Participants

The main purpose of the prospective, longitudinal and multidisciplinary cohort 4HAIE study was to evaluate the impact of living in an air-polluted environment on physiological, biomechanical, psychosocial and anthropometric variables in male and female runners and inactive controls. A total of 1314 participants were recruited from two different regions within the Czech Republic that differ in air quality [16]. The highly polluted Moravian-Silesian Region (MSR, N = 750) was compared with South Bohemia (SB, N = 564) as the control region with low air-pollution. A two-day laboratory assessment was performed by all participants including a series of physiological, biomechanical, anthropometric, cognitive and MRI assessments, which has been described in more detailed in the protocol papers [17–19]. To be included in the study, participants had to meet the following criteria: non-smoker, aged between 18–65 years at the start of the study. Runners had to meet public PA guidelines (150 min/week moderate or 75 min/week vigorous PA or an equivalent combination of the two as per WHO recommendations; be running at least 10 km per week in the last 6 weeks and plan to continue running for the next 12 months). Inactive controls did not meet the public PA recommendations but were capable of normal PA including running. Both runners and inactive participants were required to have lived in the recognized regions for at least 5 years with the prerequisite not to move residency to another region in the next 12 months. Exclusion criteria included: being a smoker, having an acute health problem (pain /injury / surgery) within the last 6 weeks preventing from physical activity, having contraindications to magnetic resonance imaging or dual-energy x-ray absorptiometry (DXA) [19].

The present study includes data from 1036 participants from the original 4HAIE data set. Only data from participants who had no missing data in the variables analysed and no detected outliers as observation via Cook´s distance (see details in statistical section) were included in the analysis. Based on this, 278 participants were excluded from the analyses. The 4HAIE study is conducted in accordance with the Declaration of Helsinki. The study was approved by the Ethics Committee of the University of Ostrava (3/2018). Participant were provided detailed information sheets and a written informed consent was obtained from all participants.

High air-polluted region versus low-polluted region

The Czech Republic follows the European Union (EU) (EU air quality standards (europa.eu)) for PM10, PM2.5, NO2, benzene, and BaP limit values, which are 40 μg/m3, 20 μg/m3, 40 μg/m3, 5 μg/m3, and 1 ng/m3, respectively. Moravian-Silesian region (MSR) defined by six districts is located at the eastern part of the Czech Republic and is among the highest air-polluted areas from fourteen high-polluted European Union regions. On the other hand, the South Bohemia region (SBR) located at the opposite part of the Czech Republic and defined by seven districts is a resource-poor area without significant sources of raw energy materials compared to MSR and is characterized as low-polluted region [20]. Long-term differences in air pollution between the MSR and the SBR published by Machaczka et al.[16] for the time period 2000–2017 showed that these EU standards were exceeded in the MSR region on average for PM10 by 12.3%, for PM2.5 by 75%, and for BaP by 160%, without exceeding of the limit values in the SB region (all data are shown in S1 Table). The cross-sectional data collection used for this study ran from 1. April 2019 to 30. August 2021. Based on data from the Czech Hydrometeorological Institute, Fig 1A and 1B show the differences in the five-year (2017–2021) average of annual mean concentrations between regions for the air pollutants PM2.5 and BaP [21].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1.

a-b. Five years (2017–2021) average of annual mean concentrations of the PM 2.5 (1a) and benzo(a)pyren (CHMU, 2021).

