Abstract
Rationale: Particulate air pollution is associated with cardiovascular morbidity. One hypothesized mechanism involves oxidative stress, systemic inflammation, and endothelial dysfunction.
Objectives: To assess an intervention's impact on particle exposures and endothelial function among healthy adults in a woodsmoke-impacted community. We also investigated the underlying role of oxidative stress and inflammation in relation to exposure reductions.
Methods: Portable air filters were used in a randomized crossover intervention study of 45 healthy adults exposed to consecutive 7-day periods of filtered and nonfiltered air.
Measurements and Main Results: Reactive hyperemia index was measured as an indicator of endothelial function via peripheral artery tonometry, and markers of inflammation (C-reactive protein, interleukin-6, and band cells) and lipid peroxidation (malondialdehyde and 8-iso-prostaglandin F2α) were quantified. Air filters reduced indoor fine particle concentrations by 60%. Filtration was associated with a 9.4% (95% confidence interval, 0.9–18%) increase in reactive hyperemia index and a 32.6% (4.4–60.9%) decrease in C-reactive protein. Decreases in particulate matter and the woodsmoke tracer levoglucosan were associated with reduced band cell counts. There was limited evidence of more pronounced effects on endothelial function and level of systemic inflammation among males, overweight participants, younger participants, and residents of wood-burning homes. No associations were noted for oxidative stress markers.
Conclusions: Air filtration was associated with improved endothelial function and decreased concentrations of inflammatory biomarkers but not markers of oxidative stress. Our results support the hypothesis that systemic inflammation and impaired endothelial function, both predictors of cardiovascular morbidity, can be favorably influenced by reducing indoor particle concentrations.
Clinical trial registered at www.clinicaltrials.gov (NCT01256957).
Exposure to particulate air pollution is associated with cardiovascular morbidity. One hypothesized mechanistic pathway involves oxidative stress, systemic inflammation, and endothelial dysfunction.
Portable air filters reduced indoor particulate air pollution, improved microvascular endothelial function, and reduced markers of systemic inflammation among healthy adults in a community heavily impacted by residential wood combustion. The cardiovascular effects of particulate matter may be mediated through systemic inflammation and impaired endothelial function, and these effects may be favorably influenced by a reduction of particle concentrations
Combustion-derived pollution is believed to play a particularly important role in the cardiovascular effects of air pollution (1), and there is now strong evidence linking traffic-related air pollution with cardiovascular morbidity and mortality (25). Although there is limited evidence to assess the impact of woodsmoke on cardiovascular health, studies of occupationally exposed populations or in controlled experimental settings suggest that short-term exposures to high concentrations of biomass emissions may also elicit a systemic inflammatory response (4, 26, 27).
Residential wood combustion (RWC) is an important source of ambient particulate matter in mid- and high-latitude climates (26). The importance of RWC as a source of air pollution is likely to increase due to the rising costs of other fuels and the promotion of wood as a “carbon neutral” and renewable fuel (28).
In this study we used portable high-efficiency particulate air (HEPA) filters in a randomized intervention crossover study design (9) to study the subclinical cardiovascular effects of PM with a diameter less than 2.5 μm (PM2.5) exposure in a woodsmoke-impacted airshed. Our main objectives were to better understand the mechanisms underlying air pollution–related cardiovascular morbidity and evaluate the potential for a simple intervention to reduce pollution-related cardiovascular health risks. HEPA filters are a potentially useful intervention because they are relatively inexpensive to purchase and operate and can effectively remove respirable particles (e.g., 99.97% of 0.3-μm diameter particles) to improve air quality inside homes, where the majority of time is spent (29–34). Our primary outcome was reactive hyperemia index (RHI), an indicator of microvascular endothelial function, because it represents an early pathology in the atherosclerotic process and predicts cardiovascular morbidity and mortality (19, 20, 35). Markers of oxidative stress (malondialdehyde [MDA], 8-iso-prostaglandin F2α, [8-isoprostane]) and inflammation (C-reactive protein [CRP], IL-6, and band cell counts) were considered exploratory endpoints to better understand potential pathways involved in endothelial dysfunction. Some of the results of this study have been previously reported in the form of an abstract (36).
This study was conducted from November 2008 to April 2009 in Smithers, British Columbia, Canada (population ∼ 5,300), where we have previously shown the outdoor air to be heavily impacted by RWC emissions (37). We recruited participants aged 19 years or older; individuals who resided in self-reported tobacco-smoking households were excluded from participating. The study protocol was approved by the research ethics boards at Simon Fraser University and the University of British Columbia, and written informed consent was obtained from all participants before enrollment. More details on the methods are available in the online supplement.
Each participant's home was monitored for two consecutive 7-day periods, during which time a HEPA filter (model 50300; Honeywell, Morristown, NJ) was operated in the main activity room and a quieter HEPA filter (Honeywell 18150) was operated in the participant's bedroom. HEPA filters were operated normally during one 7-day period and without the internal filters in place (i.e., placebo filtration) during the other period, thus blinding participants to the filters' status. The order of filtration or nonfiltration was random. Indoor pollution sampling equipment was placed in the home's main activity room.
At the end of each 7-day period, a study technician measured microvascular endothelial function and collected blood and urine samples at the participant's home. Microvascular endothelial function was measured via peripheral artery tonometry using the portable EndoPAT 2000 instrument (Itamar Medical Ltd, Cesari, Israel), which determines RHI based on a computer algorithm. Serum samples were analyzed for CRP and IL-6 by ELISA. A trained technician performed manual band cell counts on thin blood smears that were air dried, fixed with methanol, and stained with Wright stain. Band cell counts are expressed as the percent of polymorphonuclear leukocytes. Urine samples were analyzed for MDA and 8-isoprostane (not normalized to creatinine) via gas chromatography mass spectrometry and ELISA, respectively.
