Content uploaded by Christian Stab Jensen
Author content
All content in this area was uploaded by Christian Stab Jensen on Nov 06, 2018
Content may be subject to copyright.
Review Article
Systemic contact dermatitis after oral exposure to
nickel: a review with a modified meta-analysis
CHRISTIAN STAB JENSEN
1
,TORKIL MENNE
´
2
AND JEANNE DUUS JOHANSEN
1
1
National Allergy Research Centre, Gentofte Hospital, University of Copenhagen, 2900 Hellerup, Denmark,
2
Department of Dermatology, Gentofte Hospital, University of Copenhagen, 2900 Hellerup, Denmark
Systemic contact dermatitis can be elicited experimentally in nickel-sensitive individuals by oral
nickel exposure. A crucial point interpreting such experiments has been the relevance of nickel
exposure from drinking water and diet. The aim of this meta-analysis study on former nickel-
exposure investigations was to provide the best possible estimation of threshold values of nickel
doses that may cause systemic contact dermatitis in nickel-sensitive patients. 17 relevant investiga-
tions were identified, and statistical analyses were performed in a stepwise procedure. 9 studies were
included in the final dose–response analysis, which divided the studies into a homogenous middle
group of 5 studies and 2 groups of 2 studies with a higher and lower response frequency, respec-
tively, described by logistic dose–response curves shifted in parallel. On the basis of these curves,
calculations were made of the doses that, theoretically, would cause systemic contact dermatitis in
exposed nickel-sensitive patients. The results from the 2 most sensitive groups show that 1% of
these individuals may react with systemic contact dermatitis at normal daily nickel exposure from
drinking water and diet, i.e. 0.22–0.35 mg nickel.
Keywords: nickel allergy; oral exposure; systemic contact dermatitis. #2006 The Authors. Journal
compilation #2006 Blackwell Munksgaard.
Accepted for publication 31 October 2005
Systemic contact dermatitis is seen as flare-up of
previous dermatitis or de novo dermatitis similar
to allergic contact dermatitis. While systemic
contact dermatitis from medicaments is a well-
established entity, systemic reactions to metals,
particularly to nickel, has been regarded as con-
troversial (1, 2). 1 reason for this is that reactions
to a ubiquitous hapten, such as nickel, are not
easily identified with certainty because it is diffi-
cult to measure and control daily exposure
through drinking water and diet.
However, there is significant evidence that oral
intake of nickel can induce systemic contact der-
matitis in nickel-sensitive individuals. Flare-up
reactions of previous nickel patch test sites after
systemic oral exposure appear to be hapten spe-
cific (3). There appears to be a dose–response
relationship, and the overall trend from several
oral exposure studies is that the severity of the
reaction after exposure and the number of react-
ing nickel-sensitive individuals increase with
higher exposure doses (Table 1) (4–20). Until
now, nobody has tried to summarize the data
from the oral exposure studies to determine an
oral nickel exposure threshold for elicitation of
systemic contact dermatitis. A complicating fac-
tor, making it difficult to draw definite conclu-
sions, is the great variation in the protocols of the
different studies and in the degree and extension
of the individual’s clinical reactions.
Whether or not the nickel exposure from
drinking water and the diet can induce systemic
reactions is still controversial, as unphysiologi-
cally high nickel doses were used in most of the
conducted studies. The doses used in the studies
vary from 0.3 mg to 5.6 mg nickel (Table 1),
while normal daily dietary nickel intake has
been found to vary from 0.02 mg to 0.48 mg
nickel (21–32). However, some studies have
shown that an exposure dose equivalent to the
maximal nickel exposure through the diet (0.9 mg
nickel per day) (21) can induce systemic reactions
in some nickel-sensitive individuals (6, 8, 9, 13,
20). Additionally, dietary intervention studies
(5, 33–36) indicate that exposure to nickel in diet
and drinking water could be a contributing factor
to systemic contact dermatitis.
