Article Text

Environment
Air pollution and exhaled nitric oxide in Dutch schoolchildren
  1. Haitske Graveland1,
  2. Sofie A H Van Roosbroeck2,
  3. Wilhelmina M Rensen3,
  4. Bert Brunekreef1,4,
  5. Ulrike Gehring1
  1. 1Institute for Risk Assessment Sciences, Utrecht University, Utrecht, The Netherlands
  2. 2Department of Epidemiology and Social Medicine, University of Antwerp, Antwerp, Belgium
  3. 3CNR - Institute of Molecular Biology and Pathology, c/o Department of Genetics and Molecular Biology, Sapienza University, Rome, Italy
  4. 4Julius Centre for Health Sciences and Primary Care, University Medical Centre Utrecht, Utrecht, The Netherlands
  1. Correspondence to Ulrike Gehring, Institute for Risk Assessment Sciences, Utrecht University, Jenalaan 18 d, 3584 CK Utrecht, The Netherlands; u.gehring{at}uu.nl

Abstract

Background Short-term changes in air pollution exposure in children may be associated with transient increases in exhaled nitric oxide (NO), a marker of airway inflammation. Also, children living in areas with high air pollution levels and/or high traffic densities appear to have chronically increased levels of exhaled NO. No studies have simultaneously addressed the long-term and short-term associations between traffic-related air pollution and exhaled NO.

Objectives To investigate associations between exhaled NO in school children and both short-term changes in outdoor PM10 and long-term traffic exposures.

Methods Offline exhaled NO measurements were conducted in 812 children from nine Dutch schools within 400 m of motorways. Daily levels of particulate matter with a 50% cut-off aerodynamic diameter of 10 μm (PM10) were obtained from background monitoring stations. Long-term exposure to traffic-related air pollution was assessed using specific traffic-related characteristics such as total, car and truck motorway traffic and the distances of the children's homes and schools from the motorway.

Results A positive association was found between ambient PM10 concentrations on the day of exhaled NO measurement and exhaled NO (adjusted geometric means ratio (95% CI) 2.24 (1.37 to 3.65)) over the range of daily PM10 concentrations of 44 μg/m3), which was largely attributable to a pollution peak associated with high particulate matter emissions from traditional Easter fires. There were suggestive associations between exhaled NO and traffic counts only in children with asthma, which were not statistically significance.

Conclusions Short-term changes in ambient PM10 largely attributable to biomass burning are associated with increased levels of exhaled NO.

  • Air pollution
  • children
  • exhaled nitric oxide
  • particulate matter
  • traffic
  • epidemiology

Statistics from Altmetric.com

What this paper adds

  • Evidence is accumulating that short-term changes in air pollution exposure are associated with transient increases in exhaled nitric oxide (NO), a marker of airway inflammation, in children.

  • There is also some evidence that children living in areas with high air pollution levels and/or high traffic densities have chronically increased levels of exhaled NO.

  • No studies so far have simultaneously addressed the long-term and short-term associations between traffic-related air pollution and exhaled NO.

  • We investigated simultaneously the long-term and short-term association between traffic-related air pollution and exhaled NO among Dutch children attending nine schools located close to motorways with varying densities and compositions of traffic and found short-term changes in ambient PM10, but not long-term exposures to traffic to be associated with increased levels of exhaled NO.

Introduction

Airway inflammation can be detected by measuring exhaled nitric oxide (NO) a non-invasive method suitable for use in children and in large epidemiological studies. Exhaled NO is increased in children1–3 and adults4–6 with asthma. Recently there has been increased use of exhaled NO measurements in air pollution studies.7–19 From these studies, evidence is accumulating that short-term changes in air pollution exposure are associated with transient increases in exhaled NO in children.7–16 There is also some evidence that children living in areas with high air pollution levels and/or high traffic densities have chronically increased levels of exhaled NO.17–19 No studies so far have simultaneously addressed the long-term and short-term associations between traffic-related air pollution and exhaled NO.

The aim of the present study is to investigate simultaneously the long-term and short-term associations between traffic-related air pollution and exhaled NO among Dutch children attending nine schools located close to motorways with varying densities and compositions of traffic.

Methods

Study design and study population

Between January and April 2005, we studied the respiratory health of children attending nine schools located within 400 m of motorways in the Netherlands by inviting all children in grades 3–7 (generally between 7 and 11 years of age) to participate in the study. Participation included a standardised questionnaire on asthma, allergies and related symptoms, and the measurement of exhaled NO performed at the school. The study protocol was approved by the medical ethics committee of Utrecht University.

