Long-Term Exposure to Air Pollution and Vascular Damage in Young Adults : Epidemiology

Secondary Logo

Journal Logo

Air Pollution: Original Article

Long-Term Exposure to Air Pollution and Vascular Damage in Young Adults

Lenters, Virissaa; Uiterwaal, Cuno S.b; Beelen, Roba; Bots, Michiel L.b; Fischer, Paulc; Brunekreef, Berta,b; Hoek, Gerarda

Author Information
doi: 10.1097/EDE.0b013e3181dec3a7


The association between exposure to air pollution and excess cardiovascular mortality has been well documented.1–3 Recently, the association with cardiovascular morbidity has been demonstrated.4,5 To elucidate the underlying pathophysiologic mechanisms, Künzli et al6 first investigated an association between long-term exposure to air pollution and a preclinical outcome: carotid artery intima-media thickness, a measure of atherosclerosis. Subsequently, living close to a major road was found to be associated with a higher degree of coronary artery calcification,7 and long-term exposure to air pollution was associated with carotid artery intima-media thickness, but not with other measures of atherosclerosis (specifically, coronary artery calcification and ankle-brachial index).8 These 3 studies involved older study populations, with mean ages around 60 years.

Although most studies have focused on exposures to particulate matter less than 2.5 or 10 μm in aerodynamic diameter (PM2.5, PM10), some recent studies have emphasized traffic-generated pollution,9,10 as this more accurately captures intraurban variability in exposure. The aim of this study was to investigate the association between exposure to ambient and traffic-related air pollution and early vascular damage. We estimated individual residential exposures to nitrogen dioxide (NO2), black smoke, PM2.5, and sulfur dioxide (SO2) and to various proxies of traffic emissions using an exposure assessment method with fine spatial resolution.11 We then examined the associations with 3 established markers of vascular damage in a cohort of young adults, with inherently low cardiovascular risk profiles: carotid artery intima-media thickness, pulse wave velocity, and augmentation index. Carotid artery intima-media thickness captures arterial wall thickening and can be used to detect preclinical atherosclerosis. The other 2 outcomes assess arterial stiffness—pulse wave velocity as a direct measure and augmentation index as a surrogate measure. Augmentation index captures the contribution of wave reflections on the aortic waveform, and augmented values indicate increased ventricular workload and impeded coronary perfusion.12 Carotid artery intima-media thickness and pulse wave velocity independently predict coronary heart disease and cardiovascular mortality.13 Augmentation index is a less established marker, but it has been linked to cardiovascular disease risk.14


Study Population and Health Outcomes

The Atherosclerosis Risk in Young Adults study is a population-based, prospective cohort study designed to investigate early-life determinants of cardiovascular risk, including vascular damage. The study design and vascular measurement procedures have been extensively described elsewhere.15,16 In brief, all young adults born between 1970 and 1973 with a complete medical record from the Municipal Health Service, and who attended secondary school in the city of Utrecht in the Netherlands were invited to participate. Of 4207 students identified as eligible, 820 (20%) were willing to participate, and ultimately 750 subjects (352 men and 398 women) underwent measurements of cardiovascular risk factors between October 1999 and December 2000. This study was approved by the Medical Ethical Committee of the University Medical Center Utrecht and all participants gave written informed consent.

The outcomes investigated were common carotid artery intima-media thickness, aortic pulse wave velocity, and augmentation index. Carotid artery intima-media thickness was measured in all 750 participants with high-resolution B-mode ultrasonography of the right and left common carotid arteries using a 7.5 MHz linear array transducer (Acuson Aspen, Mountain View, CA). Thickness was assessed as both the mean and maximum of 8 predefined angles capturing the media-adventitia interface of the near and far arterial walls. Mean carotid artery intima-media thickness is the more commonly assessed outcome. However, maximum carotid artery intima-media thickness more likely captures thickness measurements at the site of plaques and may more accurately predict an individual's cardiovascular disease risk.17 Because of equipment failure, pulse wave velocity measurements were available only for a subset of 524 participants. Aortic pulse wave velocity and augmentation index were assessed using the SphygmoCor device (AtCor Medical, Sydney), and pulse waves were determined by applanation tonometry of the radial artery (Millar Instruments, Houston, TX). Pulse wave velocity was calculated as D/t, with (D) representing the distance traveled by the pulse wave from the carotid to femoral arterial sites, and (t) the wave transit time. Pulse wave velocity was measured over 10 consecutive heartbeats to cover one respiratory cycle, and this was performed 3 separate times to calculate a mean pulse wave velocity value. Ascending aortic pressure was derived from the central pressure waveform using a validated transfer function.18 Augmentation index was calculated as the difference between early and late pressure peaks, divided by pulse pressure, and expressed as a percentage.

