- Email: [email protected]
Plasma epinephrine concentration in healthy men: Correlation with systolic pressure and rate-pressure product
352
lACC Vol. 5. No.2 February 1985:352-6
REPORTS ON HYPERTENSION
Plasma Epinephrine Concentration in Healthy Men: Correlation With Systolic Pressure and Rate-Pressure Product LAWRENCE R. KRAKOFF, MD, FACC, STANLEY DZIEDZIC, PHD, MD, SAMUEL J. MANN, MD, KATHERINE FELTON, MS, KAREN YEAGER, RN New York. New York
Relations among plasma epinephrine, norepinephrine and renin activity and systolic pressure, diastolic pressure, heart rate and the product of heart rate and systolic pressure (rate-pressure product) were evaluated in 31 healthy men whose arterial pressure spanned the range from normal to mildly elevated. Measurements were made during 60 minutes with the patient in the supine position and during lO to minutes of quiet standing. In the supine position, highly significant regressions were found between systolic pressure or rate-pressure product and plasma epinephrine, but not between these variables and norepinephrine or renin activity. A weakly significant correlation was also found between heart rate and norepinephrine. On standing, norepinephrine and epineph-
The determinants of systolic and diastolic arterial pressures in the early phases of the hypertensive process remain to be characterized. This problem is compounded by difficulty in establishing an exact classification such that' 'normotensive" and "hypertensive" groups can be easily distinguished for comparison (1,2). Epidemiologic studies (3) have avoided such an arbitrary division and tend to rely on regression relations between arterial pressure as a continuous variable and subsequent rates of appearance of cardiovascular disease (3). On this basis, we evaluated a group of healthy, well characterized men with a range of arterial pressures varying between normal and that which is considered to be mildly hypertensive. Plasma norepinephrine, epinephrine concentration and renin activity were measured, and correlations examined with respect to systolic pressure, diastolic pressure, heart rate and heart rate X systolic pressure (rate-pressure product). Medicine. From the Hypertension Division. Mount Sinai School of Medicine, New York, York. New York. This study was supported by Grants HLB 13595 and HLB 22514 from the U.S. Public Health Service. Washington. D.C. 21. 1984; revised manuscript received August Manuscript received May 21, 20. 1984. 20, 1984, accepted September 5. 1984. Krakoff. MD. Hypertension DiviAddress for reprints: Lawrence R. Krakoff, Medicine. One Gustave L. Levy Place, Place. New sion. Mount Sinai School of Medicine, York. New York 10029. If') © 1985
by the American College of Cardiology
rine increased significantly. In this position, rate-pressure product was significantly related by regression analysis only with plasma epinephrine. Weakly significant correlations between systolic pressure and epinephrine and between heart rate and norepinephrine and epinephrine were also found. Plasma renin activity was not significantly correlated with arterial pressure, heart rate or rate-pressure product in either position. These results suggest that plasma epinephrine is a determinant of systolic pressure when postural reflexes are minimized and that epinephrine may participate in control of cardiac work load, as reflected by rate-pressure product in the absence of exercise or definable stress. (J Am Coli CardioI1985;5:352-6)
Methods Study subjects. The subjects for this study were 31 white or Hispanic men ranging in age from 18 to 55 years. years, who were selected for the protocol because their arterial pressures on outpatient examination spanned the normal to mildly elevated range. None of the subjects had taken antihypertensive medication for at least 6 months before the study. Complete medical history and physical examinations were normal except for the presence of borderline diastolic arterial pressure (90 to 100 mm Hg) on several occasions in 13 of the 31 subjects. When appropriate, screening examinations for causes of secondary hypertension were performed; all were normal. No subject had hyperglycemia, hyperlipidemia, hypokalemia, elevation of serum urea nitrogen or creatinine, proteinuria or electrocardiographic abnormalities other than increased voltage in the precordial leads. Half of these subjects engaged in regular recreational physical exercise such as tennis or jogging. In the 2 year follow-up period for these subjects since these studies were performed, some have been placed on antihypertensive drug therapy. No one in the group has had evidence of coronary heart disease, heart failure or any complication of hypertensive cardiovascular disease. All subjects were studied as outpatients and permitted to continue their usual diet. 07:15-1097/85/$3.30 0735-1097/85/$3.30
lACC Vol. 5. No.2 lACC Vol. No.2
KRAKOFI' KRAKOFF ET AL. EPINEPHRINE. BLOOD PRESSURE. RATE-PRESSURE PRODUCT
February 1985:352-6
Study protocol. On the day of the study, subjects were permitted a clear fluid breakfast. They had been requested not to smoke or drink caffeinated beverages from the previous evening at II :00 PM until arrival at the laboratory. On arrival, each subject was weighed and then placed in the supine position. The cuff on an Arteriosonde was placed on one arm, leads for an electrocardiogram and cardiac tachometer were taped in place on the chest and an indwelling catheter was inserted into an antecubital or forearm vein on the side opposite to the pressure cuff. For I hour, arterial pressure and heart rate were measured at 2 minute intervals while the subject remained supine. A blood sample for plasma catecholamines and renin activity was then taken. The subject then stood upright for 10 minutes until a second blood sample was taken. For each subject, the blood pressure and heart rate were averages of the last 10 determinations in the supine position and of 5 determinations during