Effects of Ketamine on Ventricular Conduction, Refractoriness, and Wavelength : Potential Antiarrhythmic Effects: A High-resolution Epicardial Mapping in Rabbit Hearts

The principles for the care and treatment of experimental animals complied with the national guidelines of the French Ministry of Agriculture. We used 15 New Zealand rabbits weighing 2.8–3.6 kg. After anesthesia with etomidate (1 mg/kg given intravenously followed by an infusion of 0.1 mg [center dot] kg sup -1 [center dot] min sup -1), the trachea was intubated, and the lungs were mechanically ventilated with 100% oxygen (Logic 07; ATM Ohmeda, Coignieres-Maurepas, France). The thorax was surgically opened by a midsternal incision. The aorta and the heart were exposed and, after anticoagulation with heparin (1,000 IU), they were rapidly removed and placed in cold perfusion fluid (10 [degree sign] Celsius). The aorta was immediately cannulated and the heart connected to a Langendorff perfusion system using Tyrode's solution. The heart was perfused with a constant flow of 40–50 ml/min (Watson-Marlow 101U pump; Falmouth Cornwall, UK), resulting in a pressure of 70 +/- 10 mmHg (Gould P23 transducer, Oxnard, CA; CGR monitor, St. Cloud, France). The Tyrode's solution consisted of 130 mM NaCl, 20.1 mM NaHCO³, 4 mM KCl, 2.2 mM CaCl², 0.6 mM MgCl², 1.2 mM NaH²PO⁴, and 12 mM glucose. The solution was saturated with a mixture of 95% O²and 5% CO²and the pH was adjusted at 7.40 +/- 0.02.

Five Langendorff-perfused hearts were kept intact (nonfrozen heart). In 10 other hearts, an endocardial cryotechnique was used to freeze the complete right ventricle, the interventricular septum, and the endocardial and intramural layers of the free wall of the left ventricle (frozen heart). [16,24,25] Briefly, a cryoprobe was inserted through the pulmonary artery into the right ventricle, filled with liquid nitrogen (-192 [degree sign] Celsius), and maintained in place until the right ventricle was completely frozen. The heart was then immersed in a tissue bath containing perfusion fluid at 30 [degree sign] Celsius. The cryoprobe was placed into the left ventricular cavity through the left atrium and the coronary circulation was temporary discontinued. The cryoprobe was filled with liquid nitrogen and maintained in place for 3 min. Thereafter, the coronary perfusion was restored, the probe was removed, and the heart withdrawn from the tissue bath. Then the temperature of the heart was kept constant at 37 [degree sign] Celsius throughout the experiment. As a result of this procedure, only a thin epicardial layer of the free wall of the left ventricle, approximately 1 mm thick, survived because the endocardial and intramural layers were completely destroyed. Previous investigations showed that in this thin surviving layer, refractoriness and conduction velocity are not affected by the procedure and remained stable for many hours, suggesting adequacy of the circulation in the epicardial layer. [13,24] This cryotechnique was used to avoid epicardial breakthrough of longitudinal wavefronts from deeper layers and to allow complete and bidimensional epicardial mapping of electrical activation. At the end of the experiments, the hearts were dissected to verify the efficacy of the cryoprocedure.

Five nonfrozen hearts were perfused at 37 [degree sign] Celsius in spontaneous sinus rhythm. After a 60-min stabilization period, they were given 50, 100, 150, and 200 micro Meter racemic ketamine, with each concentration infused for 20 min. After the last concentration, ketamine was washed out for 40 min to verify the return to baseline values. Ketamine was obtained commercially (50 mg/ml) in water, with chlorobutanol (5 mg/ml) as a preservative (Panpharma, Fougeres, France). We did not use a chlorobutanol-treated group because the highest concentration of chlorobutanol given with ketamine was 5.5 micro gram/ml, whereas others have shown that chlorobutanol has no significant cardiac effect at 15 micro gram/ml. [26] In nonfrozen hearts, the spontaneous sinus cycle length (RR interval, in milliseconds) and the occurrence of asystole or spontaneous arrhythmias were recorded at baseline, after each concentration of ketamine, and after washout. If sustained monomorphic tachycardias occurred, they were to be terminated by overdrive pacing, and atrial or ventricular fibrillation were to be stopped by administration of potassium chloride.

Ten frozen hearts were treated with 50, 100, 150, and 200 micro Meter ketamine.

