British Journal of Anaesthesia

This Article Abstract Full Text (PDF) E-Letters: Submit a response to the article Alert me when this article is cited Alert me when E-letters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (41) Request Permissions Disclaimer Google Scholar Articles by Zaugg, M. Articles by Spahn, D. R. Search for Related Content PubMed PubMed Citation Articles by Zaugg, M. Articles by Spahn, D. R. Social Bookmarking          What's this?

British Journal of Anaesthesia, 2002, Vol. 88, No. 1 101-123
© 2002 The Board of Management and Trustees of the British Journal of Anaesthesia

Modulation of ß-adrenergic receptor subtype activities in perioperative medicine: mechanisms and sites of action

M. Zaugg^*,1, M. C. Schaub², T. Pasch¹ and D. R. Spahn³

¹Department of Anesthesiology, University Hospital Zurich, Rämistrasse 100, CH-8091 Zurich, Switzerland. ²Institute of Pharmacology and Toxicology, University of Zurich, Switzerland. ³Department of Anesthesiology, University Hospital Lausanne, Switzerland*Corresponding author

	Abstract

Top Abstract Choice of literature Clinical uses of perioperative... The present armamentarium of... Mechanisms and sites of... Tolerability of perioperative... Conclusions Addendum References

 
This review focuses on the mechanisms and sites of action underlying ß-adrenergic antagonism in perioperative medicine. A large body of knowledge has recently emerged from basic and clinical research concerning the mechanisms of the life-saving effects of ß-adrenergic antagonists (ß-AAs) in high-risk cardiac patients. This article re-emphasizes the mechanisms underlying ß-adrenergic antagonism and also illuminates novel rationales behind the use of perioperative ß-AAs from a biological point of view. Particularly, it delineates new concepts of ß-adrenergic signal transduction emerging from transgenic animal models. The role of the different characteristics of various ß-AAs is discussed, and evidence will be presented for the selection of one specific agent over another on the basis of individual drug profiles in defined clinical situations. The salutary effects of ß-AAs on the cardiovascular system will be described at the cellular and molecular levels. ß-AAs exhibit many effects beyond a reduction in heart rate, which are less known by perioperative physicians but equally desirable in the perioperative care of high-risk cardiac patients. These include effects on core components of an anaesthetic regimen, such as analgesia, hypnosis, and memory function. Despite overwhelming evidence of benefit, ß-AAs are currently under-utilized in the perioperative period because of concerns of potential adverse effects and toxicity. The effects of acute administration of ß-AAs on cardiac function in the compromised patient and strategies to counteract potential adverse effects will be discussed in detail. This may help to overcome barriers to the initiation of perioperative treatment with ß-AAs in a larger number of high-risk cardiac patients undergoing surgery.

Br J Anaesth 2002; 88: 101–23

Keywords: sympathetic nervous system, adrenergic block; complications, side-effects; heart, heart rate, perioperative; heart, cardioprotection; sympathetic nervous system, beta-adrenergic receptor

	Choice of literature

Top Abstract Choice of literature Clinical uses of perioperative... The present armamentarium of... Mechanisms and sites of... Tolerability of perioperative... Conclusions Addendum References

 
Literature relevant to the topic of the review was identified by literature search of Medline (1966 to March 2001) using ‘beta-blocker’, ‘beta-adrenergic antagonist’, ‘beta-adrenergic receptor’, ‘cardioprotection’, ‘side-effect’, ‘perioperative’ and combinations of these terms as keywords. The reference lists of relevant articles were further reviewed and personal files were searched to identify additional citations.

	Clinical uses of perioperative ß-adrenergic antagonists

Top Abstract Choice of literature Clinical uses of perioperative... The present armamentarium of... Mechanisms and sites of... Tolerability of perioperative... Conclusions Addendum References

 
Current clinical uses of ß-adrenergic antagonists (ß-AAs) include treatment of arterial hypertension, primary and secondary prevention of myocardial infarction in patients with coronary artery disease, treatment of atrial and ventricular arrhythmias, and, most recently, the treatment of the failing heart.^{28 29 37 60 111 121 167} ß-AAs belong to the group of first-line antihypertensive agents, decreasing cardiac death by 30%,¹⁷⁶ and long-term treatment with ß-AAs after myocardial infarction reduces total mortality by more than 30%.⁷⁴ Decreased elevation of the MB heterodimer of creatine kinase after coronary interventions associated with improved intermediate-term survival was reported in patients with prior ß-AA therapy compared with those not on ß-AAs.¹⁹⁸ In addition, it has been estimated from crude annualized mortality rates, derived from trials with inhibitors of angiotensin-converting enzyme (ACE) and with ß-AA conducted in heart-failure patients, that ß-AAs are more than twice as effective as ACE inhibitors in terms of average reduction in mortality.^34 65 Thus, ß-AAs are highly potent cardiovascular drugs.

In spite of this overwhelming evidence, ß-AAs are underused in current clinical practice, and physicians prescribe ß-AAs only to approximately 50% of patients qualifying for this therapy.^{21 122 130 242} Notably, medical contraindications do not appear to explain the low use of ß-AAs, and it has been speculated that pharmaceutical industry competitiveness may have contributed to it by leading to exaggeration of the side-effects of ß-AAs (harmful lipid profile, decreased sexual function, potential for precipitating congestive heart failure, decreased exercise performance).^112 210 The proof of the concept of ß-adrenergic antagonism in cardioprotection, however, is now also firmly established in patients with coexisting disease states that were traditionally considered as contraindications. Particularly, patients older than 80 yr with heart failure (ejection fraction less than 20%), non-Q-wave infarction, diabetes or chronic obstructive pulmonary disease have a disproportionate high benefit from postinfarction ß-adrenergic antagonism.^74 205 Therefore, anaesthetists should assume the role of primary caregivers and initiate treatment with ß-AAs in surgical patients with well-defined indications for ß-AAs who are admitted to the hospital without proper treatment.^78 232

Because gaining control over the autonomous nervous system constitutes a significant part of perioperative medicine,^59 184 ß-adrenergic antagonism has been used traditionally to maintain blood pressure and heart rate within baseline values in various perioperative settings. In particular, ß-AAs were successfully used to blunt haemodynamic responses to intubation,^44 154 at the time of emergence caused by decreasing anaesthetic depth,⁶⁷ and during electroconvulsive therapy.^46 262 High-dose ß-AA treatment is used to maintain deliberate hypotensive anaesthesia,^105 219 and has been used most recently to enable multiple-vessel coronary artery bypass grafting (CABG) on the beating heart.^170 152 Esmolol-enriched normothermic blood also resulted in better myocardial protection compared with crystalloid cardioplegia in patients undergoing CABG surgery.¹⁹ In addition, perioperative ß-AAs reduce the incidence of atrial fibrillation after cardiac surgery,^94 191 as well as after thoracotomy for lung resection.¹⁰⁰ Nonetheless, it was not until the late 1970s that it became generally accepted that patients taking ß-AAs preoperatively should be continued on ß-AA treatment perioperatively. In one study, Slogoff and colleagues²⁰² reported pre-bypass ischaemia in patients undergoing CABG surgery in 26% of patients with propranolol treatment continued in full dosage until operation, in 50% of patients with propranolol withdrawal and in 70% of patients with no ß-AA treatment.

