Journal of Gerontology: BIOLOGICAL SCIENCES
2003, Vol. 58A, No. 7, 585–591
Copyright 2003 by The Gerontological Society of America
Exercise Heart Rate Variability of Older Women
in Relation to Level of Physical Activity
Sylvia Reland,1 Nathalie S. Ville,2 Sara Wong,3 Hervé Gauvrit,3 Gaëlle Kervio,1 and François Carré1
2
1
Groupe de Recherche Cardiovasculaire, Unité de Biologie et Médecine du Sport, Rennes, France.
Laboratoire de Physiologie et Biomécanique de l’Exercice Musculaire, UFR STAPS Université de Rennes II, France.
3
Laboratoire du Traitement du Signal et de l’Image, INSERM Université de Rennes I, France.
The purpose of this study was to determine the effect of the level of physical activity in older
women on heart rate (HR) response to its neural control at rest and during exercise by using heart
rate variability (HRV) analysis. Electrocardiogram (ECG) was recorded in 3 (low, moderately,
and highly) active groups of older women at rest and during submaximal exercise. Spectral HRV
indexes were obtained from the ECG signal. At rest, highly active subjects have low HR without
any alteration of HRV. During incremental submaximal exercise, parasympathetic modulations
of HR decreased only in the highly active subjects ( p , .01) without any alteration of HR,
compared with the other groups. In older women, the effects of the level of physical activity on
HR and HRV are dissociated. Quite a high level of physical training induces a higher sensitivity
of sinus node response to the autonomic nervous system during exercise.
H
EART rate (HR) is primarily controlled by autonomic
nervous system input to the sinus node. Its responses
to sympathetic and parasympathetic drive are explored
by a useful noninvasive tool: heart rate variability (HRV)
(1). HR is influenced by physiological factors such as
cardiorespiratory fitness and aging (2). Indeed, young and
older endurance-trained athletes generally present resting
bradycardia concomitant with an alteration of autonomic
balance illustrated by a high resting HRV (2). However
there were some discrepancies concerning the aging process
on resting HR in healthy sedentary people, which is
decreased (3), increased (4), or remains unmodified (5).
Generally, aging leads to a decrease in parasympathetic
influence on the heart, which is manifested by a decrease in
HRV (6). A low level of resting HRV has been identified as
a risk marker for all causes of mortality (7). The incidence
of clinically significant cardiovascular disease in women
increases with age, particularly after menopause (8). As the
postmenopausal women have both increased incidence of
cardiac event and a low physical activity level (9), it is
interesting to investigate the effects of physical activity
levels on HR and HRV in older women.
HR increases linearly with exercise intensity, and heart
rates at rest and at given levels of exercise are lower in both
young and older exercise-trained subjects (5). The effects
of physical training on parasympathetic and sympathetic
control of heart rate are traditionally based on HRV studies
conducted at rest or during postural change in older people
(2,10,11). Physical exercise, particularly in young and
middle-aged people, is used to study HRV during stress,
which greatly modifies the cardiovascular system homeostasis and induces autonomic alterations (12–14). However,
little is known about HRV in relation to physical activity
level during exercise in older people, particularly in women.
Moreover, these studies lead to conflicting results, probably
reflecting some differences in experimental protocols (15,16).
All these data underscore the need to better understand
the effects of physical activity levels on HR and HRV in
older women during exercise. Therefore the aim of this
study was to assess, in healthy postmenopausal women, the
HR responses to its neural control. To this end, HRV
analysis was used in accordance with different levels of
physical activity, at rest, and during moderate-intensity
physical exercise.
METHODS
Subjects
Thirty healthy postmenopausal women (aged 60 to 70
years) participated in the present investigation. The nature,
purpose, and risks of the study were explained to each
subject. Written informed consent was obtained from the
subject, and ethics committee approval was received. The
participants were all nonsmokers, normotensive, and free of
cardiovascular disease by history and clinical examination,
which included resting blood pressure measurement and
electrocardiogram (ECG). None were taking cardioactive
medication. Self-reported usual physical activity level was
assessed using a physical activity questionnaire (17). A
MET (Metabolic Equivalent Task) value was assigned to
each activity. The number of hours spent in each activity per
week was multiplied by the appropriate MET value and the
subject body mass to obtain a value of energy expenditure
(kcal/wk). The sum of the activity values was an estimation
of weekly exercise energy expenditure. Three groups were
then identified. Subjects reporting less than 1000 kcal/wk
were classified as low active, which corresponds to the
recommendations for minimal activity required (150 kcal/
585
RELAND ET AL.
