Rate-dependence of class III actions in the heart

Geometrical considerations in cardiac electrophysiology and arrhythmogenesis

The rate of repolarization (RRepol) and so the duration of the cardiac action potential are determined by the balance of inward and outward currents across the cardiac membrane (net ionic current). Plotting action potential duration (APD) as a function of the RRepol reveals an inverse non-linear relationship, arising from the geometric association between these two factors. From the RRepol-APD relationship, it can be observed that a longer action potential will exhibit a greater propensity to shorten, or prolong, for a given change in the RRepol (i.e. net ionic current), when compared with one that is initially shorter. This observation has recently been used to explain why so many interventions that prolong the action potential exert a greater effect at slow rates (reverse rate-dependence). In this article, we will discuss the broader implications of this simple principle and examine how common experimental observations on the electrical behaviour of the myocardium may be explained in terms of the RRepol-APD relationship. An argument is made, with supporting published evidence, that the non-linear relationship between the RRepol and APD is a fundamental, and largely overlooked, property of the myocardium. The RRepol-APD relationship appears to explain why interventions and disease with seemingly disparate mechanisms of action have similar electrophysiological consequences. Furthermore, the RRepol-APD relationship predicts that prolongation of the action potential, by slowing repolarization, will promote conditions of dynamic electrical instability, exacerbating several electrophysiological phenomena associated with arrhythmogenesis, namely, the rate dependence of dispersion of repolarization, APD restitution, and electrical alternans.

A Critical Reconsideration of the Clinical Effects and Treatment Recommendations for Sodium Channel Blocking Drug Cardiotoxicity

The cardiac sodium channel is comprised of proteins that span the cardiac cell membrane and form the channel pore. Depolarisation causes the proteins to move and open the sodium channel. Once the channel is open (active conformation), sodium ions move into the cell. The channel then changes from the active conformation to an inactive conformation - the channel remains open, but influx of sodium ions ceases. Recovery occurs as the channel moves from the inactive conformation back to the closed conformation and is then ready to open following the next depolarisation. Sodium channel blocking drugs (NCBDs) occupy receptors in the channel during the active and inactive conformations. The drug dissociates from most of the channel receptors during recovery, but the time it takes the drug to dissociate slows recovery. The slowed recovery prolongs conduction time, the main toxicity of NCBD overdose. Conduction time is further prolonged if heart rate increases as there are more available active and inactive conformations/unit time, which increases channel receptor binding sites for the NCBD. In addition to prolonging conduction time, NCBDs also decrease inotropy. Treatment of NCBD cardiotoxicity has been based on in vitro and animal experiments, and case reports. Assumptions based on this evidence must now be reassessed. For example, canines consistently develop ventricular tachycardia (VT) when tricyclic antidepressants (TCAs) are administered. Much of the literature discussing NCBD cardiotoxicity assumes that TCA poisoning induces VT in humans with the same regularity that occurs in canines. Seemingly, in support of this assumption was the finding that patients with remote myocardial infarction developed VT when therapeutically ingesting a NCBD. However, conduction is prolonged in myocardium that is or has been ischaemic. NCBD prolong conduction more in previously ischaemic myocardium than in normal myocardium, which causes nonuniform conduction and allows the development of re-entrant arrhythmias such as VT. Although some nonuniform conduction may occur in the healthy heart following a NCBD overdose, there is no evidence that nonuniform conduction occurs to the extent that it will cause re-entrant arrhythmias in this setting. Using various animal models and a variety of NCBDs, sodium ions, bicarbonate ions and alkalosis have been compared for the treatment of ventricular arrhythmias, hypotension and mortality. The results of these experiments have been extrapolated to NCBD overdose in humans. Animal models and single treatment approaches may have narrowed our scope. More recent evidence indicates that properties of each individual NCBD may require unique treatment. There is limited evidence that glucagon, which increases initial sodium ion influx into the cardiac cell, should be considered early in the treatment of cardiotoxicity. Another consideration may be treatment of NCBD with faster kinetics. Conduction time is decreased if a NCBD occupying the receptor is replaced by a NCBD that moves off and on the receptor more quickly. There is less evidence for this treatment, as risk may be greater. With greater understanding of the sodium channel and NCBDs, we must reassess our approach to the treatment of patients with healthy hearts who overdose on NCBD.

Effects of bimakalim on human cardiac action potentials: comparison with guinea pig and nicorandil and use-dependent study