standing. Measurements. Plasma catecholamine concentrations, norepinephrine and epinephrine were determined in duplicate by radioenzymatic assay as previously described (4,5). (4,5). The coefficient of variation for interassay agreement is 9% 9% for norepinephrine and II % for epinephrine. Plasma renin activity was determined by radioimmunoassay of generated 5%. angiotensin I (6). (6). Coefficient of variation is 5%. Statistical analysis. Data were processed for statistical analysis by the SAS system of the CUNY computer center employing the GLM program for regression analysis. In each phase of the study, the analysis employed considered systolic or diastolic pressure and heart rate or rate-pressure product (systolic pressure x heart rate) (7) as the dependent variables for correlation in a multiple linear regression model with plasma norepinephrine, plasma epinephrine and plasma renin activity or their logarithms as independent variables (for example, systolic pressure = f30 + f31 . plasma norepinephrine + f32 . plasma epinephrine + f3, . plasma renin activity). The overall relation was considered significant for
less than or equal to 0.0 I. Specific correlation coefficients (r) for individual relations between the dependent and independent variables were also calculated. l.tF LtF
Results Supine measurements. Descriptive characteristics of the 31 men who participated in this study are given in Table I. Although six were mildly overweight (body mass index 25 to 29), 29), none was obese. The range of diastolic arterial pressure in the supine position varied from normal to mild elevation (defined as 90 to 104 mm Hg). The low average supine heart rate may reflect the degree of physical fitness of this group in agreement with the frequent history of regular exercise. The distribution of supine systolic and diastolic pressures is shown in Figure I. Measurements on standing. As indicated in Table I, diastolic pressure and heart rate increased significantly during standing. Measurements of plasma norepinephrine and epinephrine concentration in the two positions are shown in Figure 2. Standing significantly increased both plasma catecholamines. In the supine position plasma renin activity was 1.4 ± 0.2 ng/ml ng/ml per h. A significant increase (+0.8 ± 0.3 ng/ml ng/ml per h, p < 0.02 by paired t test) occurred during standing. Correlations among blood pressure, heart rate, plasma catecholamines and renin activity. Table 2 presents the statistical analysis for relations among the dependent and independent variables that were analyzed. In the supine position, a highly significant relation was found between systolic pressure or rate-pressure product and independent histograms of systolic (a) Figure 1. Frequency histograms (a) and diastolic (b) (b) pressures in in 31 31 subjects in in the the supine position. position. The midpoint midpoint of of pressures each interval is is given on on the the abscissa, abscissa, the the number number (n) (n) of of subjects on the ordinate. DISTRIBUTION OF ARTERIAL PRESSURE
12 12
a Supine systoliC pressure
10 10
of Study Group (n (n Table 1. Clinical Characteristics of
31 )) 31
Mean ±± SE Age (yr) Height (m) Weight (kg) Body mass index (kg/m") Supine Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Heart rate (beats/min) Standing Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Heart rate (beats/min)
8
Range
N N
6 4
36 36 1.77 74.4 23.9
± ± II ± 0.0\ 0.01 ±
± 1.3 \.3 ± 0.4
IX 1.67 167 60.1 19.9 19.9
to to to to
55 55 1.87 I.X7 89.3 28.9 28.9
± 2 117 ± 2 79 79 ± 2 2 59 ± ± 2
9X to 147 64 to 100 35 to 84 X4
122 \22 ± 2 89 2* 89 ± 2* 71 ± 7\ ± 2*
ISO 103 to 150 61 to 114 44 99 44 to 99
test. SE *p < 0.0 I compared with supine value by paired I tes!. error.
~ standard =
353
2
0 '----"'9""5-1""0'"--5-----'I""I5'---""'12"-5----'1'"'35"---"'14""5o 95 105 115 125 135 145 16 16
b Supine diastolic pressure
14 14 12 12 10 10 N N
8 6 4 2
o'---......-"""-------''''''----''''''-0 65 65
75 75
85 85
95 95
Hg) PRESSURE (mm Hgl
354
KRAKOFF ET AL. EPINEPHRINE, BLOOD PRESSURE, RATE-PRESSURE PRODUCT
No.2 lACC Vol. 5, NO.2 February 1985:352-6 February
POSTURE AND CATECHOLAMINES
SUPINE PLASMA EPINEPHRINE AND SYSTOLIC ARTERIAL PRESSURE
600
150 r - - - - - - - - - - - - - - - - - - - - ,
•
500 :x:'" 140
400
E
ci. a.
E E
UJ 0:: ll:
300
130
•
;:)
en en 120 UJ
200
•
0:: ll:
a.
u ~ ...J ....J
._.__.-x ._.----1:
100
o
•
•
•
•
•
• • • • • • I • • • •
en >- 100 en
variables, Highly significant individual correlations between systolic pressure or rate-pressure product and plasma epinephrine were also found, For the log-transformed independent variables, systolic pressure was significantly cor0.001) only with epinephrine (Fig. related (F = 8.18, P < 0,001) 3). Neither diastolic pressure nor heart rate was significantly related to the three independent variables in the supine position by multiple regression analysis. A weakly significant correlation was evident between heart rate alone and plasma norepinephrine. norepinephrine, Table 2 also displays the statistical relations for the standing position. Only rate-pressure product is significantly related to independent variables in the model employed. A highly significant correlation was found between rate-
•
•
9 i
,1
5 10
II
20
1I
I
40
80
0.0001).
pressure product and plasma epinephrine. Weakly significant individual correlations were observed between systolic pressure and plasma epinephrine and between heart rate and plasma norepinephrine or epinephrine, epinephrine. Correlation with age and weight. In the supine position, neither age nor weight was significantly correlated with either plasma catecholamine concentration or plasma renin activity. During standing, there was a significant positive correlation between age and plasma epinephrine (r = 0.45, p < 0.05), but not between age and norepinephrine or renin activity. Weight was not significantly correlated with standactivity, ing plasma catecholamine concentration or renin activity.
