Myocardial Depressant Effects of Sevoflurane: Mechanical and Electrophysiologic Actions In Vitro

According to a protocol approved by the Yonsei University College of Medicine Animal Research Committee, the heart was removed from female guinea pigs (weighing 300-400 g) or Sprague-Dawley rats (weighing 300-400 g) after 50 mg/kg intraperitoneal pentobarbital sodium injection), and papillary muscles were isolated from the right ventricles. The excised muscle was mounted horizontally in a tissue bath with a volume of 1.5 ml. The bath was superfused at 37 degrees C at a rate of 8 ml/min with modified Tyrode's solution (in mM): 143 Na sup +, 5 K sup +, 2 Ca²+, 127 Cl sup -, 1.2 MgSO⁴, 25 HCO³sup -, 11 glucose, 0.10 ethylenediaminetetraacetic acid. Solution was circulated through the bath from reservoirs (37-38 degrees C) through which 95% O²/5% CO sub 2 was bubbled, maintaining pH at 7.4+/-0.5. The muscles were stimulated through a pair of field electrodes along the lateral wall of the bath by 0.5-ms voltage pulses maintained at approximately 110% of threshold intensity, adjusted as necessary to maintain the same stimulus to action potential latency (usually 5-10 ms). The tendinous end of the papillary muscle was attached to a Grass FT03 force transducer (Quincy, MA) while the other end of the muscles was pinned to the bottom of the tissue bath. Muscle length was adjusted to the lowest resting force at which maximum twitch force was obtained (measured in milliNewtons (mN)). The muscles were stimulated at a frequency of 0.5 Hz for 40-60 min during the stabilization period before beginning the experiments. After each experiment, the cross-sectional area of each muscle was calculated from the muscle length, weight, and density (1.04 g/ml) assuming a cylindrical form. Mean cross-sectional areas were 0.93+/-0.07 (mean +/-SEM, n = 31; range 0.29-1.94 mm²) and 1.19+/-0.18 mm²(mean+/-SEM, n = 12; range 0.38-2.38 mm²) in guinea pigs and rats, respectively.

Membrane potential was monitored by a conventional 3-M potassium chloride-filled glass microelectrodes (10-20 M Omega) attached to a VF-1 preamplifier (World Precision Instruments, Sarasota, FL). Force, membrane potential, and its rate of rise during the action potential (dV/dt) were monitored on a digital storage oscilloscope, while force and membrane potential were continuously recorded on a Grass model 79E polygraph. After a 15-min rest, a rested-state contraction was elicited followed by stimulation rates of 0.1, 0.5, 1, 2, and 3 Hz applied sequentially. Each stimulation rate was maintained until an unchanging, steady-state contraction was achieved before proceeding to the next higher rate. After study under control conditions, characteristics were measured after 15 min in the presence of anesthetic. Recovery responses were measured after washout of the inhalational anesthetic for at least 20 min. Because higher stimulation rates (2 and 3 Hz) occasionally caused dislodgment of the recording microelectrode from the muscle, only results from impalements that were maintained throughout the whole sequence (i.e., control, drug exposure, and washout periods) were tabulated. AP amplitude, maximum rate of depolarization (dV/dt^max), and AP duration at 50% and 90% repolarization (APD⁵⁰and APD⁹⁰, respectively) were measured. Effects on contractile function are presented as alteration in dF/dt^max.

Evoked force and action potentials were also studied in muscles in 26 mM Potassium sup + Tyrode solution (Potassium sup + substitution for Sodium sup +), which caused partial depolarization to -40 to -45 mV and inactivated fast sodium channels. Concurrent application of 0.1 micro Meter isoproterenol results in an increase in number of Calcium²+ channels via the beta-adrenergic pathway to permit the measurement of slow action potentials. Under these conditions, contractility, Calcium²+ current, [18]and Calcium²+ uptake into the SR are enhanced. [19,20]The sequential increase in stimulation rate after rest up to 3 Hz was performed as in the normal Tyrode solution. The pattern of force after rest and up to 0.5-1 Hz stimulation consistently demonstrated a distinct late phase of force development [21]; the rate of force development after rest and up to 0.5-1 Hz showed distinct early and late maxima, termed dF^E/dt^maxand dF^L/dt^max, respectively. [14,22]Late peaking force was not present at 2 and 3 Hz; consequently, no value of dF^L/dt^maxwas noted. Slow action potential characteristics, amplitude, dV/dt^max, APD⁵⁰, and APD⁹⁰were determined as in normal Tyrode's solution. Because of the difficulty of maintaining intracellular microelectrode impalements and appropriate stimulus-action potential latencies at 2 and 3 Hz stimulation, accurate measurements for slow action potential dV/dt^maxcould only be observed for stimulation frequencies up to 1 Hz.

