Journal of the Autonomic Nervous System, 22 (1988) 107-114 107 Elsevier

JAN 00797

Effect of changes in cardiac autonomic balance on blood pressure regulation in man

M i c h a e l L. S m i t h , H o w a r d M . G r a i t z e r , D o n n a L. H u d s o n a n d P e t e r B. R a v e n

Department of Physiology, Texas College of Osteopathic Medicine, Fort Worth, TX 76107 (U.S.A.)

(Received 28 June 1987) (Revised version received 16 November 1987)

(Accepted 17 December 1987)

Key words: Sympathe t ic ; Pa rasympa the t i c ; Exercise t ra ining; Lower b o d y negat ive pressure

Abstract

The role of cardiac autonomic balance in fitness-related differences in blood pressure regulation was evaluated by comparing the cardiovascular responses to lower body negative pressure (LBNP) in 10 trained and 10 untrained men. Cardiac autonomic balance was quantified as the ratio of resting heart rate to intrinsic heart rate, and was significantly lower in the trained subjects (0.68 + 0.03) than the untrained subjects (0.81 + 0.03) indicating a greater parasympathetic dominance at rest in the trained subjects. Arterial pressure decreased significantly more during LBNP in the trained subjects and was due to lower chronotropic and vasoconstrictor respones in these trained subjects. 'Cardiac autonomic balance' was equilibrated between the groups by partial parasympathetic blockade with atropine sulfate in the trained subjects and partial sympathetic blockade with metoprolol tartrate in the untrained subjects. Equilibration of cardiac autonomic balance eliminated the group differences in blood pressure maintenance, and chronotropic and vasoconstrictor responsiveness during LBNP. It was hypothesized that the elevated tone of parasympathetic control of the heart rate of the trained subjects resulted in an attenuation of blood pressure regulation.

I n W o d u ~ o n

The adap t a t i ons to endurance exercise t ra in ing inc lude a s ignif icant reduc t ion in rest ing hear t ra te which is thought to be due in pa r t to an aug- men ted pa r a sympa the t i c act ivi ty and a decreased sympa the t i c ac t iv i ty [4,14,20,23]. This t rans la tes in to an a l te red au tonomic ba lance or ' t u n i n g ' tha t reflects an exace rba ted p a r a s y m p a t h e t i c domi - nance at rest. Also associa ted wi th endurance ex- ercise t ra in ing is an a t t enua ted b l o o d pressure

Correspondence: Michael L. Smith. Present address: Cardio- vascular Physiology, Research Division, McGuire V.A. Medi- cal Center, 1201 Broad Rock Blvd., Richmond, VA 23249, U.S.A.

r egu la to ry sys tem [11,17,22]. However , it is not k n o w n whether changes in au tonomic ba lance af- fect the homeos t a t i c con t ro l of b l o o d pressure , and hence, whe ther the dif ferences in b l o o d pres- sure regula t ion obse rved be tween exerc ise- t ra ined and un t r a ined subjects are a t t r ibu tab le to a dif- ference in a u t o n o m i c ba lance . Therefore , the pur - pose of this inves t iga t ion was to de te rmine whether the dif ferences in b l o o d pressure regula t ion ob- served be tween exerc ise- t ra ined and un t ra ined men are a t t r i bu tab le to di f ferences in au tonomic bal- ance cont ro l l ing the hear t . In this s tudy, ' c a rd i ac au tonomic ba l ance ' was quant i f i ed as the ra t io of rest ing hea r t ra te ( H R r ) to in t r ins ic hear t ra te (HR0) , which factors H R 0 out of the H R r value and should reflect the ba l ance of the pa ra sym-

0165-1838/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)

108

pathetic and sympathetic neural influences on heart rate. This cardiac autonomic balance was equilibrated between a group of exercise-trained and a group of untrained men by reducing the parasympathetic dominance of the trained men and reducing sympathetic influences in the un- trained men with partial pharmacologic blockade. The cardiovascular responses to a hypotensive stress (lower body negative pressure, LBNP) were evaluated in each subject during the unblocked and partially blocked states.

