To the Editor:
Chronic rhinosinusitis (CRS) is a debilitating condition affecting millions of adults and is associated with depression, anxiety, impaired sleep, and low quality of life (1). Although its pathogenesis remains unclear, recent epidemiological studies have implicated environmental exposures in CRS (2). Indeed, airborne particulate matter ⩽2.5 μm in aerodynamic diameter (PM2.5) exacerbates lower airway conditions causing inflammation (3). Whether PM2.5 has similar effects in the upper airway, as might be expected by the “unified airway” concept, has not been demonstrated. In prior mouse studies, we found that PM2.5 exposure causes type 2 eosinophilic sinusitis, but human data are limited (4). Thus, the purpose of this study was to determine whether airborne PM2.5 exposure is associated with the development of CRS. Data were extracted from the outpatient otolaryngology clinics at an academic medical center and analyzed via a case–control approach. The study was approved by the institutional review board of the Johns Hopkins University School of Medicine.
Cases were defined as new patients aged ⩾18 years diagnosed with a CRS ICD-10 code by a board-certified otolaryngologist using nasal endoscopy and computed tomography (CT) scans. Patients who gave a history of environmental allergy were excluded. Two control subjects without such diagnosis codes and with clear sinus CT scans were selected for each case using the nearest neighbor strategy to match for age, sex, race, and date of CRS diagnosis. Clinical characteristics were extracted, together with the onset of CRS, defined as diagnosis date. Ambient PM2.5 exposure levels were estimated based on validated prediction models (5). Briefly, machine learning approaches that incorporated meteorological measurements, land-use terms, satellite-based measurements, and simulation outputs from a chemical transport model were used to predict daily PM2.5 concentrations. We calculated mean PM2.5 exposures for each patient based on their residential address postal code at 12, 24, 36, and 60 months before the diagnosis date. A Bayesian space-time downscaler model was used to estimate daily ozone (8-h max) exposure using monitoring data from the National Air Monitoring Stations and Local Air Monitoring Stations.
Conditional logistic regression models were used to determine the association between long-term PM2.5 exposure and CRS. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were obtained by adjusting for covariates and potential confounding factors including age, sex, race, body mass index, alcohol consumption status, smoking status, hypertension, diabetes, chronic obstructive pulmonary disease, and asthma. We used 5 μg/m3 as the scale to facilitate comparison with previous studies. To determine if any associations were specific to anatomy, we examined CRS subtypes based on anatomy involved as secondary outcomes, including chronic maxillary, frontal, ethmoidal, and sphenoidal sinusitis, based on sinus CT scans. We also examined severe disease (presence in all four sinuses). Statistical analyses were conducted using STATA (version 16.0; Stata Corp.) and R (version 4.1; R Development Core Team).
A total of 6,102 subjects met inclusion criteria: 2,034 cases and 4,068 controls, 90% residing in the Northeast. Age, sex, and race were similar between groups (Table 1). The mean (SD) PM2.5 exposures during the 12-, 24-, 36-, and 60-month periods before diagnosis were 10.1 (1.7), 10.4 (1.9), 10.6 (1.9), and 10.9 (1.9) μg/m3, respectively. Cases had a higher prevalence of obesity (36.2% vs. 28.9%) and current smoking (66.7% vs. 57.8%).
Table 1.