When isolated strips of muscle are rapidly cooled, a contracture usually develops because of loss of Calcium²+ from the SR. [23,24]Rapid cooling contractures (RCCs), which have been used a measure of SR Calcium²+ content, [23]were triggered by rapidly changing the temperature of 95% O²/5% CO²perfusing solution from 37 degrees Celsius to 0 degree C (flow rate of approximately 50 ml/min), which resulted in a change in temperature to < 5 degrees C achieved within 2-4 s. [23,24]Both isometric force and bath temperature were recorded continuously during these experiments. After a 15-min rest at 37 degrees Celsius, rapid cooling was induced and maintained for approximately 60 s. After the measurement of rested state RCC, the chamber was reperfused at 37 degrees C. After complete rewarming to 37 degrees C, muscle stimulation was steadily increased from 0.1 Hz to 2 Hz until maximal and stable contractions after 2 Hz were elicited for 30 s. Rapid cooling was induced again after abrupt cessation of 2 Hz stimulation; cooling caused a transient contracture, which peaked after 10-12 s and gradually declined to near the diastolic resting force at 50-60 s. After control measurements, experiments were performed after exposure to sevoflurane for 15 min at 37 degrees C. Recovery responses were also observed after washout for 20 min.

Whole-cell voltage clamp measurements were applied to isolated myocytes to examine the effect of sevoflurane on separated calcium and potassium currents, which play key roles in myocardial excitation-contraction coupling. According to an established protocol, guinea pig ventricular myocytes were isolated enzymatically. [24]English Shorthair guinea pigs (weighing approximately 300 g) were anesthetized with 50 mg/kg sodium pentobarbital according to the guidelines of the University of Virginia Animal Research Committee. The heart was quickly excised and perfused (30 mmHg perfusion pressure) using a Langendorff perfusion system. All solutions were passed through a 0.2-micro meter filter, equilibrated with 95% O²and 5% CO²to pH 7.4 and maintained at 37 degrees C. The heart was perfused for 5 min with solution identical to the modified Tyrode solution, except for the omission of calcium and addition of 5 mM pyruvic acid. The calcium-free buffer was then supplemented with 25 micro Meter CaCl², 0.5 mg/ml albumin, and 0.5 mg/ml type II collagenase (Worthington Biochemical, Freehold, NJ) and continuously recirculated for 20 min, or until the heart appeared flaccid. The ventricles were then cut off and minced with scissors into the enzyme solution contained in a thermostatically controlled funnel. The tissue was agitated by slow bubbling of the solution with 95% O²/5% CO²for 30 min. The resulting slurry was filtered through 200-micro meter nylon mesh and centrifuged for 2 min at 80g. Cells were then washed twice in Krebs-Heinsleit buffer with 200 micro Meter CaCl²and finally resuspended in the buffer with 1 mM CaCl². Isolated cells were stored in a 95% air/5% CO²incubator set at 37 degrees C. Only rod-shaped cells with apparent striations that remained quiescent in solution with 2 mM CaCl²were used for experiments. Before establishing the whole-cell recording configuration, the external bathing solution contained (in mM): 140 Sodium sup +, 5 Potassium sup +, 2 Calcium sup 2+, 1 Magnesium²+, 151 Chlorine sup -, 10 N-2-hydroxyethylpiperazine-N'-2 ethanesulfonic acid (HEPES), adjusted to pH 7.4 with 1 N NaOH. For measurement of Potassium sup + currents, the patch pipette solution contained (in mM): 140 Potassium sup +, 1 Calcium sup 2+, 102 Chlorine sup -, 11 ethylene glycol-bis(beta-aminoethyl ether)-N,N,N,N-tetra-acetic acid, 10 HEPES, 5 Magnesium-adenosine triphosphate, adjusted to pH 7.3 with 1 N HCl. After establishing whole-cell voltage-clamp, 0.5 mM CdCl²was added to the external solution to eliminate any confounding Calcium²+ current. For measurement of inward Calcium²+ currents, the patch pipette solution contained (in mM): 120 Caesium sup +, 20 tetraethylammonium, 1 Calcium²+, 140 Chlorine sup -, 11 ethylene glycol-bis(beta-aminoethyl ether)-N,N,N,N-tetraacetic acid, 10 HEPES, 5 Magnesium-adenosine triphosphate, pH 7.3 with 1 N HCl. Once whole-cell recording was achieved, the bathing solution was exchanged to (in mM): 125 Caesium sup +, 20 tetraethylammonium, 2 Calcium²+, 1 Magnesium²+, 151 Chlorine sup -, 10 HEPES, 0.01 tetrodotoxin, adjusted to pH 7.4 with 1 CsOH.

