211
25 Functional Skin Testing: the SMART Procedures
Swen Malte John, Hans J. Schwanitz (†)
Contents
25.1
25.2
25.2.1
25.2.1.1
25.2.1.2
25.2.1.3
25.2.2
25.2.2.1
25.2.2.2
25.2.2.3
25.2.2.4
25.2.2.5
25.2.3
25.3
25.4
25.1
Irritant Patch Testing with Sodium Hydroxide . . . 211
Swift Modified Alkali Resistance Test (SMART) . . . 211
Methods and Participants . . . 211
Subjects . . . 211
Irritant Patch Testing . . . 212
Relevant Variables and Statistical Analyses . . . 213
Results . . . 214
Cohort and Medical Diagnoses . . . 214
Irritant Patch Testing . . . 215
30-Minute NaOH Challenge . . . 216
Reproducibility . . . 216
Constitutional Risks: Atopy . . . 216
Conclusions . . . 217
Applications of the SMART: Differential Irritation Test . . . 218
Outlook: Implications for Medicolegal Evaluations in Occupational Dermatology . . . 219
References . . . 220
Irritant Patch Testing with
Sodium Hydroxide
An archetype of chemical skin irritation tests is Burckhardt’s “alkali resistance test” which he introduced in
1947, using 0.5 M sodium hydroxide up to 8×10 min
under occlusion [9]. The test procedure was later
modified by Burckhardt and his colleagues [37, 29, 10;
Table 1]. He and his co-workers claimed that the test
was able to assess the integrity of the epidermal barrier; the test was recommended as a screening tool for
chemically phenotyping the individual. There were
some reports claiming that the technique was useful
for pre-employment testing in risk professions [14,
23]. However, the concept was controversial from the
start, findings were inconsistent, and the technique
fell into oblivion in most countries [5, 6, 17, 23, 24].
In current occupational dermatology in Germany,
alkali resistance tests given in numerous variations
still play an important role for medicolegal evaluations. However, the use of these tests for such purposes in routine diagnostics is controversial [23].
Throughout the last five decades, many scientists
have studied sodium hydroxide as an irritant and debated its value, using various test protocols (Table 1).
Recent epidemiological data from the Swiss metal
industry seem to confirm the use of sodium hydroxide for pre-employment screening [4]. In the United
States, where “alkali resistance” never caught on, two
modifications have recently been proposed, one of
which only uses skin bioengineering for the assessment of test results obtained with sodium hydroxide
[39], while the other employs only clinical evaluations
[24]. Both groups claim high reliability of the respective test procedure. However, results were obtained in
small populations, and reproducibility has been questioned [2, 22].
Unlike the popular irritant model sodium lauryl
sulfate (SLS), for which the standardization process
is far advanced [35, 27], no such efforts have yet been
undertaken concerning NaOH. Therefore, with the
special indication of standardizing diagnostics in
occupational dermatology, we re-evaluated sodium
hydroxide skin irritation in a large group of patients
with occupational dermatoses, objectifying results
using current biophysical techniques.
25.2
Swift Modified Alkali
Resistance Test (SMART)
25.2.1
Methods and Participants
25.2.1.1 Subjects
We tested 1,271 patients from various high-risk professions with a history of previous occupational eczema for medicolegal evaluations from January 1993
to April 1999 in the dermatology department of Os-
212
Swen Malte John, Hans J. Schwanitz
Table 1. Alkali resistance test and some recommended modifications
NaOH concentration in aqueous solutions: 1 normal (N) = 1 molar = 1 mol/l. FA forearm, UA upper arm, DH dorsum of hand, Th
thigh, Sh shoulder
Test site(s)
Maximum occlusion time
NaOH concentration
Parameters studied
Remarks
References
FA dorsal, upper back
8×10‘, later 3×10‘
0.5 N
Vesicles, erythema, erosions,
patchy brown
stains (stinging)
[9, 10]
Subsequent
observation of
reactions in test site
FA volar, upper back, UA
3×10‘
0.5 N
Papules, vesicles,
erosions, crusts
after 24 h
Parallel observation of three test
sites; late reactions relevant
[37]
FA volar, Sh, UA
8×10‘, later 6×10‘
0.5 N
≥10 Red spots
or vesicles
No glass blocks
[29]
Th
5×5‘
0.5 N
≥10 Nitrazine
yellow-positive erosions
Introduction of
nitrazine yellow
[26]
Sh, FA, DH
5×5‘
0.5 N
a. Erythema >1 cm
If a–c not fulfilled,
break up according to Locher. Late
readings (24 h)
[36]
b. Erosion >2 mm
c. Three nitrazine
yellow-positive erosions
div.
