Effects of Aerobic Exercise Training on Hemodynamic Responses

During Psychosocial Stress in Normotensive and Borderline

Hypertensive Type A Men: A Preliminary Report

ANDREW SHERWOOD, PHD, KATHLEEN C. LIGHT, PHD, AND JAMES A.

BLUMENTHAL, PHD

This study assessed the effects of aerobic exercise training on cardiovascular responses to a 5-
min reaction time competition task. Twenty-seven Type A men (aged 30-56) participated in
this randomized study in which 14 underwent supervised aerobic training and 13 strength
training, with sessions scheduled three times per week for 12 consecutive weeks. Aerobic
exercise training was associated with a 13.6% increase in VO2max compared to 2.9% for the
strength group. The effects of aerobic exercise training were most evident in subjects whose
initial casual blood pressure readings fell in the borderline hypertensive range (N = 5). Those
individuals exhibited a general reduction in diastolic blood pressure (i.e., during rest, compe-
tition, and recovery) which was associated with a fall in both heart rate and total peripheral
vascular resistance. Furthermore, diastolic pressure reactivity to the competition task was
attenuated in borderline hypertensive subjects who underwent aerobic conditioning. These
data are interpreted as preliminary findings suggesting that borderline hypertensives may be
particularly responsive to the cardiovascular benefits of aerobic conditioning. For patients who
have progressed to this stage of hypertensive disease, aerobic exercise may be of ameliorative
value.

INTRODUCTION ergy delivery by the cardiovascular sys-

In a recent epidemiological study of a
large cohort of Harvard alumni, habitual
physical activity was shown to be a life-
style characteristic associated with in-
creased longevity, due primarily to a de-
crease in mortality from cardiovascular
and respiratory diseases (1). Aerobic ex-
ercise training may be especially benefi-
cial to cardiovascular health because of
the resulting physiological adaptations,
with both an improved potential for en-
From the Department of Psychiatry, University of
North Carolina, Chapel Hill, North Carolina (A.S.,
K.C.L.) and Duke University Medical Center, Dur-
ham, North Carolina (J.A.B.).
Address reprint requests to Andrew Sherwood,
Ph.D., CB #7175, Medical Research Building A, Uni-
versity of North Carolina, Chapel Hill NC 27599.
Received June 28,1988; revision received Novem-
ber 9, 1988.
tem and energy utilization by the skeletal
muscles leading to increased physical
work capacity (2). Of the numerous mech-
anisms that underlie these adaptations,
modifications in autonomic nervous sys-
tem control of the cardiovascular system
are of particular interest in the context of
the present study. Resting and exercise
bradycardia is characteristic of the exer-
cise-trained individual and is considered
to be at least partially attributable to a
decreased level of sympathetic and in-
creased level of parasympathetic influ-
ences on the heart (3). The converse state
of autonomic balance, with heightened
sympathetic and reduced parasympa-
thetic control of the heart, associated with
abnormally high cardiac output (CO), is
responsible for elevated blood pressure
(BP) in some cases of borderline hyperten-
sion, a condition that frequently pro-
gresses to established hypertension (4).

Psychosomatic Medicine 51:123-136 (1989) 123

0033-3174/89/5102-0123$O2.OO/O
Copyright© 1989 by the American Psychosomatic Society

In contrast to the beneficial aspects of
aerobic exercise, psychological stress is
generally considered to be deleterious to
cardiovascular health. Several models
have been proposed that implicate the
sympathetic nervous system in the path-
ological process by which behavioral in-
fluences may contribute to both coronary
heart disease (CHD) (5, 6) and hyperten-
sion (7). According to these models, indi-
viduals who repeatedly exhibit pro-
nounced sympathetically mediated
arousal of the cardiovascular system, in
association with their responses during
psychological stress, may be at particular
risk for the development of cardiovascu-
lar disease (8). It seems plausible that
aerobic exercise training may ameliorate
the risk of cardiovascular disease associ-
ated with behavioral influences, both by
attentuation of sympathetic responses
and/or through modification of personal-
ity traits and behavior patterns.
Relatively few studies have been con-
ducted to examine directly the effects of
aerobic exercise on physiological reactiv-
ity during psychological stress. Positive
findings from psychophysiological studies
include reports that heart rate (HR) re-
sponses during psychological stress are
attenuated in physically fit individuals
(9-11). Other studies have found negative
results regarding differences in HR reac-
tivity but have found that HR recovery
from stress tends to occur more rapidly in
the aerobically fit (12-15). Myocardial
contractility and systolic blood pressure
(SBP) responses during stress, in addition
to HR, were found in a recent study to be
less pronounced in college students cate-
gorized as high in aerobic fitness, relative
to low fit students (16). Diastolic blood
pressure (DBP) increases during stress
have also been found to be attenuated in
association with physical fitness, in a
124
A. SHERWOOD et al.
study which otherwise reported negative
findings (17). Methodological differences
between studies which may partly ac-
count for the inconsistency of findings to
date include cross-sectional versus longi-
tudinal designs, differences in the labo-
ratory stressors employed, and differences
in the definition and measurement of
physical fitness. Also, benefits may differ
depending on whether subjects under
study have borderline or sustained hyper-
tension versus normal resting BPs. Per-
sons with borderline hypertension are a
population of special interest: two recent
five-year prospective follow-up studies
have shown that a high DBP response
during and after stress is a powerful pre-
dictor of the development of sustained
hypertension in such individuals (18, 19).
Overall, further research seems justified
to help clarify if, how, and in which in-
dividuals exercise might modify stress re-
activity.
In a previous study of Type A men (20)
we found that a 12-week program of aero-
bic exercise was associated with signifi-
cant attenuation of HR and BP responses
to mental arithmetic. In the present study,
on a new cohort of Type A men, some of
whom were borderline hypertensive, we
provided a more comprehensive assess-
ment of cardiovascular function, using
impedance cardiography to assess the
hemodynamic basis of BP responses. Car-
diovascular responses to a reaction time
task, in which pairs of subjects were in
direct competition for a monetary bonus,
were monitored both before and after a
12-week supervised program of either
aerobic or strength training. A number of
studies have shown that resting blood
pressure is reduced in hypertensive and
borderline hypertensive individuals fol-
lowing aerobic exercise training (21, 22).
Since greater reactivity during mental
Psychosomatic Medicine 51:123-136 (1989)
AEROBIC TRAINING EFFECTS ON STRESS RESPONSES