Administration of perioperative ß-AAs controls haemodynamic variables and successfully decreases the incidence of ischaemic events in patients with or at risk of coronary artery disease. This is particularly relevant as patients with perioperative ischaemia have a nine-fold increase in the risk of developing a serious adverse cardiac outcome during hospitalization and more than a two-fold increase in the risk of dying prematurely over the first 6 months after surgery.¹⁴⁰ Patients with a postoperative in-hospital myocardial infarction have a 28-fold increase in the rate of subsequent cardiac complications within 6 months, a 15-fold increase within 1 yr and a 14-fold increase within 2 yr. Stone and colleagues²¹² administered a single oral preoperative dose of one of three different ß-AAs (labetalol, oxprenolol, atenolol) to patients with mild uncontrolled hypertension undergoing non-cardiac surgery. The incidence of myocardial ischaemia was 28% in the untreated controls compared with 2% in the ß-AA-treated patients, based on ECG criteria (P<0.001). Wallace and colleagues²³¹ administered i.v. atenolol preoperatively and i.v. and oral atenolol for up to 7 days postoperatively in patients with or at risk of coronary artery disease. Myocardial ischaemia was assessed by continuous three-lead Holter monitoring. Intraoperative myocardial ischaemia was reported to be 12% in each of the control and atenolol-treated groups. Conversely, the incidence of myocardial ischaemia was reported to be 34% in the control vs 17% in the treated patients in the first 48 h after surgery (P<0.008) and 39 vs 24% over days 0–7 (P<0.029). More recent studies by Raby and colleagues¹⁷⁹ and Urban and colleagues²²⁵ confirm the powerful anti-ischaemic effect of perioperative ß-adrenergic antagonism and re-emphasize the importance of stress-induced increases in heart rate in the pathogenesis of perioperative myocardial ischaemia. The association between the occurrence of perioperative myocardial ischaemic events, perioperative tachycardia and an adverse long-term cardiac outcome led to therapeutic trials with ß-AAs. However, it should be noted at this point that the association between postoperative myocardial ischaemia and adverse cardiac events does not necessarily imply a causal relationship and that postoperative myocardial ischaemia may represent only a manifestation of the underlying cardiac disease. Factors other than reduced ischaemia may contribute significantly to the improvements in outcome observed after administration of ß-AAs. These will be discussed extensively in this article. Using preoperative (1 h before surgery) and postoperative (until 7 days after surgery) atenolol (Table 1), Mangano and colleagues¹⁴² demonstrated in a well-designed study a significant reduction in postoperative myocardial ischaemia in patients with or at risk of coronary artery disease. This reduction in ischaemic events was associated with a 55% decrease in overall mortality and a 65% decrease in cardiac mortality at 2 yr for the atenolol-treated patients. The protective effects of atenolol were evident in these patients 6 months after surgery (overall mortality 0 vs 8%, P<0.001) and were preserved over the 2-yr follow-up period (overall mortality 10 vs 21%, P<0.019). Recently, Poldermans and colleagues¹⁷³ randomized patients undergoing vascular surgery with mild to moderate positive stress echocardiography to preoperative (from 1 week before surgery) and postoperative (until 30 days after surgery) bisoprolol treatment or placebo (Table 1). After the inclusion of 112 patients, this study was halted for ethical reasons associated with large differences in morbidity and mortality rates between the placebo and bisoprolol arms of the study. Notably, the study reports a 10-fold decrease in the 30-day perioperative incidence of death from cardiac causes and non-fatal myocardial infarction in bisoprolol-treated patients (3.4 vs 34%, P<0.001). A critical evaluation of these studies has been published recently in the British Journal of Anaesthesia and can be recommended as additional reading.⁹⁶ Taken together, these data indicate that ß-adrenergic antagonism remains the sole proven pharmacological means of reducing perioperative cardiovascular short- and long-term cardiac morbidity and mortality in patients with or at risk of coronary artery disease. In the light of the benefits of perioperative ß-AAs and the exceptionally low complication rate associated with the perioperative use of ß-AAs, future trials have to clarify whether the cumulative morbidity and mortality associated with sophisticated and expensive preoperative testing can be justified in high-risk cardiac patients undergoing surgery.^{128 144 190}

View this table: [in this window] [in a new window]   Table 1 Perioperative administration of atenolol and bisoprolol. *If p.o. administration is not feasible in the perioperative period, esmolol, metoprolol or atenolol should be administered i.v. to maintain a heart rate of 50–80 beats min–1

 
New aspects of perioperative ß-adrenergic antagonism have emerged recently. Johansen and colleagues^102 103 demonstrated that esmolol could potentiate the reduction in minimum alveolar concentration for isoflurane by alfentanil (–26%) and decrease anaesthetic requirements for skin incision during propofol/nitrous oxide/morphine anaesthesia (–27%) in patients. The clinical utility of this effect was subsequently demonstrated by Zaugg and colleagues²⁵⁴ in a study with elderly surgical patients that evaluated three anaesthetic regimens, two of them with atenolol. High-dose intraoperative administration of atenolol decreased isoflurane requirements by 37% and still allowed an adequate depth of anaesthesia, as assessed by bispectral analysis (mean bispectral index 50–60). Pre- and postoperative atenolol as well as high-dose intraoperative atenolol also decreased requirements for intraoperative fentanyl (–27%) and postoperative morphine (–40%). As a consequence, extubation time and recovery in the postanaesthesia care unit were significantly faster in patients treated with atenolol.

	The present armamentarium of ß-adrenergic antagonists

Top Abstract Choice of literature Clinical uses of perioperative... The present armamentarium of... Mechanisms and sites of... Tolerability of perioperative... Conclusions Addendum References

 
Although all ß-AAs are able to antagonize the transduction of the ß-adrenergic receptor signal (ß-AR), this class of drugs is far from being homogeneous. Recently, antisense oligonucleotides against ß₁-adrenergic receptor (ß₁-AR) mRNA, which suppress protein translation at the ribosomes, have been constructed and used successfully to treat hypertensive rats.²⁵⁸ Currently used ß-AAs, however, competitively antagonize ß-ARs and can be roughly classified into three generations depending on their ancillary properties (Table 2).^28 29 The main ancillary properties of individual agents include partial agonist activity (intrinsic sympathomimetic activity), ß-receptor subtype specificity (ß₁ vs ß₂), lipophilicity and membrane-stabilizing activity (Table 3). Other ancillary properties include vasodilator effects [ß₂ (celiprolol)-, anti-₁ (carvedilol)- or nitric oxide (NO)-mediated effects (nipradilol, nebivolol)], class III anti-arrhythmic activity (sotalol), antioxidant effects (carvedilol) and stereoselective hepatic metabolism (carvedilol, metoprolol). Accordingly, carvedilol by oral administration exerts equal effects on - and ß-ARs, whereas carvedilol by i.v. administration exerts more ß-AR effects than -AR effects because of decreased stereoselective hepatic metabolism of the ß-AR-specific S-isomer of carvedilol.¹⁶⁴ Interestingly, stereoselective metabolism of metoprolol may result in insufficient ß-adrenergic antagonism in ‘poor metabolizers’ (S/R isomer ratio <1).^{201 203}

View this table: [in this window] [in a new window]   Table 2 Classification of ß-adrenergic antagonists

 

View this table: [in this window] [in a new window]   Table 3 Ancillary properties of clinically used ß-adrenergic antagonists. +=effect present; –=effect absent