A subset of participants underwent repeated measurements on a second occasion to evaluate reproducibility (n = 21, 25, and 23 for carotid artery intima-media thickness, pulse wave velocity, and augmentation index, respectively). The intraclass correlation coefficient was 0.84, 0.67, and 0.65, respectively.

Systolic and diastolic blood pressure (BP) were measured twice during 2 visits within a 3-week period using a semi-automated device (Dynamap, Critikon, Tampa, FL). Mean arterial pressure was calculated as the mean of systolic blood pressure (SBP) minus diastolic blood pressure (DBP), and pulse pressure was calculated as (2 × DBP + SBP)/3.

Participants completed a standardized written questionnaire concerning demographic, lifestyle, and cardiovascular disease risk factors. The exact residential address in the year 2000 was available for all participants, and for 554 participants their childhood address (last known address between ages 4 and 16 years) was also available.

Exposure Assessment

Individual exposure to air pollution was assigned as in previous Dutch cohort studies.11,19 Exact procedures are described in the eAppendix (https://links.lww.com/EDE/A388). Briefly, overall concentrations of NO2, PM2.5, black smoke, and SO2 for the year 2000 at the home address were characterized as the sum of a regional, an urban, and a local traffic component. We previously demonstrated that the spatial contrast in air pollution is stable for at least 10 years,11 although absolute levels have changed, especially for SO2. A separate exposure assessment was performed to calculate background NO2 and SO2 concentrations for the childhood home addresses, making use of air-pollution data, and urban-predictor data from the year 1978. Sufficient PM2.5, PM10, and black smoke data were not available for this period.

The regional component was determined by interpolation of regional background concentrations. For the urban component, regression models were developed using data on 10 categories of land use in 100-m grids, in addition to the population density and land-use predictors used in the previous study.11 Contributions from traffic on roads near the home address were estimated for NO2, PM2.5, and black smoke as by Beelen et al,11 and were added to the background (sum of regional and urban) concentrations.

We also evaluated 4 traffic indicator variables: (1) an indicator for living within 100 m of a motorway or within 50 m of a major road (with an average of >10,000 motor vehicles per 24 hours) and traffic intensity (2) in a 100-m buffer around the home, (3) on the nearest road, and (4) on the nearest major road.11

Statistical Analysis

Multiple linear regression analyses were performed to examine associations of vascular damage outcomes with air pollution exposures, specifically overall and background concentrations, and traffic indicators in combination with background black smoke concentrations.

We evaluated a basic model (1st model), which was age- and sex-adjusted for all 3 outcomes, and further adjusted for mean arterial pressure in the pulse wave velocity analyses (because pulse wave velocity intrinsically relates to distending pressure20). We then adjusted for potential covariates (2nd model) selected a priori: body mass index (BMI); pack-years of active smoking; whether either parent smoked at home during childhood (dichotomous); alcohol intake (units/day); highest education (3 categories) and highest profession (5 categories) attained, as proxies of socioeconomic status; diabetes mellitus (dichotomous); and percentage of persons with a low and high income in the neighborhood. Carotid artery intima-media thickness was adjusted for pulse pressure (instead of mean arterial pressure) in the adjusted models.20 Augmentation index was additionally adjusted for heart rate. Subsequently, we added covariates that are possibly involved in the causal mechanisms of vascular damage: hypertension (defined as blood pressure >140/90 mm Hg), HDL, and LDL cholesterol (mmol/L), and family history of CVD (dichotomous). Since we cannot distinguish whether these cardiovascular risk factors act as confounders or mediators, we added these covariates in a third model to minimize the chance of residual confounding, and to explore their influence on estimates.

Effect estimates are presented as percent change of arithmetic mean levels of the vascular damage outcomes with 95% confidence intervals (CIs), calculated as the absolute change in outcome divided by the study population mean (Table 1). The absolute change was calculated by multiplying the regression slope by the difference between the fifth and 95th percentiles of the exposure variable's distribution. This approximated 25 μg/m3 for NO2, 10 μg/m3 for black smoke, 5 μg/m3 for SO2, and 5 μg/m3 for PM2.5.

