Author + information
- Received July 31, 2001
- Revision received January 3, 2002
- Accepted January 18, 2002
- Published online April 17, 2002.
- ↵*Reprint requests and correspondence:
Dr. John D. Parker, Division of Cardiology, Mount Sinai Hospital, University of Toronto, 600 University Avenue, Suite 1609, Toronto, Ontario, Canada M5G 1X5.
Objectives The goal of this work was to study the effects of short-term infusion of dobutamine on efferent cardiac sympathetic activity.
Background Increased efferent cardiac sympathetic activity is associated with poor outcomes in the setting of congestive heart failure (CHF). Dobutamine is commonly used in the therapy of decompensated CHF. Dobutamine, through its effects on excitatory beta-receptors, may increase cardiac sympathetic activity.
Methods Seven patients with normal left ventricular (LV) function and 13 patients with CHF were studied. A radiotracer technique was used to measure cardiac norepinephrine spillover (CANESP) before and during an intravenous infusion of dobutamine titrated to increase the rate of rise in LV peak positive pressure (+dP/dt) by 40%.
Results Systemic arterial pulse pressure increased significantly in response to dobutamine in the normal LV function group (74 ± 3 mm Hg to 85 ± 3 mm Hg, p = 0.005) but remained unchanged in the CHF group. Dobutamine caused a significant decrease in LV end-diastolic pressure in the CHF group (14 ± 2 mm Hg to 11 ± 2 mm Hg, p = 0.02), an effect not observed in the normal LV group. In the normal LV function group, CANESP did not change in response to dobutamine (75 ± 22 pmol/min vs. 72 ± 22 pmol/min, p = NS). In contrast, dobutamine infusion was associated with a significant reduction in CANESP in patients with CHF (199 ± 43 pmol/min to 128 ± 30 pmol/min, p < 0.0009).
Conclusions Dobutamine infusion caused a significant sympatholytic response in patients with CHF. This sympathetic withdrawal response is probably related to reduction of LV filling pressures and/or activation of ventricular mechanoreceptors with dobutamine infusion.
The detrimental effect of long-term oral positive inotropic therapy in patients with advanced congestive heart failure (CHF) is well established (1–3). Different classes of inotropic agents including beta-adrenergic receptor agonists and phosphodiesterase inhibitors have been evaluated in clinical trials with neutral or negative effects (1–6). Despite the failure of oral positive inotropic agents, intravenous positive inotropic therapy continues to be used extensively in the treatment of decompensated CHF. Dobutamine, a relatively selective beta1-agonist, is one agent in widespread clinical use for this indication (7,8). It is also used in intermittent therapeutic regimens, although its efficacy and safety in this setting remain controversial (9). Multiple case-control studies and case series suggested that this approach improved functional capacity and reduced hospital admissions (10–13). However, a single randomized trial, presented only in abstract form, described an increase in mortality (14).
An important potential mechanism for adverse effects of beta-agonist therapy in patients with CHF is via sympathetic activation. Beta-agonists directly stimulate adrenergic receptors on cardiac myocytes with possible adverse consequences including increased arrhythmias and myocardial ischemia (15). Beta-agonist therapy may also increase endogenous norepinephrine release through stimulation of facilitative prejunctional beta-receptors at several sites within the sympathetic nervous system (16,17). We have previously demonstrated that a beta2-agonist increases cardiac sympathetic activity, presumably through an excitatory prejunctional mechanism (18).
The purpose of the current study was to explore the effects of acute beta-agonist administration on cardiac sympathetic activity. We hypothesized that dobutamine would cause an augmentation in cardiac sympathetic activity through activation of sympathoexcitatory beta-adrenergic receptors. To test this hypothesis we measured cardiac norepinephrine spillover (CANESP) responses during the acute administration of dobutamine in patients with CHF and normal left ventricular (LV) function.
