Monitoring of peanut-allergic patients with peanut-specific IgE

Monitoring of peanut-allergic patients with peanut-specific IgE Rozita Borici-Mazi, M.D., Jorge A. Mazza, M.D., David W. Moote, M.D., and Keith B. Payton, M.D. ABSTRACT Peanut allergy affects 1% of the population. Double-blind placebo-controlled food challenges are gold standard for diagnosis. Serum peanut-specific IgE (PN-IgE) is used in clinical practice as an additional diagnostic and monitoring tool. The purpose of this study was to characterize the clinical features of a peanut-allergic patient s cohort and determine the optimum frequency of measuring PN-IgE to predict the outcome of future peanut challenges. Retrospective chart review was performed of peanut-allergic patients followed up and serially tested for PN-IgE with a qualitative antibody fluorescent-enzyme immunoassay performed at the Immunology Laboratory, London Health Sciences Center, from 1997 to 2004. One hundred eighteen patients (median age at first reaction to peanut, 1.5 years; median baseline PN-IgE, 18.75) were reviewed. Younger age at first reaction and first PN-IgE measurement predicted slower decline of PN-IgE values (p 0.001 and p 0.044). At 2 and 5 years post initial measurement, 12.9 and 66%, respectively, of all patients had a significant decrease of PN-IgE values. Twenty percent of the patients experienced elevation of PN-IgE levels during follow-up. For most patients with significant history of reaction to peanuts and positive skin-prick test, it is probably adequate to measure serum PN-IgE levels every 3 5 years to screen for development of tolerance and predictproperty the outcome of future peanut challenges. More frequent measurements might be considered in older patients with lower initial PN-IgE levels. (Allergy Asthma Proc 29:329 335, 2008; doi: 10.2500/aap.2008.29.3117) Key words: Long-term follow up, monitoring, natural history, oral challenge, peanut allergy, peanut-specific IgE, peanut tolerance practice, peanut-specific IgE (PN-IgE) is used to quantify the degree of sensitization and predict the development of tolerance to ingested peanut, thus reducing the number of double-blind placebo-controlled food challenge. 7 Recently, it has been shown that the peanut allergy is not necessarily a life-long condition. In a study of Skolnick et al., 8 21.5% of their peanut-allergic patients outgrew their allergy. Sixty-seven percent of their patients with PN-IgE levels 2 kilounits of antibody per liter (ku A /L) and 61% with levels 5 ku A /L had nega- Several longitudinal studies in the United States, tive challenge results. A second review of peanut-allergic England, and France have indicated that 1% of patients who underwent peanut challenges showed that the general population is allergic to peanuts. 1 3 A survey the level of PN-IgE of 5 ku A /L was associated with of primary school children, from kindergarten to 50% chance of outgrowing allergy. 9 The follow-up of grade 3, in Montreal, Canada, showed that the peanut peanut-allergic patients with PN-IgE, to identify those allergy prevalence was as high as 1.7%. 4 Recently, the patients who have the higher chance of passing future rate of peanut allergy among the children has increased. peanut challenges has been widely used. So far, only a 5,6 few studies have looked at the serial measurements of The gold standard for diagnosing peanut allergy is PN-IgE in peanut-allergic individuals. Vander Leek et the double-blind placebo-controlled foodoceanside challenge, Publications al. 10 compared two separate measurements of PN-IgE at which can be time-consuming and difficult. In clinical least 10 months apart in 18 peanut-allergic patients and found no significant difference. In 2004, van Odijk et al. published data on the change of PN-IgE over a 5-year interval in relation to peanut intake, symptoms, and age at first reaction. Thirty-four percent of their patients increased their PN-IgE, 28% had similar levels, whereas 38% decreased their PN-IgE. 11 The aim of this study was to characterize the clinical features of a peanut-allergic patient cohort and determine the