The aim of the conducted study was to deter-
mine the best possible threshold values for oral
nickel exposure which would elicit allergic
Contact Dermatitis 2006: 54: 79–86 #2006 The Authors
Printed in Singapore. All rights reserved Journal compilation #2006 Blackwell Munksgaard
CONTACT DERMATITIS
Table 1. Characteristics of the different studies
Criteria’s of inclusion Study design Group size Dose (mg Ni) Type of exposure and conditions Definition of positive reactions Time before evaluation References
PPT, HNA and HE DB, SE 1 12 0, 5.6 NiSO
4
in capsule, not fasting FUE, WHE and EB 6 + 30 hr Christensen and
Mo
¨ller (4)
PPT and HE DB, SE 1 28 0, 2.5 NiSO
4
in capsule, not fasting FUE and/or WHE 2–72 hr Kaaber et al. (5)
PPT, HNA and HE DB, SE 1 11 0, 0.6, 1.2, 2.5 NiSO
4
in capsule, not fasting FUE , WHE and EB 48–72 hr Kaaber et al. (6)
NPT* and HNA NB, SE 1 4, 1 7 4.0 NiSO
4
in capsule, not fasting FUE, WHE and EB 72 hr Veien and Kaaber (7)
PPT and HE DB, DE 1 10 0, 0.5 NiSO
4
in capsule, not fasting WHE 0–48 hr Jordan and King (8)
PPT and HE NB, SE 3 5 0.6, 1.25, 2.5 NiSO
4
in capsule, over night
fasting and 1 hr after
FUL, WHE and EB 0–72 hr Cronin et al. (9)
PPT and HE DB, DE 1 22 0, 2.0, 4.0 NiSO
4
in capsule, not fasting WHE 24 + 72 hr Burrows et al. (10)
Not documented DB, SE 1 49 0, 2.24 Not documented Not documented Not documented Bedello et al. (11)
Not documented DB, SE 1 20 0, 2.4 Not documented Not documented Not documented Sertoli et al. (12)
PPT, HNA and HE DB, SE
and DE
16, 2 10 0, 0.4, 2.5, 5.6 NiSO
4
in capsule, not fasting FUE and/or FUL and/or EK 72 hr Gawkrodger et al. (13)
PPT, HNA
and/or HE
NB, SE 1 10, 1 19 0, 2.5 NiSO
4
in capsule, not fasting FUE and/or FUPT and/or EB 24 + 48 hr Roduner et al. (14)
PPT and HNA DB, SE 1 131 0, 2.5 Ni-salt in tablet; not fasting FUE and/or FUPT and/or EB 0–96 hr Veien et al. (15)
PPT and HNA NB, SE 1 25 2.24 NiSO
4
in capsule, not fasting FUE and/or FUPT and/or EB 0–48 hr Santucci et al. (16)
PPT and ECZ DB, SE 1 19, 1 9 0, 2.5 NiSO
4
in capsule, not fasting FUE and FUPT 0–24 hr Mo
¨ller et al. (17)
PPT, HNA and HE NB, SE 2 20 0.8†
61
Ni in water, fasting 12 hr
before and 4 hr after
FUE and/or WHE and/or EK 0–72 hr Nielsen et al. (18)
PPT DB, SE 2 10, 1 9 0, 1.0, 4.0 NiSO
4
in capsule, over night
fasting and 1 hr after
FUPT 24 hours Hindse
´n et al. (19)
PPT and HE DB, SE 4 10 0, 0.3, 1.0, 4.0 NiSO
4
in capsule, fasting
12 hr before
FUE and/or FUPT and/or EB 24 hr Jensen et al. (20)
Inclusion criteria: ECZ ¼eczema; HE ¼hand eczema; HNA ¼history of nickel allergy; NPT ¼negative patch test; PPT ¼positive patch test; design DB ¼double blinded; NB ¼not
blinded; SE ¼single-exposure; DE ¼double exposure over 2 days. Definition of positive reaction: FUE ¼flare-up of former eczema; FUPT ¼flare-up of former patch test; WHE ¼
worsening of hand eczema; EB ¼erythema on the body.
*Patients with negative routine patch tests were challenged orally with nickel.
†Average exposure dose. Persons were exposed to 12 mg/kg bodyweight.
80 JENSEN ET AL. Contact Dermatitis 2006: 54: 79–86
contact dermatitis in nickel-sensitive individuals
by combining the data from published oral expo-
sure studies in a meta-analysis.
Methods and Statistical Analyses
Relevant studies published from January 1966 to
November 2004 were identified by searching
MEDLINE, EMBASE and BIOSIS using the
words: nickel, oral, challenge, exposure, hyper-
sensitivity, sensitivity, allergy, systemic provo-
cation and contact dermatitis in different
combinations. Additionally, the databases were
searched by the use of the ‘Related Articles’ func-
tion and by searching the publications by specific
authors in the scientific area. Finally, relevant
references were localized by the use of the refer-
ence lists in review articles. Though probably
most of the relevant publications were localized
by these methods, this study is not a formal
meta-analysis (or series of meta-analyses), which
includes all publications on the subject.
Standard statistical methods based on the
binomial distribution of the number of positive
reactions were used for the calculation of confi-
dence intervals for response frequency and for
tests of differences in response between doses
within the same study or between 2 studies at
the same dose.