During each 7-day period, PM2.5 filter samples were collected indoors and outdoors using Harvard Impactors (Air Diagnostics and Engineering, Harrison, ME). Filters were analyzed for PM2.5 mass concentration and the woodsmoke tracer levoglucosan (26), and we partitioned indoor PM2.5 concentrations into indoor- and outdoor-generated components by first calculating the PM2.5 infiltration efficiency (Finf, the fraction of the outdoor concentration that penetrates indoors and remains suspended) for each home during HEPA filtration and placebo filtration using indoor and outdoor measurements made with nephelometers (Radiance Research, Seattle, WA) (38). Indoor temperature and relative humidity were logged continuously using HOBO data loggers (Onset Computer Corporation, Pocaseset, MA) in a subset (n = 13) of homes. Each participant recorded their locations and proximity to potential sources of PM exposure at 60-minute resolution.
Outcome variables were skewed, so we log-transformed before analysis (for band cells, which had 0 values, we added 0.5 before log-transforming). As a sensitivity analysis we also modeled RHI without log-transforming. We used mixed models to account for measurements clustered within individuals and individuals clustered within homes. All models were adjusted for sex, age, body mass index (BMI), and temperature. We explored effect modification by filtration/placebo order, age (> or ≤ 43 yr, the median age), sex, overweight (BMI > or ≤ 25 kg/m2), time spent indoors at home (> or ≤ 75%), and use of a wood stove.
We enrolled a total of 56 participants from 31 homes. Before analysis, we excluded eight participants who did not have complete PM2.5 and Finf data to allow for direct comparisons of effects between different exposure indicators. In addition, before analysis we removed one pregnant participant, one participant with Raynaud syndrome, and one participant who reported being highly exposed to environmental tobacco smoke the night before a technician visit.
The final study population for analysis consisted of 45 participants, from 25 homes, with complete paired HEPA and non-HEPA period data (Table 1). The mean age for the included participants was 43.0 ± 9.9 years (range, 20–63), there was a nearly even sex balance (53% female), and most (89%) of the participants reported working or volunteering outside the home. Twenty-three participants in 13 homes reported using a wood stove. Compared with the 45 participants with complete data, the 11 excluded participants were more likely to be female (8 out of 11, or 73%).
Variable | Mean ± SD or No. (%) |
|---|---|
| Age, yr | 43.0 ± 9.9 |
| BMI, kg/m2 | 25.7 ± 3.5 |
| Female | 24 (53) |
| Asthma | 2 (4) |
| Hypertension | 1 (2) |
| Diabetes | 0 (0) |
| Employed outside home | 40 (89) |
| Wood stove used in home | 23 (51) |
Outdoor temperatures during 2-week home monitoring sessions ranged between −10.7 and 4.9°C, and outdoor temperatures were similar during HEPA and non-HEPA periods (Table 2). Averages of Finf and all indoor concentrations were significantly lower during HEPA filtration, with nearly 60% reductions in average concentrations of indoor PM2.5 components and a 75% reduction in average indoor levoglucosan (Table 2). HEPA filters reduced indoor PM2.5 concentrations in 24 of 25 homes, and concentration changes during HEPA filtration in individual homes ranged between −15.7 and 2.2 μg/m3. PM2.5 and levoglucosan concentrations outdoors were similar under HEPA and non-HEPA conditions (Table 2). During both HEPA and non-HEPA periods, indoor-generated PM2.5 accounted for an average of 67% of the total indoor concentration. Consistent with our previous findings in this region (37), relatively high outdoor levoglucosan/PM2.5 ratios (mean > 5%; Table 2) and high PM2.5-levoglucosan correlations (Spearman r ≥ 0.82; see Table E1 in the online supplement) indicated a major contribution of woodsmoke to outdoor PM2.5 concentrations. Lower levoglucosan/PM2.5 ratios (mean ≤ 1%; Table 2) and correlations (r ≤ 0.53; Table E1) indoors indicated a smaller PM2.5 contribution from woodsmoke to indoor concentrations. Indoor-generated PM2.5 concentrations were generally higher in the 13 homes where participants reported burning wood (Figure 1). Median within-participant changes in indoor PM2.5, indoor-generated PM2.5, and levoglucosan were −7.5 μg/m3, −6.3 μg/m3, and −44 ng/m3 in wood-burning homes, whereas in non–wood-burning homes the median changes were −6.2 μg/m3, −2.1 μg/m3, and −58 ng/m3, respectively.
HEPA Off | HEPA On | ||||||
|---|---|---|---|---|---|---|---|
| Variable | Mean ± SD | Median | Mean ± SD | Median | Paired t Test P Value | ||
| 7-d Average outdoor temperature, °C | −2.5 ± 4.6 | −2.3 | −3.6 ± 6.1 | −1.7 | 0.32 | ||
| 7-d Average indoor temperature,* °C | 19.7 ± 1.4 | 19.4 | 19.8 ± 1.7 | 19.4 | 0.75 | ||
| 7-d Average indoor relative humidity,* % | 35.1 ± 3.3 | 36.0 | 35.3 ± 3.4 | 33.7 | 0.90 | ||
| PM2.5 outdoors, μg/m3 | 10.8 ± 5.0 | 9.0 | 9.8 ± 4.2 | 8.9 | 0.26 | ||
| PM2.5 infiltration efficiency (unitless) | 0.34 ± 0.17 | 0.30 | 0.20 ± 0.17 | 0.13 | <0.01 | ||
| PM2.5 indoors, μg/m3 | 11.2 ± 6.1 | 10.5 | 4.6 ± 2.6 | 3.9 | <0.01 | ||
| PM2.5 outdoor-generated, μg/m3 | 3.5 ± 2.3 | 3.6 | 1.5 ± 0.9 | 1.4 | <0.01 | ||
| PM2.5 indoor-generated, μg/m3 | 7.6 ± 6.6 | 6.3 | 3.0 ± 2.8 | 2.1 | <0.01 | ||
| Levoglucosan outdoors,† ng/m3 | 613 ± 548 | 415 | 530 ± 358 | 471 | 0.18 | ||
| Levoglucosan indoors, ng/m3 | 127 ± 191 | 73 | 33 ± 39 | 19 | 0.01 | ||
| Levoglucosan/PM2.5 outdoors,† % | 5.1 ± 2.8 | 5.3 | 5.3 ± 1.8 | 5.1 | 0.79 | ||
| Levoglucosan/PM2.5 indoors, % | 1.0 ± 1.1 | 0.7 | 0.9 ± 1.3 | 0.7 | 0.61 | ||
Figure 1. Distributions of indoor particulate matter (PM) with a diameter less than 2.5 μm (PM2.5) and levoglucosan concentrations by use of a wood-burning stove during periods without high-efficiency particulate air (HEPA) filtration (upper plot) and with HEPA filtration (lower plot). P values are for two-sample t tests comparing wood-burning and non–wood-burning homes. Note: outliers not shown. Lines in the boxes are the median concentrations.
