Increased Cardiac Adrenergic Drive Precedes Generalized Sympathetic Activation in Human Heart Failure
Background Previous studies with radiotracer methods have indicated increases in cardiac norepinephrine (NE) and renal NE spillover in patients with severe congestive heart failure (CHF). However, data on the regional sympathetic profile in early stages of CHF are limited. In this study, sympathetic function in the heart, kidneys, and skeletal muscle was evaluated in patients with mild-to-moderate CHF and compared with that in patients with severe CHF and healthy subjects.
Methods and Results Total body and regional NE spillover from the heart and kidney was assessed with isotope dilution with steady state infusions of [3H]NE. Sympathetic nerve traffic to the skeletal muscle vascular bed (MSA) was recorded intraneurally. Cardiac NE spillover in patients with mild-to-moderate CHF (n=21) was increased threefold versus that in healthy subjects (n=12, P<.05), whereas total body and renal NE spillover and MSA did not differ from those in healthy subjects. In the severe CHF group (n=12), cardiac NE spillover was increased fourfold (P<.05), and total body and renal NE spillover and MSA were high compared with both mild-to-moderate CHF subjects and healthy subjects (P<.05 for both). Fractional extraction of [3H]NE across the heart was reduced by ≈40% in both CHF groups versus control subjects (P<.05).
Conclusions These results indicate a selective increase in cardiac adrenergic drive (increased amounts of transmitter available at neuroeffector junctions) in patients with mild-to-moderate CHF. This increase appears to precede the augmented sympathetic outflow to the kidneys and skeletal muscle found in advanced CHF.
There is ample evidence for increased sympathetic nerve activity in patients with severe CHF.1 2 3 Previous studies with steady state infusions (isotope dilution) of tritiated NE in CHF patients have shown increased overall and regional NE spillover from the heart and kidneys.4 5 6 7 8 The elevated cardiac NE spillover results from both increased neuronal release (augmented nerve firing) and impaired neuronal uptake of NE.5 7 9 10 Direct recording of sympathetic nerve activity to the skeletal muscle vascular bed (MSA) is also augmented in CHF,11 whereas intraneural recording of sympathetic nerve activity to the skin remains normal.12 Thus, sympathetic activation in CHF is nonuniform, and there is a disproportionate increase in cardiac NE spillover compared with that in other organs.5 6 8 Accumulating evidence suggests that adrenergic mechanisms have important prognostic and pathophysiological significance in CHF.13 14 15 16 17 Furthermore, increased cardiac NE spillover in CHF has been associated with both malignant ventricular arrhythmias18 and adverse prognosis.17
In the early stages of CHF, in which long-term mortality is still high and arrhythmia-related deaths are common,19 cardiac and renal NE kinetics have not been studied. This investigation was therefore undertaken to study changes in regional sympathetic nerve function in the heart, kidneys, and skeletal muscle in a well-defined cohort of patients with mild-to-moderate CHF. A specific aim was to identify a putative increase in cardiac NE spillover, representing a possible compensatory mechanism for the reduced left ventricular function. Regional sympathetic function was estimated in these patients and compared with that in patients with severe CHF. These two CHF groups were then compared with age-matched patients with stable angina pectoris who had normal left ventricular function and with healthy subjects. Total body and regional NE spillover from the heart and kidneys was assessed through isotope dilution of [3H]NE.20 Sympathetic nerve activity to skeletal muscle vascular bed was recorded intraneurally in the peroneal nerve.21
Heart Failure Patients
Twenty-one patients with symptomatic mild-to-moderate CHF and 12 patients with severe CHF were examined. Functional capacity in the mild-to-moderate CHF group corresponded to NYHA functional class II or III, and a reduction in left ventricular ejection fraction (echocardiography22 ) of <45% was mandatory for inclusion in the study (Table 1⇓). Patients in the NYHA class III group were also classified as being in NYHA class IIIA, indicating that they can walk ≤200 m on flat ground without chest discomfort. Patients with significant noncardiac disease were excluded. All patients in the mild-to-moderate CHF group had been receiving stable medical treatment for >3 months. All mild-to-moderate CHF patients were receiving furosemide and an ACE inhibitor; 11 subjects were also receiving digitalis.
Patients in the severe CHF group were recruited during evaluation for heart transplantation (Table 1⇑). They were all markedly symptomatic despite optimal pharmacological treatment, including diuretics, digitalis, and ACE inhibition. Patients in this group were classified as being in NYHA class IIIB or IV, the former indicating marked chest discomfort after walking <200 m on flat ground. All patients were investigated while in a stable clinical condition without recent (<3 months) acute ischemic events. Coronary angiography was performed to identify the presence of significant coronary artery disease as a possible cause of CHF.