Recording and Induction of Ventricular Dysrhythmias. High-resolution mapping of epicardial excitation was performed by applying to the epicardial surface a spoon-shaped electrode (Figure 1) containing 256 unipolar electrodes at regular distances of 2.25 mm. The computerized mapping system allows simultaneous recording, storage, and automatic analysis of all 256 electrograms and on-line presentation of color-coded activation maps (Maptech system, Maastricht, The Netherlands). [27,28] Programmed electrical stimulation was performed using a programmable constant-current stimulator that delivered square pulses lasting 2 ms at twice the diastolic threshold for both regular stimulation and induction of premature beats (Maptech system). Bipolar stimulation could be performed through any pair of electrodes in the spoon electrode. The stimulation protocol was only performed in frozen hearts and consisted of (1) application of one, two, and three premature stimuli (S2, S3, and S4, respectively) delivered with a decreasing coupling interval after 10 basic stimuli (S1-S1) at a 300-ms interval; and (2) application of trains of 10 stimuli at a regular cycle length, which was progressively decreased at 10-ms steps until 1-to-1 capture of the ventricle failed (F^max). After the inducibility of ventricular dysrhythmias was assessed on baseline in the 10 frozen hearts, 50, 100, 150, and 200 micro Meter ketamine were successively infused to the circuit for 20 min. Then the inducibility of dysrhythmias was tested again using the same protocol as during baseline. Once the protocol was completed, normal Tyrode solution was infused for approximately 40 min to return to baseline conditions to rule out the possibility of deterioration over time (washout). Thereafter, any occurrence of inducible ventricular dysrhythmias was also recorded and analyzed.

Electrophysiologic Measurements. The following parameters were measured in frozen hearts at baseline, after each ketamine concentration, and after washout: longitudinal ventricular conduction velocity (theta L, expressed in centimeters per second), transverse ventricular conduction velocity (theta T, expressed in centimeters per second), anisotropic ratio (theta L/theta T), VERP (expressed in milliseconds), and ventricular wavelength (lambda, expressed in millimeters). During the experiment, ventricular conduction velocities were recorded using the same pacing site, which was located at the center of the thin surviving layer of the left ventricle. As previously described by Clerc [29] and Spach et al., [30] cardiac tissue has a different axial resistance along and perpendicular to the fiber axis of the myocardial fibers. This different axial resistance results in direction-dependent differences in conduction velocity (anisotropic conduction) at baseline. Therefore, pacing at the center of the thin surviving layer of the left ventricle produced an ellipsoidal spread of propagation. It has been shown that the long axis of this ellipsoid is parallel to the fiber orientation and has a fast conduction (longitudinal conduction, theta L), whereas the short axis is perpendicular to the fiber orientation and has a slow conduction (transverse conduction, theta T). [13] Conduction velocity was defined as the distance traveled by the wavefront normal to the isochrones per unit of time. In each experiment, both longitudinal and transverse conduction velocities and anisotropic ratio were measured after 10 basic stimuli (S1-S1) at 1,000-ms intervals. In addition, to test the use dependency of ketamine, the longitudinal and transverse ventricular conduction velocities and anisotropic ratio were measured after 10 basic stimuli at 900-, 800-, 700-, 600-, 500-, 400-, 300-, 250-, and 200-ms intervals. Use dependency of ketamine is defined as the ketamine-induced changes in the rate-dependent ionic membrane properties, including conduction and refractoriness. The VERP was defined as the shortest S1-S2 interval that still resulted in a propagated premature impulse during regular pacing with a S1-S1 interval of 300 ms. The VERP was determined by decreasing the coupling interval of the premature stimulus in 1-ms steps. To test the spatial dispersion of refractoriness, VERP was measured on five different epicardial sites: center of left ventricular area (main pacing site), apex, near the free wall of the left ventricle, near the left anterior descending coronary artery, and near the base of the heart (Figure 1). Wavelength (lambda) was defined as the product of the conduction velocity and the effective refractory period. Because ventricular conduction is anisotropic, lambda was calculated for longitudinal (lambda L) and transverse (lambda T) directions, using theta L and theta T, respectively, measured at a pacing cycle length (PCL) of 300 ms. This PCL was chosen because it is comparable to the normal rabbit heart rate.

Definition of Ventricular Arrhythmias. We defined ventricular arrhythmias as ventricular premature beats, ventricular fibrillation, and sustained and nonsustained ventricular tachycardia. Finally, monomorphic and polymorphic tachycardia were distinguished.