[More] [Minimize]Participants' activity patterns were similar between HEPA and non-HEPA periods, as were durations spent cooking or exposed to environmental tobacco smoke (Table 3). The HEPA-related differences in biological measurements were generally in the hypothesized directions, with increases in median RHI and decreases in median CRP, band cell counts, IL-6, and MDA during periods of HEPA filtration. There was an increase in median concentrations of 8-isoprostane during HEPA filtration (Table 3). Only CRP and RHI were correlated (Spearman r: −0.31, P = 0.04) during “baseline” (non-HEPA periods); endpoints were not correlated during HEPA filtration periods.
HEPA Off | HEPA On | ||||||
|---|---|---|---|---|---|---|---|
| Variable | Mean ± SD | Median | Mean ± SD | Median | Paired t Test P Value | ||
| Room temperature during RHI measurement, °C | 19.3 ± 1.4 | 19 | 19.1 ± 1.0 | 19 | 0.44 | ||
| % of Time indoors at home | 77.0 ± 13.2 | 78.4 | 76.0 ± 12.8 | 75.0 | 0.45 | ||
| % of Time at work | 14.8 ± 11.7 | 16.0 | 16.3 ± 11.9 | 17.4 | 0.29 | ||
| % of Time in transit | 5.0 ± 5.5 | 3.1 | 5.4 ± 5.6 | 3.1 | 0.49 | ||
| % of Hours with ETS exposure reported | 0.1 ± 0.4 | 0.0 | 0.1 ± 0.6 | 0.0 | 0.76 | ||
| % of Hours cooking | 7.2 ± 5.1 | 6.8 | 7.8 ± 4.9 | 8.8 | 0.35 | ||
| Systolic blood pressure,* mm Hg | 112.4 ± 10.8 | 113 | 112.2 ± 11.5 | 112 | 0.88 | ||
| Diastolic blood pressure,* mm Hg | 68.6 ± 7.6 | 68 | 68.4 ± 8.2 | 67 | 0.80 | ||
| RHI | 2.06 ± 0.63 | 1.93 | 2.28 ± 0.72 | 2.32 | 0.03 | ||
| CRP, mg/L | 1.00 ± 0.78 | 0.83 | 0.78 ± 0.74 | 0.48 | 0.06 | ||
| IL-6, pg/mL | 6.11 ± 19.34 | 1.66 | 4.12 ± 8.73 | 1.18 | 0.26 | ||
| Band cells,† % of PMN | 4.62 ± 3.49 | 4.00 | 3.57 ± 2.84 | 3.00 | 0.08 | ||
| Malondialdehyde, μM | 2.64 ± 1.78 | 2.14 | 2.61 ± 3.34 | 1.83 | 0.94 | ||
| 8-isoprostane,‡ pg/mL | 8.78 ± 12.29 | 3.57 | 10.90 ± 14.32 | 4.58 | 0.48 | ||
In our mixed model analysis HEPA filtration was associated with a 9.4% (95% CI, 0.9–18%) increase in RHI and a 32.6% (4.4–60.9%) decrease in CRP (Figure 2). Similar to the crude results in Table 3, when RHI was modeled on the original scale HEPA filtration was associated with an RHI change of 0.22 (0.02–0.41). With the exception of 8-isoprostane, HEPA filtration and air pollution concentration effects on other endpoints were generally in the expected directions but with confidence intervals that included the null. For CRP, IL-6, and MDA, there was some suggestion of an association with total indoor PM2.5 and indoor-generated PM2.5, but no evidence of a relationship with outdoor-generated (infiltrated) PM2.5 or indoor levoglucosan. Band cells were the only outcome for which there was any evidence of an indoor levoglucosan effect, with an 11.3% (5.0–17.7%) decrease in band cells per standardized reduction in levoglucosan. Band cell effect estimates were moderately sensitive to the summand used before log-transforming, with summands of 0.1 and 1 leading to modeled decreases of 13.2% (3.8–22.5%) and 10.1% (5.0–15.3%), respectively, per standardized levoglucosan reduction. As expected due to the crossover study design, model results were not sensitive to adjustment for age, BMI, or sex. Results were also insensitive to adjustment for indoor temperature at the time of sample collection or the percent of time spent indoors at home. Based on continuous indoor nephelometer light-scattering data, there was no clear influence of PM2.5 averaging time on the effect estimates (Figure E1).
Figure 2. Model estimates of exposure reduction effects on health indicators. *Exposure contrasts for continuous exposure metrics are the median within-participant change between non–high-efficiency particulate air (HEPA) filtration and HEPA filtration periods to allow for a comparison of effect sizes between exposure metrics with different distributions. The exposure contrasts are: indoor particulate matter with a diameter less than 2.5 μm (PM2.5) = −6.6 μg/m3; indoor-generated PM2.5 = −4.4 μg/m3; outdoor-generated PM2.5 = −1.3 μg/m3; indoor levoglucosan = −57.6 ng/m3. CRP = C-reactive protein; 8-iso = 8-iso-prostaglandin F2α; MDA = malondialdehyde; RHI = reactive hyperemia index.
