INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

SEVOFLURANE IS A NEW INHALATIONAL volatile anesthetic that is rapidly gaining extensive clinical use because of its desirable properties of a low blood-gas partition coefficient and nonpungent character. The cardiovascular side effects are less than those of halothane (22) or enflurane (33) and seem comparable to those exerted by isoflurane (35). Depressing cardiovascular effects by sevoflurane in in vitro studies have been shown in guinea pig, rat, and dog ventricle and in human atrium (2, 10, 16-19, 38). The negative inotropic effect has been attributed to a decrease of transsarcolemmal Ca²⁺ influx, and the effect on SR Ca²⁺ release seems to be modest (2, 17, 38). Studies in skinned (37) and intact (23, 24, 28) cardiac fibers provided ample evidence that halothane, enflurane, and isoflurane decrease myofibrillar Ca²⁺ sensitivity to some extent. There is no definitive information on whether sevoflurane may also decrease myofibrillar Ca²⁺ sensitivity, although recently a depressing effect on cross-bridge cycling kinetics was shown in skinned rat cardiac fibers (40). The aim of this study was to evaluate effects of sevoflurane on contractility and relaxation in isolated ventricular myocardium of the ferret because the ferret myocardium shares unique properties with human ventricular myocardium (8, 9, 46). Furthermore, the effects of sevoflurane were compared with those of decreased extracellular calcium concentration to identify mechanisms of drug action in intact muscle, in particular to determine whether sevoflurane alters myofibrillar Ca²⁺ sensitivity in intact cardiac fibers.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

This study was approved by the Animal Care and Use Committee of the Mayo Foundation, with protocols completed in accordance with National Institutes of Health guidelines and in accordance with the "Guide for the Care and Use of Laboratory Animals" (Institute of Laboratory Animal Resources Commission on Life Sciences, National Research Council). Adult male ferrets (weighing 1,100-1,500 g and 16-19 wk of age) were anesthetized with pentobarbital sodium (100 mg/kg ip). Suitable papillary muscles were carefully excised and mounted vertically in a temperature-controlled (30°C) muscle chamber that contained a physiological salt solution made up in ultrapure water (Nanopure Infinity, Barnstead, Dubuque, IA) and with the following composition (in mM): 137.5 Na⁺, 5.0 K⁺, 2.25 Ca²⁺, 1.0 Mg²⁺, 124.5 Cl, 1.0 SO₄², 20.0 acetate, 10.0 glucose, 5.0 MOPS, pH 7.40, bubbled with 100% O₂. Experiments were carried out at 30°C and at a stimulus frequency of 0.25 Hz; isolated papillary muscle function is stable for many hours in these conditions. Suitable preparations were selected on the basis of previously used criteria (24, 26, 27): the length at which twitch active force is maximal (L_max)  3.5 mm, a mean cross-sectional area  1.2 mm², and a ratio of resting to total force  0.25. Table 1 summarizes the muscle characteristics in control conditions at L_max. The tendinous end of each muscle was tied with a thin, braided polyester thread (size 6.0, Deknatel surgical suture) to the lever of a force-length servo-transducer (26). The ventricular end of each muscle was held in a miniature Lucite clip with a built-in platinum punctate electrode; two platinum wires were arranged longitudinally, one along each side of the muscle, and served as anode during punctate stimulation. Rectangular pulses of 5-ms duration were delivered by a Grass S88D stimulator at a stimulus interval of 4 s. Stimuli at 10-20% above threshold (range 4-12 V) were used to minimize the release of endogenous norepinephrine by the driving stimuli. The muscles were stimulated and made to contract in alternating series of four isometric and four isotonic twitches for 1-2 h. At the end of this stabilization period, muscles had reached steady state, and initial muscle length was set at L_max. Sevoflurane was delivered as previously published (26) for other anesthetics. The concentration of sevoflurane was measured continuously with an anesthetic agent monitor (Ohmeda 5330, Madison, WI). Gas chromatography (Hewlett-Packard 5880A) measurements showed that 1% (vol/vol) sevoflurane corresponded to 0.18 mM in fluid at 30°C. The concentration of sevoflurane in fluid and the calculated partial pressure of sevoflurane in fluid followed closely imposed changes of anesthetic vapor concentration in the gas phase. After the sevoflurane administration was discontinued, its concentration in liquid declined rapidly and was always undetectable at 20 min.

