Effects of Hypothermia, Potassium, and Verapamil on the Action Potential Characteristics of Canine Cardiac Purkinje Fibers 

This study was approved by the Medical College of Wisconsin Animal Care Committee and conformed with standards set forth in the Guide for Care and Use of Laboratory Animals.*

Adult mongrel dogs (10–22 kg, of either sex)(n = 45) were anesthetized with 30 mg/kg pentobarbital sodium. The heart was quickly excised, and the anterior false tendon with attached papillary muscle from the left ventricle was removed and immersed in modified Krebs' solution (22 degrees Celsius) equilibrated with 97% Oxygen²and 3% CO². Small (< 1-cm²) preparations with free-running strands of Purkinje fibers were dissected from this tissue and pinned to the silicone elastomere floor of a 2-ml chamber and superfused at a rate of 4 ml/min with modified Krebs' solution (37 degrees Celsius) containing 2.3, 3.9, or 6.8 mM KCl with or without 1 micro Meter verapamil and equilibrated with a 97% Oxygen²-3% CO²gas mixture. The millimolar composition of Krebs' solution was 137 NaCl, 12 NaHCO³, 1.8 NaH²PO⁴, 1.8 CaCl², 0.5 MgCl², 5.5 glucose, and 0.05 EDTA, with a pH of 7.4.

Each preparation was stimulated at a constant rate (1 Hz) with the use of bipolar-silver wire endocardial surface electrodes. The stimuli were square-wave pulses lasting 2 ms at 1.5 times threshold. Transmembrane action potentials were recorded with conventional microelectrode techniques. Action potential changes stabilized within 5–8 min. Glass microelectrodes (15–30-M Omega resistance) were coupled by Silver-AgCl wire to a preamplifier (World Precision Instruments, New Haven, CT). Action potential signals were recorded on frequency-modulated tape (AR Vetter, Rebersburg, PA) for later analysis of maximum diastolic potential (MDP), maximum rate of phase 0 depolarization (V^max), and APD at 50%(APD⁵⁰) and at 95%(APD sub 95) repolarization. These values were displayed and measured electronically directly off the digital oscilloscope (Nicolet 310). The V^maxwas determined with a differentiator exhihiting a linear response from 100–1,000 V/s. The “zero” potential was obtained at the beginning and at the end of the experiments by withdrawing the microelectrode from the inside of the fiber. Because the ground connection between the bath and the circuit was made through a direct Silver-AgCl connection, the change in temperature may slightly influence the half cell potential in the bath, and this change will sum with the real MDP changes. To verify the significance of this phenomenon we performed additional experiments (n = 8) in which the microelectrode was placed into the bath, the temperature was changed from 37 degrees Celsius to 28 degrees Celsius, and the Silver-AgCl bath junction potential change was measured. The half cell potential caused a hyperpolarization to be measured by the microelectrode over this temperature range by -2.1 plus/minus 0.02 mV and -4.2 plus/minus 0.05 mV at 32 degrees Celsius and 28 degrees Celsius respectively. All measured MDP values at 32 degrees Celsius and 28 degrees Celsius were corrected for the above differences, by subtracting the temperature-induced bath potential from the actual measured potential, and all the statistical analyses were performed with these corrected values.

The tissue bath was surrounded by a thermostatically controlled water bath, maintained at a constant temperature of 37 plus/minus 0.02 degrees Celsius. The low bath superfusate temperatures (32 and 28 plus/minus 0.5 degrees Celsius) were attained by readjusting the setting of the thermostat. The temperature in the tissue bath was gradually decreased from 37 degrees Celsius to 25 degrees Celsius over a 20–30-min period. The temperature of the solution was measured by a small, rapidly responding, custom made thermistor probe placed less than 2 mm from the preparation. The fluid level in the chamber was kept at a constant height (4 mm) by continuous suction.