https://doi.org/10.1371/journal.pone.0315767.g001

Cardiac autonomic regulation

Cardiac autonomic regulation was assessed by heart rate variability (HRV) measures and for the purpose of this study HRV was expressed by the time-domain most robust vagal modulation variable root mean square value of successive differences of the normal RR intervals (rMSSD) [22]. 10-min of recording in supine position was performed after waking in the morning (until 7:00 am) before breakfast in a quiet room without any external disturbances at the start of the second-day laboratory assessment. Participants were also instructed the evening before to remain in bed upon waking up and limit their activities to only essential tasks, such as using the bathroom, until the researcher arrived to carry out the HRV measurement. The first 5 minutes of the recording were used to achieve steady-state conditions and this part was disregarded for the analysis. ECG data obtained from a Bittium Faros device (Bittium Inc, Oulu, Finland) from two electrodes placed on a chest belt were then imported into Kubios HRV Premium software (Kubios Oy, Kuopio, Finland). The length of the analysed record was 5 minutes according to task force recommendations [23]. Before extracting the R-wave time points, the R-waves were interpolated to 2000 Hz. Analysis began with visual inspection for artefacts and through an automated artefact detection algorithm, which was also applied by the Kubios software. All visual inspections and edits were consulted and agreed by the team of 3 researchers. Artefacts included missing, redundant, or misaligned rhythm detections, as well as ectopic rhythms such as premature ventricular contractions or other arrhythmias which were filtered out from the RR interval time series used for HRV analysis. Slow nonstationary oscillations in the RR intervals were detrended using smoothness priors [17].

Graded exercise test (GXT)

As a measure of CRF level, maximum oxygen uptake (VO_2peak) was obtained from a GXT test. The protocol of the GXT started with 3 min of walking at 5.0 km/h to familiarise subjects with the treadmill. The GXT protocol then commenced at 6.0 km/h, with speed subsequently increasing by 1.0 km/h every minute until volitional exhaustion at an inclination remaining at 1% throughout the test. Expired air was continuously monitored to analyse O₂ and CO₂ concentrations during the GXT with a breath-by-breath system (Blue Cherry, Geratherm Medical AG, Germany) which was calibrated according to the manufacturer’s guidelines. The highest average O₂ consumption measured over a 30 s period prior to completing the GXT was used to determine VO_2peak. More detailed information regarding exercise performance have been presented recently [24]. Participants who did not pass the Physical Activity Readiness Questionnaire [25], were not allowed to perform the GXT unless explicit permission (given by a medical doctor) was provided. Blood pressure (BP) was also checked before the GXT. In case of BP values exceeding 150/90 mm Hg, participants were not allowed to perform the GXT, but continued in the study protocol. Each participant was advised not to participate in any vigorous activity 24 hours prior to the test [17].

Anthropometry

Dual-energy X-ray absorptiometry (DXA) was used as the gold standard method for estimating visceral adipose tissue (VAT). The measurements were executed using the Hologic Horizon A bone densitometer (Hologic Discovery A, Waltham, Massachusetts, USA). For the DXA measurements, the height (InBody 370 stadiometer, Biospace, South Korea) and weight (InBody 770 bioimpedance, BioSpace, South Korea) of the participants were used. A whole-body scan was used for the analysis of body composition, including segmental analyses [17].

Statistical analysis

A linear regression model with multiple independent variables and one dependent variable was employed. The dependent variable in the model was the heart rate variability variable rMSSD. Independent variables included CRF level expressed as maximal oxygen uptake (O_2peak), visceral adipose tissue (VAT) and age, sex, region of residency, and socioeconomic status represented by education level were used as covariates. For the analysis, the level of education was divided into the three following categories: Education level 1 (basic, apprentice, secondary vocational without GSCE), Education level 2 (secondary school diploma–general and vocational), Education level 3 (higher vocational–post secondary, university and higher). All numerical data were controlled for normal distribution using the Shapiro-Wilk test, which yielded negative results. A logarithmic transformation (log10) was applied to achieve normal distribution at a significance level of 0.01 for rMSSD and VAT (data before transformation is given in S2 Table). Additionally, the residuals vs. fitted plot was utilized to assess the assumptions of linearity and homoscedasticity in the linear model. Cook’s distance was used to identify influential observations within the model. Observations with Cook’s distance exceeding the value two have a greater potential to impact the estimated coefficients of the model as well as homoscedasticity were excluded from further analyses. The provided summary includes information about model coefficients, standard errors, t-statistics, p-values, and other statistical indicators at a significance level of 0.05.

