DIFFERENTIAL EFFECTS OF SELF - PACED AND DEVICE - GUIDED SLOW DEEP BREATHING ON PHYSIOLOGICAL OUTCOMES By Kevin Lawrence Kelly A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Physiology Masters of Science 2015 ABSTRACT DIFFERENTIAL EFFECTS OF SELF - PACED AND DEVICE - GUIDED SLOW DEEP BREATHING By Kevin Lawrence Kelly Slow breathing exercises have been used for thousands of years in practices such as yoga and medit ation, and recent scientific studies demonstrate that they do reduce stres s and lower blood pressure. Yet r ecently, device - guided breathing maneuvers that seek to replicate self - paced breathing practices have raised doubts about the effectiveness of slow breathing techniques . It is hypothesized that the act of reducing breathing rate is not the only factor contributing to stress and blood pressure reduction, and that other factors such as devices themselves and the baseline personality traits and state of mind at the time of the maneuver are also important in determining the effectiveness. The present study found that changes in blood flow to the forearm, heart rate, and blood pressure were affected differently by device - guided maneuvers and self - paced maneuvers. In addition, several psych ological metrics are predictive of different physiological respo nses to each of these practices. ABSTRACT DIFFERENTIAL EFFECTS OF SELF - PACED AND DEVICE - GUIDED SLOW DEEP BREATHING By Kevin Lawrence Kelly Slow breathing exercises have been used for thousands of years in yoga and meditation practices, and have been successful in reducing sympathetic nerve activity and lowering blood pressure. Recently, computerized devices aimed to replicate self - guided bre athing have been used. There have been mixed results regarding the efficacy of these devices. It is hypothesized that the act of reducing breathing rate is not the only factor contributing to physiological changes, and that factors such as external aid and psychological metrics are also important in determining the response s to breathing rate changes. Therefore we compared device - guided (n=10) and self - paced (n=11) slow breathi ng maneuvers in young healthy males and found differences between conditions in forearm vasodilation as measured by venous occlusion plethysmography ( 111.78±7.82% and 96.24±5.82%, p=0.067) and changes in systolic blood pressure as measured by finometry ( - 3.47±1.43 mmHg and - 0.45±2.35 mmHg , p=0.138 ). In addition, several psychological metrics are predictive of differences in physiological respo nses to each of these practices: trait anxiety correlates with a lack of reduction in mean arterial pressure ( MAP ) in the self - guided intervention (r=0.452, p=0.81); acti ng without awareness correlates with a lack of reductio n in MAP in the self - guided intervention (r=0.683, p=0.01); and acting without awareness correlates with a reduction in MAP in the device - guided intervention (r= - 0.472, p=0.084). It is concluded that these two breathing practices are inherently different breathing maneuvers, given the differences in physiological outcomes and psychological predictors. iii To those who have supported me iv ACKNOWLEDGEMENTS I would like to thank Dr. Wehr wein for helping me through my M every opportunity to be successfu l. I also would like to thank my committee members, Dr. Kreulen, Dr. Parameswaran, and Dr. Root - Bernstein f or their input and support throughout the process of the research presented within. Finally, I want to thank the department as a whole for providing me with a quality education over the course of both my M B degrees. v TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ................. v i i LIST OF FIGURES ................................ ................................ ................................ ............... v i KEY TO ABBREVIATIONS ................................ ................................ ................................ ix LITERATURE REVIEW ................................ ................................ ................................ ...... 1 Breathing Practices ................................ ................................ ................................ ........... 1 Yoga and self - paced breathing ................................ ................................ ................... 1 D evice - guided breathing ................................ ................................ ............................ 2 Physiological Mechanisms of Respiratory Control ................................ .......................... 3 Central control of respiration ................................ ................................ ..................... 3 Respiratory - sympathetic c oupling ................................ ................................ ............. 4 Psychological Correlates to Physiological Outcomes ................................ ....................... 6 Specific Aim I ................................ ................................ ................................ ................... 7 Specific Aim II ................................ ................................ ................................ .................. 8 CURRENT STUDY ................................ ................................ ................................ ............... 11 Methods ................................ ................................ ................................ ............................. 11 Subject inclusion/exclusion criteria ................................ ................................ ............ 11 IRB approval ................................ ................................ ................................ ............... 12 Software/hardware ................................ ................................ ................................ ...... 12 Electrocardiography (ECG) ................................ ................................ ........................ 12 Pneumobelt ................................ ................................ ................................ ................. 12 Finometer ................................ ................................ ................................ .................... 13 Venous occlusion plethysmography ................................ ................................ ........... 13 Study conditions ................................ ................................ ................................ .......... 14 Psychological surveys ................................ ................................ ................................ . 14 Results ................................ ................................ ................................ ............................... 18 Breathing Manipulation ................................ ................................ .............................. 18 Specific Aim I ................................ ................................ ................................ ............. 19 Specific Aim II ................................ ................................ ................................ ............ 21 Discussion ................................ ................................ ................................ ......................... 36 Conclusion ................................ ................................ ................................ ........................ 41 REFERENCES ................................ ................................ ................................ ...................... 42 vi LIST OF TABLES Table 1: Demographics ................................ ................................ ................................ .......... 11 vii LIST OF FIGURES Figure 1: Central Respiratory Control Diagram ................................ ................................ .... 9 F igure 2: Timeline for Studies ................................ ................................ ............................... 1 6 Figure 3 : Respiratory Rate ................................ ................................ ................................ ..... 23 Figure 4 : Tidal Volume ................................ ................................ ................................ .......... 24 Figure 5 : Ventilation ................................ ................................ ................................ .............. 25 Figure 6 : Systolic Blood Pressure ................................ ................................ .......................... 