Portal de Periódicos da CAPES

Cardiovascular Responses to Sevoflurane

1995; Lippincott Williams & Wilkins; Volume: 81; Issue: Supplement Linguagem: Inglês

10.1097/00000539-199512001-00003

1526-7598

Thomas J. Ebert, Christopher P. Harkin, Michael Muzi,

Cardiac Ischemia and Reperfusion

Sevoflurane is a halogenated volatile anesthetic with a low blood:gas solubility and minimal pungency. If sevoflurane is shown to be safe and efficacious, it holds great promise for rapid induction of anesthesia when administered by mask, as well as for rapid elimination and awakening from anesthesia. Although sevoflurane has been in clinical use in Japan for several years and has been administered without evidence of adverse cardiovascular effects, additional clinical evaluation of the cardiovascular effects of sevoflurane in animals, human volunteers, and patients has been required prior to its approval for use in the United States. This review summarizes the effects of sevoflurane on the cardiovascular system. It is based on both published research and unpublished Food and Drug Administration Phase I and III single and multicenter trials of sevoflurane sponsored by Abbott Laboratories. This review will, wherever statistically possible, compare and contrast these data with other approved volatile anesthetics. The review is structured in sections that focus on different components of the circulatory system most relevant to the clinical anesthesiologist. Data derived from experimental animal preparations are presented separately from studies in both human volunteers and patient populations. Heart Rate and Rhythm Considerable interest has been generated concerning the effects of the newer volatile anesthetics on heart rate. This interest is in part because of the data indicating that desflurane (and to a lesser extent isoflurane) can initiate large increases in heart rate [1-3] that might predispose select patient populations to myocardial ischemia. In 1981, the initial human study in nonpremedicated volunteers evaluated 2%-3% sevoflurane over a 1-h period [4]. Heart rate was unchanged from conscious baseline throughout the study period. Further evidence of stable heart rates during the administration of sevoflurane was provided several years later by Manohar and Parks [5] in chronically instrumented healthy swine. Sevoflurane was given at 1 and 1.5 minimum alveolar anesthetic concentration (MAC) both with and without 50% nitrous oxide, and heart rate remained stable and unchanged from awake baseline throughout each anesthetic period. Despite the apparent stable heart rates in human volunteers and swine receiving sevoflurane, several studies in chronically instrumented dogs indicated that sevoflurane was associated with increases in heart rate [6,7]. Bernard et al. [6] and Harkin et al. [7] noted 30%-40% increases in heart rate from the awake state during the administration of 1.2-2.0 MAC sevoflurane. These increases were significantly larger than heart rate increases produced by equianesthetic concentrations of isoflurane. Additional data from nonpremedicated children also suggest that sevoflurane may be associated with small increases in heart rate [8]. Lerman et al. [8] administered sevoflurane via a mask and reported heart rate data from a measurement period immediately prior to tracheal intubation. In neonates and infants up to 3 yr of age, there were no changes in heart rate when sevoflurane was administered at approximate 1.0 MAC. However, in children 3-5 and 6-12 yr of age, heart rate increased approximate 3-10 bpm from awake baseline. Bradycardia did not occur in any of the 90 neonates, infants, and children under study. In two patients nodal rhythms were noted. In recent studies in volunteers and in patients, the administration of sevoflurane has been associated with stable and even lower heart rates compared with isoflurane. For example, in healthy, nonpremedicated young humans, aged 19-30 yr, sevoflurane (without adjuvants) was administered in doses ranging from 0.4 to 1.2 MAC (1%-3%) [9]. Heart rate was extremely stable throughout the administration period, and this was consistent with the 1981 work of Holaday and Smith [4]. When retrospectively compared with an identical study in an identical population receiving isoflurane, heart rates were lower during the administration of sevoflurane Figure 1. A direct comparison of heart rate responses to sevoflurane and isoflurane in elective surgery has been performed in healthy ASA grade I-II adult patients premedicated with intravenous midazolam and anesthetized with sodium thiopental [10]. The anesthetic gas concentrations were freely adjusted to maintain systolic pressure within +/- 20% of baseline, and the investigators noted that heart rates during the surgical period were significantly lower in patients receiving sevoflurane than in those receiving isoflurane Figure 2.Figure 1: Heart rate responses to increasing concentrations of sevoflurane, isoflurane, and desflurane in healthy volunteers. At lower MACs, neither desflurane nor sevoflurane was associated with increases in heart rate, whereas isoflurane caused an initial increase in heart rate that was sustained with increasing MAC. Desflurane at 1.5 MAC was associated with a significant increase in heart rate. *P < 0.05 compared with sevoflurane. [Adapted from Ebert and Muzi [2] and Ebert et al. [9].]Figure 2: Heart rate responses to sevoflurane and isoflurane anesthesia for elective surgical cases. Sevoflurane was associated with significantly lower heart rates (*P < 0.05) than isoflurane from the period immediately prior to surgical incision to 60 min after the initiation of surgery. [Adapted from Frink