1 h; for 6 days
0.03 N
TWL + assessment Claimed reproducof refractory period ibility + intraindivual constancy
[32,33]
FA
1×5‘
0.2 N
SSWL 5‘ after
occlusion
Biophysical assessment only,
no clinical skin
signs induced
[39]
FA, volar, dorsal
20×1‘
1.0 N
Time interval to ≥1
nitrazine yellowpositive erosion
(erosion time)
Clinical assessment only, erosion
time not altered
by petrolatum
pretreatment
[24]
nabrueck University. Of these patients, aged 17 – 73
years (314 female, 258 male), 572 fulfilled the criteria
mentioned below and were subsequently accepted
for the study. History taking and detailed dermatological examination of the complete integument were
performed by physicians in a specialist training programs for occupational dermatology or dermatologists. For the most part, there were abundant prior
medical records available and assessments from the
(former) workplace. Additionally, in most of these patients, epicutaneous patch testing was performed and
prick tests as well as serologic investigations (e.g., IgE
and sIgE) for assessing inhalatory atopy.
25.2.1.2 Irritant Patch Testing
Irritant patch testing was conducted after informed
consent was obtained. The study was approved by the
Ethical Committee of the University of Osnabrueck.
Measurements were only taken in clinically healthy
skin. Skin lesions had to have healed at least 3 weeks
25 Functional Skin Testing: the SMART Procedures
before the investigation. Participants were requested
not to use soap or creams/emollients in the areas of
investigation 24 h prior to each examination.
NaOH Challenge
Sodium hydroxide (33 µl, 0.5 M) was pipetted to
the test site (mid-volar forearm) and covered by a
2.5×3×1,0-cm glass block according to Locher [26].
The glass block was fixed with nonocclusive tape under slight pressure to assure uniform spreading of the
solution; 0.9% NaCl (33 µl) served as a control in an
adjacent area and was covered by a glass block in the
same fashion. After 10 min, the solution was gently
wiped off with a swab. Clinical and biophysical readings were done in the test and control area 10 min after the end of the provocation phase. Then a second,
identical 10-min provocation phase with consecutive
clinical and biophysical assessment after 10 min was
conducted. Another clinical and biophysical reading
was done after 24 h.
Clinical skin changes in the test areas were recorded using a five-grade ordinal scale:
1 = “Nil”
2 = “Soap effect”
3 = “Minimal erythema and/or minimal vesiculation
and/or maximally one erosion”
4 = “Marked erythema and/or marked edema and/or
marked vesiculation and/or 2 erosions”
5 = “Very marked erythema/vesiculation/edema and/
or 5 erosions or necrosis”
The test was stopped immediately if after the first
provocation phase there were marked clinical skin
changes (grade 4 [n=51] or grade 5 [n=0]) or subjective discomfort (n=0).
Skin Bioengineering
In a previous pilot study with 92 similar patients, we
showed that two 10-min provocation periods with
0.5 M NaOH provided significant information on
individual skin sensitivity [23]. Briefly, evaporimetric measurements of SSWL/TWL were performed 2–
10 min after the end of NaOH-provocation in 2-min
intervals. Transepidermal water loss (TWL) allows an
estimation of water evaporation through the stratum
corneum providing information on epidermal barrier function [28]. If evaporimetric measurements are
performed immediately after aqueous solutions have
been applied to the skin, initially the SSWL (skin
surface water loss; equals excess water loss from skin
surface hydration plus TWL [40]) is being measured.
The experiments revealed that a 10-min interval after the end of each of the NaOH provocation phases
was a suitable time period for clinical and biophysical patch test reading. At this time, excess water loss
from skin surface hydration was already minimal so
that roughly only TWL has been estimated.
In the present study, therefore, TWL and also
relative skin moisture (RSM) were routinely assessed.
TWL was measured using the ServoMed evaporimeter EP1 (ServoMed, Stockholm, Sweden) in a perspex-incubator applying the ServoMed gold-plated
protection cover (steel grid) and a rubber stopper
as an insulating probeholder according to the ESCD
guidelines [28]. Relative skin moisture was estimated
by capacitance using the corneometer C 820 (Courage & Khazaka, Cologne, Germany). Measurements
were conducted in triplicate, the median was then
taken as the RSM value.
The investigations took place in an air-conditioned laboratory in steady state conditions (ambient
temperature 20–21° C, relative humidity 40%–45%).
Acclimatization of participants was at least 15 min,
usually 30 min. Measurements were conducted after
acclimatization in the prospective test and control
areas (ex-ante readings), and then 10 min after each
provocation phase and finally after another 24 h.
25.2.1.3 Relevant Variables and Statistical
Analyses
Clinical Diagnoses and Atopy
Diagnosis of clinical atopy usually had to be made retrospectively because skin lesions were either healed
on investigation or reduced to minor residuals; if
florid skin changes were detected on first clinical examination subjects were asked to return after healing
for patch testing. If in the medical records previous
atopic dermatitis (n=92), palmar or plantar eczema
[31] in its various manifestations (n=93), or previous
flexural eczema (n=208) was documented, the respective subjects were grouped as “atopic skin disposition”
(n=248; it should be noted that there were frequent
combinations of the above-mentioned skin manifestations). The other main groups of diagnoses were
pure “irritant contact dermatitis” (without atopic skin
manifestations in the history, n=138) or pure “allergic
contact dermatitis” (n=130).
