influence of exercise training on heart rate variability in post-menopausal women with elevated...

Influence of exercise training on heart rate variabilityin post-menopausal women with elevated

arterial blood pressure

K. P. Davy, W. L. Willis and D. R. SealsHuman Cardiovascular Research Laboratory, Departments of Kinesiology and Medicine

(Cardiology and Geriatric Medicine), University of Colorado, Boulder, CO, USA

(Received 25 February 1996; accepted August 1996)

Summary. Low heart rate variability (HRV) has been reported to be an independentrisk factor for the development of coronary heart disease in women and has recentlybeen identified as a risk factor for cardiac sudden death and all-cause mortality. We haverecently demonstrated that endurance-trained post-menopausal women demonstratehigher levels of HRV than sedentary control subjects. The purpose of the present studywas to test the hypothesis that 12 weeks of regular aerobic exercise would increase HRVin sedentary post-menopausal women with elevated arterial blood pressure (BP) (i.e.either high normal BP or stage I hypertension). A secondary aim was to test thehypothesis that the increase in HRV with exercise training, if observed, would beassociated with an increase in spontaneous cardiac baroreflex sensitivity (SBRS), animportant physiological determinant of HRV. To accomplish these aims, we studiedeight sedentary post-menopausal women (age554?561?3 years) before and after 12weeks of aerobic exercise training (3?360?3 days per week at 70%62% of maximal heartrate for 4363 min per day). Maximal oxygen uptake and body weight did not change(P.0?05) with training, but percentage fat (35?562?6% vs. 34?562?3%, P,0?05) de-creased and treadmill time to exhaustion increased (9?860?5 vs. 11?360?5 min, P,0?05).Supine resting levels of heart rate, RR interval and the standard deviation of the RRinterval (time domain measure of HRV) were unchanged (all P.0?05) from baselinelevels after 12 weeks of aerobic training. Similarly, the high-frequency, low-frequencyand total power of HRV (frequency domain measures) were also unchanged frombaseline (all P.0?05). SBRS was also not different before and after aerobic exercisetraining (1062 vs. 1363 ms mmHg21 respectively, P.0?05). In contrast, systolic anddiastolic BP were reduced approximately 8 and approximately 5 mmHg with training(both P,0?05) respectively. These results indicate that 12 weeks of moderate-intensityaerobic exercise training does not increase HRV or SBRS, despite producing a clinicallysignificant reduction in BP at rest in post-menopausal women with elevated BP. Con-

Correspondence: Kevin P. Davy, PhD, University of Colorado, Department of Kinesiology, Campus Box 354,Boulder, CO 80309, USA.

Clinical Physiology (1997) 17, 31–40

© 1997 Blackwell Science Ltd 31

sidered together with our previous findings in female master endurance athletes, thesefindings suggest that more intense and prolonged exercise training may be required toproduce increases in HRV and SBRS in sedentary post-menopausal women.

Key words: ageing, exercise training, heart rate variability, vagal tone.

Introduction

Coronary heart disease (CHD) prevalence increases with age in men and women(Kannel et al., 1976; Manolio et al., 1993). Women demonstrate the greatest increase inCHD prevalence with advancing age, with CHD accounting for 50% of all deaths inwomen over 50 years of age (Eaker et al., 1993; Manolio et al., 1993). The dramaticincrease in CHD-related risk with age in women appears to occur after the time ofmenopause caused, in part, by an elevation in arterial blood pressure (BP) (Eaker et al.,1993). Approximately 35–40% of women aged 55–65 years with CHD have beenreported to be disabled by their illnesses (Wenger, 1993). The prevalence of cardiacsudden death (CSD), which is often the first manifestation of CHD, also rises withadvancing age and the level of BP (i.e. the severity of hypertension) (Gilman et al., 1994). Low heart rate variability (HRV) has been reported to be an independent risk factorfor the development of CHD (Liao et al., 1995) in men and women and has recently beenidentified as a risk factor for CSD (Moolgard et al., 1991) and all-cause mortality (Tsujiet al., 1994). Low HRV is thought to reflect cardiac autonomic balance in which vagalmodulation of heart rate is low (Katona & Jih, 1975; Schwartz et al., 1992) and suscep-tibility to arrhythmia generation is elevated (Schwartz et al., 1992). Heart rate variabilitydecreases with age (Hellman & Stacy, 1976; Korkushko et al., 1991) and is reduced inelevated BP states (Novak et al., 1994) and, therefore, is thought to contribute to theage- and hypertension-related increase in ventricular arrhythmias and CSD (Gilman etal., 1994). We have recently shown (Davy et al., 1996) that endurance-trained post-menopausalwomen demonstrate higher levels of HRV compared with less active women. Theelevated HRV in these endurance athletes was associated with elevated levels of cardiacbaroreflex sensitivity, a key physiological determinant of cardiac vagal modulation ofheart rate in humans (Kollai et al., 1994). However, because these women were highlytrained and had other unique constitutional features, it is not possible to extrapolatethese findings to the possible effects of endurance training on HRV in sedentarypost-menopausal women. As such, it is not known whether regular aerobic exercise canincrease cardiac vagal modulation of heart rate, as reflected by elevations in both timeand frequency domain measures of HRV, in middle-aged and older adults with elevatedBP. Accordingly, the primary aim of the present investigation was to test the hypothesisthat a moderate programme of regular aerobic exercise would increase HRV in seden-tary post-menopausal women with elevated BP. A secondary aim was to test the