stress has been reported to occur in bor-
derline hypertensives (23-25), we were
particularly interested in assessing
whether subjects in this category would
exhibit heightened reactivity to the com-
petition task; and, if so, whether atten-
uation of cardiovascular reactivity follow-
ing aerobic training might be particularly
evident in borderline hypertensive sub-
jects.
METHODS
after instrumentation, just before the initial rest
period of the pre-training laboratory competition
task test session. Subjects were categorized as bor-
derline hypertensive if resting SBP exceeded 140
mm Hg and/or DPB exceeded 90 mm Hg during the
latter period while all other subjects were consid-
ered normotensive. According to these criteria, in
the Strength training group, six subjects were cate-
gorized as borderline hypertensive and seven as
normotensive, while in the Aerobic group there
were five subjects classified as borderline hyperten-
sive and nine normotensive. Group mean casual
blood pressures taken at the screening physical ex-
amination and immediately after instrumentation,
before the competition task protocol, are summa-
rized in Table 1.

Subjects
Data based upon 27 men, aged between 33 and 56
years (mean = 41.4 years) and who were initially
rated as Type A by the Structured Interview (26),
are described in the present study. (Note: thirty-four
men comprised the initial sample, but three failed
to complete the exercise training program, one failed
to show for follow-up testing, and physiological
measurement problems led to the loss of data on
another three.) All subjects were employed and had
at least a high school education Clinical manifesta-
tions of CHD were absent in all subjects as deter-
mined by medical history, physical examination,
and exercise treadmill testing.
A number of subjects participating in this study
could be considered borderline hypertensive accord-
ing to casual (ausculatatory) BP readings taken dur-
ing the initial screening physical examination. Some
subjects also showed borderline hypertensive pres-
sures during the first three seated readings taken
Exercise Intervention Program
Subjects were randomly assigned to either an
Aerobic exercise training group (N=14) or a Strength
and Flexibility training group (N=13). Subjects in
the Aerobic group attended three supervised exer-
cise training sessions per week for 12 consecutive
weeks. Aerobic exercise sessions began with a 15-
min stretching period followed by 35 min of contin-
uous aerobic exercise which included walking, jog-
ging, and stair-climbing at an intensity of at least
70% of subjects' initial maximal aerobic power
(V02max). Subjects were instructed to monitor their
HRs by taking their radial pulses. A HR monitor
(Exersentry 3) was also worn twice a week during
the first two weeks of the training program, and once
a week thereafter, to ensure that subjects' HRs were
kept within the prescribed training ranges.
Subjects in the Strength group participated in 20
min of flexibility exercises, followed by 30 min of

TABLE 1. Means and SEMs for SBP and Casual DBP in Borderline Hypertensive and Normotensive
Subject Groups
Aerobic Training Group Strength Training Croup

Borderline Borderline ..
, Normotensives, Normotensives
hypertensives . hypertensives
(N = 9) (N=7)
(N = 6)
Casual auscultatory BPa Systolic (mm Hg) 140 ±7.6 122 ± 2.2 146 + 6.4 132 ± 2.0
Diastolic (mm Hg) 98 ± 2.7 79 ± 2.8 93 + 3.3 86 ±1.8
BP readings taken immediately Systolic (mm Hg) 140 ±4.0 122 ±2.3 138 ±6.0 122 ±2.5
6
after instrumentation Diastolic (mm Hg) 96 ± 3.0 82 ±1.2 90 ± 3.7 80 + 2.2
Baseline BPs, reported under Results, are lower since they were recorded following 15 min of quiet rest.
a
Taken during screening physical examination.
b
Prior to onset of competition task protocol.