 
Recent research in transgenic animal models also emphasizes the importance of the two-state model of ß-AR activation in characterizing ß-AAs (Fig. 004F1). This model proposes an equilibrium between an inactive and an active conformation of the receptor, which is differentially modulated by various ligands (concept of inverse agonism: neutral antagonist vs inverse agonist).¹⁸ Notably, it predicts spontaneous activation of ß-ARs, which was verified recently for the ß₂- but not the ß₁-AR.²⁶¹ This model also explains the inability of some ß-AAs with pronounced neutral antagonism to block the effects of receptor overexpression fully, as neutral antagonists counteract activation by endogenous catecholamines but not activation by spontaneous transition into the active receptor conformation.¹³⁴ The physiological consequences are not yet determined fully but may be of clinical relevance with respect to the tolerability of various ß-AAs and the treatment of the withdrawal syndrome. Finally, there is growing evidence that ß-ARs differentially couple to various G-proteins depending on the specific properties of the ligand, thereby stimulating differential cellular responses.²³³

View larger version (24K): [in this window] [in a new window]   Fig 1 Two-state model of ß-adrenergic receptor activation by competitive ligands. Signalling at the receptor includes binding of the ligand to the extracellular binding domain, transduction of the signal through conformational changes of the receptor and activation of the effector (G-protein complex). In the absence of a ligand, the receptor can undergo spontaneous transition from the inactivated to the activated state. Ligands can be classified into agonists, neutral antagonists and inverse agonists according to their tendency to shift this equilibrium. The agonist shifts the equilibrium towards the active state and the inverse agonist shifts it to the inactive state. Although most ß-adrenergic antagonists (ß-AAs) act as inverse agonists, some ß-AAs with weak inverse agonism may be classified as neutral antagonists.18 The relative degree of inverse agonism increases in the following order: bucindolol<carvedilol <propranolol<metoprolol.249 This model explains the observed lower tolerability of patients treated with inverse ß-AAs: they shift the receptor population almost completely to the inactive state, particularly when the sympathetic basal tone is low. On the other hand, ß-AAs with weak inverse agonism leave a sizeable fraction of the receptor in the active state, thus explaining their better tolerability in clinical use.

 
Interestingly, a meta-analysis of randomized controlled trials revealed differential effects on cardiovascular events, such as reinfarction and sudden cardiac death, metoprolol being more effective than atenolol or propranolol. This led the authors to conclude that the so-called class effect of ß-AAs may be less important than ancillary properties.²⁰⁴ However, the mechanistic concept of the class effect is greatly supported by the observation that selective as well as non-selective ß-AAs decrease mortality significantly in chronic heart failure.¹²¹ Nonetheless, ancillary properties are important with respect to the side-effects and tolerability of the specific agents, which will be discussed separately.

	Mechanisms and sites of action

Top Abstract Choice of literature Clinical uses of perioperative... The present armamentarium of... Mechanisms and sites of... Tolerability of perioperative... Conclusions Addendum References

 
The following sections will focus on the mechanisms and sites of action elicited by ß-AAs.

Cardiac considerations
Bradycardia, the link to many cardioprotective effects of ß-adrenergic antagonists
Elevated heart rate is a well-established independent predictor of coronary artery disease and cardiovascular morbidity and mortality.¹⁶⁸ Also, delayed decrease in heart rate after graded exercise predicts cardiovascular mortality.³⁸ Bradycardia is suggested to be one important mechanism of cardioprotection elicited by ß-AAs. Importantly, a negative chronotropic response to ß-AAs is preserved among diabetic patients with progressive autonomic dysfunction.¹¹¹

In the perioperative period, increased heart rate is strongly associated with myocardial ischaemia.^{141 143} Accordingly, myocardial oxygen balance is closely related to heart rate and must be examined on a beat-to-beat basis (Fig. 004F2). Increased heart rate results in elevated myocardial oxygen demand via the Bowditch effect, which is, however, nearly offset by decreased oxygen demand caused by the lower ventricular wall tension at higher heart rates. Because increased heart rate is usually accompanied by increased inotropy and the length of diastole is significantly decreased in tachycardia, myocardial oxygen balance can deteriorate seriously at higher heart rates in patients with coronary artery disease. Increased left ventricular stiffness further exacerbates impairment of ventricular filling. Importantly, the heart rate at which patients are considered at risk of developing ischaemia is not absolute and must be individualized. Patients with severe angiographic narrowing of the coronary arteries show a gradual decrease in cross-sectional area by 32% when heart rate reaches 90 beats min^–1.¹⁶⁰ Therefore, it is not surprising that patients with coronary artery disease and a heart rate greater than 100 beats min^–1 almost inevitably develop myocardial ischaemia.⁶⁹

View larger version (19K): [in this window] [in a new window]   Fig 2 Myocardial oxygen balance. In patients with coronary artery disease, tachycardia decreases myocardial oxygen supply and concomitantly increases oxygen demand.

 
The deleterious effects of an increased heart rate on infarction size have been reported. Augmentation of heart rate after experimental coronary occlusion in dogs, by ventricular pacing, isoproterenol or atropine, leads to increases of 40, 70, and 40% respectively in myocardial necrosis when compared with control.¹⁹⁹ Consistent with this notion, thiopental given during coronary occlusion doubles infarction size by increasing heart rate. Notably, tachycardia also accentuates endomyocardial to epimyocardial maldistribution of ventricular blood flow in ischaemia.¹² ß-AAs effectively reverse all these untoward effects by lowering heart rate.²²⁷ ß-AAs also exert a beneficial effect in coronary artery disease by decreasing the stiffness of atherosclerotic plaques, which results in increased tensile strength.¹²⁷ The stiffness of fibrous caps of human atherosclerotic plaques is directly related to heart rate, and increased heart rate promotes the fissuring of atherosclerotic plaques. ß-AAs prevent the rupture of vulnerable atherosclerotic plaques, which leads to less inflammation in the plaque and decreases the gradual narrowing of the vessel lumen.^63 66 178 Furthermore, tachycardia causes activation of platelets.⁵⁴ When coronary blood flow increases, platelets can be traumatized and activated across the coronary bed, particularly at sites with significant narrowing. Histopathological analyses of perioperative myocardial infarction stress the importance of plaque disruption and thrombosis as pivotal steps in the pathogenesis of perioperative myocardial infarction.⁴⁸ Alternatively, long-duration subendocardial ischaemia and subsequent non-Q-wave infarction resulting from prolonged tachycardia were also proposed as the underlying mechanism of perioperative myocardial infarction.¹²⁴ The view that postoperative myocardial ischaemia is a mere manifestation of the underlying cardiac disease is specifically supported by the findings of the post-mortem study by Dawood and colleagues.⁴⁸ In this study, fatal postoperative myocardial infarctions were associated with evidence of unstable plaques in 55% of the patients. In contrast, imbalance of myocardial oxygen supply/demand may play a causal role in the pathogenesis of postoperative cardiac events. This view is supported by the fact that most postoperative myocardial infarctions are non-Q-wave infarctions that are preceded by long-duration (>2 h) postoperative ischaemia.¹²⁵ Because cardiac complications are preceded by long-duration ST-segment depression rather than elevation, it seems plausible that the cascade of events leading to postoperative cardiac complications does not begin with acute coronary occlusion but with long-duration subendocardial ischaemia. This is further supported by a recent study by Badner and colleagues,⁸ which determined the incidence of postoperative myocardial infarction after non-cardiac surgery in a large group of patients at high risk. This study also reports the preponderance of non-Q-wave infarction, which differs from that seen in non-surgical patients presenting to the emergency room. From a mechanistic point of view, non-Q-wave infarctions result from prolonged ischaemia rather than from total occlusion of the coronary arteries. Certainly, further studies are needed to elucidate the role of postoperative myocardial ischaemia in the cascade of events leading to perioperative cardiac morbidity and mortality.

