Research Article | Published:

Effects of ultrafine and fine particulate and gaseous air pollution on cardiac autonomic control in subjects with coronary artery disease: The ULTRA study

Journal of Exposure Science and Environmental Epidemiology volume 16, pages 332341 (2006) | Download Citation

Subjects

Abstract

Previous studies have shown an association between elevated concentrations of particulate air pollution and cardiovascular morbidity and mortality. Therefore, the association between daily variation of ultrafine and fine particulate air pollution and cardiac autonomic control measured as heart rate variability (HRV) was studied in a large multicenter study in Amsterdam, the Netherlands, Erfurt, Germany, and Helsinki, Finland. Elderly subjects (n=37 in Amsterdam, n=47 in both Erfurt and Helsinki) with stable coronary artery disease were followed for 6 months with biweekly clinical visits. During the visits, ambulatory electrocardiogram was recorded during a standardized protocol including a 5-min period of paced breathing. Time and frequency domain analyses of HRV were performed. A statistical model was built for each center separately. The mean 24-h particle number concentration (NC) (1000/cm3) of ultrafine particles (diameter 0.01–0.1 μm) was 17.3 in Amsterdam, 21.1 in Erfurt, and 17.0 in Helsinki. The corresponding values for PM2.5 were 20.0, 23.1, and 12.7 μg/m3. During paced breathing, ultrafine particles, NO2, and CO were at lags of 0–2 days consistently and significantly associated with decreased low-to-high frequency ratio (LF/HF), a measure of sympathovagal balance. In a pooled analysis across the centers, LF/HF decreased by 13.5% (95% confidence interval: −20.1%, −7.0%) for each 10,000/cm3 increase in the NC of ultrafine particles (2-day lag). PM2.5 was associated with reduced HF and increased LF/HF in Helsinki, whereas the opposite was true in Erfurt, and in Amsterdam, there were no clear associations between PM2.5 and HRV. The results suggest that the cardiovascular effects of ambient ultrafine and PM2.5 can differ from each other and that their effect may be modified by the characteristics of the exposed subjects and the sources of PM2.5.

Introduction

There is growing epidemiological evidence on adverse effects of particulate air pollution on cardiovascular health. These effects include increased hospital admissions and mortality (Schwartz, 1999; Samet et al., 2000; Brunekreef and Holgate, 2002). The pathophysiological link between air pollution and the cardiovascular system is not known, but it has been proposed that high particle concentrations, particularly in the ultrafine range, are able to provoke alveolar inflammation, which in turn might release mediators capable of exacerbating lung disease and increasing blood coagulability in susceptible individuals (Seaton et al., 1995). Alternatively, it has been hypothesized that an increase in air pollution levels could modify autonomic nervous control of the heart in subjects with existing cardiovascular disease and thus lead to increased morbidity and mortality (Watkinson et al., 1998; Stone and Godleski, 1999; Godleski et al., 2000). In addition, it has been shown that inhaled ultrafine particles diffuse rapidly into the systemic circulation, and this process could account for extrapulmonary effects of air pollution (Nemmar et al., 2002).

Heart rate variability (HRV) is a measure of the autonomic control of the heart. A decreased overall HRV is a predictor of mortality in subjects with prior myocardial infarction (Tsuji et al., 1994; Task Force, 1996). The results of previous smaller studies conducted among elderly suggest an association of particulate air pollution with a decrease in overall HRV, but the results for short-term components of HRV are somewhat inconsistent (Liao et al., 1999; Pope et al., 1999, 2004; Gold et al., 2000; Creason et al., 2001; Devlin et al., 2003; Holguin et al., 2003). In these studies, particulate air pollution has been measured as PM2.5, PM10, or concentrated ambient air pollution particles. No previous study has investigated effects of ultrafine particles on HRV.

The aim of the present multicenter study was to investigate effects of air pollution, especially ultrafine particulate air pollution, on HRV in elderly subjects with stable coronary artery disease in three European cities. We hypothesized that an increase in daily concentrations of particulate air pollution would be associated with decreased HRV, especially with decreased overall variability and vagal input. The study was conducted in the framework of the EU-funded ULTRA (Exposure and risk assessment for fine and ultrafine particles in ambient air) study (Pekkanen et al., 2000).