© 1997 Blackwell Science Ltd, Clinical Physiology, 17, 31–40

32 K. P. Davy et al.

hypothesis that the increase in HRV with exercise training, if observed, would beassociated with an increase in cardiac baroreflex sensitivity, an important physiologicaldeterminant of HRV (Kollai et al., 1994). To accomplish these aims, we measured time[standard deviation (SD) of the RR interval] and frequency (high- and low-frequencyand total power) domain measures of HRV and spontaneous cardiac baroreflex sensi-tivity (SBRS) before and after 12 weeks of aerobic exercise training in eight sedentarypost-menopausal women with high normal BP or stage I hypertension. BP was alsomeasured to determine whether regular aerobic exercise of this intensity and durationhad selective effects on HRV and BP in this population.

Subjects and methods

SUBJECTS

Eight sedentary post-menopausal women aged 50–60 years volunteered to participatein the present investigation. None of the women had participated in a programme ofregular exercise for at least one year. All of the women demonstrated high normal BP[systolic BP (SBP)=130–139 mmHg and/or diastolic BP (DBP)=85–89 mmHg (n=5)] orstage I hypertension [SBP=140–159 mmHg and/or DBP=90–99 mmHg (n=3)] on threeconsecutive occasions, as defined by the Joint National Committee on the Detection,Evaluation and Treatment of High Blood Pressure (1993). Three of the subjects werereceiving calcium channel blockers and two were receiving diuretic therapy. In addition,three of the women were on hormone replacement therapy. Medications remainedunchanged during the course of the study in all subjects. All of the women were free ofovert CHD, as indicated by physical examination and resting and exercise electrocar-diograms. The nature, purpose and risks of the study were explained to each subject beforewritten informed consent was obtained. The experimental protocol was approved by theHuman Research Committee at the University of Colorado at Boulder.

MEASUREMENTS AND PROTOCOL

HRV and BP. These measurements have been described previously (Davy et al., 1996).The RR interval was recorded from the electrocardiogram lead that produced the largestR wave amplitude, and BP was measured continuously using the finger photoplethys-mography technique (Finapres model 2300, Englewood, CO, USA). Arterial BP meas-ured using this technique has been shown to be a valid and reliable method for assessingbeat-by-beat changes in BP under the present study conditions (Parati et al., 1989). A pneumobelt was placed around the upper abdomen to measure respiratory excur-sions. Subjects were studied after a 12-h fast and 24–48h after their last exercise session.Each subject rested quietly for at least 15 min in the supine posture to obtain stablebaseline control levels of all haemodynamic variables. Beat-to-beat measurements of


Exercise training and heart rate variability 33

RR interval, BP and respiratory excursions were made during 5-min periods of con-trolled frequency breathing (15 breaths min21). Subjects inspired (2s) and expired (2s)in synchrony with a clock to achieve 15 breaths min21. Absolute casual BP levels were determined before and after the exercise trainingprogramme in the supine position by auscultation over the brachial artery using amercury sphygmomanometer according to published guidelines (Perloff et al., 1993 ).The first and fifth phase of the Korotkoff sounds were used to determine SBP and DBPrespectively.

