[Skip to Content]
[Skip to Content Landing]
Figure 1.
CONSORT Diagram
CONSORT Diagram

BP indicates blood pressure.

Figure 2.
Changes in Brachial Blood Pressure (BP) Levels
Changes in Brachial Blood Pressure (BP) Levels

A, BP levels represent overall changes during all days of any filtration type compared with the 3-day sham filtration. B, BP levels represent individual daily and 3-day mean changes during each separate filtration type compared with the corresponding sham filtration day. C, BP levels represent overall changes during all days of any filtration type compared with the 3-day sham filtration among all, 19 obese (body mass index [calculated as weight in kilograms divided by height in meters squared]≥30), and 21 nonobese participants. Error bars indicate 95% CIs; Δ, difference. DBP indicates diastolic BP; HE, high-efficiency; LE, low-efficiency; and SBP, systolic BP.

aP < .05, test of fixed effects from the adjusted mixed model.

Figure 3.
Change in Secondary Outcomes (Aortic Hemodynamics) by Filtration Type During Each 3-Day Intervention
Change in Secondary Outcomes (Aortic Hemodynamics) by Filtration Type During Each 3-Day Intervention

Results represent individual daily and 3-day mean changes during each separate filtration intervention type compared with the sham filtration. Augmentation and pulse pressures are measured for the aorta. Error bars indicate 95% CI; Δ, difference. AIx@75 indicates augmentation index standardized to a heart rate of 75 beats/min; HE, high-efficiency; LE, low-efficiency; and PWV, pulse-wave velocity.

aP < .05, test of fixed effects from the adjusted mixed model.

