The Diagnostic Value of Delayed-Type Reactions to Perennial Aeroallergens for Atopic Disease

Abstract

Abstract: Background:

Delayed-type reactions to aeroallergens have been observed, however, their clinical significance continues to be debated.

Objective:

We assessed the prevalence and significance of delayed-type reactions to aeroallergens in atopic patients.

Methods:

Retrospective study including 266 patients with history or evidence of atopic disease (atopic dermatitis [AD], allergic rhinitis, and/or allergic asthma) and tested via either the intradermal skin test (IDT) or atopy patch test for common aeroallergens, specifically house dust mites (Dermatophagoides farinae, Dermatophagoides pteronyssinus) and perennial molds (Aspergillus fumigatus, Penicillium notatum). All patients were tested via IDT with both immediate (15 minute) and delayed (2 and 4 days) readings. Delayed reading was considered positive if the IDT injection site demonstrated at least 5 mm induration 48 hours after inoculation.

Results:

In total, 195 (73.3%) patients demonstrated an immediate-type reaction, whereas 118 (44.4%) had a delayed-type reaction. In total, 75 (28.2%) patients experienced both immediate- and delayed-type reactions, 43 (16.2%) reacted delayed-type only, and 85.3% of delayed-type reactions to individual aeroallergens were associated with eczematous lesions predominantly in air-exposed areas.

Conclusion:

Delayed-type reactions to aeroallergens are prevalent and clinically significant as a component of extrinsic AD and atopic diseases. The data support delayed reading of the IDT to guide diagnosis and management in these patients.

Capsule Summary

• The prevalence of delayed-type reactions to aeroallergens in intradermal testing among atopic patients (44%, cohort n = 266) is much higher than previously observed in the literature (6%).

• These delayed-type reactions may be either treated as an airborne contact dermatitis and/or a good marker for extrinsic atopic dermatitis (AD), directly impacting clinical decision making in the use of biologics to treat recalcitrant AD.

• These reactions can be severe and may be the only kind observed, yet intradermal tests to aeroallergens and evaluation of their delayed reactions after 1–4 days are uncommon in practice. Findings support the importance of delayed reading of the intradermal skin test to guide diagnostic and therapeutic steps in AD.

INTRODUCTION

It has been observed that patients with atopic disease and a predominantly type 2 immune response not only display the expected immediate-type I IgE-mediated symptoms, but also the delayed-type IV cellular hypersensitivity reaction to aeroallergens.^1,2 Owing to the potential clinical significance of these delayed-type reactions, some observers have suggested that patients should be patch tested for aeroallergens when indicated.³ Notably, the atopy patch test (APT) has demonstrated statistically significant results when performed in patients with atopic dermatitis (AD) sensitive to common dust mite species.⁴

Although prior studies have attempted to characterize the prevalence of delayed-type reactions to aeroallergens via APT,^1,5–7 few have done so in a large atopic population and no recent reports have established the significance of delayed reading of the intradermal skin test (IDT) in this context. The purpose of this study is to evaluate how many atopic patients reveal delayed-type reactions to common perennial aeroallergens via IDT—particularly dust mites and molds—and to evaluate how clinically relevant these reactions are.

House dust mites (HDMs) Dermatophagoides farinae and Dermatophagoides pteronyssinus are a known major cause of allergic diseases throughout the world.⁸ The prevalence of mold allergy in patients with atopy is significant as well, and Aspergillus fumigatus and Penicillium notatum are particularly relevant to induce contact allergy due to their preference for indoor home environments.^9,10 As such, these perennial aeroallergens are tested regularly in clinical practice for sensitization in atopic individuals. When skin prick tests (SPTs) to these common perennial aeroallergens are negative, we usually perform IDTs if an underlying airborne allergy and atopy are suspected based on the patient's history and clinical appearance.

Intradermal testing introduces a higher number of allergens just below the epidermis and is more sensitive than prick testing, particularly with molds.¹¹ In these patients, we observe negative SPT and specific IgE (CAP-RAST) but clearly positive immediate and sometimes delayed-type reactions via IDT, which indicates a problem with the sensitivity of SPT and IgE tests in the detection of triggers for dermatitis. Notably, the IDT has been found to detect allergy in 20% of negative SPT cases, most frequently for indoor allergens.¹²

Although clinicians routinely perform IDT in patients with negative SPT, especially those under high suspicion of allergy, delayed reading is not common practice. This is most relevant in the clinical setting as delayed-type reactions may be the only kind observed in some patients with AD and respiratory allergies.^5,13 Similar to delayed-type reactions to aeroallergens that present as dermatitis, delayed reactions manifesting as respiratory allergies are non-IgE T cell mediated with the recruitment of eosinophils, peaking 3–5 days from exposure.^14,15 To avoid severe allergic reactions, IDTs are never performed if the SPT is highly positive to the suspected allergen.

