Biphasic Defibrillation Waveforms Reduce Shock-Induced Response Duration Dispersion Between Low and High Shock Intensities
Abstract Mechanisms underlying defibrillation threshold reduction with biphasic waveforms remain unclear. The interaction of local shock-induced voltage gradients, which change with distance from the shocking electrode, and the state of membrane repolarization results in different cellular responses that may influence the success of defibrillation. We used intracellular microelectrodes and S1S2 pacing protocols in myocardial cell aggregates to determine the effects of shock intensity and waveform on refractory period responses during simulated fibrillation (3 s of S1 pacing at 180-ms cycle length). We simulated defibrillation by electric field stimulation S2 using 8-ms monophasic (MS2) and 4/4 biphasic (BS2) waveforms (65% total tilt) delivered at intensities of 1.5, 3, and 5 times S1 diastolic threshold, or ≈2 to 7 V/cm. Responses following MS2 varied with S2 intensity and coupling interval (P<.001). When averaged over the last 10 ms of the refractory period, MS2 produced a negligible response (8.8±1.4 ms) at 1.5 times diastolic threshold and a prolonged response (53.0±3.1 ms) at 5 times diastolic threshold (P<.01). In contrast, BS2 response duration did not change significantly (P=NS) between 1.5 times diastolic threshold (35.1±12.6 ms) and 5 times diastolic threshold (46.2±2.7 ms). Our results suggest that biphasic waveforms not only prolong response duration at low shock intensity but reduce dispersion of refractoriness produced by differing local potential gradients generated by defibrillation shocks compared with monophasic waveforms. Preventing dispersion of refractoriness and prolonging shock-induced responses may improve biphasic waveform efficacy at low shock intensity.
Results from human trials1 2 confirm animal experiments3 4 showing that biphasic waveforms lower defibrillation threshold compared with monophasic waveforms. The decreased defibrillation threshold with biphasic waveforms allows implantation of automatic implantable cardioverter-defibrillators in patients for which the defibrillation threshold is too high with monophasic waveforms5 and also permits the development of smaller defibrillators. However, mechanisms for the improved efficacy of biphasic waveforms remain unclear.
Refractory period membrane excitation leading to action potential prolongation (“extension of refractoriness” hypothesis) has been proposed as a mechanism for defibrillation.6 7 8 9 10 By prolonging the action potential and thus refractoriness, the shock blocks fibrillation wavefront(s) and stops fibrillation. It is also important that the defibrillating shock avoids conditions that are likely to lead to refibrillation. Dispersion of refractoriness is one known facilitator of reentry leading to fibrillation. A defibrillation shock creates differing potential gradients in different regions of the heart, with regions of higher intensity near the electrodes.11 Therefore, dispersion of refractoriness can be produced by inhomogeneous action potential prolongation caused by the interaction of different intensity shocks with different stages of membrane repolarization. The “upper limit of vulnerability” hypothesis12 13 states that the shock must also be strong enough to avoid dispersion of refractoriness leading to refibrillation.
Computer model studies in which the ventricular action potential was simulated in an isolated membrane patch have suggested that at the same S1S2 coupling interval, the duration of the shock-induced response is intensity dependent8 but that a significant interaction occurs between S2 shock intensity and coupling interval in determining shock-induced response duration.14 Because of this interaction, monophasic waveforms produced significant dispersion of repolarization for cells exposed to differing shock intensity and coupling intervals. Biphasic waveforms reduced the dispersion of refractoriness by prolonging APD between low– and high–current density regions more uniformly than did monophasic waveforms.
Experimental studies using current injection stimulation of myocardial cells paced at simulated normal sinus rhythm9 indicated that the results obtained in the computer model might explain results obtained in actual myocardial cells. However, cellular responses to current injection, where all portions of the sarcolemma are either depolarized or hyperpolarized by the stimulus, may differ significantly from responses to electric field stimulation, where opposite ends of the cell are subjected to opposite membrane polarization during the stimulus.15 16 17
A preliminary study18 suggested that similar responses to refractory period stimulation are also produced by electric field stimulation, which more closely simulates defibrillation fields. However, during fibrillation in both humans19 and animal models, CL is very short (100 to 200 ms), and there is no diastolic interval between action potentials. Because action potentials during early ventricular fibrillation show decreased amplitude and duration compared with action potentials during normal sinus rhythm,19 20 21 22 23 24 refractory period responses to S2 stimulation may also differ considerably.
Biphasic waveforms do not enhance action potential prolongation compared with monophasic waveforms when delivered to fibrillating hearts at intensities (5 V/cm) that are above the monophasic waveform defibrillation threshold.23 However, because the biphasic waveform defibrillation threshold is lower than the monophasic waveform threshold, there is a “window” of low shock intensities between 2 and 3 V/cm25 and 5 and 9 V/cm,26 27 in which the biphasic waveform produces successful defibrillation while the monophasic waveform fails to defibrillate. The “extension of refractoriness” hypothesis8 9 predicts that biphasic waveforms will enhance action potential prolongation only in this window of low intensities and that cellular responses in this window of low intensities may help to explain the improved defibrillation efficacy of biphasic waveforms.