Psychosomatic Medicine 51:123-136 (1989) 125


Nautilus circuit training. Subjects participated in the
supervised group sessions two to three times per
week for 12 consecutive weeks. The Strength train-
ing group was employed to serve as a control for the
effects of social stimulation and attention, since both
groups were supervised. Cardiovascular adaptations
characteristic of aerobic conditioning are not pro-
duced by Nautilus circuit training (27). All training
sessions were concluded with a 10-min "cool-down"
period. Subjects in the Strength group were re-
quested not to engage in any aerobic exercise during
their participation in this study. Subjects in both
groups were instructed to maintain their dietary
habits until the completion of the study, and no
suggestions for dietary modification were offered.
All participants underwent comprehensive phys-
iological evaluations as well as behavioral and psy-
chophysiological assessments. Evaluations were
conducted prior to the beginning of the exercise
training program and after three months of exercise
conditioning. The focus of the present report is on
the hemodynamic adjustments to the psychosocial
challenge presented by performance on a reaction
time task in which subjects were placed in direct
competition with each other.
Treadmill Exercise Test
During a one-week evaluation period, pre- and
post-exercise training, subjects' V02max was meas-
ured during a symptom-limited maximum treadmill
(Pacesetter R-9) exercise test following the Duke
protocol of 1-Met/mi n increments (28). Oxygen con-
sumption (VO2) was measured using a Beckman Met-
abolic Cart (Sensormedics), and HR was measured
from the continuous electrocardiogram.
Psychosocial Stress Test Protocol
Subjects participating in this study were tested in
pairs and were exposed to essentially identical ex-
perimental protocols on each of two sessions. The
same two subjects were never paired together on
both the pre- and post-exercise training sessions.
After signing an informed consent form and fol-
lowing instrumentation for physiological recording,
subjects were seated 6 feet apart, facing a projector
screen, in an electrically shielded, sound-attenuated
experimental chamber. At this time three BP read-
ings were taken in the presence of a technician,
using our automated auscultatory system. The ex-
perimental session was initiated by a 15-min rest
126
A. SHERWOOD et al.
period, during which subjects were asked to relax
and remain quiet. At the end of this period a tech-
nician entered the room and instructed subjects on
the nature of'the competition task.
The competition task was 5 min long and required
subjects to depress a button, as quickly as possible,
upon identifying that a target letter was present in a
live-letter word, projected onto the screen in front
of them. A total of 17 words were presented at
varying intervals, with a mean intertrial interval of
18 sec and a range of 8-38 sec. Each subject was
designated a target letter during the pre-task instruc-
tion period. However, it was explained at this time
that, in order to make the task more difficult, after
every few trials (two to five trials) a technician, who
sat between the subjects throughout the task, would
say "Switch," at which point each subject should
respond only to the target letter to which his oppo-
nent had just previously been responding. Subjects'
reaction times for each trial were displayed on dig-
ital stopclocks (Lafayette model 54519-A). In order
to monitor their performance in an overt manner,
the technician recorded each subject's reaction
times on a clipboard scoresheet. It was explained
that, in this "winner-take-all" competition, a $20
bonus would be awarded to the subject with the
lower total reaction time score over the entire task.
Since, it was explained, any incorrect response
would lead to a 2-sec penalty for that trial, it was
emphasized that subjects should not give up, even
if they felt they were behind, as a single mistake by
their opponent could profoundly affect the compe-
tition outcome. The task was designed so that sub-
jects could be required to both respond, either re-
spond or neither respond on a given trial, but both
subjects would need to respond to 9 of the 17 trials
if all responses were to be correct
Following completion of the competition task,
subjects were told that they would be informed of
the outcome of the contest at the end of the experi-
ment and were asked again to relax quietly for a
second 15-min rest (recovery) period. At the end of
this period subjects were asked to complete a brief
questionnaire, asking them to indicate on a seven-
point scale how hard they tried (one=not at all;
4=moderately; 7=very, very much) during the com-
petition.
Physiological Measurements
Arterial BP was measured noninvasively using
the auscultatory method. A standard inflatable BP
cuff, positioned around the subject's left arm, was
rapidly inflated and then slowly deflated at a linear
Psychosomatic Medicine 51:123-136 (1989)
AEROBIC TRAINING EFFECTS ON STRESS RESPONSES
rate of 3 mm Hg/sec, using a custom-designed and
-built automated device. Kortkoff (K) sounds were
detected using a piezoelectric microphone posi-
tioned over the brachial artery and under the lower
edge of the cuff. Cuff pressure and the microphone
output were displayed on adjacent channels of a
Beckman Dynograph chart recorder, permitting the
determination of SBP, as the cuff pressure corre-
sponding to the onset (Phase I), and DBP, as the
disappearance (Phase V) of K-sounds. Mean arterial
blood pressure (MAP) was derived for each set of
pressure readings by computing V> pulse pressure
added to DBP.
Impedance cardiography was utilized to permit
noninvasive monitoring of cardiac performance (29).
A Minnesota Impedance Cardiograph (model 304B)
was used in conjunction with a tetrapolar electrode
band configuration. All four electrodes (Contact
Products No. M6001) were used in conjunction with
an electrode gel (Aquasonic 100) to facilitate electri-
cal transmission. The inner two recording electrode
bands were positioned around the base of the neck
and around the thorax over the tip of the xiphoid
process. The outer two current electrode bands were
positioned to encompass the neck and thorax, at
least 4 cm away from each of the recording elec-
trodes. The EKG was recorded independently using
Beckman biopotential electrodes. The basal thoracic
impedance (Zo), the first derivative of the pulsatile
impedance (dZ/dt) and the EKG waveforms were
recorded on a FM tape recorder (TEAC model
MR30). Processing of the impedance cardiogram was
accomplished using a computer-based system devel-
oped in the laboratory. This system, which has pre-
viously been described in detail (30), was used to
derive CO, HR, stroke volume (SV), pre-ejection
period (PEP), and left ventricular ejection time
(LVET). Each of these indices was generated to rep-
resent minute-mean values, based upon 30-sec con-
tinuous data samples, taken from within the 1-min
periods of interest.
Total peripheral resistance (TPR) of the systemic
vasculature was derived on the basis of the concur-
rently recorded blood pressure and cardiac output
measurements using the equation: TPR (dyne-
sec cm"5) = (MAP/CO)*80.
RESULTS
Treadmill Exercise Test
Training effects on V02max were eval-
uated using a three-way repeated meas-
Psychosomatic Medicine 51:123-136 (1989)
ures analysis of variance (ANOVA) (BP
status x Exercise Group x Session), which
gave rise to a Group x Session effect
(F(l,23) = 9.27, p < 0.01]. Tukey's Hon-
estly Significant Difference (HSD] test re-
vealed that this interaction emerged be-
cause there was a significant (+13.6%)
improvement in V02max in the Aerobic
group (mean pre-training = 33.76 ml/kg/
min; post-training = 38.36 ml/kg/min),
while the change in the Strength group
(+2.9%) was nonsignificant (mean pre-
training = 34.08 ml/kg/min; post-training
= 35.07 ml/kg/min). Training effects on
HR responses to the treadmill test session
were evaluated using a four-way repeated
measures ANOVA (BP status x Exercise
group x Session x Phase (Rest, Exercise
at 3.5, 5, and 8 Mets)). Post-training HRs
were significantly lower at rest and dur-
ing exercise in the Aerobic group only
(Groups X Sessions F(l,21) = 7.34, p <
0.02).
Cardiovascular Responses during the
Competition Task, Pre- and Post-
training
Cardiovascular responses were sum-
marized into five phases within sessions:
Baseline (average of the last 5 min of the
initial 15-min rest period); Comp 1 (first
minute of the competition task); Comp 2
(second minute of the competition task);
Recovery 1 (average of the first 5 min
following competition); and Recovery 2
(average of min 11-15 following competi-
tion). The first 2 min of Competition were
selected because, during this time, sub-
jects were most uncertain about their
chances of winning. Although the result-
ing cell sizes were small, BP status was
included as grouping factor in the ANO-
VAs and analyses of covariance (ANCO-
VAs) described below, in order to provide
a preliminary assessment of the possibil-
127

