The Pennsylvania State University. The Graduate School. Department of Cellular and Molecular Physiology THE ROLE OF THE VESTIBULOSYMPATHETIC REFLEX

Transcription

1 The Pennsylvania State University The Graduate School Department of Cellular and Molecular Physiology THE ROLE OF THE VESTIBULOSYMPATHETIC REFLEX IN BLOOD PRESSURE REGULATION IN HUMANS A Dissertation in Physiology by Damian Joseph Dyckman 2009 Damian Joseph Dyckman Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy May 2009

2 The dissertation of Damian Joseph Dyckman was reviewed and approved* by the following: ii Chester A. Ray Professor of Medicine, and Cellular and Molecular Physiology Dissertation Advisor Chair of Committee Charles H. Lang Professor of Cellular and Molecular Physiology, and Surgery James A. Pawelczyk Associate Professor of Physiology, and Kinesiology Lawrence I. Sinoway Professor of Medicine Leonard S. Jefferson, Jr. Professor of Cellular and Molecular Physiology Head of the Department of Cellular and Molecular Physiology *Signatures are on file in the Graduate School

3 ABSTRACT iii Abstract Chapter 2 Activation of the vestibular otolith organs with head-down rotation increases muscle sympathetic nerve activity (MSNA) in humans. Previously, we demonstrated this vestibulosympathetic reflex (VSR) elicits increases in MSNA during baroreflex unloading (i.e., lower-body negative pressure) in humans. Whether such an effect persists during baroreflex loading is unknown. We tested the hypothesis that the ability of the VSR to increase MSNA is preserved during baroreflex unloading and inhibited during baroreflex loading. Ten subjects (26±1 yr) performed 3 trials of headdown rotation (HDR) to activate the VSR. These trials were performed after a period of sustained saline (control), nitroprusside (baroreflex unloading: µg/kg/min), and phenylephrine (baroreflex loading: µg/kg/min) infusion. Nitroprusside infusion decreased ( 7±1 mmhg; p<0.001) and phenylephrine infusion increased ( 8±1 mmhg; p<0.001) mean arterial pressure at rest. HDR performed during the control ( 3±2 bursts/min, 314±154 arbitrary units (au) total activity, 41±18% total activity; p<0.05) and nitroprusside trials ( 5±2 bursts/min, 713±241 au total activity, 49±20% total activity; p<0.05) increased MSNA similarly despite significantly elevated levels at rest (13±2 to 26±3 bursts/min) in the latter. In contrast, HDR performed during the phenylephrine trial failed to increase MSNA ( 0±1 bursts/min, -15±33 au total activity, -8±21% total activity). These results confirm previous findings that the ability of the VSR to increase MSNA is preserved during baroreflex unloading. In contrast, the ability of the VSR to increase MSNA is abolished during baroreflex loading. These results

4 iv provide further support for the concept that the VSR may act primarily to defend against hypotension in humans. Abstract Chapter 3 Orthostatic intolerance is a common problem following bed rest. The mechanism for this is equivocal. Vestibular reflexes contribute to orthostatic blood pressure regulation. We hypothesized that sympathetic nerve responses to otolith stimulation would be attenuated by prolonged head-down bed rest (HDBR) and that this attenuation would be associated with increased orthostatic intolerance. Arterial blood pressure, heart rate, muscle sympathetic nerve activity (MSNA), and peripheral vascular conductance were measured during head-down rotation (HDR; otolith organ stimulation) in the prone posture before and after short-duration (24-hrs; n=22) and prolonged (36±1 days; n=8) HDBR. Head-up tilt at 80 o was performed to assess orthostatic tolerance. After short-duration HDBR, MSNA responses to HDR were preserved ( 5±1 bursts/min, 53±13 % burst frequency, 65±13% total activity; p<0.001). No association was observed between the vestibulosympathetic reflex and head-up tilt duration. After prolonged HDBR, MSNA responses to HDR were attenuated ~50%. MSNA increased by 23±13% burst frequency and 34±22% total activity during HDR. Moreover, these results were observed in 3 subjects tested again at 75±1 days of HDBR. This reduction in MSNA responses to otolith organ stimulation at 5 weeks was associated with reductions in head-up tilt duration. These results indicate that prolonged HDBR (~5 wks) attenuates the vestibulosympathetic reflex and contributes to orthostatic intolerance following HDBR in humans. These results suggest a novel mechanism in the development of orthostatic intolerance in humans.

5 Abstract Chapter 4 Classically, the glycerol dehydration test (GDT) has been v used to test for the presence of Ménière s disease and can cause acute alterations in vestibular reflexes in both normal and pathological states. The vestibulosympathetic reflex (VSR) elicits increases in muscle sympathetic nerve activity (MSNA) and peripheral vasoconstriction. We hypothesized that the GDT would attenuate the VSR through an acute fluid shift of the inner ear. Nine male subjects (27±1 years) performed head-down rotation (HDR), which engages the VSR, before and after administration of either the GDT or saline. MSNA (microneurography), arterial blood pressure, and leg blood flow (venous occlusion plethysmography) were measured during HDR. Before drug administration, HDR significantly increased MSNA in burst frequency ( 5±1 bursts/min, 8±1 bursts/min; p<0.01) and total activity ( 44±13%, 77±17%; p<0.01), and decreased calf vascular conductance ( 15±6%, 20±3%; p<0.01), in both the saline and glycerol trials, respectively. Post-saline, HDR still significantly increased MSNA ( 6±2 bursts/min, 83±20% total activity; p<0.01) and decreased calf vascular conductance ( 21±4%, p<0.01), which was not significantly different from pretesting. In contrast, post-gdt resulted in an attenuation of MSNA ( 3±1 bursts/min, 22±3% total activity) and reduction in calf vascular conductance ( 7±4%) during HDR. These results suggest that a fluid shift of the inner ear via glycerol dehydration attenuates the VSR. These data provide support that dynamic fluid shifts can have a significant effect on the VSR.

6 TABLE OF CONTENTS vi LIST OF FIGURES...ix LIST OF TABLES...x ACKNOWLEDGEMENTS...xi Chapter 1 Introduction Vestibular Contributions to Blood Pressure Regulation Relation to Orthostasis Specific Aims and Hypotheses...6 Chapter 2 Effect of Baroreflex Loading on the Responsiveness of the Vestibulosympathetic Reflex in Humans Introduction Methods Subjects Measurements Microneurography Hemodynamic Measurements Experimental Design Saline Infusion (Trial 1) Nitroprusside Infusion (Trial 2) Phenylephrine Infusion (Trial 2) Data Analysis Statistical Analysis Results Hemodynamic Responses Muscle Sympathetic Nerve Responses Discussion Mechanism Previous Studies/Clinical Implications Limitations Conclusions...22 Chapter 3 Prolonged Bed Rest Attenuates the Vestibulosympathetic Reflex: Implications for Orthostatic Intolerance Introduction Methods Experimental Design Hour HDBR...26

7 Subjects Bed Rest Protocol Prolonged HDBR Subjects Bed Rest Protocol Measurements Microneurography Limb Blood Flow Head-up Tilt Hemodynamic Measurements Data Analysis Statistical Analysis Results Hour HDBR Muscle Sympathetic Nerve Responses Hemodynamic Responses Limb Blood Flow Responses Head-up Tilt Prolonged HDBR Muscle Sympathetic Nerve Responses Hemodynamic Responses Limb Blood Flow Responses Head-up Tilt Discussion Mechanisms Clinical Implications/Significance Summary...43 Chapter 4 Glycerol-Induced Fluid Shift Attenuates the Vestibulosympathetic Reflex in Humans Introduction Methods Subjects Experimental Design Vestibular Activation Drug Administration Measurements Microneurography Limb Blood Flow Hemodynamic Data Analysis Statistical Analysis Results Saline Trial...52 vii

8 4.3.2 Glycerol Trial Discussion Mechanisms Clinical Implications Summary...59 Chapter 5 Discussion Conclusions Future Directions...62 Bibliography...64 Appendix A Raw Data for Baroreflex Loading Study (Chapter 2)...72 Appendix B Raw Data for Prolonged Bed Rest Study (Chapter 3)...91 Appendix C Raw Data for Glycerol-Fluid Shift Study (Chapter 4) viii

