| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Whether propofol contributes a direct negative inotropic effect is controversial. Our principal aim in this study was to determine whether negative inotropic effects of propofol occur at clinically relevant concentrations. We constructed the concentration-response relationship for the negative inotropic effects on intact, isolated, stimulated rat ventricular myocytes. Contraction was measured as cell shortening by using an optical system. Propofol was applied as dilutions of the commercial preparation in physiological saline solution. The drug vehicle had a minimal effect on myocyte contractility. Propofol produced a concentration-dependent reduction in evoked contraction at concentrations greater than 5 µM. The maximum effect was observed at >100 µM, with the K0.5 calculated to be 34.5 µM (95% CI, 21.854.7 µM). In further experiments, we investigated the relationship between changes in contractility and changes in Ca2+ transient (measured by using fura-2 fluorescence) after the application of propofol. By using the shift in the relationship of the cell length to fura-2 fluorescence ratio in the relaxation phase of a contraction as an index of Ca2+ response of the myofilaments, we demonstrated that some of the negative inotropic effect of propofol may be caused by a reduction in myofilament Ca2+ sensitivity. We confirmed this by comparing the reduction in contractility in the presence of propofol with that caused by reducing the extracellular Ca2+ concentration. We observed that, for a decrease in the fura-2 fluorescence ratio of 21%, propofol caused a 12% (95% CI, 2% to 22%) greater reduction in contractility than predicted from reducing the extracellular Ca2+ concentration. However, the K0.5 for the negative inotropic effect of propofol we observed is more than 80 times the 50% effective concentration value for anesthesia. The potential relevance of these findings for clinical use of propofol in humans is discussed.
Implications: By using intact, isolated rat heart ventricle cells, we investigated the mechanisms and concentration dependence of the depressant effect of propofol on contractility of the heart. We conclude that direct effects of propofol on the heart are unlikely to be of significance at the clinical dosage usually given.
Cardiovascular depression after the administration of propofol has been attributed to relaxation of peripheral vascular smooth muscle (1) and a direct negative inotropic effect on the heart (2). The relative contribution of these two proposed mechanisms is disputed, with the greatest area of contention being the clinical relevance of the described effects of propofol on the myocardium (3). The situation is confused by the shortcomings of the various experimental preparations used. The use of a different, and limited, range of concentrations of propofol by various workers, compounds the problem. A negative inotropic effect of propofol will be due either to a reduced Ca2+ transient or a decrease in the sensitivity of the myofilaments to Ca2+. The Ca2+ transient is dependent on the membrane Ca2+ current (ICa) and release of Ca2+ from the sarcoplasmic reticulum (4,5). Our aim in this study was to describe the concentration-response relationship for the negative inotropic effect of propofol on intact, stimulated rat ventricular myocytes and to determine whether reduced contractile responses were associated with a reduction in the Ca2+ transient. We wished to ascertain to what degree the observed effects occurred at concentrations of propofol relevant to clinical practice.
Rats (weighing 300350 g) of either sex were killed (under UK Home Office license) by a blow to the head and subsequent cervical dislocation. The heart was rapidly excised and retrograde perfused with a series of solutions based on a nominally Ca2+ free isolation solution (see below for composition). The heart was first perfused with the isolation solution supplemented with 750 µM CaCl2 for several minutes to clear it of blood and then, perfused with the nominally Ca2+ free isolation solution plus 100 µM EGTA for 4 min. The heart was next perfused with the isolation solution supplemented with 1 mg/mL collagenase (Type 1, Worthington Biochemical, Lakewood, NJ), 0.1 mg/mL protease (Type XIV; Sigma Chemical, St. Louis, MO) and 80 µM CaCl2 for 10 min, after which the ventricles were removed, finely chopped, and agitated gently in enzyme solution (supplemented with 1% bovine serum albumin) for 5-min intervals. Dissociated cells were harvested at the end of each 5-min digestion, and the remaining tissue subjected to further enzyme treatment. The isolation solution was composed of (in mM): 130 NaCl; 5.4 KCl; 1.4 MgCl2; 0.4 NaH2PO4; 5 HEPES; 10.0 glucose; 20.0 taurine; 10.0 creatine; and 7.1 pH at 37°C (NaOH). After isolation, cells were perfused with a physiological saline solution composed of (in mM): 140 NaCl; 5.4 KCl; 1.4 MgCl2; 0.4 Na2HPO4; 5 HEPES; 10 glucose; 1 CaCl2; and 7.4 pH at 30°C (NaOH). Solutions were delivered to the