The Ib Phase of Ventricular Arrhythmias in Ischemic In Situ Porcine Heart Is Related to Changes in Cell-to-Cell Electrical Coupling
Background This study was designed to test the hypothesis that the loss of cell-to-cell electrical interaction during ischemia modulates the amplitude of ischemia-induced TQ-segment depression (ie, the injury potential) and the occurrence of ventricular fibrillation (VF) during the so-called Ib phase of ventricular arrhythmias.
Methods and Results Regional ischemia was induced by 60 minutes of mid–left anterior descending coronary artery ligation in open-chest swine (n=10). Cell-to-cell electrical uncoupling was defined as the onset of the terminal rise in whole-tissue resistivity (Rt). Local activation times and TQ-segment changes (injury potential) were determined from unipolar electrograms. Extracellular K+ ([K+]e) and pH (pHe) were measured with plunge-wire ion-selective electrodes. VF occurred in 6 of 10 pigs during regional no-flow ischemia between 19 and 30 minutes after the arrest of perfusion. The occurrence of VF was positively correlated to the onset of cell-to-cell electrical uncoupling (R2=.885). Cell-to-cell electrical uncoupling superimposed on changes of [K+]e and pHe contributed to the failure of impulse propagation between 19 and 30 minutes after the arrest of perfusion. During ischemia, maximum TQ-segment depression was −10 mV at 19 minutes, after which TQ-segment depression slowly recovered. The onset of the TQ-segment recovery was correlated to the second rise in Rt (R2=.886).
Conclusions In the regionally ischemic in situ porcine heart, loss of cell-to-cell electrical interaction is related to the occurrence of VF and changes in the amplitude of the injury current. Cellular electrical uncoupling contributes to failure of impulse propagation in the setting of altered tissue excitability as a result of elevated [K+]e and low pHe. These data indicate that Ib arrhythmias and ECG changes during ischemia are influenced by the loss of cell-to-cell electrical interaction.
Ventricular arrhythmias occurring secondary to an arrest of myocardial perfusion are the determinant cause of death in more than one half of the fatalities associated with myocardial infarction and are an important cause of sudden cardiac death.1 Two distinct phases of ventricular arrhythmias occur during the first 30 minutes after induction of regional ischemia by acute occlusion of a coronary artery in canine and porcine hearts.2 3 4 5 6 These periods of ventricular arrhythmia, separated by an arrhythmia-free interval, were first described by Haase and Schiller5 and further characterized and called type Ia and type Ib by Kaplinsky and colleagues4 7 and others.6 7 8 Type Ia arrhythmias occur 2 to 10 minutes after the onset of ischemia, with a peak frequency at 5 to 6 minutes.4 The second wave of arrhythmias, type Ib, occurs later, peaking 12 to 30 minutes after coronary artery ligation.4 Previous studies showed a loose association in the occurrence of arrhythmias with some electrophysiological changes, including excitation threshold, refractoriness, and conduction times.2 Type Ia arrhythmias are not likely to be due to changes of passive, cable-like properties such as cell-to-cell electrical uncoupling but rather occur secondary to depressed excitatory current, increased refractory periods, and conduction delay.9 10 11 In contrast, the onset of cell-to-cell electrical uncoupling 10 to 20 minutes after the arrest of perfusion suggests a possible role of impaired intercellular communication in the initiation of type Ib arrhythmias.9
Intercellular coupling with low-resistance pathways (ie, gap junctions) is essential for impulse propagation.9 The conductance of the gap junctions is modulated by intracellular calcium,12 13 magnesium,12 protons,12 14 15 and lipid metabolites,16 which accumulate in the cytosol or sarcolemma during ischemia. Increasing concentrations of these compounds, along with decreasing intracellular concentrations of ATP,17 decrease gap junctional conductance and increase intercellular electrical resistance. Increasing gap junctional resistance (resulting in cell-to-cell electrical uncoupling with increased tissue resistance) and the formation of activation delay and conduction block were shown to be potential causes of ventricular arrhythmias.9 18
Although much discussion exists regarding the association of cardiac arrhythmias with the degeneration of cellular coupling,9 10 18 19 20 21 22 no studies have definitively demonstrated an association of VF with the onset of cell-to-cell electrical uncoupling in a whole-animal model of regional ischemia. The purpose of this series of experiments was to establish the time course of conduction delay, conduction block, and tissue resistance (an indirect measure of cellular coupling) during 60 minutes of regional ischemia in the in situ porcine heart. After this elucidation, our aim was to determine any existing correlation between the occurrence of VF during the Ia and Ib phases with these changes or ionic changes occurring simultaneously. Our results suggest that changes in tissue resistivity (reflecting cell-to-cell electrical uncoupling) play a prominent role in the second phase of conduction slowing and blocking and the subsequent genesis of the Ib phase of ventricular arrhythmias.