586
Table 1. Anthropometric and Physiological Data of Low,
Moderately, and Highly Active Subjects
Low (n ¼ 12) Moderate (n ¼ 9)
Age (y)
67.3 6 1.2
Mass (kg)
57.8 6 2.2
Height (cm)
159.3 6 1.4
BMI (kg/m2)
22.7 6 0.6
Body fat percentage (%) 33.5 6 0.7
Activity questionnaire
707.0 6 73.7
(kcal/wk)
V_ O2max (ml/min/kg)
19.2 6 0.8
% predicted V_ O2max
94.6 6 2.1
ATV_ O2 (ml/min/kg)
12.3 6 0.5
HRmax (bpm)
151.1 6 2.84
Wmax (watts)
81.7 6 3.8
High (n ¼ 9)
68.8
57.7
157.6
23.2
33.0
1496.1
6
6
6
6
6
6
1.1
65.0
2.5
57.2
2.3
160.3
0.7
22.2
0.8
32.9
91.6* 3445.0
6
6
6
6
6
6
1.3
1.2
1.0
0.3
1.7
409.3 ,à
22.7
115.4
13.01
150.7
95.6
6
6
6
6
6
0.7*
1.4§
0.6
4.81
5.8*
6
6
6
6
6
1.4 ,à
5.8 ,à
0.7 ,à
4.37{,jj
5.7 ,à
31.8
151.2
19.1
164.6
137.8
* p , .05 between moderately and low active groups.
p , .001.
à
p , .001.
§
p , .01.
{
p , .05 between highly and moderately active groups.
jj
p , .05 between highly and low active groups.
BMI ¼ body mass index; V_ O2max ¼ maximal oxygen uptake; ATV_ O2 ¼
oxygen uptake at anaerobic threshold; HRmax ¼ maximal heart rate; Wmax ¼
maximal exercise power.
day) to optimize health (18). This group (n ¼ 12) was not
engaged in any regular physical activity. Subjects reporting
more than 2000 kcal/wk, which corresponds to a physical
activity level above which there is a reduced age-adjusted
relative risk of myocardial infarction, were considered
highly active (19). They (n ¼ 9) came from a cycle touring
club and had cycled regularly (in excess of 80 kilometers per
week) for more than 4 years. Subjects reporting physical
activity levels of 1000–2000 kcal/wk were considered
moderately active. This group (n ¼ 9) had regularly
practiced voluntary gymnastics for more than 4 years.
Physiological capacity was assessed by a maximal oxygen
uptake (V_ O2max) measurement, a classical index of
maximal cardiovascular function. The low active group
V_ O2max was lower than predicted (20). The moderately
active V_ O2max was 100%–120% of that predicted, while
the highly active V_ O2max was higher than 130% of
predicted uptake. Body fat percentage was measured using
the skinfold thickness method (21). Anthropometric and
physiological data is described in Table 1.
Experimental Procedure
All subjects performed a graded maximal exercise test
to measure V_ O2max (ml/min/kg), HRmax (beat/min),
maximal exercise power (Wmax, watts), and to reveal
medical exclusion criteria for the study. The anaerobic
threshold was calculated with Beaver’s method (20) by 2
technicians (N.S.V. and F.C). The graded exercise test was
performed on an ergocycle (ERG 900, Marquette Hellige,
Milwaukee, WI), with continuous recording 12-lead ECG
(Cardio System, Marquette Hellige) and breath-by-breath
gas exchange analysis (Oxycon Delta, Jaeger, Hoechberg,
Germany).
The subject sat quietly on the ergocycle for 2 minutes,
connected to the gas analyzer. Then, after a 3-minute
warming-up period at 20 watts, the work rate was increased
10 watts every minute until exhaustion. The mean V_ O2 and
HR as measured during the last 30 seconds of each stage
were taken into account. Arterial blood pressure was
automatically measured every 2 minutes. The exercise test
was stopped when at least three classical criterions of V_ O2
max were attained (22).