Therefore, the goal of the present study was to test the following hypotheses for electric field simulation with S2 shocks delivered under fibrillation conditions: (1) The shock intensity/coupling interval–dependent action potential prolongation, seen in the computer model, occurs in actual cardiac cells. (2) The clinically used 8-ms 65%-tilt biphasic waveform, which decreases defibrillation threshold compared with the monophasic waveform,2 produces a longer shock-induced response duration when delivered to cells in the window of low intensities of enhanced biphasic waveform defibrillation effectiveness. (3) Within this low intensity window, the biphasic waveform reduces the dispersion in shock-induced response durations between low and high voltage gradients compared with the monophasic waveform.
In intact hearts, S2 stimuli with the S1S2 coupling intervals used in this study, which were delivered at the relevant shock intensities and were above the lower limit of vulnerability and below the upper limit of vulnerability, produce fibrillation. This complicates the measurement of S2 response duration. Therefore, myocardial cell aggregates, a preparation consisting of tightly coupled myocardial cells28 having typical ventricular action potential shape, were used as the experimental model. This model has been used successfully in many of our previous studies examining the effects of electrical stimulation on myocardial stimulation and dysfunction16 29 30 31 and therefore provides a wealth of control data on which to base the interpretation of the present study.
Materials and Methods
Myocardial cell aggregates were cultured by using techniques previously described.9 In short, cells from 9- to 11-day chick embryos were suspended, at a concentration of 106/mL, in L-15 (GIBCO) containing 10% fetal calf serum. The suspension of cells was transferred to a shaker bath at 37°C (68 to 70 rpm) in which cells formed aggregates (≈100 to 200 cells) with a diameter ranging from 90 to 150 μm.
Twenty minutes before experimentation, aggregates were placed in 60-mm tissue culture dishes that were coated with Cell-Tak adhesive (Collaborative Research Inc). A Plexiglas experimental chamber with a 2.3×3.3-cm rectangular opening was inserted in the culture dish. Platinum/platinum-black electrodes were attached to the long sides of the rectangular opening, 2.2 cm apart, so that a uniform electric field was produced throughout the dish. The culture dish was then placed on a 37°C heated stage of an inverted microscope.9 16
Electrolyte Solutions and Drugs
Cells were superfused at 37°C with modified Krebs’ solution containing (mmol/L) NaCl 110, KCl 4, NaH2PO4 1.2, NaHCO3 32, MgSO4 1.2, CaCl2 2.5, dextrose 5.5, and pyruvate 2.0. Insulin (10 U/L) was added to the working solution just before use. The solution was bubbled continuously with 95% O2/5% CO2 to maintain a pH of 7.4.
The whole-cell recording solution used in the pipette for cellular impalement consisted of (mmol/L) KCl 140, MgCl2 2, EGTA 11, CaCl2 1, and HEPES 10, adjusted to pH 7.2 with 3N KOH.
Electric field stimulation was delivered to the cells via platinum/platinum-black electrodes. A Macintosh computer with labview 2.2.1 software (National Instruments) synthesized the 10-ms rectangular S1 waveform and the 8-ms monophasic and biphasic (with the polarity reversed after 4 ms) S2 waveforms with a total tilt of 65% (Fig 1⇓). The biphasic pulse was a single capacitor waveform; the trailing voltage of phase 1 was equal to the leading voltage of phase 2. The signal was amplified by a custom DC-coupled power amplifier (Cardiac Pacemakers, Inc).
Transmembrane potentials were acquired with a single microelectrode (tip resistance, 2 to 4 MΩ) using an Axoclamp-2A (Axon Instruments, Inc) in the bridge mode. A silver/silver chloride wire served as the indifferent electrode. Reduction of the shock artifact in the microelectrode system was achieved by applying a train of low-amplitude stimuli (approximately one-half threshold intensity) while positioning the indifferent electrode to produce the minimum artifact in the recording. The data were simultaneously displayed on an analog oscilloscope, acquired by a Macintosh computer at a sampling rate of 5 KHz, and stored on disk for off-line analysis. All data from each aggregate were obtained with a single penetration.
Cell aggregates (n=7) from different hearts were impaled by guiding the pipette tip to the cell membrane until an increase in the voltage indicated contact of the cell membrane with the pipette tip. Then low-level suction was applied to the pipette to form a gigaohm-level seal. Penetration to the interior of the cell was achieved by applying a higher level suction.
Cells were stimulated by use of an S1S2 protocol. “Control” conditions simulating normal sinus rhythm were defined by pacing the cells at a CL of 600 ms for 10 s.