Twenty-one healthy age-matched subjects were included as control subjects for the MSA recordings. Another group comprised 12 healthy male subjects. They were catheterized as control subjects for the NE kinetic measurements. All healthy subjects had a normal physical examination and were not receiving any medication (Table 1⇑).
Thirteen patients with stable angina pectoris were also included as an older age group for studies of cardiac and total body NE kinetics. These patients had no signs or history of CHF and showed normal left ventricular ejection fraction on contrast ventriculography (Table 1⇑).
All investigations were performed at Sahlgrenska University Hospital. The study protocol was approved by the local ethical and isotope committees. All subjects gave written consent to participate in the study.
Catheterization was performed in the morning after an overnight fast, and medication was withdrawn 12 hours before the investigation. MAP and heart rate were recorded continuously via an indwelling catheter in the left radial or brachial artery. Cardiac pressures and flows were obtained with a balloon-tipped thermodilution catheter via the right internal jugular vein. In the same vein, a coronary sinus catheter (Wilton-Webster) was inserted (each patient had two sheaths). Coronary blood flow was determined by the retrograde thermodilution technique according to Ganz et al.23 When central hemodynamics had been obtained, the right renal vein was catheterized through replacement of the thermodilution catheter with a 7F catheter. This experimental set-up enabled simultaneous arterial, renal vein, and coronary sinus blood sampling. Renal blood flow was calculated from the infusion clearance and renal fractional extraction of para-aminohippurate. To confirm the CHF diagnosis, all mild-to-moderate CHF patients with normal baseline hemodynamics underwent supine cycling exercise during catheterization.
All patients received a continuous infusion of tritiated NE (ring-labeled [3H]NE, 40 Ci/mmol; New England Nuclear) delivered via an antecubital vein at a rate of 1.3 μCi/min. The infusate contained acetic acid (2 mmol/L) and ascorbic acid (1 mmol/L) to stabilize the [3H]NE.
Total body spillover of NE into plasma (SPTB) and total body plasma clearance of NE (CLTB) were calculated according to Esler et al20 :CL_|<|TB|>||<|=|>|I/|<|[|>|^|<|3|>|H|<|]|>||<|\cdot|>|NE_|<|a|>| (mL/min)SP_|<|TB|>||<|=|>|CL_|<|TB|>||<|\cdot|>|NE_|<|a|>| (pmol/min)where I is the infusion rate of [3H]NE (dpm/min), [3H]NEa is the arterial tritiated NE concentration (dpm/mL), and NEa is the arterial endogenous NE concentration (pmol/mL).
Organ spillover (heart, kidney) of NE into plasma (SPorgan) and the organ fractional extraction of NE (EX) were calculated24 :SP_|<|organ|>||<|=|>||<|[|>|(NE_|<|v|>||<|-|>|NE_|<|a|>|)|<|+|>|NE_|<|a|>||<|\cdot|>|EX_|<|organ|>||<|]|>||<|\cdot|>|QEX_|<|organ|>||<|=|>|(|<|[|>|^|<|3|>|H|<|]|>|NE_|<|a|>||<|-|>||<|[|>|^|<|3|>|H|<|]|>|NE_|<|v|>|/|<|[|>|^|<|3|>|H|<|]|>|NE_|<|a|>|where Q is the organ plasma flow (mL/min), NEv is the venous NE concentration (pmol/mL), and [3H]NEv is the venous concentration of tritiated NE (dpm/mL).
Blood samples were transferred immediately into ice-cold tubes containing reduced glutathione and heparin. Samples were centrifuged at 4°C, and plasma was separated for storage at −80°C until assayed. Plasma concentrations of endogenous NE were determined by liquid chromatography with electrochemical detection. Timed collection of eluate leaving the detection unit allowed separation of [3H]NE for subsequent counting by liquid scintillation spectroscopy. Interassay coefficients of variation were 4.6% for NE and 3.2% for [3H]NE.
Sympathetic Nerve Recordings
Microneurographic recordings were performed 24 hours after catheterization. Multiunit postganglionic sympathetic nerve activity was recorded with a tungsten microelectrode inserted into a muscle-innervating fascicle of the peroneal nerve at the fibular head. A reference electrode was inserted subcutaneously 1 to 2 cm from the recording electrode. Details regarding the recording technique and the criteria for MSA have been described previously.21 The number of MSA bursts in the mean voltage neurogram were counted by inspection of the mean voltage neurogram. The nerve activity was expressed as the average burst frequency (bursts/min) and the burst incidence (bursts/100 heartbeats).
