Results

Among the 3153 Ring grade ≥2 FIA cases reported to the AVN, 2217 (70.3%) involved plant-derived elicitors, all ages combined. Of these, 1694 (76.4%) were due to PAs. Four cases were excluded (age not specified) resulting in 1690 cases. Among these, adolescents (N = 237; 14%) and adults (N = 516; 30.5%) together accounted for 44.5% of cases. Additionally, FIA induced by PAs occurred in 909 (53.8%) cases in children and in 28 (1.7%) cases in infants (Fig. 1).

Fig. 1: Flowchart of the food-induced anaphylaxis cases collected by the Allergy-Vigilance Network: 2002–2024 and included in our final analysis (highlighted in green); Undetermined category includes idiopathic or unknown allergens, food coloring, and additives/preservatives.

Time trend of all FIA attributed to plant allergens relevant to PBDs over 2 decades (2002–2024)

Among adolescents, PA-related FIA remained stable until approximately 2017, after which cases declared increased significantly (post-breakpoint APC 16.7%; 95% CI 6.3–28.1; P = 0.011). In contrast, cases declared remained stable among adults (−0.8% per year; 95% CI –2.1 to 0.5; P = 0.230) (Fig. 2A). At the individual-allergen level, peanut showed a significant increase in adolescents (APC 5.5%; 95% CI 1.9–9.2; N = 79; P = 0.020) whereas trends for cashew, pistachio, and walnut were based on few cases (N = 10–14) with wide CIs (Fig. 2B). In adults, no allergen increased significantly, whereas significant decreases were observed for lupin (APC −12.2%; 95% CI −17.3 to −6.9; N = 37; P < 0.001) and buckwheat (APC −6.7%; 95% CI −11.0 to −2.1; N = 44; P = 0.024) (Fig. 2C). These per-allergen trends were based on small annual case counts and should be interpreted with caution.

Fig. 2: (A) Annual total FIA cases attributed to plant allergens (PAs) reported to the Allergy-Vigilance Network, 2002–2024, by age group. Points are annual counts; lines are fitted Poisson trends with 95% confidence bands. A segmented model was fitted to both age groups; a significant breakpoint was identified only in adolescents (≈2017). For adults, no significant breakpoint was identified, and the trend is shown as a single fitted slope. Annual percent change (APC) in FIA attributed to specific PA among (B) Adolescents and (C) Adults. For clarity, only allergens with ≥10 reported cases over the 2002–2024 study period are displayed. Error bars represent 95% confidence intervals. Trend analyses included cases with a documented year. N represents number of cases. P values (per-allergen APC) were derived from Poisson regression and Benjamini-Hochberg adjusted within each age group

Age distribution of food-induced anaphylaxis due to plant allergens

Fig. 3 shows the distribution of PA families within each age group. In adolescents, legumes (48.5%) and tree nuts (21.9%) accounted for the majority of PA-related FIAs while in adults, cereals (29.7%), legumes (26.6%), and tree nuts (21.1%) predominated.

Fig. 3: Age distribution of food-induced anaphylaxis cases attributed to plant allergens relevants to plant-based diets reported to the Allergy-Vigilance Network (2002–2024), stratified by age group. P-values shown were calculated using Fisher's exact test, with Benjamini-Hochberg correction for multiple comparisons across allergen families. Only comparisons reaching statistical significance are indicated

Supplemental Fig. 1 illustrates the relative contribution of specific allergens within each PA families to FIA. In adolescents, peanut (70.4%) predominated among legumes, whereas peanut, soy, and lupin contributed similarly in adults (26–30%). Other legumes commonly used in PBDs, such as peas (1.7% in adolescent vs 1.5% in adults) and lentils (2.9% in adults), were uncommon elicitors of FIA in both age groups. Tree nut-related FIA in adolescents was mainly due to cashew (26.9%), hazelnut (21.2%), pistachio (19.2%) and walnut (19.2%), while hazelnut (36.7%) and almond (18.3%) were the leading elicitors in adults. Sesame (40.6% vs 47.8%) and pine nut (40.6% vs 21.7%) were the most common seed allergens in adolescents and adults, respectively. Wheat was the dominant cereal allergen (78.3% vs 90.8%), and buckwheat accounted for nearly all pseudocereal-related FIA (100% vs 91.7%) in adolescents and adults respectively.

Clinical phenotype and immunological profile of the most frequent plant allergens: comparison according to age group

Fig. 4 depicts the most frequent PAs responsible for FIA in adolescents and adults. Eleven allergens represent 85% (640/753) of PA-related FIA.