The study population consisted of 20 patients. Seven subjects with normal LV function (LV ejection fraction: 56 ± 2%, mean age: 62 ± 3 years) underwent coronary angiography to investigate a chest pain syndrome. Four of these had coronary artery disease. In this group, six of the patients received beta-blockers which included atenolol (n = 3, daily dose 50 mg), metoprolol (n = 1, daily dose 100 mg), nadolol (n = 1, daily dose 40 mg) and propranolol (n = 1, daily dose 80 mg), and one patient received an angiotensin receptor blocker. Thirteen patients had CHF (LV ejection fraction: 21 ± 3%, mean age: 62 ± 2 years). The etiology of the heart failure was ischemic in nine patients and idiopathic in four patients. The medical therapy consisted of metoprolol (n = 4, mean dose 81 mg), atenolol (n = 1, daily dose 50 mg), carvedilol (n = 1, daily dose 25 mg), digoxin (n = 7) and angiotensin-converting enzyme inhibitor or angiotensin receptor antagonist (n = 11). The patients were on a stable dose of beta-blockers for a minimum of eight weeks before participation in the study. Eight patients had New York Heart Association functional class II symptoms; four had functional class III symptoms and one patient had functional class IV symptoms.
All medications were held on the morning of the study. The University of Toronto Ethical Review Committee for experimentation involving human subjects approved the protocol. Written informed consent was obtained from all participants.
Hemodynamic and coronary flow measurements
A diagnostic left and right heart catheterization from the femoral approach was performed without sedation. After the diagnostic procedure, the pulmonary artery catheter was left in place. A 7F coronary sinus thermodilution catheter (type CCS-7U-90B, Webster Laboratories, Baldwin Park, California) was inserted from an antecubital vein and positioned under fluoroscopic guidance in the coronary sinus for flow measurements and blood sampling. A 7F micromanometer tipped catheter (Millar Industries, Houston, Texas) was placed in the LV. Femoral artery pressure was monitored via an 8F side-arm sheath (Terumo Medical Corp., Elkton, Maryland). Cardiac output was measured by the Fick method. The electrocardiogram (ECG), right atrial pressure, pulmonary artery pressure, femoral artery pressure, LV pressure and its first derivative ([dP/dt] continuous electronic differentiation) were recorded on a strip chart recorder. For each variable, the results were expressed as an average measurement of 10 cardiac cycles in patients with sinus rhythm and 15 cardiac cycles in patients with atrial fibrillation. Left ventricular pressure and the ECG were digitally recorded at 300 Hz with a Macintosh personal computer equipped with a multichannel analog-to-digital converter. Data files were stored to disk for later analysis. Left ventricular peak +dP/dt values were calculated off-line with customized software developed in Labview (Version 3.0, National Instruments Corp., Austin, Texas). Coronary sinus blood flow measurements were performed in duplicate at each measurement point according the method of Ganz et al. (19).
Norepinephrine spillover measurements
Sympathetic activity was estimated by the measurement of CANESP and total body norepinephrine spillover (20). For these measurements, tritiated norepinephrine (1 to 1.2 μCi/min with a 16 μCi priming bolus of L-[2,5,6-3H] norepinephrine; New England Nuclear, Boston, Massachusetts) was infused into a peripheral vein to steady-state concentration in plasma. Norepinephrine clearance and spillover rates were calculated as previously described (21,22).
Analysis of plasma catecholamines
Plasma catecholamine concentrations were measured by high-performance liquid chromatography (HPLC) with electrochemical detection. Fractions from the HPLC effluent containing tritium-labeled norepinephrine were assayed by liquid scintillation spectroscopy. These analyses were performed by established methods in our laboratory by personnel blinded to patient status (21,22).
After the diagnostic heart catheterization and insertion of catheters for hemodynamic monitoring, the patient was left undisturbed for a minimum of 20 min for tritium-labeled norepinephrine to reach steady-state. Baseline hemodynamic and coronary blood flow measurements were obtained along with measures of CANESP and total body norepinephrine spillover. Subsequently, an intravenous infusion of dobutamine was initiated starting at 2.5 μg·kg−1·min−1. The dobutamine infusion rate was increased until LV +dP/dt had increased by 40%. Hemodynamic, coronary sinus blood flow, CANESP and total body norepinephrine spillover were reassessed 20 min after this increase in LV +dP/dt had been achieved. The dobutamine infusion was then discontinued, and recovery measurements were performed once LV +dP/dt had returned to baseline values.