Table 1.  
Participant Characteristics
Participant Characteristics
Table 2.  
Cardiovascular Outcome and Exposure Assessment Results Under 3 Scenarios
Cardiovascular Outcome and Exposure Assessment Results Under 3 Scenarios
1.
Cohen  AJ, Brauer  M, Burnett  R,  et al.  Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015.  Lancet. 2017;389(10082):1907-1918. doi:10.1016/S0140-6736(17)30505-6PubMedGoogle ScholarCrossref
2.
Brook  RD, Rajagopalan  S, Pope  CA  III,  et al; American Heart Association Council on Epidemiology and Prevention, Council on the Kidney in Cardiovascular Disease, and Council on Nutrition, Physical Activity and Metabolism.  Particulate matter air pollution and cardiovascular disease: an update to the scientific statement from the American Heart Association.  Circulation. 2010;121(21):2331-2378. doi:10.1161/CIR.0b013e3181dbece1PubMedGoogle ScholarCrossref
3.
Di  Q, Wang  Y, Zanobetti  A,  et al.  Air pollution and mortality in the Medicare population.  N Engl J Med. 2017;376(26):2513-2522. doi:10.1056/NEJMoa1702747PubMedGoogle ScholarCrossref
4.
Brook  RD, Newby  DE, Rajagopalan  S.  The global threat of outdoor ambient air pollution to cardiovascular health: time for intervention.  JAMA Cardiol. 2017;2(4):353-354. doi:10.1001/jamacardio.2017.0032PubMedGoogle ScholarCrossref
5.
Giles  LV, Barn  P, Künzli  N,  et al.  From good intentions to proven interventions: effectiveness of actions to reduce the health impacts of air pollution.  Environ Health Perspect. 2011;119(1):29-36. doi:10.1289/ehp.1002246PubMedGoogle ScholarCrossref
6.
Matz  CJ, Stieb  DM, Davis  K,  et al.  Effects of age, season, gender and urban-rural status on time-activity: Canadian Human Activity Pattern Survey 2 (CHAPS 2).  Int J Environ Res Public Health. 2014;11(2):2108-2124. doi:10.3390/ijerph110202108PubMedGoogle ScholarCrossref
7.
Klepeis  NE, Nelson  WC, Ott  WR,  et al.  The National Human Activity Pattern Survey (NHAPS): a resource for assessing exposure to environmental pollutants.  J Expo Anal Environ Epidemiol. 2001;11(3):231-252. doi:10.1038/sj.jea.7500165PubMedGoogle ScholarCrossref
8.
Bräuner  EV, Forchhammer  L, Møller  P,  et al.  Indoor particles affect vascular function in the aged: an air filtration–based intervention study.  Am J Respir Crit Care Med. 2008;177(4):419-425. doi:10.1164/rccm.200704-632OCPubMedGoogle ScholarCrossref
9.
Allen  RW, Carlsten  C, Karlen  B,  et al.  An air filter intervention study of endothelial function among healthy adults in a woodsmoke-impacted community.  Am J Respir Crit Care Med. 2011;183(9):1222-1230. doi:10.1164/rccm.201010-1572OCGoogle ScholarCrossref
10.
Morishita  M, Thompson  KC, Brook  RD.  Understanding air pollution and cardiovascular diseases: is it preventable?  Curr Cardiovasc Risk Rep. 2015;9(6):1-9. doi:10.1007/s12170-015-0458-1PubMedGoogle ScholarCrossref
11.
Chen  R, Zhao  A, Chen  H,  et al.  Cardiopulmonary benefits of reducing indoor particles of outdoor origin: a randomized, double-blind crossover trial of air purifiers.  J Am Coll Cardiol. 2015;65(21):2279-2287. doi:10.1016/j.jacc.2015.03.553PubMedGoogle ScholarCrossref
12.
Karottki  DG, Spilak  M, Frederiksen  M,  et al.  An indoor air filtration study in homes of elderly: cardiovascular and respiratory effects of exposure to particulate matter.  Environ Health. 2013;12(1):116. doi:10.1186/1476-069X-12-116PubMedGoogle ScholarCrossref
13.
Kajbafzadeh  M, Brauer  M, Karlen  B, Carlsten  C, van Eeden  S, Allen  RW.  The impacts of traffic-related and woodsmoke particulate matter on measures of cardiovascular health: a HEPA filter intervention study.  Occup Environ Med. 2015;72(6):394-400. doi:10.1136/oemed-2014-102696PubMedGoogle ScholarCrossref
14.
Padró-Martínez  LT, Owusu  E, Reisner  E,  et al.  A randomized cross-over air filtration intervention trial for reducing cardiovascular health risks in residents of public housing near a highway.  Int J Environ Res Public Health. 2015;12(7):7814-7838. doi:10.3390/ijerph120707814PubMedGoogle ScholarCrossref
15.
Sacks  JD, Stanek  LW, Luben  TJ,  et al.  Particulate matter-induced health effects: who is susceptible?  Environ Health Perspect. 2011;119(4):446-454. doi:10.1289/ehp.1002255PubMedGoogle ScholarCrossref
16.
Pope  CA  III, Hansen  ML, Long  RW,  et al.  Ambient particulate air pollution, heart rate variability, and blood markers of inflammation in a panel of elderly subjects.  Environ Health Perspect. 2004;112(3):339-345. doi:10.1289/ehp.6588PubMedGoogle ScholarCrossref
17.
Wang  C, Tu  Y, Yu  Z, Lu  R.  PM2.5 and cardiovascular diseases in the elderly: an overview.  Int J Environ Res Public Health. 2015;12(7):8187-8197. doi:10.3390/ijerph120708187Google ScholarCrossref
18.
Peng  RD, Bell  ML, Geyh  AS,  et al.  Emergency admissions for cardiovascular and respiratory diseases and the chemical composition of fine particle air pollution.  Environ Health Perspect. 2009;117(6):957-963. doi:10.1289/ehp.0800185PubMedGoogle ScholarCrossref
19.
US Department of Health and Human Services.  A Profile of Older Americans: 2016. Administration on Aging, US Dept of Health and Human Services; 2016. https://www.acl.gov/sites/default/files/Aging%20and%20Disability%20in%20America/2016-Profile.pdf. Accessed July 5, 2018.
20.
Lewtas  J.  Air pollution combustion emissions: characterization of causative agents and mechanisms associated with cancer, reproductive, and cardiovascular effects.  Mutat Res. 2007;636(1-3):95-133. doi:10.1016/j.mrrev.2007.08.003PubMedGoogle ScholarCrossref
21.
Rahimi  K, Emdin  CA, MacMahon  S.  The epidemiology of blood pressure and its worldwide management.  Circ Res. 2015;116(6):925-936. doi:10.1161/CIRCRESAHA.116.304723PubMedGoogle ScholarCrossref
22.
Brook  RD, Bard  RL, Burnett  RT,  et al.  Differences in blood pressure and vascular responses associated with ambient fine particulate matter exposures measured at the personal versus community level.  Occup Environ Med. 2011;68(3):224-230. doi:10.1136/oem.2009.053991PubMedGoogle ScholarCrossref
23.
Dvonch  JT, Kannan  S, Schulz  AJ,  et al.  Acute effects of ambient particulate matter on blood pressure: differential effects across urban communities.  Hypertension. 2009;53(5):853-859. doi:10.1161/HYPERTENSIONAHA.108.123877PubMedGoogle ScholarCrossref
24.
Giorgini  P, Rubenfire  M, Das  R,  et al.  Particulate matter air pollution and ambient temperature: opposing effects on blood pressure in high-risk cardiac patients.  J Hypertens. 2015;33(10):2032-2038. doi:10.1097/HJH.0000000000000663PubMedGoogle ScholarCrossref
25.
Brook  RD, Sun  Z, Brook  JR,  et al.  Extreme air pollution conditions adversely affect blood pressure and insulin resistance: the Air Pollution and Cardiometabolic Disease Study.  Hypertension. 2016;67(1):77-85. doi:10.1161/HYPERTENSIONAHA.115.06237PubMedGoogle ScholarCrossref
26.
Cai  Y, Zhang  B, Ke  W,  et al.  Associations of short-term and long-term exposure to ambient air pollutants with hypertension: a systematic review and meta-analysis.  Hypertension. 2016;68(1):62-70. doi:10.1161/HYPERTENSIONAHA.116.07218PubMedGoogle ScholarCrossref
27.
Michigan Department of Transportation. 2013 Average daily traffic (ADT) maps. https://gis-mdot.opendata.arcgis.com/datasets/2013-traffic-volumes. 2013. Accessed July 5, 2018.
28.
Pickering  TG, Hall  JE, Appel  LJ,  et al.  Recommendations for blood pressure measurement in humans and experimental animals, part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research.  Circulation. 2005;111(5):697-716. doi:10.1161/01.CIR.0000154900.76284.F6PubMedGoogle ScholarCrossref
29.
Vlachopoulos  C, Aznaouridis  K, Stefanadis  C.  Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis.  J Am Coll Cardiol. 2010;55(13):1318-1327. doi:10.1016/j.jacc.2009.10.061PubMedGoogle ScholarCrossref
30.
Chuang  HC, Ho  KF, Lin  LY,  et al.  Long-term indoor air conditioner filtration and cardiovascular health: a randomized crossover intervention study.  Environ Int. 2017;106:91-96. doi:10.1016/j.envint.2017.06.008PubMedGoogle ScholarCrossref
31.
Li  H, Cai  J, Chen  R,  et al.  Particulate matter exposure and stress hormone levels: a randomized, double-blind, crossover trial of air purification.  Circulation. 2017;136(7):618-627. doi:10.1161/CIRCULATIONAHA.116.026796PubMedGoogle ScholarCrossref
32.
Weichenthal  S, Hoppin  JA, Reeves  F.  Obesity and the cardiovascular health effects of fine particulate air pollution.  Obesity (Silver Spring). 2014;22(7):1580-1589. doi:10.1002/oby.20748PubMedGoogle ScholarCrossref
33.
Franklin  SS, Larson  MG, Khan  SA,  et al.  Does the relation of blood pressure to coronary heart disease risk change with aging? the Framingham Heart Study.  Circulation. 2001;103(9):1245-1249. doi:10.1161/01.CIR.103.9.1245Google ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    October 2018

    Effect of Portable Air Filtration Systems on Personal Exposure to Fine Particulate Matter and Blood Pressure Among Residents in a Low-Income Senior Facility: A Randomized Clinical Trial

    Author Affiliations
    • 1Department of Family Medicine, College of Human Medicine, Michigan State University, East Lansing
    • 2School of Public Health, University of Michigan, Ann Arbor
    • 3Community Health Nursing, Ann Arbor, Michigan
    • 4Division of Cardiovascular Medicine, University of Michigan, Ann Arbor
    JAMA Intern Med. 2018;178(10):1350-1357. doi:10.1001/jamainternmed.2018.3308
    Key Points

    Question  Can portable air filtration systems reduce personal exposures to fine particulate matter air pollution and blood pressure levels among elderly adults living in a typical US urban location?

