Role of the Endocardial Endothelium in the Negative Inotropic Effects of Thiopental

Muscle preparation, treatment, and experimental apparatus have been described in detail before. [1]Briefly, beating hearts were excised from cats that were anesthetized with 40 mg/kg sodium pentobarbital given intraperitoneally. Papillary muscles were isolated from the right ventricle. The muscles were mounted vertically in a 7-ml organ bath filled with Krebs-Ringer solution containing 118 mM NaCl, 4.7 mM KCl, 4.7 mM MgSO⁴, 1.2 mM H²O, 20 mM NaHCO³, and 1.25 mM CaCl²[centered dot] H²O (pH = 7.4; 35 degrees Celsius) and bubbled with a gas mixture of 95% O²and 5% CO².

The lower end of the muscle was held by a phosphor-bronze clip, and the upper tendinous end was attached with a Tevdek 7.0 braided thread to an electromagnetic force-length transducer. Muscles were stimulated electrically at 0.2 Hz at a voltage of 10% above threshold by rectangular pulses lasting 5 ms through two longitudinally arranged platinum electrodes. After a stabilization period of 2 h at 35 degrees Celsius (the temperature at which the actual experiments were performed), further stabilization of the muscles was continued for 1 h at the muscle length at which the maximal active tension was developed (l^max), until steady state was obtained.

Myocardial performance was assessed from preloaded isotonic and isometric twitches at 35 degrees Celsius and at l^max. Isometric parameters measured included total tension, time to total tension (TtTT), time to half relaxation (t1/2), maximal rate of force increase (df^max), and force decline (df^min). Isotonic parameters measured were peak shortening, time to peak shortening (TtPS), and peak velocity of lengthening and shortening. Muscle cross-sectional area was calculated by dividing the lightly blotted wet weight of the muscle at the end of the experiment by its length at l^max, assuming a cylindrical shape and a specific gravity of 1.0. Tension measurements were normalized by muscle cross-sectional area and length measurements by l^max.

The animal care and investigations conformed to institutional guidelines and to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication 85-23, revised 1985).

To evaluate a possible role of EE in the negative inotropic effects of thiopental, effects of increasing doses of thiopental on isometric and isotonic muscle contraction parameters were measured under different experimental conditions. In a first protocol, the effects of increasing doses of thiopental were evaluated in the presence and absence of an intact EE layer. Subsequently, whether observed different effects of thiopental with and without EE could be modulated by acting on the NO pathway was determined. Therefore, in a second protocol, the inotropic effects of thiopental were evaluated in muscles with and without EE after pretreatment of the muscles with L-NAME (blocking of nitric oxide synthase and nitric oxide production). In the third protocol, the inotropic effects of thiopental were evaluated in muscles with and without EE after the muscles were pretreated with L-arginine (a precursor of nitric oxide).

The EE was selectively damaged by a 1-s immersion of muscles in 0.5% Triton X-100 while in their working position, immediately followed by an abundant wash with Krebs-Ringer solution at 35 degrees Celsius. This procedure was previously shown to induce a characteristic alteration of the isometric and isotonic twitch. [1]Damage to the EE resulted in an immediate and irreversible abbreviation of the muscle twitch, and the onset of tension decline occurred earlier during the isometric twitch with a concomitant decrease in peak twitch tension, but without change in the early phase of the twitch. Maximal unloaded shortening velocity (Vmax) remained unchanged. These observations differ from other negative inotropic interventions such as decreasing extracellular calcium or reduction of cyclic adenosine monophosphate-mediated effects (beta-blocking agents), which are all associated with changes in Vmax and in the early phase of the isometric twitch. Morphologic observations, including light microscopy, scanning electron microscopy, and transmission electron microscopy, confirmed that this procedure selectively damaged the EE but without damaging the underlying myocardium. [20]The 1-s Triton exposure was only a negligible fraction of the time needed to damage the subjacent myocardium. The effects of selective EE damage on the mechanical performance of isolated cardiac muscle were confirmed in different animal species [21-23]with various chemical, [1]physical, [20]and pharmacologic [24,25]damaging procedures, and at different stimulation frequencies, muscle lengths, temperatures, and extracellular calcium concentrations, [1]thereby confirming that the observed effects express an EE-mediated effect and are not caused by a specific method used.

