Mechanisms of blood pressure regulation in humans elucidated by alterations in gravity and maximal exercise


Material Information

Mechanisms of blood pressure regulation in humans elucidated by alterations in gravity and maximal exercise
Physical Description:
xiv, 166 leaves : ill. ; 29 cm.
Engelke, Keith A
Publication Date:


Subjects / Keywords:
Research   ( mesh )
Blood Pressure -- physiology   ( mesh )
Exercise   ( mesh )
Gravitation   ( mesh )
Hypotension, Orthostatic -- prevention & control   ( mesh )
Department of Physiology thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Physiology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1994.
Bibliography: leaves 135-165.
Statement of Responsibility:
by Keith A. Engelke.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002324332
oclc - 50445785
notis - ALS7884
System ID:

Full Text








To my parents who taught me that with faith, patience,

perseverance, and humor, no challenge is too daunting; and to



I am greatly indebted to the many individuals who have

provided friendship, guidance, and support over the past four

years. First, I would like to express my sincere

appreciation to my primary advisor and friend, Victor A.

Convertino, Ph.D., who challenged me to achieve to the

highest standard and provided invaluable advice and

direction. I am also indebted to my advisory committee co-

chairman, Charles E. Wood, Ph.D., who gave me an opportunity

to broaden my perspective and encouraged me to strive for

excellence. Together, Drs. Convertino and Wood have not only

provided outstanding training, but have also set an enviable

example of professional responsibility and dedication. My

sincere thanks are also extended to the other members of my

advisory committee, Scott K. Powers, Ph.D. and Thomas J.

Wronski, Ph.D., for their time and support of this project

and my career.

I must also thank all the dedicated professionals at

NASA-Kennedy Space Center, NASA-Ames Research Center, Humana-

Lucerne Hospital, and Brooks Air Force Base who helped make

this project a success. I am particularly indebted to the

excellent people in the Kennedy Space Center Biomedical

Operations and Research Office including Dr. Wycke Hoffler,


Don Doerr, Sandy Reed, Art Maples, Marc Duvoison, Barry

Slack, Dick Triandifils, Jill Polet, Mike Merz, Anne Bowers,

and Mary Lou Lasley. In addition, I would also like to thank

those at Ames Research Center who provided invaluable support

including Dee O'Hara, Elizabeth Lowe, and Drs. Joan Vernikos,

Charles Wade, and John Greenleaf. I am also indebted to Dr.

Craig Crandall for his friendship and many hours of hard work

on the Ames bedrest studies and to Dr. Dave Ludwig for

assistance with statistical analyses.

An additional thank you goes to all my colleagues,

classmates, and professors at the University of Florida

including Tim Cudd, Randy Braith, Christine Saoud, Yeoung-

Choy Kam, Maria Castro, Melanie Pecins-Thompson, Mike Welsch,

Jane Eason, Cathy Golden, and Drs. Maureen Keller-Wood,

Wendell Stainsby, and Michael Pollock for their support

during this challenging and rewarding adventure.

Finally, I would like to thank the Florida Space Grant

Consortium for providing the financial support necessary to

allow me to dedicate the totality of my efforts to this



ACKNOWLEDGEMENTS .......................................... iii

LIST OF TABLES ............................................ vii

LIST OF FIGURES .......................................... viii

KEY TO ABBREVIATIONS ........................................ x

ABSTRACT ................................................. xiii

CHAPTER 1 INTRODUCTION ...................................... 1

CHAPTER 2 BACKGROUND REVIEW ................................. 5

2.1 A Brief Review of the Baroreflexes..................... 5
2.1.1 Arterial Baroreceptors ......................... 5
2.1.2 Cardiopulmonary Baroreceptors ................. 10
2.2 Methods of Testing Baroreflex Functionin Humans ...... 11
2.2.1 Carotid-Cardiac Baroreflex..................... 12
2.2.2 Aortic Baroreflex ............................. 13
2.2.3 Cardiopulmonary Baroreflex..................... 14
2.2.4 Integrated Baroreflex Function ................ 15
2.3 Baroreflex Plasticity ................................ 16
2.3.1 Baroreflex Resetting .......................... 16
2.3.2 Altered Baroreflex Sensitivity ................ 18
2.4 Physiology of Altered Baroreflex Function ............ 19
2.4.1 Blood Volume .................................. 19
2.4.2 Hormonal Stimuli .............................. 22
2.4.3 Ionic Sensitivity ............................. 28
2.4.4 Mechanical Stimuli ............................ 28
2.5 Baroreflex Responses to Altered Orthostatic
Conditions ........................................... 29
2.5.1 Anti-Orthostasis (g-g) ........................ 29
2.5.2 Orthostasis (1-g) ............................. 32
2.6 Effects of Exercise on Baroreflex Function ........... 39
2.6.1 Endurance Exercise Training ................... 39
2.6.2 Dynamic Maximal Exercise ...................... 41
2.7 Summary .............................................. 44
2.8 Conclusions .......................................... 45

CHAPTER 3 GENERAL METHODOLOGY .............................. 49

3.1 Experimental Design .................................. 49
3.2 Experimental Procedures and Protocols ................ 49
3.2.1 Study 1.. ....................................... 49

3.2.2 Study 2 ....................................... 50
3.3 Subjects ............................................. 52
3.4 Measurements and Techniques .......................... 53
3.4.1 Heart Rate and Blood Pressure ................. 53
3.4.2 Carotid Baroreflex Function ................... 54
3.4.3 Tests of Orthostatic Responses ................ 56
3.4.4 Blood and Plasma Volume ....................... 58
3.4.5 Vasoactive Hormones ............................ 60
3.4.6 Forearm Blood Flow ............................ 62
3.4.7 Leg Compliance ................................ 64
3.4.8 Leg Volume .................................... 64
3.4.9 Exercise bout ................................. 65
3.4.10 Valsalva Maneuver ............................. 66
3.4.11 Statistical Analyses .......................... 67

PARAPLEGIC SUBJECTS .............................. 70

4.1 Abstract ............................................. 70
4.2 Introduction ......................................... 71
4.3 Methods .............................................. 73
4.4 Results .............................................. 74
4.5 Discussion ........................................... 80

AFTER 16 DAYS OF 60 HEAD-DOWN TILT ............... 90

5.1 Abstract ............................................. 90
5.2 Introduction ........................................... 91
5.3 Methods .............................................. 93
5.4 Results .............................................. 93
5.5 Discussion ........................................... 98


6.1 Abstract ............................................ 106
6.2 Introduction ........................................ 107
6.3 Methods ............................................. 109
6.4 Results ............................................. 110
6.5 Discussion .......................................... 116

CHAPTER 7 SUMMARY ......................................... 125

7.1 Summary ............................................. 125
7.2 Conclusions ......................................... 132

LIST OF REFERENCES ........................................ 135

BIOGRAPHICAL SKETCH ....................................... 166






Table 4.2.

Table 5.1.

Table 6.1.

Physical characteristics of subjects ........... 52

Baseline hemodynamic, vasoactive hormone, plasma
volume, and vascular resistance values during
control and 24 h after exhaustive exercise ..... 75

Mean hemodynamic responses to the Valsalva
maneuver ....................................... 78

Comparison of mean heart rate and blood pressure
changes to the Valsalva maneuver between
experimental conditions ........................ 95

Reproducibility of baseline measures between
experimental conditions ....................... 110



Figure 2.1.

Figure 3.1.

Figure 4.1.

Figure 4.2.

Figure 4.3.

Figure 4.4.

Figure 5.1.

Figure 5.2a.

Figure 5.2b.

Figure 6.1a.

Figure 6.1b.

Figure 6.2a.

Cardiovascular reflex responses to an
orthostatic challenge ........................ 48

Representative mean arterial pressure and heart
rate responses to a 15-s Valsalva maneuver at
an expiratory pressure of 30 mmHg ............ 68

Comparison of forearm vascular resistance in
the supine posture and after 15 min of 70
head-up tilt in the control and exercise
conditions ................................... 76

Mean heart rate and systolic blood pressure
reponses to the Valsalva maneuver in the
control and 24 h post-exercise conditions .... 77

Carotid-cardiac baroreflex stimulus-response
relationships, plotted over the range of
pressure from which maximum slopes were
derived, during control and 24 h after
exercise ..................................... 79

Circulating hormone levels during supine
baseline and at the end of 70 HUT during
control and post-exercise conditions ......... 81

Mean HR and MAP responses to a 15-s Valsalva
maneuver on day R-1 at an expiratory pressure
of 30 mmHg in control and post-exercise
conditions ................................... 96

Ratio of the unit change in HR to unit change
in MAP during phase II of the Valsalva
maneuver ..................................... 97

Change in MAP from pre-Valsalva baseline during
late phase II ............................... 97

Maximal LBNP tolerance time before and after
HDT ......................................... 112

Difference in LBNP tolerance time from pre-HDT
for each of the seven subjects .............. 112

MAP response at peak LBNP ................... 113


Figure 6.2b.

Figure 6.3.

Figure 6.4.

Figure 7.2.

Figure 7.2.

Heart rate response at peak LBNP ............ 113

Range of forearm vascular responses to LBNP
before and after 16 days HDT in control and
exercise conditions ......................... 114

Carotid-cardiac baroreflex stimulus-response
relationships before, on day 16, and just prior
to re-ambulation from 16 days HDT in the
control and exercise conditions ............. 115

Reflex responses to orthostasis following
microgravity exposure ....................... 126

Specific components of the orthostasis-stimulus
cardiovascular-reflex response cascade which
are enhanced following completion of acute,
maximally intense exercise .................. 128


Ang II




beats -mmHg-1


















angiotensin II

analysis of variance

atrial natriuretic peptide

arginine vasopressin

beats per millimeter of mercury

blood pressure

beats per minute


carotid-cardiac baroreflex


corticotropic releasing hormone

diastolic blood pressure



etyylenediaminetetraacetic acid

forearm vascular resistance



head-down tilt

heart rate

head-up tilt















msec mmHg-1




kilopond meters per minute

liters per minute

lower body negative pressure

leg comliance

leg volume

mean arterial pressure

milli-equivalents per day



milliliters per millimeter of

millimeters of mercury


millisecond per millimeter of


nanogram per milliliter per

plasma renin activity

peripheral resistance unit

plasma volume

second (s)

systolic blood pressure

standard error

sympathetic nerve activity

stroke volume











rpm revolutions per minute

VM Valsalva maneuver

VO2max maximum oxygen uptake

yr year


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Keith A. Engelke

April, 1994

Chairman: Charles E. Wood
Cochairman: Victor A. Convertino
Major Department: Physiology

To test the hypothesis that maximal exercise could

enhance the function of mechanisms responsible for blood

pressure control, two repeated-measure (cross-over design)

experiments were conducted in which heart rate (HR), blood

pressure (BP), forearm vascular resistance (FVR), and

vasoactive hormone responses were measured during an

orthostatic challenge in seventeen volunteers prone to

postural hypotension on two occasions: 1) 24 hours post-

exercise and 2) without exercise (control). Isolated

carotid-cardiac (CCB) and integrated baroreflex sensitivity,

plasma volume (PV), and leg vascular responses were also



Study 1 investigated the effect of maximal exercise on

blood pressure stability during 700 head-up tilt (HUT) in 10

paraplegic subjects. During HUT, the reduction in systolic

blood pressure (SBP) observed in the control condition was

larger than that following exercise. Maintenance of SBP

during post-exercise HUT was associated with increased FVR

and accentuated baroreflex sensitivity, and was independent

of differences in plasma hormone levels, leg compliance, and

blood volume.

In Study 2, cardiovascular responses during lower body

negative pressure (LBNP) to pre-syncope were measured in 7

subjects following 16 days of 60 head-down tilt (HDT). PV,

CCB, and integrated baroreflex sensitivity were also

measured. During LBNP 24 hours after exercise, time to pre-

syncope was unchanged from pre-HDT levels compared to a 2-min

decrease observed in controls. HR was similar in both

conditions as was leg volume. Improved blood pressure

stability and maintenance of orthostatic tolerance post-

exercise was associated with greater FVR, restored PV, and

enhanced CCB and integrated baroreflex sensitivity.

Thus, maximal exercise enhanced BP regulation by

improving control of vascular resistance and increasing

baroreflex sensitivity and plasma volume. These adaptations

appear to be effective in ameliorating microgravity-induced

orthostatic hypotension.



Compromised blood pressure control, manifested as

hypotension during an orthostatic challenge, has been

consistently observed following exposure to microgravity and

its ground-based analogs (Blomqvist and Stone, 1983;

Convertino et al., 1990a, 1991; Greenleaf et al., 1989;

Hoffler, 1977; Johnson et al., 1977b; Stegemann et al.,

1975). Although not completely understood, an array of

mechanisms and their interactions underlying orthostatic

hypotension including inadequate venous return and cardiac

output due to reduced blood volume (Convertino, 1990, 1991;

Greenleaf et al., 1989; Sandler, 1986), attenuated

autonomically-mediated cardiovascular reflex responsiveness

(Convertino et al., 1990a; Eckberg and Fritsch, 1991, 1992;

Fritsch et al., 1992), and increased venous compliance of the

lower extremities (Buckey et al., 1992; Convertino et al.,

1989b; Hoffler, 1977) have been observed following prolonged

spaceflight, bedrest, and chair confinement. These

adaptations suggest that gravity has a significant influence

on the functional capacity of blood pressure control


Attempts to minimize or prevent orthostatic instability

induced by exposure to weightlessness through application of

conventional endurance-type exercise training programs have

proven ineffective (Convertino, 1990; Greenleaf et al.,

1989). It is likely that this is due to a failure to

understand the mechanisms underlying orthostatic hypotension

and the way in which they are affected by exercise.

Recently, single bouts of intense exercise have been shown to

increase plasma volume (Gillen et al., 1991) and sensitivity

of the autonomically-mediated carotid-cardiac baroreflex

(Convertino and Adams, 1991; Somers et al., 1985) and to

decrease the incidence of syncopal episodes following

exposure to groundbased simulations of microgravity such as

water immersion (Stegemann et al., 1975) and bedrest

(Convertino, 1987b). Taken together, these observations

suggest acute, maximally intense exercise may provide a

specific physiological stimulus that can acutely attenuate or

reverse microgravity-induced cardiovascular adaptations

associated with orthostatic hypotension.

