Comparative Effects of Bupivacaine and Ropivacaine on Intracellular Calcium Transients and Tension in Ferret Ventricular Muscle

All experiments conducted in the current study strictly conformed to the Guiding Principles for the Care and Use of Animals  approved by the Council of the Physiologic Society of Japan, Tokyo, Japan. Male ferrets (body weight, 800–1200 g) were anesthetized with pentobarbital sodium (100 mg/kg intraperitoneal administration). Then, the hearts were quickly removed and perfused through the aorta with the normal Tyrode solution (135 mm Na⁺, 5 mm K⁺, 2 mm Ca²⁺, 1 mm Mg²⁺, 102 mm Cl⁻, 20 mm HCO³⁻, 1 mm HPO⁴²⁻, 1 mm SO⁴²⁻, 20 mm acetate, 10 mm glucose, 5 U/l insulin, and pH 7.35 at 30 ± 0.5°C when equilibrated with 5% CO²-95% O²). Papillary muscles with a suitable size (length, 3.0–4.5 mm; diameter, 0.5–1.0 mm) were dissected from the right ventricle. Extreme care was taken during dissection to avoid overstretching of the muscle.

Experiments were carried out according to our previously reported procedure.21–23Briefly, the papillary muscle was mounted horizontally in an experimental chamber perfused with the normal Tyrode solution, where one end of the muscle was connected to a tension transducer (BG-10; Kulite Semiconductor Products, Leonia, NJ) and the other to a motor (JCCX-101A; General Scanning, Watertown, MA). A pair of platinum electrodes was placed parallel to the muscle for electrical stimulation. The muscle was stimulated with a square pulse at 1.2-fold threshold with 5 ms duration at 0.2 Hz. The muscle was slowly stretched from the slack length to the length at which developed tension reached the maximum (L  ^max). The diameter of the preparation was measured to obtain tension per cross-sectional area at L  ^max(in mN/mm²).

After twitch tension had been stabilized, the stimulation was interrupted and the Ca²⁺-sensitive photoprotein aequorin was microinjected into 50–100 superficial cells of the muscle. Aequorin was purchased from John R. Blinks, M.D., Ph.D. (Professor Emeritus, Friday Harbor Laboratories, University of Washington, Friday Harbor, WA). Aequorin signals were detected with a photomultiplier (EMI 9789 A; Ruislip, UK) and converted to [Ca²⁺]ⁱusing an in vitro  calibration curve:

I/I  ^max=[(1+K  ^R[Ca²⁺])/(1+K  ^TR+K  ^R[Ca²⁺])]ⁿ 

where I  indicates aequorin light intensity and I  ^maxindicates peak light intensity at the saturating [Ca²⁺]ⁱ(pCa, 4.5). The latter was calculated from the total light obtained by quickly destroying the cell membrane in the HEPES-Tyrode solution containing 1% (vol/vol) polyethylene glycol mono-p-isooctylphenyl ether (Triton X-100; Nacalai Tesque, Kyoto, Japan) at the end of the experiments. The composition of the HEPES-Tyrode solution was as follows: 128 mm Na⁺, 5 mm K⁺, 2 mm Ca²⁺, 1 mm Mg²⁺, 117 mm Cl⁻, 1 mm SO⁴²⁻, 20 mm acetate, 5 mm HEPES, 10 mm glucose, 5 U/l insulin, and pH 7.40 (adjusted with NaOH) at 30 ± 0.5°C when equilibrated with 100% O². The constants used in the current study were as follows: n , 3.14; K  ^R, 4,025,000; K  ^TR, 114.6.24,25 

After injection of aequorin, the solution was changed to the HEPES-Tyrode solution and experiments were performed. We measured the following parameters in the absence and presence of 10, 30, and 100 μm bupivacaine (Sigma, St. Louis, MO) or ropivacaine (gift from AstraZeneca, Cheshire, UK) during twitch: the peak values of systolic [Ca²⁺]ⁱand tension, the time for aequorin light to reach its peak from the onset of stimulus (time to peak light), the time for aequorin light to decay from 75 to 25% of the peak (decay time, DT), the time for tension measured from the onset of stimulus to the peak (time to peak tension), and the time for tension to decrease from the peak to 50% (relaxation time). We confirmed in vitro  that bupivacaine or ropivacaine at concentrations up to 100 μm did not affect the aequorin light signals in the presence of various concentrations of free Ca²⁺(pCa, 9.0–4.5) (data not shown). For both Ca²⁺transients and tension, sixty-four records of the signals were averaged to improve the signal-to-noise ratio.