[More] [Minimize]We explored modification of the HEPA effect by HEPA order (filter installed first or placebo filtration first), age (> or ≤ 43 yr), sex, overweight status (BMI > 25 or ≤ 25), percent of time spent indoors at home (> 75% or ≤ 75%), and wood stove use (Figures 3 and 4). Although interactions were not statistically significant, with the exception of 8-isoprostane effects were generally more pronounced among males (n = 21) and overweight participants (n = 25) (Figure 3). Inflammatory effects, but not RHI effects, were generally more pronounced among participants 43 years of age or younger (Figure 3). There was also a general pattern across endpoints of more pronounced effects among 23 subjects living in homes with wood-burning stoves (Figure 4). The order of HEPA filtration did not modify the HEPA effect consistently across endpoints.
Figure 3. Model estimates of high-efficiency particulate air (HEPA) filtration effects on reactive hyperemia index (RHI) (upper panel) and blood and urine markers (lower panel) stratified by age, sex, body mass index (BMI), and time spent indoors at home. CRP = C-reactive protein; 8-iso = 8-iso-prostaglandin F2α; MDA = malondialdehyde.
[More] [Minimize]
Figure 4. Model estimates of exposure reduction effects on reactive hyperemia index (RHI) (upper panel) and blood and urine markers (lower panel) stratified by use of a wood-burning stove. *Exposure contrasts for continuous exposure metrics are the median within-participant change between non–high-efficiency particulate air (HEPA) filtration and HEPA filtration periods to allow for a comparison of effect sizes between exposure metrics with different distributions. The exposure contrasts are: indoor particulate matter with a diameter less than 2.5 μm (PM2.5) in wood-burning homes = −7.5 μg/m3; indoor PM2.5 in non–wood-burning homes = −6.2 μg/m3; indoor-generated PM2.5 in wood-burning homes = −6.3 μg/m3; indoor-generated PM2.5 in non–wood-burning homes = −2.1 μg/m3. CRP = C-reactive protein; 8-iso = 8-iso-prostaglandin F2α; MDA = malondialdehyde.
[More] [Minimize]We used HEPA filters in a randomized crossover design to evaluate the relationship between relatively low PM2.5 concentrations and microvascular endothelial function, our primary endpoint, and oxidative stress and systemic inflammation, our secondary endpoints, among healthy adults in an airshed heavily influenced by residential wood combustion. Consistent with previous results from this region (29, 37), the infiltration of outdoor PM2.5 was relatively low, and the majority of indoor PM2.5 was produced by indoor sources. HEPA filters reduced average indoor PM2.5 and levoglucosan concentrations by approximately 60 and 75%, respectively. These reductions were anticipated based on numerous previous studies of HEPA filter effectiveness (30), including recent work in this region by Barn and colleagues (29), who concluded that HEPA filters effectively reduce PM exposures during periods of residential wood combustion.
Our RHI findings are similar to work by Brauner and colleagues (9), who also used a HEPA filter intervention design to investigate the subclinical cardiovascular health effects of traffic-related air pollution exposure among healthy older couples in Copenhagen. Their RHI results were quantitatively similar to ours, despite studying older participants (median age: 67 yr) exposed to an urban air pollution mixture. In their study, HEPA filtration reduced geometric mean indoor PM2.5 concentrations by 7.9 μg/m3 (from 12.6 to 4.7 μg/m3) and was associated with an 8% increase in RHI, very similar to our observed 6.6 μg/m3 reduction in median indoor PM2.5 concentration and 9.4% increase in RHI. Brauner and colleagues (9) also evaluated several elements in the PM2.5 samples and found that only potassium, which is present in relatively high concentrations in biomass smoke (26), was independently associated with RHI. They reported no associations with CRP, IL-6, or 8-isoprostane. Our study provides the first evidence of a link between air pollution and endothelial dysfunction in a woodsmoke-impacted airshed. In addition, although limited evidence suggests that people with diabetes may be more susceptible to endothelial dysfunction related to air pollution (7, 10, 11), our results provide additional evidence of endothelial effects among healthy individuals (8, 12, 39).
The mechanism(s) through which PM may affect endothelial function is not fully understood. Endothelial dysfunction is characterized by a reduction in the bioavailability of nitric oxide (NO) and other vasodilators, which occurs through scavenging by reactive oxygen species (ROS) and/or reduced synthesis (21, 35). ROS can be produced directly by the redox potential of the particles or through the activation of inflammatory cells (40). Inflammation may also play a role in the reduction of NO synthesis. For example, both CRP (23) and IL-6 (24) have been shown to decrease expression of NO synthase in human aortic endothelial cells. In our study, there was some indication of associations between air pollution and inflammatory markers CRP, IL-6, and band cells, although the results were not entirely consistent across all exposure metrics. IL-6 is one of several cytokines that initiates the acute-phase inflammatory response, which involves the release of CRP and other proteins (41, 42). Band cells are immature polymorphonuclear leukocytes, and elevated numbers of band cells indicate stimulation of the bone marrow (4, 43). For both CRP and IL-6, there was some evidence of associations with total indoor PM2.5 and indoor-generated PM2.5, but less so for outdoor-generated PM2.5 or levoglucosan. The lack of effects for outdoor-generated PM2.5 and levoglucosan is possibly due to the low indoor concentrations of these constituents (due to low infiltration in these tightly sealed homes), which limited the exposure gradients introduced by HEPA filtration (Table 2).
