Clinical relevance of food additives in adult patients with atopic dermatitis

Clinical relevance of food additives in adult patients with

atopic dermatitis

M. WORM, I. EHLERS, W. STERRY and T. ZUBERBIER

Department of Dermatology and Allergy, ChariteÂ Clinic, Humboldt University, Berlin, Germany

Summary

Background Adverse reactions to food play an important role in the pathogenesis of atopic

dermatitis (AD). In infancy and childhood, food allergies are observed in up to 30%,

whereas nonallergic hypersensitivity reactions (pseudoallergic reactions) towards food

additives have been reported to occur between 2 and 7%. By contrast, sensitizations

towards food allergens are rarely of clinical relevance in adults and little data is available on

nonallergic hypersensitivity reactions. To date the role of pseudoallergic reactions as an

aggravating factor in AD of adult patients remains controversial. However, many adult

patients report on food-related aggravation of the disease and nonallergic hypersensitivity

reactions have been incriminated repeatedly.

Objective To elucidate the relevance of food additives in adult patients suffering from

AD.

Methods Fifty patients were monitored over 4 weeks under regular diet followed by 6

weeks of a diet omitting known pseudoallergens. Skin status of patients was assessed every

2 weeks by a standardized scoring, and serum eosinophilic cationic protein (ECP) was

determined before and after diet.

Results Nine of ®fty patients dropped out, 26 showed a signi®cant improvement of the

Costa-score by 57%. In 23/26 patients a corresponding reduction of serum ECP level by

52% on average was determined. Responder patients (24/26) were orally challenged with

food rich in pseudoallergens followed by double-blind exposure to food additives (n� 15).

A worsening of the eczema was seen in 19/24 patients after intake of pseudoallergen-rich

food and in 6/15 patients after exposure to food additives.

Conclusion These results indicate that a subgroup of adult patients with AD clinically

improve on low-pseudoallergen diet but only a small subgroup respond to oral provocation

with food additives.

Keywords: atopic dermatitis, eosinophil cationic protein, food additives, nonallergic

hypersensitivity, pseudoallergy

Clinical and Experimental Allergy, Vol. 30, pp. 407±414. Submitted 23 February 1999;

revised 29 April 1999; accepted 7 May 1999.

Introduction

Atopic dermatitis (AD) is a chronic remittent, pruritic skin

disease with typical clinical appearance and distribution of

skin lesions. The incidence of AD has been continuously

increasing in recent decades and it is estimated that 10±15%

of the population of industrialized countries are currently

affected by AD [1]. In infancy and early childhood, food

allergies are observed in up to 30% of these patients [2].

However, sensitizations towards the classical food aller-

gens, such as cows' milk, egg, wheat, soy and nuts are rarely

of clinical relevance in older children or adults. Apart from

allergic mechanisms, also nonallergic hypersensitivity reac-

tions (pseudoallergic reactions) against food in adults

have been previously reported to cause an exacerbation of

disease in some patients [3±5]. The eliciting factors of

these hypersensitivity reactions are food additives, vasoac-

tive substances (e.g. histamine) and naturally occurring

Clinical and Experimental Allergy, 2000, Volume 30, pages 407±414

407q 2000 Blackwell Science Ltd

Correspondence: M. Worm, Hautklinik der ChariteÂ, Schumannstr. 20±21,

D-10117 Berlin, Germany.

substances such as salicylates, benzoates and other com-

pounds (such as aromatic compounds) in fruits, vegetables

and spices [6]. Pseudoallergens have been shown to induce

histamine, tryptase and other pro-in¯ammatory mediators

by mast cells in vivo [7]. The mechanisms are only partly

understood, but they are apparently based on direct activa-

tion of mast cells. The pseudoallergic reaction is not

immunoglobulin (Ig) E-mediated and does not need a

prior sensitization. Since in AD, mast cell numbers are

increased in lesional skin and due to the close vicinity of

in®ltrating T cells, a direct T-cell activation is feasible [8].

