Can a specific breathing pattern change the autonomic response to a psychological stressor?
Yes has been my experience. You too can directly experience this through using learning to use the breath.
This a recent study from my good friend Aaron Cawley at Bristol University conducted at Bristol university. We had over 40 participants. With the aim to prove that with conscious use of the breath, not only can we induce physiological stress but then soothe our nervous systems response to psychological stress.
Why is this important? It is the same nervous system that perceives our Physical, Mental, Emotional and Social environments. Internally and externally. (Link to future post) Adapting to stress in one area can benefit perception and resilience to stress in another.
This study was done with 15 minutes of Conscious Connected Breathwork (CCB). Inhaling actively through the mouth. Then keeping the breath connected without a pause passively exhaling though the mouth. Connecting in to initiating the next inhale with out having to force or finish the exhale. Punctuated with three Breath retention. Inducing aerobic exercise and nervous system arousal similar to cardio, with out the muscular effort or force on the exhale. The breath retention take the participant to their perceived edge of first for breath. Then release slowly from this heightened state of “threat”. One of the key aspects of this sequence is creating Eustress to the system. Short term intentional stress. This is beneficial for growth and resilience. It is the same body and breath we use to perceive our Physical, Mental, Emotional and Social environments,
Note:
This is stimulating way to use CCB. There are other ways, longer inhale and exhale duration, and breathing through the nose which can have soothing effects to the nervous system (link to future post)
I will be adding my thoughts, expansions and questions to the text.
© Bristol University
This research is opensource. Please cite fairly if used.
1
Can a specific breathing pattern change the autonomic response to a psychological stressor?
Aaron Cawley – Third Year Neursocience Student. Bristol University
2
Contents
List of abbreviations…………………………………………………………………………………………………….3
Abstract……………………………………………………………………………………………………………………….5
- Introduction……………………………………………………………………………………………………..6
1.1. Background…………………………………………………………………………………………..6
1.1.1. Overview of the Autonomic Nervous System…………………………….6
1.1.2. Psychological stress and ANS response……………………………………..6
1.1.3. Pathologies related to dysregulated ANS response……………………7
1.1.4. How patterns of respiration can influence the ANS…………………. 7
1.1.4.1. Respiratory. Sinus arrhythmia & Heart rate variability………………………………7
1.1.4.2. Blood gas reflexes…………………………………………………………………………………….8
1.1.5. Measuring the ANS response to psychological stress…………………9
1.1.5.1. Serial sevens………………………………………………………………….9
1.1.5.2. HRV……………………………………………………………………………….9
1.1.5.3. Infrared thermography………………………………………………….9
1.1.6. Motivation, aims and hypothesis…………………………………….………10
1.1.7. Outline of study……………………………………………………………………….10 - Methods and materials…………………………………………………………………………………..11
2.1. Ethics……………………………………..…………………………………………………………..11
2.2. Recruitment and subjects…………………….……………………………………………..11
2.3. Procedure…………………………………………………………………………………………..11
2.4. Mental arithmetic task………………………………………………………………………..12
2.5. Breathing method……………………………………………………………………………….12
2.6. Skin temperature………………………………………………………………………………..13
2.7. Heart rate variability…………………………………………………………………………..13
2.8. Pilot experiment procedures…………………………………….………………………..13
2.8.1. Thermal imaging pilot study……………………………..……………………..13
2.8.2. Blood Gas normalisation pilot study………………………………………..14
2.9. Data processing and analysis………………………………………………………………15 - Results……………………………………………………………………………………………………………16
3
3.1. The breathing method significantly increased HRV during psychological stress and the increase was sustained after the rest period………………16-19
3.2. Gas levels were also shown to normalise to baseline before 15-minutes post breathing method………………………………………………………………………20-21
3.3. No change in cheek SST was shown in the main study………………………..22
3.4. The forehead did not show to be a more sensitive variable to measure when studying ANS response to psychological stress……………………………..22 - Discussion………………………………………………………………………………………………………25
4.1. The breathing method altered the autonomic response to psychological stress…………………………………………………………………………………………………25-26
4.2. Thermal imaging failed to detect changes in autonomic balance………..26 - General limitations…………………………………………………………………………………………27
- Conclusion……………………………………………………………………………………………………..27
- Areas for future research……………………………………………………………………………….27
- References………………………………………………………………………………………………..28-30
- Appendices…………………………………………………………………………………………………….31
- Acknowledgments………………………………………………………………………………………….32
4
List of abbreviations
ANS – Autonomic nervous system
MAT – Mental arithmetic task
HRV – heart rate variability
NA – Nucleus Ambiguous
LC – Locus coeruleus
PGi – Nucleus paragigantocellularis
CVMNs – Cardiac vagal motor neurons
SARs – slow adapting stretch receptors
MAP – mean arterial pressure
CO – Cardiac output
SST – Skin surface temperature
PIS – Patient information sheet
bpm – beats per minute
RMSSD – Root mean square of successive differences
ANOVA – Analysis of variance
RTN – Retrotrapezoid nucleus
Pa – Partial pressure of a gas
5
Abstract
The autonomic nervous systems (ANS) response to psychological stress is a key factor in the pathogenesis of many conditions including panic and general anxiety disorder (GAD). In this paper forty conscious, and healthy subjects were split into test and control groups. The ANS response to a mental arithmetic task (MAT) was quantified by Heart rate variability (HRV) and Infrared thermography from the cheek. This paper aims to investigate the extent to which conscious manipulation of breathing patterns can alter the ANS response to a psychological stressor. A significant increase in HRV compared to control was observed immediately after the breathing technique was performed compared to before (+21.32 ± 10.03, p<0.05). The HRV increase was sustained after a 15-minute recovery period (+20.50 ± 8.25, P=0.05). The control group showed no significant changes in HRV. Infrared thermography data produced no significant results. The ANS response to psychological stress has been well characterised previously, however the dysregulation of this stress response in conditions such as GAD is still an area of active research. In the context of the ANS response in psychological stress this paper demonstrates that, through the breathing practices discussed, parasympathetic ANS output can be increased during induced psychological stress.