Cardiac autonomic regulation and cardiorespiratory fitness

We found significant association between HRV and CRF (expressed as maximal oxygen uptake; O_2peak). This finding is in line with Medeiros et al. [29] who also found a significant relationship between CRF assessment (by the Leger shuttle run test) and cardiac autonomic regulation (expressed by rMSSD). Similarly, Kiviniemi et al. [30] revealed a significant correlation between CRF and cardiac autonomic regulation (assessed by rMSSD variables) in 3144 participants (1761 females). Our results and results above indicate that improving the CRF level could be one of the best options for achieving improvements in cardiac autonomic modulation. Cardiorespiratory fitness is partly genetically determined, but can be improved substantially by regular physical activity or exercise [31]. The Danish Health Examination Survey study by Eriksen et al. [32] of 16 025 participants aged 18–65 years (DANHES) showed that CRF level, obtained from a GXT and expressed as VO_2max, was inversely associated with total daily sitting time. The harmful effects of a lack of leisure time physical activity (LTPA) on CRF were presented in the results of the National Health and Nutrition Examination Survey (NHANES) in US adults aged 20–49 years [33]. Furthermore, the DANHES study, observed that participants who reported lower amounts of LTPA also had lower CRF levels and vice versa. Lastly, the Finish population-based study MOPO also indicated a positive association between the amount of the physical activity and cardiac autonomic regulation in 3395 adolescent males [34]. The individual CRF level was estimated indirectly via the Polar Fitness Test^TM which is based on resting heart rate, heart rate variability, gender, age, body weight, height and the self-assessment of the level of long-term physical activity. Participants who reported “high” or “top” amounts of PA/week had significant higher rMSSD compared to the “low” and “moderate” groups.

Although we did not take the intensity of the physical activity into account in our analysis, Kim et al.[35] showed that exercise intensity could have beneficial effects on cardiac autonomic regulation. It was observed that high-intensity interval training significantly increased CRF levels and rMSSD compared to the control/no exercise group in 34 male regular smokers [35]. However, no significant differences were observed between the moderate training group and control/no exercise group for either CRF or rMSSD [35] highlighting the importance of exercise intensity.

Cardiac autonomic regulation and long-term exposure to air-pollution

Our analyses did not find any association between HRV and residing for at least 5-years in a high or low-polluted region within the Czech Republic. These results are contrary to the results of other studies evaluating short-term exposure to ambient air-pollution. For example, the population-based ARIC study by Liao et al. [36] demonstrated significant adverse effects of higher levels of PM₁₀, ozone, carbon monoxide, nitrogen dioxide, and sulphur dioxide on cardiac autonomic control, however did not control for CRF. Similar acute, negative trends were observed for HRV after exposure to low-levels of PM_2.5 in a small sample of participants (n = 10) [37].

The potential positive acute effects of physical activity on cardiac autonomic function in a polluted environment have been observed previously [38]. Cole Hunter et al. [38] examined the acute effects of exercising in low and high-traffic environments on cardiac autonomic function. The authors found that physical activity performed even in high traffic-induced air pollution environments still had a beneficial acute effect on 24-hour cardiac autonomic regulation. It was speculated that physical activity may offset the impact of high air pollution on parasympathetic modulation of the heart even at higher levels of air pollution exposure [38]. However, the CRF level of the participants was not mentioned, and so it is unclear whether being physically active or also having a certain, baseline level of fitness is the cause for such beneficial effects [38]. Definitely, there is a lack of long-term exposure to ambient air-pollution studies. One rare study evaluated the short-term and 1-year long effect of the exposure to PM_2.5 and black carbon (BC) particulate air-pollution in an elderly cohort of the Normative Aging Study [39]. This study included result from 540 males and HRV time and frequency domain variables were obtained from 12 min recordings in a sitting position where the last 7 min were used for final analyses. The results of this study confirmed adverse effects of a short-term exposure to air-pollution on heart rate variability. On the other hand, evaluation of 1-year long exposure to PM_2.5 and BC showed similar patterns which did not provide any significant strong association between pollutant and heart rate variability and provided inconsistent results [39]. It also needs to be critically noted that CRF was not identified as an important confounding factor in this study [39]. Therefore, the long-term exposure to air pollution or the controlling of CRF may explain why no association was found with HRV and air pollution in the current study. Moreover, the discrepancy between our findings and studies reporting a negative impact of air pollution on heart rate variability could be also attributed to methodological differences. In particular, our approach of dividing participants into experimental and control groups solely on the basis of regional pollution differences is relatively new in this field. This methodological distinction may contribute to the different results compared to studies that directly measured exposure to pollutants or quantified individual exposure levels. In addition, our study cohort, which included healthy runners and inactive individuals aged 18–65 years, stratified by sex and age, differs from other studies that predominantly included smaller samples, individuals at increased cardiovascular risk or with pre-existing cardiovascular disease, or older adults.