et al. [10].]When opioid adjuvants are included in the clinical care of patients during anesthesia, the heart rate differences between sevoflurane and isoflurane appear to be negligible. In a Phase III multicenter study, elderly ASA grade I-III patients with a mean age of 71 yr (range 57-93 yr) were randomized to receive either isoflurane (n = 64) or sevoflurane (n = 62) in conjunction with routine adjuvants (fentanyl, midazolam, and nitrous oxide). The average MAC-hour of anesthesia was 1.6 and 1.4 h for sevoflurane and isoflurane, and there were no significant differences in heart rate between study groups. In two additional multicenter studies of patients with heart and vascular disease, similar heart rate responses to sevoflurane and isoflurane when combined with fentanyl were noted. In one study, patients with existing coronary artery disease (CAD) or at high risk for CAD based on coexisting medical conditions underwent noncardiac surgery; in the other study, patients with CAD underwent elective coronary artery bypass graft procedures. In both studies, the incidence of tachycardia and bradycardia with sevoflurane was infrequent and not different from that in patients randomized to receive isoflurane. Additionally, there were no significant differences in heart rates between treatment groups before, during, and after anesthesia. Desflurane and, to a lesser extent, isoflurane have been associated with periods of tachycardia when first administered into the inspired gas after intravenous induction of anesthesia and after increasing the inspired concentration during steady state periods of anesthesia [1-3]. During similar rapid manipulations, the administration of sevoflurane has not led to increases in heart rate in nonpremedicated volunteers [9]Figure 3. The summary of all adverse heart rate experiences in the Abbott Clinical Program revealed no significant differences between isoflurane and sevoflurane in the incidence of tachycardia or bradycardia during the perioperative period in adults Figure 4. The summary from studies in the pediatric population indicated a lower incidence of bradycardia and arrhythmias in infants and children receiving sevoflurane compared with those receiving halothane.Figure 3: Neurocirculatory responses of healthy volunteers to a rapid advancement of the anesthetic vaporizer. The rapid increase in the inspired concentration of three volatile anesthetics was initiated after a 30-min stabilization period at either 0.8 (sevoflurane) or 1.0 minimum alveolar anesthetic concentration (MAC). The down arrow indicates the point of advancement of the vaporizer from 2% to 3% sevoflurane, from 7% to 11% desflurane, or from 1.2% to 1.8% isoflurane. The response to desflurane was unique and was associated with large increases in sympathetic nerve activity leading to substantial increases in heart rate and mean blood pressure. [Adapted from Ebert and Muzi [2] and Ebert et al. [9].]Figure 4: The total number of adverse experiences with regard to the incidence of bradycardia and tachycardia in more than 2800 adults and nearly 1500 pediatric patients. Bradycardia was defined as a larger than 20% decrease in heart rate from ward baseline. Tachycardia was defined as a larger than 20% increase in heart rate from ward baseline. These data represent the adverse experiences that, in the judgment of the clinical investigator, were directly related to the study drug administration. In adults the comparative drug was isoflurane, and in the pediatric population the comparative drug was halothane. The only significant differences were noted in the incidence of bradycardia in the pediatric studies, which was less with sevoflurane than with halothane. *P < 0.05. The adverse experiences related to increases and decreases in blood pressure in the adult patients enrolled in the Abbott Clinical Program are presented as well. Hypertension and hypotension were defined as a systolic pressure decrease greater than 20% above or below ward baseline.The volatile anesthetics can alter the dose ("sensitize the heart") of epinephrine that causes cardiac arrhythmias. The arrhythmogenic threshold of epinephrine was determined in dogs during 1.3 MAC sevoflurane, enflurane, and isoflurane anesthesia and retrospectively compared with halothane [11]. Threshold was defined as the dose of epinephrine that produced four or more premature ventricular contractions (PVCs) within 15 s. This was 5.2, 8.6, and 9.8 micro gram centered dot kg-1 centered dot min-1 for enflurane, sevoflurane, and isoflurane, respectively, and these doses were substantially larger than the epinephrine dose causing arrhythmias with halothane. In ASA grade I or II patients undergoing transphenoidal surgery, the arrhythmogenicity of epinephrine applied to the oral and nasal mucosa has been determined during the administration of 1-1.3 MAC sevoflurane or isoflurane [12]. No PVCs were noted with either anesthetic when epinephrine doses were less than 5 micro gram/kg. The frequency of PVCs did not differ between anesthetics when larger doses of epinephrine (5 to 14.9 micro gram/kg) were given. Sevoflurane has been administered without sequelae to patients undergoing surgical resection of a pheochromocytoma [13]. Arterial Blood Pressure Blood pressure responses to the volatile anesthetics are a function of their effects on cardiac output and vascular resistance. Each of these two major components of blood pressure homeostasis is influenced by the direct effects of the anesthetic on the heart and vascular smooth muscle, and by the indirect effects of the anesthetic on the autonomic nervous system. All potent volatile anesthetics alter these factors and, in