Actions of Phenylephrine, Isoproterenol, and Epinephrine Halothane on Endocardial Conduction and Activation in Canine Left Ventricular Papillary Muscles 

The experiments performed in this study were reviewed and approved by the Animal Care Committee of the Medical College of Wisconsin. Left ventricular papillary muscle preparations were dissected from the hearts of 30 adult mongrel dogs killed during halothane-oxygen anesthesia. The muscles, measuring about 12 x 20 mm, were pinned to the floor of a 12-ml tissue chamber and superfused at 20 ml/min with 37 [degree sign] Celsius Tyrode's solution equilibrated with 97% O²- 3% CO²of the following composition (mM): NaCl 137, KCl 4.0, CaCl²1.8, MgCl sub 2 0.5, NaH²PO⁴0.9, NaHCO³, 16, dextrose 5.5, and Na ethylenediamine-tetraacetic acid (50 micro Meter) added as an antioxidant. All preparations were stimulated orthodromically at 150 beats/min (Figure 1) using square wave constant current 2-ms pulses applied via bipolar platinum wire electrodes placed on the false tendon 1 or 2 mm proximal to its attachment at the tip of the muscle.

After equilibration for 1 h, a mapping procedure was performed to locate a PMJ using a twisted pair of 75-micro meter teflon insulated silver wire electrodes mounted perpendicular to the endocardium in a three- axis micro-manipulator to explore the muscle surface. One of the wires was cut short and positioned 2 mm above the other wire to act as a reference and reduce the large field effect of myocardial activation. [15] The monopolar extracellular signal from this surface x-y exploring electrode was amplified differentially by a custom-built multichannel alternating current (AC) amplifier, monitored on a digital oscilloscope, sampled at 10 kHz per channel by A/D conversion, stored in the memory of an Hewlett Packard (Palo Alto, CA) computer and intermittently printed. Electrograms were obtained from 80–120 sites on each muscle over about 2 h, using a 1-mm spacing between sites in a grid pattern of approximately 8 rows by 12 columns. Activation maps for the Purkinje and muscle layers were constructed, as shown in Figure 1, by measuring the intervals from onset of the stimulus artifact to the maximum negative deflections of the Purkinje (P) and muscle (V) components of each electrogram. These times on the electrogram from each site correspond to the phase 0 upstroke of Purkinje and ventricular muscle action potentials recorded at each location. [15] The criteria [17] used to select the PVJ were that the site exhibit 1) earlier activation and 2) shorter delay between P and V components relative to muscle activation in the surrounding endocardium, and 3) a uniphasic all negative V (muscle) deflection, indicating propagation away from the site in all directions through the muscle after conduction from the P layer. We also required that an intracellular microelectrode [25] advanced progressively from the surface at the PVJ encounter superficial Purkinje, characteristic double upstroke “transitional” and deeper ventricular muscle action potentials. [15,17] Where two or more sites toward the muscle origin (cardiac apex) exhibited simultaneous early V activation, the PVJ closest to the papillary muscle tip was selected.

After locating a PVJ, 3 or 4 additional 75-micro meter wire electrodes were aligned every 1 or 2 mm between the proximal false tendon attachment at the papillary muscle tip (site A in Figure 1) and at the distal (apical) PVJ (site E in Figure 1). The signals were sampled (10 kHz) after every tenth stimulus and recorded each minute during catecholamine exposures. Purkinje conduction velocity was measured between P deflections of the proximal and PVJ electrograms, using the fixed distance (5–10 mm) between the electrodes. The PVJ delay was measured to the nearest 0.2 ms from the P-V interval of the PVJ electrogram. Ventricular muscle conduction velocity was measured for propagation in the retrograde direction from the junctional V wave back to the V deflection of the proximal electrogram. Experiments in which the sequence or relative timing of P or V signals in the 5 or 6 electrograms changed in a sudden discontinuous manner indicative of a change in activation pattern during drug exposure were discarded. The velocities measured assume linear propagation between the recording sites and do not exclude possible changes in the direction of conduction, which would require development of isochronal activation maps at 1- or 2-min intervals during catecholamine exposure. Therefore, we also evaluated changes in the activation times for the Purkinje and ventricular muscle responses at the PVJ (P and V at site E) and the false tendon attachment (V at site A) relative to the earliest Purkinje response (P at site A in Figure 1) to assess conduction independent of any assumed directional effects.

