INTRODUCTION — The tools available for diagnosing food allergy include the clinical history, physical examination, trial elimination diets, diet diaries, skin prick testing (SPT), allergen-specific serum immunoglobulin E (IgE) testing, and, for certain foods, IgE testing for individual allergen proteins (component testing). Clinician-supervised oral food challenges remain the gold standard and are often required to confirm or rule out the diagnosis because of the limitations in the diagnostic accuracy of the other methods available and potential overdiagnosis of food allergies.
Improved or new testing methodologies are needed for determining the presence and severity of a food allergy and the likelihood of resolution of an allergy. This topic reviews improvements in available diagnostic tools and new testing methods that are in development. Current diagnostic tools are discussed separately, as is the initial evaluation of a patient with suspected food allergy. (See "Diagnostic evaluation of IgE-mediated food allergy" and "History and physical examination in the patient with possible food allergy".)
OVERVIEW — The role of medical testing for food allergy must be thought of in the context of the suspected mechanism of pathogenesis (ie, IgE or non-IgE mediated) and the limitations of available testing. IgE-mediated food allergy is more common, better understood, and more actively researched. There is also active research on testing for non-IgE-mediated and "IgE-associated" diseases that are believed to be driven, wholly or partly, by dietary proteins (eg, eosinophilic esophagitis [EoE], atopic dermatitis). (See "Diagnostic evaluation of IgE-mediated food allergy" and "History and physical examination in the patient with possible food allergy".)
There is a persistent and wide gap between the number of individuals who believe they have food allergies and the true prevalence (approximately 30 and 5 percent, respectively) [1]. Part of this discrepancy stems from confusion regarding allergic (ie, immune-mediated) reactions versus other adverse reactions to foods (eg, intolerances). Diagnostic test limitations, including the low specificity of specific IgE for true allergy (particularly when used qualitatively), also contribute to overdiagnosis.
Research efforts are focused on improving diagnostic tests and on developing tests that have better prognostic performance to better guide clinical management of patients. To address limitations, investigators are attempting to characterize the patient immune response in ever greater detail to find phenotypes that more closely correlate with the clinical disease aspect of interest.
Instead of measuring the IgE response to complex allergen extracts, investigators are evaluating the specificity of that response on the level of individual allergens ("component-resolved diagnosis") or the epitopes of those allergens ("epitope mapping" or "profiling"). Additionally, the fraction of total IgE that is allergen specific, the affinity of the allergen-specific IgE, and the presence and specificity of other, potentially inhibitory antibodies (eg, immunoglobulin G4 [IgG4]) can be measured. Methods of assessing the patient's effector cell response to allergen (eg, "basophil activation testing" [BAT]) are also under investigation.
Advances in precision medicine have been employed to better understand the characteristics of the individual patient's immune response to foods in an attempt to better predict and potentially avoid the development of food allergies [2].
TECHNOLOGICAL ADVANCES — Several technical developments have gone hand-in-hand with the search for better diagnostic tools for food allergy. Current diagnostic techniques based upon crude allergen extracts suffer from several factors affecting specificity, including potential cross-contamination with other allergens [3] and the presence of highly cross-reactive carbohydrate epitopes [4]. There has been a massive effort to identify, clone, and express major allergens from hundreds of sources. Assays are also being developed in tandem with the molecular characterization of allergens. As an example, microarray-based assays allow numerous proteins to be probed simultaneously, thus requiring minute amounts of patient serum. In addition, recombinant allergens can be used for traditional skin prick testing (SPT) or other functional tests, such as basophil activation.
The primary biomarkers for allergic disease are still mainly IgE antibodies, measured in various ways as detailed above, and basophils measured in functional tests, but epigenetics, specifically DNA methylation patterns, are under investigation as potential biomarkers for allergen exposure [5].
Allergen microarrays — The use of microarrays for food allergy diagnosis has two direct applications: evaluation of sensitization to multiple allergens using protein arrays and evaluation of allergen epitope recognition patterns using peptide arrays [6]. In addition to simply detecting sensitization, novel microarrays are also under study to distinguish between binding to epitopes that are associated with allergic disease and epitopes associated with tolerance.