F (model) F
NE
E
PRA
0.26 0.31 0.22 0.29
0.65* 0.15 0.44t 0.65
<0.1 <0.1 <0.1 <0.1
Supine 6.96* 0.41 2.30 6.85* Standing 1.04
0.30 1.62 5.44*
160
Figure 3. Relation between systolic arterial pressure and plasma epinephrine in 31 subjects in the supine position (r = 0.68, p =
Correlation Coefficients (r)
Diastolic pressure Heart rate Rate-pressure product
.-J
PLASMA EPINEPHRINE (pg/mll (pg/ml)
Table 2. Statistical Analysis
Systolic pressure
• •
STAND
Figure 2. Mean plasma norepinephrine (NE) and epinephrine (E) concentrations in supine and standing positions. Results are expressed as mean ± standard error. Significant increases occurred in both plasma catecholamine concentrations on standing.
Systolic pressure Diastolic pressure Heart rate Rate-pressure product
••
•
I-
.---- E SUPINE
110
0
P
•
• •
0.12
0.22 0.36t 0.36t
0.37t
<0.1 0.42t 0.56*
0.20
<0.1 <0.1 0.16
*p < 0.01; to.OI to.OI < p < 0.05. The F value is given for the entire statistical model (see Methods), whereas the r value is shown for individual correlations. E = plasma epinephrine; NE = plasma norepinephrine; PRA = plasma renin activity.
lACC Vo!. Vol. 5. No.2 JACC
KRAKOFF ET AL. EPINEPHRINE. BLOOD PRESSURE. RATE-PRESSURE PRODUCT
February 1985:352-6
Discussion Plasma catecholamines and sympathoadrenal function. Plasma norepinephrine and epinephrine are measures of sympathoadrenal function. Various stimuli cause significant increases in plasma catecholamine concentration (8-10). (8- 10). Increased sympathetic tone is necessary for the maintenance of arterial pressure in response to upright posture; in normal subjects standing causes a doubling of plasma norepinephrine. Patients with idiopathic orthostatic hypotension have no postural increase in the concentration of this catecholamine (8,11). Many previous studies have explored possible relations between sympathetic function, as reflected by plasma norepinephrine, and essential hypertension. Although there is a slight trend toward elevation of plasma norepinephrine in hypertensive compared with control groups, when many studies are pooled interpretation is difficult because of variation in control groups and categorization of the hypertensive subjects as to etiology of disease or cardiac status (12). Furthermore, plasma norepinephrine reflects spillover from that which is released from nerve endings and escapes neural re-uptake or local metabolism (13). Unless sympathetic function is activated by a potent stimulus, plasma norepinephrine remains far below its apparent physiologic threshold for effect as a circulating hormone (14). Plasma epinephrine in the normal and hypertensive subject. Circulating epinephrine reflects secretion from the adrenal medullae. Supine plasma epinephrine concentration is normally in a remarkably low range; the mean concentration is about half the apparent physiologic threshold for effects on systolic pressure and heart rate as determined by infusion of exogenous epinephrine (15). Infusion studies may overestimate the threshold for endogenous epinephrine unless the preinfusion plasma level of the amine is far below threshold concentration (that is, such that adrenergic receptors are unoccupied before infusion). It is then possible that plasma epinephrine concentration has some physiologic effect at levels found in the circulation in unstressed states. (16,17) 6,17) have reported elevations Previous publications (J of plasma epinephrine concentration in patients with established hypertension compared with normal subjects. High levels of plasma epinephrine concentration have also been found in "borderline hypertensives," but no correlation (18,19). with arterial pressure has been described (l8.J 9). More recently, it has been reported (20) that supine plasma epinephrine concentration was significantly higher in a small number (n = 9) of subjects with "high normal" arterial pressure compared with subjects with a "low normal" pressure drawn from a large healthy working population in Western Australia. However, the only significant regression relation was between systolic pressure and plasma epinephrine in the "low" group. Amann et at. aL (21) reported that in patients with essential hypertension, plasma epinephrine
355
concentration is positively correlated with alpha receptormediated vasoconstriction in the forearm as revealed by blockade with intraarterial prazosin. In normotensive subjects, the correlation is not observed. Results of this study. Our study evaluated well characterized normal volunteers and patients referred for borderline arterial pressure or uncomplicated mild hypertension. Although this group may be viewed as selected and not representing an unbiased sample of the population at large, it is free of factors that might alter plasma catecholamine concentration independent of changes in arterial pressure, such as coronary heart disease or diabetes (22). Our results indicate that plasma epinephrine concentration is highly correlated with systolic arterial pressure in the supine position when reflexes mediating postural adjustment are less active. During both supine and upright posture, the heart rate x systolic pressure product correlated positively with plasma epinephrine but not norepinephrine. In contrast, diastolic pressure was not significantly correlated with plasma catecholamine concentration or renin activity as an independent variable in either the supine or the standing position. Role of circulating epinephrine in control of blood pressure. These results do not establish the extent to which circulating epinephrine itself controls systolic arterial pressure or rate-pressure product. It may be that plasma epinephrine reflects global sympathoadrenal tone to a greater extent than does plasma norepinephrine, which spills into the circulation from the large number of sympathetic nerve terminals as well as the adrenal medullae. The specific role of circulating epinephrine need not be limited to direct effects on cardiac beta-receptors increasing heart rate and contractility or vascular beta2-receptors causing vasodilation. Recent studies (23) suggested that circulating epinephrine may be taken up and secreted by adrenergic neurons as a "co-transmitter" or participate in presynaptic regulation of adrenergic neurosecretion. Implications. Systolic arterial pressure is a well defined and independent risk factor for future cardiovascular disease (24,25). In the past, elevation of systolic pressure has been viewed as secondary to structural alterations in large arteries, that is, evidence of early atherosclerosis. Our results support the view that sympathoadrenal mechanisms may actively participate in determining systolic arterial pressure at rest and, as reflected by plasma epinephrine concentration, may control the work of the heart (that is, rate-pressure product) far more than was previously appreciated. Further studies are needed to characterize the determinants of plasma epinephrine in normal and hypertensive subjects and to establish the causal physiologic relations among circulating epinephrine, cardiovascular function and long-term cardiovascular risk. We thank Ellen Elting for technical expertise and Ruth Gonsky and Elaine Grohman for careful preparation of this manuscript.
