IgE to Bet v 1/Bet v 2 in UAS

Authors


  • Positivity to apple is the best predictor.

Some patients with birch pollen allergic hypersensitivity also refer symptoms of the ‘oral allergy syndrome’ when they eat fresh fruit, certain vegetables and hazelnuts (1–4). Cross-reactivity between apples and birch pollen occurs not only at the serologic level but also at the cellular level (4). In order to evaluate the relationship between specific IgE levels for Bet v 1/Bet v 2 with 11 types of food (apple, celery, carrot, hazelnut, tomato, potato, cucumber, peanut orange, kiwi and turkey meat) serum IgE levels were determined by FEIA (Pharmacia, Upjohn) in 33 patients with specific IgE to birch pollen.

Twenty patients (61%) were positive for Bet v 1, four patients (12%) for Bet v 2, and six patients (18%) for both allergens. Twelve out of the 30 patients (40%) sensitized to recombinant allergens reported symptoms of the ‘oral allergy syndrome’.

A significant correlation (Spearman's rank test) was identified between Bet v 1 IgE and apple IgE (r = 0.37, P = 0.03). Significant positive correlations (P < 0.01) were also found for Bet v 2 IgE and 10 of the 11 foods (not for turkey meat).

As shown in Table .1, all patients with IgE specific only for Bet v 2 also had IgE for the same three foods (apple, peanut and carrot), while all those with IgE for both Bet v 1 and Bet v 2 had IgE for the same five foods (apple, peanut, carrot, tomato and potato). Specific IgE was observed most frequently for apple and was present in 93% (28/30) of the patients sensitized to either or both recombinant birch pollens.

Table 1.  Frequency of positivity (%) for the 11 foods tested in the patients: only Bet v 1 + (n = 20), only Bet v 2 + (n = 4), or both Bet v 1 + and Bet v 2 + (n°= 6)
FoodBet v 1+Bet v 2+Bet v 1+/v 2+
Apple
Peanut
Carrot
Tomato
Potato
Celery
Hazel nut
Orange
Cucumber
Kiwi
90
35
40
15
10
35
45
15
10
5
100
100
100
75
75
75
75
50
50
50
100
100
100
100
100
83
67
83
50
33
Turkey meat52517

Symptoms of the ‘oral allergy syndrome’ were reported by 35% (7/20) of the patients positive for Bet v 1 alone, by 50% (2/4) positive for Bet v 2 alone, and by 50% for both recombinant allergens (3/6), but by none of the three negative patients . The univariate analysis by logistic regression indicated that six foods (apple, hazelnut, tomato, potato, peanut, orange) were significant predictive factors (P < 0.05) for the appearance of the ‘oral allergy syndrome’ However, according to the Bonferroni correction only the P-value reached with apple should be considered significant (P = 0.006). The relative risk of developing ‘oral allergy syndrome’ was calculated by considering two threshold values of serum IgE levels (negative = < 0.35 kU/l and positive > 3.5 kU/l) against apple. This risk was estimated to be 1.44 in patients negative for apple, compared to that of apple-positive patients, and increased up to 1.99 in patients with IgE levels above 3.5 kU/l for apple.

Acknowledgments

We are indebted to Virginia Monafo MD for her co-operation and substantial contributions.

Accepted for publication 24 September 2001

Ancillary