26 Figure 7 : Diastolic Blood Pressure ................................ ................................ ........................ 27 Figure 8 : Mean Arterial Blo od Pressure ................................ ................................ ................ 28 Figure 9 : Forearm Vasodilation ................................ ................................ ............................. 29 Figure 10 : Heart Rate ................................ ................................ ................................ ............. 30 Figure 11 : State Anxiety ................................ ................................ ................................ ........ 31 Figure 12: Trait Anxiety ................................ ................................ ................................ ........ 32 Figure 13: Cognitive Failure ................................ ................................ ................................ .. 33 Figure 14: Observational Mindfulness ................................ ................................ ................... 34 Figur e 15: Acting Without Awareness ................................ ................................ .................. 35 viii KEY TO ABBREVIATIONS AHA American Heart Association BMI Body Mass Index CFQ Cognitive Failure Questionnaire CVLM Caudal Ventrolateral Medulla DBP Diastolic Blood Pressure DG Device - Guided ECG Electrocardiogram IRB Institutional Review Board FFMQ Five - Faceted Mindfulness Questionnaire MAP Mean Arterial Pressure MSNA Muscle Sympathetic Nerve Activity NTS Nucelus Tractus Solitarius pFRG Parafacial Respiratory Group PreBotC PreBotzinger Complex RSA Respiratory Sinus Arrhythmia RTN Re trotrapezoidal Nucleus RVLM Rostroventrolateral Medulla SBP Systolic Blood Pressure SDB Slow Deep Breathing SE Standard Error Self Self - Guided Breathing ix STAI - S State Trait Anxiety Inventory State STAI - T State Trait Anxiety Inventory Trait V E To tal Ventilation VOP Venous Occlusion Plethysmography VRG Ventral Respiratory Group 1 LITERATURE REVIEW Slow - paced breathing is a respiratory maneuver , utilized during relaxation techniques and meditative practices, which claim s to promote a reduction in stress. Given its long history of use in yoga and meditation , there is an abundance of anecdotal evidence relating to stress reduction associated with slow breathing . W ith the potential medical applications of slow - paced breathing , such as a reduction in blood pressure in hypertensive patients , it is important to give these p ractices a scientific framework. This will lead to a better understand ing of slow - paced breathing viability as a method to reduce stress and how to best advise patients to achieve the maximum benefit from these techniques. Breathing Practices Yoga and self - paced breathing Slow - breathing is an anchor in yoga - physical changes in body posture. Due to the positive anecdotal evidence surrounding yoga and meditative slow breathing, several studies have focused on determini ng efficacy of slow breathing on reducing sympathetic nervous system activity and blood pressure. It has been shown that three months of regular slow breathing practice reduces sympathetic nerve activity and inc rease parasympathetic nerve activity (Pal, Velkumary et al. 2004) . S everal studies have demonstrated a reduction in blood pressure in response to a variety of yoga slow breathing techniques (Sundar, Agrawal et al. 1984, McCaffrey, Ruknui et al. 2005, Raghuraj and Telles 2008, Upadhyay Dhungel, Malhotra et al. 2008, Bhavanani, Madanmohan et al. 2012) . 2 Device - guided breathing Given the positive benefits noted above, the medical device industry has attempted to extract and standardize the breathing aspect of these practices by inventing devices that coach patients through a slow breathing practice with tones and monitoring. Recently , a study concluded that device - guided breathing practices were not effective at reducing blood pressure aft er an eight week period in white, hypertensive male and female patients (Landman, Drion et al. 2013) . Although the conclusions of that study have been challenged due to low sample sizes and potentially unsuccessful randomization (Huang a nd Subak 2014) . Several o ther device - guided studies have also supported the idea that slow deep breathing is not feasible as a method to reduce blood pressure (Altena, Kleefstra et al. 2009, Anderson, McNeely et al. 2010, Hering, Kucharska et al. 2013) . On the other hand, many groups have shown a reduction in blood pressure using device - guided practices (Grossman, Grossman et al. 2001, Schein, Gavish et al. 2001, Viskoper, Shapira et al. 2003, Elliot, Izzo et al. 2004, Meles, Giannattasio et al. 2004, Schein, Gavish et al. 2009, Oneda, Ortega et al. 2010, Bertisch, Schomer et al. 2011) . Similarly, several studies have concluded that muscle sympathetic nerve activity (MSNA) was si gnificantly lower after an acute use of device - guided breathing (Hering, Kucharska et al. 2013, Harada, Asanoi et al. 2014) , although this was not seen in chronic use (Hering, Kucharska et al. 2013) . Despite the evidence that self - paced breathing exercises, such as yoga breathing, lowers blood press ure and despite mixed results from device - guided studies, t he American Heart Association (AHA) recently issued a position statement providing moderate support to device - guided breathing while withholding support of yoga breathing (Brook, Appel et al. 2013) . Notably, however, resistance to supporting yoga breathing is based on a lack of good studies and not to 3 negative findings, per se. Rather than working from data that does not take into account all of the variables , it is important to perform well - controlled studi es on slow breathing practices to determine whether there are significant differences between these two slow deep breathing practices and their effects on the cardiovascular system. Physiological Mechanisms of Respiratory Control In order to carry out useful studies of the effects of controlled breathing on heart - related functions, it is necessary to understand some of the key mechanisms by which the two physiological systems interact. Central control of respiration See figure 1. Inspiration and expiration involve two areas in the medulla that rhythmically generate motor efferent activity to muscles involved in respiration , the Pre - Botzinger Complex (P reBotC) and the parafacial respiratory group/retrotrapezoidal nucleus (pFRG/RTN) . Th ese two areas in the brainstem P reBotC and pFRG/RTN are phase - locked, or synchronized, through reciprocal excitatory coupling to ensure proper breathing patterns, or eupnoea. (Mellen and Thoby - Brisson 2012) . The P reBotC is involved in generating i nspiration . This was established by observing that PreBotC activity is coupled to respiratory patter n s (Smith, Greer et al. 1990, Smith, Ellenberger et al. 1991, Richter and Spyer 2001, Ramirez, Zuperku et al. 2002) and lesion studies s howing that destruction of the P reBotC leads to ataxic breathing patterns (Gray, Janczewski et al. 2001) . While much of the rhythmicity of the P reBotC is still unclear and debated, it is agreed upon that about half of this area in the medulla is comprised of pa cemaker neurons that initiate the 4 inspiratory patt erning. From there, inspiratory - related neural activity is relayed to premotor neurons in the rostral ventrolateral medulla (RVLM) , which continues to project to the phrenic nerve (diaphragm) and intercosta l nerves (external intercostal muscle s) to cause contraction of these muscles and thus inhalation (Feldman and Del Negro 2006) . Under rest or light activity, the lungs will deflate through passive recoil of the ribcage and diaphragm. Under conditions that require increased ventilation or conscious breathing, active expiration will arise from a second area in the medulla, the pFRG/R TN . The pFRG/RTN also exhib its rhythmicity similar to the P reBotC, but is lesser known due to its recent identification (Mellen, Janczewski et al. 2003, Feldman and Del Negro 2006, Janczewski and Feldman 2006) . Activity in the pFRG/RTN is relayed to the central ventrolateral medulla, before co ntinuing on to the intercostal nerves, which inn ervate the internal intercostal muscles as well as the abdominal muscles. Contraction of b oth of these muscle groups result s in active expiration. Respiratory - sympathetic coupling The central respiratory neural network is anatomically and physiologically linked to