Materials and Methods

Twenty healthy men, age 20-31 years, par- ticipated in this study. Each subject underwent standard medical screening procedures and pro- vided informed written consent for experiments that had been approved by the institutional review board. The subjects were divided into two groups of 10 according to their fitness levels as de- termined by their maximal aerobic capacity (VO2max). The endurance exercise-trained sub- jects had been in training regularly for more than two years and had a 902max of > 60 m l / m i n . kg -~. The untrained subjects were sedentary indi- viduals with a VO2max of < 45 ml /min • kg -1.

The VO2max was determined by use of a graded exercise test on a treadmill. Ventilation volumes and expired 02 and CO 2 content were used to calculate on-line breath-by-breath measurements of oxygen consumption (902). Attainment of .pre- dicted maximal heart rate and a plateauing of VO 2 during the final stages were used as criteria for determining maximal effort. A plateau or a final minute increase in 902 of < 50 ml /min was observed in 17 of the 20 subjects. The other 3 subjects had final stage increases ranging from 70 to 140 ml/min.

Total blood volume (BV) and plasma volume (PV) were determined prior to the experimental procedures using the carbon monoxide (CO) dilu- tion technique described by Myhre et al. [16]. Total hemoglobin, carboxyhemoglobin and hematocrit were determined in venipucture sam- ples made prior to and after 10 min of breathing on a closed system containing 50 ml CO mixed in

100% 02 . These measurements were used to calculate BV, PV, and PV/kg b.wt. The coefficient of variation for control values was 3.9% as based on 8 serial measurements made on a 70 kg man with a mean BV of 5.72 1.

Assuming HR r reflects the product of HR 0 and the balance of autonomic influences, the ratio of H R r / H R 0 will predict the degree of parasym- pathetic dominance of autonomic balance on the resting heart rate, where smaller values reflect a greater parasympathetic dominance. Intrinsic heart rate was obtained during complete cardiac auto- nomic blockade using 0.04 mg/kg atropine sulfate and 0.2 m g /k g metoprolol tartrate. The blockades were confirmed in two steps. First, the subject performed a Valsalva maneuver by producing an expiration pressure of 60 Torr while blowing into a closed tube. Absence of the chronotropic changes during the Valsalva maneuver provided evidence for complete physiologic blockade of both muscarinic and fll-adrenergic receptor popula- tions. In addition, the flradrenergic receptor agonist isoproterenol was given in stepwise doses from 1 to 50 /tg with no change in heart rate considered further verification of the full metoprolol blockade of 131 receptors of the heart.

On a separate day, cardiac autonomic balance was equalized in the two subject groups by using partial sympathetic blockade in the untrained sub- jects and partial parasympathetic blockade in the trained subjects. This equilibrated the HRr rela- tive to HR 0 in all subjects. The cardiovascular responses of each untrained subject to progressive LBNP were studied during control condition and during partial /3~-adrenergic receptor blockade with metoprolol tartrate. The metoprolol was indi- vidually titrated using 0.05 mg/kg progressive in- creases in dose, into a venous cannula placed in an arm vein to a dose producing a 10% decrease in heart rate. The cardiovascular responses of each trained subject to LBNP were studied during con- trol condition and during partial parasympathetic blockade using atropine sulfate. The atropine was titrated, using 0.005 mg/kg progressive increases in dose, to a dose producing a 10% increase in heart rate.

LBNP was performed by having the subject lie in the supine position with his lower extremeties

inside of a wooden chamber and an airtight seal engaged at the iliac crests. Specific description of the LBNP protocol has been previously reported [17]. The negative pressure was increased progres- sively through - 8, - 16, - 32, and - 40 Torr stages. All physiologic measurements were made at rest, - 1 6 and - 4 0 Torr. All LBNP procedures were performed at a time during which the sub- jects had no physical activity, stimulatory drinks or medications for at least 12 h prior to the test.