To better understand the association between mechanisms

of blood pressure regulation and microgravity-induced

orthostatic hypotension, the two studies included in this

dissertation were conducted to examine effects of an exercise

stimulus on basic mechanisms associated with blood pressure

regulation. Plasma volume, leg vascular responses, arterial

baroreflex sensitivity, plasma levels of vasoactive hormones,

and cardiovascular responses to changes in intra-thoracic

pressure and orthostatic stress were measured with and

without exercise treatment in subjects prone to postural

hypotension. The first study (Study 1) was designed to

evaluate the effect of a bout of graded exercise on cardiac

and vascular responses to orthostasis in subjects whose blood

pressure control system had been impaired due to prolonged

confinement to wheelchairs. In this study, we tested the

hypothesis that maximal exercise would provide blood pressure

homeostasis during head-up tilt.

The purpose of the second study (Study 2) was to

determine the effects of exhaustive exercise on reversing

cardiovascular adaptations associated with orthostatic

intolerance following 16 days of 60 head-down tilt. We

hypothesized that one exercise bout would be effective in

attenuating or reversing decrements in blood volume,

baroreflex sensitivity, cardiovascular function, and

orthostatic stability following head-down tilt and enhance

blood pressure control during lower body negative pressure.

The mechanisms underlying enhanced blood pressure

control during recovery from a bout of maximal exercise are

unclear. This dissertation attempts to elucidate the

contribution of changes in blood volume and alterations in

responsiveness of autonomically-mediated baroreflexes to

blood pressure maintenance during orthostasis. Therefore,

Chapter 2 provides a background review of baroreceptor

function and stimuli which alter baroreflex activity.

Chapter 3 describes the general experimental design and

methodology used in this project. Chapter 4 evaluates the

effect of maximal exercise on the blood pressure response to

head-up tilt of subjects permanently confined to the seated

posture. Chapter 5 describes the effect of exercise on

integrated cardiovascular reflex response to a change in

intra-thoracic pressure elicited by the Valsalva maneuver and

Chapter 6 examines the influence of a single bout of leg

exercise on cardiac and vascular response to lower body

negative pressure. Chapter 7 provides a general summary of

the two projects and offers overall conclusions. The results

of these studies should provide a better understanding of

adaptive and integrative processes of components important in

blood pressure regulation.


2.1 A Brief Review of the Baroreflexes

2.1.1 Arterial Baroreceptors

Arterial baroreceptors are mechano-receptors located in

the walls of large systemic arteries including the aortic

arch, brachiocephalic artery, and carotid sinuses (Boss and

Green, 1956; Green, 1954). Afferent signals from the arch of

the aorta are transmitted through the left and right aortic

(depressor) nerves to the vagus nerve and ultimately to the

nucleus tractus solitarius in the medullary area of the brain

stem (Boss and Green, 1956; Nakayama, 1965; Nonindez, 1935).

Sensory input from the carotid sinus region travels to the

nucleus tractus solitarius as well, but via the sinus

(Hering's) and glossopharyngeal nerves.

Barosensitive nerve endings are found in areas of the

arteries with large quantities of elastic tissue (Muratori,

1967). Approximately 40% of the tissue comprising the walls

of the aortic arch is an elastin-collagen mixture (Bader,

1963) which is almost free of smooth muscle (Gregoreva,

1962). Similarly, the carotid sinus is thinner (Adams, 1958;

Addison, 1944; Rees, 1968; Rees and Jepson, 1970), contains

less smooth muscle (Addison, 1944; Bagshaw and Fischer, 1971;

Muratori, 1967; Rees and Jepson, 1970), and shows a higher

elastin content (Addison, 1944; Rees and Jepson, 1970) than

other areas of the carotid artery.

Although their name implies a pressure-sensitive

quality, baroreceptors are stretch receptors which respond to

deformation of the vessel wall in which they are located

(Hauss et al., 1949; Angell-James, 1971). There is strong

experimental evidence which shows that the degree of wall

deformation determines the electrical activity of the carotid

sinus and aortic arch baroreceptors. Hauss et al. (1949)

demonstrated that the reflex fall of blood pressure produced

by an increase in carotid sinus pressure is abolished if the

stretching of the carotid artery is prevented by a plaster

cast applied to the outside of the sinus region.

Additionally, Angell-James (1971) reported that increased

baroreceptor activity produced by elevation of intrathoracic

pressure can be prevented by simultaneously increasing the

extramural pressure by the same amount. Experiments in man

have shown that changing pressure in a chamber surrounding

the neck results in reflex changes in heart rate and blood

pressure and provide further evidence that altered transmural

pressure is the stimulus for baroreceptor activiation

(Bevegard and Shepherd, 1966; Ernsting and Parry, 1957).

Koch (1931) was the first to demonstrate that when

carotid sinus pressure was changed in a stepwise manner, mean

arterial pressure exhibited an inverse sigmoidal response to

the change in intrasinus pressure. These findings were later

substantiated by Bronk and Stella (1932) who observed that

the impulse frequency in Hering's nerve exhibited a positive

sigmoidal relationship to changes in sinus pressure. These

early findings identified that baroreceptors can be

characterized as having a threshold pressure at which

receptor discharge is initiated, a pressure range for which

the discharge rate increases with a rise in mean arterial

blood pressure, and an asymptotic saturation pressure beyond

which there is little increase in baroreceptor activity

(Koushanpour, 1991).

The aortic and carotid baroreceptors exhibit different

threshold and saturation characteristics. Carotid sinus

baroreceptors are silent at arterial pressures between 0 and

60 mmHg, but above 60 mmHg, they respond progressively and

reach maximum discharge capacity at approximately 180 mmHg

(Koushanpour, 1991). Aortic baroreceptors respond in a

manner similar to that of the carotids except that they

exhibit a threshold pressure approximately 30 mmHg higher,

i.e., 90 mmHg (Koushanpour, 1991). Therefore, in the normal

operating range of approximately 100 mmHg (80 to 180 mmHg),

slight changes in pressure elicit strong baroreceptor-

mediated autonomic reflexes to return arterial pressure to

within homeostatic limits.

Effective baroreceptor function is necessary to respond

to transient alterations in arterial pressure (Brown, 1980).

In the face of increasing pressure, baroreceptor-generated

signals ascend afferent pathways and enter the nucleus

tractus solitarius where secondary signals inhibit the

medullary vasoconstrictor center and excite the vagal center

stimulating vasodilation and decreased myocardial inotropic

and chronotropic response (Brown, 1980). These actions lead

to lowered peripheral resistance, cardiac output, and

ultimately, lower blood pressure. Conversely, a sudden fall

in arterial pressure leads to reflex actions which increase

cardiac output and sytemic resistance to raise blood


The bradycardic response to baroreceptor stimulation in

humans is mediated through vagal cholinergic mechanisms.

Several investigators (Eckberg et al., 1971; Pickering et

al., 1972; Simon et al., 1977; Takeshita et al., 1979) have

demonstrated that elongation of the R-R interval which

accompanied a rise in arterial pressure following

administration of phenylephrine was not reduced by

propranolol (Jose and Taylor, 1969), but was abolished by

atropine. Others have reported similar observations in

response to stimulation of the carotid baroreceptors by neck

suction (Eckberg, 1977; Eckberg et al., 1976).

In contrast, there is a lack of consensus regarding the

autonomic mechanisms mediating the tachycardic response to

arterial baroreceptor unloading. It has been observed that

the early tachycardia observed after administration of

vasodilators was unaffected by propranolol, but abolished by

atropine, suggesting a predominant vagal mediation of this

response (Leon et al., 1970; Mancia et al., 1979; Mroczek et

al., 1976; Pickering et al., 1972). Contrary to these

reports, others have demonstrated that the increase in heart

rate produced during infusions of nitroglycerin was reduced

by atropine, but could only be abolished by combined

administration of atropine and a beta-adrenergic blocker

(Goldstein et al., 1975; Robinson et al., 1966). In

addition, it has been demonstrated that during lower body

negative pressure, tachycardia was diminished 52% by

propranolol with the remaining response abolished by atropine

(Bjurstedt et al., 1977). Therefore, it appears there is a

significant vagal component to the cardioacceleration which

accompanies baroreceptor unloading. However, an increase of

sympathetic cardiac influence may contribute to the more

sustained component of baroreflex-mediated tachycardia

(Mancia and Mark, 1983). Taken together, these results

suggest redundancy in mechanisms by which the autonomic

nervous system mediates baroreflex-induced tachycardia.

The baroreceptor system markedly reduces daily variation

in arterial pressure. This phenomena is readily demonstrated

in sinoaortic dennervated animals who exhibit elevated blood

pressure and increased blood pressure liability (Cowley et

al., 1973). Controversy exists regarding the persistence of

this hypertension with some authors arguing it eventually

subsides (Guyton et al., 1974) while others suggest elevated

blood pressure persists (Scher and Ito, 1978; Alexander,

1979; Touw et al., 1979; Werber and Fink, 1979).

2.1.2 Cardiopulmonary Baroreceptors

The atria (Paintal, 1953) and pulmonary vessels

(Coleridge and Kidd, 1963; Ledsome and Kan, 1977) have

numerous mechano-receptors in their walls which also play a

role in the control of cardiovascular function (Mark and

Mancia, 1983). As is the case with arterial baroreceptors,

afferent signals from the cardiopulmonary receptors are

transmitted to the cardiovascular centers in the medulla via

either vagal (Donald and Edis, 1971; Thoren et al., 1976) or

sympathetic (Malliani et al., 1975) nerve fibers. Because

these receptors are located on the venous (low pressure) side

of the circulation, they are not well-suited for detecting

small and/or rapid fluctuations in arterial blood pressure.

However, information regarding even minor volume changes is

provided by these receptors and they elicit reflexes parallel

to the arterial baroreflexes to make the total reflex system

much more potent for control of blood pressure (Persson,


Several investigators have demonstrated that low

pressure receptors exert an important influence on peripheral

vascular resistance during alterations of venous return and

cardiac filling pressures in humans. Roddie et al. (1957)

observed reflex dilation in skeletal muscle during leg

raising. Others (Johnson et al., 1974; Zoller et al., 1977)

have noted that as central venous pressure decreased due to

application of low levels (-5 to -20 mmHg) of lower body

negative pressure, forearm vascular resistance increased.

Low pressure receptors may also mediate alterations in

splanchnic resistance (Abboud et al., 1979; Johnson et al.,

1974) and venous tone (Abboud and Mark, 1979; Epstein et al.,

1968; Samueloff et al., 1966).

Much effort has been expended in an attempt to determine

the influence of cardiopulmonary receptors on heart rate.

Although this concept, first proposed by Bainbridge (1915)

after he observed an increase in heart rate during saline

infusions in dogs, is relatively well-supported by data from

animal models (Bishop et al., 1976; Gupta, 1975; Vatner et

al., 1975;), the existence of such a reflex in humans remains

controversial (Doyle et al., 1951; Johnson et al., 1974;

Takeshita et al., 1979; Zoller et al., 1972). In contrast,

there is compelling evidence to suggest that low pressure

baroreceptors participate in the reflex control of renin and

vasopressin secretion (Epstein et al., 1981; Epstein et al.,

1975; Epstein and Saruta, 1971; Kiowski and Julius, 1978).

2.2 Methods of Testing Baroreflex Function in Humans

Both invasive and non-invasive experimental techniques

are employed to assess the function of human baroreceptors.

Additionally, there are methods available to isolate specific

baroreceptor populations as well as measure integrated

baroreflex function. The method by which the reflex function

is analyzed is important because the technique utilized will

influence the quantity and quality of the collected data

(Mancia and Mark, 1983).

2.2.1 Carotid-Cardiac Baroreflex

In humans, invasive assessment of the carotid baroreflex

involves measurement of heart rate and/or blood pressure

during direct electrical stimulation of the carotid sinus

nerve (Wallin et al., 1975), temporary interruption of nerve

traffic with anesthesia (Guz et al., 1966), or permanent

denervation via surgical ablation (Angell-James and Lumley,

1964). However, there are several limitations to these

techniques including concerns regarding completeness of the

section or blockade, non-selectivity of anesthesia, and

unanticipated side effects due to the anesthetic or surgical

procedure utilized. Thus, these approaches are rarely used

in the research laboratory.

A simple, non-invasive method to assess carotid

baroreceptor function involves measurement of heart rate and

blood pressure during carotid massage (Roddie and Shepherd,

1957). While this method is widely used clinically, it is of

little use as a research tool due to the inability to

quantify or repeat the stimulus presented to the

baroreceptors. Carotid baroreceptor function is also

assessed by recording changes in heart rate or blood pressure

while carotid distending pressure is altered by a device

which transmits pressure and/or suction to the area

surrounding the carotid sinuses (Ernsting and Parry, 1957;

Stegemann et al., 1974). Recent advances in this technique

allows rapid changes in neck pressure and suction to be

delivered to the anterior 2/3 of the neck for 10-15 seconds

(Eckberg and Eckberg, 1982; Sprenkle et al., 1986). The

carotid-cardiac baroreflex stimulus-response curve is

calculated by plotting changes in R-R interval (or blood

pressure) against estimated carotid sinus pressure (Sprenkle

et al., 1986; Ludwig and Convertino, 1991). The primary

criticism of this method is the inability to quantify the

exact amount of pressure and/or suction transmitted through

the tissue of the neck to the carotid sinus. Despite the

work of Ludbrook et al. (1977) who determined that 86% of the

applied pressure and 64% of the suction is transmitted

directly to the sinuses, it is likely that variability exists

in the amount of pressure/suction reaching the sinuses due to

differences in body composition of subjects and design of the

neck collar device used by investigators.

2.2.2 Aortic Baroreflex

The currently accepted method to assess the isolated

aortic baroreflex in humans requires activation of the three

baroreceptor populations followed by manipulations which

subsequently attenuate or abolish their influence (Ferguson

et al., 1985; Sanders et al., 1988). Briefly, the carotid

sinus, aortic, and cardiopulmonary baroreceptors are

activated via a phenylephrine-induced increase in arterial

pressure. During this increase of arterial pressure, carotid

sinus baroreceptor activity is countered by application of

neck pressure at a level equal to the elevation of blood

pressure. Increased cardiopulmonary baroreceptor activity is

negated by lower body negative pressure. Therefore, with the

carotid sinus and cardiopulmonary baroreceptor activities

clamped at baseline levels, changes in heart rate can be

attributed to the aortic baroreceptors. Sensitivity of the

aortic baroreflex is then assessed by calculating the ratio

of the change in heart rate to the change in mean arterial


2.2.3 Cardiopulmonary Baroreflex

Cardiopulmonary baroreceptor function is evaluated by

examining reflex changes in peripheral vascular conductance

stimulated by changes in central venous pressure. Typically,

techniques which alter venous return and atrial filling are

utilized including lower body positive (Bevegard et al.,

1977a) and negative (Brown et al., 1966; Stevens and Lamb,

1965; Wolthuis et al., 1974) pressure, leg elevation (Roddie

and Shepherd, 1956; Roddie and Shepherd, 1957), head-out

water immersion (Gauer and Henry, 1963; Epstein, 1976), tilt

(Roddie and Shepherd, 1957), and respiratory maneuvers

(Roddie and Shepherd, 1958; Roddie et al., 1957). Reflex

changes in peripheral blood flow are then measured, most

simply via plethysmography.

