The effect of air pollution exposure on the risk of outpatient visits for periodontitis: a time-series study

ABSTRACT This study aimed to determine if air pollution affected the risk of periodontitis outpatient visits. We collected the records of 56,456 periodontitis outpatient visits in Hefei, China, from January 1, 2014 to December 31, 2021. The relationship between air pollution and periodontitis outpatient visits was evaluated using distributed lag nonlinear and generalized linear models. Additional analyses were performed, stratifying the data by age, season, and sex. Subgroup analyses showed a significantly higher risk of periodontitis outpatient visits due to NO2 exposure during the warm season compared with the cold season. Moreover, O3 exposure was associated with a lower risk of periodontitis outpatient visits in the cold season. The findings suggest that NO2 exposure is associated with an increased risk of periodontitis outpatient visits, whereas O3 exposure is associated with a decreased risk of periodontitis outpatient visits. Season is found to be an effect modifier in these associations.


Introduction
Periodontitis is a chronic infectious disease caused by plaque biofilm formation involving four types of periodontal supporting tissues, including gingiva, periodontal membrane, alveolar bone, and cementum.Periodontitis progressively destroys periodontal supporting tissues and clinically manifests as periodontal pocket formation with bleeding, attachment loss, and alveolar bone resorption (Pihlstrom et al. 2005).
Periodontitis is a global public health problem that significantly affects the quality of life of patients and imposes a heavy health and economic burden on patients, their families, and the society (Peres et al. 2019).In addition, periodontal disease is closely related to systemic health.Evidence from population studies, clinical investigations, and in vitro animal experiments all highlight the importance of oral health for systemic health (Peres et al. 2019).Thus, the investigation of risk factors for periodontitis is vital not just for taking timely preventive measures, but also for enhancing overall health (Kapila 2021).
Recent studies have shown that, in addition to bacterial factors, host factors play an important role in the development of periodontitis, which is the main mechanism linking periodontal and systemic diseases (Cardoso et al. 2018).In addition to host factors, pro-inflammatory events caused by environmental factors can exacerbate the adverse effects on inflamed periodontal tissues.For example, the impact of tobacco smoke on periodontitis has been demonstrated in several studies, making tobacco smoking one of the systemic promoters of periodontitis (Baumeister et al. 2021).Researchers from the United States Environmental Protection Agency concluded that exposure to environmental tobacco smoke could be harmful, most likely due to the presence of nitrogen oxides in the smoke (Anyanwu 1999).Numerous substances, including nitrogen dioxide (NO 2 ), a major contributor to air pollution, are included in nitrogen oxides.Similar to the exposure characteristics of tobacco smoke, air pollutants can come into prolonged contact with and remain in the tissues of the oral cavity, and other harmful components of these air pollutants, such as particulate matter (PM) 2.5 , carbon monoxide (CO), ozone (O 3 ), sulfur dioxide (SO 2 ), and PM 10 (Fu et al. 2018), have the potential to adversely affect the oral cavity, teeth, and their supporting tissues (Yang et al. 2015;Zhao et al. 2018;Glodkowska and Emerich 2020;Li et al. 2023;Marruganti et al. 2023).To investigate the underlying mechanism, researchers hypothesised that air pollutants may play a role in the pathogenesis of periodontal disease by affecting local and systemic inflammation, oxidative stress and oral microbial composition (Yang et al. 2015;Wu et al. 2021;Li et al. 2023;Marruganti et al. 2023).A study by Marruganti et al. (2023) suggests that air pollutants may act as novel indicators of periodontitis risk.The other study by Yang et al. (2015) suggests that patients with periodontal disease are more sensitive to systemic inflammatory responses induced by PM 2.5.
Based on the above evidence, air pollutants may be involved in periodontitis development and progression.Therefore, the present study aimed to investigate the potential relationship between air pollution and outpatient visits for periodontitis using the time-series analysis.Furthermore, we sought to determine whether this relationship varied by sex, age, and season.

Location of the study
Hefei is a sub-center of the Yangtze River Delta urban agglomeration and a primary city in the Wanjiang River urban belt.It is located in the center of Anhui Province in East China and is the economic, political, and cultural center of the region.By the end of 2021, it covered a total area of 114,496 km 2 and had a resident population of 9.47 million.Recently, Hefei has experienced an air pollution problem due to the city's rapid rise in residential population and economic development.