Previous studies characterizing the prevalence of delayed-type reactions have provided a range of results. The most notable use of the IDT was in a prospective study of 292 patients (200 atopic, 92 nonatopic by history) that found the prevalence to be 6% in their cohort with no significant difference between atopic and nonatopic groups.¹⁶ Conducted in 1967, this study utilized nonspecific “dust extract” formulations by Hollister-Stier and found only 17% of their atopic patients to have immediate-type sensitivity. In addition, results of the delayed reactions were not stratified among the individual atopic diseases, disallowing differentiation of prevalence between those with AD and those with respiratory allergies.

Other studies have observed the prevalence of delayed-type reactions to aeroallergens in various settings among patients experiencing AD solely via APT.^1,9,17 Some found the prevalence of delayed-type sensitivity to HDMs to be as low as 26.8% in patients with AD¹⁷ and others as high as 86.5% (78.7% in those with respiratory atopy only, 77.5% overall).¹³ One study considered delayed-type reactions to be positive only if wheal diameter increased between 48- and 96-hour readings in an attempt to rule out irritant reactions.⁴ The reliability of the APT and its allergens is further questioned when considering the proposed 500 Da limit of a compound to allow transcutaneous absorption of the allergens.¹⁸

Overall, the diagnostic ability of APT in determining AD caused by HDMs has been demonstrated to be “moderate” and correlated with the quality of the allergens, characterized by a 0.54 sensitivity and 0.72 specificity.¹⁹ The lack of standardization of the APT may be contributing to the variation in these findings.

In a recent letter, we presented the case of a 32-year-old woman office administrator with allergic perennial rhinoconjunctivitis, asthma, and pruritic eczematous lesions on her scalp, hands, and extremities.²⁰ This patient demonstrated a clear positive infiltrate only to A. fumigatus and P. notatum in delayed reading of the IDT, whereas prick, immediate IDT, and history-driven patch testing with 120 allergens read at 2 and 4 days were negative.

Her symptoms improved initially with allergen avoidance and resolved with the addition of dupilumab, suggesting that the patient's delayed-type sensitivity to molds—her sole sensitivity—was a component of her extrinsic AD.²¹ As such, this case serves as a representation of the population of interest here.

In this study, we hypothesize that delayed-type reactions to common airborne allergens such as HDMs (D. farinae, D. pteronyssinus) and molds (A. fumigatus, P. notatum) are an important activator of atopic symptoms such as dermatitis, allergic rhinitis (AR), and asthma with high eosinophilic infiltrate. The dermatitis may be an activation of underlying AD by these aeroallergens or a form of systemic or localized allergic contact dermatitis on skin exposed to them.

This type of diagnosis requires other causes of dermatitis to be excluded, particularly allergic contact sensitization to both common and uncommon contact allergens indicated by patient history tested via patch testing. Here, we evaluate the prevalence and clinical significance of delayed-type reactions to these perennial aeroallergens in atopic patients without other clinically relevant contact sensitization that can explain the skin symptoms.

METHODS

Selection of Patient Population

This retrospective study includes 266 patients with history or clinical evidence of atopic disease and completed testing via SPT and either IDT or APT. These patients were seen and tested at the Dermato-Allergy division of the Department of Dermatology at the University of Minnesota between June 1, 2019, and June 1, 2021. Atopic disease was defined as AD, AR, and/or allergic asthma and diagnosed according to the established Hanifin and Rajka criterion for AD²² and Allergic Rhinitis and its Impact on Asthma initiative guidelines for AR and allergic asthma²³ as well as clinical history and allergy testing. In total, 71.4% had a history of AD, 73.7% had AR, and 28.2% had allergic asthma.