Muscles contracted isotonically at the preload of L_max throughout the experiment. After 12-15 min in each sevoflurane concentration [or extracellular Ca²⁺ concentration ([Ca²⁺]_o), see below], a series of variables of contraction and relaxation were determined from three types of contraction. The first contraction was an isotonic twitch at the preload of L_max from which were measured the maximal amount of shortening (DL), peak velocity of shortening and of lengthening, and corresponding times to peak values measured from the stimulus. The second contraction was a "zero-load-clamp" contraction; that is, an isotonic twitch at the preload of L_max at which load was rapidly (<3 ms) decreased electronically to zero during the latent period (5). Muscles shortened in unloaded conditions, and maximal unloaded velocity of shortening (MUVS) and time to MUVS (TMUVS) were measured. Theoretical and technical considerations of the zero-load-clamp technique to obtain MUVS have been discussed earlier (5). The third contraction was an isometric twitch, from which we measured peak developed force (DF), maximal rate of rise (+dF/dt) and fall (dF/dt) of force, corresponding time to peak values, and time to half isometric relaxation (RTH) measured from the time to peak force (TPF). Each of these three contractions was separated by seven isotonic twitches at the preload of L_max to eliminate effects of loading history of preceding contractions (39). Load-sensitivity of relaxation was determined as in earlier studies (27), from an isometric twitch and six afterloaded isotonic twitches, each with a different afterload. In brief, the ratio of time to initiation of isometric relaxation in afterloaded isotonic twitches relative to time for the isometric twitch to decline to corresponding force levels was plotted against the ratio of force of the afterloaded isotonic twitches relative to peak developed force of the isometric twitch. The slope of this time ratio-vs.-force relationship is a quantitative measure of load sensitivity of relaxation.

To minimize effects of release of endogenous catecholamines, experiments were conducted in the presence of (±)-bupranolol hydrochloride (5 × 10⁷ M), a competitive -blocking agent that is more potent than propranolol and devoid of agonistic effects in heart muscle (31). Waveforms of force, length, velocity, and dF/dt were recorded on a four-channel digital oscilloscope (Nicolet 4094A, Madison, WI) and on a four-channel pen recorder (Honeywell 1400, Minneapolis, MN). All waveforms of interest were transferred to a desktop computer by software programs written in WFBASIC (Blue Feather Software, New Glarus, WI).

Two protocols were used to examine the mechanism of sevoflurane's inotropic effect. Each muscle served as its own control. In group 1 muscles (n = 9), we measured the effects of incremental concentrations of sevoflurane on variables of contraction and relaxation. After 12-15 min in each concentration, steady state was reached, and the test contractions required for analysis of contractility and of relaxation were recorded, after which the concentration of sevoflurane was increased. Sevoflurane was applied in incremental concentrations of 0.7, 1.4, 2.1, 2.7, 3.4, 4.1, and 4.8% (vol/vol). These concentrations correspond to 0.25, 0.50, 0.75, 1.00, 1.25, 1.50, and 1.75 minimum alveolar concentration (MAC) in the ferret. MAC is an anesthetic half-maximal effective dose (ED₅₀) as defined by Eger et al. (12). One MAC is the concentration of anesthetic at which 50% of the animals respond to a standardized supramaximal stimulus with "gross purposeful muscular movements of body or extremities" (12). It is a measure of anesthetic equipotency. Sevoflurane MAC in the ferret was calculated to be 2.7% (vol/vol) from the following data. The MAC values for isoflurane and halothane in the ferret are 1.52% and 1.01% (vol/vol) (36). The ratio of isoflurane MAC to halothane MAC in the ferret is 1.5, similar to that found in human, dog, and horse (41). Sevoflurane MAC values for human, dog, and horse are 2.05, 2.36, and 2.31, respectively (1, 32, 42). The halothane MAC values for human, dog, and horse are reported to be 0.75, 0.86, and 0.88, respectively (41). We calculated the MAC value for sevoflurane in the ferret assuming that the relative potency ratio of sevoflurane to halothane is close to that in human (2.73), dog (2.74), and horse (2.63) as well, which brings us to an estimated MAC of 2.7% (vol/vol) in the ferret as used in this study. After the highest concentration, the vaporizer was turned off, and the reservoir bag was emptied. Muscle recovery was then followed under identical conditions, and variables of contraction and relaxation were recorded at 15 min and 30 min of recovery.

