Asthma in children exposed to nitrogen dioxide in ice arenas

Area and arenas

In the planning of this study a survey of all 286 Swedish indoor ice arenas was made in 1998. This showed that two thirds of the arenas used propane-powered ice-resurfacing machines. Based on this survey a region in the central part of Sweden with a large number of indoor ice arenas was selected. The region consists of small communities and rural areas except for the city of Gävle, which has ∼67,000 inhabitants. Thus, 15 arenas at 14 different locations were included, nine using propane-powered and six using electric-powered machines. Two of the electric-powered arenas were located in the city of Gävle.

Study population

All children in hockey and figure-skating clubs in Sweden are registered from the age of 10 in a mandatory insurance file. From this file it was possible to identify 2,231 children, aged 10–16 yrs who were presently active in clubs using the selected arenas or had been active during any of the 3 yrs prior to the study. In the winter of 1999 a questionnaire was sent to the selected children and their parents. The response rate following two postal reminders was 77.8%, similar for children attending electric-powered and propane-powered arenas. One hundred and nine of the children who attended electric arenas reported that their main activity was on outdoor rinks and were excluded from analysis. Furthermore, children in electric arenas with mixed exposure to both types of resurfacing machines were also excluded, leaving a total of 1,535 children for the final analysis.

Questionnaire and health outcome measurements

The questionnaire had 57 questions, including 10 on symptoms of atopic diseases identical to those in the International Study of Asthma and Allergies in Childhood (ISAAC-1) for children aged 13–14 yrs 9. Furthermore, questions on “physician-diagnosed asthma” and asthma medication were added. To estimate the exposure to NO₂, questions on physical exercise in ice arenas, training intensity and use of more than one arena were included. Questions on living conditions, household pets (past and present), household smoking (past and present) and parental history of allergic disease were also included. The parents were given instructions to complete the questionnaire together with their child.

Nitrogen dioxide monitoring

At the time of distribution of the questionnaires ambient NO₂ was measured inside and outside all the selected arenas. In addition to the propane- or electric-powered resurfacing machines, all arenas used ice edgers powered by combustion engines. The resurfacing machines were used daily on average ∼1 h and the edgers were usually used once or twice a week for ∼15 min. The measurements were started on the day before the “special ice-care day”, when the resurfacing machine was operated for a longer time and the edger was used. NO₂ concentrations were determined by passive diffusion samplers at three different locations in each arena during 3 consecutive days. Each sampler was in use for 1 day only. At each arena two samplers were located near the rink (goal and rink-side) and one outside.

The samplers were in position during open hours, usually 13 h·day⁻¹. To control for passive diffusion of NO₂ into the sampler when not used, a control sampler not removed from its sealed tube was used during all 3 days in each arena. The sampler was developed by the Swedish Environment Research Institute and gives time integrated NO₂ concentrations 10. The relative sd between duplicate samplers has been estimated as <4% and the concentrations are within ±15% of those obtained by volumetric techniques.

Statistical analysis

Comparisons between children exposed to propane-powered and electric-powered resurfacing machines were performed by estimating odds ratios (OR) and 95% confidence intervals (CI), using logistic regression analysis. Adjustment was made for sex and family history of allergic disease. Other potential confounders, such as keeping furred pets, a smoking mother or condensation on window panes during wintertime, did not influence the OR and were not included in the regression models. Individuals not responding to a particular question were excluded from analysis of that question. However, missing values were ≤1% in any single question.

Ethical considerations

The Ethics Committee of Karolinska Institute, Stockholm, Sweden, approved the study and all families gave informed consent before participation.