Demographic and Clinical Characteristics of Participants
| Characteristics | Control Subjects (n = 4,068) | Patients with CRS (n = 2,034) |
|---|---|---|
| Age, yr | 51.9 ± 17.4 | 51.1 ± 16.0 |
| Sex, M | 1,771 (43.5) | 840 (41.3) |
| Race | ||
| White | 2,694 (66.2) | 1,333 (65.5) |
| African American | 861 (21.2) | 425 (20.9) |
| Hispanic/Latino ethnicity | 177 (4.4) | 110 (5.4) |
| Other | 336 (8.3) | 166 (8.2) |
| 12-mo PM2.5 average, μg/m3 | 9.9 ± 1.9 | 10.1 ± 1.7 |
| 24-mo PM2.5 average, μg/m3 | 10.2 ± 2.0 | 10.4 ± 1.9 |
| 36-mo PM2.5 average, μg/m3 | 10.4 ± 2.1 | 10.6 ± 1.9 |
| 60-mo PM2.5 average, μg/m3 | 10.7 ± 2.1 | 10.9 ± 1.9 |
| 12-mo ozone average, ppb | 40.8 ± 2.1 | 40.7 ± 1.8 |
| 24-mo ozone average, ppb | 41.0 ± 1.9 | 41.1 ± 1.6 |
| 36-mo ozone average, ppb | 41.0 ± 1.9 | 40.9 ± 1.5 |
| 60-mo ozone average, ppb | 41.0 ± 2.2 | 40.9 ± 1.8 |
| BMI, kg/m2 | ||
| Underweight (<18.5) | 136 (3.3) | 32 (1.6) |
| Normal weight (⩾18.5 to <25) | 1,491 (36.7) | 623 (30.6) |
| Overweight (⩾25 to <30) | 1,265 (31.1) | 643 (31.6) |
| Obese (⩾30) | 1,176 (28.9) | 736 (36.2) |
| Current smoking status | ||
| Never-smoker | 1,245 (30.6) | 608 (29.9) |
| Current smoker | 2,353 (57.8) | 1,357 (66.7) |
| Former smoker | 478 (11.8) | 172 (8.5) |
| Current alcohol consumption | 1,237 (30.4) | 505 (24.8) |
| Hypertension | 1,610 (39.6) | 974 (47.9) |
| Diabetes mellitus | 435 (10.7) | 179 (8.8) |
| COPD | 136 (3.3) | 47 (2.3) |
| Asthma | 356 (8.8) | 303 (14.9) |
| Nasal polyps | 0 (0.0) | 82 (4.0) |
Definition of abbreviations: BMI = body mass index; COPD = chronic obstructive pulmonary disease; CRS = chronic rhinosinusitis; PM2.5 = particulate matter ⩽2.5 μm in aerodynamic diameter.
Data are shown as mean ± SD or n (%).
The P values for differences of means of proportions comparing the CRS and control groups were calculated using one-way ANOVA for continuous variables and chi-square test for categorical variables.
CRS was more likely to be diagnosed in patients exposed to higher concentrations of PM2.5 across all windows of exposure (e.g., 12 months; OR, 1.29; 95% CI, 1.07–1.55 per 5-μg/m3 PM2.5 increase; Table 2). This association was present across all anatomic locations, particularly for ethmoid sinusitis (OR, 2.90; 95% CI, 2.00–4.21 for each 5-μg/m3 12-month PM2.5 increase). When evaluating severe sinusitis involving all four sinuses, there was a progressive increase in the odds of developing CRS peaking at 36-month exposure (OR, 4.65; 95% CI, 1.37–15.73 for each 5-μg/m3 60-month PM2.5 increase).
Table 2.
Conditional Logistic Regression Analyses for the Association between Long-Term PM2.5 Exposure and Diagnosis of CRS
| Pollution | Model 1* | Model 2† | Model 3‡ |
|---|---|---|---|
| CRS (N = 6,102) | |||
| PM2.5 12-mo | 1.18 (0.99–1.41) | 1.26 (1.05–1.52) | 1.29 (1.07–1.55) |
| PM2.5 24-mo | 1.43 (1.18–1.73) | 1.54 (1.26–1.88) | 1.56 (1.28–1.91) |
| PM2.5 36-mo | 1.25 (1.03–1.53) | 1.33 (1.08–1.63) | 1.35 (1.10–1.66) |
| PM2.5 60-mo | 1.34 (1.08–1.67) | 1.42 (1.13–1.77) | 1.44 (1.15–1.81) |
| Maxillary sinusitis (n = 4,092) | |||
| PM2.5 12-mo | 1.35 (1.09–1.68) | 1.39 (1.11–1.74) | 1.41 (1.12–1.77) |
| PM2.5 24-mo | 1.69 (1.33–2.14) | 1.76 (1.37–2.25) | 1.77 (1.38–2.28) |
| PM2.5 36-mo | 1.50 (1.17–1.91) | 1.54 (1.20–2.00) | 1.55 (1.20–2.01) |
| PM2.5 60-mo | 1.58 (1.21–2.06) | 1.60 (1.21–2.12) | 1.61 (1.22–2.14) |
| Frontal sinusitis (n = 1,644) | |||