Data collection
The First Affiliated Hospital of Anhui Medical University, Second Affiliated Hospital of Anhui Medical University, First People's Hospital of Hefei, and Second People's Hospital of Hefei provided clinical data for the study, which included patient age, sex, address, and dates of outpatient visits for periodontitis from 1 January 2014 to 31 December 2021.The inclusion criteria were as follows: (1) patients residing in the Hefei city district to ensure that all selected patients belonged to this region, and (2) those who had their initial visit recorded in dental and emergency department, with a primary diagnosis of periodontitis.The diagnostic criteria for periodontitis followed the 1999 classification of periodontal diseases (Armitage 1999), which was based on plaque and dental calculus accumulation, periodontal pocket formation, and periodontal attachment loss.The exclusion criteria included the following: (1) patients with periodontitis who were not residents or who had been diagnosed with other diseases and were referred from other departments, and (2) repeat or follow-up patients who visited the four hospitals in a short period of time.
The National Environmental Testing Center of China (http://www.cnemc.cn/)provided data on air pollutants, which included statistics on the daily 24 h average concentrations of NO 2 , SO 2 , CO, PM 10 , PM 2.5 and the 8 h maximum concentration of O 3 .Daily meteorological statistics, including relative humidity, mean temperature, and wind speed, were obtained from the China Weather Data Network (http://data.cma.cn/).

Statistical analysis
A distributed lag nonlinear model together with a quasi-Possion generalized linear model was used to evaluate the impact of air pollutants on the risk of outpatient visits for periodontitis.These two functions, which describe a conventional exposure and an additional lagged response correlation, allow simultaneous characterization of complex nonlinear and lag relationships.Natural cubic spline (ns) functions were used to adjust meteorological variables and long-term trends.To address multicollinearity concerns, covariates with correlation coefficients greater than 0.7 were excluded from the model using Spearman's correlation analysis.The final model is as follows: where, E(Y t ) denotes the expected number of outpatients with periodontitis at time t of observation; t denotes the time of observation; l denotes the number of days lagged; α represents the intercept; βZ_(t,l) is the cross base used to calculate the effect of air pollutants; Z is determined by each pollutant (SO 2 , NO 2 , O 3 , PM 10 , PM 2.5 , and CO); and β is the coefficient.Mean temperature, relative humidity, and average wind speed were used to control the impact of meteorological factors on the incidence of periodontitis.ns is a natural cubic spline function with df as its degree of freedom.Time is a date variable for controlling temporal trends and seasonal fluctuations.DOW stands for the day of the week and controls natural fluctuations.Based on previous observations and studies, the df for meteorological factors was set at 3, and the df for time at 8 or 6.
Based on prior relevant literature (Gao et al. 2020;Wang et al. 2021) and the quasi-Akuchi Information Criteria, the final lag days for air pollutants were selected.Additionally, stratified subgroup analyses were performed based on patient age (<65 and ≥65 years), sex (male and female), and season of clinic visit (warm season from April to September and cool season from October to March), and a subgroup analysis was conducted.

Sensitivity analysis
To evaluate model robustness, Spearman correlation coefficients were applied to a two-pollutant model.Additionally, the df was modified for the calendar year (6-8 df/year).
The Spearman correlation coefficient of air pollutants is shown in Supplementary Material Figure S1.Considering the effect of multicollinearity, correlation coefficients greater than 0.7 were not included in the model.The temporal distribution of air pollutants (SO 2 , NO 2 , CO, and O 3 ) from 1 January 2014 to 31 December 2021, is presented in Supplementary Figure S2.

Relationship between air pollutant concentrations and risk of outpatient visits for periodontitis
The exposure-effects curve is shown in Figure 1.An increase in NO 2 concentration was associated with a higher risk of outpatient visits for periodontitis, whereas an increase in O 3 concentration was associated with a lower risk of outpatient visits for periodontitis.No significant association was observed between SO 2 and CO levels and the risk of outpatient visits for periodontitis.Additionally, PM 2.5 and PM 10 pollutants did not reveal any significant association (Supplementary Figure S3).