In total, 28 patients who were initially suspected of atopic disease due to their history were ultimately diagnosed with nonallergic skin diseases such as irritant contact dermatitis after testing. Patients who had used topical steroids, antihistamines, phototherapy, or immunosuppressive medications within the past 4 weeks or had previously diagnosed skin diseases other than AD were excluded. All patients were referred to and selected in the Dermato-Allergy clinic. The high number of eczematous diseases in the sample population is explained by the primary referral being for the evaluation of these diseases. This study was determined to be exempt from review by the University of Minnesota IRB.

Skin Prick Testing

In total, 263 of the patients were tested via an SPT atopic panel to characterize immediate-type sensitization to HDMs (D. farinae, D. pteronyssinus) and perennial and seasonal molds (A. fumigatus, P. notatum, Alternaria alternata, Cladosporium herbarum) as well as other common aeroallergens at standard concentrations (ALK-Abelló, Nieuwegein, The Netherlands). These included pet dander (cat and dog) and various tree (pecan, maple, oak, sycamore, ash, white ash, river/red birch, black walnut, common cottonwood, American elm), grass (june, orchard, redtop, timothy, johnson), and weed (ragweed, mugwort, common cocklebur, lamb's quarters, rough redroot pigweed, dock/sorrel) pollens.

The SPT was read after 15 minutes, and wheal diameter was evaluated and compared with the positive histamine control (1 mg/mL; ALK). All readings were performed by the senior investigator. Of the total 266 patients in the study, 3 had resulted negative via SPT by their referring provider and opted not to repeat the procedure at our clinic. These patients instead received IDT and the negative SPT results were not included in analysis.

Intradermal Testing

All 266 patients were tested via IDT with both immediate (15 minute) and delayed (2 and 4 days) readings. Both HDM (10,000 AU/mL; ALK) and mold (1:20 w/v; ALK) species were tested as 1:10 dilution in addition to pet dander (cat: 10,000 AU, dog: 1:100 w/v; ALK) when relevant. Allergen formulations and histamine controls were injected intradermally into the anterior surface of the forearm. Reactions were considered positive only with a papule or wheal >5 mm in diameter.

Delayed reading was considered positive if the IDT injection site demonstrated at least 5 mm induration 48 hours after inoculation. For safety reasons, we do not perform intradermal testing in patients highly positive to aeroallergens via SPT and instead substitute the APT for evaluation of delayed reactions. In this study, no SPT reaction was severe enough to require substitution.

Patch Testing

Given the limited literature regarding delayed reading of the IDT in this context, 44 patients were tested via APT (prick test solutions diluted in 50% petrolatum, 50% glycerin vehicle (v/v); ALK) when clinically indicated either congruently with IDT or alone on occasions when IDT was unavailable to assess delayed-type sensitivity to aeroallergens. Patches were applied using Finn Chambers (Aqua SmartPractice or Chemotechnique/Dormer IQ Ultra) on healthy normal skin on the patient's back and removed after 2 days. Reactions were read at 2 and 4 days and evaluated using the European Task Force on Atopic Dermatitis guidelines.²⁴

In total, 200 patients were concurrently tested according to their clinical history and necessity via the comprehensive patch testing series of the Dermato-Allergy Clinic at M Health Fairview Clinics and Surgery Center, Minneapolis, MN. These series include the standardized allergens of the North American Standard series (80 allergens; NACDG 2020) and additional series and allergens according to the clinical needs with standardized allergens from SmartPractice or Chemotechnique/Dormer.²⁵

Statistical Analysis

Participant demographic and clinical characteristics were summarized using mean (SD) for continuous and count (%) for categorical variables. One-sample tests of proportion and respective CI (95%) were employed to help determine whether our patients were experiencing clinically significant reactions. We hypothesized that at least 75% of delayed-type reactions identified via IDT or APT in this study were positive to either dust mites or mold species only and were, therefore, not the result of over-reactivity in sensitive individuals. Owing to the limited literature concerning delayed-type reactions to aeroallergens observed via IDT, this value was selected based on clinical judgment. Analysis was completed in R version 4.1.1 (R Core Team, Vienna, Austria).

RESULTS

Prevalence of Immediate and Delayed-Type Reactions

In total, 266 patients (age: range [18–91], median 46.5; gender: 214 female, 52 male) met the criteria for this study. Of them, 195 (73.3%) demonstrated an immediate-type reaction to HDM, mold, or pet dander (Table 1), corroborated by their clinical symptoms. In total, 118 (44%) had a delayed-type reaction (Table 2). Notably, 75 (28.2%) patients experienced both immediate and delayed-type reactions and 43 (16.2%) reacted delayed-type only. The frequency of concurrent immediate and delayed-type reactions to different aeroallergens observed via IDT is shown in Figure 1.