Isolated myocytes were permitted to settle to the bottom of a recording chamber mounted on an inverted microscope where bathing solutions could be exchanged. Standard whole-cell voltage-clamp methods were employed as described by Hamill et a1. [25]; an interval of 4-6 min was allowed after initiating the whole-cell recording configuration to establish a stable baseline. Voltage-clamp measurements were performed using the Axopatch 200 (Axon Instruments, Foster City, CA) patch clamp amplifier. Patch electrodes were prepared from borosilicate glass model KIMAX-51 (American Scientific, Charlotte, NC) and were typically less than 1 M Omega when filled with internal solution. After the fabrication of pipettes with a two-stage micropipette puller, pipette tips were heat-polished with a micro-forge. All myocyte experiments were conducted at room temperature (20-22 degrees C).

Data acquisition was performed using a pCLAMP system version 5.5.1 (Axon Instruments) coupled with an IBM-compatible, 386-based microcomputer. For measurement of I^Ca, unless otherwise noted, myocytes were voltage-clamped at -80 mV and P/n analysis was used to estimate leakage and capacitative current. [26]By this method, the ensemble of the currents elicited by n = 4 hyperpolarizing subpulses of amplitude P/4 were added to the current evoked by a depolarizing test voltage (V^T) from the holding potential where P = depolarizing test voltage - holding potential. To assess the anomalous rectifying Potassium sup + current (I^K1) and delayed outward Potassium sup + current (I sub K), ramp currents were elicited by sweeping the voltage from -100 to +90 mV at a rate of 21.1 mV/s. Given the profound resting Potassium sup + conductance, P/-4 leakage analysis was not practical. Instead, sevoflurane was applied only to myocytes with resting membrane potentials, determined as the voltage where the net transmembrane current was 0, more negative than -75 mV. Current records were analyzed off-line using a custom program capable of preparing current-voltage (I-V) relationships and nonlinear curve fitting. [27].

For the papillary muscle experiments, the perfusate was equilibrated with sevoflurane by passing the 95% O²/5% CO²gas mixture (flow rate: 1 l/min) through a sevoflurane vaporizer (Sevotec 3, Ohmeda, West Yorkshire, UK) before bubbling the solution. Anesthetic concentrations in the superfusate were confirmed by gas chromatography (error+/- 4%). At 37 degrees C, the perfusate contained 0.21 and 0.43 mM sevoflurane when equilibrated with 1.7% and 3.4% sevoflurane (approximately 0.9 and 1.8 human minimum alveolar concentration [MAC] [28]), respectively. Assuming the published sevoflurane water/gas partition coefficient of 0.37, these aqueous concentrations are approximately 80% of predicted, and are equivalent to 1.4 or 2.8% sevoflurane (approximately 0.75 and 1.5 MAC), which are the values noted in this report. Because the aqueous/gas partition coefficient (P^c) of sevoflurane at room temperature (approximately 22 degrees C) has not been reported, we estimated the aqueous/gas partition coefficient with the use of the gas chromatograph. Test solution of volume V^Lcontaining a constant concentration C^L1of sevoflurane via the vaporizer set at 1.7% was withdrawn into a glass syringe and allowed to equilibrate with a volume V sub G of air. According to the conservation of mass: Equation 1.

Rearranging the terms: Equation 2.