The preparations were allowed to equilibrate for about 1 h. To determine the effects of hypothermia on action potential characteristics, action potentials were recorded at 37 degrees Celsius, 32 degrees Celsius, and 28 degrees Celsius (plus/minus 0.5 degrees Celsius). To determine the effects of various [Potassium sup +]^oon action potentials, experiments were performed with 2.3, 3.9, and 6.8 mM [Potassium sup +] in the superfusate. Verapamil (Sigma, St. Louis, MO) prepared as stock solution (100 micro Meter) was added to measured volumes to achieve the desired 1 micro Meter concentration in the superfusate. The same set of action potential measurements were performed with 1 micro Meter verapamil in the superfusate. Tissues were exposed to each [Potassium sup +]^oand to verapamil (1 micro Meter) for 20 min before measurement of action potential characteristics.

Data are expressed as means plus/minus standard error of the mean. Statistical analysis was performed by paired and unpaired t tests and with one-way analysis of variance (analysis of variance repeated measures and factorial analysis), as appropriate, with P < 0.05 considered statistically significant.

Typical effects of hypothermia on the action potential in canine Purkinje fibers are illustrated in the upper panel of Figure 1. Each preparation was stimulated at a constant rate of 1 Hz. Recordings were taken from the same cell at various temperatures at a [Potassium sup +]^oof 3.9 mM. Hypothermia significantly increased APD. The bottom panel of Figure 1shows the changes of action potential in hypothermia in the presence of 1 micro Meter verapamil in the superfusate. In the presence of verapamil the Purkinje fiber's action potential is shorter than at the same temperature without the drug (dashed action potential was recorded at 37 degrees Celsius in the absence of verapamil). The slope of phase 2 repolarization is increased by verapamil, a finding that is consistent with the blockade of a slow inward current such as that carried by a Calcium²+ ion. Hypothermia, on the other hand, appears to decrease the slope of phase 2, suggesting that low temperature affects the Calcium²+-dependent ionic mechanisms in opposite direction.

The effects of two stages of hypothermia (32 degrees Celsius and 28 degrees Celsius) on the MDPs of Purkinje fibers were recorded during the exposure to low (2.3 mM), normal (3.9 mM), and high (6.8 mM)[Potassium sup +]^oin the superfusate (Figure 2). MDPs were higher (hyperpolarized) at low, and lower (depolarized) at high [Potassium sup +]^othan at normal [Potassium sup +]^o(P < 0.05), regardless of the temperature. During the 20–30 min of gradual cooling from 37 degrees Celsius to 28 degrees Celsius, MDP decreased at normal and high [Potassium sup +]^o(P < 0.05), and did not change at low [K sup +]^o.

At 37 degrees Celsius and in the presence of verapamil MDP was lower (difference not significant, P > 0.05 vs. control;Figure 2). In hypothermia and at [K sup +]^ogreater or equal to 3.9 mM loss of membrane potential was parallel to this of control Purkinje fibers, and reached significant depolarization at 28 degrees Celsius (P < 0.05). At [K sup +]^o2.3 mM there was no effect of temperature on MDP in verapamil-superfused Purkinje fibers.

The effects of hypothermia, Potassium sup + variation, and verapamil on V^maxare summarized in Figure 3. Hypothermia decreased V^maxat each [K sup +]^o(P < 0.05). Despite significant differences in MDP at various [K sup +]^oat 37 degrees Celsius (Figure 2), the respective V^maxvalues were not different (Figure 3), although there were trends for V^maxto be lower at 6.8 and higher at 2.3 mM [K sup +]^othan at 3.9 mM [K sup +]^o(P > 0.05). At [K sup +]^oless or equal to 3.9 mM with verapamil in the superfusate no additional effect, besides that of temperature, was noted on V^max(Figure 3). At 6.8 mM [K sup +]^oin hypothermia, V^maxwas lower in the presence of verapamil (P < 0.05).

To quantify the effects of changing MDP on V^max, we examined eight additional Purkinje fiber preparations by gradually increasing [K sup +]^oin the superfusate from 0.8 to 10 mM; MDP and V^maxvalues were measured at 37 degrees Celsius and at 30 degrees Celsius (Figure 4). At both temperatures MDP decreased (less negative) directly with increasing [K sup +]^obetween 2.3 and 10 mM (P < 0.05); below 2.3 mM MDP did not further change (P > 0.05). Only between 3.9 and 8.5 mM [K sup +]^oMDP was lower at 30 degrees Celsius (P < 0.05), V^maxdecreased moderately between 2.3 and 6.8 mM [K sup +] sub o, followed by a rapid decrease when [K sup +]^oincreased above 6.8 mM (MDP less or equal to -80 mV)(Figure 4). Compared with 2.3 mM [K sup +]^o, at 0.8 mM [K sup +]^othere was a significant decrease in V^maxat 37 degrees Celsius, and no change in V^maxat 30 degrees Celsius. By varying the [K sup +]^o, V^maxat 37 degrees Celsius and 30 degrees Celsius resulted in similar pattern of changes, yet amplitudes were different.