| PM2.5 12-mo | 1.18 (0.83–1.69) | 1.30 (0.90–1.88) | 1.28 (0.88–1.87) |
| PM2.5 24-mo | 1.41 (0.96–2.09) | 1.57 (1.05–2.35) | 1.59 (1.06–2.40) |
| PM2.5 36-mo | 1.45 (0.97–2.18) | 1.62 (1.07–2.46) | 1.64 (1.07–2.51) |
| PM2.5 60-mo | 1.26 (0.81–1.97) | 1.40 (0.89–2.22) | 1.43 (0.90–2.28) |
| Ethmoidal sinusitis (n = 1,851) | |||
| PM2.5 12-mo | 2.49 (1.77–3.51) | 2.74 (1.91–3.94) | 2.90 (2.00–4.21) |
| PM2.5 24-mo | 2.81 (1.91–4.13) | 3.09 (2.06–4.63) | 3.39 (2.23–5.15) |
| PM2.5 36-mo | 2.85 (1.92–4.24) | 3.05 (2.01–4.62) | 3.31 (2.15–5.08) |
| PM2.5 60-mo | 2.87 (1.85–4.43) | 3.04 (1.92–4.81) | 3.27 (2.03–5.25) |
| Sphenoidal sinusitis (n = 948) | |||
| PM2.5 12-mo | 1.23 (0.80–1.89) | 1.37 (0.87–2.16) | 1.43 (0.90–2.28) |
| PM2.5 24-mo | 1.33 (0.80–2.20) | 1.50 (0.88–2.56) | 1.61 (0.93–2.77) |
| PM2.5 36-mo | 2.16 (1.26–3.70) | 2.46 (1.40–4.33) | 2.64 (1.48–4.69) |
| PM2.5 60-mo | 1.41 (0.78–2.57) | 1.57 (0.84–2.94) | 1.68 (0.89–3.17) |
| Severe sinusitis§ (n = 369) | |||
| PM2.5 12-mo | 4.00 (1.71–9.34) | 3.49 (1.41–8.67) | 4.07 (1.53–10.82) |
| PM2.5 24-mo | 3.72 (1.39–9.99) | 3.67 (1.24–10.8) | 4.30 (1.36–13.61) |
| PM2.5 36-mo | 6.69 (2.98–15.06) | 6.80 (2.80–16.55) | 7.91 (3.06–20.42) |
| PM2.5 60-mo | 4.06 (1.44–11.48) | 3.98 (1.26–12.55) | 4.65 (1.37–15.73) |
Definition of abbreviations: CRS = chronic rhinosinusitis; PM2.5 = particulate matter ⩽2.5 μm in aerodynamic diameter.
Data are shown as odds ratio (95% confidence interval).
Adjusted for age, sex, and race.
Further adjusted for body mass index, current alcohol consumption status, and current smoking status.
Further adjusted for comorbidity (history of hypertension, diabetes, chronic obstructive pulmonary disease, and asthma).
Indicates concomitant chronic maxillary, ethmoid, sphenoid, and frontal sinusitis.
In this study, we demonstrate that long-term PM2.5 exposure is significantly associated with CRS diagnosis, especially in the ethmoid sinus, which anatomically has the most airflow, and in the most severe cases. Our study provides the strongest evidence to date that PM2.5 exposure is associated with CRS in a well-characterized cohort, consistent with mouse models, supporting a role for long-term PM2.5 exposure in causing eosinophilic rhinosinusitis. In addition, these associations were robust after adjusting for ozone. These data are consistent with prior studies that used less strict inclusion criteria (self-reported CRS), had unclear timing of disease onset, or examined other settings (established disease). For example, Bhattacharyya used the National Health Interview Survey to identify an association between air quality improvements and the declining prevalence of self-reported sinusitis (acute or chronic) (6). Mady and colleagues correlated PM2.5 levels to an increased need for sinus surgery and medication use in patients with established CRS (7). Others have demonstrated that patients with CRS exposed to higher PM2.5 levels had sinonasal biopsies with increased eosinophilic inflammation (8). Soldiers exposed to dust and air pollution in combat zones show increased sinonasal disease rates (9).
Strengths of this study include a large, well-defined patient population (objective testing, otolaryngologist diagnosis); matching strategy for controls, who had clear sinus imaging; adjustment for confounding variables; and assessment of exposure with high spatial and temporal resolution.
