The Occurrence of Food Hypersensitivity Reactions and the Relation to the Sensitization to Grass and Trees in Atopic Dermatitis Patients 14 Years of Age and Older.

BACKGROUND: Although epidemiologic data are scarce, there is no doubt that the increase in pollen allergies is going to be followed by an increase in the so-called pollen-related food allergies. The aim of this study was the evaluation of the food hypersensitivity reactions in atopic dermatitis patients and the relation to the sensitization to grass and trees.
METHODS: The complete dermatological and allergological examinations were performed in all included patients; the occurrence of food hypersensitivity reactions was recorded and the sensitization to inhalant allergens (grass and trees) was examined (skin prick test, and specific IgE). The statistical evaluation of the relation between the sensitization to these inhalant allergens and the occurrence of food hypersensitivity reactions was performed.
RESULTS: A total of 321 patients were included, with an average age of 26.7 years (±9.2 years) and with an average SCORAD of 33.2 (±13.3) points. The significant relation was recorded between the patients suffering from sensitization to grass and the reactions to peanuts, celery and walnuts; another significant relation was demonstrated between the patients suffering from sensitization to trees and reactions to apple, peanuts, and walnuts. In patients suffering from sensitization to grass and trees, the occurrence of food hypersensitivity reactions to tomatoes, kiwi, spices, oranges, capsicum, tangerines, and carrot was higher also, but the relation was not significant.
CONCLUSION: The significant relation was found between the reactions to peanuts and walnuts and sensitization to grass and trees, another significant relation was found between the sensitization to grass and celery and between the sensitization to trees and apple.

Introduction

The presence of atopic disorders implies an increased risk of food allergy representing only a small percentage of all adverse reaction to foods. The term food hypersensitivity reactions represents the umbrella term for food allergy and for nonallergic food hypersensitivity (= food intolerance). The term food allergy is used to describe the clinical symptoms that are mediated by the immune system; number of IgE-, cellular-, and mixed IgE- and cell-mediated food hypersensitivity disorders have been described.[123] Food intolerance is a nonallergic hypersensitivity to food that does not include the immune system even though the symptoms are similar to those of IgE-mediated allergic reactions. An impaired histamine degradation based on reduced diaminooxidase activity, and the resulting histamine excess may cause numerous symptoms mimicking an allergic reaction.[4] Food allergy is highly associated with atopic dermatitis and is one of the most common triggers of potentially fatal anaphylaxis in the community. Sensitization to food allergens can occur in the gastrointestinal tract (class 1 food allergy) or as a consequence of cross reactivity to structurally homologous inhalant allergens (Class 2 food allergy). Although epidemiologic data are scarce, there is no doubt that the increase in pollen allergies is going to be followed by an increase in the so-called pollen-related food allergies.[56789] Cross-reaction is based on the binding of an IgE antibody to homologous allergen structures – shared linear or – in most cases – conformational epitopes (i.e., structural similarities).[10111213] Reports on specific food allergies linked with specific aeroallergens are inconsistent. This apparent inconsistency is not surprising, as today the majority of allergic patients are sensitized toward pollen or other inhalant allergens from more than one plant species, and therefore, there is a plethora of possible cross-reactions. Moreover, geographic differences and different nutritional habits may also play an important role in this context.[13] Recently, European Academy of Allergy and Clinical Immunology (EAACI) published evidence-based guidelines for food allergy and anaphylaxis.[14] The aim of this present work was to extend the knowledge on cross-reactivities between inhalant and food allergies.[14]

Materials and Methods

During the period 2008–2016, 321 patients suffering from atopic dermatitis at the age of 14 years and older were examined. All these patients were examined in the Department of Dermatology, Faculty Hospital Hradec Králové, Charles University of Prague, Czech Republic. The diagnosis of atopic dermatitis was made with the Hanifin-Rajka criteria.[15] Exclusion criteria were long-term therapy with cyclosporin or systemic corticoids, pregnancy, breastfeeding. Patients with atopic dermatitis having other systemic diseases were excluded from the study as well. Complete dermatological and allergological examinations were performed in patients included in the study. The following parameters were followed: sensitization to mixture of grass and trees and the occurrence of food hypersensitivity reactions. The statistical evaluation of the relation between the sensitization to grass and trees and the occurrence of food hypersensitivity reactions was performed.