6
Introduction
1.1. Background
1.1.1. Overview of the Autonomic Nervous System (ANS)
The ANS is a complex system that acts largely unconsciously to maintain homeostasis or prepare for an anticipated change in environmental conditions. Neural networks collect sensory information which is integrated in processing areas that coordinate and instigate appropriate physiological and physiological responses. The ANS has two branches, they are both constituently active to some degree, but it is the balance of their activity which dictates physiological and mental states. The sympathetic branch is associated with increased alertness and physiological adaptions which facilitate exertions. The ANS plays a major role in producing the peripheral component of the fear response. The parasympathetic branch produces largely opposing effects in the body (Levenson, 2014).
1.1.2. Psychological stress and ANS response
Hans Selye was an endocrinologist who conducted research on organism’s response to stressors (Tan and Yip, 2018). He proposed the idea that a stresser is a non-specific demand which causes a deviation from a sensory set point. This deviation from the set point promotes a response or a change in rhythms to maintain physiological and psychological homeostasis; psychological homeostasis is defined as a situation where an individual’s desires have been met completely). Therefore, a psychological stressor could be defined as any stimulus that makes an individual’s conscious experience less desirable to them.
The amygdala and the orbitofrontal cortex appear to have pivotal roles in integrating sensory information which is then compared with past and present experiences to coordinate an appropriate physiological response to psychological stress (Baleydier and Maugiere, 1980). Once a threat has been detected the locus coeruleus (LC) is the primary effector and is important in the initiation of the mammalian fear response. For example, LC noradrenaline release produces tachycardia via inhibition of cardiac vagal motor neurons (CVMNs) (Deutch and Charney, 1996).
7
1.1.3. Pathologies related to dysregulated ANS response
Some pathologies involve the ANS response to psychological stress to acute and sometimes specific stimuli or contexts, like panic and general anxiety disorder. A maladaptive change to the threat appraisal process leads to an unnecessarily large sympathetic response. Resulting in tachycardia, hyperventilation, high blood pressure and a sensation of impending doom to varying degrees. Changes in the LC noradrenergic system or its modulation by higher cortical areas is suggested to play a crucial role the pathogenesis of these conditions due to its role in stress-responsiveness and mediation of fear conditioning (Abercrombie et al., 1995).
1.1.4. How patterns of respiration can influence the ANS
There exists a number of brainstem reflexes that alter the output from both branches of the ANS. The sensory information these negative feedback circuits receives, can be changed by voluntarily altering breathing patterns.
1.1.4.1. Respiratory sinus arrhythmia (RSA) & Heart rate variability (HRV)
Heart rate variability (HRV) is the variation in consecutive heartbeat intervals. RSA is a process by which HRV synchronises which respiration. During an inhale slow adapting stretch receptors (SARs) increase their activity, increasing inhibitory signals to inspiratory centres via pulmonary vagal afferents to prevent excessive lung inflation. Some of these inhibitory signals also inhibit CVMNs which remove inhibitory cardiac vagal modulation of the sinus node. This results in an increase in mean arterial pressure (MAP) and cardiac output (CO). Stretch induced tonic inhibition of the vagal output to the heart is decreased during exhalation resulting in bradycardia and a lower MAP (Shykoff et al., 1991). RSA has been suggested to improve pulmonary gas exchange efficiency by matching pulmonary capillary bed perfusion with each respiratory cycle (Hayano et al., 1996). This is also an energy saving mechanism through suppressing dispensable heart beats during the ebb of perfusion.
