This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bou-Abboud, E. Articles by Nattel, S. Search for Related Content PubMed PubMed Citation Articles by Bou-Abboud, E. Articles by Nattel, S. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB SODIUM SODIUM BICARBONATE

(Circulation. 1996;94:1954-1961.)
© 1996 American Heart Association, Inc.

Articles

Relative Role of Alkalosis and Sodium Ions in Reversal of Class I Antiarrhythmic Drug–Induced Sodium Channel Blockade by Sodium Bicarbonate

Elias Bou-Abboud, MSc; Stanley Nattel, MD

the Department of Pharmacology, University of Montreal (E.B.-A., S.N.); the Department of Medicine, Montreal Heart Institute and University of Montreal, the Department of Pharmacology and Therapeutics, McGill University, and the Research Center of the Montreal Heart Institute (S.N.); Montreal, Quebec, Canada.

Correspondence to Stanley Nattel, MD, Montreal Heart Institute, 5000 Belanger St, Montreal, Quebec, H1T 1C8, Canada.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Background Hypertonic sodium salts are used to treat sodium channel–blocking drug cardiotoxicity. The relative roles of alkalinization and increased sodium concentration ([Na⁺]_o) for various drugs are incompletely known.

Methods and Results The effects of four class I drugs on action potential characteristics of canine Purkinje fibers at equi-effective concentrations (disopyramide 30 µmol/L, mexiletine 80 µmol/L, flecainide 7 µmol/L, imipramine 5 µmol/L) were studied in the presence of normal Tyrode solution and one altered solution (increased [Na⁺]_o, increased bicarbonate concentration, or both) in each experiment. Combined increases in sodium and bicarbonate concentration significantly reduced the depressant effects of flecainide, imipramine, and mexiletine on phase 0 upstroke (V_max) but did not alter the effects of disopyramide. The effects of sodium bicarbonate were entirely due to alkalinization in the case of imipramine, but both alkalinization and increased [Na⁺]_o contributed to the interaction with flecainide and mexiletine. The reversal of V_max depression by increased [Na⁺]_o and pH was due in part to hyperpolarization. In addition, alkalosis directly reversed the hyperpolarizing shift in V_max inactivation caused by flecainide and imipramine without altering the shift caused by disopyramide and mexiletine.

Conclusions Increases in sodium bicarbonate concentration reverse the effects of class I antiarrhythmic drugs to a varying extent, with drug-specific contributions of the sodium and bicarbonate moiety. The molecular basis for this drug specificity remains to be elucidated, but it has important potential implications for the use of hypertonic sodium salts to treat cardiotoxicity caused by sodium channel–blocking drugs.

Key Words: ions • sodium • antiarrhythmia agents • action potentials • conduction • excitation

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Hypertonic sodium salts, in particular sodium bicarbonate and sodium lactate, have long been known to counteract cardiotoxic effects of class I antiarrhythmic drugs.^{1 2 3} Initially used to reverse adverse effects of quinidine and procainamide, they have been found to be useful in treating cardiotoxicity caused by a wide range of sodium channel–blocking drugs including tricyclic antidepressants^{4 5 6 7} and class Ic antiarrhythmic agents.^{8 9 10 11}

Prudhommeaux et al¹² provided the first evidence that alkalinization is responsible for the beneficial actions of sodium bicarbonate against imipramine toxicity. Conflicting results have been presented regarding the relative importance of pH change and sodium loading in the reversal of sodium channel blocker cardiotoxicity by sodium salts. In the case of tricyclic antidepressants, results have been presented suggesting a primary role for pH change,^{12 13 14} a primary role for sodium loading,¹⁵ and a contribution of both.¹⁶ Studies of O-desmethyl-encainide toxicity suggested that only the sodium moiety contributed to the reversal caused by hypertonic sodium bicarbonate administration.¹⁷

The present study was designed to test the hypothesis that various sodium channel blockers in clinical use respond to the moieties of sodium bicarbonate in drug-specific ways, ie, that increases in extracellular sodium concentration ([Na⁺]_o) will antagonize the sodium channel–blocking actions of some agents, that increases in extracellular pH (pH_o) will antagonize the actions of other agents, and that the actions of some compounds will respond to changes in both [Na⁺]_o and pH_o. To eliminate as much as possible interpreparation variability in action potential parameters and drug response, we exposed each preparation under continuous stable impalement of a single cell to at least four conditions: (1) control ("normal") superfusate, (2) a superfusate with increased [Na⁺]_o, pH_o, or both, (3) and (4), the same solutions as in (1) and (2), but containing a sodium channel blocker at a concentration causing 50% reduction in maximum phase 0 upstroke velocity (V_max) in control superfusate. Four drugs were chosen for study: three sodium channel–blocking antiarrhythmic agents (mexiletine, rapid kinetics, class 1a of the Singh-Vaughan Williams classification; disopyramide, intermediate kinetics, class 1b; flecainide, slow kinetics, class 1c) and a tricyclic antidepressant (imipramine, intermediate kinetics). Data were obtained under physiological conditions of temperature and ion concentrations with the use of the standard microelectrode technique, with V_max as an indicator of net phase 0 inward current.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Mongrel dogs of either sex (12 to 25 kg) were anesthetized with sodium pentobarbital (30 mg/kg IV). The hearts were rapidly removed through a lateral thoracotomy and placed into chilled, oxygenated, standard Tyrode solution. All usable false tendons from both ventricles were removed along with their muscle attachment. The standard Tyrode solution contained (mmol/L): NaCl 120, NaHCO₃ 22, KCl 4, MgCl₂ 0.5, CaCl₂ 1, NaH₂PO₄ 0.9, dextrose 10; aerated with 95% O₂/5% CO₂ (pH=7.35 in the bath). Preparations were pinned to the Sylgard-covered floor of a 30-mL Lucite tissue chamber and superfused with normal Tyrode solution, which was identical to standard Tyrode solution except that NaHCO₃ was reduced to 20 mmol/L and sucrose (40 mmol/L) was added to allow for the desired changes in Na⁺ and HCO₃^- concentrations in altered solutions while maintaining constant osmolarity (see Table 1). The bath temperature was maintained at 37°C by a heating element and proportional power supply. All preparations were stimulated at 2 Hz for 1 hour in the presence of standard Tyrode solution to allow for equilibration before study.

