Grantee Research Project Results
Final Report: Epidemiological Studies on Extra Pulmonary Effects of Fresh and Aged Urban Aerosols from Different Sources
EPA Grant Number: R832415C002Subproject: this is subproject number 002 , established and managed by the Center Director under grant R832415
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
Center: Rochester PM Center
Center Director: Oberdörster, Günter
Title: Epidemiological Studies on Extra Pulmonary Effects of Fresh and Aged Urban Aerosols from Different Sources
Investigators: Peters, Annette , Utell, Mark J. , Zareba, Wojciech , Phipps, Richard , Wichmann, Heinz-Erich , Henneberger, Alexandra , Breitner, Susanne , Stoelzel, M , Rückerl, Regina
Institution: GSF-National Research Center for Environment and Health , University of Rochester
EPA Project Officer: Chung, Serena
Project Period: October 1, 2005 through September 30, 2010 (Extended to September 30, 2012)
RFA: Particulate Matter Research Centers (2004) RFA Text | Recipients Lists
Research Category: Human Health , Air
Objective:
The objective of the epidemiological study was to examine the effect of fine and ultrafine particles on systemic responses, endothelial and cardiac function.
The specific aims of the study were to:
- Aim #1: Determine the effect of ambient fine and ultrafine particles on an acute phase reaction in the blood of subjects with type 2 diabetes mellitus (T2D), impaired glucose tolerance (IGT) and potential genetic susceptibility (gen. susc.).
- Aim #2: Determine the effect of ambient fine and ultrafine particles on pro-thrombotic states of the blood in the above subject panels.
- Aim #3: Determine the effect of ambient fine and ultrafine particles on endothelial dysfunction as a key element of coronary vulnerability in a subset of the above subject panels.
- Aim #4: Determine the effect of ultrafine particles on cardiac function as characterized by ECG measures of autonomic function and repolarization in a subset of the above subject panels.
We extended aim #1, aim #2 and aim#3 and determined also the effects of air temperature. We extended aim#4 and determined also the effects of day-time noise.
Summary/Accomplishments (Outputs/Outcomes):
As part of the Rochester Particulate Matter Center investigations, a prospective panel study was conducted in Augsburg, Germany, between March 19th 2007 and December 17th 2008. Participants were recruited from the follow up examination of the KORA (Cooperative Health Research in the Region of Augsburg) survey 2000(Holle, Happich et al. 2005). The study consisted of three different subgroups: 1) individuals with type-2 diabetes (T2D), 2) individuals with an impaired glucose tolerance (IGT) indicating an enhanced risk of T2D yet without medication, 3) participants with a potential genetic susceptibility (Gen. susc.) on the detoxification or inflammation pathways. The potential genetic predisposition was defined by having the null polymorphism for GSTM1 and either two minor alleles of the single nucleotide polymorphism (SNP) rs1205 located in the C-reactive protein gene or at least one minor allele of the SNP rs1800790 located in the fibrinogen gene FGB. The individuals were invited to participate in up to seven visits each scheduled every four to six weeks on the same weekday and the same time of the day. At each visit (base program) a blood sample was withdrawn. A subset was invited to an add-on program that included the measurements of endothelial function, blood pressure, cardiac function as well as individual exposure measurements. The add-on program was conducted in up to four visits, each between 7:30 a.m. and 3 p.m. In this period participants pursued their daily routines. All their activities and whereabouts were recorded in a diary.
Ambient and individual exposure
Hourly data on ambient air pollution and meteorology were collected at an urban background monitoring site south of the city centre on an hourly basis. For each person and visit, the individual 24h-average of each air pollutant and meteorological variable preceding the visit (lag0: 0-23h before the examination) for up to 4 days (lag1: 24-47h, lag2: 48-71h, lag3: 72-95h, lag4: 96-119h) before the examination were calculated if more than two thirds of the hourly measurements were available for this period. Additionally, we calculated the mean of these 5 days (5-day average).
In the add-on program individual measurements of ultrafine particles using a portable condensation particle counter (CPC), of noise using a noise dosimeter as well as of air temperature and relative humidity using a data logger were conducted. Twentyfour hour and 5 min-averages were determined if at least 2/3 of the values in a 24h or 5min segment were available, respectively.
Statistical analyses
The associations between air pollution or air temperature and health outcomes were analyzed using additive mixed models with random participant effects. An appropriate covariance structure (either compound symmetry or first order autocorrelation) was chosen in order to account for dependencies between repeated measurements. A confounder model was built for each health outcome separately, adjusting for long-term (e.g. day of the study) as well as short- term (e.g. day of the week) time trends and meteorology (air temperature, relative humidity, barometric pressure). Penalized splines were used to allow for non-linear confounder adjustment. Model selection was done by minimizing Akaike's information criterion. Data were analyzed with SAS statistical package (version 9.2; SAS Institute Inc., Cary, NC, USA).
Results
Descriptives
In total, 275 individuals participated in 1,797 visits with a mean of 6.5 visits per person. One hundred and twelve persons participated in the add-on program completing 385 visits. The different number counts of valid examination parts are shown in Table 1.
Table 1. Number counts of valid examination parts
| Blood withdrawl | ECG recordings | Endothelia I Function | Blood pressure | Person CPC | Personal temperature | Person humidity | Personal noise | |
|---|---|---|---|---|---|---|---|---|
| Participants | 274 | 110 | 100 | 110 | 110 | 112 | 111 | 111 |
| Visits | 1,766 | 364 | 354 | 340 | 337 | 380 | 367 | 343 |
| For abbreviations see Appenix A3. | ||||||||
The study population consisted of elderly people, who were on average overweight. Nearly all participants were medicated. The panel of the gen. susc. persons was younger and had a lower BMI compared to the individuals with T2D or IGT. They also reported less accompanying cardiovascular diseases and took less medication (Table 2).