optimum frequency of PN-IgE measurement to monitor for development of tolerance and predict the outcome of future peanut challenges. From the Division of Allergy and Clinical Immunology, Department of Medicine, University of Western Ontario, London, Canada Address correspondence and reprint requests to Rozita Borici-Mazi, M.D., University of Western Ontario, 60 Rollscourt Place, London, Ontario N6G 1M2, Canada E-mail address: rboricim@uwo.ca Copyright 2008, OceanSide Publications, Inc., U.S.A. METHODS Study Population Charts of peanut-allergic patients followed up at the Allergy Unit of London Health Sciences Center from 1997 to 2004 were reviewed. Patients were enrolled in Allergy and Asthma Proceedings 329

the study if they were 15 years old at the time of their first reaction to peanuts and had had at least two serial measurements of PN-IgE at least 1 year apart. Patients were diagnosed with peanut allergy if they had a history of an acute reaction to peanut within 30 60 minutes after exposure and a positive skin-prick test (SPT) and/or PN-IgE. In 14 cases, patients were diagnosed as peanut sensitive based on positive SPT and radioallergosorbent test (RAST) to peanut without prior exposure. All peanut-sensitive patients and their families had implemented peanut avoidance measures. Open peanut challenges were offered during the follow-up if PN-IgE levels 2 ku A /L. The Ethics Committee of the University of Western Ontario approved this study and oral consent was obtained from the participants. Chart Review Patients charts were reviewed for symptoms of the initial reactions and number of accidental exposures to peanuts reported during follow-up visits. AdverseProperty re- this of definition, the patients were divided into two actions to peanuts were classified into (1) skin symptoms (including generalized hives/angioedema, contact urticaria, and worsening dermatitis), (2) respiratory symptoms, subdivided into mild (rhinitis and conjunctivitis) and severe (mouth swelling, throat tightness, laryngeal edema, cough, and wheezing), (3) gastrointestinal symptoms (nausea, vomiting, abdominal cramps, and diarrhea), and (4) anaphylactic shock (decreased blood pressure and impaired level of consciousness). Additional information recorded included gender, age at first reaction and first PN-IgE measurement, age at last PN-IgE measurement, and follow-up time. All available SPT results and PN-IgE values during follow-up were recorded. Associated Atopy and Other Food Allergies OceanSide Publications group 3, 100 ku A /L. Comparisons of survival curves Associated atopic conditions such as rhinitis, asthma, were made using log rank statistics. The frequencies of and atopic dermatitis were reviewed. The occurrence other atopic conditions and food allergies were calculated using the 2 -test. of other food allergies was documented and their diagnosis was based on a significant history of an acute reaction to one or more foods and positive SPT and/or specific IgE; selected open food challenges were performed as well. Laboratory Studies Skin tests were performed with standardized puncture method using Morrow-Brown style prick test needles (Allersharp; Quorum Pharmaceuticals, Mississauga, ON, Canada). A glycerinated extract of peanut along with positive (histamine) and negative (saline buffer solution) controls were used to perform SPTs. The wheal and flare reaction to peanut was measured at 15 minutes. The SPT to peanut was considered positive if the largest wheal diameter was 3 mm larger than the negative control. PN-IgE levels were measured with qualitative antibody fluorescent-enzyme immunoassay, Pharmacia CAP System RAST FEIA, performed at the Immunology Laboratory at London Health Sciences Center. The assay had a lower limit of detection of 0.35 ku A /L and an upper limit of 100 ku A /L. The positive results were rated based on CAP-RAST index as defined by manufacturer: 0.36 0.69, 0.70 3.49, 3.50 17.49, 17.50 49.99, 50.0 100, and 100 ku A /L. Values of 100 were arbitrarily assigned to 106 for statistical purposes. Statistical Analysis The primary end point was the time to a