general, do so in a dose-related fashion. The question arises whether sevoflurane alters blood pressure in any unique way compared with equianesthetic concentrations of other volatile anesthetics. In experimental animal preparations, differences between anesthetics are small and may be species specific. For example, in a study of mongrel dogs, increasing concentrations of sevoflurane decreased blood pressure to a greater extent than did isoflurane [14]. In three subsequent studies in chronically instrumented dogs, the blood pressure-decreasing effects of sevoflurane and isoflurane were virtually indistinguishable [6,7,12]. However, in chronically instrumented swine, a retrospective comparison suggested that the mean arterial pressure (MAP) decrease during the administration of sevoflurane was less than that with equianesthetic concentrations of halothane and isoflurane [5]. In spontaneously breathing rats, sevoflurane was noted to decrease blood pressure less than halothane; this was due to a better preservation of cardiac output with sevoflurane [15]. The experiences with sevoflurane in unstimulated human volunteers and pediatric patients are also inconsistent. In the pediatric population, with subjects ranging in age from newborn to 12 yr, the decrease in systolic blood pressure associated with approximate 1 MAC sevoflurane was inversely related to age [8]Figure 5. Additionally, the decrease with sevoflurane was less than the decrease noted in an earlier study using equianesthetic concentrations of desflurane [16,17]. In contrast, in nonpremedicated human volunteers, there were no differences in the effect of sevoflurane in concentrations up to 1.2 MAC on blood pressure when directly compared with desflurane and retrospectively compared with isoflurane [2,9]Figure 6. Sevoflurane at 1.2 MAC was associated with a approximate 30% decrease in MAP.Figure 5: Systolic blood pressure responses to the administration of sevoflurane in each of six pediatric age groups. Awake, resting baseline data are provided as well as responses to approximate 1 minimum alveolar anesthetic concentration (MAC) sevoflurane before and after skin incision. Systolic pressure decreased significantly during administration of sevoflurane in all groups except the children 5-12 yr of age (P < 0.05). Systolic pressure returned toward awake values after skin incision but remained significantly less than awake values in the neonate and 6- to 12-mo age groups (P < 0.025). [Adapted from Lerman et al. [8].]Figure 6: Mean arterial pressure and forearm vascular resistance responses to the administration of desflurane, isoflurane, and sevoflurane in healthy volunteers. With increasing minimum alveolar anesthetic concentration, there were progressive decreases in blood pressure with each of the volatile anesthetics. Forearm vascular resistance was, in general, progressively decreased with increasing minimum alveolar anesthetic concentrations of each of the volatile anesthetics; however, this decline was less in the group receiving sevoflurane. [Adapted from Ebert and Muzi [2] and Ebert et al. [9].]Additional Phase I data comparing sevoflurane with isoflurane in subjects not undergoing surgery have been generated by Malan and colleagues [18]. Their studies were conducted in healthy, young, (19- to 34-yr-old) nonpremedicated volunteers receiving either isoflurane (n = 6) or sevoflurane (n = 7) without adjuvants. The study was designed such that if MAP decreased to less than 50 mm Hg at any steady state concentration of anesthetic, the protocol was terminated for that subject. The data indicate that at higher MAC, the incidence of hypotension was more frequent with isoflurane than with sevoflurane. However, in several clinical studies of ASA grade I-II [19] and II-IV (on file, Abbott Laboratories, Abbott Park, IL) surgical patients who were randomized to receive either isoflurane or sevoflurane as their primary anesthetic, there were no significant differences in the blood pressure responses and the incidence of hypertension and hypotension (defined as 20% change from ward baseline) during both simple and complex surgical procedures Figure 4 and Figure 7. These data need to be qualified however, since most clinical protocols allowed the investigator to vary the anesthetic concentration to maintain stable blood pressure during the surgical case and fluid administration was not a controlled variable. Moreover, the use of other anesthetic adjuvants was included as part of the routine management of these patients. Thus, differences in blood pressure between anesthetics may have been difficult to detect due to routine clinical management that was guided by blood pressure.Figure 7: The blood pressure and heart rate responses of patients with coronary artery disease (CAD) or at high risk for the presence of CAD undergoing elective noncardiac surgery and randomized to receive sevoflurane (n = 19) or isoflurane (n = 18). Although there was a tendency for heart rates to be slightly faster during the time period after intubation up to the period shortly after surgical incision in the isoflurane group, there were no statistical differences between anesthetics throughout the period of observation.Cardiac Output and Peripheral Resistance To better evaluate possible differences between the blood pressure effects of the volatile anesthetics, a careful examination of the two key determinants of blood pressure, cardiac output and peripheral resistance, is required. Sevoflurane and isoflurane have been compared in a study using a basal anesthetic of alpha-chloralose in artificially ventilated rats [20]. The concentration of the volatile anesthetic was titrated to