autonomic outflow to the rest of the body. The most obvious case of this is evident in variations in heart rate that align with breathing patterns, called respiratory sinus arrhythmia (RSA). RSA causes heart rate to increase during periods of inspiration and decrease during periods of expiration, allowing for more efficient gas exchange. This is modulated via changes in parasympathetic cardiac vagal outflow, causing an increase in vagal tone durin g expiration and a withdrawal during inhalation , controlled by pulmonary stretch afferents and the Hering - Breuer reflex (Yasuma and Hayano 2004) . Less evidently, the central respiratory neural network outflow. To be more specific, the sympathetic nervous system is more active during the 5 inspiratory expiratory phase. An example of this was shown cleverly through stimu lation of the baroreceptors by applying a baroreflex - triggering procedure of external neck pressure or suction during different phases of the respiratory cycle and observing that blood pressure responses, and therefore sympathetic responses, were greatest during inspiratory phases (Eckberg and Orshan 1977, Eckberg, Kifle et al. 1980, Wallin and Eckberg 1982, Eckberg 2003) . This respiratory observing that MSNA quiescence occurs during exhalation (Mozer, Fadel et al. 2014) . Beyond the acute effects of respiratory - sympathetic coupling on physiological outcomes mentioned earlier, there are also several relevant chronic effects. It has been shown that, when broken in tertiles based on breathing rate, those who breathe at a faste r rate also have a higher MSNA burst incidence (Narkiewicz, van de Borne et al. 2006) . It has also been shown that higher breathing rate s are correlated with an increase in total peripheral resistance in young males (Charkoudian, Gusman et al. 2010) . The anatomical basis for the above relationships is based in central connections between the respiratory and autonomic neurons. The nucleus tractus solitarius (NTS) is a structure in the dorsolateral medulla, and is considered an integrative center for various processes related to the sympathetic nervous system, including baroreflex and chemoreflex control. The NTS can be regulator of sympathetic outflow through direct projections to the RVLM (Zoccal, Furuya et al. 2014) . The RVLM, found in the medulla, houses presympathetic neurons, which are the point of origin for excitatory inputs to the preganglionic sympathetic neurons that are responsible for controlling arterial pressure (Guertzenstein and Silver 1974, Ross, Ruggiero et al. 1984) . 6 The ventral respiratory group neurons (VRG, which is where the P reBotc and BotC are located) are intermingled with the RVLM (Moraes, Bonagamba et al. 2011) . An anatomical link between these two areas was established when it was discovered that some axons from the BotC lied close to presympathetic neurons in the RVLM , suggesting communication between these two areas (Sun, Minson et al. 1997) . Even further, catecholeminergic neurons, or C1 neurons, in the RVLM were shown to be phase - locked with the re spiratory cycle (Moraes, da Silva et al. 2013) . Psychological Correlates to Physiological Outcomes There are many psychological metrics are used in the assessment of various traits related to personality and higher levels of brain functioning. In the current study, we focused on metrics that were deemed to be pertinent to assessing differences between conditions. These included the State - Trait Anxiety Inventory (STAI), Five Facet Mindfulness Questionnaire (FFMQ), Cognitive Interference Questionnaire (CIQ), Penn State Worry Questionnaire (PSWQ), and Cogni tive Failures Questionnaire (CFQ). The STAI is broken down into two sections, both of which assess anxiety levels within the participant. The first section , STAI - state, looks at how anxious the participant is in the current moment, and is capable of chang ing rapidly in response to new events. The second, STAI - trait, assesses the anxiety levels as a personality trait, meaning that it is stable through time and is subject to little, change, if any. The FFMQ assesses at various aspects of mindfulness. It is b roken down into five categories: observational, or how aware the subject is of your body and your surroundings; descriptive, or how well the subject is capable of articulating their feelings and thoughts; acting with(out) awareness, or how unaware the subj ect is of their actions both before and during their completion; non - judgemental, or how self - critical the subject is of their 7 thoughts and feelings; and non - reactionary, or how capable the subject is at stepping away from their emotions and feelings witho ut becoming overwhelmed by them. Finally, the CIQ measures the relative amount that the subject thought of other topics during the the participant worries and how much worry affects their lives. The CFQ measures the relative frequency at which the participant forgets what they are doing, how frequently they fail to notice their surroundings. There is an increasing amount of evidence that points to the involvement of the state of mind on p hysiological outcomes . It has been shown that a high level of trait mindfulness is related to a reduction in blood pressure, lower IL - 6 levels (Tomfohr, Pung et al. 2015) , lower cortisol levels (Brown, Weinstein et al. 2012, Jacobs, Shaver et al. 2013) , and lower activation of the amygdala in response to stressors (Creswell, Way et al. 2007) . Taken together, these studies indicate that a Another psychological trait potentially linked to autonomic activity is cogni tive interference. It has been sh own that a decrease in the amount of attention that can be spared, or an increase in the amount of cognitive interference occurring, results in an increase in blood pressure in response to negative, stressful images (Okon - Singer, Mehnert et al. 2014) . Specific Aim I To determine if device - guided and self - guided breathing result in different physiologic al outcomes. W e hypothesize that self - guided breathing and device - guided breathing are 8 fundamentally different practices, and thus will result in differing degrees of physiological responses , such as reduction in blood pressure. The results from pursuing this aim will help better understand if it is appropriate to use devices to assist in slow breathing maneuvers. This is especially relevant for clinicians and choosing the best intervention for their patients. Specific Aim II To determine if psychological metrics are correlated to a reduction in blood pressure during slow pace breathing. Psychological traits such as mindfulness and attention are important in modulating the autonomic nervous system and blood pressure. We hypothesize that various levels of m indfulness , cognitive interference and related psychological metrics will be correlated with the degree to which blood pressure is affected in both device and self - guided slow pace breathing practices. These results will be important in predicting a person - pace d breathing maneuvers based on their psychological traits. Further, it may also explain the mixed results found with device - guided studies. 9 Figure 1: Central Respiratory Control Diagram. 10 Figure 1: Central Respiratory Control Diagram . neurons. Inspiratory neurons project to the Rostroventrolateral Medulla (RVLM), which relays inspiratory signals to the diaph ragm and external intercostal muscles via the phrenic and inter costal nerves, respectively, to cause inhalation. The Retrotrapezoid Nucleus/ Parafacial Respiratory Group (RTN/pFRG) houses expiratory pacemaker neurons that function under active expiration situations. The expiratory neurons project to the Caudal Ventrol ateral Medulla, which relays exhalation signals to the abdominal and internal intercostal muscles, triggering exhalation. The RTN/pFRG and PreBotC reciprocally excite each another, leading to phase locking and ensuring proper breathing patters, or eupnoea. The Nucleus Tractus Solitarius (NTS) receives input from baroreceptors and chemoreceptors. The se input s are then translated to sympathetic nerve activity increases or de creases through the RVLM. The - sympathetic coupling. 