Protein allergen microarrays have been developed with up to 5000 allergens on a single assay [7-15]. Overall, microarray assays correlate well with serum IgE measurements. However, they tend to be somewhat less sensitive since they contain many sequenced epitopes but not all that a patient may be exposed to when ingesting the fresh food. There are a few protein microarray panels that focus on food allergy specifically, although none have US Food and Drug Administration approval yet. This approach is promising, but more prospective studies showing clinical application need to be performed.
Epitope pattern analysis by microarray has been applied to food allergy. However, existing literature on the correlation between epitope recognition and clinical reactivity or prognosis is scant. Several studies linking epitope recognition patterns and clinical allergy using methods other than peptide and protein microarray have been performed previously [16,17]. Microarray is a versatile tool to define and identify binding patterns and informative epitopes [18,19].
An advantage of the microarray approach is that it allows for practical, whole-sequence epitope analysis using individual patient serum [18,19]. As an example, in the case of peanut allergy, there is a high degree of heterogeneity in epitope recognition that is not practical to measure with older methods [20].
Additional innovations include multiple antibody isotype detection using different fluorochrome channels and more sophisticated statistical analyses that can detect multiple epitopes recognized by a population of patients even when the epitopes share overlapping sequences [21]. Measurement of antibody binding to bead immobilized epitopes in combination with machine learning has shown promising results [22].
Microarray technology allows for evaluation of a patient's B cell repertoire to a given set of allergens. Peptide and protein microarrays may be particularly useful in better understanding associated antibody repertoire changes in both natural and immunotherapy-induced tolerance.
Additional recombinant-based approaches have begun to address this problem [23]. As an example, the microarray technique was employed in investigation of conformational food allergen epitopes, determining IgE clustering epitopes on recombinant alpha-lactalbumin in patients with cow's milk allergy [24]. Preventative and therapeutic strategies can be developed based upon these mapped IgE epitopes.
A further development of protein microarray to better identify allergenic epitopes and help differentiate from epitopes that are not associated with clinical disease has been the coupling with a T7-bacteriophage-dependent technique. This technique results in longer peptide sequences of higher quality [25,26].
Basophil activation testing — Basophil activation testing (BAT) allows for functional testing to allergen exposure in vitro. Basophils are specifically sensitized effector cells that are readily accessible from the peripheral blood. Basophil responsiveness may have higher specificity than specific IgE testing when judging clinical reactivity because activation of these cells is a biologic response [27,28]. These tests use either autologous donor cells or cells from established cell lines.
The most common form of the BAT involves direct activation with antigen [29]. Flow cytometry is used to identify the population of basophils and measure their activation based upon upregulation of cell-surface proteins (eg, CD63 and CD203c) that are associated with release of mediators [30].
Mast cell activation testing — Mast cells are considered the main effector cells in an allergic reaction. Primary human blood-derived mast cells generated from CD117+ precursors are passively sensitized with serum of allergic patients and then incubated with allergen [31].
Gene-level testing for food allergies — Genetic screening to identify persons affected by or at risk for a disease is used in the diagnostic evaluation for many disorders. Testing of genes and gene interactions, as well as testing of epigenetic changes that do not affect DNA directly, is primarily used in research settings to improve the understanding of disease mechanisms and develop biomarkers to identify at-risk populations. A better understanding of the complex relationship between genetic predisposition and environmental factors including the microbiome can help identify biomarkers and genetic patterns and may lead to precise diagnostics and targeted therapies in the future.
One study investigating genome-wide DNA methylation of blood mononuclear cells detected a signature of 96 CpG sites that predicted clinical food challenge outcomes [32]. A trial investigating the association of certain genetic variants with challenge-proven peanut allergy identified alleles at the human leukocyte antigen (HLA) DR and HLA-DQ locus as genetic determinants for peanut allergy [33]. By pooling of data using meta-analysis, the C11orf30 locus reached genome-wide significance as a gene locus associated with food allergy. The same study identified several genes involved in epigenetic regulation of gene expression [34].