9 LIST OF FIGURES ix Figure 1.1: Diagram of short-term blood pressure mechanisms 2 Figure 2.1: Representative neurogram during infusion trials 14 Figure 2.2: MSNA burst frequency and total activity during infusions trial and HDR. 15 Figure 2.3: Change in total activity during HDR for each infusion. 16 Figure 3.1: MSNA burst frequency and total activity during HDR before and after 24-hr HDBR.. 31 Figure 3.2: Change in calf blood flow and calf vascular conductance during HDR before and after 24-hr HDBR 33 Figure 3.3: Individual responses for head-up tilt duration and change in MSNA during HDR following 24-hr HDBR. 34 Figure 3.4: Representative neurogram MSNA response to HDR before and after prolonged HDBR Figure 3.5: MSNA burst frequency and total activity during HDR before and after prolonged HDBR Figure 3.6: Change in popliteal blood flow and leg vascular conductance during HDR before and after prolonged HDBR Figure 3.7: Individual responses for head-up tilt duration and change in MSNA during HDR following prolonged HDBR. 39 Figure 4.1: MSNA burst frequency and total activity during HDR before and after drug administrations of saline and glycerol.. 52 Figure 4.2: Change in calf vascular conductance during HDR before and after administration of saline and glycerol 53 Figure 4.3: MSNA burst frequency during cold pressor test before and after administration of saline and glycerol 54

10 LIST OF TABLES x Table 2.1: Hemodynamic responses during each infusion trial 13 Table 3.1: Hemodynamic responses to HDR following HDBR 32 Table 4.1: Drug administration volume alterations and hemodynamic responses to HDR... 51

11 ACKNOWLEDGEMENTS xi The author would like to thank the following individuals for their guidance, support, encouragement, and participation throughout this endeavor. Doctoral Committee Dr. Chester Ray Dr. Charles Lang Dr. James Pawelczyk Dr. Lawrence Sinoway Experimental Assistance Charity Sauder Nathan Kuipers Matthew Kearney Amy Fogelman Erin Muldoon Thad Wilson Kevin Monahan GCRC Nurses and Staff Family and Friends The author would like to thank all of his family and friends for their support, encouragement, and understanding through these long years and sometime difficult times. Without their help, none of this could be accomplished. The author pays special consideration to his parents, Dennis and Julia Dyckman. Research Volunteers The author would like to pay special thanks to all the subjects that participated in this study. Without their participation this research endeavor could not have been completed.

12 1 Chapter 1 Introduction Blood pressure regulation occurs through a series of complex short-term and long-term mechanisms. Short-term regulation of blood pressure is regulated by sympathetic activity. Sympathetic nerve activity is regulated through a variety of reflexes demonstrated in Figure 1.1. Long-term blood pressure regulation occurs through renal mechanisms of fluid control while acute changes in blood pressure are primarily regulated by the arterial baroreflex. In addition to the arterial baroreflex, other contributory reflexes, the muscle mechanoreflex, low-pressure cardiopulmonary baroreflexes, and central and peripheral chemoreflexes, modify acute blood pressure regulation. These regulatory mechanisms converge in the brain stem, are integrated with central command, and modify sympathetic outflow, which elicits changes to peripheral vasoconstriction and cardiac output. The product of cardiac output and systemic vascular resistance determines mean arterial pressure. Recently, another regulatory reflex has been demonstrated to contribute to blood pressure regulation, the vestibulosympathetic reflex.

13 2 Figure 1.1: Diagram of short-term blood pressure regulatory mechanisms. Afferent nerve signals are generated through the vestibular apparatus, arterial baroreceptors and chemoreceptors, low-pressure cardiopulmonary baroreceptors, and skeletal muscle mechanoreceptors. These converge on the brain stem and are integrated with central command. Sympathetic outflow is generated both to the heart, for contractility and rate, and the periphery, for changes in vascular resistance. 1.1 Vestibular Contributions to Blood Pressure Regulation The vestibulosympathetic reflex (VSR) is a vestibular-mediated contribution to sympathetic nerve activity. Initial studies by Doba and Reis (21) first demonstrated the presence of the VSR in anesthetized cats. Their studies transected the vestibular nerve

14 and followed with passive head-up tilts and demonstrated a loss of orthostatic blood 3 pressure control (21). Similar studies were repeated in awake cats and were further expanded to demonstrate that direct electrical stimulation of the vestibular nerve can modify sympathetic nerve traffic (42, 43, 87, 92). Through the animal studies, it appears the VSR is a powerful reflex that contributes to blood pressure regulation during position changes, especially times of orthostatic stress (88). Vestibular contributions to autonomic control of blood pressure have been evident in humans, although not clearly defined. This evidence ranges from patients with peripheral and central vestibular abnormalities manifesting symptoms and signs of autonomic dysfunction to people experiencing simple motion sickness that elicits nausea, sweating, and increased blood flow to muscles (5, 27). Demonstrating the VSR in our laboratory has been accomplished through the use of head-down rotation (33, 60, 62-64, 72). Essandoh et al. (24) first showed alterations in peripheral vasoconstriction during head-down neck flexion. Our laboratory has further demonstrated changes in muscle sympathetic nerve traffic with increasing vasoconstriction during head-down rotation (33, 61, 72). Other studies have verified the presence of the VSR in humans using both galvanic stimulation (3) and off-axis rotation (40). The VSR appears to be otolithmediated, therefore, dependent on gravitational inputs. The purpose of this research is to clarify the role of the VSR in humans, especially during conditions that are known to produce orthostatic hypotension.

15 1.2 Relation to Orthostasis 4 Orthostatic intolerance is the inability to maintain blood pressure and cerebral perfusion upon standing. Orthostatic hypotension is nearly synonymous with intolerance, although hypotension is a decrease in blood pressure during orthostatic stress that leads to intolerance. This loss of blood pressure control during standing is a significant problem in the elderly, demonstrating increased morbidity and mortality (50). Orthostatic intolerance is also a significant problem experienced by up to two-thirds (10) of all astronauts upon returning to earth. The mechanism of this condition is not clearly understood. Data exist that demonstrate many factors contribute to this problem ranging from hypovolemia, changes in baroreceptor activation, sympathetic withdrawal, and an inability to increase peripheral vascular resistance (10, 18, 23, 26, 34, 83). Studies have demonstrated that the most severe cases of orthostatic intolerance result from an inability to augment total peripheral resistance (10, 26, 47, 83). Since mechanism for development of orthostatic intolerance post spaceflight is not completely understood, effective countermeasures to prevent orthostatic intolerance are still unavailable. In addition, vestibular alterations are known to occur in microgravity contributing to space motion sickness and spatial perception changes because the vestibular system is gravity dependent (20, 32, 52, 91). At least 50% of astronauts experience space motion sickness in the first few days (20). Characterization of the vestibular changes in humans included altered otolith-ocular and otolith-spinal reflexes (52). Head-down bed rest (HDBR) serves as a model to study cardiovascular alterations that occur as a result of spaceflight and HDBR has been associated with an increased incidence of orthostatic

16 5 intolerance (25). A study was able to demonstrate that otolith function was modified in a thirty-day bed rest in older men, which showed altered nystagmic pattern and dissociation of the components of the caloric reaction (29). The VSR has not been investigated as a possible contributory mechanism for the development of orthostatic intolerance. The purpose of these studies is to investigate the role of the VSR in blood pressure regulation in humans. Specifically, these studies will characterize the VSR in humans during various physiological stressors that have been associated with developing orthostatic hypotension.

17 6 1.3 Specific Aims and Hypotheses Aim #1 To determine if baroreflex loading (acute hypertension) changes the response of the VSR. Hypothesis #1 The ability of the VSR to modulate MSNA would be maintained during conditions where hypotension risk is elevated (baroreflex unloading), but would be attenuated during baroreflex loading. Aim #2 To determine the response of the VSR during simulated microgravity, 6 o head down bed rest (HDBR). Hypothesis #1 The VSR would be preserved following short-term HDBR and no relationship between the VSR and orthostatic intolerance. Hypothesis #2 The VSR would be attenuated following prolonged HDBR and would be associated with changes in orthostatic intolerance. Aim #3 To determine if an acute fluid shift via glycerol dehydration could modify the response of the VSR. Hypothesis #1 Acute glycerol dehydration will attenuate the ability of the VSR.