Fig. 4: Proportion of food-induced anaphylaxis (FIA) cases attributed to the eleven most frequent plant allergens in adolescents and adults. Percentages are calculated using the total number of FIA cases within each age group for the respective allergen as the denominator. The total number of cases for each allergen across adolescents and adults is shown on the x-axis labels. P-values shown were calculated using Fisher's exact test, with Benjamini-Hochberg correction for multiple comparisons across allergen. Only comparisons reaching statistical significance are indicated

Clinical phenotype of these allergens are presented in descending order of frequency, as shown in Fig. 4. Additionally, Fig. 5 summarizes our findings and observed patterns of FIA due to PAs in adolescents and adults, including age-associated phenotypic presentation and related clinical considerations.

Fig. 5: Summary of the study findings, translated into the situation of adolescents/adults seeking to adopt plant-based diets (PBDs). (A) Overview of the observed patterns of food-induced anaphylaxis (FIA) associated with plant allergens relevant to PBDs (PAs) in adolescents and adults and possible clinical considerations. (B) Age-specific phenotypes of FIA associated with the main PAss in adolescents versus adults. aPAs with possible risk of reactions following hidden exposures; bReflects the adult phenotype; the small number of adolescent cases limits recommendations for this age group

Wheat

One-hundred fifty-seven cases of wheat-induced anaphylaxis were reported. Most occurred in adults (N = 139; 88.5%) while adolescents accounted for 11.5% (N = 18) of cases (Table 1). Grade 2 reactions occurred in 66.7% of adolescents and 49.6% of adults, grade 3 in 33.3% and 48.2%, respectively, and grade 4 occurred only in adult (2.2%). No fatalities were reported. Cofactors were frequently reported, 94.4% in adolescents and 90.5% in adults (Supplemental Table 2). Physical exercise (PE) was the most common cofactor, reported in 70.6% adolescents and 50% adults. In addition, combinations of cofactors, medications and alcohol, were also observed in adults (Supplemental Fig. 2A). A prior wheat-allergy diagnosis was uncommon in both adolescents (5.9%) and adults (10.9%), whereas self-reported symptoms following wheat consumption, including recurrent urticaria or anaphylaxis, were frequent (64.7% and 60.3%, respectively). Sensitization to Tri a 19 was more frequent in adults (97.7%) vs adolescents (72.7%; P = 0.009), whereas Tri a 14 sensitization was more common in adolescents vs adults (27.3% vs 2.3%, P = 0.009; Supplemental Table 2). Most reactions (78.9%) occurred after full serving-size meals (Supplemental Table 3), commonly bread or pizza. In most cases (96.6%), wheat was identifiable in the consumed food products while it was not readily recognizable in only 5 cases (soup packets, soy sauce, and meat preparations) (Supplemental Table 4).

Food-induced anaphylaxis attributed to plant allergens in adolescents and adults (N = 753)
Allergens | Total number of cases | AdolescentsN (%) | AdultsN (%)
Legumes | 252 | 115 (45.6) | 137 (54.7)
Peanut | 123 | 81 (65.9) | 42 (34.1)
Soy | 52 | 16 (30.8) | 36 (69.2)
Lupin | 46 | 9 (19.6) | 37 (80.4)
Fenugreek | 6 | 3 (50) | 3 (50)
Mung beans | 5 | 2 (40) | 3 (60)
Lentils | 4 | 0 (0) | 4 (100)
Peas | 4 | 2 (50) | 2 (50)
Beans | 4 | 0 (0) | 4 (100)
Chickpeas | 2 | 1 (50) | 1 (50)
Mix of legumes | 2 | 0 (0) | 2 (100)
Broad beans | 1 | 0 (0) | 1 (100)
Carob | 1 | 0 (0) | 1 (100)
Alfalfa sprouts | 1 | 1 (100) | 0 (0)
Flageolet beans | 1 | 0 (0) | 1 (100)
Cereals | 176 | 23 (13.1) | 153 (86.9)
Wheat | 157 | 18 (11.5) | 139 (88.5)
Barley | 7 | 0 | 7 (100)
Maize | 7 | 1 (14.3) | 6 (85.7)
Oats | 3 | 2 (66.7) | 1 (33.3)
Rice | 2 | 2 (100) | 0 (0)
Tree nuts | 161 | 52 (32.3) | 109 (67.7)
Hazelnut | 51 | 11 (21.6) | 40 (78.4)
Cashew | 29 | 14 (48.3) | 15 (51.7)
Walnut | 25 | 10 (40) | 15 (60)
Almond | 24 | 4 (16.7) | 20 (83.3)
Pistachio | 14 | 10 (71.4) | 4 (28.6)
Brazil nut | 8 | 2 (25) | 6 (75)
Mix of tree nuts | 6 | 1 (16.7) | 5 (83.3)
Macadamia | 3 | 0 (0) | 3 (100)
Pecan | 1 | 0 (0) | 1 (100)
Seeds | 101 | 32 (31.7) | 69 (68.3)
Sesame | 46 | 13 (28.3) | 33 (71.7)
Pine nut | 28 | 13 (46.4) | 15 (53.6)
Sunflower seeds | 11 | 1 (9.1) | 10 (90.9)
Mustard | 7 | 3 (42.9) | 4 (57.1)
Flax seed | 4 | 0 (0) | 4 (100)
Pumpkin seed | 4 | 2 (50) | 2 (50)
Chia seeds | 1 | 0 (0) | 1 (100)
Pseudocereal | 63 | 15 (23.8) | 48 (76.2)
Buckwheat | 59 | 15 (25.4) | 44 (74.6)
Quinoa | 3 | 0 (0) | 3 (100)
Amaranth | 1 | 0 (0) | 1 (100)