Association between air pollution exposure and risk of outpatient visits for periodontitis under lagged model
The air pollutants, NO 2 and O 3 , are significantly associated with periodontitis outpatient visits, while the particulate matter (PM 2.5 and PM 10 ) did not exhibit any association (Supplementary Figure S4).

Effect of NO 2
Figure 2 depicts that a 10 ug/m 3 increase in NO 2 concentration exhibited a significant adverse effect on the risk of outpatient visits for periodontitis.The single-day lag association was significant from lag 0 (relative risk (RR) = 1.014, 95% confidence interval (CI): 1.006-1.022)to lag 2 (RR = 1.005, 95% CI: 1.002-1.008).Moreover, on all cumulative lag days, a significant adverse impact on periodontitis outpatient visits were observed.The cumulative lag effect peaked at lag 0-4 days (RR = 1.030, 95% CI: 1.016-1.044).The single-day and cumulative effects of NO 2 are described in Table S1.The results of the stratified analyses by sex, age, and season were shown below.The singleday lag impact of NO 2 on periodontitis visits was significant at day 1 for men and lag0-lag2 for women.For women, the effect reached its peak at lag0 (RR = 1.018, 95% CI: 1.008-1.028)(Figure3).The single-day lag effect of NO 2 on the risk of outpatient visit for periodontitis for patients aged <65 years was significant at lag0-lag2 and reached its peak on lag0 (RR = 1.012, 95% CI: 1.004-1.021);for patients aged ≥65 years, the single-day lag effect was significant on days 0-1 and reached its peak on day 0 (RR = 1.023, 95% CI: 1.007-1.039)(Figure 4).At a single-day lag of 5-7, NO 2 significantly affected periodontitis outpatient visits both during cold and warm seasons.However, the impact of NO 2 on outpatient visits for periodontitis during the warm season was detrimental (RR > 1) and that during the cold season was protective (RR < 1) (Figure 5).Nevertheless, further subgroup analyses indicated significant statistical differences between the two seasonal categories.NO 2 exposure was linked to higher risk of periodontitis for outpatients in the warm season more than those in the cold season.Conversely, no significant statistical difference was observed between the two gender and age categories (Supplementary Table S1).

Effect of SO 2
Figure 2 depicts that no significant difference was observed between the single-day and cumulative lag in SO 2 on the risk of outpatient visits for periodontitis.Additionally, SO 2 exhibited a protective effect on periodontitis outpatient visits during the cold season at single-day lags of 6 and 7 days (RR = 0.976, 95% CI: 0.960-0.993;RR = 0.968, 95% CI: 0.939-0.998)(Figure 3-5).None of the results from the further subgroup analyses showed statistical significance, as revealed in Supplementary Table S2.

Effect of O 3
Figure 2 depicts that O 3 exhibited a protective effect on the risk of outpatient visits for periodontitis at single-day lags of 0, 1, 2, and 6 days, as well as on all cumulative lag days.In the sex-stratified analysis, O 3 exhibited no significant effect on the risk of outpatient visit for periodontitis in men, but it exhibited a significant protective effect on the risk of outpatient visits for periodontitis in women at single-day lags of 1, 5, 6, and 7 days (Figure 3). Figure 4 Additionally, O 3 exhibited a protective effect during the cold season but had no significant effect during the warm season (Figure 5).The results of further subgroup analyses revealed that there was a significant difference between the two seasonal categories.O 3 exposure was associated with a lower risk of periodontitis outpatient visit for periodontitis in the cold season (Supplementary Table S3).

Effect of CO
Neither the single-day nor cumulative lag revealed any significant effects of CO on the risk of outpatient visits for periodontitis (Figure 2).Moreover, the stratified analyses based on age and sex did not reveal significant findings.In the stratified analysis based on the season, CO had a significantly negative impact on the risk of outpatient visits for periodontitis during the warm season at single-day lags of 3 and 4 days and a positive impact during the cold season at single-day lags of 4, 5, and 6 days (Figure 5).Further subgroup analyses revealed a significant difference between two different seasons.CO exposure was significantly correlated with elevated outpatient risk of periodontitis in the warm season, in contrast to the cold season (Supplementary Table S4).