In each muscle of group 2 (n = 8), variables of contraction and relaxation and load sensitivity of relaxation were determined in each step of consecutive cumulative concentration-effect experiments, the first for [Ca²⁺]_o and the second for sevoflurane. The [Ca²⁺]_o-effect protocol was carried out from 0.45 to 2.25 mM in steps of 0.45 mM. After the [Ca²⁺]_o-effect experiment, the bathing solution was changed, and the sevoflurane concentration-effect protocol was started with a new (zero-anesthetic) control. The following incremental sevoflurane concentrations were used: 0.7, 1.4, 2.1, 2.7, 3.4, 4.1, 4.8, and 5.4% (vol/vol). Recording of test contractions for analysis of contraction and relaxation was started after 12-15 min, when twitch height in a particular [Ca²⁺]_o or sevoflurane concentration was stable.

This study was carried out in identical conditions as those previously published (24, 26, 27), except for minor differences in the physiological salt solution used. The current study used the organic pH buffer MOPS, a convenient and reliable alternative to the more traditional bicarbonate-CO₂ buffer. MOPS buffer is widely used in muscle tissue studies because it does not require bubbling with an O₂-CO₂ mixture, and MOPS does not affect smooth muscle contractility (30). MOPS does not affect cardiac muscle contractility in the conditions of our experiments, because 1) peak developed force and other measures of contractility in control conditions are similar to those found in previous studies in which we used a bicarbonate-CO₂ buffer system, and 2) muscles maintained their contractile performance for many hours in these conditions.

Statistical analysis. The relationship between contractile variables and sevoflurane concentration was subjected to least squares linear regression analysis (11) for each individual muscle. This allowed for subsequent calculation of ED₅₀, the sevoflurane concentration required to decrease the amplitude of contractile variables (DF, DL, MUVS) by 50%. Concentration-effect relationships between sevoflurane concentration and variables of contractility and relaxation were tested for differences with repeated-measures ANOVA followed by Dunnett's test for pairwise comparison with control.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Table 2 summarizes the absolute values of variables of contraction and relaxation at the onset of the experiments of group 1 (n = 9) and group 2 (n = 8) muscles.

Effects of sevoflurane on contraction and relaxation. Figures 1 and 2 illustrate the key observations relating to the concentration-effect experiments to sevoflurane (n = 9). Sevoflurane caused a reversible concentration-dependent decrease in DF, DL, and MUVS (Fig. 1, A and B). Sevoflurane abbreviated TPF and RTH (Fig. 1C). Sevoflurane slightly decreased the duration of isotonic shortening (TDL) (Fig. 1D). Sevoflurane slightly increased TMUVS of zero-load clamped twitches only at two concentrations (Fig. 1B). Sevoflurane decreased the maximal +dF/dt and dF/dt in isometric twitches (Fig. 2A), and peak velocities of shortening and of lengthening in isotonic twitches at the preload of L_max, all in a concentration-dependent fashion (Fig. 2B). Except for those listed below, variables of contraction and relaxation returned to control values 15 and 30 min after sevoflurane was discontinued. The variables dF/dt, +dF/dt, and V had recovered to values above control (V in 7 out of 9 muscles, dF/dt in 8 out of 9 muscles, and +dF/dt in all muscles). Relaxation of ferret papillary muscle is sensitive to load during control conditions, and sevoflurane had no significant effect on load sensitivity of relaxation. Table 3 lists the ED₅₀ values of sevoflurane for several variables calculated from least squares linear regression. The ED₅₀ value for MUVS is larger than that for DF and DL (P < 0.001). The ED₅₀ value (DF) for sevoflurane was statistically significantly larger than that for halothane (P < 0.001), enflurane (P < 0.001), and isoflurane (P < 0.02) obtained in identical experimental conditions (26).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Effects of sevoflurane on variables of contractility. A: effects of sevoflurane on peak developed force of isometric twitches (DF) and peak isotonic shortening (DL) of isotonic preloaded twitches. B: effects of sevoflurane on maximal unloaded velocity of shortening (MUVS) and on time to MUVS (TMUVS) of zero-load-clamped twitches. C: effects of sevoflurane on time to peak force (TPF) and on time to half isometric relaxation (RTH) of isometric twitches. D: effects of sevoflurane on time to peak shortening of isotonic preloaded twitches (TDL). All data are means ± SD (n = 9). * P < 0.05; # P < 0.01; repeated-measures ANOVA followed by Dunnett's test for pairwise comparison with corresponding control.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Effects of sevoflurane on maximal rates of force development (+dF/dt) and force decline (dF/dt; A) and on peak isotonic velocity of shortening (+V) and of lengthening (V; B). All data are means ± SD (n = 9). * P < 0.05; # P < 0.01; one-way repeated-measures ANOVA followed by Dunnett's test for pairwise comparison with corresponding control.