Data are presented as the mean value ± SEM. The analysis was performed using SAS (release 8.1, SAS Institute Inc., Cary, North Carolina). Between-group comparisons of baseline characteristics were performed with unpaired ttest. The effects of dobutamine were analyzed using a general linear modelling procedure. The model allowed for analysis of the effects of dobutamine within groups and also determined if there was significant group by treatment interactions. A value of p ≤ 0.05 was required for statistical significance.
Baseline hemodynamic and neurochemical characteristics are summarized in the Table 1. The CHF group had reduced cardiac index and elevated central filling pressures.
The final dobutamine infusion rate was 2.7 ± 0.2 μg/kg per min in the normal LV function group and 5.6 ± 0.6 μg/kg per min in the CHF group (p = 0.002, normal LV vs. CHF group). The LV peak +dP/dt increased by 38 ± 8% in the normal LV function group and by 42 ± 5% in the CHF group (Table 2). The analysis revealed that there was no significant interaction between the effect of dobutamine and patient group (p = NS, normal vs. CHF). In the normal LV function group, systemic arterial pressure parameters increased in response to dobutamine, with the increase in pulse pressure being the most prominent (74 ± 3 mm Hg to 85 ± 3 mm Hg, p = 0.005). Dobutamine infusion caused smaller increases in systemic arterial pressure parameters in the CHF group, none of which were statistically significant. There was no significant interaction between the systemic arterial blood pressure effects of dobutamine and patient group. Dobutamine did cause a significant decrease in LV end-diastolic pressure during dobutamine infusion in the CHF group (14 ± 2 mm Hg to 11 ± 2 mm Hg, p = 0.02). There was no significant change in LV end-diastolic pressure in the normal LV function group; however, there was a significant interaction between the effect of dobutamine and patient group, indicating that the LV end-diastolic pressure response in the CHF group was significantly different from the response in the normal LV function group. Hemodynamic measurements returned to baseline after discontinuation of dobutamine.
Cardiac sympathetic responses
In the normal LV function group, dobutamine caused no change in CANESP (0 ± 11%, p = NS). In the CHF group, there was a large and highly significant reduction in CANESP with dobutamine (−36 ± 8%, p < 0.0009). There was a significant group by treatment interaction indicating that the response of CANESP in the CHF group was significantly different from the response observed in the normal LV function group (p = 0.01; normal LV function vs. CHF group) (Fig 1). There was no difference between recontrol CANESP measurements and baseline measurements.
Generalized sympathetic responses
Dobutamine had no significant effect on total body norepinephrine spillover in either group (Table 3). In both the normal LV function group and the CHF group, dobutamine infusion was associated with a significant increase in total body norepinephrine clearance (Table 3).
Effects of beta-blocker treatment on responses to dobutamine
Six patients in the CHF group were being treated with beta-blockers. There was no difference in the final dobutamine infusion dose between the patients on beta-blockers and those who were not (5.8 ± 1.1 μg/kg per min vs. 5.3 ± 0.6 μg/kg/min, no beta-blocker vs. beta-blocker, p = NS). Heart rate and cardiac output increased significantly in both groups with dobutamine infusion with no effect of beta-blockade on the magnitude of response. Similarly, the CANESP responses to dobutamine were not affected by beta-blocker treatment (−39 ± 12% vs. −33 ± 10%, no beta-blocker vs. beta-blocker, p = NS).
This investigation provides the first human in vivo description of the effects of dobutamine on CANESP rate, an indirect index of cardiac sympathetic efferent neuronal activity. Dobutamine was given intravenously and in clinically relevant doses. We have demonstrated that dobutamine caused a significant reduction in CANESP, an accepted index of efferent postganglionic nerve activity in patients with CHF.
Sympathoexcitatory beta-adrenergic receptors
In this investigation we hypothesized that dobutamine administration would cause an increase in cardiac sympathetic activity through its effects, primarily, on beta1-adrenergic sympathoexcitatory receptors located on intramural and juxtacardiac sympathetic ganglia and efferent postganglionic neurons. These receptors have been described at a number of levels within the intrathoracic sympathetic nervous system. Classically, prejunctional beta2-adrenergic receptors are described that facilitate neuronal norepinephrine release. These receptors, and their functional significance, have been described in a number of animals and humans (16,17). In a previous report from our laboratory, we documented the presence and functional importance of cardiac, sympathoexcitatory beta2-adrenergic receptors in humans with normal ventricular function (18). In that study, we demonstrated that salbutamol infusion increased CANESP by 124% and that the increase in norepinephrine release mediated by beta2-adrenergic receptor stimulation made an important contribution to the inotropic responses observed (18). Recently, intracardiac, pericardiac and intrathoracic sympathetic ganglia have been described that contain both beta1- and beta2-adrenergic receptors that are involved in the control of cardiac sympathetic responses (16,17,23). These cardiac and intrathoracic sympathetic neurons have been termed the “intrinsic cardiac sympathetic nervous system” and are felt to play an important role in the peripheral control of cardiac sympathetic responses (23). Animal experiments have demonstrated that beta1-adrenegic receptors in these neuronal systems can modulate an increase in cardiac sympathetic activity (23).
