Termination of Reentry by Lidocaine in the Tricuspid Ring In Vitro
Role of Cycle-Length Oscillation, Fast Use-Dependent Kinetics, and Fixed Block
We hypothesized that drugs with rapid recovery kinetics from use-dependent sodium channel block could promote oscillatory termination of reentry by enhancing interval-dependent conduction. Mechanisms of termination were related to properties of the reentrant circuit. Nine adjustable reentrant preparations were used in which the canine atrial tricuspid ring was cut and then reconnected electronically by sensing activation on one side of the cut and pacing the other after an adjustable delay. The cycle length and diastolic interval during reentry were manipulated by changing this delay. Lidocaine (1.28×10−5 mol/L) significantly increased refractoriness (94±39 ms) and the slope of the conduction curve (−0.12±0.07) at the site of block during pacing. Lidocaine terminated sustained reentry by two mechanisms. Early termination resulted from increased cycle length oscillation and refractoriness (reproducible in each experiment) but only at short delays with short initial diastolic intervals. The range of delays showing this mechanism of termination was 100±48 ms. Increased cycle-length oscillation resulted from an increased slope of the conduction curve. In eight experiments, lidocaine terminated reentry by causing fixed block after 50 minutes of drug superfusion, which prevented reentry at all delays. Fixed block occurred at one of two vulnerable sites and was transiently reversed by acetylcholine. Termination due to refractory block occurred only when the initial diastolic interval was short, and termination due to fixed block developed when there was a susceptible region with a low safety factor for propagation. Fast recovery from sodium channel block promotes oscillatory termination by increasing the slope of the conduction curve.
Development of a rational and efficient approach to the clinical use of antiarrhythmic drugs has been hampered by an incomplete understanding of the basic mechanisms of antiarrhythmic drug action. This report addresses three aspects of this problem: (1) how the kinetics of use-dependent channel blockade contribute to antiarrhythmic drug action, (2) the mechanisms of drug-induced termination of reentrant tachycardias, and (3) the characteristics of the reentrant circuit that determine whether a drug will terminate the tachycardia.
Use-dependent block of sodium channels results in greater depression of excitability and conduction at fast rates for all class I antiarrhythmic drugs.1 2 3 4 However, the kinetics for onset and recovery from use-dependent block are faster for class IB drugs, such as lidocaine, than for class IA or class IC drugs.2 5 The implications for antiarrhythmic drug action have not been fully explored. Fast-recovery kinetics accentuates interval-dependent changes in conduction during the first several hundred milliseconds after a previous action potential.6
We hypothesized that this change in the shape of the conduction curve by lidocaine could promote oscillatory termination of reentry, because increasing interval-dependent conduction promotes CL oscillation during reentry.7 Increased oscillations have been seen before drug-induced termination of ventricular and supraventricular tachycardias in patients,8 9 10 11 12 which suggests that enhancement of interval-dependent conduction may be a clinically important mechanism for antiarrhythmic drug actions.
The duration of the shortest excitable gap and the lowest safety factor for propagation in the reentrant circuit are two characteristics of reentrant circuit that may determine the response to drugs.13 Prolongation of refractoriness will only terminate reentry when the DI and excitable gap are relatively short initially.14 Termination due to fixed conduction block depends on the presence of specific regions susceptible to complete block in a unique and bounded limb of the circuit.
Studies were conducted using the canine atrial tricuspid ring in vitro.7 15 The model involves reentry around a fixed anatomic barrier with a partially excitable gap. There is interval-dependent conduction during the relative refractory period in fast-response tissue. A modification called “adjustable reentry” allowed us to manipulate the CL and the duration of DI to determine mechanisms of tachycardia termination by lidocaine.14 16
Materials and Methods
Nine healthy mongrel dogs weighing 15 to 20 kg were anesthetized with pentobarbital 30 mg/kg IV. The heart was rapidly excised and dissected in cold Tyrode's solution equilibrated with 95% O2/5% CO2. The atrial tricuspid ring (see Fig 1⇓) was dissected and mounted, endocardium upward, in a special tissue bath, as previously described.7
The composition of Tyrode's solution was (mmol/L) NaCl 125, NaHCO3 24, NaHPO4 1.8, MgCl2 0.5, CaCl2 1.8, dextrose 5.5, and KCl 4.0. The solutions were bubbled with 95% O2/5% CO2 before entering the tissue bath and multiple sites around the circular tissue bath. In seven experiments, a fixed concentration of acetylcholine was present throughout the experiment.7 The concentration in experiments 1 through 7, respectively, was 5×10−8, 1×10−8, 1.5×10−7, 1×10−7, 2×10−7, 4×10−7, and 2×10−7 mol/L. The temperature of the tissue bath was maintained at 34±0.1°C, and the pH was between 7.30 and 7.38, with variation of <0.02 within each experiment.