achieve a MAP decrease to 70 and then 50 mm Hg. Once these pressures were achieved, measurements of cardiac output and calculations of systemic vascular resistance were made. It was fortuitous that the calculated MAC of the two anesthetics responsible for each step decrease in MAP were of equal potency. To achieve 70 mm Hg pressure, approximate 0.7 MAC was used and the effects of sevoflurane and isoflurane on cardiac output and systemic vascular resistance were identical. At 50 mm Hg (approximate 1.7 MAC), isoflurane was associated with a larger decrease in peripheral resistance and a smaller decrease in cardiac output than sevoflurane [20]. In spontaneously breathing rats, 1.0 MAC sevoflurane reduced blood pressure and cardiac output less than 1.0 MAC halothane [15]. In dogs, the cardiac output and peripheral resistance responses to sevoflurane and isoflurane at 1.2 and 2.0 MAC were virtually identical [6]. Neither anesthetic changed cardiac output at 1.2 MAC, whereas both significantly decreased cardiac output (approximate 17%) at 2 MAC. Peripheral resistance significantly and progressively decreased with increasing MAC [6]. In a Phase I human study that used thermodilution measurements of cardiac output, there were no differences in the cardiac output and systemic vascular resistance responses between sevoflurane and isoflurane, although the protocol design was such that subjects with excessive decreases in blood pressure during the anesthetic administration were dropped from the group means [18]. This could have obscured differences between anesthetics. Regional Blood Flow The hepatic circulation is one of the more important regional circulations because of the possibility that depression of liver blood flow may contribute to the development of anesthetic-related hepatic injury and may influence the pharmacokinetic profile of drugs metabolized in the liver. In two separate studies in chronically instrumented dogs, sevoflurane administered up to 2.0 MAC preserved hepatic arterial blood flow despite decreasing cardiac output and blood pressure [21,22]. This preservation of flow was similar to that achieved with isoflurane, whereas the administration of halothane and enflurane was associated with decreases in hepatic arterial flow (approximate 50% and approximate 40%, respectively, at 1.5 MAC). Similarly, in both artificially ventilated and spontaneously breathing rats, sevoflurane better maintained liver blood flow than equianesthetic concentrations of halothane [15,20]. Portal venous blood flow also was preserved with the administration of sevoflurane up to 1.0 MAC [21,22]. Renal blood flow was unchanged with 1.0 MAC sevoflurane but decreased (approximate 18%) with 1.0 MAC halothane [15]. Splenic, pancreatic, and lung blood flows were better preserved with sevoflurane than with isoflurane, whereas intestinal and muscle blood flows did not differ [20]. In humans, muscle blood flow can be estimated with forearm plethysmography [23]. In nonpremedicated human volunteers, increasing the MAC of sevoflurane increased forearm blood flow and decreased forearm vascular resistance less than did increasing the MAC of isoflurane [2,9]Figure 6. The halogenated volatile anesthetics generally increase cerebral blood flow (CBF) and decrease cerebral metabolic rate (CMRO2) in a dose-related fashion. In swine, sevoflurane had insignificant effects on brain blood flow and cerebral vascular resistance when retrospectively compared with halothane and isoflurane [24]. In spontaneously breathing rats, 1.0 MAC sevoflurane increased CBF (approximate 35%), but this was significantly less than increases (approximate 64%) produced by 1.0 MAC isoflurane [15]. Because the cerebral metabolic demands were unknown in these studies, the adequacy of blood flow and the possibility of overperfusion could not be determined. Scheller et al. [25,26] have provided additional comparisons of sevoflurane and isoflurane in rabbits and dogs and noted that, although neither anesthetic had significant effects on CBF at 0.5-1.0 MAC, both anesthetics increased intracranial pressure (ICP). These data suggest that sevoflurane and isoflurane caused cerebral venodilation [27]. However, in the Abbott Clinical Program, sevoflurane up to 1.5 MAC has not altered ICP dynamics. In one small preclinical study of eight patients with supratentorial mass lesions, it was possible in all patients (n = 4) to maintain ICP within 5 mm Hg of awake baseline during the administration of 1.0 MAC sevoflurane by establishing hypocapnia. This control was only possible in one of four patients receiving isoflurane despite aggressive hyperventilation. In terms of CMRO2, sevoflurane acts much like isoflurane and reduces CMRO2 by 50% at 2.0 MAC. Sevoflurane has not been associated with epileptiform activity on the electroencephalogram during either normocapnia or hypocapnia and appears much like isoflurane. In dogs, sevoflurane at 2.15 MAC produced burst suppression and significant decreases (approximate 35%) in CMRO2 without changing CBF from baseline, and these effects did not differ from those of isoflurane [25]. Because of the lower blood:gas solubility of sevoflurane and its favorable effects on cerebral physiology, the use of sevoflurane may have an important role in neurosurgical procedures when intraoperative or early postoperative awakening is planned for neurologic evaluation. Autonomic and Baroreflex Effects In rats, direct measures of efferent sympathetic and parasympathetic nerve activity to the heart have been made during the administration of sevoflurane [28]. With increasing concentrations of sevoflurane, cardiac sympathetic nerve traffic decreased but parasympathetic