Methods

The ULTRA study was carried out in three European cities: Amsterdam, The Netherlands, Erfurt, Germany, and Helsinki, Finland. The study period was in Amsterdam (3 November 1998–18 June 1999), in Erfurt (14 October 1998–4 April 1999), and in Helsinki (2 November 1998–30 April 1999). The study protocol was approved by ethical committees in each study center. All subjects gave a written informed consent.

All methods used in the ULTRA study were conducted according to Standard Operating Procedures (SOP) developed for the ULTRA study (Pekkanen et al., 2000). In each city, a panel of subjects with coronary artery disease was followed up for 6 months with biweekly clinical visits and daily symptom diaries. The clinical visit included an ambulatory electrocardiogram (ECG) recording during a standardized protocol (5-min periods of rest in supine position, rest in supine position with paced breathing (frequency of 0.2 Hz, 2.5 s inhalation and 2.5 s exhalation), standing, a 6-min light exercise with a bicycle ergometer with a target heart rate at 90–100/min, and a 10 min recovery in supine position). The detailed description of the methods has been published (Pekkanen et al., 2000).

In each city, concentrations of ambient air pollutants were measured at a fixed monitoring site with special emphasis on measurements of particle number concentrations (NC) (Pekkanen et al., 2000). Aerosol spectrometers were used to measure particle NC of ultrafine particles, that is, particles in size range of 0.01–0.1 μm, and of accumulation mode particles, that is, particles in size range of 0.1–1.0 μm (Tuch et al., 2000; Khlystov et al., 2001; Ruuskanen et al., 2001). Harvard impactors were used to measure PM2.5 (Janssen et al., 2000). Data on meteorological variables, PM10, NOX, CO, SO2, and O3 were collected from existing networks in Amsterdam by the Royal Dutch Meteorological Institute and the National Institute of Public Health and the Environment, and in Helsinki by Helsinki Metropolitan Area Council. In Erfurt, the study personnel collected the data, except CO data, which were collected by Thüringer Landesanstalt für Umwelt. To describe coarse particles, PM10−2.5 was calculated by subtracting PM2.5 from PM10. All variables are 24-h means from noon-to-noon and the descriptive statistics of the data have been published (Pekkanen et al., 2000).

There were 37 panelists in Amsterdam and 47 panelists in both Erfurt and Helsinki. The criteria for being included in the study included a self-report on a physician-diagnosed coronary artery disease (e.g., angina pectoris, a past myocardial infarction, PTCA (percutaneous transluminal coronary angioplasty), or a coronary by-pass surgery) and being a non-smoker. The exclusion criteria included a recent, less than 3 months, myocardial infarction, stroke or by-pass surgery, unstable angina pectoris, having a cardiac pacemaker, type I diabetes, and poor cooperation. For each subject, the visit was aimed to be always on the same weekday at the same time. The medication of the subjects was not changed for the clinical visits. In Amsterdam, the subjects were recruited in cooperation with the municipal health center and by using an information letter, in Erfurt through a cardiologist's practice, and in Helsinki through a patient organization. The subjects usually travelled by car to the clinical visits: in Amsterdam and Erfurt in 75% of the visits and in Helsinki in 97% of the visits (Pekkanen et al., 2000).

Two-channel ambulatory ECGs were recorded with Oxford MR-63 tape recorders (Oxford Instruments, Abington, UK). Standard electrode positions recommended by the manufacturer were used. The ECG recordings were analyzed with the Oxford Exel Medilog II V7.5 system (Oxford Instruments, Abington, UK). All tapes were analyzed in one core laboratory (Kuopio University Hospital). A trained research assistant reviewed and when necessary interactively edited the automatically predetermined classification of the QRS complexes. Only normal-to-normal beat (NN) intervals between 300 and 2000 ms with NN ratios between 0.8 and 1.2 were included for analyses of HRV. Recordings with frequent extrasystolia (>10% from total beats) were excluded.

Both time domain and frequency domain measures of HRV were obtained. The time domain measures used for the analyses were standard deviation of the NN (SDNN) interval and the square root of the mean of the sum of the squares of the differences between adjacent NN intervals (r-MSSD). The former is a measure of overall HRV and the latter assesses the short-term variation in heart rate. The frequency domain measures included power spectrum densities of two frequency bands: low-frequency (LF) power (0.04 to 0.15 Hz) and high-frequency (HF) power (0.15 to 0.4 Hz). The HF component of HRV is thought to reflect mainly vagal (parasympathetic), and the LF component both sympathetic and parasympathetic part of the autonomous nervous input to the heart (Task Force, 1996).