Table 1.

Immediate-Type Reactions

Allergen	Sensitization Total n = 266	Prevalence (%)	SPT	IDT*
Dermatophagoides farinae	117	44.0	24	94
Dermatophagoides pteronyssinus	120	45.1	33	88
Aspergillus fumigatus	50	18.8	2	48
Penicillium notatum	51	19.2	4	47
Pet dander (cat)	25	9.4	24	2
Pet dander (dog)	7	2.6	4	4
Seasonal molds	15	5.6	15	—
Weed pollens	16	6.0	16	—
Tree pollens	15	5.6	15	—
Grass pollens	13	4.9	13	—

Sensitization total refers to the immediate-type positivity in the sample per allergen. SPT ≥3 mm; IDT ≥5 mm at 15 minutes.

^* 81.2% of reactions were positive to either dust mites or mold species only (95% CI 76%–100%, P = 0.018).

IDT, intradermal skin test; SPT, skin prick test.

Table 2.

Delayed-Type Reactions

Allergen	Sensitization Total n = 266	Prevalence (%)	APT*	IDT†
Dermatophagoides farinae	76	28.6	3	73
Dermatophagoides pteronyssinus	84	31.6	6	78
Aspergillus fumigatus	28	10.5	0	28
Penicillium notatum	29	10.9	0	29
Pet dander (cat)	0	0	—	0
Pet dander (dog)	1	0.4	—	1

Sensitization total refers to the delayed-type positivity in the population per allergen. APT ≥ weak positive (+); IDT ≥5 mm at 48 hours.

^* 100% of reactions were positive to either dust mites or mold species only (95% CI 98%–100%, P < 0.001).

^† 94.8% of reactions were positive to either dust mites or mold species only (95% CI 92%–100%, P < 0.001).

APT, atopy patch test; IDT, intradermal skin test.

Figure 1.

Concomitant reactions observed via intradermal skin testing ordered by frequency of occurrence. Lines connected vertically represent a patient demonstrating sensitivity to multiple allergens, whereas the bar above them represents the frequency of that individual combination occurring in the study. Most frequent was delayed type to HDMs only (29 patients), followed by immediate type to HDMs only (25). *Indicates a delayed-type reaction pattern that is positive to either dust mites or mold species only. HDM, house dust mite.

Of the 266 patients, 190 presented with AD (71.4%), 197 with AR (74.1%), and 75 with allergic asthma (28.2%). The occurrence of delayed-type sensitivity among the observed atopic diseases AD (89 patients, 46.8%), AR (83, 42.3%), and allergic asthma (32, 42.7%) was evenly distributed.

Patch Testing

Of the 44 patients tested for delayed-type sensitivity via APT, 9 (20.5%) were positive and only for HDMs. Of the 266 patients tested via IDT, 73 (27.4%) had a delayed reaction to D. farinae and 78 (29.3%) to D. pteronyssinus (Table 2).

Sensitivity to common contact allergens identifiable by standard series patch testing was also observed in the atopic population. Of the 200 patch-tested patients, 42 (21.0%) demonstrated sensitivity to fragrances, 19 (45.2%) of whom also experienced delayed-type reactions to an aeroallergen. Among the 88 patients who demonstrated delayed-type sensitivity to aeroallergens and were also patch tested, 26 (29.5%) were sensitive to nickel sulfate and/or cobalt (Table 3). Comparatively, 24 (16.7%) of 144 patients with only immediate-type sensitivity to aeroallergens were sensitive to nickel sulfate and/or cobalt. Overall, 72 (81.8%) of the 88 patients did not react or had sensitivities determined to be nonclinically contributory to their presenting symptoms after patch testing.

Table 3.

Delayed-Type Reaction Prevalence in Patch Testing Results

Allergen	Positivity		Delayed-Type Prevalence (%)
	All Patients n = 200	Delayed Type n = 88
Nickel sulfate	50	25	28.4
Cobalt	20	10	11.4
Fragrance mix	42	19	21.6
Formaldehyde*	21	13	14.8
Other preservatives	41	23	26.1
Adhesives	4	3	3.4
Other†	6	2	2.3
No reaction	73	23	26.1

Delayed-type refers to patients exhibiting delayed-type sensitization to common house dust mite or mold species who were also patch tested. Prevalence refers to the proportion of positive delayed-type reactions for each allergen among the delayed-type sample.

