Specific IgE as the best predictor of the outcome of challenges to baked milk and baked egg
2019; Elsevier BV; Volume: 8; Issue: 4 Linguagem: Inglês
10.1016/j.jaip.2019.10.039
ISSN2213-2201
AutoresRachel De Boer, Natalia Cartledge, Sophia Lazenby, Aurelio Tobı́as, Susan Chan, Adam Fox, Alexandra F. Santos,
Tópico(s)Celiac Disease Research and Management
ResumoClinical Implications•Specific IgE performed better than skin prick test to predict reactivity to milk and egg during challenges, with 50% positive predictive value cutoffs being useful to support the decision of whether to challenge to baked milk and egg to assess for tolerance. •Specific IgE performed better than skin prick test to predict reactivity to milk and egg during challenges, with 50% positive predictive value cutoffs being useful to support the decision of whether to challenge to baked milk and egg to assess for tolerance. Cow's milk and egg allergies are common in childhood, affecting about 1.9% to 3%1Kattan J. The prevalence and natural history of food allergy.Curr Allergy Asthma Rep. 2016; 16: 47Crossref PubMed Scopus (30) Google Scholar and 0.5% to 2.5%2Rona R.J. Keil T. Summers C. Gislason D. Zuidmeer L. Sodergren E. et al.The prevalence of food allergy: a meta-analysis.J Allergy Clin Immunol. 2007; 120: 638-646Abstract Full Text Full Text PDF PubMed Scopus (1056) Google Scholar of young children, respectively, but are often outgrown.3Wood R.A. Sicherer S.H. Vickery B.P. Jones S.M. Liu A.H. Fleischer D.M. et al.The natural history of milk allergy in an observational cohort.J Allergy Clin Immunol. 2013; 131: 805-812Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar,4Sicherer S.H. Wood R.A. Vickery B.P. Jones S.M. Liu A.H. Fleischer D.M. et al.The natural history of egg allergy in an observational cohort.J Allergy Clin Immunol. 2014; 133: 492-499Abstract Full Text Full Text PDF PubMed Scopus (192) Google Scholar Oral food challenge (OFC) is the criterion standard to diagnose milk and egg allergies and to assess resolution; however, OFCs are resource-intensive and involve the risk of inducing an allergic reaction of unpredictable severity. There has been a paradigm shift in the management of milk and egg allergies in view of recent studies showing that about 77% to 83% of milk-allergic patients tolerate baked milk (BM) and 75% of egg-allergic patients tolerate baked egg (BE),5Lemon-Mule H. Sampson H.A. Sicherer S.H. Shreffler W.G. Noone S. Nowak-Wegrzyn A. Immunologic changes in children with egg allergy ingesting extensively heated egg.J Allergy Clin Immunol. 2008; 122: 977-983.e1Abstract Full Text Full Text PDF PubMed Scopus (404) Google Scholar,6Nowak-Wegrzyn A. Bloom K.A. Sicherer S.H. Shreffler W.G. Noone S. Wanich N. et al.Tolerance to extensively heated milk in children with cow's milk allergy.J Allergy Clin Immunol. 2008; 122: 342-347.e2Abstract Full Text Full Text PDF PubMed Scopus (437) Google Scholar with many centers offering OFCs to assess tolerance to BM and BE as a means of encouraging inclusion of these foods to help broaden the diet, improve quality of life, and possibly assist tolerance acquisition.6Nowak-Wegrzyn A. Bloom K.A. Sicherer S.H. Shreffler W.G. Noone S. Wanich N. et al.Tolerance to extensively heated milk in children with cow's milk allergy.J Allergy Clin Immunol. 2008; 122: 342-347.e2Abstract Full Text Full Text PDF PubMed Scopus (437) Google Scholar,7Leonard S.A. Sampson H.A. Sicherer S.H. Noone S. Moshier E.L. Godbold J. et al.Dietary baked egg accelerates resolution of egg allergy in children.J Allergy Clin Immunol. 