Statistical Analyses

The frequency domain variables were log-transformed for the analyses. The HRV variables SDNN, HF, and LF/HF were used as the main end points. We chose SDNN as it describes overall HRV and HF, because it is a good marker of vagal control of heart rate and also highly correlated with r-MSSD. LF/HF is a tool to describe sympathovagal balance. In addition, r-MSSD, and HF in normalized units (HFnu=HF/(LF+HF)) were used in selected analyses.

The data were analyzed by using the statistical package SAS and S-Plus (SAS Institute Inc., 1989; S-PLUS 2000, 1999). For the exposure variables, lag 0 was defined as the 24-h period from the previous day noon to the noon of the day of the clinical visit. The 5-day average was calculated as the mean of lag 0–lag 4.

A basic model for each panel was built without any air pollution variable in S-Plus using preset rules (Pekkanen et al., 2000). The basic model for the Amsterdam panel included linear terms for time trend, temperature (lag 2), relative humidity (lag 3), and barometric pressure (lag 3), and weekday as a categorical variable. The basic model for the Erfurt panel included linear terms for time trend, relative humidity (lag 3) and barometric pressure (lag 2), temperature (lag 3) was modeled with linear, squared, and cubic terms, and weekday as a categorical variable. The basic model for the Helsinki panel included linear terms for time trend, relative humidity (lag 1) and barometric pressure (lag 1), temperature (lag 3) was modeled with linear and squared terms, and weekday as a categorical variable. The polynomial terms were chosen based on the shape of the association and the statistical significance of the polynomial terms. Results were very insensitive to alternative model specifications. Further adjustment for heart rate did not affect the pollution effect estimates.

In final statistical analyses, individual pollutants were added to the basic model one at a time. A mixed model was used (PROC MIXED in SAS) taking into account repeated observations and assuming constant correlation within a subject (Littell et al., 1999). The effect estimates are given for an increase of 10,000 particles/cm3 for ultrafine particles, 1000 particles/cm3 for accumulation mode particles, 10 μg/m3 for PM2.5, PM10−2.5, and NO2, and 1 mg/m3for CO.

Pooled effect estimates were calculated as a weighted average of the center-specific estimates using the inverse of the center-specific variances as weights. The heterogeneity between centers was tested with χ2 test (Normand, 1999). When significant heterogeneity (P<0.1) between the centers was observed, a pooled effect estimate was calculated using random-effects model (Berkey and Laird, 1986).

As sensitivity analyses two-pollutant models were constructed. The pollutant pairs were PM2.5 with ultrafine particles, NO2, CO, and O3, and ultrafine particles with NO2, CO, and O3.

Results

Altogether, there were a total of 131 subjects and 1431 clinical visits (Table 1). From a total of 1394 ECG recordings, 58 recordings were excluded based on exclusion criteria. From the remaining, in 1266 recordings the quality of the recording was high enough for a successful HRV analysis (success rate 91.8%). The subjects in the Erfurt panel were younger, but their overall HRV was lower than in the Amsterdam and Helsinki panels. In Erfurt, there were also more subjects with a history of a coronary by-pass surgery or PTCA.

Table 1: Description of the study population and heart rate variability (HRV) variables during the paced breathing.

During the study periods, NC of ultrafine particles were rather similar in the three cities, while the levels of accumulation mode particles and PM2.5 were clearly lower in Helsinki than in Amsterdam and Erfurt (Table 2). In contrast, concentration of PM10−2.5 was lower in Erfurt than in Amsterdam and Helsinki. Concentrations of gaseous air pollutants were highest in Amsterdam; in Erfurt and Helsinki, the concentrations were rather similar. During the study period, the temperature was the lowest in Helsinki, the mean temperature was below 0°C.

Table 2: Descriptive statistics of 24-h mean levels of air pollutants and temperature.