Lower body negative pressure (LBNP) studies have

contributed greatly to the understanding of the role of

cardiopulmonary receptors in man. The primary advantage of

this techique is that the magnitude of the stimulus can be

controlled. Negative pressure of less than -20 mmHg

decreases cardiac filling pressure without detectable changes

in arterial pulse pressure, mean pressure, arterial dt/dP, or

heart rate. Thus, mild lower body negative pressure

minimizes the stimulus to low-pressure receptors without

significant changes to determinants of arterial (high

pressure) baroreceptor activity (Johnson et al., 1974; Zoller

et al., 1972).

2.2.4 Integrated Baroreflex Function

Global baroreflex function has been assessed by a

variety of methods. However, because these techniques elicit

a simultaneous response from many or all baroreceptor

populations, conclusions cannot be made regarding the

stimulus-response relationship of a specific baroreflex arc.

The most common technique to assess integrated baroreflex

function involves measuring heart rate in the presence of

either a pressor or depressor agent and then calculating the

ratio of the change in heart rate to the change in blood

pressure (Mancia and Mark, 1983) The slope of this

relationship is indicative of baroreflex sensitivity. A

limitation of this method is that only baroreflex control of

heart rate can be assessed since vasoactive drugs are

administered. Other techniques which cause a change in heart

rate and blood pressure such as lower body negative pressure

(Smith et al., 1988), head-up tilt (Smith et al., 1987a), and

the Valsalva maneuver (Smith et al., 1987b) have also been

used to assses global baroreflex function.

2.3 Baroreflex Plasticity

2.3.1 Baroreceptor Resetting

Both human and animal studies have demonstrated that the

operating range of baroreflexes is labile. Persistent

alterations of mean arterial pressure result in a relative

shift of the pressure-stimulus baroreceptor-response curve

along the stimulus (pressure) axis in the direction of

increased or decreased pressure without a change in the slope

(sensitivity or gain) of the steepest part of the curve

(Hayward et al, 1993; Igler et al., 1981; Munch and Brown,

1985; Sleight et al., 1977). This change in baroreflex

function is referred to as resetting. McCubbin et al. (1956)

was the first to demonstrate this phenomenon. Using

electroneurographic techniques to record carotid and aortic

baroreceptor activity in renal hypertensive dogs, they

demonstrated that when the baroreceptors were exposed to

prolonged increases of arterial pressure, the threshold and

operating range of the reflex were displaced to a higher

point on the pressure-stimulus baroreceptor-firing curve

(McCubbin et al., 1956). Subsequently, resetting has been

demonstrated in a variety of species and preparations

(Dorward et al., 1982; Kreiger, 1970; Kunze, 1981; Salgado

and Kreiger, 1978; Sleight et al., 1977). Alterations in

reflex function act to maintain, rather than oppose, the

prevailing pressure (McCubbin et al., 1956) possibly by

influencing the control of heart rate (Chen et al., 1982;

Dorward et al., 1985; Fritsch et al., 1989), arterial

pressure (Kunze, 1981; Tan and Zucker, 1989), and/or

sympathetic nerve activity (Dorward et al, 1985; Heesch and

Carey, 1987; Undesser et al., 1975).

In early studies, resetting was reported to occur over

hours or days. However, recent investigations have

demonstrated a more rapid resetting of baroreceptors.

Although the time course of resetting varies due to

differences in experimental preparations, baroreceptor

activity generally reaches a new steady-state within 5-20

minutes (Mifflin and Kunze, 1982; Munch et al., 1983).

Within this time, the entire pressure-baroreceptor activity

relationship shifts toward the new prevailing pressure

(Coleridge et al., 1981; Coleridge et al., 1984; Dorward et

al., 1982; Heesch et al., 1984a; Heesch et al., 1984b; Munch

et al., 1983; Munch and Brown, 1985; Undesser et al., 1984).

This resetting of baroreceptors allows them to respond

effectively to changes in arterial pressure and buffer such


2.3.2 Altered Baroreflex Sensitivity

In contrast to the frequently observed baroreceptor

resetting following exposure to static (Heesch et al., 1984a;

Heesch et al., 1984b; Mifflin and Kunze, 1982; Munch et al.,

1983; Munch and Brown, 1985) or sinusoidal (Kunze, 1981)

elevations in arterial pressure, resetting appears to be

markedly attenuated or absent when pressure changes are

pulsatile in nature. In support of this hypothesis, Chapleau

et al. (1987) have demonstrated that when carotid sinus

pressure is low, pulsations increase baroreceptor activity

without resetting of the baroreceptor stimulus-response

function. Similarly, when Mendelowitz and Scher (1988)

exposed a carotid sinus preparation to elevated pulsatile

pressures, the resulting blood pressure was lower throughout

the period of pulsation compared to that observed when the

carotid sinus was exposed to a static pressure profile.

Chapleau et al. (1987) also noted that increased baroreflex

sensitivity persisted for several minutes after cessation of

the pulsing period. Furthermore, they reported that the

magnitude of the effect was directly related to duration of

the pulsing period and to frequency and amplitude of the

pressure pulses (Chapleau et al., 1987).

A complete explanation for these observations remains to

be found. It appears that increases in static pressure

elicits resetting of the baroreflex possibly reflecting a

decreased responiveness of the medullary barosensitive

neurons to sustained baroreceptor discharge. In contrast,

phasic afferent activity which accompanies pulsatile

increases in mean arterial pressure may minimize central

adaptation to continued baroreceptor discharge. As a result,

there is a greater reflex inhibition of sympathetic activity

and subsequent maintenance or enhancement of baroreflex

activity to pulsatile pressure increases (Chapleau et al.,


2.4 Physiology of Altered Baroreflex Function

2.4.1 Blood Volume

Several investigators have observed that alteration or

redistribution of central blood volume is associated with

changes in baroreflex responses (Bevegard et al., 1977b;

Billman et al., 1981; Harrison et al., 1986; Mack et al.,

1987). Experiments in rats have demonstrated that

hypervolemia decreased the blood pressure response to

bilateral carotid occlusion, while a moderate hypovolemia

tended to increase the blood pressure response to this same

stimulus (Castenfors and Sjostrand, 1973). Vatner et al.

(1975) noted that elevated mean arterial pressure stimulated

by acute volume loading in conscious dogs elicited a

decreased heart rate response and concluded that the

sensitivity of the arterial baroreceptor reflex had been

reduced. A similar conclusion was reached by others who

observed nearly identical results in rabbits (Lubrook and

Graham, 1984; Stinnett et al., 1976), dogs (Wang et al.,

1992; Dujardin, 1980), and non-human primates (Billman et al,

1981; Cornish et al., 1989). Using human subjects, Pickering

et al. (1971) altered effective circulating volume with

upright tilt and observed a diminishment of baroreceptor

sensitivity compared to supine control. Similarly, Harrison

et al. (1986) used head-up and head-down tilt to acutely

alter central blood volume and noted that the heart rate

response to neck suction was greater in the head-down

position. Blood volume may also influence the function of

the cardiopulmonary baroreflex (Mack et al., 1987; Thompson

et al., 1990).

Conversely, Eckberg et al. (1972, 1976) compared

baroreflex sensitivity in the upright and supine postures and

detected no difference between conditions. Consistent with

these data is the observation that increasing central blood

volume by applying lower body positive pressure (Bevegard et

al., 1977a) or by raising the lower limbs and trunk

(Takeshita et al., 1979), has no effect on arterial

baroreceptor responsiveness. Similarly, Thompson et al.

(1990) reported that acute alterations in plasma volume had

no influence on the carotid-cardiac baroreflex sensitivity.

Decreasing central blood volume by applying non-hypotensive

(-10 to -40 mmHg) lower body negative pressure also failed to

alter baroreceptor control of heart rate (Takeshita et al.,

1979), but did augment carotid baroreflex control of vascular

resistance (Bevegard et al., 1977a; Mancia and Mark, 1983).

An explanation of these discrepancies remains elusive,

but it appears that they can be partly attributed to species

differences and dissimilarities among experimental

techniques. Of particular importance is the magnitude of

blood volume alteration induced by each experimental approach

(Barajanzi and Cornish, 1987). It is well-established that

changes in blood volume alter the activity of

mechanoreceptors in the atria and vena cava (Thorn et al.,

1976). Reflexes from cardiopulmonary receptors also modulate

arterial baroreceptor function (Ludbrook and Graham, 1984).

Takeshita et al. (1979) suggested that, in humans, variations

of central venous pressure and cardiopulmonary activity

within a certain physiological range do not influence

arterial baroreceptor control of sinus node activity.

Therefore, it is possible that circulating volume must be

altered above or below a threshold before high pressure

baroreceptor function is influenced (Thompson et al., 1990).

2.4.2 Hormonal Stimuli

Circulating hormones such as norepinephrine (NE),

arginine vasopressin (AVP), and angiotensin II (Ang II), in

addition to actions on vascular smooth muscle and fluid

regulation, may act in the central nervous system to modulate

the regulation of peripheral sympathetic nerve activity (SNA)

possibly through a baroreflex-dependent mechanism (Bishop et

al., 1987; Ferario et al., 1987). In addition to NE, Ang II,

and AVP, atrial natriuretic peptide (ANP), insulin, and

corticotropin releasing factor (CRF) may also influence

arterial baroreflex control of arterial blood pressure.

Norepinephrine. The proximity of baroreceptor afferents

and sympathetic efferents in the aortic arch and carotid

sinus regions provides an anatomical basis for sympathetic

efferent modulation of baroreceptor activity (Rees, 1967;

Reis and Fuxe, 1968; Bock and Gorgas, 1976). Activation of

sinoaortic sympathetic nerves via electrical stimulation or

topical application of NE alters baroreceptor discharge and

produces reflex changes in arterial blood pressure. For

example, painting concentrated solutions of NE on the carotid

sinus lowers blood pressure (Kirchheim, 1976). Likewise,

activating the sinus efferent nerves reduces blood pressure

(Kedzi, 1954) and attenuates the response to carotid

occlusion (Wurster and Trobiani, 1973).

Although these reports suggest NE modifies baroreceptor

activity, what is not fully understood is the mechanism by

which these changes occur (Munch et al., 1987). It is

unclear if NE acts indirectly via contraction of local

vascular smooth muscle or directly through activation of

adrenergic receptors on the barosensitive nerve endings.

Both neurogenic and exogenic NE can alter mechanical

properties of the arterial wall smooth muscle in which the

nerve endings are embedded (Aars, 1971; Aars, 1981; Bagshaw

and Peterson, 1967). This hypothesis is also supported by

Munch and Brown (1985) who demonstrated that NE constricted

rat aortic arch and subsequently decreased activity of the

baroreceptors. In contrast, it has also been observed that

NE enhances baroreceptor activity independent of changes in

smooth muscle tone (Goldman and Saum, 1984; Kunze et al.,

1984). Thus, it appears that NE may have two modes of action

which lead to altered baroreceptor discharge characteristics.

Angiotensin II. Inappropriate heart rate responses to

Ang II-induced increases in arterial pressure as well as

similar increases in arterial pressure in intact and

sinoaortic denervated dogs during infusions of Ang II have

led investigators to conclude that this agent alters

baroreflex response (Guo and Abboud, 1984; Matsukawa and

Reid, 1990; Reid, 1984). Additionally, it has been observed

that there is a reduction in both arterial baroreflex

inhibition of sympathetic nerve activity and overall vagal

efferent activity during systemic administration of Ang II

(Guo and Abboud, 1984; Lumbers et al., 1979) suggesting that

Ang II resets the baroreflex operating range toward higher

pressures (Matsukawa and Reid, 1990).

The mechanisms by which Ang II acts to reset the

arterial baroreflexes remain unclear (Reid, 1984).

Angiotensin II may act centrally to reset the operating point

of the arterial baroreflex or to decrease the sensitivity of

this relationship (Reid, 1984). In the non-human primate,

infusions of Ang II decreased the sensitivity of the cardiac

baroreflex during changes in arterial pressure (Garner et

al., 1987). This observation has been substantiated by

others (Guo and Abboud, 1984; Lee and Lumbers, 1981).

Conversely, studies in dogs and rabbits failed to detect any

alteration in the slope of the mean arterial pressure-heart

rate relationship during changes in arterial pressure (Brooks

and Reid, 1986; Matsumura et al., 1989). Interestingly,

resetting of the reflex was prevented in rabbits and rats

with lesions in the area postrema (Fink et al., 1980)

suggesting that Ang II may modify baroreceptor afferent

input, and thus baroreceptor activity, in this region of the


Vasopressin. Studies in humans, dogs, and rabbits have

shown that increased circulating arginine vasopressin may act

in the central nervous system to augment the

sympathoinhibitory influence of the arterial and

cardiopulmonary baroreflexes (DiCarlo et al., 1989; Floras et

al., 1987; Hasser et al., 1987). This effect has been

directly determined by measuring sympathetic nerve activity

during intravenous infusions of AVP. In both the

anesthetized and conscious rabbit, progressive infusions of

AVP produced large decreases in lumbar and renal SNA prior to

any detectable increase in arterial pressure (Undesser et

al., 1985). When the arterial and cardiopulmonary

baroreflexes had been rendered non-functional by anesthesia

or surgery, increased circulating AVP had no effect on renal

SNA, suggesting that inhibition of renal SNA during AVP

infusion is due to a facilitation of the arterial baroreflex

(Undesser et al., 1985).

The mechanism of action of AVP on baroreceptor function

remains to be elucidated. Baroreceptor-mediated bradycardia

was prevented by blocking the vascular AVP receptor with a Vi

receptor antagonist (Bishop et al., 1987). Ablation of the

area postrema returned the slope of the mean arterial

pressure-heart rate relationship during AVP infusion to that

observed with phenylephrine infusion (Undesser et al., 1985)

confirming the hypothesis that AVP acts at the area postrema

to modify baroreflex responsiveness. Subsequent studies have

shown that endogenous increases in AVP due to osmotic and

reflex stimuli facilitate the arterial and cardiopulmonary

reflexes (DiCarlo et al., 1989; Hasser et al., 1988).

Therefore, it appears reasonable to conclude that AVP plays a

role in regulation of arterial baroreflexes, possibly by

regulating the operating point of the reflex via actions in

the brainstem.