2012; 130: 473-480.e1Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar In this study, we aimed to define predictors of clinical reactivity during OFCs to milk or egg, both baked and nonbaked, to guide clinical decision making about when to refer for OFC to confirm tolerance. Clinical records of patients who underwent OFC to cow's milk (baked or fresh) or egg (baked or cooked) between January 2014 and December 2016 were reviewed to assess OFC outcomes compared with skin prick test (SPT) (Stallergenes, Antony, France/ALK-Abelló, Hørsholm, Denmark) and specific IgE (sIgE) (ImmunoCAP, Thermofisher, Uppsala, Sweden) tested before OFC. Referrals for OFCs were made at clinician's discretion, depending on the clinical history and allergy test results. Absence of reaction in the past year, SPT wheal size less than 5 mm and sIgE less than 2 KU/L, or discrepancy between history and allergy test results were the main indications for OFC. OFCs consisted of 4 doses of the challenge food administered openly after a baseline set of observations and physical examination that were repeated about 20 minutes after each dose up to a cumulative amount of 3.6 g of milk protein for BM, 8.33 g of milk protein for fresh milk (FM), 3.7 g of egg protein for BE, and 5.56 g of egg protein for cooked egg (CE) OFCs, as long as no reaction developed. If patients developed signs of an allergic reaction, the OFC was stopped, medication given, and the patients were advised to avoid the food in the diet. See this article's Online Repository at www.jaci-inpractice.org for characteristics of the study population (Table E1) and details about the OFC procedure (Table E2 and Figures E1 and E2) and statistical analyses. Over the 3-year period, 462 children underwent OFC to milk or egg and 94 (20%) had a positive OFC (Figure 1). Three (2%) of milk and 21 (7%) of egg challenges were inconclusive, due to refusal to eat the entire portion of challenge food and were excluded from the analysis. Overall, the OFCs were well tolerated, with only 6 (1%) requiring intramuscular adrenaline (see Table E3 in this article's Online Repository at www.jaci-inpractice.org): 2 to BM, 1 to FM, and 3 to BE. There was no significant difference in the presence of intermittent or persistent asthma (defined as per Global Initiative for Asthma guidelines) between children who reacted and those who tolerated the food (FM: P = .481, BM: P = .921, CE: P = .476, BE: P = .628) or between children who experienced anaphylaxis and those who experienced milder symptoms during OFC. The OFC was used as the criterion standard to define allergy to each of the foods tested. We compared the results of SPT and sIgE between subjects who had negative and positive OFC (see Tables E4 and E5 in this article's Online Repository at www.jaci-inpractice.org). Children who reacted to BM had higher SPT wheal size to cow's milk extract and higher milk sIgE. Children who reacted to FM also had higher milk sIgE. In children who reacted to BE, SPT wheal size to egg extract, SPT wheal size to raw egg, and sIgE to egg white were higher compared with those who did not react. Only SPT wheal size to raw egg and the difference in SPT wheal size to egg extract and raw egg were significantly different between CE-allergic and CE-tolerant patients. Data for sIgE to milk and to egg individual allergen components were limited because these are not routinely used in our current clinical practice. Receiver-operator characteristic curve analyses were performed to assess the utility of each test to predict the outcome of OFC and cutoff points with 100% and 50% positive predictive value (PPV) were determined where possible (see Figure E3 and Table E6 in this article's Online Repository at www.jaci-inpractice.org; see also Table I).Table IOptimal cutoffs for SPT and sIgE to cow's milk or egg white and 100% PPV cutoffs (to confirm allergy) and 50% PPV cutoffs (to determine whether to challenge) for sIgE to cow's milk or egg white∗Optimal cutoffs were defined as the best balance between sensitivity and specificity and calculated on the basis of the largest Youden index.Allergy testsBM allergyFM allergyBE allergyCE allergySPT to cow's milk/egg white extract AUC ROC 95% CI0.66 (0.54-0.78)0.56 (0.37-0.76)0.6 (0.51-0.68)0.62 (0.41-0.82) Optimal cutoffs2 mm2 mm4 mm3 mmS = 76%S = 29%S = 57%S = 43%Sp = 55%Sp = 84%Sp = 62%Sp = 81%PPV = 27%PPV = 33%PPV = 39%PPV = 20%NPV = 91%NPV = 81%NPV = 77%NPV = 93%SIgE to cow's milk/egg white AUC ROC 95% CI0.72 (0.60-0.84)0.72 (0.59-0.85)0.74 (0.64-0.83)0.75 (0.42-1.00) Optimal cutoffs3.06 KU/L0.3 KU/L2.81 KU/L0.94 KU/LS = 75%S = 89%S = 67%S = 67%Sp = 69%Sp = 54%Sp = 81%Sp = 83%PPV = 92%PPV = 28%PPV = 84%PPV = 20%NPV = 35%NPV = 96%NPV = 61%NPV = 98% 100% PPV cutoffs84.9 KU/L (100% PPV)6.60 KU/L (100% PPV)NDND 50% PPV cutoffs9.34 KU/L3.31 KU/L1.61 KU/LNDAUC ROC, Area under the receiver-operator characteristic curve; ND, not determined; NPV, negative predictive value; S, sensitivity; Sp, specificity.