MAXIMAL OXYGEN CONSUMPTION AND BODY FATNESS

Maximal oxygen uptake was measured during graded treadmill exercise to exhaustionbefore and after the 12-week period of regular aerobic exercise, as described in detailpreviously (Stevenson et al., 1994). Briefly, oxygen consumption, carbon dioxide pro-duction, pulmonary ventilation and the respiratory exchange ratio were determined byuse of on-line computer-assisted open circuit spirometry. Expired oxygen and carbondioxide gas fractions were measured using a Perkin-Elmer MGA-1100 mass spectrome-ter (Panoma, CA, USA) calibrated with gases of known concentrations. Pulmonaryventilation was determined using a flow turbine (VMM-2, Interface Associates, LagunaNiguel, CA, USA) calibrated with a 3-l syringe. The criteria used to ensure a validmaximal oxygen consumption have also been described (Stevenson et al., 1994). Bodyfat percentage was estimated from the sum of five skinfolds (Jackson et al., 1980). In ourlaboratory, similar levels of body fat percentage are obtained from skinfold measure-ments compared with hydrodensitometry in post-menopausal women (K. P. Davy et al.unpublished observations).

BEAT-TO-BEAT RR INTERVAL AND BP ANALYSES

The procedures for obtaining and analysing the RR interval for power spectral analysishave been described in detail by our laboratory previously (Davy et al., 1996). Theelectrocardiogram, beat-to-beat BP and respiratory signals were recorded on FM tapeand later digitized (CODAS, Dataq Instruments, Akron, OH, USA) at 1000 Hz with alaboratory computer. Heart rate, mean RR interval and the SD of the RR intervals (timedomain measure of HRV) were computed from the beat-to-beat RR intervals. Fre-quency domain measures of HRV were obtained using power spectral analysis based onthe Welch algorithm (Welch, 1967). Equidistant samples were obtained by interpolatingeach 300-s time series at 4Hz. The segments were detrended, Hanning filtered, and afast-Fourier transformation was used to estimate the frequency components. The spec-trum estimate was derived by averaging the periodograms. The spectral power at eachof the following frequency bandwidths was calculated: high frequency (0?15–0?40 Hz);low frequency (0?06–0?14 Hz); and total (0?01–0?40 Hz). We (K. P. Davy et al., unpub-lished observations) have found no significant differences in high frequency (13706277



vs. 12076445 ms2 Hz21), low frequency (27066847 vs. 283261283 ms2 Hz21) and totalpower (37516992 vs. 283261284 ms2 Hz21) of HRV, or time domain measures of HRV(data not shown), in eight middle-aged and older women studied in our laboratory ontwo different occasions. Similar stability over time has been reported by other investi-gators using this technique (Kleiger et al., 1991). Spontaneous cardiac baroreflex sensitivity was determined by the sequence method(Blaber et al., 1995), as described previously by our laboratory (Davy et al., 1996). Thesequence method has been shown to be a valid and reproducible method for determiningcardiac baroreflex sensitivity in humans (Parlow et al., 1995). Sequences of at least threeor more increasing or decreasing SBP that were accompanied by directionally similarchanges in the RR interval were considered to be baroreflex responses. A linearregression was performed on each of these sequences for each data segment. Theaverage of the individual linear regressions was used as the baroreflex slope for eachsubject and posture.

EXERCISE TRAINING PROGRAMME

After completing the initial round of testing, the eight subjects performed moderate-in-tensity exercise training for 12 weeks at home on their own. Subjects were instructed towalk 3–4 days per week for at least 30–45 min per day at an intensity corresponding to60–75% of their individually determined maximal heart rate. Compliance with theprogramme was documented using heart rate monitors (Polar High Performance HeartRate Monitor, Vantage XL, Polar CIC, Port Washington, NY, USA) and daily activitylogs. The monitors and activity logs were returned to the laboratory biweekly forassessment of compliance.

STATISTICAL ANALYSIS

A paired t-test was used to assess differences in the dependent variables before and after12 weeks of regular aerobic exercise. The significance level was set a priori at P,0?05.All data are presented as means6SE.