8
1.1.4.2. Blood gas reflexes The primary role of respiration is the maintenance of partial pressures (PaGas) of O2 and CO2 in the systemic blood system. Hyperventilation increased the removal of CO2 and increases blood O2 saturation. Breath holding allows PaCO2 to build up and O2 saturation to fall (Achenbach-Ng et al., 1994).
The central chemoreflex serve to regulate PaCO2, and was stimulated in artificially ventilated dogs during hyperoxic progressive hypercapnia. An intensification of RSA without an increase in vagal tonic firing was observed (Yasuma and Hayano, 2001). Central chemoreflex stimulation causes an uncoupling of RSA and vagal activity. Central chemoreflex activation increases the central respiratory drive, RSA is intensified in order to further match respiration and pulmonary perfusion to improve efficiency of CO2 expulsion. In the experiment previously mentioned no significant increase in HR or MAP was measured while tidal volume and respiratory rate did.
The peripheral chemoreceptors located in the carotid bodies are most sensitive to decreases in arterial PaO2 and are the most efficient respiratory stimulus. Isocapnic hypoxia was shown to significantly increase HR and MAP, tidal volume and respiratory rate; whilst RSA was markedly reduced. Increased ventilation and CO permit improved delivery of O2 to vital organs for survival. At increased heart rates diastolic filling becomes a limiting factor which makes fluctuations in beat-to-beat intervals disadvantageous for increasing CO (Yasuma and Hayano, 2001).
9
1.1.5. Measuring the ANS response to psychological stress
1.1.5.1. Serial sevens The serial sevens mental arrhythmic task (MAT) induces sustained and consistent psychological stress resulting in a measurable change to subjects’ autonomic profiles towards an increase in sympathetic output and parasympathetic inhibition. The use of serial sevens while measuring heart rate variability is well documented in research (Abdul Wahab and Mat Zin, 2012).
1.1.5.2. HRV
HRV is mainly a manifestation of respiratory sinus arrhythmia which is well established to have linear relations with cardiac vagal outflow in spontaneously breathing anesthetized dogs (Katona and Jih, 1975) and is commonly considered as an index of parasympathetic tone (Eckberg, 1983). An increase in sympathetic outflow markedly reduces heart period oscillations (Taylor et al, 2001), therefore HRV can be used to quantify the sympathetic-parasympathetic balance of cardiac modulation. HRV has been previously used to investigate the ANS response to psychological stress in humans (Schunack, 2008).
1.1.5.3. Infrared thermography Both direct post ganglionic sympathetic innovation of vascular smooth muscle and adrenal medullary release of noradrenaline and adrenaline acting on a1-adrenal receptors result in vasoconstriction of subcutaneous arterioles (Tansey, Roe and Johnson, 2014). The resulting change in skin surface temperature (SST) can be measured as a proxy measurement of the sympathetic autonomic response to psychological stressor. Thermal imaging of skin on the face has already been classified to give information about an individual’s arousal during affective visual stimuli (Nhan and Chau, 2010).
10
1.1.6. Motivation, aims and hypothesis A recent study (Kox et al., 2014) investigated the effects of a training intervention which involved cold exposure, meditation, yoga and breathing exercises. The study provided evidence that trained individuals could voluntarily activate their sympathetic nervous system by performing a learned breathing method. It would be of interest to investigate how the learned breathing method can affect the ANS.
There is an increasing prevalence of anxiety related mental health problems with 15% of the UK population being affected in 2007 compared to 13.3% in 1993 (Halliwell, 2009). Psychotherapy has been shown to help 2 in every 3, but waiting times are between 4 and 61 days depending on the region in the UK. The economic burden is substantial with the NHS planning on spending £13 billion on mental health services in 2019/20 which is 14% of the NHS allocation. More research conducted on the ANS responsiveness to psychological stress could lead to greater understanding of how to prevent, as well as treatment related pathologies.
The aim of this study was to investigate a specific breathing methods effects on the ANS response to psychological stress.
It was hypothesised that the group that performs the breathing exercise would have an altered autonomic response to a psychological compared to the control.
1.1.7. Outline of study
The study presented in this paper investigated one of the breathing methods used in the Kox et al., (2014) study to determine if it had any effect on the ANS response to acute psychological stress in conscious healthy human volunteers. The study procedure conducted, and the observed results are presented. Followed by a discussion around the possible mechanisms by which this breathing method produced its effect.
11 - Methods and materials
2.1. Ethics
This experiment was approved by the ethics committee at the University of Bristol. Ethics application was submitted by Prof. Mark Cannell (ref no: 97142).