There are at least three possible explanations for the observation that HEPA filtration, but not PM2.5, was associated with changes in RHI and CRP. First, the lack of measurement error in the binary intervention variable may have allowed us to observe associations that were masked by error in the continuous pollution concentration variables. Second, the observed HEPA effects may be due to specific components of the PM mixture, such as ultrafine (< 100 nm-diameter) particles. HEPA filters are believed to effectively remove particles in the ultrafine range (10–100 nm) (44), and ultrafine particles may play an important role in the inflammatory and endothelial effects of PM (1, 45, 46). Finally, the averaging period for the PM2.5 measurements (7 d) may not have matched the relevant exposure-response period for some of these outcomes (41), although continuous indoor measurements did not reveal a clear influence of averaging times on the PM2.5 associations (Figure E1). Repeated measurements of outcomes during the 7-d monitoring periods, which would have allowed us to evaluate the time course of the biological responses, were not feasible in this study.
Although the literature is not totally consistent (47, 48), our results add to a growing body of evidence linking short-term PM exposure with a systemic inflammatory response (1). Traffic-related air pollution has been studied more extensively in relation to inflammation (3, 6), but there is also some evidence linking high concentrations of biomass smoke with a systemic inflammatory response. In an experimental crossover study, Barregard and colleagues (27) administered clean air and woodsmoke at PM2.5 mass concentrations of 240 to 280 μg/m3 to healthy adult volunteers. They reported significant associations between woodsmoke and serum amyloid A, an acute-phase inflammatory protein, 8-isoprostane, and plasma factor VIII. Swiston and colleagues (4) studied 52 seasonal forest-fire fighters and reported significant increases in circulating white blood cells, band cells, IL-6, and monocyte chemotactic protein-1 levels after fire fighting. PM levels, estimated from measurements of carbon monoxide, were estimated in the 1,000 to 2,000 μg/m3 range.
In our study, band cells were the only endpoint for which there was persuasive evidence of an association with levoglucosan, a marker of woodsmoke PM. Similar to our results and those of Swiston and colleagues (4), Tan and colleagues (43) reported an association between air pollution from biomass combustion and increased circulating band cells. They studied 30 men in Singapore exposed to biomass smoke during the 1997 Southeast Asian smoke haze. PM10 concentrations, which averaged 125 μg/m3 during the event, were significantly associated with band cells at 0- at 1-day lags. The associations with band cells in these three studies suggest that this biomarker may be particularly sensitive to biomass smoke exposure.
There was limited evidence of more pronounced effects among participants residing in wood-burning homes, males, and participants with BMI greater than 25 kg/m2. For the systemic inflammation markers, there was also some indication of more pronounced effects among younger participants. The findings in wood-burning homes were unexpected given the lack of associations with the woodsmoke tracer levoglucosan for all endpoints but band cells. This discrepancy may be explained by the presence of some other (non-woodsmoke) indoor PM2.5 source in wood-burning homes, which is supported by the observation that during HEPA filtration wood-burning homes experienced much larger reductions in indoor-generated PM2.5, but similar reductions in indoor levoglucosan, compared with homes where wood was not burned. Alternatively, the participants residing in these homes may have been more sensitive to the cardiovascular impacts of PM exposure.
Despite some inconsistency, previous research has suggested that older individuals may be more susceptible to the cardiovascular effects of air pollution (1). For example, in contrast to the results of their HEPA intervention study (9), Brauner and colleagues found that RHI and biomarkers of inflammation and oxidative stress were not associated with traffic-generated PM in a controlled exposure study among 29 healthy young (median age, 25 yr) adults (47). Sex has also not been definitively identified as an effect modifier. Nevertheless, our results are consistent with several previous studies that have reported short-term air pollution effects on endothelial function and inflammation among young male participants (6, 8, 12, 39, 43), and one study suggesting that the inflammatory effects of chronic PM exposure are more pronounced in men (5). The existing evidence for BMI/obesity as an effect modifier is somewhat stronger and is consistent with our results. Schneider and colleagues (10) reported greater effects of short-term PM2.5 on flow-mediated dilatation among persons with type 2 diabetes and among those with BMI greater than 30 kg/m2. Similarly, Dubowsky and colleagues (3) reported stronger associations between 5-day PM2.5 concentrations and both CRP and IL-6 among older adults with BMI greater than 30 kg/m2, whereas Chen and Schwartz (49) found that that metabolic syndrome modified the association between annual PM10 concentrations and white blood cell counts.
We did not find persuasive evidence that any exposure metrics were associated with MDA or 8-isoprostane, two products of lipid peroxidation that have been assessed in previous air pollution studies (9, 27, 50–52). The experimental work of Barregard and colleagues (27) provides the only published evidence of an association between biomass smoke and systemic oxidative stress, whereas some experimental and observational studies of the urban pollution mixture have reported associations with oxidative stress markers among young adults and children (50, 53–56). The lack of observed effects in our study may have been due to other factors, such as diet (57). In addition, the 8-isoprostane results may have been influenced by the lack of normalization to urinary creatinine and the use of ELISA, which is a less specific and less quantitative assay than gas chromatography mass spectrometry (51, 58).
Some additional limitations of this study should be noted. First, our measure of microvascular endothelial function, RHI, has not been widely used for research or clinical purposes. Nevertheless, this measure is predictive of adverse cardiovascular events (59). Although RHI does not directly distinguish between endothelium-dependent and endothelium-independent effects, inhibition of endothelial nitric oxide synthase attenuates the RHI response, suggesting that this measure is indicative of endothelial function (60). Moreover, Bonetti and colleagues (61) have reported a relationship between RHI and coronary artery endothelial function, whereas Kuvin and colleagues (62) demonstrated a correlation between RHI and endothelium-dependent brachial artery flow-mediated dilatation. In addition, RHI is associated with traditional cardiovascular risk factors, including diabetes mellitus, BMI, cholesterol, and smoking (63). Administration of sublingual nitroglycerin, which would have allowed us to assess endothelium-independent effects on the RHI response (61), was not feasible in this residence-based study.
