|Home | About | Journals | Submit | Contact Us | Français|
Sesame and coconut are emerging food allergens in the US. We sought to examine whether children allergic to peanuts and tree nuts are at increased risk of having an allergy to sesame or coconut. We performed a retrospective chart review of children who underwent skin prick testing (SPT) to sesame and coconut and identified 191 children who underwent SPT to sesame and 40 to coconut. Sensitization to sesame was more likely in children with positive SPT to peanuts (odds ratio [OR] = 6.7, 95% confidence interval [CI] [2.7–16.8], P<0.001) and tree nuts (OR = 10.5, 95% CI [4.0–27.7], P<0.001). Children with histories of both peanut and tree nut reaction were more likely to have a history of sesame reaction (OR = 10.2, 95% CI [2.7–38.7], P<0.001). Children with sensitization or allergy to peanuts or tree nuts were not more likely to be sensitized or allergic to coconut. In conclusion, children with peanut or tree nut sensitization were more likely to be sensitized to sesame but not coconut. Children with clinical histories of both peanut and tree nut allergy were more likely to be allergic to sesame.
Pediatric food allergy is increasing in prevalence, and peanuts and tree nuts are among the most common causes of severe allergic reactions (1–3). Children with peanut or tree nut allergies may be at higher risk of developing allergies to previously less well recognized foods such as sesame (4–6) and coconut (7). With the high prevalence of peanut and tree nut allergy in the US and concern for potential cross-reactivity with sesame, allergists suggest screening peanut and tree nut allergic children for sesame allergy (8). With the recent consideration of coconut as a tree nut in food labeling practice, some allergists are testing for coconut as well (9).
Sesame is an emerging food allergen in the US and carries a significant risk of severe reactions including anaphylaxis. In a case series from Israel, 30% of patients with sesame allergy had anaphylaxis as their first presenting symptom, and all were less than one year of age (10). Of significant clinical concern is that sesame allergy can begin in children who are too young to describe their symptoms, and the risk of severe allergic reactions is high (10–12). The prevalence of sesame allergy varies geographically, with increased allergy in regions of greatest exposure. In Israel, where sesame is regularly consumed in the diet, sesame is the third leading cause of food allergy after egg and cow’s milk, and the second leading cause of anaphylaxis after milk (13). With immigration and globalization, more American children are being exposed to sesame through their diets, making sesame an emerging cause of severe allergy in the US. In addition to sensitization via direct exposure to sesame, there is hypothesized cross-reactivity between nuts and sesame (14–16).
Coconut is an allergen that is also gaining more notice in the US. Immediate hypersensitivity to coconut is a relatively new phenomenon, first described in two adult patients with walnut allergy (17). Coconut allergy remains rare, with only a small number of reported cases (7, 17–19). However, there is new interest in this allergen since coconut is now considered a tree nut for food labeling purposes (9).
There are few studies examining the risk of sesame and coconut sensitization and clinical allergy among nut allergic patients. We sought to determine the prevalence of sesame and coconut sensitization among our population and the risk of sensitization and clinical allergy to sesame and coconut in our nut allergic patients.
This study was a retrospective chart review of skin prick test (SPT) results at Children’s Hospital Boston and several affiliated outpatient clinics from December 2006 through March 2008. Patients ranged in age from 6.6 months to 19.6 years (median, 4.0 years). Subjects underwent SPT to various allergens. All subjects avoided short-acting antihistamines for 72 hours and long-acting antihistamines (loratadine, cetirizine, fexofenadine, and hydroxyzine) for 10 days prior to SPT. Standardized extracts were used for testing to peanut, hazelnut, cashew, Brazil nut, almond, walnut, pistachio, pecan, sesame, and coconut. Control tests for SPTs were performed with histamine (positive control) and normal saline (negative control). Wheal diameters were measured at 15 minutes. A positive SPT was defined as a wheal diameter at least 3 millimeters larger than the diameter of the negative control.
The study population consisted of all patients who underwent SPT to sesame and all patients who underwent SPT to coconut. Data were also collected on SPT results to peanut and tree nuts, if performed. Tree nuts in our population were defined as hazelnut, cashew, Brazil nut, almond, walnut, pistachio, or pecan. Clinical history of allergic reaction to peanut, tree nuts, sesame, and coconut was collected for all patients. Symptoms of allergic reaction included urticaria, eczema exacerbation, non-specific dermatitis, angioedema, vomiting, diarrhea, repeated coughing or sneezing, wheezing, or anaphylaxis. History of allergic reaction was determined by patient or parent/guardian reporting as documented by the physician in the written medical record.