Results are presented as mean±SD unless stated otherwise. Group comparisons were made with one-way ANOVA, with the Scheffe´ posthoc test to identify differences among the various groups. A value of P<.05 was considered statistically significant. Nonnormal distributed NE kinetic data were transformed logarithmically before analysis.
The heart rate was higher and MAP was lower in patients with severe CHF than in healthy subjects, patients with angina, or patients with mild-to-moderate CHF (Table 1⇑). In the mild-to-moderate CHF group, approximately two thirds had normal cardiac pressures and cardiac output. In the severe CHF group, intracardiac pressures were substantially higher and cardiac index was lower than in patients with mild-to-moderate CHF. Coronary blood flow was similar among the study groups. In contrast, renal blood flow was reduced by 38% and 52% in those with mild-to-moderate CHF and severe CHF, respectively, compared with healthy subjects (P<.05), which is consistent with a higher renal vascular resistance in CHF (Table 2⇓). Systemic and coronary vascular resistances were similar in CHF patients, and coronary vascular resistance in CHF patients did not differ significantly from that in healthy subjects (Table 2⇓).
Cardiac NE Kinetics
Cardiac NE spillover was increased threefold in patients with mild-to-moderate CHF (P<.05) and fourfold (P<.05) in the severe CHF group compared with both healthy subjects and angina patients (Fig 1⇓, Table 2⇑), whereas the difference between the two CHF groups was not statistically significant. There also was an increase in the venoarterial gradient for NE across the heart in both CHF groups compared with both healthy subjects and angina patients (P<.05). Fractional extraction of [3H]NE across the heart was significantly reduced in both CHF groups: 57±14% among patients with severe CHF and 63±11% in mild-to-moderate CHF, respectively, compared with 87±4% in healthy subjects and 76±13% in stable angina patients (P<.05 for both; Fig 2⇓).
Renal NE Kinetics
There was a significant increase in the venoarterial gradient of endogenous NE concentration across the renal circulation in patients with severe CHF compared with both healthy subjects and the mild-to-moderate CHF group (P<.05). In the mild-to-moderate CHF group, the renal venoarterial gradient of NE was similar to that in healthy subjects (Table 2⇑). Renal NE spillover was similar in control subjects and mild-to-moderate CHF patients, whereas an increase of 125% (P<.05) was observed in patients with severe CHF compared with healthy subjects (Fig 3⇓, top, and Table 2⇑). Renal extraction of tritiated NE and para-aminohippurate did not differ between groups.
Total Body NE Kinetics
Total body NE spillover in mild-to-moderate CHF patients did not differ significantly from the control groups (Fig 3⇑, middle, and Table 2⇑). In contrast, total body NE spillover was increased by ≈100% (P<.05) in patients with severe CHF compared with both control subjects and patients with mild-to-moderate CHF. Arterial plasma NE concentrations were elevated in both CHF groups (Table 2⇑). There was, however, a reduction in total body NE clearance in the two CHF groups (1.82±0.54 and 2.03±0.56 L/min for severe and mild-to-moderate CHF patients, respectively) compared with healthy control subjects (2.81±0.10 L/min) and angina patients (2.20±0.45 L/min; P<.05 for both).
Muscle Sympathetic Nerve Activity
Mild-to-moderate CHF patients and healthy subjects showed similar MSA burst frequency values (Fig 3⇑, bottom, and Table 2⇑). In contrast, there was a 100% increase (P<.05) in MSA burst frequency among severe CHF patients compared with healthy subjects and a 67% increase (P<.05) compared with mild-to-moderate CHF (Fig 3⇑, bottom, and Table 2⇑). When MSA was corrected for heart rate (burst incidence), MSA was similar in mild-to-moderate CHF patients (64±22 bursts/100 heartbeats) and healthy subjects (63±17 bursts/100 heartbeats), whereas the severe CHF group showed a 46% increase in MSA (92±7 bursts/100 heartbeats) compared with both healthy subjects and mild-to-moderate CHF patients (P<.05 for both).