18 7 Chapter 2 Effect of Baroreflex Loading on the Responsiveness of the Vestibulosympathetic Reflex in Humans 2.1 Introduction Animal studies have established the existence of a powerful vestibular-mediated reflex that contributes critically to the maintenance of arterial blood pressure (BP) in the upright posture. Doba and Reis (21) first reported that bilateral transection of the vestibular nerve in cats resulted in persistent hypotension during upright tilt. Subsequent studies demonstrated that direct electrical stimulation of the vestibular nerve elicits pronounced effects on sympathetic nervous system outflow and vascular resistance in the cat (42, 43, 87, 92). Thus, it appears that this vestibular-mediated reflex exerts its effect in part via the sympathetic arm of the autonomic nervous system. Studies in humans have provided further support for the existence of a vestibulosympathetic reflex (VSR) (5, 40, 61, 91). Using head-down rotation (HDR) as a model to activate the vestibular otoliths, direct measurement of sympathetic outflow (muscle sympathetic nerve activity; MSNA) has been repeatedly demonstrated to increase and elicit peripheral vasoconstriction (33, 60, 62-64, 72). We have previously demonstrated that sympathetic activation during HDR is independent of central command, neck muscle afferents, visual inputs or other non-specific receptors activated during head movements (62-64, 72). Thus it appears that during the transition from the supine to upright posture vestibular activation contributes to BP regulation. Importantly,

19 8 this integrative response likely involves other powerful neurocardiovascular reflexes such as the baroreflexes. Therefore, orthostatic BP regulation depends on the interaction of various neurocardiovascular reflexes (12, 76). Previously, we have established that the ability of the VSR to increase MSNA during vestibular activation (HDR) is well preserved during orthostatic stress imposed using lower-body negative pressure (LBNP) (60). These data suggest that baroreflex unloading does not modulate the sensitivity of the VSR. Moreover, as the ability of the VSR to modulate MSNA is maintained during baroreflex unloading these data are consistent with the concept that the VSR is a powerful reflex that defends against hypotension (21, 36, 61). Whether this ability of the VSR to modulate MSNA persists during baroreflex loading is unknown. Previous animal studies suggest that raising BP (baroreceptor loading) attenuates the vestibulosympathetic responses (43). Thus, based on these previous studies, we developed and tested the hypothesis that the ability of the VSR to modulate MSNA would be maintained during conditions where hypotension (cerebral hypoperfusion) risk is elevated (baroreflex unloading), but not when it is decreased (baroreflex loading). 2.2 Methods Subjects Ten young healthy volunteers (5 men and 5 women; age: 26±1 years; height: 176.0±2.6 cm; weight: 68.4±3.8 kg) participated in the study. All subjects were non-

20 smokers, non-obese, normotensive, and not taking any medications that may influence 9 the results of the study. The Institutional Review Board of the Pennsylvania State University College of Medicine approved the experiment and written informed consent was obtained from all subjects prior to testing Measurements Microneurography Multifiber recordings of MSNA were obtained from a tungsten microelectrode inserted in the peroneal nerve behind or lateral to the knee, as previously described (60). A reference electrode was placed subcutaneously 2-3 cm from the recording electrode. Previously identified criteria for an adequate MSNA signal were applied to ensure proper recording (79). The nerve signal was amplified (20,000 50,000 times), fed through a bandpass filter with a band width of 700-2,000 Hz, integrated using a 0.1 s time constant (University of Iowa Bioengineering, Iowa City, IA), and recorded digitally (16SP Powerlab, ADInstruments, New Castle, Australia). The mean voltage neurogram was routed to a computer screen and a loudspeaker for monitoring during the study. Sympathetic recordings that demonstrated possible electrode site shifts, altered respiratory patterns (e.g., breath holding, inspiratory gasp, and hyperventilation), or electromyographic artifact during experimental intervention were excluded from analysis.

21 Hemodynamic Measurements 10 Heart rate was derived from an electrocardiogram. BP was measured continuously by a finger photoplethysmography (Finapres, Ohmeda, Englewood, CO) during each trial. Respiration pattern was measured using impedance plethysmography Experimental Design The purpose of this study was to determine if vestibular (i.e., otolith organ) activation elicits increases in MSNA during baroreceptor loading and unloading. Subjects were instrumented for the study (BP, heart rate, and respiration) and a catheter was inserted in an antecubital vein. Then, subjects were placed in the prone position and an appropriate MSNA recording site was established. The protocol consisted of 3 individual trials (Trials 1, 2, and 3). The trials were not randomized, but each trial was separated by at least 15 minutes, where resting BP and MSNA returned to resting values. All experiments were performed in a dimly lit, quiet laboratory maintained at C Saline Infusion (Trial 1) During this trial, which served as a control trial, saline was infused intravenously throughout. Subjects performed HDR in the prone position as previously described (72). Briefly, after a 3-minute baseline period with the head in the baseline chin-up neckextended position, the chin support was removed and the head was passively rotated to

22 11 the point of maximal rotation. This position was maintained for 3 minutes followed by the subject s head being returned to the baseline chin-up neck-extended position for 3 minutes of recovery. This trial was therefore 9 minutes in duration Nitroprusside Infusion (Trial 2) Nitroprusside was infused intravenously during this trial. It was performed identically to Trial 1 except in the ~10 minute period before the baseline data collection nitroprusside was titrated to induce a sustained decrease in mean arterial pressure of ~10 mmhg. To accomplish this, nitroprusside infusion commenced at a dose of 0.2 mg/kg/min for 3 minutes. After this period the dose was titrated upwards by 0.2 µg/kg/min every 3-minutes until the desired effect on BP was obtained ( µg/kg/min). Three minutes after obtaining a sufficient sustained decrease in BP, the baseline period began followed by 3 minutes of HDR and subsequently a 3-minute period of recovery. Once the desired dose of nitroprusside was determined that infusion rate was continued until the end of the trial (end of the 3 minute recovery period). Subsequent trials did not commence until heart rate and BP returned to baseline (minimum of 15 minutes) Phenylephrine Infusion (Trial 2) Trial 3 was identical to Trial 2; with the exception that phenylephrine was infused instead of nitroprusside. The intravenous infusion of phenylephrine commenced at a rate

23 12 of 0.2 µg/kg/min. After 3 minutes this dose was increased in 0.2 µg/kg/min increments at 3-minute intervals until a mean BP increased ~10 mmhg ( µg/kg/min). Once this dose was established the infusion was continued until the HDR protocol was complete (3 minute baseline period followed by 3 minutes of HDR and 3 minutes of recovery) Data Analysis All data were digitally recorded at 100 Hz for later off line analysis. MSNA was expressed as bursts per minute and total activity (sum of area underlying individual bursts per minute). Sympathetic bursts were identified from the mean voltage neurogram and the sum of the area under each burst, expressed in arbitrary units (au) was assessed by a computer program (Chart 5, ADInstruments). Each neurogram was normalized by assigning the tallest burst an amplitude of 1000 and by setting the baseline to zero during the resting (baseline) portion of the saline (control) trial. MSNA comparisons were made between the three-minute average resting (baseline) level and during the first minute of HDR. Blood pressure and heart rate were also averaged per minute Statistical Analysis To identify possible differences between each trial infusion, a two-within (infusion trial, intervention), repeated-measures analysis of variance (ANOVA) was used. Tests for simple effects were used to identify if there were differences in baseline when

24 the interaction term was significant (41). A significance level of p<0.05 was used for all tests. Values are presented as mean±se Results Hemodynamic Responses Hemodynamic changes are listed in Table 2.1. Saline Infusion SNP Infusion PE Infusion Variable Baseline HDR Baseline HDR Baseline HDR Systolic (mmhg) 128±4 127±4 126±5 127±5 136±2* 134±3 Diastolic (mmhg) 76±3 75±3 70±3* 70±4 83±3* 83±3 PP (mmhg) 52±4 52±4 56±4 57±4 53±3 51±3 MAP (mmhg) 91±2 89±3 84±3* 84±4 99±3* 99±3 HR (beats/min) 63±2 64±2 76±3* 77±3 57±2* 57±2 * P<0.001 vs. saline infusion baseline, P<0.001 vs. SNP baseline. Values expressed as mean ± SE. SNP (Nitroprusside), PE (Phenylephrine), PP (Pulse Pressure), MAP (Mean Arterial Pressure), HR (Heart Rate). The saline infusion trial served as the control to compare to the drug infusions. Nitroprusside infusion decreased mean arterial pressure (91±2 to 84±3 mmhg; p<0.001) at rest (baseline). This was associated with a compensatory increase in heart rate (63±2 to

25 14 76±3 beats/min; p<0.001) at rest (baseline). Phenylephrine infusion increased mean arterial pressure (91±2 to 99±3 mmhg; p<0.001) and decreased heart rate (63±2 to 57±2 beats/min; p<0.001) at rest (baseline) Muscle Sympathetic Nerve Responses Figure 2.1. Representative neurograms for each trial from a single subject are presented in

26 15 Figure 2.1: Representative neurogram of one test subject during each infusion trial. Baselines and HDR time periods are 30 sec. Resting (baseline) MSNA levels were increased between the saline and nitroprusside (SNP) infusion trials, but were decreased between the saline and phenylephrine (PE) infusion trials. MSNA increased significantly between baseline and HDR during the saline and nitroprusside infusion trials, but did not increase significantly during the phenylephrine infusion trial. The scales are the same for each neurogram. Nitroprusside infusion increased MSNA burst frequency (13±2 vs. 26±3 bursts/min before and during infusion, respectively; p<0.001) and total MSNA (699±129 vs. 1815±201 au; p<0.001) at rest (baseline). Phenylephrine infusion decreased MSNA burst frequency (13±2 vs. 2±1 bursts/min; p<0.001) and total MSNA (699±129 to 112±66 au; p<0.001) (Figure 2.2). Figure 2.2: MSNA burst frequency (bursts/min) and total activity (a.u.) measured during each infusion and HDR. MSNA increased significantly during the saline infusion and the nitroprusside (SNP) infusion, but did not increase significantly during the phenylephrine (PE) infusion. * P<0.05 vs. baseline.