An additional limitation was that we were not able to quantify air pollution exposure outside the home, where on average our study participants spent 25% of their time. Although time spent outside the home reduced the effectiveness of the in-home air cleaner intervention, pollution exposures outside the home are unlikely to explain the observed associations because of the crossover study design and the similarity in time-location patterns between HEPA and non-HEPA periods.
Carryover of effects between “treatments” is a concern in crossover study designs (64). However, in this study the 7-day exposure periods were long relative to the expected response time of the biological measurements (41). Therefore, our exposure periods were probably sufficient to wash out any effects from the previous exposure scenario. Moreover, carryover effects would likely have caused an underestimation of the effects (i.e., a bias toward no effect) and are therefore unlikely to be responsible for the observed associations.
In conclusion, portable HEPA filters reduced average indoor PM2.5 concentrations by 60% and were associated with improved endothelial function and decreased concentrations of inflammatory biomarkers, but not markers of oxidative stress, among healthy adults residing in a woodsmoke-dominated airshed. There was limited evidence that effects were more pronounced among participants residing in homes that burned wood, males, younger participants, and overweight participants. Our results support the hypothesis that systemic inflammation and impaired endothelial function, both predictors of cardiovascular morbidity, can be favorably influenced by a reduction of indoor particle concentrations.
The authors thank the study participants and our field technicians, Thurza Aspinall and Amelia Mattson. The British Columbia Ministry of Environment and the Bulkley Valley District Hospital generously provided laboratory access. They also thank Dr. Winnie Chu and the University of British Columbia School of Environmental Health Laboratory staff for their assistance with sample preparation and analysis.
| 1. | Brook RD, Rajagopalan S, Pope CA, Brook JR, Bhatnagar A, Diez-Roux AV, Holguin F, Hong YL, Luepker RV, Mittleman MA, et al. Particulate matter air pollution and cardiovascular disease: an update to the scientific statement from the American Heart Association. Circulation 2010;121:2331–2378. |
| 2. | Bai N, Khazaei M, van Eeden SF, Laher I. The pharmacology of particulate matter air pollution-induced cardiovascular dysfunction. Pharmacol Ther 2007;113:16–29. |
| 3. | Dubowsky SD, Suh H, Schwartz J, Coull BA, Gold DR. Diabetes, obesity, and hypertension may enhance associations between air pollution and markers of systemic inflammation. Environ Health Perspect 2006;114:992–998. |
| 4. | Swiston JR, Davidson W, Attridge S, Li GT, Brauer M, van Eeden SF. Wood smoke exposure induces a pulmonary and systemic inflammatory response in firefighters. Eur Respir J 2008;32:129–138. |
| 5. | Hoffmann B, Moebus S, Dragano N, Stang A, Mohlenkamp S, Schmermund A, Memmesheimer M, Brocker-Preuss M, Mann K, Erbel R, et al. Chronic residential exposure to particulate matter air pollution and systemic inflammatory markers. Environ Health Perspect 2009;117:1302–1308. |
| 6. | Riediker M, Cascio WE, Griggs TR, Herbst MC, Bromberg PA, Neas L, Williams RW, Devlin RB. Particulate matter exposure in cars is associated with cardiovascular effects in healthy young men. Am J Respir Crit Care Med 2004;169:934–940. |
| 7. | O'Neill MS, Veves A, Zanobetti A, Sarnat JA, Gold DR, Economides PA, Horton ES, Schwartz J. Diabetes enhances vulnerability to particulate air pollution-associated impairment in vascular reactivity and endothelial function. Circulation 2005;111:2913–2920. |
| 8. | Briet M, Collin C, Laurent S, Tan A, Azizi M, Agharazii M, Jeunemaitre X, Alhenc-Gelas F, Boutouyrie P. Endothelial function and chronic exposure to air pollution in normal male subjects. Hypertension 2007;50:970–976. |
| 9. | Brauner EV, Forchhammer L, Moller P, Barregard L, Gunnarsen L, Afshari A, Wahlin P, Glasius M, Dragsted LO, Basu S, et al. Indoor particles affect vascular function in the aged: an air filtration-based intervention study. Am J Respir Crit Care Med 2008;177:419–425. |
| 10. | Schneider A, Neas L, Herbst MC, Case M, Williams RW, Cascio W, Hinderliter A, Holguin F, Buse JB, Dungan K, et al. Endothelial dysfunction: associations with exposure to ambient fine particles in diabetic individuals. Environ Health Perspect 2008;116:1666–1674. |
| 11. | Liu L, Ruddy TD, Dalipaj M, Szyszkowicz M, You HY, Poon R, Wheeler A, Dales R. Influence of personal exposure to particulate air pollution on cardiovascular physiology and biomarkers of inflammation and oxidative stress in subjects with diabetes. J Occup Environ Med 2007;49:258–265. |
| 12. | Tornqvist H, Mills NL, Gonzalez M, Miller MR, Robinson SD, Megson IL, MacNee W, Donaldson K, Soderberg S, Newby DE, et al. Persistent endothelial dysfunction in humans after diesel exhaust inhalation. Am J Respir Crit Care Med 2007;176:395–400. |