In general, the correlations between NC of ultrafine particles and accumulation mode particles, PM2.5, or PM10−2.5 were rather low, and the correlations of NC of ultrafine particles with NO2 were the highest (Table 3). NC of accumulation mode particles was highly correlated with PM2.5. Correlations between PM10−2.5 and other pollutants were low. In Erfurt, all correlations between different air pollutants were higher than in other cities. The mean temperature was inversely correlated with all pollutants but PM10−2.5.

Table 3: Spearman correlations between particulate air pollution, temperature, and relative humidity.

As breathing frequency strongly affects HRV, measurements from the paced breathing period were considered to be most reliable and are reported in detail. In the pooled analyses, an increase in the NC of ultrafine particles was consistently associated with a decrease in LF/HF (Table 4, Figure 1). The effect estimates were negative for all the lags and all but one lag (lag 0) were statistically significant. The largest estimate was for the 2-day lagged NC of ultrafine particles. There was no heterogeneity between the centers. Elevated levels of NC of ultrafine particles were not associated with a decrease in SDNN or HF in the pooled analysis. In Amsterdam, NC of ultrafine particles was associated with increased HF (Table 5, Figure 1). An increase in NC of ultrafine particles at lag 2 was also statistically significantly associated with an increase in HFnu in all centers, which is in accordance with the observed association with decreased LF/HF (Table 5).

Table 4: Association between number concentration of ultrafine particles, accumulation mode particles, PM2.5, PM10−2.5, NO2, CO, and HRV during paced breathing.
Figure 1
Figure 1

Association between ultrafine particles, PM2.5, and HRV. Center-specific and pooled estimates with 95% confidence intervals (CI). In the group of three dots (•), the first on the left-hand side refers to lag 0, the second to lag 2 and the third to 5-day average concentration of the air pollutant.

Table 5: Center-specific associations between ultrafine particles (lag 2, per 10,000/cm3), PM2.5 (lag 2, per 10 μg/m3), and HRV during paced breathing.

Neither accumulation mode particles nor PM2.5 showed a clear pattern in the associations with HRV variables (Tables 4 and 5, Figure 1). Significant heterogeneity existed between the centers. In Helsinki, elevated concentrations of PM2.5 were associated with decreased HF and increased LF/HF, whereas the opposite was true in Erfurt. No such associations were observed in Amsterdam. Variations in PM10−2.5 concentrations were not significantly associated with HRV variables (Table 4).

Elevated concentrations of NO2 and CO were associated with decreased LF/HF and SDNN at lags of 1–2 days in the pooled analyses (Table 4). Elevated concentrations of CO were also associated with decreased HF. No heterogeneity was observed between the three centers. Only one center-specific estimate was statistically significant: in Erfurt, lag 1 of CO was associated with decreased LF/HF (data not shown).

Despite the fairly high correlations between some pollutants, introducing two pollutants in a model at a time did not change much the results of the associations between particulate air pollution and HRV or between gaseous air pollution and HRV. Only when having ultrafine particles and NO2 in the same model the effect estimates were unstable. O3 was not associated with HRV in the two-pollutant models.

The results for the spontaneous breathing period, in general, reflected those of the paced breathing period, but they were less significant. No consistent associations were observed during the standing period. No clear differences were observed between men and women. Excluding the subjects exposed to environmental tobacco smoke at home from the analyses did not affect the effect estimates.

Discussion

The present study is the largest study conducted until today on the effects of particulate air pollution on cardiac autonomic control measured as HRV. Among elderly subjects with stable coronary heart disease in the Netherlands (Amsterdam), Germany (Erfurt), and Finland (Helsinki), the most consistent result was a decrease in LF/HF ratio in association with the elevated NC of ultrafine particles and levels of NO2 and CO, suggesting a change in sympathovagal balance towards vagal tone. In contrast, effects of PM2.5 on HRV were not consistent across the centers; PM2.5 was associated with reduced HRV only in Helsinki.