Cardiac output (0) was determined using the CO 2 rebreathing technique described by Klausen [10] and based on the Defares method of predic- ting mixed venous CO 2 [3]. The coefficient of variation for resting measurements was 1.9% around a mean value of 5.45 1/min as determined from 8 serial measurements on a 70 kg male. A semiautomated Jones rebreathing valve-bag sys- tem was utilized during the 0 determinations. The concentration of the measured gases (rebreathe, end-tidal, and mixed-venous) was determined by use of an infrared gas analyzer (Beckman, model LB-2) and recorded on a multi-channel pen recorder. VO 2 was measured over a 30-s period immediately before 0 measurements using a polarographic oxygen analysis (Beckman, model OM-11) of expired air in phase with the CO2 analysis and volume determination using a turbine volume meter (Pneumoscan, model S-301). The electrocardiogram was monitored on an ECG oscilloscope and recorded on a strip-chart re- corder. Ten-second rhythm strips were obtained every minute for determination of heart rate. Indi- rect blood pressure measurements were recorded every minute by means of a semi-automated elec- trosphygmomanometer (Narco, model 700). Prior to each test, an auscultatory measurement was made simultaneous to an automated measurement to verify the accuracy of the chart recording. Mean arterial blood pressure (MBP) was calcu- lated as diastolic blood pressure (DBP) plus one- third pulse pressure. Peripheral vascular resistance (PVR) was calculated from MBP and 0.

Forearm blood flow (BF) was determined using the plethysmographic technique described by Whitney [24]. A Mercury-in-silastic strain gauge was placed in a mid-forearm position and con- nected to a plethysmographic pre-amplifier, the

109

output of which was recorded on the chart re- corder. Forearm vascular resistance (FVR) was calculated from MBP and BF. Leg volume changes, used as an index of the level of LBNP stress, were determined from changes in leg circumference measured with a plethysmographic strain gauge and the calculated resting leg volume. Previously, we found that moderate levels of muscle tension c~m profoundly affect the bood pressure response to LBNP [19]; therefore, the electromyographic activity of the abdominal muscles and legs was monitored to insure that the subject remained relaxed.

A 3-factor (2 x 2 × 3) analysis of variance with repeated measures design was used to statistically analyze the data. A Student -Newman-Keuls 'post-hoc' analysis was used to analyze main ef- fect differences between groups and conditions. A probability of P < 0.05 was considered significant.

Results

The baseline values for each variable during the control and partial blockade conditions are sum- marized in Table I for the trained (n = 10) and untrained (n = 10) subjects. During the control condition, H R was significantly lower in the trained subjects while stroke volume (SV) was significantly greater in the trained subjects. No other group differences were observed. After the partial blockade, only SBP was different between the groups (trained > untrained). Heart rate was significantly affected by the partial blockade with the HR lower in the untrained subjects and higher in the trained subjects, resulting in the disap- pearance of the group difference observed during control condition. Consequently, the SV group difference was eliminated by the equilibration of cardiac autonomic balance. The H R r / H R 0 ratio was significantly less in the trained subjects during the control condition (0.68 + 0.03 vs 0.81 + 0.04 for the untrained subjects), indicating a greater parasympathetic dominance in these subjects; whereas, the selective partial blockade resulted in equal cardiac autonomic balance in the two sub- ject groups (Table II).

110

TABLE I

Baseline values during the control or unblocked, and partial autonomic blockade conditions in 10 trained (T) and 10 untrained (UT) men

Values represent mean _+ S.E.M. HR, heart rate; SBP, systolic blood pressure; MBP, mean blood pressure; Q, cardiac output; SV, stroke volume; BF, forearm blood flow: FVR, forearm vascular resistance; PVR, peripheral vascular resistance.

Subjects Control Blockade

HR T 54 _+2 61 _+3 t (beats /min) UT 70 5:4 * 64 5:3 t SBP T 126 -+2 125 -+2 (Torr) U T 121 d:2 119 + 2 * MBP T 94 + 2 92 + 2 (Torr) U T 93 +1 90 +1

(~ T 5.85-0.4 5.7-+0.2 ( l /min) UT 6.0 5- 0.3 5.7 5= 0.2 SV T 97 5-8 91 5-5 (ml) U T 84 5 - 4 " 86 +5 BF T 5.2 5- 0.9 5.6 + 1.0 (ml /min -100 m1-1) U T 4.45-0.7 4.45-0.8 FVR T 19 5-3 20 5-3 (PRU) UT 21 _+3 23 5-3 PVR T 17 5-1 16 5-1 (PRU) UT 16 5-1 16 5-1

* Significant difference between groups (P < 0.05); t signifi- cant difference when compared to control condition (P < 0.05).

The cardiovascular responses to LBNP during the two conditions are presented in Figs. 1 and 2. The data represent the changes from baseline to - 1 6 Torr and from baseline to - 4 0 Torr.