Electrical Recording and Stimulation
Activation sequences were recorded using 10 unipolar platinum electrodes spaced equidistantly in a circle around the ring as indicated in Fig 1⇑. Electrograms were amplified with a bandwidth of 0.5 to 500 Hz (Bloom Associates). Signals were recorded on FM tape and on chart paper at 100 to 200 mm/s. Signals were also displayed on an eight-channel storage oscilloscope (5111, Tektronix). An Apple Macintosh II computer connected to an analog-to-digital converter board (National Instruments) and custom software provided on-line display of CLs measured at the site of block.
MAPs were recorded using Ag/AgCl-tipped Franz MAP catheters (EP Technologies), positioned near the site where the unidirectional block and termination of reentry occurred using a micromanipulator, and maintained at this site throughout the experiment. The site of block was determined before placing the MAP catheter. The minimum pressure needed to obtain the recording was applied. If positioning the MAP catheter altered conduction or reentry CL, the catheter was removed and repositioned. APD was measured at 90% repolarization. When action potential amplitude varied because of incomplete repolarization or variations in the peak depolarization voltage, the durations were measured at the voltage that corresponded to 90% repolarization for the largest action potential. DI was the interval between the end of the preceding action potential and the next activation measured at the same voltage used to measure APD (APD+DI=CL).
Stimulation was performed through separated polytetrafluoroethylene-coated bipolar silver wire electrodes placed between the recording electrodes and controlled by a constant current stimulator (Bloom Associates). Pacing stimuli were 2 ms in duration and three times diastolic threshold current.
Adjustable Reentry Preparation
A diagram of the preparation is shown in Fig 1⇑. Premature stimuli were delivered from several sites to determine the site with the longest ERP, which was susceptible to unidirectional block for initiation of reentry. A pacing electrode for initiating reentry was located 1 to 2 cm away. The ring was then divided 1 to 2 cm from the pacing electrode on the side opposite the site of block. Adjustable reentry was established by sensing activation on one side of the cut and pacing the other side after an adjustable electronic delay.16 A long delay resulted in a long CL and DI during reentry. Decreasing delay decreased the tachycardia CLs and the DIs. To ensure that the electronic connection did not contribute to CL oscillation or termination of reentry, we sensed a sharp narrow electrogram, displayed the sensing signal to ensure consistent timing relative to the electrogram, and stimulated with four times diastolic threshold current to minimize the stimulus to activation latency as confirmed by a nearby electrode.14
Conduction times and refractory periods were measured during pacing at CLs of 350 and 500 ms in the control period. Measurements were repeated in the presence of drug at a CL of 350 ms except when consistent capture was not possible. A CL of 500 ms was then used. Data for the same CL before and after drug exposure are reported. Stimuli were delivered at the electrodes used to connect the ring during adjustable reentry so that the paced impulse was conducted in the same direction as during reentry. ERP was measured at the site of stimulation and at the site of block. ERPs was defined as the longest coupling interval of a premature stimulus that failed to capture at the site of stimulation. ERPb was defined as the longest coupling interval of the premature impulse that failed to propagate measured at the site of block. Conduction times were measured as the difference between activation times at two adjacent electrodes along a line parallel to the direction of propagation around the ring. The APD restitution curve and the conduction curve are plots of APD and conduction times elicited by a premature stimulus (S2) as functions of the preceding DI for S1S2 intervals from 1000 ms to ERPs. The slope of the conduction curves were calculated for specified ranges of DI using simple linear regression.
Reentry was induced either by bursts of rapid pacing or by a single premature stimulus. Tachycardias were studied over a wide range of adjustable delays by establishing reentry at a long delay and reducing delay in decrements of 20 to 30 ms. The bifurcation cascade is a plot of 30 to 40 consecutive reentrant CLs at steady state for each tested delay. The bifurcation threshold was defined as the longest delay that showed persistent CL alternation. The termination threshold was defined as the longest delay that produced spontaneous termination of reentry. Reentry always terminated spontaneously at delays shorter than the termination threshold. Reentry was classified as sustained reentry if it did not terminate spontaneously during at least 2 minutes of observation at a given delay.
Lidocaine 1.28×10−5 mol/L was washed in during tachycardia at a delay 10 to 20 ms longer than the bifurcation threshold. When the tachycardia terminated, it was reinduced at least twice at the same delay. Then the delay was increased to the longest delay studied in the control period (usually 251 ms) to determine whether the tachycardia was inducible and sustained.
Fixed block was identified by the inability of impulses to conduct in either direction through the site of block during pacing at a constant CL of 800 to 1000 ms or after 2-s pauses.14 16 If the drug caused fixed conduction block in the reentrant circuit, several drops of acetylcholine (1×10−4 mol/L) were administrated directly into the 30-mL tissue bath. The effect of acetylcholine dissipated within 5 minutes. Lidocaine was then washed out by superfusing with drug-free Tyrode's solution.
