Effects of Chronic Estrogen Replacement Therapy on Beat-to-Beat Blood Pressure Dynamics in Healthy Postmenopausal Women
Abstract Recent data showing gender differences in autonomic control of heart rate and acute estrogen effects on vasodilatation suggest that estrogen may influence autonomic regulation of heart rate and blood pressure. We aimed to determine the effect of postmenopausal estrogen replacement therapy on autonomic control of beat-to-beat heart rate and blood pressure dynamics. Subjects included 20 healthy postmenopausal women aged 60 to 75 years with normal exercise tolerance tests, 10 of whom were taking oral estrogen for 13±3 (±SEM) years. Six healthy premenopausal women were also studied. Continuous electrocardiographic and noninvasive radial artery blood pressure measurements and intermittent forearm blood flow recordings (by venous-occlusion plethysmography) were obtained before and after a 20-minute, 60° head-up tilt and a 420-kcal meal during periods of spontaneous and metronomic breathing (at 0.25 Hz). Low-frequency (0.01- to 0.15-Hz) and high-frequency (0.15- to 0.50-Hz) heart rate and blood pressure spectral powers were computed with a fast Fourier transform. Cardiovascular and heart rate spectral power responses to upright tilt and meal digestion were the same in postmenopausal estrogen users and nonusers. However, during spontaneous breathing the blood pressure spectral power responses to upright tilt and meal ingestion were significantly different between the two groups of women. The low-frequency systolic pressure power response to upright tilt was smaller in estrogen users than nonusers (P=.01). After meal ingestion nonusers had an early postprandial fall (20 to 30 minutes after the meal) and late rise (50 to 60 minutes) in low-frequency systolic and diastolic pressure powers, which were significantly attenuated in estrogen users (P<.02). Healthy premenopausal women had a response similar to that of postmenopausal estrogen users. Estrogen may attenuate the low-frequency vasomotor response to posture change and meal digestion in healthy postmenopausal women. This effect of estrogen may represent damping of vasomotor instability after menopause.
Postmenopausal estrogen administration appears to have an important protective effect against the development of cardiovascular disease in women1 which may be mediated in part through an increase in high-density lipoprotein and decrease in low-density lipoprotein cholesterol.2 Recent studies suggest that estrogen may also have a direct effect on vascular function. Functional estrogen receptors have been found in human vascular smooth muscle.3 In both the tail artery of rats4 and coronary5 or forearm6 arterial circulation of humans, acute intravenous administration of estradiol results in vasodilatation. Also, estrogen acutely reverses abnormal acetylcholine-induced vasoconstriction in atherosclerotic coronary arteries.5 7 Short-term estrogen replacement therapy has been shown to improve endothelium-dependent vasodilation in the brachial arteries of hypercholesterolemic postmenopausal women.8 These data suggest that estrogen plays an important role in vascular physiology. However, little is known about the long-term effects of oral estrogen on beat-to-beat vasomotor activity or its relationship to the autonomic regulation of cardiovascular function.
The purpose of this study was to determine the effect of postmenopausal estrogen replacement therapy on autonomic control of beat-to-beat cardiovascular dynamics in response to two standardized physiological conditions: head-up tilt and meal digestion. These conditions were chosen because they represent common daily activities that are often associated with impaired autonomic responses in healthy elderly individuals.9 10
A total of 25 healthy postmenopausal women between the ages of 60 and 75 years were recruited from the local community through newspaper advertisements and the Harvard Cooperative Program on Aging subject registry. All of these potential subjects underwent a detailed history, physical examination, electrocardiogram (ECG), blood screen (including lipoprotein profile), and exercise stress test to assure they were in excellent health without evidence of coronary artery disease, hypertension, hyperlipidemia, or other diseases or risk factors possibly affecting cardiovascular function. None of the subjects smoked cigarettes, drank alcohol, or took cardioactive medications other than estrogen. All subjects were free of hot flashes or other postmenopausal symptoms.
One potential subject was excluded because of evidence of a previous silent myocardial infarction on her ECG. Eight women (six taking estrogen replacement and two not) had positive or equivocal exercise ECGs and were referred for exercise thallium scans. Four of these women had normal scans and were included in the study. Four were ultimately excluded because they did not undergo the thallium scan.
The final sample included 20 healthy women, 10 of whom were taking estrogen replacement therapy for at least 6 months and 10 who were not. Four women on estrogen were also taking progesterone. Subject characteristics are summarized in Table 1⇓.
After the results from the 20 postmenopausal women were analyzed, a comparison group of 6 healthy premenopausal women aged 21 to 30 years was recruited from the local community. Except for exercise stress tests these subjects were screened in the same fashion as the older women. The study was approved by the Institutional Review Board of the Hebrew Rehabilitation Center for Aged, and all subjects provided written informed consent.
