Prevalence of allergic and nonallergic rhinitis in a rural area of northern China based on sensitization to specific aeroallergens

Traditionally, rhinitis is divided into two types according to its etiology; allergic rhinitis (AR) and nonallergic rhinitis (NAR). Nowadays, most epidemiologic studies of AR are based on mostly self-reported prevalence of symptoms and often lack objective measures. Moreover, epidemiologic evidence of NAR is scant. Thus, we aimed to perform a population-based cross-sectional study in a rural community of northern China, to assess the prevalence of clinical AR and NAR, based on both subjective and objective measures.

As an inflammatory disease of nasal mucosa, rhinitis is often defined by its clinical manifestation, such as rhinorrhea, sneezing, nasal congestion and itching. Although sometimes considered as a trivial disease, because it is not associated with mortality, it can lead to a great financial burden and tremendously impair a patient’s quality of life; negatively impacting many aspects of life such as sleeping, working performance, emotion, socializing, etc.

According to the latest Allergic Rhinitis and its Impact on Asthma (ARIA) guidelines [1], classic symptoms of AR include sneezing, rhinorrhea, nasal congestion, and nasal pruritus, alongside with ocular symptoms such as redness and itching of eyes and lachrymation. Detectable immunoglobulin E (IgE) against relevant aeroallergens is also measurable, as the symptoms of AR are a result a hypersensitivity reaction caused by specific inhalant allergens, mediated by specific immunoglobulin E (IgE).

Globally, AR affects 10–40% adults and 2–25% children [2]. Evidence indicates that prevalence of AR is high and increasing throughout the world; with 23–30% in Europe [3], 12–30% in the United States [4], and 11.1–17.6% in China [5, 6]. According to several relevant epidemiologic studies, the prevalence of self-reported AR varies from 10 to 50% [6‐13]; likely because of the coexistence of AR and NAR and many studies have evaluated the prevalence of AR solely based on questionnaire surveys, telephone interviews and self-reported symptoms. Objective tests such as skin prick test (SPT) and serum specific IgE were often not included in the data, leading to a higher prevalence. Indeed, one recent study by Zhang and the colleagues [11] to investigate the prevalence of allergic rhinitis (AR) and its associated risk factors in over 4000 3–5 years old preschool children in Beijing has reported that while the self-reported prevalence of AR was 48%, the prevalence of clinical AR, diagnosed based on SPT results, was 14.9%.

NAR, on the other hand, is a group of heterogeneous diseases, characterized by clinical nasal symptoms with negative specific allergen tests [14]. It is based on exclusive diagnosis, being nonatopic with nasal symptoms in response to inhalant and irritant triggers in the absence of a specific cause [14]. Although the mechanisms underlying NAR remain unclear, it is generally thought that innervation of the nasal mucosa may be involved [14‐16]. In recent years, there has been renewed interest in NAR with the prevalence reported to be 7% in the United States [17]. However, there’s little consensus on the diagnostic criteria and lack of specific tests, and NAR is thus often misdiagnosed as AR.

Few epidemiologic studies have been reported on NAR, and most of the studies have concentrated solely on the self-reported prevalence. The aim of the present study was therefore to assess the prevalence of clinical AR and NAR, using both subjective and objective tests, and the associated risk factors in Chinese subjects from a rural community in Beijing, China.