A systematic review cited strong evidence for a role of genetic variants in filaggrin (FLG), HLA, and interleukin (IL) 13 in the development of food allergies, making these genes a potential target for diagnostic tests [35].
The role of the human microbiome as an important player in the development of food allergies, and also as potential target for therapeutic interventions in food allergies, has resulted in multiple studies investigating the microbiome in food allergy [36,37].
IgE-MEDIATED FOOD ALLERGY
Diagnosis — A key concept regarding IgE-mediated food allergy is the distinction between sensitization and allergy. Sensitization refers to the presence of IgE capable of binding to a particular antigen, as evidenced by detection in serum by an allergen binding assay or indirectly by mast cell degranulation induced by skin prick testing (SPT). Mast cell activation by non-IgE-mediated mechanisms is theoretically possible with SPT, but this is not regarded as a common problem, unlike with drug allergy. Nonspecific IgE binding and technical restrictions are potential limitations of serum binding assays.
More fundamentally, however, production of specific IgE is necessary for immediate hypersensitivity but is not sufficient for clinical allergy. Individuals may have positive testing (SPT or in vitro) and yet fully tolerate the antigen in the diet.
Component-resolved diagnosis — Several plant foods contain proteins that are homologous to plant aeroallergens. This cross-reactivity contributes to the overdiagnosis of food allergies because it results in positive tests that may not be clinically relevant. A "component-resolved" or "molecular allergen" diagnostic approach, in which specific IgE is measured to defined allergens individually rather than as a whole extract, can enhance specificity and improve understanding of cross-reactivity and cosensitization [38]. Component testing is available for plant- and animal-derived allergens. Component-resolved testing is reviewed in greater detail separately. (See "Pathogenesis of oral allergy syndrome (pollen-food allergy syndrome)" and "Component testing for pollen-related, plant-derived food allergies" and "Component testing for animal-derived food allergies".)
Allergen epitope profiling — Food allergens must at least partially survive digestion and absorption from the gastrointestinal tract to be immunogenic. This fact has led to the hypothesis that persons who have IgE antibodies recognizing a greater number or a specific pattern of sequential epitopes (eg, those not easily destroyed by denaturation and partial digestion) are more likely to have clinical allergy rather than asymptomatic IgE sensitization [39]. Multiple methods to identify IgE epitopes have been reported, including spot assays, microarrays, arrays using color-coded beads that are precoated with antibodies, and arrays in combination with basophil activation testing (BAT) [40].
Multiple studies have found correlations between epitope patterns, defined using peptides, and clinical disease. Examples of investigations supporting the concept that IgE epitope-recognition patterns can predict clinical outcomes are:
●Researchers analyzed IgE binding of eight sequential epitopes of the major peanut allergens, Ara h 1, 2, and 3. Ninety-three percent (n = 14) of patients with clinical reactivity to peanut recognized at least one of the immunodominant epitopes of Ara h 1 or 2 in contrast to 12 percent (n = 2) of peanut-tolerant subjects with IgE sensitization [41]. Furthermore, reactive patients recognized a greater number of epitopes than tolerant patients. Testing for IgE-binding epitopes may decrease the need for oral challenges to determine if sensitized patients are clinically reactive or tolerant.
●Another group identified specific epitopes of major wheat allergens, omega-5 gliadin and a high-molecular-weight glutenin subunit, that were recognized by 97 percent (n = 29) of patients with wheat-dependent exercise-induced anaphylaxis but were not recognized by wheat-sensitized patients with atopic dermatitis [42,43]. (See "Exercise-induced anaphylaxis: Clinical manifestations, epidemiology, pathogenesis, and diagnosis".)
However, one study has provided evidence that short sequential epitopes contribute little to IgE binding to two major food allergens, Ara h 2 and Pen a 1, from peanut and shrimp, respectively [44].
The combination of measuring IgE and IgG4 responses on bead immobilized epitopes and machine learning has greatly increased the ability to make clinically relevant predictions on food allergy [22,45].
IgE antibody mapping to epitopes that are associated with allergic disease versus tolerance is under investigation to decrease potential overdiagnosis of food allergies [22].