Comparison of effects of sevoflurane with those of decreased [Ca²+]_o. Figure 3 illustrates the concentration-effect relationships between variables of contraction (DF, DL, and MUVS) and [Ca²⁺]_o (Fig. 3A) and sevoflurane concentration (Fig. 3B). Sevoflurane is most depressant on DF, less on DL, and least on MUVS. Almost the same order of sensitivity is found when these variables are measured in [Ca²⁺]_o-effect experiments in the same muscles: DF is decreased most, and MUVS and DL are decreased to the same extent in lower [Ca²⁺]_o. Changes in [Ca²⁺]_o or in sevoflurane concentration had effects on DF, DL, and MUVS that were statistically different among these variables. Force development changed more than did DL and MUVS with changes either of [Ca²⁺]_o or of sevoflurane concentration. Differences between DL and MUVS were statistically significant for changes in sevoflurane concentration but not for changes in [Ca²⁺]_o.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Concentration-effect graphs (means ± SE) of DF, DL, and MUVS as a function of [Ca2+]o (A) and of sevoflurane concentration (B) in 2.25 mM [Ca2+]o. Results (means ± SE) are from the same muscles (n = 8). Slopes of linear relationship between [Ca2+]o or sevoflurane concentration and DF, DL, and MUVS, respectively, were tested for significant differences with one-way repeated-measures ANOVA. Pairwise comparison between DF, DL, and MUVS was carried out with Student's t-test with Bonferroni correction. ***P < 0.001 (DF vs. DL); +++P < 0.001 (DF vs. MUVS); ###P < 0.001 (DL vs. MUVS).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   A: dependence of TPF and of RTH on peak DF in isometric twitches in two different conditions: changes in [Ca2+]o from 0.45 to 2.25 mM in 0.45 mM increments and of sevoflurane concentration in 0.6 or 0.7% increments in 2.25 mM [Ca2+]o. Values (means ± SE) in the graph are from the same muscles (n = 8). Levels of significance (P) above or below data points reflect differences of slopes from zero for each of the four data groups (one-sample t-tests); NS, not significant. Slopes of the relationship of TPF vs. DF and of RTH vs. DF in sevoflurane were different from those in different [Ca2+]o (Student's paired t-tests). B: representative single-muscle example of specific effects of sevoflurane on isometric relaxation. Two isometric twitches were superimposed, one in sevoflurane 4.9% (vol/vol) in 2.25 mM [Ca2+]o, and one in 0.90 mM [Ca2+]o. From the same contraction amplitude, the muscle relaxed faster in the presence of sevoflurane than in low [Ca2+]o. Muscle characteristics: length at maximal twitch force (Lmax), 5 mm; mean cross-sectional area, 0.42 mm2, ratio of resting to total force at Lmax, 0.24.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Dependence of time to peak shortening (A) and of maximal velocity of isotonic relaxation (B) during preloaded isotonic twitches on contraction amplitude (peak shortening) at various [Ca2+]o (0.45-2.25 mM in 0.45 mM increments) and at various sevoflurane concentrations [0-5.4% (vol/vol) in 0.6 or 0.7% increments] in 2.25 mM [Ca2+]o. Values (means ± SE) are from the same muscles (n = 8). Levels of significance (P) above or below data points reflect differences of slopes from zero for each of the four data groups (one-sample t-tests); NS, not significant. Slopes of the relationship of TDL vs. DL and of V vs. DL in sevoflurane were different. C: representative single-muscle example of specific effects of sevoflurane on relaxation in preloaded isotonic twitches. From a same-peak shortening amplitude, muscle relaxed sooner and faster in the presence of sevoflurane (in 2.25 mM [Ca2+]o) than in low [Ca2+]o. Muscle characteristics: length at Lmax, 6.43 mm; mean cross-sectional area, 0.48 mm2; ratio of resting to total force at Lmax, 0.22.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Sevoflurane exerts a reversible concentration-dependent negative inotropic effect that is less than that of halothane, enflurane, and isoflurane. The negative inotropic effect of sevoflurane cannot be entirely explained by decreased intracellular Ca²⁺ availability. This is the first study that suggests that sevoflurane decreases myofibrillar Ca²⁺ sensitivity in intact ventricular muscle. ED₅₀ values (in MAC) for sevoflurane for DF, DL, and MUVS are higher (Table 3) than those for halothane (0.85 ± 0.03, 0.94 ± 0.11, and 1.13 ± 0.10), enflurane (0.86 ± 0.12, 0.96 ± 0.19, and 1.34 ± 0.28) and isoflurane (1.32 ± 0.33, 1.44 ± 0.25, and 1.97 ± 0.66) obtained in identical conditions (26). This indicates that the negative inotropic effect of sevoflurane is less than that of halothane, enflurane, or isoflurane. These findings are in contrast to results of in vitro experimental studies (10, 15, 17, 29, 38) that reported that the myocardial depressant effect of sevoflurane is similar or even slightly greater than that of isoflurane. Yet, in the isolated working rat heart (43) and in isolated canine ventricular muscle strips (18), sevoflurane depresses myocardial function less than does isoflurane. The negative inotropic effect of sevoflurane is more pronounced in heavily loaded (isometric) than in lightly loaded (zero-load-clamped) contractions (Table 3). This observation was extended and confirmed in group 2, muscles in which we compared the slopes of sevoflurane concentration-effect curves with the low-[Ca²⁺]_o-effect curves. The slopes of the [Ca²⁺]_o-effect curves differed in the same order (DF > DL = MUVS) as the slopes in the sevoflurane effect curves DF > DL > MUVS. On the basis of this analysis, the negative inotropic effect of sevoflurane can be accounted for by a decrease of intracellular calcium availability, which is in agreement with observations that sevoflurane decreases transsarcolemmal Ca²⁺ influx (2, 18-20, 29, 38). In an effort to discern subtle differences between sevoflurane's effect in heavily loaded vs. lightly loaded contractions, the anesthetic potency ratios W = slope (sevoflurane)/slope (log[Ca²⁺]_o) for DF, DL, and MUVS were calculated. The absolute values for the anesthetic potency ratio W were lower for MUVS than for DF or DL. The fact that MUVS is less sensitive to sevoflurane than is DF suggests that sevoflurane might also decrease Ca²⁺ sensitivity. The negative inotropic effect of sevoflurane is more pronounced in isometric conditions when the native myofibrillar Ca²⁺ sensitivity is high (21, 25), whereas in unloaded contractions the native myofibrillar sensitivity is low. Sevoflurane's potency ratios did not differ from those of isoflurane but differed from those of halothane and enflurane. Therefore, the relative contribution of a decreased myofibrillar Ca²⁺ sensitivity to the negative inotropic effect of sevoflurane may be similar to that of isoflurane.

