Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Apr 10.
Published in final edited form as: J Alzheimers Dis. 2019;71(3):1015–1025. doi: 10.3233/JAD-190563

Fine Particulate Matter Exposure and Cerebrospinal Fluid Markers of Vascular Injury

Rachel M Shaffer a, Lianne Sheppard a,b, Elaine R Peskind c,d, Jing Zhang e, Sara D Adar f, Ge Li c,d,g,*
PMCID: PMC7147945  NIHMSID: NIHMS1577569  PMID: 31476158

Abstract

Background:

Cerebrovascular diseases play an important role in dementia. Air pollution is associated with cardiovascular disease, with growing links to neurodegeneration. Prior studies demonstrate associations between fine particulate matter (PM2.5) and biomarkers of endothelial injury in the blood; however, no studies have evaluated these biomarkers in cerebrospinal fluid (CSF).

Objective:

We evaluate associations between short-term and long-term PM2.5 exposure with CSF vascular cell adhesion molecule-1 (VCAM-1) and e-selectin in cognitively normal and mild cognitive impairment (MCI)/Alzheimer’s disease (AD) individuals.

Methods:

We collected CSF from 133 community volunteers at VA Puget Sound between 2001–2012. We assigned short-term PM2.5 from central monitors and long-term PM2.5 based on annual average exposure predictions linked to participant addresses. We performed analyses stratified by cognitive status and adjusted for key covariates with tiered models. Our primary exposure windows for the short-term and long-term analyses were 7-day and 1-year averages, respectively.

Results:

Among cognitively normal individuals, a 5 μg/m3 increase in 7-day and 1-year average PM2.5 was associated with elevated VCAM-1 (7-day: 35.4 (9.7, 61.1) ng/ml; 1-year: 51.8 (6.5, 97.1) ng/ml). A 5 μg/m3 increase in 1-year average PM2.5, but not 7-day average, was associated with elevated e-selectin (53.3 (11.0, 95.5) pg/ml). We found no consistent associations among MCI/AD individuals.

Conclusions:

We report associations between short-term and long term PM2.5 and CSF biomarkers of vascular damage in cognitively normal adults. These results are aligned with prior research linking PM2.5 to vascular damage in other biofluids as well as emerging evidence of the role of PM2.5 in neurodegeneration.

Keywords: Alzheimer’s disease, biomarkers, cellular adhesion molecules, cerebrospinal fluid, cerebrovascular disorders, dementia, particulate matter

INTRODUCTION

Growing evidence suggests that cerebrovascular damage, such as microvascular injury and stroke, may contribute to cognitive impairment and dementia [1, 2]. Specifically, based on the vascular contributions to cognitive impairment and dementia hypothesis, a substantial proportion of neurodegeneration and cognitive decline is actually due to vascular insults, which can also occur concomitantly with Alzheimer’s disease (AD) pathology [3].

Vascular disease morbidity and mortality is strongly linked to fine particulate matter air pollution (PM2.5) exposure [4, 5]. There are several plausible mechanisms by which PM2.5 could impact vascular health, including systemic oxidative stress and inflammation leading to endothelial damage [4]. Vascular cell adhesion molecule-1 (VCAM-1) and e-selectin are adhesion molecules expressed by endothelial cells in response to injury and inflammation [6, 7]. Elevations in these markers reflect vascular injury and are associated with hypertension, atherosclerosis, and dysregulation of cerebral blood flow [8, 9]. Prior epidemiological studies indicate associations between exposure to air pollution, such as PM2.5, and these and other biomarkers of vascular damage in plasma and serum [10–21]. PM2.5 has also been linked to cerebrovascular damage, such as stroke [22].

Cerebrospinal fluid (CSF), which is derived from blood plasma and circulates within the cerebral ventricular system, can be used to detect biochemical changes and understand pathological disease processes in the brain during life. For example, alterations in e-selectin and VCAM-1 have been detected in the CSF of individuals with AD and vascular dementia [6, 7, 23–29]. Yet, despite the well-recognized link between PM2.5 and vascular injury and the role of vascular injury in dementia, no prior studies have evaluated the association between PM2.5 and these biomarkers of endothelial injury in the CSF of adult populations.

To address this scientific gap, we conduct a novel study to evaluate the associations between short-term (days/months) and long-term (years) PM2.5 exposure with both VCAM-1 and e-selectin in the CSF. Our interest in short-term time windows was informed by prior research indicating associations between PM2.5 exposure over days or weeks and changes in these biomarkers of vascular injury in the blood [10–21]. We investigated long-term time windows to evaluate a potential chronic inflammatory response, which has also been suggested by studies of PM2.5 and blood-based biomarkers [20]. Since individuals with mild cognitive impairment (MCI) or AD often also already exhibit vascular pathologies, we stratified our analyses by cognitive status (cognitively normal versus MCI/AD). Our a priori focus was on the cognitively normal subgroup, because we hypothesized that the disease process in existing MCI/AD cases would likely drive endothelial injury to a relatively greater extent than PM2.5.

Given the hypothesized role of cerebrovascular injury and neuroinflammation in dementia and neurodegeneration [3, 30], understanding whether PM2.5 is linked with CSF biomarkers of vascular injury may provide evidence to support the link between air pollution exposure and cognitive impairments and dementia [31–33].

METHODS

The cohort included 133 individuals enrolled in studies at the University of Washington (UW) Alzheimer’s Disease Research Center (ADRC) between 2001–2012 (Supplementary Table 1). All procedures were approved by the UW Institutional Review Board (IRB) (01–8926-V & 01173), and all individuals or legal authorized representatives for AD participants provided written consent prior to enrollment in the study. Participants were classified as having no cognitive impairment, MCI, or AD at consensus conference based on clinical evaluations supported by neuropsychological testing. Exclusion criteria included major neurological diagnoses (other than MCI/AD) that could affect cognitive function (stroke, Parkinson’s disease, multiple sclerosis, history of moderate or severe head injury), major psychiatric disorder (schizophrenia, major affective disorder, posttraumatic stress disorder), unstable medical conditions, illegal drug use, and alcohol use disorder.