Spontaneous sinus cycle length (RR) in the nonfrozen hearts, and VERP, longitudinal, and transverse ventricular conduction velocities and anisotropic ratio in frozen hearts were expressed as means +/- SD. Wavelengths were expressed as both the absolute value (in millimeters) and the percentage of change of mean values. All parameters were analyzed using a two-way analysis of variance for repeated measures followed by a Neuman-Keuls test and Bonferroni's correction. Variables used to determine the dispersion of refractoriness using analysis of variance were the dose and the site at which VERP was measured. In addition, we compared, in each heart, the area under the curve of VERP plotted for each site. This was designed to corroborate the results of analysis of variance. P < 0.05 was considered significant.

Ketamine induced a dose-dependent lengthening of the spontaneous sinus cycle length (RR). The value of the RR interval was 325.6 +/- 41.3 ms at baseline and increased to 374.6 +/- 50.8 ms (P <0.01), 427.6 +/- 57.6 ms (P < 0.01), 464.6 +/- 60.6 ms (P < 0.005), and 502.6 +/- 70.1 ms (P < 0.005) after 50, 100, 150, and 200 micro Meter ketamine, respectively. After washout, all preparations returned to their baseline values (353.8 +/- 56.5 ms). No dysrhythmia occurred spontaneously during and after ketamine administration.

Conduction velocities, anisotropic ratio, and VERP were measured during ventricular pacing at a PCL ranging from 1,000–200 ms for the former and of 300 ms for the latter. During the study of conduction velocities and anisotropic ratio, all hearts could be paced until a PCL of 300 ms, whatever the dose of ketamine. At a PCL of 250 ms, all hearts could be paced after 50 micro Meter ketamine, nine hearts after 100 and 150 micro Meter, and seven hearts after 200 micro Meter ketamine. At a PCL of 200 ms, seven hearts could be paced after 50 micro Meter ketamine, only three hearts after 100 and 150 micro Meter, and none after 200 micro Meter ketamine. After washout, all hearts recovered their ability to be paced until 200 ms.

No ventricular dysrhythmia could be induced at baseline and after increasing doses of ketamine. Table 1reports the effects of ketamine on ventricular longitudinal and transverse conduction velocity and on anisotropic ratio. Ketamine induced a dose-dependent slowing of longitudinal conduction velocity that was significant from 50 micro Meter (P < 0.05). Transverse conduction velocity was also decreased in a dose-dependent manner, with statistical significance reached at 100 micro Meter (P < 0.05). No significant use dependency was observed, either on theta L or theta T (Figure 2). Anisotropic ratio was unchanged whatever the dose or the PCL (Table 2).

The VERP was prolonged in a dose-dependent manner, which was significant from 50 micro Meter ketamine (P < 0.05;Table 3). There was no significant intersite variability when assessed by the comparison of mean values of VERP (Table 4) and the comparison of areas under the curves of VERP in each site. Figure 3shows four different activation maps recorded during a regular pacing at 1,000 ms in the same heart on baseline, after 100 and 200 micro Meter ketamine, and after 200 micro Meter ketamine at a PCL of 400 ms.

In Figure 4, ketamine-induced changes in ventricular lambda are shown. Baseline values of lambda L and lambda T were 114.9 +/- 15.5 mm and 58.5 +/- 9.3 mm, respectively. Table 3shows the dose effect of ketamine on the absolute value of lambda L and lambda T. Ventricular lambda L was increased by 2.9% to 7.9%, whereas lambda T was increased by 4.5% to 8.2% after ketamine administration (Figure 4). However, this increase in lambda L and lambda T was not significant.

The present study shows that ketamine induced a significant bradycardia in isolated intact hearts and did not facilitate the occurrence of spontaneous arrhythmias. In frozen hearts, ketamine slowed ventricular epicardial longitudinal and transverse conduction in a dose-dependent manner. Anisotropic ratio was unchanged, and no use dependency was observed. Furthermore, ketamine induced a dose-dependent prolongation of VERP, with no increased dispersion of refractoriness, whereas it did not significantly modify ventricular wavelength. Finally, no arrhythmia could be induced by pacing.

