Regulation of Norepinephrine Release by β²-Adrenergic Receptors during Halothane Anesthesia

Approval for the study was obtained from Vanderbilt University Animal Care Committee. Seven female mongrel dogs (27 plus/minus 1.9 kg, mean plus/minus standard error of the mean) were studied. At least 7 days before the experiment, the animals underwent surgery for the placement of an ultrasonic flow probe (Transonic Systems, Ithaca, NY) in the left external iliac artery and insertion of intravascular catheters bilaterally in the femoral artery and vein. Catheters were placed in the right femoral artery and vein and advanced so that the catheter tips were in the distal aorta and inferior vena cava. Another right femoral venous catheter was advanced to the level of the intrathoracic inferior vena cava for measurement of central venous pressure. Implantation of the left femoral arterial and venous catheters differed from those on the right in that the catheters were introduced through side tributaries to preserve blood flow through the femoral vessels of the hind limb to be studied. These catheters were advanced into the external iliac vessels, but to avoid disrupting flow measurements they were left well distal to the flow probe. Anesthesia for the surgery was performed with thiopental (20-25 mg/kg) followed by maintenance of anesthesia with halothane.

To allow observed changes in NE kinetics to be attributed to the anesthetic state, it was essential that the baseline control values be measured in calm, resting, conscious dogs. In five or six training sessions, each lasting about 1 h, the dog was familiarized with the experimental room and procedures. Relaxed posture, slow deep respirations, and spontaneous eye closure were used as indicators of adequate training.

On each occasion the dogs were studied first awake and then during halothane anesthesia (1.0 minimum alveolar concentration [MAC] = 0.9%) so that each dog acted as her own control. Systemic and local hind-limb NE spillover and clearance rates were measured by isotope dilution [6]in awake and halothane-anesthetized dogs as we have previously described. [5,10]For each set of measurements a separate 50-min intravenous infusion of NE-L-[ring-2,5,6-sup 3 Hydrogen] ([sup 3 Hydrogen]NE) was performed. [sup 3 Hydrogen]NE of specific radioactivity 43.7-56.9 Ci/mmol (New England Nuclear, Wilmington, DE) was diluted to 1 micro Ci/ml in 0.9% sodium chloride with 1 mg/ml ascorbic acid and infused into the right femoral vein; 15 micro Ci in the 1st min was followed by 0.6 micro Ci/min thereafter. We have previously shown this regimen to be a true tracer dose, devoid of pharmacologic effects. [5]Thirty minutes was allowed for plasma [sup 3 Hydrogen]NE to reach steady state [5]before hemodynamic measurement and blood sampling procedures were performed.

After 30 min of [sup 3 Hydrogen]NE infusion when the dog was conscious, systolic and diastolic aortic arterial blood pressure, heart rate, central venous pressure, and hind-limb blood flow were measured. Simultaneous arterial and venous blood samples were then taken from the study hind limb (left) for measurement of endogenous and [sup 3 Hydrogen]NE concentrations. Blood was collected in cooled ethyleneglycol-bis-(beta-aminoethyl ether) tetra-acetic acid and glutathione (CAT-A-KIT blood collection tubes, Amersham, Arlington Heights, IL) and centrifuged, and the plasma was stored at -20 degrees Celsius until assayed. NE concentrations were measured as previously reported [18]by high-performance liquid chromatography; the effluent coinciding with the NE peak was collected and counted by liquid scintillation. This approach allowed the assay of plasma [sup 3 Hydrogen]NE concentration without interference from tritiated metabolites.

During the above control period, the arterial catheter was maintained patent by a continuous infusion of 0.9% sodium chloride at a rate of 60 ml/h. When all of the baseline control measurements had been taken, isoproterenol was then infused intraarterially into the study limb at two dosages, 30 and 80 ng/min. During intraarterial administration of isoproterenol, the total flow rate through the arterial catheter was maintained constant at 60 ml/h by adjustment of the sodium chloride infusion rate. Each dosage of isoproterenol was administered for 7 min; hind-limb blood flow was measured; and simultaneous arterial (10 ml) and venous (10 ml) blood samples were again drawn at isoproterenol dosages of 30 and 80 ng/min.

