Air pollution exposure is associated with restrictive ventilatory patterns

To the Editor:

Exposure to ambient air pollution is associated with a substantial burden of morbidity and mortality worldwide [1]. In a recent paper, Adam et al. [2] showed significantly impaired levels of forced expiratory volume in 1 s (FEV₁) and forced vital capacity (FVC) associated with exposure to the ambient air pollutants nitrogen dioxide (NO₂) and particles with a 50% cut-off aerodynamic diameter of 10 μm (PM₁₀) in 7613 adults included in the European Study of Cohorts for Air Pollution Effects (ESCAPE). Effect estimates for FVC were of similar magnitude (for NO₂) or larger (for PM₁₀) than those for FEV₁. In line with these findings, Forbes et al. [3] showed negative associations of PM₁₀ and NO₂ with the level of FEV₁ in 40 329 adults included in the Health Survey for England between 1995 and 2001, whereas no significant associations with FEV₁/FVC were observed. In 1997, the Swiss Study on Air Pollution and Lung Disease in Adults (SAPALDIA), including 9651 adults, showed negative associations of ambient air pollutants NO₂ and PM₁₀ with both FEV₁ and FVC [4]. The effect estimates for FVC were stronger than for FEV₁ for various pollutants, and this was consistently the case in most subgroups (according to smoking status and respiratory symptoms). Reduced FVC, with FEV₁ being normal or reduced to a lesser degree than FVC, suggests restrictive rather than obstructive lung disease (in which FEV₁ specifically is reduced, resulting in a low FEV₁/FVC ratio). Thus, findings from several European studies suggest that restrictive rather than obstructive ventilatory patterns associate with long-term low levels of exposure to ambient air pollution. A study with slightly different findings is the German Study on the influence of Air Pollution on Lung Function, Inflammation and Ageing (SALIA), including 2593 women. This study also found negative associations of NO₂ and PM₁₀ exposure with both FEV₁ and FVC, yet the effects estimates for FEV₁ were stronger than for FVC, and consequently there were small significant negative associations with the FEV₁/FVC ratio [5]. A review article concluded that despite biological plausible mechanisms, there is suggestive, but not conclusive evidence that chronic exposure to air pollution is associated with the prevalence and incidence of chronic obstructive pulmonary disease (COPD), a disease characterised by airway obstruction [6]. Thus far, no studies have focused explicitly on whether air pollution exposure is associated with obstructive or restrictive ventilatory patterns.

In the current study we looked explicitly at both obstructive and restrictive ventilatory patterns in relation to air pollution exposure, in the large Dutch population-based LifeLines Cohort Study (air pollution data for >60 000 adults, mainly from the northern Netherlands) [7]. Within LifeLines, each residential address at the baseline visit (2007–2013) was geocoded and annual average exposure to NO₂, PM_2.5, PM₁₀ and PM_2.5 absorbance (PM_2.5abs) was assigned to each address (X,Y coordinate) individually using land-use regression models developed within the ESCAPE study [2]. Associations of air pollutants with lung function levels at the baseline visit (pre-bronchodilator FEV₁, FVC and FEV₁/FVC) were assessed with linear regression models adjusted for sex, height, age, smoking status, pack-years, body mass index (BMI), environmental tobacco smoke exposure, highest level of obtained education, income and asthma. The asthma definition was based on a doctor diagnosis or a combination of symptoms and the use of asthma medication. Effect modification was tested by assessing interactions of air pollution levels with smoking (never/ever), sex, overweight (BMI ≥25 kg·m⁻²), asthma and airway obstruction (FEV₁/FVC <70%).