CSF samples were obtained through lumbar puncture with standardized procedures as published previously [34]. All samples were kept frozen at −80°C prior to analysis at the VA Puget Sound. E-selectin and VCAM-1 were measured with the Human Premixed Multi-Analyte Kit (R&D Systems, Minneapolis, MN, USA) [35]. Apolipoprotein (APOE) genotype was assessed through a restriction digest method [36]. All assays were performed blinded to clinical diagnosis.

Short-term (days, month) average PM2.5 exposure was estimated from a single central site monitor in the Seattle-Beacon Hill neighborhood (AQS ID 530330080) using measurements from a mass-based federal reference method (FRM) [37]. When no FRM data were available, we used data from a federal equivalent method (FEM) instrument, the tapered element oscillating microbalance (TEOM), at the same or neighboring site that were calibrated to the FRM. PM2.5 may vary by temperature and season. Thus, to control for the temporal confounding, daily PM2.5 concentration was “pre-whitened” (pre-adjusted) for temperature (3 degrees of freedom (df)) and time (8 df/year), and these residuals were used in the inferential analyses. This pre-whitening method has been demonstrated as an effective approach for adjusting for seasonality in evaluating the health effects of short-term pollutant exposures in cohort studies [38]. In contrast to these short-term exposure estimates from the central site, long-term (years) average PM2.5 exposure was estimated through yearly national model predictions based on participant address, which were geocoded using ArcMap version 10.5.1. More specifically, PM2.5 exposure was estimated with universal kriging using a land use regression for the mean model and an exponential covariance for the geostatistical smoothing. Input data came from a national network of monitoring stations, as described previously [39]. Yearly averages were estimated as a weighted average based on the date of lumbar puncture.

All inferential analyses were stratified by cognitive status (cognitively normal versus MCI/AD). For short-term PM2.5 exposures, we focused on the effects of 7-day averages (prior to CSF draw) with VCAM-1 and e-selectin. This window was selected based on prior studies of PM2.5 and blood biomarkers of vascular injury. Sensitivity analyses considered alternative exposure averaging periods (2-day, 5-day, 1-month prior to CSF draw), other pre-whitening spline specifications (12 df/year for time), and restriction to exposure data from the Beacon Hill monitor only. Exploratory analyses considered potential interaction effects of age [40, 41] and diabetes e [16, 18, 42–44]; however, because of small sample size in the diabetes subgroup, we were unable to perform the latter analysis and therefore considered effect modification of age only. For long-term PM2.5 exposures, we focused on the effects of 1-year averages with VCAM-1 and e-selectin. This time window allowed us to investigate potential chronic inflammatory responses. We also conducted sensitivity analyses to consider other exposure averaging periods (5-year, 10-year, 20-year), adjustment for year, and to drop individuals for whom we only had P.O. box address information.

We conducted multivariable adjusted linear regression for all analyses. Based on information from relevant scientific literature, we included the following covariates and/or precision variables in a tiered model approach: age (years) [27, 45]; smoking status [27]; sex [27, 45]; education (years) [46, 47]; apolipoprotein E4 (APOE ε4) status (at least 1 copy of E4 versus no copies of E4) [27, 45], body mass index (BMI) [27], hypertension [45, 48–51], coronary heart disease [45], and diabetes [45, 52, 53]. The latter three variables are potential intermediates in the causal pathway and therefore were not included in the main model. While we report results from all models below, our a priori model (model 2) adjusted for age, smoking, and sex. To allow for comparison to prior studies on air pollution and vascular injury as well as on AD and CSF biomarkers, we provide raw numerical estimates of mean change (i.e., the estimated beta regression coefficient) and as well as percent change estimates scaled to the mean outcome value for each subgroup. All exposure effect estimates were scaled to 5 μg/m3 for reporting.

All data analysis was performed using R version 3.6.0.

RESULTS

Descriptive statistics

Among the cognitively normal individuals (n = 73), average (standard deviation (SD)) age was 71.7 (8.0) years and average (SD) years of education was 15.9 (2.5) (Table 1). There were roughly equal proportions of males and females (males = 50.7%), and slightly more than half of this subgroup were past smokers (53.4%) (none were current smokers). Most cognitively normal individuals (69.9%) did not have any copies of APOE ε4 allele. Average (SD) CSF VCAM-1 and e-selectin concentrations among this subgroup were 126.4 (56.8) ng/ml and 62.6 (48.4) pg/ml, respectively. Inter-individual variability in exposure was higher for the short-term averages than for the long-term averages.

Table 1.

Descriptive statistics on UW ADRC cohort, stratified by cognitive status

Cognitively Normal (n = 73) MCI/AD (n = 60)
Covariates mean (SD)/n(%)
Age (y) 71.7 (8.0) 69.9 (9.7)
Education (y) 15.9 (2.5) 16.2 (3.0)
Male 37 (50.7) 33 (55.0)
Former smoker 39 (53.4) 26 (43.3)
APOE ε4 status
 0 51 (69.9) 18 (30.0)
 1 22 (30.1) 40 (66.7)
 NA 0 (0.0) 2 (3.3)
Body mass index (BMI) (kg/m2) 26.3 (3.3) 25.4 (3.8)
Coronary artery disease 0 (0.0) 2 (3.3)
Diabetes 1 (1.4) 1 (1.7)
Hypertension 7 (9.6) 10 (16.7)
Outcome variables (mean (sd))
VCAM-1 (ng/ml) 126.4 (56.8) 139.2 (57.6)
e-selectin (pg/ml) 62.6 (48.4) 58.9 (44.0)
PM2.5 (μg/m3) average (mean (sd))
Short-term1
2-day 7.8 (4.2) 7.5 (4.2)
5-day 7.7 (2.9) 7.6 (3.9)
7-day2 7.8 (2.6) 7.6 (3.5)
30-day 7.7 (2.1) 7.4 (1.8)
Long-term
1-year2 8.3 (1.4) 7.8 (1.4)
5-year 8.8 (1.5) 8.2 (1.5)
10-year 9.6 (1.6) 8.9 (1.6)
20-year 11.2 (1.7) 10.5 (1.8)
Open in a new tab
1

Short-term exposure values are unadjusted (not pre-whitened).

2

Primary exposure period for inferential analyses.