Sevoflurane abbreviated both TPF and RTH. The acceleration of isometric relaxation is not necessarily a consequence of the concomitant decrease in peak force, because RTH is unchanged in control conditions over the range of extracellular concentrations 0.45-2.25 mM. Starting from the same peak force either in low [Ca²⁺]_o or in sevoflurane, force declined faster in the presence of sevoflurane. There is strong evidence that isometric relaxation in cardiac muscle is controlled by the contractile proteins themselves, whereas cell relengthening rate in isolated myocytes (similar to isotonic lengthening in this study) is limited by the rate of decrease of the intracellular Ca²⁺ transient (3, 44). The rate-limiting step for isometric relaxation is therefore in the kinetics of contractile proteins. A faster isometric relaxation, as seen with sevoflurane, would then most likely result from 1) a decreased affinity in troponin C for Ca²⁺ in its low-affinity Ca²⁺-specific site; 2) an effect on the thin filament troponin-tropomyosin complex that results in a decreased gain of force per Ca²⁺ bound to troponin; 3) an increase in cross-bridge turnover, resulting in a shorter average cross-bridge cycle duration; or any combination of these effects. A recent study compared mechanisms of sevoflurane and halothane on cross-bridge cycling kinetics in neonatal and adult skinned rat cardiac fibers (40). The rate of force redevelopment after a release-stretch cycle was decreased by sevoflurane. When interpreted in a two-state cross-bridge model, this finding suggests a decrease in the cross-bridge apparent attachment rate with no change in the cross-bridge detachment rate (40). This would keep cross bridges attached in the force-generating state for a shorter period of time and shorten the average cross-bridge lifetime, and fewer cross-bridges would be attached at any given time. Consequently, these changes will become manifest as a faster isometric relaxation in intact fibers as observed in this study. Currently it is not known whether sevoflurane has additional effects on further mechanisms listed above. Preliminary data (4) on the intracellular Ca²⁺ transient in intact muscle fibers support the view that sevoflurane decreases myofibrillar Ca²⁺ sensitivity.