Statistical comparisons between the drug period and the control period were made using paired two-tailed Student's t tests. Values of P<.05 were considered significant. Group data are summarized as mean±SD. Exponential time constants for the conduction curves were determined by fitting the data to the following equation: CT=Ae−(DI/B)+K, where CT is conduction time, B is the time constant, and A and K are other constants, We calculated the slopes of a limited portion of the conduction and restitution curves using simple linear regression.
Electrophysiological Effects of Lidocaine During Pacing
Table 1⇓ summarizes the electrophysiological effects of lidocaine during pacing at a constant CL in five experiments. Comparisons were made at a CL of 350 ms in two experiments and at 500 in three experiments. In the other four experiments, lidocaine produced conduction block before the pacing protocol was completed. Lidocaine significantly increased conduction time around the circuit. It also significantly increased both ERPs and ERPb but not APD during pacing. The difference between ERPb and APD (measured at 90% repolarization) was 44±43 ms in the control period and was increased 134±76 ms by lidocaine, indicating an increase in postrepolarization refractoriness.
The conduction time in the area with the slowest conduction increased 23±16 ms, which was 59% of total increase in conduction time, even though its conduction time was only 38% of the tissue conduction time in control and the length of this region averaged only 24% of the tissue path length.
Fig 2⇓ illustrates the effect of lidocaine on ERP. Recording electrodes from around the ring are displayed with a time line, with a MAP recording taken from between sites 2 and 3 shown at the bottom of the recording. In the control period (Fig 2A⇓) a premature impulse initiated at site 9 with a coupling interval of 200 ms reached site 2 with a coupling interval of 260 ms and did not block. Therefore, the ERP at this site was <260 ms during control. After lidocaine (Fig 2B⇓), a premature impulse with a coupling interval of 422 ms at site 2 did block long after repolarization was complete. Therefore, the ERP at this site was >422 ms, and lidocaine increased the ERP by >162 ms (422 to 260 ms).
Effects of Lidocaine on the Conduction Curve and APD Restitution Curve
The effects of lidocaine on the conduction and APD restitution curves measured at the site of slowest conduction are shown in Fig 3⇓. Conduction time was constant at long DIs, reflecting full recovery of excitability. The increasing portion of the conduction curve reflects interval-dependent conduction at shorter DIs. Lidocaine shifted the conduction curve upward at all DIs. The drug also extended the range of interval-dependent conduction by >350 ms. Therefore, it increased the slope of the conduction curve for DIs between 130 and 500 ms (zone 2). The exponential time constant of the conduction curve increased from 33 to 127 ms in this experiment.
Conduction curves across the area of slowest conduction could be measured in the control period and in the presence of lidocaine but before conduction block developed in four experiments. In the control period, the transition from interval-dependent conduction (negative slope) to constant conduction (zero slope) occurred at a DI of 173±46 ms. This transition point was increased 253±194 ms by lidocaine. The range of DIs between this transition point in the control period and this transition point in the presence of lidocaine defines zone 2 for the pair of curves illustrated in Fig 3⇑, top. The slope of the conduction curve in zone 2 was zero in control by definition. It was increased to −0.12±0.07 after lidocaine administration. The mean time constant for recovery of conduction time increased from 37±12 to 118±41 ms.
Lidocaine had very little effect on the APD restitution curve (Fig 3⇑, bottom). The shapes of the APD restitution curves were almost identical before and after lidocaine in zones 3 and 2. The slopes of the curves in zone 2 could be estimated in three experiments and were unchanged. The mean slope was 0.156±0.094 in the control period and 0.156±0.094 in the presence of lidocaine.
Termination of Reentry due to Refractory Block During the Control Period
Reducing the electronic delay during adjustable reentry decreased the CL and DI (Table 2⇓). Reducing the delay in the control period to a critical value resulted in termination of reentry at least three times in all experiments. An example is shown in Fig 4⇓. CL was constant, and reentry was sustained at delays of 251 ms (Fig 4⇓, top left) and 181 ms (Fig 4⇓, bottom left). The bifurcation threshold was 171 ms (not shown). The termination threshold was at a delay of 91 ms (Fig 4⇓, top right). Termination of reentry resulted from oscillations of CL and APD, leading to a critically short DI proximal to the site of block (Fig 4⇓, top right). The DI in the MAP recording preceding the beat that blocked was 121 ms, which is shorter than DIs not associated with block. This DI was also shorter than DIs observed during sustained reentry at longer delays, as shown in the left panels of Fig 4⇓. Termination of reentry was attributed to refractory block, because it was associated with a critical reduction of the DI just proximal to the site of block. After spontaneous termination, reentry could easily be reinduced at the same delay, but it terminated again. Sustained reentry could be induced at a longer delay.