The protocol consisted of a 60° head-up tilt test followed by a standardized liquid meal. Data were collected during intervals of spontaneous or metronomic breathing before and during the tilt and meal studies. During metronomic breathing subjects breathed for 10 minutes in time to tape-recorded cues at a rate of 15 breaths per minute (0.25 Hz).
All subjects were studied between 7 and 11 am after an overnight fast. After their arrival at the clinical research laboratory subjects rested supine on a tilt table, with their feet against a foot rest and waist secured by a loose belt around the table. The transducer from a noninvasive continuous blood pressure (BP) recording device (Colin Electronics Inc) was placed over the radial artery on the right wrist for continuous BP measurement. The right arm was kept at the level of the right atrium on an adjustable support throughout the study. Bipolar electrodes were placed on the chest for continuous recording of the ECG. A strain-gauge and upper arm and wrist pneumatic cuffs were attached to the left arm for arterial blood flow measurements with the use of venous-occlusion plethysmography.11 Forearm vascular resistance was calculated as the ratio of mean arterial BP to forearm blood flow.
After a 20-minute equilibration period the tilt table was inclined to 60° head-up, where it remained for 20 minutes. BP and heart rate (HR) were measured continuously for spectral analysis (see below), and forearm blood flow was measured at 3 and 13 minutes during the tilt.
After the tilt study, subjects rested quietly in the supine position for 20 minutes while basal measurements were obtained. Then subjects were served a standardized liquid meal containing 74% carbohydrate, 6% fat, 20% protein, and 420 kcal, served at 23°C. They sat up to consume the meal over a 10-minute period and then returned to the supine position for another 70 minutes while BP, HR, and forearm blood flow measurements were obtained. Ambient room temperature was maintained constant at 22±2°C throughout each study.
Spectral Analysis of HR and BP Variability
Autonomic control of HR and BP during spontaneous and metronomic breathing in the pretilt, tilt, premeal, and postmeal segments of the study was assessed with spectral analysis techniques. Continuous ECG and BP data were digitized at 250 Hz, and then each heartbeat was annotated with an automated arrhythmia detection algorithm. Each beat annotation was verified by visual inspection. If ectopic beats were present, they were removed and normal beats added with the use of linear interpolation. After proper calibration of a BP signal, a peak-trough detection algorithm was run on the BP waveform to obtain instantaneous systolic and diastolic pressures. For each segment of the study 8-minute sections of stationary continuous ECG and BP data were selected for analysis. Continuous BP data could not be obtained from one estrogen user because of previous wrist surgery that prevented the radial artery tonometer from detecting the pressure wave.
HR was calculated as the reciprocal of the RR interval. Instantaneous, uniformly sampled HR and BP values during each 8-minute segment were obtained by resampling each time series at 2 Hz. Then each time series was analyzed with a fast Fourier transform algorithm yielding a 512-point power spectrum. Total spectral power, an index of overall HR or BP variability, was computed (in arbitrary units) for the entire 0.01- to 0.50-Hz frequency spectrum. Low-frequency power, a measure of baroreflex control of HR or vasomotor activity, was calculated for the 0.01- to 0.15-Hz frequency band. High-frequency power, representing respiratory (vagally mediated) modulation of HR and BP, was determined for the 0.15- to 0.50-Hz band at each time segment.
Changes in all cardiovascular variables and spectral indexes in response to the tilt and meal were compared between women taking and not taking estrogen with the use of two-factor (treatment and time) repeated-measures ANOVA. Tilt and meal responses were also analyzed separately during periods of spontaneous and metronomic breathing. When significant differences in response were found, post hoc analyses were performed with Student’s t tests for determination of the time points at which these differences occurred. BP spectral responses to meal digestion were compared with those of healthy premenopausal women with the use of a similar two-factor (group and time) repeated-measures ANOVA with post hoc Student’s t tests. Statistical significance was set at an α level of .05 and adjusted for multiple comparisons. All data are expressed as mean±SEM.
Baseline Subject Characteristics
Women taking estrogen replacement therapy had been exposed to oral estrogens for an average of 13±3 years. These subjects were otherwise similar to untreated postmenopausal women in age, height, weight, body mass index, low- and high-density lipoprotein levels, basal BP, HR, and resting HR standard deviation during spontaneous breathing (Table 1⇑). Although forearm blood flow tended to be higher and forearm vascular resistance lower in estrogen users, this difference was not statistically significant.
Spectral Analysis of BP Variability
Basal supine systolic and diastolic spectral powers were similar between estrogen users and nonusers under both spontaneous and metronomic breathing conditions. However, during spontaneous breathing the low-frequency BP power responses to upright tilt and meal digestion were significantly different between the two groups of women (Tables 2⇓ and 3⇓).
