Introduction

Allergic rhinitis (AR), which is one of the more prevalent respiratory diseases, can cause sneezing, runny nose, postnasal drip, and nasal congestion concomitantly with eye symptoms such as redness and watery eyes1. Airborne allergens are generally considered to be the most common trigger2. Statistical data from the Korean Ministry of Health and Welfare indicate that the number of patients affected by AR in the Republic of Korea has increased over the past 10 years. This may in part be attributable to the recent finding from the Health Effects Institute in the USA that the annual average fine dust concentration in Korea ranked second to Turkey among nations in the Organization for Economic Co-operation and Development. Moreover, there has been a trend toward increasing fine dust concentration in Korea since 2011.

Nasal congestion, one of the common symptoms of AR, can induce the obstruction of nasal airway, leading to mouth breathing1, which has been reported to have adverse effects by changing composition of saliva and oral normal flora inducing gingival inflammation, halitosis, and an altered dentofacial growth pattern3,4,5. In particular, periodontitis, which is characterized by inflammation of the periodontal tissue (including gingiva and alveolar bone), is associated with a significantly higher gingival inflammation index in patients with mouth breathing than in their counterparts with closed-mouth breathing5,6,7. Thus, there appears to be a connection in AR patients between periodontal diseases and mouth breathing induced by nasal congestion. The oral cavity is anatomically adjacent to the nasal airway, which may allow conditions in one to impact those in the other. For example, especially for maxillary molars and sinus floor located in maxilla with anatomical proximity, odontogenic maxillary sinusitis is one of the common diseases8.

According to position paper from the ‘Journal of Periodontology’, the prevalence of periodontal disease in the adult population has been reported to be more than 20% including 50% of those who aged 55–64 suffering from moderate to severe periodontitis9. Periodontitis is a major cause of tooth loss among middle-aged people and older, and can influence quality of life while simultaneously conferring a large social cost for treatment10. In this context, given that AR is an emerging and tentative risk indicator for periodontitis, investigations of the association between AR and periodontitis may contribute to improved public welfare. There have been very few studies of this association, and there is no consistency among these studies. For example, Hung et al. reported that there was increased risk of periodontal disease in patients with AR in Taiwan11, whereas Friedrich et al. demonstrated an inverse association between periodontitis and respiratory allergies including AR12.

The purpose of the present study was to determine whether an association exists between AR and periodontitis among the population of Korea, adjusting for the impact of exogenous variables, which can have influence on the outcome variable, including sociodemographic features, systemic health status, and oral hygiene behaviors.

Results

Characteristics of the study population

Based on previous studies, there may be an association between sociodemographic variables and factors related to systemic diseases, and periodontitis11,13. The characteristics of the study population, which are listed in Table 1, were well-balanced with respect to sex, with 49.75% being male. Mean age of study population was 46.93 years. Education level was classified into three groups: lower than high school, high school, and higher than high school, accounting for 22.54%, 27.44%, and 50.03% of the study population, respectively. The rate of alcohol consumption (defined as at least once per month) was 59.39%, and 22.62% were smokers. The most common systemic diseases were hypertension (HTN, 28.13%), diabetes mellitus (DM, 10.79%), and osteoporosis (OP, 6.23%); rheumatoid arthritis (RA), angina pectoralis, chronic obstructive pulmonary disease (COPD), and myocardial infarction were found in 1.72%, 1.60%, 0.98%, and 0.87%, respectively. The rate of auxiliary oral hygiene device use (AOHD) was 51.96%. A history of treatment for periodontitis (HTP) was recorded for 22.81% of the study population, and AR group took 15.32% of the study population.

