The timeline of blood pressure changes and hemodynamic responses during an experimental noise exposure

Environmental Research 163 (2018) 249–262

Contents lists available at ScienceDirect

Environmental Research journal homepage: www.elsevier.com/locate/envres

The timeline of blood pressure changes and hemodynamic responses during an experimental noise exposure

T

⁎

Katarina Paunovića, , Branko Jakovljevića, Vesna Stojanovb a

Institute of Hygiene and Medical Ecology, Faculty of Medicine, University of Belgrade, Dr Subotića 8, 11000 Belgrade, Serbia Multidisciplinary Center for the Diagnostics and Treatment of Arterial Hypertension, Clinical Center of Serbia; Faculty of Medicine, University of Belgrade, Pasterova 2, 1100 Belgrade, Serbia b

A R T I C L E I N F O

A B S T R A C T

Keywords: Blood pressure Chronology Hemodynamics Noise Vascular resistance

Background: Noise exposure increases blood pressure and peripheral vascular resistance in both genders in an experimental setting, as previously reported by the authors. Objectives: The aim of this re-analysis was to present the minute-by-minute timeline of blood pressure changes and hemodynamic events provoked by traffic noise in the young and healthy adults. Methods: The experiment consisted of three 10-min phases: rest in quiet conditions before noise (Leq = 40 dBA), exposure to recorded road-traffic noise (Leq = 89 dBA), and rest in quiet conditions after noise (Leq = 40 dBA). Participants’ blood pressure, heart rate, and hemodynamic parameters (cardiac index and total peripheral resistance index) were concurrently measured with a thoracic bioimpedance device. The raw beat-to-beat data were collected from 112 participants, i.e., 82 women and 30 men, aged 19–32 years. The timeline of events was created by splitting each experimental phase into ten one-minute intervals (30 intervals in total). Four statistical models were fitted to answer the six study questions what is happening from one minute to another during the experiment. Results: Blood pressure decreased during quiet phase before noise, increased in the first minute of noise exposure and then decreased gradually toward the end of noise exposure, and continued to decline to baseline values after noise exposure. The cardiac index showed a gradual decrease throughout the experiment, whereas total vascular resistance increased steadily during and after noise exposure. Conclusions: The timeline of events in this 30-min experiment provides insight into the hemodynamic processes underlying the changes of blood pressure before, during and after noise exposure.

1. Introduction Human reactions to noise in an experimental setting were a focal point of research for a long time. The biochemical, endocrine and cardiovascular changes (Ising et al., 1999; Ising and Michalak, 2004) provoked by short-term exposures to loud sounds in the laboratory helped researchers understand noise as an environmental stressor, explore the physiological processes underlying these reactions, and explain the effects of noise on one's mental health, annoyance, sleep, and cardiovascular system (Babisch, 2002). In a similar attempt to comprehend blood pressure changes caused by noise, a team of researchers conducted an extensive experimental study in Belgrade in 2012, when young healthy volunteers with normal blood pressure were exposed to recorded road-traffic noise for ten minutes and had their cardiovascular and hemodynamic parameters monitored before, during and after noise

exposure. The experiment showed a significant increase in blood pressure and peripheral vascular resistance and a decrease in heart flow and blood volume during noise exposure in comparison to quiet conditions before or after noise (Paunovic et al., 2011; Paunović et al., 2013, 2014). Encouraged by these findings the team wanted to take a closer look at the chronological course of the events happening throughout the experiment. We were not satisfied with reporting only the average values of the investigated cardiovascular parameters before, during and after noise exposure, because they provided no insight into the timeline of the events during the experiment. We were, in short, wondering when the changes of blood pressure started, how long they lasted, and when they eventually ended. We expected not only to observe the physiological processes in every minute of the experiment, but also to analyze them in relation to the preceding minutes and between the

Abbreviations: Leq, equivalent noise level; dBA, A-weighted decibel unit ⁎ Corresponding author. E-mail address: [email protected] (K. Paunović). https://doi.org/10.1016/j.envres.2018.01.048 Received 15 December 2017; Received in revised form 29 January 2018; Accepted 31 January 2018 0013-9351/ © 2018 Elsevier Inc. All rights reserved.

Environmental Research 163 (2018) 249–262

K. Paunović et al.

pumped blood volume, ml/m2; calculated as stroke volume normalized for body surface area), cardiac output (minute volume of heart, l/min), and cardiac index (indicator of global heart flow, l/(min*m2); calculated as cardiac output normalized for body surface area). The validity of impedance cardiography to measure stroke volume and cardiac output in healthy adults was confirmed in comparison to other standard methods (direct Fick method, dye dilution, carbon-dioxide rebreathing) (Woltjer et al., 1997). In addition, the thoracic electrical bioimpedance device measures heart rate (beat/min) and blood pressure (mmHg) in two ways: by oscillometry (a cuff is placed on patient's left forearm) and during every heart beat (continuous beat-to-beat measurement on patient's right hand; two ring cuffs are placed on person's index finger and middle finger). Mean arterial pressure (mmHg) is calculated as a sum of systolic pressure and doubled diastolic pressure, divided by three. The device performs the oscillometric blood pressure measurements every three minutes, making such recording inapplicable for the timeline analysis; they were, thus, excluded from the presented study. Finally, based on previous cardiac parameters, heart rate and blood pressure, the device calculates total peripheral resistance (the resistance of all peripheral vasculature in the systemic circulation, dyne*s/cm5), as well as total peripheral resistance index (indicator of systemic vascular resistance, dyne*s*m2/cm5; calculated as total peripheral resistance normalized for body surface area). Body surface area is calculated from the Mosteller formula (Mosteller, 1987) as a square root of a product of body weight (in kilograms) and body height (in centimeters) divided by 3600. The state of blood inside the vascular system, i.e., hemodynamics is determined by mean arterial pressure, cardiac output and peripheral vascular resistance (Paunović et al., 2014). Once the experiment was completed, data were exported from the thoracic electrical bioimpedance device, assembled into a database and analyzed for outliers. At this stage, data from 10 participants were excluded due to the errors in the recordings. Later, a detailed data analysis revealed errors in the measurement of three participants and missing data in five participants, who were thus excluded from the timeline analysis. Finally, the minute-by-minute timeline of all the events during the experiment was created from the data collected from 112 participants, i.e., 82 women and 30 men. Given the limited space, we were not able to describe all the investigated cardiovascular parameters; we have decided to report the timeline of blood pressure (obtained from the continuous beat-to-beat measurement), and its hemodynamic constituents, i.e., peripheral vascular resistance, cardiac output, and heart rate.

experimental phases, hoping to understand blood pressure changes and hemodynamic responses in short-term stressful events. We decided to perform a new analysis of the existing data, aiming to assess the minuteby-minute timeline of the changes in blood pressure and other cardiovascular and hemodynamic parameters provoked by recorded traffic noise in the young and healthy adults. 2. Materials and methods The experimental study was performed in 2012 in the collaboration between the Institute of Hygiene and Medical Ecology, Faculty of Medicine, and the Multidisciplinary Center for the Diagnostics and Treatment of Arterial Hypertension, Clinical Center of Serbia. The study design is described in details previously (Paunovic et al., 2011; Paunović et al., 2013, 2014). In short, 132 young healthy volunteers, i.e., medical students and young medical doctors (aged 19–32 years), signed an informed consent form to participate in the experiment in which they would be exposed to loud recorded road-traffic noise for ten minutes and would concurrently have their cardiovascular and hemodynamic parameters monitored by a non-invasive method. The expected results of the experiment were not discussed with the participants before, during, or after the testing procedure, to prevent from the possible anticipation-bias. The study was approved by the Ethics Committee of the Clinical Center of Serbia in 2009. Before enrolling the study, all participants underwent a medical examination and had their weight and height measured in light clothes and barefoot. Body mass index (BMI) was calculated as body weight (in kilograms) divided by squared body height (in meters). At this point, some volunteers were excluded from the study due to the presence of diabetes (n = 1), kidney diseases (n = 1), obesity (n = 3), hypertension (n = 5), and arrhythmia (n = 2) (Paunović et al., 2014). Participants fulfilled a questionnaire containing socio-demographic data (age, gender), smoking habits (non-smoker / current smoker / ex-smoker), regular engagement in physical activity for at least 30 min most days per week (yes / no), and some eating habits, such as daily use of coffee (yes / no), and daily addition of salt to food during meals (yes / no) (Paunović et al., 2014). None of the participants followed any specific dietary regimen. 2.1. Experimental procedure All eligible participants were asked to avoid smoking, drinking coffee or intensive physical activity for at least two hours before the testing procedure. The experimental procedure consisted of three phases. At the beginning before noise exposure, participants rested for 10 min in quiet conditions (Leq = 40 dBA). In the second phase, participants listened to the recorded road-traffic noise (Leq = 89 dBA) for 10 min. After noise exposure, participants remained lying for another 10 min in quiet conditions (Leq = 40 dBA). Two loudspeakers were placed at both sides of a subject's head at 30 cm distance. Equivalent noise levels were measured with Hand Held Noise Level Analyzer Type 2250 ‘Brüel & Kjær’ at the level of participants’ ear. Participants were lying on their back with their arms and legs outstretched and with the thoracic electrical bioimpedance device connected during the whole course of the experiment (Paunović et al., 2014). Thoracic electrical bioimpedance device (Task Force® Monitor, CNSystems Medizintechnik AG, Graz, Austria) consists of impedance cardiography electrodes (ICG), electrocardiography electrodes (ECG) and two sets of cuffs for blood pressure measurement. Impedance cardiography (ICG) electrodes measure the change in thorax impedance during the flow of the alternating current through the thorax. Subsequently, the impedance change during a cardiac cycle serves to estimate thoracic fluid content (the electrical conductivity of the chest cavity, determined by fluids in the thorax, 1/kOhm), and to compute several cardiac parameters, such as: stroke volume (the amount of pumped blood during each systole, ml), systolic index (indicator of