View this table: [in this window] [in a new window]   Table 1. Composition of Various Test Superfusates Used

Glass microelectrodes filled with 3 mol/L KCl and with tip resistances of 8 to 20 M were coupled by a silver–silver chloride junction to a high-input impedance amplifier (WPI KS-700, World Precision Instruments). Amplified waveforms were digitized (Tecmar Labmaster A/D board), and standard action potential characteristics were analyzed on-line with an IBM PC according to previously described methods.¹⁸ An electronic differentiating circuit was calibrated to provide an output of 0.5 V per 100 V/s rate of change in transmembrane potential with output linearly related to rate of voltage rise between 100 and 1000 V/s. A peak memory unit was used to capture maximal output from the differentiator, which was then converted into digital format and analyzed with an IBM personal computer and customized software routines. A bipolar Teflon-coated platinum electrode was used to deliver square-wave pulses (2-ms duration, 1.5 times late-diastolic threshold current) controlled by a programmable stimulator (Bloom Instruments) to stimulate the preparation. The pH was measured with an IL1312 analyzer (Coulter Electronics of Canada), and ionized [Ca²⁺] was measured with an ion-sensitive electrode (Statprofile 9 Gas Analyzer System, Nova Biomedical).

Solutions
To isolate the role of sodium and/or pH, the normal solution was modified to contain a high sodium concentration and/or increased pH while maintaining other ionic constituents and osmolarity constant. Sucrose, a metabolically inert substance,¹⁹ was added to solutions with a lower ionic content (as shown in Table 1) to keep osmolarity constant among all solutions used. The Na⁺, HCO₃^-, and pH of the normal and various test solutions are listed in Table 1, with other constituents being identical to the standard solution mentioned above. The solutions with increased Na⁺ and/or HCO₃^- concentration will be referred to as "altered Tyrode solutions."

Experimental Protocols
Modulation of Drug Actions on Action Potential Parameters by Test Solution
Each cell was studied under both normal and one modified solution under control conditions, and then after the addition of a study drug. The order in which solutions were studied was varied to avoid bias from time-dependent electrophysiological changes. Stable continuous impalement of a single cell throughout each experiment was required. Preliminary experiments were used to establish the concentration of each study drug needed to reduce V_max at a frequency of 2 Hz by 50% in normal Tyrode solution. Concentrations reported in the literature to reduce V_max by 30% to 70% were used in initial experiments and then modified as necessary until a concentration was found for each drug that reduced V_max by 50% at 2 Hz. The concentrations defined were then used for all protocols described. The equi-effective concentrations identified were imipramine, 5 µmol/L; disopyramide, 30 µmol/L; mexiletine, 80 µmol/L; and flecainide, 7 µmol/L. Enough time was allowed for steady-state changes after any solution change. Steady-state conditions required 30 minutes after changing Tyrode solution and after the addition of flecainide or mexiletine, 60 minutes after the addition of disopyramide, and 90 minutes after the addition of imipramine. Action potential parameters were determined after 3 minutes of pacing at interstimulus intervals of 300, 400, 500, 800, and 1000 ms under each condition. Qualitatively similar results were obtained at all cycle lengths, and a cycle length of 500 ms was selected as in previous work²⁰ for primary analyses. The following parameters were monitored: MAP (membrane potential at the onset of phase 0), AMP, APD₅₀ and APD₉₅, and V_max. Percent reversal of drug effect on V_max by an altered Tyrode solution was calculated on the basis of the reduction of drug effect caused by the solution divided by the drug effect in normal Tyrode solution, times 100%.