China, a country of diverse regions with a huge population of 1.3 billion, provides a wide-ranging patient base for investigating the epidemiology of rhinitis. Despite, a progressively increasing prevalence of AR over the last few decades [5] and availability of this large patient base, there is a marked paucity in epidemiological data; particularly the prevalence of AR and NAR in rural Chinese populations. Moreover, most of the available studies have reported their findings for prevalence of AR purely on the basis of the presence of self-reported nasal symptoms; with very few studies taking into account the importance of using objective tests for providing more accurate estimates, because of the cost and the low compliance of the subjects. In this respect Wang and colleagues [23] have reported the prevalence of AR in the northern China to be 9.2% when both the nasal symptoms and the result of sIgE were considered. Another study by Zhang and colleagues [11] combining a typical medical history of AR symptoms and diagnostic skin prick tests (SPT) reported that the prevalence of self-reported AR in Chinese children was 3.26 greater than clinically diagnosed AR. Indeed, one early nationwide study using telephone interviews for assessing prevalence of AR across China demonstrated that the prevalence of self-reported AR varied from 8.7% in Beijing to 24.1% in Urumqi (southwestern of China) [6]. It is possible that the wide variation in prevalence of self-reported AR varied might lie in the difficulty of understanding the questionnaires and the coexistence of NAR, which was not specifically addressed. Furthermore, it is possible that the variability may additionally result from differences in both training and comprehension of the disease by investigators in different centers, as adopting face-to-face interviews has shown the prevalence of AR to vary from 6.24% [13] to 32.5% [7]. In this respect, no comparative epidemiological data have been reported for prevalence of NAR in China. To our knowledge the present study is the first to provide such epidemiological data on both clinically diagnosed AR and NAR in a rural area of northern China, using a combination of self-reported AR and objective assessment of sensitization to specific aeroallergens by SPT.

Our study has demonstrated that the prevalence of clinical AR was 15.41%, which is similar to an AR prevalence of 16.67% found in a population-based survey of AR in Quebec in 2008 [24]. Moreover, the prevalence of NAR was 31.37% based on typical nasal symptoms as well as negative SPTs results. Similarly, age- and gender-standardized prevalence of AR was 16.78% and of NAR was 24.60%. Studies from the US and Europe have reported the prevalence of NAR to range from 7 to 19% in adults [17, 25, 26]. A study by Rondón and colleagues [27], involving a cohort of 428 randomly selected Spanish patients attending an allergy service, has further reported that local allergic rhinitis (LAR), which has a similar clinical profile to AR in nonatopic patients, is a distinct and prevalent entity, as indicated by diagnosis of 63.1% patients with AR, 11.2% with NAR and 25.7% with LAR. Thus, it is possible that in the absence of specific nasal provocation tests some subjects with local allergic rhinitis (LAR) may be misdiagnosed as NAR patients.

While two studies have previously reported the prevalence of self-reported AR in the rural areas of northern China to be between 6.2 and 10.8% [7, 23], another more recent multicenter study has indicated that the prevalence of self-reported AR was significantly increased over a 6-year period from 11.1% in 2005 to 17.6% in 2011 [9]. Indeed, we have previously have previously compared the prevalence of self-reported and skin prick test-confirmable AR among adults in urban and rural areas of China, and demonstrated that the prevalence of confirmable AR was similar between rural and urban areas in China, although there was a higher prevalence of self-reported AR in rural areas [7]. In the current study prevalence of epidemiologic self-reported AR was found to be 46.77% compared with 15.41% (16.78% after standardization for age and gender) clinical AR. These findings are in accordance with the findings of Zhang and colleagues [11], who also demonstrated prevalence of epidemiologic and adjusted clinical AR in preschool children in Beijing to be 48% and 14.9%, respectively. It is possible that the relatively high prevalence of self-reported AR in the present study compared to other studies in China [7, 9, 28] may partly be related to the socio-economic background of many of the participants, who were villagers with very low income and poor educational background, and found it difficult to fully understand the questionnaires, even under supervision of the investigators. Nevertheless, the present study indicated that although both clinical AR and NAR patients had severe nasal symptoms, which exerted negative effects on their QoL, the severity of the nasal symptoms in AR patients was significantly greater than in NAR patients.

Sensitization patterns for AR have been shown to vary from place to place, and are influenced by climate change [29, 30], gender [31], age [32, 33], etc. Race may also play a role because studies in Chinese [8, 34] and Korean [33] AR patients have shown dust mites to be the most prevalent sensitizing allergens, whereas studies in Europeans have pollen as the prevalent sensitizing allergens [3]. In the present study, Blattella germanica and the two house dust mites (Der p and Der f) were found to be the most prevalent allergens responsible for clinical AR, which is consistent with the findings of Lou and colleagues [8]. Indeed, these findings are also in agreement with the findings of a previous study, in which we demonstrated that Blattella germanica, Der p, Der f, as well as Altenaria tenuis were the most common sensitizing allergens in a cohort of over 10,000 AR out-patients [35]; and therefore suggest that a combination of standardized questionnaires and specific allergen tests may provide accurate and reliable estimates of prevalence of AR (and NAR) in population-based surveys.