The study was approved by the Ethics Committee of Faculty Hospital, Hradec Králové, Charles University in Prague, Czech Republic. There was no conflict of interest. CONSORT statement and STROBE statement guidelines were followed.[16]

Evaluation of parameters monitored

Sensitization to grass and trees

It was determined according to the specific IgE (sIgE) level and the skin prick test (SPTs) results. Commercial extracts Alyostal (Stallergens, France) was used for SPT. The serum level of the sIgE was measured with the method of CAP (system FEIA-Pharmacia Diagnostics, Uppsala, Sweden). The level of specific IgE higher than 0.35 U/ml was assessed as positive. The sensitization was confirmed in positive results in sIgE and/or SPT examination.

The diagnosis of food hypersensitivity reactions

It was made according to the patient's history. The patients answered whether they had suffered from immediate or late food reactions (oral allergy syndrome [OAS], gastrointestinal problems, the occurrence of skin problems, and respiratory problems). The most frequent food allergens were mentioned and patients answered whether they had suffered from some reactions to these foods, potentially they should have mentioned other foods with recorded reactions. The answers concerning the possible food reactions reflected the patient's history; the occurrence of food hypersensitivity reactions was confirmed in cases of unclear history by oral challenge tests.[171819]

Statistical analysis

We analyzed the data to determine whether the occurrence of sensitization to grass and trees was associated with the occurrence of food hypersensitivity reactions. Pairs of these categories were entered in the contingency tables, and the Chi-square test for the relationship of these variables was perfomed with the level of significance set to 5%. We used the coefficient of concordance (CC) and if the CC was positive, the dependence was direct. If the CC was negative, the dependence was indirect.

Results

A total of 321 patients were examined, 105 men and 216 women with the average age of 26.7 years (±9.2 years) and with the average SCORAD of 33.2±13.3 points.

The sensitization to grass was recorded in 215 patients (67%), to trees in 160 patients (50%); no sensitization to grass and trees was recorded in 98 patients (31%). The occurrence of food hypersensitivity reactions was recorded in 255 patients (79%), no food hypersensitivity reaction was recorded in 66 patients (21%). The characteristic of patients with the occurrence of the followed parameters is summarized in Table 1.

Table 1

The characteristics of patients included in the study

Total no. of patients	321
Gender
Men	105
Women	216
Average age	26.7 (±9.2) years
Average SCORAD	33.2 (±13.3) points
Sensitization to grass	215 (67%) patients
Sensitization to trees	160 (50%) patients
No sensitization to grass or trees	98 (31%) patients
Food hypersensitivity reaction	255 (79%) patients

SCORAD: Atopic dermatitis score

These symptoms of food hypersensitivity were recorded – OAS in 178 patients (55%), pruritus in 129 patients (40%), worsening of atopic dermatitis in 84 patients, (26%), urticaria in 26 patients (8%), gastrointestinal symptoms in 31 patients (9.6%), rhinitis in 24 patients (7.4%), respiratory problems in 12 patients (3.7%), and contact allergic reaction in 15 patients (4.6%) [Table 2]. More than one kind of reactions were described in some patients (for example pruritus and OAS).

Table 2

The symptoms of food hypersensitivity reactions in 321 patients

Symptom	Number of patients (%) (n=321)
Oral allergy syndrome	178 (55.4)
Pruritus	129 (40)
Worsening of atopic dermatitis	84 (26)
Gastrointestinal problems	31 (9.6)
Urticaria	26 (8)
Rhinitis	24 (7.4)
Breathlessness	12 (3.7)
Contact allergic reaction	15 (4.6)

The occurrence of food hypersensitivity reaction to single food was recorded to walnut in 119 patients (37%), to peanut in 88 patients (27%), tomato in 62 patients (19%), kiwi in 57 patients (18%), apple in 51 patients (15.8%), spices in 59 patients (18.3%), orange in 49 patients (15.2%), capsicum in 32 patients (10%), celery in 35 patients (11%), strawberry in 10 patients (3%), and to carrot in 18 patients (6%).