Results

The average age and height of the women in the present study was 54?561?3 years and1?6262?0 m respectively. There were no significant changes in body weight (75?763?9kg vs. 75?164?0 kg) or maximal oxygen uptake (23?261?1 vs. 24?261?1 ml kg–1 min–1)from pre- to post-exercise training in these women. However, body fat percentage(35?562?6% vs. 34?562?3%) and treadmill time to exhaustion (9?860?5 min vs. 11?360?5min) decreased and increased (both P,0?05) respectively. Compliance with the homeexercise programme was excellent; the subjects exercised 3?360?3 days per week for



4363 min per day at an intensity of 7062% (range=66–79%) of maximal heart rate. Thisintensity corresponded to 5963% (range=47–75%) of maximal oxygen consumption. Table 1 shows the time and frequency domain measures of HRV in the women fromthe present study. Supine resting levels of HRV were unchanged (P.0?05) after exercisetraining in these women. The slightly higher mean levels of the frequency domainmeasures of HRV after exercise training were the result of increases with exercisetraining in one subject (Fig.1). This individual did not differ from the other subjects inher baseline characteristics or with regard to her other responses or compliance to theexercise programme. Spontaneous cardiac baroreflex sensitivity was 1062 ms vs. 1362ms (P.0?05) at baseline compared with after exercise training, respectively, in thesewomen. Supine resting levels of SBP and DBP decreased (both P,0?05) from 13862 mmHgto 13064 mmHg and from 8862 mmHg to 8362 mmHg, respectively, after 12 weeks ofregular aerobic exercise in these women.

Discussion

The primary new finding from the present study was that a moderate programme ofaerobic exercise, which produced an increase in exercise tolerance and reductions in BP,failed to produce an increase in HRV or SBRS in sedentary post-menopausal womenwith elevated resting BP. In the present study, there were no significant changes in either the time or frequencydomain measures of HRV or SBRS after 12 weeks of exercise training in these previouslysedentary middle-aged and older women. One woman did demonstrate large increasesin the frequency domain measures of HRV. However, this subject was unremarkablefrom the other subjects in every way; thus, her post-training responses remain unex-plained. There are at least two possibilities, which might explain the apparent lack of

Table 1. Absolute levels of baseline and post-exercise train-ing time and frequency domain measures of heart ratevariability

Variable Baseline Post

HF (ms2 Hz21) 2946130 3616241LF (ms2 Hz21) 238647 2716108Total power (ms2 Hz21) 5096163 6166340Heart rate (beats min21) 6563 6362 Mean RR interval (ms) 928637 967637 SD of RR interval (ms) 2864 3166

Values are means6SE, n58 subjects. HF, high-frequencypower; LF, low-frequency power; SD of RR interval, stand-ard deviation of RR interval.



Fig. 1. Individual changes from baseline to post-exercise training in the high frequency, low frequency andtotal power of HRV in the women from the present study.



influence of exercise training on cardiac vagal modulation of heart rate in the remainingwomen. First, it is possible that regular aerobic exercise is a physiological stimulus thatcan increase HRV in sedentary post-menopausal women, but that the ‘dose’ of exercisetraining used in the present study was insufficient to elicit such changes. That is, theintensity, duration, frequency and/or length of training may have been below ‘threshold’.This could explain why we previously found elevated HRV in highly trained femaleendurance athletes, although other explanations are also possible (see below). Futurelongitudinal studies should consider employing a greater exercise training stimulus inorder to test this hypothesis in sedentary women. Second, it is possible that physical activity per se has little or no influence on cardiacvagal modulation of heart rate and, by definition, some other factor(s) must be impor-tant. For example, the endurance-trained women we studied previously (Davy et al.,1996; Stevenson et al., 1994) demonstrated lower levels of body weight and fat, andhigher levels of total blood volume and maximal oxygen consumption compared withsedentary women. In that study, the high-frequency power of HRV was significantlycorrelated with body weight (r=–0?71), body mass index (r=–0?65), total blood volume(r=0?72) and maximal oxygen consumption (r=0?73) (K. P. Davy, unpublished observa-tions). In this regard, Aronne et al. (1995) have reported increases and decreases in HRVwith hypocaloric dieting and overfeeding-induced changes in body weight, respectively,in humans. Thus, it is possible that in the present investigation the changes in body weightand/or composition or some other constitutional factor were insufficient to inducechanges in HRV. Future studies will be necessary to address the potential roles of suchfactors in determining HRV. Whether changes in maximal oxygen consumption andtotal blood volume are necessary before changes in HRV can be observed with exercisetraining also remains unclear. In the present study, decreases in BP were observed without concomitant improve-ments in HRV. This observation clearly indicates that the moderate-intensity exerciseprogramme employed was a sufficient stimulus for reducing BP, but was unable to evokeincreases in cardiac vagal modulation of heart rate. Thus, it appears that the training-re-lated mechanisms responsible for producing reductions in BP are different from thosethat determine HRV. Regular aerobic exercise has been shown to lower casual SBP and DBP levels byapproximately 10 mmHg under resting conditions (Hagberg et al., 1989; Martin et al.,1991; Seals & Reiling, 1991). In the present investigation, we observed reductions ofapproximately 8 mmHg and approximately 5 mmHg for systolic and diastolic BP,respectively, in women with high normal or stage I hypertension. Based on the resultsof several clinical trials (Coope & Warrender, 1986; National High Blood PressureEducation Program Working Group, 1994; The Systolic Hypertension in the ElderlyProgram Cooperative Research Group, 1993), this magnitude of reduction in BP wouldbe expected to have a significant effect in lowering CVD risk in these women. Impor-tantly, the decreases in BP in the present study appear to be largely the result of exercisetraining per se because only small, albeit significant, reductions in body fat occurred, and