In a different set of experiments, we tested the effects of 0.6 μm BAY K 8644 (Sigma, St. Louis, MO) on bupivacaine or ropivacaine (100 μm)-induced decreases in Ca²⁺transients and tension. Previous studies suggest that BAY K 8644 dramatically increases the open probability of sarcolemmal Ca²⁺channels in cardiac muscle, resulting in marked increases in systolic [Ca²⁺]ⁱand developed tension.26In this series of experiments, we measured the peak values of Ca²⁺transients and tension.

Concentration-dependent effects of the anesthetic were assessed using one-way analysis of variance with repeated measures and the Scheffépost hoc  test, and unpaired Student t  test was employed to compare the effects of bupivacaine and ropivacaine. The least-square method was used for correlations between the anesthetic-induced reduction of peak systolic [Ca²⁺]ⁱand that of tension. Paired Student t  test was used in the experiments with BAY K 8644. Statistical significance was verified at P < 0.05 in all cases. All data were expressed as mean ± SD, with “n” representing the number of muscles.

Figure 1shows typical chart recordings for changes in [Ca²⁺]ⁱand tension during twitch in the absence (control) and presence of 100 μm bupivacaine (A) or ropivacaine (B). Under the control condition without an anesthetic, the values of peak systolic [Ca²⁺]ⁱand tension were similar to those previously reported by us.21–23Bupivacaine suppressed Ca²⁺transients and tension more markedly than ropivacaine (note that the difference in the inhibitory effect was particularly noticeable on tension).

Figures 2A and Bsummarize the concentration-dependent inhibitory effects of the anesthetics on peak systolic [Ca²⁺]ⁱand tension. Bupivacaine significantly decreased [Ca²⁺]ⁱat 30 and 100 μm and tension at 10–100 μm in a concentration-dependent manner (fig. 2A). Ropivacaine significantly decreased [Ca²⁺]ⁱand tension at 30 and 100 μm in a concentration-dependent manner (fig. 2B). Conversion of absolute values of [Ca²⁺]ⁱinto percentage changes in [Ca²⁺]ⁱrevealed that the inhibitory effect on [Ca²⁺]ⁱwas significantly greater for bupivacaine at 30 and 100 μm (fig. 2C). Likewise, the inhibitory effect on tension was significantly greater with bupivacaine at all concentrations used (fig. 2D). We plotted the percentage reduction of [Ca²⁺]ⁱversus  that of tension with the anesthetics and found that a significant linear relationship was present between the parameters for both bupivacaine and ropivacaine (fig. 2E). The slopes were significantly different (P < 0.05), with ≈1.0 and ≈1.3 for ropivacaine and bupivacaine, respectively.

Table 1summarizes the effects of bupivacaine and ropivacaine on the time courses of Ca²⁺transients and tension. The local anesthetics prolonged time to peak light and DT, and the effects were statistically significant with 100 μm bupivacaine for both parameters. Time to peak tension tended to be shortened with bupivacaine or ropivacaine. Relaxation time tended to decrease with bupivacaine or ropivacaine, similar to the case when tension is decreased with decreases in the extracellular Ca²⁺concentration.21However, the changes of time to peak tension and relaxation time did not reach statistical significance.

Figure 3shows the influence of BAY K 8644 on the inhibitory effects of 100 μm bupivacaine (A) or ropivacaine (B) on peak systolic [Ca²⁺]ⁱand tension. BAY K 8644 only partially suppressed the inhibitory effects of bupivacaine, whereas it almost completely abolished those of ropivacaine. It should be stressed that the recovery of tension with BAY K 8644 was smaller in magnitude than that of [Ca²⁺]ⁱin the presence of bupivacaine.

In the current study, we demonstrated in insolated ferret ventricular muscle that bupivacaine and ropivacaine suppressed Ca²⁺transients and tension during twitch, with the effects being more pronounced with bupivacaine, particularly on tension. We discuss the current results focusing on the effects of the local anesthetics on intracellular Ca²⁺regulation and myofibrillar activation.

Earlier studies with guinea pig atrial3and rat ventricular muscles20suggest that bupivacaine at a concentration of 10 μm decreases peak twitch tension by ≈30%. Although a different animal species was used in the current study under experimental conditions different than these previous studies, our findings on bupivacaine are in reasonable agreement with those of the previous studies (i.e. , ≈20% decrease) (figs. 2A and D). Taking into account the results of figure 2D(tension), it is reasonable to conclude that bupivacaine exerts a cardiodepressant effect approximately twofold greater than ropivacaine in ferret ventricular muscle.