The institution of bradycardia was recently found to cause restoration of contractile function in a canine model of mitral regurgitation-induced left ventricular dysfunction.¹⁶¹ In this model, optimized myocardial Ca²⁺ handling and bioenergetics are direct consequences of bradycardia and are suggested to be responsible for the observed improvement in contractility.²⁵⁷

At the cellular level, rapid electrical stimulation of contraction reduces the density of ß-ARs and their responsiveness,¹¹⁶ which appears to be associated with disassembly of microtubules secondary to undue micromechanical stress.²⁴⁸

Cardioprotective effects of ß-adrenergic antagonists not apparently associated with bradycardia
Stangeland and colleagues²⁰⁷ addressed the important question of whether decreased heart rate is the only mechanism responsible for cardioprotection elicited by ß-AAs. They treated anaesthetized cats with alinidine (a clonidine analogue that decreases heart rate independently of ß-ARs) or timolol. Heart rate was similarly reduced by 40 beats min^–1 in the treatment groups compared with the control group, and regional ischaemia was induced by occluding the left descending coronary artery. After 6 h of ischaemia, the necrotic tissue was measured and expressed as a percentage of necrotic tissue in the area at risk. Notably, alinidine significantly decreased necrosis from 87% present in the control group to 77% (P<0.01), whereas timolol decreased necrosis to 65% (P<0.001). This observation clearly indicates that mechanisms other than decreased heart rate contribute substantially to cardioprotection by ß-AAs.

ß-Adrenergic signal transduction in cardiomyocytes. Biological responses mediated by ß-ARs involve positive chronotropy, dromotropy, inotropy and cardiomyocyte growth and death (Fig. 004F3). ß-ARs are members of the G-protein-coupled superfamily, which share the characteristic feature of the seven-transmembrane-spanning domains. In healthy mammalian cardiomyocytes, ß₁-ARs constitute around 70–80% of the ß-ARs in human and rat hearts.^52 236 In many disease states with heightened sympathetic drive, ß₁-ARs are down-regulated by phosphorylation (desensitization), translocation (sequestration) and finally by degradation of the receptor.⁸⁸ Conversely, ß₂-ARs do not decrease in number; however, they show some loss of contractile response to agonist stimulation as a result of the up-regulation of ß-AR kinases (ß-ARK) and Gi proteins (Table 4).²⁶ In the failing human heart, ß₂-ARs represent 40% of ß-ARs and are of great importance in mediating inotropic and chronotropic responses.²⁵ Key steps in signal transduction of ß₁- and ß₂-ARs involve coupling to G-proteins and activation of the cAMP/protein kinase A (PKA) pathway, which leads to phosphorylation of target proteins such as phospholamban, the ryanodine receptor, troponin I and L-type Ca²⁺ channels.^209 244 However, apart from changes in myocardial contractile function, ß-ARs exert important effects on cellular metabolism, growth and death (gene expression) through the activation of PKA and protein kinase C (PKC).⁹⁰ Because the ß₁-AR and the ß₂-AR share only 54% of amino acid sequences overall, it is possible that ß-AR subtypes couple to distinct signal transduction pathways.^243 245 Although both ß₁- and ß₂-ARs increase the contractile response and hasten relaxation in ventricular myocytes, several striking differences with respect to G-protein binding characteristics and signal transduction downstream from the receptor have been revealed. In contrast to ß₁-ARs, ß₂-ARs exhibit dual coupling to Gs and Gi that can completely negate Gs-mediated responses. Also characteristic of ß₂-mediated signalling is the exceptionally modest increase in cAMP and the compartmentalized increase in PKA activity, which is restricted to the vicinity of L-type Ca²⁺ channels.^124 260 Finally, ß₂-ARs may also bind to Gq-activating phospholipase C (Fig. 004F3). These data indicate that ß-AR subtypes differentially modulate cardiac function and cardiomyocyte phenotype. Therefore, the subtype specificity of various ß-AAs affects biological responses significantly.

View larger version (40K): [in this window] [in a new window]   Fig 3 ß-Adrenergic signalling cascades in cardiomyocytes. (A) Binding of an agonist (L) to either the ß1- or the ß2-adrenergic receptor (ß1-AR, ß2-AR) stimulates Gs protein, which dissociates from the receptor and binds to adenylate cyclase (AC), causing production of cAMP from ATP and activation of protein kinase A (PKA). G-protein-coupled receptors interact by their intracellular loop 3 with the heterotrimeric G complex and promote GDP release. PKA phosphorylates the voltage-dependent L-type Ca2+ channels, the Na+/H+ exchange channels and the Na+/K+ pump at the sarcolemma, phospholamban (PLB) and the ryanodine receptor (RYR) at the sarcoplasmic reticulum (SR) and cardiac troponin-I (cTnI) in the sarcomeres, leading to increased inotropic and lusitropic (relaxation) responses. In contrast, the ß2-AR is also able to couple to Gi or Gq proteins. Gi inhibits adenylate cyclase (AC) and opposes the effects of Gs. Gq activates phospholipase C (PLC). By splitting phosphatidylinositol bisphosphate, PLC liberates the two intracellular second messengers diacylglycerol (DAG) and inositol trisphosphate (IP3). IP3 binds to the IP3 receptor (IP3R), which releases Ca2+ from the SR. Ca2+ combines with calmodulin (CaM) and directly activates the sarcolemmal Ca2+ pump as well as several CaM-dependent protein kinases (CaMKs). This signalling pathway leads to phosphorylation of PLB, ventricular myosin light chain 2 (MLCV2) and the Na+/Ca2+ exchanger. DAG and CaM together activate PKC, which in turn phosphorylates the mitochondrial ATP-dependent K+-channel and MLCV2. In its unphosphorylated state, the regulatory protein PLB is bound to the SR Ca2+-pump (SERCA2), inhibiting its activity. When phosphorylated by PKA and/or CaMK, it dissociates from SERCA2, relieving the inhibitory effect. On the other hand, direct phosphorylation of RYR at Ser-2809 dissociates the regulatory component FKBP (FK506 binding protein), leading to increased activity of the RYR channel. PKA and PKC both affect gene expression in the cell nucleus via the common MAPK (mitogen-activated protein kinases) signalling pathway: Ras (monomeric GTPase), Raf (a MAPKKK), MEK (mitogen-activated ERK activating kinase) and ERK (extracellular signal regulated kinase). To protect the cardiomyocyte from ß-adrenergic overstimulation, a negative feed-back loop (not shown in the diagram) is built in, which degrades cAMP to AMP by means of CaM-activated phosphodiesterase III. (B) Diagram showing in more detail the dissociation of the heterotrimeric G complex from the ß-AR upon binding of the agonist (L). After this dissociation, the adenylate cyclase (AC) becomes activated by binding to the -subunit. Desensitization of the ß1- and ß2-ARs is mediated by phosphorylation of the intracellular C-terminal part of the receptor by either PKA or the ß-adrenergic receptor kinase (ß-ARK, GRK2 and 3). Binding of arrestin and clathrin to the phosphorylated ß-AR mediates internalization to the endosome. After dephosphorylation, ß-ARs may be either degraded or resensitized. Note that an additional negative feed-back loop leads to inhibition of the ß-ARK via increased Ca2+-calmodulin (CaM).17 58 90 147 148 171 189