On completion of the awake stage, the [sup 3 Hydrogen]NE and isoproterenol infusions were stopped, and anesthesia was induced with thiopental 20-25 mg/kg, followed by tracheal intubation and pulmonary ventilation with halothane (1.0 MAC) in oxygen. Continuous measurement of end-tidal concentrations of carbon dioxide and halothane (254 monitor, Datex), with regular arterial blood gas analysis, allowed adjustment of respiratory, rate to maintain arterial carbon dioxide tension between 30 and 40 mmHg. Throughout anesthesia, 0.9% saline was administered at 4 ml *symbol* kg sup -1 *symbol* h sup -1. All blood samples were replaced with twice their volume of 0.9% saline.

Sixty minutes was allowed for equilibration after induction of anesthesia before the [sup 3 Hydrogen]NE infusion was recommenced. The study protocol was repeated with measurements made at baseline and then during two intraarterial infusion rates of isoproterenol, 30 and 80 ng/min.

NE pharmacokinetic parameters were calculated as follows: Equation 1but also Equation 2Therefore, Equation 3Processes of NE removal and of NE release occur simultaneously in any vascular bed, so that simple measurement of the net arteriovenous increment of NE concentration underestimates regional NE spillover. A correction factor must be added to allow for the NE removal during passage through the hind limb. Equation 4where FE = fractional extraction of NE by the hind limb; A* = arterial concentration of [sup 3 Hydrogen]NE (disintegrations per minute per milliliter); V* = venous concentration of [sup 3 Hydrogen]NE (disintegrations per minute per milliliter), and; Equation 5where A = arterial concentration of endogenous NE (ng/ml); V = venous concentration of endogenous NE (ng/ml); and Q with dot = hind-limb plasma flow (ml/min). Equation 6where hematocrit is expressed as a fraction and hind-limb NE clearance is expressed in milliliters per minute. Plasma NE clearance from the whole body (i.e., systemic clearance) was calculated at steady state as: Equation 7where systemic NE clearance is expressed in milliliters per minute and [sup 3 Hydrogen]NE infusion rate is expressed in disintegrations per minute per minute. The rate at which NE entered plasma for the whole body (i.e., systemic spillover) was calculated at steady state as Equation 8.

Conversion of blood flow to plasma flow required determination of arterial hematocrit. Hematocrit samples were collected in capillary tubes prepared with heparin; spun for 10 min in a Micro Hematocrit centrifuge; and read on a Micro-capillary Reader (both from Damon, IEC Division, Needham Heights, MA). Measurements were made at three times: (1) at the beginning of the awake stage; (2) at the end of the awake stage, before the anesthesia stage; and (3) on completion of the study. The average of the first and second measurements was used for calculation of the hind-limb plasma flows during the awake stage, and the average of the second and third measurements was used for the anesthesia stage. Data were analyzed by repeated measures analysis of variance followed by Student's paired t test or Wilcoxon's signed rank test as appropriate. P < 0.05 was accepted as the minimal level of significance.

The number of observations at each data point varies as indicated in the figures. Five dogs received both isoproterenol dosages, one only 30 ng/min, and another only 80 ng/min. In addition, arterial pressure measurements were unavailable for one dog during halothane anesthesia because of occlusion of the catheter. Table 1shows the baseline (i.e., before intraarterial infusion of isoproterenol) values for the various hemodynamic and NE pharmacokinetic parameters both awake and during 1.0 MAC halothane anesthesia. The low values during the awake period for systolic and diastolic blood pressure, heart rate, and arterial NE concentration confirm that we achieved our objective of a calm resting state in the dogs. Halothane anesthesia resulted in a significant decrease in systolic blood pressure (P < 0.001) without a change in diastolic blood pressure or heart rate. Consistent with our previous report, [5]arterial NE concentration was decreased by halothane because of a 38% reduction (P < 0.01) in systemic NE spillover accompanied by a relatively small decrease in systemic NE clearance. There was no significant difference between hind-limb blood flow awake and during halothane anesthesia, probably because of a significant decrease in hind-limb vascular resistance produced by halothane (Table 1).