After immersion in Triton, the muscles were allowed to stabilize for another hour before the protocols were started. In muscles not subjected to Triton treatment, the same total stabilization time was maintained to exclude different time courses in muscles with and without EE.

The possible role of EE was studied by performing parallel experiments in two groups of randomly assigned muscles. In the first group (group A, n = 8), EE was undamaged. In the second group (group B, n = 8), EE was selectively damaged by immersing muscles for 1 s in 0.5% Triton X-100 while in their working position, immediately followed by an abundant wash with Krebs-Ringer solution at 35 degrees Celsius. After immersion in Triton, the muscles were allowed to stabilize for another hour. In each group, cumulative concentration responses were obtained for increasing doses of thiopental (1.5, 3, 6, 9, 12, and 24 micro gram/ml, which corresponds to 6 x 10 sup -6 M, 11 x 10 sup -5 M, 22 x 10 sup -5 M, 34 x 10 sup -5 M, 45 x 10 sup -5 M, and 9 x 10 sup -4 M, respectively). Muscles were allowed to stabilize for 30 min at each concentration before measurements were recorded for analysis.

The influence of pretreating the muscles with L-NAME on the inotropic effects of increasing doses of thiopental in muscles with and without EE was studied by performing parallel experiments in two groups of randomly assigned muscles. In one group (group C, n = 8), EE was undamaged. In the other group (group D, n = 8), EE was selectively damaged. Muscles were then pretreated by adding 10 sup -3 M N^G-nitro-L-arginine methyl ester (L-NAME; blocking of nitric oxide synthase and production) to the bath solution. A stabilization period of 30 min was allowed after administering L-NAME before muscles were exposed to increasing doses of thiopental. Muscles were allowed to stabilize for 30 min at each concentration before measurements were recorded for analysis.

The influence of pretreatment of the muscles with L-arginine on the inotropic effects of increasing doses of thiopental in muscles with and without EE was studied by performing parallel experiments in two groups of randomly assigned muscles. In one group (group E, n = 8), EE was undamaged. In the other group (group F, n = 8), EE was selectively damaged. Muscles were then pretreated by adding 5 x 10 sup -4 M L-arginine (a precursor of nitric oxide) to the bath solution. A stabilization period of 30 min was allowed after administering L-arginine before muscles were exposed to the increasing doses of thiopental. Muscles were allowed to stabilize for 30 min at each concentration before measurements were recorded for analysis.

Data from the different muscle groups were compared using a multivariate analysis of variance for repeated measurements to determine the effects of inter-muscle variation, the effects of the different experimental conditions, and the effects of increasing doses of thiopental. Interaction analysis compared the effects of thiopental in the different experimental conditions and assessed the influence of these conditions on the magnitude of the effects of thiopental. Analysis of variance was followed by a Dunnett multiple comparison test. All values are expressed as mean +/- SEM. Statistical significance was accepted at P < 0.01.

Data on muscle length at L^max, mean cross-sectional area, peak developed total tension, and ratio of resting to peak total tension at L^max(RT/TT) are summarized in Table 1. Among the six muscle groups, there were no statistically significant differences in any of the muscle characteristics.

(Figure 1) shows a representative example of the effects of increasing doses of thiopental (1.5, 3, and 6 micro gram/ml) on an isometric and an isotonic twitch. Twitches obtained in the presence of an intact endocardial endothelium (+EE) are illustrated in the upper part of Figure 1, whereas the effects obtained after damage of the endocardial endothelium (-EE) are illustrated in the lower panel of Figure 1. As is apparent from the lower panel of Figure 1, damage of the EE induced specific changes, which were described in detail before. [1,2]Isometric twitch duration decreased together with a decrease in total tension. Similarly, isotonic twitch duration decreased with a concomitant decrease in peak shortening. In the presence of an intact EE (upper), increasing doses of thiopental caused a dose-dependent decrease in total tension (left) and peak shortening (right). After removal of the EE (lower), increasing doses of thiopental hardly affected total tension and peak shortening.

(Figure 2) shows mean data of contractile parameters of isometric and isotonic twitches of all muscles. On each panel of Figure 2, the two lines represent the dose-response curve for thiopental in the presence (squares, filled lines) of EE and after its removal (circles, dashed lines). "Base" represents the first baseline measurement, whereas EE represents the measurement obtained 30 min later. For the muscles with intact EE, this measurement was identical to the baseline measurement, but for the muscles with damaged EE, parameters of contraction were altered. Condition "EE" represents the values to which the increasing doses of thiopental are compared in both groups. The filled squares and circles represent a statistically significant difference with respect to data at condition "EE."