Atrial Natriuretic Peptide. Atrial natriuretic peptide

is released into the plasma in response to increased atrial

stretch following blood volume expansion (Ballermann and

Brenner, 1986). Although this peptide hormone primarily

influences salt and water balance, pharmacological

concentrations of ANP decreased resting sympathetic nerve

activity (Imaizumi et al., 1987; Niijima, 1989) and shifted

the baroreflex curve to a lower arterial pressure range

(TerKonda and Bishop, 1987). These findings are supported by

other investigators who have shown reflex tachycardia to ANP-

induced reductions in arterial pressure to be reduced

(Ackermann et al., 1988; Ferrari et al., 1990), although this

effect may be somewhat variable (Marks et al., 1990; Volpe et

al., 1987). Decreases in heart rate to phenylephrine or

volume expansion are enhanced by the presence of ANP (Ferrari

et al., 1990; Marks et al., 1990; Volpe et al., 1987). Ebert

and Cowley (1988) infused ANP into human subjects and

observed that cardioacceleration during neck pressure is

attentuated without a change in resting blood pressure or

heart rate, while others (Imam et al., 1989) have shown that

changes in plasma norepinephrine to lower body negative

pressure are reduced in the presence of ANP. These findings

are consistent with the observation of Floras (1990) who

noted attenuated SNA activity in the presence of ANP in

response to reductions in central venous pressure and a cold

pressor test.

The mechanisms responsible for these observations remain

questionable. The observation that ANP attenuates

sympathetic nerve activity following reductions in central

venous pressure (Floras, 1990) suggests that afferent

cardiopulmonary baroreceptor input may be altered.

Alternatively, ANP may modulate central nervous system

processing of afferent information resulting in alteration of

baroreflex activation threshold (Bishop et al., 1991).

However, other mechanisms may also be involved as illustrated

by the observation that the effects of vagal stimulation are

augmented by ANP and the reduction in heart rate induced by

ANP is blocked by atropine (Ackermann et al., 1988; Atchison

and Ackermann, 1990). These findings suggest that ANP

influences parasympathetic control of arterial pressure. In

addition, ANP interferes with release of norepinephrine from

the nerve terminal (Nakamaru and Inagami, 1986).

Other Hormones. Other hormones have been suggested to

alter sympathetic nerve activity and thereby influence

baroreflex function. Increases in insulin elevate SNA

(Morgan et al., 1990) possibly through a central action

(Sakaguchi et al., 1988). Little is known about the direct

effects of insulin on the function of the baroreflexes,

although Hilsted (1982) observed that animals and humans with

impaired insulin release have reduced baroreflex function.

However, this finding may be associated with peripheral

neuropathy (Hilsted, 1982).

Several studies have suggested that hormones associated

with the pituitary-adrenal axis regulate sympathetic activity

and affect baroreflex function. Glucocorticoids act

centrally to suppress sympathetic outflow (Brown and Fisher,

1986), while CRF increases sympathetic outflow as well as

release of norepinephrine and epinephrine into the plasma

(Fisher and Brown, 1984). CRF also interferes with the

arterial baroreflexes, reducing the gain and range of the

reflex curve and increasing the midpoint of the pressure

range (Fisher, 1989).

2.4.3 Ionic Sensitivity

Hodgkin and Katz (1949) theorized that reduced

extracellular sodium ion concentration could lead to a change

in the sensitivity of baroreceptors to suprathreshold

pressures. Indeed, Kunze and Brown (1978) and Kunze et al.

(1977) have demonstrated that lowering extracellular sodium

concentration elevates the operating point, increases the

minimum pressure at which the carotid sinus reflex is

elicited, and reduces sensitivity of the reflex.

Baroreceptors are also calcium and potassium-sensitive and

there is evidence to suggest that sodium, calcium, and

potassium interact to influence baroreceptor activity (Brown,

1980). However, despite evidence supporting the hypothesis

that small changes in electrolyte concentrations alter

baroreflex function, there is little understanding of the

underlying mechanisms of action (Brown, 1980).

2.4.4 Mechanical Stimuli

There is general agreement that baroreceptors exhibit

plasticity in response to persistent mechanical stimuli,

possibly due to relaxation of the viscoelastic medium which

couples receptor elements to external stimuli (Hubbard, 1958;

Lippold et al., 1960). However, interpretation of these data

is made difficult because investigators have exposed

baroreceptors to a variety of stimuli including highly

damped, sinusoidal, and static pressure profiles in an

attempt to elicit alterations in reflex function (Coleridge

et al., 1981; Coleridge et al., 1984; Heesch et al, 1984a;

Heesch et al, 1984b; Kunze, 1981; Mifflin et al., 1982; Munch

et al., 1983; Munch and Brown, 1985; Tan and Zucker, 1989).

Regardless, it appears the rate and degree of adaptation is

proportional to the intensity of the intravascular pressure

stimulus (Eckberg, 1977).

2.5 Baroreflex Responses to Altered Orthostatic Conditions

2.5.1 Anti-Orthostasis (4-g)

True Microgravity. Only limited physiological data on

the effects of true microgravity were collected during the

early phases of the US and Russian manned spaceflight

programs (Pestov and Geratewohl, 1975). More extensive

studies were made during the American Apollo, Skylab, and

Space Shuttle missions and the Russian Salyut and Soyuz

flights (Blomqvist and Stone, 1983). In addition, earth's

gravitational field has been cancelled for up to 45 seconds

in aircraft following parabolic flight patterns (Sandler,

1979). Therefore, although data obtained during actual

weightlessness are crucial in the study of effects of gravity

on cardiovascular function, results must be interpreted with

caution due to the small number of subjects that can be

studied in true weightlessness and the limited time of

exposure and range of methods available (Blomqvist and Stone,


Only one study has assessed the baroreflex control of

heart rate following exposure to real microgravity. Using a

neck collar device, the carotid stimulus-cardiac response

relationship was measured before and after 4-5 days of

spaceflight (Fritsch et al., 1992). The results indicated

that baseline R-R interval was reduced following flight

suggesting tachycardia at rest. Additionally, there was a

tendency for the slope of the estimated carotid sinus

pressure-heart rate relationship to be decreased post-flight.

Despite the large variation of response in a relatively small

number of subjects, the investigators concluded that the

carotid-cardiac baroreflex function was impaired following 4-

5 days of microgravity exposure (Fritsch et al., 1992).

Ground-Based Analogs. Technical and economic

constraints have hindered the conduction of rigorously

controlled experiments during actual microgravity exposure.

Consequently, a number of ground-based analogs utilizing both

animal and human subjects have been developed which elicit

adaptations in the cardiovascular system similar to those

observed in true microgravity. However, conclusions drawn

from ground-based simulations may be limited due to the

presence of residual hydrostatic gradients. Regardless,

there is no conclusive evidence that zero gravity has any

unique or specific effects on the cardiovascular system that

cannot be reproduced during earth-bound simulations

(Blomqvist and Stone, 1983).

Changes in baroreflex sensitivity in human subjects has

been assessed using the 60 head-down tilt bedrest and

prolonged chair confinement models of microgravity

(Convertino et al., 1990a, 1991, 1992; Eckberg and Fritsch,

1992). Both Convertino et al. (1990a, 1991, 1992) and

Eckberg and Fritsch (1992) have determined carotid-cardiac

baroreflex gain of humans using the neck pressure/neck

suction technique. During a 30-day bedrest study, Convertino

et al. (1990a) noted a significant reduction in maximal slope

of the carotid-cardiac baroreflex at bedrest day 12 which

persisted until day 30. These observations are corroborated

by data collected during 10 days of head-down tilt bedrest

(Eckberg and Fritsch, 1992) and 4-5 days of Space Shuttle

flight (Fritsch et al., 1992). In both cases, there was a

tendency for a progressive reduction in maximal slope of the

carotid-cardiac baroreflex response over the course of

microgravity exposure. Interestingly, although data from

Convertino et al. (1990a) and Eckberg and Fritsch (1992)

verify that simulated microgravity elicits changes in

baroreflex gain similar to those noted following spaceflight

(Fritsch et al., 1992), the time course of change appears to

be different. Bedrest-induced reductions of baroreflex gain

occurred after 9-12 days (Convertino et al., 1990a; Eckberg

and Fritsch, 1992) whereas true microgravity apparently

elicited a reduction in reflex sensitivity after only 4-5

days (Fritsch et al., 1992).

Prolonged confinement to the seated posture has also

been used to investigate the effects of partial gravity on

baroreflex sensitivity. In these studies, baroreflex

desensitization appears to have occurred despite the presence

of a gravity vector along the long axis of the body and its

concomitant hydrostatic influence (Blomqvist and Stone,

1983). Convertino et al. (1991) measured vagally-mediated

carotid-cardiac baroreflex sensitivity in quadriplegic

subjects confined to wheelchairs for an average of 35 months

and reported that the stimulus-response relationship of this

reflex was decreased compared to that of age and weigh-

matched ambulatory controls. Similarly, it appears that

baroreflex function may have been compromised in healthy

subjects immobilized in the sitting posture for four to ten

days as suggested by a smaller increase in heart rate for a

given drop in blood pressure, i.e., decreased AHR/AMAP, and

increased incidence of hypotension during standing following

prolonged chair confinement (Lamb et al., 1964; Lamb et al.,

1965). Russian investigators have found similar responses

(loffe, 1968).

2.5.2 Orthostasis (1-g)

The term "orthostasis" is derived from the greek words

"ortho", meaning "caused by erect posture" and "stasis",

meaning "stationary" and can therefore describe responses to

any perturbation that produces effects similar to those

induced by the upright posture (Convertino, 1987a). The

primary hemodynamic effect of orthostasis is translocation of

blood from the central veins and heart to large veins in the

lower extremities caused by gravity or similar stimulus

(Blomqvist and Stone, 1983). "Orthostatic intolerance" refers

to the inability of cardiovascular reflexes to maintain

arterial pressure for adequate cerebral profusion due to

accumulation of blood in the lower extremities (Convertino,

1987a). Since mean arterial blood pressure is the product of

heart rate, stroke volume, and peripheral resistance, changes

in the physiological condition which affects one of these

variables, without appropriate compensation from the others,

will lead to an alteration in mean arterial pressure. If the

change is severe enough to reduce cerebral perfusion,

orthostatic intolerance and syncope may occur.

The incidence of orthostatic intolerance and syncope

during lower body negative pressure, standing, or head-up

tilt following exposure to both simulated and actual

microgravity is well documented (Berry, 1976; Blomqvist and

Stone, 1983; Bungo et al., 1985; Convertino et al., 1990a;

Greenleaf et al., 1989; Stegemann et al., 1975). Although

the array of mechanisms which cause orthostatic hypotension

are not entirely understood, inability to maintain blood

pressure during orthostasis appears to be a result of an

inadequate venous return due to reduced blood volume

(Blomqvist and Stone, 1983), depressed stimulus-response

relationship of the carotid-cardiac baroreflex (Blomqvist and

Stone, 1983; Convertino et al., 1990a, 1992; Eckberg and

Fritsch, 1992), altered cardiopulmonary (Convertino et al.,

1994a) and aortic (Crandall et al., 1993) baroreflexes, and

altered autonomic function (Eckberg and Fritsch, 1991).

Additionally, increased venous compliance of the lower

extremities (Buckey et al., 1992; Convertino et al., 1989a,

1989b; Convertino, 1994; Hoffler, 1977) and decreased cardiac

output due to diminished vascular volume (Blomqvist and

Stone, 1983; Greenleaf et al., 1989; Sandler, 1986) may also

contribute to reductions in orthostatic tolerance following

exposure to actual or simulated microgravity.

Increased heart rate during a provocative orthostatic

test has been the earliest and most consistently observed

alteration in cardiovascular function following microgravity

exposure (Hyatt, 1971; McCally and Wunder, 1971; McCally et

al., 1966, 1968; Miller et al., 1964b; Stevens et al., 1966).

Higher heart rates and lower stroke volumes for a given level

of orthostatic stress may not be indicative of a lower

orthostatic tolerance (Harper and Lyles, 1988; Hyatt et al.,

1969). Miller et al. (1964a) observed significantly elevated

heart rates during 30-min of 900 head-up tilt after bedrest,

but only 42% of subjects actually exhibited pre-syncopal

signs or symptoms. Fortney et al. (1991) determined

tolerance (defined as the level of LBNP tolerated without

symptoms of syncope) after 13 days of head-down bedrest and

reported that, although all subjects had elevated heart rates

during LBNP after bedrest, only half had a reduction in the

level of negative pressure tolerated. Thus, elevated heart

rate may be a compensatory response to gravitational stress,

but it does not necessarily indicate reduced tolerance.

The cardiovascular reflex response to an orthostatic

challenge is largely dependent on the magnitude of the volume

shift induced by experimental perurbation (Blomqvist and

Stone, 1983). Therefore, it is important that experimental

procedures which elicit a reproducible and quantifiable

volume redistribution are employed to study mechanisms

underlying cardiovascular responses to orthostasis (Blomqvist

and Stone, 1983). These techniques include upright tilt,

passive standing, and lower body negative pressure.

Head-Up Tilt. Head-up tilt involves rapidly elevating a

table to a position where the head is higher than the feet

while the subject is supported by a saddle or harness that

allows the legs to hang passively. This technique minmizes

muscular activity and provides a relatively reproducible

orthostatic stimulus. However, one limitation to this

technique is that head-up tilting can only produce a maximal

gravitational stimulus of 1-g (Convertino, 1987a).

The hemodynamic responses to varying degrees of head-up

tilt have been reported by several investigators.

Essentially, as subject is progressively shifted from the

supine to upright posture, blood pressure remains stable or

increases slightly, heart rate increases, and stroke volume

and cardiac output fall (Blomqvist and Stone, 1983; Hyatt,

1971; Melada et al., 1975; Poliner et al., 1980).

Physiologic responses to head-up tilt are altered after

exposure to microgravity (Blomqvist et al., 1980; Butler et

al., 1991; Greenleaf et al., 1985; Greenleaf et al., 1989;

Hoffler, 1977; McCally et al., 1966; Vogt, 1967). Hoffler

(1977) subjected 18 Gemini astronauts to 25-min of 700 head-

up tilt and noted that supine resting and tilt-induced heart

rates increased an average of 29% and 51%, respectively after

flight. In addition, there was a 70% increase in leg volume

and a marked narrowing of the pulse pressure during post-

flight tilting (Hoffler, 1977). Similarly, Vogt (1967)

reported greater increases in heart rate and a faster decline

in blood pressure during tilt after 12-h of recumbency

compared to control and Miller et al. (1964a) observed

significantly elevated heart rates during 30-min 90 head-up

tilt after bedrest. In general, maximal heart rate during

700 head-up tilt is increased 15-35 bpm before, and 20-60 bpm

following, bedrest (Chobanian et al., 1974; Hyatt, 1971;

Lancaster and Triebwasser, 1971; Thorton and Hoffler, 1977;

Vogt, 1967).