 

View this table: [in this window] [in a new window]   Table 4 Major changes in components of ß-adrenergic signalling in the failing human heart28 147 171

 
Little is known about the role of ß₃-ARs in cardiomyocytes. Whereas ß₃-ARs are known to exert important physiological effects in brown adipose tissue, gut relaxation and vasodilation, ß₃-ARs mediate negative inotropy by a NO-dependent pathway in cardiomyocytes.^70 226 Studies evaluating the distribution and quantification of ß₁/ß₂/ß₃-AR subtypes in heart tissue have revealed subtype proportions for the left porcine ventricle as follows: ß₁:ß₂:ß₃=72%:28%:0.25%.¹⁵¹ This implies that, in the normal myocardium, ß₃-AR may be of less importance. However, recent observations in heart failure patients demonstrate that opposite changes in the abundance of ß₁-AR (down-regulation) and ß₃-AR (up-regulation) occur and may play a role in the progressive functional degradation in the failing human heart.¹⁵⁵

Cell death signalling: apoptosis and necrosis. Catecholamines, although beneficial in the short-term cardiovascular response, exert significant cardiac toxicity. The toxic effects of catecholamines on cardiomyocytes have been known since the beginning of the 20th century.⁸² However, necrotic and apoptotic cell death has been closely related to enhanced ß-adrenergic signalling only recently.¹⁴⁵ Apoptotic cardiomyocyte death by activation of the ß-adrenergic signalling pathway was reported in norepinephrine-stimulated adult rat ventricular myocytes.³⁹ Zaugg and colleagues²⁵⁵ further demonstrated that apoptotic cardiomyocyte cell death is dissociated from ß₂-ARs and selectively mediated by ß₁-ARs in adult ventricular myocytes (Fig. 004F4). This is in line with clinical observations that cardiac lesions associated with massive catecholamine bursts were prevented with atenolol in patients with subarachnoid haemorrhage.⁴³ Communal and colleagues⁴⁰ and Chesley and colleagues³⁶ further showed that ß₂-AR stimulation may protect cardiomyocytes from apoptosis-inducing stimuli. The abilities of ß-AR stimulation and tachycardia to induce cardiomyocyte apoptosis were addressed by Shizukuda and colleagues²⁰⁰ in an in vivo rat model. Rats were treated with placebo or isoproterenol to establish whether catecholamines per se in the absence of significant increases in systolic load and tachycardia induce myocardial damage via apoptosis. After only 24 h of isoproterenol treatment, a significant increase in apoptotic events was detected. Animals exposed to ventricular pacing to induce tachycardia equivalent to that produced by isoproterenol treatment did not show an increase in apoptosis. The authors concluded that apoptotic cardiomyocyte death resulting from isoproterenol treatment may not be explained by increased heart rate alone. Conversely, in a canine model, rapid ventricular pacing per se increased apoptotic cell death and led to cardiac myopathy.¹³⁶ Importantly, decreased apoptosis has been reported in carvedilol- and propranolol-pretreated rabbit hearts subjected to ischaemia–reperfusion injury.²⁵² Because apoptotic cell death occurs in only a few hours, apoptosis may be an important mechanism for loss of viable cardiomyocytes and myocardial dysfunction in the immediate perioperative period.²⁵³

View larger version (53K): [in this window] [in a new window]   Fig 4 Cardiotoxicity of catecholamines. Adult rat ventricular myocytes (ARVMs) grown on coverslips were exposed to norepinephrine (NE) (10 µmol litre–1) alone or in the presence of atenolol (AT) (10 µmol litre–1) for 12 h and subjected to TUNEL (terminal dUTP nick end labelling) staining, which is specific for apoptotic cell death. (A) Mean percentage of TUNEL-positive ARVMs on coverslips. Data are mean (SEM). *P<0.0001 vs control; P<0.0001 vs NE. (B) Control ARVMs with rod-shaped morphology. (C) NE-exposed ARVMs with rounded morphology and black apoptotic nuclei. (D) NE+AT-treated ARVMs. Note preservation of rod-shaped morphology with NE treatment (reproduced with permission from Circulation255).

 
New insight from gene-targeted animals. While experimental results indicate incontrovertibly that enhanced ß₁-AR signalling is exceptionally cardiotoxic, data on the effects of the ß₂-AR with respect to beneficial and detrimental effects are contradictory. Recent research in the field of heart failure tried to construct genetically altered mouse models mimicking increased sympathetic nervous system activity (Table 5). In particular, transgenic mouse models with cardiac-specific overexpression of various G-proteins and ß-AR subtypes were constructed.^{3 45 61 71 134 153} Mice with Gs overexpression typically develop a characteristic hypertrophic cardiomyopathy at 15 months of age. Sections of these hearts reveal hypertrophic myocytes with increased cross-sectional areas and an increased number of apoptotic myocytes. Importantly, propranolol, a non-selective ß-AA, abolished the hypertrophic response and the development of dilated chambers, thereby improving survival. Similar results were reported for mice overexpressing Gq- and ß₁-AR. Notably, mice with only five-fold overexpression of the ß₁-AR develop fatal cardiomyopathy,⁶¹ whereas mice with 30- to 60-fold overexpression of the ß₂-AR exhibit enhanced cardiac function and do not develop overt cardiomyopathy.¹³⁴ Although no long-term toxic effects were reported by some authors in mice with 200-fold overexpression of the ß₂-AR,^209 244 increased susceptibility to ischaemic injury⁴² and augmented afterload⁵⁶ were clearly observed. Nonetheless, it was suggested that manoeuvres that serve to augment ß₂-adrenergic signalling, which improves systolic and diastolic function, may offer a potential therapeutic approach in patients suffering from impaired cardiac function. For this purpose, pharmacological means and ultimately in vivo gene transfer strategies were proposed and investigated.^{110 150 196} Accordingly, the contractility of single myocytes isolated from the ventricles of rabbits chronically paced to produce heart failure can be functionally restored by adenovirus-mediated transfer of ß₂-ARs.⁴ Consistent with this notion, a dual-expressing ‘designer’ mouse with cardiac-specific Gq expression and concomitant ß₂-AR expression at low (30x), medium (150x) and high levels (1000x) was constructed recently.⁵⁵ Gq mice with low concomitant expression of ß₂-ARs, i.e. a 30-fold increase in ß₂-AR compared with wild type, displayed rescue of hypertrophy and ventricular function. Importantly, these effects occurred in the absence of any improvement in basal or agonist-stimulated adenylate cyclase (AC) activity, indicating the restoration of a compartmentalized ß₂-AR–AC signalling pathway.^124 260 The summarized experimental results, which clearly demonstrate beneficial effects of modest ß₂-adrenergic signalling in an animal model of heart failure, have found their clinical counterpart very recently.³⁰ Studies of ß₂-AR gene variations in twins revealed that specific ß₂-AR polymorphisms, which resulted in enhanced down-regulation of the ß₂-AR, increased cardiac dimensions (septum thickness, posterior wall thickness, left ventricular mass).