With an intact EE, thiopental induced a dose-dependent decrease in total tension from 44.8 +/- 8.4 mN/mm²at baseline to 22.8 +/- 3.8 mN/mm²at a concentration of 24 micro gram/ml. This change in total tension was significant starting from a concentration of 1.5 micro gram/ml (P = 0.0001; filled squares in Figure 1). Time to half isometric relaxation also decreased in a dose-dependent manner from 454 +/- 26 ms at baseline to 363 +/- 28 ms at a concentration of 24 micro gram/ml (P = 0.0001, filled circles).

Damage to the EE decreased peak isometric total tension from 46.1 +/- 8.2 to 37.6 +/- 7.7 mN/mm²(p = 0.0002). Onset of isometric relaxation occurred earlier, with a decrease in t1/2 from 444 +/- 21 to 388 +/- 19 ms (P = 0.004). Increasing doses of thiopental up to 6 micro gram/ml did not change total tension significantly. Total tension at 6 micro gram/ml was 35.3 +/- 6.5 mN/mm²and t1/2 was 386 +/- 18 ms. Only at thiopental concentrations greater than 6 micro gram/ml was a dose-dependent decrease in total tension and t1/2 found (filled circles). Interaction analysis revealed that this different dose-dependent effect of thiopental in the presence and the absence of EE was statistically significant (P = 0.0009 for total tension and P = 0.0001 for t1/2).

The different effects of thiopental in muscles with and without EE was also present for TtTT (P = 0.0014). Maximal rate of force increase (df^max) and force decline (df^min) decreased with increasing doses of thiopental, but this effect was similar in the presence or the absence of EE (P = 0.6206 and P = 0.1484, respectively).

A similar phenomenon was found for the parameters of isotonic twitch (Figure 1). In the presence of EE, thiopental induced a dose-dependent decrease in peak muscle shortening from 0.110 +/- 0.010 mm at baseline to 0.062 +/- 0.009 mm at a thiopental concentration of 24 micro gram/ml (P = 0.007). Similarly, TtPS decreased in a dose-dependent manner from 252 +/- 14 ms at baseline to 230 +/- 14 ms at the highest thiopental concentration (P = 0.0008).

Damage to the EE also altered isotonic twitch parameters, with a decrease in peak shortening from 0.116 +/- 0.008 mm to 0.098 +/- 0.007 mm (P = 0.005) and an earlier isotonic relaxation with a decrease in TtPS from 248 +/- 17 ms to 224 +/- 14 ms (P = 0.007). In the absence of EE, increasing doses of thiopental up to 6 micro gram/ml did not change peak shortening. Only at higher doses of thiopental was a dose-dependent decrease in peak shortening found (Figure 1, filled circles). This different effect of thiopental in groups A and B was statistically significant (P = 0.0046). The different effects of thiopental in groups A and B was also apparent for TtPS and peak velocity of shortening (P = 0.0016 and P = 0.0015, respectively). Peak velocity of lengthening decreased with increasing doses of thiopental, but this effect was not different in both groups (P = 0.1052).

(Figure 3) shows mean data of contractile parameters of isometric and isotonic twitches of all muscles in protocol 2. On each panel of Figure 3, the two lines represent the dose-response curves for thiopental in muscles with intact EE (squares, filled lines) and in muscles in which EE had been removed (circles, dashed lines). "Base" represents the first baseline measurement, "EE" represents the measurement obtained 30 min later (one group with intact EE, the other group with damaged EE), and L-NAME represents the data obtained after administering 10 sup -3 M n^G-nitro-L-arginine methyl ester. The filled squares and circles represent a statistically significant difference with respect to data at condition "EE."

Damage to EE resulted in the typical alterations of isometric and isotonic twitch parameters, which were described before. Administration of L-NAME slightly increased total tension (P = 0.035) and there was a tendency for an increase in peak shortening (P = 0.072). None of the other parameters of isometric and isotonic contraction were significantly altered.

In muscles with and without EE, increasing doses of thiopental up to 6 micro gram/ml did not alter contraction parameters. Only at doses greater than 6 micro gram/ml was muscle contraction depressed in a dose-dependent manner. Interaction analysis showed that the effects of increasing doses of thiopental in groups C and D were similar (P = 0.184 for total tension and P = 0.091 for t1/2). This phenomenon was also apparent for TtTT (P = 0.076), df^max, and df sub min.