Cellular/functional assays — The use of BAT in food allergy diagnosis is increasingly under investigation. One study examined its performance for predicting challenge outcomes in a group of 71 children with hen's egg or cow's milk allergy previously diagnosed by challenge or convincing history [46]. Performance characteristics using optimal cutoff points were comparable with earlier studies following the same design for skin or IgE testing. Similar results were seen in children with peanut allergy [47].
Other studies suggest that BAT is also comparable with skin or in vitro (CAP-FEIA) specific IgE testing in its ability to distinguish clinical allergy from sensitization alone in patients with food pollen allergy syndrome [48-51]. One study examined the use of BAT with recombinant allergens for the diagnosis of food allergy in three groups of 20 birch-allergic patients with clinical allergy to apple, carrot, or celery [50]. Using the recombinant profilin allergens (Mal d 1, Dau c 1, and Api g 1), they achieved sensitivities and specificities of 75 and 68 percent, 75 and 77 percent, and 65 and 100 percent for apple, carrot, and celery, respectively. These results were comparable with CAP-FEIA but less sensitive than prick-prick skin testing using fresh fruit or vegetable.
BAT may also help differentiate between sensitization and clinical allergy in patients with a history of clinical reactivity to one food allergen and a positive test to another cross-reactive allergen. Ses i 6 is a seed storage protein that has partial immunologic cross-reactivity with a walnut protein, Jug r 4. Basophils passively sensitized with sera from sesame-allergic, walnut-sensitized patients are activated by Ses i 6 but not Jug r 4 [52]. Similarly, BAT was superior to skin prick testing and peanut and component-specific IgE in distinguishing between peanut allergy and tolerance, reducing the number of oral food challenges needed [53].
BAT on a large cohort of children participating in the Learning Early About Peanut Allergy (LEAP) trial has shown high specificity (97 percent) and sensitivity (100 percent) in identifying children with severe allergic reactions to peanuts in this cohort, making it a clinically relevant test [54].
T cell responses to food allergens may also eventually prove useful in determining if a patient is sensitized or clinically allergic. As an example, T cell proliferation to crude peanut extract and major peanut allergens was studied in children with peanut allergy (n = 18) or peanut sensitization (n = 7) and nonallergic adults (n = 11) [55]. The T cell response to crude peanut extract was stronger in children with peanut allergy than those with peanut sensitization or adults without peanut allergy. Only the children with peanut allergy had detectable interleukin (IL) 13 production in response to major peanut allergens (Ara h 1, Ara h 3, and Ara h 6).
While atopic persons display a significant skewing towards a T helper cell type 2 (Th2) phenotype, allergen-specific T cells are rare. A specific Th2 memory cell type has been reported (Th2a-cell) that secretes IL-5 and IL-9 and is almost exclusively found in allergic persons [56].
The presence of regulatory T cell subsets early in life are a marker for the development of food allergies [57].
RNA tests — Nonencoding microRNAs (miRNAs) are other potential eosinophilic esophagitis (EoE) biomarkers. Some of these miRNAs are differentially expressed in EoE, either constitutively or specifically with inflammation, compared with normal healthy controls and those with reflux esophagitis [58]. Several of these miRNAs are detectable in peripheral blood, making them promising noninvasive alternatives to endoscopic biopsy for diagnosing and/or monitoring disease.
The role of microRNAs in food allergy increasingly under investigation. MicroRNA 193a-5p, which is involved in regulation of IL-4, is downregulated in the peripheral blood mononuclear cells (PBMCs) of children with milk allergy [59]. Additionally, programmed cell death protein 1 (PD-1) is involved in inhibiting CD4+ T cell responses [60].
The use of messenger RNA (mRNA) in the diagnosis of food allergy is limited. Transcriptional profiles of specific cell populations, such as PBMC, may play a role in the diagnosis of different food allergies. PBMC response to hen's egg protein assessed by microarray showed expression of genes associated with allergic inflammation in children with egg allergy but not children tolerant to egg but with other food allergies [61].
Prognosis — Beyond accurate diagnosis of food allergy, there are some closely related prognostic questions confronting food-allergic patients, their caregivers, and health care providers. Two of the most common questions are "Who is at risk for anaphylaxis?" and, for children, "Will the allergy be outgrown?"