traffic was unchanged, which might explain, in part, the absence of tachycardia with increasing MAC of sevoflurane. Similar data with other volatile anesthetics with this experimental preparation have not been reported. In humans, direct measures of sympathetic nerve traffic to blood vessels supplying skeletal muscle can be made with a technique called sympathetic microneurography [29,30]. This nerve traffic contributes to the regulation of muscle blood flow, which represents about 40% of the resting cardiac output. During steady state periods of administration of sevoflurane and isoflurane ranging from 0.4 to 1.2 MAC, basal levels of sympathetic traffic and plasma catecholamines were unchanged from conscious baseline Figure 8. This contrasts with the increased level of sympathetic nerve traffic and plasma norepinephrine associated with the administration of desflurane at higher MAC. Desflurane has also been associated with periods of substantial sympathoexcitation, tachycardia, and hypertension when it is first introduced into the inspired gas shortly after intravenous induction of anesthesia [2] and during periods of steady state anesthesia when the inspired concentration of desflurane is abruptly increased by 1% or 3.6% [2,3,31]Figure 3. Isoflurane can provoke a response that is qualitatively similar to that of desflurane when abruptly increased, but the magnitude of the response is far less than that with desflurane [3,32]. The administration of sevoflurane in rapidly increasing inspired concentrations has not been associated with any signs of sympathetic activation in humans [9]Figure 3.Figure 8: Sympathetic nerve activity and plasma norepinephrine concentration responses to steady state periods of anesthesia with desflurane, isoflurane, or sevoflurane in healthy volunteers. During the administration of isoflurane and sevoflurane, there were no significant changes in basal levels of sympathetic nerve activity and circulating norepinephrine concentrations. However, at desflurane concentrations above 0.5 minimum alveolar anesthetic concentration, there were increases in sympathetic outflow leading to significant increases in plasma norepinephrine. Bursts/100HB = bursts per 100 heartbeats. [Adapted from Ebert and Muzi [2] and Ebert et al. [9].]The baroreflex network is an important short-term pressure-regulating system that helps maintain blood flow to vital organs in the face of hemodynamic variations. The functional performance of the baroreflex feedback system can be assessed by studying the response to hypo- or hypertensive challenges. There are two main effector limbs by which the baroreflex maintains blood pressure. The first is the cardiac limb, in which blood pressure perturbations induce reflex changes in the cardiac (R-R) interval, thereby altering heart rate and cardiac output. This response is mediated primarily via the vagus or parasympathetic branch of the autonomic nervous system. The second effector limb of the baroreflex is the sympathetic nervous system, in which blood pressure changes initiate reflex adjustments in peripheral sympathetic outflow to adjust vascular tone. Both effector limbs can be directly quantified in animals and humans by measuring the change in heart rate and efferent sympathetic nerve activity in response to a given change in blood pressure. In humans, reflex heart rate and sympathetic nerve activity responses (quantitated as the baroslope) to pressure perturbations are diminished with increasing MAC of sevoflurane, and this decrease in sensitivity appears to be generally similar to that produced by isoflurane and desflurane Figure 9 and Figure 10.Figure 9: Summary data of the baroreflex regulation of heart rate (R-R interval) in response to a decreasing pressure stimulus (sodium nitroprusside) or in response to an increasing pressure stimulus (phenylephrine). These data were acquired in healthy volunteers who were randomized to receive either isoflurane, desflurane, or sevoflurane. With increasing minimum alveolar anesthetic concentration (MAC), each of the volatile anesthetics led to a progressive reduction in the cardiac baroslope (an index of baroreflex sensitivity derived by relating changes in mean pressure to changes in R-R interval). There were no statistical differences between anesthetics.Figure 10: The sympathetic baroreflex function of healthy volunteers randomized to receive isoflurane, desflurane, and sevoflurane. The slope (sensitivity) is the relationship between decreasing diastolic pressure and increasing sympathetic nerve activity directed to the vasculature of skeletal muscle. The reflex regulation of sympathetic outflow was fairly well preserved with increasing minimum alveolar anesthetic concentration (MAC) of anesthetic, the exception being a slightly greater depression of the sympathetic baroslope at 1 MAC sevoflurane compared with desflurane (P < 0.05).The reflex control of the mesenteric capacitance veins has been evaluated in rabbits [33]. Reflex control of vein diameter, blood pressure, and heart rate in response to brief periods of bilateral carotid occlusion or aortic depressor nerve stimulation was significantly reduced (approximate 80%) by sevoflurane, and this attenuation was similar to the inhibition of these reflex responses produced by equianesthetic concentrations of halothane, isoflurane, and desflurane. Myocardial Function/Contractility There have been several in vivo investigations assessing the effects of sevoflurane on ventricular function. Bernard et al. [6] compared the effects of sevoflurane and isoflurane in chronically instrumented dogs. They reported that these anesthetics produced essentially equivalent