significant change of PN-IgE levels, which was calculated based on Kaplan-Meiers survival analysis. A significant change in PN-IgE levels was arbitrarily defined as the amount of decline/elevation of PN-IgE that allowed the second value to be rated in the lower/higher CAP- RAST class as defined by the manufacturer. Based on groups: (1) decline in PN-IgE and (2) increased PN-IgE. Other definitions used were (1) age at first reaction patient s age at first reaction to peanut, (2) age at first and last PN-IgE measurement patient s age at first and last PN-IgE measurement, and (3) follow-up time the time between first PN-IgE level and the level that showed significant change or the last PN-IgE level if such change did not occur. When the rate of decline on PN-IgE values was calculated for all of the patients, the serial measurements of PN-IgE were censored at the values that represented a significant increase of PN-IgE to minimize the confounding factors. For this analysis, patients were arbitrarily grouped based on initial PN-IgE value into group 1, 17.5 ku A /L; group 2, 17.5 100 ku A /L; and RESULTS A total of 118 patients were reviewed, 75 male patients (63.55%) and 43 female patients. The age at first reaction ranged between 3 months and 15 years with a median value of 1.5 years (Table 1). Median baseline PN-IgE value was 18.75 ku A /L with a range 0.36 to 100 ku A /L. As seen on the Table 1, 94 patients were diagnosed based on a significant history of an allergic reaction after exposure to peanuts, positive SPTs, and PN-IgE. Ten patients had a positive history and positive PN-IgE alone; of these, 6 patients did not undergo skin testing and 4 patients had a negative SPT to peanuts. Among patients with negative SPT to peanuts, two had PN-IgE of 100 ku A /L and their SPT converted to positive in retesting; the other two were not 330 May June 2008, Vol. 29, No. 3

Table 1 Patient characteristics Males n 75 (63.55%) Females n 43 (36.45%) Age at first reaction Age at first PN-IgE measurement Age at last PN-IgE measurement Baseline PN-IgE value Time to decrease in PN-IgE value (all patients) Follow up time No. diagnosed by History, SPT and PN-IgE n 94 History and PN-IgE n 10 SPT and/or PN-IgE/both n 12 PN-IgE n 2 Median (Range) 1.5 yr (3 mo 15 yr) 6.0 yr (1 25 yr) 8.0 yr (1 27 yr) 18.75 ku A /L (0.36 to 100 ku A /L) 49 mos 25.11 mos (13 83.5 mos) retested and one of them had symptoms on accidental Table 2 Associated other atopic conditions and FAs exposure to peanuts. Fourteen patients were not exposed to peanuts. Of those, 12 had positive SPTs Property and of No. of PN-IgEs and two had positive PN-IgEs only. This Patients (%) group s mean PN-IgE value was 25.06 ku A /L with a Associated Other Atopic Conditions range between 1.26 and 100 ku A /L. Asthma 51 (43) Seventy-three patients (61%) manifested other food Rhinitis 45 (38.1) allergies, the most common being tree nuts, eggs, and Atopic dermatitis 25 (21.2) milk (see Table 2). Tree nut allergy was present in Environmental allergies 73 (61.8) 35.6% of our peanut-allergic patient group. As shown Solitary food allergy 30 (25.4) in Table 2, asthma was the most common associated Solitary peanut allergy 4 (3.4) atopic condition; 30 patients suffered from food allergies only, whereas four manifested solitary peanut al- Other FAs Tree nuts 42 (57.3) lergy. Eggs 36 (49.3) One hundred four patients reacted significantly after Milk 27 (37) first known exposure to peanuts. Table 3 summarizes Vegetables 10 (13.7) the clinical characteristics of initial reactions to peanuts. Fifty-nine patients (56%) manifested skin symp- Fish/shellfish 7 (9.6) toms only, whereas six patients (5.76%) developed OceanSide anaphylactic shock after ingesting peanuts. The Publications Soya 7 (9.6) Grains 4 (5.5) Meats 3 (4.1) demographic features of the anaphylactic patients are presented in Table 4. The baseline PN-IgE level of patients that manifested skin symptoms only on exposure to peanuts (n 57; median PN-IgE, 18.7 ku A /L; SD 42.0) was not significantly different from the rest of the patients (n 44; median PN-IgE, 12.6 ku A /L; SD 42.1; p 