Biophysical Parameters
∆-TWL and ∆-RSM, respectively, were calculated
in regard to the ex-ante-values and the values in the
NaCL controls using the following mathematical
term: ∆-TWLNaOH, 10 or 20 min = (TWLNaOH, 10 or 20 min –
TWLNaOH, 0 min) – |TWLNaCl, 10 or 20 min – TWLNaCl, 0 min|.
213
214
Swen Malte John, Hans J. Schwanitz
Fig. 2. Time course of biophysical parameters with respect to the
result of visual scoring (negative/positive) after 2×10 min NaOH
challenge. The medians of ∆−TWL and ∆−RSM measurements
are shown (Pos TWL10’ NaOH n=287, Neg TWL10’ NaOH n=285; Pos
TWL20’ NaOH n=238, Neg TWL20’ NaOH n=283; Pos TWL24 h n=212,
Neg TWL24 h n=199; identical n for RSM values)
Fig. 1. Age and sex distribution of cohort (n=572)
For the second difference (|TWLNaCl, 10 or 20 min
– TWLNaCl, 0 min |), only the total positive amount was
used; this is of relevance when the second difference
is negative. In these 176 cases (TWLNaCl, 10 min), and 102
cases (TWLNaCl, 20 min), respectively, the ex-ante value
in the control area was slightly higher then after 10 or
20 min of NaCl application. ∆-TWL and ∆-RSM will
be negative if the difference of values in the NaOH
areas (TWLNaOH, 10 min – TWLNaOH, 0 min) is smaller than
the respective difference in the control areas. This
was only rarely the case in robust skin without skin
changes, due to the variance of measurements (∆TWLNaOH, 10 min: n=37; ∆-TWLNaOH, 20 min: n=46).
Statistical Analyses
Besides descriptive statistical analysis, data evaluation focused on the estimation of a predictive (critical) value of the investigated biophysical parameters;
for this purpose, receiver operating characteristic
(ROC) curves were developed, as described in detail
by Green and Swets [20], and Lange and Weinstock
[25]. For ROC curves and other investigations, the
clinical score (five-point ordinal scale) was dichotomized (clinical grades 1, 2 vs grades 3–5). Differences
in TWL, RSM between NaOH, and controls were
tested using bivariate two-tailed nonparametric tests
(Mann-Whitney U or Kruskal-Wallis test). In crosstabulations, the chi-square statistics or the McNemar-test (linked samples), respectively, were used to
analyze the differences between the observed and the
expected values. Using the Cohens Kappa index [11],
the degree of agreement of clinical and biophysical
parameters and reproducibility were calculated. Correlations between the various clinical, demographical,
and biophysical parameters were analyzed using the
two-tailed Spearman rank correlation test. The size of
estimated effects was judged with Cohen’s classification [12].
An error probability of <5% was considered statistically significant. For statistical analyses, the statistical software package SPSS (version 9.0 D, SPSS Inc.,
Munic, Germany) was employed.
25.2.2
Results
25.2.2.1 Cohort and Medical Diagnoses
The age and sex distribution of the study population
is given in Fig. 1. The most frequent profession was
“hairdresser” (22.6%), followed by various “professions in the health sector” (16.9%), mainly nursing.
This explains the dominance of women in the young
age groups. Male-dominated jobs followed in third
and fourth positions: “metal worker” (7.0%) and
“brick layer” (6.4%).
In 84.6%, the result of medicolegal evaluation was
that skin disease was considered to have been induced
by the job; overall the most relevant single factor for
the elicitation of the dermatoses was wet work.
As pointed out above, diagnosis had to be made
in retrospect because skin lesions were either healed
on investigation or reduced to minor residuals. The
most frequent primary diagnosis was allergic contact dermatitis (n=165; 28.9%), followed by irritant
dermatitis (n=158; 27.6%) and various atopic skin
manifestations (e.g., atopic dermatitis; atopic palmar or plantar eczema). Due to the frequent overlap
in the pathogenesis of occupational skin diseases in
205 cases, more than one diagnosis was made; most
frequently atopic manifestations were diagnosed together with allergic or irritant contact dermatitis. If
all cases in which atopic skin manifestations were diagnosed alone or in combination are taken together,
248 patients (43.4%) were considered atopic (“atopic
skin disposition”). 153 (26.7%) had a history of inha-
25 Functional Skin Testing: the SMART Procedures
Fig. 3. Association of ∆-TWL after 2×10 min NaOH challenge
and clinical grading (n=572; in 51 cases test was stopped after 10 min in accordance with break-up criteria; then 10-min
readings were used])
latory atopy (rhinitis or asthma), which was mostly
associated with a history of some kind of atopic skin
manifestations.