ity that there may be differential effects
of exercise training associated with preex-
isting BP status. Four-way repeated meas-
ures ANOVAs (BP status x Groups x Ses-
sions x Phases) were adopted as the pri-
mary means of statistical analysis and
were conducted for each cardiovascular
measure. The Greenhouse-Geisser cor-
rection procedure for repeated measures
was applied, and adjusted p values only
are reported. Tukey's HSD tests were ap-
plied where appropriate to identify the
basis of main and interactive effects. The
effects of exercise training on cardiovas-
cular reactivity (defined as change from
resting Baseline to Competition) and re-
covery rates were assessed using AN-
COVA tests with Baseline values as the
covariate.
BP responses. Mean SBP and DBP, pre-
and post-training, are presented in Figure
1. For DBP there was a significant Phases
effect (F(4,92) = 18.95, p < 0.01), which
was found to be due to a rise in DBP
during the Competition task, with levels
returning to within Baseline range by the
first Recovery period. DBP was signifi-
cantly higher overall for borderline hy-
pertensives (BP status F(l,23) = 5.55, p <
0.05). There was a general fall in DBP
following training (Sessions F(l,23) = 4.85,
p < 0.05), but this effect was primarily
due to the dramatic change occurring in
borderline hypertensive subjects who
underwent aerobic exercise training (BP
status X Exercise groups X Sessions
F(l,23) = 6.28, p < 0.02), for whom post-
training DBP levels were reduced to nor-
motensive control levels. Furthermore, a
significant BP status X Exercise Groups X
Sessions X Phases interaction (F(4,92) =
2.65, p < 0.05) supported the effect sug-
gested in Figure 1, that aerobic training
also attenuated the DBP response to the
Competition task in borderline hyperten-
128
A. SHERWOOD et al.
sive men. This reduction in DBP reactiv-
ity during stress was confirmed by an
ANCOVA adjusting for Baseline levels,
which gave rise to a significant BP status
x Exercise Groups x Sessions effect
(F(l,22=6.23, p < 0.05) showing that DBP
responses were reduced both during
(Comp min 1 and 2) and following (Recov-
ery 1 and 2) the Competition task.
SBP showed a significant Phases effect
(F(4,92) = 60.17, p < 0.001), accounted for
by a rise in SBP during Competition
which remained elevated during Recov-
ery 1, but returned to Baseline by Recov-
ery 2. A significant main effect of BP sta-
tus (/(1,23) = 11.91, p < 0.002) indicated
that SBP was higher in the borderline
hypertensive subjects across all phases of
the stress test protocol, both before and
after training. There were no other signif-
icant main or interactive effects for SBP.
However, it is of note, as shown in Figure
1, that though nonsignificant, mean SBP
in the borderline hypertensives who
underwent Aerobic training showed a
tendency to fall post-training. The AN-
COVA test indicated that neither training
program significantly affected SBP reac-
tivity to the Competition task or rate of
recovery of SBP to Baseline levels.
Myocardiai responses. HR means are
presented in Table 2. The Competition
task gave rise to reliable increases in HR
as indicated by a Phases effect (F(4,92) =
36.66, p < 0.001), with HR remaining el-
evated above Baseline during the first Re-
covery period, returning to Baseline
by Recovery 2. Normotensive subjects
showed greater HR reactivity to the Com-
petition task than the borderline hyper-
tensives (ANOVA BP status X Phases
F(4,92) = 5.37, p < 0.001; ANCOVA BP
status main effect F(l,22) = 5.66, p < 0.05).
Post-training HRs were generally lower in
both exercise groups (Sessions F(l,23) =
Psychosomatic Medicine 51:123-136 (1989)
AEROBIC TRAINING EFFECTS O N STRESS RESPONSES

STRENGTH GROUP AEROBIC GROUP

155

150

145

140

X 135
E
E.
o. 130
m
CO

125

120

115

110

105

100

95

90

Bose Compete Recov Recov Base Compete Recov Recov

10-15 1 2 1-5 10-15 10-15 I 2 1-5 10-15

Fig. 1. SBP and DBP means for borderline hypertensive and normotensive men during resting baseline
(Base), competition task (Compete), and post-task recovery (Recov), before and after 12 weeks of
aerobic or strength training.

Psychosomatic Medicine 51:123-136 (1989) 129

 A. SHERWOOD et al.

TABLE 2. Mean HR, CO, SV, PEP, and TPRS and Total Responses during the Stress Test Protocol,
Before and After Exercise Training
Aerobic Training Croup Strength Training Croup

Borderline Borderline
Normotensives Normotensives
hypertensives hypertensives
(n = 9) (n = 7)
(n = 5) (n = 6)

Pre- Post- Pre- Post- Pre- Post- Pre- Post-

training training training training training training training training
HR(bpm)
Baseline 67 61 68 59 69 62 71 70
Comp 1 77 75 86 79 75 73 94 93
Comp 2 79 72 85 75 76 72 92 89
Recov 1 75 68 73 66 72 67 78 75
Recov 2 72 66 71 63 70 65 75 73

CO (liters/mm)
Baseline 3.57 3.95 5.00 4.97 4.35 4.78 4.39 4.38
Comp 1 5.09 4.63 6.59 6.54 4.83 5.86 7.34 7.64
Comp 2 5.35 4.88 6.75 6.49 5.14 6.05 7.54 7.57
Recov 1 4.15 4.42 5.53 5.57 4.76 5.20 4.95 4.72
Recov 2 3.59 3.94 4.86 4.85 4.58 4.60 4.54 4.28