2.2. Recruitment and subjects
Participants were recruited through responses to advertisements and being directly contacted by the experimenter. Potential participants were sent a patient information sheet (PIS) and given at least 24 hours to decide if they were willing to sign a consent form. Risk factors were given in the PIS and participants could not continue with the study if they met any predetermined exclusion health criteria (pregnancy, diabetes, epilepsy, cardiovascular or respiratory disease apart from asthma). Subjects with asthma were required to have their inhaler close by during the experiment.
Informed consent documents, health and study questionnaires were completed directly before commencing experiment procedure.
Forty student volunteers from the University of Bristol ranging from 20 to 24 years of age (Male=22 Female=18).
2.3. Procedure
Subjects were asked to arrive at the study having not eaten, smoked or drank alcohol 3 hours before.
Subjects were sent a google sheet via email and asked to choose a time slot for when they could attend the study. Alternating test and control conditions between each participant was used to assign subjects to each group, the random nature of which time slot the subject chose was deemed satisfactory to control for selection bias.
Subjects were tested for one hour individually. Upon subject’s arrival, experimental procedure and instructions for mental arithmetic task were described. The test group was taught the breathing method at this point. The subject then laid in a recumbent position with the upper body propped up at a ~35° angle (figure 1). Measurement devices were set up, after which a 6-minute habituation period started. HRV and SST from the cheek were recorded at baseline and during each 2-minute MAT. As seen on the timeline shown in Figure 2 the MAT was performed directly before and after the
12
conditioning period to see if the breathing method caused any significant changes in autonomic stress response compared to the control. The MAT was also performed after each group had a 15 minute ‘rest period’. During the control group conditioning period and both groups rest periods participants conducted spontaneous breathing. The post rest measurements were used to see if the breathing method caused any lasting changes after subjects had been able to recover (See appendix for timeline).
2.4. Mental arithmetic task (MAT)
In order to induce psychological stress, the participants were asked to subtract serial sevens from a three-digit number for 2 minutes. Participants were not aware performance was not being tested they were asked to count out loud as accurately and rapidly as possible. A recorded metronome (120bpm) was playing during the arithmetic task to induce mild harassment (LaVeau et al., 1989).
2.5. Breathing method The breathing technique involved hyperventilating for an average of 20 breaths (“Ventilation phase”). All breathing was done though the mouth, the inhale was forceful and active whereas the exhale was passive. Subsequently the subject held their breath at functional residual capacity ~1-3 minutes (“retention phase”). The
Figure 1: Showing the position subjects were in during the experiment.
13
retention time was completely dictated by the subject, their instruction was to hold for as long as possible but not to the point of panic. The first inhale at the breath hold breaking point was deep and held for 10s while contracting abdominals, tucking their chin and trying to create pressure in the top of the neck where the spine meets the skull. The ventilation and retention phase were both completed three times.
2.6. Skin temperature A Flir ONE iOS thermal imaging camera was used to measure the SST on the subject’s right cheek. The temperature was taken from a boxed area (figure 2) from the bottom of the right ear to the corner of the mouth to give consistency of measurements despite variation in facial features. The average SST over two minutes was recorded during each 2-minute MAT.
2.7. Heart rate variability (HRV)
Measurements were taken from the right-hand index finger with a finger pulse transducer (FPT). The device connected directly in to a PowerLab Pod Port which displayed a pulse signal in labchart. HRV was calculated by entering interval duration between neighbouring pulse signals into an excel file and the ‘Root mean square of successive differences’ (RMSSD) was used to calculate HRV during baseline and each 2-minute MAT.
2.8. Pilot experiment procedures Recruitment and consent procedures were the same as describes above.
2.8.1 Thermal Imaging pilot study The study (n=8 Female=3, Male=5) compared two groups; data was collected from one group using the same method at the main study. The second group data was the average SST from the forehead (figure 2). This study aimed to see if the breathing method alone caused changed in SST and if the forehead could be a better area to detect changes in the ANS. The timeline and measurement points were the same as the main experiment apart from no MAT was given during measurement periods.
14
2.8.2. Blood Gas normalisation pilot study This experiment was conducted (n=1, male) to investigate if blood gases normalise before the final. End tidal volume was measured using a ML206 Gas analyser and data was collected from Labchart. Assuming perfect gas diffusion results will have been an equivalent approximation of systemic arterial blood gas partial pressures.
Figure 2: The right photo shows the area where average skin temperature was measured from on the cheek. Left photo showing the area measured from the forehead
15
2.9. Data processing and analysis All data sets were first analysed with groups separated, in a one-way repeated measure analysis of variance (ANOVA) to test for significant within group effects. Significant differences were analysed with simple effects analysis using Bonferroni correction for multiple comparisons. Delta-HRV data was then analysed in a three-way factorial repeated measures ANOVA. To investigate as to whether group and sex produced a main effect, and if there was any significant interaction between the two. Simple effects analysis using Bonferroni correction for multiple comparisons was then conducted to see if any changes in HRV were significantly different between groups.