Previous findings suggest that the acute inhibitory effects of bupivacaine on Ca²⁺transients in cardiac muscle are primarily attributable to its blocking effects on sarcolemmal Na⁺and Ca²⁺channels.27–30It is therefore reasonable to assume that the blocking effects of bupivacaine on cardiac Ca²⁺channels, as well as on Na⁺channels,15are stronger than those of ropivacaine and that this may at least in part be an underlying factor in its greater inhibitory effect on Ca²⁺transients and, therefore, on tension (fig. 2). It is also known that bupivacaine exerts a cardiodepressant effect via  inhibition on the SR function (i.e. , inhibition of Ca²⁺release or Ca²⁺sequestration).17–19Here we discuss the possible effects of bupivacaine and ropivacaine on the SR function.

The aequorin method is a powerful technique to determine the time courses of Ca²⁺transients during twitch (which primarily reflects the SR function) and has been widely used to investigate pharmacological actions of various drugs, including anesthetics.31,32In the current study, time to peak light was prolonged with bupivacaine or ropivacaine and the effect was statistically significant with 100 μm bupivacaine (table 1). This suggests that the local anesthetics, especially bupivacaine, may directly inhibit the SR Ca²⁺release channels,18resulting in slowing of the rise in [Ca²⁺]ⁱupon twitch. Also, both bupivacaine and ropivacaine prolonged DT, and the effect was statistically significant with 100 μm bupivacaine. It is well established that DT is under the strong influence of Ca²⁺removal by the SR Ca²⁺pump. Takahashi et al.  19reported that bupivacaine reduces Ca²⁺uptake by the SR Ca²⁺pump in rabbit masseter muscle. Although higher concentrations were used on different preparations in their study, it is reasonable to assume that the prolonged DT observed with bupivacaine or ropivacaine in the current experiments results from inhibition of the SR Ca²⁺pump. Further studies are needed to establish the effects of bupivacaine and ropivacaine on the SR function and the resultant changes in the time courses of Ca²⁺transients; however, considering the significant effects of bupivacaine on time to peak light and DT, it is reasonable to conclude that the overall effects on the SR function may be substantially greater with bupivacaine than with ropivacaine.

Our regression analysis revealed that there exists a linear relationship between the percentage reduction of peak systolic [Ca²⁺]ⁱand that of tension for both bupivacaine and ropivacaine, and that the slope of the linear regression line was significantly steeper for bupivacaine (fig. 2E). We previously found in ferret ventricular muscle that when the extracellular Ca²⁺concentration was decreased to decrease systolic [Ca²⁺]ⁱ([Ca²⁺]ⁱvaried between ≈1.5 and ≈2.5 μm) the slope of the linear regression line became ≈1.0 (data analyzed from Komukai et al.  21). Therefore, it is reasonable to assume that the cardiodepressant effect of an anesthetic predominantly results from inhibition of intracellular Ca²⁺regulation when the value of the linear regression line is 1.0, whereas the effect involves inhibition of myofibrillar activation (i.e. , the reaction beyond Ca²⁺regulation), as well as that of Ca²⁺transients, when the value is greater than 1.0. Recently, we reported that bupivacaine markedly reduces myofibrillar activation via  inhibition on the actomyosin interaction in rat ventricular muscle.20The expression profiles of contractile proteins may vary in rat and ferret (e.g. , V1 and V3 myosin heavy chains are expressed in rat and ferret, respectively);33,34however, it is likely that bupivacaine suppresses actomyosin interaction in ferret ventricular muscle as well. Here, care should be taken in the interpretation of figure 2E, as the slope of the linear relationship is an approximate measurement of Ca²⁺sensitivity of tension, which may be affected, to some extent, in a condition where the rise of [Ca²⁺]ⁱis vastly altered. Further studies are awaited to establish the Ca²⁺sensitivity measurement with isolated cardiac preparations during physiologic twitch where the effects of bupivacaine and ropivacaine can be examined in a more systematic manner.