Logistic regression analysis was applied to esti-
mate dose–response relations and for comparison
of dose–response relations between studies. These
methods are based on the assumption that
patients are randomly divided into the various
dose groups. It should be pointed out that the
dose–response modelling analysis does not apply
to the experimental situation where the same
group of patients is tested with different doses,
including a dose of 0 mg nickel (placebo). In this
case, it would be a natural assumption that if a
patient has shown a positive reaction to a given
dose, that person will also show a positive reac-
tion to all higher doses. This may be checked by
inspection of the individual responses to the
doses in question, but the reported response
rates do not contain this information except
when none or all patients react.
Paired responses to 2 different doses were com-
pared by McNemar’s test, while Fisher’s test or
Chi-square tests were used for unpaired responses
(data not shown). Multiple logistic regression
analysis was applied to determine which studies
deviate significantly from the mean under the
assumption that the slopes of the dose–response
curves were similar. However, since it is not cor-
rect to include the same patients more than once
in this analysis, it was chosen only to use the
response rate closest to 50% from the studies
where a single group of patients were tested
with more than 1 nickel dose. The analysis was
performed according to the backwards elimina-
tion procedure as well as the forward selection
procedure. In the backwards elimination, the
study that deviates the least as well as insigni-
ficantly (P>0.05) from the mean of the others is
regarded as representing the mean, and in the
next step the least non-significantly deviation
group among the remaining studies is accepted
as representing the mean, etc. This pooling of
studies stops when only significantly deviating
studies remain. These studies may also be identi-
fied by the forward selection procedure, where
the study (or the nickel dose–response effect)
that deviates most and significantly from the
mean is selected first. In the next step, the most
significant among those remaining is selected, etc.
The selection stops when none of the remaining
studies are statistically significant, i.e. P>0.05.
Finally, estimates of ED
50
and other lower
(threshold) doses were calculated from the logis-
tic dose–response curves by the use of Fieller’s
method. The statistical analyses were with
APL2000
1
software (Maneugistics, Rockville,
MD, USA) validated relative to SAS
1
version
8.2 procedures (Proc Freq and Proc Univariate,
SAS Institute, Cary, USA, 2001).
Results
From the literature search, 17 studies on oral
nickel exposure in nickel-sensitive patients were
identified (Table 1). Additional 2 studies were
found (37, 38), but these were excluded because
the nickel-exposure doses used varied both intra-
and interindividually. The aim of the statistical
analysis was to conduct a comparative quantita-
tive dose–response analysis including as many of
the identified studies, which was proven difficult
because of the heterogeneity amongst the identi-
fied studies. The 17 studies included 496 indi-
viduals and varied in size from 11 to 131
individuals. Different protocols have been used
in the various studies, and so the observed differ-
ences between the studies may be caused by dif-
ferences in testing and evaluation methods as well
as differences in the tested patient populations.
The distribution of the number of positive
reactions with 95% confidence limits plotted
against the dose in logarithmic scale showed
that the limits in most cases were rather wide
due to the relative small number of patients
(data not shown). There was a clear tendency to
increasing reaction rate with increasing dose, but
a more detailed analysis was needed to evaluate
Contact Dermatitis 2006: 54: 79–86 SYSTEMIC CONTACT DERMATITIS 81
how the dose–response relations might differ
between the studies.
Some of the studies had positive responses to
placebo [studies 8, 10, part of 13 (the study con-
sisted of both single and double exposure), 15, 17
and 20]. A comparison of these reactions demon-
strated differences between single studies and
groups of studies both among studies using
double exposure and single exposure (data not
shown). Since the majority of the studies used
single exposure, it was more rational to continue
the analyses with these studies. This meant that
the data from studies 8, 10 and parts of 13 were
not included (Table 1).
A stepwise multiple logistic regression analysis
was performed on the relevant data from 401
patients in the 15 studies where single exposure
was used. The result showed a statistically signif-
icant dose–response effect (P<0.0001), and it
was found that the studies 9, 14, 15 and 17
deviated significantly from the remaining 11
studies, which showed an acceptable agreement
(Fig. 1). The 4 deviating studies could be divided
into 1 group with 3 studies (studies 14, 15 and 17)
and 1 separate study (study 9). The group com-
prising the 11 studies had an intermediary dose–
response curve, while the group of 3 had a lower
dose–response curve, and the separate study had
a higher response rate. The hypothesis of parallel
dose–response curves was acceptable. However,
the analysis has 2 weaknesses. The data included
2 studies without placebo testing (studies 16 and
18), and it contained 3 studies (studies 15, 17 and
20) which had positive responses to placebo.
