2004 Progress Report: Inflammatory Responses and Cardiovascular Risk Factors in Susceptible PopulationsEPA Grant Number: R827354C002
Subproject: this is subproject number 002 , established and managed by the Center Director under grant R827354
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
Center: Airborne PM - Rochester PM Center
Center Director: Oberdörster, Günter
Title: Inflammatory Responses and Cardiovascular Risk Factors in Susceptible Populations
Investigators: Wichmann, Heinz-Erich , Bedada, Getahun Bero , Cyrus, Josef , Henneberger, Alexandra , Peters, Annette , Pitz, Mike , Rückerl, Regina , Socher, Martin , Stölzel, Matthias , Yue, Wei
Current Investigators: Wichmann, Heinz-Erich , Peters, Annette
Institution: GSF - Forschungszentrum fur Umwelt und Gesundheitand Ludwig Maximilian University, Neuherberg, Germany
EPA Project Officer: Chung, Serena
Project Period: June 1, 1999 through May 31, 2005 (Extended to May 31, 2006)
Project Period Covered by this Report: June 1, 2004 through May 31, 2005
RFA: Airborne Particulate Matter (PM) Centers (1999) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Particulate Matter , Air
The associations observed in the Rochester Particle Center Study seem to be attributable to particulate air pollution; however, the health impact of particles might differ depending on the physical and chemical properties of the particles. As the size and the composition of the particles are determined by the sources of the particles, it is important for regulatory purposes to better understand the relative importance of sources in association with health effects. In addition, some air pollutants might be surrogates for the true acting but unmeasured agents. For example, SO2 concentrations at very low concentrations seem to exhibit strong effects on heart rate variability in 24-hour EKG recordings and with respect to an acute phase reaction in the Rochester Particle Center Study. Given the very low concentrations observed in Erfurt, Germany, it seems unlikely that the measured SO2 concentrations are responsible for the observed associations. In Year 6 of the project, we worked on the following specific objectives: (1) blood markers of exacerbation of chronic inflammation and altered vascular function are elevated in association with ambient particles; (2) cardiac function in patients with chronic obstructive pulmonary disease (COPD) is altered in association with ambient air pollution as seen in coronary artery disease (CAD) patients; and (3) particles from traffic and other combustion sources are associated with vascular and cardiac effects.
Year 1 to 5
Epidemiological studies conducted as part of the Rochester Particle Center showed effects of ambient particles with respect to cardiac responses and blood biomarkers. In a study in patients with CAD, the autonomic control of the heart was altered in association with PM2.5 and organic carbon (OC) and elemental carbon (EC) concentrations of PM2.5. These findings highlight the importance of the carbonaceous component in particles. Furthermore, we were able to detect changes in the repolarization of the heart in association with PM2.5 and the number concentrations of accumulation mode particles (AP) (Henneberger, et al., 2004). Regarding arrhythmia, the number of supraventricular and ventricular runs showed strong effects correlated to AP and ultrafine particles as well. Thereby, we found first evidence that particles also might increase cardiac vulnerability and might modify the cardiac substrate. In addition to the cardiac responses, there also was evidence for an increase in C-reactive protein (CRP) concentrations and a shift to a more pro-coagulating state of the blood. An area not covered by this study so far was changes in the endothelial dysfunction, which were suggested by the results of the human exposure studies in core 3.
Year 6 Activities
Two epidemiological studies were conducted to assess short-term health effects of fine and ultrafine particles in 61 patients with CAD and in 39 patients with COPD in Erfurt, Germany, as part of the Rochester Particle Center. Twelve clinical visits, including ECG measurements and blood withdrawals, were scheduled. Ninety-eight percent of all scheduled ECG recordings and 94 percent of all scheduled blood withdrawals were realized. A detailed description of the study and the results of the study in patients with CAD are presented in the appendix of the Year 4 progress report. Ethylenediaminetetracetic acid and citrate plasma samples were ascertained for future analyses and stored either at –80 °C or in liquid nitrogen. These samples were used for the analyses described in this report.