Passive Standing. Passive standing is a widely used

technique to evaluate the cardiovascular responses to

orthostasis (Blomqvist and Stone, 1983). Although economical

and easy to conduct, this method is less than ideal because

it promotes muscle contraction in the legs which have the

potential to reduce pooling of blood in the lower extremities

and increase venous return, thus enhancing atrial filling and

cardiac output. Such muscle contractions are impossible to

control from subject to subject during any given test

protocol (Blomqvist and Stone, 1983). Regardless, changes in

heart rate to this procedure are frequently used as an index

of the cardiovascular response to an orthostatic stress

before and after microgravity exposure. In general,

increases in heart rate average 35 bpm during 5-min stand

tests before bedrest (Hyatt et al., 1975; Lancaster and

Triebwasser, 1971) and nearly 70 bpm after two weeks of

bedrest (Harper and Lyles, 1988). Furthermore, the fall in

stroke volume and cardiac output are about twice as great

(Harper and Lyles, 1988). Interestingly, Convertino et al.

(1990a) have also demonstrated a significant linear

correlation between microgravity-induced changes in carotid-

cardiac baroreflex gain and changes of systolic blood

pressure during standing after 30 day bedrest. Baroreflex

gain was reduced more in subjects who fainted during standing

after bedrest than in those who did not (Convertino et al.,


Lower Body Negative Pressure. LBNP has been used

extensively as a research tool to evaluate the cardiovascular

response to orthostatic stress (Hoffler, 1977; Wolthuis et

al., 1974). This technique has the inherent advantage of

producing a gravity-independent redistribution of venous

volume. Furthermore, because subjects remain supine during

LBNP, measurement of physiologic responses is facilitated and

the likelihood of confounding activity in skeletal muscle is

minimized (Convertino, 1987a).

Several investigators have demonstrated that LBNP at -40

to -50 mmHg results in blood pressure and heart rate changes

similar to those seen with passive standing and head-up tilt

(Hoffler, 1977; Musgrave et al., 1969; Wolthuis et al.,

1974). Changes in maximal orthostatic capacity following

bedrest or true microgravity exposure has also been

determined with LBNP (Hordinsky et al., 1981; Sandler et al.,

1985). Hoffler (1977) reported that 18 Apollo crewmembers

exhibited elevations in heart rate that averaged 42% higher

after flight during LBNP to -50 mmHg. Additionally, pulse

pressure was reduced by 25% during the orthostatic stress

with 7 astronauts experiencing pre-syncopal symptoms during

their immediate post-flight LBNP test (Hoffler, 1977). Data

from groundbased experiments suggest that prior to bedrest,

80-90% of all subjects can tolerate a minimum of 15 minutes

of -50 mmHg without experiencing syncopal symptoms.

Following bedrest, however, time to syncope is reduced by

nearly 50% (Melada et al., 1975; Miller et al., 1964b;

Sandler, 1980; Sandler and Winter, 1978; Stevens et al.,

1966). Bedrested subjects exhibit heart rate responses to

LBNP which are similar to those observed in flight crews

following exposure to true microgravity with heart rate

usually doubling during LBNP following weightlessness

(Greenleaf et al., 1976, 1982). Interestingly, females

undergoing LBNP have shown greater increases in heart rate

both before and after bedrest (Sandler and Winter, 1978).

2.6 Effects of Exercise on Baroreflex Function

2.6.1 Endurance Exercise Training

It is well-known that endurance (aerobic) exercise

training of moderate intensity and long duration increases

maximal oxygen uptake and blood volume (Saltin et al., 1968).

Conversely, microgravity induces a significant reduction in

both of these parameters (Convertino, 1987b; Fischer et al.,

1967; Johnson et al., 1977a; Leach and Johnson, 1984).

Interestingly, it has been reported that the incidence of

orthostatic intolerance following exposure to actual

spaceflight (Convertino, 1990; Hoffler, 1977) and simulated

microgravity (Hyatt and West, 1977; Sandler, 1986) is closely

related to the magnitude of blood volume loss. Therefore, it

seems reasonable to consider that adherence to an endurance

exercise training program while inhabiting a low gravity

environment would maintain or increase blood volume and help

defend against post-exposure loss of orthostatic tolerance.

However, data from both spaceflight and ground-based

experiments do not support this hypothesis.

Extensive endurance exercise training was used during

the U.S. Skylab program and, although the conditioning

program was effective in maintaining a high level of crew

physical fitness (Convertino, 1990), plasma volume was

reduced and orthostatic instability was increased following

return to earth (Convertino, 1990; Johnson et al., 1977b).

Greenleaf et al. (1989) assessed the effects of exercise

training during 30 days of bedrest on orthostatic

hypotension. Cardiovascular responses and time to pre-

syncope during 600 head-up tilt were measured in three groups

of subjects who had performed no exercise, endurance

exercise, or resistive exercise during the bedrest period.

Only subjects who participated in the endurance exercise

training program maintained plasma volume at pre-bedrest

levels. However, post-bedrest tilt tolerance times were

reduced by the same magnitude in all groups (Greenleaf et

al., 1989). Since blood volume was maintained in the

endurance-trained subjects, these data provide evidence that

physiological mechanisms other than vascular volume must be

important in determining orthostatic tolerance following

prolonged exposure to microgravity.

In addition to reduced blood volume, baroreflex

responses are impaired by microgravity and its analogs

(Convertino et al., 1990a, 1991a, 1992; Eckberg and Fritsch,

1992; Fritsch et al., 1992; Nicogossian et al., 1991) with

the degree of impairment related to the duration of exposure

(Convertino et al., 1990a; White et al., 1991). Altered

baroreflex function has also been associated with orthostatic

instability following bedrest (Convertino et al., 1990a).

The results from spaceflight (Johnson et al., 1977b) and

simulated microgravity (Greenleaf et al., 1989) studies

suggest that training regimens designed to defend aerobic

capacity and increase blood volume may not provide the

appropriate stimuli to reverse or attenuate baroreflex

impairment (Convertino et al., 1990b; Seals and Chase, 1989;

Tatro et al., 1992).

2.6.2 Dynamic Maximal Exercise

There is evidence to suggest that graded, dynamic

maximal exercise can acutely restore various cardiovascular

and metabolic capacities attenuated by exposure to simulated

microgravity or physical deconditioning. One bout of maximal

exercise following 10 days of detraining returned insulin

sensitivity to normal levels (Heath et al., 1983).

Similarly, a single bout of exhaustive treadmill exercise at

the end of 10 days of bedrest restored maximal oxygen uptake,

heart rate, blood pressure, time to fatigue, and orthostatic

stability to pre-bedrest levels within two hours of

ambulation (Convertino, 1987b). Furthermore, performance of

exhaustive exercise provided protection against orthostatic

instability and syncope following water immersion (Stegemann

et al., 1975).

There are also reports that a single bout of dynamic

exercise enhances blood pressure regulation during the post-

exercise recovery period. This observation has been

attributed to an alteration in sensitivity of cardiovascular

reflexes mediated by arterial baroreceptors. Somers et al.

(1985) reported a 50% increase in the gain of the cardiac

baroreflex stimulus-response relationship 60 minutes after

completion of fatiguing cycle exercise. Convertino and Adams

(1991) observed a 92% increase in baroreflex sensitivity 3

hours after cessation of maximal leg exercise with the effect

maintained for up to 24 hours. Also, a 32% increase in

carotid baroreflex gain has been reported 24 hours after

completion of a bout of resistance exercise in bedrested

subjects (Convertino et al., 1992).

The mechanisms responsible for increased carotid

baroreflex sensitivity following a single bout of intense

exercise are unclear. Several investigators have reported

rapid resetting of the baroreflex at the. onset of exercise

(Ebert, 1986; Hales and Ludbrook, 1988; Rowell, 1986;

Shepherd and Mancia, 1986) due to either changes in central

modulation of the reflex arc or to changes at the site of the

mechanoreceptors (Chapleau and Abboud, 1989; Chapleau et al.,

1989), but this seems an unlikely or incomplete explanation

because resetting involves a shift of the stimulus-response

curve along the stimulus axis without a change in sensitivity

(gain) of the reflex. Similarly, alterations in ionic

environment and circulating hormones, which have been shown

to change arterial baroreceptor sensitivity (Holmes and

Ledsome, 1984; Munch and Brown, 1985; Munch et al., 1987;

Tomomatsu and Nishi, 1981), do not appear to explain changes

in baroreflex sensitivity since elevations in these agents

caused by exercise are transient and return to baseline

levels before changes in baroreflex function occur. Changes

in vascular volume caused by acute exercise (Gillen et al.,

1991) should have little contribution to the observed changes

in baroreflex function since acute and chronic changes in

blood volume are not related to alterations in the carotid-

cardiac vagal baroreflex response (Convertino, 1992;

Convertino et al., 1990a; Thompson et al., 1990). Input from

muscle chemosensitive (metaboreceptor) afferents may also

modulate autonomically-mediated blood pressure control

mechanisms following exercise, possibly in the central

cardiovascular centers of the brain stem (O'Leary, 1993), but

this hypothesis awaits further research.

There is evidence to suggest that loading arterial

baroreceptors by periodic elevation of systolic blood

pressure may provide a stimulus to maintain or increase

sensitivity of the baroreflexes. The observation of a

reduced baroreflex gain following long-term restriction from

a 1-g environment via spaceflight (Fritsch et al., 1992),

bedrest (Convertino et al., 1990a, 1992; Eckberg and Fritsch,

1992), or chair confinement (Convertino et al., 1991) may

underscore the importance of periodic loading and unloading

of baroreceptors in maintaining reflex integrity. This

hypothesis is supported by the observation that pulsatile

increases in arterial pressure altered carotid baroreceptor

sensitivity; an effect which persisted for several minutes

after removal of the stimulus (Chapleau and Abboud, 1989).

Additional support comes from Eiken et al. (1992) who used a

combination of submaximal exercise and lower body positive

pressure to load the baroreceptors and observed an enhanced

gain of the carotid-cardiac baroreflex.

The mechanisms by which pulsatile elevations of pressure

induce changes in baroreflex sensitivity remain unclear.

Chapleau and Abboud (1989) noted pulse pressure-induced

baroreflex sensitization could not be explained by changes in

viscoelastic properties of the carotid sinus nor by release

of endothelial or ionic substances. Interestingly, increased

pulse pressure appears to inhibit sympathetic nerve activity

within the medullary cardiovascular centers possibly leading

to central modulation of the baroreflex (Chapleau et al.,

1989). Thus, increased pulse pressure which accompanies

increased arterial pressure elicited by dynamic exercise may

provide a stimulus which acts in the central nervous system

to heighten the sensitivity of the carotid-cardiac


2.7 Summary

Reflex responses of the cardiovascular system to

orthostasis are illustrated in Figure 2.1. With the onset of

orthostatic stress, large quantities of blood accumulate in

the lower extremities due to an increased head-to-toe

hydrostatic gradient, thus compromising venous return, atrial

filling,and cardiac output. Subsequent reductions of central

venous and arterial pressures elicit strong cardiopulmonary

and arterial baroreceptor responses which initiate a cascade

of neuro-humoral responses to restore blood pressure

homeostasis. Failure of these mechanisms to respond with

sufficient speed and magnitude will result in orthostatic

intolerance and possible syncope.

Also indicated in Figure 2.1 are components of the blood

pressure control system whose function is enhanced following

performance of a single bout of maximal exercise.

Specifically, it has been demonstrated that exhaustive

exercise enhances carotid-cardiac baroreflex sensitivity and

increases plasma volume, however, the importance of these

adaptations to blood pressure stability during provocative

orhtostatic testing has not been determined. Therefore, data

presented in the following chapters provides evidence that

exercise-induced increases of baroreflex sensitivity and

plasma volume are associated with improved blood pressure

control and increased orthostatic tolerance. These data also

identify other blood pressure regulatory variables and their

interactions influenced by exercise as well as identify areas

for future research.

2.8 Conclusions

1. Arterial baroreceptors are mechano- (stretch) receptors

which are stimulated by deformation of the vascular wall

in which they are located. These receptors respond to

changes in intravascular pressure by initiating actions

which return mean arterial pressure to homeostatic


2. The bradycardic response to baroreceptor stimulation is

mediated via vagal cholinergic mechanisms. Controversy

remains as to mechanisms mediating the tachycardic

response, although evidence exists which suggests both

sympathetic and parasympathetic involvement.

3. Cardiopulmonary baroreceptors are located on the low

pressure (venous) side of the circulation and repond to

alterations in blood volume by eliciting reflex changes

in systemic resistance.

4. Baroreflex function is evaluated via invasive and non-

invasive techniques which test either integrated

baroreceptor function or isolated reflex arcs. The

method utilized to assess reflex function will determine

the conclusions which can be drawn from the data.

5. The operating range of baroreflex stimulus-response

relationships is labile and readily shifts in the

direction of prevailing arterial pressure. There is

evidence to suggest that increases in intravascular

pressure which are pulsatile in nature prevent

baroreflex resetting, but elicit changes in baroreceptor

sensitivity. This may also be the case with activation

of muscle chemosensitive receptors.

6. Circulating hormones such as norepinephrine, angiotensin

II, vasopressin, and atrial natriuretic peptide modify

baroreceptor activity.

7. Baroreflex function is sensitive to changes in the

gravitational environment. This is readily demonstrated

by observations of altered baroreflex function following

prolonged spaceflight, bedrest, and chair confinement.

Altered baroreflex activity may play a role in

development of orthostatic hypotension following

microgravity exposure.

8. Unlike conventional endurance exercise training which

has no apparent effect on baroreceptor sensitivity, a

single bout of maximal dynamic exercise appears to

provide a stimulus which enhances sensitivity of the

carotid-cardiac baroreflex and increases plasma volume.

Although the mechanism of this action remains unclear,

recent evidence in animal models suggests that pulsatile

increases of arterial pressure may play a role. Also,

afferent input from muscle chemosensitive receptors may

influence sensitivity of baroreflexes, possibly through

action in the medullary cardiovascular control centers.



SVenous Pooling .(- Leg

Venous Return

Central Venous

Baroreceptor -

a ,

Mean Arterial


- Aortic t


Muscle Chemoreceptors

/ Na+





t Heart Rate

t Stroke volume

t Veno- A Arterial
constriction t Constriction
I -- --

Plasma Volume t, A Vasocontrictive
A Reserve


Output -

Mean Arterial Total Peripheral
Pressure --- Resistance


Figure 2.1. Cardiovascular reflex responses to orthostasis.
Asterisks indicate where maximal exercise is known to impact
mechanisms responsible for blood pressure control. Questions
marks identify where exercise may potentially exert an
effect, but confirmation awaits future research.