Table 1 Characteristics of the study population.
Full size table

Distribution of variables according to HTP

Table 2 presents the distribution of variables according to the presence (HTP group) or absence of HTP (non-HTP group). Education level (< high school), DM, HTN, OP, and age (≥ 65 years) were statistically significantly higher in the HTP group. A diagnosis of AR was recorded for a statistically significantly greater proportion of patients in the non-HTP group than in the HTP group (17.55% vs 11.07%, P = 0.0002). The proportions of subjects with HTP among those with (AR group) and without AR (non-AR group) were calculated as 0.1572 and 0.2416, respectively (Table 3). Thus, the risk of periodontal disease was 1.536-fold higher in the non-AR group than in AR subjects.

Table 2 Distribution of variables according to history of treatment for periodontitis.
Full size table
Table 3 Proportion of patients with a history of treatment for periodontitis among patients with (AR group) and without allergic rhinitis (non-AR group).
Full size table

Evaluation of associations

Associations between multiple factors and HTP were evaluated in two steps to consider the effect of potential exogenous variables of concern (Table 4). It has been suggested that one of the ways of control the exogenous variables is to put this variable into study design as an independent variable14,15. Univariate analysis was performed for each variable, revealing P values of < 0.1 for education level, alcohol consumption, DM, HTN, OP, age, body mass index (BMI), and AR (Fig. 1). Multiple logistic analysis using these variables and sex suggested that there were statistically significant associations between HTP and several of these variables, including education level (higher than high school), with an odds ratio (OR) of 0.70, and presence of OP or AR, with ORs of 1.48 and 0.68, respectively. Statistically significant association was not found from multiple logistic analysis between HTP and other variables such as alcohol consumption, DM, HTN, age and BMI.

Table 4 Results of univariate and multivariate analyses to adjust for exogenous variables.
Full size table
Figure 1
figure 1

Variables included in this study. Line arrow suggests that each variable have association with periodontitis, supported by previous studies. Variables with white color are included in multiple logistic analysis, selected from univariate analysis. Variables with grey color are excluded from multiple logistic analysis. AOHD use of auxiliary oral hygiene devices, AR allergic rhinitis, BMI body mass index, COPD chronic obstructive pulmonary disease, DM diabetes mellitus, HTN hypertension, HTP history of treatment for periodontitis, OP osteoporosis, RA rheumatoid arthritis.

Full size image

Subanalysis based on the characteristics of the study population

One of the characteristics of this study population was the distribution of education level according to age (Table 5). More than 50% of the subjects aged at least 65 years did not graduate from high school. This was adjusted for by performing multivariate subanalysis with discrimination of subjects based on an age of 65 years (Table 6). A statistically significant association was found between AR and HTP among those younger than 65 years (OR = 0.62, P = 0.0057). In addition, the ORs for this association for education level (higher than high school) and OP among those aged younger than 65 years were 0.65 (P = 0.0241) and 1.99 (0.0108) with statistical significance, respectively. No such association was found for the older age group.

Table 5 Distribution of education level by age group.
Full size table
Table 6 Results of multivariate subanalysis based on age.
Full size table

Discussion

The findings of these analyses of data from 6129 subjects extracted from the Seventh Korea National Health and Nutrition Examination Survey (KNHANES VII-1, 2016) demonstrated that for patients aged younger than 65 years, HTP was less prevalent among those with a diagnosis of AR, indicating that patients with AR had a lower risk of periodontitis. This finding is supported by that of a similar study by Friedrich et al., which suggested that there was an inverse association between periodontitis and allergic respiratory diseases12. Moreover, Grossi et al. reported a negative association between history of allergies and the severity of bone resorption based on a cross-sectional study of 1361 patients16.

The present study also revealed associations between HTP, education level, and presence of OP. This is in line with the finding in another study finding a significant inverse association between education level and risk of periodontitis17. Furthermore, OP is considered as a risk factor for the progression of preexisting periodontitis18.

These statistical phenomena can be interpreted in the context of the T-cell-mediated immune response. Upon recognizing an antigen, naïve T cells differentiate into several kinds of T cells, including T-helper (Th) cells and regulatory T cells (Treg)19. There are three types of Th cell: Th1, Th17, Th2, and Treg19.