Peanut

Of 123 peanut-induced anaphylaxis cases, 81 (65.9%) occurred in adolescents and 42 (34.1%) in adults. (Table 1). Most reactions were grade 2–3 in both age groups, with grade 4 occurring in 9.9% of adolescents vs 4.8% of adults (P = 0.641) including 4 fatalities in adolescents vs 1 in adult. Cofactors were reported more frequently in adults (50%) than in adolescents (25.3%; P = 0.016; Supplemental Table 5), with alcohol being the main trigger in adults (47.4%), whereas diverse cofactors were reported in adolescents (Supplemental Fig. 3A). A higher proportion of adolescents had a previously diagnosed peanut-allergy compared with adults (71.8% vs 46.2% respectively; P = 0.012; Supplemental Table 5). Most reactions followed small exposures (<1 teaspoon; 54.9%), mainly from peanut-containing dessert or snacks (Supplemental Table 3). In 22.3% of cases, peanut was not identifiable in the consumed product due to contamination or labeling issues (Supplemental Table 4).

Buckwheat

Among 59 buckwheat-induced anaphylaxis cases reported, most occurred in adults (N = 44; 74.6%) vs adolescents (N = 15; 25.4%; Table 1). Reactions were predominantly grade 2 (73.3% in adolescents vs 61.4% in adults) with grade 3 occurring in 26.7% vs 38.6% respectively. No grade 4 reactions were observed. Cofactors were reported in 26.7% of adolescent and 35.7% of adult cases (Supplemental Table 6). PE was the most common cofactor in adolescents (50%), whereas adults more frequently reported multiple cofactors, alcohol, and medications (Supplemental Fig. 2B). Previous buckwheat-allergy diagnosis was more frequent in adolescents (46.7%) vs adults (13.9%; P = 0.026; Supplemental Table 6). Most reactions (59%) followed ingestion of a full-serving portions (eg, buckwheat crepes, bread, or pastries), while 23.1% occurred after very small exposures (<1 teaspoon, Supplemental Table 3). Reactions attributed to hidden exposure due to either contamination or lack of labelling occurred in 9 cases (16.7%; Supplemental Table 4).

Soy

Fifty-two cases of soy-induced anaphylaxis were reported, mainly in adults (N = 36; 69.2%) vs adolescents (N = 16; 30.8%; Table 1). The majority of reactions were grade 2 (81.2% in adolescents and 80.6% in adults), while grade 3 reactions accounted for 18.8% and 19.4% of cases, respectively, with no fatalities reported. Cofactors were reported with similar frequency in adolescents (37.5%) and adults (44.4%; Supplemental Table 5). PE was the most frequent cofactor in adolescents (50%). In adults, PE was relatively less common (25%), while medications and combination of cofactors (∼60%) predominated (Supplemental Fig. 3B). Prior soy-allergy diagnosis was uncommon in both groups (∼7–13%). Adolescents more frequently reported a known peanut and/or other legume-allergy diagnosis (50%) compared to adults (5.6%; P < 0.001). Sensitization to seed storage proteins (Gly m 5/6) was also more frequent in adolescents (45.5%) compared to adults (4%; P = 0.018). In contrast, birch pollen allergy (BPA) was more frequently reported in adults than in adolescents (64.5% vs 18.8%; P = 0.005) and were associated with a sensitization to Gly m 4, a pathogenesis-related protein 10 (PR-10) (84% vs 45.5%; P = 0.059; Supplemental Table 5). Soy beverages were the most frequently implicated products in adults (70.6%), whereas a broader range of food products was reported in adolescents, including soy beverages (37.5%), meat preparations (18.8%), meat alternatives (12.5%), and ultra-processed soy products (12.5%). FIA induced by soy beverages were more frequently reported in patients sensitized exclusively to Gly m 4 (50%), while those sensitized to Gly m 5 and/or Gly m 6 reported a broader range of food sources (Supplemental Table 7). Majority of reactions (60.7%) occurred after ingestion of large amount (1/2 cup–1 plate; Supplemental Table 3). Soy was identifiable (95.7%), with contamination or mislabeling reported in only 2 cases (Supplemental Table 4).