27 16 MSNA increased during HDR in the saline ( 3±2 bursts/min, 314±154 au total activity, 41±18% total activity; p<0.05) and nitroprusside trials ( 5±2 bursts/min, 713±241 au total activity, 49±20% total activity; p<0.05) (Figure 2.2 and 2.3). In contrast, HDR performed during the phenylephrine trial did not increase MSNA ( 0±1 bursts/min, -15±33 au total activity, -8±21% total activity) (Figure 2.2 and 2.3). Figure 2.3: Change in total activity (% of baseline) during HDR for each infusion. No differences were noted between the saline and nitroprusside (SNP) infusion trials. Phenylephrine (PE) infusion trial was significantly different than the saline trial. * P<0.05 vs. saline trial (control).

28 2.4 Discussion 17 The major finding from the present study is that the ability of the VSR to elicit increases in MSNA through HDR is abolished during baroreflex loading. These data are consistent with prior animal studies (43) and provide further experimental support for the concept that the VSR is a powerful reflex system capable of defending against acute risks associated with hypotensive challenges (40, 82, 91, 92). Our study demonstrates that during the steady-state infusion of nitroprusside, MSNA increases further during HDR (similar to the saline infusion), despite elevated MSNA at rest. Previously, we established that the ability of the VSR to elicit increases in MSNA through HDR was unaltered during baroreflex unloading induced using lower body negative pressure (60). Additionally, the preserved ability of the VSR to elicit increases in MSNA during HDR during baroreflex unloading suggests that the neural pathways mediating increases in MSNA are sufficiently distinct such that unloading of the baroreflex does not influence the magnitude of response of the VSR. Moreover, these data obtained during baroreceptor unloading are consistent with the observations that HDR performed during head-up-tilt is associated with increases in systemic vascular resistance in subjects with neurogenic orthostatic hypotension (7) Mechanism In contrast to these observed responses during baroreflex unloading, the ability of the VSR to mediate increases in MSNA during baroreflex loading was prevented. In fact when HDR was performed during phenylephrine infusion, where the resting level of

29 MSNA was reduced, HDR was unable to elicit further increases in MSNA. The 18 mechanism(s) underlying these effects are unknown. However, animal studies have shown that the ability of the VSR to stimulate increases in sympathetic outflow was abolished after BP at rest was increased by infusion of an α-adrenergic agonist (43). These data, in contrast to the data obtained during baroreflex unloading, suggest that a reflex interaction occurs. It seems the integration between the two reflexes, especially during the two different stimuli (i.e., baroreflex unloading and loading), is unclear in humans. According to animal data, the neurons in the rostral ventrolateral medulla appear to be inhibited by baroreceptor stimulation and those neurons are needed for relaying the vestibular signals (43). Our results suggest the baroreflex stimulation (loading) causes a large elevated inhibitory signal that impairs downstream vestibular effects on the peripheral vasculature, while baroreflex unloading allows the vestibular stimulation to enhance the overall generation of MSNA. Therefore, this interaction between the reflexes (baroreflex and VSR) occurs in a manner as to not compromise BP regulation at times when hypotensive risk is high (baroreflex unloading). This study provides data to suggest this pathway could be present in humans. Other data derived from animal studies may also be important in regards to the present findings. For instance, it has been demonstrated that the VSR mediates increases in renal sympathetic nerve activity during hypergravity in rats (31). These increases were attributed to a rapidly acting vestibular-mediated feed-forward mechanism to prevent or attenuate decreases in BP when animals were subjected to gravitational stress. When both the baroreflex and VSR are intact, a modest pressor effect occurs. It was suggested that the true magnitude of the vestibular-mediated feed-forward response might be

30 underestimated due to baroreflex restraint, which was similar to the cat data and could 19 provide an explanation for our data in humans. Collectively, these findings and our discussion are consistent with the suggestion that the VSR is a reflex system designed to respond to acute hypotensive challenges and therefore, may be used as evidence for significant integration between these reflexes Previous Studies/Clinical Implications Our previous studies demonstrated preserved MSNA responses during HDR while mean arterial pressure was elevated during isometric handgrip (59) and mental stress (15). We believe our current data does not directly contradict those previous findings, although the MSNA responses differed. During exercise the baroreflex is reset around a higher prevailing level of blood pressure (58). In contrast, during phenylephrine infusion we would not expect such an effect. This may help explain why the ability of the VSR to modulate increases in MSNA is preserved during exercise, but not baroreflex loading. Additionally, Anderson et al. (1) demonstrated that mental stress was able to increase MSNA further when MSNA was suppressed and blood pressure was elevated during a steady-state phenylephrine infusion. Our previous study also demonstrated an additive interaction between mental stress and the VSR (13, 15). However, mental stress could possibly reset the baroreflex operating point, similar to what could occur during exercise (14). Thus, it may be important to consider the nature of the stimulus applied as well as any potential effect of the stimulus on the set point of the baroreflex.

31 In contrast to the compelling and definitive body of experimental evidence 20 available from animals, there is less evidence that vestibular activation contributes directly to BP control in humans. More definitive evidence for a critical role of the vestibular system in orthostatic BP control in humans may be obtained in patients with altered vestibular inputs, through vestibular damage (86). These patients experience symptoms associated with orthostasis that can result in lightheadedness or presyncope (90). Yates et al. (89) demonstrated an attenuated increase in BP in vestibular-deficient patients during linear acceleration as compared to healthy controls. These results demonstrate that the loss of vestibular inputs blunt the increases in BP observed during gravitational stress. Consistent with this concept, when older adults perform HDR, the resultant increase in MSNA is blunted as compared to responses observed in young adults (65). Interestingly, not only is the increase in MSNA during HDR blunted, but this response occurs in the presence of a decreased systemic BP (65). In addition, Wilson et al. (85) demonstrated that HDR attenuated an increase in cerebral vascular resistance only during lower-body negative pressure suggesting that the VSR acts to redistribute blood flow throughout the body to maintain consciousness especially in times of orthostatic stress Limitations Several limitations deserve mention. First, using pharmacological substances to load and unload baroreceptors introduces unavoidable criticism that these substances directly or indirectly influence the results. However, we do not believe this is the case as

32 21 the observed responses to HDR during the nitroprusside infusion provided results that are nearly identical to our previous data in which no pharmacological substances were used to unload the baroreflex (60). We cannot exclude the possibility that phenylephrine did not exert some direct effect on responses independent of baroreceptor loading. However, Somers et al. (75) utilizing steady-state phenylephrine infusions to examine the interaction of the baroreflex and chemoreflex was able to demonstrate increased MSNA during hypercapnia and a cold-pressor test despite the steady-state phenylephrine infusion, demonstrating the sympathetic nerve responses were still intact. In addition, despite decreased MSNA at rest during the phenylephrine infusion, our results show a tendency to decrease in total activity (au) during HDR. This observation is consistent with the results by Somers et al. (75), which demonstrated the combination of hypoxia and phenylephrine infusion tended to decrease in total activity. It is noted that the trials were performed in the same order. Repeated HDR has demonstrated a consistent and similar increase in MSNA (33). Also, the drugs used in the present study have a short half-life and the time between trials allowed MSNA and blood pressure to return to baseline levels. In addition, another limitation could be with the use of HDR to stimulate the VSR it is difficult to quantify the stimulus to the otolith organs. To minimize this concern, repeated HDR maneuvers were performed to the same degree of head rotation within subjects. Additionally, HDR is a complex stimulus that activates many different sensory receptors. However, because previous studies have demonstrated that other inputs during HDR (such as neck afferents, baroreceptors, central command, visual inputs) do not influence MSNA responses (62-64, 72), it is likely that the changes in MSNA we observed are directly attributable to stimulation of the VSR through HDR. It

33 is possible that some of these other sensory inputs could have an impact on the 22 integration of the baroreflex and VSR. Finally, studying vestibular deficient patients may provide more definitive insight into the interaction between the VSR and the baroreflexes Conclusions In summary, VSR-mediated increases in MSNA during baroreflex unloading are preserved whereas they are abolished during baroreflex loading. These results are consistent with prior animal studies. Collectively, these data provide further evidence for the concept that the VSR is a powerful neurocardiovascular reflex that is particularly important at times when immediate homeostatic control of the organism is most at risk (i.e., during acute hypotensive challenge).