Among the dementia/MCI individuals (n = 60), average (SD) age was 69.9 (9.7) years, and average (SD) years of education was 16.2 (3.0). There were slightly more males than females in this subgroup (males = 55.0%), and 43% were past smokers (none were current smokers). In contrast to the cognitively normal subgroup, most individuals in this subgroup (66.7%) had one or more copies of APOE ε4. Average (SD) CSF VCAM-1 and e-selectin concentrations in this subgroup were 139.2 (57.5) ng/ml and 58.9 (44.0) pg/ml, respectively. Inter-individual variability in exposure was higher for the 2–7 day averages than for the long-term averages.

VCAM-1

Among the cognitively normal subgroup, a 5 μg/m3 increase in 7-day average PM2.5 was associated with increased CSF VCAM-1 (a priori adjusted: 35.4 (9.7, 61.1) ng/ml) (Fig. 1; Supplementary Table 2). Results were robust to restriction to Beacon Hill monitor data only and using a 12-df spline adjustment for year. Effect estimates for the 2-day and 5-day average periods were consistent with the 7-day average effect; however, estimates were attenuated and consistent with a range of effects for the 1-month average period. There was no evidence of effect modification by age.

Fig. 1.

Fig. 1.

Open in a new tab

Estimated associations between PM2.5 and VCAM-1 among cognitively normal individuals (a priori adjusted model)1.

1Model adjusted for age, smoking, and sex.

We also estimated positive associations between 1-year average PM2.5 and CSF VCAM-1 among cognitively normal individuals (Fig. 1; Supplementary Table 2). Based on our a priori adjustment model (model 2), a 5 μg/m3 increase in 1-year average PM2.5 was associated with a 51.8 (6.5, 97.1) ng/ml increase in CSF VCAM-1. Estimates were slightly strengthened with increased covariate adjustment. Results were somewhat attenuated but overall robust to sensitivity analyses evaluating other long-term exposure periods (5-year, 10-year, 20-year) and dropping PO box individuals. Effect estimates were strengthened when year of cohort enrollment was added to the analytical model.

Among the MCI/AD subgroup, estimates for 7-day and 1-year average PM2.5 were positive yet consistent with a wide range of effects (7-day: 16.2 (−8.5, 40.8) ng/ml; 1-year: 17.1 (−33.9, 68.2) ng/ml) (Supplementary Table 4). There was no evidence of effect modification by age. In sensitivity analyses, 2-day average PM2.5 was associated with increased VCAM-1 (21.7 (1.2, 42.2) ng/ml), but other results were consistent with the primary analyses. Sensitivity analyses evaluating other long-term exposure periods (5-year, 10-yeaer, 20-year), adding adjustment for enrollment year, and dropping individuals with PO box addresses only were also consistent with the primary results.

E-selectin

Our analysis of 7-day average PM2.5 and e-selectin was inconclusive (−1.8 (−27.1, 23.4) pg/ml) in the cognitively normal subgroup (Fig. 2; Supplementary Table 3). We observed similar results in sensitivity analyses. There was no evidence of effect modification by age.

Fig. 2.

Fig. 2.

Open in a new tab

Estimated associations between PM2.5 and E-selectin among cognitively normal individuals (a priori adjusted model)1.

1Model adjusted for age, smoking, and sex.

However, we estimated positive associations between 1-year average PM2.5 and e-selectin. Based on our a priori model, a 5 μg/m3 increase in 1-year average PM2.5 was associated with a 53.3 (11.0, 95.5) pg/ml increase in CSF e-selectin (Fig. 2; Supplementary Table 3). In sensitivity analyses evaluating other long-term exposure periods, results from the 5-year and 10-year averages were consistent with results from the 1-year average; restriction to individuals without PO box addresses strengthened the observed associations. Effect estimates were slightly attenuated when year of cohort enrollment was added to the analytical model; however, conclusions from the a priori model were consistent the primary model.

Among the MCI/AD subgroup, estimates for 7-day and 1-year average PM2.5 were consistent with a wide range of effects (7-day: −9.5 (−28.2, 9.1) pg/ml; 1-year: 23.0 (−15.1, 61.2) pg/ml) (Supplementary Table 5). Sensitivity analyses evaluating other exposure periods, adding adjustment for enrollment year, and dropping individuals with PO box addresses only were consistent with the primary results.

DISCUSSION

To our knowledge, this is the first study to evaluate the association between PM2.5 exposure and CSF biomarkers of endothelial injury. Among cognitively normal individuals, we estimate that 7-day and 1-year average PM2.5 exposure is associated with increased CSF VCAM-1, and that 1-year average PM2.5 is associated with increased CSF e-selectin. Among individuals with existing MCI/AD, associations were inconclusive for both short-term and long-term PM2.5 exposure and the selected biomarkers.

Some prior epidemiological studies have evaluated the association between PM2.5 and these biomarkers of endothelial injury in other biofluids, such as serum and plasma. While CSF and blood are not directly comparable, we provide results from these prior studies as context for our findings (Table 2). Most of these studies report associations between short-term exposure to PM2.5 and VCAM-1, in alignment with our results. However, our percent change estimate for VCAM-1 among cognitively normal individuals (28.0 (7.7, 48.3)%) was larger than those reported for comparable time periods in prior work. To our knowledge, only one prior study has considered PM2.5 in relation to e-selectin [20]; estimates from this study are several orders of magnitude larger than what we observed in our analyses of PM2.5 and e-selectin (7-day: −1.8 (−27.1, 23.4) pg/ml; 1-year: 53.3 (11.0, 95.5) pg/ml).

Table 2.