0.387). All patients were followed up with serial testing of PN-IgE for 25.11 months (median value) ranging between 13 and 83.25 months. Ninety-four patients (79.6%) had a gradual decline of PN-IgE values, whereas 24 patients (20.4%) increased their PN-IgE during follow-up (Table 5). The patients who showed an elevation of PN-IgE were significantly younger at the time of their first reaction to peanuts compared with patients who showed a decline in PN-IgE (p 0.025) and their baseline PN-IgE values were significantly lower (p 0.01). Sesame 3 (4.1) Fruits 2 (2.7) Total individual FAs 141 FA food allergy. The patients who experienced a decline in PN-IgE levels were divided into three groups based on their baseline PN-IgE level. As shown in Table 6, the patients with a baseline PN-IgE level of 17.5 100kU A /L showed the highest rate (73%) of significant decline at 5 years, whereas those with initial PN-IgEs of 100 ku A /L had the least decline in PN-IgEs. In the latter group, only five patients (8%) decreased their initial PN-IgE level to the next RAST class (50 100 ku A /L). At 2 and 5 years post initial PN-IgE measurement, 12.9 and 66.2% of all patients, Allergy and Asthma Proceedings 331

Table 3 Clinical characteristics of initial reaction to ingested peanuts Organ System Involved No. of patients Skin 59 Gastrointestinal 3 Respiratory 3 mild 3 severe Skin gastrointestinal 10 Skin respiratory 12 mild 6 severe Respiratory gastrointestinal 1 Skin respiratory gastrointestinal 1 Anaphylactic shock 6 Total 104 Skin: Contact rash or hives, generalized urticaria/angioedema, and worsening dermatitis. Gastrointestinal: Nausea, vomiting, abdominal cramps, and diarrhea. Respiratory: Mild symptoms, rhinitis and conjunctivitis; severe symptoms, throat tightness, mouth swelling, laryngeal edema, cough, and wheezing Anaphylactic shock: decreased blood pressure and decreased level of consciousness. Table 4 Demographic features and workup results Property of patients of who presented with anaphylactic shock after ingestion of peanuts Age (yr) Sex Accidental Exposure Skin Test Initial PN-IgE (ku A /L) Last PN-IgE (ku A /L) Associated Allergic Conditions 3 M Yes, A 18.3 12.8 Rhinitis, asthma, and sesame allergy 15 F Yes, A 100.0 100.0 Fish, shellfish, and tree nuts allergy 3 M No 9.85 5.33 Asthma, environmental, fish, tree nuts, and carrot allergy 3 F Yes, A Not done 2.94 0.99 Rhinitis, asthma, environmental, and egg allergy 1.5 F Yes, S RE 100.0 100.0 Rhinitis and environmental allergy 3 M Yes, A 7.7 5.78 Rhinitis, environmental, egg, and tree nuts allergy Table 5 Clinical characteristics of patients OceanSide grouped into Publications (a) patients with gradual decline of PN-IgE and (b) patients with PN-IgE elevation during follow-up PN-IgE Gradual PN-IgE Elevation p Decline Group Group Value Median/Mean Value Median/Mean Value Age first reaction (yr) 2.47/2.24 1.65/1.20 0.025 Age first RAST (yr) 7.15/4.95 6.09/4.26 0.349 Baseline RAST (ku A /L) 50.52/45.43 18.91/23.27 0.001 Last RAST (ku A /L) 44.47/44.76 31.60/34.17 0.378 Age at Last RAST (yr) 9.51/4.98 8.43/4.64 0.315 respectively, experienced a significant decline of their PN-IgE levels. The survival curves representing the PN-IgE level declines in each group of patients at 5 years post initial PN-IgE measurement are presented in Fig. 1. Additional analysis of the data from the serial PN-IgE measurements showed that the younger age at first reaction and PN-IgE measurement predicted a longer decline interval of PN-IgE levels (Fig. 2, A and B). Gender was not a predicting factor of PN-IgE decline interval (results not shown). Subgroup analysis of patients with atopic dermatitis showed that their decline in PN-IgE levels was not different from the rest of the patients (p 0.608). 332 May June 2008, Vol. 29, No. 3