356
KRAKOFF ET AL. EPINEPHRINE, BLOOD PRESSURE, RATE-PRESSURE PRODUCT
References I. Pickering G. The Nature of Essential Hypertension. New York: Grune & Stratton, 1961. 2. Hypertension Detection and Follow-Up Program Cooperative Group: The Hypertension Detection and Follow-Up Program: a progress reI977;40(suppl 1):1-106-9. port. Circ Res I977;40(suppl 3. The Multiple Risk Factor Intervention Trial Group: Statistical design considerations in the NHLI Multiple Risk Factor Intervention Trial. 1 Chronic Dis 1977;30:261-75.
JACC Vol. 5, No.2
February 1985:352-6
14. Silverberg AB, Shah S, Haymond MW, Cryer P. Norepinephrine: I978;234:E252-6. hormone and neurotransmitter in man. Am 1 Physiol I978;234:E252-6. 15. Clutter WE, Bier DM, Shah SD, Cryer PE. Epinephrine plasma metabolic clearance rates and physiologic threshold for metabolic and hemodynamic actions in man. 1 Clin Invest 1980;66:94-101. 16. Franco-Morselli R, Elghozi lL, 10ly E, di Giulio S, Meyer P. Increased plasma adrenaline In in benign essential hypertension. Br Med 1 1977;2: 1251-4.
4. Passon PG, Peuler lD. A simplified radiometric assay for plasma norepinephrine and epinephrine. Anal Biochem 1973;51 :618-31.
17. Bertel 0, Buhler F, Kiowski W, Lutold B. Decreased betaadrenoreceptor responsiveness as related to age, blood pressure and plasma catecholamines in patients with essential hypertension. Hypertension 1980;2: 130-8.
5. Glass DC, Krakoff LR, Contrada R, et al. Effect of harassment and competition upon cardiovascular and plasma catecholamine responses in type A and type B individuals. Psychopathol 1980;17:453-63.
18. de Champlain 1, J, Nadeau RA, Lavallee M, Denis G. Autonomic dysfunctions in human hypertension. Hypertension 1981 ;3(suppl 11):11124-8.
6. Krakoff LR, Ribeiro AB, Gorkin lU, Felton KR. Saralasin infusion in screening for renovascular hypertension. Am 1J Cardiol 1980;45:609-13. 7. Sheffield LT. Exercise stress testing. In: Braunwald E, ed. Heart Disease: A Textbook of Cardiovascular Medicine. Philadelphia: WB Saunders, 1980:259-77.
19. Meier A, Weidmann P, Grimm M, et a!. al. Pressor factors and cardiovascular pressor responsiveness in borderline hypertension. Hypertension 1981 ;3:367-72. L1, Vandongen R, Arkwright PD, Davison L. Adrenal and 20. Beilin LJ, sympathetic nervous activity in subjects with "low" and "high" normal blood pressure. 1 Hypertension 1983; I: 13-8.
8. Cryer PE. Physiology and pathophysiology of the human sympathoadrenal neuroendocrine system. N Engll Med 1980;303:436-44.
p, Kiowski W, Buhler FR. Enhanced alpha-adren21. Amann FW, Bolli P, oreceptor-mediated vasoconstriction in essential hypertension. Hy1981;3(suppl 1):1-119-23. pertension 1981;3(suppII):i-119-23.
9. Robertson D,lohnson D, 10hnson GA, Robertson RM, Nies AS, Shand DG, Oates lA. JA. Comparative assessment of stimuli that release catecholamines in man. Circulation 1979;59:637-43. CaldwelllH, Richie lL. Com10. Stratton lR, Halter lB, Hallstrom AP, CaldweIllH, parative plasma catecholamine and hemodynamic responses to handgrip, cold pressor and supine bicycle exercise testing in normal subjects. 1 Am Coli Cardiol 1983;2:93-104. II. Ziegler MG, Lake CR, Kopin 11. The sympathetic-nervous-system 1977 ;296:293-7 . defect in primary orthostatic hypotension. N Engll Med 1977;296:293-7. 12. Goldstein DS. Plasma norepinephrine in essential hypertension: a study 1981 ;3:48-52. of the studies. Hypertension 1981;3:48-52. 13. Esler M, lackman Jackman G, Bobik A, et al. Norepinephrine kinetics in essential hypertension. Defective neuronal uptake of norepinephrine 1981;3:149-56. in some patients. Hypertension 1981;3: 149-56.