2.2. Timeline analysis In order to perform the timeline analysis, all three experimental phases that lasted for ten minutes were divided into ten one-minute intervals, and the average values of all the investigated parameters were calculated from raw beat-to-beat data. This procedure created a total of thirty average values per investigated parameter per person. Once all individual data were aggregated, the following four comparisons were made: 1) the values obtained at a given minute were compared to the values obtained at the preceding minute (Model 1; 29 pairs of values in total); 2) the values obtained at a given minute were compared to the values obtained at the first minute of the same phase (Model 2; 9 pairs of values per phase, i.e., 27 pairs of values in total); 3) the values obtained at every minute of noise exposure and quiet phase after noise were compared to the values obtained at the last minute of quiet phase before noise (Model 3; 10 pairs of values per phase, i.e., 20 pairs of values in total); and 4) the values obtained at every minute of quiet phase after noise were compared to the values obtained at the last minute of noise exposure (Model 4; 10 pairs of values in total). These models were incorporated into the following study questions: 250

Environmental Research 163 (2018) 249–262

K. Paunović et al.

Question 1) what is happening in the first experimental phase before noise? To answer, the values obtained at every minute of the quiet phase before noise were compared to the first minute of the same phase (Model 2), and to the preceding minute (Model 1). Question 2) what is happening in the second experimental phase during noise exposure? To answer, all the values obtained at every minute of noise exposure were compared to the first minute of the same phase (Model 2), and to the preceding minute (Model 1). Question 3) is there a difference between noise exposure and the quiet phase before noise? To answer, the values obtained at every minute of noise exposure were compared to the values obtained at the last minute of the quiet phase before noise (Model 3). Question 4) what is happening in the third experimental phase after noise? To answer, the values obtained at every minute of the quiet phase after noise were compared to the first minute of the same phase (Model 2), and to the preceding minute (Model 1). Question 5) is there a difference between the quiet phase after noise and the quiet phase before noise? To answer, the values obtained at every minute of the quiet phase after noise were compared to the values obtained at the last minute of the quiet phase before noise (Model 3). Question 6) is there a difference between the quiet phase after noise and noise exposure? To answer, the values obtained at every minute of the quiet phase after noise were compared to the values obtained at the last minute of noise exposure (Model 4).

distribution. All differences were tested with Student's paired-samples test for parametric data, or with Wilcoxon signed ranks test for nonparametric data. The analyses were performed using SPSS 15.0 for Windows software (SPSS Inc. 1989–2006).

3. Results The basic characteristics of the investigated men and women are presented in Table 1. Men and women shared similar smoking and eating habits, but men practiced physical activity more often than women. Given the differences in their blood pressure values at baseline, the timeline analysis was performed separately by gender. The minute-by-minute timeline of changes of systolic blood pressure in men and women is presented in Fig. 1. Tables 2, 3 present the four models exploring the changes of systolic blood pressure during the experiment in men and women respectively. In both genders, there is a significant decrease in systolic pressure within the first two minutes of the first phase of the experiment and a significant increase in systolic pressure in the first minute of noise exposure (Model 2); a decrease in systolic pressure in the first minute of quiet phase after noise is observed in men only (Model 1). In men, there are significant decreases in systolic pressure values from the first minute of the quiet phase before noise and minutes 3–9 of the same phase (Model 2), significant decreases from the first minute of noise exposure and minutes 15–20 (Model 2), and significant decreases between the first minute of the quiet phase after noise and minutes 26, 27 and 29 of the same phase (Model 2) (Table 2). Furthermore, there are significant differences between the last minute before noise and all minutes of noise exposure (Model 3), as well as significant differences between the last minute of noise exposure and all minutes of the quiet phase after noise (Model 4) (Table 2). In women, there are significant differences in systolic pressure values between the first minute of the quiet phase before noise and

2.3. Statistical analysis Descriptive statistic is presented as mean values and standard deviation for numeric variables or as percent's (relative numbers) for categorical variables. All parameters were tested by Kolmogorov-Smirnov test and almost all observed distributions corresponded to the normal Table 1 Baseline characteristics of the investigated population by gender. Characteristics

Men

Women

Number of participants (%) Age (years) (Mean ± SD) Body mass index (kg/m2) (Mean ± SD) Systolic pressure before noise exposure (mmHg) (Mean ± SD) Diastolic pressure before noise exposure (mmHg) (Mean ± SD) Current smoker (%) Daily consumption of coffee (%) Daily addition of salt to food (%) Regular engagement in physical activity (%)

30 (26.8) 24.70 ± 2.59 24.77 ± 3.26 119.09 ± 8.37 73.53 ± 5.59 7 (23.3) 17 (56.7) 11 (36.7) 13 (43.3)

82 (73.2) 24.80 ± 2.64 20.97 ± 2.82 105.56 ± 9.77 68.37 ± 8.22 14 (17.1) 59 (72.0) 28 (34.1) 15 (18.3)

SD – Standard Deviation.

Fig. 1. The timeline of changes of systolic blood pressure in men and women.

251

Environmental Research 163 (2018) 249–262

K. Paunović et al.

Table 2 Minute-by-minute analysis of systolic pressure changes during the experiment in men. Experimental phase

Minute of the experiment

Systolic pressure (mmHg) (Mean ± SD)

Model 1 – difference from the preceding minutea

Model 2 – difference from the first minute of the same phasea

Quiet before noise

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

121.28 ± 9.20 120.16 ± 8.07 119.01 ± 7.23 118.34 ± 8.46 117.64 ± 9.87 118.41 ± 9.58 117.87 ± 9.70 118.28 ± 8.46 118.61 ± 8.96 119.59 ± 9.28 124.41 ± 9.43 123.81 ± 9.46 123.01 ± 8.93 122.98 ± 8.85 122.75 ± 8.20 122.78 ± 7.81 121.77 ± 7.27 121.37 ± 7.89 121.91 ± 8.42 121.93 ± 7.71 120.34 ± 7.09 119.67 ± 7.41 119.18 ± 7.42 119.89 ± 7.62 119.69 ± 6.80 119.43 ± 7.26 118.22 ± 7.44 119.00 ± 7.83 118.24 ± 8.21 118.46 ± 8.37

Question 1

Question 1

Noise exposure

Quiet after noise

Question 2

Question 4

0.077 0.026 0.233 0.268 0.335 0.892 0.773 0.597 0.119 < 0.001 0.378 0.085 0.957 0.662 0.934 0.055 0.420 0.273 0.974 0.001 0.229 0.323 0.429 0.502 0.512 0.117 0.083 0.281 0.646