General Characteristics of the Atherosclerosis Risk in Young Adults Study Population (n = 745 Unless Otherwise Indicated)

To explore potential effect modification, we stratified by sex, smoking status (never, former, current), and categories of education. As a sensitivity analysis, we removed outliers with Studentized residuals ≥|2|. Epidemiologic analyses were performed with SAS 9.1 (SAS Institute Inc, Cary, NC), and Geographic Information System (GIS) calculations were made using ArcGIS (ESRI, Redlands, CA).


Geographic coordinates for the home address in 2000 were identified for 745 of the 750 (99%) Atherosclerosis Risk In Young Adults participants. Of these, 310 (42%) lived in the city of Utrecht, and a further 188 (25%) lived in 5 villages surrounding Utrecht. General characteristics of participants are summarized in Table 1. Very few subjects used medication for lipids, blood pressure, cholesterol, or diabetes mellitus (0.3%–1.3%).

Exposure Assignment

The distribution of long-term exposure to overall NO2, black smoke, PM2.5, and SO2 at the 2000 home address indicates considerable exposure variability within the cohort. Exposure to traffic also varied considerably (Table 2 and Fig. 1). The participants who lived in the city of Utrecht experienced higher exposures: mean (SD) = 42.0 (6.7) μg/m3 for NO2, 13.5 (3.3) for black smoke, 21.4 (1.1) for PM2.5, and 3.0 (0.3) for SO2.

Summary of Air Pollution Exposures (μg/m3) and Traffic Indicators (Motor Vehicles/24 Hours) in the Atherosclerosis Risk in Young Adults Cohort (n = 745)
Exposure distributions of NO2, black smoke, PM2.5, and SO2 concentrations and of the traffic intensity in a 100-m buffer and on the nearest road, at the home address in the year 2000.

Urban component regression models explained 88%, 58%, 11%, and 61% of the variance of concentrations of NO2, black smoke, PM10, and SO2, respectively. Correlations between overall concentrations of NO2, black smoke, and PM2.5 were all >0.5, whereas SO2 was poorly correlated with the other pollutants (r = 0.1). Total traffic intensity in a 100-m buffer was moderately correlated with overall NO2 concentrations (r = 0.6), and highly correlated with overall black smoke and PM2.5 (r = 0.9) concentrations, but not with SO2. (See eTables 1 and 2 [https://links.lww.com/EDE/A388] for complete correlation tables.)

It was possible to geocode 552 of the 554 (99%) available addresses for the year 1978. Mean (SD) estimated background concentrations were 40.2 (3.9) μg/m3 for NO2 and 30.9 (2.6) μg/m3 for SO2. Urban regression models explained 52% and 38% of the variance of NO2 and SO2 concentrations, respectively. Eighty-two (11%) participants lived at the same 2000 address as during childhood. Correlations between exposures in 1978 and exposures in 2000 were low; r = 0.24 for NO2, and r = −0.09 for SO2. This is consistent with the fact that most (89%) of the participants had moved out of their childhood home.

Associations With Vascular Damage

We found no clear association between exposure to air pollution and arterial-wall thickening. PM2.5, black smoke, and SO2 were positively associated with carotid artery intima-media thickness (Table 3). There were modest differences in estimates due to adjustments. Adjustment for area-level socioeconomic characteristics decreased associations with mean and maximum carotid artery intima-media thickness, as did adjustment for pack-years of smoking. Adjustment for BMI and LDL cholesterol strengthened associations (data not presented).

Associations Between Long-Term Exposure to Air Pollutants and Vascular Damage Outcomes

We found evidence of an association between exposure to gaseous pollutants and arterial stiffness. Exposure to NO2 was associated with a 4.9% (95% CI = 1.2% to 8.7%) increase in pulse wave velocity per 20 μg/m3 increase in background concentrations, and a 4.1% (0.1% to 8.0%) increase per 25 μg/m3 increase in overall (background plus traffic-related) concentrations. Only background concentrations of SO2 were estimated, because traffic contributes negligibly to SO2. A contrast of 5 μg/m3 corresponded to a 5.3% (0.1% to 10.4%) increase in pulse wave velocity. Estimates for effect of NO2 or SO2 on pulse wave velocity remained nearly identical after adjustment for mean or maximum carotid artery intima-media thickness. Black smoke and PM2.5 both showed a weak positive association with pulse wave velocity upon full adjustment for covariates. A 25 μg/m3 contrast in NO2 was associated with a 37.6% (2.2% to 72.9%) increase in augmentation index, and this association remained robust in all 3 confounder models (P < 0.04). Pulse wave velocity and augmentation index effect estimates generally increased upon adjustment for BMI and for percentage of persons in the neighborhood with a low and high income. Adjustment for heart rate decreased the magnitude of augmentation index estimates.