Pure “irritant contact dermatitis” – without atopic
skin manifestations in the history – was found in 138,
and pure “allergic contact dermatitis” manifested in
130 subjects.
25.2.2.2 Irritant Patch Testing
During the test period and within a follow-up of at
least 24 h, patients felt no relevant discomfort or pain
by the test. Clinical grade 5 (“very marked erythema/
vesiculation/edema and/or 5 erosions or necrosis”)
was never observed. If reactions were positive after
20 min then after 24 h sometimes minimal erythema
or solitary, small superficial erosions were detected,
but no necrosis, scarring, infection, or discoloration.
Healing was uneventful: in 43.6% of test-positive individuals skin reactions were completely reversed after 24 h. Thus, as expected, agreement between 2×10min- and 24-h observation concerning identifiable
skin changes was poor (Cohens κ=0.38). Therefore
test reading after the second 10-min NaOH-provocation phase was considered relevant for clinical outcome.
The time course of the biophysical parameters
within the observation period (24 h) with respect to
the results of visual scoring after 2×10 min NaOH
challenge is shown in Fig. 2 (clinical observations
dichotomized). The differences were statistically significant for ∆−TWL after 10 min (z=–7.3; p<0.001),
Fig. 4. ROC curve for ∆-TWL after 2×10 min NaOH and clinical outcome after 20 min (n=572). X-axis: inverse specificity
(equals false-positive classifications)
20 min (z=–12.0; p< 0.001), 24 h (z=–7.6; p<0.001)
and for ∆−RSM (10 min: z=–4.6; p<0.01; 20 min
z=–2.5; p<0.01; 24 h: z=–3.0; p<0.01); (Mann-Whitney U test). In the group of patients who reacted
clinically after NaOH challenge, there was an increase
of TWL after 20 min, which was still detectable after
24 h, whereas ∆−RSM showed after an initial increase
a decline 24 h after NaOH provocation, most likely
due the exsiccation following the barrier impairment.
Differences in measurements in the test (NaOH) and
control areas (NaCl) were for both biophysical parameters significant at all observation intervals (p<0.05;
Wilcoxon test).
If the result of visual scoring is examined in detail,
a good correlation between the degree of clinical reactivity and ∆−TWL can be demonstrated. Correlation
was best after 20 min NaOH provocation (rs=0.587;
p<0.01 [Spearman rank correlation]; Fig. 3). Correlation was poor at all measurment intervals for ∆−RSM
(20 min: rs=0.106).
No significant correlations could be detected between age or sex and clinical or biophysical findings,
nor any job-specific effects.
A relevant question is whether biophysical parameters can predict clinical outcome. In order to answer
this question, a ROC analysis was conducted. For
the parameter ∆-TWL after 2×10 min NaOH, the
ROC curve is shown in Fig. 4. As a cut-off point, a
∆-TWL of 2 g/m2h was determined; for this cut-off
point (“critical value”) TWL has a sensitivity of 76.7%
and a specificity of 74.7% (C=0.83) for prediction of
the target variable (“clinical outcome after 2×10 min
NaOH”). ROC analysis for ∆-TWL after 10 min
215
216
Swen Malte John, Hans J. Schwanitz
Fig. 5a, b. Clinical (a) and biophysical (b) response with respect to the three main clinical diagnoses (n=516). a Percent of positive
clinical reactions after 2×10 min NaOH. b Boxplot of ∆−TWL after 2×10 min NaOH; (o) extremes, (*) outliers
NaOH for the target variable was not significant
(C=0.67). Also for ∆-RSM at all measuring intervals, there was no relevant ability to significantly discriminate between positive and negative clinical test
results.
25.2.2.3 30-Minute NaOH Challenge
At the beginning of the study, in 18 patients in whom
there were no or only minor clinical reactions and
only a small increase of TWL after 2×10 min NaOH
challenge, another 10-min NaOH provocation phase
was added; this approach corresponds to Burckhardt's
most recent modification of his original method [37,
10]. Thus it could be shown that in the investigated
subgroup it was not possible to gain relevant additional information; the agreement of clinical outcome
after 20 and 30 min was high, reflected by a Cohens
kappa-coefficient κ=0.87. Significant agreement was
also demonstrated for the parameter TWL (κ=0.88;
∆-TWL dichotomized at cut-off point [2 g/m2h]).
25.2.2.4 Reproducibility
Fourteen patients were examined twice in the recruitment-phase; the median of the interval between
the two separate investigations was 2 years (range,
1–3 years). In the NaOH challenge, there were similar clinical findings after 2×10 min NaOH (κ=0.66).
However, agreement of TWL values was below the
significance level (κ=0.43; ∆-TWL dichotomized at
cut-off point [2 g/m2h]).