The relation between the occurrence of sensitization to grass and the occurrence of food hypersensitivity reaction is shown in Table 3. The significant relation was recorded between the patients suffering from sensitization to grass and the reactions to peanut (P<0.001), celery (P=0.013) and walnut (P<0.001). To show the rising occurrence of reactions to walnut, peanut, and celery in patients with sensitization to grass, we demonstrated the complement table, where the occurrence of these reactions is calculated in group of patients suffering from sensitization to grass as 100% and in group of patients without sensitization to grass as 100% [Table 3]. No significant relation was found between the sensitization to grass and reaction to strawberry (P=0.836), tomato (P=0.457), kiwi (P=0.134), apple (P=0.058), spices (P=0.650), orange (P=0.294), capsicum (P=0.158), tangerine (P = 0.381), and carrot (P=0.289) [Table 3].

Table 3

The relation between sensitization to grass and food hypersensitivity reaction

Food hypersensitivity reaction	Allergy to grass yes (%)	Allergy to grass no (%)	Total number of patients (%)
Strawberry +	7 (2)	3 (1)	10 (3)
Strawberry −	208 (65)	103 (32)	311 (97)
P	0.836
Tomato +	44 (14)	18 (6)	62 (19)
Tomato −	171 (53)	88 (27)	259 (81)
P	0.457
Kiwi +	43 (13)	14 (4)	57 (18)
Kiwi −	172 (53)	92 (29)	264 (82)
P	0.134
Apple +	40 (12)	11 (3)	51 (16)
Apple −	175 (54)	95 (30)	270 (84)
P	0.058
Spices +	41 (13)	18 (5)	59 (18)
Spices −	174 (54)	88 (28)	262 (82)
P	0.650
Orange +	36 (11)	13 (4)	49 (15)
Orange −	179 (56)	93 (29)	272 (85)
P	0.294
Tangerine +	41 (13)	16 (5)	57 (18)
Tangerine −	174 (54)	90 (28)	264 (82)
P	0.381
Carrot +	10 (3)	8 (3)	18 (6)
Carrot −	205 (64)	98 (30)	303 (94)
P	0.289
Capsicum +	25 (8)	7 (2)	32 (10)
Capsicum −	190 (59)	99 (31)	289 (90)
P	0.158
Celery +	30 (9)	5 (2)	35 (11)
Celery −	185 (58)	101 (31)	286 (89)
P	0.013* CC=0.532238
Walnut +	94 (29)	25 (8)	119 (37)
Walnut −	121 (38)	81 (25)	202 (63)
P	0.000* CC=0.431338
Peanut +	73 (23)	15 (4)	88 (27)
Peanut −	142 (44)	91 (28)	233 (73)
P	0.000* CC=0.514419

P = P-value of Chi-square test; P*= significant; CC = Coefficient of concordance

The relation between the occurrence of sensitization to trees and the occurrence of food hypersensitivity reactions is tabulated in Table 4. A significant relation was recorded between the sensitization to trees and reactions to apple (P=0.021), peanut (P=0.011), and walnut (P=0.003). Here also we cross-tabulated the sensitization to different food items to sensitization to trees [Table 4]. No significant relation was found between the sensitization to trees and reaction to strawberry (P=0.514), tomato (P=0.381), kiwi (P=0.294), apple (P=0.021), spices (P=0.488), orange (P=0.424), capsicum (P=0.455), celery (P=0.360), tangerine (P=0.057), and carrot (P=0.360).