Model 3 – difference from the last minute before noisea

Model 4 – difference from the last minute of noise exposurea

0.077 0.010 0.001 0.005 0.010 0.006 0.003 0.014 0.110 Question 2

Question 3 0.378 0.105 0.117 0.036 0.071 0.006 0.002 0.012 0.013

Question 4

Question 5 0.229 0.098 0.359 0.132 0.012 0.004 0.181 0.048 0.108

< 0.001 0.001 0.002 0.002 0.001 0.001 0.019 0.005 0.017 0.005 0.428 0.924 0.671 0.901 0.863 0.435 0.137 0.780 0.268 0.348

Question 6

0.001 0.001 0.002 0.023 0.013 0.001 < 0.001 0.013 0.001 0.004

SD – standard deviation. a Student's paired-samples test.

one minute to the other; there is a significant decrease in minute 8 before noise and a significant increase from the previous minute in minutes 12 and 16 during noise exposure (Model 1). In addition, there are significant decreases in heart rate from the first minute of noise exposure to minutes 12–15 and 19–20 of the same phase (Model 2) (Table 6). In women, heart rate values vary insignificantly from one minute to the other; there are significant decreases in minutes 5, 12 and 22, as well as increases in minutes 10, 14 and 30 of the experiment in relation to the preceding minute (Model 1). In addition, there are significant decreases in heart rate from the first minute of noise exposure to the minutes 12–13 of the same phase (Model 2), and significant decreases from the first minute of quiet phase after noise and minutes 22, 24 and 26 (Model 2). Finally, there are significant differences between the last minute before noise and all the minutes of noise exposure and the quiet phase after noise (Model 3) (Table 7). The minute-by-minute timeline of changes of cardiac index in men and women is presented in Fig. 4. Tables 8 and 9 present the four models exploring the changes of cardiac index during the experiment in men and women respectively. In both genders, there is a significant decrease of the cardiac index within the first few minutes of the first phase of the experiment and a significant increase in cardiac index in the first few minutes of noise exposure (Model 1). In men, there are significant decreases in cardiac index between the first minute of the quiet phase before noise and minutes 3–9 of the same phase (Model 2); significant decreases between the first minute of noise exposure and all the minutes of the same phase (Model 2); and significant decreases between the first minute of quiet phase after noise and almost all the minutes of the same phase (Model 2) (Table 8). In addition, there is a significant difference between the last minute before noise and the first minute of noise exposure (Model 3), and a significant difference between the last minute of noise exposure and the first minute of quiet

all the following minutes of the same phase (Model 2), as well as significant differences between the first minute of noise exposure and all the following minutes of noise exposure (Model 2) (Table 3). Furthermore, there are significant differences between the last minute before noise and minutes 11–15 and 19–20 of noise exposure (Model 3) (Table 3). The minute-by-minute timeline of changes of diastolic blood pressure in men and women is presented in Fig. 2. Tables 4, 5 present the four models exploring the changes of diastolic blood pressure during the experiment in men and women respectively. In both genders, there is a significant decrease of diastolic pressure within the first two minutes of the first phase of the experiment, a significant increase in diastolic pressure in the first minute of noise exposure (Model 1); a decrease of diastolic pressure in the first minute of quiet phase after noise is observed in men only (Model 1). In men, there are significant decreases in diastolic pressure values between the first minute of the quiet phase before noise and minutes 2–9 of the same phase (Model 2) (Table 4). Furthermore, there are significant differences between the last minute before noise and all minutes of noise exposure (Model 3), as well as significant differences between the last minute of noise exposure and almost all the minutes of the quiet phase after noise (Model 4) (Table 4). In women, there are significant decreases in diastolic pressure values from the first minute of the quiet phase before noise to all the following minutes of the same phase (Model 2) (Table 5). Furthermore, there are significant differences between the last minute before noise and all the minutes of both noise exposure and the quiet phase after noise (Model 3) (Table 5). The minute-by-minute timeline of changes of heart rate in men and women is presented in Fig. 3. Tables 6, 7 present the four models exploring the changes of heart rate during the experiment in men and women respectively. In men, heart rate values vary insignificantly from 252

Environmental Research 163 (2018) 249–262

K. Paunović et al.

Table 3 Minute-by-minute analysis of systolic pressure changes during the experiment in women. Experimental phase

Minute of the experiment

Systolic pressure (mmHg) (Mean ± SD)

Model 1 – difference from the preceding minutea

Model 2 – difference from the first minute of the same phasea

Quiet before noise

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

106.87 ± 9.72 105.61 ± 9.82 104.96 ± 10.17 104.57 ± 10.66 103.34 ± 9.80 103.03 ± 10.61 103.98 ± 11.41 103.54 ± 11.19 104.37 ± 10.07 103.94 ± 11.49 108.69 ± 12.34 106.93 ± 11.36 106.33 ± 10.84 105.62 ± 10.52 105.52 ± 10.23 105.20 ± 10.61 104.73 ± 11.14 105.15 ± 10.26 105.62 ± 9.11 105.94 ± 10.15 105.69 ± 8.98 105.20 ± 8.97 105.23 ± 8.82 106.19 ± 7.06 106.43 ± 7.34 106.41 ± 6.88 106.41 ± 7.24 105.99 ± 8.04 106.26 ± 8.58 106.55 ± 8.43

Question 1

Question 1

Noise exposure

Quiet after noise

Question 2

Question 4

0.018 0.196 0.160 0.672 0.128 0.070 0.052 0.091 0.421 < 0.001 < 0.001 0.163 0.104 0.973 0.410 0.140 0.304 0.270 0.431 0.770 0.360 0.941 0.537 0.819 0.792 0.292 0.447 0.471 0.596

Model 3 – difference from the last minute before noisea

Model 4 – difference from the last minute of noise exposurea

0.018 0.021 0.002 < 0.001 < 0.001 0.003 0.001 0.005 0.004 Question 2

Question 3 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.001 0.001

Question 4

Question 5 0.360 0.497 0.851 0.980 0.755 0.693 0.632 0.318 0.224

< 0.001 < 0.001 0.001 0.017 0.019 0.072 0.296 0.117 0.036 0.022 0.056 0.197 0.217 0.109 0.277 0.501 0.214 0.213 0.070 0.071

Question 6

0.770 0.400 0.435 0.619 0.656 0.166 0.507 0.375 0.498 0.839

SD – standard deviation. a Student's paired-samples test.

present the four models exploring the changes of total peripheral resistance index during the experiment in men and women respectively. In men, total peripheral resistance index varies insignificantly from one minute to the other through the most time of the experiment (Model 1). There are significant increases from the first minute of the quiet phase before noise and minutes 8–10 of the same phase (Model 2), significant increases from the first minute of noise exposure to minutes 15–16 and 19–20 (Model 2), and significant increases from the first minute of the quiet phase after noise and minute 24 of the same phase (Model 2) (Table 10). In addition, there are significant differences between the last minute of the quiet phase before noise and minutes 15, 19 and 20 of noise exposure (Model 3); there are significant differences between the last minute of noise exposure and minutes 21, 26, 28, and 29 of the

phase after noise (Model 4) (Table 8). In women, there are significant decreases in cardiac index between the first minute of the quiet phase before noise and all the minutes of the same phase (Model 2); significant decreases between the first minute of noise exposure and minutes 13–20 of the same phase (Model 2); and significant decreases between the first minute of quiet phase after noise and almost all the minutes of the same phase (Model 2) (Table 9). In addition, there is a significant difference between the last minute before noise and almost all the minutes of noise exposure (Model 3), and a significant difference between the last minute before noise and all the minutes of quiet phase after noise (Model 3) (Table 9). The minute-by-minute timeline of changes of total peripheral resistance index in men and women is presented in Fig. 5. Tables 10, 11

Fig. 2. The timeline of changes of systolic blood pressure in men and women.