∗ Optimal cutoffs were defined as the best balance between sensitivity and specificity and calculated on the basis of the largest Youden index. Open table in a new tab AUC ROC, Area under the receiver-operator characteristic curve; ND, not determined; NPV, negative predictive value; S, sensitivity; Sp, specificity. This was a large study of well-characterized patients, all submitted to OFC, which allowed us to assess the utility of SPT and sIgE to predict the outcome of OFCs and to identify cutoff levels with approximately 50% PPV that can support the decision of when to challenge to BM and BE, in our patient population. This is the first study looking at the clinical utility of SPT and sIgE in the context of BM and BE in our center, which is a World Allergy Organization–accredited allergy center and one of the largest food allergy centers in the world, and one of the very few studies looking at BM and BE OFCs performed in the United Kingdom or Europe. The larger proportion of patients being challenged to BM/BE reflects our recent change in practice of proactively assessing tolerance to the baked forms of these foods as opposed to strict avoidance of all forms of milk and egg advised in the past. In addition, once a BM/BE challenge has been undertaken, it is not common practice in our unit to refer straight on for an FM/CE OFC because the diet has already been expanded and the quality of life improved. Because of the higher number of patients challenged, the cutoffs generated for BM/BE are therefore more robust than for FM/CE. The low rate (18% milk, 22% egg) of positive OFCs may reflect our proficiency in predicting allergic reactivity or alternatively may reflect a more conservative approach to challenge referrals. Our overall anaphylaxis rate was lower than in other series, which is likely to reflect differing criteria for OFC referral and/or practice for administration of adrenaline. The dosing schedule for BM and BE challenges was based on age-appropriate portions of these foods to ensure a robust definition of patient phenotype and patients' safety when eating shop-bought and home-baked products that can contain variable quantity of the baked allergen. The fact that the proportion of positive OFCs was low despite the higher protein content of the challenge doses further suggests that our population was low risk compared with other published series.5Lemon-Mule H. Sampson H.A. Sicherer S.H. Shreffler W.G. Noone S. Nowak-Wegrzyn A. Immunologic changes in children with egg allergy ingesting extensively heated egg.J Allergy Clin Immunol. 2008; 122: 977-983.e1Abstract Full Text Full Text PDF PubMed Scopus (404) Google Scholar,6Nowak-Wegrzyn A. Bloom K.A. Sicherer S.H. Shreffler W.G. Noone S. Wanich N. et al.Tolerance to extensively heated milk in children with cow's milk allergy.J Allergy Clin Immunol. 2008; 122: 342-347.e2Abstract Full Text Full Text PDF PubMed Scopus (437) Google Scholar,8Bartnikas L.M. Sheehan W.J. Hoffman E.B. Permaul P. Dioun A.F. Friedlander J. et al.Predicting food challenge outcomes for baked milk: role of specific IgE and skin prick testing.Ann Allergy Asthma Immunol. 2012; 109: 309-313.e1Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar,9Kwan A. Asper M. Lavi S. Lavine E. Hummel D. Upton J.E. Prospective evaluation of testing with baked milk to predict safe ingestion of baked milk in unheated milk-allergic children.Allergy Asthma Clin Immunol. 