Modulation by Alkalinization of the V_max Inactivation Curve
Alkalinization caused a 4- to 5-mV hyperpolarization of the MAP (Table 2, A). To determine whether changes in pH have any direct effect on the voltage dependence of inactivation of V_max, we performed experiments in which V_max was determined as a function of MAP by increasing extracellular K⁺ concentration ([K⁺_o]) from 2.7 to 10 mmol/L and then returning to 2.7 mmol/L [K⁺_o]. Data were obtained in each cell under normal and high pH conditions. Some cells for which impalement was particularly stable were studied at both pH values in the absence and then the presence of imipramine, flecainide, mexiletine, or disopyramide, while in other cells stable impalement was maintained under both pH values and all [K⁺_o] conditions in either the absence or presence of drug, but not both. Alkalinization was found to cause a slight but significant decrease in ionized Ca²⁺ concentration, from 0.88±0.08 mmol/L at pH 7.3 to 0.81±0.1 mmol/L at pH 7.6 (P<.001). To evaluate the extent to which effects of alkalinization may have been due to changes in ionized [Ca²⁺], we performed experiments in which the voltage dependence of V_max was determined at pH 7.3 and 7.6, but with total [Ca²⁺] in the alkaline Tyrode solution increased by 10% to ensure that ionized [Ca²⁺] was the same at both pH values.

View this table: [in this window] [in a new window]   Table 2. Effects of Altered Tyrode Solutions on Action Potential Characteristics and Effects of Drugs Studied on Action Potential Parameters in Normal Tyrode Solution

Statistical Analysis
Group data are represented as mean±SD. Student's paired and unpaired t tests were used for single statistical comparisons between groups. Multiple group means were compared by ANOVA with a range test (Scheffe test). A value of P<.05 was considered statistically significant.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Effects of Drugs and Altered Tyrode Solutions
Successful experiments were completed in a total of 84 Purkinje fibers from 68 dogs. Table 2 (A) presents changes in action potential characteristics caused by altered Tyrode solutions. Increasing [Na⁺]_o resulted in slight but statistically significant increases in MAP, APD, and V_max. Increasing pH without altering [Na⁺]_o caused significant hyperpolarization but did not alter APD or V_max. Combined alkalosis and increased [Na⁺]_o significantly increased MAP, APD₅₀, and V_max.

All drugs studied significantly reduced MAP, APD₅₀, and V_max (Table 2, B). Flecainide, mexiletine, and imipramine significantly decreased APD₉₅, whereas disopyramide had the opposite effect, delaying full repolarization.

Modulation of Drug-Induced Electrophysiological Changes by Various Test Solutions
Fig 1 shows typical action potential recordings obtained in one cell at normal and high pH, in the absence and presence of imipramine. In the absence of the drug, alkalinization caused a slight hyperpolarization and increase in V_max without significantly changing APD₉₅. Imipramine in standard Tyrode solution reduced APD, MAP, and V_max. In the presence of imipramine, the high pH solution caused a substantial increase in V_max, along with a slight hyperpolarization and decrease in APD₉₅.

View larger version (70K): [in this window] [in a new window]   Figure 1. Representative action potential recording as obtained from a cell exposed to both normal (N) and high (H) pH in the absence and the presence (D) of 5 µmol/L imipramine. The differentiated signal showing Vmax is at the upper right.

The modulation of drug-induced changes by altered solutions is illustrated in Figs 2 to 5. Fig 2 shows values of V_max measured under four conditions for each cell and each drug-normal Tyrode solution, altered Tyrode solution (with high pH, high [Na⁺], or the combination), drug in normal Tyrode solution, and then drug in the same altered solution previously used. The asterisks in the figure indicate a significant interaction (by ANOVA) between drug actions and the solution studied, ie, statistically significant modulation of drug action by the factor modified in the altered Tyrode solution. In the case of imipramine, solutions with increased pH attenuated the actions of the drug on V_max, but increased [Na⁺] alone had no effect. For flecainide and mexiletine, changes in [Na⁺], pH, or both significantly modulated drug action. Effects of disopyramide were not altered by changes in pH or combined changes in pH and [Na⁺], although increases in [Na⁺]_o alone appeared to produce weak modulation of drug action.

View larger version (41K): [in this window] [in a new window]   Figure 2. Vmax in the presence of normal and altered (HIGH) Tyrode solution in the absence and then in the presence of disopyramide, flecainide, imipramine, and mexiletine. Each cell was studied in the presence of normal and one modified solution in the absence and then in the presence of drugs. Modified solutions contained increased Na+ alone (HNa), increased pH alone (Alk), or the combination (Comb). Data are mean±SD. *P<.05, **P<.01, ***P<.001 for significance of modulation of drug action by modified solution (ANOVA). Each set of conditions was studied in five preparations for each altered solution for disopyramide, flecainide, and mexiletine and six preparations for each altered solution for imipramine.

Modulation of drug effects by altered Tyrode solutions are further analyzed in Fig 3. Alkalinization and increased [Na⁺] reversed the effect of both mexiletine and flecainide by about 15% to 20%, with the combination being more effective than either alone in the case of mexiletine, and significantly more effective than high [Na⁺] in the case of flecainide. For imipramine, high [Na⁺] had no effect, while alkalinization and the combination both resulted in about a 35% reduction in drug effect on V_max, with either being substantially more effective than high [Na⁺] and neither significantly different from the other. High [Na⁺] reduced the action of disopyramide by a limited extent (10% on average), whereas alkalinization and the combination had no clear effect.

View larger version (20K): [in this window] [in a new window]   Figure 3. Percent reversal by modified solutions of Vmax depression caused by drugs (for equation, see text), based on raw data shown in Fig 2. Data are mean±SD. *P<.05, **P<.01, ***P<.001 for comparison shown. DIS. indicates disopyramide; FLEC., flecainide; IMI., imipramine; and MEX., mexiletine. Concentrations used were 30 µmol/L for disopyramide, 7 µmol/L for flecainide, 5 µmol/L for imipramine, and 80 µmol/L for mexiletine.