22. Hilsted 1, J, Parving H-H, Christensen Nl, Benn 1, Galbo H. Hemodynamics in diabetic orthostatic hypotension. 1J Clin Invest 1981 ;68: 1427-34. 23. Brown Ml, MJ. Macquin I. Is adrenaline the cause of essential hypertension') Lancet 1981 ;2: 1079-82. 24. Kannel WB, Dawber TR, McGee DL. Perspectives on systolic hypertension: the Framingham study. Circulation 1980;61: 1179-82. 25. Rabkin SW, Methewson FAL, Tate RB. Predicting risk of ischemic heart disease and cerebrovascular disease from systolic and diastolic blood pressures. Ann Intern Med 1978;88:342-5.
lACC Vol. 5. No.2 February 1985:352-6
REPORTS ON HYPERTENSION
Plasma Epinephrine Concentration in Healthy Men: Correlation With Systolic Pressure and Rate-Pressure Product LAWRENCE R. KRAKOFF, MD, FACC, STANLEY DZIEDZIC, PHD, MD, SAMUEL J. MANN, MD, KATHERINE FELTON, MS, KAREN YEAGER, RN New York. New York
Relations among plasma epinephrine, norepinephrine and renin activity and systolic pressure, diastolic pressure, heart rate and the product of heart rate and systolic pressure (rate-pressure product) were evaluated in 31 healthy men whose arterial pressure spanned the range from normal to mildly elevated. Measurements were made during 60 minutes with the patient in the supine position and during lO to minutes of quiet standing. In the supine position, highly significant regressions were found between systolic pressure or rate-pressure product and plasma epinephrine, but not between these variables and norepinephrine or renin activity. A weakly significant correlation was also found between heart rate and norepinephrine. On standing, norepinephrine and epineph-
The determinants of systolic and diastolic arterial pressures in the early phases of the hypertensive process remain to be characterized. This problem is compounded by difficulty in establishing an exact classification such that' 'normotensive" and "hypertensive" groups can be easily distinguished for comparison (1,2). Epidemiologic studies (3) have avoided such an arbitrary division and tend to rely on regression relations between arterial pressure as a continuous variable and subsequent rates of appearance of cardiovascular disease (3). On this basis, we evaluated a group of healthy, well characterized men with a range of arterial pressures varying between normal and that which is considered to be mildly hypertensive. Plasma norepinephrine, epinephrine concentration and renin activity were measured, and correlations examined with respect to systolic pressure, diastolic pressure, heart rate and heart rate X systolic pressure (rate-pressure product). Medicine. From the Hypertension Division. Mount Sinai School of Medicine, New York, York. New York. This study was supported by Grants HLB 13595 and HLB 22514 from the U.S. Public Health Service. Washington. D.C. 21. 1984; revised manuscript received August Manuscript received May 21, 20. 1984. 20, 1984, accepted September 5. 1984. Krakoff. MD. Hypertension DiviAddress for reprints: Lawrence R. Krakoff, Medicine. One Gustave L. Levy Place, Place. New sion. Mount Sinai School of Medicine, York. New York 10029. If') © 1985
by the American College of Cardiology
rine increased significantly. In this position, rate-pressure product was significantly related by regression analysis only with plasma epinephrine. Weakly significant correlations between systolic pressure and epinephrine and between heart rate and norepinephrine and epinephrine were also found. Plasma renin activity was not significantly correlated with arterial pressure, heart rate or rate-pressure product in either position. These results suggest that plasma epinephrine is a determinant of systolic pressure when postural reflexes are minimized and that epinephrine may participate in control of cardiac work load, as reflected by rate-pressure product in the absence of exercise or definable stress. (J Am Coli CardioI1985;5:352-6)
Methods Study subjects. The subjects for this study were 31 white or Hispanic men ranging in age from 18 to 55 years. years, who were selected for the protocol because their arterial pressures on outpatient examination spanned the normal to mildly elevated range. None of the subjects had taken antihypertensive medication for at least 6 months before the study. Complete medical history and physical examinations were normal except for the presence of borderline diastolic arterial pressure (90 to 100 mm Hg) on several occasions in 13 of the 31 subjects. When appropriate, screening examinations for causes of secondary hypertension were performed; all were normal. No subject had hyperglycemia, hyperlipidemia, hypokalemia, elevation of serum urea nitrogen or creatinine, proteinuria or electrocardiographic abnormalities other than increased voltage in the precordial leads. Half of these subjects engaged in regular recreational physical exercise such as tennis or jogging. In the 2 year follow-up period for these subjects since these studies were performed, some have been placed on antihypertensive drug therapy. No one in the group has had evidence of coronary heart disease, heart failure or any complication of hypertensive cardiovascular disease. All subjects were studied as outpatients and permitted to continue their usual diet. 07:15-1097/85/$3.30 0735-1097/85/$3.30
lACC Vol. 5. No.2 lACC Vol. No.2
KRAKOFI' KRAKOFF ET AL. EPINEPHRINE. BLOOD PRESSURE. RATE-PRESSURE PRODUCT
February 1985:352-6