3.1 Experimental Design

Both Studies 1 and 2 employed a cross-over design in

which the order of experimental treatments was randomized.

Using the subjects as their own controls minimized the

potential for heterogeneous responses to experimental

interventions across groups of different subjects and

therefore served to decrease the probability of making a Type

I error due to relatively small sample size.

3.2 Experimental Procedures and Protocols

3.2.1 Study 1

For Study 1, paraplegic subjects confined to wheelchairs

due to complete traumatic spinal transaction in the thoracic

region were recruited. Because restriction to the seated

posture has been associated with postural hypotension due to

partial removal of the hydrostatic gradient along the

longitudinal axis of the body (Blomqvist and Stone, 1983;

Lamb et al., 1965), these subjects provided an accessible

model of the long-term effects of microgravity on blood

pressure homeostasis. All subjects completed two

experimental protocols. Each protocol involved measurement

of heart rate (HR), blood pressure (BP), forearm vascular

resistance (FVR), and vasoactive hormone (NE, AVP, PRA)

responses before and during 15 minutes of 70 head-up tilt on

two days: 1) 24 hours after acute intense arm exercise, and

2) during a control (no exercise) condition. Additionally,

measurements of plasma volume, carotid-cardiac baroreflex

relationship, leg vascular compliance, and hemodynamic

responses to the Valsalva maneuver were made under the two

experimental conditions. Test days were separated by one

week and the two experimental conditions were given in random

order. This study was conducted in the hospital laboratory

at the Humana-Lucerne Spinal Injury Unit in Orlando, Florida.

3.2.2 Study 2

Study 2 required that ambulatory subjects be exposed to

two 16 day periods of 60 head-down tilt (HDT) separated by 11

months. The experimental protocol for this study consisted

of 4 days of ambulatory control (Cl-C4) followed by 16 days

of 60 HDT (BR1-BR16) and 2 days of post-HDT recovery (Rl-R2)

for a total of 22 days of confinement. During one HDT

period, each subject performed an acute bout of graded

exercise designed to elicit maximal effort twenty-four hours

prior to re-ambulation (exercise condition). Following the

other, subjects performed no exercise (control condition).

During both HDT periods, subjects lived 24 h-day-1 in

the Human Research Facility at NASA-Ames Research Center,

Moffett Field, CA and remained head-down without interruption

for all daily activities except bathing, for which they were

horizontal. The 16 day HDT period was chosen because it

represents the maximum projected duration for Extended

Duration Orbiter Space Shuttle missions. The 60 HDT posture

was chosen because changes in cardiovascular reponses

elicited by this posture are similar to those seen in actual

spaceflight (Blomqvist and Stone, 1983; Convertino et al.,

1989b). Subjects followed the same controlled diet (average

daily caloric intake of 2500-2800 kcal: 45% carbohydrate, 38%

fat, 17% protein) during both 22 day experimental periods.

Dietary sodium and potassium were held constant at

approximately 120 and 60-80 mEq-day-1 respectively, and fluid

intake was ad libitum. The photoperiod was 16 hours light to

8 hours dark with lights on at 0700 hours. All measurements

where conducted at the same time of day and in the same

sequence before, during, and after HDT.

Plasma volume and the stimulus-response relationship of

the carotid-cardiac baroreflex were measured on day C-4, day

BR-16, and day R-1. A 30 ml antecubital venous blood sample

was taken without stasis at 0700 h on these days for the

baseline determination of plasma NE. On day C-4 and day R-1,

each subject underwent a test of orthostatic tolerance using

a graded LBNP protocol to pre-syncope. An additional blood

sample was obtained immediately following termination of the

LBNP test to determine the catecholamine response to


3.3 Subiects

A total of 17 healthy, non-smoking, normotensive

volunteers (15 males, 2 females; Table 3.1) gave their written

consent to serve as subjects for these two investigations

after they had been informed of all procedures and risks. All

Table 3.1. Physical characteristics of subjects.

Study 1 Study 2

n = 10 n = 7

Age, yr 36 4 40 2

Height, cm 185 2 183 2

Weight, kg 90 7 81 2
Values are mean SE.

procedures were approved by the Human Research Review Boards

of NASA-Kennedy Space Center and NASA-Ames Research Center.

Selection of subjects was based on results of a screening

evaluation comprised of a detailed medical history, physical

examination, blood chemistry analysis, urinalysis, chest X-

ray, and resting and treadmill electrocardiogram. During an

orientation session conducted prior to the study, all

subjects were made familiar with the laboratory surroundings,

laboratory personnel, and experimental procedures and

protocols. All subjects were asked to abstain from tobacco,

alcohol, caffeine, medications, and conventional exercise for

the duration of the studies.

3.4 Measurements and Techniques

3.4.1 Heart Rate and Blood Pressure

A direct writing, multi-pen recorder was used to

transcribe the electrocardiogram (ECG) and BP recordings

during all tests. Continuous HR was computed from the R-R

interval of the ECG (Quinton Q4000 or Lifepack-6) from a

three lead electrode placement. A Finapres fingercuff BP

monitoring device (Ohmeda 2300) was used to provide

continuous beat-by-beat measurement of peripheral arterial

BP. The Finapres measures BP using a small finger cuff that

contains a photoplethysmographic volume transducer and an

inflatable air bladder. The cuff is connected to a fast-

response servo control system that instantaneously regulates

the pressure applied to the finger through the bladder and,

thus the pressure applied to the walls of the arteries. As

BP increases, the arterial wall expands, increasing the

volume of the finger. This volume differential is measured

by the plethysmographic transducer. The Finapres monitor

responds to the increasing volume by increasing cuff pressure

until the original arterial size and blood volume are again

reached. The external pressure continuously adjusted by the

cuff closely follows the intra-arterial pressure within the

finger, allowing measurement of the external pressure itself

as a function of the arterial blood pressure (Boehmer, 1987).

Thus, the Finapres provides a reliable non-invasive technique

for the measurement of changes in beat-by-beat arterial blood

pressures. Additionally, periodic BP measurements were made

with a stethoscope and sphygmomanometer to verify the

readings obtained from the Finapres.

3.4.2 Carotid Baroreflex Function

Carotid baroreceptor-cardiac reflex responses were

measured with the method described by Sprenkle et al. (1986).

Stimuli were delivered to the carotid baroreceptors by a

computer-controlled motor-driven bellows which provided

pressure steps to a silastic neck chamber covering the area

of the carotid sinuses. An initial pressure of 40 mmHg was

delivered to the chamber and maintained for four R waves.

With the next R wave, the pressure sequentially stepped to

approximately 25, 10, -5, -10, -20, -35, -50, and -65 mmHg

followed by a return to ambient pressure. Pressure steps

were triggered by R waves so that neck chamber pressures were

superimposed upon naturally occurring cardiac pulses. To

avoid respiration-related variations of cardiac vagal

outflow, neck-pressure changes were applied only while

subjects held their breath at mid-expiration. With this

system, the pressure gradient across the arterial walls was

rapidly altered to produce reflex cardiac acceleration or

slowing as measured by changes in the R-R interval. Carotid

distending pressure (CDP) was calculated as systolic pressure

minus neck-chamber pressure applied during each heart beat.

This calculation assumes complete transfer of pressure in the

neck chamber to the carotid arteries. A test session

consisted of five successful applications of neck-pressure

sequences. Each sequence lasted ~15 seconds and each test

session lasted approximately 15 minutes. Individual trials

were discarded immediately if the subject breathed during the

stimulus sequence or if the neck chamber failed to seal

adequately. Blood pressure was measured with a

sphygmomanometer prior to the application of the stimulus.

From the average of the five trials, a stimulus-cardiac

response curve was derived by plotting R-R intervals at each

pressure step against their respective CDP. This plot was

used to define threshold, saturation, and linear ranges of

the sigmoid baroreceptor-cardiac reflex response. Previous

studies have indicated that baroreceptor stimulus-sinus node

response relationships measured in this manner are highly

reproducible (Eckberg et al., 1992; Kasting et al., 1987).

Sensitivity of the baroreflex was determined by calculating

the maximum slope of this sigmoid curve. To determine the

segment of steepest slope, least squares linear regression

analysis was applied to every set of three consecutive points

on the response relationship. According to Ludwig and

Convertino (1991), the maximum slope of the stimulus-response

curve is the best indicator of baroreflex function.

R-R intervals were used to characterize the pressure

input-neural output relationship because of the relationships

that exist between R-R intervals and vagal-cardiac nerve

activity and HR. R-R intervals are highly linear functions

of vagal-cardiac nerve activity (Katona et al., 1970;

Nicogossian et al., 1991). As a result of this linear

relationship, it is possible to use changes of R-R intervals

as surrogates for changes of vagal-cardiac nerve activity,

and to compare responses before and after interventions such

as HDT, which alter baseline R-R intervals. Because the

relationship between HR and vagal-cardiac nerve activity is

curvilinear, it is extraordinarily difficult to compare

responses to baroreflex forcing when baseline HRs are

different. Eckberg et al. (1976) have found that R-R

interval responses to neck pressure sequences similar to

those employed in this study are not reduced by beta-

adrenergic blockade, but are nearly abolished by muscarinic

blockade. Therefore, this measurement of carotid-cardiac

baroreflex response focuses primarily on the vagal limb of

the baroreceptor reflex.

3.4.3 Tests of Orthostatic Responses

Cardiovascular and vasoactive hormone responses to

orthostasis were recorded in all subjects during either 70

head-up tilt (Study 1) or lower body negative pressure (LBNP;

Study 2). In either case, the duration of the orthostatic

test was determined by any one or combination of the

following criteria: (a) completion of a pre-determined time

limit; (b) onset of presyncopal symptoms such as grey-out, a

precipitous fall in systolic blood pressure greater than 15

mmHg, and/or a sudden bradycardia greater than 15 bpm between

adjacent 1-min measurements; (c) progressive diminution of

SBP below 80 mmHg; and (d) voluntary subject termination due

to discomfort such as sweating, nausea, or dizziness (Sather

et al., 1986).

70 Head-Up Tilt. A motorized tilt table elevated

subjects from the supine posture to the 700 head-up position

within 10 seconds. Subjects remained in this position for 15

minutes or until the onset of pre-syncopal symptoms. During

tilt, beat-to-beat HR was continually recorded. Blood

pressure was measured on the right arm at 2 minute intervals

while forearm blood flow was measured on the left arm

according to the protocol described below. At the conclusion

of the test, the subject was quickly returned to the supine

posture and a blood sample was drawn within 30 seconds to be

analyzed for NE, AVP, and PRA. An identical procedure was

repeated one week later to measure the same variables under

the remaining treatment condition.

Lower Body Negative Pressure. LBNP is a means of

producing circulatory alterations similar to those

encountered when changing posture from supine to upright

standing. Following instrumentation for measurement of HR,

BP, and forearm bloodflow, subjects in Study 2 were assisted

into the supine position within a wooden LBNP pressure

chamber. A foam-padded saddle was adjusted to comfortably

support and stabilize the subject in the chamber. The

location of the saddle was noted to ensure consistency in

subject positioning during all LBNP tests. A rubberized

skirt was used to secure an airtight seal between the chamber

and the waist (identified as the iliac crests) of the

subject. Orthostatic tolerance was determined by

progressively reducing the pressure around the lower body

relative to ambient pressure. The protocol consisted of a 5

minute resting period followed by decompression to -5, -10, -

15, -30, and -40 mmHg for 2, 2, 15, 15, and 5 minutes,

respectively. Supplemental 10-mmHg reductions in pressure

every 2 minutes were added until test termination.

Immediately following cessation of the test, an antecubital

blood sample was obtained for determination of NE.

3.4.4 Blood and Plasma Volume

Plasma volume was determined using a modified Evans

blue dye (T-1828; Macarthy Medical, Romford, Essex, UK)

dilution method (Greenleaf et al., 1979). After a minimum of

30 minutes supine rest, a control baseline sample was drawn

followed by an intravenous injection of 11.5 mg of dye pre-

diluted with isotonic saline (2.5 ml) through a sterile 0.45

pm filter. A 5-ml aliquot of the control baseline sample was

used to determine the pre-injection dye concentration.

Plasma (1 ml) from a 10-minute post-injection blood sample

was passed through a wood-cellulose powder (Solka-Floc SW-

40A) chromatographic column so that the dye could be absorbed

after it had been separated from the albumin by the action of

a detergent (Teepol 610 in 2% Na2HPO4). In addition,

interfering substances such as pigments, proteins, and

chylomicrons were washed from the column with 2% Na2HPO4.

The absorbed dye was eluted from the column using a 1:1

water-acetone solution (pH 7.0) and collected in a 10-ml

volumetric flask. The post-injection solution was compared

with 1-ml samples from a pre-injection time (zero control)

and a standard dye solution (1:50 dilution with distilled

water). All samples were read at 615 nm with a

spectrophotometer. Plasma volume was calculated using a

standard dilution equation as follows:

PV = [(V*D)*(St*v)]/(1.03*T)

PV = plasma volume (ml)
V = volume of dye injected (ml)
D = dilution of standard (1:250)
St = absorbance of standard
v = volume of sample extracted (ml)
T = absorbance of plasma sample
1.03 = correction factor for dye uptake by tissues

Hemoglobin concentration was determined with triplicate

measurements using using the Coulter S+4 system. Hematocrit

(Hct) was measured in triplicate with a microhematocrit

centrifuge and a micro-capillary tube reader. Total blood

volume (TBV) was calculated from the plasma volume and Hct

using the following equation (Convertino et al., 1980):

TBV (ml) = PV/(1.0 0.91 Hct)

0.91 = correction factor for venous/total body
hematocrit (Chaplin et al, 1953)

Greenleaf et al. (1979) have reported that, using these

procedures, the test-retest correlation coefficient for blood

volume was 0.969 (N = 12) and the day-to-day variation was 82

ml (1.5%, N = 17) over 4 days, 75 ml (1.5%, N = 19) over 8

days, and 56 ml (1.1%, N = 23) over 15 days.

3.4.5 Vasoactive Hormones

Antecubital venous blood samples (30 ml) were drawn

without stasis for the determination of NE, AVP, and PRA.

Immediately following each withdrawal, 21 ml of whole blood

was taken from the vacutainer and transferred to a chilled

tube containing sodium EDTA. The remaining 9 ml was

introduced into a lithium heparinized tube. Whole blood was

also taken directly from the EDTA tube for triplicate

microhematocrit (approximately 0.5 ml) and for hemoglobin

approximatelyy 0.5 ml) measures. The remaining EDTA-treated

whole blood was centrifuged at 2000 g for 15 min at 40C.