Since Na⁺ channels can be inactivated by reductions in MAP, particularly in the presence of sodium channel blockers, it is important to consider the potential role of changes in MAP in mediating alterations in V_max. Fig 4 presents the changes in MAP observed with various solutions in the presence and absence of the drugs studied. As indicated in Table 2, high [Na⁺] produced a slight but statistically significant hyperpolarization, while alkalosis and the combination produced a more pronounced hyperpolarization. Each of the drugs studied produced a slight but consistent depolarization. As shown in Fig 4, while the above changes were observed consistently, there were no significant interactions between the various test solutions and drug actions.

View larger version (45K): [in this window] [in a new window]   Figure 4. Resting membrane potential under control and drug conditions. The format is the same as in Fig 2. No significant modulation of drug action by altered solutions was observed. Numbers of preparations/group as in Fig 2.

Many drugs bind preferentially to the inactivated state of the Na⁺ channel. Changes in APD can therefore mediate alterations in drug action. Fig 5 shows changes in APD₉₅ caused by the various drugs studied in the presence of different solutions. All results shown are at a basic cycle length of 500 ms. For disopyramide, flecainide, and mexiletine, altered solutions showed either no interaction with drug effects on APD or a slight interaction of weak statistical significance. In the absence of a significant interaction, changes in APD were unlikely to have played a role in the ability of increased [Na⁺] and increased pH to attenuate the effect of flecainide on V_max or the ability of increased [Na⁺] to reverse the effect of mexiletine or imipramine on V_max. On the other hand, increased pH slightly but significantly increased the APD-reducing action of imipramine and mexiletine, and the combination of increased [Na] and pH significantly enhanced the APD-reducing properties of flecainide, imipramine, and mexiletine. By attenuating drug binding to the inactivated state of the Na channel, these effects could have contributed to reducing drug-induced V_max inhibition under these conditions.

View larger version (43K): [in this window] [in a new window]   Figure 5. APD95 as measured under control and drug conditions. Format as in Figs 2 and 4. *P<.05, **P<.01 for significance of interaction between drug effects and solution (ANOVA). Numbers of preparations/group as in Fig 2.

Modulation of V_max Inactivation Curves
Because increased pH caused substantial hyperpolarization, and class I drugs are known to cause hyperpolarizing shifts in Na channel inactivation, much of the reversal of drug action on V_max at increased pH may have been due to hyperpolarization alone. We therefore sought to determine the extent to which changes in drug effects caused by alkalosis were due to direct, activation potential–independent interactions. V_max inactivation curves were obtained by exposing preparations to varying superfusate K⁺ concentrations. Fig 6 shows examples of such curves obtained at normal and high pH values and in the absence and presence of each drug in a representative preparation.

View larger version (26K): [in this window] [in a new window]   Figure 6. Modulation by alkalinization of the voltage dependence of Vmax inactivation in the absence (circles) and in the presence (triangles) of disopyramide, flecainide, imipramine, and mexiletine. Similar results were obtained in four cells for disopyramide, four cells for flecainide, seven cells for imipramine, and three cells for mexiletine. Each set of data was fit by a Boltzmann distribution (curves shown) of the form Vmax=A(1+exp[MAP-V50]/K), where A, V50, and K are constants, with V50 being the voltage for 50% inactivation and K a slope factor. For mean values of V50 and K (slope factor) under each condition, see Table 3.

Under control conditions, alkalosis produces a small but consistent hyperpolarizing shift of the inactivation curve. All the drugs studied caused a hyperpolarizing shift in the inactivation curve. Alkalosis-induced changes in the voltage dependence of inactivation in the presence of drugs varied with the drug studied. In the presence of imipramine, alkalosis partially reversed the drug-induced hyperpolarizing shift, resulting in an alkalosis-induced change opposite in direction to that observed in the absence of the drug. In the presence of flecainide, changes in pH did not change the voltage dependence of inactivation. Alkalosis produced a modest hyperpolarizing shift in the presence of mexiletine and a strong hyperpolarizing shift in the presence of disopyramide.

The inactivation curves were analyzed quantitatively by nonlinear curve-fitting to a Boltzmann distribution, producing the type of curve fits shown in Fig 6 and the mean characterizing constants summarized in Table 3. Under control conditions, alkalosis produced a highly significant, >3-mV shift in inactivation as indicated by the changes in voltage for 50% inactivation (V₅₀). All drugs studied shifted the inactivation voltage in a hyperpolarizing way, with the most negative V₅₀ in the presence of imipramine and the least negative in the presence of disopyramide. Alkalosis caused a significant depolarizing shift in the presence of imipramine, a significant hyperpolarizing shift in the presence of disopyramide, and no significant changes in the presence of the other drugs.