^* Includes formaldehyde releasers (eg, quaternium-15).

^† Includes lanolin, polyethylene glycol, and rubbers.

AD Lesion Locations and Delayed-Type Reactions

As traditional airborne contact allergies can be associated with reactions on frequently unclothed and air-exposed areas of skin, the predominant distribution of AD lesions in patients demonstrating delayed-type sensitivity to our aeroallergens of interest was noted. “Air-exposed” areas included hands/wrists, face/scalp, neck, ears, eyelids, lips, and diffuse lesion distributions. “Nonair exposed” included the back, trunk, legs/feet, arms, shoulders, and extensor/flexor surfaces. Overall, 176 patients had documented AD lesion locations, 148 (84.1%) of whom were in air-exposed areas. Among the 176 patients, there were 150 positive delayed-type reactions to individual aeroallergens. In total, 128 (85.3%) were associated with lesions predominantly in air-exposed areas.

Of the 176 total patients, 77 (43.8%) with lesions in an air-exposed distribution demonstrated only immediate-type reactions or nonreactivity to aeroallergens. By patient history, use of aerosol-based products (sunscreens, perfumes, hair products) was not present among those also reacting to the relevant allergens in comprehensive patch testing.

Dupilumab Treatment Efficacy

Dupilumab treatment was followed in 32 patients in our study, 16 of whom had a positive delayed-type reaction to HDMs or molds. Of these patients, 22 (68.8%) showed a complete or majority improvement in their atopic symptoms after at least 6 months of treatment with dupilumab. Among the patients showing a complete/majority improvement, 13 (59.1%) demonstrated only delayed-type, 8 (36.4%) demonstrated only immediate-type, and 1 (4.5%) demonstrated both immediate- and delayed-type sensitivities to aeroallergens.

DISCUSSION

In this study, we evaluated the prevalence and clinical significance of immediate- and delayed-type reactions to perennial aeroallergens in atopic patients. We consider these to be true hypersensitivity reactions rather than the result of overly sensitive individuals reacting to multiple unrelated allergens as at least 75% were positive to either dust mites or mold species only (P < 0.05, Fig. 1). In addition, of all 266 patients tested with IDT, 73 patients (27.4%) had a delayed-type reaction to D. farinae, 78 patients (29.3%) to D. pteronyssinus, 28 patients (10.5%) to A. fumigatus, and 29 patients (10.9%) to P. notatum (Table 2). This means that each patient had their own reaction profile.

Furthermore, 85.3% of delayed-type reactions to individual aeroallergens were associated with lesions predominantly in air-exposed areas, in contrast to 43.8% of those who were nonreactive or had only immediate-type reactions. Given the contrast between air-exposed lesion prevalence in those demonstrating delayed-type sensitivity to aeroallergens and those that do not, it may be suggested that the delayed sensitivity is contributing to this distribution as an airborne contact dermatitis.

However, as many patients also experience sensitivities to preservatives and fragrances in products such as liquid run-off shampoos, the contribution of delayed-type sensitivities to aeroallergens to airborne distribution cannot be quantified. Interestingly, although it is estimated that the prevalence of nickel allergy in the general population is 8.6%,²⁶ we found that 28.4% (25/88) of patch-tested patients demonstrating delayed-type sensitivity and 17.4% (25/144) with only immediate-type sensitivity were also sensitive to nickel in our sample (Table 3). Although more extensive studies are needed to clarify the link between nickel sensitivity and atopy, a higher prevalence of these reactions has been noted among atopic patients.²⁷

In all, these associations suggest that the delayed-type reactions to aeroallergens in our study were observed in atopic individuals and could possibly be treated in part as an airborne contact dermatitis. However, although many of these patients see an improvement in their symptoms with allergen avoidance, full elimination with current techniques is difficult, particularly with HDMs. We realistically expect a reduction in symptoms but never full disappearance with avoidance. As such, it can be concluded that these sensitivities are a component but not necessarily the sole cause of atopy in these patients. Further studies are needed to correlate these results to nonatopic populations.

Notably, 43 (16.2%) patients in our study revealed only delayed-type sensitivity. One demonstrated a 1.5 cm vesicular to bullous crescendo-like delayed-type reaction to both mold species, with all other tests including delayed reading of the IDT to HDMs negative (Fig. 2). This case emphasizes that delayed-type reactions can be quite severe and may go undiagnosed if no delayed reading is performed.