253

Environmental Research 163 (2018) 249–262

K. Paunović et al.

Table 4 Minute-by-minute analysis of diastolic pressure changes during the experiment in men. Experimental phase

Minute of the experiment

Diastolic pressure (mmHg) (Mean ± SD)

Model 1 – difference from the preceding minutea

Model 2 – difference from the first minute of the same phasea

Quiet before noise

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

75.70 ± 6.44 74.47 ± 5.49 73.40 ± 5.49 72.82 ± 5.79 72.79 ± 5.96 73.10 ± 6.18 72.45 ± 6.29 72.59 ± 6.01 73.45 ± 6.19 73.95 ± 6.39 76.08 ± 6.48 76.05 ± 7.11 75.78 ± 6.87 76.32 ± 6.87 75.88 ± 5.96 76.29 ± 5.70 75.98 ± 5.33 75.99 ± 5.40 76.50 ± 5.66 76.45 ± 5.57 75.32 ± 5.80 75.19 ± 5.51 74.39 ± 5.81 75.03 ± 6.24 75.29 ± 4.80 74.84 ± 4.74 74.54 ± 5.23 74.70 ± 5.87 74.20 ± 5.36 74.65 ± 5.73

Question 1

Question 1

Noise exposure

Quiet after noise

Question 2

Question 4

0.031 0.013 0.229 0.174 0.329 0.943 0.961 0.038 0.233 0.001 0.956 0.481 0.119 0.184 0.261 0.465 0.990 0.135 0.891 0.011 0.761 0.015 0.307 0.811 0.353 0.657 0.268 0.264 0.212

Model 3 – difference from the last minute before noisea

Model 4 – difference from the last minute of noise exposurea

0.031 0.003 0.000 0.001 0.001 < 0.001 0.001 0.012 0.058 Question 2

Question 3 0.956 0.618 0.675 0.730 0.693 0.872 0.874 0.500 0.573

Question 4

Question 5 0.761 0.065 0.574 0.409 0.074 0.037 0.223 0.108 0.401

0.001 0.002 0.003 < 0.001 0.001 < 0.001 0.001 0.001 < 0.001 < 0.001 0.031 0.073 0.586 0.245 0.368 0.811 0.633 0.462 0.994 0.490

Question 6

0.011 0.044 0.005 0.095 0.050 0.003 0.001 0.008 0.002 0.033

SD – standard deviation. a Student's paired-samples test. Table 5 Minute-by-minute analysis of diastolic pressure changes during the experiment in women. Experimental phase

Minute of the experiment

Diastolic pressure (mmHg) (Mean ± SD)

Model 1 – difference from the preceding minutea

Model 2 – difference from the first minute of the same phasea

Quiet before noise

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

70.54 ± 7.84 68.72 ± 8.25 67.85 ± 8.74 67.70 ± 8.90 66.47 ± 8.58 66.05 ± 8.96 66.23 ± 9.29 66.42 ± 9.42 66.83 ± 9.20 66.60 ± 9.94 68.54 ± 9.61 68.16 ± 8.68 68.17 ± 8.46 68.32 ± 8.14 68.00 ± 8.46 67.84 ± 8.80 67.82 ± 8.72 68.34 ± 7.91 68.95 ± 7.41 69.33 ± 8.04 69.12 ± 7.63 69.44 ± 7.70 69.60 ± 7.27 69.65 ± 8.01 69.50 ± 9.90 70.52 ± 6.20 70.39 ± 6.50 69.96 ± 6.48 70.06 ± 6.80 70.43 ± 6.82

Question 1

Question 1

Noise exposure

Quiet after noise

Question 2

Question 4

< 0.001 0.006 0.298 0.021 0.087 0.841 0.291 0.156 0.448 0.001 0.188 0.983 0.599 0.287 0.490 0.555 0.097 0.034 0.130 0.733 0.342 0.600 0.382 0.325 0.311 0.888 0.431 0.740 0.333

SD – standard deviation. a Student's paired-samples test.

254

Model 3 – difference from the last minute before noisea

Model 4 – difference from the last minute of noise exposurea

< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 Question 2

Question 3 0.188 0.324 0.608 0.275 0.200 0.145 0.713 0.479 0.199

Question 4

Question 5 0.342 0.234 0.972 0.835 0.198 0.149 0.067 0.033 0.019

0.001 0.010 0.005 0.001 0.006 0.017 0.036 0.006 < 0.001 < 0.001 0.002 0.001 < 0.001 0.004 0.052 0.004 0.006 0.010 0.003 0.002

Question 6

0.733 0.865 0.663 0.770 0.595 0.994 0.759 0.928 0.791 0.374

Environmental Research 163 (2018) 249–262

K. Paunović et al.

Fig. 3. The timeline of changes of heart rate in men and women.

Table 6 Minute-by-minute analysis of heart rate changes during the experiment in men. Experimental phase

Minute of the experiment

Heart rate (beat/min) (Mean ± SD)

Model 1 – difference from the preceding minutea

Model 2 – difference from the first minute of the same phasea

Quiet before noise

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

67.79 ± 13.83 67.88 ± 13.57 67.35 ± 13.42 67.87 ± 13.31 67.52 ± 13.95 67.70 ± 14.48 67.57 ± 13.78 67.13 ± 14.07 67.58 ± 14.15 68.23 ± 14.35 69.69 ± 15.34 67.81 ± 14.63 67.50 ± 14.29 67.97 ± 14.57 67.74 ± 14.44 68.58 ± 14.58 68.67 ± 14.75 68.14 ± 14.74 67.84 ± 14.77 67.73 ± 14.93 68.19 ± 14.66 67.24 ± 14.82 66.76 ± 14.14 66.60 ± 14.05 67.77 ± 13.51 67.41 ± 14.36 67.74 ± 14.97 67.76 ± 15.10 66.99 ± 15.30 67.08 ± 15.34

Question 1

Question 1

Noise exposure

Quiet after noise

Question 2

Question 4

0.801 0.174 0.343 0.332 0.404 0.755 0.030 0.418 0.102 0.126 0.002 0.417 0.221 0.485 0.021 0.813 0.132 0.482 0.697 0.342 0.077 0.453 0.636 0.252 0.338 0.973 0.404 0.075 0.804

Model 3 – difference from the last minute before noisea

Model 4 – difference from the last minute of noise exposurea

0.801 0.174 0.914 0.513 0.999 0.950 0.322 0.803 0.540 Question 2

Question 3 0.002 0.003 0.049 0.015 0.159 0.256 0.076 0.033 0.043

Question 4

Question 5 0.077 0.084 0.224 0.748 0.169 0.161 0.880 0.276 0.377

0.126 0.597 0.269 0.691 0.406 0.614 0.492 0.893 0.563 0.460 0.951 0.178 0.114 0.223 0.877 0.261 0.209 0.670 0.193 0.257

Question 6

0.342 0.413 0.207 0.581 0.813 0.544 0.395 0.615 0.669 0.774

SD – standard deviation. a Student's paired-samples test.

4. Discussion

quiet phase after noise (Model 4) (Table 10). In women, total peripheral resistance index varies insignificantly from one minute to the other through the most time of the experiment, with significant increases in the first minute of noise exposure and the first minute of the quiet phase after noise in relation to the preceding minute (Model 1). There are significant increases from the first minute of the quiet phase after noise and minutes 24, 25, 27–30 of the same phase (Model 2) (Table 11). In addition, there are significant differences between the last minute of the quiet phase before noise and all the minutes of both noise exposure and the quiet phase after noise (Model 3) (Table 11).

To the authors’ knowledge, this is the first timeline analysis of the changes in blood pressure and other hemodynamic parameters provoked by recorded traffic noise in an experimental setting. The analysis helped us answer the questions what is happening with their cardiovascular system from one minute to another during the 30-min experiment. Question 1) what is happening in the first experimental phase before noise? Answer: there are significant decreases in both systolic and

255

Environmental Research 163 (2018) 249–262

K. Paunović et al.

Table 7 Minute-by-minute analysis of heart rate changes during the experiment in women. Experimental phase

Minute of the experiment

Heart rate (beat/min) (Mean ± SD)

Model 1 – difference from the preceding minutea

Model 2 – difference from the first minute of the same phasea

Quiet before noise

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

75.61 ± 11.62 75.04 ± 11.31 75.50 ± 11.19 75.79 ± 10.94 75.39 ± 10.52 75.26 ± 10.61 74.88 ± 10.99 75.27 ± 10.92 75.12 ± 10.88 75.84 ± 11.42 75.15 ± 11.83 73.97 ± 11.19 73.84 ± 10.95 74.35 ± 11.06 74.40 ± 11.28 74.70 ± 11.50 74.39 ± 11.00 74.24 ± 10.78 74.06 ± 10.73 74.05 ± 10.95 74.19 ± 11.19 73.39 ± 10.72 73.55 ± 10.56 73.39 ± 11.27 73.65 ± 11.15 73.66 ± 11.06 74.15 ± 11.17 74.03 ± 10.70 73.94 ± 10.84 74.93 ± 11.62

Question 1

Question 1

Noise exposure

Quiet after noise

Question 2

Question 4

0.171 0.113 0.398 0.035 0.925 0.337 0.457 0.594 0.025 0.117 0.008 0.560 0.024 0.860 0.303 0.246 0.567 0.541 0.979 0.621 0.019 0.534 0.065 0.186 0.940 0.499 0.620 0.728 0.009

Model 3 – difference from the last minute before noisea

Model 4 – difference from the last minute of noise exposurea

0.171 0.801 0.846 0.352 0.176 0.126 0.471 0.328 0.668 Question 2

Question 3 0.008 0.009 0.129 0.142 0.360 0.156 0.083 0.058 0.067

Question 4

Question 5 0.019 0.093 0.003 0.065 0.036 0.179 0.354 0.234 0.436

0.117 < 0.001 < 0.001 0.002 0.006 0.013 0.003 0.001 0.001 0.001 0.002 < 0.001 < 0.001 < 0.001 < 0.001 0.001 0.002 0.004 0.002 0.121

Question 6

0.621 0.062 0.189 0.009 0.100 0.140 0.469 0.686 0.563 0.175

SD – standard deviation. a Student's paired-samples test.