Thus, it has been shown that mast cell-derived mediators

and cytokines such as interleukin (IL)-4, IL-10 and IL-13

are capable to activate a TH2 cell response [9]. Whether also

abnormal T-cell functions, which are known to play an

important role in the pathogenesis of AD, are involved in

triggering skin in¯ammation by food additives needs to be

determined.

To date it has only been shown for type I allergens that

allergen-speci®c T cells in®ltrate lesional skin after antigen

exposure and produce various cytokines [10±14]. The initial

antigen-induced T-cell response leads to the production of

IL-4 and IL-5, whereas in chronic lesions IFNg and IL-12

production is predominant. Production of these cytokines

favours skin in¯ammation, which is clinically characterized

by worsening of the eczema. Therefore certain antigens with

occurrence of allergen-speci®c T cells in affected skin

lesions have been implicated to play an important role in

the pathogenesis of AD and can be effectively controlled by

strict avoidance of the allergen(s) [15].

While in children suffering from AD, the prevalence of

pseudoallergic reactions towards food additives has been

reported to occur between 2 and 7% [16], for adult patients,

the available data is limited. Thus the role of these non-

allergic hypersensitivity reactions as aggravating factors in

adult patients with AD is still a contentious issue. However,

many adult patients report food-related worsening of the

disease. Although nonallergic hypersensitivity reactions

against food and food additives have been incriminated,

no controlled studies about the clinical relevance of food

additives in adult patients with AD have been conducted.

In this study, adult patients suffering from AD were

followed over 6 weeks by skin-score and serum eosinophil

cationic protein (ECP) levels during the performance of an

low-pseudoallergen diet (Table 1). In patients responding to

the low-pseudoallergen diet by signi®cant improvement of

their skin status and decrease of serum ECP-levels the

clinical relevance of food additives was con®rmed by

double-blind, placebo-controlled food challenge.

Methods

Patients

Fifty individuals (15 male) ranging from 18 to 72 years

(mean age 28 years) were recruited from the outpatient

408 M. Worm et al.

q 2000 Blackwell Science Ltd, Clinical and Experimental Allergy, 30, 407±414

Table 1. Low-pseudoallergen diet

Allowed Forbidden

Basic food Additive-free bread, potatoes, All others (e.g. pasta with eggs, cake, biscuits,

rice, unprocessed cereals, ¯our (not potato chips, crisps)

self-raising), rice cakes, durum wheat

pasta (without egg)

Fats Butter, cold pressed plant oils All others (e.g. margarine, mayonnaise)

Milk products Fresh milk, cream without stabilizers, All others

white cheese, fromage frais, a small

amount of mild Gouda

Food from animals Fresh meat without seasoning All others including eggs, sea-food, smoked meat

Vegetables All except those listed as forbidden Artichokes, peas, mushrooms, spinach, rhubarb,

(e.g. lettuce, carrots, zucchini, tomatoes and tomato products, olives, sweet peppers, spinach

cabbage, broccoli, asparagus)

Fruit None All including dried fruits or fruit juices

Herbs, spices Salt, sugar, chives, onions All others including garlic and herbs

Sweets None All including chewing gum

Beverages Milk, mineral water, coffee, black tea All others including beer, wine, spirits and herbal teas

Spreadings Honey All others

Note: all food containing preservatives, dyes or antioxidants; all industrially processed food should be carefully checked regarding food

additives (Zuberbier et al. 1995) which strictly forbidden.

clinic of dermatology, allergy branch. Type I sensitizations

to major inhalation allergens were found in 41 of 50 patients.

Twenty-®ve patients showed positive skin prick tests to grass

pollen, 31 to birch pollen, 15 to mugwort pollen and 24 to

animal danders. In 18 of 41 patients with proven type

I sensitization the patient's history suggested a pollen-

associated food allergy (cross-reactivity between pollen and

certain foods, e.g. birch pollen and apple/nuts). Measurements

for total IgE ranged from 15 to over 2000 kU/L (mean level

1900 kU/L). Twenty-four of 50 patients had a positive family

history of atopy. All patients were proven to suffer from AD

as de®ned by the criteria of Hani®n and Rajka [17].

