
oxidative hair dyes using hair colouring products containing 0.1,
0.5, 1.0 or 1.5% PPD. When subjected to the SAT, all subjects
developed definite type IV allergic reactions (erythema and papules
or homogeneous infiltration) to the PPD-containing test products
within 2 days. Although the experimental products were open tested,
the nature and severity of the reactions satisfied the criteria of
a moderate and strong reaction by the ICDRG grading scale for
evaluating occlusive patch tests in 27/34 subjects (presence of
vesicles, sometimes coalescing into bullae). Most importantly, the
findings were not only rated as positive by classical
dermatological criteria, but were also noticed and reported by the
study subjects. All PPD-positive subjects felt the development of a
reaction, pruritus being the first warning symptom. As the most
severe reactions occurred on Day 2, the recommendation to perform
the SAT 2 days prior to hair dyeing is consistent with the time
course of the allergic reactions observed in our study.
Given that 4 subjects with skin phototypes V and VI displayed a
similar reactivity as the subjects with lighter skin, the SAT
appears to be able to predict contact dermatitis to hair colouring
products in individuals with darker skin types.
In the majority of cases (27/34), open application of a test
product containing as little as 0.1% PPD produced manifestations
that included all typical features of allergic contact dermatitis.
Experimental studies have shown that in a sensitized person the
exposure to a higher allergen dose on a constant surface area
results in reactions with higher severity [11]. Taking into account
ethical considerations, human elicitation studies refrain from
application of higher concentrations when a definite allergic
reaction is observed [12-14]. Given that a more severe reaction may
be expected to a product containing PPD above the first eliciting
concentration, we refrained from applying higher concentrations
whenever a subject developed a clear-cut contact allergic reaction
to the test products. Therefore, when 27 subjects completed the
study after reacting to product A (0.1% PPD), only seven subjects
remained to be tested with products containing higher PPD
concentrations. Overall, the cumulative response rates of subjects
reacting to products containing 0.1, 0.5, 1.0 or 1.5% PPD were
27/34, 30/34, 33/34 or 34/34, corresponding to 79, 88, 97 or 100%
of the study subjects, respectively.
These results confirm the data of our previous study using a
1.8% PPD-containing dark hair dye. In that study, the SAT yielded a
good predictive value for identification of hair dye contact
dermatitis. The results of the present study demonstrate that a SAT
with lower concentrations of PPD may also be predictive of contact
dermatitis to hair dyes.
Some previous studies reported that open application of PPD may
elicit contact dermatitis [15-17]. A study in which two
PPD-containing hair dyes were tested open in a limited number of
subjects found the same order of responsiveness as our present
study: 3/8 PPD-positive subjects were reported to react to the hair
colouring product containing as little as 0.1% PPD whereas 8/8
reacted to the hair colouring product containing 1% PPD [16].
Although the elicitation dose-response to PPD has been studied
under occlusive patch tests and in open application [16, 18, 19],
none of these studies compared the eliciting experimental
concentrations with those of commercial products under their
conditions of use. In contrast, our study also investigated PPD
concentrations in the eliciting commercial products (declarative
data). Since the study subjects were able to report the shades and,
in a majority of cases, even the brands of previously used
commercial products, we were able to determine the nature of the
principal colourant base (PPD or PTD) and their concentration range
in the originally eliciting commercial product. A drawback of this
approach, which relies on the recollection of events that may have
occurred several years previously, is the possibility of changes in
the threshold of elicitation. Indeed, one subject from the
PPD-positive group underwent testing with all the experimental
products without developing an allergic reaction and was found
negative in 2 subsequent patch tests with 1% PPD under a Finn
chamber. This subject had either lost her contact sensitivity to
PPD or her level of sensitivity may now have been below the
threshold of testing. Another subject reacted only to product D
(dark shades prototype) while reporting a history of contact
dermatitis to medium shade hair dyes. A use test confirmed that at
the time of the study there was no evidence of contact dermatitis
to medium shades. However, this finding is not surprising, since
considerable variation of the elicitation thresholds under patch
test [20-23] or use conditions [24-26] is a well known phenomenon
which may ultimately progress to loss of contact sensitisation or a
state of clinical tolerance [27-29].
Our results demonstrate that the eliciting concentration of PPD
when applied under conditions of the SAT is equal to or lower than
the PPD concentration in the colourant base of the respective
commercial product responsible for elicitation of previous
reactions. This finding confirms the sensitivity and specificity of
the SAT over the entire range of PPD concentrations used in hair
colouring products.
Overall, our previous and present studies confirm that the SAT
application of the neat colourant reliably detects contact
dermatitis to PPD-containing hair colouring products. The excellent
predictive value of the SAT implies that mixing with the developer
is not required for the efficiency of the skin sensitivity test. In
addition, our data suggests that the colourant base alone and not
products of the oxidative reaction are responsible for the allergic
reaction.
Our results also suggest that the SAT is highly specific:
although some control subjects showed low grade erythema, not a
single reaction to the experimental and control products was scored
as positive in PPD-negative subjects, or to the control products in
PPD-positive subjects.
Finally, it is noteworthy that a control product and its
corresponding test product containing sodium metabisulfite produced
strongly positive reactions in a sodium metabisulfite-positive
subject of the control group. This accidental finding confirms the
predictive value of the SAT and suggests that the test may be
efficient not only for detection of hair dye contact dermatitis
related to PPD, but also for detection of allergies related to
other ingredients.
Overall, our data demonstrate an excellent sensitivity and
specificity of the SAT over the complete range of PPD
concentrations used in hair colouring products. Our results
indicate that the SAT using the neat colourant base is sufficient
to detect and predict contact sensitivity to hair colouring
products and is therefore an important tool for the secondary
prevention of hair dye contact dermatitis.