ECG recordings for HRV measurement were always
performed 1 week after the maximal exercise test. Subjects
were instructed to refrain from any excessive physical
activity and ingesting caffeine or alcohol for at least 24
hours before testing. Trials were held in the morning, 3
hours after breakfast. Before measurement, subjects rested
for 10 minutes in a supine position in a quiet room. All
recordings were performed by the same technicians (S.R.
and F.C.). Lighting was kept constant and noise levels were
minimized during the test. The subjects performed 2 tests,
including 1 familiarization test and 1 experimental test at
intervals of 1 week. During the experimental test, a 3-lead
ECG was recorded at rest, in a seated position on the
ergocycle during a 6-minute period, and during dynamic
exercise performed at 30% and 60% of individual Wmax,
each over a period of 10 minutes. No attempt was made to
influence breathing frequency or tidal volume.
ECG Data Analysis
The ECG was sampled at 1000 Hz with the PowerLab
acquisition system (ADInstruments Pty. Ltd., Castle Hill,
Australia) installed in a Macintosh computer (Power Mac,
Cupertino, CA). The accuracy of the measurements in the
present study was within 1 millisecond. The ECG recordings for the 6-minute long resting period and the last 6
minutes of all exercises were then analyzed. The detection
of the QRS complex was conducted using Gritzali’s
algorithm (23). The RR intervals sequence was defined by
the duration between 2 consecutive R peaks. These data
were edited to eliminate any glitches due to premature
cardiac contraction using the procedure reported by Bruggeman and Andersen (24). Each RR interval was visually
validated by 2 confirmed experts (S.R. and S.W.) before
temporal and spectral analysis. For each RR sequence, the
mean RR interval was calculated at the time of each
recording (RR) and represented mean heart rate. Prior to
power spectrum density estimation, the RR sequence, which
is intrinsically nonevenly spaced data, was linearly interpolated to obtain a series of uniformly sampled data. The
retained sampling rate was set to 2 Hz. Using a sliding
window of 64 seconds’ duration, a time-varying autoregressive modeling of the interpolated RR sequence was performed
for estimating its power spectrum (ms2) to eliminate the slight
discrepancies of the sequence. Low frequency (LF) was
defined as 0.04–0.15 Hz, high frequency (HF) was defined as
0.15–0.4 Hz, and total power (TP) was the sum of HF þ LF
(25). During exercise, as recommended and because of the
higher ventilation rate, HF was defined as 0.15–0.5 Hz (26).
The spectral power of the RR series in these frequency bands
was then calculated and averaged over the duration of each
recording. HF corresponds to the respiratory sinus arrhythmia
and represents vagally mediated modulations in heart rate. LF
is influenced by both sympathetic and parasympathetic
activities (26).
HEART RATE VARIABILITY OF OLDER WOMEN
HF and LF are expressed in absolute terms (ms2). LF and
HF are also expressed in normalized units (NU), which
represent the relative value of each power component in
proportion to the total power minus very low frequency
(VLF) value (%). LF/HF is a marker of sympathovagal
balance (25).
Statistical Analysis
Results were presented as mean 6 standard error (SE) of
the mean. A one-way analysis of variance (ANOVA) was
performed to compare anthropometric and physiological
data.
HRV parameters were systematically tested for normality.
If the test was not passed, the natural log transformation was
done before performing the statistical test.
The influence of physical activity levels on HRV was
assessed during the experimental test. A two-way repeated
measures ANOVA was performed on each variable (3
groups 3 3 conditions). The 3 conditions for the test were
the seated resting position, 30%, and 60% Wmax. The
location of significant pair-wise differences was determined
using a Tukey’s post hoc procedure.
Pearson’s correlation coefficient was used to assess
relationships in the whole population between HRV indexes
and V_ O2max and between HRV indexes and oxygen uptake
at anaerobic threshold (ATV_ O2).
A p value .05 was considered significant. Statistical
analyses were performed with Statistica software version
5.97 (StatSoft, Inc., France).
RESULTS
There were no significant differences in age, height,
weight, body mass index, or body fat percentage between
the 3 groups. The score for the physical activity questionnaire, Wmax, and V_ O2max, were the highest in the highly
active group (p , .001), and higher in the moderately active
group compared with the low active group ( p , .05).
ATV_ O2 and HRmax were significantly higher only in the
highly active group (p , .001 and p , .05, respectively)
(Table 1).