Complete data were available for 51 855 adults with a median (range) age of 44 (18–89) years, mean FEV₁ 3546 mL (102% predicted), FVC 4597 mL (112% pred) and FEV₁/FVC 77%. Median (range) levels of exposure were 15.7 (8.4–50.8 µg·m⁻³ NO₂, 15.4 (14.8–20.2) µg·m⁻³ PM_2.5, 24.0 (23.7–31.7) µg·m⁻³ PM₁₀ and 0.95 (0.84–2.65) 10⁻⁵ m⁻¹ PM_2.5abs. Exposure levels were highly correlated (Spearman's r 0.49–0.93). Exposure to ambient air pollutants NO₂, PM₁₀ and PM_2.5abs was associated with significantly lower FVC levels (table 1). Exposure to NO₂ was associated with a lower level of FEV₁, whereas no significant associations of the other air pollutants with FEV₁ levels were observed. For all air pollutants the negative (nonsignificant) estimated effects on the level of FEV₁ were smaller than on the level of FVC, which resulted in positive associations with their ratio (FEV₁/FVC). PM_2.5 exposure was not associated with any of the lung function variables investigated.

Associations between PM₁₀ and PM_2.5abs with FVC were generally stronger in females, subjects with BMI ≥25 kg·m⁻², asthma and subjects with airway obstruction (table 1), yet without significant interactions. Associations between NO₂ and FVC were similar in males and females, subjects without and with overweight, never- and ever-smokers, subjects without and with asthma, and without and with airway obstruction.

Importantly, in the current study we confirmed findings from previous European studies suggesting that long-term exposure to low levels of ambient air pollution is associated with restrictive ventilatory patterns [2–4]. Due to the cross-sectional design of the current study it was not possible to pinpoint whether observed associations are due to reduced lung growth, accelerated decline in lung function, or both. In the ESCAPE study, significant associations between ambient air pollutants and FVC and FEV₁ levels were observed, yet there were no associations with annual lung function decline [2]. These observations suggest that associations between air pollution exposure and lung function levels in adulthood are rather reflecting reduced lung growth than accelerated lung function decline. This is in line with observations in children showing reduced age-related increase in FVC associated with air pollution exposure [8, 9]. Moreover, in childhood, estimated effects of exposure to various air pollutants were already found to be larger for FVC than for FEV₁ [10, 11]_. Additionally, decreases in air pollution levels have been associated with increased growth of FEV₁ and FVC in children, with largest effects on FVC [12], and with improvements in FVC, but not in FEV₁, in adults [13].

The current study suggests that associations of PM₁₀ and PM_2.5abs with impaired lung function levels in adults are stronger in females, subjects with higher BMI and with underlying respiratory diseases such as asthma or COPD, whereas such differential effects were not observed for NO₂ exposure. The ESCAPE study found stronger effects of NO₂ exposure on FEV₁ and FVC in females, subjects with asthma and in obese subjects, with the latter showing significant interaction with NO₂ [2]. In general, and in line with our observations, most studies thus far have not observed clear differences in effect estimates between smokers and nonsmokers [2, 3, 5]. Identifying susceptible subgroups in the population may be important from a public health perspective, e.g. exposure limits may be determined based on the most susceptible subgroups. Moreover, identifying such (genetically) susceptible subgroups may pinpoint potential (biological) mechanisms underlying observed associations [14], which will give rise to further studies in experimental settings.

Restrictive ventilatory patterns with exposure to ambient air pollution may reflect pulmonary fibrosis, a disease which has thus far largely been understudied in relation to air pollution exposure in large-scale epidemiological studies [15]. Yet, further studies are warranted that include additional measurements, such as total lung capacity, in subsets of individuals with low FVC, to confirm true restrictive ventilatory effects of air pollution exposure.

To conclude, low levels of ambient air pollution are associated with restrictive ventilatory patterns. Both epidemiological and experimental studies are warranted to confirm true restrictive ventilatory effects of air pollution exposure and to elucidate underlying biological mechanisms leading to a disease that is largely understudied in relation to air pollution exposure, i.e. pulmonary fibrosis. Identifying susceptible subgroups and underlying biological pathways may contribute to lowering the substantial burden of respiratory disease related to ambient air pollution exposure.