experimental chamber by magnetic drive gear pumps (Micropump, Concord, CA) and solution level and temperature maintained by feedback circuits (6). The clinically available preparation of propofol was used (Zeneca Pharma, Wilmslow, UK). In this, propofol was supplied prepared in a 10% soybean emulsion. This solution was diluted with the physiological saline solution to achieve the desired final concentration of propofol. Cells were transferred to a small tissue chamber (0.1 mL volume) attached to the stage of an inverted microscope equipped for epifluorescence (Cairn Research, Faversham, Kent, UK). The cells were allowed to settle for several minutes onto the glass bottom of the chamber before being superfused at a rate of approximately 3 mL/min with the physiological saline solution. Cells were maintained at 30°C and stimulated to contract by field stimulation at a frequency of 1 Hz. Cell length was recorded by using an optical system based on a photodiode array (7) and displayed on a chart recorder (2600S; Gould, Oxnard, CA). A sample and hold circuit (8) was used to display the shortening of a cell during each contraction (twitch shortening) on the chart recorder. This has the effect of excluding changes in resting cell length, although the time course of the twitch is recorded faithfully. Cell length, and twitch shortening were recorded by using a pulse code modulator (DR-890 Neuro-Corder; Neuro Data Instruments, New York, NY) coupled to a standard VHS video recorder. We studied the effect of 5150 µM propofol on contractility by applying propofol for 60 s at each concentration. The responses to a maximum of three concentrations of propofol were recorded from individual cells. In control experiments, we investigated the effects of the drug vehicle, soybean emulsion, at concentrations equivalent to those present in the propofol experiments, on the contraction of isolated myocytes. In some experiments, freshly dissociated cells were loaded with the acetoxy methyl (AM) ester form of fura-2 (Molecular Probes, Eugene, OR) to measure changes in intracellular Ca2+. Preliminary experiments demonstrated that neither propofol nor its vehicle influenced the fura-2 fluorescence signal. To load cells with fura-2-AM, 50 µg of dye was dissolved in 50 µL of dimethyl sulfoxide (DMSO) yielding a 1-mM stock solution. The amount of 6.25 µL of this stock solution was added to 2 mL of cells (final fura-2-AM concentration, 3 µM) and agitated gently at room temperature for 10 min. Once dye loading was complete, the cell suspension was centrifuged at 30g for 30 s and the pellet resuspended in the physiologic saline solution. The cells were left for 30 min before use to allow deesterification of the dyes. For fluorescence measurements, a 150-W xenon lamp (Ealing Electro-optics, Watford, UK) was used as the source of ultraviolet light. A wheel containing six band-pass filters (Cairn Research) spinning at 100 revolutions per second provided alternate beams of 340- and 380-nm light. The emitted fluorescence was collected by the objective lens, filtered at 510 nm and detected by a photomultiplier tube. The 510-nm fluorescence emitted during 340-nm illumination was electronically divided by the fluorescence emitted during 380- nm illumination to give a fluorescence ratio, which has an approximately linear relationship to Ca2+ concentration in the physiological range. Fluorescence measurements were recorded by using the pulse code modulator and video recorder as previously mentioned. Results were presented as mean ± SEM unless otherwise indicated. Statistical comparisons of data were with paired or unpaired Students t-tests, or nonparametric equivalents if a preliminary normality test was failed. A Bonferroni correction was applied when multiple comparisons were made. A P value < 0.05 was considered significant. A dose-response curve was fitted (with 95% confidence limits) to the contractility data using the equation: y = 100. [propofol]n/([propofol]n + K0.5), where y is the percentage decrease in contraction, n is the Hill coefficient, and K0.5 is the propofol concentration causing a 50% decrease in contraction.
Propofol produced a concentration-dependent reduction in evoked contraction. The effect was observed at propofol concentrations >5 µM and reached a maximum at >100 µM propofol. The time course and pattern of the effect are shown in Figure 1. From these traces it can be seen that the negative inotropic effect developed gradually over the first 30 s after application. Once the depression of contraction had reached its peak level, the negative inotropic effect was maintained during application of propofol. When propofol was washed from the cell, contraction returned to control values. The effect of the drug vehicle (Intralipid) was studied at two dilutions equivalent to those present when the propofol concentration was 34.5 µM and 150 µM. In the presence of the larger concentration of Intralipid, contractions were 97 ± 2% (n = 7, P > 0.05) with no depression seen at the smaller concentration.