View this table: [in this window] [in a new window]   Table 3. Effect of Increasing pH on Voltage Dependence of Vmax Inactivation

As mentioned in "Methods," alkalosis caused a small but significant reduction in ionized [Ca²⁺]. To determine pH effects in the absence of changes in [Ca²⁺], we performed five additional experiments in which additional Ca²⁺ was added to the high pH superfusate to maintain a constant ionized [Ca²⁺]. Fig 7 shows typical inactivation curves at normal and high pH with a constant ionized [Ca²⁺]. With the latter variable controlled, alkalosis no longer produced a perceptible shift in inactivation. Mean data for the five experiments are provided in Table 3. When results at high pH in the presence of antiarrhythmic drugs are corrected for changes induced by altered [Ca²⁺], the depolarizing shift (ie, reversal of drug-induced hyperpolarizing shift) in the presence of imipramine is increased, and a small depolarizing shift appears for flecainide. The changes observed with mexiletine and disopyramide decrease greatly, suggesting that they are largely attributable to pH effects on ionized [Ca²⁺] when the latter is not corrected.

View larger version (14K): [in this window] [in a new window]   Figure 7. Vmax-MAP relationship after increasing [CaCl2] in the high pH solution (closed circles) by 10% compared with the normal solution (open circles) to keep ionized [Ca2+] constant. Note the lack of a hyperpolarizing shift when ionized calcium was kept constant. Similar results were obtained in a total of five experiments.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Hypertonic sodium salts are commonly used to treat cardiotoxic effects of sodium channel–blocking drugs. The major novel finding of the present study is that there is great interdrug variability in the extent of reversal of drug action on V_max by increased [Na⁺]_o, increased pH, and the combination and in the relative role of the Na⁺ and HCO₃^- moieties. The potential mechanisms of reversal appear to be complex and can include changes in diastolic potential, in APD, in ionized [Ca²⁺], and in the direct interaction between drugs and the sodium channel as reflected in the V_max inactivation curve.

Comparison With Previous Studies of Modulation of Antiarrhythmic Drug Action by Sodium Salts
Several previous studies have evaluated the effects of sodium salts on Na⁺ channel–blocking drug action. Sodium lactate reverses the effect of quinidine on V_max, an action largely attributed to the Na⁺ moiety.²¹ Jensen and Katzung^{22 23} showed that increases in [Na⁺]_o reverse the depressant effects of phenytoin on V_max and conduction. Kohlhardt²⁴ found that decreases in [Na⁺]_o enhance the actions of a variety of Na⁺ channel blockers.

Relatively few studies have attempted to separate the respective roles of pH change and the Na⁺ moiety in the ability of Na salts to reverse Na⁺ channel–blocking drug cardiotoxicity. In vivo studies have suggested that pH change is the primary factor in NaHCO₃ reversal of amitriptyline cardiotoxicity¹⁴ and that the Na⁺ moiety may be of prime importance for O-desmethyl-encainide¹⁷ and desipramine.¹⁵ In vitro studies pointed toward a role for both the Na⁺ moiety and pH change in the interaction¹⁵ between [NaHCO₃] and amitriptyline on V_max in canine cardiac Purkinje fibers.¹⁶ The present study is the first to compare systematically the response to increased [Na⁺] alone, increased pH alone, and the combination for a variety of Na⁺ channel–blocking drugs. Our results indicate important variability among the responses of different drugs to a combination of increased [Na⁺] and pH, as occurs with NaHCO₃ administration, as well as important variability in the respective role of [Na⁺] and pH.

Potential Underlying Mechanisms
Our data point to a variety of potential mechanisms for the modulation of drug action by changes in [Na⁺]_o and pH. In addition to direct alterations of the drug-receptor interaction, drug effects may be altered by changes in activation potential (Fig 4), in APD (Fig 5), and in ionized Ca²⁺ concentration (Table 3).

Changes in pH have complex effects on the direct interactions between sodium channels and blocking drugs. Acidosis slows the unbinding of a variety of drugs from the Na⁺ channel,^{25 26 27} but complex changes in state-dependent block occur and result in no significant change in net blocking action of lidocaine as a result of acidosis when holding potential is controlled.²⁸ External alkalosis enhances a rapid component of block by lidocaine, disopyramide, and mexiletine,^{29 30} an effect interpreted as due to enhanced interaction of uncharged drug molecules with the inactivated³⁰ or activated²⁹ state, respectively. On the other hand, block by imipramine is reduced by alkalosis, apparently as a result of more ready dissociation of uncharged drug from the various states of the channel. (F. Bou-Abboud, MD; V. Nesterenko, PhD; S. Nattel, MD; unpublished data, 1996). The nature of the drug-pH interaction may also depend on the relative magnitude of intracellular and extracellular pH change, which may affect the blocking actions of various drugs in qualitatively different ways.³²

Electrostatic repulsion has been postulated to explain the reduction in Na⁺ channel drug-blocking action when [Na⁺]_o is increased,³³ and changes in the magnitude of I_Na as a result of altered [Na⁺]_o have been thought to be important.²⁴ The results of studies of cocaine block of batrachotoxin-altered Na⁺ channels in lipid bilayers favor an electrostatic "knock-off" type of mechanism for the interaction between Na⁺ and blocking drugs, with a straightforward competitive interaction less likely.³⁴ Our laboratory has shown that [Na⁺]_o modulates the interaction between flecainide and its Na⁺ channel receptor, even under conditions that prevent Na⁺ flux across the channel.³⁵