Three separate groups of 10 muscles were studied to measure the effects of exposure to 5 micro Meter epinephrine, 5 micro Meter phenylephrine, or 1 micro Meter isoproterenol on conduction and activation. Two trials of exposure to one adrenergic agonist were performed in each group, once in the presence of 0.4 micro Meter halothane and once in the absence of anesthetic. After initial drug-free control measurements, 5 of the preparations in each agonist group were first equilibrated for 30 min with halothane dissolved in the superfusate from a reservoir preequilibrated with the agent by passing the CO²/O²mixture through a calibrated vaporizer. Measurements were again made with halothane just before switching to a solution containing halothane and the agonist, and measurements were made at each minute during 15 min trials of agonist exposure. Agonist washout values were obtained over 15 min (with halothane), and the anesthetic was removed by superfusion with drug-free Tyrode's solution for at least 45 min before the trial of agonist exposure in the absence of halothane. The other 5 muscles in each group were studied in the opposite order by measuring response to the agonist alone, followed by the response in the presence of halothane to balance any potential time-dependent effect in experiments lasting 6 or 7 h. Halothane concentrations in the tissue chamber were sampled in duplicate and determined by gas chromatography.

The data are reported as values of the mean +/- SEM except as indicated in the figures. The values obtained in all groups were evaluated by repeated-measures ANOVA, and within-group means were compared using the Waller-Duncan least significant difference method. [26] The actions of halothane were compared with preceding control using paired t tests. A P level <or= to 0.05 was considered significant.

The actions of 0.4 mM halothane alone on conduction after equilibration for 30 min were evaluated using values from all preparations as summarized in Table 1. The effects of halothane were relatively greater at the PVJ, which exhibited increased (5.0%) conduction delay, than in the ventricular muscle or Purkinje fiber layers, which exhibited smaller insignificant velocity decreases of 3.9% and 1.6%, respectively. Halothane slightly delayed activation of muscle at the PVJ and delayed activation of muscle at the false tendon attachment by about 4%.

The effects of beta- and alpha¹-adrenoreceptor activation are illustrated in Figure 2and Figure 3and those of epinephrine in Figure 4. In each figure, the error bars for the values at each time (the least significant difference [LSD] value) are similar to a confidence interval such that mean values that do not exhibit overlapping errors differ significantly (P <or= to 0.05). The characters in each figure indicate the time of onset of significant change from the time 0 control value just before agonist exposure without or with halothane. Isoproterenol alone (solid lines in Figure 2) progressively but modestly increased Purkinje fiber conduction velocity from a preceding control value of 1.93 +/- 0.14 m/s (at time 0) by a maximum of 6% to 2.05 +/- 0.14 m/s (P <or= to 0.01) at 5 min of exposure. The conduction time across the PVJ was rapidly (by 1 min) and markedly (by up to 18% at 3 min) shortened from 3.07 +/- 0.33 ms to 2.52 +/- 0.25 ms (P <or= to 0.01) after exposure to the beta-agonist in concert with a substantial increase (by up to 21%) of conduction velocity in the underlying ventricular muscle, from 0.88 +/- 0.06 m/s to 1.07 +/- 0.09 m/s (P <or= to 0.01) at 2 min. The actions of isoproterenol in the presence of 0.4 mM halothane (dashed lines in Figure 2), including the increases of Purkinje and ventricular muscle velocities and the reduction of PVJ delay, did not differ from those of the beta-agonist in the absence of anesthetic.