    Findings  In this randomized, double-blind crossover intervention study, short-term use of portable air filtration systems reduced personal exposures to fine particulate matter and systolic blood pressure in senior citizens living in a low-income residence.

    Meaning  The use of portable air filtration systems is potentially cardioprotective against exposures to fine particulate matter and warrants further research.

    Abstract

    Importance  Fine particulate matter (smaller than 2.5 μm) (PM2.5) air pollution is a major global risk factor for cardiovascular (CV) morbidity and mortality. Few studies have tested the benefits of portable air filtration systems in urban settings in the United States.

    Objective  To investigate the effectiveness of air filtration at reducing personal exposures to PM2.5 and mitigating related CV health effects among older adults in a typical US urban location.

    Design, Setting, and Participants  This randomized, double-blind crossover intervention study was conducted from October 21, 2014, through November 4, 2016, in a low-income senior residential building in Detroit, Michigan. Forty nonsmoking older adults were enrolled, with daily CV health outcome and PM2.5 exposure measurements.

    Interventions  Participants were exposed to the following three 3-day scenarios separated by 1-week washout periods: unfiltered air (sham filtration), low-efficiency (LE) high-efficiency particulate arrestance (HEPA)–type filtered air, and high-efficiency (HE) true-HEPA filtered air using filtration systems in their bedroom and living room.

    Main Outcomes and Measures  The primary outcome was brachial blood pressure (BP). Secondary outcomes included aortic hemodynamics, pulse-wave velocity, and heart rate variability. Exposures to PM2.5 were measured in the participants’ residences and by personal monitoring.

    Results  The 40 participants had a mean (SD) age of 67 (8) years (62% men). Personal PM2.5 exposures were significantly reduced by air filtration from a mean (SD) of 15.5 (10.9) μg/m3 with sham filtration to 10.9 (7.4) μg/m3 with LE fitration and 7.4 (3.3) μg/m3 with HE filtration. Compared with sham filtration, any filtration for 3 days decreased brachial systolic and diastolic BP by 3.2 mm Hg (95% CI, −6.1 to −0.2 mm Hg) and 1.5 mm Hg (95% CI, −3.3 to 0.2 mm Hg), respectively. A continuous decrease occurred in systolic and diastolic BP during the 3-day period of LE filtration, with a mean of 3.4 mm Hg (95% CI, −6.8 to −0.1 mm Hg) and 2.2 mm Hg (95% CI, −4.2 to −0.3 mm Hg), respectively. For HE filtration, systolic and diastolic BP decreased by 2.9 mm Hg (95% CI, −6.2 to 0.5 mm Hg) and 0.8 mm Hg (95% CI, −2.8 to 1.2 mm Hg), respectively. Most secondary outcomes were not significantly improved.

    Conclusions and Relevance  Results of this study showed that short-term use of portable air filtration systems reduced personal PM2.5 exposures and systolic BP among older adults living in a typical US urban location. The use of these relatively inexpensive systems is potentially cardioprotective against PM2.5 exposures and warrants further research.

    Trial Registration  ClinicalTrials.gov identifier: NCT03334565

    Introduction

    The World Health Organization attributes more than 4 million deaths per year to ambient fine (<2.5 μm in diameter) particulate matter (PM2.5).1 Short-term exposures (eg, days) increase risks for numerous cardiovascular (CV) events, including myocardial infarction, stroke, and heart failure.2 Longer-term exposures (eg, years) amplify this risk and potentiate development of chronic cardiometabolic conditions (eg, type 2 diabetes, hypertension).2

    The extremely high air pollution levels across Asia are a major public health threat.1 However, mounting evidence indicates that even low levels of PM2.5 exposure within World Health Organization air quality guidelines of less than 10 μg/m3 pose significant health risks.1,3 Despite improvements in air quality during prior decades, the range of PM2.5 concentrations across the United States remains associated with excess mortality.3 As such, Brook et al4 and Giles et al5 have advocated more testing of preventive strategies that individuals can use to protect their health.

    With the US population spending nearly 90% of their time indoors—70% of this in their own residence6,7—portable residential air filtration units may be a practical tool for reducing PM2.5 exposures. A growing body of studies shows that high-efficiency particulate arrestance (HEPA) filtration can reduce indoor PM2.5 concentrations and may deliver health benefits. Although some trials demonstrated improvements in surrogate CV outcomes, including vascular function and blood pressure (BP),8-11 overall evidence remains mixed.12-14 Few studies have been performed in the United States with pollution levels more representative of urban environments faced by millions of at-risk individuals. In addition, a paucity of data exists among the elderly, the fastest growing vulnerable population, who are particularly susceptible to adverse health effects of PM2.5 exposure.15-19

    In this context, we tested the capacity of 2 inexpensive, commercially available air filtration systems to reduce PM2.5 exposures among elderly adults in a low-income senior residence in a typical urban US environment (Detroit, Michigan). We hypothesized that air filtration would reduce personal PM2.5 exposure, which differs from indoor levels due to several factors, including daily activities,20 thereby yielding improvements in CV health. Change in BP was selected as the primary end point because high BP is the leading cause of global morbidity and mortality1,21 and because PM2.5 exposure has been shown to increase BP in our study location22-24 and across global environments.25,26

    Methods
    Study Population

    The Reducing Air Pollution in Detroit Intervention Study (RAPIDS) enrolled 40 nonsmoking adults not receiving supplementary oxygen and living in a government-subsidized, low-income residential building for senior citizens in Midtown Detroit; participants received an in-residence air filtration intervention. The building is near a major state highway (approximately 100 m, with 21 900 vehicles/d),27 major interstate highways (approximately 800 m, with 133 000 vehicles/d),27 and several large industrial facilities. Individual residences used the same floor plan (approximately 46.8 m2) and hydronic baseboard heating. Participants were not restricted from going outdoors or opening windows during the interventions. The study protocol is found in Supplement 1. The study was approved by the institutional review board of the University of Michigan, and participants signed a written informed consent document during screening visits.