The distribution of carotid artery intima-media thickness was similar in the full study population and in the population with valid pulse wave velocity measurements. The direction and magnitude of effect estimates were similar in the 522 participants with valid pulse wave velocity measurements, that is, the adjusted change in mean carotid artery intima-media thickness per 5 μg/m3 PM2.5 was 0.94% (95% CI = −2.59% to 4.47%) in the entire cohort and 0.48% (−3.64% to 4.60%) in the subset of subjects with valid pulse wave velocity measurements.

The adjusted associations between traffic indicators and vascular damage outcomes were inconsistent in direction of effect (Table 4).

Associations Between Long-Term Exposure to Air Pollutants and Vascular Damage Outcomes

Stratifying by sex, smoking status, and education categories yielded inconsistent results (Fig. 2). Percent change in vascular damage outcomes tended to be higher in women and among those with less education, although the pattern differed across pollutants and outcomes. Greater increases in pulse wave velocity appeared in the never and former versus current smokers, and in never smokers for augmentation index, whereas the opposite pattern occurred in subgroups for the outcome carotid artery intima-media thickness. Augmentation index displayed effect modification by sex (P = 0.003). None of the other interaction terms reached a significance level of P < 0.05. Effect estimates did not substantially change upon removal of pulse wave velocity outliers (n = 18; 3%) or augmentation index outliers (n = 130; 18%), which reinforced that the association between SO2 and pulse wave velocity is robust.

Percent change (filled circles) and 95% CI (vertical lines) in mean carotid artery intima-media thickness, pulse wave velocity, and augmentation index associated with a 5–95th percentile exposure contrast in all participants and in subgroups. Estimates are adjusted for all covariates (age, sex, blood pressure, BMI, pack-years of smoking, parental smoking at home, alcohol intake, education, highest profession, diabetes, percentage of low and high income households in neighborhood, hypertension, HDL cholesterol, LDL cholesterol, family history of CVD; the augmentation index models are additionally adjusted for heart rate); the null effect line is indicated.

In 2-pollutant models (data not shown), with pulse wave velocity as the outcome, NO2 and SO2 were more strongly associated with arterial stiffness than were black smoke and PM2.5. When all 4 pollutants were included in the pulse wave velocity regression analysis, NO2 and SO2 retained significance (P = 0.02 and 0.05, respectively); as did NO2 in the 4-pollutant model for augmentation index (P = 0.02).

We did not find evidence of an independent association of childhood (1978) exposures to NO2 and SO2 with vascular damage. The change in mean carotid artery intima-media thickness associated with childhood exposures was 0.4% (−3.1% to 3.9%) for NO2 and 1.4% (−2.1% to 4.9%) for SO2; change in pulse wave velocity was −1.0% (−5.7% to 3.7%) for NO2 and −2.9% (−7.6% to 1.8%) for SO2; and change in augmentation index was 31.2% (−18.0% to 80.3%) for NO2 and 5.2% (−44.0% to 54.3%) for SO2.


Exposure to NO2 was weakly associated with increased arterial stiffness (pulse wave velocity, augmentation index) in a cohort of young adults. We did not find associations between air pollution and carotid artery intima-media thickness. Traffic indicators were not associated with vascular damage outcomes. Carotid artery intima-media thickness, pulse wave velocity, and augmentation index are intermediate endpoints of cardiovascular disease, and reflect functional and structural arterial characteristics. To our knowledge, no previous studies have assessed how exposure to ambient air pollution influences pulse wave velocity or augmentation index. Furthermore, no studies investigating the relationship between air pollution and preclinical cardiovascular disease have investigated such a young population as this cohort, nor such a comprehensive set of pollutants and traffic-related exposures.