Sensitivity analysis
To evaluate model robustness, sensitivity analyses were carried out, primarily using a two-pollutant model, as well as by modifying the annual degrees of freedom (df 6-8).No significant changes were observed in the results, indicating that the model was robust and the results were reliable (Supplementary Table S5, Table S6).

Discussion
Periodontal diseases are highly prevalent, affecting over 90% of the worldwide population (Pihlstrom et al. 2005).The scientific investigation of periodontal diseases has become a major area of interest for scientists worldwide.Several studies have shown shared epidemiological associations and underlying inflammatory mechanisms between periodontitis and other chronic inflammatory diseases (Cardoso et al. 2018;Kapila 2021).Evidence is increasing that short-term exposure to air pollutants is associated with the development of many chronic diseases (World Health Organization 2021).Nevertheless, studies on the relationship between air pollutants and periodontitis are limited.Our study suggested a potential correlation between air pollutants and periodontitis.
The results of this study, which are consistent with those of earlier investigations into bacterial infectious diseases such as pulpitis (Li et al. 2023) and tuberculosis (Huang et al. 2020) in China, revealed the significant impact of NO 2 on the increased risk of outpatient visit for periodontitis.Stratified and subgroup analyses based on seasons revealed that NO 2 exposure was significantly correlated with elevated outpatient risk of periodontitis in the warm season, in contrast to the cold season.This could be explained by the fact that people spend more time outdoors during the warmer months, thereby increasing their exposure to air pollution and worsening its negative effects.Regarding the mechanisms, the host's immune response may play a crucial role in mediating the effects of air pollutants.Growing evidence shows that air pollutants may impact systemic health by influencing biological processes, such as stimulating the immune system and triggering proinflammatory responses (Zhao et al. 2019).This viewpoint is supported by certain investigations, suggesting that exposure to air pollutants raises the risk of developing immune and inflammatory conditions, such as rheumatoid arthritis (Wu et al. 2021).As systemic and oral health have a mutually reinforcing relationship and their causal factors can intersect to some extent, exposure to air pollution can also possibly affect periodontal health by initiating localized periodontal inflammation and oxidative stress.However, the exact biological mechanisms need to be confirmed by performing more experiments in the future.
NO 2 , along with NO, belongs to the NOx class of extremely reactive gases.NOx is one of the causes of acid rain and poses a threat to human health.Based on the existing literature, when NO 2 is dissolved in water, nitrous acid and nitric acid are produced (2NO 2 + H 2 O-HNO 2 + HNO 3 ) (Elsayed 1994), which may lower dental plaque pH when inhaled, and initiate periodontal disease.In contrast, the acidified plaque can increase nitrite concentration and nitrogen oxide production in the plaque, which diffuses into the adjacent periodontal tissue, causing further deterioration (Takahama and Hirota 2010).Additionally, a lowered pH in the oral environment leads to an imbalance in the typical oral flora, allowing pathogenic periodontitis bacteria, including Porphyromonas gingivalis and Prevotella intermedia to proliferate (Aamdal-Scheie et al. 1996).Air contaminants, such as NO 2 , when inhaled into the mouth, worsens hypoxia in the gingival sulcus or periodontal pocket, creating an oxygen-free environment conducive to the survival and proliferation of pathogenic bacteria, primarily subgingival non-adhesive facultative anaerobic bacteria, which may speed up tissue destruction in periodontitis (Gui et al. 2019).Furthermore, several studies have demonstrated that the presence of NO (one of the nitrogen oxides) in saliva serves as a reliable marker of the degree of periodontal disease (Parwani et al. 2012;Sukuroglu et al. 2015).Low levels of NO can keep tissues in a state of homeostasis, but when NO is produced locally at high concentrations, it can be cytotoxic against cells infected with fungi, bacteria, and protozoa, as well as tumor cells and cells near the production site, causing tissue destruction (Kendall et al. 2001;Parwani et al. 2012).Additionally, the production of NO, or peroxynitrite, has been linked to the pathophysiology of several inflammatory diseases and may trigger unfavorable host responses, including protein damage and inflammatory cytokine