It is an open question whether it from a biolo-
gical and statistical point of view is meaningful to
compare the dose–response relations from all the
single-exposure studies, when the response rates
to placebo is positive in some studies, and when
some studies have no placebo testing included.
Therefore, after exclusion of these 5 studies, an
additional analysis was performed on the data
from the remaining 9 studies comprising 171
patients. This altered the result of the stepwise
multiple logistic regression analysis. In a homo-
geneous group (studies 5, 6, 7, 11 and 13), an
intermediary dose–response curve was found,
while 2 groups each comprising 2 studies had a
higher (studies 4 and 14) or a lover dose–response
curve (studies 12 and 19) (Fig. 2). These 2 groups
included only 1 of the deviating studies from the
previous analysis.
On the basis of the result from the last analysis,
exposure doses corresponding to response rates
of 50% (ED
50
), 25% (ED
25
), 10% (ED
10
), 5%
(ED
5
) and 1% (ED
1
) of exposed nickel-sensitive
eczema patients were calculated (Table 2). These
calculations predict that an oral nickel exposure
with 0.22 mg, 0.35 mg or 0.53 mg nickel
(depending on which dose–response curve is
used) will make 1% of nickel-sensitive patients
respond. Similarly, 2% of these patients will react
if they are exposed orally to nickel doses of
0.29 mg, 0.45 mg or 0.70 mg, 5% will react to
0.41 mg, 0.65 mg or 1.0 mg nickel, while an
exposure to 0.55 mg, 0.87 mg or 1.33 mg nickel
will make 10% of the patients react with signs of
a systemic contact dermatitis.
Discussion
An extensive literature search identified dose–
response data from 17 published studies of oral
0.1 1 10
Ni dose lo
g
scale (m
g
)
0
20
40
60
80
100
Per cent
0.2 0.5 2 5 Study 20
Study 19
Study 18
Study 17
Study 16
Study 15
Study 14
Study 13
Study 12
Study 11
Study 9
Study 7
Study 6
Study 5
Study 4
Fig. 1. Observed rates of response with 95% confidence
limits and fitted dose–response curves in studies with single
exposure. Studies 4, 5, 6, 7, 11, 12, 13, 16, 18, 19 and 20 form
a homogenous group with a common dose–response curve
(in black), studies 14, 15 and 17 form another homogeneous
group (in green) and study 9 forms a third group (in red),
which are all significantly different.
10
Ni dose in lo
g
scale (m
g
)
0
20
40
60
80
100
Per cent
210.50.20.1 5
Study 19
Study 14
Study 13
Study 12
Study 11
Study 7
Study 6
Study 5
Study 4
Fig. 2. Observed rates of response with 95% confidence
limits and fitted dose–response curves in studies with single
exposure and no placebo reactions. Studies 5, 6, 7, 11 and 13
form a homogenous group with a common dose–response
curve (in black), studies 4 and 14 form another homogenous
group (in green), and studies 12 and 19 form a third group
(in red), which are all significantly different.
82 JENSEN ET AL. Contact Dermatitis 2006: 54: 79–86
nickel exposure on patients sensitive towards
nickel. These studies varied in criteria of inclu-
sion, design of the exposure protocol, numbers of
exposure boluses, how the same patients were
tested, inclusion of and reactions to placebo, fast-
ing condition prior to exposure, definition of
positive reaction to exposure and the time of
evaluation after exposure, which made a strict
meta-analysis including all studies impossible.
Therefore, it was necessary to modify the analysis
and exclude data from studies which differed
significantly from the majority to be able to
determine the best possible threshold values for
oral nickel exposure, which would elicit allergic
contact dermatitis in nickel-sensitive individuals.
In the overall evaluation of the identified
studies, there seem to be a tendency that increas-
ing exposure doses results in increasing reaction
rates, but a more detailed analysis was needed to
evaluate how the dose–response relations might
differ between the studies.
After conducting several analyses and exclud-
ing studies with double exposure, studies with
responses to placebo and studies without placebo
testing, the final model included 9 studies with
oral nickel exposure conducted on a total of 171
individuals. The result was a homogeneous group
including 5 studies (studies 5, 6, 7, 11 and 13)
with an intermediary dose–response curve, while
2 groups each including 2 studies had a higher
(studies 4 and 14) or a lower dose–response
curve.