During fibrillation, CL between action potentials of myocardial cells becomes very short and relatively uniform (CL, 215±28 ms in humans19 ). The fibrillation activation-front pathways leading to this rapid excitation vary from beat to beat32 ; thus, cells are excited at different fiber orientations with slightly different timing, and relative contributions of ionic channels to fibrillation action potentials vary between beats. During the first few seconds of fibrillation, there is still some perfusion, because 5 to 10 s is required for blood pressure to drop to the mean circulatory pressure and for flow to cease. Therefore, after short fibrillation durations, ischemia and increased extracellular potassium are less likely to influence the response of myocardial cells to a defibrillation shock than are cellular conditions imposed by the preceding period of rapid excitation, which produces action potential shortening and alters the ionic currents. Therefore, we used rapid S1 pacing to simulate fibrillation. Three seconds of “simulated fibrillation” were used so that effects of refractory period stimulation, independent of the ischemic effects that develop with longer periods of fibrillation, could be examined. The S1 pacing stimuli consisted of seventeen 10-ms monophasic rectangular pulses delivered at a basic CL of 180 ms at 1.5 times excitation threshold. This was the shortest CL that could be achieved without producing alternans.
The S2 (8-ms monophasic or 4/4 biphasic pulses) shocks were delivered at intensities of 1.5, 3, and 5 times diastolic threshold of the 8-ms monophasic waveform. The S1 refractory period was defined as the longest S1S2 coupling interval that did not produce an action potential when the monophasic truncated exponential S2 was delivered at 1.5 times diastolic threshold. The S1 action potential was scanned by the S2 in steps of 5 ms, from 10 ms inside to 5 ms outside the refractory period for each S2 test pulse and amplitude.
The shock-induced response duration was measured at 90% repolarization with the aid of a custom computer program. Response duration was defined as the time interval between the beginning of the S2 stimulus and 90% repolarization of the S2 response, as shown in Fig 2A⇓. The action potential amplitude required for determining 90% repolarization was measured as the difference between the takeoff potential and the transmembrane potential at the beginning of the plateau phase, as measured 20-ms after the beginning of the stimulus. This time was chosen because the stimulus artifact is complete, as shown in Fig 2⇓.
The S1S2 coupling intervals for each cell were normalized by subtracting the S2 monophasic waveform refractory period (measured at 1.5 times diastolic threshold) from the measured S1S2 coupling interval for each episode to produce FNCI.9 FNCI of 0 ms corresponds to the end of the S1 action potential refractory period to a monophasic S2 of 1.5 times diastolic threshold.
A total of 122 of 148 S1S2 episodes were included in the analysis (61 episodes each for the monophasic and biphasic test pulses). All data used in the study were paired for monophasic and biphasic waveforms at the same intensities and coupling intervals within each cell. Thirteen pairs of episodes were eliminated for one of the following reasons: (1) After the data were normalized for coupling interval, some episodes did not occur in a sufficient number of cells to be statistically evaluated because they were too early or too late. (2) Some episodes had poor quality recordings.
Results are shown as mean±SEM. Significance of differences was determined by ANOVA and multiple comparisons using Student-Newman-Keuls and Bonferroni tests. Differences were considered significant at P<.05.
Transmembrane potential recordings were obtained during 122 S1S2-simulated fibrillation episodes (CL, 180 ms). Control recordings for action potential characterization were also taken during S1 pacing at a CL of 600 ms in each aggregate. Mean stimulation voltage threshold for the rectangular 10-ms monophasic pulse at CLs of 600 and 180 ms was 10.3±0.9 and 10.6±1.0 V, respectively. Mean stimulation threshold for the truncated exponential 8-ms monophasic test pulse was 13.2±1.4 V. Under control conditions (CL, 600 ms), the resting potential was −84±0.8 mV, APD at 90% repolarization was 152±7.2 ms, and action potential amplitude was 125±0.7 mV. Under simulated fibrillation (CL, 180 ms), APD was reduced to 128±3.5 ms (P<.007), and action potential amplitude was reduced to 119±0.8 mV (P<.01).
Fig 2⇑ shows responses to S2 stimuli delivered to a typical cell at a coupling interval of 140 ms, which corresponded to 87% repolarization. S2 stimuli were delivered at 1.5, 3, and 5 times the diastolic threshold of the MS2 waveform. At an intensity of 1.5 times diastolic threshold, the monophasic pulse produced only a passive membrane response, as shown by the dotted line in Fig 2A⇑. Therefore, this coupling interval was defined as being within the S1 refractory period. At the higher intensities of 3 and 5 times diastolic threshold, the MS2 produced action potential–type responses at the same coupling interval. These responses produced a shock-induced response duration of 70 ms.
In contrast to the passive response produced by MS2 at 1.5 times diastolic threshold, BS2 produced a response with a plateau that produced a response duration of 73 ms. Therefore, even at the low intensity of 1.5 times diastolic threshold, BS2 produced a response duration similar to that produced by the higher intensity MS2. At 3 times diastolic threshold, the BS2 shock–induced response duration was 62 ms, and at 5 times diastolic threshold, it was 65 ms. In Fig 2⇑, a comparison of panels A and B shows that the biphasic waveform (panel B) produces a more uniform response at all intensities tested than does the monophasic waveform (panel A).
