Should Norepinephrine, Rather than Phenylephrine, Be Considered the Primary Vasopressor in Anesthetic Practice?

The induction of general anesthesia is associated with sympatholysis1 and a decrease in circulating norepinephrine (NE) and epinephrine (E) concentrations.2,3 Yet, the associated hypotension is commonly treated with phenylephrine (PE), a synthetic vasoconstrictor.4,5 Theoretically, NE might better combat this general anesthesia–induced hypotension by restoring decreased circulating concentrations of this catecholamine and maintaining cardiac output (CO). However, NE is rarely used in these circumstances. In this article, I will propose that patients might be better served by using NE rather than PE as the primary vasopressor to combat hypotension during general anesthesia. PHARMACOLOGY The pharmacology of PE and NE is well known and is summarized in Table 1.6–8 PE is now readily accepted as a first-line agent to combat hypotension from both general and spinal anesthesia.4,9 In contrast, NE has been viewed with some trepidation. This is reflected in the moniker, “leav-em-dead” when referring to its common trade name, Levophed.10 Thus, NE use has been largely confined to “sicker patients” and restricted to cardiac anesthesia and the management of sepsis.11,12Table 1.: Pharmacology of Phenylephrine Versus NorepinephrinePE, because of α1 selectivity, has purely vasoconstrictor effects on vascular beds.6,13 Arteriolar vasoconstriction results in an increase in arterial blood pressure, which in turn results in a baroreceptor-triggered, vagally mediated, decrease in heart rate and CO. In venous capacitance vessels, such vasoconstriction may increase vascular return and so potentially improve CO. However, this effect on CO may be impaired with PE because of an increase in venous resistance, limiting venous return to the heart.13,14 An increase in venous resistance is not found with NE, which has both α1 and β1 effects as well as some β2 effects,7 resulting in a β-induced venous vascular relaxing effect on venous resistance. Hence, NE better enhances venous return to the heart and so CO.14,15 The β1 effects also result in a positive inotropic effect on the myocardium with little to no chronotropic effect when compared with E.6 Thus, NE increases arterial blood pressure through arteriolar vasoconstriction and an increase (or maintenance) of heart rate, stroke volume, and CO.16–19 NE is the neurotransmitter of the sympathetic nervous system.20 Plasma levels are reported as being 264 ± 162 pg/mL (mean ± SD; n = 27) before induction of anesthesia with thiopental and 127 ± 78 pg/mL afterward, during isoflurane and sufentanil anesthesia.2 Only 2% of circulating NE is attributed to release from the adrenal medulla.20,21 Plasma NE concentrations are dependent on the net amount of “spillover” from sympathetic nerve terminals and NE clearance by organ tissue.20 The net NE spillover into surrounding venous beds is a function of immediate neuronal reuptake. Spillover may be increased by disease (hypothyroidism, depression, renal failure) and drugs (chlorthalidone) or decreased by them (hyperthyroidism, clonidine, desipramine, and PE).20 Plasma clearance by the lungs (45%), kidneys (8%), and hepatomesenteric system (25%) is dependent on CO.20 Hence, cardiac failure increases plasma NE levels.20,22 NE is metabolized by monamine oxidase and catechol-O-methyltransferase7 to vanillylmandelic acid. NE has a half-life of 1 to 2 minutes.7,23 On the basis of human in vivo and in vitro vasoconstrictor studies,24 investigators have regarded the potency ratio of NE to PE to be 20:1.15,25 Hence, IV NE (5 μg/mL) was compared with PE (100 μg/mL) administration in a recent study investigating hypotension from spinal anesthesia in obstetric patients.15 Ephedrine is a third vasopressor that has α1, β1, and β2 effects.7 It can be used to increase heart rate and arterial blood pressure. These cardiovascular effects have been shown to be variable and delayed, and tachyphylaxis occurs readily, so ephedrine cannot be used as a continuous infusion.26 This is ascribed to the fact that ephedrine acts indirectly, primarily by releasing endogenous catecholamines from adrenergic nerve endings.27 This indirect mechanism of action makes ephedrine a less reliable vasopressor than direct acting agents such as NE and PE.26 This is because ephedrine’s vasopressor responses may vary depending on preoperative medications such as clonidine28 and the type of anesthesia administered.28 Kanaya et al. compared the pressor response from ephedrine (7.5 mg) administered before and during the administration of propofol (100 μg/kg/min) or sevoflurane (2%) as the sole anesthetic. Compared with the awake state, propofol anesthesia significantly enhanced the pressor effect from ephedrine, whereas sevoflurane anesthesia did not. The authors attributed this to differing effects of propofol and sevoflurane on sympathetic nerve activity.28 Ephedrine also has a significantly delayed onset of action when compared with PE in the treatment of hypotension from spinal anesthesia