Cycle length dynamics and spatial stability at the onset of postinfarction monomorphic ventricular tachycardias induced in patients and canine preparations

Universal mechanisms for self-termination of rapid cardiac rhythm

Excitable media sustain circulating waves. In the heart, sustained circulating waves can lead to serious impairment or even death. To investigate factors affecting the stability of such waves, we have used optogenetic techniques to stimulate a region at the apex of a mouse heart at a fixed delay after the detection of excitation at the base of the heart. For long delays, rapid circulating rhythms can be sustained, whereas for shorter delays, there are paroxysmal bursts of activity that start and stop spontaneously. By considering the dependence of the action potential and conduction velocity on the preceding recovery time using restitution curves, as well as the reduced excitability (fatigue) due to the rapid excitation, we model prominent features of the dynamics including alternation of the duration of the excited phases and conduction times, as well as termination of the bursts for short delays. We propose that this illustrates universal mechanisms that exist in biological systems for the self-termination of such activities.

Alternans amplification following a two-stimulus protocol in a one-dimensional cardiac ionic model of reentry: from annihilation to double-wave quasiperiodic reentry

Electrical pacing is a common procedure in both experimental and clinical settings to study and/or annihilate anatomical reentry. A previous study [Comtois and Vinet, Chaos 12, 903 (2002)] has described new ways to terminate reentry in a one-dimensional loop model by a protocol consisting of only two stimulations. Annihilation of the reentrant activity was much more likely with these new scenarios than through a unidirectional block. This paper investigates the sensitivity of these scenarios of annihilation to the length of the pathway. It shows that double-pulse stimulation can stop the reentry if the circuit is shorter than a limiting length. Beyond this upper limit, stimulation rather yields sustained double-wave reentry. The same dynamical mechanism, labeled alternans amplification, is found to be responsible for these two types of post-stimulus dynamics.

Myocardial electrical alteration in canine preparations with combined chronic rapid pacing and progressive coronary artery occlusion

Our objective was to create an animal preparation displaying long-term electrical alterations after chronic regional energetic stress without myocardial scarring. An Ameroid (AM) constrictor was implanted around the left circumflex coronary artery (LCx) 2 wk before chronic rapid ventricular pacing (CRP) was initiated at 240 beats/min for 4 wk (CRP-AM). Comparisons were made with healthy canines and canines with either AM or CRP. Unipolar electrograms were recorded from 191 sites in the LCx territory in open-chest, anesthetized animals during sinus rhythm and while pacing at 120–150 beats/min, with bouts of transient rapid pacing (TRP; 240/min). In CRP-AM and AM, ST segment elevation was identified at central sites and ST depression at peripheral sites, both increasing with TRP. In CRP-AM and CRP, the maximum negative slope of unipolar activation complexes was significantly depressed and activation-recovery intervals prolonged. Areas of inexcitability as well as irregular isocontour patterns displaying localized activation-recovery intervals shortening and gradients >20 ms between neighboring sites were identified in one-third of CRP-AM at slow rate, with increasing incidence and magnitude in response to TRP. In CRP-AM, programmed stimulation-induced marked conduction delay and block as well as polymorphic ventricular tachycardias, which stabilized into monomorphic tachycardias with the use of lidocaine or procainamide. Whole cell Na+ current and channel protein expression were reduced in CRP-AM and CRP. Despite complete constrictor closure, small areas of necrosis were detected in a minority of CRP-AM. Long-term electrical alterations and their exacerbation by TRP contribute to arrhythmia formation in collateral-dependent myocardium subjected to chronic tachycardic stress.

Functional reentry in cultured monolayers of neonatal rat cardiac cells

Previous studies of reentrant arrhythmias in the heart have been performed in computer models and tissue experiments. We hypothesized that confluent monolayers of cardiac cells can provide a simple, controlled, and reproducible experimental model of reentry. Neonatal rat ventricular cells were cultured on 22-mm-diameter coverslips and stained with the voltage-sensitive dye RH-237. Recordings of transmembrane potentials were obtained from 61 sites with the use of a contact fluorescence imaging system. An electrical field stimulus, followed by a point stimulus, induced 39 episodes of sustained reentry and 21 episodes of nonsustained reentry. Sustained reentry consisted of single-loop (n = 18 monolayers) or figure-of-eight (n = 4) patterns. The cycle length, action potential duration at 80% repolarization, and conduction velocity were (in means +/- SE) 358 +/- 33 ms, 118 +/- 12 ms, and 12.9 +/- 1.0 cm/s for single loop and 311 +/- 78 ms, 137 +/- 18 ms, and 7.8 +/- 1.3 cm/s for figure-of-eight, respectively. Electrical termination by 6- to 13-V/cm field pulses or 15- to 20-V point stimuli was successful in 60% of the attempts. In summary, highly stable reentry can be induced, sustained for extensive periods of time, and electrically terminated in monolayers of cultured neonatal rat cardiac myocytes.