Table 4

The relation between the occurrence of sensitization to trees and the food hypersensitivity reaction

Food hypersensitivity reaction	Allergy to trees yes (%)	Allergy to trees no (%)	Total number of patients (%)
Strawberry +	6 (2)	4 (1)	10 (3)
Strawberry −	154 (48)	157 (49)	311 (97)
P	0.514
Tomatoe +	34 (10)	26 (8)	62 (19)
Tomatoe −	126 (39)	133 (42)	259 (81)
P	0.381
Kiwi +	32 (10)	25 (8)	57 (18)
Kiwi −	128 (40)	136 (42)	264 (82)
P	0.294
Spices +	27 (8)	32 (10)	59 (18)
Spices −	133 (41)	129 (40)	262 (82)
P	0.488
Orange +	27 (8)	22 (7)	49 (15)
Orange −	133 (41)	139 (43)	272 (85)
P	0.424
Tangerine +	41 (13)	16 (5)	57 (18)
Tangerine −	154 (48)	110 (34)	264 (82)
P	0.057
Capsicum +	18 (6)	14 (4)	32 (10)
Capsicum −	142 (44)	147 (45)	289 (90)
P	0.455
Celery +	20 (6)	15 (5)	35 (11)
Celery −	140 (44)	146 (45)	286 (89)
P	0.360
Carrot +	11 (3)	7 (2)	18 (6)
Carrot −	149 (46)	154 (48)	303 (94)
P	0.325
Walnut +	72 (22)	47 (15)	119 (37)
Walnut −	88 (27)	114 (35)	202 (63)
P	0.003* CC=0.329877
Peanut +	54 (17)	34 (10)	88 (27)
Peanut −	106 (33)	127 (40)	233 (73)
P	0.011* CC=0.31103
Apple +	33 (10)	18 (6)	51 (16)
Apple −	127 (40)	143 (44)	270 (84)
P	0.021* CC=0.347323

P = P-value of Chi-square test; P*= significant; CC = Coefficient of concordance

Discussion

The aim of our study was to find out the occurrence of food hypersensitivity reactions in atopic dermatitis patients and the relation to the sensitization to grass and trees. The epidemiologic data about pollen-related food allergies are scarce; however, there is little doubt that the increase in pollen allergy is going to be followed by an increase in the so-called pollen-related food allergies.[6789] We enrolled 321 patients from Czech Republic (Central Europe) suffering from atopic dermatitis aged 14 years and more. The sensitization to grass was recorded in 215 patients (67%), to trees in 160 patients (50%). The occurrence of food hypersensitivity reactions was recorded in 255 patients (79%). The most frequent reactions were observed with walnut (in 119 patients, 37%), and peanut (in 88 patients, 27%). The most often recorded symptoms in our study were oral allergy syndrome (OAS) in 178 patients (55%), pruritus in 129 patients (40%) and worsening of atopic dermatitis in 84 patients (26%). We confirmed the significant relation between the patients suffering from sensitization to grass and the reactions to peanut, celery, and walnut and between the patients suffering from sensitization to trees and reactions to apple, peanut, and walnut. In patients with sensitization to grass the occurrence of reaction to walnut was observed in 43% of patients, to peanut in 34% and to celery in 14%; in patients without sensitization to grass the reaction to walnut was observed in 23%, to peanut in 14%, and to celery in 5%. In patients with sensitization to trees the occurrence of reaction to walnut was observed in 45% of patients, to peanut in 34% and to apple in 21%; in patients without sensitization to trees the reaction to walnut was observed in 29%, and to peanut in 21% and to apple in 11%. We also observed that in patients suffering from sensitization to grass the occurrence of food hypersensitivity reactions to tomato, kiwi, apple, spices, orange, tangerine, and capsicum was higher in comparison to patients without sensitization to grass; however, the difference was not statistically significant. On the other hand, in patients with sensitization to grass, there was lower occurrence of reactions to carrot than in patients without this sensitization; however, the difference was not statistically significant. As well, we observed that in patients suffering from sensitization to trees the occurrence of food hypersensitivity reactions to tomato, kiwi, apple, carrot, orange, tangerine, and capsicum was higher in comparison to patients without sensitization to trees; however, the difference was not statistically significant. On the other hand, in patients with sensitization to trees, there was lower occurrence of reactions to spices than in patients without this sensitization; however, the difference was not statistically significant. The most important panallergens are lipid transfer proteins (LTP), profilins and pathogenesis-related proteins (PR-10).[20212223242526272829303132333435] Nonspecific LTPs (nsLTPs) have only been identified in seed plants but are not even present in algae.[24] Allergenic LTP has been identified in trees and weed pollens of vegetable foods, fruits, and latex. LTPs may induce sensitization through the oral route and may determine systemic symptoms also on oral ingestion of both fresh and processed foods as beer, fruit juices, wine, etc.[25] Profilin, instead, comprises a family of highly conserved proteins, which are present in all eukaryotic cells and are involved in processes related to cell motility.[2627] The first allergenic profilin was described in birch pollen and was designated Bet v 2.[28] Profilin sensitization varies between 5% and 40% among allergic individuals and sensitization to Bet v 2 was observed in about 20% of patients allergic to birch pollen in different studies.[28293031] Allergenic profilins were identified in tree and grass pollens, in weeds, in plant-derived foods, as well as in latex.[32] The ingestion of raw fruits in profilin sensitized patients usually determines reactions restricted to the oral cavity (OAS).[3334] The PR-10 family consists of the Bet v 1 homologue, a protein with unknown function induced in stress conditions.[35] Bet v 1 is the major allergen of birch and about 50%–70% of birch pollen allergic patients, usually after respiratory sensitization, report symptoms after ingestion of a large spectrum of fruits and vegetables such as apple, hazelnut, peach, and potato.