The effect of S2 shock intensity and waveform on response duration was determined during both early and late repolarization. Fig 3⇓ shows the mean±SEM response duration produced by MS2 and BS2 test pulses at coupling intervals ranging from 10 ms inside the S1 refractory period (FNCI, −10 ms) to 5 ms outside the refractory period (FNCI, +5 ms). The corresponding (mean±SEM) S1 membrane potentials (S2 takeoff potential) and percent S1 repolarization at FNCI values of −10, −5, 0, and +5 ms were −44.2±0.8 (66% repolarization), −55.2±0.7 (75% repolarization), −63.7±0.6 (82% repolarization), −71±1.4 (89% repolarization) mV, respectively. Fig 3A⇓ shows responses produced by S2 at 1.5 times diastolic threshold. MS2 (solid line), produced a negligible response for coupling intervals inside the refractory period, ie, FNCI of ≤0 ms. For example, the response produced at FNCI of 0 ms was only 10.0±1.9 ms in duration. Outside the refractory period (FNCI, +5 ms), a short action potential was induced that was 68.5±4.7 ms in duration. In contrast to the monophasic waveform, BS2 produced a prolonged response even when delivered within the S1 refractory period. At this low intensity, the BS2 response was 19.0±4.6 ms at FNCI of −10 ms (66% repolarization) and 60.0±3.0 ms at FNCI of 0 ms (82% repolarization). Two-way ANOVA showed that at 1.5 times diastolic threshold, response duration varied interactively as a function of both waveform and coupling interval (P<.001). At coupling intervals of −5 and 0 ms, BS2 produced a significantly longer response (P<.05) than did MS2 (60.0±7.4 versus 10.0±1.9 ms at FNCI of 0 ms).
Fig 3B⇑ shows responses to MS2 and BS2 delivered at 3 times diastolic threshold. At this shock intensity, responses to MS2 and BS2 were similar at all coupling intervals; eg, there was no interaction between waveform and coupling interval (by ANOVA). Both waveforms produced prolonged responses at FNCI of 0 ms (MS2, 60.6±5.2 ms; BS2, 51.0±6.4 ms; P=NS). At FNCI of −5 ms, the responses were also similar (MS2, 28.0±3.3 ms; BS2, 25.2±0.6 ms; P=NS) but were significantly shorter than those produced at FNCI of 0 ms.
At the higher intensity of 5 times diastolic threshold (Fig 3C⇑), both MS2 and BS2 produced long responses when delivered up to 10 ms into the S1 refractory period. For MS2 (solid line) at FNCI of −10 ms, the response duration was 49.2±4.4 ms, which was significantly longer (P<.003) than that produced by 1.5 times diastolic threshold (MS2, 10.5±2.9 ms).
At 5 times diastolic threshold, BS2 (Fig 3C⇑, dotted line) also prolonged response duration at all coupling intervals. However, in contrast to results obtained at the low intensity of 1.5 times diastolic threshold, the responses produced by BS2 were slightly but significantly shorter than those produced by MS2 at all coupling intervals (P<.001 by ANOVA). At this higher intensity, MS2 produced an 11-ms longer response duration than did BS2 at 66% repolarization and a 13-ms longer response duration at 82% repolarization. The mean response duration averaged over all coupling intervals was 55.2±1.7 ms for MS2 and 43.8±2.1 ms for BS2. There was no statistically significant interaction between waveform and coupling interval at this intensity.
Fig 4⇓ shows the response duration for monophasic (panel A) and biphasic (panel B) waveforms as a function of S1S2 coupling interval and shock intensity. As shown in panels A and B, outside the refractory period (FNCI, >0 ms), both MS2 and BS2 produced long responses at both 1.5 and 5 times diastolic threshold, as would be expected for pacing during diastole. MS2 shown in Fig 4A⇓, when delivered within the refractory period (FNCI, ≤0 ms), produced a short response at 1.5 times diastolic threshold and a long response at 5 times diastolic threshold. The mean response duration for FNCI of ≤0 ms was 8.8±1.4 ms at 1.5 times diastolic threshold and 53.0±3.1 ms at 5 times diastolic threshold (P<.001 by two-way ANOVA). Therefore, there was a large dispersion in response duration produced between high and low S2 intensities.
Fig 4B⇑ shows the mean S2 response duration for the biphasic waveform. In contrast to the results shown in Fig 4A⇑, BS2 produced longer responses at both low (1.5 times diastolic threshold) and high (5 times diastolic threshold) intensity for all measured coupling intervals within the S1 refractory period (FNCI, ≤0 ms). The mean response duration for FNCI of ≤0 ms was similar (35.1±12.6 ms at 1.5 times diastolic threshold and 46.2±2.7 ms at 5 times diastolic threshold, P=NS by ANOVA).
All measured response durations for the lowest and highest intensities within the window (1.5 and 5 times diastolic threshold) at FNCI of ≤0 ms (representing the coupling intervals during the relative refractory period) were grouped. The mean response duration for the biphasic waveform (39.9±3.2 ms) was significantly longer than that for the monophasic waveform (26.9±4.4 ms, P<.02). These differences are depicted in the box plot shown in Fig 4C⇑. The box plot shows that the median response duration is short (15.0 ms) and that the dispersion is very large (50% of values have a range of 45 ms) for the monophasic waveform. In contrast, for the biphasic waveform, the median response duration is much longer (40.0 ms), and the dispersion is markedly reduced (50% of values fall within a range of only 22 ms). Because BS2 produced long responses at both low and high intensity, the response duration, as a function of both coupling interval and S2 intensity, had a dispersion that was less than one half that of the monophasic waveform. Therefore, the BS2 response was both longer and had a smaller dispersion than that for MS2.