Although statistically imprecise, our carotid artery intima-media thickness effect estimates of 0.5% to 2.5% increase in mean and maximum carotid artery intima-media thickness per 5 μg/m3 PM2.5 are in line with previous studies. In 798 older adults of mean age 59 years, Künzli et al6 reported that carotid artery intima-media thickness adjusted for covariates increased 3.9% to 4.3% per 10 μg/m3 contrast in PM2.5. Diez-Roux et al8 reported a 1% to 4% increase in carotid artery intima-media thickness adjusted for covariates, including cardiovascular risk factors, per 10 μg/m3 contrast in PM2.5 and PM10 (n = 5172). They did not find an association with 2 other measures of atherosclerosis—coronary artery calcification and ankle-brachial index. An earlier study of 4994 adults found that living close to a major road was associated with coronary artery calcification.7 Although our point estimates for PM2.5 and carotid artery intima-media thickness are similar to these 2 studies, our cohort experienced a shorter duration of “lifetime” exposure, and a smaller contrast in exposure to PM2.5. Although the mean individually assigned PM2.5 exposures were similar, the standard deviations were not (mean [SD] 20.7 (1.2) μg/m3, compared with 20.3 (2.6) and 21.7 (5.0) μg/m3 in the other 2 studies6,8).

Our estimates may differ from those previously reported because our study population is younger, has a lower cardiovascular risk profile, and is probably more residentially mobile. Carotid artery intima-media thickness progression rates accelerate with age, and comparable increases correspond with higher cardiovascular risks in younger versus older adults.21 The mean carotid artery intima-media thickness and also its variation among subjects, were clearly smaller in this study of young adults than in previous studies. The signal-to-noise ratio could be high in a young and healthy population such as ours. This hypothesis is further supported by the smaller variability of carotid artery intima-media thickness in the cohort compared with pulse wave velocity and augmentation index. Considering that vascular measurements were performed on only one occasion, intraindividual variability may also have attenuated the associations. However, the reproducibility of carotid artery intima-media thickness measurements was better than for pulse wave velocity and augmentation index. Carotid artery intima-media thickness exhibits negligible short-term variation, in contrast with pulse wave velocity and augmentation index, which are affected by changes in blood pressure. The lack of residential mobility data may, therefore, have limited our ability to find an association with carotid artery intima-media thickness, as this outcome is more likely to reflect long-term exposure. Finally, we may not have detected an association between air pollution and carotid artery intima-media thickness due to insufficient power. This may be the case for PM2.5, for which the exposure contrast was more limited compared with the contrast for the other pollutants.

We found a 4% to 5% increase in pulse wave velocity associated with exposure to NO2 and SO2, and a 38% increase in augmentation index with NO2, but no associations with other pollutants. These effect estimates generally remained stable upon adjustment although for NO2, adjustment strengthened the associations with pulse wave velocity. Adjustment for BMI and the percentage of persons with a low income in the neighborhood both increased the magnitude of pulse wave velocity and augmentation index associated with exposure to NO2. As NO2 was only moderately correlated with percentage of low-income persons in the neighborhood (r = 0.42), we do not think it is likely that the NO2 associations in the fully-adjusted models merely reflect collinearity with the confounders. SO2 was not correlated (r = 0.02) with neighborhood income. Beelen et al2 found slightly stronger effects for NO2, compared with PM2.5, black smoke, and SO2, on cardiovascular mortality in a large Dutch cohort. Likewise, higher relative risks for cardiopulmonary mortality have been reported for NO2 versus PM1010 and for total suspended particles, black smoke, and SO2.22 Therefore, there is some evidence that a pollution mixture represented by NO2 may be associated with higher cardiovascular risks than mixtures represented by other pollutants such as particulate matter. However, one hypothesized reason is that NO2 reflects traffic exposures, and our results do not indicate an association between traffic-related exposure metrics and either pulse wave velocity or augmentation index. We found an independent, robust association between exposure to SO2 and pulse wave velocity. SO2 in the Netherlands is primarily from industrial sources and SO2 levels have substantially decreased over the past couple of decades.11 Several mortality studies have not demonstrated that SO2 is associated with cardiopulmonary mortality.2,22,23 In contrast, Pope et al1 found that SO2 was associated with cardiopulmonary mortality, whereas NO2 was not. Hedley et al24 document a 2% decline in cardiovascular deaths following an abrupt reduction in ambient SO2 in Hong Kong of approximately 20 μg/m3 because of policies limiting sulfur content in fuel used by power plants and vehicles. SO2, at current low levels, is unlikely to be causally related to morbidity and mortality. Even so, the associations found in our study add to the evidence that SO2 is a useful indicator of the mixture of ambient air pollution.