Potential Significance of Our Findings
Sodium salts have been used widely to treat cardiotoxic effects of a variety of Na⁺ channel–blocking drugs.^{1 4 8 9} Our results indicate that drugs may differ widely in the response to Na⁺ salts and that caution is therefore appropriate in their choice and use. For example, the effects of disopyramide were not altered by NaHCO₃, suggesting that the latter might not be effective in treating disopyramide cardiotoxicity. Since fluid, acid-base, and electrolyte abnormalities can result from the administration of hypertonic NaHCO₃ solutions, use of the latter might be counterproductive in a patient intoxicated with disopyramide. On the other hand, NaHCO₃ strongly inhibited the actions of imipramine, with actions fully attributable to pH change. Thus, hyperventilation might be a reasonable and rapidly reversible alternative to hypertonic Na⁺ salts in patients with imipramine overdose on a mechanical ventilator.

Our results point to the complexity of the drug-pH and drug-[Na⁺]_o interaction. Net effects will depend on direct interactions between pH, [Na⁺]_o, and drugs at the level of the Na⁺ channel receptor but also potentially on indirect changes caused by alterations in ionized [Ca²⁺], resting potential, and APD.

Study Limitations
We used standard microelectrode techniques, with V_max as an indicator of maximum phase 0 I_Na. The relationship between V_max and I_Na is nonlinear,^{36 37} although the discrepancy decreases as temperature approaches the physiological range.³⁷ On the other hand, V_max is likely to be related linearly to net phase 0 inward current.³⁸ Standard microelectrode techniques allow for the use of highly physiological conditions, which do not permit accurate voltage clamp of I_Na with present technology. Reduced temperature, almost always needed for studies of whole-cell I_Na, can cause qualitative and quantitative changes in I_Na-blocking drug action,³⁹ as can the reduced [Na⁺]_o that is usually necessary. Standard microelectrode techniques also allow for the evaluation of important physiological variables, such as resting potential and APD, which are not monitored during voltage clamp studies. Finally, multicellular preparations in general (and V_max in particular) are much more stable over time than single cells and their I_Na. This allows for multiple interventions to be studied under stable conditions in the same cell, resulting in more controlled and powerful statistical comparisons.

These experiments were conducted in canine cardiac Purkinje fibers. Extrapolation to changes in human ventricular muscle, particularly in the presence of cardiac disease, requires caution. It is difficult to relate directly the drug concentrations we used to a corresponding clinical concentration because of important potential interspecies differences in sensitivity to antiarrhythmic drugs.⁴⁰ There is theoretical and experimental evidence that in vivo changes in QRS duration are proportional to the inverse of the square root of changes in V_max.⁴¹ On the basis of this relation, concentrations that cause a 50% decrease in V_max (as used in the present study) should correspond to those producing a 41% prolongation in QRS duration in vivo, of the same order as seen in many patients with mild to moderate toxicity due to sodium channel–blocking drugs.^{5 6 7 8 9 10 11} The concentrations that we studied were in the clinically toxic but relevant range for flecainide (concentration studied, 7 µmol/L; therapeutic range, 0.5 to 2.4 µmol/L),⁴² disopyramide (concentration studied, 30 µmol/L; therapeutic range, 8.8 to 23.6 µmol/L),⁴² and imipramine (concentration studied, 5 µmol/L; toxic tricyclic antidepressant concentrations >3.6 µmol/L).⁴³ For mexiletine, the concentration we studied (80 µmol/L) was well in excess of the therapeutic range (4.5 to 11 µmol/L),⁴² but clinical doses of mexiletine are limited by noncardiovascular toxicity, and our choice of concentration was justified on the pharmacological basis of selecting equi-effective sodium channel–blocking concentrations.

Conclusions
We have shown that the actions of a variety of class I drugs are modulated in different ways by changing extracellular [Na⁺], pH, and the combination. Caution is warranted when using hypertonic NaHCO₃ to treat class I cardiotoxicity, since such therapy may not be beneficial for all drugs. The relative role of the Na⁺ moiety and pH varies among drugs, and for some agents more specific therapy may be provided by altering one or the other rather than both. Finally, the mechanisms by which [Na⁺] and pH can alter class I cardiotoxicity are potentially complex and varied.

	Selected Abbreviations and Acronyms

 

AMP = action potential amplitude APD = action potential duration APD50 = APD at 50% repolarization APD95 = APD at 95% repolarization MAP = membrane activation potential Vmax = maximal rate of voltage rise during phase 0 of action potential

	Acknowledgments

 
This study was supported by grants from the Medical Research Council of Canada, the Quebec Heart Foundation, and the Fonds de Recherche de l'Institut de Cardiologie de Montreal. The authors thank Emma De Blasio for technical support, Dr Rene Cardinal for useful discussion and suggestions, and Carolyn Gillis for typing the manuscript.

Received November 13, 1995; revision received April 17, 1996; accepted April 23, 1996.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Wasserman F, Brodsky L, Dick MM, Kathe JH, Rodensky PL. Successful treatment of quinidine and procaine amide intoxication: report of three cases. N Engl J Med. 1958;259:797-803.
   