The NC of ultrafine particles was rather similar across the study centers. This finding is in accordance with the results from a previous study performed during wintertime 1996–1997 in Helsinki, Erfurt, and Alkmaar, a medium-sized city about 50 km northwest of Amsterdam (Ruuskanen et al., 2001). In that study, mean NC of ultrafine particles ranged from 16,200 (Helsinki) to 18,300 (Alkmaar) particles/cm3. In the present study, differences in PM2.5 and PM10−2.5 concentrations between the centers were substantially larger than the differences in NC of ultrafine particles. The low PM10−2.5 concentration in Helsinki can, at least partly, be explained by snow cover on the ground during most of the time of the study period. On the other hand, in Amsterdam there is little snow cover, and this allows particles from the ground to resuspend in the air with wind blow. Ultrafine particles and PM2.5 were correlated only in Erfurt. This suggests that particle number and particle mass characterize separate aspects of the urban air pollution mixture. In addition, the difference in the correlations between PM2.5 and NC of ultrafine particles is partly explained by the different contribution of wind direction and other meteorological variables like relative humidity, and by the location of Erfurt in a valley compared to maritime climate of Amsterdam and Helsinki (de Hartog et al., 2005).

Stone and Godleski (1999) have described two possible mechanisms by which inhaled particles could affect HRV and thereby increase cardiovascular risk. They suggest that the adverse effects of particles could be mediated either through a sympathetic stress response leading to decreased HRV and tachyarrythmias or through a vagal response leading to increased HF, increased HRV, and bradyarrythmias. Therefore, these two mechanisms would have opposite effects on HRV. The present observation of an association between elevated levels of ultrafine particles and decreased LF/HF and increased HFnu would tend to support this vagal response.

Although most previous studies have observed associations between particle mass and decreased HRV (Liao et al., 1999; Gold et al., 2000; Creason et al., 2001; Devlin et al., 2003; Holguin et al., 2003; Pope et al., 2004), which would suggest a sympathetic response, the present result is supported by two previous smaller studies. In the study of Pope et al. (1999), r-MSSD, which is closely correlated with HF, increased with elevated levels of PM10 among elderly subjects. Tarkiainen et al. (2003) found an increase in r-MSSD in relation to acute CO exposure among subjects with stable coronary artery disease. These observations together with the present results give some support to the hypothesis that air pollution could lead to increased cardiac vagal control, too.

The picture is, however, not completely clear, as we observed little association of ultrafine particles with HF, which is a widely accepted marker of cardiac vagal control. In contrast, the interpretation and physiological basis of the LF component is more controversial (Task Force, 1996; Eckberg, 1997; Houle and Billman, 1999; Malliani, 1999). The LF component is affected by both the sympathetic and parasympathetic nervous system and, as such, does not accurately reflect changes in the sympathetic activity. There was also some suggestion of decreased HRV, especially in association with NO2, which would suggest a sympathetic, not a vagal, response. It is therefore possible that the effect of air pollution is mediated through several mechanisms and the final outcome is best described by measures of the balance between the sympathetic and vagal components, like LF/HF or HFnu.

The air pollution effects on HRV observed in the present study were delayed a day or two after exposure, which is in agreement with the studies on the respiratory effects of particulate air pollution. The lag structure indicates that the air pollution effect to the heart might not be mediated alone directly through the nervous system, which acts rapidly. Instead, the effect in HRV could be a response to a systemic stress, for example, an inflammatory reaction and systemic oxidative stress.

Consistent with previous studies (Liao et al., 1999; Gold et al., 2000; Creason et al., 2001; Devlin et al., 2003; Holguin et al., 2003; Pope et al., 2004), in Helsinki elevated concentrations of PM2.5 were significantly associated with reduced HF and PM2.5 also tended to be associated with reduced SDNN. No such associations were, however, observed in Amsterdam or Erfurt. In contrast, the opposite was true in Erfurt. We have previously shown in Helsinki among this same panel an association between elevated levels of PM2.5 and increased risk of myocardial ischemia (Pekkanen et al., 2002), which supports the present association between PM2.5 and reduced HRV, as both end points are associated with disease severity or increased mortality (Kawaguchi et al., 1993; Tsuji et al., 1994; Task Force, 1996; Lai et al., 2004).

We used fixed site measurements to estimate exposure to particulate air pollution. Previously, it has been reported from the present study that fixed site measurements of PM2.5 describe well variations in personal exposure to PM2.5 in Amsterdam and Helsinki (Janssen et al., 2000). As spatial variation in NC of ultrafine particles is larger than in PM2.5 concentration, it is possible that the exposure to ultrafine particles is not estimated as well as to PM2.5 when using fixed measurement site.

