Abstract
Background
Local allergic rhinitis (LAR) is characterized by chronic rhinitis with localized nasal allergic response to allergens in the absence of systemic immunoglobulin (IgE)-mediated disease assessed by skin prick test (SPT) or specific serum IgE level. Nasal provocation test (NPT) is the gold standard for the diagnosis of LAR. Nasal eosinophilia is a known inflammatory marker in allergic rhinitis. We hypothesized that nasal eosinophilia can be used as a screening tool for LAR.
Objective
To determine the applicability of nasal eosinophilia as a diagnostic tool for LAR.
Methods
Forty-eight perennial nonallergic rhinitis (NAR) patients with SPT negative were recruited. Nasal cytology analysis was performed. NPTs with 3 allergens (mixed mites, mixed cockroaches, and Bermuda grass) were performed to diagnose LAR. Nasal symptoms combined with nasal patency were used to determine the results of NPT. The sensitivity, specificity, positive predictive value, and negative predictive value of nasal eosinophilia as a diagnostic tool of LAR were calculated.
Results
LAR was diagnosed in 41.6% of the NAR patients. Nasal eosinophilia was found in 58% of the NAR patients. The sensitivity of nasal eosinophilia for diagnosing LAR was 80%, the specificity was 57.14%, the positive predictive value was 57.14%, and the negative predictive value was 80%. The most common allergen of LAR was mixed mites.
Conclusion
Nasal eosinophilia on nasal cytology was a good screening tool for diagnosing LAR because of its high sensitivity and simplicity of the procedure. Unfortunately, the low specificity of nasal eosinophilia makes the NPT necessary for confirmation of LAR.
Keywords
Introduction
Local allergic rhinitis (LAR) is characterized by chronic rhinitis with localized nasal allergic response to allergens in the absence of systemic immunoglobulin (IgE)-mediated disease assessed by skin prick test (SPT) or specific serum IgE level. 1 Nasal provocation test (NPT) is the gold standard for the diagnosis of LAR. 2 NPT activates mast cells and eosinophils which leads to the fast release of nasal allergen-specific IgE, tryptase, and eosinophil cationic protein in LAR. 3 One study showed that 54% of perennial nonallergic rhinitis (NAR) patients had positive nasal response to Dermatophagoides pteronyssinus and 22% of the patients with nasal response to D. pteronyssinus had nasal-specific IgE to D. pteronyssinus. 4 Another study of antigen-specific IgE on mucosal brush biopsy of the inferior turbinate detected at least 1 antigen-specific IgE in all NAR patients and the IgE to cockroach was present in 90% of the NAR patients. 5 Diagnosis of LAR in NAR patients is important because the LAR and NAR have different clinical phenotypes and treatments. 6 Unfortunately, the NPT and measurement of nasal-specific IgE are not readily available to most clinicians. Nasal cytology is simple and inexpensive. It has been used for differentiating between inflammatory and noninflammatory rhinopathies. 7
The presence of nasal eosinophils is an important biomarker in allergic rhinitis (AR). 8 The number of nasal eosinophils correlates well with the severity of nasal symptoms and signs in AR. 9 NAR with eosinophilia syndrome (NARES) is characterized by chronic rhinitis in the absence of systemic allergy with nasal cytology analysis of greater than 20% eosinophils. 10 Therefore, nasal eosinophilia was defined if the nasal cytology showed greater than 20% eosinophils in this study. We hypothesized that the presence of nasal eosinophilia could be used as a screening tool for LAR. This study aimed to evaluate the possibility of nasal eosinophilia as a screening tool for LAR in chronic rhinitis patients with negative SPT.
Methods
The NAR patients at the outpatient clinic of the Otorhinolaryngology Department were recruited. Inclusion criteria were patients aged 20 to 70 years who had a clinical history of perennial rhinitis for at least 1 year with a negative SPT and without other known etiologies. SPT was performed with common local aeroallergens including Orris root, Cotton Linters, Acacia spp., Kapok, Bermuda grass, Johnson grass, Careless weed, American cockroach, German cockroach, Cat hair, Dog epithelia, mixed feathers, D. pteronyssinus, Dermatophagoides farinae, and Mold mix (ALK-Abelló, Port Washington, NY, USA). Histamine (1 mg/mL) and glycerinated saline were used as a positive and a negative control, respectively. A positive SPT was defined as a wheal with diameter of ≥3 mm. Patients were excluded if they had acute upper or lower respiratory tract infection, history of nasal or sinus surgery, nasal polyps, atrophic rhinitis, moderate-to-severe nasal septum deviation, asthma, severe cardiopulmonary disease, pregnancy, lactating woman, or smoking. Patients with oral systemic corticosteroid use within 4 weeks, nasal steroid use within 2 weeks, or antihistamine or decongestant use within 1 week were excluded. The Ethics Committee approved the study (IRB 103/58). The study was registered in Thai Clinical Trial Registry (TCTR 20170219001). A written informed consent was obtained from every patient.