Table 6 Patients Yearly decline rate of PN-IgE after initial measurement Total n 118 Baseline PN-IgE Level (ku A /L) <17.5 n 52 17.5 100 n 30 >100 n 32 At 1 yr (95% CI) 4.7% (0.7 8.7%) 6.3% (0.0 13.3%) 3.5% (0.0 10.0%) 3.3% (0.0 9.8%) At 2 yr (95% CI) 12.9% (6.0 19.8%) 16.3% (5.0 27.6%) 12.3% (0.0 25.4%) 7.9% (0.0 18.7%) At 3 yr (95% CI) 28.9% (17.7 40.0%) 21.0% (7.1 34.9%) 48.8% (24.7 73.0%) 22.3% (1.4 43.2%) At 4 yr (95% CI) 47.5% (31.1 63.8%) 50.3% (25.6 75%) 59.1% (32.7 85.5%) 22.3% (1.4 43.2%) At 5 yr (95% CI) 66.2% (42.1 90.3%) 50.3% (25.6 75.0%) 72.7% (44.7 100%) NA Property of Figure 1. Decline rate (%) of PN-IgE as determined by baseline values (a) 17.5 ku A /L, (b)19.5 100 ku A /L, and (c) 100 ku A /L (106 values given for statistical purposes). Event-significant decline of PN-IgE value as defined. DISCUSSION This study was a retrospective chart review of a peanut-allergic patient cohort seen at our Allergy Clinic. These patients were diagnosed and followed up longitudinally based on significant history of reaction to peanuts, positive SPT, and serial measurements of PN-IgE. An important finding of this study was OceanSide that the Publications features of initial reaction to peanuts did not predict the magnitude of PN-IgE levels. Fifty-six percent IP: of212.209.13.15 our patients (59 of 104) manifested skin symptoms only at their first reaction to peanuts and their baseline PN-IgE levels did not differ significantly from those who manifested additional symptoms such as respiratory, gastrointestinal, or combination. Our findings are different from those reported by Vanderleek et al., 10 who showed a significant difference in peanut IgE levels between 11 patients who reacted with skin symptoms only and the other 51 patients who reacted with skin symptoms initially but had subsequent gastrointestinal and respiratory complaints on repeated exposures. Contrary to our study, they included not only symptoms during first reaction to peanuts, but also during subsequent exposures and peanut challenges. Second, as previously shown, PN-IgE levels vary greatly among patients, so the number of participants included in the analysis will affect the results as well. Figure 2. Decline rate (%) of PN-IgE as determined by (A) age at first reaction and (B) age at first PN-IgE measurement. Younger age at first reaction ( 2 years old) and first PN-IgE measurement ( 5 years old) predicted longer times for significant decline of PN-IgE (p 0.044 and p 0.001, respectively). Twenty- three percent of our patients had atopic dermatitis, less commonly than previously reported. 8,9,12 Despite that, the baseline PN-IgE value (median, 18.5 ku A /L) of our cohort was high. This could have been caused by selection bias; therefore, patients Allergy and Asthma Proceedings 333

with higher baseline PN-IgE were more likely to get reaction to peanuts and the referral to the allergist repeated measurements and were selected for this review. and possibly because of reluctance of physicians and Despite higher baseline PN-IgE values, a sub- parents to perform blood tests in very young chil- group analysis of atopic dermatitis patients showed dren, which is a common occurrence in clinical practice. that atopic dermatitis had no effect on PN-IgE decline Our statistical analysis indicated that the pea- over time when compared with the patients without nut-allergic patients who were 5 years old at the atopic dermatitis. This finding could be of clinical significance time of first measurement of PN-IgE, had a lower when following up such patients. rate of significant decline in peanut-specific levels at During our longitudinal follow-up, we looked at 5 years when compared with older patients. Similarly, the repeated measurements of PN-IgE levels and van Odijk et al. have looked at PN-IgE levels at estimated the amount of change over time. To do baseline and 5 years later in 86 peanut-sensitive that, we used an arbitrary definition of PN-IgE significant patients and reported that patients who were 6 change, which does not necessarily translate years old at the time of initial tests were more likely into clinical significance but it was rather used here to have higher titers of PN-IgE when tested 5 years to measure the amount of decline. We found that a later than their older peers also tested after a 5-year great majority of our patients experienced a decline interval. 