2. Bellet S, Hamdan G, Somlyo A, Lara R. A reversal of the cardiotoxic effects of procaine amide by molar sodium lactate. Am J Med Sci. 1959;237:177-189.[Medline] [Order article via Infotrieve]
   
3. Bellet S, Hamdan G, Somlyo A, Lara R. The reversal of cardiotoxic effects of quinidine by molar sodium lactate: an experimental study. Am J Med Sci. 1959;237:165-176.[Medline] [Order article via Infotrieve]
   
4. Brown TCK, Barker GA, Dunlop ME, Loughnan PM. The use of sodium bicarbonate in the treatment of tricyclic antidepressant induced arrhythmias. Anaes Intens Care. 1973;1:203-210.[Medline] [Order article via Infotrieve]
   
5. Bismuth C, Bodin F, Pebay-Peroula F, Frejaville J-P. Intoxication par l'imipramine avec insuffisance cardiaque aigue: Interet du lactate de soude. Presse Med. 1968;48:2277-2278.
   
6. Callaham M. Tricyclic antidepressant overdose. JACEP. 1979;8:413-425.[Medline] [Order article via Infotrieve]
   
7. Hoffman JR, McElroy CR. Bicarbonate therapy for dysrhythmia and hypotension in tricyclic antidepressant overdose. West J Med. 1981;134:60-64.[Medline] [Order article via Infotrieve]
   
8. Chouty F, Funck-Brentano C, Landau JM, Lardoux H. Efficacite de fortes doses de lactate molaire par voie veineuse lors des intoxications au flecainide. Presse Med. 1987;16:808-810.
   
9. Pentel PR, Goldsmith SR, Salerno DM, Nasraway SA, Plummer DW. Effect of hypertonic sodium bicarbonate on encainide overdose. Am J Cardiol. 1986;57:878-880.[Medline] [Order article via Infotrieve]
   
10. Camous JP, Ichai C, Meyer P, Gibelin P, Baudouy M, Varenne A, Morand PH. Traitement des intoxications aux nouveaux anti-arythmiques (cibenzoline, flecainide, propafenone). Presse Med. 1987;41:2076.
    
11. Winkelmann BR, Leinberger H. Life-threatening flecainide toxicity: a pharmacodynamic approach. Ann Intern Med. 1987;106:807-814.
    
12. Prudhommeaux J-L, Lechat P, Auclair M-C. Etude experimentale de l'influence des ions sodium sur la toxicite cardiaque de l'imipramine. Therapie. 1968;23:675-683.[Medline] [Order article via Infotrieve]
    
13. Brown TCK. Tricyclic antidepressant overdosage: experimental studies on the management of circulatory complications. Clin Toxicol. 1976;9:255-272.[Medline] [Order article via Infotrieve]
    
14. Nattel S, Mittleman M. Treatment of ventricular tachyarrhythmias resulting from amitriptyline toxicity in dogs. J Pharmacol Exp Ther. 1984;231:430-435.[Abstract/Free Full Text]
    
15. Pentel P, Benowitz N. Efficacy and mechanism of action of sodium bicarbonate in the treatment of desipramine toxicity in rats. J Pharmacol Exp Ther. 1984;230:12-19.[Abstract/Free Full Text]
    
16. Sasyniuk BI, Jhamandas V. Mechanism of reversal of toxic effects of amitriptyline on cardiac Purkinje fibers by sodium bicarbonate. J Pharmacol Exp Ther. 1984;231:387-394.[Abstract/Free Full Text]
    
17. Bajaj AK, Woosley RL, Roden DM. Acute electrophysiologic effects of sodium administration in dogs treated with O-desmethyl encainide. Circulation. 1989;80:994-1002.[Abstract/Free Full Text]
    
18. Nattel S. Relationship between use-dependent effects of antiarrhythmic drugs on conduction and V_max in canine cardiac Purkinje fibers. J Pharmacol Exp Ther. 1987;241:282-288.[Abstract/Free Full Text]
    
19. Bailey JC. Electrophysiological effects of hypertonic sucrose solutions on canine cardiac Purkinje fibers. Circ Res. 1981;49:1112-1118.[Abstract/Free Full Text]
    
20. Nattel S, Elharrar V, Zipes DP, Bailey JC. pH-dependent electrophysiological effects of quinidine and lidocaine on canine cardiac Purkinje fibers. Circ Res. 1981;48:55-61.[Abstract/Free Full Text]
    
21. Cox AR, West TC. Sodium lactate reversal of quinidine effect studied in rabbit atria by the microelectrode technique. J Pharmacol Exp Ther. 1961;131:212-222.[Abstract/Free Full Text]
    
22. Jensen RA, Katzung BG. Electrophysiological actions of diphenylhydantoin in rabbit atria. Circ Res. 1970;26:17-27.[Abstract/Free Full Text]
    
23. Katzung BG, Jensen RA. The depressant action of diphenylhydantoin on electrical and mechanical properties of isolated rabbit and dog atria: dependence on sodium and potassium. Am Heart J. 1970;80:80-88.[Medline] [Order article via Infotrieve]
    
24. Kohlhardt M. A quantitative analysis of the Na⁺-dependence of V-_max of the fast action potential in mammalian ventricular myocardium: saturation characteristics and the modulation of a drug-induced I_Na blockade by [Na⁺]_o. Pflugers Arch. 1982;392:379-387.[Medline] [Order article via Infotrieve]
    