Local intraarterial infusion of isoproterenol resulted in a significant dosage-related increase in hind-limb blood flow in all dogs (Figure 2). Because arterial blood pressure did not increase, this isoproterenol-induced increase in hind-limb blood flow resulted from reduced hind-limb vascular resistance (Figure 2) and can be attributed to postjunctional beta²-adrenergic receptor stimulation. This response was not significantly affected by halothane anesthesia (Figure 2), indicating a lack of effect of halothane on postjunctional beta²receptor-mediated vasodilation.

Stimulation of prejunctional beta²-adrenergic receptors by intraarterial isoproterenol markedly increased local hind-limb NF spillover in awake dogs (Figure 3). Isoproterenol, 80 ng/min, increased hind-limb NE spillover 288% from an awake baseline spillover of 5.2 plus/minus 0.8 to 13.2 plus/minus 1.7 ng/min (P < 0.02). Halothane anesthesia produced a 43% decrease in hind-limb spillover compared with the awake baseline value (P < 0.05). Furthermore, halothane abolished isoproterenol-mediated facilitation of NE release; NE spillover after infusion of isoproterenol. 80 ng/min, was 4.8 plus/minus 2.3 ng/min during 1.0 MAC halothane compared with 13.2 plus/minus 1.7 ng/min in the awake state (P < 0.05) (Figure 3). NE clearance by the hind limb was unchanged during isoproterenol infusion and was not significantly affected by halothane anesthesia (Figure 3).

(Figure 4) shows the effect of halothane anesthesia and intraarterial administration of isoproterenol on systemic NE kinetics. Systemic NE spillover did not significantly change during 30-ng/min administration of isoproterenol but increased 28 plus/minus 10% (P < 0.05) when isoproterenol 80 ng/min was administered. Figure 5shows the arterial blood pressure and heart rate measurements during the study. There was no significant change in systolic blood pressure during intraarterial isoproterenol infusion, both awake and during halothane anesthesia, but diastolic blood pressure was significantly reduced (by 9 mmHg awake and 4 mmHg during halothane) only during infusion of the greater dosage of isoproterenol (80 ng/min).

In addition to the observed increase in hind-limb spillover, there was also a smaller increase in systemic spillover (Figure 6). However, when the contribution of hind-limb NE spillover is subtracted from systemic NE spillover, the systemic NE spillover at 80-ng/min isoproterenol infusion (172 plus/minus 22 ng/min) is not significantly different from baseline (152 plus/minus 29 ng/min). Thus, although the increase in hind-limb spillover contributes to the increase seen in systemic spillover and may account for that increase, systemic spillover does not contribute to the measured increase in hind-limb spillover. Thus, the increase in systemic spillover cannot account for the change seen in hind-limb spillover, because at the 30-ng/min isoproterenol dosage, hind-limb spillover increased by 98 plus/minus 26% (P < 0.05), whereas there was no significant increase in systemic spillover (Figure 6). At the greater isoproterenol dosage, 80 ng/min, when significant increase in systemic spillover was seen, that increase was only 28 plus/minus 10% in contrast to the 288 plus/minus 150% increase in hind-limb spillover.

This study was designed to test the hypothesis that prejunctional beta²-adrenergic receptors facilitate neuronal NE release in response to stimulation by isoproterenol and that this effect is abolished by halothane anesthesia. We have clearly shown that intraarterial isoproterenol administration is a potent stimulus for NE release and that halothane markedly impairs the prejunctional beta²receptor-induced increase in NE release but has no effect on postsynaptic vasodilation.

Anesthetic agents produce cardiovascular depression in part through their effects on the sympathetic nervous system. We have previously shown that inhalational and intravenous anesthetics decrease the rate of NE release into the circulation. [5,10]Prejunctional receptors modulate NE release; they may be targets for blood-borne humoral agents such as epinephrine, for example, and hence they act as a link in presynaptic feedback control of NE release. There is now a large body of evidence that epinephrine can augment NE release in humans, [11,19,20]Small, repetitive increases in circulating epinephrine from the adrenal medulla in response to stress for example, can augment NE release by stimulating prejunctional beta²-adrenergic receptors that facilitate NE release from sympathetic nerve endings. Sustained indirect aftereffects of epinephrine also may occur by the neuronal uptake of epinephrine and later release as a cotransmitter when autofacilitation of NE release by NE may result. Several lines of experimental evidence have thus led to the hypothesis that epinephrine stimulates prejunctional beta²-adrenergic receptors that facilitate NE release from sympathetic nerve endings and that epinephrine augments NE release in humans. [11,19,20]Systemic infusion of epinephrine, in dosages that result in epinephrine concentrations seen under physiological stress conditions, has been shown to produce a delayed or protracted pressor response, indicating the functional importance of prejunctional beta²-adrenergic receptors. [19]Prejunctional beta²receptors may play an important role in the regulation of NE release in vivo and in particular during anesthesia and surgery.