This study demonstrates a threefold increase in cardiac NE spillover in treated patients with stable mild-to-moderate CHF without showing significant changes in either total body and renal NE spillover or MSA. These findings indicate that there is an increased cardiac adrenergic drive (ie, increased amounts of transmitter available at neuroeffector junctions for receptor binding) in these patients. Furthermore, cardiac fractional extraction of tritiated NE was reduced markedly in mild-to-moderate CHF, consistent with impaired NE uptake, which suggests a primary abnormality in cardiac sympathetic function. Thus, in this study group, an increase in cardiac adrenergic drive precedes the more generalized hyperactivity of sympathetic discharge that prevails in advanced CHF. The increased concentrations of NE at cardiac receptor sites in both mild-to-moderate and severe CHF may constitute the prerequisite for several secondary adverse effects for the failing heart, such as β-receptor downregulation/desensitization, pathological cardiac growth, depletion of NE stores, and changes in G proteins.16 25 26 These changes thus provide pathophysiological mechanisms by which a high cardiac adrenergic drive could contribute to worsening of cardiac function and progression of CHF. Furthermore, the documented association among malignant arrhythmias, myocardial ischemia, and cardiac NE spillover provides links to clinical events and prognosis.17 18 27 28 The prognostic aspects are of particular interest because cardiac NE spillover (above the median of 310 pmol/min) specifically was found to be the most important predictive variable for death among CHF patients awaiting cardiac transplantation.17 Because 11 of the 21 mild-to-moderate CHF patients had cardiac NE spillovers of >310 pmol/min, the important question arises of whether this subgroup also carries a worse prognosis.
Cardiac, renal, and total body NE spillover values in the present severe CHF group were high as was MSA, which is consistent with earlier studies in this patient category.4 5 6 7 8 11 These indexes of sympathetic outflow were significantly increased versus those of both control subjects and the present mild-to-moderate CHF group. In contrast, our patients with mild-to-moderate CHF showed no significant changes in total body and renal NE spillover or MSA. This contrasting regional sympathetic profile in severe CHF versus mild-to-moderate CHF patients suggests the presence of different sympathoexcitatory mechanisms within the heart versus those in the kidneys and skeletal muscle vascular bed.
The measurement of cardiac NE spillover indicates the amount of transmitter available at the myocardial neuroeffector junction and does not reflect only neuronal release of NE.24 29 A decreased neuronal uptake of NE may also result in elevated NE spillover.7 29 Therefore, both increased neuronal release and decreased uptake of NE could contribute to increased NE spillover. In the present study, cardiac fractional extraction of [3H]NE was substantially reduced in both CHF groups. Because uptake of NE in the heart is predominantly neuronal,24 29 a decreased fractional NE extraction would indicate that neuronal uptake is impaired in mild-to-moderate CHF. Reduced NE reuptake may thus be one important mechanism contributing to a selective increase in NE spillover from the heart and therefore also to an augmented cardiac adrenergic drive. A reduction in cardiac fractional extraction in the present study may be confounded by the use of older subjects30 in the CHF groups versus the younger healthy control group. However, there remained a marked difference between both CHF groups and age-adjusted angina patients with nonfailing hearts for [3H]NE extraction (and cardiac NE spillover), suggesting marked alteration of cardiac NE extraction in the early stages of CHF.
Although a decreased cardiac removal of NE has been demonstrated in several CHF models,6 7 8 9 10 the underlying cause has not been unraveled. With infusions of [3H]NE to steady state and desipramine, a neuronal reuptake blocker, cardiac neuronal uptake of NE was shown to be largely intact in untreated patients with severe CHF.5 In a recent study, however, the reduction in cardiac NE fractional extraction after desipramine infusion was lower among treated CHF patients than among healthy subjects, which is consistent with decreased efficiency of the active neuronal NE reuptake mechanism.7 It was also demonstrated that although cardiac extraction of NE was reduced, the increase in NE spillover originated from both increased neuronal release and reduced NE reuptake.7 However, cardiac NE spillover was much higher in untreated CHF patients5 than in treated CHF patients, indicating that most of the increase in NE spillover in the former group was due to augmented NE release, whereas cardiac NE spillover in the treated group was more dependent on reduced NE reuptake.7 In the present study, most of the increase in cardiac NE spillover in our treated mild-to-moderate CHF patients is therefore likely to have been the result of impaired NE uptake. A contribution from increased neuronal release cannot, however, be ruled out.