Study design

The study was divided into three sections. In order to

monitor the individual course of disease, the skin status of

each patient was documented over 4 weeks without dietary

intervention in phase 1. At the beginning and at the end of

this phase blood samples were drawn for determining serum

parameters (see below). In phase 2, the patients were set on

a low-pseudoallergen diet for 6 weeks. The diet was

introduced and explained to the patients by a nutritionist.

To ensure patients complied with the diet, they were told to

keep a diary, which was examined by a nutritionist at every

visit. At the end of this phase blood samples were drawn

again. In phase 3, oral provocation tests were performed

in the responder group (see below) to verify the success

of the diet. A group of non-responder patients were also

exposed to the oral provocation tests as a control group.

The patients were asked to stop antihistamine intake (if

used regularly) at least 5 days before the challenge. The

procedure of provocation was as follows.

First, pseudoallergen-rich food (Table 2) was given over

2 days. The meals contained the formerly forbidden food

additives and major naturally occurring pseudoallergens. If

no worsening of eczema with an increase of Costa Score

above 10 points occurred, the improvement of skin status

was judged not be related to the diet and further provocation

tests were not performed. In case of a positive reaction

(worsening of Costa Score above 10 points), pseudoallergic

reactions were suspected to be responsible and oral chal-

lenges with food additives followed after a washout time of

48 h, in order to identify the eliciting agents. The food

additives, as listed in Table 3, were given in capsules to

guarantee double-blind, placebo-controlled provocation

tests. Dosages were based on former provocation tests in

patients with suspected pseudoallergy [6]. Capsules contain-

ing all food additives were given in one single administra-

tion of challenge, the identical amount of placebo capsules

were given at another time point. The order was random-

ized. The time of observation was 48 h after each provoca-

tion. Before and after each provocation test, the skin status

was documented.

Food additives in atopic dermatitis 409


Table 2 Pseudoallergen-rich food (2 days)

1.

Breakfast

50 g muesli with 3±4 (25 g) dried apricots (containing sulphites) and 150 g fruit yoghurt (containing sorbic acid)

Lunch

150±200 g potato salad (containing benzoic and sorbic acid); 150 g ®sh salad (containing benzoic acid); 200 g fruit jelly (containing

colouring agents)

Tea

One chocolate bar, 330 mL cola (low calorie)

Supper

Two slices of bread, 20 g margarine (containing preservatives), three 30 g matured cheese (matured Gouda cheese, matured Camembert,

Blue cheese); 200 mL red wine

2.

Breakfast

Two slices of bread, 20 g margarine (containing preservatives), 40 g jam (containing preservatives

Lunch

100 g tomatoes, 50 g celery, 100 g sweet pepper, instant broth (containing glutamate), garlic, ®ve olives (containing colouring), with chips

or crisps; seasoning: paprika, thyme, oregano, pepper, nutmeg

Tea

150 g gooseberries, strawberries and red currents (fresh or frozen) with cinnamon

Lunch

Two slices of bread, 20 g margarine (containing preservatives), one slice of ham, one burger, tomato salad from 200 g tomatoes, oil,

vinegar, honey and dill; one apple

Skin status

The skin status was documented throughout the entire study

by the same dermatologist. The clinical scoring system used

was modi®ed from the method of Costa et al. [18]. In

accordance to their scoring system 10 intensity criteria

(extent of erythema, oedema, vesicles, crusts, excoriations,

scales, licheni®cation, pigmentation, pruritus, loss of sleep)

and 10 topography items were used. The score assessed the

intensity of single criteria and the affected skin area by a

scoring system of 0±3. In contrast, the Costa Score assesses

the intensity criteria from 0 to 7 and the distribution from 0

to 4. To compensate for the smaller range of our score

regarding the intensity criteria, we calculated the total score

by counting the sum of intensity criteria twice.

The change of skin status by dietary intervention was

measured by comparison of the scores before and after diet.

The mean of the three scores of phase 1 (without interven-

tion) was used as baseline, and compared with the score

after 6 weeks of diet. An improvement of > 35% was de®ned

as success of diet.