The Th1/Th2 hypothesis is one of theories regarding the mechanism of immune regulation and is based on homeostasis between the activities of Th1 and Th2 cells20, in which they act as cross-inhibitors for each other, thus maintaining a balance in their activities19,21. Th1/Th2 immune responses can account for various diseases22. Th1-related cytokines are connected to autoimmune-related pathology20. They promote inflammation pathways through secretion of the cytokine interferon gamma (IFN-γ), which activates macrophages, in turn suppressing Th2 activity19,20. In some studies it was suggested that Th1 cells and IFN-γ contribute to the breakdown of periodontal tissue by stimulating monocytes and macrophages23. Meanwhile, Th2-related cytokines are involved in the genesis of allergic diseases22. Th2 cells inhibit Th1 cells via the production of interleukin (IL)-10; they also stimulate antibody formation by B cells through the production of IL-4 and IL-520,21. One of the main roles of Th2 cells is the production of the immunoglobulin E (IgE)-synthesizing cytokines IL-4 and IL-13; IgE is involved in the allergic reaction24.

Periodontitis is regarded as an infectious pathology of periodontal tissue with several specific characteristics25. Subgingival pathogens can interact with and invade periodontal tissues25. Although bacterial pathogens are considered to initiate the periodontal disease, the host response appears to be related to destruction of gingival tissue and bone26. Invasion of these antigens can cause an inflammatory reaction and generation of immune responses, including innate and adaptive immune responses26. Periodontal tissue breakdown occurs mainly via cellular immune responses together with proinflammatory mediators such as tumor necrosis factor, IL-1β, and IL-17, which promote degradation of gingival tissue and bone resorption19,27. As a part of this mechanism, activated lymphocytes including Th1 and Th17 play important roles in the loss of bone through a RANKL-dependent mechanism28.

AR, a symptomatic pathologic state of the nose caused by exposure to allergens, is an IgE-mediated hypersensitivity reaction29. The pathogenesis of AR starts with the dendritic-cell-induced activation of Th2 cells, which themselves induce the production of IL-4 and IL-13, and ultimately IgE30. Specific IgE antibodies formed by B cells become attached to mast cells to enable cross-linking between the two31. This results in release of histamine, leukotrienes, and prostaglandins from mast cells, causing typical immediate reactions of AR such as sneezing, itching, and running of the nose or blockage of the nasal epithelium30. As mentioned above, overactivation of Th1 or Th2 cells can cause disease, and either pattern may inhibit the other20. In this context, AR with a Th2-dominant state may cause the down-regulation of the Th1 pathway, resulting in suppression of periodontal tissue destruction by proinflammatory cytokines.

Children account for almost 40% of all AR patients, with adults accounting for only 10–30%32,33,34. Conversely, several studies suggest that the prevalence of periodontitis increases with age35,36,37. The differences in the distribution of AR and periodontitis prevalence with age may support the explanation for the present findings. It can be hypothesized that diagnosed AR in young age may affect the occurrence of periodontitis in older age. In the context of immunology, a statistical association between periodontitis and Th2-related diseases such as asthma and atopic dermatitis is required to support that found in the present study between AR and periodontitis.

The Th1/Th2 theory is just one of the ways of explaining and understanding the process of immune regulation; however, that theory is still considered controversial, with limitations and discrepancies. Further research based on large-scale human studies are needed to support its validity20.