HRV components are shown in Tables 2 and 3. Resting
RR was significantly longer in the highly active group
versus the low active group ( p , .05). RR decreased
between rest and 30% Wmax, then decreased further at 60%
Wmax in all groups ( p , .001). There was no significant
difference between groups concerning RR at 30% or at 60%
Wmax (Table 2). Figure 1 illustrates RR alterations induced
by exercise in a highly active subject.
587
Table 2. RR and Spectral Indexes of Heart Rate Variability in Low,
Moderately, and Highly Active Groups
RR (ms)
HF (ms2)
LF (ms2)
TP (ms2)
LF/HF
Low
1
828 6 25* 125.9 6 50.4 642.5 6 208.8 768.4 6 255.1 5.9 6 0.9
1 vs 2
à
2
563 6 16
7.9 6 1.5
19.9 6 4.8
27.7 6 5.7
2.8 6 0.6
2 vs 3
§
§
§
3
456 6 10
3.2 6 0.7
2.7 6 0.6
5.9 6 0.9
1.3 6 0.4
1 vs 3
{
{
{
{
{
Moderate
1
900 6 42
1 vs 2
2
642 6 27
2 vs 3
§
3
525 6 17
1 vs 3
{
151.6 6 32.6 514.2 6 147.2 655.8 6 177.6 3.3 6 0.4
25.1 6 8.4
51.0 6 15.9
76.1 6 19.5 3.2 6 1.2
§
§
8.4 6 2.7
9.8 6 3.4
18.2 6 5.2
1.8 6 0.6
{
{
{
High
1
921 6 42* 193.0 6 59.9 524.4 6 95.4
1 vs 2
2
605 6 11
12.9 6 2.4
38.4 6 8.2
2 vs 3
§
§
§
3
472 6 16
2.3 6 1.8
3.1 6 0.6
1 vs 3
{
{
{
717.4 6 150.5 4.4 6 1.2
51.3 6 9.4
3.0 6 0.5
§
5.4 6 0.9
1.7 6 0.2
{
Notes: Condition 1, at rest; Condition 2, during exercise at 30% Wmax;
Condition 3, at 60% Wmax.
* p , .05 between low and highly active groups.
p , .01 between rest and 30% Wmax.
à
p , .001.
§
p , .001 between 30% and 60% Wmax.
{
p , .001 between rest and 60% Wmax.
HF ¼ high frequency; LF ¼ low frequency; TP ¼ total power; LF/HF ¼
low frequency/high frequency.
At rest in a seated position, LF peak was centered at
approximately 0.07 Hz, and HF peak was approximately 0.25
Hz. During exercise, LF peak was less sharp. However, no
changes occurred in central frequency. On the contrary, HF
became a well-defined peak and its central frequency increased ( p , .01). Figure 2 illustrates the typical spectral data
alterations obtained by exercise in the same highly active
subject.
No significant difference was observed between groups at
rest or during exercise, concerning all spectral HRV indexes
(Tables 2 and 3).
During exercise, all the absolute spectral HRV indexes
decreased from rest to 30% Wmax and from rest to 60%
Wmax in the 3 groups ( p , .001). From 30% to 60% Wmax,
LF (p , .001) and TP (p , .01) decreased in the 3 groups
while HF decreased only in the highly active one ( p , .001).
Table 3. Indexes of Heart Rate Variability in Normalized Units in Low, Moderately, and Highly Active Groups
Low
LF NU (%)
HF NU (%)
Moderate
High
1
2
3
1
2
3
1
2
3
83 6 2*
17 6 2*
68 6 3 32 6 3 47 6 6
53 6 6
75 6 3
25 6 3
67 6 5
33 6 5
54 6 6
46 6 6
76 6 3
24 6 3
72 6 3
28 6 3
57 6 4
43 6 4
Notes: Condition 1, at rest; Condition 2, during exercise at 30% Wmax; Condition 3, at 60% Wmax.
* p , .001 between rest and 60% Wmax.
p , .01 between 30% and 60% Wmax.
LF ¼ low frequency; HF ¼ high frequency; NU ¼ normalized units.
588
RELAND ET AL.
Figure 2. Spectrum heart rate variability obtained in the same highly active
subject as in Figure 1 at rest in a seated position (top), during exercise at 30%
Wmax (maximum exercise power, in watts) (middle), and at 60% Wmax
(bottom).
