Sex-Related Differences in Autonomic Modulation of Heart Rate in Middle-aged Subjects
Background Women have worse outcomes when they experience acute myocardial infarction (MI), but the reasons for this sex-related difference are not well understood. Because cardiovascular neural regulation plays an important role in cardiac mortality, we studied possible sex-related differences in the autonomic modulation of heart rate (HR) in middle-aged subjects without known heart disease.
Methods and Results Baroreflex sensitivity (BRS) and HR variability were studied in randomly selected, age-matched populations of middle-aged women (n=186; mean age, 50±6 years) and men (n=188; mean age, 50±6 years) without hypertension, diabetes, or clinical or echocardiographic evidence of heart disease. BRS measured from the overshoot phase of the Valsalva maneuver was significantly lower in women (8.0±4.6 ms/mm Hg, n=152) than in men (10.5±4.6 ms/mm Hg, n=151) (P<.001), and the low-frequency component of HR variability measured from ECG recordings also was lower in women (P<.001), whereas the high-frequency component was higher in women than in men (P<.001). The ratio between the low- and high-frequency oscillations also was lower in the women (P<.001). The increase of HR and decrease of high-frequency component of HR variability in response to an upright posture were smaller in magnitude in women than in men (P<.01 for both). After adjustment for differences in the baseline variables, such as blood pressure, HR, smoking, alcohol consumption, and psychosocial score, the sex-related differences in BRS and HR variability still remained significant (P<.001 for all). Women with estrogen replacement therapy (n=46) had significantly higher BRS and total HR variance than the age-matched women without hormone treatment (P<.01 for both), and the BRS and HR variability of the women with estrogen therapy did not differ from those of the age-matched men.
Conclusions Baroreflex responsiveness is attenuated in middle-aged women compared with men, but the tonic vagal modulation of HR is augmented. Hormone replacement therapy appears to have favorable effects on the cardiovascular autonomic regulation in postmenopausal women.
Previous studies have shown that the short-term prognosis for women who have experienced an acute myocardial infarction (MI) is worse than that for men,1 but reasons for this difference are not well understood.
Experimental and clinical studies demonstrate that cardiovascular autonomic regulation plays an important role in cardiac mortality.2 3 4 The clinically applicable methods for assessing this regulation are measurement of heart rate (HR) variability and evaluation of baroreflex sensitivity (BRS).3 4 5 HR variability reflects primarily tonic autonomic modulation, whereas the baroreflex-mediated response of HR to changes in arterial blood pressure indicates the capacity of reflex autonomic modulation. Low HR variability and BRS are related to increased risk of cardiac mortality,2 3 4 and experimental studies indicate that when measured before acute MI, impaired reflex vagal activity increases the risk of mortality after the occurrence of MI but not the tonic activity.6 7 We compared possible sex-related differences in BRS and HR variability in randomly selected, age-matched population samples of middle-aged subjects.
Originally, 300 women and 300 men between 40 and 60 years of age were selected at random from the Social Insurance Register, which covers all the inhabitants of the Oulu district. All those on antihypertensive or cardiac medication were excluded if their names were filed in the reimbursement register maintained by the Social Insurance Institute. Eventually, 266 women and 258 men volunteered to take part, and they filled out a standardized health questionnaire on their past medical history, cardiac symptoms and medication used, smoking habits, alcohol consumption, physical activity,8 and psychosocial type.9 A standard 12-lead ECG was recorded, laboratory tests were conducted, and echocardiography was performed. Subjects with symptoms of angina pectoris, cardiac medication, ECG abnormalities, diabetes, any other medical disease requiring continuous medication, echocardiographic evidence of heart disease, or atrial fibrillation were excluded. Standardized Holter ECG recordings in the supine and sitting postures and assessment of BRS by the Valsalva maneuver then were performed, after which those subjects with inadequate Holter recordings because of premature beats or technical artifacts were excluded, and the analyses of HR variability were conducted on 186 women (mean age, 50±6 years) and 188 men (mean age, 50±6 years). After further exclusion of subjects with an inadequate blood pressure response during the Valsalva maneuver, BRS could be reliably assessed in 152 women and in 151 men. A Hewlett-Packard 77020A ultrasound system was used for M-mode, two-dimensional, and Doppler echocardiographic examinations of each subject, all performed by the same experienced cardiologist.