Hypothermia increased early (APD⁵⁰) and late (APD⁹⁵) stages of repolarization (Table 1and Table 2). Figure 5and Figure 6compare the relative changes in APD⁵⁰and APD⁹⁵at various temperatures and [K sup +]^owith and without verapamil in the superfusate. In hypothermia the increase in APD⁵⁰(at 28 degrees Celsius) and APD⁹⁵(at 32 degrees Celsius and 28 degrees Celsius) was inversely related to [K sup +]^o; at 6.8 mM [K sup +] sub o in hypothermia APD⁵⁰and APD⁹⁵were always closest to normothermic control. Regardless of temperature, verapamil did not affect APD⁵⁰or APD⁹⁵at 2.3 mM [K sup +]^o. At 28 degrees Celsius and [K sup +]^ogreater or equal 3.9 mM verapamil significantly shortened the APD⁵⁰(Figure 5) and the APD⁹⁵(Figure 6). At 28 degrees Celsius and 6.8 mM [K sup +]^owith verapamil, the APD⁵⁰was 21% shorter whereas the APD⁹⁵was 12% longer than normothermic control. Without the verapamil at 28 degrees Celsius, there were no longer relative differences between the increases in APD⁵⁰and APD⁹⁵, regardless of the [K sup +]^o, indicating that hypothermia prolonged early and late repolarization similarly (Figure 7). After verapamil was introduced to the superfusate, the relative changes in APD became dependent on the actual [K sup +]^o; at 2.3 mM [K sup +] APD⁵⁰and APD⁹⁵were unchanged (Figure 7(A)); at 3.9 mM [K sup +]^othe primary effect was on APD⁵⁰(Figure 7(B)); at 6.8 mM APD⁵⁰and APD⁹⁵decreased, with APD⁵⁰> APD⁹⁵(P < 0.05)(Figure 7(C)).

Induced hypothermia is commonly used to protect myocardium and other tissues against ischemia during cardiopulmonary bypass. [1] At the same time hypothermia is known for its arrhythmogenic potential, including ventricular fibrillation. [1–3] The common mechanisms underlying cardiac arrhythmias are the loss of the MDP in Purkinje fibers, [5] diastolic Calcium²+ overload, [16] and altered dispersion of repolarization, [11] all being reported in hypothermia. [4,13,14,20,21] Hypokalemia has been encountered during hypothermia, [6,7] and rebound hyperkalemia has occurred during rewarming. [7] This study examined electrophysiologic alterations of MDP, V^max, and APD during hypothermia and hypothermic hypokalemia and their reversibility after verapamil treatment and Potassium sup + supplementation.

In the current study, the shift in MDP caused by variation of [K sup +]^obetween 2.3 and 10 mM was consistent with values predicted by the Nernst equation [14]: depolarization was present at high, and hyperpolarization at low [K sup +]^o(Figure 2and Figure 4). Below [K²+]^o2.3 mM, specifically at 0.8 mM, MDP did not further hyperpolarize, but also did not depolarize as could be expected. [14] We do not have a plausible explanation for the disparity between ours and previously reported findings. [14].