Acknowledgements
We thank Anne-Lise Garcel from L’Oréal Research and Development,
Asnières, France for the statistical evaluation of data.
References
1 Maouad M, Fleischer AB, Sherertz EF, et al.
Significance-prevalence index number: a re-interpretation and
enhancement of data from the North American Contact Dermatitis
Group. J Am Acad Dermatol 1999; 41: 573-6.
2 Marks JG, Belsito DV, DeLeo VA, et al.
North American Contact Dermatitis Group patch-test results, 1998 to
2000. Am J Cont Dermatitis 2003; 14: 59-62.
3 Calnan C. Adverse reactions to hair products. In:
Zwiak C, ed. The science of hair care. New York and Basel:
Marcel Dekker Inc, 1986: 409-23.
4 Krasteva MK, Cristaudo A, Hall B, et al.
Contact sensitivity to hair dyes can be detected by the consumer
open test. Eur J Dermatol 2002; 12: 322-6.
5 Frosch PJ, Burrows D, Camarasa JG, et al.
Allergic reactions to a hairdressers’ series: results from 9
European centres. Contact Dermatitis 1993; 28: 180-3.
6 Rietschel RL, Fowler JF. Allergy to preservatives
and vehicles in cosmetics and toiletry. In: Rietschel RL,
Fowler JF, eds. Fisher’s contact dermatitis. 5th edition.
Philadelphia: Lippincott Williams & Wilkins, 2001: 211-59.
7 Krasteva M, Kehren J, Sayag M, et al.
Contact dermatitis. II. Clinical aspects and diagnosis. Eur J
Dermatol 1999; 9: 144-59.
8 Devos SA, Van Der Valk PG. Epicutaneous patch
testing. Eur J Dermatol 2002; 12: 506-13.
9 Johansen JD, Bruze M, Andersen KE, et al.
The repeated open application test: suggestions for a scale
evaluation. Contact Dermatitis 1997; 39: 95-6.
10 Fregert S. Manual of contact dermatitis. 2nd edition.
Copenhagen: Munksgaard, 1981.
11 Friedmann PS, Moss C, Shuster S, et al.
Quantitation of sensitization and responsiveness to
dinitrochlorobenzene in normal subjects. Br J Dermatol 1983;
109(Suppl 25): 86-8.
12 Basketter D, Horev L, Slodovnik D, et al.
Investigation of the threshold for allergic reactivity to chromium.
Contact Dermatitis 2001; 44: 70-4.
13 Bruze M, Johansen JD, Andersen KE, et al.
Deodorants: An experimental provocation study with cinnamic
aldehyde. J Am Acad Dermatol 2003; 48: 194-200.
14 Svedman C, Bruze M, Johansen JD, et al.
Deodorants: An experimental provocation study with
hydroxycitronellal. Contact Dermatitis 2003; 48: 217-23.
15 Epstein WL, Taylor MK. Experimental sensitization
to paraphenylenediamine and paratoluenediamine in man. Acta
Dermatovenereol 1979; 59: 55-7.
16 Ishihara M, Nogami T, Itoh M, et al.
Basic studies on contact dermatitis due to hair colourings and cold
permanent wave solutions. Japanese Cosmetic Journal 1983; 7:
272-82.
17 Reiss F, Gahwiler M, Lustig B. Sensitivity to
hair dyes. J Allergy 1957; 28: 134-41.
18 Hextall JM, Alagaratnam NJ, Glendinning AK,
et al. Dose-time relationships for the elicitation of contact
allergy to para-phenylenediamine. Contact Dermatitis 2002; 47:
96-9.
19 McFadden JP, Wakelin SH, Holloway DB,
et al. The effect of duration on the elicitation of
para-phenylenediamine contact allergy. Contact Dermatitis 1998; 39:
79-81.
20 Hindsen M, Bruze M, Christensen OB. Individual
variation in nickel patch test reactivity. Am J Contact Dermatitis
1999; 10: 62-7.
21 Ayala F, Balato N, Lembo G. Statistical
evaluation of the persistence of acquired hypersensitivity by
standardized patch tests. Contact Dermatitis 1996; 34: 354-8.
22 Valsecchi R, Rossi A, Bigardi A, et al.
The loss of contact sensitization in man. Contact Dermatitis 1991;
24: 183-6.
23 Nielsen NH, Linneberg A, Menné T, et al.
Persistence of contact allergy among Danish adults: an 8-year
follow-up study. Contact Dermatitis 2001; 45: 350-3.
24 Epstein WL. Poison oak and poison ivy dermatitis. In:
Adams RM, ed. Occupational skin disease. Philadelphia: WB
Saunders Company, 1990: 536-42; 2nd edition.
25 Seo KI, Eun HC. Loss of contact sensitization
evaluated by laser doppler blood flowmetry and transepidermal water
loss measurement. Contact Dermatitis 1996; 34: 233-6.
26 Mastrolonardo M, Lopalco PL, Diaferio A.
Topical immunotherapy with contact sensitizers: a model to study
the natural history of delayed hypersensitivity. Contact Dermatitis
2002; 47: 214-214.
27 Chan YC, Ng SK, Goh CL. Positive patch test
reactions to para-phenylenediamine, their clinical relevance and
the concept of clinical tolerance. Contact Dermatitis 2001; 45:
217-20.
28 Krasteva M, Kehren J, Ducluzeau MT,
et al. Contact dermatitis. I. Pathophysiology of contact
sensitivity. Eur J Dermatol 1999; 9: 65-77.
29 Saint-Mézard P, Bérard F, Dubois B,
et al. The role of CD4+ and CD8+ T cells in contact
sensitivity and allergic contact dermatitis. Eur J Dermatol 2004;
14: 131-8.
|
Printable version
Version PDF