Analysis of HR Variability
Each subject was monitored for 30 minutes with an ambulatory ECG recorder for 15 minutes while lying quietly and breathing normally and for 15 minutes after assuming a sitting posture. The ECG data were transferred from the Del Mar Avionics ECG scanner (model 500) to a microcomputer for the analysis of HRV with the use of a custom-made program described previously in detail.10 Premature beats and noise were excluded both automatically and manually, and only segments with >90% qualified beats were included in the analysis. An autoregressive model was used to estimate the power spectrum densities of two frequency bands: low-frequency (LF) power (0.04 to 0.15 Hz) and high-frequency (HF) power (0.15 to 0.4 Hz). The spectral components were calculated as absolute units and normalized units as described previously.10 The standard deviation of RR intervals and the percent of successive RR intervals >50 ms (pNN50) were used as time domain measures. The spectral and time domain measures were analyzed from segments of 512 consecutive beats.
The subjects performed the Valsalva maneuver in the sitting position by blowing into a rubber tube connected to an aneroid manometer and maintaining a pressure of 40 mm Hg for 15 seconds. The test was performed three times at 5-minute intervals and the data collected and analyzed as described previously in detail.5 Noninvasive arterial pressure was measured on a beat-to-beat basis with the use of the Finapres finger-cuff method. The slope of the linear relationship between the length of the RR interval (in milliseconds) and the preceding systolic blood pressure value (in mm Hg) was calculated with the use of a linear least mean squares fitting method. The baroreflex slope was determined in a time window ranging from the beat when the systolic blood pressure exceeded that at the end of the Valsalva strain to the beat after the maximum systolic pressure overshoot. Only regression lines with a correlation coefficient >.8 and with a blood pressure change >15 mm Hg were accepted for analysis.
The differences in continuous variables between the women and men were assessed with the use of the Mann-Whitney two-sample test. Categorical variables were compared with the use of the χ2 test. Significant differences in baseline variables between women and men were taken into account by means of ANCOVA after logarithmic transformation of the data that were not normally distributed.
Blood pressure was higher in the men than in the women (systolic blood pressure, 146±19 mm Hg versus 139±20 mm Hg, P<.001; diastolic blood pressure, 89±11 mm Hg versus 82±12 mm Hg, P<.001), and there were more smokers and drinkers among the men (P<.05 for both). The leisure time physical activity score did not differ, but the Framingham psychosocial score was lower among men than women (26.1±5.4 versus 27.8±5.1, P<.001). There were no other significant differences in the baseline variables between women and men.
The BRS and HR variability results are shown in Table 1⇓. The mean BRS was significantly lower in women than in men. Similarly, the LF power of HR variability analyzed as normalized units was lower in women, but the HF power and pNN50 were higher in women. The ratio between LF and HF components was significantly lower in women, as was the standard deviation of RR intervals. When sex-related responses to upright posture were compared, the women had an attenuated increase in HR (P<.01) and a smaller decrease in the HF component (P<.01).
After adjustment for differences in mean HR, blood pressure, smoking, alcohol consumption, and the psychosocial score, the sex-related differences in BRS (F=22.8, P<.001) and standard deviation of RR intervals (F=6.7, P<.01) became even more marked and the differences in the HF and LF components still remained significant (P<.001 for both).
When BRS and measures of HR variability were compared between age-matched postmenopausal women with and without estrogen replacement therapy (Table 2⇓), the BRS and standard deviation of RR intervals were significantly higher in women who were on hormone therapy.
The results indicate that there are sex-related differences in the cardiovascular autonomic regulation. The HR response to an abrupt rise in blood pressure and LF modulation of HR are lower in women than in men, whereas HF modulation of HR is higher in women. The HR response to the overshoot phase of the Valsalva maneuver describes the reflex vagal activity in response to a rapid increase of arterial blood pressure,5 and the LF component of HR variability is determined by spontaneous oscillation of blood pressure, also reflecting the baroreflex-mediated control of HR. These observations show that the baroreflex responsiveness is attenuated, whereas the tonic vagal activity is augmented in women compared with men. Women also had smaller HR and HF changes in response to an upright posture, suggesting that the vagal responses to cardiovascular unloading are attenuated in women. Similar sex-related differences in BRS have been described previously in a smaller sample of young women and men.11 Blood pressure and some lifestyle variables potentially influencing the cardiac autonomic control differed between women and men, but the sex-related differences in BRS and HR variability still remained significant after adjustment for these, showing that the observed differences are related to the sex of the subject. BRS and total HR variance were higher in postmenopausal women who were on estrogen replacement therapy compared with those without hormone therapy, suggesting that hormonal factors may partly explain the sex-related differences in autonomic modulation of HR.