2016; 12: 54Crossref PubMed Scopus (9) Google Scholar Consistent with previous studies, sex, age, and atopic comorbidities including the presence of asthma were not able to predict clinical reactivity during OFC.6Nowak-Wegrzyn A. Bloom K.A. Sicherer S.H. Shreffler W.G. Noone S. Wanich N. et al.Tolerance to extensively heated milk in children with cow's milk allergy.J Allergy Clin Immunol. 2008; 122: 342-347.e2Abstract Full Text Full Text PDF PubMed Scopus (437) Google Scholar,8Bartnikas L.M. Sheehan W.J. Hoffman E.B. Permaul P. Dioun A.F. Friedlander J. et al.Predicting food challenge outcomes for baked milk: role of specific IgE and skin prick testing.Ann Allergy Asthma Immunol. 2012; 109: 309-313.e1Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar,9Kwan A. Asper M. Lavi S. Lavine E. Hummel D. Upton J.E. Prospective evaluation of testing with baked milk to predict safe ingestion of baked milk in unheated milk-allergic children.Allergy Asthma Clin Immunol. 2016; 12: 54Crossref PubMed Scopus (9) Google Scholar We found that SPT wheal size and sIgE levels were generally higher for children who reacted during OFCs. Surprisingly, SPT did not perform as well as we expected and this could be in part because in our clinic, SPT is at the core of the decision of whether to refer for OFC. Many studies have looked at the ability of allergy tests to predict positive challenges (eg, 95% PPV), which is valuable to confirm allergy, but fewer studies have looked at cutoffs to determine whether to challenge to assess resolution. We identified 50% PPV cutoffs for sIgE milk and egg white in relation to tolerance to BM and BE that can support timely decisions on referral for OFC in the future. Further studies are needed to validate our cutoffs, ideally in a larger prospective study to help establish even more reliable predictors of challenge outcomes. The study focused on a selected population of children referred for an OFC within our service and does not include all children being assessed for milk and egg allergies; therefore, the cutoff points may have limited generalizability. Compared with other studies,5Lemon-Mule H. Sampson H.A. Sicherer S.H. Shreffler W.G. Noone S. Nowak-Wegrzyn A. Immunologic changes in children with egg allergy ingesting extensively heated egg.J Allergy Clin Immunol. 2008; 122: 977-983.e1Abstract Full Text Full Text PDF PubMed Scopus (404) Google Scholar,6Nowak-Wegrzyn A. Bloom K.A. Sicherer S.H. Shreffler W.G. Noone S. Wanich N. et al.Tolerance to extensively heated milk in children with cow's milk allergy.J Allergy Clin Immunol. 2008; 122: 342-347.e2Abstract Full Text Full Text PDF PubMed Scopus (437) Google Scholar our cutoffs tended to have lower negative predictive value and higher PPV, suggesting that our population was more likely to react at a given sIgE level compared with other populations. To include a population of children that is more representative of all children with suspected food allergy seen in specialized clinics, we would have had to include highly sensitized patients. However, this would have meant that we had to either challenge children at high risk of reaction, which has ethical limitations, or determine the allergic status of children solely on the basis of SPT and sIgE results and not on the criterion standard OFC, which has its own limitations. This is why our data are important, because they reflect the decision-making process that takes place in a real-life clinic scenario. Only an unbiased approach in a purpose-designed diagnostic study in which patients undergo all tests including OFC would allow for a precise determination of global diagnostic cutoff points for the various tests. The ongoing BAT2 study (NCT03309488) will define more generalizable cutoffs for the diagnosis of BM and BE that allow dispensing OFCs with a higher degree of certainty. Skin prick test (SPT) was performed as previously described,E1Santos A.F. Douiri A. Bécares N. Wu S.Y. Stephens A. Radulovic S. et al.Basophil activation test discriminates between allergy and tolerance in peanut-sensitized children.J Allergy Clin Immunol. 