decreases in myocardial contractility, as evaluated by peak positive left ventricular change in pressure per unit of time (dP/dt). Left ventricular dP/dt is an isovolumic index of contractile state and is dependent on changes in heart rate and preload [34]. In dogs, sevoflurane and isoflurane produced very similar heart rate and preload changes; this result implies that sevoflurane decreases the intrinsic inotropic state to a similar degree as isoflurane. Harkin et al. [7] evaluated the effects of sevoflurane on both systolic and diastolic function in chronically instrumented dogs. They demonstrated that sevoflurane produced dose-dependent decreases in myocardial contractility as evaluated by the dP/dt and by the slope (MW) of the regional preload recruitable stroke work relationship, a relatively heart rate- and load-independent index of contractile state in vivo [35]Figure 11. At 1.0 MAC sevoflurane, contractile indices were reduced about 25%. They further determined that this decrease in contractility was independent of autonomic nervous system tone, as the decrease in MW was nearly identical in autonomically intact and blocked animals. These investigators also reported that sevoflurane caused diastolic dysfunction by producing dose-dependent increases in the time constant for isovolumic relaxation, decreases in peak negative left ventricular dP/dt, and decreases in rapid ventricular filling. Diastolic mechanisms have been shown to play an important role in overall cardiac performance, and diastolic dysfunction may be an important contributing factor in the pathogenesis of congestive heart failure [36]. Their retrospective analyses indicated that the effect of sevoflurane on contractility indices and diastolic mechanics was qualitatively similar to the effects of isoflurane and desflurane [7]Figure 12.Figure 11: Myocardial contractility indices from chronically instrumented dogs. For these measurements, pharmacologic blockade of the autonomic nervous system was established to eliminate neural or circulating humoral influences on the inotropic state of the heart. The conscious control data were assigned 100%, and subsequent reductions in the inotropic state are depicted for both 1 and 1.5 minimum alveolar anesthetic concentrations of sevoflurane, desflurane, and isoflurane. There were no differences between these three volatile anesthetics. MW = slope of the regional preload recruitable stroke work relationship; dP/dt50 = change in pressure per unit of time. [Adapted from Pagel et al., Acta Anaesthesiol Scand 37:203, 1993, and Harkin et al. [7].]Figure 12: The absolute coronary flow rate and coronary flow reserve as measured from the isolated nonworking perfused rat heart prior to and during the administration of sevoflurane and isoflurane. The response to adenosine was assumed to represent the maximum vasodilatory capacity of the coronary vessels. This remained fairly constant throughout the experimental protocol regardless of the administration of the volatile anesthetic. The middle line in each tracing represents coronary flow determined during the anesthetic administration. The vertical distance between this line and the adenosine line represents the coronary flow reserve for each condition. The reduction in coronary reserve with isoflurane suggests that isoflurane is a more potent coronary vasodilator than sevoflurane. [Adapted from Larach et al. [40].]Kikura and Ikeda [37] compared the effects of sevoflurane/nitrous oxide and enflurane/nitrous oxide on myocardial contractility in ASA grade I and II patients prior to elective surgery. These investigators used transthoracic (in conscious patients) and transesophageal echocardiography (in anesthetized patients) to assess left ventricular end-systolic wall stress (LVESWS, a quantitative measure of afterload) and the velocity of circumferential fiber shortening (Vcfs) corrected for heart rate. The LVESWS-Vcfs relationship has been shown to be a relatively preload-independent index of myocardial contractility that incorporates changes in afterload into its calculation [38]. Although different patient groups were studied in the conscious and anesthetized states, Kikura and Ikeda's results suggested that patients anesthetized with 1.5 MAC sevoflurane/nitrous oxide had approximate 15%-20% smaller decreases in myocardial contractility compared with those anesthetized with 1.5 MAC enflurane/nitrous oxide [37]. Because the linearity of the LVESWS-Vcfs relationship degenerated at higher anesthetic concentrations, it was difficult to quantify this index of contractility. However, changes in more conventional echocardiographically derived measures of contractile function, including Vcfs alone, fractional shortening, and ejection fraction, supported the conclusion that, in humans, sevoflurane produced smaller decreases in myocardial contractility than did enflurane. Malan et al. [18,39] recently reported results from an investigation of the cardiovascular effects in healthy human volunteers. They assessed myocardial contractility using an echocardiographic transthoracic left parasternal view in awake subjects and a transesophageal short access view in sevoflurane-anesthetized subjects. With increasing doses of sevoflurane from 0 to 2.0 MAC, no changes in myocardial contractility, as evaluated by fractional area of contraction and Vcfs, were observed. Coronary Flow and Resistance Several animal studies have evaluated myocardial perfusion using radionucleotide microspheres and examined whether sevoflurane disrupts the link between coronary blood flow and myocardial oxygen consumption. Manohar and Parks [5] observed that sevoflurane produced dose-dependent and