During tilt, low- and high-frequency systolic BP power increased significantly under both spontaneous (P≤.0002) and metronomic (P=.01) breathing conditions. However, upright tilt resulted in a significantly smaller increment in low-frequency systolic BP power in estrogen users compared with nonusers during spontaneous breathing (P=.01, Table 2⇑).
Diastolic BP power also increased during tilt in all frequency bands under spontaneous breathing conditions (P≤.0005) and only in the low-frequency band with metronomic breathing (P=.04). There was no difference between estrogen users and nonusers in the diastolic BP power response.
Meal digestion resulted in significant changes in BP power only during spontaneous breathing conditions (Table 3⇑). There was an early withdrawal and late enhancement of low-frequency systolic and diastolic BP power in the untreated group that was significantly attenuated in estrogen users (Fig 1⇓, treatment×meal interaction: P=.006 for systolic power and P=.02 for diastolic power). By 50 to 60 minutes after the meal, women not taking estrogen demonstrated high-amplitude low-frequency Mayer wave–like oscillations in both systolic and diastolic BP, which were much less prominent in estrogen-treated women (see example in Fig 2⇓).
Healthy premenopausal women had a BP spectral power response to the meal that was similar to that of postmenopausal estrogen users. They also had no evidence of enhanced low-frequency Mayer wave–like BP oscillations after the meal.
Cardiovascular and HR Spectral Responses to Tilt and Meal Studies
The mean cardiovascular responses to upright tilt and meal digestion were the same in women with and without estrogen replacement therapy. BP did not change significantly in either study, but HR increased significantly (P=.0001) after both the tilt and meal stimuli. During upright tilt forearm vascular resistance increased significantly (P=.0001) and to a similar extent in both groups of women. There was no statistically significant postprandial change in forearm vascular resistance in either group of women. The results were similar during the spontaneous and metronomic breathing phases of each study. The sinus rhythm HR spectral powers for low- and high-frequency bands were the same in estrogen users and nonusers for each study condition.
To our knowledge this is the first human study to examine the effect of chronic oral estrogen replacement therapy on autonomic control of beat-to-beat HR and BP dynamics under several standardized physiological conditions. Our findings suggest that estrogen attenuates the amplitude of low-frequency BP oscillations in postmenopausal women. These oscillations, called Mayer waves, represent rhythmic changes in vasomotor tone that are thought to be mediated by the sympathetic control loop of the baroreflex.12
The effect of estrogen in reducing arterial BP oscillations may occur at any of several points along the baroreflex arc. Given the known presence of estrogen receptors in vascular smooth muscle3 and the recently described inhibitory effect of 17β-estradiol on calcium influx in vascular smooth muscle cells,13 estrogen may inhibit the vascular response to baroreflex-mediated sympathetic activation. It is also possible that the favorable effect of estrogen on circulating lipoproteins may retard the development of atherosclerosis and preserve vascular compliance, thereby preventing large beat-to-beat fluctuations in BP. However, in the present study women with and without estrogen therapy had similar serum lipoprotein levels.
It is well established that arterial baroreceptor denervation increases BP variability in both animals and humans.14 Since aging reduces baroreceptor sensitivity,15 it is not surprising that untreated postmenopausal women show exaggerated BP oscillations in response to sympathetic nervous system activation. The observed damping of BP oscillations in estrogen-treated women suggests that estrogen might act by increasing baroreflex gain, thereby improving beat-to-beat control of vasomotor activity.
Only indirect evidence from studies of gender differences in baroreflex activity suggests that estrogen may play a role in autonomic control of cardiovascular function in humans. Women have a blunted HR response to bolus injections of phenylephrine16 and upright tilt17 compared with men with similar age and BP. Muscle sympathetic nerve activity at rest is also lower in healthy women compared with men.18 Furthermore, we have shown a larger ratio of high- to low-frequency HR variability (suggesting relatively greater respiratory than baroreflex control of HR) in women compared with men across the entire adult age spectrum.19 Our current data suggest that estrogen plays a greater role in the beat-to-beat regulation of BP than HR. Therefore, previous observations of reduced HR and sympathetic nerve responses in women compared with men may be due to the damping effect of estrogen on short-term BP fluctuations and consequent diminution of reflex HR and sympathetic nervous system responses.
The apparent inverse relationship between estrogen treatment and beat-to-beat vasomotor activity is biologically plausible, given the potent effects of estrogen replacement on reducing symptoms of vasomotor instability (hot flashes) during menopause. However, the time scale of hot flashes is much longer, on the order of minutes, compared with the 10-second BP cycles observed in the present study.