Fig. 4. The timeline of changes of cardiac index in men and women.

diastolic pressure from the first to the tenth minute of the experiment in both genders. Blood pressure decreased within the first few minutes of this phase and remained low until the end of the phase. A similar pattern of events is observed for the cardiac index in both genders. Heart rate remained stable during the first experimental phase, whereas total peripheral resistance index increased in the last three minutes of this phase in men. These events come out of the fact that participants were lying down and resting for ten minutes in a quiet and comfortable setting. Question 2) what is happening in the second experimental phase during noise exposure? Answer: there are significant increases in both systolic and diastolic pressure in the first minute of noise exposure in comparison to the preceding minute (i.e., the last minute of the quiet

phase) in both genders. From that moment on, the values remained similar to the preceding minute of the same phase. Nevertheless, in comparison to the first minute of noise exposure (i.e., the eleventh minute of the experiment), systolic pressure decreased gradually until the twentieth minute of the experiment in women, and from the fifteenth to the twentieth minute in men. A gradual decrease in cardiac index during noise exposure is observed in both genders. Heart rate decreased during the first few minutes of noise exposure in both genders, whereas total peripheral resistance index increased in the last six minutes of this phase in men. Question 3) is there a difference between noise exposure and the quiet phase before noise? Answer: systolic and diastolic pressure values obtained at every minute of noise exposure were significantly higher 256

Environmental Research 163 (2018) 249–262

K. Paunović et al.

Table 8 Minute-by-minute analysis of cardiac index changes during the experiment in men. Experimental phase

Minute of the experiment

Cardiac index (l/(min*m2)) (Mean ± SD)

Model 1 – difference from the preceding minutea

Model 2 – difference from the first minute of the same phasea

Quiet before noise

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

3.78 ± 0.81 3.76 ± 0.77 3.67 ± 0.80 3.66 ± 0.80 3.52 ± 0.81 3.49 ± 0.77 3.57 ± 0.81 3.52 ± 0.77 3.51 ± 0.79 3.55 ± 0.80 3.72 ± 0.86 3.65 ± 0.85 3.61 ± 0.82 3.58 ± 0.83 3.55 ± 0.80 3.57 ± 0.81 3.55 ± 0.78 3.52 ± 0.73 3.52 ± 0.79 3.50 ± 0.75 3.57 ± 0.71 3.50 ± 0.72 3.48 ± 0.72 3.48 ± 0.75 3.53 ± 0.74 3.53 ± 0.75 3.42 ± 0.70 3.42 ± 0.66 3.39 ± 0.67 3.37 ± 0.69

Question 1

Question 1

Noise exposure

Quiet after noise

0.384 < 0.001 0.584 0.054 0.498 0.633 0.001 0.764 0.054 0.013 0.050 0.084 0.328 0.166 0.493 0.291 0.272 0.914 0.236 0.027 0.026 0.424 0.820 0.489 0.729 0.498 0.130 0.121 0.301

Question 2

Question 4

Model 3 – difference from the last minute before noisea

Model 4 – difference from the last minute of noise exposurea

0.384 < 0.001 0.001 < 0.001 0.002 < 0.001 < 0.001 < 0.001 < 0.001 Question 2

Question 3 0.050 0.013 0.012 0.002 0.004 0.004 0.002 0.001 0.001

Question 4

Question 5 0.026 0.037 0.009 0.075 0.038 0.010 0.274 0.084 0.032

0.013 0.060 0.187 0.393 0.924 0.671 0.934 0.490 0.539 0.314 0.650 0.330 0.227 0.085 0.396 0.215 0.111 0.501 0.177 0.087

Question 6

0.027 0.981 0.623 0.172 0.474 0.273 0.133 0.750 0.688 0.378

SD – standard deviation. a Student's paired-samples test. Table 9 Minute-by-minute analysis of cardiac index changes during the experiment in women. Experimental phase

Minute of the experiment

Cardiac index (l/(min*m2)) (Mean ± SD)

Model 1 – difference from the preceding minutea

Model 2 – difference from the first minute of the same phasea

Quiet before noise

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

4.29 ± 0.93 4.21 ± 0.90 4.18 ± 0.88 4.16 ± 0.88 4.09 ± 0.87 4.05 ± 0.90 4.02 ± 0.88 4.06 ± 0.88 4.05 ± 0.89 4.07 ± 0.89 4.07 ± 0.92 4.04 ± 0.91 4.00 ± 0.89 3.99 ± 0.86 3.98 ± 0.86 3.98 ± 0.87 3.94 ± 0.85 3.95 ± 0.85 3.93 ± 0.86 3.93 ± 0.86 3.93 ± 0.85 3.90 ± 0.85 3.89 ± 0.84 3.86 ± 0.85 3.86 ± 0.84 3.87 ± 0.88 3.88 ± 0.90 3.91 ± 0.91 3.89 ± 0.91 3.93 ± 0.93

Question 1

Question 1

Noise exposure

Quiet after noise

Question 2

Question 4

< 0.001 0.190 0.084 0.001 0.251 0.013 0.329 0.450 0.291 0.950 0.139 0.001 0.365 0.831 0.869 0.001 0.622 0.114 0.758 0.721 0.117 0.340 0.044 0.272 0.831 0.943 0.821 0.367 0.034

SD – standard deviation. a Student's paired-samples test.

257

Model 3 – difference from the last minute before noisea

Model 4 – difference from the last minute of noise exposurea

< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 Question 2

Question 3 0.139 0.001 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001

Question 4

Question 5 0.117 0.042 0.001 0.005 0.035 0.074 0.062 0.021 0.514

0.950 0.302 0.005 0.001 0.001 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.001

Question 6

0.721 0.251 0.069 0.007 0.057 0.148 0.183 0.175 0.063 0.792

Environmental Research 163 (2018) 249–262

K. Paunović et al.

Fig. 5. The timeline of changes of total peripheral resistance index in men and women.

Table 10 Minute-by-minute analysis of total peripheral resistance index changes during the experiment in men. Experimental phase

Minute of the experiment

Total peripheral resistance index (dyne*s * m2/cm5) (Mean ± SD)

Model 1 – difference from the preceding minutea

Model 2 – difference from the first minute of the same phasea

Quiet before noise

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

1977.58 ± 398.01 1959.72 ± 394.37 1986.06 ± 393.17 1989.80 ± 441.48 2061.90 ± 449.00 2080.68 ± 442.52 2028.51 ± 478.38 2056.96 ± 458.48 2080.47 ± 472.20 2079.47 ± 477.35 2066.45 ± 499.98 2093.98 ± 504.48 2104.56 ± 495.10 2130.16 ± 518.53 2131.05 ± 491.26 2132.14 ± 495.59 2127.59 ± 452.45 2127.66 ± 467.83 2155.31 ± 519.48 2158.52 ± 500.93 2069.53 ± 416.53 2107.56 ± 439.23 2100.04 ± 419.84 2116.06 ± 433.22 2102.70 ± 447.02 2084.03 ± 460.38 2130.35 ± 458.07 2142.03 ± 433.90 2147.08 ± 452.33 2168.50 ± 449.51