production (Paquette and Williams 2000;Brennan et al. 2003).Our findings suggested that O 3 exposure may have a preventative effect on periodontal health, leading to a lower risk of outpatient visits for periodontitis.These results are in line with previous findings on systemic diseases associated with periodontitis, such as ischemic stroke (Wing et al. 2017) and gestational diabetes mellitus (Zhang et al. 2020).O 3 is a powerful bactericidal agent that can kill various Gram-positive and Gram-negative bacteria viruses, fungi, and spores (Ramirez-Pena et al. 2022).As periodontitis is primarily caused by anaerobic bacteria, the bactericidal effects of O 3 could play a protective role in reducing the risk of outpatient visits for periodontitis.Currently, O 3 is used in dentistry to treat early caries, infections of the root canals, and periodontal pockets.It is also used as a rinse for dental avulsions to speed up the healing of epithelial wounds, such as those caused by ulcers and herpetic lesions, and to disinfect dentures (Deepthi and Bilichodmath 2020).The study's subgroup analysis revealed that O 3 exposure during the cold season was linked with a lower risk of outpatient visits for periodontitis.However, the current literature is inadequate, and additional studies are necessary to confirm this association.
Our results demonstrated no significant association between SO 2 , PM 2.5 , and PM 10 , with the risk of outpatient visits for periodontitis.This finding is inconsistent with the results of a recent study reported in Korea, wherein researchers found a positive association between PM 10 and periodontitis prevalence and an inverse association with NO 2 , using simple and multiple regression analysis (Marruganti et al. 2023).The possible explanations were as follows: (1) the differences in meteorological and air pollutants concentrations between the two regions were significant, and (2) Unlike the time series analysis method used in this study, regression analyses did not consider the lagged effect of air pollutants.However, given this inconsistency, larger scale and more comprehensive studies should be performed.Regarding the effect of SO 2 , a previous research had reported that endogenous SO 2 has antioxidant, anti-inflammatory, and anti-hypertensive effects in mammals, despite its toxic effects (Wang et al. 2014).In macrophages, reduction of aspartate aminotransferase AAT2 activates the NF-κB signaling pathway, whereas SO 2 effectively rescues NF-κB activation (Zhu et al. 2020).It was speculated that the lack of a notable impact of SO 2 in this study was related to its dual effect.
This study has some limitations.First, it may contain an ecological fallacy due its design.Second, although data were collected from four reputable dental hospitals in Hefei, the findings may not be fully representative of the entire city.This investigation also presents several advantages.First, a new statistical model is employed to examine the correlation between exposure to air pollutants, lagged effects, and periodontitis outpatient visits.Second, the identified high-risk seasons could aid in developing more precise preventative measures.
In conclusion, a higher risk of outpatient visit for periodontitis is associated with NO 2 exposure, whereas a lower risk of periodontitis outpatient visit is associated with O 3 exposure.Moreover, early preventive measures, such as wearing a mask on heavily polluted days, especially during the warm season, must be recommended to reduce outpatient visits for periodontitis.

Figure 1 .
Figure 1.Association between the risk of periodontitis among outpatients and NO 2 , SO 2 , O 3 , and CO levels.The green line denotes the relative risk (RR), and the shaded region denotes the 95% confidence interval (CI).

Figure 2 .
Figure 2. Single-day (a) and cumulative relative risks (b) of periodontitis incidence associated with an increase of 10 μg/m 3 (or 1 mg/m 3 ) in NO 2 , SO 2 , O 3 , and CO at lag 0-7 days.

Figure 3 .
Figure 3. Lag-specific relative risks (95% confidence interval) of periodontitis per 10(or 1)-unit increase in the daily concentrations of air pollution in models stratified by sex.

Figure 4 .
Figure 4. Lag-specific relative risks (95% confidence interval) of periodontitis per 10(or 1)-unit increase in the daily concentrations of air pollution in models stratified by age.

Figure 5 .
Figure 5. Lag-specific relative risks (95% confidence interval) of periodontitis per 10(or 1)-unit increase in the daily concentrations of air pollution in models stratified by season.

Table 1 .
Summary statistics of periodontitis, air pollutions, and meteorological variables in Hefei, China from 2014 to 2021.