11 CURRENT STUDY Methods Subject inclusion/exclusion Participants in this study were young healthy male s , 18 - 25 years of age, non - obese (BMI<30), non - hypertensive (BP <140/90) , moderately active (less than 60 min, 3 times per week) , and yoga/meditation naï ve. Participants could not be smokers or have any chronic diseases related to the cardiovascular or respiratory systems, n or be taking any medication that could potentially alter cardiovascular function. On the day of the study, participants were asked to refrain from caffeine or alcohol in the past 6 hours and food in the past 3 hours, and intense exercise 24 hours . Prior to arrival, participants were randomly sorted into either device - guided or self - guided breathing groups. Participants were screened and consented prior to enrollment. Ta ble 1: Demographics. Demographics Device (n=10) ±SE Self (n=11) ±SE P - value Age 20.01 ±0.4 20.17 ±0.43 0.78 BMI 23.92 ±0.96 24.59 ±0.77 0.59 Heart Rate 60.2 ±2.71 63.18 ±3.30 0.5 Systolic 117.46 ±1.91 121.3 ±2.58 0.26 Diastolic 67.41 ±1.45 65.2 ±2.77 0.5 Breathing Rate 15.39 ±1.28 14.35 ±1.03 0.53 12 IRB approval Board, IRB approval #: 14 - 691. All participants read and signed a consent form outlining their rights as a participant. Software/ h ardware All physiological data was collected and analyzed using Labchart (version 7.3.7, ADInstruments) in conjunction with Powerlab 8/35 (ADInstruments). The inductotrace, finometer, and plethysmograph (see below) were all connected to the Powerlab unit with BNC cables. ECG was connected using a BNC cable allowing communication between the main unit and the bio amp unit. Electrocardiography (ECG) ECG was obtained using a 4cm x 3.5cm electrode (3M Red Dot 2560). The electrodes were attached to an ADInstruments 5 - lead shielded bio amp cable, which was plugged into a ADInstruments Dual Bio Amp/Stimulator unit. Three leads were placed in total. The ne gative lead was positioned just under the right clavicle. The positive lead was positioned just above the left iliac crest. The ground was placed just above the right iliac crest. This positioning is known as Lead II. The skin was cleaned using an alcohol swab prior to lead attachment. Pneumobelt A d ouble pnuemobelt set up from Inductotrace was used to monitor chest wall and abdominal movements associated with breathing (model number 10.9000 from Ambulatory Monitoring Inc.) Appropriate transducer band size s were determined by measuring the circumference of the 13 respectively. Pneumobelts were calibrated using a one - liter spiro bag after the participant was l ying d own, using a nose clip to prevent nose breathing. Participants fully inflate d and deflated the bag 4 times and these values were used to convert raw voltage readings to liters. Data from both calibrated belts were summed to determine total air flow. Fino meter The finometer MIDI ( Finapres Medical Systems) was used to assess beat - to - beat blood pressure continuously throughout the study. Appropriate finger cuff sizes were determined by using the cuff size guide apparatus provided by the company. The f inger cuff was placed on the left hand middle finger intermediate phalanx and adjusted for proper readouts. To verify and calibrate the finometer readouts, manual blood pressure was taken two to three times before and after the study. Participants were asked to provided a small heating pad. Venous occlusion p lethysmography Forearm blood flow was assessed using Venous Occlusion Plethysmography (VOP) . Two rapid cuff inflators (E20, Hokanson) were at tached to one inflator air source (AG 101, Hokanson). One the wrist, and the other rapid cuff inflator was attached to an adult arm blood pressure cuff, which w as fastened around the upper arm. The rapid cuff inflator attached to the cuff around the bicep was on a n 8 sec inflate/7 sec deflate timer, while the wrist cuff stayed inflated for 4 minutes. 14 Mercury strain gauges (Hokanson) were used to assess the expans ion of the forearm. To select the appropriate sized strain gauge, the forearm circumference was measured two cm distal from the elbow fold. This wa s also the location of the strain gauge when fitted onto the participant. The measured size of the forearm was then subtracted by two and the resulting value was the size of the strain gauge used (cm). The cables were taped to avoid any unnecessary movement. The strain gauge was attached to a plethysmograph (EC6, Hokanson). Th e plethysmograph was calibrated prior to fitting of the strain gauge to the participant. VOP d ata was analyzed in Labchart by measuring the slope of each tracing during the inflation phase of the upper arm cuff over eight cycles of VOP and the individual slopes were averaged. Study c onditions Participants in the device - guided breathing condition were instructed to synchronize their breathing to a series of tones administered through headphones, which results in a breathing rate of 6 breaths/min. Participants in the self - guided condition were instructed to reduce their br eathing to a rate slower than normal, but still maintain a breathing pattern that is comfortable to them in both rate and depth. Self - guided participants also had headphones on, but no audio was played. All studies occurred in the same room with controlle d temperature and humidity, dimmed lighting, and quiet conditions. Psychological s urveys A State - Trait Anxiety Survey - State (STAI - S) was administered prior to instrumentation. Another STAI - S was administered after de - instrumentation. Following the second STAI - S, the participant was given the following surveys, in order: STAI - Trait, Five Facet Mindfulness Questionnaire, 15 Cognitive Interference Questionnaire, Penn State Worry Questionnaire, and Cognitive Failures Questionnaire . 16 Figure 2 : Timeline for studies. 17 Figure 2 Participants were screened and consented, then filled out an STAI - S, followed by intake forms. Next, the participant was instrumented with ECG, pnuemobelts, finometer, and VOP equipment, and the appropriate calibrations completed. To begin data c ollection, the participant was asked to breathe normally for 10 minutes, then slow their breathing down for 15 minutes either by self - guiding or device - guiding, then breathe normally again for 10 minutes. Data collection stopped after the end of the second endogenous breathing, and the participant was deinstrumented and asked to fill out the remaining psychological surveys. Throughout the entirety of data collection, the ECG, pneumobelts , finometer were continuously recording data. VOP cycling began 2 minut es before the end of a condition and ended 2 minutes after the beginning of the next. 18 Result s Breathing Manipulation B aseline b reathing rate was not different between the self - guided and device - guided groups ( 14.36±1.03 vs 15.39±1.29 breaths per minute , p=0.27). Breathing was reduced to 8.36±0.74 breaths per minute (BPM) in self - guided (p<0.01 compared to baseline) and 6.37±0.10 BPM in device - guided ( p>0.01 compared to baseline, p<0.01 between groups ). During recovery, breathing rate w as lower from baseline levels in self - paced ( 13.25±0.716 BPM, p=.030) and device - guided ( 13.67±1.11 BPM, p=0.020, Figure 3). Recovery breathing was not significant between conditions (p=0.378 ) . Breathing data represents the mean breath rate of the entire 1 0 or 15 min period. During baseline, self - guided and device - guided conditions had similar tidal volumes of 0.460.04 L and 0.42 ± 0.05 L, respectively (p=0.31 ). During condition breathing, the device - guided group had higher tidal volumes of 1.17±0.116 L (p> 0.01 compared to baseline) compared to self - guided conditions with an average of 0.710±0.0924 L ( p<0.01 compared to baseline, p>0.01 between groups ). During the recovery period, both self - and device - guided groups returned to basel ine tidal volume levels o f 0.47 ± 0.075 L and 0.49 ±0.087 L, respectively (p>0.0 5 between groups, Figure 4). Total ventilation (V E ) was similar for both self - and device - guided breathing during baseline at 6.172±0.35 L/min and 6.176±0.82 L/min, respectively (p=0.498 ) . V E was increase d in the device - guided condition during condition breathing at 7.41±0.69 L/min (p=0.056 compared to baseline) compared to self - guided condition breathing at 5.430± 0.364 L/min (p>0.01 ). During recovery, both groups returned to baseline ventilation levels and did not differ from each other: 19 5.919±0.721 (p=0.322 compared to baseline) and 6.335±1.000 (p=0.300 compared to baseline) L/min respectively (p=0.368 between conditions, Figure 5). Specific Aim I Aim I: To determine if device - guided and self - guided breathing result in different physiological outcomes. Self - guided and device - guided conditions do not statistically differ (p> 0.05) nor come statistically close to differing from each other (p>0.1) in any study phase (baseline, condition start, condition end, recovery start, recovery end) unless explicitly stated. Similarly, within - condition differences to baseline during study phases are either statistically insignificant (p>0.05) and/or are not statistically close to significance (p>0.1) unless explicitly stated. Changes in systolic blood pressure (SBP) from baseline for self - guided breathing during the start of condition, end of condit ion, start of recovery, and end of recovery was - 1.4 ±1.6 , - 3.5 ±2.4 (p=0. 