Electrophysiologic effects of K(ATP) channel openers (KCOs) are rarely studied for tissue and species specificity, and use-dependent investigations in human tissues are lacking. We therefore investigated in vitro the concentration-dependent effects of the KCO bimakalim [from 10 nM to 10 microM, at 1,000 ms of cycle length (CL) and 37 degrees C] on human (atrium, n = 4, and ventricle, n = 6) and guinea pig (atrium, n = 7, and ventricle, n = 6) transmembrane action potential (AP). The frequency relation (from CL 1,600 to 300 ms, 31 degrees C) of human atrial AP duration 90% (APD90) shortening (10 microM vs. baseline, n = 7) also was determined. A parallel study was performed with the KCO nicorandil (from 10 nM to 1 mM, n = 3) in human atrial APs, at 31 degrees C. Resting membrane potential and maximal upstroke velocity of AP were not modified by bimakalim at maximal concentration, whereas AP amplitude was decreased in both guinea pig preparations (p < 0.05); APD90 was shortened in all tissues (p < 0.01). Median effective concentration (EC50) for APD90 shortening at 37 degrees C was 0.54 and 2.74 microM in atrial and ventricular human tissue, respectively, and 8.55 and 0.89 microM in atrial and ventricular guinea pig tissue, respectively. In human atrial tissue at 31 degrees C, EC50 with bimakalim was 0.39 microM; a much higher value was seen with nicorandil (210 microM). Bimakalim (10 microM)-induced APD90 shortening as a function of stimulation rate was greatest at longest CL. Evidence is provided for (a) species (human vs. guinea pig) and tissue (atrium vs. ventricle) differential AP sensitivity to bimakalim; (b) an approximately 500-fold higher efficacy of bimakalim versus nicorandil to shorten human atrial APD90; and (c) normal use-dependence of human atrial APD90 shortening with bimakalim at 10 microM.

Dynamic analysis of dofetilide-induced changes in ventricular repolarization

OBJECTIVE

To use dynamic electrocardiographic (ECG) techniques to study the influence of heart rate on dofetilide-induced QT prolongation among healthy volunteers.

BACKGROUND

The extent to which heart rate modulates QT prolongation induced by the new class III antiarrhythmic drug dofetilide is a matter of debate.

METHODS

Ten healthy volunteers underwent two 24-hour ECG recordings, one in the absence of dofetilide and the other after a single oral dose of 0.5 mg dofetilide. Two 4-hour periods were defined during the second recording: Dh, which corresponded to stable high concentration of the drug, and D1, which corresponded to low concentration of the drug. Corresponding baseline recording periods, Ch and C1, matched by time with Dh and D1 were selected from the control ECG recording in the absence of dofetilide. QT versus R-R relations were compared in the presence and absence of dofetilide. The QT versus R-R relation slope was used as an index of the rate dependence QT prolongation. Rate-independent changes in QT duration were also analyzed.

RESULTS

During Dh, dofetilide induced a mean 12% lengthening of ventricular repolarization. Dynamic ECG analysis showed that this prolongation increased as R-R cycles became longer, a phenomenon known as reverse rate dependence. However, QT prolongation persisted at the shortest (600 ms) R-R cycle length that could be analyzed. During D1, dynamic ECG analysis showed a persistent, although small, effect of dofetilide on both QT prolongation (3%) and reverse rate dependence of this effect.

CONCLUSIONS

Dofetilide prolongs QT duration, and this class III effect is influenced by heart rate. Although dofetilide-induced QT prolongation decreases when the R-R cycle shortens, this reverse rate dependence is only partial because marked QT prolongation persists at an R-R cycle of 600 ms. The results of our study indicated that dynamic ECG techniques can be useful in detection of subtle, drug-induced changes in the duration of ventricular repolarization.

Sparfloxacin but not levofloxacin or ofloxacin prolongs cardiac repolarization in rabbit Purkinje fibers

Sparfloxacin, a fluoroquinolone antibacterial, has been reported to prolong cardiac repolarization in some patients. In this study, we have investigated the in vitro cardiac electrophysiological effects of two other fluoroquinolones, levofloxacin and ofloxacin, and compared them with those exerted by sparfloxacin. Cardiac action potentials have been recorded from rabbit Purkinje fibers using conventional glass microelectrodes. The influence of a sudden decrease in stimulation rate on repolarization is examined. It is found that ofloxacin and levofloxacin (1-100 microM) do not alter the action potential parameters even at a concentration as high as 100 microM. The stimulation rate is without effect on repolarization. On the contrary, sparfloxacin (1-100 microM) lengthens concentration-dependently the duration of action potential, this effect being significant from the concentration of 10 microM. A non significant decrease in maximal rate of rise of phase 0 depolarization was observed at the concentration of 100 microM. Under low stimulation rate, the sparfloxacin-induced prolonging effect was magnified and early afterdepolarizations occurred in one of seven fibers from the concentration of 30 microM and in four other fibers at the concentration of 100 microM. These results suggest that levofloxacin and ofloxacin had no effect on cardiac cellular electrophysiology whereas sparfloxacin exerts pure class III electrophysiological effects, which can explain the prolongation of QT interval observed clinically in some patients and might become arrhythmogenic in the presence of other predisposing factors.

Changes in repolarization dynamicity and the assessment of the arrhythmic risk

At the present time, the assessment of the arrhythmic risk from surface ECG recordings is built on time-domain and frequent-domain analysis of high resolution ECG acquisition together with interlead variability of QT interval duration (QT dispersion). The corresponding raw ECG tracings are obtained in resting conditions. However, the dynamic aspects of the ECG signal is a rapidly evolving matter of interest. In addition to the beat-to-beat oscillations of the ventricular repolarization amplitude (QT alternans), there is growing evidence that the patterns of QT interval shortening with increasing heart rate are linked to susceptibility to ventricular arrhythmias. In this report, we will mainly address the association between QT dynamicity and the risk of developing torsades de pointes.