2014; 134: 645-652Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar using metal lancets and commercial allergen extracts (Stallergenes, Antony, France/ALK-Abelló, Hørsholm, Denmark) as well as fresh milk (FM) and raw egg prepared on the day of the SPT. Whole milk was used for SPT to FM. In the SPT to raw egg, the preparation included both egg white and egg yolk. The difference in millimeters between the wheal diameters to FM and milk extract and between raw egg and egg extract was calculated. Specific IgE to milk or egg white was tested using ImmunoCAP (Thermofisher, Uppsala, Sweden). The oral food challenges (OFC) referrals were done at the clinician's discretion, depending on the clinical history and allergy test results. Patients underwent open OFC using standardized updosing schedules (Table E2). Fresh full fat cow's milk and scrambled egg or omelette cooked on the cooker for 5 minutes were used for FM and cooked egg OFC, respectively. Standardized recipes for milk- and/or egg-containing cupcakes were used for baked milk and baked egg OFCs (Figures E1 and E2). OFCs were considered negative when the child successfully consumed the total cumulative dose without displaying any signs of an allergic reaction and were considered positive if objective signs of an allergic reaction developed. If so, the OFC was stopped, and patients were treated according to local guidelines and at the discretion of the clinician. Tolerant children were recommended to incorporate the tolerated food form into their diet at least twice weekly, and allergic children were recommended to avoid the culprit food. For determining the significance of differences between groups, the Wilcoxon rank sum test was used. The performance of allergy tests was examined against the allergic status using receiver-operating characteristic curve analyses, with their respective sensitivity, specificity, and predictive positive and negative values. Optimal cutoff points were based on the largest Youden index and the 100% and 50% positive predictive value cutoffs were determined by approximation. Statistical analyses were done using Stata statistical software; release 15 (StataCorp, Collegue Station, Texas), and P values of less than .05 were considered statistically significant .Figure E2Recipe for standard homemade cupcakes for baked egg challenge.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure E3Receiver-operator characteristic (ROC) curves for the different tests regarding (A) BM, (B) FM, (C) BE, and (D) CE challenge outcomes. BE, Baked egg; BM, baked milk; CE, cooked egg; sIgE, specific IgE.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Table E1Baseline clinical characteristics of the study population assessed for milk or egg allergiesClinical featureMilk allergy (n = 181)Egg allergy (n = 281)Median age (y)5.36.16Other food allergy172 (97)210 (80)Atopic eczema134 (75)159 (61)Allergic rhinitis53 (30)57 (22)Persistent asthma36 (20)26 (10)Intermittent asthma20 (11.2)51 (19)Values are n (%) unless otherwise indicated. Open table in a new tab Table E2Scheduled doses of milk and egg challenge foodsChallenge foodDoses of challenge foodDose of food protein (g)BM∗For recipes used for BM and BE challenge foods, see Figure E2.Dose 1: ¼ fairy cake0.33 g milk proteinDose 2: ½ fairy cake0.65 g milk proteinDose 3: ¾ fairy cake0.98 g milk proteinDose 4: 1 fairy cake1.3 g milk proteinTotal cumulative dose: 2½ fairy cakes3.26 g milk proteinFMDose 1: 5 mL0.17 g milk proteinDose 2: 10 mL0.34 g milk proteinDose 3: 30 mL1.02 g milk proteinDose 4: 200 mL6.8 g milk proteinTotal cumulative dose: 245 mL8.33 g milk proteinBE∗For recipes used for BM and BE challenge foods, see Figure E2.Dose 1: ¼ fairy cake0.47 g egg proteinDose 2: ½ fairy cake0.94 g egg proteinDose 3: ¾ fairy cake1.41 g egg proteinDose 4: 1 fairy cake1.875 g egg proteinTotal cumulative dose: 2½ fairy cakes3.7 g egg proteinCEDose 1: 0.5 g0.06 g egg proteinDose 2: 4 g0.5 g egg proteinDose 3: 10 g1.25 g egg proteinDose 4: 30 g3.75 g egg proteinTotal cumulative dose: 44.5 g5.56 g egg proteinBE, Baked egg; BM, baked milk; CE, cooked egg.