    Intervention Study Design

    RAPIDS was a randomized, double-blind, 3-way crossover intervention study conducted from October 21, 2014, through November 4, 2016 (excluding December 1 through April 30). Interventions included 3 blinded scenarios in computer-generated random order: unfiltered ambient air exposure (sham filtration), low-efficiency (LE) HEPA-type filtration, and high-efficiency (HE) true-HEPA filtration of ambient air using air purifier systems in the bedroom and main living space of each residence. The LE filter removes 99% of particles at 2.0 μm in size, whereas the HE filter removes 99.97% of particles at 0.3 μm in size. Each scenario lasted 3 days, separated by 1-week washout periods.

    On Monday during each study week, an unblinded team member placed randomized portable air filter systems (HAP424-U; Holmes), with a clean air delivery rate of 3.29 m3/min for smoke, in each participant’s residence. Participants, health technicians, and the data analysists (S.D.A. and J.D.) were blinded to intervention ordering. Participants wore personal air monitors starting at 8:00 am and carried them for 72 hours. Each participant underwent CV outcome testing in a fasting condition (>8 hours) at the same time between 8:00 and 10:00 am on 3 consecutive days starting 24 hours after filter system placement (Tuesday through Thursday). Daily PM filter samples were collected in each participant’s residence throughout each 3-day filtration period, during which time no filtration, LE recirculating filtration (HAPF30D-U2 HEPA-type filter; Holmes), or HE recirculating filtration (HAPF300D-U2 true-HEPA filter; Holmes) was used. For the sham condition, the air filtration systems (ie, HAP424-U) were operated without any filter element.

    Cardiovascular Outcome Measurements

    The primary outcome was brachial BP. Although this outcome included systolic BP (SBP) and diastolic BP (DBP), we specifically powered the trial based on a change in SBP because in prior studies in Detroit22,23 that had been completed at the time of designing this study, ambient PM2.5 had shown stronger and more consistent association with SBP compared with DBP. A repeated-measures design with 40 participants provided 90% power to detect a 1.4–mm Hg difference in SBP between active filtration interventions (HE and LE together) vs sham filtration. Secondary outcomes included noninvasive aortic hemodynamics, pulse-wave velocity (PWV), and heart rate variability (HRV). First, participants rested while seated for 5 minutes, and then CV outcome measurements were performed each morning in the following order: BP (approximately 5 minutes), PWV (approximately 10 minutes), and HRV (approximately 6 minutes). The measurement protocol is detailed in Supplement 1 and briefly reviewed below.

    Brachial BP

    Five-minute resting seated BP, a well-established causal factor for CV events,21 was measured according to guidelines28 using an automatic validated device (BpTRU; http://www.medsource-sw.com/blood-pressure/item-bpm-100/). The mean of the last 5 of 6 automated BP measurements (taken at 60-second intervals) was recorded.

    Central Aortic Hemodynamics and PWV

    A detection system (SphygmoCor System; http://atcormedical.com) was used to measure aortic augmentation pressure and augmentation indices (alone and controlled to a heart rate of 75 beats/min), metrics of aortic arterial pressure-wave reflection, and aortic systolic and pulse pressure. Carotid-femoral PWV was also determined by applanation tonometry using this system.29

    Cardiac Autonomic Function

    Participants rested supine for 6 minutes of continuous electrocardiographic monitoring (Evo Holter system; Spacelabs Healthcare). Time domain (SD of normal-to-normal R-R intervals) and frequency domain (high and low frequencies) HRV metrics were analyzed using echocardiographic analysis software (Pathfinder system; https://www.spacelabshealthcare.com:443/).

    Exposure Assessment
    Indoor Air Sampling

    Indoor PM2.5 samples were collected in the living room at the furthest point from the air filtration unit. Twenty-four–hour indoor PM2.5 samples were collected daily on polytetrafluoroethylene (PTFE [Teflon; Pall Laboratory]) filters using cyclone sample inlets at a flow rate of 16.7 L/min.

    Personal Air Sampling

    Each participant wore a battery-powered personal particulate monitor (pDR-1500; Thermo Scientific) that collected particles on 37-mm PTFE filters for subsequent gravimetric analysis. The monitor also continuously recorded PM2.5 concentration, relative humidity, and temperature. Participants were instructed to place monitors on a nearby nightstand or equivalent while sleeping.

    Outdoor Air Sampling

    We collected 24-hour ambient PM2.5 samples daily on PTFE filters using a dichotomous sequential air sampler (Partisol-Plus Model 2025, Rupprecht and Patashnick, Inc). All samples were processed and analyzed in class 100 ultraclean rooms at the Michigan State University Exposure Science Laboratory and the University of Michigan Air Quality Laboratory.

    Statistical Analysis

    Summary statistics were calculated as the overall mean (SD) of the mean value for each participant under each intervention scenario. We tested whether LE and HE filtration resulted in CV outcomes statistically better than sham filtration using mixed models of the following formula:

    CVit = bo + [(β11 × LEit) + (β12 × HEit)] + (β2 × CONFOUNDERit) + εit

    where CVit is the continuous CV health outcome in individual i at time t; bo is the overall intercept; β11 and β12 are the overall effects of LE vs unfiltered air and HE vs unfiltered air, respectively; CONFOUNDERit is a vector of time-varying covariates that may confound associations of interest (eg, intervention sequence, calendar time, month of intervention, temperature, or day of intervention); and β2 is the associated effect of these confounders. As a balanced design in which every participant contributes information to every intervention, time-invariant characteristics such as sex, race, and age cannot confound the associations of interest. Within the error term of this model (εit), we accounted for the repeated nature of the samples from each participant and allowed for increased correlation among observations from the same participant that are closer in time.