Pulse wave velocity and augmentation index are closely related.12 Augmented wave reflections may reflect decreased aortic compliance or increased peripheral resistance. Our results agree with recently observed augmented augmentation index values in 12 healthy volunteers experimentally exposed for 1 hour to diesel exhaust while exercising,25 and with increased augmentation index associated with short-term exposure to metal-rich PM2.5 associated with welding (n = 26).26

We found clearer associations for pulse wave velocity and augmentation index than for carotid artery intima-media thickness. In our cohort, age-sex adjusted mean carotid artery intima-media thickness and pulse wave velocity additionally adjusted for mean arterial pressure were not correlated (Pearson's correlation coefficient = 0.03, P = 0.51).20 Such correlation was also not found in 564 hyper- and normotensive patients of mean age 58.2 years after adjustment for age and blood pressure.27 It has not been established whether arterial stiffness precedes atherosclerotic lesions, or vice versa, although the relationship between these vascular changes is synergistic. If functional changes precede structural changes, this could explain why air pollution was associated with arterial stiffening and wave reflections, but not with carotid artery intima-media thickness. Perhaps shorter-term recent exposures over days or months alters endothelial function in conduit or elastic arteries and has an intermediate effect of impairing arterial compliance. Decreased distensibility without accompanying increased intima-medial thickening has been observed in pediatric populations.28 Studies in middle-age and older populations27,29 support the opposite hypothesis—that elevated arterial stiffness is evident only after marked increases in carotid artery intima-media thickness. Most studies have investigated the predictive value of pulse wave velocity and augmentation index in high-risk patients, such as hypertensive or elderly populations.30–32 The determinants of pulse wave velocity and augmentation index in young adults have not been well characterized.20

Künzli et al6 reported that women, elderly, hyperlipidemic, and never-smokers were more susceptible to long-term exposure to air pollution. Studies investigating mortality outcomes have also found evidence that less-educated people are more susceptible to the damaging effects of air pollution.1,33 Augmentation index was modified by sex; increases were observed only in women with NO2 exposure, and higher estimates were generally observed in women compared with men across all pollutants. We did not find that the other vascular damage outcomes were modified by sex, education, or smoking status, possibly because of limited statistical power. Pulse wave velocity and augmentation index effect estimates tended to be elevated in former and never smokers, perhaps due to less “noise” induced by damage due to smoking.

It has been proposed that systemic inflammation is the primary mechanism by which inhaled air pollution induces cardiovascular disease.34 Studies on animal models have shown that exposure to particulate matter accelerates progression of atherosclerotic plaques.35,36 Human studies have indicated that acute exposure to high concentrations of particles is associated with increased blood clotting factors,37 endothelial dysfunction and brachial artery vasoconstriction,38 and inflammation markers, such as C-reactive protein.39 A recent study40 found that in young men in Paris where 1- and 5-day exposure to ambient air pollution was characterized, NO2 and SO2 were associated with endothelial function, whereas PM and black smoke were not.

Nondifferential misclassification of exposure may explain why we were unable to reproduce the associations between air pollution and preclinical atherosclerosis, or find more robust associations between other exposures such as traffic indicators and arterial stiffening. The exposure assignment methodology assumes that spatial patterns and temporal trends between areas are homogenous over time. The validity of using data for the year 2000 to estimate long-term exposure is supported by an earlier study,11 in which nearly the same exposure assignment methodology was applied to a large cohort. Comparing 1975 and 1996, correlations between average annual concentrations at monitoring sites in the Netherlands were high for NO2, black smoke, and SO2. However, for our study, participants' exposures determined for their home addresses in 1978 and 2000 did not correlate well because most participants had moved since 1978. Young adults in their late 20s probably move more frequently than middle-aged and older individuals. A limitation of this study is the incomplete information on residential mobility. Except for the small subset of subjects (11%) who lived at the same address at age 16 years and at age 27–30 years, we do not know how long subjects had lived at their year 2000 address. There were too few participants with the same childhood and adulthood address to test the robustness of our findings within a subgroup with a certain long-term residency and thus exposure.

There is growing interest in how exposures during childhood and adolescence influence determinants of health later in life. We did not find an association between air pollution estimated for 1978, when participants were 5–8 years of age, and vascular damage. This may be due to the fact that the assessment of childhood exposures did not include traffic data and was therefore more imprecise. Or, it may be because developing vessels might have the capacity to more readily recover from exposures to air pollution. In light of the current interest in air pollution control measures and the public health burden of cardiovascular diseases, we conclude that even low levels of air pollution may cause early vascular damage.