Continuous outcomes such as measurements of the blood coagulability, heart rate variability, and lung function measures were analyzed based on linear regression models considering repeated measurements for the subjects. The distribution of the residuals was checked carefully and additional analyses converting the continuous measurements into binary variables were conducted in case the residuals were not approximately normally distributed. These variables were analyzed using logistic regression analyses. Special care was given to adjust for time trend, season, and meteorological parameters. Core confounder models have been developed as part of the analyses performed during Years 3 and 4 of the Rochester Particle Center activities. In particular, several approaches to model the dose-response functions were applied, including parametric, semi-parametric, and non-parametric methods. The lag structure of the association between the air pollutants and the outcomes was analyzed to evaluate the time lags between exposure and response. Based on the experimental and clinical data collected in the other cores, specific hypotheses were formulated before the analyses and then tested in the epidemiological data (specific aims 1 and 2). The results obtained for the different sources will be compared to the results of the contributing particle fractions or gaseous pollutants (specific aim 3). It is unlikely that there is sufficient power to test for differences between regression coefficients for single pollutants and for specific sources. However, the biomarkers of cardiac function exhibit different response profiles when PM2.5, ultrafine particles, and OC or EC are considered. Additional information on source contributions will help to elucidate the role of different particle properties responsible for cardiovascular disease exacerbation via different mechanisms.
Specific Aim 1 - Blood Biomarker
For the CAD panel, an additional marker of inflammation was determined by the immunology core. Soluble CD40 ligand had been selected as a marker for exacerbation of chronic inflammation and altered vascular function (Phipps, 2000).
Our findings suggest an increase in sCD40L in association with ambient air particles, particularly with elevated levels of ultrafine particles and accumulation mode particles. For platelets, the effects were limited to ultrafine particles showing an immediate as well as a 3-day delayed decrease. The regression of leukocytes showed consistently negative associations for ultrafine particles, AP, and PM2.5, with lag 0 and for AP in addition with lag 3 and the 5-day average. As the effects seemed to be limited to the 24 hours prior to the blood withdrawal, we split the 24 hours into four 6-hour periods and analyzed the results for ultrafine particles. Although the effect for sCD40L was most prominent for the time period 12 to 17 hours prior to the blood withdrawal, platelets and leukocytes showed an immediate decrease in the first 5 hours and a delayed one between 18 to 24 hours (Figure 1).
Figure 1. Effects of UFP on Blood Cells and sCD40 Ligand in 6 Hour Periods Prior to the Blood Withdrawal
Erythrocytes and hemoglobin in contrast seemed to react more to the larger particle size fractions PM2.5 and PM10, showing a decrease in association with higher levels of air pollution. The largest negative effects for the erythrocytes were seen for PM10 for lag 2 and the 5-day-average exposure, a finding that is reflected to a lesser extent in the results of the hemoglobin.
For the COPD panel, differential hemograms were available. Preliminary results suggest no effect of particulate matter on all leukocytes combined. However, an increase in neutrophilic bandform granulocytes was observed in association with PM10 and AP immediately as well as with a 5-day average. Other leukocytic cell rows were either unaffected or showed small decreases. These results may provide evidence for a stimulation of the bone marrow by particulate matter.
Specific Aim 1 - Interdependence of Blood Markers and Cardiac Function
Additionally, associations within ECG recordings (time- and frequency-domain of heart rate variability (HRV) and repolarization parameters) and associations within blood markers (acute phase response, endothelial cell activation, and coagulation state markers) as well as associations between ECG recordings and blood markers were analyzed using generalized estimating equation models adjusting for repeated measurements. Within the ECG recordings, strong significant associations were found between time- and frequency-domain parameters, and moderate but also significant associations between frequency-domain and repolarization parameters. Within blood markers, strong but significant associations existed between CRP and fibrinogen, D-dimer, E-selectin, ICAM-1, and SAA.
Between ECG recordings and blood markers, repolarization parameters and acute phase response proteins showed moderate but significant associations. HRV parameters and endothelial cell activation markers were significantly but only weakly associated. The results indicate the interplay between the autonomic nervous system and myocardial substrate as well as interactions of the acute phase response with endothelial cell activation and coagulation state. Although ECG parameters and blood markers seem to vary independently, there was the suggestion for a link between systemic inflammation and repolarization as well as endothelial dysfunction and HRV.
