Introduction

According to the Guidelines for the Clinical Management of Allergic Conjunctival Disease (ACD) of the Japanese Ophthalmology Society (“the guidelines”) [1], allergic conjunctivitis (AC) is defined as non-proliferative conjunctivitis associated with a type I allergy mediated by IgE antibodies. The guidelines indicate that a definitive diagnosis of ACD requires positive results for serum antigen-specific IgE (s-sIgE) antibodies and eosinophils in a conjunctival smear. A commercial kit that measures total IgE antibody levels in lacrimal fluid has recently become available; it can be used to confirm local production of IgE antibodies in the eye, but this test cannot be used for the definitive diagnosis of ACD.

Allergic diseases have various phenotypes [2]. In 1975, Huggins et al. [3] reported that allergic rhinitis (AR) has one entity, i.e., local allergic rhinitis (LAR). LAR is described as a phenotype of AR in the Global Atlas of Allergic Rhinitis and Chronic Rhinosinusitis of the European Academy of Allergy and Clinical Immunology [4]. In 1993, Leonardi et al. [5] suggested a probable entity of local allergic conjunctivitis (LAC), but the concept of LAC has not been widely studied in the ophthalmologic community.

We performed total tear IgE (t-tIgE) and specific serum IgE (s-sIgE) testing in patients clinically diagnosed using subjective symptoms (itchy, red, or watery eye) and clinical findings based on the guideline [1], and evaluated the results for the purpose of seeking the presence of LAC cases in these patients.

Moreover, we also evaluated cytological analysis of eosinophils in the conjunctival scraping specimens in patients who showed negative for both t-tIgE and s-sIgE to make clear the pathophysiology of LAC in this study.

Materials and methods

This study followed the ethical principles of the Declaration of Helsinki and was approved by the Ethics Committee of Clinical Research Network Fukuoka (No. 17-E01). The study was registered in the UMIN Clinical Trials Registry (UMIN Trial ID: UMIN000027651).

This study included 148 patients (33 men and 115 women; 5–88 years old; mean age, 65.3 years) who visited the Yamana Eye Clinic between December 2013 and March 2017, with a clinical diagnosis of AC on the basis of subjective and clinical symptoms (Table 1). Patients without t-tIgE and s-sIgE test results were excluded.

Table 1 Age groups of the subjects, n = 148
Full size table

The following tests were performed according to the guidelines [1]. T-tIgE levels were measured with the Allerwatch® Tear IgE (Wakamoto Pharmaceutical Co., Ltd., Tokyo, Japan) (AW) kit, using immunochromatography [6, 7]. S-sIgE antibody levels against eight allergens, included Dermatophagoides pteronyssinus, cockroach, cat dander, dog dander, Japanese cedar, orchard grass, ragweed, and mugwort, were measured with the ImmunoCAP® Rapid (Thermo Fisher Diagnostics K.K., Tokyo, Japan) (ICR) kit, using a fluorescent enzyme immunoassay [8, 9]. Antibody levels against 39 allergens (including the eight allergens measured with ICR) were assessed with the View Allergy 39® (Thermo Fisher Diagnostics K.K., Tokyo, Japan) (View39) kit, using a chemiluminescent enzyme immunoassay (Table 2). The ICR uses immunochromatography based on a sandwich method, with colloidal gold as the labeled substance; the View39 also uses immunochromatography based on a sandwich method, but with β-galactosidase as the labeled substance.

Table 2 Allergens included in the specific IgE View Allergy 39 test kit
Full size table

All patients underwent both t-tIgE and s-sIgE testing. Of the patients who tested positive for t-tIgE but had no detectable s-sIgE antibodies, more than half underwent serum total IgE (s-tIgE) testing.

Conjunctival cytology using spatula was carried out on the palpebral conjunctiva of patients with negative results in all tests; the specimens were examined for the presence of eosinophils using Hansel staining and an optical microscope (Olympus Corporation, Tokyo, Japan).

Results

A total of 83 patients (56.1%) tested positive for t-tIgE in the AW. In the ICR, 97 patients (65.5%) were positive for at least one of the eight allergens, whereas 51 (34.5%) were negative for all allergens. Of the 83 patients who tested positive for t-tIgE in the AW, 19 (22.9%) were negative for all eight allergens examined with the ICR. Of these, five had detectable s-sIgE antibodies in View39, whereas 14 (16.9%) had no detectable s-sIgE antibodies (Table 3). Of these 14 patients, nine underwent s-tIgE testing; all showed a level that was below the reference range (< 170 IU/mL; Table 4).

Table 3 The results of total tear IgE and specific serum IgE test
Full size table
Table 4 Clinical details of patients who were positive for total tear IgE and showed no detectable specific serum IgE antibodies
Full size table

Of 32 patients (21.6%) with negative results in both the AW and the ICR, four had detectable s-sIgE antibodies in View39 against allergens that were not examined in the ICR. In total, 28 patients (18.9%) were negative in all three tests (AW, ICR, and View39).

Of these 28 patients who were negative for t-tIgE and s-sIgE antibodies, 21 underwent conjunctival cytology of the palpebral conjunctiva to evaluate the presence of eosinophils. Eosinophils and eosinophilic granules were detected in four (19%) and two (9.5%) patients, respectively. Of the 28 patients, 19 underwent s-tIgE testing; of these, 17 had a normal value (< 170 IU/mL) and two had a level ≥ 170 IU/mL.