Table 2. Baseline description of the three panels of the study population
| TD20 N=83) | IGT (N=104) | Gen. susc. (N=88) | po-va | ||||
|---|---|---|---|---|---|---|---|
| Mean | (SD) | Mean | (SD) | Mean | (SD) | ||
| Age (years) | 67.5 | (7.3) | 65.6 | (9.2) | 55.9 | (12.1) | <.0001b |
| BMI (kg/m2)a | 31.6 | (5.0) | 29.6 | (5.6 | 26.9 | (5.0) | <.0001b |
| N | (%) | N | (%) | N | (% | ||
| Men | 49 | (59.0) | 59 | (56.7) | 49 | (55.7) | 0.90c |
| Smoking | |||||||
| Never Smoker | 32 | (38.6) | 57 | (54.8) | 44 | (50.0) | |
| Ex-Smoker | 48 | (57.8) | 46 | (44.2) | 39 | (44.3) | 0.08d |
| Occassional Smoker | 3. | (3.6) | 1 | (1.0) | 5 | (5.7) | |
| History of | |||||||
| Coronary Heart Disease | 7 | (8.0) | 0.65C | ||||
| Myocardial infarction | 9 | (10.8) | 11 | (10.6) | 3 | (3.4) | 0.042c |
| Hypertension | 68 | (81.9) | 67 | (64.4) | 35 | (39.8) | <.0001c |
| Angina Pectorisa | 6 | (7.3) | 9 | (8.7) | 3 | (3.4) | 0.31d |
| Medication Use | |||||||
| Agents acting on rennin angiotensin-system | 54 | (61.5) | 54 | (39.4) | 21 | (23.9) | <.0001c |
| Antidiabetics | 47 | (56.6) | 3 | (2.9) | 2 | (2.3) | <.0001d |
| Antihypertensives | 66 | (79.5) | 61 | (58.7) | 31 | (35.2) | <.0001c |
| Antiinflammatory agents | 22 | (26.5) | 27 | (26..0) | 12 | (13.6) | 0.06c |
| Beta Blockers | 39 | (47.0) | 30 | (28.9) | 15 | (17.1) | 0.0001c |
| Calcium channel blockers | 14 | (16.9) | 19 | (18.3) | 7 | (8.0) | 0.10c |
| Diuretics | 42 | (50.6) | 35 | (33.7) | 14 | (15.96) | <.0001c |
| Nitrates | 4 | (4.8) | 2 | (1.9) | 0 | (0) | 0.10d |
| Statins | 31 | (35.4) | 23 | (22.1) | 13 | (14.8 | 0.002c |
| No medication use at all | 2 | (2.4) | 7 | (6.7) | 18 | (20.5) | <.0001d |
| Being employed | 14 | (16.9) | 30 | (28.9 | 52 | (59.1) | <.0001c |
| a For one participants the value is missing; p-value for hertogeneity between panels determind by b ANOVA, c Chi-square test, d Fisher's exact test. For appreviations see Appendix A3 | |||||||
Table 3 shows, for each panel separately, a description of the outcome variables which have been used for the different study aims. Mean levels of all blood markers differed significantly between the panels. For the other outcome variables there was only a significant difference between the panels for pulse pressure and the standard deviation of all normal-to-normal intervals (SDNN). A description of ambient and individual air pollutants and meteorological variables measured during the study period is shown in Table 4 and Table 5.
Table 3. Description of repeated measurements of blood markers, blood pressure, and ECG parameters
| T2D | IGT | Gen. Susc. | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| N | Mean | (SD) | N | Mean | (SD) | N | Mean | (SD) | p-valueb | ||
| Blood markers | |||||||||||
| IL-6 (pg/ml) | 522 | 2.1 | (5.5) | 675 | 1.3 | (0.9) | 569 | 1.0 | -1.0 | <0.0001 | |
| CRP (ng/ml) | 522 | 2.9 | (7.4) | 675 | 2.2 | (3.8) | 569 | 1.4 | (2.3) | <0.0001 | |
| MPO (ng/ml) | 522 | 16.2 | (7.8) | 675 | 16.2 | (6.8) | 569 | 14.5 | (12.3) | 0.002 | |
| CD40L (ph/ml) | 522 | 764.4 | (532.4) | 675 | 797.1 | (488) | 569 | 1,000.8 | (773.1) | <0.0001 | |
| Fibrinogen (g/l) | 521 | 3.7 | (0.8) | 675 | 3.7 | (0.6) | 569 | 3.3 | (0.5) | <0.0001 | |
| PAI-1 (ng/ml) | 522 | 4.6 | (3.7) | 675 | 4.7 | (3.7) | 569 | 3.6 | (3.1) | <0.0001 | |
| Blood pressure | |||||||||||
| SBP (mmHg) | 110 | 133.1 | (16.8) | 109 | 134.0 | (19.7) | 152 | 126.6 | (17.9) | 0.12 | |
| DBP (mmHg) | 110 | 79.1 | (8.5) | 109 | 81.0 | (11.9) | 152 | 79.3 | (10.2) | 0.61 | |
| pp (mmHg) | 110 | 54.0 | (13.1) | 109 | 53.0 | (12.6) | 152 | 47.3 | (11.9) | 0.02 | |
| 1h-ECG intervalsa | |||||||||||
| HR (beats/min) | 597 | 77.3 | (14.4) | 606 | 80.8 | (14.1) | 880 | 77.7 | (11) | 0.28 | |
| T-wave amplitude (µv) | 585 | 290.1 | (108.7) | 600 | 319.3 | (141.9) | 875 | 364.0 | (134.5) | Infinate Likelihood | |
| T-wave complexity (%) | 597 | 17.0 | (8.4) | 606 | 17.9 | (6.4) | 880 | 18.7 | (9.7) | 0.51 | |
| QTc (ms) | 597 | 440.0 | (26.1) | 606 | 447.0 | (21.8) | 880 | 436.3 | (23.8) | 0.08 | |
| SDNN (ms) | 597 | 77.5 | (26) | 606 | 76.0 | (28.2) | 879 | 85.0 | (30) | 0.04 | |
| RMSSD (ms) | 597 | 32.6 | (32.2) | 606 | 35.2 | (31.2) | 880 | 27.4 | (16.1) | 0.32 | |
| SDNN (ms) | 597 | 77.5 | (26) | 606 | 76.0 | (28.2) | 879 | 85.0 | (30) | 0.04 | |
| LF (ms) | 595 | 6.3 | (7.1) | 606 | 9.2 | (10.7) | 879 | 12.8 | (12.2) | <0.0001 | |
| HF (ms) | 595 | 4.4 | (9.8) | 606 | 5.1 | (8.3) | 879 | 3.6 | (4.3) | 0.57 | |
| LF/HF Ratio | 595 | 3.0 | (2.4) | 606 | 3.4 | (2.4) | 879 | 4.7 | (3.2) | 0.001 | |
| a Descriptive statistics for five-minute intervals were similar. b P-value of fixed groujp effect in mixed effects model. For abbreviations see Appendix A3 | |||||||||||
Table 4. Description of ambient air pollutants and meteorological variables (March 14th, 2007 to December 17th, 2008)
| average | N | Mean | SD | Min | 25% | Median | 75% | Max | IQR | |
|---|---|---|---|---|---|---|---|---|---|---|
| PM10 (µg/m3) | 1h 24h | 15,466 645 | 19.3 18.3 | 14.1 12.0 | 0.0 2.0 | 8.4 10.1 | 15.3 15.8 | 24.4 24.0 | 159.8 86.5 | 16.0 14.0 |
| PM2.5 (µg/m3) | 1h 24h | 15,461 644 | 13.7 13.7 | 11.2 10.0 | 0.0 1.6 | 5.8 6.7 | 10.9 11.3 | 18.1 17.8 | 106.5 65.8 | 12.3 11.1 |
| UFP (n/cm3) | 1h 24h | 14,699 611 | 9,516 9,537 | 6,902 4,417 | 937 1,879 | 4,892 6,305 | 7,629 8,890 | 12,049 12,027 | 80,858 26,503 | 7,157 5,722 |
| ACP | 1h 24h | 14,699 611 | 2,060 2,068 | 1,535 1,213 | 88 291 | 1,020 1,179 | 1,657 1,855 | 2,615 2,712 | 17,377 8,120 | 1,595 1,533 |
| BC (µg/m3) | 1h 24h | 13,359 550 | 1.8 1.8 | 1.5 1.1 | 0.3 0.4 | 0.9 1.1 | 1.3 1.5 | 2.1 2.2 | 21.4 7.3 | 1.2 1.2 |
| Ozone (µg/m3) | 1h 24h | 15,429 641 | 45.9 45.9 | 33.3 22.7 | 3.0 3.0 | 15.0 27.2 | 44.0 49.5 | 69.0 62.6 | 158.0 97.6 | 54.0 35.4 |