Fig 5⇓ shows the relation between response duration and intensity for MS2 and BS2 at FNCI of 0 ms. The differences in the mean response duration among the different levels of intensities and between waveforms were statistically significant by two-way ANOVA (P<.001). The effects of different levels of intensity depended on which waveform was present. The monophasic waveform response duration increased by 50.6±4.0 ms between 1.5 and 3 times diastolic threshold and then reached a plateau. The Student-Newman-Keuls test for multiple comparisons showed a statistically significant difference (P<.05) between MS2 at 1.5 times diastolic threshold and at 3 and 5 times diastolic threshold. In contrast, the response duration produced by the biphasic waveform did not change significantly between 1.5 and 5 times diastolic threshold (P=NS) but maintained the long response duration of >40 ms, which was attained by MS2 only at high intensities.
Defibrillation Success and Action Potential Prolongation With Monophasic Waveforms
During fibrillation in humans19 and in animals,24 action potentials immediately follow one another, with little or no period of diastole. Therefore, in contrast to cardiac pacing, defibrillation involves interaction of the shock with cells primarily during their action potentials, ie, during some phase of their refractory period. In addition, during fibrillation, the circus movement produces multiple small wavelets of excitation.33 Thus, action potential propagation throughout the heart is erratic, and different regions of the heart are in different phases of the action potential at the time of the shock. Because of these two factors, the response of cells to shocks delivered during the refractory period rather than during diastole is important in understanding the mechanisms of defibrillation.8 9
To achieve successful defibrillation, the shock must halt activation wavefront(s) in all or in a critical mass of the myocardium.34 Cells early in their absolute refractory period will remain refractory to incoming fibrillation wavefront(s) for a critical postshock period of time regardless of the extended refractory period produced by the shock. Therefore, reentry of fibrillation wavefront(s) cannot take place in those regions. The vulnerable regions are those in which fibrillation wavefront(s) are about to reenter, ie, those that are in the last few milliseconds of their refractory period,35 as shown by the hypothetical situation in Fig 6⇓. The solid line in this figure shows the last three S1 action potentials (CL, 180 ms) that simulate fibrillation and the prolonged response induced by BS2. The dashed line represents the continuation of the S1 responses simulating fibrillation and represents the timing of wavefront arrival leading to local excitation if the simulated fibrillation had continued. Because the S2 shock prolonged the total APD, the cell would have been refractory at the time of the next incoming fibrillation wavefront. If this phenomenon occurred in all or a critical mass of tissue, fibrillation would be terminated.
During defibrillation, not only does the shock interact with cells in all phases of their action potentials, but the local voltage gradient distribution throughout the heart varies so that regions near the shock electrodes are exposed to relatively high shock intensities while those regions far from the electrodes are exposed to low intensities.7 26 27 For successful defibrillation, the shock must extend the refractory period in both high- and low-intensity regions. It has been previously shown that successful defibrillation requires a local voltage gradient of 6 to 7 V/cm in low-intensity regions.27 This corresponds to local shock intensities of 3 to 6 times the diastolic excitation threshold, which are the intensities used in the present study (excitation threshold is ≈1 to 2 V/cm36 ).
The experimental results described in the present study are consistent with the results of our previous computer model of S2 stimulation of the ventricular cell membrane patch.8 14 In both the experimental and theoretical models, at low intensity (1.5 times diastolic threshold), MS2 response duration did not increase until action potentials were produced for S1S2 coupling intervals outside the refractory period (compare Fig 3A⇑ with Fig 3⇑ of Jones et al14 ). In both models, monophasic waveform stimulation within the refractory period produced significant action potential prolongation at the higher intensities of 3 to 5 times diastolic threshold. Response duration as a function of shock intensity, for S2 delivered at ≈80% to 90% S1 repolarization, reached a plateau by 3 to 5 times diastolic threshold (compare Fig 5⇑ with Fig 1⇑ of Jones and Jones8 ).
The present results using electric field stimulation under fibrillation conditions are consistent with an earlier study showing lack of refractory period extension by using intracellular current injection S2, at 1.5 times diastolic threshold (MS2), delivered to cells during slow S1 stimulation, simulating normal sinus rhythm.9 This similarity demonstrates that intracellular current injection and electric field stimulation produce similar S2 responses in spite of their very different physiological effects on the cell. Consistent with our results (shown in Fig 3⇑), low-intensity MS2 stimulation (1.4 V/cm) in rabbit papillary muscle also does not produce refractory period responses. Only at high intensities above the defibrillation threshold7 (8.4 V/cm37 ) is S1 repolarization delayed by S2 stimuli delivered during the refractory period.