34 23 Chapter 3 Prolonged Bed Rest Attenuates the Vestibulosympathetic Reflex: Implications for Orthostatic Intolerance 3.1 Introduction Orthostatic intolerance (OI) is the inability to maintain blood pressure and cerebral perfusion while in the upright position. Moreover, the failure to maintain blood pressure in the upright posture is associated with increased mortality (50). Head-down bed rest (HDBR) is used to simulate microgravity, physical deconditioning, and to elicit OI (25). Mechanisms believed to contribute to the development of OI with HDBR include the following: impaired vagal baroreflex responses (17), an inadequate increase in sympathetic discharge (38, 71), cardiac atrophy (48), an inability to increase peripheral vascular resistance (10, 49), and hypovolemia (57). Studies clearly indicate that alterations in sympathetic nerve activity contribute to the development of OI (25, 38, 49, 71). Hypotensive episodes during head-up tilt appear to be closely related to lack of increased MSNA (49). It has been reported that MSNA is attenuated during tilt following 14 days of HDBR in subjects that experienced OI (71). Despite considerable research on this topic, the mechanisms for OI have been enigmatic. One mechanism for OI after bed rest that has not been examined is alterations in the vestibulosympathetic reflex (VSR). Data exist demonstrating the presence of a vestibular-mediated sympathetic reflex that contributes to blood pressure regulation in both humans and animals (21, 33, 40, 72, 91, 92). Bilateral transection of the vestibular nerve results in persistent hypotension

35 during upright tilt in the cat (21). Additionally, direct electrical stimulation of the 24 vestibular nerve elicits sympathetic nerve activation and increased vascular resistance in the cat (42, 43, 87, 92). In humans, using head-down rotation (HDR) as a model to activate the vestibular otolith organs, we have demonstrated marked increases in MSNA and peripheral vasoconstriction (33, 60, 62-64, 72). Other studies have supported the presence of the VSR in humans (3, 24, 40, 82). Studies examining changes to the vestibular system following bed rest are sparse. Altered nystamic pattern indicating changes in vestibular function have been observed after seven days of HDBR (11). However, currently no studies have examined if the VSR is altered following HDBR and possibly contribute to OI. The purpose of this study was to examine the vestibulosympathetic reflex and OI following short-duration and prolonged 6 o HDBR. It was hypothesized that the MSNA response to HDR would be attenuated following prolonged HDBR and would be associated with changes in OI. There would be no relationship between the VSR and OI following short-term HDBR because the lack of adequate time to elicit functional changes in the vestibular system. The results of this study support our hypothesis and the concept that alterations in the VSR contribute to OI in humans following HDBR. 3.2 Methods The Institutional Review Board of the Pennsylvania State University College of Medicine and the University of Texas Medical Branch in Galveston, TX (Study 2),

36 approved all experiments and written informed consent was obtained from all subjects before testing Experimental Design Two studies were performed. One study consisted of 24-hour HDBR conducted at the General Clinical Research Center at the Milton S. Hershey Medical Center in Hershey, PA. The second study consisted of prolonged HDBR (36±1 days) and was conducted in association with the University of Texas Medical Branch Flight Analogs Research Center, in Galveston, TX. All subjects were instrumented for the study (BP, heart rate, and respiration) and then were placed in the prone position after which an MSNA recording site was established. All experiments were performed in a dimly lit, quiet laboratory maintained at C. Vestibular otolith activation, tested in both studies before and after HDBR, consisted of subjects performing HDR in the prone position as previously described (72). This maneuver engages the otolith organs, but not the semicircular canals when the head becomes stationary. Previous studies have eliminated other possible factors contributing to sympathetic activation during HDR (33, 59, 60, 62, 72). After a 3-minute baseline period with the head in the chin-up neck extended position, the chin support was removed and the head was passively rotated to the point of maximal rotation. This position was maintained for 3 minutes followed by the subject s head being passively returned to the chin-up neck-extended position for 3 minutes of recovery. During the head movements,

37 continuous blood pressure measurements, heart rate, MSNA, and limb blood flow of the contralateral leg were recorded Hour HDBR Subjects Twenty-two young healthy volunteers (12 men and 10 women; age: 24±1 years; height: 174.4±2.6 cm; weight: 71.7±3.4 kg) participated in the study. All subjects were non-smokers, non-obese, normotensive, and not taking any medications that would influence the results of the study Bed Rest Protocol Subjects abstained from caffeine (12 hr) before and during the study. Subjects performed HDR as described above. Following the HDR protocol, the subjects perform head-up tilt (HUT) at 80 degrees for up to 30 minutes to assess their orthostatic tolerance. Following the HUT protocol, the instrumentation was removed and subjects were placed at 6 o HDBR for 24 hours. During this time, subjects were not allowed to sit-up or get out of bed. Their sleep schedule was maintained to what they received the night before testing. After 24 hours, the subjects were brought back into the testing room to undergo the same experimental protocol as before bed rest. During posttesting, the subjects did not sit-up or get out of bed rest during this time, but were brought back up to the zerodegree level for the HDR protocol.

38 3.2.3 Prolonged HDBR Subjects Eight healthy volunteers (5 men and 3 women; age: years; height: cm; weight: kg) participated in the study. A suitable nerve signal could not be obtained in one subject and was excluded from the MSNA analysis. Popliteal blood flow data was not successfully obtained in a different subject and was excluded from blood flow analysis. All subjects were nonsmokers, in good health with no history of cardiovascular, neurological, gastrointestinal, or musculoskeletal problems and were not taking any medications Bed Rest Protocol This study was part of a day HDBR study conducted at the University of Texas Medical Branch Flight Analogs Research Center in Galveston, Texas, sponsored by NASA. The subjects were tested approximately 6±1 days before HDBR and underwent HDR protocol and HUT as described in our 24-hr HDBR study. After the pretest period, the subjects were placed in 6 o HDBR for 90 days. Participants lived in a special research unit for the entire study and were fed a carefully controlled diet. Every day, they were awake for 16 hours and lights out (asleep) for 8 hours. During the bed rest time, they participated in a number of other tests run by a variety of researchers. Subjects were tested using the HDR protocol at 36±1 days of HDBR. Three subjects repeated the

39 protocol again at 75±1 days of HDBR. In addition, the subjects performed head-up-tilt at 45 days (n=3) or 60 days (n=4) of HDBR Measurements Microneurography Microneurography was used to assess MSNA as previous described.(72) Briefly, multifiber recordings of MSNA were obtained from a tungsten microelectrode inserted in the peroneal nerve behind the knee. A reference electrode was placed subcutaneously 2-3 cm from the recording electrode. Previously identified criteria for an adequate MSNA signal were applied to ensure proper recording (79). The nerve signal was amplified (20,000 40,000 times), fed though a bandpass filter with a band width of 700-2,000 Hz, integrated using a 0.1 s time constant (University of Iowa Bioengineering, Iowa City, IA), and recorded digitally (16SP Powerlab, ADInstruments, New Castle, Australia). The mean voltage neurogram was routed to a computer screen and a loudspeaker for monitoring during the study. Sympathetic recordings that demonstrated possible electrode site shifts, altered respiratory patterns (e.g., breath holding, inspiratory gasp, and hyperventilation), or electromyographic artifact during experimental intervention were excluded from analysis.

40 Limb Blood Flow 29 Leg blood flow was assessed in two ways. Venous occlusion plethysmography (Hokanson EC 4 plethysmography, D.E. Hokanson, Bellevue, WA) was used to assess calf blood flow of the contralateral leg (24 hr HDBR), as previously described (33) and Doppler ultrasound (Phillips Medical Systems, iu22, with a 5-12 MHz linear probe) was used to assess popliteal blood flow (Prolonged HDBR). Limb vascular conductance was calculated as the ratio of limb blood flow to mean arterial pressure. Using Doppler ultrasound, popliteal blood flow was determined by mean blood velocity and diameter of the vessel. Leg vascular conductance was calculated as the ratio of popliteal blood flow to mean arterial pressure Head-up Tilt Passive HUT was performed at 80 o for up to 30 minutes to assess OI. HUT was performed during pretesting, after 24-hours HDBR, and either at 45 days (n=3) or 60 days (n=4) of HDBR. The tilt test was terminated if the subject completed 30 minutes of head-up tilt, the subjects began to feel presyncopal symptoms including lightheadedness, nausea, vomiting, excessive heat, or sweating, the subjects systolic blood pressure decreased >25 mmhg or below 70 mmhg, or the subjects diastolic blood pressure decreased >15 mmhg.