The dose-response relationship for the negative inotropic effect of propofol is summarized in Figure 2 in which the data are fitted with a typical dose-response curve. The K0.5 for this effect of propofol was calculated as 34.5 µM (95% CI, 21.854.7 µM) and the Hill coefficient calculated as 1.7.
In seven cells, Ca2+ transients and contraction were measured simultaneously during application of propofol at concentrations of 34.5 µM and 100 µM. In Figure 3 representative recordings are shown illustrating changes in contractility occurring in parallel with a reduction in the Ca2+ transient after application of propofol. The mean decrease in contraction was 45.5 ± 7.6% for 34.5 µM and 60.4 ± 5.5% for 100 µM propofol (P < 0.001 for each concentration versus control; P < 0.01 for difference between the two concentrations; paired Students t-test), whereas the corresponding decrease in fura-2 fluorescence ratio was 18.0 ± 5.3% for 34.5 µM and 30.0 ± 4.5% for 100 µM (P < 0.05 for each concentration versus control and for the difference between concentrations; Wilcoxons signed rank test). The time course and pattern of changes in the Ca2+ transient mirrored those in contraction. However, the reduction in Ca2+ transient was proportionately less than the reduction in contraction. This may indicate that only part of the negative inotropic effect of propofol is caused by a reduction in Ca2+ transient, with possibly some contribution from a reduction in the sensitivity of the myofilaments to Ca2+.
To obtain an index of the Ca2+ response of the myofilaments, we plotted cell length against fura-2 fluorescence ratio during the relaxation phase of a contraction (9). Figure 4A illustrates such plots from contractions recorded in the absence and presence of propofol. Figure 4B shows data during the final phase of relaxation on expanded scales. The solid lines are the result of linear regression of these data. In this cell and six others analyzed in a similar way, the relationship between the fura-2 fluorescence ratio and cell length in the presence of propofol is shifted down. The slope of the regression line under control conditions was -12.9 ± 3.4 µm/fura-2 fluorescent unit (µm/Fr) but was significantly reduced by 34.5 µM propofol (-8.6 ± 2.3 µm/Fr, P = 0.022, paired Students t-test). In the presence of 100 µM propofol the mean slope was -7.1 ± 1.8 µm/Fr compared with -13.4 ± 3.5 µm/Fr (P = 0.017, paired Students t-test).
In an additional seven cells we compared the effects of propofol on the relationship between fura-2 fluorescence ratio and contractility to the effect of reducing the extracellular Ca2+ concentration (to 0.5 mM and 0.8 mM). The results from these experiments are presented in Figure 5. These experiments demonstrate that for a given reduction in fura-2 fluorescence ratio, propofol reduced contractility more than reducing the extracellular Ca2+ concentration. The data support an action of propofol on myofilament calcium sensitivity.