25.2.2.5 Constitutional Risks: Atopy
If the degree of clinical reactivity after NaOH challenge is analyzed with respect to the presence of
“atopic skin disposition” there is a significant negative
association with the clinically unresponsive grades
“nil” and “soap effect” and a corresponding positive association with marked clinical reactivity (X32 =12.17;
p=0.007). This phenomenon is further elucidated by
the post-hoc comparison of the adjusted, standardized residuals (Table 2).
There was no such association with the other main
clinical diagnoses “irritant dermatitis” and “allergic
contact dermatitis” (Fig. 5a). Statistical analysis of the
cross-tabulation showed again that distribution was
unequal (X22=9.65; p<0.01). By post-hoc-comparison
of the adjusted, standardised residuals it was obvious
that this was only due to the parameter “atopic skin
disposition” (Table 3).
There was also a significant positive association
of “atopic skin disposition” and the variable ∆−TWL
at all measuring intervals (10 min NaOH: z=–2.69,
p<0.01; 20 min NaOH z=–3.17, p<0.01; 24 h z=–2.2,
p=0.02 [Mann-Whitney U test]). When the association with the other main clinical diagnoses was
evaluated, again there were differences detectable for
this variable; however, this was significant only after
20 min NaOH (Fig. 5b): X22=10.36, p=0.006 (Kruskal-Wallis test); in separate Mann-Whitney U tests
it could be confirmed that differences in distribution
were explained only by the parameter “atopic skin
disposition.” For the variable ∆−RSM, a significant
association with clinical diagnoses, single or in combination, was not detectable.
25 Functional Skin Testing: the SMART Procedures
Table 2. Association between the clinical response after 2×10 min NaOH and the parameter “atopic skin disposition.” Clinically
unresponsive grades 1 and 2 (“nil”, “soap effect”) are comprised
Atopic skin disposition
No
Clinic 20’ count NaOH
Nil/soap effect
Mild erythema
Count
181
104
Corrected residuals
3.3
–3.3
Count
Corrected residuals
65
58
–1.0
1.0
Marked erythema/
Count
78
86
Edema/vesicle
Corrected residuals
–2.8
2.8
Count
324
248
total
Total
Yes
285
123
164
572
Table 3. Association between the clinical response after 2×10 min NaOH and the three main clinical diagnoses
Clinical outcome 20“ NaOH
Negative
Diagnosis
Atopic skin disposition
Irritant contact dermatitis
Allergic contact dermatitis
total
25.2.3
Positive
Total
248
Count
104
144
Corrected residuals
–3.1
3.1
Count
76
62
Corrected residuals
1.7
–1.7
138
Count
73
57
130
Corrected residuals
1.9
–1.9
Count
253
263
Conclusions
Burckhardt's alkali resistance test – even in the less
rigid modifications [26] – was shown to induce colliquation necrosis in 1% [36]. The above-proposed
modification with only two 10-min NaOH challenges
did not produce relevant volunteer discomfort when
the above-defined break-up criteria were used. This is
a relevant finding, when considering that, unlike previous studies employing NaOH where small groups
of healthy individuals were studied [39, 24], we examined a large cohort of patients with former occupational dermatitis. Even in these patients with a high
likelihood of increased skin sensitivity, the test was
well tolerated and yet revealed remarkable individual
differences.
Minimal clinical reactions (erythema) were taken
as sufficient sign of positive skin reactivity, and clinical findings were objectified by current biophysical
techniques. TWL proved the biophysical variable
of relevance; whereas RSM was not equally useful.
The TWL reading after the second challenge period
(2×10 min NaOH) provided the most relevant re-
516
sults, longer provocations (30 min) or late readings
(24 h) did not yield more information, neither clinically nor biophysically. Prediction of clinical outcome
by TWL20 min NaOH was good. Thus both parameters
(clinical grading and TWL) can serve as internal controls, opening up a kind of stereoscopic view of the
test result. If in an individual clinical and biophysical findings are in agreement, measurements are to
be considered relevant; if not, especially if there are
clinical changes without a corresponding increase in
TWL, the investigation may be repeated. Generally,
results should be interpreted considering the complete range of anamnestic and physical findings in
the individual, the test providing valuable additional
information.
In the investigated cohort of 572 thoroughly examined patients with a history of former occupational
skin disease, the test was able to distinguish atopics
from nonatopics, thus identifying constitutional risks.