Yet another explanation for a lower extraction of tritiated NE across the heart may be a mismatch of perfusion versus innervation within the diseased myocardium,5 31 preferentially in patients with coronary artery disease. Furthermore, other mechanisms also must be considered, such as changes in the myocardial ultrastructure with prolonged diffusion distance due to hypertrophy and/or increased fibrosis. Functional impairment or decreased density of membrane NE transporters in CHF may also contribute.10
Changes in myocardial structure may also influence cardiac afferent and autonomic reflex control through, for example, sympathoexcitatory mechanosensitive and chemosensitive receptors described within the myocardium.32 33 Dilatation of the heart could thus be a possible stimulus for myocardial mechanosensitive sympathetic afferents. There was, however, no correlation between left ventricular volumes or pulmonary capillary wedge pressure and cardiac NE spillover, arguing against this reflex mechanism being a significant determinant of efferent cardiac sympathetic nerve activity. Stimulation of chemosensitive sympathetic afferents by myocardial ischemia, bradykinin, and prostaglandins, eliciting a cardiocardiac excitatory reflex,33 have been demonstrated in animal models. Such a reflex could contribute to a cardioselective increase in NE release, at least in ischemic CHF. This is supported by findings of a selective increase in cardiac NE spillover in patients with unstable angina pectoris.27 28 The significance of these sympathoexcitatory afferents in human CHF remains to be elucidated. In addition, most of our patients had angiographically normal coronary vessels.
Several investigators have found reduced reflex responses in both arterial and cardiopulmonary baroreceptor systems in CHF patients.34 In addition, in a recent study, impaired baroreceptor function was also found in patients with mild CHF.35 A reduced afferent negative feedback via these reflex arches to vasomotor centers may constitute a pathophysiological mechanism for the sympathoexcitation in CHF.34 However, we were unable to demonstrate any correlation between baseline central hemodynamics and cardiac NE spillover in the mild-to-moderate CHF group, arguing against baroreceptor-mediated cardiac sympathoexcitation at this early stage. Indeed, a selective increase in cardiac sympathetic tone is probably not explained by a reduced arterial or cardiopulmonary baroreceptor control because these systems normally involve augmented nerve efferent traffic to both the vascular bed in skeletal muscle and the kidneys.34 Therefore, we did not identify any increased sympathetic nerve activity to the kidneys and the skeletal muscle during baseline conditions, which is consistent with a normal baroreceptor reflex control. This finding is also supported by a recent study that showed a normal renal blood flow response in CHF.36 In addition, the high renal vascular resistance and low renal blood flow found in both CHF groups in the present study did not correlate with renal NE spillover, suggesting that other, intrarenal mechanisms, such as local angiotensin II, may have an important role for renal vascular tone.
In summary, the present study demonstrates a selective increase in cardiac NE spillover in patients with mild-to-moderate CHF. This increase is to a major degree caused by the markedly reduced fractional extraction of [3H]NE across the heart, which is indicative of impaired NE uptake mechanisms in early stages of CHF. These findings imply that there are increased amounts of the transmitter at neuroeffector junctions, resulting in an increased cardiac adrenergic drive. Whether therapeutic interventions aimed at blocking this sympathoexcitation in mild-to-moderate CHF patients would be beneficial may be resolved in clinical trials with β-blockers and other sympathetic antagonists.
Selected Abbreviations and Acronyms
|CHF||=||congestive heart failure|
|MAP||=||mean arterial blood pressure|
|MSA||=||muscle sympathetic nerve activity|
This study was supported by grants from Sahlgrenska University Hospital, The Swedish Medical Research Council (3546, 9047, and 11133), and The Swedish Heart and Lung Foundation. Dr Bergmann-Sverrisdottir was supported by The Nordic Academy for Advanced Studies. The authors are grateful for the invaluable technical assistance of Anneli Ambring, Elsa Bang, Gun Bodehed, Gunilla Fritzon, and Douglas Hooper.
- Received February 20, 1996.
- Revision received June 6, 1996.
- Accepted June 17, 1996.
- Copyright © 1997 by American Heart Association
Davis D, Baily R, Zelis RJ. Abnormalities in systemic norepinephrine kinetics in human congestive heart failure. Am J Physiol. 1988;254:E760-E766.
Hasking GJ, Esler MD, Jennings GL, Burton D, Korner PI. Norepinephrine spillover to plasma in patients with congestive heart failure: evidence of increased overall and cardiorenal sympathetic nervous activity. Circulation. 1986;73:615-621.
Meredith IT, Eisenhofer G, Lambert GW, Dewar EM, Jennings GL, Esler M. Cardiac sympathetic nervous activity in congestive heart failure: evidence for increased neuronal norepinephrine release and preserved neuronal uptake. Circulation. 1993;88:136-145.