Hazelnut

Fifty-one hazelnut-induced anaphylaxis cases were reported, 78.4% (N = 40) in adults and 21.6% (N = 11) in adolescents (Table 1). Reaction severity was similar between groups, with grade 2 predominating (63.6% in adolescents vs 65% in adults), grade 3 occurring in 36.4% and 30%, respectively, and grade 4 observed only in adults (5%) including 1 fatality. Cofactors were reported at similar frequencies in adolescents (54.5%) and adults (63.2%; P = 0.729; Supplemental Table 8). PE (33.3%) and peak pollen season (33.3%) were the main cofactors in adolescents, whereas alcohol (45.8%) predominated in adults (Supplemental Fig. 4A). Prior hazelnut-allergy diagnosis was more frequent in adolescents vs adults (40% vs 14.7%; P = 0.175), as were BPA (54.5% vs 34.3%; P = 0.296), and oral allergy syndrome (OAS) (50% vs 20.5%; P = 0.104). Among the cases with available CRD data — a minority of patients — sensitization to Cor a 1 and to Bet v 1 (both PR-10) were common in both age groups. Sensitization to seed storage proteins (Cor a 9/14) was observed in 2 adolescents (28.6%; Supplemental Table 8). Most reactions occurred after small exposures (<1 teaspoon-1 tablespoon; 70.6%), typically via hazelnut-containing desserts, snacks, or native hazelnut (Supplemental Table 3). Hazelnut was identifiable in most cases with unrecognized exposure occurring in only 3 cases (1 chocolate and 2 home-made cakes without labeling; Supplemental Table 4).

Lupin

Among 46 cases of lupin-induced anaphylaxis, most occurred in adults (N = 37; 80.4%), with a female predominance (81.1%), while 19.6% (N = 9) occurred in adolescents. Reactions were mainly grade 2 in both groups (100% in adolescents vs 81.1% in adults). Grade 3 reactions occurred in 18.9% of adults, with no fatalities reported. Cofactors were infrequent (12.5% in adolescents vs 20.6% in adults; Supplemental Table 5). Previous lupin-allergy diagnosis was reported in 28.6% of adolescents and 33.3% of adults and prior peanut or other legume-allergy diagnosis was more common in adolescents than in adults (37.5% vs 9.4% respectively, P = 0.082). Most reactions (66.7%) followed consumption of a full serving of lupin flour-containing pastries, with only 1 case reported after ingestion of a small amount (<1 teaspoon; Supplemental Table 3). In 19% of the cases, reactions were suspected to be induced by lupin in industrial food products although it was not clearly identifiable (Supplemental Table 4).

Sesame

Of the 46 sesame-induced anaphylaxis cases, 71.7% (N = 33) occurred in adults and 28.3% (N = 13) in adolescents (Table 1). Reactions were predominantly grade 2 (69.2% in adolescents vs 54.5% in adults), with grade 3 observed in 30.8% vs 45.5% respectively. No grade 4 or fatalities occurred. Cofactors were reported at similar frequencies in adolescents (38.5%) and adults (35.5%; Supplemental Table 9), with PE being common in adolescents (40%), whereas multiple cofactors (36.4%) and alcohol (27.3%) predominated in adults (Supplemental Fig. 5A). Previous sesame-allergy diagnosis was more frequent in adolescents (36.4%) than in adults (10.7%; P = 0.083). BPA was reported in 27.3% of adolescents and 3.7% of adults (P = 0.065; Supplemental Table 9). Most reactions (60%) followed exposure to small amounts (<1 teaspoon to 1 tablespoon), typically via sesame paste (tahini or hummus), whereas 40% occurred after consumption of full serving-size sesame-containing meals (eg, burgers, biscuits; Supplemental Table 3). Reactions due to hidden exposure was seen in 14.6% of cases, all attributed to mislabeling (Supplemental Table 4).