Clinical severity — The ability to identify those at higher risk for severe reactions and/or reactive at lower thresholds of peanut exposure would potentially allow for more efficient targeting of resources to those individuals. However, reaction severity cannot be reliably predicted from food-specific serum IgE concentrations. As in the case of accurate diagnosis, cross-reactivity between inhalant and food proteins has a large impact on the poor association between measured specific IgE to whole allergen source extracts and the clinical phenotype.
Additional reasons for poor reliability of food-specific serum IgE concentrations in determining allergy severity may include variability in target organ sensitivity (eg, the presence of asthma, which is associated with more severe reactions), the patient's general health, and the dose of allergen ingested.
Some studies with larger samples or more homogeneous study populations have found a greater risk of severe reactions in those with higher peanut-specific IgE concentration [62,63].
Component-resolved diagnosis — The pattern of specific IgE reactivity to defined allergens can help define risk for anaphylaxis for at least some allergens, such as peanut. As an example, sensitization to pollen-related proteins, such as Bet v 1 homologs (PR-10 proteins) or profilins, is generally associated with limited (oropharyngeal) reactivity, whereas sensitization to more stable proteins, such as the seed storage or lipid transfer proteins, is often associated with systemic reactions [38,64-68]. The usefulness of component testing for animal food allergens (eg, cow's milk, hen's egg) is less well established. Component testing is discussed in greater detail separately. (See "Component testing for pollen-related, plant-derived food allergies" and "Component testing for animal-derived food allergies".)
Attempts to determine the probability of a positive challenge based upon specific IgE values alone versus specific to total IgE ratios have not been successful [69].
Allergen epitope profiling — A positive relationship between the diversity of the IgE response to both peanut and cow's milk proteins and reaction severity has been demonstrated. The total number of peanut or cow's milk proteins (epitopes) recognized by patient IgE directly correlates with severity of clinical symptoms during the respective food challenge [70-72].
Similarly, in a retrospective study, IgE from patients with more severe reactions recognized a greater number of peanut allergen epitopes compared with those who had milder reactions [20]. In these same subjects, commercially available tests for peanut-specific IgE levels did not distinguish between the two groups.
Cellular/functional assays — As an ex vivo test, BAT holds the potential to supplement or replace more invasive allergy tests. BAT may be useful for identifying patients with a more severe phenotype of food allergy. As an example, basophils of patients with more persistent and severe cow's milk allergy had basophils that were more reactive to milk in vivo, whereas natural resolution of the allergy was associated with basophil suppression [73]. BAT can also discriminate between children with cow's milk allergy that is still present and those who have outgrown their cow's milk allergy but are still sensitized [74]. In addition, BAT is useful in the identification of different allergy phenotypes [75]. Prospective studies are needed to validate this observation.
A major limitation of BAT is the need for fresh blood and the short viability and potential nonreactivity of patient basophils. Mast cell activation testing (MAT) uses patient serum, which activates a human mast cell line, leading to more stable and reproducible results. Using MAT, one study was able to identify patient at risk for severe peanut-related reactions [31].
Disease persistence — An important question for patients and clinicians managing the allergic patient is whether or not the allergy is likely to persist. Differences in patient IgE responses may be helpful in making this determination. This will be increasingly important as new therapies become available that focus on patients not likely to outgrow their allergy spontaneously. (See "Food allergy in children: Prevalence, natural history, and monitoring for resolution".)
Allergen epitope profiling — Differences in epitope-specific IgE are seen between patients with persistent versus transient food allergy:
●Two studies of cow's milk allergy, one focusing on epitopes of whey proteins [76] and one on alpha s1-casein [16], showed that patients with transient milk allergy recognized fewer sequential epitopes than those with persistent allergy.
●A study defining IgE and IgG4 binding cow's milk allergen epitopes using peptide microarray described 10 epitopes that appear to distinguish between tolerant and reactive patient groups [77]. In another study, persistent milk allergy was associated with stable IgE epitope recognition, in contrast to transient allergy, which was associated with falling IgE but increasing IgG4 epitope recognition [78].