Effects of increasing doses of thiopental on peak shortening and TtPS were also similar in groups C and D (P = 0.083 and P = 0.172, respectively). Peak velocity of shortening changed with increasing doses of thiopental in a similar way in groups C and D (P = 0.325). The effects of increasing doses of thiopental on peak velocity of lengthening also were similar in the different experimental conditions.

(Figure 4) shows mean data of contractile parameters of isometric and isotonic twitches of all muscles of protocol 3. On each panel of Figure 4, the two lines represent the dose-response curve for thiopental in muscles with intact EE (squares, filled lines) and in muscles without EE (circles, dashed lines). "Base" represents the first baseline measurement, "EE" represents the measurement obtained 30 min later (one group with intact EE, the other group with damaged EE), and "L-arg" represents the data obtained after administering 5 x 10 sup -4 M L-arginine. The filled squares and circles represent a statistically significant difference with respect to data at condition EE.

Damage to the EE resulted in the typical alterations of isometric and isotonic twitch parameters, described before. Administering L-arginine did not change any of the parameters of isometric or isotonic twitch significantly. In the presence of an intact EE, thiopental caused a dose-dependent decrease in total tension (Figure 4, filled squares). This effect was apparent from the lowest dose of 1.5 micro gram/ml, whereas in muscles in which EE had been removed the negative inotropic effects of thiopental were only apparent at higher doses. Interaction analysis revealed a significant difference in the effects of thiopental between groups E and F (P = 0.0073). A similar phenomenon was observed for t1/2 (P = 0.0061) and TtTT (P = 0.0092). The effects of increasing doses of thiopental on df^maxand df^minwere comparable for the different experimental conditions.

A similar observation was made for the parameters of isotonic twitch. The effects of increasing doses of thiopental on peak shortening TtPS were different between groups E and F (P = 0.0086 and P = 0.0093, respectively). Peak velocity of shortening and lengthening changed with increasing doses of thiopental in a similar way in groups E and F (P = 0.415).

The data from protocol 1 showed that negative inotropic effects of low concentrations of thiopental (1.5 to 6 micro gram/ml) depend on the presence of an intact EE. The negative inotropic effects of these low doses of thiopental could be abolished by pretreating the muscles with the nitric oxide synthase inhibitor L-NAME. Interaction analysis revealed that the effects of increasing doses of thiopental in the muscles with an intact EE (group A) were significantly different from the effects observed in muscles with EE pretreated with L-NAME (group C; P = 0.0084 for total tension and P = 0.001 for t1/2), but were similar to the effects of thiopental in muscles without EE (group B) (P = 0.4888 for total tension and P = 0.3768 for t1/2). This phenomenon was also apparent for TtTT (differences in effects of thiopental among groups B, C, and D: P = 0.3651, and among groups A, C, and D: P = 0.0076). The effects of increasing doses of thiopental on df^maxand df^minwere not altered in the different experimental conditions.

A similar observation was made for the parameters of isotonic twitch. The effects of increasing doses of thiopental on peak shortening and TtPS in the muscles with an intact EE (group A) were significantly different from the effects observed in muscles with EE pretreated with L-NAME (group C; P = 0.0096 for peak shortening and P = 0.0083 for TtPS) but were similar to the effects of thiopental in muscles without EE (group B; P = 0.634 for peak shortening and P = 0.726 for TtPS). The effects of increasing doses of thiopental on peak velocity of lengthening were similar for the different experimental conditions.

Pretreatment of the muscles with L-arginine tended to accentuate the negative inotropic effects of thiopental at lower doses (1.5 to 6 micro gram/ml) and to attenuate the effects at higher doses (9 to 24 micro gram/ml). Interaction analysis, however, could not reveal a statistically significant difference in the effects of thiopental between muscles with an intact EE (group A) and muscles with an intact EE pretreated with L-arginine in any of the parameters of isometric and isotonic twitch.

(Figure 5) shows these different effects of increasing doses of thiopental on isometric total tension in the various experimental conditions.