On the basis of the result of the last modified
meta-analysis, theoretical exposure doses predict
that an oral nickel exposure with 0.22 mg,
0.35 mg or 0.53 mg nickel (depending of which
dose–response curve is used) will make 1% of
nickel-sensitive patients respond. Similarly, 10%
of these patients will react if they are exposed
orally to nickel doses of 0.55 mg, 0.87 mg or
1.33 mg nickel (Table 2).
It should be emphasized that the tested individ-
uals included in the studies do not represent
the majority of the population but are persons
allergic to nickel, who have consulted a
dermatologist. In the majority of the cases, the
patients had symptoms which could indicate a
systemic component in their disease, and many
patients had had chronic hand eczema (Table 1)
which flared up after the oral exposure. Using
the 2 ‘most sensitive’ statistical calculations, the
model predicts that 1% of these patients can be
expected to get a systemic reaction to the nickel
content of a normal diet, e.g. exposures between
0.22 mg and 0.35 mg nickel and that 10% of
them would react to exposures between 0.55 mg
and 0.89 mg nickel which theoretically could be
attained by consuming food with a high level of
nickel, by drinking water contaminated with
nickel from the pipes or taps and/or by drinking
a large volume of water containing a high level of
nickel while fasting.
Nickel exposure from drinking water and other
beverages is usually considered to be negligible,
although exceptions occur. In most studies, the
nickel levels in groundwater and drinking water
have been in the range of 1–10 mg per litre
(39–45). However, some studies have found
water samples containing higher levels.
Andersen et al. (44) found the nickel levels in
most of the tap water samples from the
Copenhagen area to be lower than 35 mg per
litre, but there were considerable variations in
the amount of nickel at different times of day,
from tap to tap and in hot versus cold water.
A statutory order on water quality and inspection
of water supply facilities from 2001, the Danish
Ministry of Environment (46) lowered the max-
imum level of nickel permitted in drinking water
from 50 mg per litre to 20 mg per litre. A recent
survey (47) found that the nickel concentrations
in the water in 2 out of 20 water supply facilities
and in 2 out of 37 sources exceeded the demands
to the quality of the drinking water. Andersen
et al. (44) found that an additional factor which
increases the nickel level in drinking water is
that the longer the water is allowed to be in the
pipes and the taps the more nickel will be
dissolved from the metal to the water. They
found concentrations of nickel up to 490 mg per
Table 2. Estimated doses of ED
50
and lower responses with 95% confidence limits
Studies 12 and 19 Studies 5, 6, 7, 11 and 13 Studies 4 and 14
Ni doses (mg) Estimate 95% limits Estimate 95% limits Estimate 95% limits
ED
50
1.27 0.77–172 2.00 1.47–2.35 3.06 2.15–4.30
ED
25
0.83 0.37–1.19 1.32 0.68–1.69 2.01 1.13–2.74
ED
10
0.55 0.17–0.86 0.87 0.31–1.26 1.33 0.53–1.94
ED
5
0.41 0.10–0.70 0.65 0.18–1.04 1.00 0.31–1.57
ED
2
0.29 0.05–0.54 0.45 0.09–0.81 0.70 0.16–1.21
ED
1
0.22 0.03–0.45 0.35 0.05–0.67 0.53 0.09–1.00
Contact Dermatitis 2006: 54: 79–86 SYSTEMIC CONTACT DERMATITIS 83
litre in the first 250 ml of drinking water. After
having the tap run for 5 min, they found much
lower nickel levels, which could indicate that
nickel is released from the pipes and taps and
concentrated in still water.
If only the maximum level of nickel permitted
in drinking water in Denmark at the moment is
obtained (20 mg per litre), a person can consume
up to 40–60 mg nickel per day through the drink-
ing water. According to the findings in the final
model, this will be the equivalent of 1/4 of the
dosage of nickel which will make 1% of the most
sensitive nickel-allergic patients seen by derma-
tologists react with a systemic contact dermatitis
response. If a relatively large part of the daily
water intake is consumed on an empty stomach,
e.g. if a person starts the day with a cup of coffee
or tea before eating breakfast, the uptake and
contribution of nickel from the drinking water
will probably be of relatively more importance
and therefore contribute more than 1/4 of the
dosage that can result in a systemic response.