for cesarean delivery. Dyer et al.8 demonstrated that the maximum effect on arterial blood pressure occurred 90 ± 38 seconds (mean ± SD) after ephedrine (10 mg) administration and 62 ± 35 seconds after PE (80 μg) administration. Ephedrine is no longer recommended in such circumstances because of the greater fetal acidosis encountered when compared with PE.29,30 Ephedrine is still used to combat hypotension from general anesthesia.31 Ephedrine (100 μg/kg) when compared with PE (2 μg/kg) resulted in statistically faster heart rates 65 ± 11 vs 57 ± 9 bpm (mean ± SD), respectively, for similar increases in mean arterial blood pressure (MAP).32 In addition, Goertz et al.33 showed that ephedrine (100 μg/kg) resulted in significantly increased heart rates (which might not be clinically important) from a baseline of 61 (55–67) to 65 (68–73) bpm (mean [95% confidence interval]) in patients experiencing hypotension from combined general and high thoracic anesthesia. This was associated with an increase in left ventricular performance as measured by an increase in left ventricular area ejection fraction from 33% to 49%. Given this hemodynamic profile, ephedrine should be considered for treating hypotension from general anesthesia when an increase in heart rate and ventricular performance is required. PERIOPERATIVE CIRCULATING CATECHOLAMINE LEVELS To inform the administration of NE to combat hypotension, it is useful to understand how NE and E levels change in the perioperative period. Catecholamine levels are critically dependent on the type and depth of the anesthetic agent administered, the patients’ concurrent diseases, the muscle relaxants used, and the stage of the surgical procedure. Induction of anesthesia with thiopental, diazepam, and midazolam is associated with significantly decreased circulating NE and E concentrations.2,3,34 Further, NE reduction by thiopental is enhanced in uremic patients.35 Ketamine induction is associated with increased levels of catecholamines and thus rarely associated with hypotension.36 Volatile anesthetic administration with 1.5 minimum alveolar concentration desflurane is associated with significantly higher NE levels than 1.5 minimum alveolar concentration isoflurane and sevoflurane administration.37 Rapid increases in desflurane and isoflurane concentrations increase circulating catecholamines (with associated hypertension), whereas sevoflurane does not.38,39 Succinylcholine34 increases NE and E, pancuronium decreases NE concentrations,40 whereas rocuronium and vecuronium have no effect.41 After anesthesia induction with thiopental or propofol, the stress response of intubation increases NE and E plasma concentrations to levels often higher than preinduction levels.42,43 At surgical incision, there is again an increase in circulating NE and E to levels equal to44 or higher than45 found preoperatively. These levels can be modulated by different anesthetic techniques such as neurolept anesthesia or a volatile-based anesthesia.23,46 Intraoperative NE and E levels are also dependent on the type of surgery. In spine surgery (with propofol [6–12 mg/kg/h] or isoflurane [0.6%–0.7%]), lower levels of circulating catecholamines have been demonstrated when compared with preinduction baseline levels.47 During major abdominal surgery (gastrectomy/colectomy, n = 7) under sevoflurane (1.5%–3%) with N2O (70%), significantly higher catecholamine levels are found when compared with preinduction levels.45 Emergence and extubation in such patients is associated with levels of NE and E, 3 and 5 times higher than preinduction values, respectively.45 COMPARISON OF NOREPINEPHRINE AND PHENYLEPHRINE TO MANAGE HYPOTENSION FROM GENERAL ANESTHESIA A few studies comparing NE and PE administration to combat hypotension from general anesthesia have been described in cardiac and general surgical patients (Table 2).15,25,48–50Table 2.: Comparison of Norepinephrine with Phenylephrine to Treat Hypotension from General and Spinal Anesthesia AdministrationIn cardiac surgical patients, Goertz et al.48 studied the effect of treating anesthetic-induced hypotension in patients with coronary artery disease or aortic stenosis. Hypotension was defined as a >10% decrease from MAP baseline. This was treated with either a bolus dose of PE (1 μg/kg) or a bolus dose of NE (0.05 μg/kg) in a crossover design. In coronary artery bypass graft patients, PE administration resulted in impairment of global left ventricular function, which remained unchanged with NE. In contrast, in patients with aortic stenosis, where PE administration resulted in higher MAP levels, there was no detected difference in left ventricular function48 (Table 2). To assess which vasopressor would be best suited to treat hypotension in cardiac surgical patients with pulmonary hypertension, Kwak et al.49 performed a study, with significant limitations in its design, warranting cautious interpretation (Table 2). They compared NE with PE for the treatment of general anesthesia–induced systemic hypotension defined as a systolic blood pressure (SBP) <100 mm Hg. If