    In sensitivity analyses, we explored the inclusion of additional adjustment for outdoor PM2.5 exposure, examined different covariance structures, and tested whether intervention effects varied during the 3 days of sampling or by personal characteristics (ie, obesity, defined as body mass index [calculated as weight in kilograms divided by height in meters squared] ≥30, and sex) using interaction terms in our models. We also tested the effects of any filtration vs no filtration. We used SAS software (version 9.3; SAS Institute Inc) to implement these models; P < .05 indicates significance.

    Results

    Forty participants (25 men [62%] and 15 women [38%]; mean [SD] age, 67 [8] years) were enrolled based on our a priori power analyses for the primary outcome. Participants were predominantly African American (38 [95%]) (Table 1), and nearly one-half had class I obesity (19 [48%]). Mean brachial SBP and DBP during sham scenarios were 133.2 (17.1) and 82.1 (10.6) mm Hg, respectively. Nearly all participants contributed 3 BP measurements per intervention scenario, totaling 359 of 360 expected measurements (Table 2 and Figure 1). Less complete information was available on the secondary outcomes, leaving 34 participants and 208 measurements for secondary analyses.

    The mean outdoor PM2.5 concentration during our study was 9.3 (3.2) μg/m3, with similar levels across different interventions. Mean indoor and personal PM2.5 concentrations without LE or HE air filtration were 17.5 (13.0) μg/m3 and 15.5 (10.9) μg/m3, respectively. Mean personal PM2.5 exposure reductions by LE and HE filtration (compared with sham filtration) were 31% and 53% (P < .05), respectively (Melissa M. Maestas, PhD, R.D.B, R.A.Z., et al; unpublished data; December 2017) (Table 2).

    Compared with the sham scenario, using any air filtration for 3 days decreased brachial SBP by 3.2 mm Hg (95% CI, −6.1 to −0.2 mm Hg) and brachial DBP by 1.5 mm Hg (95% CI, −3.3 to 0.2 mm Hg) (Figure 2A). Figure 2B illustrates the temporal variation of brachial SBP and DBP during the interventions and shows that LE filtration reduced mean SBP by 3.4 mm Hg (95% CI, −6.8 to −0.1 mm Hg) and mean DBP by 2.2 mm Hg (95% CI, −4.2 to −0.3 mm Hg). High-efficiency filtration decreased mean SBP by 2.9 mm Hg (95% CI, −6.2 to 0.5 mm Hg) and mean DBP by 0.8 mm Hg (95% CI, −2.8 to 1.2 mm Hg), respectively. These reductions did not differ significantly between HE and LE air filtration (P = .75 for SBP and P = .14 for DBP).

    In post hoc exploratory analyses, we evaluated for potential effect modifiers of the BP responses. The 19 individuals with obesity experienced significantly greater decreases in SBP (−7.5 mm Hg; 95% CI, −12.0 to −3.1 mm Hg) and DBP (−2.9 mm Hg; 95% CI, −5.6 to −0.2 mm Hg) with filtration compared with the 21 nonobese participants (SBP, −0.4 mm Hg [95% CI, −3.7 to 4.5 mm Hg]; DBP, −0.6 mm Hg [95% CI, −3.1 to 2.0 mm Hg]) (P < .001 for interaction for brachial SBP; P = .01 for interaction for brachial DBP) (Figure 2C). Except for PWV, HE and LE filtration also improved all secondary outcomes more for obese participants than for nonobese participants, although these differences only met statistical significance for aortic pulse pressure (decrease by 4.6 mm Hg; 95% CI, −7.3 to −1.8 mm Hg). The intervention effects did not consistently differ by other factors, including sex and day of intervention.

    Among secondary CV study outcomes, central aortic, aortic augmentation pressure, pulse pressure, and augmentation index controlled to a heart rate of 75 beats/min tended to decrease during the filtration interventions. For instance, compared with no filtration, any filtration reduced pulse pressure by 1.9 mm Hg (95% CI, −3.7 to −0.01 mm Hg), but remaining secondary outcomes were not statistically different from no association (Figure 3). No significant consistent differences between interventions were observed for PWV and cardiac HRV variables (eTable in Supplement 2).

    Discussion

    Fine particulate matter air pollution is the fifth leading risk factor for global morbidity and mortality.1 Even low levels across the United States pose significant public health risks3; however, no proven personal strategy exists to protect at-risk individuals.4 We demonstrate herein that 2 relatively inexpensive (<US $70), commercially available portable air filtration systems can significantly decrease SBP and 24-hour mean personal PM2.5 exposure in elderly adults in a typical urban US location (Detroit). There was also a concomitant reduction in DBP; however, this change did not reach statistical significance. We further demonstrate that health benefits may appear more rapidly than previously known,12,16 with BP reductions manifesting within 3 days. Notably, 24-hour mean reductions in personal PM2.5 exposures occurred despite air filtration systems being fixed to indoor in-residence locations. The effectiveness of this intervention on personal exposures is important because in real-world scenarios, people spend a variable portion of time outside of their residence.

    Previous Air Filter Studies

    Several prior studies tested air filter interventions in mostly younger adults living in exceptionally polluted regions (Asia) or locations heavily affected by wood smoke.9,11,13,30 Improvements in BP and metrics of vascular function have been reported.9,11 A recent study among students in Shanghai31 demonstrated reductions in circulating stress hormone levels (eg, catecholamines, corticosteroids) and other markers of adverse systemic responses by metabolomic profiling. Conversely, we conducted our trial among an elderly population at greater risk of health effects of PM2.5 exposure and in a typical US city facing more widely relevant PM2.5 levels (ie, compliant with National Ambient Air Quality Standards) from urban sources. The few other HEPA-intervention studies conducted in less-polluted environments12-14 reported mixed changes in health end points. Although a study of elderly adults in Denmark reported improved microvascular function,8 BP levels were not affected, and experimental limitations prevented replication of their results in a follow-up trial.12 Blood pressure was also only a secondary outcome and not repetitively determined for 3 days using a fully automated device as in our trial, which likely contributed to our ability to detect significant reductions. Last, a recent in-home air filtration study of 21 middle-aged adults living near a Massachusetts roadway14 reported negative findings. The smaller sample size, younger age, lack of air filters placed in participants’ bedrooms (important for the success of 24-hour mean exposure reduction),12 and single measurement of follow-up BP may have contributed to their null findings. Nevertheless, together with our new findings, the overall body of evidence highlights the need for further large-scale investigations to fully understand the potential health benefits of air filtration systems.