Prevalence rates of sensitization and clinical reaction for each food allergen were calculated. We created 2 × 2 tables to determine cross-reactivity among two different foods. We calculated odds ratios (OR) and 95% confidence intervals (CI) from 2 × 2 tables. ORs indicated the risk of sensitization to one food in the pair if sensitized to the other food in the pair. Statistical significance was assessed by Pearson’s Chi-square or Fisher’s exact method, as appropriate. A P-value of <0.05 was considered statistically significant.
For sesame and coconut, the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of a positive skin test was determined with regards to a history of clinical allergic reaction.
The study population consisted of all patients who underwent SPT to sesame or coconut (N = 212). Several children underwent SPT to both sesame and coconut (N = 19). Therefore, of the total study population, 191 children underwent SPT to sesame and 40 to coconut. The prevalence of sesame and coconut sensitization in this pediatric population was 36.6% and 20.0%, respectively.
The proportion of sesame sensitization was 53.3% (N = 40) among peanut sensitive children, 57.7% (N = 41) among tree nut sensitive children, and 68.9% (N = 31) among those sensitized to both peanuts and tree nuts. Sensitization to sesame was more likely in children with positive SPT to peanuts (OR = 6.7, 95% CI [2.7–16.8], P<0.001) and tree nuts (OR = 10.5, 95% CI [4.0–27.7], P<0.001). Further, children sensitized to both peanuts and tree nuts were more likely to be sensitized to sesame (OR = 6.1, 95% CI [2.9–12.6], P<0.001). There was no significant association between sensitization to peanuts or tree nuts and coconut (Table II).
The proportion of sesame allergy was 13.2% (N = 9) among peanut allergic children, 14.8% (N = 4) among tree nut allergic children, and 50.0% (N = 5) among those allergic to both peanuts and tree nuts. Children were not more likely to have an allergic reaction to sesame if they had a history of allergic reactions to peanuts (OR = 1.4, 95% CI [0.6–3.5], P=0.443) or tree nuts (OR = 2.1, 95% CI [0.6–7.0], P=0.207). However, children with a history of allergic reactions to both peanuts and tree nuts were more likely to have an allergic reaction to sesame (OR = 10.2, 95% CI [2.7–38.7], P<0.001). There was no increased risk of coconut allergic reaction among peanut or tree nut allergic children (Table III).
Among children with a positive SPT to sesame, 14.3% (N = 10) had a clinical history of allergic reaction to sesame. Additionally, among children with a positive SPT to coconut, 25.0% (N = 2) had a history of clinical allergic reaction to coconut (Table I). Correlation of SPT result and clinical history of allergic reaction demonstrated that sesame SPT had low sensitivity and specificity but high NPV while coconut had high specificity and NPV (Table IV).
The prevalence of peanut and tree nut allergies in American children is increasing (1–3). The emergence of sesame as a cause of allergic reactions and anaphylaxis is becoming increasingly recognized as well. Laboratory data suggest mechanisms for possible allergen cross-reactivity among nuts and sesame and coconuts (14, 16, 20, 21). However, little is known about the risk of clinical allergy to sesame or coconut in nut allergic children. The aim of this study was to determine whether children with allergies to peanuts or tree nuts were at increased risk of having sensitization as well as allergic reactions to sesame or coconut. We addressed this question by examining both SPT data and clinical reaction history. Our study is one of the largest evaluating SPT data and the largest evaluating clinical history of allergic reactions. Furthermore, this is the first study evaluating SPT and clinical history of cross-reactivity for coconut in peanut and tree nut allergic children.
We demonstrated that sensitization to sesame is common among children sensitized to peanuts, tree nuts, or both. Children did not appear to be at a higher risk of sensitization to coconut if they were sensitized to peanuts, tree nuts, or both. We then investigated whether these patterns of sensitivities from our SPT data were reflected in the clinical history of allergic reactions. In contrast to our SPT data, children with clinical allergies to peanuts or tree nuts were not more likely to have a reaction to sesame. However, children with a history of clinical reactions to both peanuts and tree nuts were more likely to have allergic reactions to sesame. Consistent with our SPT data, children did not appear to be at a higher risk of reaction to coconut if they had an allergic reaction to peanuts, tree nuts, or both.