The alpha¹-agonist phenylephrine, in the absence of halothane (solid lines in Figure 3), did not significantly alter either Purkinje or ventricular muscle conduction velocity. Phenylephrine alone decreased the PVJ conduction time in a slow progressive manner by a maximum of 7% at 15 min, from 3.60 +/- 0.37 ms to 3.33 +/- 0.36 ms (P <or= to 0.05). On the other hand, phenylephrine with halothane rapidly but transiently decreased Purkinje velocity by up to 12%, from the control value of 1.92 +/- 0.14 m/s (at time 0) with halothane to a minimum of 1.69 +/- 014 m/s (P <or= to 0.01) at 4-min exposure, without any change of conduction in the underlying muscle layer. Phenylephrine tended to increase the PVJ conduction time with halothane from the third to fifth min, but this change (+ 2%) was not significant. Thereafter, the PVJ delay decreased (P <or= to 0.01) to 3.51 +/- 0.38 ms at 15-min phenylephrine exposure, by 8% relative to the control of 3.81 +/- 0.40 ms. The conduction velocity in the Purkinje fiber layer was lower (P <or= to 0.01) from 3–10 min of exposure to phenylephrine with halothane than without halothane, and the transient decrease at maximum phenylephrine effect with halothane (-0.22 +/- 0.06 m/s at 4 min, -12% vs. time 0) was different (P <or= to 0.01) from the change (-0.01 +/- 0.03 m/s, -1% vs. time 0) in the absence of anesthetic.

Epinephrine alone (solid lines in Figure 4) rapidly and transiently decreased conduction velocity in the Purkinje fiber layer by up to 11% at the time (5 min) of maximum effect, from 1.83 +/- 0.12 m/s (time 0) to 1.64 +/- 0.13 m/s (P <or= to 0.01). Purkinje velocity increased (P <or= to 0.05) from this minimum value to 1.72 +/- 0.13 m/s by 15 min, a value greater than that at the nadir of effect but less than the control value before epinephrine exposure. The junctional conduction time decreased by 3 min after epinephrine exposure and by as much as 10%, from 4.38 +/- 0.58 ms to 3.97 +/- 0.51 ms (P <or= to 0.01), at 15-min epinephrine exposure. This improvement in junctional conduction occurred in concert with rapid (by 2 min) increase of muscle velocity by up to 35%(from 0.76 +/- 0.07 m/s to 1.03 +/- 0.14 m/s (P <or= to 0.01)) at 6 min. The increase of muscle velocity produced by epinephrine alone did not persist, and velocity decreased significantly to 0.86 +/- 0.09 m/s after 15 min. Epinephrine in the presence of halothane (dashed lines in Figure 4) rapidly (by 3 min) and markedly (by 22%) decreased Purkinje fiber conduction velocity from 1.80 +/- 0.09 m/s (at time 0 with halothane) to 1.39 +/- 0.10 m/s (P <or= to 0.01) after 5 min, and thereafter the velocity increased toward control (to 1.55 +/- 0.12 m/s) at 15 min despite continuing epinephrine exposure with halothane. Epinephrine progressively decreased PVJ conduction time in the presence of halothane from 4.26 +/- 0.52 ms to 3.92 +/- 0.44 ms at 10 min (P <or= to 0.01). Simultaneously epinephrine increased conduction velocity in ventricular muscle (within 2 min) from 0.70 +/- 0.05 m/s to 0.93 +/- 0.10 m/s (+ 32%, P <or= to 0.01) at 6 min. Compared with the actions of epinephrine in the absence of halothane, the Purkinje conduction velocities were lower (P <or= to 0.01) from the third min throughout exposure to epinephrine with halothane. In addition, the depression of Purkinje conduction velocity at maximum epinephrine effect with halothane (-0.40 +/- 0.08 m/s at 5 min, -22% vs. time 0) was larger (P <or= to 0.01) than the change (-0.19 +/- 0.07 m/s, -11% vs. time 0) without halothane. The PVJ conduction time was decreased by epinephrine about the same with and without halothane. The time-dependent increases of ventricular muscle conduction velocity produced by epinephrine were shifted to lower values (P <or= to 0.05 at all times) in the presence of the anesthetic. However, the maximum increase (32%) produced by epinephrine with halothane (+0.23 +/- 0.06 m/s) did not differ significantly from that found (35%) without halothane (0.27 +/- 0.09 m/s).