| Air temperature (oC) | 1h 24h | 15,398 636 | 10.8 10.9 | 7.9 7.3 | -8.4 -5.8 | 4.7 5.0 | 10.8 11.0 | 16.5 17.0 | 33.8 27.0 | 11.8 12.0 |
| Relative Humidity (%( | 1h 24h | 15,398 636 | 76.9 77.0 | 18.3 12.6 | 21.0 32.4 | 63.3 68.1 | 81.3 77.6 | 92.8 86.9 | 100.0 100.0 | 29.5 18.8 |
| Barometric pressure (hPa) | 1h 24h | 15.398 636 | 961.2 961.3 | 7.9 7.6 | 927.8 933.9 | 956.8 957.1 | 961.4 961.3 | 965.9 965.9 | 985.6 983.5 | 9.1 |
| For abbreviations see Appendix A3 | ||||||||||
Table 5. Description of individual air pollutants and meteorological variables (April 23rd, 2007 to December 9th, 2008)
| average | N | Mean | SD | Min | 25% | Median | 75% | Max | IQR | |
|---|---|---|---|---|---|---|---|---|---|---|
| PNC (n/cm3, individual) | 5min visit | 21,918 337 | 21,347 21,175 | 38,086 18,272 | 524 2,947 | 6,162 11,780 | 10,882 17,145 | 21,702 24,965 | 697,779 248,083 | 15,540 13,186 |
| Air temperature (oC; individual) | 5min visit | 28,402 380 | 21,5 21,5 | 3,9 3,0 | -0,3 12,4 | 19,6 19,7 | 21,9 21,6 | 23,.9 23,6 | 36,2 31,1 | 4,3 3,9 |
| Relative humidity (%,individual | 5min visit | 27,350 367 | 44,8 44,9 | 11,2 9,0 | 10,3 21,6 | 36,7 38,7 | 44,4 44,9 | 51,4 50,9 | 11,0 69,8 | 14,7 12,1 |
| Noise (dB(A), individual) | 5min visit | 25,227 343 | 74,7 74,7 | 82,6 79,5 | 37,0 61,1 | 60,8 68,2 | 66,6 70,0 | 71,5 72,8 | 98,9 91,1 | 71,1 71,0 |
| For abbreviations see Appendix A3 | ||||||||||
Aim #1 and aim#2: Health effects on blood markers
As blood markers of inflammation we considered high sensitive C-reactive protein (hsCRP), interleukin 6 (IL-6), and myeloperoxidase (MPO). As blood markers of coagulation we investigated high sensitivity soluble CD40Ligand (sCD40L), plasminogen activator inhibitor-1 (PAI-1). Fibrinogen is a marker of both, inflammation and coagulation.
Regression results for effects of air pollutants
The first analyses were conducted taking participants with T2D and IGT together in one group based on the recommendations by the scientific advisory board. In secondary analyses, we evaluated the effects in the three panels separately. Effects are presented as percent change of the outcome mean per interquartile range (IQR) increase in air pollutant together with 95% confidence intervals (CI).
Table 6 sums up the results for the association between 24-hour means of air pollutants on blood markers. The results show clear differences between persons with T2D or IGT and gen susc. individuals, with stronger associations for the latter ones. Except for sCD40L, associations also point in different directions for both groups.
Table 6. Overview on results of associations between air pollutants and blood markers.
| T2D or IGT | Gen Susc | |||
|---|---|---|---|---|
| Blood marker | Association | Time scale | Association | Time scale |
| IL-6 | ↑ | delayed | ↓ | delayed |
| hsCRp | -- | n.a. | ↑ | immediate and delayed |
| MPO | ↓ | immediate | ↑ | Immediate and delayed |
| sCD40L | ↓ | immediate | ↓ | immediate |
| PAI-1 | -- | n.a. | ↓ | immediate |
| Fibrinogen | ↑ | delayed | -- | n.a. |
| ↑ significant positive associations | ↑ mainly non-significant positive associations | |||
| ↓ significat negative associations | ↓ mainly on-significan negative data | |||
| For abbreviations see Appendix A3 | ||||
The clearest association was found for hsCRP and MPO in gen. susc. individuals. For hsCRP, positive associations with air pollutants were seen for almost all lags. The pattern was similar, however not as strong, for MPO (Figure 1). The coagulation markers sCD40L and PAI-1 showed an immediate decrease in association with particles, especially with PM2.5. No association was seen for later lags in gen. susc. persons (data not shown). Results for the group of individuals with T2D or IGT showed small and mostly non-significant estimates. However, separate analyses for the panel with IGT and the panel with diabetic individuals led to significant increases only in persons with IGT as shown for fibrinogen in Figure 2.
Figure 1 Associations between PM2.5 and UFP and CRP and
MPO, respectively, in gen. susc. persons.
Figure 2. Associations between PM2.5 and fibrinogen in
individuals with IGT and/or T2D.
For abbreviations see Appendix A3
Title: Short-term effects of air temperature on blood markers of coagulation and inflammation in potentially susceptible individuals (Schäuble et al. 2012)
Figure 3 shows the association between a 5°C air temperature decrement and blood markers. We observed immediate, lagged and cumulative increases in fibrinogen and PAI-1 in association with a drop of air temperature only in participants with T2D or IGT. A 5°C decrease in the 5-day average of temperature led to the strongest increase in fibrinogen (0.8% [0.3, 1.3%]) and PAI-1 (10.1% [6.4, 13.8%]), respectively. We observed effects of a 5°C temperature decrease on IL-6 with a lag of one (6.2% [0.4, 12.4%]) up to 4 days (5.9% [0.3, 11.9]) as well as for the 5-day average (8.0% [0.5, 16.2%]) only in gen. susc. participants. In the same panel, hsCRP decreased significantly with almost all lags and the 5-day average (-9.5% to -6.5%) in association with a 5°C temperature decrement, whereas we found no temperature effects on hsCRP in participants with T2D or IGT.
Figure 3. Air temperature effects on blood markers.
For abbreviations see Appendix A3
Discussion
Our analyses of associations between air pollutants and blood markers confirm the hypothesis that oxidative stress plays a role in the mechanism linking air pollution and cardiovascular disease. Furthermore, the changes in sCD40Ligand and PAI-1 point towards the coagulatory system. However, results are not quite conclusive as sCD40L promotes thrombus formation while PAI-1 inhibits fibrinolysis. Both markers showed a decrease for the same time lag. One explanation could be that this reaction is a case of regulation and counter-regulation in the very tightly regulated system of coagulation.