Results from prior epidemiological studies evaluating effects of PM2.5 on blood VCAM-1 or E-selectin

Author Study Cohort/Location PM2.5 increment (μg/m3) Blood biomarker Avg. Period: Effect Estimate (% change (95% CI) or mean change (95% CI)
Bind et al. [10] Veterans Administration Normative Aging Study / Boston, MA, USA 7.1 Plasma VCAM-1 4-h, 24-h, 3–28 day: 2–5%1
Wilker et al. [13] Veterans Administration Normative Aging Study / Boston, MA, USA 4.3 Serum VCAM-1 7-day: 2.5 (0.6, 4.5)%
Pope et al. [21] Provo, UT, USA 10 Plasma VCAM-1 24-h: 0.5%1; (also reported as 2.3 (0.3, 4.3) ng/ml increase)
O’Neill etal. [18] Boston, MA; diabetics only 7.6 Plasma VCAM-1 All diabetics
Same-day: 6.9 (−2.9, 17.6)%
2-day: 8.2 (−1.4, 18.7)%
3-day: 6.9 (−1.7, 16.3)%
4-day: 6.5 (−1.2,14.7)%
5-day: 8.6 (0.1, 17.8)%
6-day: 11.8(3.5,20.7)%
Excluding statin users
Same-day: 10.3 (−0.6, 22.4)%
2-day: 15.0 (3.8, 27.5)%
3-day: 14.6 (3.9, 26.3)%
4-day: 15.2 (4.5, 26.8)%
5-day: 16.2 (5.8, 27.6)%
6-day: 17.7 (7.8, 28.5)%
Madrigano et al. [17] Veterans Administration Normative Aging Study / Boston, MA, USA 10 Plasma VCAM-1 1-day: 1.0 (−1.1, 3.2)%
2-day: 1.7 (−0.9, 4.3)%
3-day: 0.4 (−2.6, 3.3)%
Liu et al. [19] Shanghai, China 27.4 Serum VCAM-1 24-h: 12%2
Delfino et al. [14] Los Angeles, CA, USA 11.53 Plasma VCAM-1 Same-day: 3.4 (−7.5, 14.2) ng/ml
2-day: 8.7 (−8.6, 26.0) ng/ml
3-day: 6.1 (−18.6, 30.7) ng/ml
4-day: −2.8 (−31.8, 26.2) ng/ml
Hajat et al. [20] Multi-Ethnic Study of Atherosclerosis (MESA) / Baltimore, MD; Chicago, IL; Winston-Salem, NC; Los Angeles, CA; New York, NY; St. Paul, MN, USA 5 Serum E-selectin Same-day: 600 (60, 1140) pg/ml
Day-prior: 390 (−130, 910) pg/ml
2-day: 440 (−200, 1080) pg/ml
3-day: −20 (−740, 690) pg/ml
4-day: −370 (−1160, 420) pg/ml
5-day: −330 (−1160, 500) pg/ml
1-year: 1100 (−700, 2800) pg/ml
Open in a new tab
1

Estimated from figure; exact numbers not provided in manuscript.

2

Confidence interval not provided.

3

Estimates for this study were based on PM0.25–2.5 μg/m3.

Results from experimental studies evaluating these biomarkers in other biofluids have been mixed. In mice, 9-month exposure to PM2.5 was not associated with increased VCAM-1 in the temporal cortex [54]. In one in vitro study using a human umbilical vein cell line, PM2.5 exposure resulted in a dose-dependent increase in VCAM-1 expression. This elevated VCAM-1 was attenuation by co-treatment with a scavenger of reactive oxidative species, demonstrating the role of oxidative stress in PM2.5-related endothelial inflammation [55]. Another study using the same cell line yielded inconsistent effects on VCAM-1 but a dose-dependent increase in expression of E-selectin [56].

Further context for our results can be obtained through comparison with prior limited research on other predictors for these CSF biomarkers; these studies report effect estimates in the range of our estimate associated with a 5 μg/m3 increase in 1-year PM2.5 (51.8 (6.5, 97.1) ng/ml). In a previous study utilizing a cohort with partial overlap to ours, individuals with diabetes had higher CSF VCAM-1 than those without the condition (61.0 (28.0, 95.0) ng/ml). Similarly, an increase in age from 50 to 75 years was associated with a 66.0 (30.0, 102.0) ng/ml increase in CSF VCAM-1 [45]. In our own dataset using the a priori model among the cognitively normal cohort, a 25-year increase in age was associated with an 87.5 (51.7, 123.3) ng/ml increase in CSF VCAM-1. It should also be noted that another study, also using a cohort with partial overlap to ours, did not report differences in CSF VCAM-1 or e-selectin by smoking status [27]. This is perhaps counterintuitive given the known links between smoking and endothelial injury [57]. Yet, while smoking is often used to represent high air pollution exposure scenarios, some studies do not report links to changes in serum VCAM-1 [58, 59], indicating the possibility of different mechanisms of action or the presence of adaptation.

Nevertheless, our results indicating positive associations between PM2.5 and CSF biomarkers of endothelial injury among cognitively normal individuals are consistent with much of the prior epidemiological research and the general scientific consensus that PM2.5 is associated with inflammation and endothelial injury [4]. Our novel findings suggest potential cerebrovascular and neuroinflammation effects from PM2.5 among cognitively normal individuals; however, we were unable to determine whether these changes also occurred in the systemic circulation. These results are aligned with research suggesting that short-term PM2.5 is linked to increased cerebrovascular resistance and decreased cerebral blood flow velocity [60]. Given the growing recognition of the role of vascular injury and neuroinflammation in neurodegeneration [3, 30], including recent findings suggesting associations between CSF VCAM-1 and CSF tau [61, 62], these cerebrovascular and inflammation changes may provide evidence to support the link between PM2.5, cognitive decline, and dementia [31–33].

Our inconclusive results among individuals classified as MCI/AD are not surprising. Increasing evidence indicates the importance of vascular pathways and the presence of endothelial dysfunction in dementia [3, 6, 7, 23, 27, 30]. Thus, in individuals with MCI or AD, existing pathological processes related to cerebrovascular injury may play a more important role in mediating VCAM-1 and e-selectin expression than PM2.5 exposure—and thus would obscure any potential PM2.5 effects.