The detection of constitutional risks, especially atopy,
by irritant patch testing is a controversial issue [3–5, 6,
10, 15, 16, 29, 34]. The discrepancies of the investigators’ results may be explained by the kind and size of
217
218
Swen Malte John, Hans J. Schwanitz
cohorts tested, choice of irritants, dose, method, and
body site as irritant reactivity is a complex phenomenon, which is multifactorially influenced, depending on barrier function, inflammatory reactivity, and
restitutional capacity [1, 17]. The absence of a gold
standard for the assessment of skin hyper-reactivity is
closely related to the fact that there is still no uniform
pathogenic concept of this phenomenon. However,
there is no doubt that an individual genetic predisposition is of relevance. Recent results from a large questionnaire investigation in Denmark show that the
concordance rate of hand eczema was almost twice
as high in homozygotic compared to heterozygotic
twins in both sexes [8]. Also, reactivity to various irritants such as SLS, benzalkonium chloride and sapo
kalinus showed a significantly higher concordance
rate in identical twins than in fraternal twins [21];
similar findings were obtained with NaOH [19].
The above-proposed NaOH test is based on existing procedures, but is less time-consuming, less
harmful, and has a proven efficacy. The test was found
to be reproducible, though further experience has to
be gained for confirmation. We call our up-dated
version of Burckhardt’s test “swift modified alkali resistance test (SMART)”. Figure 6 gives an overview
on the recommended test protocol based on the obtained findings.
Fig. 6 Flow chart of the SMART. Clinical and biophysical
readings are to be conducted 10 min after each provocation
phase
25.3
Applications of the SMART:
Differential Irritation Test
It is beyond dispute that individuals with hyperirritable skin do exist [1, 17, 18, 24]. Using a test battery of various irritants, including sodium hydroxide,
Frosch identified patterns (cluster) of individual skin
susceptibility; in a group of 44 healthy volunteers
25% were deemed hypoirritable, 61% were normal,
and 14% were hyperirritable [18]. Frosch also dem-
onstrated that the functional integrity of the stratum
corneum is critical for the degree of individual irritability. The phenomenon of hyperirritability is related
to a genetic predisposition, but not necessarily identical with atopy. Besides this primary hyperirritability,
it is also widely accepted that there may develop secondary hyperirritability after former dermatitis. In
occupational dermatology, some patients with former,
healed dermatitis complain of experiencing ongoing
increased skin sensitivity. However, the clinician frequently in these cases cannot detect any skin impairment.
With the aim of objectifying secondary hyperirritability we conducted a pilot study. For this purpose,
the SMART was applied simultaneously to two different body areas in a comparative fashion:
– One area that was previously exposed to irritants
at the workplace: the back of the hand
– while the other was not: the ventral forearm
Forty-eight patients with completely healed, former
occupational eczema (aged 19–68 years; 22 female, 26
male) were investigated, 31 healthy volunteers (aged
21–58; 21 female, 10 male) who were not previously
exposed to relevant private or occupational skin hazards served as controls. The SMART was performed
as described above in parallel in the two test sites; furthermore, late readings were performed after 24 and
48 h. In addition to the SMART an SLS test (1%, 24 h
occlusion) was conducted in the two test sites, both
tests were clinically and biophysically monitored. The
study is described in detail elsewhere [23]. In summary, the test revealed that in the control group there
was reactivity only in the forearm in a minority of
(sensitive) individuals, as an indication of constitutionally impaired barrier function. However, there
was no reactivity in the dorsum of the hands in any of
the controls. Moreover, in the test group with former
(healed) dermatitis a subgroup of four persons (8.3%)
was detected where relevant clinical (and biophysical)
reactivity to the SMART occurred only in the dorsum
of the hands; these patients claimed to have observed
a remaining increased skin sensitivity.
It has long been known that there are marked regional variations of skin reactivity; they are attributed
to differences in keratinization and in the density
of epidermal shunts such as hair follicles and sweat
ducts [17]. Various studies, using different irritants,
have convincingly demonstrated that the back of the
hand is a relatively robust area, even in skin-sensitive
individuals [1, 13, 17, 30]. This is confirmed by our
data, comparing skin reactivity to the SMART in the
25 Functional Skin Testing: the SMART Procedures
Table 4. Interpretation of the differential irritation test (DIT), the SMART being applied to the forearm and back of hands simultaneously for objectifying secondary hyperirritability
forearm and the dorsum of the hand. In the minority of cases, where the normal hierarchy of skin sensitivity is absent (isolated reactivity in the hands), we
claim that this is strong evidence for remaining acquired hyperirritability. Table 4 outlines the rationale
of the above-described concept of comparative (differential) irritation testing in separate body locations.
Assessment of remaining subclinical hyperirritability
and differentiating between primary and secondary
hyperirritability is of great importance for medicolegal evaluations in occupational dermatology (i.e.,
claims, prognosis). Thus far, the differential irritation
test (DIT) is the first systematic methodical approach
to objectify secondary, subclinical skin hyperirritability for these purposes.