    Biological Mechanisms

    Short-term increases in ambient PM2.5 concentrations promote elevated arterial BP in areas with poor or good air quality.22-26 Autonomic imbalance favoring sympathetic activation and vascular dysfunction have been implicated.2,24,25 Our study could not elucidate the precise mechanism because HRV metrics and arterial function variables were not consistently improved with filtration. This finding may be owing to inadequate power to evaluate secondary end points or research technique shortcomings or because other pathways may be responsible. Nonetheless, our results support a trend toward improved central aortic hemodynamics and arterial compliance with parallel decreases in brachial BP. In follow-up analyses, we will evaluate via metabolomic profiling whether stress hormones (eg, activation of the hypothalamic-pituitary-adrenal axis) or other hemodynamically active mediators were mechanistically involved.31 We will also investigate whether microvascular tone, assessed by retinal arterial photography, played a mediating role.

    Clinical Implications

    For this study, BP was only monitored during a 3-day period of air filtration. However, the observed 3.2-mm Hg reduction in SBP could possibly be sustained for more prolonged interventions. Even such modestly lower BP levels, if maintained for the long term (eg, months to years), could result in an approximate 16% decrease in composite CV events based on epidemiologic calculations.21 Given the size of the population affected by PM2.5 exposure,1 widespread use of economical exposure-reduction solutions could potentially deliver substantial improvements in global public health.4 We recognize this possibility is only speculation, and we aim to launch a follow-up study specifically to evaluate the efficacy and health benefits of longer-term interventions. Our results also showed that participants with obesity may exhibit greater decreases in BP from air filtration. This finding is consistent with that of a recent review in which 11 of 14 panel studies showed stronger associations between PM2.5 exposure and acute changes in physiological measures of CV health among obese participants, including BP.32 Because the prevalence of obesity has more than doubled since 1980, improved understanding of the interactions among air pollution, obesity, CV health, and intervention tools will be required to tackle this important public health issue.

    A recent study of the Medicare population3 demonstrated that the adverse effects of PM2.5 exposure are more pronounced among self-identified racial minorities and people with low income. Seniors in urban low-income housing are particularly vulnerable to air pollution, and an economical and easily implemented intervention is needed to reduce their PM exposure. This group is understudied; in 2015, almost 4.2 million seniors lived below the poverty level and another 2.4 million were classified as near poor.19 To our knowledge, this study is the first to focus on low-income housing facilities in an urban US environment and on personal PM2.5 exposures.

    Limitations

    Although the slightly more expensive HE system was more effective in reducing personal exposures, this intervention did not clearly yield superior CV health outcomes, likely because the study was a priori powered to detect a difference in SBP while using any air filtration (HE and LE pooled together) vs sham filtration. This power may have therefore been inadequate to detect statistically significant changes in secondary CV outcomes. As such, the differences shown in Figure 3 represent secondary (hypothesis-generating) end points only, and a larger follow-up trial is needed to determine whether the small (statistically insignificant) differential BP response unexpectedly favoring LE over HE filtration was a result of chance alone, as we suspect.

    Although this study showed that interventions as short as 24 hours can reduce BP, previous studies10 suggest that a longer intervention (eg, 9 days) might have demonstrated more robust changes and detectable improvements in our secondary CV outcomes. A preliminary study in Taiwan31 conducted for 1 year suggests that this may indeed be the case. Longer-term studies of months to years in duration are ultimately required to determine whether health benefits of air filtration persist and could thereby potentially translate into reductions in overt CV events (eg, myocardial infarctions).

    The study size and design were inadequate for assessing effect modification by multiple factors. Our objective was to model a real-world scenario as much as possible; thus, we did not exclude participants based on the presence of many comorbidities, including hypertension or use of specific antihypertensive medications. The heterogeneous nature of the participants may have also led to variability in responses owing to differences in underlying disease states and medications. Furthermore, although the participants were nonsmokers and the residential building was a nonsmoking building, we did not assess effects of secondhand smoke. In future analyses, we plan to assess PM2.5 components and their sources (eg, smoking) related to changes in CV outcomes.

    Whether such in-home interventions would be less effective among more free-living adults, such as those who spend more time outside their residence, also remains unknown. Considering these limitations, larger trials are required to determine optimal populations to target and the comparative effectiveness among various strategies (eg, face masks) for intervention.

    Finally, the primary study end point was brachial BP. Although SBP significantly decreased, the reduction in DBP after any filtration did not reach statistical significance, most likely because we specifically powered the study based on changes in SBP and it may have therefore been underpowered to detect a decrease in DBP. Our prior studies in Detroit had suggested that ambient PM2.5 may have a more consistent association with SBP compared with DBP.22,23 However, recent studies have found that ambient PM2.5 can increase SBP and DBP.24 In this study, DBP was also significantly decreased by LE filtration. These findings suggest that future trials with appropriate power are warranted to determine whether air filtration indeed lowers DBP and not just SBP. Regardless, the fact that SBP alone was reduced in this study is still of clinical relevance because SBP is well established as a stronger and more important determinant of CV risk in elderly people than DBP.33

    Conclusions

    In this trial, use of indoor portable air filtration for 3 days led to significant reductions in SBP in elderly adults. Our findings suggest that this relatively inexpensive and practical approach may be an effective tool for reducing PM2.5-related health effects. Future studies are required to better understand how to optimally deploy this personal-level intervention in real-world settings, how it performs among different populations (eg, patients with established cardiovascular diseases), and its efficacy over longer time frames (eg, >9 days).