L B N P responses under control conditions Mean and systolic blood pressure decreased

significantly more in the trained subjects (Fig. 1). However, no differences in the cardiac output

TABLE II

Cardiac autonomic balance ( H R r / H R o ) during control and partial blockade conditions in 10 trained (T) and 10 untrained (UT) men

Values represent mean + S.E.M.

Subjects Control Blockade

T 0.68 + 0.03 0.76 + 0.03 UT 0.81 + 0.04 * 0.74 + 0.03

• Indicates significant difference (P < 0.05).

u l

8e

z

C PB

°iu. i -10

-20

A 0 c

-1.0 ,,~ ,,i I - ,,-i 0 ,o -2.0

I r

0 -3.0

c P8

n,,

w ix D.

8 o ¢, o m

o I- ll-

C PB

~ -10

<1 . ~ - 2 0

2 -30

+30 t 2

E .

" AN I +20

g 111 ' "

~ + 1 0 I II

I--

• 1- 0 • T tJ T

C P B

Fig. 1. Mean (-+S.E.M.) blood pressure, cardiac output and heart rate responses from 0 to - 1 6 Torr LBNP (hatched portion of bars) and 0 to - 4 0 Torr LBNP (full bars) in 10 trained (T) and 10 untrained (U) men during control condition (C) and partial blockade (PB) of the parasympathetic nervous sytem in T subjects and the sympathetic nervous system in U subjects. * indicates significant difference between groups; t indicates significant difference between PB and C conditions.

All significant differences represent P < 0.05.

(Fig. 1) or stroke volume (Fig. 2) responses were observed ( P > 0.05). The heart rate responses were similar (P > 0.05) between groups (Fig. 1), but the heart rate response per change in blood pressure was significantly lower in the trained subjects (Table III). The regional vasoconstrictor responses were not different ( P > 0.05) between groups as evidenced by the similar forearm blood flow and resistance responses. The diastolic blood pressure decreased significantly in the trained subjects (5 + 1 Torr), but increased in the untrained subjects (1 + 1 Torr). Furthermore, peripheral vascular re- sistance appeared to increase more in the un- trained subjects (0.10 < P < 0.15) suggesting that the untrained subjects responded with a greater vasoconstriction.

111

0

E

" ' -20 =E ..-i ,,.i O

u.I -40 w, O E I.- u)

-60

C PB T U T

~'r"l I I

G PB T U T U

o ~ -2 , , , J r ,

0 u.

+30

~ +20 ,~ t t l

: i Z

~ +10 =m n. w O m t , L E

. +10

O. +8

T U T U T U T U C PB C PB

Fig. 2. Mean (+ S.E.M.) stroke volume, blood flow and resis- tance responses from 0 to - 16 Torr LBNP (hatched portion of bars) and 0 to - 4 0 Torr LBNP (full bars) in 10 trained (T) and 10 untrained (U) men during control condition (C) and partial blockade (PB) of the parasympathetic nervous system in T subjects and the sympathetic nervous system in U sub- jects. * indicates significant difference between groups; 1" indi- cates significant difference between PB and C conditions. All

significant differences represent P < 0.05.

LBNP responses during selective partial blockade When cardiac autonomic balance was equi-

librated by partial blockade, the decreases in both mean and systolic blood pressure were not signifi-

T A B L E I I I

Chronotropic responsioeness (A HR /ASBP) to LBNP during control and partial blockade conditions in 10 trained (T) and 10 untrained (UT) men

Values represent m e a n + S.E.M.

Control Blockade

AHR/ASBP T 0 . 9 1 + 0 . 3 1 . 2 6 + 0 . 4

U T 1 .62+0 .3 * 1 .23+0 .3

• Ind ica tes s ignif icant difference be tween groups ( P < 0.05).

cantly different between groups (Fig. 1). Cardiac output and stroke volume responses were not af- fected, however, the heart rate response was sig- nificantly augmented in the trained subjects, rela- tive to control condition, resulting in a signifi- cantly greater heart rate response than in the untrained subjects (Fig. 1). The resulting zaHR/ ASBP and AHR/AMBP ratios were not signifi- cantly different (Table III). The vasoconstrictor responses were similar to the control condition with the eXception of an augmented PVR increase in the trained subjects (P < 0.05). This resulted in identical resistance increases in the two subject groups during the equilibrated autonomic balance (i.e. partial block condition).