This study has several limitations. First, there is possible exposure misclassification for both long-term and short-term exposures to PM2.5. With respect to long-term averages, we only had access to a single address for each individual from which to estimate exposures. This address may not represent actual participant address over longer time periods, which could have resulted in exposure misclassification, particularly for our 10-and 20-year exposure measures. However, our concerns about exposure misclassification are somewhat mitigated given that our primary exposure period for the long-term analysis was focused on 1-year average exposures. Effects of possible exposure misclassification due to use of PO box rather than residential address was addressed through sensitivity analyses, as described above, which indicated consistent or strengthened associations with restriction to residential address only. For our short-term analysis, we focused only on temporal variation. Ignoring spatial variability may result in some exposure misclassification [38, 63]; however, properly addressing spatial variation at the daily scale is a challenge that might have offset the benefits of incorporating it, and temporal variation is the more important determinant of variability at this scale. There is also possible residual confounding in our study. We did not have access to several covariates, such as socioeconomic status (SES), secondhand smoke exposure, medication use, and sleep quality, that may be linked to the exposure, outcomes, or factors that would affect our measurements of the outcomes [14, 64–71]. In one prior study among diabetics, associations between PM2.5 and plasma VCAM-1 were stronger in those not taking statins compared to the full population [18] (Table 2), yet we were not able to account for statin use. The closest approximation to SES that we had was education level, which may not fully capture the effects of SES in the causal pathway. Future studies evaluating this question should seek to integrate these variables. Finally, this cohort is a small convenience sample based in the Puget Sound region, and results may not be generalizable.

Despite these limitations, our study has important strengths. We evaluated both short-term and long-term exposure PM2.5 in relation to CSF biomarkers of endothelial injury, neither of which had been investigated previously despite the well-documented vascular effects of PM2.5. By including both of these exposure periods, we capture variation that is predominantly spatial (long-term) and temporal (short-term). With our pre-whitening approach for the short-term PM2.5 analysis, we were able to address most of the confounding on a short time scale that is difficult to remove in a cohort study, particularly those with small sample sizes. Finally, our findings are particularly meaningful because they suggest that PM2.5 may have effects on cerebrovascular injury and related neuroinflammation at even the low levels observed in the Puget Sound region. Prior research on PM2.5 also suggests effects on biomarkers of endothelial injury, functional cerebrovascular changes, vascular disease, and mortality at low exposure levels [13, 60, 72–74].

To our knowledge, our study is the first to evaluate the associations between short-term and long-term PM2.5 exposure with biomarkers of endothelial injury in the CSF. Here, we estimated that short-term and long term PM2.5 exposure was associated with elevated biomarkers of endothelial injury among cognitively normal individuals; no convincing associations were observed for MCI/AD individuals. The clinical significance of these effects among cognitively normal individuals is unknown. Nevertheless, in the context of the growing recognition of the role of vascular injury in neurodegeneration and dementia [3, 30, 62], our results may provide evidence to support the link between air pollution and cognitive decline or dementia [31–33].

Supplementary Material

supplementary

ACKNOWLEDGMENTS

RMS was supported by NIEHS T32ES015459, NIA T32AG052354, the University of Washington Retirement Association Aging Fellowship, and the Seattle Chapter of the Achievement Rewards for College Scientists (ARCS) Foundation.

The work was supported by NIEHS ES026187, NIA AG05136, R01 AG056711, U01 NS091272, the Department of Veterans Affairs and an anonymous foundation.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies.

The authors would like to thank Amanda J. Gassett and Cooper Schumacher at the University of Washington for their assistance in obtaining fine particulate matter exposure estimates and predictions for these analyses.

Footnotes

Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/19-0563r1).

SUPPLEMENTARY MATERIAL

The supplementary material is available in the electronic version of this article: http://dx.doi.org/10.3233/JAD-190563.