Figure 1. RR series obtained in a highly active subject at rest in a seated
position (top), during exercise at 30% Wmax (maximum exercise power, in
watts) (middle), and at 60% Wmax (bottom).
The LF/HF ratio (Table 2), between rest and both 30%
Wmax and 60% Wmax, only decreased in the low active
group (p , .01 and p , .001, respectively).
Concerning normalized spectral HRV values, between
rest and 30% Wmax, no significant changes were observed
in any group. From rest to 60% Wmax and from 30% to
60% Wmax, HF NU increased only in low active subjects
(p , .001 and p , .01, respectively). Consequently, the
results are inverted concerning LF NU.
No significant correlation was observed between resting
HRV indexes and V_ O2max or ATV_ O2. At 30% Wmax, LF
and TP were positively correlated with V_ O2max (r ¼ .41,
p , .05, and r ¼ .38, p , .05, respectively), and TP was
correlated with ATV_ O2 (r ¼ .38, p , .05).
DISCUSSION
The main findings of the present study performed in
postmenopausal women are that (a) the level of physical
activity influences HR at rest, without any alteration in HRV
indexes, and (b) at 30% and 60% Wmax, HR is not influenced by the level of physical activity, contrary to sinus
node response to autonomic regulation.
It is widely assumed that regular exercise training induces
a decrease in resting HR (2,5). This decreased HR can be
observed both in young and middle-aged women (14). The
origin of the training bradycardia is probably multifactorial,
with a decrease of intrinsic heart rate (27) and/or an
alteration of HRV (2,5). In women, resting HRV declines
with aging in sedentary as in highly trained endurance
populations (2). However, transversal study demonstrates
that HRV is higher in trained women than in their sedentary
peers, regardless of age (2). In older athletes, as in young
women, training effects on HRV results mainly from an
increased parasympathetic tone (2,5). LF seems also
increased in endurance-trained young and older women
(2,28), but the underlying mechanisms still need to be
clarified. Our results confirm the impact of the level of
physical activity on resting HR in older women. Nevertheless, no significant difference in the resting HRV indexes
has been clearly demonstrated between our populations.
HEART RATE VARIABILITY OF OLDER WOMEN
Moreover no correlation has been observed between
V_ O2max and resting HRV indexes. In the present study,
the highest level of physical activity is accompanied by the
highest V_ O2max values. However, the physical activity
level performed in the highly active group should be not
sufficient to induce significant alterations to resting HRV.
Thus, the lower resting heart rate in our highly active
population may be due to a lower intrinsic heart rate as
previously suggested (27).
Some stress conditions such as physical exercise alter
HRV (12–14). Relatively little and inconsistent information is reported on the effect of the level of physical
activity on HRV in older people during exercise. In our
study and regardless of group, exercise decreased HF.
Classically, the HF peak, which reflects respiratory sinus
arrhythmia, corresponds to parasympathetic drive (25).
This result confirms the withdrawal of the vagal outflow to
the heart with increasing levels of dynamic exercise
already shown by using selective parasympathetic blockade in healthy adult men (29). In low and moderately
active subjects, HF decreased only between rest and 30%
Wmax, with no further changes. However, in the highly
active group, an alteration of sinus node response still
persisted with the increase in exercise intensity between
30% Wmax and 60% Wmax, as shown by a further
decrease in HF. Our results are in agreement with the
study of Tulppo and colleagues (15) performed on young
and older male subjects. Moreover, in young active males,
although HF is reduced during exercise, it shows a further
reduction after atropine, suggesting some degree of
persistent vagal drive (30). The administration of atropine
causes an increase in resting HR in young subjects, an
effect that is blunted in the elderly (31). It suggests
a decrease with aging of vagal influence on the heart (32).
