The Association of Physical and Biological Factors of Outdoor Air with Exercise-Induced Bronchoconstriction in Children

Background: Outdoor free-running exercise test is commonly used for diagnosing asthma in children, but the real-life influence of outdoor environmental factors on test outcomes remains unclear. While temperature and humidity are known to associate with exercise-induced bronchoconstriction (EIB) in adults, the generalization of the results to children is uncertain. Although exposure to pollens or air pollutants worsens asthma symptoms, it is not known how much they affect the outcome of free-running exercise test outdoors.

Aim: The aim of our study was to identify which outdoor air physical factors are associated with a reduction in lung function after exercise and whether the study results are different in younger (lung function measured with forced oscillation technique (FOT)) and older (lung function measured with spirometry) children. We also aimed to determine which factor of outdoor air is the most powerful in affecting the severity and incidence of EIB. We also aimed to examine whether there is an increased incidence of EIB in pollen season and whether elevated pollution concentrations are associated with the increased incidence or more severe EIB.

Methods: We analyzed data from 868 reliable exercise challenge tests with FOT measurements conducted in preschool children between January 2012 and April 2015, as well as 321 exercise tests with spirometry in children aged 6-16 years between January 2012 and December 2017 at Tampere University Hospital. Children were tested because of suspicion of asthma or a need to assess the efficacy of treatment in persistent asthma. We examined the relationship between post-exercise lung function measured with FOT or spirometry with outdoor air factors temperature, relative humidity, and absolute humidity through regression analysis. We also compared the occurrence of EIB defined as a decrease in forced expiratory volume in one second (FEV1) of ≥ 15% in spirometry or as a ≥40% increase in respiratory resistance measured with FOT at 5 Hz (R5) at different levels of temperature, relative humidity, and absolute humidity. For children who underwent FOT tests (n=799) between January 2012 and December 2014, we also had data on pollen concentrations of outdoor air. We examined the relationship between EIB occurrence and pollen concentrations. Furthermore, we analyzed pollutant concentrations (PM2.5, NO2, and O3) of outdoor air at the time of the exercise tests and examined their association with the occurrence and severity of EIB, adjusting analysis for air humidity and pollen counts. The data on air humidity, temperature, pollen, and pollution levels were obtained from public registers.

Results: In regression analysis, the relative change in R5 after exercise was related to temperature (regression coefficient (β)=-0.22, p=0.020) and absolute humidity (β=- 0.89, p=0.002), but not to relative humidity. There were also more EIB reactions if absolute air humidity was <5 g/mᶟ in comparison to ≥10 g/mᶟ (p=0.008). Only absolute humidity was independently associated with change in airway resistance (p=0.009) in multivariable regression analysis in younger children with FOT measurements. In older children with spirometry, absolute humidity ≥5g/m3, but not relative humidity or temperature, was associated with the occurrence of EIB (p=0.035). In multivariable logistic regression, absolute humidity ≥5g/m3 was negatively associated with EIB (OR 0.51, p=0.026). Absolute humidity (Pearson correlation coefficient (r)= -0.12, p=0.028) and temperature (r= -0.13, p= 0.023) correlated with decrease in FEV1. In multivariable linear regression, only absolute humidity was associated with FEV1 decrease (β= -0.04, p=0.033). Increase in R5 after exercise or the frequency of EIB were not related to alder or birch pollen concentrations over 10 grains/m3 (p>0.1). Increase in R5 after exercise did not correlate with O3, NO2 or PM2.5 concentrations (p values 0.065-0.884). In multivariable logistic regression, we compared the effects of PM2.5 ≥ 10 µg/m³, absolute humidity ≥ 10 g/m³ and alder or birch pollen concentration ≥ 10 grains/m³. In this analysis, absolute humidity ≥ 10 g/m3 was associated with decreased incidence (OR 0.31, p value 0.004), and PM2.5 ≥10 µg/m³ was associated with increased incidence (OR 1.69, p value 0.036) of EIB.

Conclusions: We showed in a large real-life clinical sample that high absolute humidity is associated with lower incidence of EIB in children, and the results were similar if lung function was measured with FOT or spirometry. The commonly interpreted association between cold air and EIB seems to be entirely explained by the low absolute humidity of cold air and temperature is not independently explaining EIB. A negative test result at high absolute humidity should be interpreted with caution. In addition, outdoor air PM2.5 levels may influence EIB in children, but pollen concentration was not associated with the probability of EIB.