Serum samples

Total IgE was assessed of each patient by Pharmacia CAP

systems (Uppsala, Sweden). Blood samples were taken at

the ®rst, the third and the last visit. Serum ECP levels were

determined by an ELISA system (Pharmacia) according to

guidelines given by the manufacturer. The sensitivity of the

ELISA was 2 mg/mL.

Statistical analysis regarding the skin status and ECP

levels between responder and non-responder group was

performed using the Wilcoxon signed-rank test for unpaired

data and paired data, respectively, regarding statistical

analysis within the responder group.

Results

Of the 50 selected patients, 41 completed the study and were

thus included for evaluation: 26 patients improved on diet

(63% responder) while 15 patients did not pro®t (37% non-

responder). The nine patients who dropped out of the study

did not ®nish the diet phase, in most cases (n� 7) because of

dif®culties in adhering to the diet.

Skin status

In the responder group the mean skin score value before the

diet was 29 points and decreased after the diet to 11 points

410 M. Worm et al.


Pseudoallergens Name E-number Dose (mg)

Colouring agents

Azo-dyes Tartrazine E102 50

Sunset yellow E110 5

Azorubine E122 5

Amarante E123 5

Ponceau E124 5

Brilliant black BN E151 5

Other synthetic dyes Quinoline yellow E104 5

Erythrosine E127 5

Patent blue E131 5

Indigotine E132 5

Natural colours Iron (III) oxide E172 5

Red cochineal E120 5

Preservatives Sorbic acid E200 1000

Sodium benzoate E211 1000

P-hydroxybenzoate E214±219 1000

Sodium metabisulphite E223 50

Sodium nitrate E251 100

Antioxidants Butylhydroxyanisol (BHA) E320 50

Butylhydroxytoluol (BHT) E321 50

Propylgallate E310 50

Tocopherol E306±309 50

Taste enhancer Monosodiumglutamate E621 200

Naturally occurring

substances Salicylic acid 100

Table 3. Pseudoallergens used for provo-

cation tests (masked in gelatine capsules)

(P < 0.05). In the non-responder group no signi®cant change

of skin status or even worsening was seen: the mean value

before the diet was 27 points and after the diet 24 points

(Fig. 1). Signi®cant improvement of the Costa Score in the

responder group was also observed when responder and

non-responder group were aggregated (P< 0.05).

Serum parameter

Serum ECP levels of all patients ranged from 2 to 200 mg/

mL. Before diet the mean ECP level of the responder group

was 27 mg/mL, the mean level of the non-responder group

was also 27 mg/mL. While in the non-responder group no

signi®cant change in ECP level was observed after diet

(25 mg/mL), 23 of 26 responders showed a mean reduction

in ECP serum levels of 52% from 27 mg/mL before diet to

14 mg/mL after diet (P< 0.05) (Fig. 2).

The mean of total IgE level was 1900 KU/mL in the study

group (2266 KU/mL in the responder group and 951 KU/mL

in the non-responder group, respectively) and did not change

> 150 KU/mL in any patient during the entire study period.

Oral provocation tests

After diet, oral provocation tests were performed in 24/26

patients of the responder group (Fig. 3). Two patients

refused to be challenged. Additionally, 10 patients of the

non-responder group were exposed to provocation tests as a



70

60

50

40

30

20

10

0

70

60

50

40

30

20

10

0Before After Before After

(a) (b)

Co

sta

sco

re in

po

ints

Fig. 1 Skin score before and after the low

pseudoallergen diet in adult patients with

atopic dermatitis. (a) Responder group,

improvement > 35% of Costa Score before

and after the diet was de®ned as responder

(n� 26), mean score value before the diet

was 29 points and after the diet 11 points

(P < 0.05). (b) Non-responder group

(n� 15), mean score value before the diet

was 27 points and after the diet 24 points

1000

100

10

1

1000

100

10

1

EC

P (

µg/m

l)

Before After Before After

(a) (b)

Fig. 2. ECP values before and after the low-

pseudoallergen diet. (a) Responder group

(n� 26), mean ECP value before the diet

was 27 mg/mL and after the diet 14 mg/mL

(P < 0.05). (b) Non-responder group

(n� 15), mean ECP value before the diet

was 27 mg/mL and after the diet 25 mg/mL

control group. Both provocation tests were followed by an

observation time of 48 h.