Defibrillation Success and Uniform Repolarization With Monophasic Waveforms
In addition to the “extension of refractoriness” mechanism described above, a second mechanism for defibrillation has been proposed (the “upper limit of vulnerability” hypothesis). This hypothesis states that shocks below the upper limit of vulnerability12 cause refibrillation due to a nonuniform dispersion of refractoriness following the shock. The “upper limit of vulnerability” hypothesis suggests that in order to halt fibrillation without reinducing fibrillation, a successful shock must produce a relatively uniform extension of refractoriness for cells in various stages of repolarization and in both high– and low–shock intensity regions of the heart. Consistent with this hypothesis, successful defibrillation shocks produce a uniform response duration or “synchronized repolarization”6 in isolated rabbit hearts. The results of the present study are also consistent with this hypothesis. Fig 4C⇑ shows that with MS2, a low-intensity shock with a minimum voltage gradient of 1.5 times diastolic threshold and a maximum voltage gradient of 5 times diastolic threshold produces a large dispersion of response duration (corresponding to a large dispersion of refractoriness). Only at relatively high shock intensities, ie, 3 to 5 times diastolic threshold, can monophasic waveforms produce the required degree of uniformity of repolarization, independent of coupling interval. These results confirm the relatively high local voltage gradient of 5 to 6 V/cm in minimum–shock intensity regions of the heart required for successful defibrillation.
Therefore, two important criteria must be satisfied for successful defibrillation. First, the shock must prolong the fibrillation APD. Second, the shock must prevent the dispersion of repolarization between low- and high-intensity regions of the heart, because a wide dispersion of repolarization facilitates fibrillation.38 With monophasic waveforms, nonuniform repolarization can take place not only because of the lack of prolongation in low-intensity regions but also because of postshock action potential changes (due to transient shock-induced dysfunction) in the high-intensity regions immediately adjacent to the defibrillation electrodes.16 31 39 40
Mechanism of Defibrillation Threshold Reduction With Biphasic Waveforms
Defibrillation threshold is frequently defined by using an up-down protocol. By use of this type of protocol, many biphasic waveforms, created by reversing the polarity of the shock part way through the pulse delivery, have defibrillation thresholds that require only ≈50% of the energy needed for corresponding monophasic waveforms. The monophasic waveform requires a minimum voltage gradient in the heart of 5 to 6 V/cm,23 or 3 to 6 times diastolic threshold,36 in those regions of lowest shock intensity far from the electrodes, whereas selected biphasic waveforms require only ≈2.7 V/cm,23 or ≈1.5 times diastolic excitation threshold.
Mechanisms through which biphasic waveforms reduce defibrillation threshold can be more fully understood by examining the probability of successful defibrillation versus shock-intensity curves.3 4 These curves show that the probability of successful defibrillation is similar for monophasic and biphasic waveforms at relatively high shock intensities, for which the monophasic waveform defibrillates successfully. They also show that at very low shock intensities, neither the monophasic nor the biphasic waveform defibrillates successfully. However, there is a window of shock intensities between these values for which the biphasic waveform has a high probability of success and for which the monophasic waveform fails. This window begins at shock intensities that produce a minimum voltage gradient of ≈3 V/cm, or ≈1.5 times diastolic threshold, and ends at ≈5 to 7 V/cm, or 3 to 5 times diastolic threshold. The cellular responses produced by S2 stimulation in this shock intensity window may provide a key to understanding mechanisms underlying the lower defibrillation threshold for the biphasic waveform compared with the monophasic waveform.
The results in Figs 3⇑ and 4⇑ of the present study show that BS2 stimulation, delivered under fibrillation conditions, uniformly prolongs the refractory period and decreases dispersion of refractoriness within this window consistent with results from a preliminary experimental study at normal sinus rhythm18 and with the computer modeling study discussed previously.14 In the computer model, low-intensity 10-ms biphasic waveforms of 1.5 times diastolic threshold, when delivered during the refractory period of the previous S1 action potential, produced 62-ms longer graded responses than did a comparable monophasic waveform. This prolongation did not occur when sodium channels were blocked immediately before S2. At higher intensities of 5 times diastolic threshold, both monophasic and biphasic waveforms produced similar long responses that were independent of sodium channel blockage. The prolonged responses produced by both MS2 and BS2 at 5 times diastolic threshold are consistent with their similarly high defibrillation efficacy at shock intensities outside of the window.
The enhanced response duration produced by low-intensity biphasic waveforms compared with monophasic waveforms and the lack of prolongation with sodium channel blockage can be explained by the hypothesis that the first phase of a biphasic pulse hyperpolarizes portions of the cell membrane. This allows time- and voltage-dependent sodium channel recovery from inactivation and increases the availability of sodium channels.8 9 14 The second pulse then depolarizes the membrane and produces a longer response than could the monophasic pulse alone.