Specific Aim 1 - Endothelial Dysfunction
No measurements of endothelial dysfunction were conducted as part of the clinical examinations. Therefore, only blood biomarkers might serve as indicators of altered endothelial response. We initially proposed to measure nitrite, the oxidation product of NO to determine an increase in dilatory mediators to counterbalance the increase vasoconstriction that might have occurred in response to particle exposures. However, it became apparent that NO products could not be measured reliably in plasma samples as collected in Erfurt.
Specific Aim 2 - HRV
The effects of particulate air pollution on the autonomic nervous system as measured by heart rate (HR) and HRV in patients suffering from COPD were analyzed. Low frequency (LF) and the ratio of low to high frequency (LF/HF) increased in association with an increase in PM10, OC, and EC during the 24 hours before the ECG measurement. Consistently, there was a significant decrease in heart rate with an increase of all particles measured 0-23 hours before the ECG recording. The analysis also showed a significant increase in RMSSD in response to an increase in all particle concentrations and some gases during 48-71 hours before the ECG recording. These results are contradictory to prior findings in CAD patients and our initial hypothesis. Taking both findings into account, it is conceivable that the air pollution reaction depends on the disease status of the patient (Suh, et al., 2004) and that elevated concentrations of ambient particles are associated with a disturbance of the autonomic heart control manifested by an increased HRV in patients with COPD.
Specific Aim 3 - Effects of Traffic on Myocardial Infarction
A complete series of myocardial infarction (MI) survivors registered between 1999 and mid-2001 was interviewed to collect information on activities during the 4 days before MI onset. Analyses considered ambient particle concentrations as well as diary data. A total of 691 subjects were interviewed, and they showed a higher prevalence of time spent in traffic 1 hour before the onset of MIs than 24 to 72 hours earlier (Figure 2). Time spent in traffic was associated with MI onset 1 hour later (OR = 2.9 (95% CI: 2.2 to 3.8) (Peters, et al., New England Journal of Medicine, 2004). These associations were seen for times spent in cars (OR = 2.6 (95% CI: 1.9 to 3.6), times spent in public transport (OR = 3.1 (95% CI: 1.4 to 6.8) and on bicycles (OR = 3.9 (95% CI: 2.1 to 7.2). Ambient PM2.5 concentrations at the urban background site also suggested an association with MI onset 2 days later (RR: 1.09 for 10 µg/m3 PM2.5 (95% CI: 0.98 to 1.20) (Peters, et al., 2005).
Figure 2. Prevalence of Time Spent in Traffic in 691 MI Survivors During the 72 Hours Before MI Onset.
Specific Aim 3 - Source Apportionment
Sources of fine and ultrafine particles were analyzed using two different datasets: (1) data on elemental composition of accumulation mode and ultrafine particles; and (2) size distribution of fine and ultrafine particles. All analyses are conducted in collaboration with R827354C001.
Elemental Composition Data. Particulate matter was collected on a nine-stage low pressure cascade Berner Impactor on 8 µm polypropylene foils. Stage 1 covers the upper tail of the ultrafine range collecting particles with aerodynamic diameters between 50 and 100 nm. Stages 2 to 4 (0.10-0.80 µm) cover the accumulation mode. The elemental composition was analyzed by proton induced X-ray emission spectrometry (PIXE). Concentrations of 23 elements can be determined whose atomic number is higher than 11.
Data are now available for the first six stages from September 1, 1997, to August 1, 1999, and in addition for stages 1 and 3 until February 13, 2002, with an interruption in the summer of 2001. In a pilot study, the data of 1997 and 1998 were successfully used for source apportionment (Cyrys, et al., 2003; Stölzel, 2003).
However, we discovered changes in the concentration levels of almost an order of magnitude in the extended data set especially for elements related to crustal material. These trends cannot be explained solely by possible changes in the aerosol composition or by changing aerosol sources. They are probably due to a somewhat changing composition of the impactor foils. Currently, new calibration procedures are being developed to account for this. A revised dataset will be available soon and will be analyzed by Positive Matrix Factorization, an advanced factor analytical techniques.