The different outcomes of the studies may
reflect variations in study design and in differ-
ences in the patch tested nickel-sensitive individ-
uals. Different protocols have been used in the
various studies, and some of the studies were not
conducted in a double-blind manner. In many of
the studies, only a very small number of subjects
have been tested. The criteria for the selection of
test persons have not been standardized. In some
studies, a positive patch test to nickel was the
only inclusion criterion; 1 study (study 7) was
performed on subjects with hand eczema but
with a negative patch test, in others the subjects
had a positive patch test and hand eczema and/or
a history of nickel dermatitis. Neither have the
criteria been standardized for identifying a posi-
tive reaction to the oral challenge. The time
between challenge and evaluation is not the
same in the various studies, varying from 0 hr
to 72 hr or more. In studies with the shortest
time frames, this could have led to an underesti-
mation of the number of subjects reacting to the
oral challenge. It has not always been specified
whether the subjects had fasted overnight or
whether there were other dietary restrictions. In
some studies, a single bolus challenge was used,
while in others they used a double bolus chal-
lenge. The absorption and biokinetics following
the bolus exposure employed in most oral chal-
lenge studies may be quite different from natu-
rally occurring, dietary exposure, where nickel is
absorbed incrementally throughout the day. The
natural dietary intake of nickel is difficult to
estimate, and in most of the oral challenge stud-
ies, it was not taken into account. Total dietary
nickel intake and absorption can be highly vari-
able and difficult to estimate precisely. This
depends both on the composition of the diet
and on such factors as how the food had been
prepared, whether it was fresh or canned food
and/or whether it was contaminated during pro-
cessing or by kitchen utensils. There may also be
regional differences in the nickel content of various
foods. In most cases, there is probably only a small
amount of nickel in drinking water, but as shown in
some studies, drinking water may be contaminated
with nickel. A particular problem in the performed
modified meta-analysis is that no distinction was
made between the different end points (definition
of positive reaction to exposure) in the various
studies. 1 used end point in some studies, flare-up
of a recurrent vesicular hand eczema, which is the
most common clinical manifestation of systemic
nickel contact dermatitis, isaverycomplexdisease
and may be precipitated by a number of stimulants.
It has not been attempted to extend the logistic
dose–response analysis to include variables
describing differences between the studies such
as those just mentioned. From a statistical point
of view, the main problem is undoubtedly the
variation in background responses between the
studies and in particular how to incorporate this
heterogeneity into a general dose–response ana-
lysis. Therefore, the decision was made to exclude
studies with positive responses to placebo and
studies without placebo testing, which meant
that some reliable studies may not have been
used in the final analysis (studies 15, 16, 17, 18
and 20). However, adjusting for a placebo effect
would probably shift a theoretical dose–response
curve in parallel to the right, resulting in higher
threshold values and thus making it impossible to
note responses to the lower exposure doses.
It is believed that the analyses performed
represent a reasonable and at the moment best
possible systematic approach given the limited
and rather heterogeneous data.
In conclusion, oral exposure to nickel may
lead to systemic contact dermatitis in a dose-
dependent fashion. The amount of nickel in
food and drinking water will under normal
circumstances only provoke a systemic reaction
in a minority of the nickel-allergic patients seen
be dermatologists.
Acknowledgements
We thank Aage Vølund, PhD, Department of
Biostatistics, Novo Nordisk, Bagsværd, for assistance
with the statistical analyses of the data. The study was
supported by a grant from the Danish Environmental
Protection Agency.
84 JENSEN ET AL. Contact Dermatitis 2006: 54: 79–86
References
1. Burrows D. Is systemic nickel important? J Am Acad
Dermatol 1992: 26: 632–635.
2. Mo
¨ller H. Yes, systemic nickel is probably important! JAm
Acad Dermatol 1993: 28: 511–513.
3. Mo
¨ller H, Bjo
¨rkner B, Bruze M. Gold allergy and artificial
sweat. Contact Dermatitis 1999: 41: 57–60.
4. Christensen O B, Mo
¨ller H. External and internal exposure
to the antigen in the hand eczema of nickel allergy. Contact
Dermatitis 1975: 1: 136–141.
5. Kaaber K, Veien N K, Tjell J C. Low nickel diet in the
treatment of patients with chronic nickel dermatitis. Br J
Dermatol 1978: 98: 197–201.
6. Kaaber K, Menne
´T, Tjell J C, Veien N K. Antabuse
1
treatment of nickel dermatitis. Chelation – A new principle
in the treatment of nickel dermatitis. Contact Dermatitis
1979: 5: 221–228.
7. Veien N K, Kaaber K. Nickel, cobalt and chromium sensi-
tivity in patients with pompholyx (dyshidrotic eczema).
Contact Dermatitis 1979: 5: 371–374.
8. Jordan W P, King S E. Nickel feeding in nickel-sensitive
patients with hand eczema. J Am Acad Dermatol 1979: 1:
506–508.