View this table: [in this window] [in a new window]   Table 5 Gene-targeted mice mimicking enhanced ß-adrenergic signalling

 
In summary, the concept that ß-adrenergic signalling may not mediate deleterious effects exclusively but may also have beneficial effects in the compromised heart is based on several experimental observations. In this regard it is interesting to note that, in patients with symptomatic heart failure, pan-adrenergic antagonism using central sympatholysis (moxonidine, an ₂-agonist) was terminated because of excess mortality.¹²¹ ß₂-AR agonism as an adjunct to ß₁-AR antagonism may therefore have the potential to improve the therapeutic tolerance, particularly during the initiation of ß-AA therapy, and to improve survival in the treatment of the failing heart. Interestingly, ß₁-AR antagonism enhances the ß₂-AR-mediated inotropic response to catecholamines, which may, in part, also explain the better tolerability of selective ß₁-AAs.⁸⁴

Mechanical unloading and modulation of gene expression. ß-AAs allow the heart to ‘rest’ by emulating a state of ventricular unloading. Accordingly, the expression of tumour necrosis factor  (TNF-), an indicator of increased mechanical load, is similarly reduced by ß-AAs and by mechanical circulatory support.^174 221 High-dose atenolol also prevents angiotensin II- and tachycardia-induced activation of metalloproteinases and diastolic stiffening.¹⁹⁵ Changes in gene expression caused by ß-AAs also involve a decrease in endothelin-1⁶⁸ and sarcoplasmic reticulum proteins²¹⁷ and an increase in atrial natriuretic factor.²⁵⁰ Furthermore, carvedilol decreases Fas receptor expression (cell death signalling receptor) after ischaemia–reperfusion injury, which leads to decreased apoptotic cell death.²⁵² Taken together, these observations suggest altered gene expression as a potential site of cardioprotection by ß-AAs.

Platelet aggregability and coagulation. The adrenergic system influences coagulation and fibrinolysis, particularly during episodes of heightened adrenergic drive, but contributes much less to baseline levels of coagulatory and fibrinolytic function. Epidemiological studies revealed a significant morning increase in the incidences of infarction, sudden cardiac death and transient myocardial ischaemia, which appears to be related to increased morning catecholamine levels and coagulation.^158 238 Perioperatively, increased catecholamine levels and tissue damage greatly increase the propensity to coagulation. Propranolol decreases thromboxane synthesis and platelet aggregation in patients receiving long-term propranolol treatment.³² One study evaluating the effect of metoprolol on platelet function did not find any inhibitory effect,²³⁹ whereas another study reports decreased platelet aggregability in patients with stable angina being tested for exercise stress.²⁴¹ Metoprolol also prevents stress-induced endothelial injury by increasing prostacyclin biosynthesis⁷⁵ and decreasing epinephrine-induced increases in von Willebrand factor antigen.¹²⁶ Esmolol has in vivo inhibitory action on neutrophil superoxide generation and platelet aggregation in a canine model of myocardial ischaemia–reperfusion.¹⁸⁶ ß-AAs also decrease the affinity of low-density lipoprotein to arterial proteoglycans and endothelial wall damage by reducing plasminogen activator inhibitor-1.²¹⁶ Conversely, increased platelet aggregability and decreased platelet cAMP production were reported after timolol treatment.²⁴⁰

ß-Adrenergic receptor down-regulation and target protein hyperphosphorylation. Prolonged and intensive perioperative agonist stimulation leads to desensitization and down-regulation of ß-ARs, which may seriously impair cardiac function.^28 29 Conversely, down-regulation of the sensitivity and number of receptors may be beneficial with respect to arrhythmogenicity.²⁵¹ Tachycardia,¹¹⁶ free oxygen radicals¹³³ and increased serum levels of TNF-,⁷⁹ which are, notably, all factors significantly affected by ß-AAs, were further implicated in the down-regulation of ß-ARs and the subsequent attenuated cardiovascular response. Importantly, down-regulation occurs after only a few hours of agonist stimulation.⁸⁸ At the molecular level, the process involves uncoupling of the ß-AR from Gs-proteins by PKA and ß-adrenergic receptor kinase 1 (ß-ARK) and binding of inhibiting arrestin to the receptor, which is followed by internalization and degradation or resensitization of the receptor (Fig. 004F3B). Early uncoupling and late down-regulation of myocardial ß-ARs were reported after cardiopulmonary bypass.^72 193 Similarly, persistent down-regulation and desensitization of ß-ARs were reported after thoracotomy or laparotomy throughout the first week after surgery.⁵ Hyperphosphorylation of channels and regulatory proteins such as the sarcoplasmic ryanodine receptor (RYR) may occur, resulting in hypersensitivity to cytosolic Ca²⁺.^147 148 ß-AAs potentially prevent hyperphosphorylation in the perioperative period, which is similar to their effects in congestive heart failure.⁹¹

Anti-arrhythmic effects. Sustained arrhythmias may be haemodynamically relevant and may affect the outcome adversely.¹³ Almost half of all high-risk cardiac patients undergoing non-cardiac surgery have ventricular ectopic beats or some sort of ventricular tachycardia.¹⁶⁶ Patients undergoing cardiac surgery have a high risk of developing new-onset atrial fibrillation.⁶ From a mechanistic point of view, sympathetic tone plays an important role in most ventricular as well as atrial arrhythmias.¹⁹² ß-AAs shift the autonomic balance towards a higher vagal and lower sympathetic tone.²²⁴ Studies on infarctions in pigs clearly showed that ß-adrenergic mechanisms play a major role in ventricular fibrillation threshold during experimental coronary occlusion.¹¹⁹ This is consistent with the notion that ß-AAs prevent sudden electrical cardiac death.^62 112 Regarding atrial arrhythmias, ß-AAs may be superior to newer class-III anti-arrhythmic drugs in the treatment of perioperative atrial fibrillation, as these drugs carry the risk of drug-induced polymorphic ventricular tachycardia. Notably, ß-AAs also counteract epinephrine-induced hypokalaemia, which significantly predisposes to arrhythmias.

Bioenergetics. ß-AAs are known to reduce NADH oxidase activity in mitochondria, which may lead to an energy-sparing effect.¹⁷⁷ Also, ß-AAs shift cellular metabolism from fatty acid oxidation to glucose utilization, which effectively reduces the myocardial oxygen requirement.²² Recently, oxidative metabolism was evaluated in patients with ventricular dysfunction using C-11-acetate positron emission tomography. The results of this study showed a significant reduction in cellular oxidative metabolism under metoprolol treatment.¹⁰ Accordingly, in patients undergoing cardiopulmonary bypass, chronic propranolol treatment reduces oxygen consumption.¹⁰⁸ Interestingly, patients receiving chronic ß-AA treatment compensate for reduced arterial oxygen content by increases in cardiac output and oxygen extraction, whereas patients not receiving ß-AA treatment demonstrate only an increase in oxygen extraction.²⁰⁶ Reduced production of lactate during exercise and increased oxygen extraction as a result of decreased cardiac output were reported previously under ß-adrenergic antagonism.¹⁷⁵ ß-AAs also prevent the decrease in mitochondrial CK activity after myocardial infarction.⁹⁷