Study protocol. On the day of the study, subjects were permitted a clear fluid breakfast. They had been requested not to smoke or drink caffeinated beverages from the previous evening at II :00 PM until arrival at the laboratory. On arrival, each subject was weighed and then placed in the supine position. The cuff on an Arteriosonde was placed on one arm, leads for an electrocardiogram and cardiac tachometer were taped in place on the chest and an indwelling catheter was inserted into an antecubital or forearm vein on the side opposite to the pressure cuff. For I hour, arterial pressure and heart rate were measured at 2 minute intervals while the subject remained supine. A blood sample for plasma catecholamines and renin activity was then taken. The subject then stood upright for 10 minutes until a second blood sample was taken. For each subject, the blood pressure and heart rate were averages of the last 10 determinations in the supine position and of 5 determinations during standing. Measurements. Plasma catecholamine concentrations, norepinephrine and epinephrine were determined in duplicate by radioenzymatic assay as previously described (4,5). (4,5). The coefficient of variation for interassay agreement is 9% 9% for norepinephrine and II % for epinephrine. Plasma renin activity was determined by radioimmunoassay of generated 5%. angiotensin I (6). (6). Coefficient of variation is 5%. Statistical analysis. Data were processed for statistical analysis by the SAS system of the CUNY computer center employing the GLM program for regression analysis. In each phase of the study, the analysis employed considered systolic or diastolic pressure and heart rate or rate-pressure product (systolic pressure x heart rate) (7) as the dependent variables for correlation in a multiple linear regression model with plasma norepinephrine, plasma epinephrine and plasma renin activity or their logarithms as independent variables (for example, systolic pressure = f30 + f31 . plasma norepinephrine + f32 . plasma epinephrine + f3, . plasma renin activity). The overall relation was considered significant for
less than or equal to 0.0 I. Specific correlation coefficients (r) for individual relations between the dependent and independent variables were also calculated. l.tF LtF
Results Supine measurements. Descriptive characteristics of the 31 men who participated in this study are given in Table I. Although six were mildly overweight (body mass index 25 to 29), 29), none was obese. The range of diastolic arterial pressure in the supine position varied from normal to mild elevation (defined as 90 to 104 mm Hg). The low average supine heart rate may reflect the degree of physical fitness of this group in agreement with the frequent history of regular exercise. The distribution of supine systolic and diastolic pressures is shown in Figure I. Measurements on standing. As indicated in Table I, diastolic pressure and heart rate increased significantly during standing. Measurements of plasma norepinephrine and epinephrine concentration in the two positions are shown in Figure 2. Standing significantly increased both plasma catecholamines. In the supine position plasma renin activity was 1.4 ± 0.2 ng/ml ng/ml per h. A significant increase (+0.8 ± 0.3 ng/ml ng/ml per h, p < 0.02 by paired t test) occurred during standing. Correlations among blood pressure, heart rate, plasma catecholamines and renin activity. Table 2 presents the statistical analysis for relations among the dependent and independent variables that were analyzed. In the supine position, a highly significant relation was found between systolic pressure or rate-pressure product and independent histograms of systolic (a) Figure 1. Frequency histograms (a) and diastolic (b) (b) pressures in in 31 31 subjects in in the the supine position. position. The midpoint midpoint of of pressures each interval is is given on on the the abscissa, abscissa, the the number number (n) (n) of of subjects on the ordinate. DISTRIBUTION OF ARTERIAL PRESSURE
12 12
a Supine systoliC pressure
10 10
of Study Group (n (n Table 1. Clinical Characteristics of
31 )) 31
Mean ±± SE Age (yr) Height (m) Weight (kg) Body mass index (kg/m") Supine Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Heart rate (beats/min) Standing Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Heart rate (beats/min)
8
Range
N N
6 4
36 36 1.77 74.4 23.9
± ± II ± 0.0\ 0.01 ±
± 1.3 \.3 ± 0.4
IX 1.67 167 60.1 19.9 19.9
to to to to
55 55 1.87 I.X7 89.3 28.9 28.9
± 2 117 ± 2 79 79 ± 2 2 59 ± ± 2
9X to 147 64 to 100 35 to 84 X4
122 \22 ± 2 89 2* 89 ± 2* 71 ± 7\ ± 2*
ISO 103 to 150 61 to 114 44 99 44 to 99
test. SE *p < 0.0 I compared with supine value by paired I tes!. error.
~ standard =
353
2
0 '----"'9""5-1""0'"--5-----'I""I5'---""'12"-5----'1'"'35"---"'14""5o 95 105 115 125 135 145 16 16
b Supine diastolic pressure
14 14 12 12 10 10 N N
8 6 4 2
o'---......-"""-------''''''----''''''-0 65 65
75 75
85 85
95 95
Hg) PRESSURE (mm Hgl
354
KRAKOFF ET AL. EPINEPHRINE, BLOOD PRESSURE, RATE-PRESSURE PRODUCT
No.2 lACC Vol. 5, NO.2 February 1985:352-6 February
POSTURE AND CATECHOLAMINES
SUPINE PLASMA EPINEPHRINE AND SYSTOLIC ARTERIAL PRESSURE
600
150 r - - - - - - - - - - - - - - - - - - - - ,
•
500 :x:'" 140
400
E
ci. a.
E E
UJ 0:: ll:
300
130
•
;:)
en en 120 UJ
200
•
0:: ll:
a.
u ~ ...J ....J
._.__.-x ._.----1:
100
o
•
•
•
•
•
• • • • • • I • • • •
en >- 100 en
variables, Highly significant individual correlations between systolic pressure or rate-pressure product and plasma epinephrine were also found, For the log-transformed independent variables, systolic pressure was significantly cor0.001) only with epinephrine (Fig. related (F = 8.18, P < 0,001) 3). Neither diastolic pressure nor heart rate was significantly related to the three independent variables in the supine position by multiple regression analysis. A weakly significant correlation was evident between heart rate alone and plasma norepinephrine. norepinephrine, Table 2 also displays the statistical relations for the standing position. Only rate-pressure product is significantly related to independent variables in the model employed. A highly significant correlation was found between rate-
•
•
9 i
,1
5 10
II
20
1I
I
40
80
0.0001).
pressure product and plasma epinephrine. Weakly significant individual correlations were observed between systolic pressure and plasma epinephrine and between heart rate and plasma norepinephrine or epinephrine, epinephrine. Correlation with age and weight. In the supine position, neither age nor weight was significantly correlated with either plasma catecholamine concentration or plasma renin activity. During standing, there was a significant positive correlation between age and plasma epinephrine (r = 0.45, p < 0.05), but not between age and norepinephrine or renin activity. Weight was not significantly correlated with standactivity, ing plasma catecholamine concentration or renin activity.