In our study, the most frequent reactions were observed after ingestion of walnut (in 119 patients, 37%) and peanut (in 88 patients, 27%). Allergens of peanut belong to the cupin (Ara h 1, Ara h 3), the prolamin (Ara h 2, Ara h 6, Ara h 7, Ara h 9), the profilin (Ara h 5), the Bet v 1 (Ara h 8), the glycosyltransferase GT-C (Ara h 10, Ara h 11), and the scorpion toxin-like knottin (Ara h 12, Ara h 13) superfamilies. Ara h 8, a homolog of the major birch pollen allergen Bet v 1, and Ara h 5, a profilin, are mostly involved in pollen-associated food allergy, while the peanut nsLTP Ara h 9 is involved in the so-called nsLTP-syndrome due to the cross-reactivity with their homologs in birch and/or grass pollen or in fruits and seeds, respectively. Ara h 2 was described as the most important peanut allergen, as it was identified as a predictor of clinical reactivity to peanut.[363738] On the other hand, monosensitization to a single peanut allergen is relatively rare[39] and patients with monosensitization to Ara h 2 had a significantly lower symptoms severity score than polysensitized subjects and a lower level of allergen-specific IgE against peanut extract and Ara h 2.[40] Polysensitization to Ara h 2 and Ara h 1 and/or Ara h 3 appeared to be predictive of more severe reactions.[40414243] In peanut-allergic patients, a clinically relevant sensitization to other legumes such as soybean, lupin, lentil, or pea occurs; however, little information is available. In a group of 39 peanut-sensitized patients, 82%, 55%, and 87% of patients were also sensitized to lupine, pea, and soybean, respectively, whereas, based on DBPCFC, 29%–35% had symptoms to these beans.[44] In a recent study, Klemens and colleagues showed that 60% of soy-allergic patients had a concomitant peanut allergy and significantly more often specific IgE to soy extract, the soy 2S albumin Gly m 8 and the soy cupins Gly m 5 and Gly m 6 compared to the peanut-tolerant group.[45] Between 20% and 40% of peanut-allergic individuals have a co-existing allergy to taxonomically distantly related tree nuts.[4647] In a large study including 324 peanut-allergic patients, 86% were sensitized to tree nuts, and 34% had clinically documented allergy.[48] The numbers may even be higher than those reported in a study by Glaspole and colleagues where up to 60% of peanut-allergic individuals examined in their adult allergy clinic were allergic to one or more tree nuts with the most common reactivity of 49% to hazelnuts.[49]