41 Hemodynamic Measurements 30 Heart rate was derived from an electrocardiogram. Arterial pressure was measured continuously by a finger photoplethysmography (Finapres, Ohmeda, Englewood, CO) during each trial. Respiration pattern was measured using impedance plethysmography Data Analysis All data were digitally recorded at 100 Hz for later off line analysis. MSNA was expressed as bursts per minute and total activity (sum of the area of each individual bursts expressed in arbitrary units). Neurograms were normalized using the highest burst amplitude set to 1000 and the baseline set to zero. Sympathetic bursts were identified from the mean voltage neurogram by visual inspection and by using a computer program (Chart 5, ADInstruments) Statistical Analysis To identify possible differences within each bed rest trial, a two-way (HDBR, intervention) repeated-measures analysis of variance (ANOVA) was used. Tests for simple effects were used to identify differences in baseline.(41) MSNA responses to HDR were made between the average resting (baseline) value and the first minute of the maneuver. Data from 3 subjects at day 75 was not used in the statistical analysis. A significance level of p<0.05 was used for all tests. Values are presented as mean±sem.

42 3.3 Results Hour HDBR Muscle Sympathetic Nerve Responses The effect of 24-hour HDBR on the VSR is shown in Figure 3.1. Figure 3.1: MSNA burst frequency (bursts/min) and change in total activity (%) during HDR before and after 24-hr HDBR. Burst frequency significantly increased during HDR both before and after HDBR. Baseline MSNA was significantly increased following HDBR. The change in total activity was significantly increased during HDR, but was not significantly different from before HDBR. * P<0.001 compared to baseline, P<0.01 compared to Pre HDBR, NS (not significant). Before HDBR, MSNA significantly increased during HDR in both burst frequency (9±1 to 14±2 bursts/min, 5±1 bursts/min; p<0.0001) and total activity ( 83±19%; p<0.0001). After HDBR, MSNA at rest significantly increased (14±2 bursts/min; p<0.0001). Despite elevated baseline MSNA at rest, HDR significantly

43 32 increased MSNA in both burst frequency (19±2 bursts/min, 5±1 bursts/min; p<0.0001) and total activity ( 65±18% total activity; p<0.002). Thus there were no significant differences in the MSNA response to HDR before and after 24-hr HDBR (Figure 3.1) Hemodynamic Responses Mean arterial pressure and heart rate during HDR did not change significantly and were not significantly different before and after HDBR (Table 3.1). Table 3.1: Hemodynamic responses to HDR following HDBR. Study 1 Study 2 Variable Pre 24hr HDBR Post 24hr HDBR Pre- HDBR Day 36±1 HDBR MAP (mmhg) Baseline 87±3 90±4 101±4 94±5* HDR 88±1 90±4 101±4 94±5 (BL vs. HDR) 1±1 0±1 0±1 0±1 HR (beats/min) Baseline 57±2 61±2 68±5 73±4 HDR 60±2 63±2 70±5 78±4 (BL vs. HDR) 3±2 2±1 2±1 5±1 P<0.05, * P<0.001 compared to pre-hdbr value. Values expressed as mean ± SE. No significant differences between before and after 24-hr HDBR. BL (Baseline)

44 Limb Blood Flow Responses 33 Venous occlusion plethysmography was measured on eight subjects and the results are shown in Figure 2. Before HDBR, calf vascular conductance decreased ( 15±9%; p<0.05) during HDR. After 24-hr HDBR, calf vascular conductance decreased ( 22±5%; p<0.01) during HDR. There was no significant difference in calf vascular conductance before and after HDBR (Figure 3.2). Figure 3.2: Change in calf blood flow (%) and calf vascular conductance (%, blood flow/map) during HDR before and after 24-hrs of HDBR. Both blood flow and calf vascular conductance significantly decreased during HDR both before and after HDBR (P<0.05). Flow and conductance were not significantly different following HDBR. NS (not significant) Head-up Tilt HUT duration before and after HDBR is shown in Figure 3.3. Thirteen subjects decreased, four subjects increased, and five subjects had no change in tilt time after

45 HDBR. There was no relation between the VSR and tilt time before and after 24-hr HDBR (Figure 3.3). 34 Figure 3.3: Individual responses for head-up tilt duration and change in MSNA (total activity) during HDR following 24-hr HDBR. Thirteen subjects decreased, four subjects increased, and five subjects had no change in tilt time from pre to post. Fifteen subjects decreased and seven subjects increased their change in MSNA (total activity) during HDR following 24-hr HDBR Prolonged HDBR Figure 3.4 is a representative neurogram of the MSNA response to HDR before and after prolonged HDBR in one subject.

46 35 Figure 3.4: Representative neurogram of one test subject before and after 36 days of HDBR. Baseline MSNA at rest was increased following HDBR. MSNA increased significantly during HDR before HDBR, but at 36 days, the increase in MSNA was attenuated. The y-axis is the burst activity measured is arbitrary units and is the same between the before and after neurograms Muscle Sympathetic Nerve Responses Before HDBR, MSNA significantly increased in both burst frequency (13±3 to 21±4 burst/min, 8±2 bursts/min; p<0.05) and total activity ( 83±12%; p<0.01) during HDR (Figure 3.5).

47 36 Figure 3.5: MSNA burst frequency (bursts/min) and change in total activity (%) during HDR before and after prolonged HDBR. MSNA significantly increased during HDR before HDBR. Baseline MSNA was significantly increased following HDBR. MSNA was not significantly increased during HDR at 36±1 days. * P<0.05 compared to baseline. P<0.05 compared to Pre HDBR. At 36±1 days of HDBR, the baseline MSNA at rest was increased from 13±3 to 22±4 bursts/min (p<0.05). Unlike 24-hr HDBR, the increase in MSNA to HDR was attenuated and was not significantly increased in either burst frequency ( 3±2 bursts/min, 23±13%) or total activity ( 34±22%) after prolonged HDBR (Figure 3.5). Additionally, three subjects were retested at 75±1 days of HDBR with the results being similar to HDBR at 36±1 days. Baseline MSNA at rest increased further from 22±4 bursts/min at day 36 to 29±8 bursts/min at day 75. The increase in MSNA to HDR remained attenuated in these subjects ( 23±13% bursts/min).

48 Hemodynamic Responses 37 Before HDBR, mean arterial pressure and heart rate did not significantly change during HDR (Table 3.1). Following prolonged HDBR, mean arterial pressure at rest decreased from 101±4 to 94±5 mmhg (p<0.001), but did not significantly change during HDR. Baseline heart rate increased (68±5 to 73±4 beats/min; p<0.01) and the heart rate response to HDR was significantly increased ( 5±1 beats/min, Table 3.1) after HDBR Limb Blood Flow Responses Before HDBR, Doppler ultrasound of the popliteal artery demonstrated a tendency to decrease popliteal blood flow (117±32 to 101±27 ml/min; p=0.07) and popliteal vascular conductance ( 13±5%; p=0.09) during HDR. At 36±1 days of HDBR, baseline popliteal blood flow at rest tended to be decreased (100±17 ml/min; p=0.06). However, popliteal blood flow and vascular conductance did not significantly decrease during HDR (Figure 3.6).