SV (ml)
Baseline 57.0 64.8 74.2 82.9 61.1 78 9 62.5 63.4
Comp 1 68.9 62.2 76.0 81.5 61.7 80.6 76.6 79.2
Comp 2 71.4 68.7 79.3 88.1 64.8 82.6 79.5 84.7
Recov 1 58.7 65.3 76.3 82.2 64.7 78.7 63.4 63.1
Recov 2 53.5 60.1 69.6 76.3 63.5 71.4 61.2 58.5

PEP (msec)
Baseline 98 95 87 89 92 81 94 106
Comp 1 88 81 64 80 88 78 69 71
Comp 2 84 78 67 74 86 78 67 71
Recov 1 95 89 81 84 89 82 89 99
Recov 2 98 93 87 88 91 83 92 106

TPR (dyne-sec-cm"5)
Baseline 2640 2065 1516 1541 2199 1824 1803 1685
Comp 1 2421 1961 1304 1459 2433 1728 1338 1231
Comp 2 2315 1811 1299 1343 2218 1811 1386 1211
Recov 1 2669 1884 1421 1466 2133 1812 1857 1708
Recov 2 2882 2103 1585 1609 2170 1997 1820 1803

9.64, p < 0.005). Although the latter effect
appeared to be more pronounced in the
Aerobic group (see Table 2), there was no
significant interaction to indicate that it
was reliably more so than in the Strength
group. HR reactivity to the Competition
task and subsequent rate of HR recovery
to Baseline were not affected by either
exercise training program.
CO (see Table 2) was also significantly
elevated during the Competition task
(Phases F(4,92) = 23.22, p < 0.001) but
returned to baseline range by Recovery 1.
Normotensive subjects showed greater

130 Psychosomatic Medicine 51:123-136 (1989)

AEROBIC TRAINING EFFECTS ON STRESS RESPONSES
CO reactivity to the Competition task
than the borderline hypertensives (AN-
OVA BP status X Phases F(4,92) = 4.08, p
< 0.05; ANCOVA BP status X Phases
F(3,69) = 4.36, p < 0.05). Although exer-
cise training led to lower overall HRs,
there was no significant change in CO
post-training in either group.
SV also showed a significant Phases ef-
fect Cf(4,92) = 11.29, p < 0.001), which was
due to SV generally increasing during the
Competition task but returning to base-
line by the first recovery period (see Table
2). As would be expected from the post-
training reduction in HR, which was not
associated with an overall change in CO,
there was a significant overall increase in
SV post-training {Sessions F(l,23) = 4.66,
p < 0.05). There was also a significant
Exercise Groups x Sessions x Phases in-
teraction for SV (F(4,92) = 3.82, p < 0.05),
which multiple comparisons (Tukey's
HSD) revealed to be predominantly due
to the pre-training increase in SV in re-
sponse to the Competition task persisting
following exercise training in the Strength
group, but showing a post-training dimi-
nution in the Aerobic group.
Myocardial PEP showed a significant
decrease in response to the Competition
task (Phases F(4,92) = 25.92, p < 0.001),
indicating a significant increase in con-
tractility during Competition, but re-
turned to baseline levels by the first Re-
covery period (see Table 2). Exercise
training did not affect resting Baseline or
Recovery PEP values, but during the first
minute of competition PEP tended to
show a lesser decrease post-training in the
Aerobic group, in contrast to a greater
decrease following Strength training, giv-
ing rise to an Exercise Groups X Sessions
X Phases effect (F(4,92) = 3.37, p < 0.005).
There was also a significant BP status X
Exercise Groups x Sessions X Phases in-
Psychosomatic Medicine 51:123-136 (1989)
teraction (F(4,92) = 5.30, p < 0.01) which
post-hoc analyses showed to be primarily
due to a reduction in PEP response to
Competition in the normotensive men fol-
lowing Aerobic exercise training. This at-
tenuation of PEP reactivity in normoten-
sives who underwent Aerobic training,
but not in those who underwent Strength
training, was further supported by an AN-
COVA BP status X Exercise Groups x
Sessions interaction (F(l,22) = 4.68, p <
0.05).
Vascular responses. TPR values, pre-
sented in Table 2, decreased significantly
during the Competition task (Phases
F(4,92) = 9.63, p < 0.001); this decrease in
response to the task showed a nonsignifi-
cant trend to be more pronounced in nor-
motensive than in hypertensive subjects
(BP status X Phases F(4,92) = 2.21, p =
0.074). There was an overall decrease in
TPR following Exercise training in both
groups (Sessions F(l,23) = 4.44 p < 0.05).
There was also a significant Groups X
Sessions X Phases interaction (F(4,92) =
3.58, p < 0.05) which was associated with
the absence of a significant fall in TPR
during the first minute of Competition on
the post-training session in the Aerobic
group (while the decrease in TPR was
greater during the same period following
Strength training). TPR was generally
higher in borderline hypertensive than in
normotensive subjects (BP status F(l,23)
= 6.50, p < 0.02), and there was a nonsig-
nificant trend toward lower overall TPR
levels in hypertensives following exercise
training (BP status X Sessions F(l,23) =
3.58, p = 0.07).
Competition Task Performance
Approximately half the subjects in each
Exercise training group won the Compe-
131
 A. SHERWOOD et al.