∗ For recipes used for BM and BE challenge foods, see Figure E2. Open table in a new tab Table E3Characteristics of allergic reactions during OFCs to milk or eggSymptomOFC to milkOFC to eggNausea and abdominal pain6 (18)28 (46)Pruritus12 (36)27 (44)Rash20 (26)16 (26)Vomiting2 (6)22 (36)Urticaria4 (12)17 (28)Rhinitis7 (21)15 (25)Possible neurological symptoms∗Examples of neurological symptoms are dizziness, feeling queasy, and change in behavior.5 (15)7 (11)Stridor5 (15)6 (10%)Wheeze6 (18)1 (2)Treatment administeredOFC to milkOFC to eggAntihistamine33 (100)61 (100)Salbutamol8 (24)2 (3)Intramuscular adrenaline3 (9)3 (5)Prednisolone1 (3)2 (3)∗ Examples of neurological symptoms are dizziness, feeling queasy, and change in behavior. Open table in a new tab Table E4Allergy test results comparing children with positive and negative challenges to baked and fresh cow's milkPatient characteristicsBM challenges (n = 111)FM challenges (n = 67)Positive, (n = 20)Negative (n = 91)P valuePositive (n = 13)Negative (n = 54)P valueAge (y)4.4 (2.3-8.6)4.5 (2.3-7.1)0.8603.7 (2.6-6.2)4.3 (2.7-7.3).788History of reactions to milk35% (7)40% (36)0.70515% (2)19% (10).791SPT FM (mm)5 (4-1)7 (5.5-8.5).4732 (0-4)2 (1-3).907SPT milk extract (mm)3 (1-5)2 (0-4).0310 (0-3)0 (0-1).850SPT difference FM-milk extract (mm)3 (3-3)5 (5-5).0760 (0-2)1 (0-2).282sIgE to milk (KU/L)5.91 (4.47-10.38)1.68 (2.10-3.77).0110.75 (0.93-1.68)0.27 (0.22-0.49).011BM, Baked milk; sIgE, specific IgE.Median, interquartile range, and P values using Wilcoxson rank sum test are represented for SPT and sIgE results. P values <.05 are marked in bold. Open table in a new tab Table E5Allergy test results comparing children with positive and negative challenges to BE and CEPatient characteristicsBE challenges (n = 179)CE challenges (n = 83)Positive (n = 54)Negative (n = 125)P valuePositive (n = 7)Negative (n = 76)P valueAge (y)6.0 (3.2-8.2)4.3 (2.5-8.2).15711.6 (4.9-13.3)4.1 (2.1-8.4).046History of reactions to egg11.1% (6)13.6% (17).97428.57% (2)28% (21).974SPT raw egg (mm)11 (7-15)7 (5-9).0015.5 (3-8)3 (0-6).009SPT egg extract (mm)5 (2-8)4 (2-6).0052 (0-5)1 (0-3).283SPT difference raw egg-egg extract (mm)4 (2-6)3 (2-5).8623.5 (3-4)0 (0-2).010sIgE to egg white (KU/L)3.79 (2.76-4.83)1 (0-2.12)<.0011 (0.33-1.77)0.31 (0-0.68).180BE, Baked egg; CE, cooked egg; sIgE, specific IgE.Median and interquartile range are represented for SPT and sIgE results. P values <.05 are marked in bold. Open table in a new tab Table E6Optimal cutoffs for SPT to FM/raw egg and difference between SPT to FM and SPT to milk/egg extractAllergy testsBM allergyFM allergyBE allergyCE allergySPT FM/raw egg AUC ROC 95% CI0.56 (0.53-0.6)0.54 (0.41-0.67)0.66 (0.5-0.75)0.81 (0.75-0.87)1 mm4 mm11 mm3 mmS = 100%S = 23%S = 50%S = 100%Sp = 12%Sp = 85%Sp = 82%Sp = 62%PPV =19%PPV = 27%PPV = 53%PPV = 19% Optimal cutoffsNPV = 100%NPV = 82%NPV = 81%NPV = 100%Difference between SPT to FM/raw egg and SPT to milk/egg extract AUC ROC 95% CI0.55 (0.46-0.64)0.50.58 (0.49-0.66)0.77 (0.70-0.84)10 mm12 mm2 mm0 mmS = 13%S = 0%S = 76%S = 100%Sp = 96%Sp =100%Sp = 40%Sp = 54%PPV = 40%PPV = 0%PPV = 34%PPV = 20% Optimal cutoffsNPV = 85%NPV = 50%NPV = 80%NPV = 100%AUC ROC, Area under the receiver-operator characteristic curve; BE, baked egg; BM, baked milk; CE, cooked egg; NPV, negative predictive value; PPV, positive predictive value; S, sensitivity; Sp, specificity. Open table in a new tab Values are n (%) unless otherwise indicated. BE, Baked egg; BM, baked milk; CE, cooked egg. BM, Baked milk; sIgE, specific IgE. Median, interquartile range, and P values using Wilcoxson rank sum test are represented for SPT and sIgE results. P values <.05 are marked in bold. BE, Baked egg; CE, cooked egg; sIgE, specific IgE. Median and interquartile range are represented for SPT and sIgE results. P values <.05 are marked in bold. AUC ROC, Area under the receiver-operator characteristic curve; BE, baked egg; BM, baked milk; CE, cooked egg; NPV, negative predictive value; PPV, positive predictive value; S, sensitivity; Sp, specificity.
Referência(s)