Although a few environmental and occupational exposures such as tobacco smoke, fire smoke and dust exposure have been linked with the prevalence of CRS, the association between the prevalence and severity of CRS and outdoor air pollution is much less understood (1,4,26). To our knowledge this is the first study using mixed models to describe symptom scoring data of CRS patients in relation to outdoor air pollution exposure. We here demonstrated that exposure to O3 and PM2.5 leads to an increased CRS symptom scoring in CRS patients. Sensitivity of CRS patients to outdoor air pollution exposure depends on their gender, presence of comorbid asthma and history of sinus surgery.
In our study, we speculated that the models created for the spring–summer population were better in reflecting personal exposure to outdoor air pollution than the models fitted for the fall-winter population. It has been previously demonstrated that during colder months people spend more time inside and aerate their households less (27). Therefore, a stronger association has been observed between indoor exposure and outdoor exposure during warmer months compared to the colder months (27). It was observed that in the fall-winter population CRS symptom scores were higher than those of the spring-summer population, while the association between outdoor air pollution and increased CRS symptoms were more present in the spring-summer population. This is likely due to the fact that during the winter viral infections are more prevalent among the population which often exacerbate CRS symptoms (28). This may thus overshadow the effects of the pollutants on the symptoms, which together with the fact that the fall-winter population models contain less accurate estimates of the exposure, explain the observation of nonsignificant changes in the total CRS symptoms after exposure in this population.
The unstratified adjusted spring-summer population showed a significant increase and decrease in symptom score for O3 and NO2 exposure respectively. The increase of symptoms associated with O3 exposure is approximately three times greater than the decrease of symptoms associated with NO2 exposure when comparing their IQR estimates. O3 exposure has previously been shown to cause damage to the nasal epithelium and to cause increased nose and sinus symptoms, such as rhinorrhea, nasal dryness, nasal obstruction, epistaxis and olfactory impairment, in non-CRS patient (29–31).
The observed decrease in symptoms associated with NO2 exposure could be explained by the interaction between O3 and NO2. O3 is formed by a photochemical reaction requiring NOx and volatile organic compounds and an inverse relationship exists between O3 and NO2 (32,33).
No significant effects were observed for black carbon. Mady et al. observed more pronounced disease progression after BC exposure (17). The relative low sample size may have been the reason why we failed to detect the association between BC and the symptoms in the adjusted spring-summer population.
In our population, male patients are more sensitive to O3 and PM2.5 exposure compared to female patients. In literature, evidence of effect modification by gender on respiratory health remains uncertain as studies most often find stronger effects of pollution exposure among women, however certain studies have also suggested stronger effects among men (37). Confounding may exist because of unmeasured characteristics, including biological factors related to deposition, reactivity, and hormonal influences on chemical transport and systemic regulation as well as gender-related explanations such as exposures to indoor allergens and cleaning agents, job-related chemical exposures, and differing exposure and response to psychosocial stressors (37).
A remarkable finding of this study is that CRS patients with comorbid asthma are sensitive to outdoor air pollution while non-asthmatic CRS patients are not. This could be explained by the fact that asthma exacerbations are known to be associated with outdoor air pollution and that poor asthma control has a negative effect on CRS patient outcomes (11,38,39).
Sinus surgery significantly changes the airflow and increases the deposition of particles in the sinuses, possibly making patients more sensitive after surgery (40). However, we observed significant symptom increases after exposure to PM2.5 in the group that did not have prior sinus surgery and no significant symptom changes in the prior surgery group. This may be due to the fact that the time between the last sinus surgery and symptom reporting varies between patients and no information was available on patient’s date of last sinus surgery.
A previous study demonstrated that CRSsNP patients are more sensitive to PM and BC compared to CRSwNP patients (17). In our population the observed increases for the CRSsNP patients for BC and PM2.5 were not significant, however changes in symptom severity were higher for these pollutants in CRSsNP compared to CRSwNP patients. For O3 we did not observe an impact of prior sinus surgery.
Certain limitations of this study should be noted. A first limitation of the study relates to the fact that we calculated exposure data for the coordinates recorded on the day of entry and assumed the same location for up to the prior seven days. Besides, outdoor pollutant concentration does not necessarily reflect the actual personal exposure of the patients to pollutant since indoor pollutant concentrations have been shown to impact patients’ health variables as well (27). Secondly, patients’ self-reported outcome measures were based on total sinusitis symptoms in this analysis. In future studies with larger numbers of patients reporting on specific symptoms of CRS, it would be interesting to study how specific symptoms such as impaired smell, facial pain and nasal blockage are affected in CRS patients by air pollution. This type of analysis featuring longitudinal real-life data can have many significant implications in further understanding the real-world impacts of pollution on CRS symptom outcomes. Therefore, we suggest a follow-up study to be performed with a larger dataset on a larger geographical scale.
In conclusion, several novel findings have been observed in the present study for CRS patient symptomatology in relationship to air pollution exposure. During the spring and summer period in Belgium it an association between the total CRS symptoms and O3 as well as PM2.5 exposure was demonstrated in CRS patients. Additionally, male patients and patients with comorbid asthma appeared to be more sensitive to the exposure of several of the pollutants compared to the female CRS patients and patients without comorbid asthma.