tition task on each of the two sessions. In which appears to be characteristic of tasks
both the Aerobic and Strength groups av- that encourage effortful active coping (32),
erage reaction times were similar on both has previously been demonstrated to be
sessions (overall mean reaction time = mediated via the sympathetic nervous
330 msec). Subjects' ratings of how hard system and, more specifically, through
they tried during the Competition task stimulation of the /3-adrenergic receptors
(scale of 1 to 7) were similar in both Ex- (30, 33).
ercise groups and similar pre- and post- Systolic BP, HR, and CO reactivity dur-
training (Strength pre-training = 6.4, post-ing the competition task remained essen-
training = 6.2; Aerobics pre-training = 6.4,tially unchanged following aerobic train-
post-training = 6.4). These results indicateing. However, there was some evidence
that task performance was similar in each that /3-adrenergic responses during the
of the two groups at both pre- and post- Competition task were attenuated follow-
training test sessions. ing aerobic training, particularly at the
onset of the task, where augmented myo-
cardial contracatility, as indexed by PEP
and SV changes, was initially less pro-
DISCUSSION
nounced and the vascular vasodilatation,
indexed by TPR decrease, was also less
In this study there was a significant marked. In contrast, following training in
improvement in aerobic fitness in sub- the Strength group, myocardial and vas-
jects assigned to the Aerobic training con- cular responses to the Competition task
dition, as verified by the 13.6% average tended to remain unaltered or were ac-
increase in directly measured VO2niax. In tually augmented.
contrast, subjects who underwent 3 Regarding the mechanisms by which
months of strength training showed an aerobic exercise training may potentially
improvement in V02max of only 2.9%, alter cardiovascular responses during
which was not statistically significant. As mental stress, the modification of central
expected, subjects in the Aerobic exercise autonomic balance toward reduced sym-
condition exhibited lower HRs at rest and pathetic and increased parasympathetic
during treadmill exercise at submaximal influences has been of particular interest.
workloads following training, while no One assumption has been that mental
such effects were evident in the Strength stressors that evoke cardiovascular re-
training group. Thus, our experimental sponses mediated via the sympathetic
manipulation was successful in inducing nervous system may be attenuated if sym-
differential changes in levels of aerobic pathetic activity is generally suppressed.
fitness. However, this proposition may be too sim-
plistic, since it is generally accepted that
Consistent with previous findings, the there is a peripheral systemic contribu-
typical rise in BP during the stressful tion to training-induced adaptations in
Competition task was characterized by an cardiovascular function. A clear illustra-
increase in CO, due primarily to HR ele- tion of this is provided by evidence that
vation with some accompanying augmen- improved maximal aerobic capacity, as
tation of SV, while there was a decrease well as altered cardiovascular adjust-
in TPR ( 30, 31). This response pattern, ments to exercise, are specific to the type

132 Psychosomatic Medicine 51:123-136 (1989)

AEROBIC TRAINING EFFECTS ON STRESS RESPONSES
of exercise engaged in during the training
program. For example, attenuated HR re-
sponses to submaximal bicycle exercise
are evident during bicycle exercise test-
ing but are not pronounced during testing
of responses to exercise in untrained arm
muscles (34). This phenomenon is thought
to be a reflection of modification in both
vascular and enzymatic processes related
to aerobic metabolism in the trained skel-
etal muscles (34). On the basis of this
evidence, it is interesting to speculate that
altered reactivity during mental stress
might therefore be expected to occur only
if the hemodynamic response pattern to
the mental stressor includes peripheral
responses that invoke the trained muscle
vascular beds. Studies incorporating fore-
arm blood flow measurement have shown
that reaction time tasks and mental arith-
metic both appear to be associated with
increased blood flow to the skeletal mus-
cles of the forearm (31, 35, 36). However,
in one study where calf blood flow was
measured in addition to the more stand-
ard forearm measurement, it was found
that augmented blood flow was present in
the forearm muscle and absent in the calf
(37). If the relationship of exercise training
to performance or exercise testing gener-
alizes to mental stress testing, then it is
possible that the running exercise used
for aerobic conditioning in the present
study may lead to less conspicuous effects
on cardiovascular reactivity than might
occur following aerobic conditioning re-
sulting from upper body exercise training.
Sinyor et al. (15) have made the point that
the nature of the mentally stressful task
may be a relevant consideration in under-
standing the relation of aerobic fitness to
stress reactivity. Extending this point, for
longitudinal studies, changes in stress re-
activity associated with changes in aero-
bic fitness may depend on both the nature
Psychosomatic Medicine 51:123-136 (1989)
of the stressful task employed and the
nature of the aerobic training program.
The most dramatic effects of exercise
training in the present study were found
to occur in subjects who were classified
as borderline hypertensive. However,
since analyses using BP status were based
on small numbers of subjects per cell, the
reported effects should be viewed as pre-
liminary findings. Borderline hyperten-
sive subjects in the aerobic exercise group
exhibited a substantial reduction in DBP
following the 3-month training program.
Since CO was unaltered post-training,
both the reduction in TPR and the brady-
cardia resulting from aerobic training are
interpreted as together forming the un-
derlying hemodynamic basis for this DBP
reduction. Furthermore, DBP increase
during and following the competition task
were significantly reduced in the border-
line hypertensives who underwent aero-
bic exercise training, but not in those who
underwent strength training. Two studies
have demonstrated a predictive relation-
ship between high DBP responses during
and following mental stress and the de-
velopment of sustained hypertension over
the next five years (18, 19). The substan-
tial reduction in resting DBP together
with an attenuation of DBP reactivity dur-
ing psychological stress suggests that aero-
bic exercise training may be of ameliora-
tive benefit to borderline hypertensive
patients who might otherwise be at high
risk for the development of sustained hy-
pertension. Research supporting this pos-
sibility is provided by an animal study
that employed a rat model of borderline
hypertension and found that daily exer-
cise conveyed some resistance to the de-
velopment of hypertension associated
with exposure to chronic stress (38).
The lowered DBP levels and attenuated
DBP reactivity during stress following
133