25.4
Outlook: Implications for
Medicolegal Evaluations in
Occupational Dermatology
Since 1994 in Germany, approximately 20,000 cases
of assumed occupational dermatoses are reported annually, which makes skin disorders by far the most
frequent occupational illnesses. Medical treatment
and job rehabilitation of occupational illnesses lies
in the hands of the public employers’ liability insurance, which is a specific branch of the social welfare
system. In dermatology, about 12,000 medicolegal
evaluations are done every year in order to distinguish between occupational and nonoccupational
skin diseases. Legal requirements for accepting oc-
cupational skin disease (BK 5101 [7]) are complex.
In the context of medicolegal evaluations, the diagnostics routinely performed are epicutaneous allergy
patch testing. While this latter method is well standardized [38], irritant patch testing is not. Nevertheless, various irritation tests are regularly conducted in
such evaluations; currently, in the majority of these
tests a great spectrum of modifications of the original
Burckhardt alkali resistance test is employed. A recent survey among German occupational dermatologists frequently involved in medicolegal evaluations
showed very heterogeneous opinions concerning the
relevance of irritant patch testing in such evaluations:
one-third strongly opposed such tests, one-third was
strongly in favor of them, the remaining third was
undecided [23]. It is likely, and could be proven by
an analysis of court cases, that the outcome of evaluations may be strongly influenced by the attitude of
the dermatologists as to whether irritant patch testing is considered relevant or not. We believe that this
validates our claim that there is a need to reach an
agreement regarding a common irritation test in occupational dermatology to ensure comparability and
equal management of patients in medical opinions.
Therefore, with the special indication of standardizing
diagnostics, we regard the swift modified alkali resistance test (SMART) as a serious candidate for routine
use in occupational dermatology. The SMART-procedures (Table 5) – as tools for assessing functional
integrity of skin barrier properties – could be another
step on the long road to developing a benchmark irritant patch test in occupational dermatology.
219
220
Swen Malte John, Hans J. Schwanitz
Table 5. A potential role for sodium hydroxide irritant patch
testing in occupational dermatology
1. SMART (forearm)
Identification of constitutional risks
Standard for medicolegal evaluations?
2. DIT (forearm vs dorsum of hand)
Distinguishing primary from secondary hyperirritability
Objectifying „acquired subclinical hyperirritability“
for medicolegal purposes (i.e., claims, prognosis)
References
1. Agner T. Noninvasive measuring methods for the investigation of irritant patch test reactions. Acta Derm Venereol,
Suppl 1992; 173:1–26
2. Bangha E, Hinnen U, Elsner P. Irritancy testing in occupational dermatology: comparison between two quick tests
and the acute irritation induced by sodium lauryl sulphate.
Acta Derm Venereol (Stockh) 1996; 76:450–452
3. Basketter DA, Miettinen J, Lahti A. Acute irritant reactivity
in atopics and non-atopics. Contact Derm 1998; 38:253–
256
4. Berndt U, Hinnen U, Iliev D, Elsner P. Is occupational
irritant contact dermatitis predictable by cutaneous bioengineering methods? Results of the Swiss metalworkers
eczema study (PROMETES). Dermatology 1999; 198:351–
354
5. Björnberg A (1968) Skin reactions to primary irritants in
patients with hand eczema. Göteburg, Isaacsons
6. Björnberg A (1974) Low alkali resistance and slow alkali
neutralization. Characteristics of the eczematous subject?
Dermatol 149:90–100
7. Brandenburg S, Schwanitz HJ, John SM (1999) Empfehlungen für die Begutachtung von Berufskrankheiten nach BK
5101. Dermatosen 47:109–114
8. Bryld LE, Agner T, Kyvik KO, Brondsted L, Hindsberger C,
Menné T. Hand eczema in twins: a questionnaire investigation. Br J Dermatol 2000; 142:298–305
9. Burckhardt W. Neue Untersuchungen über die Alkaliempfindlichkeit der Haut. Dermatologica 1947; 94:8–96
10. Burckhardt W. Praktische und theoretische Bedeutung der
Alkalineutralisations- und Alkaliresistenzproben. Arch
Klin Derm 1964; 219:600–603
11. Cohen J. A coefficient of agreement for nominal scales. Educational and Psychological Measurement 1960; 20:37–46
12. Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale, Erlbaum, 1988
13. Cua AB, Wilhelm KP, Maibach HI. Cutaneous sodium lauryl sulphate irritation potential: age and regional variability.
Br J Dermatol 1990; 123:607–613
14. Czernielewski A. L’importance pratique du test de résistance de la peau à l’action des alcalis. In: Symp Dermatol
Pragae cum participatione internationali de morbis cutaneis professionalibus, Prag 1960. 1962; Exerpta Lectionum,