Regarding the panel of persons with T2D or IGT we found substantially reduced effects for the blood markers compared to the panel of gen. susc. persons. Also, the results clearly differ from the results of the genetically susceptibles. One can therefore assume that there are different mechanisms involved in those two groups. This might either be due to the underlying disease, the medication intake or both. Moreover, separate analyses showed that the associations were mostly driven by the panel of people with impaired glucose tolerance, which consists of patients with the same underlying disease, yet without the heavy medication. This confirms our initial hypothesis that medication can blunt the small associations seen between air pollution and markers for cardiovascular diseases.
Finally, the effects were mostly seen for PM2.5 and less for UFP, and most clearly for inflammatory markers. We assume that the reason for this is the inflammatory potential of particulate mass, which possibly derives from secondary organic aerosols.
The physiological mechanisms leading from temperature changes to cardiovascular events are still not clearly established. Exposure to cold can have various effects like an increase of blood pressure and heart rate (Keatinge, Coleshaw et al. 1984; Alperovitch, Lacombe et al. 2009; Hess, Wilson et al. 2009) as well as changes in blood markers. Elderly individuals seem to be more susceptible to cold temperature compared to younger individuals because it has been reported that they showed a higher increase of blood pressure after superficial skin cooling(Hess, Wilson et al. 2009) and higher longitudinal changes in blood pressure regarding seasonal temperature changes(Alperovitch, Lacombe et al. 2009). Earlier studies suggested that controlled exposure to cold air or water led to an increase in blood markers like fibrinogen(De Lorenzo, Kadziola et al. 1999) and PAI-1.(Neild, Syndercombe-Court et al. 1994) Furthermore, it has been shown that a short-term decrease in temperature was associated with increased levels of fibrinogen, IL-6 and CRP.(Schneider, Panagiotakos et al. 2008) Blood markers like fibrinogen, IL-6 and CRP(Koenig, Khuseyinova et al. 2006; Ridker 2009; Rana, Arsenault et al. 2011) have been established as predictors of cardiovascular risk and IL-6 and CRP seem to be more associated with fatal than nonfatal cardiovascular events.(Sattar, Murray et al. 2009) Several blood markers like fibrinogen, IL-6 and CRP showed seasonal variations with increases mostly in the cold season(Elwood, Beswick et al. 1993; Woodhouse, Khaw et al. 1994; Sung 2006; Rudnicka, Rumley et al. 2007; Kanikowska, Sugenoya et al. 2009) and might therefore be one possible link to the peak of cardiovascular events in winter. However, in our study temperature effects were inconsistent in participants with T2D or IGT and healthy participants with a special genetic background. Participants with T2D or IGT were older, more often overweight, more strongly medicated whereupon participants with T2D took more statins and antidiabetics compared to participants with IGT, and had a worse overall health than healthy participants with a special genetic background. A combination of these factors might have influenced the susceptibility to temperature.
Aim #3: Health effects on endothelial function and blood pressure
Markers of endothelial dysfunction (reactive hyperemia index, augmentation index, and time of reflection) were determined with the Endo-PAT2000 device (Itamar Medical, Israel) in subjects who participated in the add-on program. Our analyses of air pollution effects on markers of endothelial dysfunction showed conflicting associations across the analyzed parameters. Since this was the first time that we used the Endo-PAT2000 device in an epidemiological study we were not able assess the validity of these parameters. However, blood pressure (BP) was measured in the same subjects. Pulse pressure (PP) was calculated as difference between systolic (SBP) and diastolic blood pressure (DBP). PP is a marker of arterial stiffness which is assumed to be correlated with endothelial dysfunction. Therefore, we investigated the effects of ambient air pollutants as well as of ambient and personal temperature measurements on SBP, DBP, and PP.
Regression results for effects of air pollution measures
In each panel, we observed only weak ambient air pollution effects on BP. Results showed mainly decreases in blood pressure with strongest effects for lag0 of air pollutant. However, associations were mainly non-significant. The same was true for PP (data not shown).
Title: Short-term effects of air temperature on blood pressure and pulse pressure in potentially susceptible individuals (Lanzinger et al.)
We found immediate, lagged and cumulative ambient air temperature effects on SBP, DBP, and PP in individuals with T2D but not in persons with IGT or in healthy individuals (Figure 4). A 5°C decrease in ambient air temperature was associated with an immediate (lag0) 3.8% [1.9;5.6%] increase in SBP in participants with T2D. Accordingly, we observed an increase in DBP by 2.0% [0.5;3.6%] in association with a 5°C decrement in ambient air temperature for lag0. Furthermore, we also found immediate temperature effects on PP, showing an increase by 6.5% [2.9;10.0%] in association with a 5°C decrease. Ambient air temperature effects were similar for all time lags. The analysis of personally measured air temperature led to similar results (Figure 4). Effect modification analyses (data not shown) showed that a decrement in ambient air temperature led to an increase in SBP only in participants without intake of antihypertensive medication (5.4% [2.9;7.8%]) or diuretics (5.0% [2.8;7.3%]). Interaction effects of gender on SBP only reached borderline significance.
In general, we detected similar effect modifications for PP. However, interaction results for effects of gender on PP showed significantly stronger increases in men (8.2%[4.5;11.9%]). Moreover, ambient air temperature effects on PP were stronger in participants who spent ≥ 70% of the time indoors (9.1%[4.8;13.4%]). Furthermore, with a lag of one day, a decrease in ambient air temperature was associated with an increase in SBP (7.0% [4.4;9.6%]), DBP (4.0% [1.8;6.2%]) and PP (11.7% [6.7;16.7%]) only in individuals with a BMI ≥ 30 kg/m2. Effect modifications tended to be similar for personally measured air temperature (data not shown). However, we additionally found a 9.3% [4.6;14.0%] increase in SBP and a 14.4% [5.5;23.3%] increase in PP, respectively, in subjects aged ≥ 60 years. Younger subjects showed no changes in SBP and PP in association with temperature decreases
Figure 4. Associations between air temperature and blood pressure
as well as pulse pressure.
For abbreviations see Appendix 3
Discussion
Biological mechanisms explaining the relationship between air temperature and BP or PP are not well understood. It is assumed that several factors are involved in the increase of BP as the regulation of BP is complex. For example, decrements in air temperature activate the sympathetic nervous system leading to an increase in heart rate and a consecutive rise in BP. Cold induced activation of the sympathetic nervous system could also result in an elevated BP through vasoconstriction of the blood vessels or impaired endothelial- dependent vasodilatation.
Especially individuals with diabetes might be particularly susceptible. Vascular abnormalities like endothelial dysfunction are very common in these individuals. Amongst others, endothelial cells synthesize nitric oxide (NO) which is important for vasodilatation and protection of the blood vessels from injuries. Many subjects with diabetes have decreased levels of NO indicating an abnormal endothelium-dependent vasodilatation. Accordingly, people with diabetes are assumed to be more susceptible to the described mechanisms resulting in stronger temperature effects. Since we observed no temperature effects in subjects with IGT we hypothesize that vascular abnormalities are not that common in individuals with a pre-diabetic state leading to a potentially lower risk of endothelial dysfunction. These findings may provide an indication of the relation between lower air temperature and short term increases of cardiovascular events.