The first aim of the study was to verify the effects of clinically relevant concentrations of ketamine on ventricular theta L, theta T, and VERP. In the clinical setting, serum concentrations of ketamine are reported to range from 100 micro Meter, which corresponds to peak concentrations reached during induction of anesthesia, and 10 micro Meter, which is the concentration obtained during maintenance of anesthesia. [31] Except for the concentration of 200 micro Meter, the other concentrations used in our study are clinically relevant. Previous studies showed that ketamine alters cardiac electrophysiologic parameters by interfering with the function of several ionic channels. In isolated guinea pig hearts, Stowe et al. [5] showed that ketamine has little electrophysiologic effect at low doses. Indeed, in this study, heart rate was significantly decreased at 50 micro Meter, and atrioventricular conduction time was slowed only at 500 micro Meter. In guinea pig and rat papillary muscles and left atrial tissues, Endou et al. [2] found that 300 micro Meter ketamine decreases the maximum upstroke velocity of action potential (V^max), which is known to be correlated with the fast inward sodium current and conduction velocity in all tissues. [32] This class 1 antiarrhythmic pattern of action was confirmed by the results of the study conducted by Hara et al. [9] on guinea pig papillary muscle. However, these authors showed that this effect differs from that of pure class 1 antiarrhythmic agents by the absence of use dependency. These two studies investigated mainly the cellular or the longitudinal conduction, whereas ventricular conduction has been shown to be anisotropic. Indeed, in normal ventricular myocardium, conduction velocity is faster when parallel (longitudinal conduction) than when perpendicular to the fibers orientation (transverse conduction) because of directional differences in effective axial resistance. [29,33,34] Although longitudinal conduction is correlated to V^maxand to the fast inward sodium current, Brugada et al. [25] found that transverse conduction is an estimate of the cell-to-cell coupling. In this regard, our study is the first to evaluate the effects of ketamine on both longitudinal and transverse conduction. Our results are in accord with previously described data, because ketamine induced a dose-dependent and a non-use-dependent slowing of epicardial conduction velocity, both in the longitudinal and the transverse directions. This corresponds with the results of Nigli et al., [35] who showed that, unlike volatile anesthetics, ketamine does not impair cell-to-cell coupling.

In the present study, ketamine induced a dose-dependent prolongation of VERP. This effect is in accordance with the known effects of ketamine on the ionic currents involved in the recovery of excitability. Endou et al. [2] found that ketamine decreases the transient outward current (I^to) in a dose-dependent manner in rat single ventricular myocytes. These authors also showed that 100 micro Meter ketamine decreases the inward rectifier potassium current (I^K1) and the delayed rectifier current (I^K) in guinea pig ventricular cells. In the study by Baum, [10] 100 micro Meter ketamine also inhibited I^K1but not I^K. The difference of effect on I^Kbetween these two studies could not be explained, because 100 micro Meter ketamine was applied to guinea pig ventricular cells in both of them. This inhibitory effect of ketamine on the potassium currents might account for a delay in repolarization and the lengthening of action potential duration. However, it must be noted that action potential duration seems to be lengthened only at high concentrations of ketamine. Napolitano et al. [22] showed that the guinea pig atrial effective refractory period was unchanged throughout a ketamine concentration range of 0–50 micro Meter. Hara et al. [9] showed that action potential duration at 90% repolarization was lengthened by 100 and 300 micro Meter ketamine in guinea pig papillary muscle. Endou et al. [2] found that action potential duration at -40 and -70 mV were also lengthened by 300 micro Meter ketamine in rat left atrium, rat papillary muscle, and guinea pig left atrium. In our study, VERP was prolonged from 50 micro Meter. Although our model does not allow the study of the ionic mechanisms of this effect, because rabbit ventricular myocytes have been shown to possess I^to, I^K1, and I^K, we can suggest that the prolongation of VERP is the result of the interference of ketamine with the early and the late phases of repolarization. In conclusion, ketamine slows ventricular conduction velocity in a dose-dependent manner, with no use dependency, and prolongs VERP.

We also aimed to evaluate the effects of ketamine on anisotropic ratio and on dispersion of the refractory period, and to evaluate the possible role of ketamine in the induction of anisotropic or dispersion-based reentry. It has been established that the anisotropic conduction in the ventricular myocardium and the effects of drugs on the anisotropic properties may facilitate the occurrence of reentrant arrhythmia around functional conduction blocks. However, anisotropic reentry occurs when anisotropy is modified by a use-dependent block [12,16,17] or by a selective or preferential slowing of conduction, or unidirectional conduction block. [25] In our study, anisotropic ratio was unchanged, suggesting that the slowing of the conduction was of similar extent in both longitudinal and transverse directions. Furthermore, there was no use dependency. Thus ketamine cannot induce ventricular arrhythmias due to anisotropic reentry, because no functional conduction block could be created. It has been shown that dispersion of refractoriness may also create conditions for reentrant arrhythmias. [14,15] The results of our study show that ketamine did not induce any increase in dispersion of the refractory period. All these facts suggest that ketamine has no direct arrhythmogenic effect. Conversely, because selective prolongation of refractoriness has antiarrhythmic effects [36] and ketamine induces a uniform prolongation of VERP, we suggest that ketamine might possess genuine ventricular antiarrhythmic properties.*.