There are few published studies examining SPT among patients sensitized to peanuts, tree nuts, and sesame. In a British study, among 55 patients with peanut or tree nut sensitization, 4 (7%) had a positive sesame SPT (4). Our study, performed 15 years later in an American pediatric population, shows an even higher degree of sesame sensitization. We showed that 53.3% of peanut sensitized children and 57.7% of tree nut sensitized children have a positive SPT to sesame. The prevalence of sesame sensitization may be higher in our study due to the differing ages and geographies of the populations, or may be due to the fact that peanut, tree nut, and sesame allergies have increased over time (8). Another group examined Australian children with peanut allergies and showed there was an increased prevalence of sesame sensitization and allergy in children who remained peanut allergic compared to those who outgrew their peanut allergies (15). Interestingly, this increased prevalence occurred despite strict recommendations to avoid sesame, suggesting a role for possible antigenic cross-reactivity. This study did not examine the relationship between tree nut and sesame allergy. A recent American study showed that the proportion of sesame allergy, defined as a convincing clinical history and evidence of positive sesame-specific IgE, was 6% among peanut allergic children (5). We found a higher proportion of sesame allergy (13.2%) among our peanut allergic patients. These differences are likely reflective of our differing study populations. Further, we reported the proportion of sesame allergy was 14.8% among tree nut allergic patients and 50.0% among children allergic to both peanuts and tree nuts.
Maloney et al determined strength of association between various food allergens using serum-specific IgE measurements (22). By calculating Spearman rank order correlation coefficients, they demonstrated that there was a weak correlation between peanut and sesame (0.34) and only a moderate correlation between certain tree nuts and sesame (range 0.57–0.79). This group did not investigate SPT data, nor did they calculate correlations based on clinical history. Although we did not examine serum-specific IgE levels, these results are consistent with our findings that children with clinical allergies to peanuts or tree nuts are not more likely to be allergic to sesame.
Shared homology between sesame and peanut and tree nut allergens have been described, suggesting a mechanism for cross-reactivity and sensitization (14, 16, 20). Seed storage proteins have been implicated as the major coconut allergen. Shared homology between the seed storage proteins of coconut and several tree nuts, primarily walnut and hazelnut, have been described (21). However, we found no evidence of increased coconut sensitization or clinical allergy in tree nut allergic patients.
An important limitation of our study is that the diagnoses of food allergies were based on physician documentation of patient or parent/guardian reporting of convincing histories of reactions. For ethical and practical reasons, we did not verify these reports with double-blind placebo-controlled food challenges, which remain the gold standard for diagnosing food allergies. Additionally, the group sizes of patients sensitized or allergic to specific foods, particularly coconut, were relatively small. Larger studies will need to be performed to verify these results. Despite these limitations, this study is important because it evaluates the potential cross-reactivity of several prominent food allergens in a larger patient population and in more detail than published previously.
Our study adds to the small body of literature defining predictive values for sesame SPT (6) and is the first to define predictive values for coconut SPT. In this study, we defined SPT results as either positive or negative and did not further validate these results in double-blind placebo-controlled food challenges for the reasons described above. However, we have show in another patient population that sesame SPT is not a good predictor of true sesame allergy as defined by oral food challenge (23). In that group, although we could not define a diagnostic decision point for sesame SPT, there was a trend for more predictability for SPT compared with sesame-specific IgE and a SPT wheal size of <3 millimeters approached a NPV of 90%.
In summary, our study shows that both cross-sensitivity and clinical cross-reactivity occur between peanuts and tree nuts with sesame but not coconut. Specifically, children with sensitivity or clinical allergy to both peanuts and tree nuts are significantly more likely to be sensitized or allergic to sesame. This study fails to demonstrate a significant risk of sensitization or allergy to coconut in peanut or tree nut sensitized or allergic children. Our differing results for sesame sensitivity versus clinical reaction in peanut and tree nut allergic children, coupled with our data demonstrating the low PPV of sesame SPT, brings into question the utility of the current practice among many allergists of screening peanut allergic children for sesame sensitivity. If future studies can show SPT reliably predicts sesame allergy, screening may then be appropriate, especially given the high risk of anaphylaxis among young children on their first exposure to sesame. However, there are risks of diagnosing sesame sensitivity without a history of allergic reactions: 1) the burden of dietary restriction and avoidance measures and 2) possibly losing the ability to acquire and sustain oral tolerance. For now, we recommend considering evaluating children for sesame allergy if they are allergic to peanuts or tree nuts and have never previously ingested sesame. If sensitization is demonstrated, oral challenge to sesame may be required to document the presence of clinical reactivity before recommendations are made for avoidance. If the child is already eating sesame without reactions, the child is clinically tolerant and further testing is not recommended. There does not appear to be sufficient data at this time to recommend that all peanut and tree nut allergic children undergo testing for coconut allergy. Future studies will be needed to validate our findings in a larger patient population and to further define the predictive values of sesame and coconut SPT against the gold standard of food challenges.