hypotension occurred, patients were randomly assigned to receive either NE (8 μg/mL) or PE (40 μg/mL) infused at 50 mL/h to increase SBP >50% above baseline values. The authors considered NE to be the preferred agent to treat hypotension in patients with pulmonary hypertension because the cardiac index was maintained with NE, whereas it decreased significantly with PE administration. Further, the ratio of mean pulmonary artery pressure to mean SBP was improved (reduced) in the NE-administered group but not in the PE group.49 In general surgical patients, Goertz et al. compared the effect of NE (0.1 μg/kg) and PE (2 μg/kg) boluses on left ventricular performance when used to treat isoflurane-induced hypotension (Table 2). As predicted by their pharmacology, NE resulted in increased and PE decreased left ventricular performance despite similar arterial blood pressure control and heart rates.25 Poterman et al.50 compared the effect of treating hypotension (MAP < 80% of baseline awake) in general surgical patients anesthetized with propofol and remifentanil, using the administration of a bolus dose and then infusion of NE (10 μg + 0.05 μg/kg/min) versus PE (100 μg + 0.5 μg/kg/min). The authors found that both regimens resulted in similar increases in MAP and decreases in noninvasively measured CO and heart rate. They demonstrated slightly decreased tissue oxygenation for NE that was considered not clinically relevant (Table 2). Recently, in patients undergoing cesarean delivery using spinal anesthesia, Ngan Kee et al.15 compared the computer-controlled infusion of NE (5 μg/mL) with PE (100 μg/mL) infused to maintain SBP. Measurements taken at 5 minutes after induction demonstrated similar SBP, slower heart rate, and lower CO for PE in comparison with NE infusion (Table 2). These relatively small differences should be carefully interpreted. They may have been because of a relative overdose in PE computer-controlled infusion when compared with NE or because of the well-known physiologic differences of these 2 vasopressors on heart rate and CO. Taken together, these studies suggest that, when treating hypotension associated with general (and spinal) anesthesia, cardiac performance, stroke volume, and so CO are better maintained with NE than with PE. It should be noted that in these studies (where most patients had good left ventricular function), a modest increase in afterload (with a consequent decrease in cardiac performance) associated with PE administration4 is well tolerated. However, when left ventricular function is impaired, the enhanced afterload sensitivity of the failing myocardium to a pure vasoconstrictor may result in diminished CO and so systemic perfusion.51 Under such circumstance, NE, with its concomitant positive inotropic effect, may be advantageous.Figure 1.: Suggested treatment algorithm for hypotension associated with general anesthesia administration. *If norepinephrine is contraindicated, e.g., in cases of (1) hypertrophic obstructive cardiomyopathy and (2) carcinoid tumor, substitute with phenylephrine 100-μg bolus. #Vasopressin depletion in patients (1) taking angiotensin-converting enzyme inhibitors or angiotensin receptor blockade, (2) with sepsis, (3) with cardiac failure after cardiopulmonary bypass, and (4) with liver failure. AVP = arginine vasopressin; HR = heart rate; HTN = hypertension; IABP = intraaortic balloon pump; PRN = as necessary.Of note too are circumstances of refractory hypotension associated with general anesthesia administration in patients receiving angiotensin receptor blockade. In a recent case report, NE was shown to be effective when PE administration was not.52 Nevertheless, therapy with arginine vasopressin is recommended in patients receiving angiotensin receptor blockade,53 or where arginine vasopressin deficiency has been described as found in patients with septic shock,54 liver failure,55 or with vasodilatory shock after cardiac surgery54 (Fig. 1).5,53–55 PROLONGED INFUSION OF NOREPINEPHRINE OR PHENYLEPHRINE TO COMBAT HYPOTENSION No studies have been performed to assess the outcome from prolonged infusion of NE versus PE to treat hypotension in complex surgery. PE is often infused in such circumstances with reported deleterious effects such as acute kidney injury and the development of lactic acidosis.56 There was, however, a randomized crossover study performed after cardiac surgery, which might inform the development of lactic acidosis.57 Nygren et al.57 compared PE with NE infused to increase MAP by 30% in patients sedated with propofol. They demonstrated that PE infusion was associated with a statistically significant increase in splanchnic oxygen extraction and mixed venous-hepatic vein oxygen saturation gradient, when compared with NE despite similar cardiovascular variables (Table 3).9,57–59 This suggested that PE infusion resulted in more pronounced global splanchnic vasoconstriction. The investigators also demonstrated that PE infusion was associated with a statistically significant increase in lactate concentration from baseline not found with