The differences in the results between most previous studies and the present one, and also between the panels in the present study, might partly be explained by differences in the disease status of the subjects. In most previous studies, the subjects have been healthy or at least not severely diseased. In the present study, all subjects had coronary heart disease, and it is well known that HRV is altered among subjects with ischemic heart disease as compared to controls without the disease (Huikuri and Makikallio, 2001). Moreover, there are also obvious differences in various measures of HRV between subjects with uncomplicated coronary heart disease and those with coronary heart disease with complicated myocardial infarction. Impaired HF oscillations of the heart rate are the most prominent feature in subjects with uncomplicated coronary heart disease, suggesting mainly an impairment in vagal autonomic regulation. Subjects with prior myocardial infarction have a reduced overall HRV, and a specific spectral pattern with a reduced LF component has been observed in subjects with prior myocardial infarction and impaired left ventricular function (Huikuri and Makikallio, 2001). Therefore, it is possible that effects of air pollution on HRV can be different among diseased subjects compared to healthy subjects, and even among the diseased.

Heterogeneity in the effect of particulate air pollution has also been observed in other multicenter studies or meta-analyses, like the European APHEA-study (Katsoyannni et al., 2001; Samoli et al., 2001) and the US National Morbidity Mortality Air Pollution Study (Dominici et al., 2003). Previously, Ibald-Mulli et al. (2004) reported from this same study that ultrafine particles did not show homogeneous effects across the study centers with blood pressure; effects of ultrafine particles on blood pressure were significant only in Erfurt (Ibald-Mulli et al., 2004). In the Erfurt panel, the subjects were the youngest, but their HRV was the lowest, possibly indicating that the subjects had a more severe heart disease. In addition, particles from different sources are likely to have different health effects. In context with this same study, we have shown that long-range transported particles form a larger proportion of PM2.5 in Helsinki (50%) than in the two other centers (32% and 34%) (Vallius et al., 2005). The correlations between different measures of particulate air pollution were also clearly higher in Erfurt than in Amsterdam or Helsinki, suggesting a difference in the air pollution mixture or meteorological conditions (de Hartog et al., 2005).

Conclusion

To conclude, measurement of HRV using ambulatory ECG recordings was feasible in a large epidemiological panel study. Concentrations of ultrafine particles, NO2, and CO were associated with decreased LF/HF ratio in all the three centers, suggesting that air pollution could lead to increased cardiac vagal control. PM2.5 was associated with reduced HF and increased LF/HF in Helsinki, whereas the opposite was true in Erfurt, and in Amsterdam, there were no clear associations between PM2.5 and HRV. The results suggest that the cardiovascular effects of ambient ultrafine and PM2.5 can differ from each other and that their effect may be modified by the characteristics of the exposed subjects and the sources of PM2.5.

References

  1. , and Nonlinear growth curve analysis: estimating the population parameters. Ann Hum Biol 1986: 13: 111–128.

  2. , and Air pollution and health. Lancet 2002: 360: 1233–1242.

  3. , et al. Particulate matter and heart rate variability among elderly retirees: the Baltimore 1998 PM study. J Expo Anal Environ Epidemiol 2001: 11: 116–122.

  4. , et al. Elderly humans exposed to concentrated air pollution particles have decreased heart rate variability. Eur Respir J 2003: 21(Suppl. 40): 76s–80s.

  5. , et al. National Maps of the effects of particulate matter on mortality: exploring geographical variation. Environ Health Perspect 2003: 111: 39–43.

  6. Sympathovagal balance. A critical appraisal. Circulation 1997: 96: 3224–3232.

  7. , et al. Mechanisms of morbidity and mortality from exposure to ambient air particles. HEI research report no. 91, Cambridge, MA, USA, 2000.

  8. , et al. Ambient pollution and heart rate variability. Circulation 2000: 101: 1267–1273.

  9. , et al. Size distribution on particulate matter in three European cities. J Environ Monit 2005: 4: 302–310.

  10. , et al. Air pollution and heart rate variability among the elderly in Mexico City. Epidemiology 2003: 14: 521–527.

  11. , and Low-frequency component of the heart rate variability spectrum: a poor marker of sympathetic activity. Am J Physiol Heart Circ Physiol 1999: 276: H215–H223.