Therefore, 1/C^Gis a dependent variable, which is a linear function of the ratio of the independent variables V^Gand V^Lwith a slope of 1/C^L1and a constant equal to P/C^L1. The plot of 1/C^Gas a function of V^G/V^L(Figure 1) resulted in data points well fitted by a straight line yielding the aqueous/gas partition coefficient = 0.52 and a concentration of 0.35+/-0.04 mM with the sevoflurane vaporizer set at 1.7% (0.68 mM in air) for use in the isolated myocyte experiments. At 0.35 mM in this aqueous concentration of sevoflurane at 22 degrees C is equivalent to approximately 2.3% sevoflurane (approximately 1.2 MAC) at 37 degrees C.

Changes are expressed as percent of the average of the control and recovery values. For papillary muscle experiments, repeated measures of analysis of variance followed by Fisher PLSD multiple range test was applied to test for significant differences among the stimulation rates and among the anesthetic concentrations. For isolated myocyte studies, Students's t test was employed. Unless otherwise noted, all values are expressed as mean+/-SEM. A P value < 0.05 was considered significant.

As shown in Figure 2(A) for contractions at 0.5 Hz, application of sevoflurane (2.8%) reversibly depressed evoked tension development in guinea pig papillary muscle. The concentration-dependent reduction in peak force (PF) and dF/dt^maxwith application of sevoflurane as a function of the stimulation rate in guinea pig papillary muscle is summarized in Figure 2(B and C). While the effect of sevoflurane on PF did not appear to depend on the stimulation frequency, a frequency-dependent effect of sevoflurane was more apparent from measurements of dF/dt^max, of which there was no significant depression at higher stimulation rates (Figure 2(C)). As suggested by the force tracing (Figure 2(A)) the initial rate of force development, reflected by dT/dt^max, was less depressed than later force development. Overall, the depression of PF at rates approaching physiologic (2-3 Hz) by sevoflurane (85% and 70% of control with 1.4% and 2.8% isoflurane, respectively) was similar to that reported for isoflurane: 1.3 and 2.5% isoflurane (approximately 1 and 2 MAC, somewhat greater than the 0.75 and 1.5 MAC sevoflurane employed here) depressed PF to approximately 85% and 60% of control. [14].

In papillary muscles in 26 mM-Potassium sup + Tyrode solution, contractions after rest and up to 1 Hz exhibited a typical late-peaking response (Figure 3(A)). Sevoflurane caused dose-dependent reductions in PF (Figure 3(B)) with marked depression of the late peaking force and dF sub L/dt^max(Figure 3(B)). In contrast, there was no change in the early peaking force or dF^E/dt^maxat stimulation rates from rested-state to 1 Hz. Sevoflurane appears to selectively block the late peaking force component; an effect that becomes negligible at elevated stimulation rates (2-3 Hz) where little or no late peaking force component can be discerned. Recovery of late peaking force and dF^L/dt^maxto approximately 80% of control was observed after a 20-min washout period.

Whereas guinea pig papillary muscle, like human tissue, [29]exhibits an increase in peak tension with higher stimulation rates (positive force-frequency relationship), rat papillary muscle shows a negative force-frequency relationship, which may be related to hypoxia. [30]Because of the potential for such hypoxic change, especially at higher stimulation rates, effects of 2.8% sevoflurane were only considered at 0.1 Hz: twitch force decreased from 7.5+/-2.1 to 6.7 +/-1.6 mN/mm²; while recovery force reached only 6.0 +/-2.0 mN/mm², suggesting a statistically nonsignificant difference.

Rapid Cooling Contracture Experiments. Under control conditions when rapid cooling was preceded by a 15-min rest period, little isometric force was elicited. In contrast, a phasic contracture developed over 3-8 s when rapid cooling followed cessation of a train of contractions at a stimulation rate of 2 Hz (Figure 4). Under control conditions in six muscles studies, RCC force was 8.7+/-1.4 mN/mm², very similar to the forces of the preceding electrically-evoked 2Hz contraction of 9.1+/-1.5 mN/mm²(N = 5); the ratio of RCC force to the PF elicited by a prior stimulus 0.98. Although PF of contractions was decreased to 70% of the control level with sevoflurane (6.4+/-1.3 mN/mm²), similar to the force-frequency experiments in 5 mM Potassium sup + Tyrode solution, the RCC was markedly depressed to 0.03+/- 0.02 mN/mm², resulting in a RCC/PF ratio of 0.04. Although PF was recovered to 98% of the control value (9.0+/-1.4 mN/mm²after a 15-min washout, there was little recovery in the RCC (0.44 mN/mm sup 2 or 5% of control). Further recovery of the RCC to approximately 21% of control was reached after an additional 15-min washout period.