Figure 2.

The 1.5 cm vesicular to bullous crescendo-like delayed-type reaction to Penicillium notatum and Aspergillus fumigatus at 2 (A) and 4 (B) days after intradermal skin test injection in a patient with no other immediate- or delayed-type sensitivities.

Although immediate reading of the IDT has been shown to be an accurate and sensitive test in the setting of some aeroallergens²⁸ and our study suggests that delayed-type reactions to aeroallergens provoked by IDT are true reactions with clinical significance rather than over-reactivity in sensitive individuals (P < 0.001), the overall statistical reliability of this test with regard to delayed-type sensitivity needs to be elucidated. Even so, our study demonstrated similar findings to those of previous observers.

Guler et al found that 30% of atopic patients in their study had both positive delayed and immediate-type reactions to HDM identified via APT and prick testing, respectively.¹⁴ This is similar to the 28.2% (75 patients) with both immediate- and delayed-type reactions present in our atopic population. In addition, Darsow et al found that 17% of 314 patients with atopic eczema in their study demonstrated only delayed-type sensitivity to HDMs via APT, with 39% positive to D. pteronyssinus overall.⁵ This corresponds to our findings of 16.2% (43) and 31.6% (84), respectively (Table 2).

However, it is important to stress that IDT cannot be performed if the SPT was highly reactive for safety concerns. In those patients, APTs are the appropriate alternative to delayed reading of the IDT to evaluate for type IV hypersensitivity. As there is no current standardization for APTs, it must also be acknowledged that comparing the positivity for allergens with this test between studies can be complicated as consistent dilution methods and vehicles (in this case, dilution in petrolatum with a glycerin vehicle) are not used. In a few cases, delayed-type reactions to SPTs were also observed.

The patient presented in our recent letter²⁰ was successfully treated with dupilumab, the first biologic medication approved for the treatment of AD that selectively inhibits interleukin (IL)-4 and IL-13, cytokines that have been linked to the type-2 helper T cell (Th2)-mediated immune response in the disease.^29,30 The successful use of dupilumab in atopic patients, including among those demonstrating delayed-type hypersensitivity to aeroallergens, was noted in our study. The effectiveness of dupilumab in these patients suggests that their sensitivity to aeroallergens plays a role in their extrinsic AD.

Given how difficult full avoidance of these perennial aeroallergens can be, treatments targeting the atopic activation pathways can be of significant benefit. However, the retrospective nature and lack of controls in this study limit the generalizability of these findings. Further investigation into the efficacy of anti-IL-4 and/or IL-13 biologics in patients with delayed-type sensitivity to aeroallergens is needed.

In patients with extrinsic atopic disease who are sensitive to aeroallergens, evaluating the reaction type is particularly important with the therapeutic choices currently available. For example, it can be theorized that those with mostly immediate-type and IgE-mediated reaction patterns may be treated with antihistamines, specific immunotherapy, or anti-IgE biologics (eg, omalizumab), and patients with a predominantly delayed-type reaction pattern could benefit most from topical steroids and anti-IL-4 and/or IL-13 biologics (eg, dupilumab, tralokinumab), anti-IL-5 biologics (eg, mepolizumab), or janus kinase inhibitors.

For patients who react to no relevant allergens on history-driven comprehensive patch testing but demonstrate a delayed-type reaction to a perennial aeroallergen, the aforementioned treatments that target atopic activation pathways are indicated as the allergens are so difficult to avoid. However, further controlled studies are required to establish effectiveness of the postulated therapies in this regard, particularly with distinction between immediate- and delayed-type sensitivities.

Although most patients with delayed-type sensitivity and clinically significant AD in our population were treated with topical steroids and allergen avoidance with benefit, treatment efficacy in those not treated with dupilumab was not followed and is a limitation of this study.

CONCLUSION

The results of this study indicate that delayed-type reactions to aeroallergens are prevalent and clinically significant as a component of extrinsic atopic diseases, particularly AD. These delayed-type reactions can be quite severe and may go undiagnosed if no delayed reading is performed. We recommend the use of IDT to aeroallergens when available in atopic patients to find additional aggravating causes, enforce the diagnosis of atopy, and ultimately to guide clinical and therapeutic decision making.