    Back to top
    Article Information

    Accepted for Publication: May 27, 2018.

    Published Online: September 10, 2018. doi:10.1001/jamainternmed.2018.3308

    Open Access: This article is published under the JN-OA license and is free to read on the day of publication.

    Corresponding Author: Masako Morishita, PhD, Department of Family Medicine, College of Human Medicine, Michigan State University, 788 Service Rd, Room B117, East Lansing, MI 48824 (tomoko@msu.edu).

    Author Contributions: Drs Morishita and Brook had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Morishita, Adar, Bard, Spino, Brook.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: Morishita, Bard, Brook.

    Critical revision of the manuscript for important intellectual content: Morishita, Adar, D’Souza, Ziemba, Spino, Brook.

    Statistical analysis: Adar, D’Souza, Spino.

    Obtained funding: Morishita, Adar, Spino, Brook.

    Administrative, technical, or material support: Morishita, Ziemba, Bard, Brook.

    Supervision: Morishita, Bard, Brook.

    Conflict of Interest Disclosures: Dr Brook reported receiving an investigator-initiated research grant from RB, Inc. No other disclosures were reported.

    Funding/Support: This study was supported by research grant R01NR014484 from the National Institute of Nursing Research.

    Role of the Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: We thank the participants and the very supportive staff at the study residential facility. Ryan Crane, BS, Zachary Klaver, MSE, and Fengyao Li, MPH, Michigan State University Exposure Science Laboratory, and Susan Lustig, University of Michigan Division of Cardiovascular Medicine, contributed field work and data compilation efforts. These research assistants were compensated in part by grant R01NR014484.