Papillary Muscles. Sevoflurane (2.8%) caused no change in either action potential amplitude or dV/dt^max; however, APD⁵⁰and APD sub 90 were significantly prolonged (Figure 2(A)). Data are presented in Table 1for 0.1 Hz stimulation, but such increases of 15-20% were typically observed at all stimulation rates. No change in the resting membrane potential, which ranged from -87 mV to -92 mV, was observed at either concentration of sevoflurane. In papillary muscles in 26 mM-Potasssium sup + Tyrode solution, sevoflurane (2.8%) also altered neither the slow action potential amplitude nor dV/dt^max, but as observed in 5 mM Potassium sup + solution, the APD⁵⁰and APD⁹⁰were significantly prolonged (Figure 3(A)). For slow action potentials evoked at 0.1 Hz, both APD⁵⁰and APD⁹⁰reversibly increased by 30% (average, n = 6) in the presence of the anesthetic (Table 1), an effect observed at higher rates. No change in the resting membrane potential, which ranged from -39 mV to -44 mV, was observed with application of sevoflurane.

Isolated Myocyte Studies. The increase in the action potential duration observed in intact papillary muscle suggested that sevoflurane may decrease re-polarizing Potassium sup + currents. To examine sevoflurane effects on current flux through the resting Potassium sup + conductance (I^K1) as well as through the delayed Potassium sup + (I sub K) current, a ramp command voltage from -100 to +90 mV was applied to the isolated guinea pig myocytes at room temperature. At potentials negative to -10 mV, inward and outward Potassium sup + currents corresponded to I^K1, while outward currents positive to -10 mV corresponded to I^K. Under control conditions, peak inward I^K1evoked by a hyperpolarization to -100 mV and outward I^K1elicited at potentials from -60 to -50 mV reached -2.12+/-0.36 (n = 12) and 0.334+/-0.046 nA (n = 12), respectively, and were either not significantly reduced or only barely depressed by 0.35 mM sevoflurane (Figure 5(A)). Inward and outward components of I^K1were to 92 +/-5% and 93+/-2% (n = 4) of control after application of the anesthetic. The peak outward I^Kassessed at +90 mV reached 1.06 +/-0.36 nA (n = 4). Although sevoflurane had no significant effect on I^K1, I^Kat +90 mV was significantly reduced to 53+/- 7% (n = 4) of control by the 0.35 mM sevoflurane. Figure 5(B) summarizes the concentration-dependent depression of the ramp Potassium sup + current components.

Voltage-dependent Calcium²+ current was evoked by step depolarizations from -70 mV to +70 mV from a holding potential of -80 mV. Although a low-threshold, dihydropyridine-insensitive (T-type) calcium current (I^CA,T) was observed in response to small amplitude depolarizations to ranging from -60 to -40 mV, ventricular myocyte I^Cais dominated by high-threshold, dihydropyridine-sensitive calcium current (I^Ca,L), which can be observed exclusively at potentials more positive than -10 mV. [31]Under control conditions, peak I^Ca,T, elicited by a test pulse to -40 mV, and peak I^Ca,L, evoked by a step potential to 0 or +10 mV, reached -88+/-6 pA (n = 10) and -595 +/-79 (n = 10), respectively. As shown in Figure 6(A and B), peak I^Ca,T and I^Ca,L were modestly, yet significantly reduced by application of 0.7 mM sevoflurane to 78+/-4% (n = 9) and 75 +/-3% (n = 10) of control, respectively. With the decrease in the peak I^Ca,L, the anesthetic also enhanced the apparent rate of inactivation, as illustrated in Figure 6(C), where the peak I^Ca,L under anesthetic conditions has been normalized to the control peak I^Ca,L. This feature of anesthetic action appears to be shared with both halothane and isoflurane. [32]The voltage for peak I^Ca, 0 or +10 mV, was not affected by the anesthetic. To determine whether or not a shift in the voltage dependence of inactivation can explain, at least in part, the anesthetic effect as suggested for halothane by Baum et al., [33]we investigated the steady-state in-activation profile of I^Ca,L. Myocytes were held at a pre-pulse voltage ranging from -110 mV to +5 mV in 5-mV increments for 5 s and I^Cawas triggered by a test pulse to 0 mV. For each cell, steady-state inactivation was quantified by normalizing the peak inward current, I^Ca,L(V^p), from each prepulse voltage, V^p, to the maximum current value I^Ca,L,max. The resulting curve was well fitted by the Boltzmann function: Equation 1where Vⁿis the prepulse potential where I^Ca,L (V^p)/I^Ca,L,max is 0.5, and kⁿis the slope factor. Under the control conditions, Vⁿand kⁿwere -27.0+/-1.8 mV (n = 7) and -5.7+/-0.3 mV (n = 7), respectively (Figure 6(D)). Sevoflurane (0.7 mM) produced a shift in steady-state inactivation; Vⁿshifted by 4.6+/-0.4 mV (n = 7) toward more hyperpolarized potentials, with no significant change in kⁿ, which was -5.5+/-0.4 mV.