The actions of the adrenergic agonists on papillary muscle activation times in the presence of halothane are illustrated in Figure 5. Isoproterenol shortened activation times at each site (P <or= to 0.05) about the same in the presence and absence of halothane (data not shown). In contrast, phenylephrine with halothane delayed activation of the Purkinje fiber layer and muscle at the PVJ (P <or= to 0.05) and increased activation time toward the tip of the papillary muscle at the false tendon attachment (site A, Figure 1). Phenylephrine did not increase activation time at any site in the absence of halothane. The actions of epinephrine with halothane delayed (P <or= to 0.01) activation of the Purkinje and muscle layers at the PVJ and simultaneously shortened activation time (P <or= to 0.05) measured at the false tendon attachment site. The delays of activation at the PVJ with halothane were larger (P <or= to 0.05) than the changes without halothane (27% vs. 13% for Purkinje and 11% vs. 3% for muscle at 5 min epinephrine exposure, with and without halothane, respectively). The shortening of activation time for muscle at the false tendon site with halothane (-6% at 5 min) was not different from that without halothane (-9% at 5 min).

The present study demonstrates important tissue (Purkinje vs. myocardial) and anesthetic (halothane) dependent differences in the time course of modulation of ventricular conduction by adrenergic agonists that may potentially influence the occurrence of reentrant dysrhythmias. The results confirm by an independent method previous findings that halothane potentiates transient alpha¹-adrenoreceptor-mediated negative dromotropic effects of epinephrine on canine Purkinje fibers. [3] On the other hand, we found that epinephrine, like isoproterenol, simultaneously facilitates conduction at the PVJs and in the endocardium to about the same degree regardless of the presence or absence of halothane.

The sequential pattern of Purkinje fiber, PVJ, and ventricular activation observed was similar to that reported by others [15,16,20,21] in canine papillary muscles but may not be representative of endocardial activation in all regions of the heart. [16] The control values of Purkinje and muscle velocities and the junctional delay observed also were comparable with those reported from other laboratories. The importance of conduction delay at the PV junctions is that these variably distributed sites represent a heterogeneous resistive barrier to propagation from the Purkinje to the muscle layers that is particularly susceptible to block by agents that produce cell-to-cell uncoupling (octanol, halothane)[20,22,23] or reduce excitability (Ca sup ++, antidysrhythmics, hypoxia, and hyperkalemia). [15,27–29] The junctions are sites of extremely slow conduction in the ventricles, and some sites normally exhibit a variety of unidirectional block that permits reentrant excitation. [20,21] There were large differences in the mean junctional delay between agonist groups studied at different times, which may reflect persistence of anesthetic or adrenergic effects in some preparations that were not completely reversible. We elected not to use epinephrine in combination with adrenergic antagonists (metoprolol, prazosin) to evaluate effects of alpha¹- and beta-adrenoreceptor activation because the actions of these agents on PV junctional delay are largely unknown. Therefore, it is possible that the responses to isoproterenol and phenylephrine observed may have been influenced to some degree by simultaneous effects mediated by beta²- or alpha²- and beta sub 1 -adrenoreceptor, respectively. The methods used cannot assess potential drug effects on anisotropic properties of the Purkinje or muscle layers, which may exhibit important differences in conduction velocity responses to uncoupling agents and adrenergic agonists measured in the transverse compared with longitudinal directions relative to the axis of muscle fibers. [18,30] However, the changes of activation times at each site were consistent with the effects of each agonist on the conduction velocities measured, assuming linear propagation, and we did not observe any large changes in the activation sequence at multiple fixed recording sites. The results are limited to orthodromically conducted nonpremature impulses at one rapid pacing rate in superfused preparations and do not necessarily indicate how alpha¹- and beta-adrenoceptor activation may influence retrograde propagation of a potential reentrant impulse from muscle into the Purkinje system.