Aim#4: Associations between air pollutants, noise and ECG parameters
Regression results for effects of personally measured PNC and ambient air pollution on heart rate and heart rate variability parameters in individuals with T2D and IGT
In our analyses we observed effects of 5-minute averages of personally measured PNC on 5- minute intervals of heart rate (HR) as well as standard deviation of normal-to-normal intervals (SDNN) in 32 participants with T2D and 32 participants with IGT. In particular, we detected a concurrent -0.56% [-1.02;-0.09%} decrease in SDNN in association with an increase of 16,000 personal PNC particles/cm³. We conducted several sensitivity analyses in order to test the robustness of these findings. Effects of personal PNC on SDNN did not change when excluding participants taking statins or additionally adjusting for ambient 1h-averages of PM2.5. We observed slightly weaker PNC effects when excluding participants with an intake of beta- blockers or visits with cooking. However, PNC effects were stronger in participants who were not exposed to environmental tobacco smoke or when adjusting for personally measured 5min- averages of noise exposure (Table 7). A manuscript including these results is currently in preparation (Peters et al.).
Table 7. Concurrent changes in SDNN per 16,000 particles/cm3 increase in personally measured PNC
| %-change | (95%-CI) | |
|---|---|---|
| Main effect* | -0.56 | (-1.02:0.09) |
| without persons with an intake of beta blockers | -0.47 | (-0.98:0.04) |
| without persons with an intake of statins | -0.58 | (-1.07:0.10) |
| without visits with ETS | -0.66 | (-1.14:0.18) |
| without visits with cooking | -0.48 | (-1.03:0.10) |
| + adjustment for ambient PM2.5 (1h-average) | -0.56 | (-1.03:0.10) |
| + adjustment for personal noise (5min-average) | -1.20 | (-1.82:0.57) |
| * adjusted for long-term time-trend, ambient air temperature, and ambient relative humidity. For abbreviations see Appendex A3 | ||
Title: Acute air pollution effects on heart rate variability and effect modifications by SNPs involved in cardiac rhythm (Hampel et al. 2012a)
We assessed air pollution effects on HR and heart rate variability (HRV) on an hourly basis in individuals with T2D or IGT. Moreover, we investigated potential effect modifications by single nucleotide polymorphisms supposed to be involved in cardiac rhythm.
SNP selection
In a first step, we conducted a literature research and identified 139 SNPs which have already been shown to modulate repolarization and HRV parameters. In order to determine the influence of SNPs on ECG parameters we used regression trees for longitudinal data implemented in the R package REEMtree (Sela and Simonoff, 2010, R package version 0.82). This method alternates between estimating the regression tree, assuming that the previously estimated random effects of a mixed model are correct, and estimating the random effects, using the information of the regression tree performed in the prior step. For each ECG parameter we estimated regression trees for longitudinal data always excluding one single ECG recording in order to check the robustness of the trees. Seven, fourteen, and eleven SNPs were chosen using the tree selection procedure for HR, RMSSD, and SDNN, respectively as they occurred at least in 75% of all trees. These SNPs were used as potential air pollution effect modifiers.
Regression results
Increased black carbon (BC) levels were only marginally associated with an increase in HR (0.9%[0.0;1.8%]) with a lag of 6h. We observed an association between increases in PM10 and PM2.5 and a concurrent reduction in RMSSD (-5.3% [-9.3;-1.1%] and -7.2% [-12.2;-1.8%], respectively). Furthermore, RMSSD changed by -3.8% [-7.1;-0.5%] and -5.2% [-9.8;-0.4%] in association with elevated BC levels with a lag of 1h and 6h, respectively. Elevated PM2.5 level led to concurrent (-3.3% [-6.0;-0.7]) and lagged decreases in SDNN by about 34%. We observed similar effects of PM10 and sulfates. Increases in BC and UFP were only related with lagged decreases in SDNN showing the strongest associations with a lag of 2h (-3.7% [-5.6;-1.8%] and 1.9% [-3.4;-0.4%], respectively). We observed no significant air pollution effects on HR; therefore, we did not calculate effect modifications by SNPs for this ECG parameter. As PM2.5 showed the strongest main effects on ECG parameters and PM variables were highly correlated with BC and sulfate but uncorrelated with UFP, we only present PM2.5 and UFP effect modifications by SNPs. rs333229 was the strongest effect modifier on SDNN (Figure 5).
Figure 5. Concurrent effects of 1 h-average of air pollutants on 1 h-averages on SDNN.
For abbreviations see Appendix A3.
Throughout all lags only participants with at least one minor allele showed a reduction in SDNN in association with increases in PM2.5. Individuals with no minor allele did not react to elevated PM2.5 levels. We observed the strongest modification with a lag of 4h (p-value of interaction=0.01). Participants with one and two minor alleles exhibited a -6.6% [-10.3;-2.8%] and a -12.9% [-20.6;-5.1%] decreased SDNN, respectively. PM2.5 effect modification by rs2966762 resulted in a similar pattern with weaker effects. Borderline and significant interaction effects between rs333229 and UFP on SDNN were detected with a lag of 2h to 5h. SDNN decreased by about 2-3% in individuals with one and about 4-6% in individuals with two minor alleles. Furthermore, an increase in PM2.5 led to a 4-8% and to a 2-4% decrease in SDNN in participants with no or one minor allele of rs1871841. The concurrent response of RMSSD to increases in PM2.5 was modified by rs2096767 and rs2745967. Elevated PM2.5 levels led to a -10.0% [-15.5;-4.1%] and a -13.2% [-23.3;-1.8%] reduction in RMSSD in individuals with one and two minor alleles of rs2096767, respectively. In contrast, people with no (-13.2% [-20.3;-5.6%]) and one minor allele (-6.4% [-11.5;-1.0%]) in rs2745967 exhibited a decrease in RMSSD in association with PM2.5. Analyzing air pollution effects on RMSSD and SDNN for participants with T2D and IGT separately showed that significant main and interaction effects were only observed in individuals with IGT.
Title: Immediate Ozone Effects on Heart Rate and Repolarization Parameters (Hampel et al.
2012b)
We investigated the associations between 1h-averages of ozone and HR, Bazett-corrected QT-interval (QTc), T-wave amplitude (Tamp) and T-wave complexity (Tcomp) in 64 3 individuals with T2D or IGT and in 46 healthy 2 individuals with a potential genetic 1 predisposition. Overall, more than 2000 1h-ECG intervals were available for the analyses.