F (model) F
NE
E
PRA
0.26 0.31 0.22 0.29
0.65* 0.15 0.44t 0.65
<0.1 <0.1 <0.1 <0.1
Supine 6.96* 0.41 2.30 6.85* Standing 1.04
0.30 1.62 5.44*
160
Figure 3. Relation between systolic arterial pressure and plasma epinephrine in 31 subjects in the supine position (r = 0.68, p =
Correlation Coefficients (r)
Diastolic pressure Heart rate Rate-pressure product
.-J
PLASMA EPINEPHRINE (pg/mll (pg/ml)
Table 2. Statistical Analysis
Systolic pressure
• •
STAND
Figure 2. Mean plasma norepinephrine (NE) and epinephrine (E) concentrations in supine and standing positions. Results are expressed as mean ± standard error. Significant increases occurred in both plasma catecholamine concentrations on standing.
Systolic pressure Diastolic pressure Heart rate Rate-pressure product
••
•
I-
.---- E SUPINE
110
0
P
•
• •
0.12
0.22 0.36t 0.36t
0.37t
<0.1 0.42t 0.56*
0.20
<0.1 <0.1 0.16
*p < 0.01; to.OI to.OI < p < 0.05. The F value is given for the entire statistical model (see Methods), whereas the r value is shown for individual correlations. E = plasma epinephrine; NE = plasma norepinephrine; PRA = plasma renin activity.
lACC Vo!. Vol. 5. No.2 JACC
KRAKOFF ET AL. EPINEPHRINE. BLOOD PRESSURE. RATE-PRESSURE PRODUCT
February 1985:352-6
Discussion Plasma catecholamines and sympathoadrenal function. Plasma norepinephrine and epinephrine are measures of sympathoadrenal function. Various stimuli cause significant increases in plasma catecholamine concentration (8-10). (8- 10). Increased sympathetic tone is necessary for the maintenance of arterial pressure in response to upright posture; in normal subjects standing causes a doubling of plasma norepinephrine. Patients with idiopathic orthostatic hypotension have no postural increase in the concentration of this catecholamine (8,11). Many previous studies have explored possible relations between sympathetic function, as reflected by plasma norepinephrine, and essential hypertension. Although there is a slight trend toward elevation of plasma norepinephrine in hypertensive compared with control groups, when many studies are pooled interpretation is difficult because of variation in control groups and categorization of the hypertensive subjects as to etiology of disease or cardiac status (12). Furthermore, plasma norepinephrine reflects spillover from that which is released from nerve endings and escapes neural re-uptake or local metabolism (13). Unless sympathetic function is activated by a potent stimulus, plasma norepinephrine remains far below its apparent physiologic threshold for effect as a circulating hormone (14). Plasma epinephrine in the normal and hypertensive subject. Circulating epinephrine reflects secretion from the adrenal medullae. Supine plasma epinephrine concentration is normally in a remarkably low range; the mean concentration is about half the apparent physiologic threshold for effects on systolic pressure and heart rate as determined by infusion of exogenous epinephrine (15). Infusion studies may overestimate the threshold for endogenous epinephrine unless the preinfusion plasma level of the amine is far below threshold concentration (that is, such that adrenergic receptors are unoccupied before infusion). It is then possible that plasma epinephrine concentration has some physiologic effect at levels found in the circulation in unstressed states. (16,17) 6,17) have reported elevations Previous publications (J of plasma epinephrine concentration in patients with established hypertension compared with normal subjects. High levels of plasma epinephrine concentration have also been found in "borderline hypertensives," but no correlation (18,19). with arterial pressure has been described (l8.J 9). More recently, it has been reported (20) that supine plasma epinephrine concentration was significantly higher in a small number (n = 9) of subjects with "high normal" arterial pressure compared with subjects with a "low normal" pressure drawn from a large healthy working population in Western Australia. However, the only significant regression relation was between systolic pressure and plasma epinephrine in the "low" group. Amann et at. aL (21) reported that in patients with essential hypertension, plasma epinephrine
355
concentration is positively correlated with alpha receptormediated vasoconstriction in the forearm as revealed by blockade with intraarterial prazosin. In normotensive subjects, the correlation is not observed. Results of this study. Our study evaluated well characterized normal volunteers and patients referred for borderline arterial pressure or uncomplicated mild hypertension. Although this group may be viewed as selected and not representing an unbiased sample of the population at large, it is free of factors that might alter plasma catecholamine concentration independent of changes in arterial pressure, such as coronary heart disease or diabetes (22). Our results indicate that plasma epinephrine concentration is highly correlated with systolic arterial pressure in the supine position when reflexes mediating postural adjustment are less active. During both supine and upright posture, the heart rate x systolic pressure product correlated positively with plasma epinephrine but not norepinephrine. In contrast, diastolic pressure was not significantly correlated with plasma catecholamine concentration or renin activity as an independent variable in either the supine or the standing position. Role of circulating epinephrine in control of blood pressure. These results do not establish the extent to which circulating epinephrine itself controls systolic arterial pressure or rate-pressure product. It may be that plasma epinephrine reflects global sympathoadrenal tone to a greater extent than does plasma norepinephrine, which spills into the circulation from the large number of sympathetic nerve terminals as well as the adrenal medullae. The specific role of circulating epinephrine need not be limited to direct effects on cardiac beta-receptors increasing heart rate and contractility or vascular beta2-receptors causing vasodilation. Recent studies (23) suggested that circulating epinephrine may be taken up and secreted by adrenergic neurons as a "co-transmitter" or participate in presynaptic regulation of adrenergic neurosecretion. Implications. Systolic arterial pressure is a well defined and independent risk factor for future cardiovascular disease (24,25). In the past, elevation of systolic pressure has been viewed as secondary to structural alterations in large arteries, that is, evidence of early atherosclerosis. Our results support the view that sympathoadrenal mechanisms may actively participate in determining systolic arterial pressure at rest and, as reflected by plasma epinephrine concentration, may control the work of the heart (that is, rate-pressure product) far more than was previously appreciated. Further studies are needed to characterize the determinants of plasma epinephrine in normal and hypertensive subjects and to establish the causal physiologic relations among circulating epinephrine, cardiovascular function and long-term cardiovascular risk. We thank Ellen Elting for technical expertise and Ruth Gonsky and Elaine Grohman for careful preparation of this manuscript.