25. Grant AO, Strauss LJ, Wallace AG, Strauss HC. The influence of pH on the electrophysiological effects of lidocaine in guinea pig ventricular myocardium. Circ Res. 1980;47:542-550.[Free Full Text]
    
26. Grant AO, Trantham JL, Brown KK, Strauss HC. pH-dependent effects of quinidine on the kinetics of dV/dt_max in guinea pig ventricular myocardium. Circ Res. 1982;50:210-217.[Free Full Text]
    
27. Gilliam FR III, Rivas PA, Wendt DJ, Starmer CF, Grant AO. Extracellular pH modulates block of both sodium and calcium channels by nicardipine. J Gen Physiol. 1977;69:497-517.[Abstract/Free Full Text]
    
28. Wendt DJ, Starmer CF, Grant AO. pH dependence of kinetics and steady-state block of cardiac sodium channels by lidocaine. Am J Physiol. 1993;264:H1588-H1598.[Abstract/Free Full Text]
    
29. Sunami A, Fan Z, Nitta J-I, Hiraoka M. Two components of use-dependent block of Na⁺ current by disopyramide and lidocaine in guinea pig ventricular myocytes. Circ Res. 1991;68:653-661.[Abstract/Free Full Text]
    
30. Ono M, Sunami A, Sawanobori T, Hiraoka M. External pH modifies sodium channel block by mexiletine in guinea pig ventricular myocytes. Cardiovasc Res. 1994;28:973-979.[Abstract/Free Full Text]
    
31. Deleted in proof.
    
32. Barber MJ, Wendt DJ, Starmer CF, Grant AO. Blockade of cardiac sodium channels: competition between the permeant ion and antiarrhythmic drugs. J Clin Invest. 1992;90:368-381.
    
33. Cahalan MD, Almers W. Interactions between quaternary lidocaine, the sodium channel gates, and tetrodotoxin. Biophys J. 1979;27:39-56.[Abstract/Free Full Text]
    
34. Wang GK. Cocaine-induced closures of single batrachotoxin-activated Na⁺ channels in planar lipid bilayers. J Gen Physiol. 1988;92:747-765.[Abstract/Free Full Text]
    
35. Ranger S, Sheldon R, Fermini B, Nattel S. Modulation of flecainide's cardiac sodium channel blocking actions by extracellular sodium: a possible cellular mechanism for the action of sodium salts in flecainide cardiotoxicity. J Pharmacol Exp Ther. 1993;264:1160-1167.[Abstract/Free Full Text]
    
36. Cohen CJ, Bean BP, Tsien RW. Maximal upstroke velocity as an index of available sodium conductance: comparison of maximal upstroke velocity and voltage clamp measurements of sodium current in rabbit Purkinje fibers. Circ Res. 1984;54:636-651.[Abstract/Free Full Text]
    
37. Sheets MF, Hanck DA, Fozzard HA. Nonlinear relation between V_max and I_Na in canine cardiac Purkinje cells. Circ Res. 1988;63:386-398.[Abstract/Free Full Text]
    
38. Walton MK, Fozzard HA. The conducted action potential: models and comparison to experiments. Biophys J. 1983;44:9-26.[Abstract/Free Full Text]
    
39. Johns JA, Takafumi A, Bennett PB, Snyders DJ, Hondeghem LM. Temperature and voltage dependence of sodium channel blocking and unblocking by o-demethyl encainide in isolated guinea pig myocytes. J Cardiovasc Pharmacol. 1989;13:826-835.[Medline] [Order article via Infotrieve]
    
40. Wang Z, Pelletier LC, Talajic M, Nattel S. Effects of flecainide and quinidine on human atrial action potentials: role of rate-dependence and comparison with guinea pig, rabbit, and dog tissues. Circulation. 1990;82:274-283.[Abstract/Free Full Text]
    
41. Nattel S, Jing W. Rate-dependent changes in intraventricular conduction produced by procainamide in anesthetized dogs: a quantitative analysis based on the relation between phase 0 inward current and conduction velocity. Circ Res. 1989;65:1485-1498.[Abstract/Free Full Text]
    
42. McNeil JJ, Sloman JG. Cardiovascular diseases. In: Speight TM, ed. Drug Treatment. Principles and Practice of Clinical Pharmacology and Therapeutics. Auckland, NZ: ADIS Press Ltd; 1987:591-675.
    
43. Petit JM, Spiker DG, Ruwitch JF, Ziegler VE, Weiss AN, Biggs JT. Tricyclic antidepressant plasma levels and adverse effects after overdose. Clin Pharmacol Ther. 1976;21:47-51.
    

This article has been cited by other articles:

				E. Bou-Abboud and S. Nattel Molecular mechanisms of the reversal of imipramine-induced sodium channel blockade by alkalinization in human cardiac myocytes Cardiovasc Res, May 1, 1998; 38(2): 395 - 404. [Abstract] [Full Text] [PDF]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bou-Abboud, E. Articles by Nattel, S. Search for Related Content PubMed PubMed Citation Articles by Bou-Abboud, E. Articles by Nattel, S. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB SODIUM SODIUM BICARBONATE