The second aim of the present study was to evaluate the antiarrhythmic or arrhythmogenic property of ketamine. In frozen hearts, we studied changes in ventricular wavelength as a possible index of the antiarrhythmic or arrhythmogenic properties of ketamine. Ketamine tended to increase wavelength (although the difference was not statistically significant), an effect that indicates an antiarrhythmic property. These results do not correspond with those of Napolitano et al., [22] who showed that, in guinea pig atrial tissue, ketamine significantly decreased atrial lambda. However, these authors studied the atria, another tissue species, and another range of ketamine concentrations than we did in our study. This disagreement is likely explained by small differences in the ionic mechanisms for atrial versus ventricular repolarization. For example, slow-inward currents (due to slow decay of Na sup + or Ca²+ plateau current) may play a more important role in determining ventricular action potential duration and refractoriness, whereas outward K sup + currents are more important in atrium. [37] The antiarrhythmic potency of ketamine has been previously described and was accounted for, at least in part, by its inhibitory effects on sodium channels. [4,38] Our results show that the lengthening of the refractoriness, the absence of increase in its dispersion, and the homogeneous pattern of the slowing of conduction velocity with regard to anisotropy might play an important role in the antiarrhythmic property of ketamine. Further, from changes of lambda L and lambda T, we can postulate that the lengthening of VERP counteracts the effect of the slowing of conduction in longitudinal and transverse directions and prevents the occurrence of a circuit of reentry. Although the cellular and molecular grounds of this antiarrhythmic effect must be clarified, they might involve the effects of ketamine on various ionic currents on different structures of the heart.

Care must be taken before extrapolating these experimental result to the clinical setting. Although the concentrations of ketamine studied correspond to those used clinically, our results do not account for the complex interactions existing between the autonomic nervous system and the intrinsic cardiac electrophysiologic activity. Because of our choice of model, the sympathetic mediation of the cardiac effect of ketamine is not present and the hemodynamic influence and the possible cardiac pathologic conditions are ignored. It also appears that direct cardiac effects observed in the present study might be modified in vivo. In addition, the present study involves a thin epicardial layer. It is established that, from epicardium to endocardium, the ventricular wall is composed by at least three subtypes of myocardial cells with different electrophysiologic behaviors. [39–41] Thus the effects of ketamine on these different cell subtypes may vary qualitatively or quantitatively, so that the resulting effect may be unexpected based on our results. Finally, we measured VERP in five sites of epicardium. Therefore we cannot completely exclude possible VERP dispersion in other sites over the left ventricular epicardium.

Ketamine is usually selected for use as the sole or adjunct anesthetic agent in patients with acute circulatory failure. [42] The results of our study might have clinical relevance when these patients have healthy hearts, presuming that the other portions (M and endocardial regions) of the myocardium are affected similarly by ketamine. However, this cannot be extrapolated to situations of systemic imbalance such as hypoxia, hypocarbia, or electrolyte disturbances. This limitation must be extended to patients with pathologic cardiac conditions such as ischemia or dilated or hypertrophic myopathy. In all these situations, the electrophysiologic behavior of the myocardial cells is modified. For example, conduction is slowed and refractoriness is shortened in ischemic compared with normal myocardial cells, and these cells might be affected differently by ketamine, as it has been demonstrated with halothane. [43,44] Furthermore, Riou et al. [45] studied the effects of ketamine on the intrinsic mechanical properties of cardiac papillary muscle from healthy hamsters and those with cardiomyopathy. They showed that ketamine induces a positive inotropic effect on papillary muscle from the healthy hamsters, whereas this effect is markedly decreased or suppressed on cardiomyopathic muscles. Thus further studies are needed to investigate the cardiac effects of ketamine in the previously mentioned pathologic conditions.

In conclusion, using a high-resolution mapping system in rabbit hearts, the study shows that ketamine significantly slows longitudinal and transverse conduction velocity, without inducing conduction blocks at therapeutic concentrations. There was no significant use dependency, the VERP was prolonged in a spatial homogenous manner, and ventricular wavelength was not significantly modified. No dysrhythmia could be induced in cryotreated hearts with a thin surviving rim of epicardial tissue. No dysrhythmia was observed in nonfrozen hearts. The electrophysiologic effects of ketamine suggest, in the present rabbit heart model, that this agent is not arrhythmogenic and has antiarrhythmic properties on ventricles.

The authors thank Gilbert Saissi, M.D. (Department of Anesthesiology), the entire pharmacy staff (especially Patrick Chartreux and Christian Francou), and Robert Sempere (Medical School).