aerobic conditioning in the borderline hy-
pertensive subjects are suggestive of an
alteration in vascular function. In a recent
study it was shown that the fall in DBP
associated with epinephrine infusion was
greater in aerobically fit individuals, as
well as being enhanced in association
with improved aerobic fitness in subjects
who underwent a 4-month exercise train-
ing program (39). Since no alterations in
myocardial responses to epinephrine
were observed, these results were inter-
preted as reflecting an increased sensitiv-
ity of vascular /3-adrenergic receptors
and/or a proliferation of skeletal muscle
precapillary blood vessels associated with
aerobic fitness. Since hypertensive dis-
ease is associated with a down-regulation
of/3-receptor sensitivity (40), it is possible
that hypertensive individuals may be es-
pecially susceptible to a reversal of this
trend brought about by aerobic condition-
ing. Such alterations in vascular /3-recep-
tors provide a plausible explanation for
both the initially high DBP reactivity dur-
ing the competition stress seen in our
borderline hypertensive subjects and the
subsequent alteration of this response
only in subjects who underwent aerobic
exercise training.
Further research investigating the ef-
fects of exercise on cardiovascular re-
sponses during psychosocial stress are
clearly needed in order to fully under-
stand the mechanisms by which aerobic
exercise may reduce the stress response.
Our preliminary findings indicate that
aerobic training may be beneficial in at-
tenuating pressor responses during stress,
especially in borderline hypertensives.
These data suggest the need for more
long-term prospective studies to assess the
ameliorative benefits of increased fitness,
especially in those hypertensives who are
hyperreactive during stress. Future stud-
134
A. SHERWOOD et al.
ies should also assess the possible benefits
of exercise in other specific populations,
including blacks, adolescents, and the el-
derly. There is also a need to clarify which
forms of exercise training are most effec-
tive in modifying cardiovascular re-
sponses during psychological stress. Pos-
sible mechanisms of reactivity attenua-
tion, such as reduced catecholamine
responses and/or altered adrenergic
receptor sensitivities, need to be evalu-
ated. Ambulatory blood pressure moni-
toring studies would also provide a means
of assessing whether the overall attenua-
tion of blood pressure levels, at rest as
well as during physical and psychological
stress, observed in the present study, gen-
eralize to a similar reduction in blood
pressure responses during everyday activ-
ities.
SUMMARY
The effects of a 12-week aerobic exer-
cise training program on responses to a
psychologically challenging competition
task were evaluated in a group of Type A
men. A second matched group of men
who underwent strength training served
as a control group. Improved aerobic fit-
ness was generally associated with low-
ered heart rates and blood pressure at rest,
during exercise, and during the psycho-
logically challenging competition task.
For subjects who pre-training casual
blood pressures fell in the borderline hy-
pertensive range, aerobic conditioning led
to both a substantial post-training reduc-
tion in resting blood pressure and a sig-
nificant attenuation of initially high dia-
stolic pressure reactivity to the competi-
tion task. While an overall reduction in
blood pressure levels resulting from aero-
bic conditioning is likely to benefit most
Psychosomatic Medicine 51:123-136 (1989)
AEROBIC TRAINING EFFECTS O N STRESS RESPONSES

individuals, the present findings are ten-
tatively interpreted as suggesting that bor-
derline hypertensives are one subgroup
for whom the benefits may be of imme-
diate clinical value by abating the pro-
gress of hypertensive disease.
We thank Doris MurrelJ, Sally Schnitz,
and Robin Pomeroy for assistance in data
coJIection and analysis, Margaret WaJsh
and Gwendy Wasser for supervision of
training procedures, and Dorothy Faulk-
ner for manuscript preparation. We also
thank Dr. Mats Frederickson for his help-
ful comments on an earlier draft of the
manuscript. This study was supported
by National Institutes of HeaJth grants
HL-30675, HL-19876, HL-01096, and
AGO4238, and by a grant from the John
D. and Catherine T. MacArthur Founda-
tion Research Network on the Determi-
nants and Consequences of Health-Pro-
moting and Health-Damaging Behavior.

REFERENCES

1. Paffenbarger RS, Jr, Hyde RT, Wing AL, Hsieh C: Physical activity, all-cause mortality, and longevity of
college alumni. N Engl J Med 314:605-613, 1986
2. Astrand PO, Rodahl K: Textbook of Work Physiology: Physiological Bases of Exercise. New York,
McGraw-Hill, 1977
3. Scheuer J, Tipton CM: Cardiovascular adaptations to physical training. Annu Rev Physiol 39:221-251
1977
4. Julius S, Esler MD: Autonomic nervous cardiovascular regulation in borderline hypertension. Am J
Cardiol 36:685-696, 1975
5. Herd JA: Physiological basis for behavioral influences in arteriosclerosis. In Dembroski TM, Schmidt
TH, Blumchen G (eds). Biobehavioral Bases of Coronary Heart Disease. Basel, Karger, 1983
6. Schneiderman N: Behavior, autonomic function and animal models of cardiovascular pathology. In
Dembroski TM, Schmidt TH, Blumchen G (eds). Biobehavioral Bases of Coronary Heart Disease. Basel,
Karger, 1983
7. Obrist PA: Cardiovascular Psychophysiology: A Perspective. New York, Plenum Press, 1981
8. Krantz DS, Manuck SB: Acute psychophysiologic reactivity and risk of cardiovascular disease: A review
and methodologic critique. Psychol Bull 96:435-464, 1984
9 Cantor JR, Zillman D, Day KD: Relationship between cardiorespiratory fitness and physiological
responses to films. Percept Motor Skills, 46:1123-1130, 1978
10. Holmes DS, Roth DL: Association of aerobic fitness with pulse rate and subjective responses to
psychological stress. Psychophysiology 22:525-529, 1985
11. Turner JR, Carroll D, Costello M, Sims J: The effects of aerobic fitness on additional heart rates during
active psychological challenge. J Psychophysiol 1988, in press
12. Cox JP, Evans JF, Jamieson JL: Aerobic power and tonic heart rate responses to psychosocial stressors.
Personality Soc Psychol Bull 5:160-163, 1979
13. Sinyor D, Schwartz SG, Peronnet F, Brisson G, Seraganian P: Aerobic fitness level and reacativity to
psychosocial stress: Physiological, biochemical and subjective measures. Psychosom Med 45:205-217,
1983
14. Hollander BJ, Seraganian P. Aerobic fitness and psychophysiological reactivity. Can J Behav Sci 16:257-
261,1984
15. Sinyor D, Golden M, Steinert Y, Seraganian P: Experimental manipulation of aerobic fitness and the
response to psychosocial stress: Heart rate and self report measures. Psychosom Med 48:324-337,1986
16. Light KC, Obrist PA, James SA, Strogatz DS: Cardiovascular responses to stress. II. Relationships to
aerobic exercise patterns. Psychophysiology24:79-86,1987
17. Hull EM, Young SH, Ziegler MG: Aerobic fitness affects cardiovascular and catecholamine responses to
stressors. Psychophysiology 21:353-360, 1984