D i s c u s s i o n

Assessing "cardiac autonomic balance" A model of cardiac autonomic balance was

used in which the relationship of HR r to HR 0 was assumed to reflect the degree of parasympathetic predominance of autonomic influences on the heart. This assumption is valid in view of the reciprocal innervation and function of the two divisions of the autonomic nervous system• Gell- horn [6] has shown that the reciprocal function (or actions) of the autonomic nervous system were maintained when autonomic balance was artifi- cially changed. Kollai and Koizumi [12] have shown that autonomic nervous system activity re- suiting from perturbations of the arterial barore- ceptors has a clearly reciprocal character; whereas, chemoreceptor stimulation or ventilation changes produce non-reciprocal autonomic responses. Thus, reflex responses to stress may not always produce reciprocal actions, but the actions of the two divisions remain reciprocal in effect. The equation for quantifying autonomic balance used in the present study was compatible with the model for resting parasympathetic and sym- pathetic activity used by Katona et al. [9]. Calcu- lation of autonomic balance from the coefficients for autonomic activity obtained by Katona et al. [9] were essentially the same as a calculation of H R r / H R 0. More recently we used Katona's model to assess the mechanisms of the resting brady-

112

cardia found in elite long-distance runners, and obtained a similar compatibility between auto- nomic balance calculated from either the coeffi- cients for parasympathetic and sympathetic activ- ity or HR~ and HR 0 [20]. Therefore, assuming the normal reciprocal function of the autonomic nervous system, the cardiac autonomic balance of different individuals can be quantified and com- pared.

The equation for cardiac autonomic balance ( H R r / H R 0 ) can be described as a functional reflection of parasympathetic predominance, be- cause a smaller value can only be explained by a greater parasympathetic activity, lesser sym- pathetic activity, or a combination of these two effects. Hence, the validity of this index of cardiac autonomic balance is not influenced by the phe- nomenon of accentuated antagonism described by Levy [13]. However, the method used in our study to equilibrate cardiac autonomic balance between two groups of subjects could be affected by accentuated antagonism if the nerve traffic is altered by the autonomic blocking drugs. It is a reasonable assumption that the low doses of either atropine or metoprolol would not profoundly af- fect nerve traffic 'per se', since the drugs are principally receptor antagonists. Furthermore, the membrane stabilizing effects of metoprolol should not have influenced HR 0, as these effect are only observed at high doses or during chronic use [1]. This is not the case with atropine. Katona et al. [8] demonstrated in dogs that at lower doses, atropine has a central stimulatory effect on vagal efferent activity, while progressively higher doses produce increasing levels of vagal blockade (i.e. increases in heart rate). Hence, accentuated antagonism could influence the responses to a given stress during partial blockade with atropine.

Cardiac autonomic balance and blood pressure regu- lation

Maintenance of blood pressure during a simu- lated orthostatic stress (LBNP) was less in the trained subjects. Although there is some debate in the literature regarding the effect of exercise train- ing on blood pressure regulation [2], these observations of attenuated blood pressure regu- lation have been well supported [11,17,21,22]. In

the present study, we found that this attenuated blood pressure regulation was attributable to a lower chronotropic responsiveness to the stress (see Table III) and to a less vasoconstrictor effect (as defined by an attenuated decrease in forearm blood flow or increase in vascular resistance). These observations are consistent with previous reports of lower baroreflex chronotropic [17,21] and vasoconstrictor [15,17] responses to LBNP in trained subjects. In the present study, the trained men had a significantly lower resting heart rate (Table I) attributable to both a greater parasym- pathetic predominance of autonomic influence (Table II) and a lower intrinsic heart rate (79 _+ 3 bpm vs 87 _+ 2 bpm for the untrained men). The equilibration of cardiac autonomic balance by partial pharmacological blockade of autonomic nerves resulted in elimination of the group dif- ference in blood pressure responses (Fig. 1), and this was due primarily to a smaller decrease of pressure during LBNP in the trained subjects. Furthermore, during LBNP and partial autonomic blockade, the chronotropic response per change in blood pressure was the same among trained and untrained groups (Table III) and was due to an increase in the chronotropic response of the trained subjects. This observation suggests that 'accentu- ated antagonism' [13] did not have a significant effect on the responses to LBNP. The partial blockade with atropine in the trained subjects presumably increases parasympathetic efferent ac- tivity as shown by Katona et al. [8]. However, in the present study, the receptor blockade pre- dominated at the dose of atropine used, thereby producing the increase in resting heart rate. An increase in parasympathetic efferent activity would reduce the response of the sympathetic nervous system in the presence of ' a ccen tua t ed antagonism'. Therefore, a reduced heart rate re- sponse would be predicted; whereas, we observed an increase in the heart rate response.