pp 84–85
15. Den Arend JA, de Haan AFJ, Malten KE. Seasonal transepidermal water loss and impedance of forearm skin in
atopics and non-atopics. Contact Derm 1988; 19:376–390
16. Frosch PJ. Hautirritation und empfindliche Haut. Grosse
Verlag, Berlin, 1985
17. Frosch PJ. Clinical Aspects of Irritant Contact Dermatitis.
In: Rycroft RJG, Menné T, Frosch PJ (eds) Textbook of
contact dermatitis, 3rd edn. Springer Verlag, Berlin Heidelberg New York, 2001; pp 313–354
18. Frosch PJ, Pilz B. Irritant patch test techniques. In: Serup J,
Jemec GBE (eds) Handbook of non-invasive methods and
the skin. CRC press, Boca Raton, 1995; pp 587–591
19. Gloor M, Schnyder UW. Vererbung funktioneller Eigenschaften der Haut. Hautarzt 1977; 28:231–234
20. Green DM, Swets JA. Signal detection theory and psychophysics, revised edn. Huntington, New York, 1974
21. Holst R, Möller H. One hundred twin pairs patch tested
with primary irritants. Br J Dermatol 1975; 93:145–149
22. Iliev D, Hinnen U, Elsner P. Reproducibility of a non-invasive skin irritancy test in a cohort of metalworker trainees.
Contact Derm 1997; 36:101–103
23. John SM. Klinische und experimentelle Untersuchungen
zur Diagnostik in der Berufsdermatologie. Konzeption
einer wissenschaftlich begründeten Qualitätssicherung in
der sozialmedizinischen Begutachtung. In: Schwanitz HJ
(ed) Studien zur Prävention in Allergologie, Berufs- und
Umweltdermatologie (ABU 4). Universitätsverlag Rasch,
Osnabrück, 2001
24. Kolbe L, Kligman AM, Stoudemayer T. The sodium hydroxide erosion assay: a revision of the alkali resistance test.
Arch Derm Res 1998; 290:382–387
25. Lange N, Weinstock MA. Statistical analysis of sensitivity,
specificity, and predictive value of a diagnostic test. In: Serup J, Jemec GBE (eds) Handbook of non-invasive methods and the skin. CRC press, Boca Raton, 1995; pp 33–41
26. Locher G. Permeabilitätsprüfung der Haut Ekzemkranker
und Hautgesunder für den neuen Indikator Nitrazingelb
“Geigy”, Modizifierung der Alkaliresistenzprobe, pH-Verlauf in der Tiefe des stratum corneum. Dermatologica
1962; 124:159–182
27. Löffler H, Effendy I, Happle R. Epikutane Testung mit Natriumlaurylsulfat. Nutzen und Grenzen in Forschung und
Praxis. Hautarzt 1999; 50:769–778
28. Pinnagoda J, Tupker RA, Agner T, Serup J. Guidelines for
transepidermal water loss (TEWL) measurement. Contact
Dermatitis 1990; 22:164–178
29. Schultheiss E. Eigenuntersuchungen und Alkaliresistenzprobe. Arch Klin Exp Dermatol 1964; 219:638–648
30. Schulz D, Korting GW. Zur weiteren Kenntnis der Alkaliresistenz-Probe. Dermatosen 1987; 35:91–94
25 Functional Skin Testing: the SMART Procedures
31. Schwanitz HJ. Atopic palmoplantar eczema. Springer Verlag, Berlin Heidelberg New York, 1988
32. Spruit D, Malten KE. Injury to the skin by alkali and its
regeneration. Dermatologica 1966; 132:124–130
33. Spruit D, Malten KE. Estimation of the injury of human
skin by alkaline liquids. Berufsdermatosen 1968; 16:11–24
34. Tupker RA, Pinnagoda J, Coenraads PJ, Nater JP. Susceptibility to irritants: role of barrier function, skin dryness and
history of atopic dermatitis. Br J Dermatol 1990; 123:199–
205
35. Tupker RA, Willis C, Berardesca E, Lee CH, Fartasch M,
Agner T, Serup J. Guidelines on sodium lauryl sulfate (SLS)
exposure tests. A report from the standardization group
of the European Society of Contact Dermatitis. Contact
Derm 1997; 37:53–69
36. Ummenhofer B. Zur Methodik der Alkaliresistenzprüfung.
Dermatosen 1980; 28:104–109
37. Wacek A. Weitere Untersuchungen aus dem Gebiet der
Alkali- und Säureabwehr der Haut, Dermatologica 1953;
107:369–417
38. Wahlberg JE. Patch testing. In: Rycroft RJG, Menné T,
Frosch PJ (eds) Textbook of contact dermatitis, 3rd edn.
Springer, Berlin Heidelberg New York, 2001; pp 435–468
39. Wilhelm KP, Pasche F, Surber C, Maibach HI. Sodium hydroxide-induced subclinical irritation. A test for evaluating stratum corneum barrier function. Acta Derm Venereol (Stockh) 1990; 70:463–467
40. Wilson D, Berardesca E, Maibach HI. In vivo transepidermal water loss and skin surface hydration in association
of moisturization and soap effects. Int J Cosm Sci 1988;
10:201–211
221