parallel decreases in myocardial perfusion and myocardial oxygen consumption (estimated with rate pressure product) in chronically instrumented swine. Conzen et al. [20] derived the same conclusions in rats, and also reported that sevoflurane produces dose-dependent decreases in calculated coronary vascular resistance. Crawford et al. [15] observed that sevoflurane did not alter coronary blood flow in their rats; however, myocardial oxygen consumption also remained unchanged. These investigations demonstrated that sevoflurane maintained the coupling of myocardial perfusion to myocardial metabolic demands, but they did not provide insight as to the possibility that sevoflurane might be a substantial coronary vasodilator. A comparison of sevoflurane, halothane, and isoflurane on coronary blood flow and coronary vascular resistance in vitro has been made by Larach and Schuler [40]. They used the nonworking isolated perfused rat heart so that they could dissociate indirect processes affecting coronary vascular resistance (e.g., loading conditions, heart rate, inotropic state) from the direct anesthetic actions. They defined coronary vasodilator reserve as the difference between coronary flow during the administration of a volatile anesthetic and maximum coronary flow produced by adenosine [40]. In this model, sevoflurane caused coronary vasodilation but appeared to do so in a nonlinear manner. At higher MAC, there was a "plateau" in the vasodilator reserve with sevoflurane that was in contrast with the linear reduction in vasodilatory reserve noted with halothane and isoflurane with increasing MAC Figure 12. These results suggested that sevoflurane was a less potent direct coronary vasodilator than halothane or isoflurane in the rat model. In a separate in vivo evaluation of rat coronary hemodynamics, there was a more pronounced decrease in coronary vascular resistance with isoflurane than with equianesthetic concentrations of sevoflurane for a given reduction in myocardial oxygen demand [20]. Whether this limited effect of sevoflurane on coronary vasodilation in rats is present in other species has been explored in several whole animal investigations. Bernard et al. [6] compared the effects of sevoflurane and isoflurane on the coronary circulation in chronically instrumented dogs. They observed that isoflurane increased Doppler-derived coronary blood flow velocity, and this was associated with linear decreases in coronary vascular resistance. In contrast, the dose-response relationship of sevoflurane was not linear with increasing MAC; rather a plateau was present with no change in coronary vascular resistance between 1.2 and 2.0 MAC. Harkin et al. [7] studied dogs after autonomic blockade of the heart. Increasing concentrations of sevoflurane did not change Doppler-derived coronary blood flow velocity, but it caused decreases in coronary vascular resistance. These two studies in dogs [6,7] support earlier data suggesting that sevoflurane produces decreases in coronary vascular resistance and also provide support for the possibility that sevoflurane is a less potent coronary vasodilator than isoflurane. However, in these two studies, sevoflurane also was shown to decreased myocardial contractility and estimated myocardial oxygen consumption, raising the possibility that it may cause coronary vasodilation beyond myocardial metabolic needs, which might predispose to "coronary steal." This possibility was evaluated by Hirano et al. [41], who measured aortic and coronary sinus blood in acutely instrumented dogs and reported dose-dependent reductions in myocardial oxygen extraction and consumption with sevoflurane. They further noted that measured coronary diameter remained unchanged over the same concentrations of this anesthetic. These data imply that coronary blood flow would increase with sevoflurane, but a decrease was actually observed. They suggested that sevoflurane caused coronary vasodilation, but whether this dilation increased myocardial perfusion beyond myocardial oxygen demands remained inconclusive. However, Kersten et al. [42] provided better insight into whether the vasodilatory activity of sevoflurane is sufficient to produce coronary steal. These investigators used an extensively validated chronically instrumented canine model of multivessel coronary artery obstruction [43]. Coronary steal was defined as an absolute decrease in perfusion to ischemic areas of the myocardium that were dependent on collateral coronary blood flow during sevoflurane- or adenosine-mediated increases in flow to normal zones. Coronary perfusion pressure and heart rate were held constant during these studies [42]. They reported that increasing concentrations of sevoflurane did not reduce collateral perfusion to ischemic myocardium. In fact, this anesthetic preferentially increased collateral blood flow when aortic pressure was held constant. Thus, unlike adenosine, sevoflurane lacked the potent coronary vasodilatory properties necessary to cause coronary steal. Myocardial Ischemia/Infarction The influence of sevoflurane and isoflurane on the incidence of myocardial ischemia and infarction have been evaluated in two Phase III multicenter studies. In one Phase III study of adult patients at risk for myocardial ischemia and undergoing noncardiac surgery, sevoflurane was compared with isoflurane on the basis of incidence of intraoperative myocardial ischemia and postoperative incidence of ischemic events, including nonfatal myocardial infarction, development of unstable angina, and cardiac death. The study population consisted of patients with documented CAD or at high risk for CAD (three or more risk factors) scheduled