Studies underlying the hypothesis that enhanced refractory period responses produced by low-intensity biphasic waveforms correlate with their enhanced defibrillation efficacies at low intensity have been primarily carried out at long CLs typical of normal sinus rhythm rather than at the short CLs that occur during fibrillation. APD is a function of CL, so that during fibrillation action potentials become very short, with durations ranging from 80 to 200 ms. APD shortens with CL because of the rate-dependent activation of several membrane ionic channels; this occurrence may be important in defibrillation. One study suggesting this importance shows that a “constant repolarization time” phenomenon that was observed at shock intensities producing successful defibrillation6 occurred only under fibrillation conditions but not at slower CLs. The results of the present study address the issue of CL dependence by demonstrating that at short CLs, which simulate fibrillation, the constant repolarization time6 (equivalent to our postshock response duration) for monophasic waveforms requires high-intensity stimuli of 5 times diastolic threshold. However, biphasic waveforms prolong response duration within the low intensity window beginning at 1.5 times diastolic threshold, suggesting that it is the prolonged refractory period responses that are responsible for successful defibrillation at low intensities.
Two studies appear at first to contradict the present results but are actually consistent. One report11 showed direct excitation produced by 3-ms monophasic waveforms that was further into the refractory period than that produced by 2/1 biphasic waveforms. In contrast, our results showed a 12-ms shorter refractory period with 4/4 biphasic waveforms than with 8-ms monophasic waveforms (at equivalent intensity). Because sodium channel recovery is time and voltage dependent, the difference between these studies can be determined by the difference in first phase duration. In the Daubert study,40 the first pulse was only 2 ms, and sodium channel recovery may not have taken place. Therefore, the direct excitation efficacy of the biphasic waveform was equivalent to the higher value of a 2-ms monophasic waveform.41 With the longer first phase (4 ms) used in the present study, sodium channel recovery could occur, and the biphasic waveform was able to shorten the refractory period.
A second report23 shows that 8/8 biphasic waveforms, delivered during the fibrillation action potential at an intensity of 5 V/cm, produce less action potential prolongation than do 16-ms monophasic waveforms. Fig 3C⇑ shows longer responses for the monophasic waveform than for the biphasic waveform at the equivalent intensity of 5 times diastolic threshold. Biphasic waveforms significantly enhanced APD prolongation only in the low-intensity window. In this context, it is important to note that studies to determine mechanisms underlying the enhanced efficacy of biphasic waveforms at low shock intensities must examine differences in cellular responses between monophasic and biphasic waveforms in this low intensity window. In this voltage gradient window, the enhanced action potential prolongation observed with biphasic waveforms correlates with their enhanced defibrillation efficacy.9
In summary, the results from the present study show that at fibrillation CLs, 4/4 low-intensity (1.5 times diastolic threshold) biphasic defibrillator waveforms (65% tilt) (1) significantly enhance action potential prolongation when delivered during the S1 refractory period and (2) shorten the S1 cellular refractory period. At higher intensities (3 to 5 times diastolic threshold), both monophasic and biphasic waveforms produce significant prolongation of cellular APD. Both biphasic and monophasic defibrillator waveforms also reduce dispersion of response duration at intensities above their respective defibrillation threshold. The similarity in response of cardiac tissue to electric shocks at CLs representing normal sinus rhythm as well as fibrillation and the ability of computer models of the cardiac cell membrane to predict these results suggest that enhanced membrane responses produced by biphasic waveforms at low shock intensity are basic properties of the cardiac cell membrane.
Selected Abbreviations and Acronyms
|APD||=||action potential duration|
|FNCI||=||functionally normalized coupling interval|
|S1 and S2||=||stimuli|
This study was supported in part by US Public Health Service grant HL-24606 and by the Department of Veterans Affairs Medical Center. We thank Nettie Knight for her expert technical assistance.
- Received August 9, 1994.
- Accepted May 1, 1995.
- © 1995 American Heart Association, Inc.
Swartz JF, Fletcher RD, Karasik PE. Optimization of biphasic waveforms for human nonthoracotomy defibrillation. Circulation. 1993;88:2646-2654.
Jones JL, Swartz JF, Jones RE, Fletcher R. Increasing fibrillation duration enhances relative asymmetrical biphasic versus monophasic defibrillator waveform efficacy. Circ Res. 1990;67:376-384.
Dillon SM. Synchronized repolarization after defibrillation shocks: a possible component of the defibrillation process demonstrated by optical recordings in rabbit heart. Circulation. 1992;85:1865-1878.
Sweeney RJ, Gill RM, Steinberg MI, Reid PR. Ventricular refractory period extension caused by defibrillation shocks. Circulation. 1990;82:965-972.
Jones JL, Jones RE. Effects of monophasic defibrillator waveform intensity on graded response duration in a computer simulation of the action potential. Proc IEEE/EMB 1991;13:598-599.
Swartz JF, Jones JL, Jones RE, Fletcher R. Conditioning prepulse of biphasic defibrillator waveforms enhances refractoriness to fibrillation wavefronts. Circ Res. 1991;68:438-449.
Dillon SM. Optical recordings in the rabbit heart show that defibrillation strength shocks prolong the duration of depolarization and the refractory period. Circ Res. 1991;69:842-856.
Daubert JP, Frazier DW, Wolf PD, Franz MR, Smith WM, Ideker RE. Response of relatively refractory canine myocardium to monophasic and biphasic shocks. Circulation. 1991;84:2522-2538.