The present study shows that in the guinea pig ventricular muscle, sevoflurane depresses the contractile force in a manner quite similar to seen by Hatakeyama et al. [10]in their preparation of isolated dog ventricular tissues. The effect of sevoflurane is similar to the dose-dependent contractile depression previously reported for isoflurane in isolated ventricular tissues, [14,34-36]with intact animal studies, [8,37]and in clinical settings. [5,6,38]In these studies, sevoflurane appears to have depressant effects similar to isoflurane and desflurane, which are less profound than the effects of halothane and enflurane. The depression in peak tension to approximately 65% of control by 0.43 mM sevoflurane (equivalent to gas phase approximately 1.5 MAC) is similar to the decrease in dP/dt-50 reported in chronically instrumented dogs by 1.5 MAC to 58% of control [37].

Not only the quantitative but also the qualitative effects of sevoflurane are in many ways similar to those previously reported for isoflurane. The rapid tension development observed at higher stimulation rates in guinea pig myocardium, or observed at rest in rat ventricular myocardium, and attributable to release of Calcium²+ from the SR, appears to be modestly altered by sevoflurane, similar to the action of isoflurane. [13,14,39]Furthermore, sevoflurane, like isoflurane and enflurane, [13,14]markedly reduces force development late in the contraction, a feature that was particularly prominent when contractions were elicited in partially depolarized (26 mM Potassium sup +) isoproterenol-stimulated muscle. During prolonged rest, isolated ventricular muscle becomes depleted of its intracellular store of Calcium sup 2+, so rested-state contractions (and those at low stimulation rates) are dependent on the transsarcolemmal influx of Calcium²+ for activation. [40,41]Depression of the late tension development seen with rested-state contractions is probably mediated in part by depression of the inward Calcium²+ current (I^Ca,L), and also by the reduction of late SR release of the Calcium²+, which represents transient accumulation into the SR of the entering Calcium²+. [22,40].

The SR of isolated rat ventricle does not become depleted of Calcium²+ during rest [42,43]so that rested-state contractions in rat myocardium reflect force development mediated by Calcium²+ released from the SR, similar to contractions at 2 or 3 Hz stimulation rates in guinea pig myocardium. While the present experiments may have been complicated by rundown and hypoxia, sevoflurane (2.8%) nevertheless had minimal effect on rested-state force development at low stimulation rates, suggesting that sevoflurane has a modest effect on SR Calcium²+ release. Halothane and possibly enflurane stand in contrast as anesthetic agents that cause a more prominent decrease initial force development, [13,14]probably by activating the SR Calcium²+ release channel and depleting the SR Calcium²+ store. [16,17]Because such studies of isometric contractions in nonhuman tissue clearly differ from the ejecting human heart, extrapolation to clinical settings must be made with care. However, the relative magnitude of depressant actions of anesthetics appears consistent in varied settings, whereas insights into cellular actions may be helpful in understanding effects in ischemic and failing heart.