49 38 Figure 3.6: Change in popliteal blood flow and leg vascular conductance (%, blood flow/map) during HDR before and after prolonged HDBR. Blood flow and vascular conductance tended to decrease before bed rest (p=0.07, p=0.09, respectively) Head-up Tilt 3.7. HUT duration before and after HDBR is shown for individual subjects in Figure

50 39 Figure 3.7: Individual responses for head-up tilt duration and change in MSNA (total activity) during HDR following prolonged HDBR. Six subjects decreased, while only one subject had no change in tilt time from pre to post. Also, six subjects decreased and one subject had no change in MSNA (total activity) during HDR following prolonged HDBR. Six subjects decreased, whereas one subject had no change in tilt time after HDBR. Figure 3.7 demonstrates the individual changes in MSNA (total activity) during HDR before and after HDBR. Six subjects decreased their change in total activity during HDR after HDBR whereas one subject demonstrated no change. These data illustrate a relation between the decrease in VSR and decrease in HUT duration before and after prolonged HDBR. 3.4 Discussion The major novel finding from this study was that prolonged HDBR markedly attenuated the VSR and that decreases in the VSR were associated with decreased orthostatic tolerance. The results suggest a desensitization of the VSR with bed rest

51 deconditioning and that this physiological adaptation may serve as a mechanism for the development of OI observed following physical deconditioning and spaceflight Mechanisms The results in this study during HDR before HDBR are consistent with our previous findings (33, 61, 72). Despite the increased baseline MSNA at rest after 24- hours HDBR, HDR was still able to increase MSNA to the same extent as before HDBR indicating an intact VSR. In contrast, after prolonged HDBR the increase in MSNA during HDR was markedly attenuated. This observation suggests a clear reduction in the sensitivity of the VSR. The mechanism(s) by which these changes occur is unknown. Our data suggest a desensitization of the otolith-mediated reflex induced by HDBR. A possible mechanism to explain the change in the VSR with HDBR could be due to changes in the sensitivity of the vestibular apparatus through a possible fluid shift in the inner ear. This fluid shift could occur in combination with hypovolemia typically experienced during HDBR (25). Additionally, desensitization of the VSR could occur with prolonged HDBR through structural hair cell changes, which has been shown to occur in aging (4, 22, 67) and microgravity (19). We have reported that older subjects have marked attenuation in the VSR (65). Thus changes in the VSR could be a common mechanism explaining increased OI observed with bed rest deconditioning and with aging. Other explanations could include altered integration between the baroreflexes and VSR (31) or changes in the central processing of vestibular inputs (91).

52 3.4.2 Clinical Implications/Significance 41 Subjects performed HUT to assess orthostatic tolerance. After the 24-hour HDBR, some of the subjects decreased tilt time, whereas other subjects demonstrated an increase or no change in tilt time. There was no clear relation between the change in tilt time and the sensitivity of the VSR after 24-hours HDBR. However, during prolonged HDBR nearly all subjects had a reduction in tilt time and VSR. Additionally, the VSR remained attenuated in three subjects tested again at 75±1 days of HDBR. These results suggest that over time during HDBR, the VSR may play a prominent role in the development of orthostatic hypotension. Some studies have suggested a correlation between OI and changes in vestibular function after spaceflight (65, 84, 91). Kamiya et al. (38) demonstrated a paradoxical sympathetic withdrawal with less of an increase in MSNA during upright tilt in subjects that experience OI after 14 days HDBR. An altered VSR could provide an explanation for the attenuated increase in MSNA observed in their study. Microgravity has been demonstrated to elicit marked changes in the vestibular system with at least 50% of astronauts experiencing space motion sickness in the first few days (20) and altered otolith-ocular and otolith-spinal reflexes (52). Both morphological and physiological changes to the vestibular system have been demonstrated after spaceflight (19, 56, 68, 69, 81). Most changes to the vestibular system as a result of microgravity focus on function of the otolith organs, since they are dependent on gravitational inputs (20, 52, 91). Because the VSR is an otolith-mediated reflex (40, 64), it is reasonable to hypothesize

53 42 that the changes observed during HDBR could apply in microgravity and contribute to post-spaceflight OI which occurs in nearly two-thirds of all astronauts (10). A secondary observation from this study is that baseline MSNA at rest increased rapidly and was maintained throughout HDBR. Previous studies investigating baseline MSNA at rest after HDBR have revealed equivocal results. Both increases (37, 57) and decreases (70) in baseline MSNA at rest have been observed. Our results indicate an increase in baseline MSNA at both 24-hrs HDBR similar to data reported by Khan et al. (44), and following prolonged HDBR similar to data reported by Kamiya et al. (37) after both 60 and 120 days of HDBR. Elevated MSNA at rest has also been reported in spaceflight (47). These results are consistent with the concept that MSNA at rest should increase as a result of reduced cardiac filling pressures and central blood volume that is associated with prolonged HDBR (25, 48). It could be suggested that the elevated MSNA at rest following prolonged HDBR prevented a further increase in MSNA during HDR. However some factors argue against that reasoning. First, all subjects demonstrated a robust sympathetic response to apnea following HDBR (data not shown). Second, we have demonstrated in older subjects, when MSNA at rest is elevated to a much greater value then in this study, sympathetic responses can be increased through various maneuvers (51, 65). Changes in peripheral vascular conductance observed in these studies correspond with changes in sympathetic nerve responses. Other studies have noted reduced peripheral responses following bed rest (25, 38). In our study, elevated resting sympathetic activation could affect the sensitivity and responsiveness of the vasculature. However, this change in the vasculature did not occur following short-duration HDBR

54 because HDR maintained its ability to elicit significant vasoconstriction along with 43 preserved sympathetic responses despite elevated MSNA at rest. However, following prolonged HDBR the sensitivity of the vasculature could be altered. Our data demonstrate a relation between diminished MSNA and vasculature responses to HDR following prolonged HDBR. Therefore, an attenuation of the VSR could contribute to inadequate increases in limb vasoconstriction and lead to subsequent decreases in orthostatic tolerance Summary This study demonstrates that prolonged bed rest (~ 5 weeks) can attenuate the VSR. This attenuation in the VSR was associated with reduced orthostatic tolerance. These results suggest a novel mechanism in the development of OI.

55 44 Chapter 4 Glycerol-Induced Fluid Shift Attenuates the Vestibulosympathetic Reflex in Humans 4.1 Introduction Orthostatic intolerance is a significant problem experienced in the elderly (50) and in approximately two-thirds (10) of all astronauts upon returning to earth. The mechanism of this problem is not clearly understood. In astronauts, many factors contribute to this problem ranging from hypovolemia, changes in baroreceptor activation, sympathetic withdrawal, and an inability to increase peripheral vascular resistance (10, 18, 23, 26, 34, 45, 83). The most severe cases of orthostatic intolerance result from an inability to augment total peripheral vascular resistance (10, 26, 47, 83). In addition, space motion sickness is experienced by up to 80% of astronauts (66). It has been hypothesized that motion sickness develops from fluid shifts that alter intracranial, cerebrospinal fluid, or inner ear fluid pressures changing the response properties of the vestibular receptors (32, 73). Data has demonstrated the presence of a vestibularmediated sympathetic reflex that can contribute to autonomic control of sympathetic nerve activity and blood pressure (21, 33, 40, 72, 91, 92). Using head-down rotation (HDR) as a model to activate the vestibular otoliths, our laboratory has reported increased muscle sympathetic nerve activity (MSNA) and peripheral vasoconstriction with this maneuver (33, 60, 62-64, 72). Other studies in humans have reinforced this concept of the vestibulosympathetic reflex (VSR) (3, 24, 40, 82).

56 45 Ménière s disease serves as a model to understand inner ear fluid dynamics (2). Patients with Ménière s disease have excessive fluid accumulation within the membranous labyrinth of the inner ear (2, 77). This results in vestibular symptoms of vertigo, tinnitus, and hearing loss (2, 77). The medical management of the disease includes the use of diuretics, which decreases the fluid volume within the body and in the inner ear (35, 74, 77). Classically, the glycerol dehydration test (GDT) has been used to test for the presence of Ménière s disease (28, 54, 55). Glycerol is an osmotic diuretic that produces diuresis by fluid shifting water from the intracellular compartment into the extracellular vasculature, thereby producing extracellular volume expansion, which in turn increases glomerular filtration rate and diuresis (8). It has been demonstrated that glycerol can reduce intraocular pressure by decreasing CSF fluid volume (8), decreasing intracranial pressure (9), and causing fluid shifts within the inner ear in humans (54, 77). Glycerol can also modify the electrolyte concentration of both the perilymph and endolymph within the inner ear (39, 54, 78). Currently, no studies have investigated the direct effect of hypovolemia or fluid shifts of the inner ear on the VSR in humans. The primary goal of this study was to test if an acute fluid shift within the vestibular apparatus, through the use of glycerol, can affect the VSR. We hypothesized acute glycerol dehydration would decrease the sensitivity of the VSR and that this attenuated increase in MSNA during HDR would lead to reduced peripheral vasoconstriction.

57 4.2 Methods Subjects Nine young healthy male volunteers (age: 27±1 years; height: 180.7±2.5 cm; weight: 81.3±3.5 kg) participated in the study. Female subjects were not used in order to eliminate possible hormonal concentration alterations from dehydration. All subjects were non-smokers, non-obese, normotensive, and not taking any medications that may influence the results of the study. The Institutional Review Board of the Pennsylvania State University College of Medicine approved the experiment and written informed consent was obtained from all subjects prior to testing Experimental Design HDR was performed twice (saline and glycerol intervention) on two separate days in a randomized, crossover controlled design, with a time interval of at least one week. A testing day consisted of a control HDR trial followed by drug administration (saline or glycerol), waiting an hour, and then a second HDR trial. The order of drug administration was randomized across the two testing days. During the one-hour break between the two HDR trials, the subjects were disconnected from all instrumentation and permitted walk around. Urine was collected during this time. All experiments were performed in a dimly lit, quiet laboratory maintained at C.