Chen PS, Shibata N, Dixon EG, Wolf PD, Danieley ND. Activation during ventricular defibrillation in open-chest dogs: evidence of complete cessation and regeneration of ventricular fibrillation after unsuccessful shocks. J Clin Invest. 1986;77:810-823.
Chen P-S, Wolf PD, Melnick SD, Danieley ND, Smith WM, Ideker RE. Comparison of activation during ventricular fibrillation and following unsuccessful defibrillation shocks in open-chest dogs. Circ Res. 1990;66:1544-1560.
Jones JL, Lepeschkin E, Jones RE, Rush S. Response of cultured myocardial cells to countershock-type electric field stimulation. Am J Physiol. 1978;235:H214-H222.
Rush S, Jones JL, Jones RE, Lepeschkin E. Field modelling for defibrillation studies. In: Proceedings of the 1st Cardiac Defibrillation Conference; West Lafayette, Ind: Purdue University; 1975:103-107.
Tovar OH, Milne KB, Jones JL. Interaction of tilt and stimulus intensity on prolongation of refractory period with monophasic and biphasic defibrillating waveforms. Proc IEEE/EMB. 1993;15:846-847.
Swartz JF, Jones JL, Fletcher RD. Characterization of ventricular fibrillation based on monophasic action potential morphology in the human heart. Circulation. 1993;87:1907-1914.
Akiyama T. Intracellular recording of in situ ventricular cells during ventricular fibrillation. Am J Physiol. 1981;240:H465-H471.
Sano T, Tsuchihashi H, Shimamoto T. Ventricular fibrillation studied by the microelectrode method. Circ Res. 1958;6:41-46.
Zhou X, Wolf PD, Rollins DL, Afework Y, Smith WM, Ideker RE. Effects of monophasic and biphasic shocks on action potentials during ventricular fibrillation in dogs. Circ Res. 1993;73:325-334.
Zhou X, Daubert JP, Wolf PD, Smith WM, Ideker RE. Epicardial mapping of ventricular defibrillation with monophasic and biphasic shocks in dogs. Circ Res. 1993;72:145-160.
Witkoswski FX, Penkoske PA, Plonsey R. Mechanism of cardiac defibrillation in open-chest dogs with unipolar DC-coupled simultaneous activation and shock potential recordings. Circulation. 1990;82:244-260.
Wharton JM, Wolf PD, Smith WM, Chen P-S, Frazier DW, Yabe S, Danieley N, Ideker RE. Cardiac potential and potential gradient fields generated by single, combined, and sequential shocks during ventricular defibrillation. Circulation. 1992;85:1510-1523.
Nathan RD, DeHaan R. Voltage clamp analysis of embryonic heart cell aggregates. J Gen Physiol. 1979;73:175-198.
Jones JL, Jones RE. Decreased defibrillator-induced dysfunction with biphasic rectangular waveforms. Am J Physiol. 1984;247:H792-H796.
Jones JL, Jones RE, Balasky G. Improved cardiac cell excitation with symmetrical biphasic defibrillator waveforms. Am J Physiol. 1987;253:H1418-H1424.
Jones JL, Jones RE, Balasky G. Microlesion formation in myocardial cells by high-intensity electric field stimulation. Am J Physiol. 1987;253:H480-H486.
Damle RS, Kanaan NM, Robinson NS, Ge Y-Z, Goldberger JJ, Kadish AH. Spatial and temporal linking of epicardial activation directions during ventricular fibrillation in dogs: evidence for underlying organization. Circulation. 1992;86:1547-1558.
Allessie MA, Bonke FI, Schopman FJ. Circus movement in rabbit atrial muscle as a mechanism of tachycardia, II: the role of nonuniform recovery of excitability in the occurrence of unidirectional block, as studied with multiple microelectrodes. Circ Res. 1976;39:168-177.
Tovar O, Bransford P, Moubarak J, Milne K, Amanna A, Jones J. Correlation between shock induced response duration and defibrillation. Proc IEEE/EMB. 1994;16:21-22.
Lepeschkin E, Jones JL, Rush S, Jones RE. Local potential gradients as unifying measure for thresholds of stimulation, standstill, tachyarrhythmia and fibrillation appearing after strong capacitor discharges. Adv Cardiol. 1977;21:268-278.
Knisley SB, Smith WM, Ideker RE. Effect of field stimulation on cellular repolarization in rabbit myocardium: implications for reentry induction. Circ Res. 1992;70:707-715.
Kuo C-S, Munakata K, Reddy CP, Surawicz B. Characteristics and possible mechanism of ventricular arrhythmia dependent on the dispersion of action potential durations. Circulation. 1983;67:1356-1367.
Knisley SB, Smith WM, Ideker RE. Prolongation and shortening of action potentials by electrical shocks in frog ventricular muscle. Am J Physiol. 1994;266:H2348-H2358.
Yabe S, Smith WM, Daubert JP, Wolf PD, Rollins DL, Ideker RE. Conduction disturbances caused by high current density electric fields. Circ Res. 1990;66:1190-1203.
Jones JL, Sheffield C, Jones RE, Swartz J. Short duration biphasic defibrillator waveforms inhibit refractory period responses. Circulation. 1990;82(suppl III):III-642. Abstract.