The increase in intracellular resistance resulting from cellular electrical uncoupling has additional effects during the acute stages of ischemia. It is anticipated that the diastolic current of injury (TQ-segment depression) produced as a result of membrane potential differences in the ischemic zone relative to the normal zone will decrease as intracellular resistance increases secondary to cell-to-cell electrical uncoupling.9 22 This study confirms that relation, demonstrating a temporal correlation between the maximum amplitude of the injury current and the onset of cell-to-cell electrical uncoupling.
The experimental protocol involved inducing an episode of 1 hour of regional ischemia in the distribution of the LAD in an open-chest porcine model. Ionic measurements, local electrogram, and resistivity measurements were made throughout the central ischemic zone. Of 10 pigs studied, 1 pig fibrillated during the Ia phase (2 minutes into ischemia), could not be defibrillated, and therefore was excluded from further data analysis. Care of all pigs used in this study conformed to the Position of the American Heart Association on Research Animal Use and was done in accordance with accepted guidelines for the care and treatment of experimental animals at the University of North Carolina.
The experimental preparation is similar to that used previously23 24 in which domestic swine of either sex weighing 30 to 40 kg received sodium pentothal (25 mg/kg) followed by repeated doses of α-chloralose as necessary to maintain a deep surgical anesthesia. The animals were intubated and maintained on mechanical ventilation. Arterial blood gases were monitored frequently to maintain Po2 >80 mm Hg and Pco2 between 35 and 45 mm Hg. Catheters were inserted into the left femoral artery and vein for the infusion of fluids and withdrawal of blood samples. Body temperature was monitored, and heating blankets were used as needed to maintain a body temperature of 36°C to 37°C.
A median sternotomy was performed, and the heart was suspended in a pericardial cradle. A site along the mid-LAD was dissected free of surrounding tissue, and a snare was placed around the artery that was used later to ligate the vessel. An epicardial ischemic zone was identified by brief occlusion of the vessel at this site, and multiple electrodes were placed within the center of the ischemic zone (at least 10 mm from the nearest cyanotic border). The arterial blood pressure, a lead II ECG, and 10 ion-sensitive electrode channels were monitored continuously with a 12-channel strip-chart recorder (Western Graphtec).
After electrode placement, an 8-in-diameter watch glass was placed over the sternotomy to maintain the temperature and humidity on the heart surface. A temperature probe was placed intramurally within the ischemic zone to monitor temperature during LAD ligation. All experimental observations were made in pigs paced slightly above their intrinsic rate, which ranged from 78 to 115 bpm in this preparation. Ventricular rates did not differ significantly between pigs that fibrillated (mean rate, 100±14 bpm) and those that did not (mean rate, 103±3 bpm, P=.37).
Forty-five minutes after electrode placement, the in vivo electrode performance was assessed by methodologies used previously (see below). After an additional 30-minute interval, the LAD was ligated at the previously identified site. Data were collected at 15-, 30-, or 60-second intervals throughout a 60-minute period of LAD ligation as a digital signal by use of a multichannel analog-to-digital converter (Preston Scientific) and custom-developed data acquisition software, drtaj (University of North Carolina, Chapel Hill).
When VF occurred, DC cardioversion was attempted. In two experiments, defibrillation was successful, and VF did not occur throughout the remainder of the hour of ischemia. Data collection was interrupted in these two experiments for 12 and 25 minutes, during which time VF occurred and paced rhythm was restored. In the other four instances of VF, a paced rhythm could not be sustained for periods adequate for data collection to resume (ie, VF recurred shortly after defibrillation). In these four experiments in which defibrillation was immediately unsuccessful, data collection from the resistivity electrode continued, and efforts to defibrillate the animal were aborted to facilitate data collection.