Figure 6 shows the percent changes of the mean ECG parameters per 20µg/m³ increase in ozone together with 95%-CI. Elevated ozone levels led to a marginal concurrent 0.54% [95%-CI: -0.07;1.16%] increase in HR and to 1h4h delayed increases about 0.6-0.7% in all participants showing the strongest association with a 2h lag (0.78% [0.18;1.37%]). Ozone effects were more pronounced in individuals with T2D or IGT for all lags showing percent changes in HR about 1%. HR did not change due to ozone exposure in individuals with a potential genetic predisposition. We observed concurrent and 1h lagged T-wave flattening of -1.31% [-2.19;-0.42%] and -1.32% [-2.19;-0.45%] associated with elevated ozone levels, respectively. This association was predominantly detected in participants with T2D or IGT (concurrent and 1h lagged ozone effects: -1.99% [-3.20;-0.78%] and -2.14% [-3.32;-0.96%]), whereas healthy subjects showed a weaker reduction in Tamp. Furthermore, Tcomp increased in association with elevated ozone levels among all participants by 2.12% [0.81; 3.52%] and 1.89% [0.55; 3.26%] with a lag of 1h and 2h, respectively. Again, only participants with metabolic disorders reacted to ozone exposure. No ozone effects were seen for QTc in either subgroup.
Title: Acute effects of individual day-time noise exposure on heart rate variability (Kraus et al. 2012)
Epidemiological studies demonstrate associations between noise exposure and cardiovascular events. However, possible underlying mechanisms are poorly investigated, which is why we examined the association between individual day-time noise exposure and HR as well as HRV. Individual noise exposure was measured as A-weighted equivalent cont
2006 Progress Report
2007 Progress Report
2008 Progress Report
2009 Progress Report
2010 Progress Report
2011 Progress Report
Conclusions:
CONCLUSION
Relationship to the overall Center goals
By looking at traffic-related PM and aged aerosols characterized at an urban background site our epidemiological study contributes to the center's goal of identifying sources with higher toxicity and pathophysiological mechanisms by which ambient ultrafine and fine PM trigger cardiovascular adverse health effects.
The study was conducted in adults with diagnosed T2D or IGT which have been hypothesized to be a susceptible subpopulation with respect to PM responses.
Relevance to the Agency's mission
The results of our study highlight that personal exposures are adding important information and that general air quality conditions together with behavior-dependent exposure jointly affect health. The research indicated further that persons with metabolic disorders as well as healthy individuals with underlying genetic susceptibility may be affected by ambient particles. The studies further more highlight that regionally distributed as well as locally produced pollution are relevant targets to protect the public. Mass-based metrics such as PM2.5 and PM10 alone are not fully capturing the health relevant properties of ambient aerosols. Reducing emissions and formation of ultrafine particles may be considered a new targeted approach to regulations that reduce the amounts of PM that most directly affects health rather than the current mass-based standard. Using real-world ultrafine and fine PM, we provided data that can be used to answer the question if there is a need for an ultrafine PM standard in addition to the fine PM standard. With regard to temperature health effects, our results show that in our study area it is rather a drop in temperature than heat that is associated with cardiovascular effects. Climate change is supposed to not only increase mean temperature but to also lead to more severe temperature variation, including sudden temperature decreases. Therefore, adequate measures for climate change adaptation as well as mitigation are needed.
Potential practical applications
This research provides evidence that the average air quality in a region as well as individual behaviors determine the personal exposures. Therefore, subjects at risk may be educated to reduce the individual impact of particles in a similar manner as advice on reduction of ozone exposures. The potential synergistic effects with heat or cold exposures should be considered in prevention campaigns in the future. We provide evidence to adequately protect subpopulations potentially susceptible to air pollution exposure or temperature changes in the future. Moreover, our results will add to the body of evidence that will be the basis for the planned setting up of new air quality standards in the United States as well as in Europe.
Concluding remarks:
- Our analyses of associations between air pollutants and blood markers confirmed the hypothesis that oxidative stress plays a role in the mechanism linking air pollution and cardiovascular disease.
- Results substantially differed between persons with type 2 diabetes or impaired glucose tolerance and genetically susceptible participants, indicating that there might be different biological mechanisms ongoing.
- Effects were mostly seen for PM2.5 and less for ultrafine particles, and most clearly for inflammatory markers. One can assume that the reason for this is the inflammatory potential of particulate mass, which possibly derives from secondary organic aerosols.
- We observed differing temperature effects on blood markers in persons with metabolic disorder and genetically susceptible individuals which again probably indicates different underlying biological mechanisms with regard to systematic responses.
- We observed associations between decreases in air temperature and increases in blood pressure as well as in pulse pressure mainly in persons with type 2 diabetes, possibly due to vascular abnormalities like endothelial dysfunction which are very common in these individuals.
- Our study showed changes in ECG measures associated with personally measured particle number concentration and centrally monitored air pollutants suggesting that both freshly emitted traffic particles as well as aged aerosol in urban areas are associated with changes in cardiac rhythm.
- Our results suggested that certain polymorphisms in persons with type 2 diabetes or impaired glucose tolerance make them potentially more susceptible to air pollutants with regard to changes in heart rate variability.
- Individual day-time noise exposure was associated with immediate changes in heart rate variability. Thereby, noise at lower levels led to a parasympathetic withdrawal while changes in high noise levels were rather associated with a sympathetic activation in terms of a "fight-or-flight" response.
APPENDIX (FOR CORE 2):
A1 Quality Assurance
A1.1 Type of Data & Data Structure
Raw data from questionnaires and exposure data were available in electronic forms. Data regarding the different procedures (e.g. blood draw, starting and ending of measurement devices) as well as the activity diary of participants were documented on paper/pencil forms. Electronic raw data were imported directly into the statistical software package SAS while paper/pencil data were first double-entered in electronic database by two different assistants. Double-entries were compared and corrected if necessary.
Data are structured in four consecutive levels (Table A1.1). Each level includes different procedures so that all changes can be reproduced beginning from level 1.
Table A1.1. Description of the data structure in SAS
| Level | Content/Procedure |
|---|---|
| Level 2: Raw | Import of data into SAL; Labels and formats were assigned. Observatins without values were marked as "non applicable" or "missing" |
| Level 2: Checked | Data were checked for plausibility (including the comparison of double entered paper/pencil data) and corrected if necessary |
| Level 3: New variables & exclusions | Preparing dataset by including new variables relevant for analyses and excluding observations inelibile for analyses |
| Level 4: Final datasets | Preparing final dtasets composed of level 3 datasets |
A1.2 Data Validation
All data has been checked for plausibility and validated before preparing datasets used for analyses (level 2). In general, checks included:
- check for uniqueness (no ID twice) and completeness,
- check of panel classification
- check of electronic notes and any remarks on the paper protocols,
- check of differences between visit dates and times,
- check of number of visits per participant,
- check of extreme values of continuous variables,
- check of values of categorical variables,
- check of "if yes, then" variables,
- check of variables being contradictory,
- check for missing values.
Changes to and corrections of the data have been documented and all drop outs were listed including patient number and reason for drop out. Visits were scheduled to be on the same weekday and approximately the same time of day for each participant. However, 22 out of 1797 visits did not meet the weekday criterion (1.2%), and 18 blood withdrawals were not taken at the same time of day (1.0%). In the analyses we will account for these deviations from the study design by including the variables day of the week and time of day as confounder variables.