This study has shown that hind-limb intraarterial infusion of isoproterenol results in an increase in NE spillover into the hind limb; stimulation of prejunctional beta²-adrenergic receptors by isoproterenol facilitates NE release. This result is consistent with our recent finding of beta²-adrenergic receptor-mediated release of NE in the human forearm. [17]We have therefore demonstrated that stimulation of the prejunctional beta²receptor in humans by isoproterenol [17]and now in a dog model is a potent stimulus to NE release. A major goal of the current study was to expand the findings of presynaptic regulation of NE release by examining its modulation by halothane administration. We have demonstrated that halothane antagonized, in fact almost abolished, the isoproterenol-induced increase in NE spillover. In contrast, halothane did not alter the post-junctional effect on blood flow, so that the isoproterenol-induced increase in hind-limb blood flow was similar before and after halothane. Previous studies in vitro [21-23]have implicated the prejunctional receptor as a potential site for inhibition of sympathetic function by halothane. The current study demonstrates that halothane affects presynaptic regulation of NE release in vivo. We have thus identified a new site of action for halothane-induced sympathetic inhibition, in addition to its well-recognized actions on baroreceptor function, [3]central sympathetic control, [3,4]and ganglionic transmission. [1-3].

Perioperative myocardial ischemia is a major cause of morbidity after anesthesia and surgery. Intraoperative risk factors include hypertension, tachycardia and dysrhythmias. Anesthetic implications of the response to surgical stress induced changes in catecholamines and other hormones are important for patients with marginal cardiac function who are unable to tolerate the increased myocardial oxygen demand associated with marked increases in sympathetic stimulation. Perioperative-induced increases in stress hormones such as NE and epinephrine may precipitate myocardial ischemia, and repetitive increases in circulating epinephrine may cause hypertension indirectly, by stimulating prejunctional beta²-adrenergic receptors that facilitate NE release from sympathetic nerve endings (Figure 1). Therapeutic strategies that modulate the stress response would be expected to reduce the incidence of ischemic episodes. Therefore, the effect of anesthesia itself on presynaptic function has widespread application. We suggest that stress-mediated epinephrine through presynaptic beta²receptors facilitates and augments sustained NE release with increased risk of perioperative cardiac ischemia. Blockade of this effect by anesthetics such as halothane would be expected to reduce NE release. We also suggest that not all anesthetics may inhibit isoproterenol-induced increase in NE and also speculate that demonstration of the ability of propranolol to antagonize beta²receptor-induced increase in NE release would be of therapeutic importance as it would provide a scientific rationale for the development of selective beta²-adrenergic receptor antagonists as premedicants.

The use of the isotope dilution technique examines sympathetic function at a different site from that examined by recording sympathetic nerve traffic. [24,25]The measurement of sympathetic nerve traffic measures sympathetic nerve activity proximal to the site of NE release and is unaffected by changes in presynaptic function such as those in this study. Although halothane altered the response to presynaptic beta²-adrenergic receptor stimulation, there was no effect on the postsynaptic receptor that mediates increases in blood flow. Increase in blood flow in response to isoproterenol was identical before and after halothane, implying that the postsynaptic beta²-receptor stimulation response was unaffected by halothane and that differences in blood flow before and after halothane in response to isoproterenol could not explain the altered NE spillover shown in this study.

In summary, we have demonstrated that stimulation of prejunctional beta²-adrenergic receptors has a profound effect on NE release and that halothane acts directly at the peripheral sympathetic nerve terminal to inhibit beta²-adrenergic receptor-mediated NE release. We also have confirmed our previous finding that halothane decreases systemic spillover and clearance of NE and have identified the prejunctional beta²-adrenergic receptor as an important site of action for halothane.