Eisenhofer G, Friberg P, Rundqvist B, Quyyumi AA, Lambert G, Kaye DM, Kopin IJ, Goldstein DS, Esler MD. Cardiac sympathetic nerve function in congestive heart failure. Circulation. 1996;93:1667-1676.
Rose CP, Burgess JH, Cousineau D. Tracer norepinephrine kinetics in coronary circulation of patients with heart failure secondary to chronic pressure and volume overload. J Clin Invest. 1989;84:1267-1275.
Leimbach WN, Wallin BG, Victor RG, Aylward PE, Sundlo¨f G, Mark A. Direct evidence from intraneural recordings for increased central sympathetic outflow in patients with heart failure. Circulation. 1986;73:913-919.
Middlekauf HR, Hamilton MA, Stevenson LW, Mark AL. Independent control of skin and muscle sympathetic nerve activity in patients with heart failure. Circulation. 1994;90:1794-1798.
Swedberg K, Eneroth P, Kjekshus J, Wilhelmsen L. Hormones regulating cardiovascular function in patients with severe congestive heart failure and their relation to mortality: CONSENSUS Trial Study Group. Circulation. 1990;82:1730-1736.
Francis GS, Goldsmith SR, Levine TB, Olivari MT, Cohn JN. The neurohumoral axis in congestive heart failure. Ann Intern Med. 1984;101:370-377.
Kjekshus J. Arrhythmias and mortality in congestive heart failure. Am J Cardiol. 1990;65(19):42I-48I.
Vallbo ÅB, Hagbarth K-E, Torebjo¨rk HE, Wallin BG. Somatosensory, proprioceptive, and sympathetic activity in human peripheral nerves. Physiol Rev. 1979;59:919-957.
Schiller NB, Shah MP, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gutgesell H, Rechek N, Sahn D, Schnittger I, Silverman N, Taijik A. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. J Am Soc Echo Cardiogr. 1989;2:358-367.
Ganz W, Tamura K, Marcus HS, Donoso R, Yoshida S, Swan HJ. Measurement of coronary sinus blood flow by continuous thermodilution in man. Circulation. 1971;44:181-195.
Esler M, Jennings G, Korner P, Blombery P, Sacharias N, Leonard P. Measurement of total and organ-specific norepinephrine kinetics in humans. Am J Physiol. 1984;247:E21-E28.
Chidsey CA, Braunwald E, Morrow AG, Mason DT. Myocardial norepinephrine concentration in man: effects of reserpine and of congestive heart failure. N Engl J Med. 1963;269:653-658.
McCance A, Forfar JC. Cardiac and whole body [3H]noradrenaline kinetics in ischaemic heart disease: contrast between unstable anginal syndromes and pacing induced ischemia. Br Heart J. 1989;61:238-247.
Neri Serneri GG, Boddi M, Arata L, Rostagno C, Dabizzi P, Copp M, Bini M, Lazzerini S, Dagianti A, Gensini GF. Silent ischemia in unstable angina is related to an altered cardiac norepinephrine handling. Circulation. 1993;87:1928-1937.
Eisenhofer G, Esler MD, Meredith IT, Dart A, Cannon RO, Quyyumi AA, Lambert G, Chin J, Jennings GL, Goldstein DS. Sympathetic nervous function in human heart as assessed by cardiac spillovers of dihydroxyphenylglycol and norepinephrine. Circulation. 1992;85:1775-1785.
Esler MD, Turner AG, Kaye DM, Thompson JM, Kingwell BA, Morris M, Lambert GW, Jennings GL, Cox HS, Seals DR. Aging effects on human sympathetic neuronal function. Am J Physiol. 1995;268:R278-R285.
Hainsworth R. Reflexes from the heart. Physiol Rev. 1991;71:617-658.
Thames MD, Kunigawa T, Smith ML, Dibner-Dunlap ME. Abnormalities of baroreflex control in heart failure. J Am Coll Cardiol. 1993;22(suppl A):56A-60A.
Grassi G, Seravalle G, Cattaneo BM, Lanfrachi A, Vailati S, Giannattasio C, Del Bo A, Sala C, Bolla GB, Pozzi M, Mancia G. Sympathetic activation and loss of reflex sympathetic control in mild congestive heart failure. Circulation. 1995;92:3206-3211.
Middlekauff HR, Nitzsche EU, Hamilton MA, Schelbert HR, Fonarow GC, Moriguchi JD, Hage A, Saleh S, Gibbs GG. Evidence for preserved baroreflex control of renal cortical blood flow in humans with advanced heart failure. Circulation.. 1995;92:395-401.