11 in PN-IgE levels during the follow-up. The rate of Another interesting finding of our study was that decline was a factor of the baseline value and the age 20% of our patients experienced an elevation of PNat first reaction to peanuts. Based on the initial peanut-specific IgE during follow-up despite the implementation of levels, our patients were grouped into peanut avoidance measures. These patients were sig- Property of three categories: 17.5, 17.5 100, and 100 ku A /L. nificantly younger (median age, 1.65 versus 2.47 At 5 years, the estimated highest rate of decline was years) at the time of the first reaction to peanuts and among the patients with baseline PN-IgE values between also had significantly lower baseline PN-IgE levels 17.5 and 100 ku A /L; 73% of patients in this (median value, 18.91 ku A /L versus 50.52 ku A /L) group would show a decline to the range of 3.5 when compared with those who showed a gradual 49.99 ku A /L. On the contrary, only 8% of our patients decline in levels. There was no significant difference with baseline PN-IgE values of 100 ku A /L of the age at baseline PN-IgE measurement between would decline to the next CAP-RAST class 50 100 two groups (p 0.349). These data suggest that ku A /L at 5-year follow-up. Based on our findings, younger patients (median, 1.65 years at first reaction) we can conclude that a patient with initial PN-IgE with lower baseline PN-IgE levels (median, level between 17.5 and 100 ku A /L has a three out of 18.91 ku A /L) have not yet mounted the maximal four chance of having a decline of levels to the lower IgE-mediated response to peanuts at the time of the RAST class at the next measurement of PN-IgE 5 first evaluation; therefore, they are more likely to years apart, whereas a patient with initial PN-IgE experience elevation of levels in the next measurement. levels of 100 ku A /L is more likely to show no Another explanation could be that these chil- change in the subsequent measurement. Moreover, dren were exposed to traces of peanuts intermittently a dose that was adequate to elicit a PN-IgE one-half of patients with initial PN-IgEOceanSide values of Publications 17.5 ku A /L would have lower PN-IgE value in the elevation but not a clinical reaction. Similarly, Van next measurement, whereas the rest would be more likely to develop a plateau, where repeated measurements would show no significant change. The age at first reaction to peanuts was a significant predictor of PN-IgE decline. Patients who were 2 years old at the time of first reaction to peanuts were more likely to experience a longer decline interval of peanut-specific levels than their older peers (p 0.044). This might be explained by the fact that the same group of patients was more likely to show an elevation of PN-IgE titers initially and then experience decline. The decline of PN-IgE levels was similarly predicted by the patient s age at first PN-IgE measurement. The median age at first reaction to peanuts was 1.5 years, whereas at first PN-IgE measurement age was 6 years, 4.5 years later. We think that this delay has to do with the time gap between the first Odijk et al. documented an elevation of peanut-specific levels during their follow-up and concluded that exposure to peanuts during the study, i.e., 5 6 years since diagnosis did not seem to affect the PN- IgE levels. 11 More studies are required to elucidate the causes of PN-IgE elevation despite the implementation of peanut avoidance measures. Several studies have looked at the relationship between the PN-IgE levels and results of upcoming peanut challenges in peanut-allergic patients. 7 9,13 PN-IgE levels 15 ku A /L have been reported to have a 95% likelihood of predicting a positive peanut challenge. 12 Perry et al. recommended that for patients with a clear history of reaction to peanuts a challenge should be considered when the PN-IgE level is 2 ku A /L and the chance of passing it is approximately 50%. 7 Our cohort had a median baseline PN-IgE value 18.5 ku A /L. Among 104 patients with a clear history of 334 May June 2008, Vol. 29, No. 3