Dobutamine effects in the normal LV group
In the normal LV function group dobutamine had no significant effect on CANESP. Despite this neutral response, we cannot completely discount that a cardiac sympathoexcitatory effect occurred. Indeed, it is possible that a direct sympathoexcitatory effect was offset by afferent reflexes that were associated with a balanced withdrawal of central sympathetic outflow. There would appear to be two potential mechanisms. First, dobutamine caused an increase in systemic arterial pressure. Such stimulation of arterial baroreceptors could have caused a reflex decrease in sympathetic outflow. Second, the infusion of dobutamine may have caused activation of ventricular mechanoreceptors. Stimulation of these receptors has been shown to elicit reflex decreases in sympathetic outflow, and the activity of these receptors has been shown to increase in response to positive inotropic therapy (24,25).
Dobutamine effects in the CHF group
Contrary to our hypothesis, dobutamine caused a 36 ± 8% reduction in CANESP in the CHF group. This observation suggests that sympathoinhibitory reflex responses are more relevant than any direct sympathoexcitatory effects of dobutamine in this patient population. Three possible reflex mechanisms to account for sympathoinhibition can be suggested. As with the normal LV function group, it is possible that arterial baroreceptors played a role in the observed sympathoinhibitory response. This seems unlikely because dobutamine had essentially no effect on systemic arterial blood pressure, while somewhat greater increases in systemic arterial pressure in the normal LV function group had no effect on CANESP. Second, it is also possible that the sympathoinhibitory effect of dobutamine in patients with CHF was caused by a reduction in LV filling pressure. Dobutamine caused a significant reduction in LV end-diastolic pressure without reducing systemic arterial blood pressure. In a previous report using lower body negative suction, we demonstrated that a decrease in ventricular filling pressure in the absence of a fall in systemic arterial blood pressure was associated with a significant reduction in cardiac sympathetic activity in patients with CHF (26). In the present study, the reduction in LV filling pressure with dobutamine was somewhat smaller than that seen with lower body negative suction. Furthermore, right-sided filling pressures did not change with dobutamine, while they did decrease significantly with lower body negative suction. Therefore, although it is possible that the reduction in LV filling pressure made some contribution to the observed result, we do not believe that the entire reduction in CANESP occurred because of this small reduction in left-sided filling pressure. Ventricular mechanoreceptor activation in response to increased contractility provides a third potential explanation for the observed decrease in cardiac sympathetic activity (24,27,28). Activation of these receptors is associated with both attenuation in the sympathetic outflow to the peripheral circulation and augmentation of vagal efferent cardiac nerve activity (24). This increase in efferent cardiac parasympathetic activity could explain the differential effects of dobutamine on cardiac sympathetic norepinephrine spillover in the two groups. We have previously shown that muscarinic receptor activation caused a reduction in CANESP in patients with CHF, an effect not observed in patients with normal LV function (29). In the CHF group there was a significant increase in both heart rate and LV +dP/dt despite concurrent evidence of sympathetic withdrawal. These differential responses are not inconsistent because the former was the result of dobutamine’s postjunctional beta-agonist activity.
There have been no previous investigations of the response of the cardiac sympathetic nervous system to inotropic agents like dobutamine. In a previous study, Colucci et al. (30)demonstrated that systemic plasma norepinephrine decreases in response to short-term infusions of dobutamine in patients with CHF. These observations suggested that dobutamine decreased systemic sympathetic activity in CHF. Similarly, intracoronary and intravenous infusions of milrinone in patients with CHF have been shown to reduce systemic plasma norepinephrine concentrations (31). Importantly, it is now recognized that the observed decrease in plasma norepinephrine may have occurred secondary to increases in norepinephrine clearance. Increases in cardiac output are consistently observed in response to dobutamine and milrinone in the setting of CHF, and we have previously demonstrated that increases in cardiac output correlate closely with increases in norepinephrine clearance (32).