Neuroendocrine stress response. There is a reduction in renin activity by selective as well as non-selective ß-AAs mediated by antagonizing ß-ARs, and some studies have even reported decreased catecholamine release after initiation of ß-adrenergic antagonism.¹⁵⁷ Most studies, however, did not observe a decrease in catecholamine serum levels²⁵⁴ but rather an increase.⁹³ One study in pigs reported decreased neuropeptide Y serum levels associated with increased heart rate variability after treatment with metoprolol.¹

Preconditioning. Theoretically, ß-AAs may prevent preconditioning of the heart, which renders it more resistant to subsequent sustained ischaemia. However, metoprolol does not neutralize the favourable effects of preconditioning.²³⁷ On the contrary, nipradilol, a nitric oxide-generating ß-AA, clearly induces preconditioning by itself.⁹⁵

Non-cardiac considerations
Effects of ß-adrenergic antagonism on core components of an anaesthetic regimen
Anaesthetic and analgesic requirements. Previous studies focused on the anti-ischaemic properties of peri- and intraoperative ß-adrenergic antagonism. Recently, it was shown that esmolol can potentiate the reduction in the minimum alveolar concentration (MAC) for isoflurane (–26% at esmolol 250 µg kg^–1 min^–1) and decrease anaesthetic requirements for skin incision during propofol/nitrous oxide/morphine anaesthesia (–27% at esmolol 250 µg kg^–1 min^–1).^102 103 Esmolol also decreases nociception in a variety of experimental settings, suggesting the potential to decrease the intraoperative anaesthetic requirements.⁴⁷ Altered distribution and decreased metabolism of opioids by ß-AAs may underlie this anaesthetic-sparing effect.¹⁸² Furthermore, although ß-AAs per se do not provide analgesia or hypnosis, they are known to have central nervous system modulating activities and anxiolytic effects.^76 146 158 ß-AAs potentially affect central nervous system pathways, which include neurones in the hypothalamus, hippocampus and cerebral cortex.^120 234 Accordingly, the favourable changes in heart rate variability after ß-AA treatment are ascribed to lower activity of the central sympathetic nervous system. In mice and rats, the locus coeruleus-associated noradrenergic system participates in arousal, and ß-adrenergic antagonism within this region reduces forebrain electroencephalographic activity.¹⁶ Similarly, amphetamine-induced activation of the rat forebrain is clearly inhibited by timolol, and in humans norepinephrine is known to enhance the responsiveness of the cerebral cortex to excitatory neuronal transmission.^15 181 Notably, pure ß-adrenergic antagonism is crucial for the observed anaesthetic-sparing effect because labetalol increases the anaesthetic requirement.⁴⁷ Even though esmolol and atenolol are hydrophilic ß-AAs, they produce the same plasma/cerebrospinal fluid ratio as lipophilic ß-AAs, thereby affecting the centrally located surface ß-ARs.¹²⁰ Another mechanism that may significantly contribute to the anaesthetic-sparing effect elicited by ß-AAs involves decreased excitatory stimulation of central nervous effector sites of hypnosis and somatic response. In this case, peripheral interruption of centripetal ß-adrenergic autonomic pathways, like spinal and epidural anaesthesia, decreases afferent input and anaesthetic requirement.¹⁶³

Memory storage. ß-AAs also possess attenuating effects on memory storage. Effects of opioids on memory are known to be mediated through noradrenergic influences.⁹⁸ Importantly, propranolol was reported to impair memory storage of particularly emotional events in humans.³¹ Also, ß-AAs impair arousal-induced enhancement of working memory in elderly patients.¹⁶⁵ As intraoperative recall and subconscious processing of information is particularly increased for emotionally charged information,¹³² it is tempting to speculate that ß-AAs, as anaesthetic adjuvants, might actually decrease the risk of intraoperative awareness and recall. However, this may be different for haemodynamically compromised patients with concomitant cardiovascular medication. Nonetheless, adequate depth of anaesthesia, as indicated by the bispectral index, was achieved in a group of elderly patients using high-dose intraoperative atenolol and a restricted amount of anaesthetic.²⁵⁴

Recovery. Faster recovery from anaesthesia was reported in patients receiving propranolol or metoprolol^101 208 and in patients receiving intra- or perioperative atenolol.²⁵⁴ Titration of anaesthetics to heart rate and blood pressure without administration of ß-AAs may lead to prolonged recovery from anaesthesia as a result of ‘relative overdosing’ with administered anaesthetics (MAC_BAR> MAC_AWAKE).¹⁸³ Furthermore, specific properties of ß-AAs may alleviate recovery from anaesthesia. In cats receiving atenolol, waking times were significantly prolonged,⁹² and human sleep disturbance is a well-known side-effect of ß-AAs.¹⁵⁹

Immune response and ß-adrenergic antagonism
The perioperative stress response impairs immune competence, particularly natural killer cell cytotoxic activity.²²⁰ In the experimental setting, reduction of natural killer cell cytotoxicity was achieved by electrical stimulation of the splanchnic nerve in rats, and was completely antagonized by nadolol.¹⁰⁹ Recently, Ben-Eliyahu and colleagues¹⁴ reported that hypothermia in barbiturate-anaesthetized rats suppresses natural killer cell cytotoxic activity and thereby accelerates the spread of tumour cells. Interestingly, nadolol attenuated the effect of hypothermia on natural killer cells and increased resistance to tumour metastasis.

	Tolerability of perioperative administration of ß-adrenergic antagonists

Top Abstract Choice of literature Clinical uses of perioperative... The present armamentarium of... Mechanisms and sites of... Tolerability of perioperative... Conclusions Addendum References

 
Contraindications to the use of ß-AAs result directly from their anti-adrenergic action. Drug intolerance greater than 20% as a result of decreased contractile function and increased afterload were previously reported in first-generation compounds.²¹⁵ However, drug tolerability for second-generation compounds is 80–100% and for third-generation compounds 90–100%.²⁷ Recent research has revealed an exceptionally low complication rate associated with the use of ß-AAs in heart failure patients as well as perioperatively in high-risk cardiac patients.^{142 173 179 225 231 254} Specifically, these studies do not report an increased number of episodes with severe hypotension, bradycardia or bronchospasm. Therefore, administration of ß-AAs according to the reported dosing by Mangano and colleagues¹⁴² and Poldermans and colleagues¹⁷³ does not require additional monitoring. Even when started acutely with high doses and in combination with potentially negative inotropic agents, ß-AAs were well tolerated in compromised patients. Nonetheless, certain contraindications to ß-AAs must be considered.