no changes in body weight and maximal oxygen uptake were observed in the presentstudy. In conclusion, the results of the present study indicate that 12 weeks of moderate-in-tensity aerobic exercise training does not increase HRV or SBRS, despite producingclinically significant reductions in BP at rest in post-menopausal women with elevatedBP. Considered together with our previous findings in female master endurance athletes,these findings suggest that a greater exercise training stimulus and/or other physiologicalchanges (e.g. reductions in body weight and fat) may be required to produce increasesin HRV and SBRS in sedentary post-menopausal women.

Acknowledgments

We would like to thank Mary Jo Reiling and Heather Silverman for their technicalassistance during the study. This investigation was supported by an Andrus FoundationGrant AARPAF-OCG0937B and National Institutes of Health Grants HL-39966 andAG-06537. K.P.D. was supported by an Individual National Research Service AwardHL-08834 and a Special Emphasis Research Career Award AG-00687 from the NationalHeart, Lung and Blood Institute and National Institute on Aging, National Institutes ofHealth, Bethesda, MD, USA.

References

ARONNE L., MACKINTOSH R., ROSENBAUM M., LEIBEL R. & HIRSH J. (1995) Autonomic nervous system activityin weight gain and weight loss. Am J Physiol, 269, R222–R225.

BLABER A., YAMAMOTO Y. & HUGHSON R. (1995) Methodology of spontaneous baroreflex relationshipassessed by surrogate data analysis. Am J Physiol, 268, H1682–H1687.

COOPE J. R. & WARRENDER T. S. (1986) Randomized trial of treatment of hypertension in the elderly inprimary care. Br Med J, 293, 1145–1148.

DAVY K., MINICLIER N., TAYLOR J., STEVENSON E. & SEALS D. (1996) Elevated heart rate variability inphysically active postmenopausal women: A cardioprotective effect? Am J Physiol 271, H455–H460.

EAKER E. D., CHESEBRO J. H., SACKS F. M., WENGER N. K., WHISNANT J. P. & WINSTON M. (1993)Cardiovascular disease in women. Circulation, 88, 1999–2009.

GILMAN J., JALAL S. & NACCARELLI G. (1994) Predicting and preventing sudden death from cardiac causes.Circulation, 90, 1083–1092.

HAGBERG J. M., MONTAIN S. J., MARTIN W. H. & EHSANI A. A. (1989) Effect of exercise training in 60- to69-year-old persons with essential hypertension. Am J Cardiol, 64, 348–353.

HELLMAN J. & STACY R. (1976) Variation of respiratory sinus arrhythmia with age. J Appl Physiol, 41, 734–738.JACKSON A. S., POLLACK M. L. & WARD A. (1980) Generalized equations for predicting body density of

women. Med Sci Sport Exerc, 12, 175–182.JOINT NATIONAL COMMITTEE ON DETECTION, EVALUATION AND TREATMENT OF HIGH BLOOD PRESSURE. (1993)

The fifth report of the Joint National Committee on Detection, Evaluation and Treatment of High BloodPressure. Arch Int Med, 153, 154–183.

KANNEL W., HJORTLAND M. C., MCNAMARA P. M. & GORDON T. (1976) Menopause and risk of cardiovasculardisease: the Framingham Study. Ann Int Med, 85, 447–452.

KATONA P. & JIH F. (1975) Respiratory sinus arrhythmia: noninvasive measure of parasympathetic cardiaccontrol. J Appl Physiol, 39, 801–805.