We approached the parents of 1661 children with a written request for participation. The parents of 1215 children agreed to participate and returned the questionnaire; 1113 (92%) of these children participated in the measurement of exhaled NO. For the present analysis, we restricted our study population to those children who participated in the exhaled NO measurements. As in a previous similar study,10 only children who lived within 1000 m of the motorway (n=879) were included. The rationale for selecting schools within 400 m and homes within 1000 m of a motorway is that air pollution from motorway traffic typically reduces to background levels within the first 500–1000 m. Since the emphasis in our study was on contrasts between motorways, we needed to make sure children were included only when it could be assumed they were actually exposed to the pollution from the motorway close to their schools and homes. After excluding children with missing data for one or more confounders (n=67), 812 children remained for the present analysis.

Selection of schools

We identified city districts in the Netherlands that had a primary school within 400 m of a motorway and approached 25 schools for participation. When we selected the schools, we aimed for a maximum variation in traffic densities and a minimal correlation between car and truck traffic. We aimed for minimal correlation between car and truck traffic so we could distinguish between the effects of car traffic and the effects of truck traffic. Nine schools agreed to participate. Reasons for non-participation were mostly non-specific (reorganisations, recently involved in other studies, too busy with other school projects, etc). The schools were located along seven different motorway stretches.

Measurements of exhaled NO

On 13 different days in March and April 2005, offline measurements of exhaled NO were carried out at the schools with the balloon method according to recent European Respiratory Society/American Thoracic Society guidelines.20 All measurements were performed in the morning between 09:00 and 12:00 h. Children were seated and were asked to take a deep breath and inhale through a charcoal NO scrubber, to exclude the effect of NO in ambient air. Children were then asked to exhale immediately into a collection device employing dynamic flow restriction, using a two-way valve. Mouth pressure was monitored during measurement using a manometer. After discarding dead space air for 3–4 s, exhaled air was collected in a NO impermeable 150 ml Mylar balloon (Jurjen de Vries, Leeuwarden, the Netherlands). Balloons were sealed, stored and analysed within 6 h with a Sievers NOA 280 analyser (Boulder, Colorado, USA; sensitivity 0.5 ppb, detection range 0.5–500 000 ppb, sample flow 200 ml/min). In approximately 10% of the children a repeated sample was taken directly (within 1–2 min) after the first measurement and in approximately 5% of the children a repeated sample was taken 1 month after the first measurement. Correlation between the first measurements and these repeated measures was 0.75 (n=107) and 0.67 (n=53), respectively, indicating reasonable repeatability.

Questionnaire

We used the International Study of Asthma and Allergies in Childhood questionnaire to measure asthma, hay fever, eczema, bronchitis and related symptoms.21 The questionnaires also included questions on potential confounders such as the child's sex, age and nationality, parental asthma and hay fever, parental education, and various exposures such as current parental smoking at home, current pet ownership, the use of natural gas for cooking, and the presence of mould stains in the home.

Traffic characteristics

We used traffic characteristics such as traffic counts and distances of the children's homes and school addresses from the motorways as markers of long-term exposure to traffic. Traffic counts were obtained from the Ministry of Public Works. These were routinely collected counts for all motorways in the Netherlands using induction loops. Vehicles shorter than 5.1 m were defined as car traffic, while vehicles longer than 5.1 m were defined as truck traffic. We used the most recent data that were available for this study, that is data for 2001, and defined exposure to car and truck traffic as the car and truck traffic counts, respectively, on the motorway section closest to the children's school. The emphasis of this study was on contrasts between motorways and therefore roads other than motorways were not included in the exposure assessment. Distances of the children's homes and schools from the nearest motorway were assessed using the geographic information system ArcGIS (Environmental Systems Research Institute, Redlands, California, USA).

Ambient air pollution levels

We used data from four (urban) background sites of the National Air Quality Monitoring Network that were located close to the schools to estimate particulate matter with a 50% cut-off aerodynamic diameter of 10 μm (PM10) levels on the day of the exhaled NO measurements and the preceding 3 days. Figure 1 of the online supplement shows the location of the monitoring sites and the schools.

Data analyses

We used linear mixed models to investigate the effect of traffic characteristics and ambient air pollution levels on log-transformed levels of exhaled NO. We included measurement day as a random effect to account for the correlation between measurements that were performed on the same day and controlled for school characteristics such as downwind location (yes/no), measurement-day specific variables such as indoor NO, and individual covariates such as the child's sex, age (in years, used as continuous variable) and nationality, parental asthma and/or hay fever, parental education, and various exposures such as current parental smoking at home, current pet ownership, the use of natural gas for cooking, and the presence of mould stains in the home. All associations were presented for a maximum–minimum difference change in independent variables. Statistical significance was defined by a two-sided α level of 5%. All calculations were performed using SAS v 9.1.