Fig 5⇓ presents similar data from another experiment in a different form. The top panel shows the patterns of CL at selected delays, and the bottom panel shows a bifurcation diagram. CL was constant, and reentry was sustained at long delays. As delay was decreased, there was a well-defined bifurcation threshold marking the transition to CL oscillation and a reproducible termination threshold, below which reentry always terminated.
Two Phases and Mechanisms of Termination of Reentry by Lidocaine
Two mechanisms of termination of reentry were seen during exposure to lidocaine. There was an initial phase during which lidocaine increased ERP and CL oscillation, which led to termination of previously sustained reentry due to refractory block at least twice in each experiment but only in a limited range of delays just above the termination threshold of the control period. In eight of nine experiments, fixed block developed subsequently, which prevented reentry at all delays.
Termination due to Increased Cycle-Length Oscillation and Refractory Block
The initial phase of refractory block was most completely characterized in experiment 2, which did not proceed to fixed conduction block. Fig 5⇑, top, shows how lidocaine affected CL stability and persistence of reentry at four selected delays. At a long delay (251 ms), the tachycardia CL was increased by lidocaine but remained constant, and reentry was sustained. At a delay of 201 ms, lidocaine increased CL and produced slight alternation without terminating reentry. At a delay of 171 ms, slight CL alternation was converted to large variable oscillation, which terminated reentry after a long-short sequence of CLs. At a delay of 71 ms, reentry terminated spontaneously during the control period and could not be induced after exposure to lidocaine.
Fig 5⇑, bottom, shows the bifurcation cascade indicating the degree of CL stability during reentry for all delays tested in this experiment. Lidocaine increased the mean CL at all delays. Lidocaine shifted the bifurcation threshold from a delay of 181 ms to a delay of 221 ms. The termination threshold increased from a delay of 71 ms to 171 ms. The range between 71 and 171 ms represents delays with short CLs and DIs, for which reentry was sustained during the control period but was terminated by lidocaine.
In eight of nine experiments in which lidocaine produced fixed conduction block, the drug effects could not be evaluated at all delays. However, an initial phase of refractory block limited to shorter delays was still evident; lidocaine terminated reentry at shorter delays, but sustained reentry was inducible at longer delays, at which CL and DI were longer. The delay that demonstrated refractory block in the presence of lidocaine was 100±48 ms longer than the termination threshold during the control period. Therefore, the termination threshold was shifted toward longer delays by at least this amount.
Fig 6⇓ demonstrates this response to lidocaine for the experiment shown in Fig 4⇑. At a delay of 181 ms, lidocaine increased CL and produced CL oscillation (compare Fig 4⇑, bottom left, and Fig 6⇓, left). The oscillation resulted from variable conduction between sites 2 and 1, and termination resulted from block at this site after a long-short (433-ms–423-ms) sequence of CLs. Even though DI was increased in the presence of lidocaine from 196 to ≈270 ms, termination followed the shortest DI (264 ms) in the presence of lidocaine. The action potential that blocked in the MAP recording arose after full repolarization. Reentry could be reinduced at this delay but was nonsustained. Increasing the delay to 251 ms (Fig 6⇓, right) allowed induction of sustained reentry because the DI was longer. The increase in the critical DI for block from 121 ms in the control period (Fig 4⇑, top right) to 264 ms after lidocaine exposure (Fig 6⇓, right) reflects a drug-induced increase in postrepolarization refractoriness. Similar results were seen in the other eight experiments.
Lidocaine increased APD during reentry at a given delay. In the example at a delay of 181 ms, APD increased from 128 ms (Fig 4⇑, bottom left) to an average of 157 ms (Fig 6⇑, left) However, the increase in APD was entirely due to the increase in CL by the drug. The APD after drug was the same as the APD at a comparable CL at a longer delay in the control period. Among all experiments, lidocaine increased APD during reentry by 15±20 ms, associated with a 52±33-ms increase in CL (Table 2⇑). The DI was increased 35±26 ms.
Termination due to Fixed Block
In eight of nine experiments, lidocaine terminated sustained reentry by causing fixed conduction block after an average of 50 minutes (Table 2⇑). Fig 7⇓, top left, is from the same experiment as Figs 4 and 6⇑⇑. Fixed conduction block developed between sites 1 and 2 at the same site where refractory block and slow and variable conduction occurred earlier. In all experiments, impulses blocked at the site of fixed block, even during pacing at long CLs of 800 to 1000 ms or after a 2-second pause. In some experiments, longer pauses up to 5 seconds were inserted without observing recovery of conduction. We confirmed that fixed block occurred at one specific site (except in experiment 9) and was bidirectional by pacing on each side of the site of block and showing that all tissues could be activated but impulses from each direction blocked at this site. In experiment 9, there were two sites of fixed block. Fixed block made reentry impossible at any delay. Fixed block could be transiently reversed by a bolus of acetylcholine (Fig 7⇓, bottom left) in all eight experiments but reappeared within 5 minutes as the acetylcholine effect dissipated. Acetylcholine shortened APD, as can be seen by comparing Fig 7⇓, bottom left, and Fig 4⇑, top left. In the presence of acetylcholine, APD was 124 ms at a pacing CL of 400 ms compared with an APD of 139 ms at a slightly shorter CL of 388 ms in the control period. When lidocaine was washed out, reentry was again inducible with CLs similar to those seen in the control period (Fig 4⇑, bottom left).