A1.3 Duplicate blood samples
To test the variability in the lab procedures we obtained throughout the study period 40 duplicate blood samples for CRP, MPO, sCD40L, and PAI-1 as well as 28 and 39 duplicates for IL-6 and fibrinogen, respectively. These duplicates were blinded for the lab personnel. We determined absolute and relative differences between the duplicate blood samples. Median relative differences were 6.16% (range: 0.85 to 51.43%) for IL-6, 7.61% (0.32 to 341.77%) for CRP, 14.94% (2.16 to 199.34%) for MPO, 13.49% (0.56 to 127.26%) for sCD40L, 2.93% (0.32 to 12.38%) for fibrinogen, and 8.65% (0.00 to 110.17%) for PAI-1. Because of the quite small medians of relative differences we can assume a good determination of the blood markers.
A1.4 Data Backup
During the field phase electronic data that were made anonymous were stored regularly on the server of the Helmholtz Zentrum München where data are daily saved by the computing center. In addition, data were saved twice to DVD. Paper/pencil forms were filed in folders which were stored in lockable rooms of the Helmholtz Zentrum München. After completion of data management all electronic data (e.g. correspondence per email, miscellaneous documents, and final datasets) was stored to DVD several times which will be distributed to different persons including the principal investigator of the study. All paper/pencil forms will be stored for 10 years.
A2 Pilot study on the impact of wood combustion on the cardiac function (Hampel et al.)
A2.1 Study design
In a sub-agreement (414278-001G) we received additional funding from the University of Rochester in order to perform a pilot study on wood combustion and cardiac function (final report submitted in 2009).
Ten participants living in Augsburg were recruited from the KORA (Cooperative Health Research in the Region of Augsburg) F4 study. Between March 4 and 11, 2008 each participant was fitted once in the afternoon (around 3pm) for up to 23 hours with a long-term Holter 7-lead- ECG. The participants left the study center and pursued their daily routines. Repeated 5min- intervals of ECG parameters were available for each individual. ECG parameters of interest were HR, SDNN, RMSSD as well as low frequency (LF, 0.04-1.5 Hz) and high frequency power (HF, 0.15-0.40 Hz) in normalized units. In parallel, personal exposure was measured with a PEM sampler pDR-1200 (MIE Inc, MA, USA) for PM2.5 and a carbon monoxide (CO) Measurer (Langan Model T15n, Langan Products Inc., California, USA). Only five of the ten participants were additionally equipped with an Ultrafine Particle Counter for PNC (P-TRAK, Model 8525, TSI, Minnesota, USA). Moreover, participants filled in a personal activity diary with an accuracy of 15 minutes, including questions on the whereabouts (e.g. being in traffic) and activities (e.g. cooking, cleaning) during the measurement period. Hourly means of ambient air temperature, relative humidity, and barometric pressure were measured at a central monitoring site in the urban background of Augsburg.
A2.2 Statistical analyses
Personal PNC measurements lasted only between 8.6 and 10.8h because of limited battery life. Therefore, for consistency reasons all analyses were performed for this shorter period only in the five individuals with personal PNC measurements. Mixed models with a random participant effect and a first order autoregressive covariance structure were used to analyze the association between ECG parameters and personally measured air pollutants as well as with the indicator variable "being in traffic" (yes vs. no). A confounder model was built for each outcome separately. We considered as potential confounders 1h-averages of ambient air temperature, relative humidity, and barometric pressure measured within the same hour as the respective 5min ECG-interval. The confounders were included linearly or smoothly as penalized spline depending on the model fit. Data were analyzed with SAS statistical package (version 9.2; SAS Institute Inc., Cary, NC, USA).
A2.3 Results
Two women and three men with an average age of 65 years (range: 44-80 years) and body mass index of 29kg/m² (25-33kg/m²) were part of our study. They spent on average 103min (45-210min) outdoors and thereby were most of the time in traffic (average: 79min, range: 15-210min).
Personal PNC measurements lasted only between 8.6 and 10.8h because of limited battery life. Therefore, all further analyses were performed for this shorter period comprising 474 5min- intervals. Table A2.1 describes the ECG parameters and personal exposure measurements.
We observed extreme values for personal PM2.5 (max: 387.1µg/m³) and PNC (max: 147,386.4 particles/cm³) while a participant had lit a candle and an incense stick. While PNC values remained high for about four hours, PM2.5 only peaked briefly.
Elevated CO levels were only significantly associated with an increase in HR with a lag of 10-14min (0.64% [0.2;1.08%]). Being in traffic led to an immediate (2.92% [0.50;5.40%]) and 10-14min delayed elevation (2.93% [0.48;5.43%]) but to a 20-24min lagged decrease (2.93% [0.48;5.43]) in HR. We observed concurrent and delayed reductions in SDNN of about 0.7-0.9% in association with increases in PM2.5. However, being in traffic led to an increase in SDNN 15-19 min later. The pattern of air pollution and traffic effects on RMSSD and HF were similar but the effects tended to be stronger and more often significant for HF. An increase in PM2.5 was associated with a concurrent borderline significant decrease in both ECG parameters and with lagged decreases especially for HF. Elevated PNC levels led to 2-3% delayed reductions in RMSSD and HF showing the strongest effects with lags of 15-19min, 20-24min, and 25-29min. Moreover, we detected a -17.81% [-26.84;-7.67% ] and -13.98% [-26.18;0.24%] concurrent reduction in RMSSD and HF when being in traffic, respectively. An elevation of PM2.5 showed only a marginal significant association with LF with a delay of 5-9min (0.30% [-0.02;0.62%]), whereas an increase in PNC led to immediate and lagged increases in LF of about 0.9-1.1%. In contrast, we observed a strong concurrent decrease by -7.96% [-13.70;-2.23%] in LF when the participants were in traffic. When adjusting for HR, air pollution effects on SDNN and RMSSD tended to be somewhat stronger, whereas the association with HF slightly weakened and all PNC effects on LF became non-significant (data not shown).