Results

Characteristics of the study population

Characteristics of the study population and the prevalences of asthma, allergies and related symptoms are presented in table 1 of the online supplement. The study population consisted of slightly more girls than boys. Children were on average 8.4 years old, and the majority of the children had Dutch nationality. The geometric mean (GM) (geometric SD, GSD) of the exhaled NO levels was 12.4 ppb (1.7).

Table 1

Distribution of traffic characteristics and PM10 levels for the nine schools

Traffic characteristics, outdoor PM10 and indoor NO levels

Table 1 shows the distribution of the traffic characteristics as well as the outdoor PM10 and indoor NO levels on the days of the exhaled NO measurements for the nine schools. Correlations between traffic characteristics and PM10 levels are presented in table 2. Car traffic and truck traffic intensities were highly correlated with each other and with total traffic intensity. Correlations between all other variables were moderate or low.

Table 2

Spearman correlations between different traffic characteristics, outdoor PM10 and indoor NO

Effects of traffic characteristics and PM10 on exhaled NO

We found a statistically significant positive association between PM10 levels on the day of the exhaled NO measurements and exhaled NO levels before and after confounder adjustment (table 3). Associations with PM10 levels on the 3 days preceding the day of the exhaled NO measurements were very similar to associations with the PM10 levels on the day of the exhaled NO measurements with adjusted geometric means ratios ranging from 1.96 to 2.16 (table 3). There were positive associations between exhaled NO and total traffic, car traffic and the distance of the school from the motorway, which did not reach statistical significance.

Table 3

Crude and adjusted geometric means ratios (GMR) with 95% CIs for the associations between traffic characteristics and PM10 levels and exhaled NO

Sensitivity analyses

Ambient NO could have affected exhaled NO measurements in two ways. Environmental NO reached high levels relative to those in exhaled breath, which could have resulted in contamination of the biological samples with ambient NO.20 Furthermore, increased levels of ambient NO could have resulted in a reduction in exhaled NO due to downregulation of nitric oxide synthases by a negative feedback mechanism.22 Using NO-scrubbing filters may have prevented some, but not all, of the effect of ambient NO on measured exhaled NO. We therefore adjusted all analyses for the indoor NO concentrations during the exhaled NO measurements. The association with PM10 became slightly smaller but remained statistically significant (table 3).

During the study period, a pollution peak occurred which was associated with high particulate matter emissions from traditional Easter fires, located mostly in the east of the country. The time series of the PM10 levels during the study period together with the dates of the exhaled NO measurements and the date of the Easter fires is shown in figure 1. We evaluated to what extent the results were driven by the PM10 peak associated with the Easter fires, which may have especially affected the exhaled NO measurements in school 5, which were performed 4 days after the Easter fires. We therefore repeated all analyses without school 5, which resulted in a reduction in the range in PM10 concentrations of almost one-third. The PM10 effect largely disappeared after exclusion of school 5, whereas the associations with total traffic and car traffic remained elevated but statistically non-significant (table 3).

Figure 1

Time-series of daily PM10 levels (μg/m3) at the four monitoring sites used for estimation of background PM10 levels at the different schools. Black vertical lines indicate the dates on which exhaled NO was measured.

Effect modification by asthma

Children with asthma had significantly higher levels of exhaled NO than those without asthma (GM (GSD) 14.45 (1.8) vs 12.1 (1.6)). We explored whether children with asthma were more sensitive to the effects of traffic and PM10 by performing stratified analyses. Results are presented in figure 2. PM10 effects were basically identical in children with and without asthma. Associations with traffic intensities were slightly stronger in those with asthma compared to those without asthma. None of the associations, however, was statistically significant.

Figure 2

Adjusted* geometric means ratios with 95% CIs for the associations between traffic characteristics and PM10 levels and exhaled NO in children with and without asthma. *All effects were adjusted for individual confounders (sex, age, current parental smoking, current pet possession, parental education level, non-Dutch nationality, gas cooking, parental allergy, presence of mould stains in the home) and downwind location. Effects of traffic characteristics were additionally adjusted for outdoor PM10 on the day of exhaled NO measurements; effects of PM10 were additionally adjusted for total traffic and distance of the school from the motorway.

Discussion

This study showed that exhaled NO in school children was associated with day-to-day changes in ambient PM10, but not with traffic characteristics of the school or the distance of the child's home from the motorway. In children with asthma we found statistically non-significant positive associations between traffic counts on the motorway in the proximity of the school and exhaled NO levels.