356
KRAKOFF ET AL. EPINEPHRINE, BLOOD PRESSURE, RATE-PRESSURE PRODUCT
References I. Pickering G. The Nature of Essential Hypertension. New York: Grune & Stratton, 1961. 2. Hypertension Detection and Follow-Up Program Cooperative Group: The Hypertension Detection and Follow-Up Program: a progress reI977;40(suppl 1):1-106-9. port. Circ Res I977;40(suppl 3. The Multiple Risk Factor Intervention Trial Group: Statistical design considerations in the NHLI Multiple Risk Factor Intervention Trial. 1 Chronic Dis 1977;30:261-75.
JACC Vol. 5, No.2
February 1985:352-6
14. Silverberg AB, Shah S, Haymond MW, Cryer P. Norepinephrine: I978;234:E252-6. hormone and neurotransmitter in man. Am 1 Physiol I978;234:E252-6. 15. Clutter WE, Bier DM, Shah SD, Cryer PE. Epinephrine plasma metabolic clearance rates and physiologic threshold for metabolic and hemodynamic actions in man. 1 Clin Invest 1980;66:94-101. 16. Franco-Morselli R, Elghozi lL, 10ly E, di Giulio S, Meyer P. Increased plasma adrenaline In in benign essential hypertension. Br Med 1 1977;2: 1251-4.
4. Passon PG, Peuler lD. A simplified radiometric assay for plasma norepinephrine and epinephrine. Anal Biochem 1973;51 :618-31.
17. Bertel 0, Buhler F, Kiowski W, Lutold B. Decreased betaadrenoreceptor responsiveness as related to age, blood pressure and plasma catecholamines in patients with essential hypertension. Hypertension 1980;2: 130-8.
5. Glass DC, Krakoff LR, Contrada R, et al. Effect of harassment and competition upon cardiovascular and plasma catecholamine responses in type A and type B individuals. Psychopathol 1980;17:453-63.
18. de Champlain 1, J, Nadeau RA, Lavallee M, Denis G. Autonomic dysfunctions in human hypertension. Hypertension 1981 ;3(suppl 11):11124-8.
6. Krakoff LR, Ribeiro AB, Gorkin lU, Felton KR. Saralasin infusion in screening for renovascular hypertension. Am 1J Cardiol 1980;45:609-13. 7. Sheffield LT. Exercise stress testing. In: Braunwald E, ed. Heart Disease: A Textbook of Cardiovascular Medicine. Philadelphia: WB Saunders, 1980:259-77.
19. Meier A, Weidmann P, Grimm M, et a!. al. Pressor factors and cardiovascular pressor responsiveness in borderline hypertension. Hypertension 1981 ;3:367-72. L1, Vandongen R, Arkwright PD, Davison L. Adrenal and 20. Beilin LJ, sympathetic nervous activity in subjects with "low" and "high" normal blood pressure. 1 Hypertension 1983; I: 13-8.
8. Cryer PE. Physiology and pathophysiology of the human sympathoadrenal neuroendocrine system. N Engll Med 1980;303:436-44.
p, Kiowski W, Buhler FR. Enhanced alpha-adren21. Amann FW, Bolli P, oreceptor-mediated vasoconstriction in essential hypertension. Hy1981;3(suppl 1):1-119-23. pertension 1981;3(suppII):i-119-23.
9. Robertson D,lohnson D, 10hnson GA, Robertson RM, Nies AS, Shand DG, Oates lA. JA. Comparative assessment of stimuli that release catecholamines in man. Circulation 1979;59:637-43. CaldwelllH, Richie lL. Com10. Stratton lR, Halter lB, Hallstrom AP, CaldweIllH, parative plasma catecholamine and hemodynamic responses to handgrip, cold pressor and supine bicycle exercise testing in normal subjects. 1 Am Coli Cardiol 1983;2:93-104. II. Ziegler MG, Lake CR, Kopin 11. The sympathetic-nervous-system 1977 ;296:293-7 . defect in primary orthostatic hypotension. N Engll Med 1977;296:293-7. 12. Goldstein DS. Plasma norepinephrine in essential hypertension: a study 1981 ;3:48-52. of the studies. Hypertension 1981;3:48-52. 13. Esler M, lackman Jackman G, Bobik A, et al. Norepinephrine kinetics in essential hypertension. Defective neuronal uptake of norepinephrine 1981;3:149-56. in some patients. Hypertension 1981;3: 149-56.
22. Hilsted 1, J, Parving H-H, Christensen Nl, Benn 1, Galbo H. Hemodynamics in diabetic orthostatic hypotension. 1J Clin Invest 1981 ;68: 1427-34. 23. Brown Ml, MJ. Macquin I. Is adrenaline the cause of essential hypertension') Lancet 1981 ;2: 1079-82. 24. Kannel WB, Dawber TR, McGee DL. Perspectives on systolic hypertension: the Framingham study. Circulation 1980;61: 1179-82. 25. Rabkin SW, Methewson FAL, Tate RB. Predicting risk of ischemic heart disease and cerebrovascular disease from systolic and diastolic blood pressures. Ann Intern Med 1978;88:342-5.