Regarding the walnuts, Bet v 1-like allergen was identified in it. In the Wangorsch et al. study specific IgE levels to walnut, Bet v 1, and Jug r 5 in birch pollen allergics (n = 16) with concomitant walnut allergy were measured by ImmunoCAP. Jug r 5 and Bet v 1 allergens showed bidirectional IgE cross-reactivity by competitive ELISA and were capable of inducing histamine release from effector cells.[50] While the allergens of peanut and walnut have a high degree of homology in their amino-acid sequences, it is unknown whether this similarity is responsible for the observed co-reactivity. In the Rosenfeld et al. study, the binding of specific IgE antibodies to sequential epitopes of peanut and walnut in peanut-allergic patients with and without walnut allergy was analyzed. The results of this study indicate that although they share a rather high degree of homology with the corresponding regions of walnut allergens, the sequence stretches previously identified as sequential IgE binding epitopes of Ara h 1, Ara h 2, and Ara h 3 have no IgE binding equivalents in walnut allergens.[51] Another study investigated whether some characteristic disease entities could be identified in Europe for allergy to Rosaceae fruits. Five allergy patterns were found involving the allergen families PR-10, LTP, and profilin. In the western Mediterranean area, allergies to Rosaceae fruits are caused by monosensitization to LTP, monosensitization to profilin, or co-sensitization to both these allergens. On the contrary, monosensitization to PR-10 and to a lesser degree, co-sensitization to profilin and PR-10 is dominant in Northern and Central Europe. LTP sensitization is present both in pollinosis and non-pollinosis patients and is associated with peach allergy in particular. The disease pattern for patients sensitized to profilin is characterized by several concomitant allergies including grass and other pollens, Rosaceae and non-Rosaceae fruits. Finally, PR-10 sensitization is primarily associated to concomitant birch pollen and apple allergy.[52] In our previous studies, we have reported the relation between soy and peanuts allergy and pollen allergy.[5354]

Another important relation was confirmed between the patients suffering from reaction to apple and sensitization to trees. The OAS is recorded in the majority of patients suffering from reactions to apple. In this syndrome, a cross-reaction occurs between a pollen allergen and a homologous protein allergen in raw fruits or vegetables (e.g., birch pollen protein Bet v 1 and the homologous Mal d 1 protein in apple or Dau d 1 in carrot).[5556]

In our study, a significant relation was recorded between the patients suffering from sensitization to grass and the reactions to peanuts, celery, and walnuts. Grass pollen is one of the most important airborne allergen sources worldwide. Carbohydrate-reactive IgE antibodies have been attributed to grass pollen sensitization and found to cross-react with the glycan structures from other allergen sources, particularly vegetable foods.[5758] Another cause of extensive cross-reactivity is the group 12 allergens (profilins) that belong to a family of proteins highly conserved throughout the plants.[59] The majority of profilin-sensitized individuals do not react to corresponding foods in challenge tests.[60] Recent data confirmed that IgE to grass pollen profilin was detectable in bakers with asthma, food, and pollen allergy.[61]

A significant relation was confirmed between the sensitization to grass and the occurrence of reactions to the celery in our study. Apium graveolens, celery, is considered one of the most important plant food allergen sources in the adult Central European population.[6263646566] Among Central European patients, a predominant sensitization to Api g 1 (59%) and Api g 4 is observed.[65] Clinical allergy to celeriac is frequently associated with sensitization to Artemisia vulgaris and Betula verrucosa pollen in Central European countries, and thus, the terms celery-mugwort and celery-birch syndrome have been established.[67] Association between birch pollinosis and celery hypersensitivity is mainly attributed to Api g 1, a Bet v 1-homologous PR-10 protein.[6468] Moreover, Api g 4 and Api g 5 have been mentioned to play a role as cross-reactive molecules in this population.[656970] By contrast, a heat stable molecule involved in the celery-mugwort syndrome which might be able to trigger severe allergic reactions in Central European celeriac sensitized patients is not yet conspicuously determined.[71]

Conclusion

The significant relation was recorded between the patients suffering from sensitization to grass and the reactions to peanut, celery, and walnut and between the patients suffering from sensitization to trees and reactions to apple, peanut, and walnut. No significant relation was found between the sensitization to grass and reaction to strawberry, tomato, kiwi, apple, spices, orange, tangerine, and carrot; also no significant relation was found between the sensitization to trees and reaction to strawberry, tomato, kiwi, apple, spices, orange, capsicum, celery, tangerine, and carrot.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.