The gradual vagal withdrawal with exercise intensity
observed in our highly active subjects supports the
suggestion that regular physical activity compensates this
ageing process. We have found that quite a high level of
physical activity is necessary to obtain a significant effect
on vagal tone. Indeed, no alteration of cardiac vagal
function during exercise has been noted in the moderately
active group, despite a relatively high level of V_ O2max. In
a recent study (16), Perini and colleagues found that 8
weeks of intense aerobic training in elderly men and
women increased V_ O2max without any change in resting
or exercise HRV. However, in their population, V_ O2max
was quite similar only to that of our moderately active
subjects in which we observed similar results concerning
exercise HRV. A high level of physical activity may
therefore be required to elicit enough adaptations in the
cardiovascular system and to modify the cardiovascular
risk marker represented by HRV. Moreover, an experimental study has shown that a vagal activity that limits the
deleterious effects of sympathetic activity (33), i.e., cardiac
arrhythmia, may have a cardioprotective effect during
exercise. Thus, high cardiorespiratory fitness may induce
such an effect by increasing the sinus node response to the
cardiac vagal tone during exercise. No significant
difference was observed between our populations concerning LF alteration during exercise. LF power decreased
589
progressively with exercise intensity, which confirms
previous results obtained in young and elderly people
(12,26). The LF peak depends on responses to both
sympathetic and parasympathetic tone alterations including
baroreflex responses (12,25). This decrease in LF power
despite the well-known increase in sympathetic activity
could be explained both by a sympathetic overstimulation
of sinus node receptors, which prevents a variation of their
response (34), and by the inhibition of the baroreflex
control of HR (12). Concerning LF/HF ratio exercise time
course, controversial results are presented in the literature
(13, 26). We observed a decrease in this marker of
sympathovagal balance (25) only in low active subjects
during exercise. In the same way, we noted a weak but
significant positive correlation between LF at 30%
V_ O2max. Given their high resting LF/HF level (Table
2), these results suggest an early saturation of sinus node
receptors, primarily of sympathetic nervous receptors, in
this sedentary population.
LF and HF, expressed in NU, also emphasize the
controlled and balanced behavior of the two branches of
the autonomic nervous system (25). In our study, whatever
the population, HF NU is maintained between rest and
30% Wmax and increased between rest and 60% Wmax in
low active subjects. These results are in agreement with
previous works performed on sedentary young and elderly
people (12,26). At rest, the contribution of nonneural
components in respiratory sinus arrhythmia is negligible.
During exercise, an increasing proportion of respiratory
sinus arrhythmia is due to nonneural mechanism (35).
Thus during exercise, HF NU tends to overestimate the
parasympathetic component. This increase in HF NU may
be due partly to a mechanical stimulus linked to the
increase in respiratory activity (35,36). This result is in
accordance with a specific ventilatory mode in sedentary
subjects, as previously suggested (16). Consequently, LF
NU is maintained between rest and 30% Wmax and
decreased between rest and 60% Wmax, which confirmed
previous findings (12,26). This evolution, which is affected
by the exercise intensity, could be explained by the
remaining baroreflex responses only at the lower level
(36).
Some potential limitations in our study should be
considered. First, as the menstrual cycle can alter HRV
indexes (37), only postmenopausal female subjects were
included in this study. Some of the participants were
receiving hormone replacement therapy. The role of this
therapy in HRV is controversial (2,38), and our study
cannot be used in that discussion. A similar ratio of
women receiving hormone replacement therapy, however,
was included in each group. Thus all groups would have
been similarly affected by potential influences of hormone
replacement therapy on HRV. Moreover, individual data
of the subjects receiving hormonal therapy were dispersed
within the group. Second, our study design cannot rule out
the possibility of a genetics bias (39). Indeed, we cannot
totally exclude a possible genetic effect on resting heart
rate (i.e., intrinsic heart rate) (39). Third, in this study, we
have chosen to compare the 3 groups at the same relative
level of intensity. It would be interesting to investigate the
590
RELAND ET AL.
effects of training on HRV responses at the same absolute
levels of exertion. At last, short data acquisition presents
some limits. Indeed, the VLF band, which is defined as
below 0.04 Hz and corresponds to the variations of
thermoregulation and humoral factors (25), cannot be
analyzed, whereas humoral factors are probably important
to adapt heart rate during constant load exercise (40).
Conclusion
In postmenopausal women, the effects of the levels of
physical activity on HR and HRV are dissociated. Quite
a high level of physical training induces resting bradycardia
without any changes in HRV indexes. In contrast, during
exercise the assessment of HRV indexes appears to be more
meaningful than at rest. Indeed, a sufficient level of physical
training induces a higher sensitivity of sinus node response
to exercise stress, which could play a role in thwarting
cardiovascular accidents in older women.