There are several limitations. Our retrospective design has inherent weaknesses related to coding accuracy, inability to assess causality, and potential sampling bias. Our PM2.5 exposure model relied on mapping to the patient’s residential address but did not account for other exposures (commuting, work) or indoor exposures. In addition, data on occupational or environmental exposure to allergens or the use of medications were not available. Lastly, subgroup analysis for severe disease was limited by a small sample size, reducing precision in our associations, and may suffer from inadequate matching.
To our knowledge, this is the first study to report that long-term exposure to fine particulate matter air pollution increases the odds of developing CRS, particularly the most severe form of the disease. Ultimately, identification of environmental determinants of inflammatory disease in the upper airway may provide new strategies to mitigate resultant effects and justify public health interventions, such as the use of PM2.5 filtered facemasks, to reduce the enormous burden of this condition. Potential therapies directed at the antioxidant transcription factor, Nrf2, may warrant further investigation based on the previous human in vitro studies demonstrating restoration of ambient PM-induced sinonasal epithelial cell barrier disruption by Nrf2 activation (10).
Footnotes
Supported by National Institute of Allergy and Infectious Diseases grant R01AI143731.
Author Contributions: All authors listed made substantial contributions to the manuscript presented here. Z.Z., M.R., and J.M.P. contributed to data analysis, reporting, and drafting the work. R.J.K., S.B., N.R.L., S.E.L., A.P.L., and V.K.S. contributed to revising and final approval of the work.
Originally Published in Press as DOI: 10.1164/rccm.202102-0368LE on June 28, 2021
Author disclosures are available with the text of this letter at www.atsjournals.org.
References
- 1.Orlandi RR, Kingdom TT, Smith TL, Bleier B, DeConde A, Luong A, et al. International consensus statement on rhinology and allergy: rhinosinusitis 2021 Int Forum Allergy Rhinol 202111213–739.. [DOI] [PubMed] [Google Scholar]
- 2. Beule A. Epidemiology of chronic rhinosinusitis, selected risk factors, comorbidities, and economic burden. GMS Curr Top Otorhinolaryngol Head Neck Surg. 2015;14:Doc11. doi: 10.3205/cto000126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Losacco C, Perillo A. Particulate matter air pollution and respiratory impact on humans and animals. Environ Sci Pollut Res Int. 2018;25:33901–33910. doi: 10.1007/s11356-018-3344-9. [DOI] [PubMed] [Google Scholar]
- 4. Ramanathan M, Jr, London NR, Jr, Tharakan A, Surya N, Sussan TE, Rao X, et al. Airborne particulate matter induces nonallergic eosinophilic sinonasal inflammation in mice. Am J Respir Cell Mol Biol. 2017;57:59–65. doi: 10.1165/rcmb.2016-0351OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Zhang Z, Wang J, Hart JE, Laden F, Zhao C, Li T, et al. National scale spatiotemporal land-use regression model for PM2.5, PM10 and NO2 concentration in China. Atmos Environ. 2018;192:48–54. [Google Scholar]
- 6. Bhattacharyya N. Air quality influences the prevalence of hay fever and sinusitis. Laryngoscope. 2009;119:429–433. doi: 10.1002/lary.20097. [DOI] [PubMed] [Google Scholar]
- 7. Velasquez N, Moore JA, Boudreau RM, Mady LJ, Lee SE. Association of air pollutants, airborne occupational exposures, and chronic rhinosinusitis disease severity. Int Forum Allergy Rhinol. 2020;10:175–182. doi: 10.1002/alr.22477. [DOI] [PubMed] [Google Scholar]
- 8. Padhye LV, Kish JL, Batra PS, Miller GE, Mahdavinia M. The impact of levels of particulate matter with an aerodynamic diameter smaller than 2.5 μm on the nasal microbiota in chronic rhinosinusitis and healthy individuals. Ann Allergy Asthma Immunol. 2021;126:195–197. doi: 10.1016/j.anai.2020.10.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Parsel SM, Riley CA, McCoul ED. Combat zone exposure and respiratory tract disease. Int Forum Allergy Rhinol. 2018 doi: 10.1002/alr.22123. [DOI] [PubMed] [Google Scholar]
- 10. London NR, Jr, Tharakan A, Rule AM, Lane AP, Biswal S, Ramanathan M., Jr Air pollutant-mediated disruption of sinonasal epithelial cell barrier function is reversed by activation of the Nrf2 pathway. J Allergy Clin Immunol. 2016;138:1736–1738. doi: 10.1016/j.jaci.2016.06.027. [DOI] [PubMed] [Google Scholar]