Measurement of [K+]e and pHe
Ion-selective (K+ and H+) plunge-wire electrodes, fashioned and calibrated with previously described methods,25 26 were used to make ionic measurements from the midmyocardium in the center of the ischemic zone. These electrodes were constructed from Teflon-coated silver wire, the end of which was chloridized and covered with a cellulose acetate and titanium dioxide sponge. K+-selective electrodes were made by covering the sponge with a valinomycin-based membrane; H+-sensitive electrodes were made by covering the sponge with the H+ ionophore tridodecylamine.27 Reference electrodes were constructed in an identical manner but lacked the ion-selective membrane. Electrodes were calibrated before each experiment in standard ionic and pH buffer solutions of balanced ionic strength. Only electrodes that demonstrated <1 mV drift per hour and 90% to 100% of predicted temperature-corrected Nernstian response were used. The in vivo performance of K+ electrodes was tested with a KCl bolus as previously described.23 24 Electrodes were removed and retested in vitro after each experiment. Data were accepted only from electrodes that calibrated appropriately during in vivo and postexperimental in vitro testing. These stringent criteria led to an ≈50% acceptance rate for ion-selective electrodes; 3 to 9 K+ (median, 3) and 0 to 5 H+ (median, 3.5) electrodes met the criteria for acceptance in each experiment.
Values of aK+ were calculated from the measured voltage changes with the calibration curve for each individual electrode and the myocardial temperature and systemic [K+]e obtained from an arterial blood sample immediately before the occlusion.28 29 An activity coefficient of 0.746 was used to convert aK+ to [K+]e. Values of pHe were calculated similarly.
Measurement of Ventricular AT and TQ Depression
Two stainless steel reference electrodes, insulated to within 1 mm of the tip, were paired with each ion-selective electrode (K+ or H+) to record bipolar and unipolar electrograms at each site within the central ischemic zone; two were also placed in the normal zone remote from the area of ischemia to measure normal activation. One stainless steel electrode from each site was referenced to a common ground on the aortic root to record a unipolar electrogram. Each electrode group was inserted into the midmyocardium at a depth of 4 to 6 mm. Ten to twelve electrode groups were placed in each experiment; data from unipolar and bipolar pairs at a site were retained even if the corresponding ionic electrode failed the inclusion criteria outlined above.
Signals from the bipolar electrodes were individually filtered between 50 and 500 Hz (−1 dB). Unipolar electrodes were DC- coupled (0 to 500 Hz at −1 dB) to a common reference electrode at the aortic root. The diastolic DC potential of each unipolar electrogram was measured at end diastole. This potential was referenced to the baseline value to determine the degree of ΔTQ. TQ reversal time was defined as the time at which a minimum ΔTQ potential occurred, after which the ΔTQ values increased toward zero (illustrated in inset in Fig 8⇓).
Local AT from bipolar electrodes was defined as the peak of the high-frequency deflection; unipolar AT occurred at the point of maximum negative slope of the unipolar electrogram.30 Unipolar and bipolar electrodes from an electrode pair were simultaneously reviewed, and the AT for each electrode site was determined from a combination of the two. Local activation block was defined as the absence of a local activation spike of at least 1 mV in the bipolar electrogram and a monophasic appearance in the unipolar electrogram. Only supraventricularly paced beats were analyzed. For each electrode site, the AT was referenced to an electrode placed in the nonischemic zone. An AT at each electrode site was thus determined, and the activation delay at each electrode site was calculated at each point in time by subtracting a baseline AT from that measured at each sample.
Measurement of Whole-Tissue Resistivity, an Index of Cell-to-Cell Electrical Coupling
Whole-tissue resistivity was measured with a four-electrode technique as described previously.31 32 33 34 35 The four-electrode array was fashioned after that of Ellenby et al35 with gold-plated electrode pins 4 mm in length and 0.25 mm in diameter. These electrodes were placed in a linear array with 5-mm spacing between the innermost two pins and 2.5-mm spacing between the outer pin and its neighbor. The pins were insulated along their length except for 0.5 mm at their most distal tips. The pins were mounted on a nonconductive wafer so that the entire four-electrode array could be inserted as a unit. This wafer was sutured to the epicardium to maintain its position. Before insertion, the electrode array was calibrated in a range of NaCl solutions of known conductivity.36
The four-electrode wafer was inserted into the ischemic region; the outside pin was not <2 cm from the visible cyanotic border. Ionic and bipolar-unipolar electrode pairs were inserted within the ischemic region at least 0.5 cm from the resistivity electrode. In each experiment, an effort was made to insert the electrode array with its long axis parallel to the long axis of the ventricular muscle fibers on the epicardium.