Characteristics of the Site of Fixed Block
The sites that developed fixed block had slower conduction during the control period than other sites in the ring. There was also a greater increase and more variability of conduction time at these sites during drug exposure. In the example shown in Figs 4, 6, and 7⇑⇑⇑, fixed block occurred between sites 2 and 1. The conduction time from site 3 to 1 was 72 ms initially or 50% of the total tissue conduction time (Fig 4⇑, bottom left), and 50% of the 100-ms increase in CL occurred here (Fig 6⇑, left).
Fixed block always occurred in the anterolateral or posterolateral parts of the ring (sites B and D in Fig 1⇑) (see Table 2⇑). Lidocaine increased the mean conduction time at this site by 31±17 ms (from 60±18 ms in the control period), which accounted for 59% of the total increase in cycle length. Conduction time at this site was increased from 36% to 41% of the total conduction time by lidocaine. The smooth tricuspid anulus lies between the valve orifice and the pectinate muscle in this region and is narrower than in other parts of the ring. In five experiments, block occurred as the impulse entered this zone, and in two experiments, it occurred as it left this zone. However, once block developed, paced impulses blocked from both directions. Both sites were excluded from the circuit in experiment 2, which was the only one that did not develop fixed block.
Lidocaine terminated sustained reentry in this model in two phases by different mechanisms: (1) termination due to refractory block caused by increased postrepolarization refractoriness and CL oscillation and (2) termination due to fixed conduction block at specific vulnerable sites. Success of each mechanism is dependent on the presence of a favorable substrate. Prolongation of refractoriness terminated reentry at only shorter delays with shorter initial DIs during reentry. Fixed block terminated reentry at all delays, but block developed only if certain vulnerable sites were present. Fixed block was reversed by the neurotransmitter acetylcholine, which suggests that it may be vulnerable to autonomic manipulation.
Increasing the Slope of the Conduction Curve Promotes Oscillatory Termination
Lidocaine produced refractory block at delays that supported sustained reentry in control. Termination was attributed to refractory block when it was associated with a critically short DI near the site of block and when reentry was inducible and sustained at longer delays with longer DIs. The increase in the critical DI for block after lidocaine reflects an increase in the postrepolarization component of refractoriness. However, the increased CL oscillation also contributed to block by transiently decreasing the DI below the critical value.
CL oscillations and oscillatory termination of reentry depend on the slopes of the conduction and restitution curves that describe the variations in APD and conduction.7 The steeper the slopes, the greater the tendency for oscillation.17 18 A steeper conduction curve means that a given perturbation of the DI results in a greater change in conduction, which promotes CL oscillation. Changes in APD contribute to the magnitude of oscillation by influencing the duration of the DI during the next cycle.
In the present study, lidocaine increased the CL more than the APD; thus, DI at the site of block was increased. This would have decreased CL oscillation if there had been no change in the shapes of the conduction and restitution curves, because the slopes of both curves decrease at longer DIs. The increase in CL oscillation by lidocaine can be explained by the increase in the slope of the conduction curve for DIs between 150 and 350 ms, where oscillations developed and refractory block occurred.
Contribution of Kinetics of Use-Dependent Block to Oscillatory Termination of Reentry
Discussions of the significance of use dependence for antiarrhythmic action have usually emphasized greater sodium channel blockade at fast heart rates or in acidotic or depolarized tissue. The present study demonstrates that increasing the slope of the conduction curve by drugs with fast use-dependent kinetics may promote oscillatory termination of reentry.
The slope of the conduction curve can be increased by drugs with fast use dependence but is less pronounced with class IA or IC drugs with slower kinetics. Interval-dependent conduction results largely from changes in sodium channel availability. A drug with slow use-dependent recovery kinetics produces a small increment of channel blockade during each depolarization, followed by very slight and gradual recovery from channel blockade during the DI. Therefore, changes in channel blockade do not significantly alter the shape of the conduction curve over the first several hundred milliseconds after repolarization. For instance, procainamide shifted the resetting response curve upward without changing its shape in a study of flutter around the tricuspid ring.19 Procainamide did not change the shape of the strength interval curve in human atria.20
To produce comparable suppression of upstroke velocity, a fast kinetic drug must produce a greater increment of sodium channel blockade during each depolarization because of rapid dissociation of the drug from the channel during diastole before the next depolarization. The larger magnitude and faster rate of sodium channel unblocking during the first few hundred milliseconds after repolarization can significantly prolong the period of interval-dependent conduction. Therefore, only sodium channel–blocking drugs with fast recovery kinetics for use-dependent block will promote oscillatory termination by changes in the shape of the conduction curve.