BAY K 8644 abolished the inhibitory effects of ropivacaine on Ca²⁺transients and tension, but its effect was partial for bupivacaine, particularly on tension (fig. 3). Therefore, as reported by Hirota et al. ,35the blocking effect of bupivacaine on sarcolemmal Ca²⁺channels in rat cerebrocortical membranes may be more pronounced than that of ropivacaine in ferret ventricular muscle as well. It is also possible that the reduction of Ca²⁺entry through sarcolemmal Ca²⁺channels via  inhibition on Na⁺channels is more pronounced with bupivacaine than ropivacaine as a result of bupivacaine’s stronger blocking effect on Na⁺channels.15In any case, the current work suggests that the inhibitory effect of bupivacaine on Ca²⁺transients via  reduction of Ca²⁺entry is more marked than that of ropivacaine. Here, it should be pointed out that the remaining inhibitory effect of bupivacaine (after the addition of BAY K 8644) was clearly greater on tension (≈60%) than on Ca²⁺transients (≈15%). This appears to support the above notion that, unlike ropivacaine, bupivacaine exerts a cardiodepressant effect via  inhibition of myofibrillar activation as well as via  inhibition of Ca²⁺transients.

Here it is worth noting that local anesthetics such as bupivacaine may modulate contractile function differently in cardiac and skeletal muscle. In fact, Zink et al.  36reported that there are major functional differences in the effects of local anesthetics on the SR function as well as on myofibrillar activation. First, the authors found that bupivacaine shows skeletal muscle toxicity via  increases in myoplasmic Ca²⁺concentrations by enhancing Ca²⁺release from the SR and via  inhibition of Ca²⁺uptake by the SR Ca²⁺pump. In contrast, the current results, as well as the results of others,17,18suggest that bupivacaine inhibits Ca²⁺release as well as Ca²⁺uptake (table 1). The differences in the effect of local anesthetics on Ca²⁺release from the SR may be attributable to differential expression profiles of ryanodine receptors in cardiac (type 2 dominant) and skeletal muscle (type 1 dominant).37Second, the authors reported that bupivacaine increases Ca²⁺sensitivity of tension in skeletal muscle, opposite to its effect on cardiac muscle.20The molecular mechanisms for this discrepancy are unknown, however, differential expression profiles of contractile proteins that can potentially affect myofibrillar activation (i.e. , myosin heavy and light chains, troponin, tropomyosin), and hence resultant differential actions of local anesthetics on myofibrillar activation may be involved. Clearly, further studies are needed to clarify the differences in the local anesthetic actions on Ca²⁺release from the SR and on myofibrillar activation in cardiac and skeletal muscle.

Graf et al.  38reported that bupivacaine at lower concentrations (i.e. , 1–5 μm) than those used in the current study (10 μm and higher) significantly depresses cardiac contractions in guinea pig whole heart. As discussed above, bupivacaine shows similar potencies on isolated cardiac muscle preparations in guinea pig and ferret; species difference unlikely accounts for the difference in the depressant effect of bupivacaine. Instead, we consider that accumulation of bupivacaine inside cardiomyocytes via  coronary vessels may at least in part be involved in its greater effect reported by Graf et al.  38If this is the case, the current experiments, conducted with isolated muscle preparations, may underestimate the cardiodepressant effect of bupivacaine and, probably, of ropivacaine compared with the isolated heart and in vivo  situations. Nevertheless, considering the fact that at the upper range of clinically relevant concentrations (i.e. , 10 μm), the cardiodepression of ropivacaine is significantly less than that of bupivacaine (fig. 2D) and that similar doses are used for both compounds in the clinical settings10,11,39(i.e. , plasma concentration, ≈3-≈10 μm),40,41ropivacaine may exert less of a cardiodepressant effect in humans and may be less likely to cause cardiac arrest on accidental rapid intravenous injection; hence it may have a favorable profile as a local anesthetic.

In conclusion, the cardiodepressant effect of bupivacaine is approximately twofold greater than that of ropivacaine in ferret ventricular muscle. This is because bupivacaine exerts more marked inhibitory effects on Ca²⁺transients and reduces myofibrillar activation. The greater inhibitory effects on Ca²⁺transients of bupivacaine may result from greater inhibition of the SR function, as well as from stronger blocking effects on sarcolemmal Na⁺or Ca²⁺channels. These results provide evidence that ropivacaine may possess a more favorable profile than bupivacaine as a local anesthetic in the clinical settings.

The authors thank Masato Konishi, M.D., Ph.D., Professor, Department of Physiology, Tokyo Medical University, Tokyo, Japan for critical reading of the manuscript and Naoko Tomizawa, Technician, Department of Physiology (II), The Jikei University School of Medicine, Tokyo, Japan for technical assistance.