We thank Danielle Vienneau and Kees de Hoogh for kindly providing the CORINE Land Cover data.


1. Pope CA III, Burnett RT, Thun MJ, et al. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA. 2002;287:1132–1141.
2. Beelen R, Hoek G, van den Brandt P, et al. Long-term effects of traffic-related air pollution on mortality in a Dutch cohort (NLCS-AIR Study). Environ Health Perspect. 2008;116:196–202.
3. Laden F, Schwartz J, Speizer FE, Dockery DW. Reduction in fine particulate air pollution and mortality: Extended follow-up of the Harvard six cities study. Am J Respir Crit Care Med. 2006;173:667–672.
4. Miller K, Siscovick D, Sheppard L, et al. Long-term exposure to air pollution and incidence of cardiovascular events in women. N Engl J Med. 2007;356:447–458.
5. Tonne C, Melly S, Mittleman M, Coull B, Goldberg R, Schwartz J. A case-control analysis of exposure to traffic and acute myocardial infarction. Environ Health Perspect. 2007;115:53–57.
6. Künzli N, Jerrett M, Mack WJ, et al. Ambient air pollution and atherosclerosis in Los Angeles. Environ Health Perspect. 2005;113:201–206.
7. Hoffmann B, Moebus S, Mohlenkamp S, et al. Residential exposure to traffic is associated with coronary atherosclerosis. Circulation. 2007;116:489–496.
8. Diez-Roux AV, Auchincloss AH, Franklin TG, et al. Long-term exposure to ambient particulate matter and prevalence of subclinical atherosclerosis in the Multi-Ethnic Study of Atherosclerosis. Am J Epidemiol. 2008;167:667–675.
9. Hoek G, Brunekreef B, Goldbohm S, Fischer P, van den Brandt PA. Association between mortality and indicators of traffic-related air pollution in the Netherlands: a cohort study. Lancet. 2002;360:1203–1209.
10. Gehring U, Heinrich J, Krämer U, et al. Long-term exposure to ambient air pollution and cardiopulmonary mortality in women. Epidemiology. 2006;17:545–551.
11. Beelen R, Hoek G, Fischer P, van den Brandt PA, Brunekreef B. Estimated long-term outdoor air pollution concentrations in a cohort study. Atmos Environ. 2007;41:1343–1358.
12. O'Rourke MF, Pauca A, Jiang XJ. Pulse wave analysis. Br J Clin Pharmacol. 2001;51:507–522.
13. Bots ML, Dijk JM, Oren A, Grobbee DE. Carotid intima-media thickness, arterial stiffness and risk of cardiovascular disease: current evidence [review]. J Hypertens. 2002;20:2317–2325.
14. Weber T, Auer J, O'Rourke MF, et al. Arterial stiffness, wave reflections, and the risk of coronary artery disease. Circulation. 2004;109:184–189.
15. Oren A, Vos LE, Uiterwaal CS, et al. The Atherosclerosis Risk in Young Adults (ARYA) study: Rationale and design. Eur J Epidemiol. 2003;18:715–727.
16. Van Trijp M, Uiterwaal C, Bos W, Oren A, Grobbee D, Bots M. Noninvasive arterial measurements of vascular damage in healthy young adults: Relation to coronary heart disease risk. Ann Epidemiol. 2006;16:71–77.
17. Bots M, Evans G, Riley W, Grobbee D. Carotid intima-media thickness measurements in intervention studies: Design options, progression rates, and sample size considerations: A point of view. Stroke. 2003;34:2985–2994.
18. Chen CH, Nevo E, Fetics B, et al. Estimation of central aortic pressure waveform by mathematical transformation of radial tonometry pressure: Validation of generalized transfer function. Circulation. 1997;95:1827–1836.
19. Hoek G, Fischer P, Van Den Brandt P, Goldbohm S, Brunekreef B. Estimation of long-term average exposure to outdoor air pollution for a cohort study on mortality. J Expo Anal Environ Epidemiol. 2001;11:459–469.
20. Oren A, Vos LE, Uiterwaal CS, Grobbee DE, Bots ML. Aortic stiffness and carotid intima-media thickness: Two independent markers of subclinical vascular damage in young adults? Eur J Clin Invest. 2003;33:949–954.