In 19 of 24 patients, intake of pseudoallergen-rich food

over 2 days caused worsening of skin status. There were no

immediate-type reactions observed in any patient but solely

late-phase reactions, in most cases only after 24 h. No

patient of the non-responder group responded to exposure

of pseudoallergen-rich food.

In 15 of the 19 patients reacting to the ®rst challenge, oral

provocation tests with food additives followed. Ten non-

responder patients as a control group were exposed to food

additives and did not show a worsening of eczema. The oral

provocation tests with food additives were performed in a

double-blind, placebo-controlled fashion. At one challenge

capsules containing all food additives were given, at the

other challenge the same amount of capsules but ®lled with

silicium oxide mannit instead of food additives were

exposed. The order was randomized. Six of 15 patients

worsened after challenge with food additives, one patient

reacted to placebo. Again, no immediate-type reactions

were observed.

Discussion

In this paper, we show that 23/50 patients suffering from AD

will improve by their skin status and serum ECP levels on

performing a low-pseudoallergen diet. However, only a

small subgroup reacted to food additives proven by

double-blind placebo-controlled food challenge.

In 23 of 41 patients set on low-pseudoallergen diet, a

signi®cant improvement of the Costa Score (> 35%)

together with a decrease of serum ECP levels was observed.

Three patients improved signi®cantly by the Costa Score

(> 35%) but not by decrease in serum ECP levels. A clinical

bene®t was suggested in 19 of the former 23 patients by

open oral provocation tests with pseudoallergen-rich food.

However, double-blind, placebo-controlled food challenges

with encapsulated food additives could only prove intoler-

ance against food additives in six out of 15 tested patients.

Although challenge dosage was based on former provoca-

tion tests in patients with suspected pseudoallergy [6], in

patients with AD, the challenge might have been insuf®-

cient. This could be elucidated by performing the challenge

for a longer period of time (e.g. 2 days of repeated

challenge).

Because of the chronic undulating course of AD, studies

investigating the therapeutic effect of intervention measures

are dif®cult to assess and should be regarded critically. A

placebo effect of up to 30% should be taken into account as

according to Ehlers et al. [19], only intensive care by a

training course (without intervention measures) can con-

siderably improve the skin status and pruritus.

To con®rm the assessment of skin status by an objective

parameter ECP levels were determined in serum before

and after diet. In 23 of 26 patients of the responder group,

ECP levels were decreased by 52% after 6 weeks of

diet. Former studies indicate that serum ECP, as a mediator

of in¯ammation, correlates with the clinical assessment of

the skin [20±23]. However, patients' medication (antihista-

mines and medium to low-strength topical steroids), which

was monitored throughout the entire study did not differ

notably in either group. Taken together, there was a ten-

dency to a lesser use of drugs towards the end of the study in

both groups.

An intervention measure by diet cannot not be realized in

a double-blind, placebo-controlled study design, which is

demanded for an intervention study because of two main

reasons. Firstly, it is dif®cult to ®nd a suitable placebo for a

diet and secondly, there is no guarantee that a placebo-diet is

de®nitely without effect on the skin status in AD patients.

This becomes even more dif®cult as the placebo diet is not

to be easily identi®ed as placebo by the patient. The

therapeutic effect of such dietary intervention can therefore

only be con®rmed by provocation tests, which must be

carried out in a double-blind, placebo-controlled fashion,

which is regarded as the `gold standard' of diagnosis of food

allergies [24].

In the present study food additives were challenged in a

double-blind, placebo-controlled design whereas the provo-

cation tests with pseudoallergen-rich food were performed

in an open setting. Blinding of pseudoallergen-rich food is

very dif®cult, because it differs greatly in taste and looks

from low-pseudoallergen food.