●Another study of cow's milk allergy found that patients who were still allergic had both high- and low-affinity IgE binding, whereas those who tolerated extensively heated milk (eg, in baked goods) or who had outgrown their milk allergy had mainly low-affinity IgE binding [72]. Using a peptide assay to evaluate both specific IgE and IgG4 to a panel of milk allergen linear epitopes, the same group has shown that they can predict tolerance of fresh versus denatured forms of milk with twice the accuracy of component allergen testing [45].
●One small study of hen's egg allergy demonstrated that all patients (n = 7) with persistent allergy recognized four major IgE-binding epitopes in ovomucoid, while none of the children who outgrew their egg allergy (n = 11) did so [17].
●Another study evaluated five IgE-binding casein peptides associated with persistent cow's milk allergy that were coupled to a commercial matrix (ImmunoCAP) [79]. Milk peptide-specific IgE was significantly higher in children who did not outgrow their cow's milk allergy by a median age of eight years compared with children whose allergy resolved by a median of three or eight years of age (early or late resolution).
●A bead-based multiplex assay has been used to map IgE binding to epitopes associated with tolerance versus clinical disease [22].
●Combination of the microarray technique with phage technology allows for differentiation between clinical tolerance and disease [26].
Cellular/functional assays — The combined expression of the T regulatory and Th2 cell markers forkhead box P3 (FoxP3), IL-16, nuclear factor of activated T cells 2 (Nfat-C2), and GATA binding protein 3 (GATA3) is predictive of persistent cow's milk allergy compared with patients who had early resolution of the allergy or who were nonatopic according to neural network analysis, which mimics biologic neural networks [80].
An elevated BAT was predictive of persisting cow's milk allergy and correlated with the eliciting dose of milk in a series of children who underwent oral challenges to reassess their allergy [81].
Other assessments — Platelet-activating factor (PAF) and the enzyme that degrades it, PAF acetylhydrolase, are potential markers of anaphylaxis severity [82]. In a case-control study, there was a direct correlation between serum PAF levels and an inverse correlation between serum PAF acetylhydrolase activity and severity of anaphylaxis. A prospective study of more than 300 cases of anaphylaxis validated this association [83]. Additional mediators evaluated in this study (mast cell tryptase, histamine, anaphylatoxins [C3a, C4a, C5a], cytokines [IL-2, IL-6, IL-10], and soluble tumor necrosis factor [TNF] receptor 1) were also found to be markers of anaphylaxis severity. Some of these mediators (mast cell tryptase, histamine, cytokines [IL-2, IL-6, IL-10], soluble TNF receptor 1) were also associated with delayed deterioration. Biomarkers that may be useful in disease monitoring for EoE are discussed elsewhere in this topic. (See 'RNA tests' above and 'Biomarkers' below.)
NON-IgE-MEDIATED OR MIXED FOOD ALLERGY — Diagnostic testing for food allergy that is not directly caused by IgE is much more difficult and less well tested.
Traditional IgE testing may be helpful for allergic diseases that are associated with IgE in many patients (eg, atopic dermatitis, allergic eosinophilic gastrointestinal disorders). However, neither the sensitivity nor the specificity of the IgE testing is certain. In addition, the causative role of food antigens in these conditions is not universally accepted, although, at least for eosinophilic esophagitis (EoE), the evidence is strong. (See "Clinical manifestations and diagnosis of eosinophilic esophagitis (EoE)" and "Atopic dermatitis (eczema): Pathogenesis, clinical manifestations, and diagnosis".)
Furthermore, some forms of food allergy (eg, food protein-induced enterocolitis syndrome [FPIES]) are unequivocally provoked by food-specific immunity but are generally not associated with IgE [84]. (See "Food protein-induced allergic proctocolitis of infancy".)
This section will briefly discuss ongoing development of potential diagnostic tools for these conditions.