Cashew

Of the 29 cashew–induced anaphylaxis cases, 48.3% (N = 14) occurred in adolescents while 51.7% (N = 15) in adults (Table 1). The majority of reactions were grade 2 (78.6% in adolescents vs 53.3% in adults), while grade 3 reactions occurred in 21.4% and 46.7%, respectively. No grade 4 reactions or fatalities were reported. Cofactors were reported only in adults (38.5%) and none in adolescents (P = 0.015; Supplemental Table 8) with alcohol (60%) and medications (40%) being the main cofactors (Supplemental Fig. 4C). Previous cashew-allergy diagnosis was more frequent in adolescents than in adults (50% vs 0%; P = 0.006) (Supplemental Table 8). Majority of reactions (81.3%) followed ingestion of small amounts (<1 teaspoon-1 tablespoon; Supplemental Table 3), typically cashew in its native form or cashew-based sauces. Cashew was identifiable in nearly all cases, with only 1 reaction strongly suspected to be due to cashew despite it not being identifiable in a sauce. (Supplemental Table 4).

Pine nut

Twenty-eight pine nut–induced anaphylaxis cases were reported which were similarly distributed between adolescents (46.4%; N = 13) and adults (53.6%; N = 15; Table 1). Reactions were predominantly grade 2 (84.6% in adolescents vs 60% in adults), while grade 3 reactions occurred in 15.4% and 40%, respectively, with no grade 4 reactions observed. Cofactors were uncommon in adolescents, reported in only 1 case (7.7%), attributed to stress, whereas cofactors were present in 50% of adult cases (P = 0.030; Supplemental Table 9), mostly alcohol (33.3%; Supplemental Fig. 5B). Prior pine nut-allergy diagnosis was uncommon in both age groups. BPA was observed only in adolescents (25%), while OAS was reported in 8.3% of adolescents and 21.4% of adults (P = 0.598; Supplemental Table 9). Most reactions (78.6%) followed small exposures (<1 teaspoon–1 tablespoon), typically from native pine nuts in salads or pesto, while 3 cases (21.4%) occurred after consumption of full serving-size of pine nut-based meals (eg, sandwich, tartare, dessert; Supplemental Table 3). Pine nut was not identifiable in 19.2% of cases due to either contamination or lack of labelling (Supplemental Table 4).

Walnut

Twenty-five walnut-induced anaphylaxis cases were reported, 60% (N = 15) in adults and 40% (N = 10) in adolescents (Table 1). Grade 2 reactions were observed in 40% of adolescents and 53.3% of adults, grade 3 in 60% and 40%, respectively, and 1 grade 4 reaction (6.7%) occurred in adults, which was fatal. Cofactors were common, reported in 73.3% of adult and 50% of adolescents (Supplemental Table 8). PE was the predominant cofactor in adolescents (40%), whereas adults exhibited a broader range of cofactors (Supplemental Fig. 4B). Previous walnut-allergy diagnosis was more frequent in adolescents (62.5%) compared to adults (15.4%; P = 0.056) (Supplemental Table 8). Most reactions (92.3%) followed ingestion of small amounts of typically native walnuts (<1 teaspoon–1 tablespoon; Supplemental Table 3). In all cases except 1 involving unrecognized exposure in a salad, walnut was easily identifiable (Supplemental Table 4).

Almond

Twenty-four almond-induced anaphylaxis cases were reported, mainly in adults (N = 20; 83.3%), with few cases in adolescents (N = 4; 16.7%; Table 1). Grade 2 reactions were observed in 25% of adolescents vs 55% of adults, and grade 3 in 75% vs 45%, respectively. No grade 4 reactions or fatalities were reported. Cofactors were frequent, reported in 2 adolescent cases (66.7%, both PE-associated) and in 60% of adult cases, mostly multiple cofactors (33.3%) and alcohol (25%) (Supplemental Table 8; Supplemental Fig. 4D). BPA and OAS were each seen in 2 adolescents (66.7%). In adults, BPA was reported in 7 (38.9%) and OAS in 4 (20%) individuals (Supplemental Table 8). Reactions occurred across a wide range of exposure amounts (<1 teaspoon to 1/2 cup; Supplemental Table 3). Almond was identifiable in all but 1 case in which the patient was unaware that the consumed macarons contained almond (Supplemental Table 4).