The sample size of the present study was small, thus limiting our ability to detect relatively small differences in hemodynamic and autonomic variables between women with and without estrogen treatment. In addition, our cross-sectional study design did not permit intraindividual comparisons of cardiovascular dynamics with and without estrogen treatment. It is possible that women were placed on estrogen for undetermined reasons that could have affected our results. However, our subjects were well matched on all baseline characteristics other than estrogen use. Although a prospective randomized controlled trial of estrogen therapy would have been desirable, we could not justify the expense of such a trial without this preliminary study.
Despite these limitations, this study addresses an important question regarding the role of estrogen in the beat-to-beat autonomic regulation of cardiovascular function. Our findings suggest that estrogen may attenuate the low-frequency vasomotor response to posture change and meal digestion in healthy postmenopausal women. This effect may represent a damping of vasomotor instability associated with menopause. Additional studies would be worthwhile to evaluate the effect of estrogen on vasomotor regulation in individuals with existing cardiovascular disease and impairments in autonomic control of BP.
This work was supported by the Hebrew Rehabilitation Center for Aged and Public Health Service awards AGO4390, AGO9538, and AGO8812 from the National Institute on Aging, Bethesda, Md. Dr Lipsitz holds the Irving and Edyth S. Usen and Family Chair in Geriatric Medicine at the Hebrew Rehabilitation Center for Aged. The authors wish to thank Roberta Rosenberg, MS, coordinator of the Harvard Cooperative Program on Aging, for her expert assistance with subject recruitment. We also thank Dr Ary Goldberger for reviewing and commenting on the manuscript and Dr Ernest Gervino for performing the exercise tolerance tests.
- Received April 11, 1995.
- Revision received May 15, 1995.
- Accepted July 18, 1995.
Bush TL, Barrett-Connor E, Cowan LD, Criqui MH, Wallace RB, Suchindran CM, Tyroler HA, Rifkind BM. Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the Lipid Research Clinics program follow-up study. Circulation. 1987;75:1102-1109.
Karas RH, Patterson BL, Mendelsohn ME. Human vascular smooth muscle cells contain functional estrogen receptor. Circulation. 1994;89:1943-1950.
Shan J, Resnick LM, Liu QY, We XC, Barbagallo M, Pang PKT. Vascular effects of 17β-estradiol in male Sprague-Dawley rats. Am J Physiol. 1994;35:H967-H973.
Reis SE, Gloth ST, Blumenthal RS, Resar JR, Zacur HA, Gerstenblith G, Brinker JA. Ethinyl estradiol acutely attenuates abnormal coronary vasomotor responses to acetylcholine in postmenopausal women. Circulation. 1994;89:52-60.
Gilligan DM, Badar DM, Panza JA, Quyyumi AA, Cannon RO III. Acute vascular effects of estrogen in postmenopausal women. Circulation. 1994;90:786-791.
Lipsitz LA, Mietus J, Moody GB, Goldberger AL. Spectral characteristics of heart rate variability before and during postural tilt: relations to aging and risk of syncope. Circulation. 1990;81:1803-1818.
Lipsitz LA, Ryan SM, Parker JA, Freeman R, Wei JY, Goldberger AL. Hemodynamic and autonomic nervous system responses to mixed meal ingestion in healthy young and old subjects, and dysautonomic patients with postprandial hypotension. Circulation. 1993;87:391-400.
Whitney RJ. The measurement of volume changes in human limbs. J Physiol (Lond). 1953;121:1-27.
Pagani M, Lombardi F, Guzzetti S, Rimoldi O, Furlan R, Pizzinelli P, Sandrone G, Malfatto G, Dell’Orto S, Piccaluga E, Turiel M, Baselli G, Cerutti S, Malliani A. Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympatho-vagal interaction in man and conscious dog. Circ Res. 1986;59:178-193.
Han XZ, Karaki H, Ouchi Y, Akishita M, Orimo H. 17β-estradiol inhibits Ca2+ influx and Ca2+ release induced by thromboxane A2 in porcine coronary artery. Circulation. 1995;91:2619-2626.
Wagner CD, Persson PB. Two ranges in blood pressure power spectrum with different 1/f characteristics. Am J Physiol. 1994;267:H449-H454.
Gribbin B, Pickering TG, Sleight P, Peto R. Effect of age and high blood pressure on baroreflex sensitivity in man. Circ Res. 1971;29:424-431.
Abdel-Rahman ARA, Merrill RH, Wooles WR. Gender-related differences in the baroreceptor reflex control of heart rate in normotensive humans. J Appl Physiol. 1994;77:606-613.
Ng AV, Callister R, Johnson DG, Seals DR. Age and gender influence muscle sympathetic nerve activity at rest in healthy humans. Hypertension. 1993;21:498-503.