Vieluf et al. [4] found an exacerbation of the skin status

by single-blind, placebo-controlled challenge with food

additives in 14 of 21 patients with atopic eczema. A study

investigating the value of a histamine-free diet con®rmed a

412 M. Worm et al.


45

40

35

30

25

20

15

10

5

0

Study group

Responder

PAR challenge

PAR challenge (+)

FA challenge

FA challenge (+)

Pat

ien

ts (

n)

Fig. 3. Results of provocation tests with pseudoallergen rich food

(PAR) and food additives (FA). None of the tested patients of the

non-responder group reacted to the provocation tests with either

pseudoallergen rich food or food additives.

bene®t in 25% of patients with AD [5]. The amount of

pseudoallergens was reduced in this histamine-free diet but

not as low as in our diet. This fact might explain the

difference in responsiveness of 25% to a histamine-free

diet and 40% in a low-pseudoallergen diet, although the

authors themselves suggest as explanation a histamine

intolerance caused by a lack of diaminoxidase. However,

this assumption has not yet been con®rmed. While 19/26 of

our challenged patients responded to pseudoallergen-rich

food, con®rmed by clinical assessment, only 6/15 reacted to

provocation with food additives. This difference points to

the relevance of naturally occurring pseudoallergens.

Apart from nonallergic hypersensitivity, pollen-associated

food allergy may also play an important role for the improve-

ment in the diet, because the diet excludes fruits, many

vegetables and most seasonings which are known to cross-

react with birch and/or mugwort pollen. In the majority of

patients in the responder group, a clinically relevant pollen-

associated food allergy may be deduced from the skin prick

test and/or the patient's history. Some responder patients

already experienced positive reactions to certain pollen-

associated foods before starting with the diet and avoided

them thereafter. We believe that birch pollen-related

foods, which are easily identi®ed by the patient as cross-

allergens and will be avoided thereafter, are not responsible

for the improvement of the low-pseudoallergen diet. Also,

grass-pollen related foods, such as ¯our, peanut and soy,

may only be of minor importance as these foods were

allowed during diet or at least not de®nitely permitted.

However, spices and herbs, which cross-react with mug-

wort, are not easily identi®ed by the patient in everyday

life. The high number of mugwort-sensitized patients in the

responder group (Table 4) supports the view that patients

with mugwort-sensitization might have bene®ted from the

diet mainly because of avoidance of mugwort-associated

foods. Further studies are necessary to provide a clear

cutting between nonallergic hypersensitivity and pollen-

associated (cross-allergic) reactions against food, mainly

mugwort-associated foods, in patients with AD who respond

to a low-pseudoallergen diet.

In conclusion, this study shows that a subgroup of

patients with AD will improve on a low-pseudoallergen

diet with regard to skin status and serum ECP levels.

However, only few patients react to oral provocation with

food additives in a double-blind, placebo-controlled

manner. A follow-up after 1 year will show whether the

low-pseudoallergen diet can provide a long-lasting bene®t.

References

1 Leung DYM. Atopic dermatitis: the skin as a window into the

pathogenesis of chronic allergic diseases. J Allergy Clin

Immunol 1995; 96:302±18.

2 Burks AW, Mallory SB, Williams LW, Shirrell MA. Atopic

dermatitis: clinical relevance of food hypersensitivity reac-

tions. J Pediatr 1988; 113:447±51.

3 Ring J, Przybilla B, Schwab U, Steger O. Klinisches Spektrum

der UÈ beremp®ndlichkeits-reaktionen gegen Sul®te. Allergologie

1987; 10:100±6.

4 Vieluf D, Przybilla B, Traenckner I, Ring J. Provocation of

atopic eczema by oral challenge tests with food additives.

J Allergy Clin Immunol 1990; 85:206.

5 Wantke F, GoÈtz M, Jarisch R. Die histaminfreie DiaÈt. Hautarzt

1993; 44:512±6.

6 Zuberbier T, Chantraine-Hess S, Hartmann K, Czarnetzki BM.

Pseudoallergen-free diet in the treatment of chronic urticaria ÿ

a prospective study. Acta Derm Venereol (Stockh) 1995;

75:484±7.

7 Murdoch RD, Pollock I, Naeem S. Food additive-induced

urticaria: studies of mediator release during provocation tests.

J R Coll Physicians Lond 1987; 21:262±6.