Simple effects analysis between groups were made for delta-HRV at the following levels: 1. pre-conditioning and post-conditioning change (MAT1- MAT2) 2. post-conditioning and post-rest change (MAT2- MAT3) 3. pre-conditioning and post-rest period change (MAT1- MAT3)
Comparing the change in HRV within subjects’ controls for variation between subject absolute values.
Data are presented as mean ± SEM. The tests conducted were two-tailed at an alpha significance level of p<0.05. Analysis was carried out using SPSS version 25.0 (IBM) and data presenting figures were created with Microsoft Excel (Office 2019).
16- Results
Hypothesis: Can a specific breathing pattern change the autonomic response to a psychological stressor?
3.1. The breathing method significantly increased HRV during psychological stress and the increase was sustained after the rest period.
This experiment aimed to see if HRV during a MAT significantly changed due to the conditioning and rest periods for both test and control group. Pulse traces from five female participants were containing too much noise to calculate HRV, therefore results were not included in the study (Control: n=18 Test: n=17). The repeated measured ANVOA showed significant effects of measurement conditions for both control ((1.78,30.33) = 4.88, p<0.05) which was Greenhouse-Geisser corrected for departure from sphericity and the test group which had sphericity assumed ((3,48) = 5.22, p<0.01). Simple effects analysis showed the only significantly different conditions in the control group were baseline and MAT1 (-26.19 ± 8.78, p<0.05). HRV from the MAT2 and MAT3 were not shown to be different from baseline, each other or MAT1. These results suggest a significant decrease in HRV with the first presentation of the MAT, then habituation causes the MAT to have a less significant effect.
Test group simple effects analysis showed a significant decrease in HRV from baseline to MAT1 (-16.4 ± 5.88, p<0.05). HRV significantly increased between MAT1 and MAT2 (+22.40 ± 6.59, p<0.05), as well as MAT1 and MAT3 (+27.32 ± 6.66, p<0.01). No significant change was seen between MAT2 and MAT3. These results suggest a significant decrease in HRV due to the first presentation of the MAT. After the breathing practice HRV increased and there was no significant difference between before and after the rest period. As seen in figure 3 both groups have the same pattern of a decrease from baseline to MAT1, followed by an increase HRV after conditioning and a further increase after the rest period.
17
The changes in HRV between each MAT were then analysed in a 3-way factorial repeated measures ANVOA. This aimed to show if the increases in HRV shown in the test group were due to habituation to the MAT, and therefore were not significantly larger than the increases seen in the control group. Sex was also entered as a factor to see if there was any sex difference in HRV change. ANOVA results showed a significant main effect of group ((1,31) = 8.78, p<0.01), simple effects analysis showed only one significantly higher increase in HRV for the test group at the MAT1-MAT3 comparison (+22.10 ± 8.05, p<0.01). No interaction between sex and group was shown. When looking at figure 4 it was noted that in the test group all participants experienced either no change or an increase in HRV for MAT1-MAT2 AND MAT1-MAT3 apart from one residual (identifiable by a red arrow in figure G) that showed a decrease of 17.20 ms-1. This was identified as a possible anomaly, further investigation revealed a
Figure 3: Clustered bar chart showing the group mean HRV at all testing conditions. Control (Grey bars) and test (Blue bars) group mean heart rate variability shown at Baseline recording at rest, MAT1=preconditioning recording during MAT, MAT2= Post conditioning/Pre-Rest period recording during MAT, MAT3= Post rest period recording during MAT.
18
research diary note that subject number 6 (male) began to uncontrollably hyperventilate during the third ventilation phase, their lips went pale and they did not response to the researcher’s instructions immediately. The residual highlighted to be a possible anomalous result was subject number 6. The highlighted residual was inconsistent with all other residuals showing changes in HRV due to conditioning period, it was also shown to be a result taken from a subject who experienced panic attack like symptoms during the breathing method. There for it was seen as justifiable to remove subject 6’s data and to conduct the same statistical analysis again.
Figure 4: Scatter graph showing the distribution of residuals for changes in HRV between conditions both groups. The residual found to be an anomalous result is made identifiable by the red arrow. (MAT1-MAT2=Change in HRV between immediately pre and post conditioning period, MAT2-MAT3=Change in HRV between immediately pre and post rest period, MAT1-MAT3=Change in HRV between immediately pre-conditioning and post rest period).
19
ANOVA results with the anomaly removed (Control: n=18 Test: n=16) showed a main effect of group ((1,30) = 6.16, p<0.05) and no significant interaction between sex and group. Simple effects analysis showed that increases in HRV were significantly larger in the test group for both MAT1-MAT2 (+21.32 ± 10.03, p<0.05) and MAT1-MAT3 (+20.50 ± 8.25, P=0.05) condition comparisons (figure 5). These results suggest that practicing the breathing method significantly increased the test groups mean HRV during psychological stress immediately after the practice and this increase was sustained for at least 15 minutes. The results also suggest that sex does not have effect on changing HRV because of habituation or the practiced breathing method.