ACKNOWLEDGMENTS
We thank the medical and technical staff of the Centre Cardio-Pneumologique and of the Unité de Biologie et de Médecine du Sport (Rennes), the
volunteers for their generous cooperation with our project, and David James
for proofreading and rewriting the manuscript.
Address correspondence to Sylvia Reland, Unité de Biologie et
Médecine du Sport, CHU Pontchaillou, 2 rue Henri Le Guilloux, 35000
Rennes, France. E-mail: [email protected]
REFERENCES
1. Akelrod S. Spectral analysis of fluctuations in cardiovascular
parameters: a quantitative tool for the investigations of autonomic
control. TIPS. 1996;9:6–9.
2. Davy KP, Desouzat CA, Jonest PP, Seals RD. Elevated heart rate
variability in physically active young and older women. Clin Sci. 1998;
94:579–584.
3. Fleg JL, O’Connor F, Gerstenblith G, et al. Impact of age on the cardiovascular response to dynamic upright exercise in healthy men and
women. J Appl Physiol. 1995;78:890–900.
4. Tasaki H, Serita T, Irita A, et al. A 15-year longitudinal follow up-study
of heart rate and heart rate variability in healthy elderly persons.
J Gerontol Med Sci. 2000;55A:M744–M749.
5. Levy WC, Cerqueira MD, Harp GD, et al. Effect of endurance training
on heart rate variability at rest in healthy young and older men. Am
J Cardiol. 1998;82:1236–1241.
6. Oida E, Kannagi T, Yamori Y. Aging alteration of cardiac vagosympathetic balance assessed through the tone-entropy analysis. J Gerontol
Med Sci. 1999;54A:M219–M224.
7. Tsuji H, Vendetti F, Manders E, et al. Reduced heart rate variability and
mortality risk in an elderly cohort. Circulation. 1994;90:878–883.
8. Eaker ED, Chesebro JH, Sacks FM, Wenger NK, Whisnant JP,
Winston M. Cardiovascular disease in women. Circulation.
1993;88:1999–2009.
9. Westerterp KR, Meijer EP. Physical activity and parameters of aging:
a physiological perspective. J Gerontol Med Sci. 2001;56A:M7–M12.
10. Schuit AJ, Van Amelsvoort LGPM, Verheij TC, et al. Exercise training
and heart rate variability in older people. Med Sci Sports Exerc. 1999;3:
816–821.
11. Ueno LM, Hamada T, Moritani T. Cardiac autonomic nervous system
activities and cardiorespiratory fitness in older men. J Gerontol Med
Sci. 2002;57A:M605–M610.
12. Casadei B, Cochrane S, Johnston J, Conway J, Sleight P. Pitfalls in the
interpretation of spectral analysis of the heart rate variability during
exercise in humans. Acta Physiol Scand. 1995;153:125–131.
13. Myslivecek PR, Brown CA, Wolfe LA. Effects of physical conditioning on cardiac autonomic function in healthy middle-aged women. Can
J Appl Physiol. 2002;27:1–18.
14. Gregoire J, Tuck S, Yamamoto Y, Hugson L. Hear rate variability at
rest and exercise: influence of age, gender, and physical training. Can
J Appl Physiol. 1996;21:455–470.
15. Tulppo MP, Makikallio TH, Seppanen T, Laukkanen RT, Huikuri HV.
Vagal modulation of heart rate during exercise: effects of age and
physical fitness. Am J Physiol. 1998;274:H242–H429.
16. Perini R, Fisher NM, Veicsteinas A, Pendergast DR. Aerobic training
and cardiovascular responses at rest and during exercise in older men
and women. Med Sci Sports Exerc. 2002;34:700–708.
17. Ainsworth BE, Haskell WL, Leon AS, et al. Compendium of physical
activities: classification of energy costs of human physical activities.
Med Sci Sport Exerc. 1993;25:71–80.
18. Pate RR, Pratt M, Blair SN, et al. Physical activity and public health:
a recommendation from the Centers for Disease Control and Prevention
and the American College of Sports Medicine. JAMA. 1995;273:402–
407.
19. Paffenbarger RS, Hyde RT, Wing AL, Steinmetz CH. A natural history
of athleticism and cardiovascular health. JAMA. 1984;252:491–495.
20. Wasserman K, Hansen JE, Sue DY, Casaburi R, Whipp BR. Principles
of Exercise Testing and Interpretation Including Pathophysiology and
Clinical Applications. 3rd Ed. Baltimore: Williams & Wilkins; 1999.