Spatiotemporal dynamics of reentrant ventricular tachycardias in canine myocardial infarction: pharmacological modulation

During the transition from a slow to rapid depolarization rhythm, rate-dependent sodium channel blockade develops progressively and increases from beat to beat under procainamide but more abruptly under lidocaine. We investigated the consequences of such differences on the dynamic course and stability of reentrant tachycardias at their onset. Procainamide and lidocaine were infused to equipotent plasma concentrations in canines with three-day-old myocardial infarction. We measured the activation times (ms) and maximum slopes of negative deflections in activation complexes (absolute value: /-dV/dt(max)/ in mV/ms) in 191 unipolar electrograms recorded from ischemically damaged subepicardial muscle during programmed stimulation inducing reentrant tachycardias. Procainamide caused a greater reduction in /-dV/dt(max)/ than did lidocaine in the responses to basic stimulation, and it favored the occurrence of cycle length prolongation at tachycardia onset as the /-dV/dt(max)/ decreased progressively in successive beats. This resulted in conduction block and tachycardia termination in three of eight preparations. In contrast, lidocaine caused a greater depression in /-dV/dt(max)/ in response to closely coupled extrastimuli, but /-dV/dt(max)/ remained constant or even improved thereafter, and none of the tachycardias terminated spontaneously under lidocaine (n = 9). However, the reentrant circuits remained spatially unstable, and lidocaine favored the occurrence of cycle length dynamics displaying constant or decreasing trends. This study supports the notion that cycle length dynamics at tachycardia onset are determined by the properties of the reentrant substrate and their pharmacological modulation.

Onset dynamics of reentrant tachycardia and rate-dependent conduction changes in canine ventricular muscle: effects of Na+ and Ca2+ channel blockade

To show that cycle-length (CL) prolongation occurring at the onset of reentrant tachycardias may be associated with an increase in conduction time (CT), and to investigate the involvement of Na+ and Ca2+ channel activity, reentrant activity was induced by programmed stimulation in thin ventricular muscle slices with a central cryothermal lesion, as documented with 7 to 12 bipolar recordings. We studied the course of the CL measured in successive tachycardia beats, as well as the course of conduction times after abrupt transition from a pacing CL of 1,000 to 400 ms (pacing paradigm). The majority of the tachycardias displayed a dynamic behavior in which CL increased progressively, with an exponential rate constant of 37 +/- 35 beats (mean +/- SD), stabilizing at 325 +/- 67 ms after a total increase of 17 +/- 17 ms. In the pacing paradigm, CT was prolonged from 68 +/- 21 ms to 79 +/- 24 ms according to a biphasic course consisting of an abrupt increase in the first response to 400 ms, followed up by an exponential increase, stabilizing with a rate constant of 18 +/- 23 beats. Lidocaine 5 x 10(-5) mol/L induced an increase in steady-state CT, which was not further modified by adding verapamil 10(-5) mol/L. However, verapamil prolonged the rate constant of the exponential course by 60 +/- 40 beats. Thus, the onset dynamics of reentrant tachycardias share common features with the dynamic behavior of CT in the pacing paradigm, in which both Na+ channel activity and Ca2+-modulated cellular coupling appear to be involved.

Cycle length dynamics at the onset of postinfarction ventricular tachycardias induced in canines: dependence on interval-dependent excitation properties of the reentrant substrate

INTRODUCTION

Postinfarction monomorphic ventricular tachycardias induced by programmed stimulation may display initial cycle length (CL) variations before stabilizing.