The time constant of recovery of dV/dtmax is 10 to 40 ms in normal cells. It is increased by lidocaine to ≈100 ms in well-polarized fibers and twice that in partially depolarized fibers.1 21 These changes are comparable to the exponential time constants describing the conduction curves in this study: 37 ms for the control period and 118 ms after exposure to lidocaine. Accentuation of interval-dependent conduction by lidocaine has been reported6 22 but has not previously been related to the magnitude of CL oscillation or termination of reentry.
These data suggest that antiarrhythmic efficacy against reentrant tachycardias depends on the dynamic properties reflected in the conduction and restitution curves as well as steady state changes in conduction and ERP. In a previous study, we showed that the class III drug d-sotalol also promotes oscillatory termination in part by increasing the slope of the APD restitution curve.14 It also increased APD more than CL so that the DI decreased during reentry. This forced the reentrant impulse to operate on steeper portions of both curves. The shape of the conduction curve was not significantly changed.
Substrate for Lidocaine-Induced Fixed Conduction Block
Fixed block is distinct from refractory block in that it is time independent and does not resolve if the diastolic recovery time is increased. Fixed block resulted from a drug effect and not deterioration of the preparation, as it was resolved with washout of lidocaine. The time delay before developing fixed block is presumably related to slow equilibration of the drug concentration in the bath. Fixed conduction block always occurred at one of two critical sites in the circuit. Such vulnerable sites responded differently than other sites to moderate concentrations of lidocaine. The ability to produce fixed block at these concentrations indicates a low safety factor for propagation at these sites. A low safety factor may result from depolarization of the resting potential, poor intercellular coupling due to fiber orientation or structural discontinuities, or impedance mismatches. Which of these factors contributed to the vulnerability to fixed block in the present study is not known.
These sites did have slower conduction, and lidocaine increased the conduction time more than at other sites. There was also a greater increase in ERP at these sites. Drug-induced termination has also occurred at one of these two sites during atrial flutter around the tricuspid ring in vivo.23 24 25 26 The consistent locations of the vulnerable sites suggests that structural factors were involved. Most of the tricuspid anulus is composed of a deeper layer of circumferential fibers and a superficial layer of radial orientated fibers perpendicular to the deep layer.27 Previous structural descriptions of the ring do not reveal a specific reason for block at these sites except that the anulus is relatively narrow here (see Fig 1⇑) and is bounded by the tricuspid orifice and the irregularly oriented pectinate muscles. This may allow a relatively short line of block connecting these inner and outer fixed boundaries to terminate reentry.
The ability of acetylcholine to overcome fixed block suggests that at least moderate resting membrane potential depolarization may have been involved, although it may not be the primary factor. Acetylcholine will hyperpolarize the resting potential of depolarized cells but not of well polarized cells. Acetylcholine shortened APD. Although such an effect might be expected to shorten the ERP by a comparable amount, it does not explain how tissue that demonstrated block lasting >1000 ms after the last depolarization can be altered to recover excitability and support reentry at cycle lengths <400 ms.
The effect of acetylcholine on fixed block is an example of autonomic modulation and reversal of antiarrhythmic drug effects. Isoproterenol has been shown to reverse drug effects on APD and refractory period in humans28 and animals29 and on conduction30 in animals. Isoproterenol may also reverse drug suppression of inducible ventricular tachycardia,31 atrial fibrillation, and reciprocating atrioventricular reentrant tachycardia.28 Identification of factors that can overcome fixed block can suggest adjunctive therapy that would make the antiarrhythmic drug strategy more effective.
Comparison With Experimental Studies
In the present study, lidocaine slowed conduction, especially at sites with slower initial conduction. It also prolonged postrepolarization refractoriness at these sites. Similar effects have been previously observed in atrial and ventricular muscle, especially in settings where the safety factor for propagation was reduced. Lidocaine causes a slight decrease in conduction velocity and little change in ERP in normal ventricular or His Purkinje tissue but markedly slows conduction, produces conduction block, and prolongs the ERP in infarcted, ischemic, or hypoxic tissue.32 33 34 35 Many of these studies have also demonstrated antiarrhythmic actions against reentrant tachycardias. Other studies have shown that depression of upstroke velocity of the action potential by lidocaine is greater in depolarized or acidotic tissue.1 21
In the canine atrium, high concentrations of lidocaine prolonged the ERP more than they decreased conduction velocity, thereby increasing the electrical wavelength (conduction velocity×ERP).36 Shirayama et al5 showed that lidocaine increased postrepolarization refractoriness at rapid rates and had faster kinetics for recovery from use-dependent block than did class 1A and 1C drugs in the guinea pig atrium.