21. Durnin JV, Rahaman MM. The assessment of the amount of fat in the
human body. Br J Nutr. 1967;2:681–689.
22. Fabre C, Massé-Biron J, Ahmaidi S, Adam B, Prefaut C. Effectiveness
of individualized aerobic training at the ventilatory threshold in the
elderly. J Gerontol Biol Sci. 1997;52A:B260–B266.
23. Gritzali F. Towards a generalized scheme for QRS detection in ECG
waveforms. Signal Process. 1988;15:183–192.
24. Bruggeman D, Andersen D. Heart rate variability: how to exclude
extrasystoles from the analysis. IEEE Comp Cardiol. 1993;467–470.
25. Task force of the European Society of Cardiology and the North
American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurements, physiological interpretation, and
clinical use. Eur Heart J. 1996;17:354–381.
26. Perini R, Milesi S, Fisher NM, Pendergast DR, Veicsteinas A. Heart
rate variability during dynamic exercise in elderly males and female.
Eur J Appl Physiol. 2000;82:8–15.
27. Stein R, Medeiros CM, Rosito GA, Zimerman AI, Ribeiro JP. Intrinsic
sinus and atrioventricular node electrophysiologic adaptations in
endurance athletes. JACC. 2002;39:1033–1038.
28. Davy KP, Miniclier NL, Taylor JA, Stevenson ED, Seals DR. Elevated
heart rate variability in physically active postmenopausal women: a cardioprotective effect? Am J Physiol. 1996;271:H455–H460.
29. Warren JH, Jaffe RS, Wraa CE, Stebbins CL. Effect of autonomic
blockade on power spectrum of heart rate variability during exercise.
Am J Physiol. 1997;273:R495–R502.
30. Polanczyk CA, Rohde LEP, Moares RS, Ferlin EL, Leite C, Ribeiro JP.
Sympathetic nervous system representation in time and frequency
domain indices of heart rate variability. Eur J Appl Physiol. 1998;79:
69–73.
31. Dauchot P, Gravenstein JS. Effects of atropine on the electrocardiogram
in different age groups. Clin Pharmacol Ther. 1971;12:247–280.
32. Seals DR, Taylor JA, Ng AV, Esler MD. Exercise and aging: autonomic
control of the circulation. Med Sci Sports Exerc. 1994;5:568–576.
33. Vanoli E, De Ferrari GM, Stramba-Badiale M, Hull SS, Foreman RD,
Schwartz PJ. Vagal stimulation and prevention of sudden death in
conscious dogs with a healed myocardial infarction. Cir Res. 1991;68:
1471–1481.
34. Malik M, Camm AJ. Components of heart rate variability—what they
really mean and what we really measure. Am J Cardiol. 1993;72:821–
822.
35. Casadei B, Moon J, Johnston J, Caiazza A, Sleight P. Is respiratory
sinus arrhythmia a good index of cardiac vagal tone in exercise? J Appl
Physiol. 1996;81:556–564.
36. Bernardi L, Salvucci F, Suardi R, et al. Evidence for an intrinsic
mechanism regulating heart rate variability in the transplanted and the
intact heart during submaximal exercise? Cardiovasc Res.
1990;24:969–981.
HEART RATE VARIABILITY OF OLDER WOMEN
37. Sato N, Myake S, Akatu J, Kuo CD, Kumashiro M. Power spectral
analysis of heart rate variability in healthy young women during the
normal menstrual cycle. Psychosom Med. 1995;57:331–335.
38. Yildiri A, Kabakci G, Yarali H, et al. Effects of hormone replacement
therapy on heart rate variability in post menopausal women. Ann
Noninvasive Electrocardiol. 2001;6:280–284.
39. Busjahn A, Voss A, Knoblauch H, et al. Angiotensin-converting
enzyme and angiotensinogen gene polymorphisms and heart rate variability in twins. Am J Cardiol. 1998;81:755–760.
591
40. Kamath MV, Fallen EL, McKelvie R. Effects of steady state exercise
on the power spectrum of heart rate variability. Med Sci Sports Exerc.
1991;23:428–434.
Received December 9, 2002
Accepted March 31, 2003
Decision Editor: George E. Taffet, MD