58 At the start of each HDR trial, a blood sample was obtained for hematocrit 47 determination and an estimated change in plasma volume was calculated using van Beaumont s equation (80) Vestibular Activation Vestibular otolith activation was tested before and after drug administration by the subjects performing HDR in the prone position as previously described (72). This maneuver engages the otolith organs, but not the semicircular canals when the head becomes stationary. Subjects were placed in the prone position, instrumented for the study (BP, heart rate, venous occlusion plethysmography), and an appropriate MSNA recording site was obtained. After a 3-minute baseline period with the head in the baseline chin-up neck extended position, the chin support was removed and the head was passively rotated to the point of maximal rotation. This position was maintained for 3 minutes followed by the subject s head being passively returned to the baseline chin-up neck-extended position for 3 minutes of recovery. A cold pressor test was performed following the HDR. The cold pressor test consisted of a two-minute baseline period followed by placing the subject s hand in ice water up to the wrist for 2 minutes. During HDR and cold pressor test, arterial blood pressure, heart rate, MSNA, and venous occlusion plethysmography of the contralateral leg were continuously recorded.

59 4.2.4 Drug Administration 48 Each subject was studied twice with a minimum period of one week separating testing days. After the initial HDR trial was performed, the subjects received either an injection of 0.9% normal saline solution (which served as a sham control) or a 1.5 g/kg glycerol solution (50/50 mixed solution with cranberry juice) taken orally. Subjects were required to wait one hour during both testing days for urine collection and for the maximal effect of the glycerol before beginning posttesting. Fluid intake by the subjects was maintained to their normal use prior to the study and was not allowed during the study Measurements Microneurography Microneurography was used to assess MSNA as previous described (72). Briefly, multifiber recordings of MSNA were obtained from a tungsten microelectrode inserted in the peroneal nerve behind the knee. A reference electrode was placed subcutaneously 2-3 cm from the recording electrode. Previously identified criteria for an adequate MSNA signal were applied to ensure proper recording (79). The nerve signal was amplified (20,000 40,000 times), fed though a bandpass filter with a band width of 700-2,000 Hz, integrated using a 0.1 s time constant (University of Iowa Bioengineering, Iowa City, IA), and recorded digitally (16SP Powerlab, ADInstruments, New Castle, Australia). The mean voltage neurogram was routed to a computer screen and a loudspeaker for

60 49 monitoring during the study. Sympathetic recordings that demonstrated possible electrode site shifts, or electromyographic artifact during experimental intervention were excluded from analysis Limb Blood Flow Venous occlusion plethysmography (Hokanson EC 4 plethysmograph, D.E. Hokanson, Bellevue, WA) was used to assess calf blood flow of the contralateral leg as previously described (33). Briefly, mercury-in-silastic strain gauges were placed around the maximal circumferences of the calf. An ankle cuff was inflated to 220 mmhg to arrest circulation in the foot while a thigh cuff was inflated to 50 mmhg every 15 seconds. Calf vascular conductance was calculated as the ratio of calf blood flow to mean arterial blood pressure Hemodynamic Heart rate was derived from an electrocardiogram. Arterial blood pressure was measured continuously by a finger photoplethysmography (Finometer, FMS, Amsterdam, The Netherlands) during each trial Data Analysis All data were digitally recorded at 100 Hz for later off line analysis. The investigator analyzing the data was blinded to the subjects identity and to which drug

61 50 was received. MSNA was expressed as bursts per minute and total activity (sum of area underlying individual bursts per minute, expressed in arbitrary units). Neurograms were normalized using the highest burst set to 1000 and the baseline set to zero. This procedure was repeated for each new MSNA site obtained. Sympathetic bursts were identified from the mean voltage neurogram and the sum of the area under each burst, expressed in arbitrary units (a.u.), was assessed by a computer program (Chart 5, ADInstruments). Plasma volume changes were calculated using hematocrit (Hct) changes according to the equation by Van Beaumont (80). The equation was: % plasma volume = (100/100-Hct1) x (100(Hct1-Hct2)/Hct2) % Statistical Analysis To identify possible differences between the pre and post-drug administration, a two-way repeated-measures (drug, HDR) analysis of variance (ANOVA) was used. MSNA comparisons were made between the average resting (baseline) activity and during the first minute of HDR. The results were separated between the individual drug trials and only compared within each trial (i.e., within saline or glycerol trial). When the interaction was significant, a test for simple effects was used to identify if there were differences between baselines (41). A t-test was used to identify differences in plasma volume and urine volume between the two drug trials. A significance level of p<0.05 was used for all tests. Values are presented as mean±se.

62 Results Hemodynamic responses to HDR for both trials are listed in Table 4.1. Table 4.1: Volume alterations and hemodynamic responses to HDR. Variable Saline Glycerol Urine Volume Lost 249±44 ml 506±65 ml* Plasma Volume Lost - 4.4±1.2 % - 2.1±2.3% Pre-test Post-test Pre-Test Post-Test MAP (mmhg) Baseline 89±3 92±4 91±2 94±2 HDR 88±3 91±3 91±2 93±3 (BL vs. HDR) - 1±2-1±2 0±1 0±1 HR (beats/min) Baseline 61±5 59±4 60±4 60±3 HDR 62±4 61±4 61±4 61±3 (BL vs. HDR) 1±1 1±1 1±1 0±1 * P<0.001 compared to saline value. Values expressed as mean ± SE. No significant differences between before and after drug administration. BL (Baseline) Hemodynamic responses to HDR for both trials are listed in Table 4.1. During both the saline and glycerol trials, HDR did not significantly change MAP or HR at either pre or post-testing. Percent change in plasma volume from pre to post-testing was 4.4±1.2% and 2.1±2.3%, for saline and glycerol trials, respectively. Hematocrit was

63 stable between testing days for each subject. Urine volume collected post-drug 52 intervention was 249±44 ml and 506±65 ml, for saline and glycerol respectively, indicating a significantly greater diuresis during the glycerol test Saline Trial In the saline control trial, HDR significantly increased MSNA burst frequency (17±2 to 21±2 bursts/min, 5±1 bursts/min, p<0.001) and total activity ( 44±13%, p<0.001) during pre-testing (Figure 4.1). Figure 4.1: Change in MSNA burst frequency (bursts/min) and changes in total activity (%) during HDR before and after administration of saline and glycerol. MSNA significantly increased during HDR before the drug administrations. This increase in MSNA during HDR was significantly blunted post-glycerol administration in both burst frequency and total activity and was significantly increased in total activity post-saline administration. * P<0.05 compared to pre-drug administration.

64 53 Following saline injection, MSNA response to HDR remained the same in burst frequency (15±2 to 22±3, 6±2 bursts/min, p<0.001) and total activity ( 83±20%, p<0.001) (Figure 4.1). Calf vascular conductance significantly decreased during HDR during pre (- 15±6%, p<0.001) and post-testing (- 21±4%, p<0.001) (Figure 4.2). Figure 4.2: Change in calf vascular conductance (%) during HDR before and after administration of saline and glycerol. Calf vascular conductance significantly decreased during HDR before drug administration. This decrease in calf vascular conductance during HDR was significantly attenuated post-glycerol administration and was significantly increased post-saline administration. *P<0.05 compared to baseline. P<0.05 compared to pre-drug administration. The cold pressor test significantly increased MSNA burst frequency ( 19±3 and 16±4, p<0.001) and total activity ( 286±65% and 318±81%, p<0.001) during both pre and post-testing, respectively (Figure 4.3).

65 54 Figure 4.3: Change in MSNA burst frequency (bursts/min) during the second minute of a cold pressor test before and after drug administrations. MSNA during the cold pressor test was significantly decreased post both drug administrations. *P<0.05 compared to pre-drug administration Glycerol Trial HDR significantly increased MSNA burst frequency (15±2 to 23±2 bursts/min, 8±1 bursts/min, p<0.001) and total activity ( 77±18%, p<0.001) during pretesting in the glycerol trial, which was comparable to the saline trial (Figure 4.1). Following glycerol administration, resting MSNA at baseline was increased; however, HDR increased MSNA in burst frequency (20±2 to 23±2, 3±1, p<0.05), but not total activity ( 22±3%) (Figure 4.1). The change in burst frequency and total activity during HDR was significantly attenuated following glycerol ingestion compared to pretesting. Calf vascular conductance significantly decreased during HDR before glycerol administration (- 20±3%, p<0.001). However, reduction in calf vascular conductance was significantly