Joyner and Overholt [20] first showed that 0.2 mM octanol, a potent volatile anesthetic (MAC about 0.06 mM) and cellular uncoupling agent, produced large increases (82%) in PVJ delay but decreased conduction velocity in the muscle layer by only 10%. Ozaki et al. [31] reported that halothane, unlike lidocaine or tetrodotoxin, depressed myocardial conduction velocity in guinea pig papillary muscles (by 40–60% at 1.5 MAC and 33 [degree sign] Celsius), with little decrease in the rate of phase 0 depolarization (V^max), and suggested that the mechanism responsible for conduction slowing may involve actions other than inhibition of the fast sodium current. Freeman and Muir [22] found that the depression of conduction by 3 vol% halothane (1.15 mM at 37 [degree sign] Celsius) at the PVJs was more pronounced than in linear (false tendon) Purkinje fibers and suggested that inhibition of cell-to-cell coupling by halothane may contribute to the occurrence of dysrhythmias. Our findings that 0.4 mM halothane (at 37 [degree sign] Celsius) produced only slight (5%) increases of mean PVJ conduction time, with less depression in Purkinje and muscle fibers, indicate that clinically relevant concentrations of halothane have only modest effects on junctional conduction. In contrast, the actions of halothane on conduction were relatively modest compared with the larger time-dependent changes produced by epinephrine, isoproterenol, and phenylephrine in the Purkinje and muscle layers and at the PVJs.

The findings that epinephrine transiently decreased conduction velocity in papillary muscle Purkinje fibers (-11% at 4 min) without halothane exposure differ from our previous studies in canine false tendon fibers, in which 5 micro Meter epinephrine produced about a 5% decrease of velocity within 3–5 min after epinephrine exposure. [2,32] Joyner et al. [20] suggested that the slower velocity in the papillary muscle Purkinje layer (1.8 m/s), relative to that in free-running linear Purkinje fibers (about 2 m/s in false tendons), is a result of flow of current through the PVJs to the muscle layer “sink.” We did not simultaneously measure velocities in the two types of tissue to determine whether the relatively large negative dromotropic effect of epinephrine alone, a possible proarrhythmic action of epinephrine in the absence of halothane, reflects a real difference between the responses of false tendon and papillary muscle Purkinje fibers. In the present study, phenylephrine alone (5 micro Meter) did not slow Purkinje conduction velocity, although this agonist may slightly decrease velocity [3] in false tendon fibers at high (> 5 micro Meter) concentration (unpublished observations). Nevertheless, the findings that the maximum decreases of velocity with either epinephrine (-22%) or phenylephrine (-12%) in the presence of 0.4 mM halothane were significantly larger than in the absence of anesthetic (-11% and +1%, respectively) confirm that halothane potentiates this transient alpha¹-mediated negative dromotropic effect of catecholamines.

Few studies examine the time course of effects of beta-adrenoceptor activation on conduction in intact ventricular tissues. Recently, Munger et al. [14] examined the steady-state voltage dependence of beta-adrenergic modulation of Na sup + current using elevated extracellular K sup + to depolarize false tendon Purkinje fibers. They found that 1 micro Meter isoproterenol increased velocity about 5%(at 5.4 mM K sup +) without changing V^max, although isoproterenol decreased velocity and V^maxin fibers depolarized by high extracellular K sup +(-18% at 12 mM K sup + and -65 mV). Facilitation of conduction by the beta-agonist may be related to increased Na sup + current [33] but is more likely the result of increased cell-to-cell coupling. [13,14,34,35] The speeding of conduction, we observed after beta-activation (Figure 2) occurred within minutes of exposure to isoproterenol in Purkinje and muscle fibers and at the PVJ. This action is consistent with reports showing enhanced cell-to-cell coupling within a few minutes after exposure of ventricular tissues to isoproterenol or intracellular injection of cAMP or ATP. [13,34,36] On the other hand, isoproterenol can depress conduction in depolarized Purkinje fibers by reducing Na sup + channel availability and phase 0 Na sup + current. [14,37] We found no evidence for beta-mediated depression of conduction in our experiments on papillary muscles at normokalemic (4 mM) K sup + ion concentration although beta-adrenoreceptor activation can depress conduction in epicardial fibers at a higher (8 mM) K sup + concentration. [30] The actions of isoproterenol speeding conduction in Purkinje fibers, at the junctions and in endocardial muscle, were similar in the absence or presence of halothane and in contrast to the potentiation of alpha¹-mediated depression of conduction in Purkinje fibers exposed to halothane by the mixed alpha¹- and beta-agonist epinephrine. The effects of halothane on beta-adrenoreceptor-mediated cardiac responses are controversial and have been reported to produce right- and leftward shifts in the isoproterenol-contractile response curve. [38,39] Our preliminary studies suggest that halothane may attenuate isoproterenol-induced conduction depression in canine epicardial muscle preparations. [40]