Psychosomatic Medicine 51:123-136 (1989) 135

 A. SHERWOOD etal.

18. Falkner B, Onesti G, Hamstra B: Stress response characteristics of adolescents with high genetic risk for
essential hypertension: A five-year follow-up. Clin Exp Hyperten 3:583-591, 1981
19. Borghi C, Costa FV, Boschi S, Mussi A, Ambrosioni E: Predictors of Stable hypertension in young
borderline subjects: A five-year follow-up study. J. Cardiovasc Pharmacol 8(Suppl 5):S138-141, 1986
20. Blumenthal JA, Cox DR. Walsh MA, Emery CF, Kuhn CM, Williams RB, Williams RS: Exercise training
in healthy Type A middle-aged men: Effects on behavioral and cardiovascular reactivity. Psychosom
Med in press, 1988
21. Kenny WK, Zambraski EJ: Physical activity in human hypertension: A mechanisms approach. Sports
Med 1:459-473, 1984
22. Siskovick DS. LaPorte RE, Newman JM: The disease-specific benefits and risks of physical activity and
exercise. Public Health Rep 100180-188,1985
23. Nestel PJ: Blood pressure and catecholamine excretion after mental stress in labile hypertension. Lancet
1:692-694, 1969
24. Jem S: Psychological and Haemodynamic factors in borderline hypertension. Acta Med Scand Suppl
662:1-55, 1982
25. Falkner B, Onesti G, Angelakos ET, Fernandes M, Langman C: Cardiovascular responses to mental
stress in normal adolescents with hypertensive parents. Hypertension 1:23-30, 1979
26. Rosenman RH: The interview method of assessment of the coronary-prone behavior pattern. In
Dembroski TM, Weiss SM, Shields J, Haynes S, Feinleib M (eds). Coronary-prone Behavior. New York,
Springer, 1978
27. Hurley BF, Seals DR, Ehsani AA, Cartier LJ, Dalsky GP, Hagberg JM, Holloszy JO: Effects of high intensity
strength training on cardiovascular function. Med Sci Sports Exerc 16:483-488,1984
28. Blumenthal JA, Rejeski WJ, Walsh MA, Miller H, Roark S, Ribisl PM, Morris PB, Brubaker P, Sanders
Williams R: A randomized trial of high and low intensity exercise training in patients recovering from
myocardial infarction. Am J Cardiol 61:26-30, 1988
29. Miller JC. Horvath SM: Impedance cardiography. Psychophysiology 15:80-91, 1978
30. Sherwood A, Allen MT, Obrist PA, Langer AW: Evaluation of beta-adrenergic influences on cardiovas-
cular and metabolic adjustments to physical and psychological stress. Psychophysiology, 23:89-104,
1986
31. Brod J, Fencl V, Hejl Z, Jirka J: Circulatory changes underlying blood pressure elevation during acute
emotional stress (mental arithmetic) in normotensive and hypertensive subjects. Clinical Science
18:269-279, 1959
32. Light KC, Obrist PA: Cardiovascular responses to stress: Effects of opportunity to avoid, shock experience
and performance feedback. Psychophysiology 17:243-252, 1980
33. Langer AW, McCubbin JA, Stoney CM, Hutcheson JS, Charlton JD, Obrist PA: Cardiopulmonary
adjustments during exercise and an aversive reaction time task: Effects of beta-adrenoceptor blockade.
Psychophysiology 22:59-68, 1985
34. Clausen JP: Effects of physical training on cardiovascular adjustments to exercise in man. Physiolog
Rev 57:779-815, 1977
35. Williams RB, Bittker FE, Buchsbaum MS, Wynne LC: Cardiovascular and neurophysiologic correlates
of sensory intake and rejection. Psychophysiology 122:427-432,1975
36. Allen MT, Obrist PA, Sherwood A, Crowell MD: Evaluation of myocardial and peripheral vascular
responses during reaction time, mental arithmetic and cold pressor tasks Psychophysiology 24:648-
656, 1987
37. Rusch NJ, Shepherd JT, Webb C, Vanhoutte PM: Different behavior of the resistance vessels of the
human calf and forearm during contralateral isometric exercise, mental stress, and abnormal respiratory
movements. Circ Res 48(Suppl I):Il 18-130, 1981
38. Cox RH, Hubbard JW, Lawler JE, Sanders BJ, Mitchell VP: Exercise training attenuates stress-induced
hypertension in the rat. Hypertension 7:747-751, 1985
39. Svedenhag J, Martinsson A, Ekblom B, Hjemdahl P: Altered cardiovascular responsiveness to adrenaline
in endurance-trained subjects. Acta Physiol Scand 126:539-550,1986
40. Buhler FR, Kiowski W, van Brummelen P, Amann FW, Bertel O, Landmann R, Lutold BE, Bolli P:
Plasma catecholamines and cardiac, renal and peripheral vascular adrenoceptor-mediated responses in
different age groups of normal and hypertensive subjects. Clin Exp Hypertens 2:409-426, 1980.

136 Psychosomatic Medicine 51:123-136 (1989)