018 compared to baseline), - 3.9 ±2.2 (p=0. 003 compared to baseline), and - 1.9 ±1.7 mmHg (p=0.0332 compared to baseline), respectively. Changes in SBP from baseline in device - guided breathing during the start of condition, end of condition, start of recovery, and end of recovery was 0.0 ±1.5 , - 0.4 ±2.4 , - 0.7 ±2.3, and +0.3±1.6 mmHg , respectively . During the end of condition breathing, self - guided and device - guided breathing were not stat istically different (p=0.138 ), though the means differed by a clinically relevant 3.2 mmHg (Figure 6). Changes in diastolic blood pressure (DBP), from baseline for self - guided breathing during the start of condition, end of condition, start of recovery, an d end of recovery was +0.1 ±0.7 , - 1.8 ±0.8 (p=0.019 compared to baseline) , - 2.0 ±0.9 , and - 1.0 ±1.4 mmHg, respectively. Changes in DBP during device - guided breathing during the start of condition, end of condition, start of recovery, 20 and end of recovery was - 1 .8 ±0.6 (p=0.008 compared to baseline) , - 0.6 ±1.0 , - 0.3 ±1.0 , and - 0.6 ±1.4 mmHg, respectively . During the start of the condition, device - guided breathing had a significantly greater drop in blood pressure compared to the self - guided condition (p=0.036 between conditions) (Figure 7). Changes in mean arterial pressure ( MAP ) , from baseline for self - guided breathing during the start of condition, end of condition, start of recovery, and end of recovery was - 0.4 ±2.3 , - 2.3 ±2.0 (p=0.004 compared to baseline) , - 2.6 ±2 .1 (p=0.013 compared to baseline) , and - 1.4 ±3.1 mmHg , respectively . Changes in MAP during device - guided breathing during the start of condition, end of condition, start of recovery, and end of recovery was - 1.2 ±2.3 (p=0.045 compared to baseline) , - 0.5 ±1.9 , - 0.4 ±1.9 , and - 0.3 ±3.0 mmHg, respectively (Figure 8). Change in forearm blood flow, as percentage changes from baseline, for self - guided breathing during the start of condition, end of condition, start of recovery, and end of recovery was 108.9 ±4.2 (p=0.030 compared to baseline) , 119.8 ±8.0 (p=0.018 compared to baseline) , 111.8 ±7.8 , and 115.8 ±9.0 %, respectively. Forearm blood flow for device - guided breathing during the start of condition, end of condition, start of recovery, and end of recovery was 112.4 ±4.1 (p=0.025 compared to baseline) , 103.5 ±6.8 , 96.2 ±5.8 , and 97.2 ±3.5 %, respectively . During the recovery end phase, the self - guided condition had significantly higher forearm blood flow than the device - guided group (p=0.039) (Figure 9). Change in h eart rate , as percentage change from baseline, for self - guided breathing during the start of condition, end of condition, start of recovery, and end of recovery was 101.5 ±1.4 , 98.7 ±1.8 , 99.0 ±1.5 , and 97.5 ±1.7 %, respectively. Change in heart rate for devic e - guided breathing during the start of condition, end of condition, start of recovery, and end of recovery was 105.3 ±1.7 (p=0.007 compared to baseline) , 98.6 ±5.2 , 99.8 ±2.7 , and 99.2 ± 1.9 %, 21 respectively . At the condition start phase, device - guided breathing had a significantly higher change in heart rate than the self - guided condition (p=0.047) (Figure 10). Specific Aim I I Aim II: To determine if psychological metrics are correlated to a reduction in blood pressure during slow pace breathing. All changes in blood pressure used for testing correlation with psychological metrics are referring to a change in MAP from the average of the last two minutes of baseline to the average of the last two minutes of condition breathing. Before the study, anxiety levels as determined by the STAI - S were not different between groups, 30.54±2.23 for the self - guided group and 32.5±3.08 for the device - guided group (p=0.304 between groups ). After the study, anxiety levels decreased to 26.00± 1.24 for self - guided (p=0.024 pre vs. post) and 28.5± 1.81 for device - guided (p=0.038 pre v s. post, p=0.131 between groups) (Figure 11). Trait anxiety as determined by the STAI - T was found to be highest in participants whose blood pressure did not fall in response to self - guided breathing conditions (r=0.452, p=0.081). There was no correlation found with device - guided conditions (r= - 0.301, p=0.199) (Figure 12). Cognitive failures was found to be highest i n participants whose blood pressure did not respond to self - guided breathing conditions (r=0.498, p=0.059). There was no correlation found with device - guided conditions (r=0.0923, p=0.400) (Figure 13). Observational mindfulness was found to be highest in p articipants whose blood pressure did not respond to device - guided breathing conditions (r=0.405, p=0.123) . There was no correlation found with self - guided conditions (r= - 0.246, p=0.233) (Figure 14). 22 Acting without awareness, a trait associated with no thin king about your actions before and during their execution, was found to correlate with a lack of response in blood pressure in self - guided participants (r=0.683, p<0.01). Acting without awareness was found to correlate with a reduction in blood pressure in device - guided p articipants (r= - 0.472, p=.084) (Figure 15). 23 Figure 3 : Respiratory Rate. Respiratory rate determined using double pneumobelts . Bars represent the me an ± SE of each entire phase. (# indicates p<0.05 between the two conditions at the given phase, * indicates p<0.05 between the current phase and baseline for the same condition). Device - guided condition denoted with DG, self - paced condition denoted with Self. 0 2 4 6 8 10 12 14 16 18 Baseline Condition Recovery Breathing Rate (Breaths/Min) Respiratory Rate Self DG * * # 24 Figure 4 : Tidal Volume . Tidal volumes determined using calibrated double pnuemobelts . Values shown are the mean ± SE of the entire duration of the condition. (# indicates p<0.05 between the two conditions at the given phase, * indicates p< 0.05 between the current phase and baseline for the same condition). Device - guided condition denoted with DG, self - paced condition denoted with Self. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Baseline Condition Recovery Tidal Volume (Liters) Tidal Volume Self DG # * * 25 Figure 5 : Ventilation . Ventilation (V E ) ( breathing rate x tidal volume ) . Breathing rate and tidal volumes determined using double pnuemobelts were manually multiplied for V E calculation. Values shown are the mean ± SE of the entire duration of the condition. (# indicates p<0.05 between the two conditions at the given phase, * indicates p<0.05 between the current phase and baseline for the same condition , & indicates p=0.056 ). Device - guided condition denoted with DG, self - paced condition denoted with Self . 0 1 2 3 4 5 6 7 8 9 10 Baseline Condition Recovery Ventilation (Liters/min) Ventilation Self DG # * & 26 Figure 6 : Systolic Blood Pressure. Systolic blood pressure determined using a finometer . Values represented are the mean ± SE for the two minute period at the beginning or end of each phase. Device - guided condition denoted with DG, sel f - paced condition denoted with Self . -7 -6 -5 -4 -3 -2 -1 0 1 2 3 Baseline Condition Start Condition End Recovery Start Recovery End Change in Blood Pressure (mmHg) Systolic Blood Pressure Self DG 27 Figure 7: Diastolic Blood Pressure. Di astolic blood pressure determined using a finometer. Values represented are the mean ± SE for the two minute period at the beginning or end of each phase. Device - guided condition denoted with DG, self - paced condition denoted with Self. -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 Baseline Condition Start Condition End Recovery Start Recovery End Change in Blood Pressure (mmHg) Diastolic Blood Pressure Self DG 28 Figure 8 : Mean Arterial Blood Pressure: Mean arterial pressure was calculated by adding (2/3*Diastolic) and (1/3*Systolic). Diastolic and systolic values determined using a finometer. Values represented are the mean ± SE for the two minute period at the beginning or end of each phase. Device - guided condition denoted with DG, self - pac ed condition denoted with Self . -4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 Baseline Condition Start Condition End Recovery Start Recovery End Change in Blood Pressure (mmHg) Mean Arterial Pressure Self DG 29 Figure 9 : Forearm Vasodilation. Forearm vasodilation determined using v enous occlusion plethysmography (VOP) . Change from baseline was calculated by (phase/baseline)*100. Values represented are the mean ± SE for the two minute period at the beginning or end of each phase . Device - guided condition denoted with DG, self - pac ed condition denoted with Self . 