Another striking similarity between sevoflurane and isoflurane, a quality also shared with enflurane, seems to be its ability to inhibit the release of Calcium²+ from the SR induced by rapid cooling. [44]The RCC is thought to provide an index of the availability of activator Calcium²+ available for release from the SR. [23,45]Based on the evidence of modest depression of contractions at higher stimulation rates in guinea pig, Calcium²+ is clearly present in the SR, yet its release induced by cooling is markedly inhibited by sevoflurane. It is noteworthy that Calcium²+ release pathway from the SR activated by rapid cooling does not appear to be the Calcium²+ release channel. [46]Although the pool of Calcium²+ that generates RCCs and electrically evoked contractions appears to be the same, [47]the present observation is consistent with the view that electrically evoked contractions and RCCs may rely on ryanodine-sensitive and -insensitive pathways for Calcium²+ release from the SR and that sevoflurane preferentially blocks the RCC pathway. It is curious that the fluorinated ether anesthetics isoflurane, enflurane, and sevoflurane are all capable of markedly inhibiting RCCs [44](Lynch, unpublished results), as well as the late component of tension prominent with isoproterenol stimulation, possibly representing a distinct Calcium²+ release pathway [46,47]whose physiologic relevance is unclear.

The whole-cell voltage-clamp studies were carried out at room temperature to enhance the viability of the isolated myocytes during recording. While MAC has clearly been shown to decrease with hypothermia, [48]the intrinsic alterations in behavior of hypothermic animals precludes reliable assignment of equivalent anesthetic concentrations at different temperatures. Nevertheless, our data clearly revealed a reduction in peak I^Ca,L, alteration in inactivation, and shift in steady-state inactivation by sevoflurane, which is similar to that produced by an equivalent concentrations of isoflurane. [32]Consistent with previously published work, [49]we observed no statistical difference between the blockade of T-type and L-type I^Ca; however, the low level of T channel current in guinea pig ventricular myocytes makes it difficult to identify any differential effects. Because the inactivation of I^Ca,L depends on both calcium and voltage, [50]the extent to which the anesthetic-induced alteration in steady-state and time-dependent inactivation can account for the reduction in I^Ca,L is unclear. The relatively modest depression of I^Caby sevoflurane in this study contrasts with recent work by Hatakeyama and colleagues [11]who demonstrated a profound blockade of I^Cain canine ventricular myocytes over a lower range of anesthetic concentrations (both gas and aqueous phases). Although it is possible that canine ventricular Calcium sup 2+ channels might be inherently more sensitive to sevoflurane than those from guinea pig, a more likely explanation involves the anesthetic-induced shift in steady-state inactivation. These previous measurements of I^Cawere performed at a holding potential of -30 mV, [11]a membrane voltage at which many of the Calcium²+ channels are already inactivated. From Figure 6(D) one can estimate the impact of the shift in steady-state inactivation, which would fall from 0.436 to 0.252 at -30 mV, resulting in a 42% decrease in the measured I^Ca. In our microelectrode experiments in papillary muscle, there was no apparent decrease in the slow action potential dV/dt^max, which reflects the net depolarizing current and is employed as an estimate of I^Ca. However, the block of I^Kand/or the outward component of I^K1by sevoflurane may have masked any reduction of I^Caassessed by dV/dt sub max during measurement of slow action potentials. Delayed Potassium sup + current does not appear to deactivate completely on repolarization in response to physiologic stimulation rates, [51]so inhibition of outward current via I^Kby sevoflurane may counteract its inhibition of I^Ca, so that the net depolarizing current responsible for dV/dt shows little alteration.

The major electrophysiologic action of sevoflurane appears to be the suppression of I^K. In fact, the sevoflurane-induced depression of I^Kis similar to that reported for equipotent concentrations of halothane and isoflurane in bullfrog atrial myocytes. [52,53]Delayed Potassium sup + current from guinea pig ventricular myocytes has been previously shown to comprise two components: a slowly activating current I^Ksand a rapidly activating, inwardly rectifying current I^Kr. [54]Under the whole-cell recording conditions employed in the present study, I^Kat +90 mV is dominated by I^Ks. It has recently been suggested that I^Ksis encoded by the min K gene [55,56]and, interestingly, expression of min K in Xenopus oocytes results in a slowly activating Potassium sup + current that has been shown to be sensitive to the inhalational anesthetics. [57]The clinical significance of the anesthetic-induced depression of I^Ksmust be interpreted with caution because there is a lack of species homogeneity relative to the distribution of Potassium sup + channels in the mammalian myocardium. Although I^Ksis expressed in human atrial tissue, [58]there are conflicting reports concerning its presence in human ventricular tissue. [59,60]Future work will determine the relative sensitivity of I^Ks, I^Kr, and the transient outward Potassium sup + current, I^to, to sevoflurane.