REFERENCES

  • [1].Snyder HM, Corriveau RA, Craft S, Faber JE, Greenberg SM, Knopman D, Lamb BT, Montine TJ, Nedergaard M, Schaffer CB, Schneider JA, Wellington C, Wilcock DM, Zipfel GJ, Zlokovic B, Bain LJ, Bosetti F, Galis ZS, Koroshetz W, Carrillo MC (2015) Vascular contributions to cognitive impairment and dementia including Alzheimer’s disease. Alzheimers Dement 11, 710–717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [2].Kisler K, Nelson AR, Montagne A, Zlokovic BV (2017) Cerebral blood flow regulation and neurovascular dysfunction in Alzheimer disease. Nat Rev Neurosci 18, 419–434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [3].Gorelick PB, Scuteri A, Black SE, Decarli C, Greenberg SM, Iadecola C, Launer LJ, Laurent S, Lopez OL, Nyenhuis D, Petersen RC, Schneider JA, Tzourio C, Arnett DK, Bennett DA, Chui HC, Higashida RT, Lindquist R, Nilsson PM, Roman GC, Sellke FW, Seshadri S (2011) Vascular contributions to cognitive impairment and dementia: A statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 42, 2672–2713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Brook RD, Rajagopalan S, Pope CA 3rd, Brook JR, Bhatnagar A, Diez-Roux AV, Holguin F, Hong Y, Luepker RV, Mittleman MA, Peters A, Siscovick D, Smith SC Jr, Whitsel L, Kaufman JD (2010) Particulate matter air pollution and cardiovascular disease: An update to the scientific statement from the American Heart Association. Circulation 121, 2331–2378. [DOI] [PubMed] [Google Scholar]
  • [5].Wellenius GA, Burger MR, Coull BA, Schwartz J, Suh HH, Koutrakis P, Schlaug G, Gold DR, Mittleman MA (2012) Ambient air pollution and the risk of acute ischemic stroke. JAMA Intern Med 172, 229–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [6].Zuliani G, Cavalieri M, Galvani M, Passaro A, Munari MR, Bosi C, Zurlo A, Fellin R (2008) Markers of endothelial dysfunction in older subjects with late onset Alzheimer’s disease or vascular dementia. J Neurol Sci 272, 164–170. [DOI] [PubMed] [Google Scholar]
  • [7].Ewers M, Mielke MM, Hampel H (2010) Blood-based biomarkers of microvascular pathology in Alzheimer’s disease. Exp Gerontol 45, 75–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [8].Mulvihill NT, Foley B, Crean P, Walsh M (2002) Prediction of cardiovascular risk using soluble cell adhesion molecules. Eur Heart J 23, 1569–1574. [DOI] [PubMed] [Google Scholar]
  • [9].Tchalla AE, Wellenius GA, Travison TG, Gagnon M, Iloputaife I, Dantoine T, Sorond FA, Lipsitz LA (2015) Circulating vascular cell adhesion molecule-1 is associated with cerebral blood flow dysregulation, mobility impairment, and falls in older adults. Hypertension 66, 340–346. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Bind MA, Baccarelli A, Zanobetti A, Tarantini L, Suh H, Vokonas P, Schwartz J (2012) Air pollution and markers of coagulation, inflammation, and endothelial function: Associations and epigene-environment interactions in an elderly cohort. Epidemiology 23, 332–340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Salvi S, Blomberg A, Rudell B, Kelly F, Sandstrom T, Holgate ST, Frew A (1999) Acute inflammatory responses in the airways and peripheral blood after short-term exposure to diesel exhaust in healthy human volunteers. Am J Resp Crit Care Med 159, 702–709. [DOI] [PubMed] [Google Scholar]
  • [12].Alexeeff SE, Coull BA, Gryparis A, Suh H, Sparrow D, Vokonas PS, Schwartz J (2011) Medium-term exposure to traffic-related air pollution and markers of inflammation and endothelial function. Environ Health Perspect 119, 481–486. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Wilker EH, Alexeeff SE, Suh H, Vokonas PS, Baccarelli A, Schwartz J (2011) Ambient pollutants, polymorphisms associated with microRNA processing and adhesion molecules: The Normative Aging Study. Environ Health 10, 45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Delfino RJ, Staimer N, Tjoa T, Polidori A, Arhami M, Gillen DL, Kleinman MT, Vaziri ND, Longhurst J, Zaldivar F, Sioutas C (2008) Circulating biomarkers of inflammation, antioxidant activity, and platelet activation are associated with primary combustion aerosols in subjects with coronary artery disease. Environ Health Perspect 116, 898–906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Tornqvist H, Mills NL, Gonzalez M, Miller MR, Robinson SD, Megson IL, Macnee W, Donaldson K, Soderberg S, Newby DE, Sandstrom T, Blomberg A (2007) Persistent endothelial dysfunction in humans after diesel exhaust inhalation. Am J Resp Crit Care Med 176, 395–400. [DOI] [PubMed] [Google Scholar]
  • [16].Schneider A, Neas L, Herbst MC, Case M, Williams RW, Cascio W, Hinderliter A, Holguin F, Buse JB, Dungan K, Styner M, Peters A, Devlin RB (2008) Endothelial dysfunction: Associations with exposure to ambient fine particles in diabetic individuals. Environ Health Perspect 116, 1666–1674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Madrigano J, Baccarelli A, Wright RO, Suh H, Sparrow D, Vokonas PS, Schwartz J (2010) Air pollution, obesity, genes, and cellular adhesion molecules. Occup Environ Med 67, 312–317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].O’Neill MS, Veves A, Sarnat JA, Zanobetti A, Gold DR, Economides PA, Horton ES, Schwartz J (2007) Air pollution and inflammation in type 2 diabetes: A mechanism for susceptibility. Occup Environ Med 64, 373–379. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Liu C, Cai J, Qiao L, Wang H, Xu W, Li H, Zhao Z, Chen R, Kan H (2017) The acute effects of fine particulate matter constituents on blood inflammation and coagulation. Environ Sci Technol 51, 8128–8137. [DOI] [PubMed] [Google Scholar]
  • [20].Hajat A, Allison M, Diez-Roux AV, Jenny NS, Jorgensen NW, Szpiro AA, Vedal S, Kaufman JD (2015) Long-term exposure to air pollution and markers of inflammation, coagulation, and endothelial activation: A repeat-measures analysis in the Multi-Ethnic Study of Atherosclerosis (MESA). Epidemiology 26, 310–320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Pope CA 3rd, Bhatnagar A, McCracken JP, Abplanalp W, Conklin DJ, O’Toole T (2016) Exposure to fine particulate air pollution is associated with endothelial injury and systemic inflammation. Circ Res 119, 1204–1214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Lee KK, Miller MR, Shah ASV (2018) Air pollution and stroke. J Stroke 20, 2–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Galasko D, Montine TJ (2010) Biomarkers of oxidative damage and inflammation in Alzheimer’s disease. Biomark Med 4, 27–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Montine TJ, Markesbery WR, Morrow JD, Roberts LJ 2nd (1998) Cerebrospinal fluid F2-isoprostane levels are increased in Alzheimer’s disease. Ann Neurol 44, 410–413. [DOI] [PubMed] [Google Scholar]