  12. , and Heart rate variability in ischaemic heart disease. Auton Neurosci 2001: 90: 95–101.

  13. , et al. Effects of particulate air pollution on blood pressure and heart rate in subjects with cardiovascular disease: a multicenter approach. Environ Health Perspect 2004: 112: 369–377.

  14. , et al. Personal exposure to fine particulate matter in elderly subjects: relation between personal, indoor and outdoor concentrations. J Air Waste Management Assoc 2000: 50: 1133–1143.

  15. , et al. Confounding and effect modification in the short-term effects of ambient particles on total mortality: results from 29 European cities within the APHEA2 project. Epidemiology 2001: 5: 521–531.

  16. , et al. Predictors of disease severity and survival in patients with coronary artery disease. Coronary Artery Dis 1993: 4: 971–980.

  17. , et al. Intercomparison using laboratory-generated aerosols. Atmos Environ 2001: 35: 2045–2051.

  18. , et al. Treadmill scores in elderly men. J Am Coll Cardiol 2004: 43: 606–615.

  19. , et al. Daily variation of particulate air pollution and poor cardiac autonomic control in the elderly. Environ Health Perspect 1999: 107: 521–525.

  20. , et al. SAS® System for Mixed Models. SAS Institute Inc., Cary, NC, 1999.

  21. The pattern of sympathovagal balance explored in the frequency domain. News Physiol Sci 1999: 14: 111–117.

  22. , et al. Passage of inhaled particles into the blood circulation in humans. Circulation 2002: 105: 411–414.

  23. Tutorial in biostatistics. Meta-analysis: formulating, evaluating, combining, and reporting. Statist Med 1999: 18: 321–359.

  24. , et al. ULTRA. Exposure and Risk Assessment for Fine and Ultrafine Particles in Ambient Air. Study Manual and Data book. Publications of National Public Health Institute B9/2000, Kuopio, 2000 Available: .

  25. , et al. Particulate air pollution and risk of ST-segment depression during repeated submaximal exercise tests among subjects with coronary heart disease. The Exposure and risk assessment for fine and ultrafine particles in ambient air (ULTRA) study. Circulation 2002: 106: 933–938.

  26. , et al. Heart rate variability associated with particulate air pollution. Am Heart J 1999: 138: 890–899.

  27. , et al. Ambient particulate air pollution, heart rate variability, and blood markers of inflammation in a panel of elderly subjects. Environ Health Perspect 2004: 112: 339–345.

  28. , et al. Concentrations of ultrafine, fine and PM2.5 particles in three European cities. Atmos Environ 2001: 35: 3729–3738.

  29. , et al. Fine particulate air pollution and mortality in 20 US Cities, 1987–1994. N Engl J Med 2000: 343: 1742–1749.

  30. , et al. Investigating regional differences in short-term effects of air pollution on daily mortality in the APHEA project: a sensitivity analysis for controlling long-term trends and seasonality. Environ Health Perspect 2001: 109: 349–353.

  31. SAS Institute Inc. SAS/STAT® Software: Changes and Enhancements through Release 6.11. SAS Institute Inc., Cary, NC, 1989, pp. 531–656.

  32. Air pollution and hospital admissions for heart disease in eight US counties. Epidemiology 1999: 10: 17–22.

  33. , et al. Particulate air pollution and acute health effects. Lancet 1995: 345: 176–178.

  34. S-PLUS 2000. Guide to Statistics, Vol. 1. Data Analysis Products Division, Mathsoft, Seattle, WA, 1999, pp. 326–327.

  35. , and First steps toward understanding the pathophysiologic link between air pollution and cardiac mortality. Am Heart J 1999: 138: 804–807.

  36. , et al. Effect of acute carbon monoxide exposure on heart rate variability in patients with coronary artery disease. Clin Physiol Funct Imaging 2003: 23: 98–102.

  37. Task Force of The European Society of Cardiology and The North American Society of Pacing and Electrophysiology. Heart rate variability. Standards of measurement, physiological interpretation and clinical use. Circulation 1996: 93: 1043–1065.

  38. , et al. Reduced heart rate variability and mortality risk in an elderly cohort. The Framingham Heart Study. Circulation 1994: 90: 878–883.