Several aspects of the present study are unique. Using a simple accessible and well-bounded reentrant circuit, we could identify and record from specific sites vulnerable to block. The use of the adjustable reentry preparation allowed us to simulate different-sized circuits to promote the conditions for refractory block and to differentiate termination due to fixed or refractory block. Simultaneous recording of activation sequence and monophasic action potentials from critical sites allows correlation of cellular responses with the dynamic behavior of the reentrant impulse. An important observation in the present study was that the prolongation of ERP and the susceptibility to fixed block were greater at specific sites responsible for termination than at an arbitrary site of pacing.
Our purpose in using the atrial tricuspid ring model was not to demonstrate the efficacy of a particular drug in treating supraventricular arrhythmias but to investigate antiarrhythmic mechanisms in a simple accessible model with well-defined properties. Reentry occurs around an anatomic obstacle with a long partially excitable gap and marked interval-dependent conduction during the relative refractory period in fast response tissue. These characteristics are commonly found in human reentrant arrhythmias of atrial flutter37 38 39 and ventricular tachycardia40 41 42 due to ischemia or infarction. However, the properties of this model are not representative of all forms of reentry. Drug effects may differ in atrial and ventricular tissue and in different species. Functional barrier reentry differs from anatomic barrier reentry. The susceptibility to termination by drugs may be different in circuits that do not have areas with a narrow isthmus between inexcitable boundaries. Nevertheless, by understanding mechanisms of antiarrhythmic drug action in well-defined experimental models, it should be possible to better evaluate observations in clinical arrhythmias.
Limitations of the Present Study
It would have been preferable to report changes in the ERP and excitable gap during reentry rather than changes in APD and DI. However, ERP cannot be measured without perturbing the tachycardia and possibly terminating it artificially. Furthermore, during CL oscillation, the ERP cannot be measured and changes from beat to beat. Therefore, DI was used as a surrogate measure for excitable gap with the understanding that lidocaine changes the relationship between the ERP and APD. Thus, refractory block was identified by associating it with a critically short DI rather than being able to show that CL oscillation transiently reduced the excitable gap to zero at the site of block. The rapid production of fixed block by lidocaine in some experiments also prevented completion of resetting protocols at longer delays that would have allowed direct evaluation of the effect of lidocaine on the excitable gap. However, the observation that at short delays lidocaine terminated previously sustained reentry due to refractory block indirectly indicates that the excitable gap was reduced.
Implications of the Present Study
The present study indicates that specific characteristics of reentrant circuits influence their response to drugs. Two important characteristics are the duration of the excitable gap and the presence of vulnerable tissue with a low safety factor for propagation. Other factors yet to be identified may also be important. If a reentrant circuit has the properties of a short excitable gap and a low safety for propagation, antiarrhythmic drugs may be effective in terminating the reentrant tachycardia by increasing ERP or by facilitating the development of fixed conduction block. Moreover, special attention to the use-dependent kinetics of antiarrhythmic drugs may advance our understanding of the mechanisms of antiarrhythmic action of such agents.
Selected Abbreviations and Acronyms
|APD||=||action potential duration|
|ERP||=||effective refractory period|
|MAP||=||monophasic action potential|
This study was supported by grant HL-38386 from the National Heart, Lung, and Blood Institute, Bethesda Md.
- Received May 6, 1996.
- Accepted October 28, 1996.
Chen CM, Gettes LS, Katzung BG. Effect of lidocaine and quinidine on steady-state characteristics and recovery kinetics of (dV/dt)max in guinea pig ventricular myocardium. Circ Res. 1975;37:20-29.
Hondeghem L, Katzung BG. Test of a model of antiarrhythmic drug action: effects of quinidine and lidocaine on myocardial conduction. Circulation. 1980;61:1217-1224.
Shirayama T, Inoue D, Inoue M, Tatsumi T, Yamahara Y, Asayama J, Katsume H, Nakagawa M. Electrophysiological effects of sodium channel blockers on guinea pig left atrium. J Pharmacol Exp Ther. 1991;259:884-893.
Davis J, Matsubara T, Scheinman MM, Katzung B, Hondeghem LH. Use-dependent effects of lidocaine on conduction in canine myocardium: application of the modulated receptor hypothesis in vivo. Circulation. 1986;74:205-214.
Frame LH, Simson MB. Oscillations of conduction, action potential duration and refractoriness: a mechanism for spontaneous termination of reentrant tachycardias. Circulation. 1988;78:1277-1287.