Size Distribution Data. Another promising approach to do source apportionment is to analyze data on the size distribution of particulate matter by advanced factor analytical techniques (Kim, et al., 2004). In Erfurt, a Mobile Aerosol Spectrometer (MAS) was employed to measure the size distribution of particulate matter from September 1995 through August 2001. The MAS as well as details about the measurement campaign were described in more detail for instance by Wichmann, et al., 2000. It measured number concentrations of particles in 54 different channels, 9 in the ultrafine range, and 45 in the accumulation and fine modes. Mass concentrations of particles were computed assuming spherical particles of a constant density of 1.5 g/cm3, which was determined using concurrent PM2.5 mass measurements and the according daily spectra (Tuch, et al., 2000). These data are currently being analyzed using Positive Matrix Factorization. We use the number concentrations of particles in the ultrafine range and mass concentrations of particles in the accumulation range.
Due to the difficulties in obtaining elemental composition in particles, statistical analyses of the impact of sources on vascular and cardiac function have not begun.
We will analyze and compare the repolarization parameters as well as blood markers in the COPD panel to the results of the CAD patients (Specific Aim 2). After source apportionment is completed, we will use source indicators in epidemiological analyses assessing Specific Aim 3: particles from traffic and other combustion sources are associated with vascular and cardiac effects.
Kim E, Hopke PK, Larson TV, Covert DS. Analysis of ambient particle size distributions using UNMIX and positive matrix factorization. Environmental Science and Technology 2004;38:202-209.
Paatero P, Hopke PK. Utilizing wind direction and wind speed as independent variables in multilinear receptor modeling studies. Chemometrics and Intelligent Laboratory Systems 2002;60:25-41.
Phipps RP. Arteriosclerosis: the emerging role of inflammation and the CD40-CD40 ligand system. Proceedings of the National Academy of Sciences USA 2000;97(13):6930-6932.
Song XH, Polissar AV, Hopke PK. Sources of fine particle composition in the northeastern US. Atmospheric Environment 2001;25:5277-5286.
Tuch T, Mirme A, Tamm E, Heinrich J, et al. Comparison of two particle-size spectrometers for ambient aerosol measurements. Atmospheric Environment 2000;33(1):139-149.
Zhou L, Kim E, Hopke PK, Stanier C, et al. Advanced factor analysis on Pittsburgh particle size distribution data. Aerosol Science and Technology (in press, 2003).
Journal Articles on this Report : 1 Displayed | Download in RIS Format
|Other subproject views:||All 11 publications||11 publications in selected types||All 11 journal articles|
|Other center views:||All 104 publications||98 publications in selected types||All 90 journal articles|
||Kim E, Hopke PK, Larson TV, Covert DS. Analysis of ambient particle size distributions using Unmix and positive matrix factorization. Environmental Science & Technology 2004;38(1):202-209.||
Supplemental Keywords:autonomic control, electrocardiogram, epidemiology, inflammation, source apportionment, statistical methods, ultrafine particles,, RFA, Health, Scientific Discipline, Air, Geographic Area, particulate matter, Virology, Environmental Chemistry, Health Risk Assessment, Epidemiology, Risk Assessments, Biochemistry, Atmospheric Sciences, Molecular Biology/Genetics, International, ambient air quality, cytokine production, particle size, particulates, sensitive populations, cardiopulmonary responses, fine particles, human health effects, morbidity, ambient air monitoring, cardiovascular vulnerability, pulmonary disease, susceptible populations, COPD, epidemelogy, environmental health effects, particle exposure, Germany, human exposure, particulate exposure, lung inflamation, coronary artery disease, inhalation toxicology, PM, mortality, urban environment, aerosols, human health risk, cardiovascular disease, ultrafine particles
Progress and Final Reports:Original Abstract
Main Center Abstract and Reports:R827354 Airborne PM - Rochester PM Center
Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
R827354C001 Characterization of the Chemical Composition of Atmospheric Ultrafine Particles
R827354C002 Inflammatory Responses and Cardiovascular Risk Factors in Susceptible Populations
R827354C003 Clinical Studies of Ultrafine Particle Exposure in Susceptible Human Subjects
R827354C004 Animal Models: Dosimetry, and Pulmonary and Cardiovascular Events
R827354C005 Ultrafine Particle Cell Interactions: Molecular Mechanisms Leading to Altered Gene Expression
R827354C006 Development of an Electrodynamic Quadrupole Aerosol Concentrator
R827354C007 Kinetics of Clearance and Relocation of Insoluble Ultrafine Iridium Particles From the Rat Lung Epithelium to Extrapulmonary Organs and Tissues (Pilot Project)
R827354C008 Ultrafine Oil Aerosol Generation for Inhalation Studies