21. Lorenz MW, von Kegler S, Steinmetz H, Markus HS, Sitzer M. Carotid intima-media thickening indicates a higher vascular risk across a wide age range prospective data from the Carotid Atherosclerosis Progression Study (CAPS). Stroke. 2006;37:87–92.
22. Filleul L, Rondeau V, Vandentorren S, et al. Twenty five year mortality and air pollution: results from the French PAARC survey. Occup Environ Med. 2005;62:453–460.
23. Nafstad P, Haheim L, Wisloeff T, et al. Urban air pollution and mortality in a cohort of Norwegian men. Environ Health Perspect. 2004;112:610–615.
24. Hedley AJ, Wong CM, Thach TQ, Ma S, Lam TH, Anderson HR. Cardiorespiratory and all-cause mortality after restrictions on sulphur content of fuel in Hong Kong: an intervention study. Lancet. 2002;360:1646–1652.
25. Lundbäck M, Mills NL, Lucking A, et al. Experimental exposure to diesel exhaust increases arterial stiffness in man. Part Fibre Toxicol. 2009;6:7.
26. Fang SC, Eisen EA, Cavallari JM, Mittleman MA, Christiani DC. Acute changes in vascular function among welders exposed to metal-rich particulate matter. Epidemiology. 2008;19:217–225.
27. Zureik M, Temmar M, Adamopoulos C, et al. Carotid plaques, but not common carotid intima-media thickness, are independently associated with aortic stiffness. J Hypertens. 2002;20:85–93.
28. Groner JA, Joshi M, Bauer JA. Pediatric precursors of adult cardiovascular disease: Noninvasive assessment of early vascular changes in children and adolescents. Pediatrics. 2006;118:1683–1191.
29. Labropoulos N, Ashraf Mansour M, Kang S, Oh D, Buckman J, Baker W. Viscoelastic properties of normal and atherosclerotic carotid arteries. Eur J Vasc Endovasc Surg. 2000;19:221–225.
30. Boutouyrie P, Tropeano A, Asmar R, et al. Aortic stiffness is an independent predictor of primary coronary events in hypertensive patients a longitudinal study. Hypertension. 2002;39:10–15.
31. Sutton-Tyrrell K, Najjar S, Boudreau R, et al. Elevated aortic pulse wave velocity, a marker of arterial stiffness, predicts cardiovascular events in well-functioning older adults. Circulation. 2005;111:3384–3390.
32. Williams B, Lacy PS, Thom SM, et al; CAFE Investigators; Anglo-Scandinavian Cardiac Outcomes Trial Investigators; CAFE Steering Committee and Writing Committee. Differential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes: principal results of the Conduit Artery Function Evaluation (CAFE) study. Circulation. 2006;113:1213–1225.
33. Krewski D, Burnett R, Goldberg M, et al. Reanalysis of the Harvard Six Cities Study. Part II: Sensitivity analysis. Inhal Toxicol. 2005;17:343–353.
34. Pope C, Burnett R, Thurston G, et al. Cardiovascular mortality and long-term exposure to particulate air pollution: Epidemiological evidence of general pathophysiological pathways of disease. Circulation. 2004;109:71–77.
35. Sun Q, Wang A, Jin X, et al. Long-term air pollution exposure and acceleration of atherosclerosis and vascular inflammation in an animal model. JAMA. 2005;294:3003–3010.
36. Suwa T, Hogg JC, Quinlan KB, Ohgami A, Vincent R, van Eeden SF. Particulate air pollution induces progression of atherosclerosis. J Am Coll Cardiol. 2002;39:935–942.
37. Schwartz J. Air pollution and blood markers of cardiovascular risk. Environ Health Perspect. 2001;109:405–409.
38. Brook R, Brook J, Urch B, Vincent R, Rajagopalan S, Silverman F. Inhalation of fine particulate air pollution and ozone causes acute arterial vasoconstriction in healthy adults. Circulation. 2002;105:1534–1536.
39. Peters A, Frohlich M, Doring A, et al. Particulate air pollution is associated with an acute phase response in men; results from the MONICA-Augsburg Study. Eur Heart J. 2001;22:1198–1204.
40. Briet M, Collin C, Laurent S, et al. Endothelial function and chronic exposure to air pollution in normal male subjects. Hypertension. 2007;50:970–976.

Supplemental Digital Content

© 2010 Lippincott Williams & Wilkins, Inc.