Duff HJ, Mitchell LB, Gillis AM, Sheldon RS, Chudleigh L, Cassidy P, Chiamvimonvat N, Wyse DG. Electrocardiographic correlates of spontaneous termination of ventricular tachycardia in patients with coronary artery disease. Circulation. 1993;88:1054-1062.
Vohra J, Hunt D, Stuckey J, Sloman G. Cycle length alternation in supraventricular tachycardia after administration of verapamil. Br Heart J. 1974;36:570-576.
Rinkenberger RL, Prystowsky EN, Heger JJ, Troup PJ, Jackman WM, Zipes DP. Effects of intravenous and chronic oral verapamil administration in patients with supraventricular tachyarrhythmias. Circulation. 1980;62:996-1010.
Gavrilescu S, Cotoi S, Pop T. Monophasic action potential of the right atrium in paroxysmal atrial flutter and fibrillation. Br Heart J. 1973;35:585-589.
Task Force of the Working Group on Arrhythmias of the European Society of Cardiology. The Sicilian Gambit: a new approach to the classification of antiarrhythmic drugs based on their actions on arrhythmogenic mechanisms. Circulation. 1991;84:1831-1851.
Fei H, Frame LH. D-Sotalol terminates reentry by two mechanisms with different dependence on the duration of the excitable gap. J Pharmacol Exp Ther. 1996;277:174-185.
Frame LH, Page RL, Boyden PA, Fenoglio JJ Jr, Hoffman BF. Circus movement in the canine atrium around the tricuspid ring during experimental atrial flutter and during reentry in vitro. Circulation. 1987;76:1155-1175.
Simson MB, Shinnar M, Frame LH. Why do some reentrant tachycardias stop? Circulation. 1988;78(suppl II):II-156. Abstract.
Kus T, Derakhchan K, Bouchard C, Page P. Effects of procainamide on refractoriness, conduction, and excitable gap in canine atrial reentrant tachycardia. Pacing Clin Electrophysiol. 1991;14(suppl II):II-1707-II-1713.
Buxton AE, Marchlinski FE, Miller JM, Morrison JF, Frame LH, Josephson ME. The human atrial strength-interval relationship: influence of cycle length and procainamide. Circulation. 1989;79:271-280.
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.
Allen DJ, Brennan FJ, Wit AL. Actions of lidocaine on transmembrane potentials of subendocardial Purkinje fibers surviving infarcted canine hearts. Circ Res. 1978;4:470-481.
Spinelli W, Hoffman BF. Mechanisms of termination of reentrant atrial arrhythmias by class I and class II antiarrhythmic agents. Circ Res. 1989;65:1565-1579.
Wu KM, Hoffman BF. Effect of procainamide and N-acetylprocainamide on atrial flutter: studies in vivo and in vitro. Circulation. 1987;76:1397-1408.
Racker DK, Ursell PC, Hoffman BF. Anatomy of the tricuspid annulus: circumferential myofibers as the structural basis for atrial flutter in a canine model. Circulation. 1991;84:841-851.
Sanguinetti MC, Jurkiewicz NK, Scott A, Siegl PKS. Isoproterenol antagonizes prolongation of refractory period by the class III antiarrhythmic agent E-4031 in guinea pig myocytes: mechanisms of action. Circ Res. 1991;68:77-84.
Kupersmith J, Antman EM, Hoffman BF. In vivo electrophysiological effects of lidocaine in canine acute myocardial infarction. Circ Res. 1975;36:84-91.
Cardinal R, Janse MJ, van Eeden I, Werner G, d'Alnoncourt CN, Durrer D. The effects of lidocaine on intracellular and extracellular potentials, activation, and ventricular arrhythmias during acute regional ischemia in the isolated porcine heart. Circ Res. 1981;49:792-806.
Rensma PL, Allessie MA, Lammers WJEP, Bonke FIM, Schalij MJ. Length of excitation wave and susceptibility to reentrant atrial arrhythmias in normal conscious dogs. Circ Res. 1988;62:395-410.
Wells JL, MacLean WAH, James TN, Waldo AL. Characterization of atrial flutter: studies in man after open heart surgery using fixed atrial electrodes. Circulation. 1979;60:665-673.
Disertori M, Inama G, Vergara G, Guarnerio M, Favero AD, Furlanello F. Evidence of a reentry circuit in the common type of atrial flutter in man. Circulation. 1983;67:434-440.
Almendral JM, Stamato NJ, Rosenthal ME, Marchlinski FE, Miller JM, Josephson ME. Resetting response patterns during sustained ventricular tachycardia: relation to the excitable gap. Circulation. 1986;74:722-730.
Spear JF, Horowitz LN, Hodess AB, MacVaugh H, Moore EN. Cellular electrophysiology of human myocardial infarction. Circulation. 1979;59:247-256.