  39. , et al. Comparison of two particle size spectrometers for ambient aerosol measurements in environmental epidemiology. Atmos Environ 2000: 34: 139–149.

  40. , et al. Sources and elemental composition of ambient PM2.5 in three European cities. Sci Total Environ 2005: 337: 147–162.

  41. , et al. Cardiac arrhythmia induction after exposure to residual oil fly ash particles in a rodent model of pulmonary hypertension. Toxicol Sci 1998: 41: 209–216.

Download references

Acknowledgements

The study was carried out within the framework of the “Exposure and risk assessment for fine and ultrafine particles in ambient air”(ULTRA) project. The project was funded by the EU ENVIRONMENT and CLIMATE Research Programme Contract ENV4-CT97-0568. The project was coordinated by the Unit of Environmental Epidemiology, National Public Health Institute, PO Box 95, Kuopio 70701, Finland with funding also from Academy of Finland, the Centre of Excellence Programme 2002–2007 of the Academy of Finland (Contract 53307), and the National Technology Fund (TEKES, Contract 40715/01). Dr. Gold was supported by EPA 826780-01-0 and NIH PO1 ES09825. The contribution of following persons and institutions to the field work of the project is gratefully acknowledged: Helsinki: Aadu Mirme, Ph.D., Gintautas Buzorius, Ph.D., Ismo Koponen, M.Sc., Marko Vallius, M.Sc., Sami Penttinen, Kati Oravisjärvi, Annalea Lohila, M.Sc., Anita Tyrväinen, Helsinki Metropolitan Area Council, Helsinki (Päivi Aarnio, Lic Tech., and Tarja Koskentalo, Lic Tech.), and the Finnish Heart Association. Erfurt: Gabi Wölke, M.A., Martina Stadeler, M.D., Regina Müller, Cornelia Engel, Thomas Tuch, Ph.D., and Sabine Koett, Klaus Koschine, Mike Pitz, MA. Amsterdam: Andrey Khlystov, Ph.D., Gerard Kos, Carolien Mommers, M.Sc., Marloes Jongeneel, M.Sc., Boukje de Wit, Isabella van Schothorst, Veronique van den Beuken, M.Sc., Marieke Oldenwening, Nicole Janssen, Ph.D., Jean Pierre van Mulken, and Environmental Medicine, Municipal Health Service Amsterdam (Saskia van der Zee, Ph.D., Willem Roemer, Ph.D., and Joop van Wijnen, M.D.).

Author information

Affiliations

  1. Unit of Environmental Epidemiology, National Public Health Institute, Kuopio, Finland

    • Kirsi L Timonen
    • , Timo Lanki
    • , Pekka Tiittanen
    •  & Juha Pekkanen
  2. Department of Clinical Physiology and Nuclear Medicine, University of Kuopio and Kuopio University Hospital, Kuopio, Finland

    • Kirsi L Timonen
    • , Esko Vanninen
    •  & Tuula Tarkiainen
  3. Institute for Risk Assessment Sciences, University of Utrecht, Utrecht, The Netherlands

    • Jeroen de Hartog
    • , Bert Brunekreef
    •  & Gerard Hoek
  4. GSF, Institute of Epidemiology, Neuherberg, Germany

    • Angela Ibald-Mulli
    • , Joachim Heinrich
    •  & Annette Peters
  5. Channing Laboratory, Brigham and Women's Hospital and the Harvard Medical School, Boston, MA, USA

    • Diane R Gold
  6. GSF, Institute for Inhalation Biology, Neuherberg, Germany

    • Wolfgang Kreyling

Authors

  1. Search for Kirsi L Timonen in:

  2. Search for Esko Vanninen in:

  3. Search for Jeroen de Hartog in:

  4. Search for Angela Ibald-Mulli in:

  5. Search for Bert Brunekreef in:

  6. Search for Diane R Gold in:

  7. Search for Joachim Heinrich in:

  8. Search for Gerard Hoek in:

  9. Search for Timo Lanki in:

  10. Search for Annette Peters in:

  11. Search for Tuula Tarkiainen in:

  12. Search for Pekka Tiittanen in:

  13. Search for Wolfgang Kreyling in:

  14. Search for Juha Pekkanen in:

Corresponding author

Correspondence to Kirsi L Timonen.

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/sj.jea.7500460

Further reading