Whilst Blattella germanica has been suggested to be an important component in the screening panels for aeroallergens in self-reported AR in most regions of China [8], there is limited evidence on the direct relationship between the sensitization of Blattella germanica and clinical manifestation of AR. Furthermore, although it is possible that the high sensitization rates to Blattella germanica noted in the present study may possibly be a result of cross reactivity with IgE for Der p or Der f; a study by Sun and colleagues [36] investigating the prevalence of sensitivity to cockroach allergens and IgE cross-reactivity between cockroach and house dust mite allergens in Chinese patients with allergic rhinitis and asthma seems to suggest that this is unlikely to be the case. These authors demonstrated that while there was a relatively high prevalence of cockroach sensitization in mainland China, the levels of cockroach SPT reactions were relatively low. Furthermore, although a large majority of cockroach sensitized patients were also SPT positive to Der p, an IgE cross-inhibition study showed that only a small number of patients appeared to have Blattella germanica and/or Der p as primary sensitizing source, which could false cause false positive SPT reactions against cockroach. However, the finding from the present study that sensitization rates for Blattella germanica in self-reported AR and NAR subjects were comparable (16.6% vs 14.9%) suggests that clinical manifestation of disease, especially the symptoms, should be compared with the specific allergen tests, rather than base a diagnosis of AR purely on tests in the clinical practice.

In this study, we found that family history of AR, and comorbid asthma, CRS, or atopic dermatitis, were all associated with increased risk of both AR and NAR. Although, it is somewhat surprising that a comorbid atopic disease, such asthma or atopic dermatitis, may confer a risk for development of a noninfectious, nonallergic condition such as NAR, this finding is nevertheless in accordance with the findings of Håkansson and colleagues [37] in adults and Chawes and colleagues [38] in children. Although the precise mechanisms underlying this association are not clear, it is possible that the “united airway” hypothesis [37, 39] or presence of a link between upper and lower airways beyond allergy associated inflammation [38] may, at least in part, explain this association. It is possible also that NAR may induced by to exposure to nonspecific triggers leading to nasal hyperreactivity [17], possibly in addition to or following development of AR and asthma. However, this would need to be confirmed in large longitudinal studies of patients and families with histories of allergic airways disease and NAR.

However, the findings of the present study are somewhat limited and thus need to be confirmed in further studies in consideration of these limitations, as well as in centers in other regions across China. In particular, in the present study NAR was differentiated AR, based only on the presence of typical rhinitis symptoms in the absence of negative SPT results, without data of IgE. In this respect, it is likely that the observed prevalence of NAR was higher than the actual prevalence because under the defined criteria for NAR in the present study patients with LAR would have been undistinguishable. Indeed, a systemic review of studies on AR or NAR patients subjected to diagnostic local nasal provocation has recently demonstrated that 26.5% of patients previously considered non-allergic demonstrate local allergen reactivity [40]. Moreover, the latest position paper of the European Academy of Allergy and Clinical Immunology on NAR [17] has indicated that NAR patients should be distinguished from AR patients with an allergic reaction confined to the nasal mucosa (i.e. LAR patients), probably by detection of nasal specific-IgE and/or positive nasal allergen provocation test (NAPT), although the latter is time consuming. Also, we have adopted the SI index for reporting the positive skin prick test size in this study; whereas a positive SPT is commonly reported as at least 3 mm greater than the negative control. Although it is possible that the low rate of positive skin prick tests noted in the present study may be due to the use of SI index, it is likely, however, that this finding is accurate for this study cohort; particularly because it included nearly half of the population of the region investigated and most of the subjects were healthy individuals.

In summary, the confirmed standardized prevalence of AR and NAR were 16.78% and 24.60%, respectively. Combination of standardized questionnaires and specific allergen tests may provide more accurate estimates of prevalence of AR and NAR and associated risk factors.