for elective, noncardiac surgery (excluding cardiac, carotid, or suprarenal aortic surgery). Induction of anesthesia was consistent for each patient and included thiopental (2-5 mg/kg), fentanyl (3-5 micro gram/kg), and vecuronium (0.1-0.2 mg/kg). Additional fentanyl (1-2 micro gram/kg bolus) and vecuronium were used when necessary. Oxygen was maintained at an inspired concentration of 30%-50% (50%-70% nitrous oxide), and the end-tidal anesthetic concentration did not exceed 1.5 MAC (3% for sevoflurane and 1.8% for isoflurane). Blood pressure and heart rate were maintained within 20% of baseline by first changing the end-tidal anesthetic concentration and if unsuccessful, with pharmacologic adjuvants. Two-channel Holter electrocardiogram (ECG) recording began approximately 12 h prior to surgery and was maintained through 48 h after surgery. Twelve-lead ECG examinations were performed prior to surgery and 24 and 48 h postsurgery. Blood samples were also drawn for creatine phosphokinase (CPK-MB) analysis by a central laboratory. Results of the 12-lead ECG analyses and CPK-MB analyses were used to determine the incidence of myocardial infarction. The incidence of ischemic events detected by Holter recording did not differ between study groups, either before induction, during maintenance, or in the postoperative period Table 1. There were no differences in the incidence of myocardial injury between the sevoflurane and isoflurane groups. The percentage of patients with significant CPK-MB increases suggestive of myocardial injury averaged 10% in the sevoflurane and 11% in the isoflurane group. The percentage of patients with Q wave or ST-T wave changes suggestive of myocardial injury averaged 12% in the sevoflurane group and 10% in the isoflurane group.Table 1: The Incidence of Ischemic Events on Holter Monitoring in Patients with Documented Coronary Artery Disease (CAD) or at High Risk for CAD Undergoing Noncardiac SurgeryIn a separate Phase III multicenter, open-label, randomized, comparative trial of sevoflurane versus isoflurane as an adjunct with narcotics in adult ASA grade II-IV patients undergoing elective coronary artery bypass graft surgery, Holter monitoring up to the time of cardiopulmonary bypass (CPB) and serial 12-lead ECGs and CPK-MB blood analyses were used to detect ischemia and infarction. Two hundred seventy-three patients were enrolled; each received oral diazepam and intramuscular morphine sulfate prior to surgery, and midazolam (0.1-0.3 mg/kg), fentanyl (5-15 micro gram/kg), and vecuronium for induction of anesthesia. After tracheal intubation, the volatile anesthetic was maintained at approximate 0.5-1.0 MAC and supplemental doses (up to a maximum of 10 micro gram/kg) of fentanyl could be administered until CPB. At the onset of CPB, the volatile anesthetic was discontinued and additional fentanyl, midazolam and, vecuronium were given at the investigator's discretion. The number of grafts, duration of CPB and aortic cross-clamp, and dose of fentanyl (approximate 17 micro gram/kg) did not differ between groups. The average percentage of MAC of volatile anesthetic used prior to CPB did not differ between groups. The incidence of myocardial ischemia as detected by Holter analysis was similar between treatment groups at all periods of observation Table 2. Average CPK-MB fraction increases postoperatively did not differ between groups. The incidence of perioperative myocardial infarction by 12-lead ECG averaged 8% in the sevoflurane-treated group and 10% in the isoflurane group, and these incidences were not statistically different. Thus, it appears that in two high-risk populations for myocardial ischemia and infarction, sevoflurane did not differ from isoflurane and appeared to be safe and efficacious.Table 2: The Incidence of Ischemic Events During Holter Monitoring in Cardiac Patients Undergoing Coronary Artery Bypass SurgerySummary In conclusion, sevoflurane appears to be similar to isoflurane and desflurane with a few exceptions. Sevoflurane was not associated with increases in heart rate in adult patients and volunteers, whereas higher MACs of isoflurane and desflurane and rapid increases in the inspired concentrations of these two anesthetics have been associated with tachycardia. Increasing concentrations of sevoflurane progressively decrease blood pressure in a manner similar to the other volatile anesthetics, and in unstimulated volunteers this decrease may be slightly less than with isoflurane at a higher MAC. Sevoflurane appears similar to isoflurane in its effect on regional blood flows, including the hepatic, renal, and cerebral circulation. In animals, sevoflurane appears to be a slightly less potent coronary vasodilator than isoflurane, and in a dog model, sevoflurane has not been associated with coronary flow redistribution ("steal"). Sevoflurane decreases myocardial contractility in a manner similar to equianesthetic concentrations of isoflurane and desflurane, and does not potentiate epinephrine-induced cardiac arrhythmias. Sevoflurane reduces baroreflex function in a manner similar to other volatile anesthetics. In several multicenter studies where patients with CAD or patients at high risk for CAD were randomized to receive either sevoflurane or isoflurane for cardiac or noncardiac surgery, the incidence of myocardial ischemia, infarction, and cardiac outcomes did not differ between treatment groups. Thus, sevoflurane has not been associated with untoward cardiovascular changes in volunteers and patients undergoing elective surgery compared with other volatile anesthetics, and it appears to offer a more stable heart rate profile than either isoflurane or desflurane.