On the first visit, nasal cytology samples were collected by scraping the medial surface of the middle part of the inferior turbinate with a Rhinoprobe (Arlington Scientific, Springville, UT, USA), 2 or 3 times per each nostril. The sample of each nostril was transferred onto a labeled glass slide (right or left nostril). The slides were air-dried and stained with Wright-Giemsa method. Nasal cytology analysis was performed using an optical 1000× magnification binocular microscopy with oil immersion. The eosinophil, neutrophil, lymphocyte, and total inflammatory cells of each slide were counted and reported. Nasal eosinophilia was defined if the nasal eosinophils were >20% of the total cells presented on a slide. Nonnasal eosinophilia was defined if the nasal eosinophils were less than 20% of the total cells on both slides.
On the second visit, the baseline subjective and objective parameters of each patient were recorded. The subjective parameter was the visual analog score (VAS) of nasal and eye symptoms. Each symptom of rhinorrhea including, sneezing, nasal blockage, itchy nose, and itchy eye was recorded using a 6-point scale: 0 = no symptoms, 1 = very mild symptoms, 2 = mild symptoms, 3 = moderate symptoms, 4 = severe symptoms, and 5 = very severe symptoms. The total symptom score was calculated by the sum of all symptoms (score range 0–25). The object parameter was the nasal patency which was assessed by measuring minimal cross-sectional area (MCA) using an acoustic rhinometry (Rhinoscan SRE2000; Rhinometrics, Lynge, Denmark) according to the guidelines of the Standardization Committee on Acoustic Rhinometry. 11 The MCA of each nasal cavity was measured separately. The average MCA of both sides was calculated and analyzed.
The baseline nasal patency was recorded after the patient was at room temperature for at least 30 minutes. A bilateral nasal provocation was performed using a nasal spray (metered-dose bottle) delivering a fixed volume of 0.125 mL/puff. A nasal challenge test with diluents (buffered saline) was performed to exclude nasal hyperreactivity. The patient was evaluated at 5 minutes after being challenged with diluents. A positive NPT response was defined as an increase ≥3 of total VAS and a decrease ≥30% of the average of bilateral MCA from baseline. If the NPT response to diluents was positive, the patient was excluded from the study.
The NPT was performed with mixed mites allergen (D. pteronyssinus and D. farinae) at 2 concentrations (1000 allergy unit [AU]/mL and 5000 AU/mL). First, the patient was challenged with the lower concentration (1000 AU/mL). The NPT response was evaluated at 5, 15, and 30 minutes after each challenge. If the NPT was negative, the higher concentration of the allergen was challenged following the same protocol as previously described. 12 On the third and the fourth visits, similar NPTs were performed using mixed cockroaches (American and German cockroaches) at 2 concentrations (1:5000 w/v and 1:1000 w/v) and Bermuda grass at 2 concentrations (1000 bioequivalent allergy unit [BAU]/mL and 5000 BAU/mL), respectively. The interval between each visit was at least 7 days to the maximum of 14 days.
Statistical Analysis
Statistical analysis was performed using a computer-based program and SPSS statistical software (version 17.0. SPSS Inc. Chicago, IL, USA). Continuous data were recorded as the mean and standard deviation (SD) or median and range. Categorical data were recorded as percentages. The sensitivity, specificity, predictive positive value, negative predictive value, and likelihood ratio analyses of nasal eosinophilia to LAR were performed by using the 2-by-2 table. The prevalence of LAR was also calculated.
Results
Fifty-six patients who had negative SPT were enrolled for nasal cytology analysis. Six patients had undetermined results and 1 patient became pregnant before the second visit. On the second visit, the baseline VAS and average MCA of 49 patients were recorded. NPT with diluents was performed and none had nasal hyperreactivity. All 49 patients were challenged with dust mites. One patient developed severe dry cough at night time after the second visit. On the third visit, she had rhonchi on physical examination which responded well with oral prednisolone plus montelukast. She was diagnosed with bronchial hyperresponsiveness from NPT and therefore was excluded from the study. A total of 48 patients completed the study. The demographic data and patient characteristics are shown in Table 1. The mean age ± SD was 42 ± 13 years. Seventy five percent of patients were female, and most of them had moderate-to-severe chronic rhinitis. Nasal eosinophilia was found in 28 patients (58.3%). Twenty patients (41.6%) had positive NPTs. The nasal cytology analysis and NPT results are shown in Table 2. Nasal obstruction was the main symptom of both LAR (13/20 patients) and NAR (15/28 patients). Itchy nose and/or itchy eyes were the least concomitant symptom of both LAR (6/20 patients) and NAR (5/28 patients). There was no significant difference in the baseline nasal or eye symptoms and nasal patency between NAR and LAR. The baseline VAS of nasal and eye symptoms and MCA on the second visit in LAR and NAR are shown in Table 3. The sensitivity of nasal eosinophilia for diagnosing LAR was 80%, the specificity was 57.14%, the positive predictive value was 57.14%, and the negative predictive value was 80%. Table 4 shows the number of patients who had nasal eosinophilia and nonnasal eosinophilia in LAR and NAR
The Demographic Data and the Patients’ Characteristics.