Biomarkers — Many non-IgE-mediated food allergies are characterized by intestinal inflammation and increased intestinal permeability. As such, fecal biomarkers as an indicator for presence and extent of disease are under investigation. Using biomarkers may also obviate the need for frequent endoscopic procedures, which are more invasive and carry an increased risk for the patient. Various eosinophil mediators, including eosinophil cationic protein (ECP) in serum or stool, and eosinophil-derived neurotoxin (EDN) or eosinophil protein X (EPX) in urine, have been proposed for the diagnosis of food allergy [85,86].
Patients with FPIES, food protein-induced enteropathy (FPE), and food protein-induced allergic proctocolitis (FPIAP) have been found to have high levels of EDN as a marker of eosinophilic colitis, which also remains stable at room temperature for seven days [87].
Fecal calprotectin (FC), a cytosolic protein in neutrophils, macrophages, and monocytes, is a marker of intestinal mucosal inflammation [88].
Thymus and activation-regulated chemokine (TARC), which was initially investigated as a potential biomarker for severity of atopic dermatitis [89], is increased in patients with FPIES approximately 24 hours after exposure to the food triggering their FPIES symptoms, making it a potentially useful tool in FPIES diagnosis [90].
Atopy patch testing — Patch testing is a method for eliciting delayed-type hypersensitivity reactions in sensitized subjects. Patient skin is exposed to allergen for prolonged periods (usually 48 hours), and reactions are generally interpreted at 72 hours. Patch testing is the gold standard for the assessment of contact allergen sensitization but is only recently being evaluated for food allergies. This method continues to be studied, but there is no consensus about appropriate reagents, standardization, and reading of results. The atopy patch test (APT) has been applied to both IgE-associated (eg, atopic dermatitis and allergic eosinophilic gastrointestinal disorders) and non-IgE-associated conditions (eg, FPIES and EoE). It is generally considered more useful in non-IgE-mediated conditions since the T cell-mediated immune response, which is measured by this test, playing a large role in these disorders.
Atopic dermatitis — A number of studies have examined the use of the APT in children with atopic dermatitis [91-99].
One of the early studies examining the utility of patch testing in children with atopic dermatitis compared traditional skin testing, serum IgE testing, and patch testing to the results of 133 double-blind, placebo-controlled food challenges, 77 of which were positive [91]. The APT was positive in 55 percent of patients with positive food challenges (42 of 77), whereas IgE testing was positive in approximately 85 percent. However, in the 21 patients who experienced only late-onset symptoms (eg, worsening eczema greater than two hours following challenge), patch testing was superior to either serum IgE measurement or skin prick testing (SPT). The positive predictive value and negative predictive value for patch testing for these delayed reactions were 81 percent and 93 percent, respectively, compared with approximately 40 percent and 75 percent for IgE-based testing.
A study investigating the role of APT in the diagnosis of food additive hypersensitivity in children with atopic dermatitis identified carmine as a potential trigger [100].
Similarly, other studies have concluded that patch testing is more sensitive for the diagnosis of food-associated atopic dermatitis, particularly in younger children [92,93]. However, one smaller study failed to show that the APT improved the diagnostic accuracy of detecting hen's egg-associated eczema over SPT [94]. Another study found increased sensitivity but at the cost of lower specificity [95]. The authors concluded that combined testing was most beneficial.
The addition of the APT is unlikely to significantly reduce the need for oral food challenges, even though combined testing (APT plus SPT or specific IgE measurement) has improved sensitivity and specificity [96,97].
Gastrointestinal food allergy — The use of patch testing has also been investigated in children with gastrointestinal symptoms, including those with a diagnosis of allergic eosinophilic esophagitis (AEE, more commonly called just "eosinophilic esophagitis" [EoE or EE]) or allergic eosinophilic gastroenteritis (AEG or EG) [98,99,101-107]. The use of APT in the diagnosis and management of EoE is discussed in detail separately. Its utility in other gastrointestinal disorders is discussed below. (See "Allergy testing in eosinophilic esophagitis" and "Dietary management of eosinophilic esophagitis".)