Question 1

Question 1

Noise exposure

Quiet after noise

Question 2

Question 4

0.277 0.176 0.858 0.519 0.658 0.379 0.018 0.090 0.948 0.736 0.157 0.641 0.154 0.958 0.956 0.858 0.997 0.198 0.815 0.002 0.092 0.666 0.512 0.628 0.458 0.593 0.397 0.768 0.248

Question 2

Question 4

Model 3 – difference from the last minute before noisea

Model 4 – difference from the last minute of noise exposurea

0.277 0.722 0.706 0.452 0.686 0.356 0.023 0.006 0.008 0.157 0.230 0.051 0.048 0.048 0.081 0.080 0.022 0.028 0.092 0.266 0.041 0.273 0.531 0.444 0.537 0.441 0.183

Question 3

Question 5

0.736 0.690 0.384 0.065 0.033 0.090 0.132 0.074 0.018 0.014 0.766 0.425 0.623 0.223 0.531 0.762 0.868 0.734 0.592 0.252

Question 6

0.002 0.106 0.128 0.274 0.151 0.025 0.098 0.050 0.034 0.159

SD – standard deviation. a Student's paired-samples test.

minute of this phase (i.e., the twenty-first minute of the experiment), blood pressure decreased in the last few minutes of the experiment. Heart rate remained stable in this phase among men, with some variations among women. The cardiac index declined during the third experimental phase in both genders. Total peripheral resistance index decreased in the first minute of this phase but started to increase until the end of the experiment in women mainly. Question 5) is there a difference between the quiet phase after noise and the quiet phase before noise? Answer: systolic pressure values obtained at every minute of the quiet phase after noise were similar to systolic pressure values obtained at the last minute of the quiet phase before noise in both genders. This was also the case with diastolic

than the values obtained at the last minute of the quiet phase before noise in both genders. A similar pattern of events is observed for total peripheral resistance index in women. On the other hand, heart rate and cardiac index values obtained at every minute of noise exposure were significantly lower in comparison to the values obtained at the last minute of the quiet phase before noise in women. Question 4) what is happening in the third experimental phase after noise? Answer: there are significant decreases in both systolic and diastolic pressure in the first minute of the quiet phase after noise in comparison to the preceding minute (i.e., the last minute of noise exposure) in men only. From that moment on, the values remained similar to the preceding minute of the same phase. In comparison to the first 258

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Quiet before noise

259

SD – standard deviation. a Wilcoxon signed ranks test.

Quiet after noise

Noise exposure

Minute of the experiment

Experimental phase

1633.85 ± 460.13 1628.77 ± 470.34 1621.10 ± 456.12 1629.35 ± 508.76 1630.71 ± 496.12 1656.00 ± 547.70 1667.56 ± 547.51 1648.88 ± 539.09 1660.27 ± 521.13 1650.05 ± 533.79 1720.06 ± 588.82 1713.54 ± 581.12 1730.00 ± 596.12 1722.28 ± 578.01 1710.00 ± 573.40 1707.71 ± 591.84 1730.73 ± 592.30 1732.86 ± 573.35 1758.65 ± 593.71 1773.13 ± 621.96 1758.07 ± 573.58 1773.69 ± 577.11 1777.93 ± 569.95 1818.49 ± 604.24 1806.88 ± 599.87 1816.24 ± 610.39 1829.60 ± 660.31 1808.25 ± 671.86 1812.80 ± 646.25 1805.34 ± 656.72

Total peripheral resistance index (dyne*s * m2/cm5) (Mean ± SD)

Table 11 Minute-by-minute analysis of total peripheral resistance index changes during the experiment in women.

Question 4

Question 2

Question 1 0.968 0.844 0.776 0.254 0.768 0.075 0.862 0.166 0.436 < 0.001 0.254 0.051 0.458 0.310 0.650 0.100 0.884 0.055 0.170 0.006 0.417 0.342 0.175 0.303 0.876 0.520 0.381 0.728 0.441

Model 1 – difference from the preceding minutea

Question 4

Question 2

Question 1

0.417 0.079 0.003 0.001 0.060 0.017 0.009 0.002 0.011

0.254 0.632 0.924 0.808 0.913 0.549 0.495 0.069 0.040

0.968 0.871 0.825 0.893 0.505 0.235 0.270 0.191 0.265

Model 2 – difference from the first minute of the same phasea

Question 5

Question 3

< 0.001 0.003 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001

Model 3 – difference from the last minute before noisea

Question 6

0.006 0.190 0.855 0.942 0.469 0.941 0.445 0.376 0.171 0.282

Model 4 – difference from the last minute of noise exposurea

K. Paunović et al.

Environmental Research 163 (2018) 249–262

Environmental Research 163 (2018) 249–262

K. Paunović et al.

during noise exposure in comparison to the quiet phase before noise. We assumed it to be a compensating physiological hemodynamic response that counteracts the increased vascular resistance to bring down blood pressure to regular values. We were, therefore, not surprised to report the gradient decrease in systolic pressure during noise exposure, as early as from the second minute of noise exposure in women, and from the fifth minute of noise exposure in men. Contrary to some expectations that noise would increase blood pressure during the whole exposure period, our minute-by-minute findings show that stressful events increase blood pressure rapidly, but compensatory mechanisms take place subsequently in order to normalize blood pressure. For this reason, one of the prime distinctions of the presented study and its strength is the ability to observe the course of events from one minute to another, and not to rely on the average values obtained from tenminute reading. In the recent years, Taiwanese researchers reported a direct effect of noise in environmental and occupational settings on blood vessels. Young healthy normotensive adults, such as male workers from an automobile company, who are exposed to 79–110 dBA noise levels at their workplace (Chang et al., 2007) and volunteers of both genders who are not professionally exposed to noise (Chang et al., 2012) were under surveillance for 24 h, with their noise exposure and vascular properties monitored continually. Both occupational and environmental noise exposures were associated with higher arterial compliance, distensibility and vascular resistance, which are responsible for the variations in blood pressure during the day (Chang et al., 2007, 2012). The researchers proposed that these effects result from the activation of the myogenic responses in arteries, whose purpose is to prevent immediate damage in vessels in acute noise exposure. After prolonged noise exposure, however, there is a high possibility for irreversible structural remodeling in small and large arteries (the socalled sympatheticotonia-induced endothelial lesion), which may lead to an increase of arterial resistance and the stress-induced elevation of blood pressure (Chang et al., 2007, 2012). To the authors’ knowledge, this is the first study to report the course of events after noise exposure. In the Swedish studies, noise-induced blood pressure and hemodynamic changes persisted for a few minutes after the termination of experimental noise exposure and disappeared within five (Andrén et al., 1981) to ten minutes (Andrén et al., 1982a) of rest at background noise. On the contrary, we were able to observe that blood pressure values after noise exposure were similar or slightly higher than the values before noise exposure in both genders. As for the underlying hemodynamic parameters, there was a continuous decline in cardiac index and heart rate, as well as a continuous increase in total peripheral vascular resistance among women. These findings suggest that compensatory mechanisms take more time to counteract the effects of a stressor than just ten minutes. The presented experiment cannot be straightforwardly generalized to real-life situations. First, loud sounds such as in our experiment cannot be found easily in the everyday activities; they could be either found in an occupational setting or come from the personal listening devices with head- or earphones. A study in the USA showed that the average daily noise exposure of adults ranged from 76 to 79 dBA (equivalent noise levels for 8 h exposure) (Flamme et al., 2012). Noise exposure showed little variations across different days of the week but was significantly related to gender and occupation (Flamme et al., 2012). The World Health Organization estimated in 2004 that only 0.057% of male workers and 0.04% of female workers in the USA were occupationally exposed to noise levels of 85–90 dBA (ConchaBarrientos et al., 2004). The EU Directive on the minimum health and safety requirements regarding exposure of workers to the risks arising from noise set the so-called lower exposure action value at 80 dBA (equivalent noise level) for eight hour exposure, thus obliging the workers to wear hearing protection at these minimum noise levels (Directive, 2003/10/EC of the European Parliament and of the Council of 6 February, 2003). None of our volunteers was exposed to