The equilibration of cardiac autonomic balance also resulted in a reversal of the difference in vasoconstrictor response between the two groups (Fig. 2). This was realized by a significant increase in peripheral vascular resistance of the trained subjects. However, this effect of vasoconstrictor response was not observed in the forearm vascular

beds (Fig. 2), suggesting that differential vaso- constrictor responses occurred in different vascu- lar beds. Such differential responses of different vascular beds have been documented during LBNP

[51. Our data indicate that altering cardiac auto-

nomic balance at the receptor level can affect blood pressure regulation during hypotension. The net response to LBNP is an integrated response of several regulatory reflexes (including arterial, cardiopulmonary and ventricular mechanorecep- tors). Hence, alteration of cardiac autonomic bal- ance affected this integrated response. The mecha- nism of this effect is uncertain, but may have involved an effect of atropine in the brainstem, as similar low doses of atropine do affect autonomic activity [81. Alternatively, the change in baseline heart rate may have affected the response of the cardiopulmonary and ventricular mechanorecep- tors to the central hypovolemia induced by LBNP; however, to our knowledge, such an effect of heart rate has not been reported.

In 1957 Ernst Gellhorn [6] proposed that auto- nomic balance (or imbalance) was an important determinant of cardiovascular function and regu- lation. Gellhorn expounded on the reciprocity of autonomic innervation and effect, and noted that the autonomic nervous system does not always function as a simple excitatory-inhibitory sum- mating operation. Gellhorn [61 described a phe- nomenon, termed ' tuning' , whereby changes in the background activity of one limb of the autonomic nervous system can affect the response (heart rate or blood pressure) mediated by the other limb. More recently, Levy [13] described a similar phe- nomenon which he termed 'accentuated antag- onism'. Therefore, although the autonomic nervous system does not function exclusively in a re- ciprocal manner, the agonist-antagonistic actions of the two limbs of the autonomic nervous system do produce effects that reflect a balance between the two limbs.

Previously, we found that complete cardiac sympathetic blockade with metoprolol did not af- fect the differences in blood pressure regulation observed between trained and untrained men [7]. However, we did observe some improvement in blood pressure regulation during complete para-

113

sympathetic blockade which was attributable to a profoundly augmented vasoconstrictor response [18]. The data f rom the present investigation, in association with these previous findings, would suggest that the differences in blood pressure regu- lation between trained and untrained men was due to a skewed cardiac autonomic balance with an exacerbated parasympathet ic dominance in the trained subjects. Partial or complete removal of this pa rasympa the t i c dominance tended to eliminate the differences in blood pressure regu- lation. Therefore, highly trained endurance athletes may represent an in vivo example of an autonomic ' imbalance ' that was defined by Gellhorn [6].

Acknowledgements

This investigation was supported in part by N.H.L.B.I. Grant no. HL-34397 and D O D Con- tract no. F33615-83-D-0602-0021. We extend our appreciation to the subjects for their time and cooperation, and to the Word Processing Center of TCOM for their production assistance.

References

1 Ablad, B., Carlsson, E., Johnsson, G. and Regardh, C.-G., Metoprolol. In A. Scriabine (Ed.), Pharmacology of Antihy- pertensive Drugs, Raven, New York, 1980, pp. 247-262.

2 Convertino, V.A., Aerobic fitness, endurance training and orthostatic intolerance, Exerc. Sport Sci. Rev., 15 (1987) 522-560.

3 Defares, J.G., Determinations of PvCO 2 from the exponen- tial CO2 rise during rebreathing, J. Appl. Physiol., 13 (1958) 159-164.