In conclusion, statistical analysis of data extracted from KNHANES VII-1 (2016) revealed that there was a significant association between AR and HTP, suggesting reduced risk of periodontitis in the AR group compared with the non-AR group particularly among those younger than 65 years. Higher education level was associated with decreased risk of periodontitis and presence of OP was associated with increased risk of periodontitis. The limitation of this study was that supplementary statistical analysis about secondary effects of variables should be included in multivariate statistical analysis although subanalysis based on the distribution of education level according to age was performed. In addition, the association between variables should be considered and reflected on statistical modeling. Variables in this study were selected from the KNHANES VII-1 (2016) raw data, which were known to have association with periodontitis according to the previous studies37,38,39. In Table 4, the crude model shows the result of univariate analysis. In this part, the association between each selected variable and HTP excluding other variables was checked. Based on this result, only variables with relatively higher statistical association with HTP were included in multiple logistic analysis. In this procedure, the association between HTP and each variable was evaluated, considering influences of exogeneous variables by including them as independent variables in multivariate analysis. Additionally, this study, as a cross-sectional study, was concentrated on association between two diseases based on ever prevalence of each disease in the specific time point. As mentioned above, the average diagnosis age of AR is younger than that of periodontitis in general, however, it cannot be demonstrated that all cases in this study have same sequence of onset. Therefore, data including the sequence of diagnostic experiences of both diseases in each individual are needed. There could be limitation on interpretation of cause-and-effect relation comparing to longitudinal study. To redeem this limitation and clarify statistical causation between two diseases, study including diagnosis age is necessary. Also, further longitudinal study is required to reveal the association between these factors and periodontitis. In this study, the association between AR and HTP in Korean population was suggested, for which there have been no research results so far. The limitation of this study will be supplemented through further studies.

Methods

Study population

This study was based on data from the KNHANES VII-1 (2016), conducted by the Korea Centers for Disease Control and Prevention40. The study included 6129 subjects, all of whom were adults older than 19 years. This study was approved by Institutional Review Board (IRB) of Ewha Womans University (approval No. EUMC 2020-02-033). The research was performed in accordance with relevant guidelines and regulations.

Variables

Data on HTP, education level, income, alcohol consumption, smoking, AOHD, diagnosis of diseases such as AR, DM, HTN, COPD, RA, OP, and cardiovascular disease, and BMI were extracted from the data for the included individuals. AR was defined based on the self-report of diagnosis experience of AR by doctor ever. In case of HTP, it was also defined based on self-report of history of periodontal treatment. Investigation of medical history included in the present study were examined by questionnaire. Other factors except BMI such as age, sex, education level, and income were also examined by questionnaire. BMI was calculated by measuring height and weight of individuals. These variables except HTP and AR were selected as exogenous variables based on the previous studies suggesting that they had association with periodontitis37,38,39. As previous studies suggested, diagnosis of periodontitis was defined based on the HTP41,42. HTP included periodontal treatment other than scaling such as subgingival curettage, periodontal flap operation, and gingivectomy. Education level was divided into three groups: lower than high school, high school, and higher than high school. Income was classified into four grades (quartiles). Alcohol consumption was defined as a history of alcohol intake on one or more occasion per month within a 1-year period. AOHD, including floss, interdental brush, mouthwash, electric toothbrush, water flosser, tongue cleaner, and end-tuft brush, was ascertained by questionnaire (the “not using” group included only subjects who did not use any AOHD).

Statistical analysis

Statistical estimations were made for the Korean population based on samples from KNHANES VII-1 using a complex sample design. Statistical analysis of this study was proceeded according to the analysis guideline from Korea Centers for Disease Control and Prevention. The sample weights, assigned to each variable, were set for sample participants to reflect the Korean population. Weighted percentages were used to express the proportions of each variable among the total population. Proportion of patients with HTP among patients with and without AR was calculated based on weighted percentage of HTP and AR group. Multiple logistic regression analysis was used to analyze associations between HTP and the other variables, with adjustment for exogenous variables, providing ORs. Univariate analysis was performed before multiple logistic regression and only variables with P value less than 0.1 were included in multiple logistic regression. Confidence level for multiple logistic regression analysis was set to 95%. P-value less than 0.05 was chosen as threshold for statistical significance. Additional sub-analysis, dividing population into two group based on age 65, was performed because population with lower education level was concentrated in aged over 65 group. There was no remarkable uneven distribution among other variables. SAS for Windows (version 9.4, SAS Institute, Cary, NC, USA) was utilized.