KLEIGER R., BIGGER J., BOSNER M., CHUNG M., COOK J., ROLNITZKY L., STEINMAN R. & FLEISS J. (1991)Stability over time of variables measuring heart rate variability in normal subjects. Am J Cardiol, 68,626–630.

KOLLAI M., JOKKEL G., BONYHAY I., TOMCSANYI J. & NASZLADY A. (1994) Relation between baroreflexsensitivity and cardiac vagal tone in humans. Am J Physiol, 266, H21–H27.

KORKUSHKO O., SHATILO V., PLACHINDA Y. & SHATILO T. (1991) Autonomic control of cardiac chronotropicfunction in man as a function of age: assessment by power spectral analysis of heart rate variability. JAutonom Nerv Syst, 32, 191–198.

LIAO D., CAI J., ROSAMOND D., BARNES R. & HUTCHINSON R. (1995) Cardiac autonomic function and incidentcoronary heart disease – The ARIC Study. Circulation (suppl. I), 92, 418.

MANOLIO T. A., FURBERG C. D., SHEMANSKI L., PSATY B. M., O’LEARY D. H., TRACY R. P. & BUSH T. L.(1993) Associations of postmenopausal estrogen use with cardiovascular disease and its risk factors inolder women. Circulation, 88, 2163–2171.

MARTIN J. E., DUBBERT P. M. & CUSHMAN W. C. (1991) Controlled trial of aerobic exercise in hypertension.Circulation, 81, 1560–1567.

MOOLGARD H., SORENSEN K. & BJERREGAARD P. (1991) Attenuated 24-h heart rate variability in apparentlyhealthy subjects, subsequently suffering sudden cardiac death. Clin Autonom Res, 1, 233–237.

NATIONAL HIGH BLOOD PRESSURE EDUCATION PROGRAM WORKING GROUP. (1994) National high bloodpressure education program working group report on hypertension in the elderly. Hypertension, 23,275–285.

NOVAK V., NOVAK P., CHAMPLAIN J. & NADEAU R. (1994) Altered cardiorespiratory transfer in hypertension.Hypertension, 23, 104–113.

PARATI G., CASADEI R., GROPPELLI A., RIENZO M. D. & MANCIA G. (1989) Comparison of finger andintra-arterial blood pressure monitoring at rest and during laboratory testing. Hypertension, 13, 647–655.

PARLOW J., VIALE J., ANNAT G., HUGHSON R. & QUINTIN L. (1995) Spontaneous cardiac baroreflex in humans:Comparison with drug-induced responses. Hypertension, 25, 1058–1068.

PERLOFF D., GRIM C., FLACK J., FROHLICH E. D., HILL M., MCDONALD M. & MORGENSTERN B. Z. (1993)Human blood pressure determination by sphygmomanometry. Circulation, 88, 2460–2470.

SCHWARTZ P., ROVERE M. L. & VANOLI E. (1992) Autonomic nervous system and sudden cardiac death.Circulation, 85, I-77–I-91.

SEALS D. R. & REILING M. J. (1991) Effect of regular exercise on 24-hour arterial pressure in older hypertensivehumans. Hypertension, 18, 583–592.

STEVENSON E. T., DAVY K. P. & SEALS D. R. (1994) Maximal aerobic capacity and total blood volume in highlytrained middle-aged and older female endurance athletes. J Appl Physiol, 77, 1691–1696.

THE SYSTOLIC HYPERTENSION IN THE ELDERLY PROGRAM COOPERATIVE RESEARCH GROUP. (1993) Implica-tions of the systolic hypertension in the elderly program. Hypertension, 21, 335–343.

TSUJI H., VENDITTI F., MANDERS E., EVANS J., LARSON M., FELDMAN C. & LEVY D. (1994) Reduced heartrate variability and mortality risk in an elderly cohort. Circulation, 90, 878–883.

WELCH P. (1967) The use of fast Fourier transform for the estimation of power spectra: a method based ontime averaging over short, modified periodograms. IEEE Trans Audio Electroacoust, 15, 70–73.

WENGER N. (1993) Coronary heart disease in women: an overview (myths, misconceptions, and missedopportunities). In: Cardiovascular Health and Disease in Women (ed. Wenger, N.), LeJaq Communica-tions, Bethesda, MD, USA.



influence of exercise training on heart rate variability in post-menopausal women with elevated...

Documents