Studies on vascular endothelium have provided evidence that different anesthetic agents may interfere with endothelium-dependent regulation of vascular muscle tone. [14-19]Because vascular and endocardial endothelium form a contiguous layer and it has been shown that EE plays an important role in regulating myocardial function, [1,2]it might be hypothesized that EE is also involved in modulating the inotropic effects of anesthetic agents. Thiopental is well known to cause strong depressant cardiovascular effects in humans, [12]anesthetized animal models, [10,11]and isolated cardiac muscle. [6-9,13]The present study evaluated whether EE mediated inotropic effects of thiopental and whether these EE-dependent inotropic effects of thiopental could be modulated by acting on the nitric oxide pathway. An isolated cardiac muscle preparation was used because it offered the advantages of quantifying direct effects of drugs on contractility and of eliminating changes secondary to alterations in systemic hemodynamics and the central nervous system.

The negative inotropic effects of thiopental in isolated cat papillary muscle with intact EE were confirmed. However, as shown in Figure 1and Figure 2, two different types of response can be identified: at low doses of thiopental, a decrease in isometric peak twitch tension was noted, but maximal rate of force development and force decline were hardly affected. Only at higher doses was the decrease in peak twitch force development accompanied by a dose-dependent decrease in rate of force development and force decline. Decreased myocardial performance may present in various ways, either by a decrease in the rate of force development and peak force, by a decrease in rate of relaxation, or by a decrease in the duration of tension development. Most myocardial depressant agents, such as beta-blocking agents, induce a decline in peak twitch force development with a concomitant decrease in both maximal rate of force development and force decline. Endocardial modulation of myocardial performance, on the contrary, is characterized by a change in twitch force development with a change of twitch duration but with no effects on rate of force development and force decline. [1]When observing the myocardial depressant effects of thiopental, the effects of the low doses of thiopental seemed to strongly resemble the type of modulation by EE, whereas at higher doses myocardial depression also involved rate of force development and force decline. The role of EE in modulating the inotropic effects of thiopental was confirmed in the experiments on muscles in which EE was removed. In these muscles, the negative inotropic effects of the low doses of thiopental (1.5 to 6 micro gram/ml) were absent. When exposed to higher concentrations of thiopental (> 6 micro gram/ml), the negative inotropic effects were similar to those seen in muscles with an intact EE. These data indicated that EE is involved in regulating the inotropic effects of low concentrations of thiopental. More specifically, it appeared that the presence of an intact EE is required for the negative inotropic effects of low doses of thiopental. These low concentrations are in the range of the free plasma thiopental levels found after intravenous induction with thiopental. The peak plasma concentration of thiopental during induction of general anesthesia in humans has been estimated to be 5 x 10 sup -5 to 3 x 10 sup -4 M. [26,27]Because approximately 85% of the drug is bound to plasma proteins, the free plasma concentration is less than 5 x 10 sup -6 M, or approximately 3 to 6 micro gram/ml.

We can draw no definitive conclusions about the mechanisms for this EE-dependent modulation of myocardial function by low doses of thiopental. Cardiac endothelium releases at least three chemical messengers that may influence myocardial performance. First, EE cells express nitric oxide synthase activity (constitutive and inducible forms). [5,28]Nitric oxide release by EE cells has been shown to elevate myocardial cyclic guanosine monophosphate (cGMP), and both nitric oxide and cGMP decrease myocardial contraction in isolated ferret papillary muscles. [29]Second, cultured EE cells from sheep release large amounts of prostaglandins, as much as 20 times more than vascular endothelium. However, the role of prostaglandins in controlling myocardial performance is not well understood. [4]Third, EE cells contain endothelin-1 mRNA. [3]In the heart, endothelin-1 is highly inotropic, even at low concentrations. [30] 

The EE-dependent negative inotropic effect of thiopental thus could be explained either by stimulating an EE-derived myocardial relaxing factor, by blocking a myocardial contracting factor, or by a combination of both. The known EE-derived relaxing factors are prostacyclin and nitric oxide. In our study, we further evaluated a possible involvement of the nitric oxide pathway in the observed EE-dependent effects of thiopental. The EE-dependent negative inotropic effect appeared to be completely blocked by the nitric oxide synthase inhibitor L-NAME. Thus these data strongly suggested that the nitric oxide pathway is involved in the observed negative inotropic effects of low doses of thiopental. This was confirmed by observations in muscles with EE, after pretreatment with L-arginine (5 x 10 sup -4 M), which is a precursor of nitric oxide and known to stimulate nitric oxide formation. In these muscles, low doses of thiopental had a slightly greater negative inotropic effect than in the muscles without L-arginine.