9. Cronin E, Di Michiel A D, Brown S S. Oral challenge in
nickel-sensitive women with hand eczema. In: Nickel
Toxicology, Brown S S, Sunderman F W Jr (eds). New
York, Academic Press, 1980: 149–152.
10. Burrows D, Creswell S, Merrett J D. Nickel, hands and hip
prostheses. Br J Dermatol 1981: 105: 437–444.
11. Bedello P G, Goitre M, Cane D, Roncarolo G. Nickel: a
ubiquitous hapten. G Ital Dermatol Venereol 1985: 120:
293–296.
12. Sertoli A, Lombardi P, Francalanci S, Gola M, Giorgini S,
Panconesi E. Effect of oral administration of haptens in
sensitized subjects with contact allergic eczema. (I).
Update. G Ital Dermatol Venereol 1985: 120: 207–212.
13. Gawkrodger D J, Cook S W, Fell G S, Hunter J A. Nickel
dermatitis: the reaction to oral nickel challenge. Br J
Dermatol 1986: 115: 33–38.
14. Roduner J, Haudenschild-Falb E, Kunz E, Hunziker T,
Krebs A. Perorale Nickelprovokation bei nichtdyshidrosi-
formem und dyshidrosiformem Nickelekzem. Hautarzt
1987: 38: 262–266.
15. Veien N K, Hattel T, Justesen O, Nørholm A. Oral chal-
lenge with nickel and cobalt in patients with positive patch
tests to nickel and/or cobalt. Acta Derm Venereol 1987: 67:
321–325.
16. Santucci B, Cristaudo A, Cannistraci C, Picardo M. Nickel
sensitivity: effects of prolonged oral intake of the element.
Contact Dermatitis 1988: 19: 202–205.
17. Mo
¨ller H, Ohlsson K, Linder C, Bjo
¨rkner B, Bruze M. The
flare-up reactions after systemic provocation in contact
allergy to nickel and gold. Contact Dermatitis 1999: 40:
200–204.
18. Nielsen G D, Søderberg U, Jørgensen P J, Templeton D M,
Rasmussen S N, Andersen K E, Grandjean P. Absorption
and retention of nickel from drinking water in relation to
food intake and nickel sensitivity. Toxicol Appl Pharmacol
1999: 154: 67–75.
19. Hindse
´n M, Bruze M, Christensen O B. Flare-up reactions
after oral challenge with nickel in relation to challenge dose
and intensity and time of previous patch test reactions.
J Am Acad Dermatol 2001: 44: 616–623.
20. Jensen C S, Menne
´T, Lisby S, Kristiansen J, Veien N K.
Experimental systemic contact dermatitis from nickel: a
dose–response study. Contact Dermatitis 2003: 49: 124–132.
21. Nielsen G D, Flyvholm M A. Risks of high nickel intake
with diet. IARC Sci Publications 1984: 53: 333–338.
22. Schroeder H A, Balassa J J, Tipton I H. Abnormal trace
metals in man – nickel. J Chron Dis 1962: 15: 51–65.
23. Tipton I H, Stewart P L, Dickson J. Patterns of elemental
excretion in long term balance studies. Health Physics 1969:
16: 455–462.
24. Veien N K, Andersen M R. Nickel in Danish food. Acta
Derm Venereol 1986: 66: 502–509.
25. Larsen E H, Andersen N L, Møller A, Petersen A,
Mortensen G K, Petersen J. Monitoring the content and
intake of trace elements from food in Denmark. Food Addit
Contam 2002: 19: 33–46.
26. Fødevaredirektoratet. Nikkelallergi og mad. Ministeriet for
Fødevarer, Landbrug og Fiskeri 2004. 2. udgave. http://
www.foedevareallergi.dk.
27. Smart G A, Sherlock J C. Nickel in foods and the diet. Food
Addit Contam 1987: 4: 61–71.
28. Dabeka R W, McKenzie A D. Survey of lead, cadmium,
fluoride, nickel, and cobalt in food composites and estima-
tion of dietary intakes of these elements by Canadians in
1986–88. J AOAC Int 1995: 78: 897–909.
29. Biego G H, Joyeux M, Hartemann P, Debry G. Daily
intake of essential minerals and metallic micropollutants
from foods in France. Sci Total Environ 1998: 217: 27–36.
30. Stelz A, Lindemann P, Wojke B, Erbe T, Schulter P,
Muskat E, Most E, Pallauf J. Nutritional content in the
daily diet from selected nursing homes for the elderly in
the German state of Hessen. Second Report—minerals.
Z Erna
¨hrungswiss/Eur J Nutr 1998: 37: 269–278.