The results do not directly address the subcellular mechanisms by which alpha¹-activation depresses Purkinje conduction velocity or how halothane potentiates this adrenergic effect. It was suggested that the alpha¹-mediated effect may involve modulation of peak Na sup + current or, more likely, depression of cell-to-cell coupling because the large transient velocity decreases were not associated with decreased rate of phase 0 depolarization (V^max), an indirect measure of peak Na sup + current. [2–5] It was anticipated that the PVJs may be susceptible to block after alpha¹-activation given their relative sensitivity to cellular uncoupling agents (e.g., octanol). [20] However, junctional delay was decreased by epinephrine, in the absence and presence of halothane, probably because of actions of epinephrine on beta-receptors, whereas 5 micro Meter phenylephrine only tended to produce longer PVJ delay with halothane (Figure 3). The findings do not necessarily exclude the possibility that alpha¹-activation in the presence of halothane may block junctional conduction at higher concentrations or at sites exhibiting less than average cell-to-cell coupling. Some PVJs exhibit unidirectional block without pharmacologic depressants, [21] and alpha sub 1 -activation could modulate the number of junctions exhibiting marginal bidirectional conduction. Nevertheless, the results indicate that alpha¹-mediated conduction depression occurs only in Purkinje and not in endocardial muscle fibers and that there are important largely unknown differences in the alpha¹-adrenoreceptor effector mechanisms between the two tissues. Such differences may involve the relative densities of the three pharmacologically defined subtypes of alpha¹-receptors in the heart, [41] differences in the G-protein alpha or beta subunits [42] linking receptors to various isoforms of phospholipase C, the potential subcellular actions of the diacylglycerol and inositol trisphosphate second messengers produced, and differences between the connexin (Cx) proteins forming Purkinje (Cx40) and myocardial (Cx43) gap junctions. [43]

In conclusion, the findings give some insight into the possible changes in ventricular conduction that might occur in the heart suddenly exposed for a few minutes to “suprathreshold” arrhythmogenic concentrations of epinephrine. Such changes would likely include rapid and persistent beta-adrenoreceptor-mediated speeding of conduction at the junctions and in the myocardium in association with transient alpha¹-mediated depression of conduction in the Purkinje system, the latter potentiated by halothane. At the time (3–5 min) of maximum alpha¹-effect, the different Purkinje and myocardial responses may promote early activation of the interventricular septum and delay activation of the free walls and cardiac base, regions normally activated by short and long pathways, respectively, through the conduction system. Such activation changes may accentuate differences between the repolarization times of myocardial and Purkinje fibers in different regions of the ventricles, adding heterogeneity of repolarization to disparate conduction characteristics. This heterogeneity should facilitate reentrant excitation by ectopic or applied impulses originating in the tissue (the myocardium) having the shorter repolarization time. [44,45] An initiating impulse, perhaps resulting from beta-adrenoreceptor-stimulated automaticity or triggered activity, would be required to initiate reentrant tachycardia by this mechanism during epinephrine challenge with halothane anesthesia. This speculation is supported by older yet unexplained reports showing that epinephrine infusions (in pentobarbital anesthetized dogs) initially increase (at 2 or 3 min) and then decrease (5–15 min) vulnerability to induction of ventricular fibrillation by single premature stimuli applied to the myocardium. [46,47] In addition, some patients with exercise-related ventricular tachycardia are known to exhibit inducible dysrhythmias by programmed stimulation during epinephrine infusions but not during infusions of isoproterenol or phenylephrine alone. [48] Our findings suggest that other “sensitizing” anesthetics, such as thiopental or propofol, [49] may potentiate transient alpha¹-mediated depression of conduction in the His-Purkinje system during epinephrine infusions in electrophysiologic studies of patients with normal hearts and a history of sudden cardiac death.