Based on scientific evidence, we know that physical activity has a positive effect to increase an individual’s cardiorespiratory fitness, which has protective effects against the development of cardiovascular or other non-communicable diseases, as well as a positive effect on cardiac autonomic regulation. On the other hand, due to the lack of research investigating the engagement in physical activity within polluted environments there is currently no clear consensus on whether physical activity in an air polluted environment has a clear beneficial or adverse impact on an individual’s health. One such investigation, a systematic review conducted by Gandhi et al. [40] demonstrated that the cardiovascular benefits of outdoor physical activity can outweigh the adverse effects of air pollution. However, the authors also noted that this positive effect diminishes when physical activity is performed in environments with high levels of pollutants, particularly fine particulate matter (PM_2.5) [40]. Similarly, Andersen et al. [41] explored the combined effects of physical activity and air pollution on mortality among elderly urban residents through data from the Danish Diet, Cancer, and Health cohort. Their findings indicated that physical activity, even in areas with elevated ambient nitrogen dioxide (NO₂) concentrations, may attenuate, but does not entirely negate its beneficial effects on respiratory mortality [41]. Due to the increasing prevalence of urbanization, it is evident that further longitudinal studies are required to definitively determine whether engaging in physical activity within polluted air exerts a beneficial or detrimental impact on one´s health. Our results however, in general do not show an increased risk of being active in polluted regions at least for the investigated healthy group of participants, a finding which was independent from age and sex.

Cardiac autonomic regulation and abdominal obesity

Sympathetic nerve stimulation has been shown to increase fatty acid release, and either a decrease in sympathetic activity or an increase in parasympathetic activity was causally related to an accumulation of white adipose tissue [42]. However, in this specific cohort of participants, we did not find an association between cardiac autonomic regulation and VAT, which is in contrast to Windham et al. [11] who found a significant association between WC and cardiac autonomic regulation expressed as rMSSD [11]. Differences between our results and Windham et al. [11] could be explained by the different variables measured (waist circumference vs DXA) for expressing VAT. Research assessing the relationship between cardiac autonomic regulation and the amount of VAT is lacking. Some rare studies with inconsistent results only use the rather weak parameter BMI to express the degree of obesity.

Strength and limitations

The primary strengths of this study are the large cohort stratified for sex, age and region, which were clearly defined by differences in air pollution according to European Union standards. The study also reveals that CRF is a significant predictor of cardiac autonomic modulation, influenced by activity and age, but not by living in an area with high air pollution over the long-term.

However, there are several limitations, which have to be considered. First, we acknowledge that the cross-sectional design does not allow to us assess and discuss causalities which could be addressed by the forthcoming second wave of measurement. Second, we did not obtain direct information on lifetime exposure to air pollution on an individual level and actual air-pollution exposure may vary throughout the year and within regions. Third, the study included only healthy participants which does not allow to evaluate valuable effects of air pollution on at-risk subjects and to generalize the results to the overall population.

Practical application

Our study results showed no association between living in areas with high or low pollution and HRV. On the other hand, we found a significant and strong association between the CRF level and HRV. These findings point to the importance of a high CRF level (e.g., through increased physical activity and exercise training) to achieve improvements in cardiac autonomic regulation and other health related risk factors regardless of living in low or high air polluted regions. It is therefore, recommended to include CRF measures in all future studies as CRF is likely a strong confounder, which has been rarely included in most studies regarding health concerns.