We found moderate loss of membrane potential (depolarization) with reduction of temperature to 28 degrees Celsius but only at [K sup +]^ogreater or equal to 3.9 mM. A very small decrease in the resting membrane potential of canine Purkinje fibers was described at 25 degrees Celsius, [13] but this decrease was considerably larger below 20 degrees Celsius. [4] The decreases in MDP in hypothermia have been attributed to inhibition of adenosine triphosphate-dependent Sodium sup +-Potassium sup + pump. [22–24] In our study the Calcium²⁴-channel blockade with 1 micro Meter verapamil had no significant additional effect on MDP regardless of the temperature, and when [K sup +]^owas greater or equal to 3.9 mM moderate depolarization at 28 degrees Celsius was parallel to that of control Purkinje fibers. This tendency toward depolarization when [Potassium sup +]^owas greater or equal to 3.9 mM may not be incidental, because verapamil is known to decrease MDP at higher concentrations. [25,26] Dersham and Han described a nonsignificant decrease in MDP after 1 micro Meter verapamil, and a statistically significant decrease (from 86 to 80 mM) after 2 micro Meter. [9] Similar results after 2 micro Meter of verapamil were reported by Amerini et al. [19] The mechanism responsible for MDP decrease after verapamil was attributed to the reduced Potassium sup + conductance, [19,27] but we cannot confirm this mechanism because the increase in K sup + conductance by increasing [Potassium sup +][8] in our study did not reduce the amount of depolarization induced by verapamil. It is unlikely that small changes in MDP, present at normal serum [Potassium sup +]([nearly equal] 3.9 mM), have any clinical significance. Our findings suggest that hypothermic cardiac arrhythmias, which occur around 28 degrees Celsius, cannot be attributed to hypothermia-induced membrane depolarization, providing serum Potassium sup + is in normal range.

The magnitude of V^maxis determined by the fast inward Sodium sup + current (I^Na), and has traditionally been used as an index of Sodium sup + channel availability. [28,29] In our experiments, hypothermia decreased V^max(Figure 3), which may be explained by the temperature-mediated decrease in I^Na. [21,30] Hypothermia decreases Sodium sup +-Potassium sup + adenosine triphosphatase activity, which then results in increased [Sodium sup +] sub j and decrease in driving force for I^Na. [31–33] At [Potassium sup +]^oless or equal to 3.9 mM in the normothermic and hypothermic states, Calcium²+-channel blockade with 1 micro Meter verapamil did not alter V^max. Although verapamil in lower concentrations has no significant effect on V^max, [9,25,34,35] the Calcium²+-channel blockade may decrease V^maxwhen given in higher concentrations. [30,37] However, at 6.8 mM [Potassium sup +]^oin hypothermia, we found that verapamil decreased V^max(P < 0.05)(Figure 3). It is possible that the cumulative effect of hypothermic I^Nablockade, low MDP caused by hyperkalemia, and verapamil-induced inhibition of I^Na, [36,37] possibly potentiated by low temperature, resulted in a reduction of V^max.

Because V^maxis voltage-dependent we examined the relation between MDP and V^maxby varying the [Potassium sup +]^obetween 0.8 and 10 mM. Whereas the increase in [Potassium sup +]^obetween 2.3 and 10 mM caused an almost linear decrease in MDP (depolarization), at less than 2.3 mM, MDP ceased to decrease further, consistent with a decrease in resting Potassium sup + conductance at low [Potassium sup +]^o. [38] The decrease in temperature from 37 degrees Celsius to 30 degrees Celsius induced a leftward shift of the V sub max curve. Although there was a decreasing trend, V^maxdid not significantly change when [Potassium sup +]^owas increased from 2.3 to approximately 6 mM, a finding that is consistent with the results described in Figure 3. Therefore, V^maxwas not greatly affected when the voltage (MDP) was between -100 and -85 mV (2.3–6 mM [Potassium sup +]^o), but was markedly decreased when MDP became less negative than -80 mV. In extreme hypokalemia (0.8 mM) at 37 degrees Celsius V^max, significantly decreased, at least in part mirroring changes in the respective MDP. Our data suggest that during phase 0 of the stimulated action potential at MDP between -80 and -100 mV, the fraction of available Sodium sup + channels remains relatively constant or only slightly increases as MDP becomes more negative. Thus, when MDP reaches -80 mV, the near-maximum capacity for Sodium sup + influx through the fast Sodium sup + channels may be reached (especially in hypothermia) and cannot be further increased by more negative MDP. The decrease in V sub max when MDP drops below approximately -72 mV at 37 degrees Celsius and -80 mV at 30 degrees Celsius, is very abrupt. In addition, hypothermia significantly decreased V^maxand less affected MDP. These findings are consistent with the existence of two mechanisms that may effect Sodium sup +-channel kinetics: first, a temperature-dependent mechanism that is responsible for the decrease in V^maxduring cooling, and second, a voltage-dependent mechanism that is active only at MDP less negative than -80 mV. In deep hypothermia and at [Potassium sup +]^o6.8 mM, verapamil further decreased V^max, most likely by the interference with I^Nakinetics by voltage-dependent mechanism. The clinical significance of the decrease in V^maxby verapamil and [Potassium sup +]^oin hypothermia has to be studied, but typically the drugs that are able to decrease the I sub Na [39] induce the following chain of events: decrease in [Sodium sup +]ⁱ, increase in Calcium²+ efflux and therefore a decrease in sarcoplasmic reticulum Calcium²+ loading. These events may ultimately result in an antiarrhythmic effect of the drugs by decreasing oscillatory afterpotentials. [40].