Our results are notable because they define the dose-response relationship for the direct negative inotropic effect of propofol on the myocardium. Furthermore, we provide the first demonstration that the negative inotropic effect of propofol must be, at least in part, the result of a decrease in the myocardial Ca2+ transient cf. Li et al. (10) and Cook and Housmans (11) who reported an effect of propofol on [Ca2+]i in myocardial cell suspensions but could not correlate these changes to inotropic effects). Although there is good evidence that a reduction in ICa plays a significant role in the negative inotropic effect of propofol (1216), the results reported here (Fig. 3) suggest that propofol may also affect the sensitivity of the myofilaments to Ca2+. We examined this possibility initially by using the approach of Spurgeon et al. (9) who suggested that during the relaxation phase of a contraction, the myofilaments come into quasi-equilibrium with the intracellular Ca2+ concentration. Based on this premise, we plotted cell length against the fura-2 fluorescence ratio for contractions recorded in the absence and presence of propofol (Fig. 4) and found the relationship in the final phase of relaxation to be shifted significantly in the presence of propofol compared to control. These results suggest that in the presence of propofol there is a decrease in myofilament Ca2+ response in intact stimulated ventricular cells. We confirmed this by demonstrating that, for a given reduction in fura-2 fluorescence ratio, propofol reduced contractility more than simply reducing the extracellular Ca2+ concentration (Fig. 5). This effect will contribute to the negative inotropic action of propofol. These results contrast with the conclusions of Kanaya et al. (17) who reported that when extracellular Ca2+ was increased to 14 mM, contractions in the presence of 100 µM propofol were significantly enhanced compared to control, suggesting propofol increased myofilament Ca2+ sensitivity. However, other data from the same paper (17) illustrated that propofol (301000 µM) decreased contraction to a proportionately greater extent than the cytosolic Ca2+ transient. Furthermore, Kanaya et al. (17) reported that 30 µM propofol reduced contractility to approximately 85% of control, but had no significant effect on the magnitude of the Ca2+ transient, consistent with propofol leading to a decrease in myofilament Ca2+ sensitivity (data similar to our own findings). Our study, in which changes in cytosolic Ca2+ and contractility in individual cells are used to determine changes in myofilament Ca2+ sensitivity, suggests that propofol leads to a modest decrease in myofilament Ca2+ sensitivity that will contribute to its negative inotropic effect. Single ventricular myocytes are an ideal preparation for investigating the mechanisms involved in the inotropic action of drugs because there are few problems associated with delays in drug diffusion which can occur in multicellular preparations. Although the single cells are unloaded, Lee and Allen (18) demonstrated that the effects of inotropic interventions were independent of cell loading. However, our experiments were performed at 30°C on rat ventricular myocytes stimulated at 1 Hz to maintain cell viability and minimize leakage of the fluorescent dye. These factors, along with species differences in action potential configuration and Ca2+ handling, make it possible that quantitatively different results may be found in other species. Evidence for such differences will, therefore, be examined. From the results of the earliest clinical studies on the myocardial effects of propofol, reviewed by Sebel and Lowden (3), there has been confusion over whether the decrease in blood pressure and cardiac output observed during propofol administration is entirely caused by peripheral vascular smooth muscle relaxation, leading to a reduction in systemic vascular resistance and preload, or whether there is an additional direct negative inotropic effect. Subsequent studies have reported a reduction in myocardial contractility (19,20); however, these studies cannot distinguish between direct and secondary effects on the myocardium. Different groups have reported the effects of propofol on isolated, perfused heart preparations of the rabbit, guinea pig, and rat. In the rabbit (21) and rat (22), it was concluded that propofol had no effect of likely clinical importance, whereas Stowe et al. (23), using isolated guinea pig hearts, concluded the opposite. Contradictory conclusions were also drawn from studies using anesthetized open-chest dog models by Belo et al. (24) who found no effect of propofol and by Puttick et al. (25) and Pagel and Warltier (26) who did report an effect. Contractility, ICa, and intracellular Ca2+ concentration changes with propofol were previously reported in the literature. Puttick and Terrar (12) and Hebbar et al. (27) have shown a reduction in contractility in isolated myocytes from guinea pigs and pigs, respectively. Puttick and Terrar (12) demonstrated a concomitant reduction in ICa, whereas Hebbar et al. (27) showed no difference between the effects of propofol on myocytes from normal pigs and those with heart failure. Zhou et al. (13) found that propofol reduced contractility in isolated papillary muscles and, as with Puttick and Terrar (12), found this to be associated with a reduced ICa. All three of these groups used concentrations of propofol within the range we have used in the current study. Azuma et al. (28) used papillary muscle preparations to compare the effects of propofol (and thiamylal) on rat and guinea pig hearts. They demonstrated no effect of propofol on contractility of rat muscle at concentrations up to 600 µM. Propofol depressed contractility of guinea pig