Sevoflurane decreased maximal velocity of lengthening (relaxation) (V) in a concentration-dependent and reversible manner. Sevoflurane caused isotonic lengthening to proceed faster than the amplitude-matched low-[Ca²⁺]_o control so that the maximal velocity of lengthening in sevoflurane-exposed twitches was higher than in low [Ca²⁺]_o. When [Ca²⁺]_o is decreased, maximal isotonic relaxation velocity and load sensitivity of relaxation are decreased (45). This results predominantly from delayed isotonic lengthening and slowed Ca²⁺ uptake by the sarcoplasmic reticulum (SR) in low [Ca²⁺]_o (34). If the intracellular free [Ca²⁺] is the same either in low [Ca²⁺]_o or in normal [Ca²⁺]_o plus anesthetic, then Ca²⁺ uptake rate by the SR is enhanced in the presence of anesthetic (7), a drug-specific effect that partially overrides the decreased SR uptake rate at low intracellular [Ca²⁺]. On the other hand, if intracellular free [Ca²⁺] is higher in control [Ca²⁺]_o plus anesthetic than in low [Ca²⁺]_o alone, either a greater stimulation of the SR Ca²⁺ uptake in the former condition and/or an anesthetic-induced decrease in myofibrillar Ca²⁺ sensitivity will cause the muscle to relax faster. It is therefore possible that sevoflurane stimulates SR Ca²⁺ uptake to some extent. Several investigations show that there is either no or a modest depressing effect on SR Ca²⁺ release contributing to the overall negative inotropic effect exerted by sevoflurane (2, 17, 38). Yet there is evidence that sevoflurane might increase SR Ca²⁺ content. In rat ventricular myocytes, partial recovery of contraction during isoflurane and sevoflurane application and greater contractions immediately after washout, both more pronounced in sevoflurane, were completely abolished by pretreatment with ryanodine (10). An increased SR Ca²⁺ content would be consistent with a stimulation of SR Ca²⁺ uptake reflected by the greater V in sevoflurane than in the low [Ca²⁺]_o control. We also observed a recovery significantly above control values of the variables +dF/dt, dF/dt, and V immediately after washout of sevoflurane, a phenomenon not seen in isoflurane, enflurane, and halothane (26). DF and DL showed a tendency to increase above control values after washout, yet the increase did not reach statistical significance. We did not follow the recovery over a period longer than 30 min and therefore do not know whether the values would return to control values, but one might speculate that our observations reflect in part the same mechanism as that observed in the investigation of Davies et al. (10).

Load sensitivity of relaxation is often used in cardiac muscle mechanics to characterize physiological or pharmacological effects on relaxation. Load sensitivity of relaxation is affected by [Ca²⁺]_o (14), osmolality (13), species (6), hypoxia (45), and volatile anesthetics (27) such as halothane, enflurane, and isoflurane. Load sensitivity changes as a result of changes of the time course of the isotonic twitch in reference to the isometric twitch in identical experimental conditions. With halothane and enflurane, and to a lesser extent with isoflurane, isometric relaxation was abbreviated more than was isotonic lengthening, and a decrease in load sensitivity of relaxation resulted. With sevoflurane, isometric and isotonic relaxation are abbreviated to approximately the same extent, and this could well explain the lack of changes of load sensitivity of relaxation and the observation that load sensitivity was unchanged in some muscles, increased in a few muscles, and decreased slightly in other muscles.

In summary, sevoflurane exerts concentration-dependent, reversible negative inotropic effects on ferret ventricular myocardium. These depressant effects are less than those seen in equianesthetic concentrations of isoflurane, enflurane, or halothane. The negative inotropic effects result from a combination of a decrease in intracellular Ca²⁺ availability and a decrease in myofibrillar Ca²⁺ sensitivity. The precise locus of action for either mechanism will require further study. Sevoflurane might also slightly increase the rate of Ca²⁺ removal from the cytoplasm as reflected by a faster and earlier isotonic relaxation when compared with amplitude-matched twitches in low [Ca²⁺]_o.