The cardiac action potential plateau is determined by a delicate balance among several distinct time- and voltage-dependent ionic currents, [8,38,41–43] After upstroke of the action potential caused by the influx of Sodium sup +, the I^Nabecomes inactivated. This inactivation does not lead to immediate repolarization, because the initial depolarization reduces the inwardly rectifying Potassium sup + conductance and opens voltage-gated Calcium²+ channels thus permitting the intracellular influx of Calcium²+ by the I^Na. During the action potential plateau, while membrane conductance for all ions is reduced, several currents help to maintain the transmembrane potential at around 0 mV, including I^Ca, Sodium sup +“window.” Cl and inward (anomalous) rectifying Potassium sup + currents. In addition, currents produced by the electrogenic Sodium sup +-Potassium sup + pump and Sodium sup +- Calcium²+ exchange mechanism help to maintain the action potential plateau. The outward (delayed) rectifying Potassium sup + current (I^K) and the inward rectifying Potassium sup + current are responsible for final rapid repolarization (phase 3). Details of the ionic basis of action potential are reviewed elsewhere. [5,16,23,27,38,41,45].

The electrophysiologic mechanisms responsible for the marked prolongation of APD seen in hypothermia are not fully understood, although mechanisms based on temperature dependence of I^K[46] and I^Ca[47] have been proposed. One important mechanism may be the hypothermia-induced reduction of I^K, which is a major contributor to membrane repolarization. [46] In addition, lengthened APD in hypothermia may be attributed to delayed inactivation of the I sub Ca sup +[47,48] which will maintain the I^Cafor a longer time. [20] Finally, prolonged APD may be also attributed to increase in [Calcium²+]ⁱthrough altered Sodium sup +-Calcium²+ exchange. [33] The activity of the Sodium sup +-Potassium sup + pump is likely to be reduced in hypothermia [31,32] resulting in a net rise in [Sodium sup +]ⁱ. [33] The increase in [Sodium sup +]^jmay promote Calcium²+ influx during the plateau of the action potential or reduce Calcium²+ efflux during diastole by Sodium sup +-Calcium²+ exchange. [49] Although the activity of the Sodium sup +-Calcium²+ exchanger is temperature dependent, its relatively low Q¹⁰(the rate of exchange produced by changing the temperature 10 degrees Celsius) means that this mechanism may contribute to raising diastolic free [Calcium²+]^j. [44].