60 70 80 90 100 110 120 130 140 Baseline Condition Start Condition End Recovery Start Recovery End VOP (% Change from Baseline) VOP (%change from Baseline) Self DG 30 Figure 10 : Heart Rate. Heart rate determined using ECG lead II. Values represented are the mean ± SE for the entirety of the phase. (# indicates p<0.05 between the two conditions at the given phase, * indicates p<0.05 between the current phase and baseline for the same condition). Device - guided condition denoted with DG, self - paced condition denoted with Self . 90 92 94 96 98 100 102 104 106 108 Baseline Condition Start Condition End Recovery Start Recovery End Change in Heart Rate (% from Baseline) Heart Rate Self DG *# 31 Figure 11 : State Anxiety . State - Trait Anxiety Inventory State was administered before and after the study. Va lues shown are the mean ± SE. (# indicates p<0.05 between the two condi tions at the given time point, * indicates p<0.05 betwe en the two different time point, same condition). Device - guided condition denoted with DG, self - pac ed condition denoted with Self . 0 5 10 15 20 25 30 35 40 PRE POST STAI - s Score STAI Self DG * * 32 Figure 12 : Trait Anxiety. STAI - T administered post - study. Pearson correlations listed below . A) STAI - T scores positively correlate with change in MAP in self - paced conditions(r=0.452, p=0.081). B) STAI - T scores weakly negatively correlate with change in MAP in device - guided conditions(r= - 0.301, p=0.199). Device - guided condition denoted with DG, self - pac ed condition denoted with Self . -8 -6 -4 -2 0 2 0 10 20 30 40 50 Change in Mean Arterial Pressure (mmHg) STAI - T Anxiety Score A: Self: STAI - T -8 -6 -4 -2 0 2 0 10 20 30 40 50 Change in Mean Arterial Pressure (mmHg ) STAI - T Anxiety Score B: DG: STAI - T 33 Figure 13 : Cognitive Failure. Cognitive failure questionnaire (CFQ) administered post - study. Pearson correlations listed below . A) CFQ scores positively correlated with change in MAP in self - paced conditions (r=0.498, p=0.059). B) CFQ score did not correlate with change in MAP in device - guided conditions (r= 0 .092, p=0.400). Device - guided condition denoted with DG, self - pac ed condition denoted with Self . -8 -6 -4 -2 0 2 0 10 20 30 40 50 60 Change in Mean Arterial Pressure (mmHg) CFQ Cognitive Failure Score A: Self: CFQ -8 -6 -4 -2 0 2 0 10 20 30 40 50 60 Change in Mean Arterial Pressure (mmHg) CFQ Cognitive Failure Score B: DG: CFQ 34 Figure 14 : Observational Mindfulness. Five - facet mindfulness questionnaire (FFMQ) was administered post study. Pearson correlations listed below . A) FFMQ - observation scores weakly negatively correlated with change in MAP in self - paced conditions (r= - 0.246, p=0.233). B) FFMQ - observation scores moderately positively correlated with change in MAP in device - guided conditions (r=0.405, p=0.123). Device - guided condition denoted with DG, self - pa ced condition denoted with Self . -8 -6 -4 -2 0 2 0 1 2 3 4 5 Change in Mean Arterial Pressure (mmHg) 5 Facet Observation Score A: Self: Observational Mindfulness -8 -6 -4 -2 0 2 0 1 2 3 4 5 Change in Mean Arterial Pressure (mmHg) 5 Facet Observation Score B: DG: Observational Mindfulness 35 Figure 15 : Acting Without Awareness. Five - facet mindfulness questionnaire (FFMQ) was administered post study. Pearson correlations listed below . A) FFMQ - act without awareness scores positively correlated with a change in MAP in self - paced conditions ( r=0.683, p=0.010). B) FFMQ - act without awareness scores negatively correlated with a change in MAP in device - guided conditions (r= - 0.472, p=0.084 ). Device - guided condition denoted with DG, self - pac ed condition denoted with Self . -8 -6 -4 -2 0 2 0 1 2 3 4 5 Change in Mean Arterial Pressure (mmHg) 5 Facet Anti - Awareness Score A: Self: Acting Without Awareness -8 -6 -4 -2 0 2 0 1 2 3 4 5 6 Change in Mean Arterial Pressure (mmHg) 5 Facet Awareness Score B: DG: Acting Without Awareness 36 Discussion Slow - paced breathing has been used for thousands of years and has amassed anecdotal evidence for its effects on stress reduction. Due to recent attempts to replicate self - paced breathing maneuvers with devices, and their mixed results in recent studies, we looked at the differences between self - guided and device - guided slow breathing practices and the psychological correlates. We hypothesized that slow - breathing exercises would result in a reduction in blood pressure, that self - paced slow breathing would be more effective than device - guided breathing, and that psychological metrics could explain the variability in responses to each condition. Our data supported our hypotheses. Diastolic blood pressure, systolic blood pressure and mean arterial pressure were reduced in the self - guided condition but not the device - guided condition. Trait anxiety, cognitive interference, observational mindfulness, and acting without awareness all were predictive of blood pressure responses in either device - guided or self - guided practices. In all, device - guided slow breathing and self - guided slow breathing are inherently different maneuvers with different outcomes and psychological predictors. In both self and device - guided conditions during the slow breathing phase, breathing rate was reduced and tidal volume increased (Figure 3, 4 ). V E increased during device - guided breathing, but decreased during self - paced breathing (Figure 5 ). The device - guided group had a lower breathin g rate, higher tidal volume and higher ventilation during slow breathing than th e self - paced breathing (Figures 3, 4, 5 ). During device - guided breathing, there is a possibility that arterial oxygen and carbon dioxide levels are altered, which could cause c hanges in chemoreflex function not seen in the self - guided group. This is another aspect where the two breathing practices may differ, as self - paced breathing practices allow participants to alter their blood gas levels through modulation of their breathin g rate. 37 We attribute these differences between conditions in breathing rate, tidal volume, and ventilation to the instructions given to participants for their slow breathing phases. Device - guided participants reduced their breathing to exactly 6 breaths/m in by matching to the musical tones that were played to this cadence . Subjects w ere breathing, either inhaling or exhaling, during the entire 5 second duration of the continuous tone played during to mark the inspiratory/expiratory phases, producing a very low stand ard error for breathing rate, a higher tidal volume and higher V E . Self - paced participants varied in their slow breathing patterns based on individual preference, and were able to breathe at a comfortable depth and rate as long as it was slower t han their regular breathing rate . It is arguable that the drop in breathing rate in the device - guided condition may have induced stress in some participants, preventing reductions in blood pressure. While six breaths per minute is not necessarily strenuous for most people, perhaps there is an optimal rate for each person that should be explored further, perhaps by instructing participants to first perform self - guided breathing to find a comfortable slow breathing rate, then syncing the device to this cadenc e. The self - guided condition resulted in a reduction in systolic blood pressure with a persistence of this effect lasting into the recovery period, whereas device - guided breathing resulted in systolic blood pressure remaining at baseline levels throughout the entirety of the study (Figure 7) , suggesting that self - paced breathing on average may be more effective in reducing blood