pressure values in men. In women, however, diastolic pressure values after noise were significantly higher than values obtained at the last minute of quiet phase before noise. In women particularly, heart rate and cardiac index values obtained at every minute of quiet phase after noise were significantly lower in comparison to the values obtained at the last minute of quiet phase before noise. The total peripheral resistance index obtained after noise remained significantly higher in comparison to the last minute before noise. Question 6) is there a difference between the quiet phase after noise and noise exposure? Answer: systolic and diastolic pressure values obtained at every minute of the quiet phase after noise were significantly lower in comparison to the values obtained at the last minute of noise exposure in men only. Heart rate and cardiac index values after noise exposure were similar to the values obtained at the last minute of noise exposure. In men particularly, total peripheral resistance index obtained at the last few minutes after noise was significantly lower in comparison to the last minute of noise exposure. To the authors’ knowledge, there are few similar experimental studies exploring the effects of noise on blood pressure. Back in the 1980-ties, a group of Swedish researchers conducted a series of experiments on persons with normal blood pressure and persons with arterial hypertension. In the first case, they exposed young normotensive males to 95 dBA noise for 20 min (Andrén et al., 1982a) or young volunteers of both genders to 100 dBA noise for 10 min (Andrén et al., 1982b). In both instances, researchers reported significant increases in diastolic and mean arterial pressure during noise exposure in comparison to the resting period of 40 dBA background noise, but no changes in systolic pressure, heart rate or cardiac output (Andrén et al., 1982a, b). In the second case, they exposed middle-aged men with mild essential hypertension to 100 dBA noise for 10 min, before and after postsynaptic alpha-adrenoceptor blockade and combined postsynaptic alpha- and non-selective beta-adrenoceptor blockade (Andrén et al., 1983), or before and after postsynaptic beta-selective and non-selective beta-adrenoceptor blockade (Andrén et al., 1981). Noise exposure was followed by a significant increase in systolic, diastolic and mean arterial pressures, but no changes in heart rate or cardiac output (Andrén et al., 1981, 1983). These researchers were the first to assess hemodynamic parameters using the impedance cardiography and to show that the increase in total peripheral vascular resistance was one of the most prominent consequences of noise exposure. This effect was observed in persons with normal blood pressure with heredity for arterial hypertension (Andrén et al., 1982b), as well as in persons with mild hypertension (Andrén et al., 1981, 1983). The increase in blood pressure and total vascular resistance by noise was inhibited by labetalol (alphaand non-selective beta-adrenoceptor blocker) (Andrén et al., 1983), but not by metoprolol (beta-selective adrenoceptor blocker) or propranolol (non-selective beta-adrenoceptor blocker) (Andrén et al., 1981). A decade later, in the 1990-ties, a Japanese researcher conducted two parallel experiments. In the first, he exposed young men with normal blood pressure to intermittent noises of 80 dB, 90 dB and 100 dB lasting for 20 min each and compared them to the steady state of 30 dBA background noise (Sawada, 1993a). In the second, he exposed young men to 100 dB noise for 10 min after they completed one of three stress tests (either a cold pressor test, or an isometric handgrip test, or a mental arithmetic test) (Sawada, 1993b). He demonstrated significant increases in blood pressures and peripheral vascular resistance during noise exposure in both experiments (Sawada, 1993a, b). The experiments described above support the hypothesis that vasoconstriction (an increase in the resistance of blood vessels) is the principal hemodynamic mechanism explaining the rise in blood pressure during noise exposure. Nevertheless, other processes take place simultaneously, such as the decrease of cardiac output (the amount of blood delivered through blood vessels), as reported by both Swedish (Andrén et al., 1983) and Japanese researchers (Sawada, 1993b). In the presented study we demonstrated a decrease in cardiac index and both constituents (stroke volume and heart rate (Opie, 2004)) in women 260

Environmental Research 163 (2018) 249–262

K. Paunović et al.

exposure and the activity of the autonomous nervous system (the socalled sympathovagal balance) (Paunović et al., 2017). Given that noise sensitivity may affect the immune system (Kim et al., 2017) and plays a role in the development of the cardiovascular diseases (HeinonenGuzejev et al., 2007), this trait should be taken into account in noiserelated studies more often. Fourth, we avoided comparing the timelines of events by gender because of the baseline differences in blood pressures and hemodynamic parameters between men and women. Sex hormones, estrogens and testosterone, as well as the renin-angiotensin system and the sympathetic nervous system are known to modulate blood pressure in humans acutely and in the long-term through blood volume, sodium metabolism, the activity of the blood vessels (vasoconstriction or vasodilatation), oxidative stress, etc (Marano and Reckelhoff, 2013). The crude observation of the presented results indicates that blood pressure changes under the given stressful circumstances follow a similar pattern in both genders, thus suggesting that the underlying hemodynamic processes are identical, as well as that they start and cease at roughly the same time during the experiment. The lack of statistical significance in minute-by-minute analysis of the heart rate, cardiac index and peripheral vascular resistance in men but not in women can be explained by the small sample size. Finally, our participants may not reflect the general adult population. Given the young age, average body weight, normal blood pressure, and good general health, we were confident that their cardiovascular reactions to noise would not be biased by vascular, renal or metabolic factors leading to hypertension. Still, we excluded participants with increased blood pressure (possible preexisting hypertension), and made an attempt to eliminate the acute hypertensive effects of smoking, coffee or physical activity before the testing. A similar experiment in patients with arterial hypertension in the future would provide more information on the hemodynamic reactions under stress considering the underlying pathological processes within the cardiovascular system.

occupational noise, but some of them admitted listening to loud sounds during leisure time occasionally. As reported by the Scientific Committee on Emerging and Newly-Identified Health Risks, sound exposure from personal music players ranges from 60 to 120 dBA among regular users; given the estimated the 8-h sound exposure from 75 to 85 dBA, the expected risk of hearing impairment at these sound levels is minimal (SCENIHR, Scientific Committee on Emerging and NewlyIdentified Health Risks, 2008). Second, short-term noise exposure for only ten minutes cannot point to the long-term effects on the cardiovascular system. The presented experiment depicts rapid hemodynamic responses to a single environmental stressor, followed by the compensating processes in the cardiovascular system, but does not accurately imitate real-life situations consisting of repeated short-term exposures to a variety of stressors. We dare not speculate whether the same events occur during community noise exposure, and to what extent blood pressure vary over time. A recent meta-analysis shows that the short-term, mid-term and long-term variability in blood pressure values can be associated with adverse cardiovascular outcomes (Stevens et al., 2016). The discrepancies in the effects of noise on blood pressure in short-term and long-term noise exposure are well documented, implying that laboratory experiments should not be applied to predict long-term consequences of noise exposure in real-life settings (Ising and Michalak, 2004). In a recently published RECORD MultiSensor study, researchers were able to follow the participants in their living environments for seven days, monitoring their noise exposure and heart rate concomitantly (El Aarbaoui et al., 2017). They reported a strong positive association between short-term noise exposure and heart rate variability, an indicator of the activity of the autonomous nervous system (El Aarbaoui et al., 2017). Researchers found that the increased heart rate variability during noise exposure was indicative of an imbalance of the sympathetic and parasympathetic activity, which may lead to the development of cardiovascular diseases (El Aarbaoui et al., 2017). On the other hand, Finnish researchers studied heart rate variability in women in their daily activities and reported that women had lower blood pressure, lower heart rate, and higher heart rate variability while they visited green environments in comparison to the visits to the city center (Lanki et al., 2017). Third, we exposed the participants to the sounds from road traffic they would usually hear in the streets of the city; we were, thus, not able to separate the sources of noise and to take their particular levels or frequencies into account. Low-frequency noise is known to have a more negative impact on heart rate variability than high-frequency noise regardless of the effect on one's blood pressure or cortisol levels (Walker et al., 2016). Similarly, we were unable to consider all the aspects of the exposure to various noise sources, such as annoyance to each particular source (buses, cars, trams), annoyance by transport in general, annoyance by vibrations caused by traffic, annoyance by air pollution emitted by traffic, other emotional reactions to traffic (anger, helplessness), as well as different circumstances of noise exposure (Lercher et al., 2017). In particular, we left behind both noise annoyance and subjective noise sensitivity of our participants from the presented analysis. We reported previously that no more than every fifth woman and every fourth man were highly annoyed by noise during our experiment (Paunović et al., 2014). Knowing that noise annoyance plays a modifying role in the association between noise exposure and the development of hypertension (Babisch et al., 2013) we can only assume that an increase in participants’ annoyance level during the study would have resulted in more prominent changes in blood pressure or hemodynamic parameters induced by noise. This could have been done by involving the participants in a mental task, such as reading or calculating or answering general knowledge questions. Nevertheless, such a task would be challenging to conduct because the participants are lying on the bed with their both arms connected to the device at all times, and, furthermore, we would not be able to separate the effects of noise from the impact of the task. We also reported in the past that noise sensitivity is a moderating factor in the association between noise