Although the present data show that the negative inotropic effects of low doses of thiopental are mediated through EE, possibly by releasing nitric oxide, they do not allow us to draw conclusions about the precise mechanisms of this phenomenon. Increasingly the nitric oxide-cGMP system is recognized as involved in regulating myocardial function. Recent data have shown that nitric oxide and cGMP induced a concentration-dependent biphasic contractile response, which could be modulated by the status of the EE and by cholinergic and adrenergic stimulation. [31]Evidence also shows that the nitric oxide-cGMP system might be involved in regulating the L-type Calcium²+ current (1^Calcium) in the myocardium. [32]The level at which the nitric oxide-cGMP pathway is involved in the EE-dependent modulation of the inotropic effects of thiopental, however, remains to be established. In addition, it cannot be excluded that still other myocardial relaxing factors (prostaglandins) or myocardial contracting factors (endothelins) are involved in the observed results.

To our knowledge, no data are available until now on possible EE dependence of inotropic effects of anesthetic agents. On the other hand, several reports have discussed endothelium dependency of the vasoactive actions of anesthetics. [14-18]Underlying mechanisms for these endothelium-dependent actions remain controversial. Terasako and coworkers [19]found on canine mesenteric arteries and rat aortae that thiopental at 1 and 3 x 10 sup -4 M strongly inhibited endothelium-dependent and independent relaxation. Barbiturates therefore seemed to exert their effect at the level of the vascular smooth muscle, where they inactivated nitric oxide or inhibited its action, either by inactivating guanylate cyclase or by facilitating cGMP breakdown.

The results of the present study, on the contrary, also suggested an EE-dependent inotropic effect of thiopental, mediated by the nitric oxide pathway. Park and associates [33]showed that that treatment with NO synthase inhibitor did not significantly modify the vascular effects of thiopental 1 and 3 x 10 sup -4 M in rat aorta and pulmonary artery. Treatment with the cyclooxygenase inhibitor indomethacin, on the other hand, diminished relaxation induced by thiopental in arteries with intact endothelium but not in the denuded ones. Thus they suggested that thiopental induced the release of vasodilative cyclooxygenase metabolites such as prostacyclin from endothelium.

Apparently these results differ from the present study. However, it should be noted that although vascular endothelium and EE form a contiguous layer, morphologic and physiologic properties differ among the different types of endothelium. [34]In addition, the doses of thiopental that we used are lower than those usually reported. We found EE dependency of the inotropic effects of thiopental in concentrations less than 6 micro gram/ml, which corresponds to doses less than 22 x 10 sup -5 M. In higher doses, EE dependency was no longer apparent, and other mechanisms for the negative inotropic effects are involved. It is also possible that modulation of myocardial function by EE is more complex and that prostaglandins do not necessarily play a direct role in modulating myocardial performance by EE but rather are merely involved in the trigger mechanism for releasing other relaxing or contracting factors.

Another possibility is that the EE-dependent inotropic effects of thiopental result from an inactivation of one or more EE-dependent myocardial contracting factors. In an isolated vascular ring model, thiopental was shown to relax large and small coronary arteries preconstricted with endothelin only at higher concentrations than those observed clinically and far exceeding those we used. A possible endothelial-mediated effect, however, was not evaluated in that study. [35] 

Final elucidation of the physiologic mechanisms responsible for the observed EE-dependent negative inotropic effects of thiopental will first require identification of the different potential EE-dependent myocardial relaxing and contracting factors. In addition, evaluation of the effects of thiopental on the synthesis, release, or (in)activation of these substances at endocardial or myocardial levels will reveal the exact nature of the EE-dependent inotropic actions of thiopental.

In conclusion, our results show that the negative inotropic actions of small doses of thiopental (1.5 to 6 micro gram/ml or 6 x 10 sup -6 to 22 x 10 sup -5 M) depend on the presence of an intact EE. In muscles with an intact EE, we noted a dose-dependent decrease in myocardial function, whereas in muscles without EE, we found no change in myocardial function at small doses of thiopental. These EE-dependent negative inotropic effects of thiopental appeared to be mediated-at least in part-by the nitric oxide pathway. The present results describe in vitro observations of isolated papillary muscles in highly oxygenated aqueous buffer. The function of both EE and myocardium may differ from in vivo observations. The potential clinical implications of these findings thus require further investigation.