| 13. | Allen RW, Criqui MH, Roux AVD, Allison M, Shea S, Detrano R, Sheppard L, Wong ND, Stukovsky KH, Kaufman JD. Fine particulate matter air pollution, proximity to traffic, and aortic atherosclerosis. Epidemiology 2009;20:254–264. |
| 14. | Kunzli N, Jerrett M, Mack WJ, Beckerman B, LaBree L, Gilliland F, Thomas D, Peters J, Hodis HN. Ambient air pollution and atherosclerosis in Los Angeles. Environ Health Perspect 2005;113:201–206. |
| 15. | Kunzli N, Jerrett M, Garcia-Esteban R, Basagana X, Beckermann B, Gilliland F, Medina M, Peters J, Hodis HN, Mack WJ. Ambient air pollution and the progression of atherosclerosis in adults. PLoS ONE 2010;5:e9096. |
| 16. | Diez Roux AV, Auchincloss A, Green TL, Raghunathan T, Barr RG, Kaufman J, Astor B, Keeler J. Long-term exposure to ambient particulate matter and prevalence of subclinical atherosclerosis in the Multienthic Study of Atherosclerosis. Am J Epidemiol 2008;167:667–675. |
| 17. | Hoffmann B, Moebus S, Mohlenkamp S, Stang A, Lehmann N, Dragano N, Schmermund A, Memmesheimer M, Mann K, Erbel R, et al. Residential exposure to traffic is associated with coronary atherosclerosis. Circulation 2007;116:489–496. |
| 18. | Hansson GK, Nilsson J. Introduction: atherosclerosis as inflammation: a controversial concept becomes accepted. J Intern Med 2008;263:462–463. |
| 19. | Deanfield JE, Halcox JP, Rabelink TJ. Endothelial function and dysfunction: testing and clinical relevance. Circulation 2007;115:1285–1295. |
| 20. | Widlansky ME, Gokce N, Keaney JF, Vita JA. The clinical implications of endothelial dysfunction. J Am Coll Cardiol 2003;42:1149–1160. |
| 21. | Fichtlscherer S, Breuer S, Schachinger V, Dimmeler S, Zeiher AM. C-reactive protein levels determine systemic nitric oxide bioavailability in patients with coronary artery disease. Eur Heart J 2004;25:1412–1418. |
| 22. | Kaptoge S, Di Angelantonio E, Lowe G, Pepys MB, Thompson SG, Collins R, Danesh J. C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta-analysis. Lancet 2010;375:132–140. |
| 23. | Venugopal SK, Devaraj S, Yuhanna I, Shaul P, Jialal I. Demonstration that C-reactive protein decreases eNOS expression and bioactivity in human aortic endothelial cells. Circulation 2002;106:1439–1441. |
| 24. | Saura M, Zaragoza C, Bao C, Herranz B, Rodriguez-Puyol M, Lowenstein CJ. Stat3 mediates interelukin-6 inhibition of human endothelial nitric-oxide synthase expression. J Biol Chem 2006;281:30057–30062. |
| 25. | HEI Panel on the Health Effects of Traffic-Related Air Pollution. Traffic-related air pollution: a critical review of the literature on emissions, exposure, and health effects. HEI Special Report 17. Boston, MA: Health Effects Institute; 2010. |
| 26. | Naeher LP, Brauer M, Lipsett M, Zelikoff JT, Simpson CD, Koenig JQ, Smith KR. Woodsmoke health effects: a review. Inhal Toxicol 2007;19:67–106. |
| 27. | Barregard L, Sallsten G, Gustafson P, Andersson L, Johansson L, Basu S, Stigendal L. Experimental exposure to wood-smoke particles in healthy humans: effects on markers of inflammation, coagulation, and lipid peroxidation. Inhal Toxicol 2006;18:845–853. |
| 28. | Richter DD, Jenkins DH, Karakash JT, Knight J, McCreery LR, Nemestothy KP. Wood energy in America. Science 2009;323:1432–1433. |
| 29. | Barn P, Larson T, Noullett M, Kennedy S, Copes R, Brauer M. Infiltration of forest fire and residential wood smoke: an evaluation of air cleaner effectiveness. J Expo Sci Environ Epidemiol 2008;18:503–511. |
| 30. | Sublett JL, Seltzer J, Burkhead R, Williams PB, Wedner HJ, Phipatanakul W, Amer Acad A. Air filters and air cleaners: rostrum by the American Academy of Allergy, Asthma & Immunology Indoor Allergen Committee. J Allergy Clin Immunol 2010;125:32–38. |
| 31. | Yamada Y, Miyamoto K, Mori T, Koizumi A. Penetration of sub-micron aerosols through high-efficiency air filters. Health Phys 1984;46:543–547. |
| 32. | Fisk WJ, Faulkner D, Palonen J, Seppanen O. Performance and costs of particle air filtration technologies. Indoor Air 2002;12:223–234. |
| 33. | Klepeis NE, Nelson WC, Ott WR, Robinson JP, Tsang AM, Switzer P, Behar JV, Hern SC, Engelmann WH. The National Human Activity Pattern Survey (NHAPS): a resource for assessing exposure to environmental pollutants. J Expo Anal Environ Epidemiol 2001;11:231–252. |
| 34. | Offermann FJ, Sextro RG, Fisk WJ, Grimsrud DT, Nazaroff WW, Nero AV, Revzan KL, Yater J. Control of respirable particles in indoor air with portable air cleaners. Atmos Environ 1985;19:1761–1771. |
| 35. | Bonetti PO, Lerman LO, Lerman A. Endothelial dysfunction: a marker of atherosclerotic risk. Arterioscler Thromb Vasc Biol 2003;23:168–175. |
| 36. | Allen R, Aspinall T, Carlsten C, Karlen B, Leckie S, Mattson A, van Eeden S, Vedal S, Brauer M. The impact of portable air filters on indoor air pollution and microvascular function in a woodsmoke-impacted community [abstract]. Epidemiology 2009;20:S51. |
| 37. | Allen RW, Leckie S, Millar G, Brauer M. The impact of wood stove technology upgrades on indoor residential air quality. Atmos Environ 2009;43:5908–5915. |
| 38. | Allen R, Larson T, Sheppard L, Wallace L, Liu LJS. Use of real-time light scattering data to estimate the contribution of infiltrated and indoor-generated particles to indoor air. Environ Sci Technol 2003;37:3484–3492. |