Immediately after centrifugation, the plasma was aliquoted

for NE, AVP, and PRA and stored at -600C until hormonal

assays were performed.

Norepinephrine. Plasma NE concentrations were measured

by high performance liquid chromatography (Waters). NE was

extracted by absorbing plasma samples onto alumina.

Following washing of the absorbed alumina with a dilute

buffer solution, catecholamines were eluted from the alumina

when treated with an acidic solution. 3,4-

Dihydroxybenzylamine was used as an internal standard and

extraction efficiency of NE was based on the extraction of

known standards. After extraction, the samples were assayed

using a Waters 712 Wisp to inject the samples onto a reverse

phase C18 column. A Waters 460 electrochemical detector was

used to determine the concentrations of NE in the samples.

The auto-injector and detector were interfaced with a Digital

380 computer using Waters software. The within assay

coefficient of variability was 1.4% and between assay

coefficient of variability was 3.8%.

Arginine Vasopressin. Radio-immunoassay (RIA)

procedures were used to analyze plasma for AVP (Instar

Nuclear Corp.). To determine AVP, samples were extracted

using ODS-silica columns and then assayed using a

disequilibrium RIA procedure. The sample and first antibody

were incubated for 18 h at 2-80C. 125I tracer was added

followed by a second incubation at 2-80C. A second antibody

precipitating complex was added for phase separation. The

supernatant was then poured off and the precipitate counted

in a gamma counter. Spiked recovery was 92%, sensitivity was

0.5 pg-ml-1, within assay coefficient of variability was 2.8%

and between assay coefficient of variability was 9.9%.

Plasma Renin Activity. To determine PRA, an EDTA plasma

sample was adjusted to pH 6 with maleic acid. Half of the

sample was incubated in a 370C water bath for 2 h and the

other half was kept in an ice bath at 40C. The incubated

portion reflects the circulating level plus the quantity of

ANG I generated through the action of the renin in the plasma

sample. The net quantity of angiotensin I at 370C was

calculated by subtracting the ANG I level in the 40C sample

from the ANG I level in the 370C sample. Measurement of PRA

was performed by a RIA competitive binding procedure using a

specific antibody, a radiolabeled antigen, a pure sample of

antigen used as a reference standard, and a separation

medium. The amount of unlabeled antigen in the sample being

analyzed was determined by comparing the assay results to a

standard curve prepared with known amounts of the unlabeled

antigen. Recovery efficiency was 96%, sensitivity was 0.1

ngnml--lh-1, within assay coefficient of variability was 2.7%

and between assay coefficient of variability was 5.5%.

3.4.6 Forearm Blood Flow

Forearm blood flow (FBF) was measured using venous

occlusion plethymography employing a dual loop mercury-in-

silastic strain gauge (Whitney, 1953). The strain gauge was

calibrated before and after all experimental sessions in

which FBF was measured and the ouput of the strain gauge

circuitry was directed to a 2-channel chart recorder (Gould).

Approximately one minute prior to each blood flow

measurement, a wrist occlusion cuff was inflated to +250 mmHg

to impede blood flow to the hand. This cuff remained

inflated for the duration of blood flow determination.

Following wrist cuff inflation, a cuff placed around the arm

just above the elbow was inflated to 40 mmHg at 10 second

intervals for two minutes to prevent the efflux of venous

blood from the forearm, thus causing the strain gauge to

stretch. By treating the forearm as a cylinder in which an

increase in volume results in an increase in cylinder

circumference but not length, the amount of stretching can

then be used as a quantitative index of venous volume. Blood

flow was calculated by manually drawing a tangent to the line

of the first three pulse waves and measuring this slope in

terms of percent circumference change vs. time. Percent

volume change per minute (ml-100ml-1*min-1) was calculated

from circumference change. Therefore, the linear change in

forearm circumference during the venous occlusion was used to

calculate blood flow as follows:

V = [(2 C2) Cl-1] 100

FBF = V T-1

C1 = initial forearm circumference (cm)
C2 = change in forearm circumference (cm)
V = change in forearm volume (ml-100 ml tissue-1)
T = time for measurement (min)

Forearm blood flow measurements were averaged to provide one

value representing mean forearm blood flow for a given

perturbation. Forearm vascular resistance (FVR) was

calculated by dividing the mean arterial pressure by forearm

blood flow.

3.4.7 Leg Compliance

Compliance of the left leg was measured during supine

rest using a Whitney strain gauge placed at the point of

greatest calf circumference (Thornton and Hoffler, 1977).

Following 30 minutes of supine control, the left leg was

slightly elevated (~4 inches) at the ankle and an occlusion

cuff placed just above the knee was inflated to 40 mmHg for

120 seconds. Leg compliance was calculated by dividing the

volume change (ml*100ml-1) at a plateau (i.e., point at which

venous pressure equals cuff pressure) by the cuff pressure

and expressed as Avol%*AmmHg-1. The value for leg compliance

was multiplied by 100 for convenience.

3.4.8 Lea Volume

Changes in left leg volume during each LBNP stage were

measured with a strain gauge placed around the point of

maximal calf girth. Percent changes in calf volume (ml-100

ml-1) were calculated from circumference changes. For small

changes in gauge circumference, resultant cross-sectional

area (and volume) changes were doubled.

3.4.9 Exercise Bout

An acute bout of exercise was performed by all subjects

24 hours prior to measurement of autonomic, hormone, cardiac,

and hemodynamic activity during the orthostatic tests. For

Study 1, subjects performed a multistage graded exercise

protocol utilizing a Monark arm-crank ergometer. After a 4-

minute warm-up against no load, the work rate was set at 10

watts increased by 5 watts every 2-minutes until fatigue.

For Study 2, exercise consisted of graded work intensities to

volitional fatigue utilizing a Quinton (model 845B)

electronically-braked supine cycle ergometer. After a 4-

minute warm-up at 200 kpm*min-1, the work rate was increased

by 100 kpm-min-1 every minute until exhaustion. During both

tests, a tachometer placed at eye level assisted the subject

in maintaining a rotational cadence of 60-70 rpm. Both arm-

crank and leg exercise protocols were designed so that the

duration of the treatment was approximatley 15 minutes.

Continuous, multistage graded workrate protocols designed to

elicit maximal effort were chosen since similar tests have

provided evidence of causing acute elevations in baroreflex

sensitivity in normotensive subjects (Convertino and Adams,

1991) and hypertensive patients (Somers et al., 1985). HR

was recorded during the last 15 seconds of each minute while

BP was measured by brachial artery auscultation before and

immediately after cessation of exercise. Subjects were

verbally encouraged to achieve a maximal effort. The

exercise bout was terminated when the subject reached

volitional fatigue and was unable to maintain the required

cadence for a period exceeding 15 seconds.

3.4.10 Valsalva Maneuver

Subjects in both studies performed three Valsalva

maneuvers according to a strict protocol at a controlled

expiratory pressure of 30 mmHg. Each trial included a 30-

second baseline period of quiet breathing, a 15-second strain

period, and a 2-minute post-strain period. After instruction

in the technique and several practice trials, subjects were

asked to take a normal inhalation and blow into a mouthpiece

connected to a pressure transducer. A small leak in the

system prevented the subject from maintaining the expiratory

pressure by occluding the glottis. A pressure gauge

positioned in front of the subject provided feedback on the

expiratory pressure. The expiratory pressure was used as an

index of intra-thoracic pressure changes during Valsalva

maneuvers (Elisberg, 1963). Continuous HR was recorded with

an ECG and beat-to-beat BP responses were measured with the

Finapres. Excellent estimates of directly measured intra-

arterial pressures during Valsalva maneuvers have been

demonstrated with this device (Imholz et al., 1988). Both HR

and BP responses were saved as a digital record and data from

the three trials were averaged.

HR and BP changes during the four phases of the Valsalva

maneuver (Hamilton et al., 1936) were analyzed in a phase-by-

phase manner (Ten Harkel et al., 1990). For comparison

between experimental conditions, blood pressure responses

(AMAP) to each phase were quantified as follows (see Fig.

3.1): phase I) peak MAP during early straining baseline MAP

(point b point a); early phase II) peak phase I MAP -

lowest mid-strain MAP (point c point b); late phase II)

peak late strain MAP baseline MAP (point d point a);

phase III) peak phase II MAP MAP immediately

followingtermination of strain (point d point e); and phase

IV) peak post-strain MAP phase III MAP (point f point e).

Changes in heart rate (AHR) were also determined during all

four phases. Change in HR to change in MAP (AHR/AMAP) during

phases II and IV was used as an index of cardiac baroreflex

sensitivity because of its usefulness in describing

integrated baroreflex function (Korner et al. 1976, Porth et

al., 1984; Smith et al., 1987b). Integrated baroreflex-

mediated vasoconstriction was assessed by measuring the late

phase II rise in MAP (Korner et al., 1976; Sandroni et al.,


3.4.11 Statistical Analysis

For Study 1, standard descriptive statistics were

performed on each of the response variables of interest with

results presented as means SE. The average HR and BP to


110 -

E 100

<_ d "
n 90 -


90 -

80 -

2 E 70 -

60 -

S 50 -


3- 30 --------- |
a- 3)
0- E

x"' r-----

Begin End
Valsalva Valsalva

Figure 3.1. Representative mean arterial pressure and heart
rate reponses to a 15-s Valsalva maneuver at an expiratory
pressure of 30 mmHg. Also indicated are characteristic
sample points used for analysis (see text).

head-up tilt as well as baseline plasma volume, leg

compliance, Valsalva maneuver, and carotid-cardiac baroreflex

gain under the two treatment conditions were compared using

paired-difference t statistics. Vasoactive hormone and

forearm vascular resistance responses were compared with a

repeated measures two-way ANOVA.

For Study 2, standard descriptive statistics were also

calculated for baseline levels of HR, BP, slope of the

carotid-cardiac baroreflex, vascular resistance, leg volume,

plasma volume, and NE with results presented as means SE.

Since each subject received all levels of both factors

(subjects were crossed with treatments) and tests were

administered in a set sequence over time, the subjects acted

as blocks and multivariate profile analyses were used to

analyze these data. This approach to the analysis of

repeated measures yields exact tests of sources of variation

associated with treatment conditions including the treatment

condition by HDT interaction (i.e., the test for parallel

profiles). Statistical probabilities associated with the

main effect of HDT, LBNP, and the HDT by LBNP interaction

reflect the chances of observing a difference as large or

larger than the one observed if in fact the actual difference

was zero.


4.1 Abstract

We tested the hypothesis that a bout of graded arm

exercise designed to elicit maximal effort would increase the

sensitivity of autonomically-mediated baroreflexes and

enhance blood pressure stability during 70 HUT in subjects

prone to postural hypotension. Therefore, we measured HR,

BP, FVR, and vasoactive hormone responses before and during

15 min of 700 HUT in 10 paraplegic subjects (21-65 yr) on two

occasions: 1) 24 h after maximal arm-crank exercise (post-

exercise) and 2) without exercise (control). During HUT, HR

increased 30 bpm in both post-exercise and control, but the

reduction in SBP during control (-12.0 4.6 mmHg) was larger

(P= 0.017) than that during HUT following exercise (-0.3

4.3 mmHg). The post-exercise increase in FVR from supine to

HUT of 17.0 2.4 to 24.8 3.2 PRU was greater (P= 0.042)

than the increase observed during control (18.3 3.7 to 19.5

3.1 PRU). The gain of the carotid-cardiac baroreflex was

also increased (P = 0.049) following exercise. Responses in

norepinephrine, vasopressin, and plasma renin-angiotensin

1This work was supported in part by a NASA grant administered
under contract NAS10-11624.

induced by HUT were similar for control and post-exercise and

there was no difference in either leg compliance or plasma

volume between the two conditions. Additionally, HR and SBP

responses to phases II and IV of the Valsalva maneuver,

indices of integrated baroreflex sensitivity, were enhanced

(P < 0.05) following maximal exercise compared to control.

Thus, acute intense exercise eliminated orthostatic

hypotension in paraplegics and was associated with increased

FVR and baroreflex sensitivity and was independent of blood

volume changes.

4.2 Introduction

Cardiovascular reflexes mediated by the baroreceptors

regulate cardiac output and systemic peripheral resistance to

maintain mean arterial pressure during an orthostatic

challenge. Since blood pressure varies as the product of

heart rate, stroke volume, and vascular resistance, failure

of these fast-acting autonomic reflexes to counteract

orthostatically-induced reductions in cardiac output can lead

to hypotension and possible syncope. A disruption in the

integrity of these reflexes has been associated with

compromised blood pressure control following prolonged

bedrest (Convertino et al., 1990a) and chair confinement

(Convertino et al., 1991) as indicated by an attenuation of

orthostatic tachycardia and a reduction in the gain of the

carotid-cardiac baroreflex (Convertino et al., 1990a, 1991).

A single bout of dynamic exercise has been reported to

increase the sensitivity of cardiovascular reflexes and

enhance the maintenance of blood pressure during the post-

exercise recovery period. Acute graded exercise to

exhaustion increased carotid baroreceptor stimulus-cardiac

reflex response by 60 min post-exercise (Somers et al., 1986)

and lasted as long as 24 h after exercise (Convertino and

Adams, 1991; Convertino et al., 1992). A single bout of

intense treadmill exercise at the end of 10 days of bedrest

restored heart rate, blood pressure, and orthostatic

stability to pre-bedrest levels within 2 h of ambulation

(Convertino, 1987b) and reversed fainting episodes after 6 h

of water immersion (Stegemann et al., 1975). These

observations suggest that exercise of this type can enhance

the function of cardiovascular reflexes to maintain blood

pressure during orthostasis and may provide effective

countermeasure treatment for orthostatic hypotension.

Therefore, we tested the hypothesis that a bout of maximal

arm exercise would increase the sensitivity of autonomically-

mediated baroreflexes and enhance blood pressure stability

during an orthostatic challenge. Paraplegic subjects were

chosen because postural hypotension has been associated with

prolonged wheelchair confinement (Convertino et al., 1991).

We recently reported stable blood pressure during head-up

tilt 24 h after maximal exercise was associated with an

increased sensitivity of the carotid-cardiac baroreflex

(Engelke et al., 1992). However, increased gain of this

vagally-mediated cardiac reflex response could not fully

explain the elimination of orthostatic hypotension observed

in the post-exercise condition. Consequently, we now report

integrated arterial and cardiopulmonary baroreflex responses

as well as heart rate, blood pressure, forearm vascular

resistance, leg compliance, and vasoactive hormones during

700 HUT in 10 paraplegic subjects 24 h after a bout of

dynamic arm exercise.