Figure 5: Clustered bar chart showing the mean change in HRV between different stressing periods for control and test group after the identified anomaly was removed. Control and test group mean changes in heart rate variability when comparing different recordings taken during the three MATs. (MAT1-MAT2=Change in HRV between immediately pre and post conditioning period, MAT2-MAT3=Change in HRV between immediately pre and post rest period, MAT1-MAT3=Change in HRV between immediately pre-conditioning and post rest period). Data are presented as mean ± SEM. (*= P≤0.05)
MAT1-MAT2 MAT2-MAT3 MAT1-MAT3
20
3.2. Gas levels were also shown to normalise to baseline before 15-minute post breathing method. This experiment (n=1 male) investigated if the gas levels normalised within 15 minutes after completion of the breathing exercise. As seen in Figure 10, gas end tidal volume fluctuations are as expected with CO2% being low at the end of the ventilation phase and O2% being high, and the opposite being seen at the end of each retention phase. Interestingly CO2% decrease further at the end of each successive ventilation phase and the opposite is true for the O2%. End tidal CO2% returned to normal and slightly increases from baseline by 5 minutes after the last breath retention. O2% returned to baseline sometime between 10 and 15 minutes after finishing the breathing practice.
21
Figure 6: Scatter diagram showing changes in O2 and CO2 end tidal expired gas % compositions during breathing practice and 15-minute recovery time. The plotted lines show the pattern in changing O2% and CO2% starting at Baseline (BL) and over the duration of the breathing exercise (PV1-PR3) and for a 15 minutes rest (PR3-15PE). The dotted lines represent the baseline level. (BL=baseline, PV1=Post 1st ventilation phase (VP), PR1=Post 1st retention phase (RP), PV2 Post 2nd VP, PR2 Post 2nd RP, PV3 Post 3rd VP, PR3 Post 3rd RP, 2PE= 2-minutes post exercise, 5PE=5-minutes post exercise, 10PE=10-minutes post exercise, 15PE=15-minutes post exercise).
22
3.3. No change in cheek SST was shown in the main study.
The main experiment aimed to see if SST on the cheek during a MAT significantly changed due to the conditioning and rest periods for both test and control group. Due to problems managing infrared battery levels some participants did not have thermal data collected (control: n=16, test: n=13). ANOVA results showed no main effects of measurement condition or group (figure 7). There were still no significant effects after removal of subject 6’s data.
3.4. The forehead did not show to be a more sensitive variable to measure when studying ANS response to psychological stress.
The pilot study (forehead: n=4, cheek: n=4). aimed to enquire as to whether the breathing method and a rest period caused ant changes in SST alone and if forehead SST is a more sensitive variable. ANOVA results showed no main effects of measurement condition or main effect of group (figure 8).
These results suggest that the breathing method used in this study does not produce a significant change in SST in either the forehead or the cheek or the measurements taken were not sensitive enough to detect a change.
23
Figure 7: Clustered bar chart showing a group mean temperatures at different experimental conditions. The bars (Blue=Control Grey=Test) show the trend in SST change between baseline, and all three MATs completed during the experiment. Data are expressed as mean ± SEM.
24
Figure 8: Clustered bar chart showing a group mean temperatures at different experimental conditions. The bars (Blue=Forehead Grey=Cheek) show the trend in SST change at baseline, post completion of breathing practice and post 15-minute quiet rest. Data are expressed as mean ± SEM.
25- Discussion
4.1. The major finding in this study was that the breathing method altered the autonomic response to psychological stress. The findings presented in this study suggested that practicing the breathing method increased HRV during psychological stress. As explained in the introduction an increased HRV can be considered an indicator that ANS modulation of the heart has shifted further towards a dominant parasympathetic balance. These results are surprising because the same breathing method in a previous study provided evidence the breathing method could produce activation of the sympathetic nervous system (Kox et al., 2014).
The pilot study gave evidence that end tidal volume blood gases were normalised before MAT 3, suggesting that the change in HRV was due to a neurophysiological adaption. There is evidence for a complex network that includes inputs from structures involved with respiratory related reflexes (Smith et al., 2010). The most comprehensive current model posits a chemo sensitive region in the retrotrapezoid nucleus (RTN) to have a putative roll in integrating information from this network to modulate the sensitivity of central chemoreceptors (Smith et al., 1989; Guyenet, Stornetta and Bayliss, 2008). The ventilation and retention phases produce alterations in the information relayed to the RTN about MAP, lung inflation and systemic blood gas levels in a fluctuating fashion from one extreme to the other. It could be speculated that these large changes in sensory information produce an increase in the central chemoreflex gain.