Exposure and demographic data

Of the 1,535 children in the final analysis, 969 practised in propane arenas and 566 in electric arenas. The study group included 1,470 (95.8%) males and 65 (4.2%) females, with an average age of 13 yrs (range 9–16). Data on exposure to NO₂, demographic characteristics and potential risk factors for allergy among the participating children are presented in table 1⇓. The age and sex distribution as well as training history were similar in the two groups of children. Most children had >3 yrs of activity in the ice arenas and 79% (propane arenas) and 76% (electric arenas) were still active in the winter of 1999 when the present study was performed. Of the 328 children who had stopped sport activities in ice arenas during the last 3 yrs, change of sports was reported as the main reason (78%). This was equally distributed between the two groups. However, 10 children reported breathing problems when attending ice arenas (four propane, six electric) as the major cause. There was no significant difference in potential risk factors for bronchial asthma between groups, such as housing conditions, indoor dampness, parental smoking (past or present) or keeping furred pets (past or present). However, children in the electric-arena group tended to have a higher prevalence of parental history of asthma and atopy (41.9%) than children in the propane-arena group (38.7%).

Nitrogen dioxide monitoring

The results from NO₂ measurements are presented in table 2⇓. There was a wide range of NO₂ concentrations in the nine propane arenas (28–1,015 µg·m⁻³). The measurements were consistent within the arenas, i.e. the range of NO₂ was small in each arena. The same arena had the lowest concentrations on all 3 consecutive days and another arena had the highest levels during all 3 days. The mean and median values were higher on the second and third day in most arenas due to longer use of resurfacing and ice-edger machines on the second or third day. The mean indoor concentrations were >20-fold higher than the mean outdoors when propane-powered resurfacing machines had been used. The highest NO₂ concentrations in the electric arenas were similar to the lowest in the propane arenas. Outdoor NO₂ concentrations tended to be similar to those inside the electric arenas.

Respiratory symptoms and allergic disorders

Overall, “ever asthma” was reported by 16% of the children and physician-diagnosed asthma by 14%. Asthma medication in the past year was used by 16%. Male children tended to have a higher prevalence of “ever asthma” (M=16%, F=11%) and physician-diagnosed asthma (M=18%, F=9%). Signs of lower and upper respiratory allergic disease, “ever asthma”, physician-diagnosed asthma and asthma medication in the past year were as prevalent in the group of children attending ice arenas with propane-powered machines as in the group of children attending ice arenas with electric machines (table 3⇓). Most children with “ever asthma” had their onset of asthma before starting hockey training or figure skating in ice arenas (63% attending propane arenas, 58% attending electric arenas). When these children were excluded there was still no difference in symptom prevalence between the two exposure groups. Remission of asthma, expressed as “ever asthma” but without current symptoms of asthma and/or asthma medication during the past year, was also similar among children practising in electric arenas (16%) and in propane arenas (13%).

In order to further assess the effect of NO₂ exposure, children who had attended propane arenas were divided into two exposure groups based on the results of the NO₂ measurements (table 4⇓). Four of the propane arenas had NO₂ concentrations above the median of NO₂ (mean 510 µg·m⁻³, median 396 µg·m⁻³). The remaining nine arenas had a mean concentration of 89 µg·m⁻³ and a median concentration of 65 µg·m⁻³. Thirteen subjects attended propane arenas without known concentrations of NO₂ and are excluded from this subanalysis. Nasal symptoms, without upper respiratory infection and “ever wheeze” were more prevalent among children with high exposure levels of NO₂. There was also a tendency to increased prevalence of “ever asthma”, physician-diagnosed asthma and symptoms of asthma in the high exposure group. The onset of asthma before beginning to attend ice arenas regularly was similar in the two exposure groups (low 63%, high 62%). There was no significant difference in years of indoor hockey activities, training intensity and/or potential risk factors for development of atopic disease among children in the high and low NO₂-exposure groups.

The study area mainly consisted of small communities and rural areas except for the city of Gävle. To minimise potential effects of other sources of air pollution, further analysis was made after exclusion of the 102 children exercising in the two electric arenas in Gävle. However, the results remained unchanged. The present study was conducted during February and March, i.e. before the pollen season. Exclusion of children reporting symptoms related to furred pets and/or pollen also did not change the results.