muscle with an IC50 of 200 µM. Azuma et al. (28) attributed the difference between the two species to the different contribution of ICa to the Ca2+ transient between rat and guinea pig. These results conflict with those of several other groups. Puttick and Terrar (12) studied the effect of propofol on guinea pig ICa and found the IC50 to be approximately 100 µM, whereas the IC50 for depression of contractility was similar. Stowe et al. (23) studied the effect of contractility on perfused guinea pig hearts and found 50% depression at a similar concentration (91 µM). We find it difficult to reconcile the apparent lack of potency found by Azuma et al. (28), who used a preparation considered intermediate to those of Puttick and Terrar (12) and Stowe et al. (23), with the findings of the latter two groups. Other workers have studied the effect of propofol on the contractility of papillary muscle preparations. Cook and Housmans (11) demonstrated a reduction in contractility at concentrations of propofol >30 µM. Riou et al. (29), by using rat papillary muscle, found no effect on contractility of propofol up to approximately 60 µM. However, we believe this latter finding can be explained by the small concentration of extracellular Ca2+ (0.5 mM) used by this group. This contention is supported by the findings of Zhou et al. (13) who demonstrated a reduction in both ICa in isolated rat myocytes and contractility in rat papillary muscles with concentrations of propofol >6 µM. The mirroring of the effects of propofol on ICa and contractility in papillary muscle reported by Zhou et al. (13) confounds the explanation of Azuma et al. (28) for their findings. In summary, the paper by Azuma et al. (28) is the only work proposing major species differences in the effect of propofol on myocardial contractility. The results of these workers, however, are inconsistent with those of several other groups. Similarly, there is no good evidence to suggest that the effects of propofol on isolated myocytes do not reflect the effects of propofol on more physiologic preparations. Steady-state plasma concentrations of propofol found in anesthetized patients range from 10 to 50 µM (30,31). However, propofol in plasma is extensively bound to albumin, leaving a free drug fraction of only 2.2%. Franks and Lieb (32) calculated the 50% effective concentration (EC50) of free drug for anesthesia to be 0.4 µM. We have extrapolated Figure 2 to include this point, and this indicates that an effect was small even at 10 times this EC50 value. Our calculated K0.5 value for the negative inotropic effect of propofol was larger still at 34.5 µM, more than 80 times the EC50 value for anesthesia. Closer examination of the literature reveals that some of the discrepancies alluded to previously between studies are contradictory with regard to the clinical relevance of the effects of propofol, and can be attributed to a failure to account for plasma protein binding of propofol in interpretation of the data. For example, of those studies concluding a clinically relevant effect, the smallest studied propofol concentrations having an effect were 2.8 µM (10), 6 µM (13), 5 µM (15), and 50 µM (12). Stowe et al. (23) found that 91 µM was required to produce a 50% reduction in dLVP/dt in the isolated perfused guinea pig heart, but concluded they did see some effect at 0.5 µM which they, quite reasonably, claimed to be a clinically relevant concentration of propofol. The dLVP/dt value at 0.5 µM propofol was, however, similar to the postexposure control values and deterioration of the preparation may have accounted for this finding. The K0.5 for the effect of propofol on the depression of ICa reported by Luk et al. (14) and Yang et al. (the same group) in 1996 (16) was 54 µM. The difference between this value and our K0.5 for depression of contractility of 34.5 µM may, in part, be caused by the effects of propofol on myofilament Ca2+ sensitivity. The range of plasma concentrations of propofol we have discussed relates to steady-state values of the drug. An even greater variability of propofol concentration is likely to be found on the induction of anesthesia when, depending on the speed of induction, a concentration severalfold higher than steady-state values could be envisaged. Caution must also be used in extrapolating data derived from animals to humans; however, studies using a variety of mammalian species show remarkable consistency, and we speculate that similar dose-response relationships may be found in human tissues if these were readily available. Gelissen et al. (33) studied human atrial muscle strips and found a negative inotropic effect of propofol (IC50, 235 µM). This is an interesting finding; however, it is difficult to relate to the majority of studies in this area that focus on the effects on ventricular contractility. Similarly, our experiments and those of most investigators [Hebbar et al. (27), Riou et al. (34), and Pagel et al. (35) being exceptions] have studied only preparations derived from young healthy animals. Hebbar et al. (27) found no difference in depression of contractility induced by propofol in porcine myocytes derived from healthy pigs or those with experimentally induced heart failure, whereas Riou et al. (34) and Pagel et al. (35) obtained similar results in healthy and cardiomyopathic hamsters and dogs, respectively. This is not to say that other pathologies, or old age, might not produce more profound reductions in contractility at smaller propofol concentrations.
Supported, in part, by a British Journal of Anaesthesia project grant and the British Heart Foundation.
Presented, in part, at the Anaesthetic Research Society, St. Bartholomews Hospital, London, UK, November, 1997.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|