5. Conclusions The timeline of blood pressure changes in this experiment shows that systolic and diastolic blood pressures and heart rate increase promptly within the first minute of noise exposure and decrease gradually during and after noise exposure. Constant increases in total peripheral vascular resistance and steady decreases in cardiac index during the experiment may lie beneath the changes in blood pressure. The timeline of events in this 30-min experiment provides insight into the hemodynamic processes underlying the changes of blood pressure before, during and after noise exposure instead of relying on average values obtained from ten-minute intervals. Acknowledgement Authors are grateful to all individuals who volunteered to participate in the study. We would also like to thank the medical staff of the Multidisciplinary Center for the Diagnostics and Treatment of Arterial Hypertension, Clinical Center of Serbia, Belgrade, for their assistance in performing the experiments. Funding sources The study was financially supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia, project No. 175078. Competing interests None to declare. 261

Environmental Research 163 (2018) 249–262

K. Paunović et al.

References

Lanki, T., Siponen, T., Ojala, A., Korpela, K., Pennanen, A., Tiittanen, P., et al., 2017. Acute effects of visits to urban green environments on cardiovascular physiology in women: a field experiment. Environ. Res. 159, 176–185. Lercher, P., De Coensel, B., Dekonink, L., Botteldooren, D., 2017. Community response to multiple sound sources: integrating acoustic and contextual approaches in the analysis. Int. J. Environ. Res. Public Health 14, E663. Marano, R., Reckelhoff, J.F., 2013. Sex and gender differences in control of blood pressure. Clin. Sci. (Lond.) 125, 311–318. Mosteller, R.D., 1987. Simplified calculation of body surface area. N. Engl. J. Med. 317, 1098. Opie, L.H., 2004. Heart Physiology: from Cell to Circulation. Lippincott Williams & Wilkins. Paunovic K., Belojevic G., Stojanov V., Jakovljevic B., 2011. The effect of noise on haemodynamic parameters in young women – a pilot study. In: Proceedings of the ICBEN 2011, the 10th Congress of the International Commission on the Biological Effects of Noise. Noise as a Public Health Problem, London, UK. Paunović K., Jakovljević B., Belojević G., Stojanov V., 2017. Noise sensitivity modifies the effect of recorded noise on sympathovagal balance in young men. In: Proceedings of the 12th ICBEN Congress on Noise as a Public Health Problem, Zurich, June 18-22, 2017. Paper ID 3661. Paunović K., Stojanov V., Jakovljević B., Belojević G., 2013. Hemodynamic and blood pressure changes provoked by recorded traffic noise in normotensive men. In: Proceedings of the INTER-NOISE 2013, the 42nd International Congress and Exposition on Noise Control Engineering, Noise Control for Quality of Life. INTERNOISE, Innsbruck, Austria. Paunović, K., Stojanov, V., Jakovljević, B., Belojević, G., 2014. Thoracic bioelectrical impedance assessment of the hemodynamic reactions to recorded road-traffic noise in young adults. Environ. Res. 129, 52–58. Sawada, Y., 1993a. Hemodynamic effects of short-term noise exposure – comparison of steady state and intermittent noise at several sound pressure levels. Jpn. Circ. J. 57, 862–872. Sawada, Y., 1993b. Reproducible increases in blood pressure during intermittent noise exposure: underlying haemodynamic mechanisms specific to passive coping. Eur. J. Appl. Occup. Physiol. 67 (4), 367–374. SCENIHR – Scientific Committee on Emerging and Newly-Identified Health Risks, 2008. Scientific opinion on the Potential health risks of exposure to noise from personal music players and mobile phones including a music playing function, 23 September 2008. Available at: 〈http://ec.europa.eu/health/ph_risk/committees/04_scenihr/ docs/scenihr_o_018.pdf〉. Stevens, S.L., Wood, S., Koshiaris, C., Law, K., Glasziou, P., Stevens, R.J., et al., 2016. Blood pressure variability and cardiovascular disease: systematic review and metaanalysis. Br. Med. J. 354, i4098. Walker, E., Brammer, A., Cherniack, M.G., Laden, F., Cavallari, J.M., 2016. Cardiovascular and stress responses to short-term noise exposures – A panel study in healthy males. Environ. Res. 150, 391–397. Woltjer, H.H., Bogaard, H.J., deVries, P.M., 1997. The technique of impedance cardiography. Eur. Heart J. 18, 1396–1403.

Andrén, L., Hansson, L., Bjorkman, M., 1981. Hemodynamic effects of noise exposure before and after β1-selective and non-selective β-adrenoceptor blockade in patients with arterial hypertension. Clin. Sci. 61, 89s–91s. Andrén, L., Hansson, L., Eggertsen, R., Hedner, T., Karlberg, B.E., 1983. Circulatory effects of noise. Acta Med. Scand. 213, 31–35. Andrén, L., Lindstedt, G., Bjorkman, M., Borg, K.O., Hansson, L., 1982a. Effects of noise on blood pressure and ‘stress' hormones. Clin. Sci. 62, 137–141. Andrén, L., Piros, S., Hansson, L., Herlitz, H., Jonsson, O., 1982b. Different hemodynamic reaction patterns during noise exposure in normotensive subjects with and without heredity for arterial hypertension. Clin. Sci. 63, 371s–374s. Babisch, W., Pershagen, G., Selander, J., Houthuijs, D., Breugelmans, O., Cadum, E., et al., 2013. Noise annoyance – a modifier of the association between noise level and cardiovascular health? Sci. Total Environ. 452–453, 50–57. Babisch, W., 2002. The noise/stress concept, risk assessment and research needs. Noise Health 4, 1–11. Chang, T.Y., Liu, C.S., Hsieh, H.H., Bao, B.Y., Lai, J.S., 2012. Effects of environmental noise exposure on 24-h ambulatory vascular properties in adults. Environ. Res. 118, 112–117. Chang, T.Y., Su, T.C., Lin, S.Y., Jain, R.M., Chan, C.C., 2007. Effects of occupational noise exposure on 24-h ambulatory vascular properties in male workers. Environ. Health Perspect. 115, 1660–1664. Concha-Barrientos M., Campbell-Lendrum D., Steenland K., 2004. Occupational noise: assessing the burden of disease from work-related hearing impairment at national and local levels. Geneva, World Health Organization. WHO Environmental Burden of Disease Series, No. 9. Directive 2003/10/EC of the European Parliament and of the Council of 6 February 2003. on the minimum health and safety requirements regarding the exposure of workers to the risks arising from physical agents (noise) (Seventeenth individual Directive within the meaning of Article 16(1) of Directive 89/391/EEC). Available at: 〈http:// data.europa.eu/eli/dir/2003/10/2008-12-11〉. El Aarbaoui, T., Meline, J., Brondeel, R., Chaix, B., 2017. Short-term association between personal exposure to noise and heart rate variability: the RECORD MultiSensor Study. Environ. Pollut. 31, 703–711. Flamme, G.A., Stephenson, M.R., Deiters, K., Tatro, A., VanGessel, D., Geda, K., et al., 2012. Typical noise exposure in daily life. Int. J. Audiol. 51, S3–S11. Heinonen-Guzejev, M., Vuorinen, H.S., Mussalo-Rauhamaa, H., Heikkilä, K., Koskenvuo, M., Kaprio, J., 2007. The association of noise sensitivity with coronary heart and cardiovascular mortality among Finnish adults. Sci. Total Environ. 372, 406–412. Ising, H., Babisch, W., Kruppa, B., 1999. Noise-induced endocrine effects and cardiovascular risk. Noise Health 1, 37–48. Ising, H., Michalak, R., 2004. Stress effects of noise in a field experiment in comparison to reactions to short term noise exposure in the laboratory. Noise Health 6, 1–7. Kim, A., Sung, J.H., Bang, J.-H., Cho, S.W., Lee, J., Sim, C.S., 2017. Effects of self-reported sensitivity and road-traffic noise levels on the immune system. PLoS One 12, e0187084. http://dx.doi.org/10.1371/journal.pone.0187084.

262