Several studies have documented that children with asthma have higher exhaled NO levels just after days with increased air pollution levels.7–11 Further support for an association between short-term changes in air pollution levels and exhaled NO comes from a study that found that exhaled NO levels decreased significantly within 1 week of relocating from a polluted urban environment to a clean rural environment in a group of allergic children with asthma.12 One Canadian study among children with asthma did not find associations between exhaled NO and air pollutants measured in the days preceding the exhaled NO measurements.23 Fewer studies have included children without asthma. One study found acute increases in exhaled NO in healthy children attending summer camps in Belgium, in response to elevated ozone concentrations.13 In a small study from Seattle, ambient PM2.5 was found to be associated with exhaled NO in children with and without asthma.24 Two studies from the Netherlands found positive associations between exhaled NO and daily air pollution levels in sensitised and non-sensitised children14 and in a random sample of school children.15 A very recent French study16 found significant positive associations between 5-day average school yard and classroom PM2.5 levels and exhaled NO in both children with and without asthma. In the present study, no data on allergic sensitisation were available. The association between exhaled NO and PM10 on the previous day was very similar for children with and without asthma.

The effect of PM10 in the present study was largely attributable to a pollution peak associated with high particulate matter emissions from traditional Easter fires, located mostly in the east of the country. Acute effects of pollution peaks due to biomass burning were found in two other studies. A Californian study found significant positive associations between peak PM10 emissions from wildfires and nose, eye and throat irritation; cough, bronchitis, wheezing and asthma attacks; medication use; and physician visits in elementary and high-school children.25 Another study found significant positive associations between total suspended particle concentration peaks from preharvest sugarcane burning and hospital admissions in Araraquara, São Paulo State, Brazil.26

In the present study, before and after adjustment for ambient PM10 levels on the days preceding the exhaled NO measurements, we did not find significant associations between measures of traffic density on the nearby motorways and exhaled NO. In contrast, Dales and co-workers17 found an association between various measures of traffic density near the homes of school children living in Windsor, Ontario, after taking into account air pollution on the days preceding the exhaled NO measurements. We did not obtain traffic density measures near the homes of the study children, but this may be less important in a study where pollution exposure is likely dominated by attending school near a busy motorway. Holguin et al18 found an association between exhaled NO and traffic measures near the home only in school children with asthma, without finding a association with school-based measurements of NO2 and PM2.5. In a study from France,19 children with asthma living in a clean environment were found to have exhaled NO levels which were significantly lower than in those living in an urban, more polluted area. Our results do not conflict with other studies suggesting a relationship between exhaled NO and measures of long-term exposure to air pollution, especially in children with asthma.

All measurements were performed between January and April 2005. This means that all measurements were more or less performed in the same (winter) season. Nevertheless, seasonal confounding may be a concern. We adjusted long-term traffic effects for indoor NO and outdoor particulate matter on the same day to control for seasonal differences in these parameters. However, we cannot rule out completely that seasonal variation in factors other than these two may have biased our results.

A potential limitation of our study may be that we used traffic counts from the year 2001, which are the most recent data that were available for this study, which was carried out in 2005. However, traffic density and composition on motorways change only gradually over time. Beelen et al27 showed that traffic intensity data were highly correlated over a 10-year period. Since the emphasis in our study was on contrasts between motorways, traffic other than motorway traffic was not included. Since air pollution from motorway traffic typically reduces to background levels within the first 500–1000 m, we selected schools within 400 m and homes within 1000 m of a motorway to make sure children were included only when it could be assumed they were actually exposed to pollution from a motorway close to their schools and homes. We defined exposure to motorways by the distance of the schools and homes from the motorway and the traffic intensity on the motorway section closest to the children's school. Due to the proximity of both the children's school and their home addresses to the motorway and to each other, we believe that for the participants in the present study (variation in) traffic intensities at the school address is a good estimate for the (variation in) traffic intensities at the home address. It is therefore not possible to distinguish the effects of exposure at home from the effects of exposure at school in the present study.

A limitation of using surrogate variables for exposure to outdoor air pollution originating from traffic is that they have not been directly validated for their use as exposure measures in epidemiological studies.28 During the last decade, the use of more sophisticated geographic information system-based methods for air pollution exposure assessment such as interpolation methods (eg, kriging, inverse distance weighing), dispersion modelling and land-use regression modelling has become increasingly popular for taking into account the within-city variability of air pollution concentrations in long-term effect studies.29 The use of one of these methods, which were not available in the present study, resulting in more accurate exposure estimates may have possibly led to clearer results.

Another imitation of the present study is the lack of atopy data, which made it impossible to evaluate the impact of atopy on the association between air pollution and exhaled NO. Furthermore, no information on the use of inhalation corticosteroids is available. Inhaled steroids are known to suppress exhaled NO,20 which may have confounded the association between exhaled NO and the exposures studied.

Our study adds to the as-yet small body of evidence that shows that acute changes in exhaled NO occur not only in children with asthma but also in unselected children in response to elevated ambient particulate matter levels.

Conclusion

Short-term changes in ambient PM10 largely attributable to biomass burning are associated with increased levels of exhaled NO.

References

Supplementary materials

Related Data

Footnotes

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the medical ethics committee of Utrecht University.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.