The study in which this breathing method was originally used showed a gradual successive decrease in blood O2 saturation levels at the end of each retention phase (Kox et al., 2014) which could be due to repeated acute periods of hypoxia which has been shown to desensitize the peripheral chemoreflex (Bascom et al., 1990) Decreasing peripheral chemoreceptor stimulation has been shown to desensitize the central chemoreflex to changes in partial pressure of (Pa)CO2 (Blain et al., 2010). One might expect hypoxia induced desensitization of peripheral chemoreceptors to increase end retention phase PaCO2. Yet the opposite is true at the end of each breath hold the PaCO2 decreases which could be attributed to an increased central
26
chemoreflex gain (Kox et al., 2014). These speculations must be made with caution as these measurements are taken from individuals who had undergone an intense multimodal training program, however the pattern of gas changes in the study aligns with results gathered from the pilot study conducted in this paper.
Increasing the central chemoreflex gain increases sensitivity to rising PaCO2 during exhalations (Dempsey et al., 2012). Stimulation of the central chemoreflex has shown to increase RSA with a decoupling of HRV and vagal tone (Yasuma and Hayano, 2001). This effect is accentuated by the removal of peripheral chemoreflex RSA inhibition during exhalation. It could be further speculated that any reduction in HRV due to higher cortical areas inhibiting CVMNs in response to psychological stress could be counteracted by an increase in the central chemoreflex gain produced by performing the specific breathing method mentioned in this study.
4.2. Thermal imaging failed to detect changes in autonomic balance
These results are not consistent with the study hypothesis. This suggests that the methods used in this experiment were not sensitive enough, or there was no detectable change to SST on the cheek or forehead in response to psychological stress. There is evidence to suggest facial SST can be used to measure changes in autonomic balance, but it is hard to determine the direction of change due to the complexity and lack of understanding that varying emotions due to subjective cognitive appraisal has on facial SST changes.
27 - General limitations The major limitation of this study is HRV recorded for 2 minutes may give limited information on the state of the ANS in response to stress. HRV consists of high and low frequency bands. By measuring the average HRV for a 2-minute period both of these frequency bands are recorded together. The two bands provide separate metrics for sympathetic (high frequency) and parasympathetic (low frequency) function. Without the analysis of HRV frequency components it raises the risk of leading to wrong conclusions on what your results say about the ANS state. Spectral analysis of HRV would allow the analysis of different frequency bands analysis. Typically, short term HRV is recorded for 5 minutes this would supply more robust results.
- Conclusion
This study has provided evidence that by manipulation of breathing patterns the ANS response to psychological stress can be consciously altered towards being proportionally more parasympathetic. Therefore, this study has successfully highlighted conscious manipulation of the breathing rhythms to be an area of interest, for further study into preventative and therapeutic interventions for ANS responsivity dysregulation related pathologies. - Areas for future research
Further research to show the effects of this breathing method on the gain of the reflexes involved with respiration could provide evidence to support the findings in this paper while furthering our understanding of how different breathing methods can modulate the ANS output.
For future thermal imaging of the face the periorbital and supraorbital vessels which feed the area around the eyes have been shown to produce reliable increases in temperature in response to psychological stress (Levine, Pavlidis and Cooper, 2001).
28 - References
Abercrombie ED, Zigmond MJ. Modification of central catecholaminergic systems by stress and injury: functional significance and clinical implications. In: Bloom FE, Kupfer DJ, eds. Phychpharmacology: The fourth Generation of Progress. New York, NY: Raven Press; 1995:355-363
Achenbach-Ng, J., Siao, T., Mavroudakis, N., Chiappa, K. and Kiers, L. (1994). Effects of Routine Hyperventilation on PCO2 and PO2 in Normal Subjects. Journal of Clinical Neurophysiology, 11(2), pp.220-225.
BALEYDIER, C. and MAUGUIERE, F. (1980). THE DUALITY OF THE CINGULATE GYRUS IN MONKEY. Brain, 103(3), pp.525-554.
Bascom, D., Clement, I., Cunningham, D., Painter, R. and Robbins, P. (1990). Changes in peripheral chemoreflex sensitivity during sustained, isocapnic hypoxia. Respiration Physiology, 82(2), pp.161-176.
Blain, G., Smith, C., Henderson, K. and Dempsey, J. (2010). Peripheral chemoreceptors determine the respiratory sensitivity of central chemoreceptors to CO2. The Journal of Physiology, 588(13), pp.2455-2471.
Deutch, A. and Charney, D., 1996. A Functional Neuroanatomy of Anxiety and Fear: Implications for the Pathophysiology and Treatment of Anxiety Disorders. Critical Reviews™ in Neurobiology, 10(3-4), pp.419-446.