Discussion

The guidelines define ACD as “conjunctival inflammatory diseases associated with type I allergy accompanied by some subjective and objective symptoms” [1]. The presence of clinical symptoms, type I allergic diathesis, and a type I allergic reaction in the conjunctiva are required for a diagnosis of AC. Type I allergic diathesis can be confirmed via positive results for serum antigen-specific IgE antibody levels or a skin reaction with presumed antigens, an increase in serum total IgE antibody levels, the presence or the absence of a family history of allergic diseases, and comorbidities of other allergic diseases. The guidelines state that the presence of eosinophils in the conjunctiva must be shown for the definitive diagnosis of ACD.

Leonardi et al. [5] stated: “In patients with rhino conjunctivitis or rhinitis, and in 10 normal subjects, results of conjunctival provocation tests (CPT) and nasal-specific provocation test (NPT) were in 100% agreement.” Conversely, in patients with only conjunctivitis, little correlation was found between the results of CPT and NPT (K = 0.3). Tear-specific IgE was the only positive diagnostic sign of antigen sensitivity in 35% of VKC (vernal keratoconjunctivitis) patients and 30% of AC patients. It was suggested from this that the conjunctiva can be a uniquely sensitized target organ in allergic patients [5]. In the field of otolaryngology, the entity of LAR was first reported by Huggins et al. in 1975 [3] and subsequently by Rondón et al. in 2010 [10, 11] as well as by Campo et al. [12]. Rondón and colleagues [10, 11] reported that atopic AR was identified in 63%, non-allergic AR in 11%, and LAR in 26% of Spanish adult patients with rhinitis. LAR is characterized by symptoms of rhinitis (sneezing, watery rhinorrhea, and nasal congestion), eosinophilic infiltration of the nasal mucosa, and IgE antibodies localized in the nasal mucosa. However, LAR shows no elevation of serum IgE levels; it is likely to progress to atopic AR accompanied by asthma or other complications. After that Rondón and colleagues reported that a similar rate of development of AR with systemic atopy was detected in patients and controls (9.7% vs 7.8%, long-rank P = 0.623). In five patients, conversion to systemic atopy occurred > 10 years (3%). And then Rondón et al. [13] concluded as follow: LAR is a well-differentiated clinical entity with a low rate of development of systemic atopy, a natural evolution toward worsening and a risk factor for suffering asthma. Campo et al. [14] reviewed the diagnosis, classification and etiology, therapeutic options, and they concluded as follows: LAR rapidly evolves toward the clinical worsening and the association to asthma and conjunctivitis implying that an early diagnosis and the initiation of specific therapies are crucial for controlling the disease and potentially preventing its comorbidities.

Kato et al. [15] and Gelardi et al. [16] reported on the etiology of LAR. Kato investigated the pathophysiology of mice with allergic rhinitis that initially sensitized with ragweed pollen through the nasal route. He concluded that Th2 cells cause nasal symptoms due to the accumulation of Th2 cell in the nasal region on the pathology of LAR. Gelardi hypothesized that the nasal IgE production represents a form of spontaneous immune response.

In this study, among 83 patients positive for t-tIgE, 14 (16.9%) had no detectable s-sIgE antibodies; however, not all s-sIgE antibodies could be examined. Of these 14 patients, nine underwent s-tIgE testing; all showed a level that was below the reference range. Therefore, the prevalence of LAC was 16.9% in this study. Atopic bronchial asthma [17,18,19], atopic dermatitis (AD) [20, 21], and AR [22] have various phenotypes. LAC might be a candidate of novel phenotype in AC.

Several ophthalmologic studies have reported local cytokine dynamics in the eye, including the palpebral conjunctiva and lacrimal fluid [23,24,25,26,27,28]. However, few studies have been reported for the dissociation of serum and tear IgE antibody levels in patients with AC. Aghayan-Ugurluoglu and colleagues [29] reported that serum and tears of many of the pollen-allergic individuals with conjunctivitis exhibited specificity for the very same pollen allergens, but no allergen-specific IgE antibodies were detected in tears of non-atopic individuals. They also stated that IgA antibodies in sera and tears of patients with allergic conjunctivitis were mainly directed against non-allergenic moieties and showed specificities that were significantly different from those of IgE antibodies [29]. Whereas these results were interesting to consider the local allergic reactions in clinical situation of ACD, their report did not directly support our findings. Pathophysiological property of LAC has not been clarified, because the prevalence of LAC among AC is limited and several steps of diagnostic measures, serum, and tear antigen-specific IgE tests and conjunctival cytology test added to clinical evaluation, are required for the diagnosis of LAC. However, Leonardi et al. [30] reported that while normal subjects resulted negative for the presence of specific IgE both in serum and in tears, of the 10 VKC patients, six resulted positive to specific IgE in serum and/or tears, and in three of these six patients, specific IgE was found positive only in tears. Also, study population was different from our study; it was revealed that the dissociation of serum and tear IgE might be observed in some patients among ACD. Future study will be focused on immunological analysis of ocular surface in LAC to find out its allergological specificities compared with other ACD.

We also found cases who showed positive eosinophils in the conjunctival scrapings, but who had negative t-tIgE and s-sIgE tests among several clinically diagnosed AC patients. There seems no study focusing on these specific cases that showed only positive conjunctival eosinophils without antigen-specific IgE elevation in both serum and tears. Although we considered the possibility of the presence of non-IgE-type AC in these cases, pathological bases of this phenotype will also be clarified in the future, since non-IgE-type AC is considered rarer in AC cases.

Conclusion

AC has been considered as an ocular manifestation of type I allergy mediated by IgE antibodies. However, the results of the present study suggest the existence of LAC, as a novel phenotype of AC. We also found non-IgE-type AC as another unique clinical entity of AC in this study.