31. Ysart G, Miller P, Croasdale M, Crews H, Robb P, Baxter M,
de L’Argy C, Harrison N. 1997 UK Total Diet Study –
dietary exposures to aluminium, arsenic, cadmium, chro-
mium, copper, lead, mercury, nickel, selenium, tin and zinc.
Food Addit Contam 2000: 17: 775–786.
32. UK EGVM (Expert group on Vitamins and Minerals).
Intakes of nickel from food in the UK. Revised Review of
Nickel. EVM/99/24.REVISEDAUG2002. UK. August 2002.
33. Veien N K, Hattel T, Justesen O, Nørholm A. Dietary
treatment of nickel dermatitis. Acta Derm Venereol 1985:
65: 138–142.
34. Veien N K, Hattel T, Laurberg G. Low nickel diet: an open,
prospective trial. J Am Acad Dermatol 1993: 29: 1002–1007.
35. Antico A, Soana R. Chronic allergic-like dermatopathies
in nickel-sensitive patients. Results of dietary restrictions
and challenge with nickel salts. Allergy Asthma Proc 1999:
20: 235–242.
36. Nielsen G D, Jepsen L V, Jørgensen P J, Grandjean P,
Brandrup F. Nickel-sensitive patients with vesicular hand
eczema: oral challenge with a diet naturally high in nickel.
Br J Dermatol 1990: 122: 299–308.
37. Santucci B, Manna F, Cannistraci C, Cristaudo A,
Capparella R, Bolasco A, Picardo M. Serum and urine
concentrations in nickel-sensitive patients after prolonged
oral administration. Contact Dermatitis 1994: 30: 97–101.
38. Nielsen G D, Soderberg U, Jorgensen P J, Templeton D M,
Rasmussen S N, Andersen K E, Grandjean P. Absorption
and retention of nickel from drinking water in relation to
food intake and nickel sensitivity. Toxicol Appl Pharmacol
1999: 154: 67–75.
39. McNeely M D, Nechay M W, Sunderman F W Jr
Measurements of nickel in serum and urine as indices of
environmental exposure to nickel. Clin Chem 1972: 18: 992–995.
40. Punsar S, Era
¨metsa
¨O, Karvonen M J, Ryhanen A, Hilska P,
Vornamo H. Coronary heart disease and drinking water.
A search in two Finnish male cohorts for epidemiologic
evidence of a water factor. J Chron Dis 1975: 28: 259–287.
41. Clemente G F, Cigna Rossi L, Santaroni G P. Nickel in food
and dietary intake of nickel. In: Nickel in Environment,
Nriagu J O (ed). New York, John Wiley & Sons, 1980:
493–498.
42. Schuhmann H. The nickel contamination of surface and
potable water and some of its causal factors (Ger.). In: 3.
Spurenelement-Symposium Nickel, Anke M, Schneider H-J,
Contact Dermatitis 2006: 54: 79–86 SYSTEMIC CONTACT DERMATITIS 85
Bru
¨ckner C (eds). Jena, Karl-Marx-Universita
¨t, Leipzig,
Friedrich-Schiller-Universita
¨t, 1980: 155–161.
43. Me
´ranger J C, Subramanian K S, Chalifoux C. Survey for
cadmium, cobalt, chromium, copper, nickel, lead, zinc, cal-
cium, and magnesium in Canadian drinking water supplies.
J Assoc Off Anal Chem 1981: 64: 44–53.
44. Andersen K E, Nielsen G D, Flyvholm M A, Fregert S,
Gruvberger B. Nickel in tap water. Contact Dermatitis
1983: 9: 140–143.
45. Andersen J N, Stamer C. Fjernelse af metaller fra grund-
vand ved traditionel vandbehandling pa
˚danske
vandværker. Arbejdsrapport fra Miljøstyrelsen nr 1999: 117.
46. Miljø-og Energiministeriet. Bekendtgørelse om vandkvalitet
og tilsyn med vandforsyningsanlæg nr. 871 af 21. september
2001.
47. Nielsen K, Andersen A. Metal release to drinking water
(Metalafgivelse Til drikkevand), Environmental Project
No. 603, Danish EPA 2001. FORCE Institute. http://
www.mst.dk/udgiv/publikationer/2001/87-7944-526-8/html.
Address:
Christian Stab Jensen
National Allergy Research Centre
Gentofte Hospital
University of Copenhagen
Niels Andersensvej 65
DK-2900 Hellerup, Denmark
Tel: +45 39 77 73 00
Fax: +45 39 77 71 18
e-mail: csj@ssi.dk
86 JENSEN ET AL. Contact Dermatitis 2006: 54: 79–86