reaction to peanuts, only 14 patients (13.5%) had a decline of PN-IgE to 2 ku A /L during 5-year follow-up and open peanut challenges were offered. The survival curves of PN-IgE levels over a 5-year period showed an overall 12.9 and 66.2% decline rate at 2 and 5 years post initial measurement and there were significant differences among patient groups based on initial PN-IgE values, age at first reaction to peanuts, and first measurement of PN-IgE levels, as previously mentioned. With the data from this study, we can recommend that for most of the patients with significant history of reaction to peanuts and positive SPT, it probably is adequate to check PN-IgE at baseline and every 3 5 years to monitor the PN-IgE decline to levels with higher prediction of negative future peanut challenges. More frequent measurements might be considered in older kids with lower baseline PN-IgE levels. This study was a retrospective review and as such has several limitations. The decision to order PN-RAST Property of REFERENCES testing of the patients was arbitrary. Some patients had lack of regular yearly follow-up, so the intervals between PN-RAST testing were not uniform. At last, while all patients avoided peanuts, the other atopic conditions/food allergies were not controlled uniformly as they would be in a prospective study. Despite the limitations, this is the first study that reviews data from monitoring peanut- allergic patients with PN-IgE to identify in a timely fashion those patients with the greatest likelihood of a future negative challenge. It would be interesting to see if our data can be replicated in other peanut-allergic patient cohorts. Our recommendation would be of assistance for allergists, pediatricians, and family physicians wishing to minimize the frequency of blood taking, which can be especially unpleasant for younger children. OceanSide Publications 1. Sicherer SH, Munoz-Furlong BA, Burks AW, and Sampson HA. Prevalence of peanut and tree allergy in US determined by a random digital telephone survey. J Allergy Clin Immunol 103: 559 562, 1999. 2. Kanny G, Moneret-Vautrin DA, Flabbee J, et al. Population study of food allergy in France. J Allergy Clin Immunol 108: 133 140, 2001. 3. Emmet SE, Angus FJ, Fry JS, and Lee PN. Perceived prevalence of peanut allergy in Great Britain and its association with other atopic conditions and with other household members. Allergy 54:380 385, 1999. 4. Sicherer SH, Munoz-Furlong A, and Sampson HA. Prevalence of peanut and tree nut allergy in the United States determined by means of a random digit dial telephone survey: A 5-year follow-up study. J Allergy Clin Immunol 112:1203 1207, 2003. 5. Grundy J, Matthews S, Bateman B, et al. Rising prevalence of allergy to peanut in children: Data from 2 sequential cohorts. J Allergy Clin Immunol 110:784 789, 2002. 6. Kagan RS, Joseph L, Dufresne C, et al. Prevalence of peanut allergy in primary-school children in Montreal, Canada. J Allergy Clin Immunol 112:1223 1228, 2003. 7. Perry T, Matsui Elisabeth, Conover-Walker MK, and Wood RA. The relationship of allergen-specific IgE and oral food challenge outcome. J Allergy Clin Immunol 114:144 149, 2004. 8. Skolnick HS, Conover-Walker MK, Koerner CB, et al. The natural history of peanut allergy. J Allergy Clin Immunol 107:367 374, 2001. 9. Fleischer DM, Conover-Walker MK, Christie L, et al. The natural progression of peanut allergy: Resolution and the possibility of recurrence. J Allergy Clin Immunol 112:183 189, 2003. 10. Vander Leek TK, Liu AH, Stefanski K, et al. The natural history of peanut allergy in young children and its association with serum peanut-specific IgE. J Pediatr 137:749 755, 2000. 11. van Odijk J, Bengtsson U, Borres MP, et al. Specific immunoglobulin E antibodies to peanut over time in relation to peanut intake, symptoms and age. Pediatr Allergy Immunol 15:442 448, 2004. 12. Sampson HA, and Ho DG. Relationship between food-specific IgE concentrations and the risk of positive food challenges in children and adolescents. J Allergy Clin Immunol 100:444 451, 1997. 13. Carr WW. Clinical pearls and pitfalls: Peanut allergy. Allergy Asthma Proc 26:145 147, 2005. e Allergy and Asthma Proceedings 335