pressure. The reduction in diastolic and mean arterial blood pressure seen in the device - guided group at the beginning of the con dition (Figures 6, 8) is likely an immediate response to the deep breathing practice that is quickly compensated for through baroreflex mechanisms over the course of condition breathing. Self - guided reductions in diastolic and mean arterial pressure do 38 not immediately present themselves at the onset of condition breathing, but drop during the condition breathing and persist through recovery (Figures 6, 8 ) . This could also potentially be due to a change in b aro reflex function , including a change in the operating point and/or a change in sensitivity, though this has not been confirmed and requires further investigation . Supporting the idea of differential baroreflex influence on changes in blood pressure, there is an increase in heart rate alongside the decrease in mean arterial and diastolic blood pressure that occurs immediately after the onset of slow breathing in the device - guided group (Figure 10) , which would be expected due to baroreflex functioning . In self - paced breathing, heart rate does not change instead of increasing, potentially allowing for a decrease in diastolic and mean arterial blood pressure. This is suggestive of a change in the operating point of the baroreflex, as otherwise heart rate would increase. This would need to be confirmed in future studies. These data concerning blood pressure reductions during self - paced breathing maneuvers (Figures 6, 7, 8) support previous studies on the acute effects of yoga breathing (Raghuraj and Telles 2008, Bhavanani, Madanmohan et al. 2012) , and though the current study was an acute study, its findings also support previous studies on the effects of chronic yoga breathin g (Sundar, Agrawal et al. 1984, McCaffrey, Ruknui et al. 2005, Upadhyay Dhungel, Malhotra et al. 2008) . The results regar ding device - guided breathing and lack of reduction in blood pressure (Figure 6, 7, 8) are difficult to compare to previous studies. This is because the present study used young, healthy men, and previous studies used older, hypertensive men (and in some cases, women). Setting this aside, our data supports previous studies that show that dev ice - guided breathing does not reduce blood pressure (Altena, Kleefstra et al. 2009, Anderson, McNeely et al. 2010, Hering, Kuchars ka et al. 2013) , and contradicts the studies that show a reduction in blood pressure (Grossman, Grossman et al. 2001, Schein, Gavish et al. 2001, Viskoper, Shapira et al. 2003, 39 Elliot, Izzo et al. 2004, Meles, Giannattasio et al. 2004, Schein, Gavish et al. 2009, Oneda, Ortega et al. 2010, Bertisch, Schomer et al. 2011) . The increase in forearm blood flow in the self - guided condition (Figure 9) indicates that there is a reduction in sympathetic outflow, resulting in a decrease in vasoconstriction in the forearm. Th is is consistent with previous findings of reductions in sympathetic nerve activity in self - paced breathing pract iced (Pal, Velkumary et al. 2004) . There is no increase in forearm blood flow in the device - guided group (Figure 9) , indicating no change in sympathetic outflow . Because forearm blood flow is determined by vasodilation and vasoconstriction, it is intimately related to sympathetic nerve activity. Thus, the lack of increase in forearm blood flow contradicts previous work that shows a decrease in MSNA activity due to device - guided breathing slow breathing (Hering, Kucharska et al. 2013, Harada, Asanoi et al. 2014) . We examined several psychological metrics in attempts to reveal predictor variables for success in reduction of blood pressure in response to these breathing practices. Both groups had a reduction in state anxiety after the st udy compared to before (Figure 11 ), despite no correlation between reduction of anxiety and reduction in blood pressure (data not shown). T rait anxiety was predictive of blood pressure reductions in both groups (Figure 12) . It appears that the more anxious a person is, the worse they will respond with a self - paced practice, but they will perform better on a device - guided maneuver. We suggest that this is because highly anxious people need guidance in performing a task so that they are validated that they are doing what is being asked. High levels of cognitive failures were predictive of poor responses in blood pressure in self - paced groups, but were not correlated with device - guided groups (Figure 13 ) . This makes sense, as those with high levels of co gnitive failures may forget to follow instructions and not adhere to the protocol, resulting in blunted results. This goes hand - in - hand with acting without awareness, 40 as those with higher levels of this trait are not attentive to what they are doi ng and th us may forget. Interestingly, those who act without awareness performed better on device - guided breathing (Figure 15) , suggesting that not thinking about the protocol while doing it is best. Higher levels of observational mindfulness, or your awareness to your body and surroundings, were prognostic of poor responses to device - guided breathing (Figure 14) . We suggest that this is due to a focus shift from the body to an external device, which could be potentially invasive to a highly mindful person. 41 Conclusion Our data shows that there are physiological and psychological differences between device - guided and self - paced breathing practices. This provides support to the idea that devices used to help reduce breathing rate in hopes to replicate the successes of sel f - paced breathing may not be appropriate in some cases. Our psychological data suggests that there are several metrics that a patient could be screened for prior to prescribing a slow - paced breathing practice in order to improve efficacy and successfully r educe blood pressure. Further studies should be done to understand the mechanisms behind the phenomenon of divergent physiological responses to these two seemingly similar maneuvers. 42 REFERENCES 43 REFERENCES Altena, M. R., N. Kleefstra, S. J. Logtenberg, K. H. Groenier, S. T. Houweling and H. J. Bilo (2009). "Effect of device - guided breathing exercises on blood pressure in patients with hypertension: a randomized controlled trial." Blood Press 18 (5): 273 - 279. Anderson, D. E., J. D. McNeely and B. G. Windham (2010). "Regular slow - breathing exercise effects on blood pressure and breathing patterns at rest." J Hum Hypertens 24 (12): 807 - 813. Bertisch, S. M., A. Schomer, E. E. Kelly, L. A. Baloa, L. E. Hueser, S. D. Pittman and A. Malhotra (2011). "Device - guided paced respiration as an adjunctive therapy for hypertension in obstructive sleep apnea: a pilot feasibility study." Appl Psychophysiol Biofeedback 36 (3): 173 - 179. Bhavanani, A. B., Madanmohan and Z. Sanjay (2 012). "Immediate effect of chandra nadi pranayama (left unilateral forced nostril breathing) on cardiovascular parameters in hypertensive patients." Int J Yoga 5 (2): 108 - 111. Brook, R. D., L. J. Appel, M. Rubenfire, G. Ogedegbe, J. D. Bisognano, W. J. Elli ott, F. D. Fuchs, J. W. Hughes, D. T. Lackland, B. A. Staffileno, R. R. Townsend and S. Rajagopalan (2013). "Beyond medications and diet: alternative approaches to lowering blood pressure: a scientific statement from the american heart association." Hypert ension 61 (6): 1360 - 1383. Brown, K. W., N. Weinstein and J. D. Creswell (2012). "Trait mindfulness modulates neuroendocrine and affective responses to social evaluative threat." Psychoneuroendocrinology 37 (12): 2037 - 2041. Charkoudian, N., E. Gusman, M. J. J oyner, B. G. Wallin and J. Osborn (2010). "Integrative mechanisms of blood pressure regulation in humans and rats: cross - species similarities." Am J Physiol Regul Integr Comp Physiol 298 (3): R755 - 759. Creswell, J. D., B. M. Way, N. I. Eisenberger and M. D. Lieberman (2007). "Neural correlates of dispositional mindfulness during affect labeling." Psychosom Med 69 (6): 560 - 565. Eckberg, D. L. (2003). "The human respiratory gate." J Physiol 548 (Pt 2): 339 - 352. Eckberg, D. L., Y. T. Kifle and V. L. Roberts (1980 ). "Phase relationship between normal human respiration and baroreflex responsiveness." J Physiol 304 : 489 - 502. Eckberg, D. L. and C. R. Orshan (1977). "Respiratory and baroreceptor reflex interactions in man." J Clin Invest 59 (5): 780 - 785. Elliot, W. J., J. L. Izzo, Jr., W. B. White, D. R. Rosing, C. S. Snyder, A. Alter, B. Gavish