  • [25].Montine TJ, Peskind ER, Quinn JF, Wilson AM, Montine KS, Galasko D (2011) Increased cerebrospinal fluid F2-isoprostanes are associated with aging and latent Alzheimer’s disease as identified by biomarkers. Neuro-molecular Med 13, 37–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Montine TJ, Quinn J, Kaye J, Morrow JD (2007) F2-isoprostanes as biomarkers of late-onset Alzheimer’s disease. J Mol Neurosci 33, 114–119. [DOI] [PubMed] [Google Scholar]
  • [27].Li G, Xiong K, Korff A, Pan C, Quinn JF, Galasko DR, Liu C, Montine TJ, Peskind ER, Zhang J (2015) Increased CSF e-selectin in clinical Alzheimer’s disease without altered CSF Abeta42 and tau. J Alzheimers Dis 47, 883–887. [DOI] [PubMed] [Google Scholar]
  • [28].Montine TJ, Beal MF, Cudkowicz ME, O’Donnell H, Margolin RA, McFarland L, Bachrach AF, Zackert WE, Roberts LJ, Morrow JD (1999) Increased CSF F2-isoprostane concentration in probable AD. Neurology 52, 562–565. [DOI] [PubMed] [Google Scholar]
  • [29].Montine KS, Quinn JF, Zhang J, Fessel JP, Roberts LJ 2nd, Morrow JD, Montine TJ (2004) Isoprostanes and related products of lipid peroxidation in neurodegenerative diseases. Chem Phys Lipids 128, 117–124. [DOI] [PubMed] [Google Scholar]
  • [30].Zlokovic BV (2011) Neurovascular pathways to neurodegeneration in Alzheimer’s disease and other disorders. Nat Rev Neurosci 12, 723–738. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Block ML, Calderon-Garciduenas L (2009) Air pollution: Mechanisms of neuroinflammation and CNS disease. Trends Neurosci 32, 506–516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Power MC, Adar SD, Yanosky JD, Weuve J (2016) Exposure to air pollution as a potential contributor to cognitive function, cognitive decline, brain imaging, and dementia: A systematic review of epidemiologic research. NeuroToxicology 56, 235–253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Jayaraj RL, Rodriguez EA, Wang Y, Block ML (2017) Out-door ambient air pollution and neurodegenerative diseases: The neuroinflammation hypothesis. Curr Environ Health Reports 4, 166–179. [DOI] [PubMed] [Google Scholar]
  • [34].Peskind ER, Riekse R, Quinn JF, Kaye J, Clark CM, Farlow MR, Decarli C, Chabal C, Vavrek D, Raskind MA, Galasko D (2005) Safety and acceptability of the research lumbar puncture. Alzheimer Dis Assoc Disord 19, 220–225. [DOI] [PubMed] [Google Scholar]
  • [35].Pan C, Korff A, Galasko D, Ginghina C, Peskind E, Li G, Quinn J, Montine TJ, Cain K, Shi M, Zhang J (2015) Diagnostic values of cerebrospinal fluid t-tau and Abeta(4)(2) using Meso Scale Discovery assays for Alzheimer’s disease. J Alzheimers Dis 45, 709–719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].Hixson JE, Vernier DT (1990) Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res 31, 545–548. [PubMed] [Google Scholar]
  • [37].Noble CA, Vanderpool RW, Peters TM, McElroy FF, Gemmill DB, Wiener RW (2001) Federal reference and equivalent methods for measuring fine particulate matter. Aerosol Sci Technol 34, 457–464. [Google Scholar]
  • [38].Szpiro AA, Sheppard L, Adar SD, Kaufman JD (2014) Estimating acute air pollution health effects from cohort study data. Biometrics 70, 164–174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [39].Sampson PD, Richards M, Szpiro AA, Bergen S, Sheppard L, Larson TV, Kaufman JD (2013) A regionalized national universal kriging model using Partial Least Squares regression for estimating annual PM2.5 concentrations in epidemiology. Atmos Environ 75, 383–392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [40].Kress BT, Iliff JJ, Xia M, Wang M, Wei HS, Zeppenfeld D, Xie L, Kang H, Xu Q, Liew JA, Plog BA, Ding F, Deane R, Nedergaard M (2014) Impairment of paravascular clearance pathways in the aging brain. Ann Neurol 76, 845–861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [41].Reiter RJ (1995) Oxidative processes and antioxidative defense mechanisms in the aging brain. FASEB J 9, 526–533. [PubMed] [Google Scholar]
  • [42].O’Neill MS, Veves A, Zanobetti A, Sarnat JA, Gold DR, Economides PA, Horton ES, Schwartz J (2005) Diabetes enhances vulnerability to particulate air pollution-associated impairment in vascular reactivity and endothelial function. Circulation 111, 2913–2920. [DOI] [PubMed] [Google Scholar]
  • [43].Zanobetti A, Schwartz J (2001) Are diabetics more susceptible to the health effects of airborne particles? Am J Resp Crit Care Med 164, 831–833. [DOI] [PubMed] [Google Scholar]
  • [44].Zanobetti A, Schwartz J (2002) Cardiovascular damage by airborne particles: Are diabetics more susceptible? Epidemiology 13, 588–592. [DOI] [PubMed] [Google Scholar]
  • [45].Li G, Shofer JB, Petrie EC, Yu C-E, Wilkinson CW, Figlewicz DP, Shutes-David A, Zhang J, Montine TJ, Raskind MA, Quinn JF, Galasko DR, Peskind ER (2017) Cerebrospinal fluid biomarkers for Alzheimer’s and vascular disease vary by age, gender, and APOE genotype in cognitively normal adults. Alzheimers Res Ther 9, 48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [46].Ott A, Breteler MMB, van Harskamp F, Claus JJ, van der Cammen TJM, Grobbee DE, Hofman A (1995) Prevalence of Alzheimer’s disease and vascular dementia: Association with education. The Rotterdam study. BMJ 310, 970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [47].Sharp ES, Gatz M (2011) The relationship between education and dementia: An updated systematic review. Alzheimer Dis Assoc Disord 25, 289–304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [48].Auchincloss AH, Diez Roux AV, Dvonch JT, Brown PL, Barr RG, Daviglus ML, Goff DC, Kaufman JD, O’Neill MS (2008) Associations between recent exposure to ambient fine particulate matter and blood pressure in the Multi-Ethnic Study of Atherosclerosis (MESA). Environ Health Perspect 116, 486–491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [49].Chen H, Burnett RT, Kwong JC, Villeneuve PJ, Goldberg MS, Brook RD, van Donkelaar A, Jerrett M, Martin RV, Kopp A (2014) Spatial association between ambient fine particulate matter and incident hypertension. Circulation 129, 562–569. [DOI] [PubMed] [Google Scholar]
  • [50].Posner HB, Tang MX, Luchsinger J, Lantigua R, Stern Y, Mayeux R (2000) The relationship of hypertension in the elderly to AD, vascular dementia, and cognitive function. Neurology 58, 1175–1181. [DOI] [PubMed] [Google Scholar]