NE (Table 3).Table 3.: Comparison of Norepinephrine with Phenylephrine Infusion to Manage Hypotension in ICU PatientsComparative NE and PE prolonged infusion studies, however, have been performed in septic intensive care unit (ICU) patients. These may not be fully representative of prolonged intraoperative infusion of NE or PE, but they are all we have to compare. This is because CO is invariably increased in septic ICU patients (often maintained using inotropes), whereas intraoperatively CO may be depressed through anesthetic administration and/or hypovolemia. Morelli et al.58 evaluated PE replacement in patients with septic shock managed initially with a NE infusion. Replacement with PE infused for 8 hours, and titrated to maintain MAP at 65 to 75 mm Hg, resulted in statistically significant increases in serum lactate concentrations and a statistically significant decrease in both creatinine clearance and indocyanine green clearance. Morelli et al.60 subsequently performed a randomized control trial comparing PE and NE infusion in septic shock patients (where mixed venous oxygen saturation was maintained using dobutamine). They found no difference in any cardiovascular variables or indices of splanchnic hypoperfusion as they had previously found (Table 3). The authors noted that when PE was compared with NE administration, there was a seemingly different requirement for renal replacement (7 vs 2 patients, P < 0.133).60 Despite the study being underpowered, they speculated that the β-adrenergic effect from NE infusion might better preserve renal function in this setting. They suggested the need for further studies because they could not exclude that hepatosplanchnic function might worsen if PE infusion occurred over longer periods.60 Reinelt et al.59 studied the effects of replacing NE with PE to achieve comparable hemodynamic end points, again in septic shock patients (Table 3). They demonstrated statistically significant decreases in splanchnic blood flow rates (1.25 vs 0.85 L/min/m2), hepatic venous oxygen saturation (54% to 41%), and splanchnic lactate uptake rates (690 to 248 μmol/min/m2) from PE replacement, whereas cardiac index remained similar (4.3 vs 4.3 L/min/m2; Table 3). Taken together these limited data from ICU patients with normal or increased CO suggest that in the treatment of hypotension, NE infusion could be preferred over PE infusion to ensure better maintenance of hepatosplanchnic and renal perfusion and avoid the development of lactic acidosis. Randomized controlled studies should be performed to establish whether this is also the case for prolonged intraoperative infusion during surgery. It is noteworthy that in the updated Surviving Sepsis Campaign guidelines12 and in a recent review on management of septic shock,61 NE was the primary vasopressor recommended to combat hypotension. PE is only to be used if arrhythmias are associated with NE administration, CO is high, and therapy with NE, dobutamine, and low-dose vasopressin have failed to maintain arterial blood pressure.62 Should NE be recommended as the primary vasopressor, there is a concern that NE should not be given through a peripheral IV, which is routinely done with PE. The concern is that extravasation could result in serious arterial and venous constriction with attendant tissue necrosis. This is not based on fact. There have been no comparative studies of this potential; however, NE and PE vasoconstriction have been compared in ex vivo human radial arteries.19 NE was found to be 7× more potent than PE in radial arteries. The relative vasoconstrictor potential in human saphenous veins in vivo has also been studied.24 The relative potency of PE versus NE for venoconstriction was found to be 76%. In clinical practice, NE is administered in a concentration (5 μg/mL) 20× less than PE (100 μg/mL)15 Thus, there is likely to be less local arterial or venous vasoconstrictor potential from extravasation with NE administration in these concentrations than with PE. Should extravasation occur, this can be managed with phentolamine.7 CONCLUSIONS There are limited studies suggesting that NE may have an advantage over PE to treat hypotension after anesthesia induction or during prolonged surgery. Those that can be applied to this situation suggest that NE would better maintain CO and splanchnic perfusion. These findings are consistent with the theoretical advantage of administering an endogenous catecholamine, NE, rather than a synthetic pure vasoconstrictor, PE, to combat hypotension. Nevertheless, more comparative outcome studies need to be performed to address this important question. In the interim, recognizing that hypotension is multifactorial, a proposed algorithm (Fig. 1) for the management of hypotension is suggested based on the patient’s heart rate,5,63 a good surrogate for CO, and so systemic perfusion. DISCLOSURES Name: Berend Mets, MBChB, PhD, FRCA, FFA(SA). Contribution: This author wrote the manuscript. Attestation: Berend Mets approved the final version of the manuscript. This manuscript was handled by: Martin London, MD.