and H. R. Black (2004). "Graded blood pressure reduction in hypertensive outpatients associated 44 with use of a device to assist with slow breathing." J Clin Hypertens (Greenwich) 6 (10): 553 - 559; quiz 560 - 551. Feldman, J. L. and C. A. Del Negro (2006). "Looking for inspiration: new perspectives on respiratory rhythm." Nat Rev Neurosci 7 (3): 232 - 242. Gray, P. A., W. A. Janczewski, N. Mellen, D. R. McCrimmon and J. L. Feldman (2001). "Normal breathing requires preBotzinger complex neurokinin - 1 receptor - expressing neurons." Nat Neurosci 4 (9): 927 - 930. Grossman, E., A. Grossman, M. H. Schein, R. Zimlichman and B. Gavish (2001). "Breathing - control lowers blood pressure." J Hum Hypertens 1 5 (4): 263 - 269. Guertzenstein, P. G. and A. Silver (1974). "Fall in blood pressure produced from discrete regions of the ventral surface of the medulla by glycine and lesions." J Physiol 242 (2): 489 - 503. Harada, D., H. Asanoi, J. Takagawa, H. Ishise, H. Uen o, Y. Oda, Y. Goso, S. Joho and H. Inoue (2014). "Slow and deep respiration suppresses steady - state sympathetic nerve activity in patients with chronic heart failure: from modeling to clinical application." Am J Physiol Heart Circ Physiol 307 (8): H1159 - 116 8. Hering, D., W. Kucharska, T. Kara, V. K. Somers, G. Parati and K. Narkiewicz (2013). "Effects of acute and long - term slow breathing exercise on muscle sympathetic nerve activity in untreated male patients with hypertension." J Hypertens 31 (4): 739 - 746. Huang, A. J. and L. L. Subak (2014). "What constitutes an adequate evaluation of device - guided breathing?" JAMA Intern Med 174 (4): 637. Jacobs, T. L., P. R. Shaver, E. S. Epel, A. P. Zanesco, S. R. Aichele, D. A. Bridwell, E. L. Rosenberg, B. G. King, K. A . Maclean, B. K. Sahdra, M. E. Kemeny, E. Ferrer, B. A. Wallace and C. D. Saron (2013). "Self - reported mindfulness and cortisol during a Shamatha meditation retreat." Health Psychol 32 (10): 1104 - 1109. Janczewski, W. A. and J. L. Feldman (2006). "Distinct r hythm generators for inspiration and expiration in the juvenile rat." J Physiol 570 (Pt 2): 407 - 420. Landman, G. W., I. Drion, K. J. van Hateren, P. R. van Dijk, S. J. Logtenberg, J. Lambert, K. H. Groenier, H. J. Bilo and N. Kleefstra (2013). "Device - guide d breathing as treatment for hypertension in type 2 diabetes mellitus: a randomized, double - blind, sham - controlled trial." JAMA Intern Med 173 (14): 1346 - 1350. McCaffrey, R., P. Ruknui, U. Hatthakit and P. Kasetsomboon (2005). "The effects of yoga on hypert ensive persons in Thailand." Holist Nurs Pract 19 (4): 173 - 180. Meles, E., C. Giannattasio, M. Failla, G. Gentile, A. Capra and G. Mancia (2004). "Nonpharmacologic treatment of hypertension by respiratory exercise in the home setting." Am J Hypertens 17 (4): 370 - 374. 45 Mellen, N. M., W. A. Janczewski, C. M. Bocchiaro and J. L. Feldman (2003). "Opioid - induced quantal slowing reveals dual networks for respiratory rhythm generation." Neuron 37 (5): 821 - 826. Mellen, N. M. and M. Thoby - Brisson (2012). "Respiratory ci rcuits: development, function and models." Curr Opin Neurobiol 22 (4): 676 - 685. Moraes, D. J., L. G. Bonagamba, D. B. Zoccal and B. H. Machado (2011). "Modulation of respiratory responses to chemoreflex activation by L - glutamate and ATP in the rostral ventr olateral medulla of awake rats." Am J Physiol Regul Integr Comp Physiol 300 (6): R1476 - 1486. Moraes, D. J., M. P. da Silva, L. G. Bonagamba, A. S. Mecawi, D. B. Zoccal, J. Antunes - Rodrigues, W. A. Varanda and B. H. Machado (2013). "Electrophysiological prop erties of rostral ventrolateral medulla presympathetic neurons modulated by the respiratory network in rats." J Neurosci 33 (49): 19223 - 19237. Mozer, M., P. Fadel, C. Johnson, N. C. B Wallin, J. Drobish, M. Joyner and E. Wehrwein (2014). "Acute slow - paced breathing increases periods of sympathetic nervous system quiescence." The FASEB Journal 28 (1): Supplement 1170.1112. Narkiewicz, K., P. van de Borne, N. Montano, D. Hering, T. Kara and V. K. Somers (2006). "Sympathetic neural outflow and chemoreflex sensitivity are related to spontaneous breathing rate in normal men." Hypertension 47 (1): 51 - 55. Okon - Singer, H., J. Mehnert, J. Hoyer, L. Hellrung, H. L. Schaare, J. Dukart and A. Villringer (2014). "Neural control of vascular reactions: impact of emotion and attention." J Neurosci 34 (12): 4251 - 4259. Oneda, B., K. C. Ortega, J. L. Gusmao, T. G. Arauj o and D. Mion, Jr. (2010). "Sympathetic nerve activity is decreased during device - guided slow breathing." Hypertens Res 33 (7): 708 - 712. Pal, G. K., S. Velkumary and Madanmohan (2004). "Effect of short - term practice of breathing exercises on autonomic funct ions in normal human volunteers." Indian J Med Res 120 (2): 115 - 121. Raghuraj, P. and S. Telles (2008). "Immediate effect of specific nostril manipulating yoga breathing practices on autonomic and respiratory variables." Appl Psychophysiol Biofeedback 33 (2) : 65 - 75. Ramirez, J. M., E. J. Zuperku, G. F. Alheid, S. P. Lieske, K. Ptak and D. R. McCrimmon (2002). "Respiratory rhythm generation: converging concepts from in vitro and in vivo approaches?" Respir Physiol Neurobiol 131 (1 - 2): 43 - 56. Richter, D. W. and K. M. Spyer (2001). "Studying rhythmogenesis of breathing: comparison of in vivo and in vitro models." Trends Neurosci 24 (8): 464 - 472. 46 Ross, C. A., D. A. Ruggiero, D. H. Park, T. H. Joh, A. F. Sved, J. Fernandez - Pardal, J. M. Saavedra and D. J. Reis (1984) . "Tonic vasomotor control by the rostral ventrolateral medulla: effect of electrical or chemical stimulation of the area containing C1 adrenaline neurons on arterial pressure, heart rate, and plasma catecholamines and vasopressin." J Neurosci 4 (2): 474 - 49 4. Schein, M. H., B. Gavish, T. Baevsky, M. Kaufman, S. Levine, A. Nessing and A. Alter (2009). "Treating hypertension in type II diabetic patients with device - guided breathing: a randomized controlled trial." J Hum Hypertens 23 (5): 325 - 331. Schein, M. H., B. Gavish, M. Herz, D. Rosner - Kahana, P. Naveh, B. Knishkowy, E. Zlotnikov, N. Ben - Zvi and R. N. Melmed (2001). "Treating hypertension with a device that slows and regularises breathing: a randomised, double - blind controlled study." J Hum Hypertens 15 (4): 271 - 278. Smith, J. C., H. H. Ellenberger, K. Ballanyi, D. W. Richter and J. L. Feldman (1991). "Pre - Botzinger complex: a brainstem region that may generate respiratory rhythm in mammals." Science 254 (5032): 726 - 729. Smith, J. C., J. J. Greer, G. S. Liu an d J. L. Feldman (1990). "Neural mechanisms generating respiratory pattern in mammalian brain stem - spinal cord in vitro. I. Spatiotemporal patterns of motor and medullary neuron activity." J Neurophysiol 64 (4): 1149 - 1169. Sun, Q. J., J. Minson, I. J. Llewellyn - Smith, L. Arnolda, J. Chalmers and P. Pilowsky (1997). "Botzinger neurons project towards bulbospinal neurons in the rostral ventrolateral medulla of the rat." J Comp Neurol 388 (1): 23 - 31. Sundar, S., S. K. Agrawal, V . P. Singh, S. K. Bhattacharya, K. N. Udupa and S. K. Vaish (1984). "Role of yoga in management of essential hypertension." Acta Cardiol 39 (3): 203 - 208. Tomfohr, L. M., M. A. Pung, P. J. Mills and K. Edwards (2015). "Trait mindfulness is associated with bl ood pressure and interleukin - 6: exploring interactions among subscales of the Five Facet Mindfulness Questionnaire to better understand relationships between mindfulness and health." J Behav Med 38 (1): 28 - 38. Upadhyay Dhungel, K., V. Malhotra, D. Sarkar an d R. Prajapati (2008). "Effect of alternate nostril breathing exercise on cardiorespiratory functions." Nepal Med Coll J 10 (1): 25 - 27. Viskoper, R., I. Shapira, R. Priluck, R. Mindlin, L. Chornia, A. Laszt, D. Dicker, B. Gavish and A. Alter (2003). "Nonpha rmacologic treatment of resistant hypertensives by device - guided slow breathing exercises." Am J Hypertens 16 (6): 484 - 487. Wallin, B. G. and D. L. Eckberg (1982). "Sympathetic transients caused by abrupt alterations of carotid baroreceptor activity in huma ns." Am J Physiol 242 (2): H185 - 190. Yasuma, F. and J. Hayano (2004). "Respiratory sinus arrhythmia: why does the heartbeat synchronize with respiratory rhythm?" Chest 125 (2): 683 - 690. 47 Zoccal, D. B., W. I. Furuya, M. Bassi, D. S. Colombari and E. Colombari (2014). "The nucleus of the solitary tract and the coordination of respiratory and sympathetic activities." Front Physiol 5 : 238.