4 Ekblom, B., Kilbom, A. and Soltysiak, J., Physical training, bradycardia, and autonomic nervous system, Scand. J. Clin. Lab. Invest., 32 (1973) 251-256.

5 Essandoh, L.K., Houston, D.S., Van Houtte, P.M. and Shepherd, J.T., Differential effects of lower body negative pressure on forearm and calf blood flow, J. Appl. Physiol., 61 (1986) 994-998.

6 Gellhorn, E., Autonomic Imbalance and the Hypothalamus, University of Minnesota Press, Minneapolis, 1957.

7 Hudson, D.L., Smith, M.L., Graitzer, H.M. and Raven, P.B., Fitness-related difference in response to LBNP during sympathetic blockade with metoprolol, Fed. Proc. Fed Soc. Exp. Biol. Med., 45 (1986) 643.

8 Katona, P.G., Lipson, D. and Dauchot, P.J., Opposing central and peripheral effects of atropine on parasym-

114

pathetic cardiac control, Am. J. PhysioL, 232 (1977) H146-151.

9 Katona, P.G., McLean, M., Dighton, D.H. and Guz, A., Sympathetic and parasympathetic cardiac control in athle- tes and nonathletes at rest, J. Appl. Physiol., 52 (1982) 1652-1657.

10 Klausen, K., Comparison of CO2 rebreathing and acetylene methods for cardiac output, J. Appl. Physiol., 20 (1975) 763-766.

11 Klein, K.E., Wegmann, H.M. and Kuklinski, P., Athletic endurance training--advantage for space flight? The sig- nificance of physical fittness for selection and training of space lab crews, Aviat. Space Environ. Med., 48 (1977) 215-222.

12 Kollai, M. and Koizumi, K., Reciprocal and non-reciprocal action of the vagal and sympathetic nerves innervating the heart, J. Auton. Nerv. Syst., 1 (1978) 33-52.

13 Levy, M.N., Sympathetic-parasympathetic interactions in the heart, Circ. Res., 29 (1971) 437-445.

14 Lin, Y. and Horvarth, S.M., Autonomic nervous control of cardiac frequency in the exercise-trained rat, J. Appl. Phys- iol., 33 (1972) 796-799.

15 Mack, G., Shi, X., Tripathi, A. and Nadel, E., Diminished baroreflex control of forearm vascular resistance in physi- cally fit humans, J. Appl. Physiol., 63 (1987) 105-110.

16 Myhre, L.G., Brown, D.K., Hall, F.G. and Dill, D.B., The use of carbon monoxide and T-1824 for determining blood volume, Clin. Chem., 14 (1968) 1197-1205.

17 Raven, P.B., Young, D.R. and Blomqvist, C.G., Physical fitness and cardiovascular responses to lower body negative pressure, Z Appl. Physiol., 56 (1984) 138-144.

18 Smith, M.L., Graitzer, H.M., Hudson, D.L. and Raven, P.B., Fitness-related differences in response to LBNP dur- ing parasympathetic blockade with atropine, Fed. Proc. Fed Am. Soc. Exp. Biol. Med., 45 (1986) 643.

19 Smith, M.L., Hudson, D.L. and Raven, P.B., Effect of muscle tension on the cardiovascular responses to lower body negative pressure in man, Med. ScL Sports Exerc., ] 9 (1987) 436-443.

20 Smith, M.L., Hudson, D.L., Graitzer, H.M. and Raven, R.B., Exercise training bradycardia: The role of autonomic balance, Med. Sci. Sports Exerc., 20 (1988) in press.

21 Stegemann, J., Busert, A. and Brock, D., Influence of fitness on the blood pressure control system in man, Aerospace Med., 45 (1974) 45-48.

22 Tipton, C.M., Matthes, R.D. and Bedford, T.G., Influence of training on the blood pressure changes during lower body negative pressure in rats, Med. Sci. Sports Exerc., 14 (1982) 81-90.

23 Tipton, C.M., Matthes, R.D., Tcheng, T.-K., Dowell, R.T. and Vailas, A.C., The use of the Langendorff preparation to study the bradycardia of training, Meal Sci. Sports Exerc., 9 (1977) 220-230.

24 Whitney, R.J., The measurement of volume changes in human limbs, J. Physiol. (Lond.), 121 (1953) 1-27.