Previous studies have observed similar sex-related differences in HR variability,12 13 14 but no previous population-based attempt has been made to evaluate BRS in healthy subjects. The present data show that in addition to sex-related differences, there are wide interindividual variations in the reflex response to a rise in blood pressure. Most notably, the values for BRS were lower (<3 ms/mm Hg) in some healthy subjects than those reported in high-risk patients after MI.6 MI results in further reduction of BRS,6 and it is obvious that subjects with low sensitivity before the occurrence of MI will have a very low BRS after MI.
The overshoot phase of the Valsalva maneuver, when the arterial blood pressure rises quickly, represents a physiological challenge for the baroreflex. Its validity and reproducibility in assessment of BRS have been previously described,5 but the inadequate increase of blood pressure after the release of Valsalva strain limits its use in a small proportion of the subjects. Phenylephrine testing has been used more commonly in previous smaller studies for determining BRS, but the invasive character of this test limits its application in population studies.
Experimental and clinical studies have convincingly demonstrated that impaired BRS and reduced HR variability increase the risk of cardiac mortality,2 3 4 6 and it has been demonstrated recently that reduced BRS and LF spectral component of HR variability but not a reduced HF component are associated with an increased risk of life-threatening arrhythmias after MI.15 16 Experimental studies have shown that both BRS and HR variability measured after MI predict the occurrence of post-MI ischemia-induced ventricular fibrillation,6 7 but at variance with BRS, HR variability does not predict post-MI ischemia-induced ventricular fibrillation when measured before MI.6 7 Therefore, it is possible that women who have a lower baroreflex responsiveness before the occurrence of an acute ischemic event will be at a higher risk of mortality during or after such an event.
This work was supported by grants from the Finnish Foundation for Cardiovascular Research, Helsinki, Finland, and the Medical Council of the Academy of Finland, Helsinki.
- Received January 3, 1996.
- Revision received May 5, 1996.
- Accepted May 14, 1996.
- Copyright © 1996 by American Heart Association
Greenland P, Reicher-Reiss H, Goldbourt U, Behar S, and Israeli SPRINT Investigators. In-hospital and 1-year mortality in 1524 women after myocardial infarction: comparison with 4315 men. Circulation. 1991;83:484-491.
Schwartz PJ, Billman GE, Stone HL. Autonomic mechanisms in ventricular fibrillation induced by myocardial ischemia during exercise in dogs with a healed infarction: an experimental preparation for sudden cardiac death. Circulation. 1984;69:780-790.
Tsuji H, Venditti F Jr, Manders ES, Evans JC, Larson MG, Feldman CL, Levy D. Reduced heart rate variability and mortality risk in an elderly cohort. Circulation. 1994;90:878-883.
Airaksinen KEJ, Hartikainen JEK, Niemela¨ MJ, Huikuri HV, Mussalo HM, Tahvanainen KUO. Valsalva manoeuvre in the assessment of baroreflex sensitivity in patients with coronary artery disease. Eur Heart J. 1993;14:1519-1523.
Schwartz PJ, Vanoli E, Stramba-Badiale M, DeFerrari GM, Billman GE, Foreman RD. Autonomic mechanisms and sudden death: new insight from analysis of baroreceptor reflexes in conscious dogs with and without myocardial infarction. Circulation. 1988;78:969-979.
Grimby G. Physical activity and muscle training in the elderly. Acta Med Scand. 1992;711:213-237.
Haynes SG, Levine S, Scotch N, Feinleib M, Kannel WB. The relationship of psychosocial factors to coronary heart disease in the Framingham study, I: methods and risk factors. Am J Epidemiol.
Huikuri HV, Niemela¨ MJ, Ojala S, Rantala A, Ika¨heimo MJ, Airaksinen KEJ. Circadian rhythms of frequency domain measures of heart rate variability in healthy subjects and patients with coronary artery disease: effects of arousal and upright posture. Circulation. 1994;90:121-126.
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.
Mølgaard H, Hermansen K, Bjerregaard P. Spectral components of short-term RR interval variability in healthy subjects and effects of risk factors. Eur Heart J. 1994;15:1174-1183.
Bigger JT Jr, Fleiss JL, Steinman RC, Rolnitzky LM, Schneider WJ, Stein PK. RR variability in healthy, middle-aged persons compared with patients with chronic coronary heart disease or recent acute myocardial infarction. Circulation. 1995;91:1936-1943.
Hohnloser SH, Klingenheben T, van de Loo A, Hablawetz E, Just H, Schwartz PJ. Reflex versus tonic vagal activity as a prognostic parameter in patients with sustained ventricular tachycardia or ventricular fibrillation. Circulation. 1994;90:1068-1073.