METHODS AND RESULTS

To show that tachycardia onset dynamics depend on rate-dependent electrical properties of the reentrant substrate, we extracted activation times and maximum negative slopes of local activation complexes (-dV/dt(max)) from 191 unipolar electrograms recorded in the anterior left ventricular wall of anesthetized, 3-day-old infarct canine preparations. Measurements were made of the responses to programmed stimulation, as well as in early and later beats of tachycardias, which displayed either a constant trend in CL (group A, n = 5 preparations) or one in which CL prolongation occurred according to an exponential course before stabilizing (group B, n = 9). Stimulation protocols inducing the tachycardias were more aggressive and their CL was significantly shorter (CL = 159 +/- 24 msec) in group A than in group B (stabilized CL = 206 +/- 34 msec). Reentrant activity occurred in subepicardial areas in which the absolute value of -dV/dt(max) (absolute value(-dV/dtmax)) was heterogeneously depressed (<2 mV/msec). Absolute value(-dV/dtmax) was reduced and activation delay increased in the successive responses to extrastimuli. Further reductions in absolute value(-dV/dtmax) (10% to 23%) were shown to occur between early and later beats in 5 of the 9 tachycardias in group B (no change in the 4 others), and they were associated with localized prolongation of conduction times in reentrant pathways. In contrast, absolute value(-dV/dtmax) improved in all group A tachycardias (7% to 25%).

CONCLUSION

This study provides evidence that the onset dynamics of postinfarction ventricular tachycardias are determined by interval-dependent electrical changes occurring in the reentrant substrate.

The class III effect of azimilide is not associated with reverse use-dependence in open-chest dogs

Certain class III antiarrhythmic agents manifest loss of effect at short cycle lengths (CLs). This effect may limit their efficacy in the presence of tachycardia. We studied the frequency-dependent effect of azimilide (NE-10064), a new class III agent, on the right ventricular monophasic action potential (APD90) in 12 open-chest dogs. The monophasic action-potential duration at different pacing CLs (140-400 ms), during sinus rhythm, and ventricular fibrillation CL (VFCL) from left epicardial electrograms were recorded before and after increasing doses of intravenous azimilide. At pacing CL of 400 ms, APD90 was significantly prolonged after 7, 17, and 30 mg/kg of azimilide by 5.4, 7.7, and 10.7%, respectively. The extent of APD90 prolongation was independent of rate. Azimilide increased the APD90 by similar amounts at CL of 400 ms and at the fastest possible stimulation rate maintaining 1:1 capture (mean, 171 +/- 23 ms): by 2.6 +/- 8.6% and 5.6 +/- 5.9% at 2 mg/kg, 5.4 +/- 4.8% and 4.8 +/- 4.7% at 7 mg/kg, 7.7 +/- 5.6% and 9.9 +/-4.5% at 17 mg/kg, and 10.7 +/- 2.6% and 19.3 +/- 11.9% at 30 mg/kg, respectively. Azimilide caused no changes in arterial blood pressure or heart rate. Azimilide prolongs APD90 even at very short CLs. The absence of reverse use-dependence of effect on APD90 may have clinical importance.

Underdetection of ventricular tachycardia using a 40 ms stability criterion: effect of antiarrhythmic therapy

Inappropriate shocks can complicate cardioverter defibrillator therapy. Among solutions proposed to avoid oversensing are algorithms to reduce inappropriate detection of atrial fibrillation (AF) or sinus tachycardia. In patients not on antiarrythmic drugs, an interval stability criterion of 40 ms has been validated with the Medtronic PCD to discriminate ventricular tachycardia (VT) from AF. With this algorithm, VT is considered stable if no interval varies from one of the three preceding intervals by more than 40 ms. If an interval does not fulfill this criterion, the VT event counter is reset to zero. The aim of this study was to investigate the incidence of underdetection when this criterion is applied in patients treated with antiarrhythmic drugs. We studied 132 sustained monomorphic VTs induced in 42 patients during 101 electrophysiological studies (EPS). EPS were performed without treatment (group I, 24 patients, 44 VTs); on Class Ia drug (group II, 17 patients, 24 VTs); Class Ic drug (group III, 22 patients, 39 VTs); or sotalol (group IV, 17 patients, 25 VTs). The endocardial electrogram of all VT episodes was digitized and the stability algorithm was applied. The reset arrhythmias were distributed among no delay, small, moderate (< 10 s) and important (> 15 s) delay in VT detection. The relation between drug use and reset was analyzed. Reset was found in 86 (65%) of induced VTs. No difference in heart rate or induction mode was shown between reset and nonreset VTs. There was a significative association between drug use and reset probability (Chi2 significantly different, P < 0.05). In patients treated with Class Ic drugs, the probability of finding an important delay in VT detection was 12.5% versus 0% in nontreated patients or in patients treated with sotalol. We conclude that a stability criterion of 40 ms is probably safe in nontreated patients but should be used with caution in patients treated with antiarrhythmics, especially in the presence of Class Ic drugs.