The sensitivity of APT for diagnosing food allergy-related gastrointestinal symptoms increased significantly in one study when fresh foods were used for testing rather than commercial food extracts [101]. Sensitivity for fresh food versus commercial extract, respectively, was 64 percent versus 6 percent for cow's milk and 84 percent versus 5 percent for hen's egg. In this study, the specificity of APT was high regardless of the preparation used for testing (95 percent for cow's milk and 100 percent for hen's egg). Results from another study suggest that a ready-to-use APT for cow's milk may have better sensitivity and test accuracy than an APT prepared in clinic [99]. Overall, further study is needed to confirm the utility of patch testing in the diagnosis of specific food allergies in eosinophilic gastrointestinal disorders. (See "Clinical manifestations and diagnosis of eosinophilic esophagitis (EoE)" and "Eosinophilic gastrointestinal diseases".)
A retrospective study evaluated patients with gastrointestinal cow's milk allergy diagnosed by elimination followed by oral challenge [102]. These patients' symptoms were largely consistent with EoE or AEG, although they did not have biopsy-confirmed diagnoses. Of the 35 patients, 24 were found to be milk allergic by challenge. Only three of these patients had milk-specific IgE, but 19 of 24 were patch-test positive. Among the non-cow's milk allergic-group (n = 11), none were IgE positive, and one was patch-test positive, resulting in sensitivity and specificity of 79 percent and 91 percent, respectively, for the APT.
A study of 19 patients suspected of FPIES compared results of patch testing with open challenge results [107]. Twelve of these patients had positive reactions to a total of 16 foods. In all, 33 challenges were compared with patch test results. The patch test had a sensitivity and specificity of 100 percent and 71 percent, respectively. APT will be a significant tool for the safe introduction of foods in children diagnosed with FPIES if this high degree of sensitivity holds true in larger studies. (See "Food protein-induced allergic proctocolitis of infancy".)
Ultrasonographic diagnosis has been studied in symptomatic patients with non-IgE-mediated food allergy to help differentiate between an acute non-IgE food reaction and gastroenteritis. Intestinal vessel density measured using abdominal ultrasound and Doppler imaging showed significant thickening of the intestinal wall, similar to that seen in inflammatory gastrointestinal diseases [108].
T cell stimulation tests — One approach for the development of diagnostic tests in the case of non-IgE-associated food allergies thought to be exclusively mediated by cellular immunity (eg, FPIES) involves measuring T cell responses to food allergens. Patients with late-onset cow's milk reactions and negative-specific IgE testing produce higher levels of interferon (IFN) gamma than either tolerant controls or IgE-mediated reactors [109]. Peripheral blood mononuclear cells (PBMCs) from infants with delayed gastrointestinal symptoms due to milk secrete large amounts of tumor necrosis factor (TNF) alpha, peaking at days 1 and 5 after stimulation with milk proteins [110]. Infants with immediate hypersensitivity symptoms have peak TNF-alpha secretion at 24 hours but low levels by day 5. Changes in PBMC responses in patients who have outgrown their gastrointestinal allergy to milk may be related to the induction of tolerance-inducing regulatory T cells [111].
The relationship between the gastrointestinal allergic symptoms of patients in these studies and the well-defined clinical syndrome FPIES is not entirely clear, and this work has not been widely replicated. In addition, T cell-based tests are relatively laborious and require skilled technicians. However, an in vitro test would be useful for selected cases of FPIES that are suspected to be provoked by multiple food antigens.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Allergy diagnostic testing".)
SUMMARY
●Oral food challenges are often required to confirm or rule out the presence or resolution of a food allergy because of the limitations in the diagnostic accuracy of skin prick testing (SPT) and serum tests. (See 'Introduction' above.)
●Improved or additional testing methods are needed for determining the severity of a food allergy, sensitization versus clinical allergy, and likelihood of resolution of an allergy. (See 'Introduction' above and 'IgE-mediated food allergy' above.)
●Research is promising for improved diagnostics for immunoglobulin E (IgE) mediated allergy, using recombinant allergens, IgE-binding epitopes, and microarrays. (See 'IgE-mediated food allergy' above.)
●Diagnostic tests for non-IgE-mediated food allergy remain to be refined and validated before they will be available in the clinic. (See 'Non-IgE-mediated or mixed food allergy' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