Table A2.1. Description of 474 5 min-intervals of ECG parameters and personally measured air pollutants in five participants
| Mean | SD | Min | 25% | Median | 76% | Max | IQR | ||
|---|---|---|---|---|---|---|---|---|---|
| HR (beats/min) | 89.9 | 16.5 | 58.1 | 77.1 | 86.9 | 102.0 | 139.3 | 24.6 | |
| SDNN (ms) | 46.4 | 20.3 | 7.6 | 32.1 | 42.2 | 56.6 | 138.9 | 24.5 | |
| RMSSD (ms) | 18.4 | 8.7 | 4.2 | 12.1 | 16.6 | 235 | 47.2 | 11.4 | |
| HF (ms3) | 77.9 | 72.0 | 1.6 | 27.4 | 56.2 | 104.0 | 422.3 | 76.6 | |
| HF (nu) | 14.0 | 8. | 1.9 | 8.8 | 12.3 | 16.8 | 68.1 | 8.0 | |
| LF (ms3) | 495.5 | 549.9 | 7.3 | 153.2 | 284.3 | 565.2 | 2962.6 | 412.0 | |
| LF (nu) | 74.3 | 15.1 | 7.8 | 67.2 | 79.4 | 85.2 | 94.7 | 17.9 | |
| PM2.5 (µg/m3) | 13.2 | 36.8 | 0.7 | 2.0 | 5.0 | 8.3 | 387.1 | 5.4 | |
| in traffica | 4.7 | 3.3 | 0.8 | 2.5 | 3.4 | 6.2 | 14.6 | 3.7 | |
| not in trafficv | 14.9 | 40.0 | 0.7 | 3.0 | 5.4 | 9.3 | 387.1 | 6.2 | |
| CO (ppm) | 0.7 | 0.8 | 0.0 | 0.3 | 0.5 | 0.8 | 9.5 | 0.5 | |
| in traffica | 1.0 | 0.8 | 0.0 | 0.5 | 0.8 | 1.3 | 3.9 | 0.8 | |
| not in trafficb | 0.6 | 0.7 | 0.0 | 0.3 | 0.5 | 0.7 | 9.5 | 0.4 | |
| PNC (c/cm3) | 19,304 | 32,561 | 9,945 | 4,205 | 6,624 | 13,787 | 147,386 | 9,582 | |
| in traffica | 7,450 | 5,263 | 9,945 | 3,693 | 5,720 | 10,844 | 2,6971 | 7,151 | |
| not in trafficb | 21,674 | 35,222 | 1,449 | 4,284 | 7,007 | 15,316 | 14,7386 | 11,0.2 | |
| a 75 5-min intervals, b 395 5mn-intervals For abbreviations see Appendix A3 | |||||||||
A2.4 Conclusion
We observed very rapid changes in HRV within 30min in association with personal measurements of PNC and PM2.5 based on 5min-segments. This immediate reaction indicates that the link between cardiovascular disease and air pollution might, at least in part, be mediated by the autonomic nervous system in response to direct reflexes from receptors in the lungs.
References:
Alperovitch, A., J. M. Lacombe, et al. (2009). "Relationship between blood pressure and outdoor temperature in a large sample of elderly individuals: the Three-City study." Arch Intern Med 169(1): 75-80.
Babisch, W. (2003). "Stress hormones in the research on cardiovascular effects of noise." Noise & Health 5(18): 1-11.
Babisch, W., H. Fromme, et al. (2001). "Increased catecholamine levels in urine in subjects exposed to road traffic noise: the role of stress hormones in noise research." Environment International 26(7-8): 475-481.
De Lorenzo, F., Z. Kadziola, et al. (1999). "Haemodynamic responses and changes of haemostatic risk factors in cold-adapted humans." QJM 92(9): 509-513.
Elwood, P. C., A. Beswick, et al. (1993). "Temperature and risk factors for ischaemic heart disease in the Caerphilly prospective study." Br Heart J 70(6): 520-523.
Henry, J. P. (1992). "Biological basis of the stress response." Integrative Physiological and Behavioral Science 27(1): 66-83.
Hess, K. L., T. E. Wilson, et al. (2009). "Aging affects the cardiovascular responses to cold stress in humans." J Appl Physiol 107(4): 1076-1082.
Holle, R., M. Happich, et al. (2005). "KORA--a research platform for population based health research." Gesundheitswesen 67 Suppl 1: S19-25.
Ising, H., M. Ising, et al. (2003). Verstärkung der Schadwirkungen von Kraftfahrzeug-Abgasen durch lärmbedingte Erhöhung von Stresshormonen. Berlin, Eigenverlag Verein WaBoKu.
Kanikowska, D., J. Sugenoya, et al. (2009). "Seasonal variation in blood concentrations of interleukin-6, adrenocorticotrophic hormone, metabolites of catecholamine and cortisol in healthy volunteers." Int J Biometeorol 53(6): 479-485.
Keatinge, W. R., S. R. Coleshaw, et al. (1984). "Increases in platelet and red cell counts, blood viscosity, and arterial pressure during mild surface cooling: factors in mortality from coronary and cerebral thrombosis in winter." Br Med J (Clin Res Ed) 289(6456): 1405-1408.
Koenig, W., N. Khuseyinova, et al. (2006). "Increased concentrations of C-reactive protein and IL-6 but not IL-18 are independently associated with incident coronary events in middle- aged men and women: results from the MONICA/KORA Augsburg case-cohort study, 1984-2002." Arterioscler Thromb Vasc Biol 26(12): 2745-2751.
Neild, P. J., D. Syndercombe-Court, et al. (1994). "Cold-induced increases in erythrocyte count, plasma cholesterol and plasma fibrinogen of elderly people without a comparable rise in protein C or factor X." Clin Sci (Lond) 86(1): 43-48.
Rana, J. S., B. J. Arsenault, et al. (2011). "Inflammatory biomarkers, physical activity, waist circumference, and risk of future coronary heart disease in healthy men and women." Eur Heart J 32(3): 336-344.
Ridker, P. M. (2009). "C-reactive protein: eighty years from discovery to emergence as a major risk marker for cardiovascular disease." Clin Chem 55(2): 209-215.
Rudnicka, A. R., A. Rumley, et al. (2007). "Diurnal, seasonal, and blood-processing patterns in levels of circulating fibrinogen, fibrin D-dimer, C-reactive protein, tissue plasminogen activator, and von Willebrand factor in a 45-year-old population." Circulation 115(8):996-1003.
Sattar, N., H. M. Murray, et al. (2009). "Are markers of inflammation more strongly associated with risk for fatal than for nonfatal vascular events?" PLoS Med 6(6): e1000099.
Schneider, A., D. Panagiotakos, et al. (2008). "Air temperature and inflammatory responses in myocardial infarction survivors." Epidemiology 19(3): 391-400.
Sung, K. C. (2006). "Seasonal variation of C-reactive protein in apparently healthy Koreans." Int J Cardiol 107(3): 338-342.
Woodhouse, P. R., K. T. Khaw, et al. (1994). "Seasonal variations of plasma fibrinogen and factor VII activity in the elderly: winter infections and death from cardiovascular disease." Lancet 343(8895): 435-439.
Journal Articles:
No journal articles submitted with this report: View all 19 publications for this subprojectSupplemental Keywords:
Health, RFA, Scientific Discipline, Air, PHYSICAL ASPECTS, Health Risk Assessment, Physical Processes, Risk Assessments, particulate matter, Epidemiology, human exposure, long term exposure, aersol particles, atmospheric particles, ambient particle health effects, exposure, atmospheric aerosol particles, PM, atmospheric particulate matter, acute cardiovascular effects, cardiovascular disease, human health riskProgress and Final Reports:
Original AbstractMain Center Abstract and Reports:
R832415 Rochester PM Center Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
R832415C001 Characterization and Source Apportionment
R832415C002 Epidemiological Studies on Extra Pulmonary Effects of Fresh and Aged Urban Aerosols from Different Sources
R832415C003 Human Clinical Studies of Concentrated Ambient Ultrafine and Fine Particles
R832415C004 Animal models: Cardiovascular Disease, CNS Injury and Ultrafine Particle Biokinetics
R832415C005 Ultrafine Particle Cell Interactions In Vitro: Molecular Mechanisms Leading To Altered Gene Expression in Relation to Particle Composition
The perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.