4.3 Methods

Eight male and two female sedentary paraplegics subjects

[mean (SE) age of 36 4 yr, height of 185 2 cm, and

weight of 90 7 kg] who had experienced traumatic spinal

cord transaction in the thoracic region (range of injury: TI-

T12) gave their written consent to serve as subjects for this

investigation after they had been informed of all procedures

and risks. Subjects had been confined to their wheelchairs

for an average of 118 21 mo and none had attained the

upright posture for at least 6 mo prior to the beginning of

the study. Subjects completed two experimental protocols

which involved measurement of HR, systolic and diastolic BP,

FVR, and vasoactive hormone responses before and during 15

min of 70 HUT. Additionally, measurements of plasma volume,

carotid-cardiac baroreflex relationship, leg compliance, and

responses to the Valsalva maneuver were made.

Upon arrival at the laboratory, the subject assumed the

supine posture on a tilt table. After 30 min quiet rest, a

blood sample was obtained to determine plasma volume and pre-

tilt plasma levels of NE, AVP, and PRA. Following the blood

sample, a test was conducted for measurement of the carotid

baroreceptor stimulus-cardiac response relationship. Leg

compliance, and HR and BP responses to the Valsalva maneuver

were also determined at this time. Following these

measurements, each subject underwent an orthostatic test on

the tilt table. Subjects remained in the head-up position

for 15 min or until the onset of pre-syncopal symptoms. At

the conclusion of the tilt, the subject was quickly returned

to the supine posture and a blood sample was drawn within 30

s to be analyzed for NE, AVP, and PRA.

4.4 Results

Exercise bouts. The final work rate at volitional

fatigue averaged 38 1 watts and was attained after a mean

time of 14 1 min. HR, systolic and diastolic BP at

termination averaged 175 2 bpm, 165 2 mmHg, and 71 1

mmHg, respectively.

Response to 70 HUT. Supine hemodynamics and vasoactive

hormones were not different between treatments (Table 4.1).

During HUT, HR was elevated (P = 0.001) by 29 bpm and 30 bpm

under control and exercise conditions, respectively.

However, during HUT in the control condition, SBP decreased

(P = 0.025) from 118 5 to 106 9 mmHg, while there was

essentially no change in SBP during HUT 24 h after maximal

exercise (116 5 to 113 5 mmHg). Average reduction in SBP

during control HUT (-12.0 4.6 mmHg) was larger (P = 0.017)

than that during HUT following exercise (-3.1 3.9 mmHg).

DBP during HUT was not altered in either condition.

Table 4.1. Baseline hemodynamic, vasoactive hormone, plasma
volume, and vascular resistance values during control and 24
hours after exhaustive exercise.


HR, bpm 61

SBP, mmHg 118

DBP, mmHg 77

NE, pg-ml-1 322

AVP, pg-ml-1 1.9

PRA, pg-ml-1 0.77

PV, ml 3020

FVR, pru 18
Values are mean SE.


[ 4

t 5



1 0.1





60 6

116 4

76 4

304 57

1.9 0.1

0.77 0.2

3191 210

17 2

Leg Compliance. Left calf compliance in the post-

exercise condition was not different (P = 0.438) than that

recorded during control period (1.8 0.2 vs. 2.1 0.4


P value









30- [ Supine *
30 700 HUT


Control Exercise

Figure 4.1. Comparison of forearm vascular resistance (FVR)
in the supine posture (open bars) and after 15 min of 70
head-up tilt (HUT; closed bars) in the control and exercise
conditions. Asterisks indicate P < 0.05 vs. corresponding
values. Bars with 'T' lines represent means SE.

Baroreflex Responses. As illustrated in Figure 4.1, maximal

exercise elicited greater (P = 0.042) FVR during HUT compared

to the control condition. Pre-tilt baseline FVR was similar

between control and exercise conditions, and was highly

correlated (r= 0.84; P= 0.03). Mean HR and BP responses to

the Valsalva strain in the control and post-exercise

condition are illustrated in Figures 4.2A and 4.2B,

respectively. Comparisons between HR and SBP changes during

the four Valsalva maneuver phases are presented in Table 4.2.

During late phase II, ASBP was four-fold greater (P = 0.002)

24 h post-exercise than ASBP measured at a similar time in

the non-exercise condition. Also, the AHR/ASBP during phase

IV was 40% greater (P = 0.023) in the post-exercise condition



110 m' a n

Begin Valsalva End Valsalva



a V
v 00



Begin Valsalva

________________ A

End Valsalva

Figure 4.2. Mean heart rate (broken line) and systolic blood
pressure (solid line) responses to the Valsalava maneuver in
the control (panel A) and 24 h post-exercise (panel B)
conditions. Panel A also indicates the four phases of the
maneuver while panel B illustrates the characteristic sample
points used for analysis (see text).

^- SBP

-- r 110



70 S


tp 170-

1 160-

S 150-

* 140

0 120

W 110-


100 0k


-80 4.)


- 60


Table 4.2. Mean hemodynamic responses

to the Valsalva

Baseline HR, bpm

Baseline SBP, mmHg

Phase I

AHR, bpm

ASBP, mmHg

AHR/ASBP, b-mmHg-1

Phase II

AHR, bpm

early ASBP, mmHg

AHR/ASBP, b-mmHg-1

late ASBP, mmHg

Phase IV

AHR, bpm

ASBP, mmHg

AHR/ASBP, b-mmHq-1
Values are mean SE.


66 2

137 4












63 3

134 4










1 .0

(1.01 .14) than that during the control period (.72 .11).

There was no distinction between groups in AHR/ASBP during

phase I. Phase III was brief and characterized by reductions

in SBP and HR brought about by a decrease in mechanical

compression of the thoracic cavity occurring as a result of

the termination of the strain.





P value










.7 1.0 .

Mean carotid baroreflex stimulus-response relationships

for all subjects during control and after exercise are

illustrated in Figure 4.3. Mean maximum slope of the

carotid-cardiac baroreflex reponses was increased (P = 0.049)

24 h following intense exercise (6.2 1.7 msec-mmHg-1)

compared to the control condition (3.3 0.6 msec-mmHg-1).

0 Control
U 1100' Exercise

| 1,


900 -i i
95 115 135

Estimated Carotid Pressure, mmHg

Figure 4.3. Carotid-cardiac baroreflex stimulus-response
relationships, plotted over the range of pressures from which
maximum slopes were derived, during control (open circles)
and 24 h after exercise (closed circles). Bars with 'T'
lines represent means SE.

Circulating Volume and Hormone Responses. NE, AVP, and

PRA increased from supine to HUT (P < 0.05), but were similar

for control and post-exercise (Fig. 4.4). Plasma volume

measured 24 h after acute exercise (3020 132 ml) was

indistinguishable (P = 0.459) from the volume measured on the

control day (3191 210 ml).

4.5 Discussion

To test the hypothesis that acute exercise designed to

elicit maximal effort can sensitize autonomically-mediated

cardiovascular reflex responses and ameliorate hypotension

during an orthostatic challenge, we measured cardiac

baroreflex responses and changes in HR, BP, and FVR to 15 min

of 700 HUT in paraplegic subjects 24 h after intense arm-

crank exercise. We also measured the plasma concentration of

vasoactive hormones immediately following HUT. The primary

finding of this study was that 24 h after exercise, the

elimination of orthostatic hypotension during 70 HUT was

associated with increased baroreflex control of heart rate

and FVR, and independent of changes in blood volume and leg

compliance compared to the control condition. Further, the

observed improvement in systemic resistance occurred without

alterations in circulating NE, AVP, and PRA.

A 10% expansion in plasma volume was induced 24 h after

the performance of maximal leg exercise in ambulatory subjects

(Gillen et al., 1992). If hypervolemia occurred in our

subjects, it is possible that the improved blood pressure

maintenance 24 h after exercise was associated with enhanced

cardiac filling (Frank-Starling effect). Against expectations

Pooled SEI (33.2)









Pooled SEI (0.22)

Figure 4.4. Circulating hormone levels during supine (0)
baseline and at the end of 70 HUT during control (open
circles, solid lines) and post-exercise conditions (closed
circles, broken lines). Asterisks indicate P < 0.05 compared
with corresponding supine values.

'""" Exercise
"0-- Control

Pooled SEj (0.58)

00 HUT

700 HUT

00 HUT

700 HUT"






0 7

00 HUT

70 HOT

plasma volume measured 24 h post-exercise in our subjects was

not different than in the control condition. The lack of a

hypervolemic effect is unclear, but may be associated with

the lesser muscle mass and time to exhaustion involved in

arm-cranking compared to leg exercise. In any event,

improved blood pressure stability during HUT 24 h post-

exercise could not be explained by blood volume changes.

Cardiac and hemodynamic responses during orthostatism

may be influenced by leg venous compliance. An inverse

relationship between leg compliance and tolerance to lower

body negative pressure has been reported (Luft et al., 1976),

suggesting that the capacity to maintain venous return,

cardiac output and systemic arterial pressure during an

orthostatic challenge is dependent on the amount of blood

pooled in the lower extremities. As expected with the

abolishment of sympathetic outflow to leg veins and

arterioles as well as muscle tone in these patients, leg

compliance in the present study was similar between control

and exercise conditions. This finding eliminated the

possibility that maintained blood pressure during HUT

following exercise could be explained by differences in leg

compliance. Unexpectedly, mean leg compliance measured in

our paraplegic subjects (1.8 ml-mmHg-1) is approximately 30%

to 60% less than values (2.6 4.4 ml-mmHg-1) reported for

able-bodied subjects (Convertino et al., 1988, 1989b). These

comparisons raise the possibility that marked reductions in

venous compliance in wheelchair-confined subjects may provide

some protection against orthostatic challenge.

Since circulating vasoactive hormones, blood volume, and

leg compliance were unaltered by exercise, increased

sensitivity of reflexes mediated by baroreceptors probably

represented the most contributing mechanism to the

differences in maintenance of arterial pressure during HUT 24

h after exercise. Among these reflexes, it has been

demonstrated that the sensitivity of the carotid-cardiac

baroreflex can be increased for as long as 24 h following

completion of a single bout of maximal exercise (Convertino

and Adams, 1991). Our present data not only confirm this

finding but also provide evidence to suggest that baroreflex

responses associated with control of vascular resistance are

increased by exercise as well. SBP was maintained during 15

min of 70 HUT in the post-exercise condition at a HR which

was associated with hypotension in the control tilt. The

absence of a difference in orthostatic tachycardia between

the two experimental conditions implies that factors other

than baroreflex control of cardiac responses were altered by

exercise. Our observation of a greater FVR during post-

exercise HUT compared to control suggests that the

responsiveness of vasoconstrictor mechanisms were enhanced by

acute intense exercise and contributed significantly to post-

exercise maintenance of blood pressure during tilt. However,

the specific mechanisms) responsible for the increased

vasoconstriction remains to be identified.

With a cross-over experimental design, we observed no

distinguishable difference in the average baseline FVR

between the control and exercise conditions, and test-retest

comparisons for individual subjects were significantly

correlated. These findings support the notion that occlusion

plethysmographic measurement of FVR was highly reproducible

from day-to-day, and indicates that increased baroreflex

response of FVR during HUT following exercise in this study

was the result of the exercise treatment rather than some

factor associated with methodology or time.

The mechanism by which FVR response during HUT was

increased by acute exercise is unclear. Alterations in

circulating volume can change cardiopulmonary baroreceptor

control of vascular responses (Thompson et al., 1990). Since

vascular volume was not altered by the acute exercise in this

study, it is unlikely that increased FVR during HUT after

exercise could be explained by this mechanism. An

alternative explanation for greater post-exercise FVR during

tilt might be that exercise elicited a greater response of

neural and humoral vasoactive agents such as NE, AVP, and PRA

to orthostasis. It is widely documented that orthostasis

causes increased plasma levels of NE (Blomqvist and Stone;

Convertino, 1987a), AVP (Convertino, 1987a), and PRA

(Blomqvist and Stone; Convertino, 1987a; Shvartz et al.,

1981) and that these hormone levels are associated with

orthostatic tolerance (Shvartz et al., 1981). Elevations in

plasma NE, AVP, and PRA in our subjects at the end of tilt

were similar to those reported in paraplegics (Guttmann et

al., 1963; Mendelsohn and Johnston, 1971; Poole et al., 1987)

and there were no differences in these responses between the

exercise and control conditions. Therefore, increased FVR

during HUT following acute intense exercise could not be

explained by increased levels of these circulating vasoactive

hormones. However, the observation that FVR was

significantly increased during post-exercise HUT at similar

levels of circulating NE, AVP, and PRA may reflect that the

exercise bout induced acute vascular receptor

hypersensitivity to these neuroendocrine agents. This

possibility remains speculative without further

investigations designed to pharmacologically assess

alterations in vascular receptor response.

Tachycardia mediated by the baroreflexes provides a

means to buffer transient changes in arterial blood pressure

(Blomqvist and Stone, 1983). Data from both human and animal

models have demonstrated that impairment of the carotid-

cardiac reflex is associated with less tachycardia and

occurrence of orthostatic hypotension during upright posture

(Convertino et al., 1990a, 1991; Cowley et al., 1973). A

positive correlation has been reported between the degree of

impairment of baroreflex function and incidence of syncope

during standing in healthy men following prolonged

confinement to bed (Convertino et al., 1990a). Similarly,

sinoaortic denervated dogs illustrated smaller heart rate

increases and greater blood pressure reductions during

upright posture than dogs with intact baroreflexes (Cowley et

al., 1973). These results suggest that if reductions in

baroreflex function lead to attenuated cardiac responses

during an orthostatic challenge, then enhancement of

baroreflex sensitivity should result in greater tachycardia

and blood pressure stability. Using a neck cuff device, the

influence of a single bout of exhaustive exercise on the

cardioacceleratory limb of the carotid-cardiac baroreflex was

assessed from changes in heart rate from baseline values

induced by an elevation of ~40 mmHg in cuff pressure. The

consequent 40-mmHg reduction in carotid distending pressure

elicited a calculated elevation in heart rate of 5 bpm (from

62 to 67) during control and 7 bpm (from 61 to 68) 24 h post-

exercise (Fig. 4). Thus, although increased carotid-cardiac

baroreflex sensitivity was relatively large and consistent,

it appears unlikely that vagally-mediated alterations in

heart rate can completely explain the tachycardia noted

during HUT. This raises the possibility that other factors

responsible for cardioacceleration, possibly sympathetically-

mediated baroreflexes, were influenced by exercise.

We used the unit change in HR to the unit change in SBP

(AHR/ASBP) during phases I, II, and IV of the Valsalva

maneuver as an index of integrated arterial baroreflex

stimulus-response relationship. The AHR/ASBP represents an

index of non-specific baroreflex control of heart rate since

pressure reductions are likely to influence heart rate

through interaction of cardiopulmonary, aortic, and carotid