The Nasal Cytology Analysis and NPT Results (n = 48).
Abbreviation: NPT, nasal provocation test.
The Baseline VAS of Nasal and Eye Symptoms and MCA on the Second Visit in LAR and NAR.
Abbreviations: LAR, local allergic rhinitis; MCA, minimal cross-sectional area; NAR, nonallergic rhinitis; SD, standard deviation; VAS, visual analog score.
The 2-by-2 Table Showing the Number of Patients Who Had Nasal Eosinophilia, Nonnasal Eosinophilia, LAR, and NAR.
Abbreviations: LAR, local allergic rhinitis; NAR, nonallergic rhinitis; NPT, nasal provocation test.
Discussion
Diagnosis of LAR in idiopathic NAR patients is important because the natural history and plan of treatment of these 2 conditions are different. Campo et al. compared the clinical phenotypes between NAR and LAR and discovered that nasal obstruction and rhinorrhea were more common in NAR patients while sneezing was more common in LAR. 6 However, there was no significant difference in nasal symptoms between NAR and LAR in this study, and nasal obstruction was the most common symptom of both conditions. Rondon et al. proposed an algorithm for diagnostic approach of chronic rhinitis which included (1) SPT, (2) serum-specific IgE level if the SPT result was negative, and (3) NPT with allergens should be performed for diagnosing LAR, if the serum-specific IgE was negative. 13 Although NPT is the gold standard diagnostic tool for AR and LAR, it requires clinician’s expertise in performing the test. To date, there is no universal NPT protocol on both concentration of the allergens and the challenging methodology. In addition, there are no standardized measurements to determine whether the result is positive or negative. In this study, the NPT was performed using nasal spray to deliver allergens with 2 different concentrations. The NPT technique and criteria of positive NPT response were based on a previous NPT study with house-dust mite, which provided a high specificity (91.22%). The study used the ≥3 increase of total VAS as the optimum cutoff value. 12 In addition, the ≥30% decrease of average MCA from baseline based on a study conducted by Kim et al. 14 was added to the criteria in this study to increase the specificity of the test.
The pathogenesis of AR initiates by allergen exposure. The allergen-presenting cells, T cells and B cells, are sensitized causing the production of specific IgE antibodies. On reexposure, the cross-linking of specific IgE on mast cells with allergen results in the release of histamine and leukotrienes that causes immediate nasal symptoms such as itching and sneezing. Late-phase allergy occurs many hours later. The infiltration of eosinophils, basophils, and lymphocytes on nasal mucosa results in nasal congestion. 15 Eosinophils also act as antigen-presenting cells that stimulate the T helper response against allergens. 16 Nasal eosinophilia has been used as a marker of AR.17–19 However, there was no study demonstrating the nasal cytologic findings in LAR.
Compared to the NPT, the nasal cytology test is simpler and less expensive and led to a hypothesis that the nasal cytology analysis could be used as a screening tool for diagnosing LAR. Ellis and Keith proposed an algorithm using nasal smear for eosinophils before NPT. The study divided the patients into 3 groups: LAR, NAR, and NARES without any evidence of the sensitivity and specificity of the test. 10 The findings of this study showed that most of the LAR patients (16/20 patients, 80%) had nasal eosinophilia. Because of the high sensitivity of nasal eosinophilia in this study (80%) and the simplicity of the test, the nasal eosinophilia could be a good screening tool for LAR diagnosis. The authors propose an algorithm (Figure 1) as follow: nasal cytology analysis should be performed in the NAR patients before NPT to reduce the cost of diagnostic tools. NPT should be considered in the patients with positive nasal eosinophilia to identify LAR patients due to its low specificity. In the nonnasal eosinophilia patients, empirical treatment for NAR seems to be reasonable due to its high negative predictive value (80%). NPT using more than 1 allergen could diagnose more LAR patients than only 1 allergen. We do NPT all year round because of perennial type of major allergens (dust mites, cockroaches, and Bermuda grass) and challenges to these aeroallergens may change depending on geographic region.
The proposed diagnostic algorithm.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Ratchadapisek Somphot Fund (RA58/078), Chulalongkorn University.