Haemodynamics
Excessive sympatholysis is undesirable in patients who depend heavily on central sympathetic tone for adequate circulatory function. While chronic administration of ß-AAs improves systolic and diastolic function in heart failure patients,^73 115 228 acute exposure to ß-AAs may lead to intolerable bradycardia and arterial hypotension and potentially result in an adverse outcome.²¹⁵ Accordingly, eye-drops with ß-AAs were implicated in the progression of ischaemic optic nerve disorders and the progression of visual loss attributable to recurrent nocturnal hypotensive episodes.⁸⁹ Therefore, patients with advanced conduction defects or symptomatic bradycardia should not receive ß-AAs without concomitant pacemaker therapy. However, patients with a resting heart rate below 60 beats min^–1 may receive therapy with caution. One study evaluating the tolerability of ß-AA titration in patients with idiopathic dilated cardiomyopathy found that generally accepted measures of the severity of heart failure were not predictive of problematic up-titration of ß-AAs.⁷ A low systolic blood pressure (<120 mmHg) was the strongest predictor of complications. The mechanisms underlying the good tolerability of initiation of ß-AAs in patients with heart failure was previously investigated by Halpern and colleagues⁸¹ (Fig. 004F5). Acute effects of metoprolol on systolic and diastolic function as well as on ventriculoarterial coupling were evaluated using volume–pressure loops. The results of these studies indicate that decreased afterload, as assessed by arterial elastance and the preservation of ventriculoarterial coupling and passive ventricular properties, explain the excellent tolerance of ß₁-AR antagonists in heart failure patients (Fig. 004F5). Notably, ß₁-AR antagonists also leave inotropic responses to ß₂-AR receptors intact and thereby produce less cardiac depression and vasoconstriction. Conversely, studies using a pulmonary artery catheter technique showed an increase in systemic vascular resistance (SVR) after metoprolol administration in chronically treated, mostly NYHA III heart-failure patients.¹²³ Similarly, acute graded administration of esmolol in patients with severe ventricular dysfunction increased SVR to compensate for the decrease in cardiac output.⁹⁹ Intraoperatively, SVR decreases significantly during undisturbed anaesthesia, but increases markedly under surgical stimulation in metoprolol as well as placebo-treated patients.¹³⁹ However, SVR depends on loading conditions and contractility, which were not measured in all these studies. Therefore, SVR may not properly reflect the effects of ß-AAs on distension in the arterial system, i.e. afterload. Accordingly, in mitral regurgitation-induced left ventricular dysfunction ß-AAs decrease left atrial hypertension independently of changes in heart rate, which is thought to result from decreased afterload.^33 228

View larger version (26K): [in this window] [in a new window]   Fig 5 Schematic depiction of left ventricular volume–pressure loops in a patient with heart failure. (A) Volume–pressure loop without ß1-adrenergic receptor (AR) antagonism. (B) Volume–pressure loop with acute exposure to ß1-A antagonism. End-systolic elastance (Ees) is decreased under ß1-AR antagonism, which is accompanied by decreased cardiac output, decreased stroke volume and decreased +dP/dtmax. In contrast, whereas the duration of active isovolumetric relaxation may increase only slightly (decrease in –dP/dtmax), passive diastolic function, as indicated by end-diastolic elastance (Eed, chamber stiffness), remains largely unaffected by ß1-AR antagonism. Importantly, afterload, as indicated by arterial elastance (Ea), is decreased by ß1-AR antagonism. Also, the ratio Ees/Ea, which represents ventriculoarterial coupling (the relationship between systolic function and afterload), is well preserved. Note that the area enclosed by the volume–pressure loops is closely related to myocardial oxygen consumption and is markedly reduced by ß-AAs.

 
Experience with recent large trials indicates that fewer than 5% of patients need to be withdrawn from ß-AAs because of worsening heart failure when they are carefully monitored. The favourable haemodynamic effects and safety of perioperative ß-AA administration is now also documented in several studies with brittle elderly surgical patients.^{142 173 179 225 231 254} As the potentially adverse clinical effects of esmolol completely disappear within minutes of its discontinuation, current experience suggests the initiation and up-titration of this short-acting ß₁-AR antagonist even in compromised patients perioperatively.⁹

Coronary vascular resistance
Sympathetic nerve stimulation in the presence of propranolol was reported to increase diastolic coronary resistance by 30%.⁶⁴ This coronary vasoconstriction is less likely to occur after administration of ß₁-AR antagonists.² More importantly, in a study that measured simultaneously total coronary flow (sinus outflow) and local tissue flow (heated thermocouples), sympathetic stimulation after ß-AA administration resulted in a decrease in sinus outflow but an increase in the nutritional microcirculatory flow.¹⁰⁷ This implies that a reduction in total coronary outflow does not necessarily parallel a decrease in tissue flow at the microcirculatory level. Also, the mechanism of the reduction in blood flow after administration of ß-AAs is likely to be a result of decreased myocardial oxygen demand. This is strongly supported by the notion that atenolol decreases myocardial blood flow by 16% and increases coronary vascular resistance by 23%.²¹¹ At the same time, however, the myocardial arteriovenous oxygen difference remains unaltered. Furthermore, when paced at the pre-atenolol heart rate, there is no decrease in coronary blood flow or increase in vascular resistance under atenolol treatment, which again clearly indicates that the observed decreases in coronary flow can be ascribed solely to decreased left ventricular work and myocardial oxygen demand. Adverse effects of ß-AAs on feed-forward coronary vasodilation (mediated by ß₂-ARs) clearly do not occur. Importantly, even in patients with vasospastic angina, administration of propranolol does not precipitate coronary spasms.⁵¹

Strategies to treat bradycardia and hypotension caused by ß-adrenergic antagonists
In general, untoward circulatory effects can be treated easily with vagolytic drugs (atropine) or can be overcome pharmacologically with inotropic agents. If atropine is not effective in treating bradycardia, i.v. glucagon 2.5 µg kg^–1 may be administered.^137 187 The haemodynamic improvements after glucagon treatment result mainly from its pronounced chronotropic effect. Importantly, ß-adrenergic agonists are not the inotropic agents of choice in treating cardiac decompensation from ß-adrenergic antagonism.¹³⁸ In the presence of fully established ß-adrenergic antagonism, high doses of catecholamines, which significantly increase afterload and pulmonary artery pressure, have to be administered to overcome the receptor antagonism.^{137 138 222} In metoprolol-treated patients, the response to phospho diesterase III inhibitor (PDE3I) is superior to the response to dobutamine. Milrinone 25 µg kg^–1 given over 10 min is well tolerated in patients with heart failure under carvedilol treatment and significantly improves the cardiac index with virtually no changes in heart rate and mean arterial pressure. PDE3Is retain their full haemodynamic effects in the face of ß-adrenergic antagonism, because their site of action is beyond ß-ARs. PDE3Is act specifically on the phospho diesterase III isoenzyme, which is anchored to the sarcoplasmic reticulum.²⁴⁶ Inhibitory effects on degradation of cAMP remain compartmentalized and thereby lead selectively to increased activity of sarcoplasmic reticulum PKA, which preferentially improves systolic and diastolic function.¹²⁹ Notably, the combination of glucagon and milrinone effectively restores cardiac output, but may increase heart rate excessively.¹⁸⁷ Interestingly, the combination of PDE3I and ß-AA, administered on a short- or long-term basis, confers additive beneficial but subtractive adverse effects (Table 6).¹⁹⁷ Accordingly, concomitant administration of PDE3Is was used to improve the tolerability of starting ß-adrenergic antagonism in patients with severe heart failure by counteracting the myocardial depressant effect of adrenergic withdrawal. The results of these studies suggest that the combined treatment may have beneficial effects beyond those produced by ß-AAs. Similarly, the use of prophylactic concomitant PDE3I administration may have the potential to facilitate the introduction of acute preoperative ß-adrenergic antagonism. Along with their favourable effects on cardiac function, PDE3Is also partially activate protein kinase G in bronchial smooth muscle, counteracting the bronchoconstriction of ß-AAs. In the future, adverse negative inotropic effects of ß-AAs may be reversed by the transient use of Ca²⁺ sensitizers, which are currently under investigation.¹⁹⁴