Influence of dofetilide on QT-interval duration and dispersion at various heart rates during exercise in humans

BACKGROUND

The objective of this study was to assess the influence of heart rate on QT-interval duration and dispersion during administration of the new selective potassium-channel blocker dofetilide in normal subjects.

METHODS AND RESULTS

Dofetilide 0.25 and 0.75 mg was administered for 4 days to 12 subjects in a randomized-sequence, double-blind, three-period, placebo-controlled, crossover study. QT-RR pairs were measured on study day 4 over a wide range of RR intervals obtained at rest and during an exercise test. QT-interval durations were calculated at seven predetermined RR intervals ranging from 400 ms (150 bpm) to 1000 ms (60 bpm) by use of monoexponential nonlinear curve fitting. QTmax and QTmin were calculated similarly, and QT-interval dispersion was measured as QTmax-QTmin at each predetermined RR interval. Minimal effects were found with 0.25 mg dofetilide. Two hours after administration of 0.75 mg dofetilide, QT interval was prolonged by 16.7 +/- 8.7% at a heart rate of 60 bpm (P < .01) and by 7.4 +/- 8.2% at a heart rate of 150 bpm (P < .05). QT prolongation at a heart rate of 150 bpm was less pronounced than at lower heart rates. Neither placebo nor dofetilide at either dose significantly increased QT-interval dispersion at any heart rate.

CONCLUSIONS

Dofetilide increases QT-interval duration but does not increase QT-interval dispersion in healthy subjects. QT-interval prolongation remains significant at high heart rates, although some degree of reverse rate dependence is observed at high concentrations.

In vitro electrophysiological detection of iatrogenic arrhythmogenicity

Cardiac arrhythmias and sudden death have been associated with both therapeutic and toxic doses of a number of cardiotropic and non-cardiac drugs. Generally the drug-induced electrocardiographic (ECG) alterations have been well described, whereas corresponding cellular electrophysiological effects are poorly documented or lacking. Taking into account the recent advances in the understanding of the mechanisms underlying arrhythmias and antiarrhythmic effects, suitable relationships can be established between ECG alterations and drug effects on cardiac action potential. Thus, a decrease in maximal upstroke velocity (Vmax) and membrane depolarisation leading to cellular inexcitability may slow conduction, prolong QRS interval duration and result in incessant wide QRS ventricular tachycardia. On the other hand, lengthening of the repolarisation phase and early afterdepolarisations (EADs) have been proposed as a mechanism for prolonged QT interval and subsequent Torsades de Pointes. A representative study aimed at detecting the arrthymogenic potentiality of a drug is given, by examining carefully the concentration- and frequency-dependent effects of four neuroleptics (sultopride, droperidol, thioridazine and clozapine) on Purkinje fibers and comparing them with the reported iatrogenic arrhythmias. The results showed that 10 to 100 microM sultopride and 0.01 to 1 microM droperidol exerted "pure" class III effects. In addition, higher concentrations (3 to 30 microM) of droperidol reversed the prolonging effect on repolarisation concomitantly with a dose- and frequency-dependent decrease in Vmax, action potential amplitude and resting membrane potential (class I effects) resulting in cellular inexcitability at 30 microM. Similar class I effects were induced by thioridazine and clozapine concomitantly with a slight prolonging effect on final repolarisation (class Ia effects). In the presence of sultopride (30 and 100 microM) and droperidol (0.3 to 3 microM), EADs developed at plateau level. Their incidence, amplitude and number were influenced by extracellular K or Mg concentration, stimulation frequency, modification of Ca entry (by nifedipine or isoproterenol). These experimental results fit well with clinical data although they need further development to precise underlying ionic mechanisms. Therefore, in vitro studies should be considered before clinical prospects for future drug development.

Rate-corrected QT interval: techniques and limitations

The duration of the QT interval on the surface electrocardiogram represents the time required for all ventricular depolarization and repolarization processes to occur. Among the many physiologic and pathologic factors that contribute to the QT interval, heart rate plays a major role. Several approaches have been used to correct the QT interval, all of which take into account the heart rate at which the interval is measured. The simplest and most common approach to correcting the QT interval is to divide its value by the square root of the preceding RR interval expressed in seconds, i.e., by using Bazett's formula. This calculation provides a corrected QT (QTc) interval that represents the QT interval normalized for a heart rate of 60 beats/min. However, several studies have shown that Bazett's correction formula is not optimal. Fridericia's cube-root formula has been shown to perform better in correcting the QT interval for heart rate. Other formulas require the measurement of several QT-RR pairs at various heart rates to obtain a reliable QTc interval and are therefore not easily usable. Any correction formula is likely to introduce an error in assessing the QTc interval. Although the importance of this error should not be minimized, the corrected QT interval remains useful in assessing the effects of drugs on the duration of repolarization. For this purpose, Fridericia's cube-root formula is preferable to Bazett's square-root formula.(ABSTRACT TRUNCATED AT 250 WORDS)