Dempsey, J., Smith, C., Blain, G., Xie, A., Gong, Y. and Teodorescu, M. (2012). Role of Central/Peripheral Chemoreceptors and Their Interdependence in the Pathophysiology of Sleep Apnea. Advances in Experimental Medicine and Biology, pp.343-349.
Fritsch, J., Eckberg, D., Graves, L. and Wallin, B. (1986). Arterial pressure ramps provoke linear increases of heart period in humans. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 251(6), pp.R1086-R1090.
Guyenet, P., Stornetta, R. and Bayliss, D. (2008). Retrotrapezoid nucleus and central chemoreception. The Journal of Physiology, 586(8), pp.2043-2048.
Halliwell, E. (2009). In the face of fear. Mental Health Foundation.
Hayano, J., Yasuma, F., Okada, A., Mukai, S. and Fujinami, T. (1996). Respiratory Sinus Arrhythmia. Circulation, 94(4), pp.842-847.
Kox, M., van Eijk, L., Zwaag, J., van den Wildenberg, J., Sweep, F., van der Hoeven, J. and Pickkers, P. (2014). 0026. Voluntary activation of the sympathetic nervous system and attenuation of the innate immune response in humans. Intensive Care Medicine Experimental, 2(Suppl 1), p.O2.
29
Katona, P. and Jih, F. (1975). Respiratory sinus arrhythmia: noninvasive measure of parasympathetic cardiac control. Journal of Applied Physiology, 39(5), pp.801-805.
LaVeau, P., Rozanski, A., Krantz, D., Cornell, C., Cattanach, L., Zaret, B. and Wackers, F. (1989). Transient left ventricular dysfunction during provocative mental stress in patients with coronary artery disease. American Heart Journal, 118(1), pp.1-8.
Levine, J., Pavlidis, I. and Cooper, M. (2001). The face of fear. The Lancet, 357(9270), p.1757.
Lang, C. and Heckmann, J. (2005). Apnea testing for the diagnosis of brain death. Acta Neurologica Scandinavica, 112(6), pp.358-369.
Nhan, B. and Chau, T. (2010). Classifying Affective States Using Thermal Infrared Imaging of the Human Face. IEEE Transactions on Biomedical Engineering, 57(4), pp.979-987.
Purwandini Sutarto, A., Abdul Wahab, M. and Mat Zin, N. (2012). Resonant Breathing Biofeedback Training for Stress Reduction Among Manufacturing Operators. International Journal of Occupational Safety and Ergonomics, 18(4), pp.549-561.
Smith, C., Forster, H., Blain, G. and Dempsey, J. (2010). An interdependent model of central/peripheral chemoreception: Evidence and implications for ventilatory control. Respiratory Physiology & Neurobiology, 173(3), pp.288-297.
Smith, J., Morrison, D., Ellenberger, H., Otto, M. and Feldman, J., 1989. Brainstem projections to the major respiratory neuron populations in the medulla of the cat. The Journal of Comparative Neurology, 281(1), pp.69-96.
Levenson, R. (2014). The Autonomic Nervous System and Emotion. Emotion Review, 6(2), pp.100-112.
Roe, S., Johnson, C. and Tansey, E., 2014. Investigation of physiological properties of nerves and muscles using electromyography. Advances in Physiology Education, 38(4), pp.348-354.
Schunack, C. (2008). Adaptability of Baroreflexsensitivity and Heart Rate Variability to Noise and Mental Stress. Klinische Neurophysiologie, 39(01).
Shykoff, B., Naqvi, S., Menon, A. and Slutsky, A. (1991). Respiratory sinus arrhythmia in dogs. Effects of phasic afferents and chemostimulation. Journal of Clinical Investigation, 87(5), pp.1621-1627.
Taylor, J., Myers, C., Halliwill, J., Seidel, H. and Eckberg, D. (2001). Sympathetic restraint of respiratory sinus arrhythmia: implications for vagal-cardiac tone
30
assessment in humans. American Journal of Physiology-Heart and Circulatory Physiology, 280(6), pp.H2804-H2814.
Tan, S. and Yip, A. (2018). Hans Selye (1907–1982): Founder of the stress theory. Singapore Medical Journal, 59(4), pp.170-171.
Yasuma, F. and Hayano, J. (2001). Augmentation of respiratory sinus arrhythmia in response to progressive hypercapnia in conscious dogs. American Journal of Physiology-Heart and Circulatory Physiology, 280(5), pp.H2336-H2341.
31 - Appendices
32 - Acknowledgments
Prof. Mark Cannell for his expertise in biomedical research and his direct and necessary guidance. Dr Peter Brennan for always being willing to help, especially with the statistical side of things. Jeremy Grattan-Kane for the inspiration and tools that helped create this research. Mr Dave Gee and Steve for all the help setting up the lab. Last but not least, thank you to all of the student participants who selflessly gave their time and energy to help make this research happen.