It is important to consider the limitations of this study. We examined the acute effects of intravenous dobutamine. The long-term effects of dobutamine infusion on cardiac sympathetic activity in patients with CHF cannot be inferred from the present results. The final infusion rate of dobutamine that was required to achieve the same increase in LV +dP/dt was almost twice that required in the normal LV function group. This is not surprising, given the well-documented beta1-receptor downregulation in the myocardium of patients with CHF (33). Direct measures of efferent sympathetic outflow to the heart are not available in humans. We have hypothesized that the reduction in CANESP that occurred in the CHF group was mediated by a reduction in central sympathetic outflow. Because cardiac efferent nerve activity cannot be recorded in vivo, we cannot be definitive about this mechanism. For example, our measure of norepinephrine spillover is an indirect estimate of changes in norepinephrine release, and we cannot rule out that some of the observed reduction in CANESP was mediated by changes in norepinephrine uptake mechanisms.
In summary, this investigation has demonstrated that a short-term infusion of dobutamine caused a significant reduction in efferent cardiac sympathetic activity in patients with CHF. These observations are relevant given the common clinical usage of dobutamine in this patient population. The authors would like to emphasize that these observations should not change present perceptions concerning the safety of using inotropic agents in the treatment of patients with CHF. At the moment, the utility of dobutamine and other inotropic agents in the short-term management of individual patients has an accepted place in clinical practice. Their utility and safety when used routinely in decompensated patients or intermittently as part of long-term maintenance care remains controversial (14,34). Our observation serves to establish that relatively selective beta1-adrenergic agonists like dobutamine do not cause an increase in cardiac sympathetic efferent neuronal activity when used in patients with chronic heart failure.
The authors thank the staff of the Bayer Cardiovascular Clinical Research Laboratory of Mount Sinai Hospital for their help in the completion of these studies.
☆ Dr. Al-Hesayen holds a research fellowship award from the Heart and Stroke Foundation of Ontario. Dr. Azevedo held a research fellowship award from AstraZeneca/Heart and Stroke Scientific Research Corporation of Canada. This study was funded by an operating grant from the Heart and Stroke Foundation of Ontario (grant No. T3696) and from Bayer Inc.
- cardiac norepinephrine spillover
- congestive heart failure
- high-performance liquid chromatography
- left ventricle/ventricular
- rate of rise in left ventricular peak positive pressure
- Received July 31, 2001.
- Revision received January 3, 2002.
- Accepted January 18, 2002.
- American College of Cardiology Foundation
- Dawson J.R,
- Canepa-Anson R,
- Kuan P,
- Reuben S.R,
- Poole-Wilson P.A,
- Sutton G.C
- Massie B.M,
- Berk M.R,
- Brozena S.C,
- et al.
- Lubsen J,
- Just H,
- Hjalmarsson A.C,
- et al.
- Tuttle R.R,
- Mills J
- Leier C.V,
- Huss P,
- Lewis R.P,
- Unverferth D.V
- Packer M
- Huang M.H,
- Smith F.M,
- Armour J.A
- Watson-Wright W,
- Boudreau G,
- Cardinal R,
- Armour J.A
- Newton G.E,
- Azevedo E.R,
- Parker J.D
- Ganz W,
- Tamura K,
- Marcus H.S,
- Donoso R,
- Yoshida S,
- Swan H.J
- Newton G.E,
- Tong J.H,
- Schofield A.M,
- Baines A.D,
- Floras J.S,
- Parker J.D
- Newton G.E,
- Parker J.D
- Azevedo E.R,
- Newton G.E,
- Floras J.S,
- Parker J.D
- Thoren P.N
- Azevedo E.R,
- Parker J.D
- Colucci W.S,
- Denniss A.R,
- Leatherman G.F,
- et al.
- Schobel H.P,
- Oren R.M,
- Roach P.J,
- Mark A.L,
- Ferguson D.W
- Newton G.E,
- Parker J.D