  • [51].Sharp SI, Aarsland D, Day S, Sønnesyn H, Ballard C (2011) Hypertension is a potential risk factor for vascular dementia: Systematic review. Int J Geriatr Psychiatry 26, 661–669. [DOI] [PubMed] [Google Scholar]
  • [52].Rajagopalan S, Brook RD (2012) Air pollution and type 2 diabetes: Mechanistic insights. Diabetes 61, 3037–3045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [53].Ahtiluoto S, Polvikoski T, Peltonen M, Solomon A, Tuomilehto J, Winblad B, Sulkava R, Kivipelto M (2010) Diabetes, Alzheimer disease, and vascular dementia. Neurology 75, 1195–1202. [DOI] [PubMed] [Google Scholar]
  • [54].Bhatt DP, Puig KL, Gorr MW, Wold LE, Combs CK (2015) A pilot study to assess effects of long-term inhalation of airborne particulate matter on early Alzheimer-like changes in the mouse brain. PLoS One 10, e0127102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [55].Rui W, Guan L, Zhang F, Zhang W, Ding W (2016) PM2.5-induced oxidative stress increases adhesion molecules expression in human endothelial cells through the ERK/AKT/NF-kappaB-dependent pathway. J Appl Toxicol 36, 48–59. [DOI] [PubMed] [Google Scholar]
  • [56].Montiel-Davalos A, Alfaro-Moreno E, Lopez-Marure R (2007) PM2.5 and PM10 induce the expression of adhesion molecules and the adhesion of monocytic cells to human umbilical vein endothelial cells. Inhal Toxicol 19(Suppl 1), 91–98. [DOI] [PubMed] [Google Scholar]
  • [57].Pittilo M (2000) Cigarette smoking, endothelial injury and cardiovascular disease. Int J Exp Pathol 81, 219–230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [58].Demerath E, Towne B, Blangero J, Siervogel RM (2001) The relationship of soluble ICAM-1, VCAM-1, P-selectin and E-selectin to cardiovascular disease risk factors in healthy men and women. Ann Hum Biol 28, 664–678. [DOI] [PubMed] [Google Scholar]
  • [59].Adams MR, Jessup W, Celermajer DS (1997) Cigarette smoking is associated with increased human monocyte adhesion to endothelial cells: Reversibility with oral larginine but not vitamin C. J Am Coll Cardiol 29, 491–497. [DOI] [PubMed] [Google Scholar]
  • [60].Wellenius GA, Boyle LD, Wilker EH, Sorond FA, Coull BA, Koutrakis P, Mittleman MA, Lipsitz LA (2013) Ambient fine particulate matter alters cerebral hemodynamics in the elderly. Stroke 44, 1532–1536. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [61].Popp J, Oikonomidi A, Tautvydaitė D, Dayon L, Bacher M, Migliavacca E, Henry H, Kirkland R, Severin I, Wojcik J, Bowman GL (2017) Markers of neuroinflammation associated with Alzheimer’s disease pathology in older adults. Brain Behav Immun 62, 203–211. [DOI] [PubMed] [Google Scholar]
  • [62].Janelidze S, Mattsson N, Stomrud E, Lindberg O, Palmqvist S, Zetterberg H, Blennow K, Hansson O (2018) CSF biomarkers of neuroinflammation and cerebrovascular dysfunction in early Alzheimer disease. Neurology 91, e867–e877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [63].Zeger SL, Thomas D, Dominici F, Samet JM, Schwartz J, Dockery D, Cohen A (2000) Exposure measurement error in time-series studies of air pollution: Concepts and consequences. Environ Health Perspect 108, 419–426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [64].Kato T, Inoue T, Morooka T, Yoshimoto N, Node K (2006) Short-term passive smoking causes endothelial dysfunction via oxidative stress in nonsmokers. Can J Physiol Pharmacol 84, 523–529. [DOI] [PubMed] [Google Scholar]
  • [65].Kosecik M, Erel O, Sevinc E, Selek S (2005) Increased oxidative stress in children exposed to passive smoking. Int J Cardiol 100, 61–64. [DOI] [PubMed] [Google Scholar]
  • [66].Montecucco F, Burger F, Pelli G, Poku NK, Berlier C, Steffens S, Mach F (2009) Statins inhibit C-reactive protein-induced chemokine secretion, ICAM-1 upregulation and chemotaxis in adherent human monocytes. Rheumatology (Oxford) 48, 233–242. [DOI] [PubMed] [Google Scholar]
  • [67].Liang YJ, Shyu KG, Wang BW, Lai LP (2008) Simvastatin inhibits C-reactive protein-induced pro-inflammatory changes in endothelial cells by decreasing mevalonate pathway products. Cardiology 110, 182–190. [DOI] [PubMed] [Google Scholar]
  • [68].Blanco-Colio LM, Martin-Ventura JL, de Teresa E, Farsang C, Gaw A, Gensini G, Leiter LA, Langer A, Martineau P, Egido J (2007) Elevated ICAM-1 and MCP-1 plasma levels in subjects at high cardiovascular risk are diminished by atorvastatin treatment. Atorvastatin on Inflammatory Markers study: A substudy of Achieve Cholesterol Targets Fast with Atorvastatin Stratified Titration. Am Heart J 153, 881–888. [DOI] [PubMed] [Google Scholar]
  • [69].Schwartz J, Park SK, O’Neill MS, Vokonas PS, Sparrow D, Weiss S, Kelsey K (2005) Glutathione-S-transferase M1, obesity, statins, and autonomic effects of particles: Gene-by-drug-by-environment interaction. Am J Resp Crit Care Med 172, 1529–1533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [70].Xie L, Kang H, Xu Q, Chen MJ, Liao Y, Thiyagarajan M, O’Donnell J, Christensen DJ, Nicholson C, Iliff JJ, Takano T, Deane R, Nedergaard M (2013) Sleep drives metabolite clearance from the adult brain. Science 342, 373–377. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [71].Hajat A, Diez-Roux AV, Adar SD, Auchincloss AH, Lovasi GS, O’Neill MS, Sheppard L, Kaufman JD (2013) Air pollution and individual and neighborhood socioeconomic status: Evidence from the Multi-Ethnic Study of Atherosclerosis (MESA). Environ Health Perspect 121, 1325–1333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [72].Pope CA 3rd, Burnett RT, Krewski D, Jerrett M, Shi Y, Calle EE, Thun MJ (2009) Cardiovascular mortality and exposure to airborne fine particulate matter and cigarette smoke: Shape of the exposure-response relationship. Circulation 120, 941–948. [DOI] [PubMed] [Google Scholar]
  • [73].Xie W, Li G, Zhao D, Xie X, Wei Z, Wang W, Wang M, Liu W, Sun J, Jia Z, Zhang Q, Liu J (2015) Relationship between fine particulate air pollution and ischaemic heart disease morbidity and mortality. Heart 101, 257–263. [DOI] [PubMed] [Google Scholar]
  • [74].Stafoggia M, Samoli E, Alessandrini E, Cadum E, Ostro B, Berti G, Faustini A, Jacquemin B, Linares C, Pascal M, Randi G, Ranzi A, Stivanello E, Forastiere F (2013) Short-term associations between fine and coarse particulate matter and hospitalizations in Southern Europe: Results from the MED-PARTICLES project. Environ Health Perspect 121, 1026–1033. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplementary
NIHMS1577569-supplement-supplementary.docx (44.9KB, docx)

RESOURCES