| 39. | Rundell KW, Hoffman JR, Caviston R, Bulbulian R, Hollenbach AM. Inhalation of ultrafine and fine particulate matter disrupts systemic vascular function. Inhal Toxicol 2007;19:133–140. |
| 40. | Kelly FJ. Oxidative stress: its role in air pollution and adverse health effects. Occup Environ Med 2003;60:612–616. |
| 41. | Gabay C, Kushner I. Mechanisms of disease: acute-phase proteins and other systemic responses to inflammation. N Engl J Med 1999;340:448–454. |
| 42. | van Eeden SF, Yeung A, Quinlam K, Hogg JC. Systemic response to ambient particulate matter: relevance to chronic obstructive pulmonary disease. Proc Am Thorac Soc 2005;2:61–67. |
| 43. | Tan WC, Qiu DW, Liam BL, Ng TP, Lee SH, van Eeden SF, D'Yachkova Y, Hogg JC. The human bone marrow response to acute air pollution caused by forest fires. Am J Respir Crit Care Med 2000;161:1213–1217. |
| 44. | Hinds WC. Aerosol technology: properties, behavior, and measurement of airborne particles. New York, NY: John Wiley & Sons, Inc., 1999. |
| 45. | Knol AB, de Hartog JJ, Boogaard H, Slottje P, van der Sluijs JP, Lebret E, Cassee FR, Wardekker A, Ayres JG, Borm PJ, et al. Expert elicitation on ultrafine particles: likelihood of health effects and causal pathways. Part Fibre Toxicol 2009;6:article no. 19. |
| 46. | Delfino RJ, Sioutas C, Malik S. Potential role of ultrafine particles in associations between airborne particle mass and cardiovascular health. Environ Health Perspect 2005;113:934–946. |
| 47. | Brauner EV, Moller P, Barregard L, Dragsted LO, Glasius M, Wahlin P, Vinzents P, Raaschou-Nielsen O, Loft S. Exposure to ambient concentrations of particulate air pollution does not influence vascular function or inflammatory pathways in young healthy individuals. Part Fibre Toxicol 2008;5:article no. 13. |
| 48. | Rudez G, Janssen NAH, Kilinc E, Leebeek FWG, Gerlofs-Nijland ME, Spronk HMH, ten Cate H, Cassee FR, de Maat MPM. Effects of ambient air pollution on hemostasis and inflammation. Environ Health Perspect 2009;117:995–1001. |
| 49. | Chen JC, Schwartz J. Metabolic syndrome and inflammatory responses to long-term particulate air pollutants. Environ Health Perspect 2008;116:612–617. |
| 50. | Bae S, Pan XC, Kim SY, Park K, Kim YH, Kim H, Hong YC. Exposures to particulate matter and polycyclic aromatic hydrocarbons and oxidative stress in schoolchildren. Environ Health Perspect 2010;118:579–583. |
| 51. | Basu S. Isoprostanes: novel bioactive products of lipid peroxidation. Free Radic Res 2004;38:105–122. |
| 52. | Moller P, Loft S. Oxidative damage to DNA and lipids as biomarkers of exposure to air pollution. Environ Health Perspect 2010;118:1126–1136. |
| 53. | Brauner EV, Forchhammer L, Moller P, Simonsen J, Glasius M, Wahlin P, Raaschou-Nielsen O, Loft S. Exposure to ultrafine particles from ambient air and oxidative stress-induced DNA damage. Environ Health Perspect 2007;115:1177–1182. |
| 54. | Vinzents PS, Moller P, Sorensen M, Knudsen LE, Hertel O, Jensen FP, Schibye B, Loft S. Personal exposure to ultrafine particles and oxidative DNA damage. Environ Health Perspect 2005;113:1485–1490. |
| 55. | Chuang KJ, Chan CC, Su TC, Lee CT, Tang CS. The effect of urban air pollution on inflammation, oxidative stress, coagulation, and autonomic dysfunction in young adults. Am J Respir Crit Care Med 2007;176:370–376. |
| 56. | Sorensen M, Daneshvar B, Hansen M, Dragsted LO, Hertel O, Knudsen L, Loft S. Personal PM2.5 exposure and markers of oxidative stress in blood. Environ Health Perspect 2003;111:161–165. |
| 57. | Brown ED, Morris VC, Rhodes DG, Sinha R, Levander OA. Urinary malondialdehyde-equivalents during ingestion of meat cooked at high or low temperatures. Lipids 1995;30:1053–1056. |
| 58. | Mayne ST. Antioxidant nutrients and chronic disease: use of biomarkers of exposure and oxidative stress status in epidemiologic research. J Nutr 2003;133:933S–940S. |
| 59. | Rubinshtein R, Kuvin JT, Soffler M, Lennon RJ, Lavi S, Nelson RE, Pumper GM, Lerman LO, Lerman A. Assessment of endothelial function by non-invasive peripheral arterial tonometry predicts late cardiovascular adverse events. Eur Heart J 2010;31:1142–1148. |
| 60. | Nohria A, Gerhard-Herman M, Creager MA, Hurley S, Mitra D, Ganz P. Role of nitric oxide in the regulation of digital pulse volume amplitude in humans. J Appl Physiol 2006;101:545–548. |
| 61. | Bonetti PO, Pumper GM, Higano ST, Holmes DR, Kuvin JT, Lerman A. Noninvasive identification of patients with early coronary atherosclerosis by assessment of digital reactive hyperemia. J Am Coll Cardiol 2004;44:2137–2141. |
| 62. | Kuvin JT, Patel AR, Sliney KA, Pandian NG, Sheffy J, Schnall RP, Karas RH, Udelson JE. Assessment of peripheral vascular endothelial function with finger arterial pulse wave amplitude. Am Heart J 2003;146:168–174. |
| 63. | Hamburg NM, Keyes MJ, Larson MG, Vasan RS, Schnabel R, Pryde MM, Mitchell GF, Sheffy J, Vita JA, Benjamin EJ. Cross-sectional relations of digital vascular function to cardiovascular risk factors in the Framingham Heart Study. Circulation 2008;117:2467–2474. |
| 64. | Grizzle JE. Two-period change-over design and its use in clinical trials. Biometrics 1965;21:467. |