8 JaÈrvikallio A, Naukkarinen A, Harvima IT, Aalto M-L,

Horshmanheimo M. Quantitative analysis of tryptase- and

chymase-containing mast cells in atopic dermatitis and

nummular eczema. Br J Dermatol 1997; 136:871±7.

9 Franji P, Oskeritzian C, Cacaraci F et al. Antigen-dependent

stimulation of bone marrow-derived mast cells of MHC II-

restricted T cell hybridoma. J Immunol 1993; 151:6318±28.

10 VanReijsen FC, Bruijnzeel-Koomen CA, De Weger RA, Mudde

GC. Retention of long-lived, allergen-speci®c T cells in atopic

dermatitis skin lesions. J Invest Dermatol 1997; 108:530.

11 Eigenmann PA, Huang SK, Sampson HA. Characterization of

ovomucoid-speci®c T-cell lines and clones from egg-allergic

subjects. Pediatr Allergy Immunol 1996; 7:12±21.

12 Van Neerven RJ, Ebner C, Yssel H, Kapsenberg ML, Lamb JR.

T-cell responses to allergens: epitope-speci®city and clinical

relevance. Immunol Today 1996; 17:526±32.

13 Werfel T, Morita A, Grewe M et al. Allergen speci®city of

skin-in®ltrating T cells is not restricted to a type-2 cytokine

pattern in chronic skin lesions of atopic dermatitis. J Invest

Dermatol 1996; 107:871±6.

14 Kapsenberg ML, Hilkens CM, Jansen HM, Bos JD, Snijders A,

Wierenga EA. Production and modulation of T-cell cytokines

in atopic allergy. Int Arch Allergy Immunol 1996; 110:107±13.



Table 4. Sensitizations to pollen (*according to SPT and/or

speci®c IgE) and pollen-associated food allergy (**according to

patient's history) in responders and non-responders

Pollen-associated

Sensitizations* to food allergy**

Birch Mugwort Grass Birch Mugwort Grass

Responder 19 11 17 9 4 6

(n� 26)

Non-responder 9 2 5 5 0 2

(n� 15)

15 Sampson HA, McCaskill CC. Food hypersensitivity and atopic

dermatitis: evaluation of 113 patients. J Pediatr 1985; 107:

669±75.

16 Fuglsang G, Madsen C, Halken S, Jorgensen M, Ostergaard

PA, Osterballe O. Adverse reactions to food additives in

children with atopic symptoms. Allergy 1994; 49:31±7.

17 Hani®n JM, Rajka G. Diagnostic features of atopic dermatitis.

Acta Derm Venereol Suppl 1980; 92:44±7.

18 Costa C, Rilliet A, Nicolet M, Saurat JH. Scoring atopic

dermatitis: the simpler the better? Acta Dermatol Venereol

1989; 69:41±5.

19 Ehlers A, Stangier U, Gieler U. Treatment of atopic dermatitis:

a comparison of psychological and dermatological approaches

to relapse prevention. J Consult Clin Psychol 1995; 63:624±35.

20 Czech W, Krutmann J, SchoÈpf E, Kapp A. Serum eosinophil

cationic protein (ECP) is a sensitive measure for disease

activity in atopic dermatitis. Br J Dermatol 1992; 126:351±5.

21 Jakob T, Hermann K, Ring J. Eosinophil cationic protein in

atopic eczema. Arch Dermatol Res 1991; 283:5±6.

22 Niggemann B, Beyer K, Wahn U. The role of eosinophils and

eosinophil cationic protein in monitoring oral challenge tests in

children with food-sensitive atopic dermatitis. J Allergy Clin

Immunol 1994; 94:963±71.

23 Walker C, KaÈgi K, Ingold P et al. Atopic dermatitis: correlation

of peripheral blood T cell activation, eosinophilia and serum-

factors with clinical severity. Clin Exp Allergy 1993; 23:145±53.

24 Burks AW, Sampson HA. Diagnostic approaches to the patient

with suspected food allergies. J Pediatr 1992; 121:S64±S71.

414 M. Worm et al.


Documents

Clinical relevance of food additives in adult patients with atopic dermatitis