We have demonstrated that hypothermia equally lengthens the APD⁵⁰and APD⁹⁵, indicating that low temperature similarly affects the ionic mechanisms that determine the early and late stages of repolarization (Table 1and Table 2and Figure 7), i.e. earlier discussed I^Kand I^Ca. The finding that hypokalemia, similarly to hypothermia, prolongs the APD is not surprising, because low [Potassium sup +]^odecreases Potassium sup + conductance. [38] APD⁵⁰and APD⁹⁵were shorter at 6.8 mM [Potassium sup +]^othan at either 3.9 or 2.3 mM [Potassium sup +]^o(Figure 5and Figure 6) as a result of improved Potassium sup + conductance. [8,40,45,50] Interventions that affect ionic currents during the repolarization phase may selectively alter the shape and duration of the cardiac action potential. Verapamil may affect both the I^Caand Potassium sup + conductance. [43,51] In our study, 1 micro Meter of verapamil significantly shortened both the APD⁵⁰and the APD⁹⁵, but only at 28 degrees Celsius and [Potassium sup +]^ogreater or equal to 3.9 mM (Figure 5and Figure 6). At this temperature and with 6.8 mM [Potassium sup +]^oand verapamil in the superfusate, the relative shortening of APD⁵⁰exceeded that of APD⁹⁵(Figure 7(C)). The observation that earlier phases of repolarization (APD⁵⁰) appear to be more readily affected by verapamil is consistent with the primary blocking action of this agent on I^Ca. At the same time less affected shortening of APD⁹⁵by verapamil, and its significant shortening at higher [Potassium sup +]^osuggests that another mechanism is involved in the lengthening of APD at low temperature, presumably hypothermia-induced reduction of I^K. [46] Also, when [Potassium sup +]^owas sufficiently high at 28 degrees Celsius, both APD⁵⁰and APD⁹⁵were shortened, indicating not only an important role of [Potassium sup +]^oin increasing Potassium sup + conductance but also an interaction between Calcium²+-channel blockade and Potassium sup + conductance. Our study indicates that hypokalemia must be corrected if the APD is to be shortened with verapamil. Not only does correction of hypokalemia affects APD through increases of Potassium sup + conductance, [38] but in addition, as Cavalie et al. have shown, I^Cainactivation (which is delayed by hypothermia) depends on voltage and is greater at a more depolarized potential, [52] as in our study at 6.8 mM.

Different arrhythmias may arise if the [Calcium²+]ⁱis increased or if cardiac action potential is lengthened. First, increase in [Calcium²+]ⁱcaused by hypothermia results in oscillatory release of [Calcium²+] from sarcoplasmic reticulum; this may generate the transient inward current allowing more Sodium sup + and Calcium²+ into the cell creating delayed afterdepolarization. [5,17,38] When delayed after-depolarization reaches certain threshold, arrhythmias may be triggered. These arrhythmias can be seen at low [Potassium sup +]^o, and when Sodium sup + extrusion from the cell is reduced, [8] because both may be encountered during hypothermia. By diminishing [Calcium²+]ⁱinflux, verapamil reduces Calcium²+ overload, delays afterdepolarization amplitude and reduces triggered automaticity. [53,54] Verapamil thus may be useful for treatment of hypothermia-induced arrhythmias based on delayed afterdepolarization. Second, reentry arrhythmias may occur during the propagation of slow action potentials through Purkinje fibers [5,8,21,31,55] especially in vivo, when, because of uneven tissue cooling, heterogeneity of repolarizations may ensue. [8] A shortening of APD with verapamil in hypothermia is both temperature and [Potassium sup +]^odependent, and this may have significant clinical implications, but only at higher [Potassium sup +]^o: the lower the temperature, the longer the APD, and the greater the shortening effect of verapamil. Because the APD is very dependent on regional myocardial temperature, verapamil selectively affects APD: a greater APD in a cooler region will be decreased more than a lesser APD in a warmer region. By reducing the regional differences in myocardial repolarization this effect may decrease the propensity for development of arrhythmogenic reentry circuits.

In conclusion, there appears to be an increased relative I sub Ca at low temperature, as evidenced by the effective shortening of APD in deep hypothermia by verapamil. Although this shortening may result from an increase in I^Ca, a decrease in Potassium sup + conductance at low temperature is more likely, as evidenced by the effective shortening of APD when the Potassium sup + conductance is increased by increasing [Potassium sup +]^O. Finally, the effectiveness of verapamil to shorten APD in hypothermia was dependent on [Potassium sup +]^Osuggesting that the increase in Potassium sup + conductance is an important factor for achieving this verapamil effect. Verapamil and Potassium sup + supplementation in hypothermia may have an antiarrhythmic effect primarily by reducing the dispersion of prolonged APD. Further studies will be needed to evaluate this antiarrhythmic effect in vivo.

The authors thank Edith Sulzer for her help in the preparation of this manuscript.

* Guide for Care and Use of Laboratory Animals. Publication 85–23. Bethesda, MD, Public Health Services, National Institutes of Health, revised 1985.