    References
    1.
    Cohen  AJ, Brauer  M, Burnett  R,  et al.  Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015.  Lancet. 2017;389(10082):1907-1918. doi:10.1016/S0140-6736(17)30505-6PubMedGoogle ScholarCrossref
    2.
    Brook  RD, Rajagopalan  S, Pope  CA  III,  et al; American Heart Association Council on Epidemiology and Prevention, Council on the Kidney in Cardiovascular Disease, and Council on Nutrition, Physical Activity and Metabolism.  Particulate matter air pollution and cardiovascular disease: an update to the scientific statement from the American Heart Association.  Circulation. 2010;121(21):2331-2378. doi:10.1161/CIR.0b013e3181dbece1PubMedGoogle ScholarCrossref
    3.
    Di  Q, Wang  Y, Zanobetti  A,  et al.  Air pollution and mortality in the Medicare population.  N Engl J Med. 2017;376(26):2513-2522. doi:10.1056/NEJMoa1702747PubMedGoogle ScholarCrossref
    4.
    Brook  RD, Newby  DE, Rajagopalan  S.  The global threat of outdoor ambient air pollution to cardiovascular health: time for intervention.  JAMA Cardiol. 2017;2(4):353-354. doi:10.1001/jamacardio.2017.0032PubMedGoogle ScholarCrossref
    5.
    Giles  LV, Barn  P, Künzli  N,  et al.  From good intentions to proven interventions: effectiveness of actions to reduce the health impacts of air pollution.  Environ Health Perspect. 2011;119(1):29-36. doi:10.1289/ehp.1002246PubMedGoogle ScholarCrossref
    6.
    Matz  CJ, Stieb  DM, Davis  K,  et al.  Effects of age, season, gender and urban-rural status on time-activity: Canadian Human Activity Pattern Survey 2 (CHAPS 2).  Int J Environ Res Public Health. 2014;11(2):2108-2124. doi:10.3390/ijerph110202108PubMedGoogle ScholarCrossref
    7.
    Klepeis  NE, Nelson  WC, Ott  WR,  et al.  The National Human Activity Pattern Survey (NHAPS): a resource for assessing exposure to environmental pollutants.  J Expo Anal Environ Epidemiol. 2001;11(3):231-252. doi:10.1038/sj.jea.7500165PubMedGoogle ScholarCrossref
    8.
    Bräuner  EV, Forchhammer  L, Møller  P,  et al.  Indoor particles affect vascular function in the aged: an air filtration–based intervention study.  Am J Respir Crit Care Med. 2008;177(4):419-425. doi:10.1164/rccm.200704-632OCPubMedGoogle ScholarCrossref
    9.
    Allen  RW, Carlsten  C, Karlen  B,  et al.  An air filter intervention study of endothelial function among healthy adults in a woodsmoke-impacted community.  Am J Respir Crit Care Med. 2011;183(9):1222-1230. doi:10.1164/rccm.201010-1572OCGoogle ScholarCrossref
    10.
    Morishita  M, Thompson  KC, Brook  RD.  Understanding air pollution and cardiovascular diseases: is it preventable?  Curr Cardiovasc Risk Rep. 2015;9(6):1-9. doi:10.1007/s12170-015-0458-1PubMedGoogle ScholarCrossref
    11.
    Chen  R, Zhao  A, Chen  H,  et al.  Cardiopulmonary benefits of reducing indoor particles of outdoor origin: a randomized, double-blind crossover trial of air purifiers.  J Am Coll Cardiol. 2015;65(21):2279-2287. doi:10.1016/j.jacc.2015.03.553PubMedGoogle ScholarCrossref
    12.
    Karottki  DG, Spilak  M, Frederiksen  M,  et al.  An indoor air filtration study in homes of elderly: cardiovascular and respiratory effects of exposure to particulate matter.  Environ Health. 2013;12(1):116. doi:10.1186/1476-069X-12-116PubMedGoogle ScholarCrossref
    13.
    Kajbafzadeh  M, Brauer  M, Karlen  B, Carlsten  C, van Eeden  S, Allen  RW.  The impacts of traffic-related and woodsmoke particulate matter on measures of cardiovascular health: a HEPA filter intervention study.  Occup Environ Med. 2015;72(6):394-400. doi:10.1136/oemed-2014-102696PubMedGoogle ScholarCrossref
    14.
    Padró-Martínez  LT, Owusu  E, Reisner  E,  et al.  A randomized cross-over air filtration intervention trial for reducing cardiovascular health risks in residents of public housing near a highway.  Int J Environ Res Public Health. 2015;12(7):7814-7838. doi:10.3390/ijerph120707814PubMedGoogle ScholarCrossref
    15.
    Sacks  JD, Stanek  LW, Luben  TJ,  et al.  Particulate matter-induced health effects: who is susceptible?  Environ Health Perspect. 2011;119(4):446-454. doi:10.1289/ehp.1002255PubMedGoogle ScholarCrossref
    16.
    Pope  CA  III, Hansen  ML, Long  RW,  et al.  Ambient particulate air pollution, heart rate variability, and blood markers of inflammation in a panel of elderly subjects.  Environ Health Perspect. 2004;112(3):339-345. doi:10.1289/ehp.6588PubMedGoogle ScholarCrossref
    17.
    Wang  C, Tu  Y, Yu  Z, Lu  R.  PM2.5 and cardiovascular diseases in the elderly: an overview.  Int J Environ Res Public Health. 2015;12(7):8187-8197. doi:10.3390/ijerph120708187Google ScholarCrossref
    18.
    Peng  RD, Bell  ML, Geyh  AS,  et al.  Emergency admissions for cardiovascular and respiratory diseases and the chemical composition of fine particle air pollution.  Environ Health Perspect. 2009;117(6):957-963. doi:10.1289/ehp.0800185PubMedGoogle ScholarCrossref
    19.
    US Department of Health and Human Services.  A Profile of Older Americans: 2016. Administration on Aging, US Dept of Health and Human Services; 2016. https://www.acl.gov/sites/default/files/Aging%20and%20Disability%20in%20America/2016-Profile.pdf. Accessed July 5, 2018.
    20.
    Lewtas  J.  Air pollution combustion emissions: characterization of causative agents and mechanisms associated with cancer, reproductive, and cardiovascular effects.  Mutat Res. 2007;636(1-3):95-133. doi:10.1016/j.mrrev.2007.08.003PubMedGoogle ScholarCrossref
    21.
    Rahimi  K, Emdin  CA, MacMahon  S.  The epidemiology of blood pressure and its worldwide management.  Circ Res. 2015;116(6):925-936. doi:10.1161/CIRCRESAHA.116.304723PubMedGoogle ScholarCrossref
    22.
    Brook  RD, Bard  RL, Burnett  RT,  et al.  Differences in blood pressure and vascular responses associated with ambient fine particulate matter exposures measured at the personal versus community level.  Occup Environ Med. 2011;68(3):224-230. doi:10.1136/oem.2009.053991PubMedGoogle ScholarCrossref
    23.
    Dvonch  JT, Kannan  S, Schulz  AJ,  et al.  Acute effects of ambient particulate matter on blood pressure: differential effects across urban communities.  Hypertension. 2009;53(5):853-859. doi:10.1161/HYPERTENSIONAHA.108.123877PubMedGoogle ScholarCrossref
    24.
    Giorgini  P, Rubenfire  M, Das  R,  et al.  Particulate matter air pollution and ambient temperature: opposing effects on blood pressure in high-risk cardiac patients.  J Hypertens. 2015;33(10):2032-2038. doi:10.1097/HJH.0000000000000663PubMedGoogle ScholarCrossref
    25.
    Brook  RD, Sun  Z, Brook  JR,  et al.  Extreme air pollution conditions adversely affect blood pressure and insulin resistance: the Air Pollution and Cardiometabolic Disease Study.  Hypertension. 2016;67(1):77-85. doi:10.1161/HYPERTENSIONAHA.115.06237PubMedGoogle ScholarCrossref
    26.
    Cai  Y, Zhang  B, Ke  W,  et al.  Associations of short-term and long-term exposure to ambient air pollutants with hypertension: a systematic review and meta-analysis.  Hypertension. 2016;68(1):62-70. doi:10.1161/HYPERTENSIONAHA.116.07218PubMedGoogle ScholarCrossref
    27.
    Michigan Department of Transportation. 2013 Average daily traffic (ADT) maps. https://gis-mdot.opendata.arcgis.com/datasets/2013-traffic-volumes. 2013. Accessed July 5, 2018.
    28.
    Pickering  TG, Hall  JE, Appel  LJ,  et al.  Recommendations for blood pressure measurement in humans and experimental animals, part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research.  Circulation. 2005;111(5):697-716. doi:10.1161/01.CIR.0000154900.76284.F6PubMedGoogle ScholarCrossref
    29.
    Vlachopoulos  C, Aznaouridis  K, Stefanadis  C.  Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis.  J Am Coll Cardiol. 2010;55(13):1318-1327. doi:10.1016/j.jacc.2009.10.061PubMedGoogle ScholarCrossref
    30.
    Chuang  HC, Ho  KF, Lin  LY,  et al.  Long-term indoor air conditioner filtration and cardiovascular health: a randomized crossover intervention study.  Environ Int. 2017;106:91-96. doi:10.1016/j.envint.2017.06.008PubMedGoogle ScholarCrossref
    31.
    Li  H, Cai  J, Chen  R,  et al.  Particulate matter exposure and stress hormone levels: a randomized, double-blind, crossover trial of air purification.  Circulation. 2017;136(7):618-627. doi:10.1161/CIRCULATIONAHA.116.026796PubMedGoogle ScholarCrossref
    32.
    Weichenthal  S, Hoppin  JA, Reeves  F.  Obesity and the cardiovascular health effects of fine particulate air pollution.  Obesity (Silver Spring). 2014;22(7):1580-1589. doi:10.1002/oby.20748PubMedGoogle ScholarCrossref
    33.
    Franklin  SS, Larson  MG, Khan  SA,  et al.  Does the relation of blood pressure to coronary heart disease risk change with aging? the Framingham Heart Study.  Circulation. 2001;103(9):1245-1249. doi:10.1161/01.CIR.103.9.1245Google ScholarCrossref
    ×