A subthreshold 10-μA DC pulse of 20-ms duration was delivered across the outer two electrodes in the array. The delivered current was measured with a current-to-voltage converter introduced between the resistivity electrode and the ground. The resulting voltage was collected with the computer software described below on a beat-to-beat basis for analysis. The voltage field resulting from the subthreshold current with the tissue as a resistor was sampled across the central two electrodes. Resistivity was calculated according to33 35
where ρ is the tissue resistivity in ohm-centimeters, V is the voltage measured across the two central electrodes, I is the current delivered, λ is a constant determined from the calibration, and d is the interelectrode distance across which the voltage is measured. Although cardiac resistivity is constant throughout the cardiac cycle,33 the current pulse was synchronized with an atrial pacing pulse delivered during end diastole to prevent interference of the pulse with the measurement of local activation (and vice versa) and TQ potentials. For standardization between experiments, resistivity is reported as percent change from baseline rather than in ohm-centimeters.
Determination of the Onset of the Third Phase of Cellular K+ Loss and Whole-Tissue Resistivity
The time point of the onset of the third phase of cellular K+ loss was determined to be the first point that demonstrated a 10% rise over the best-fit line through the plateau phase, after which the values continued to increase (inset, Fig 7⇓). The second rise time for tissue resistivity was calculated in the same manner (inset, Fig 3⇓). In one experiment, VF occurred when the resistivity had risen 9% over baseline. In this experiment, the data point before VF was considered the initiation of the second rise.
Single small electrodes were individually amplified by high-impedance amplifiers. Analog signals from all electrodes, along with a lead II ECG, were digitized and simultaneously sampled at a sampling frequency of 1000 Hz for 1 second at 15-, 30-, or 60-second intervals with a Micro VAX II /GPX computer. Data collection was interrupted during VF and was restarted when defibrillation was achieved. When defibrillation was unsuccessful, data collection was restarted to determine whether the slope of the second rise in resistivity during VF differed from that during ischemia.
For each variable studied, mean values of each electrode type from the ischemic zone were first calculated for each experiment individually. These mean values were then pooled to calculate a mean for the experimental group as a whole. Values were assigned to the nearest half-minute interval (ie, if an observation was made between 1:15 and 1:44, it was assigned to 1:30) for the purposes of calculating means.
Linear regression plots and equations were created with a least-squares algorithm on the Macintosh computer program cricket graph (Computer Associates). Probability values were obtained for linear regressions by determining the level of significance at which the slopes of the plots differed from a slope of zero (ie, a nonrelation). To test the difference in minimum TQ-segment depression in pigs in which VF occurred versus those in which there was no VF, the Wilcoxon rank-sum test was applied. The Wilcoxon rank-sum test was also used to evaluate mean changes in parameters over an interval of interest. In analyses of completers (defined as having data at both the beginning and the end of an interval; when an animal fibrillated during an interval of interest, its data were subsequently absent), a Wilcoxon signed-rank test was applied. Because the sample sizes were small, both testing techniques are nonparametric; therefore, the analysis was performed on the ranked data. Statistical analysis was completed by use of a general linear model procedure in sas, release 6.04 (SAS Institute) and statxact, version 2.0 (Cytel Software). Data are presented as mean±SD unless otherwise indicated. A value of P<.05 was considered statistically significant.
Occurrence of VF and Ventricular Arrhythmia
A total of 10 pigs were studied (1 pig was later excluded owing to fibrillation during the Ia phase). VF occurred at least once in seven experiments. Fig 1⇓ summarizes the timing of the VF episodes. Other than the single episode of VF during the Ia phase, the onset of VF was clustered between 19 and 30 minutes of ischemia. We defined a “window of fibrillation” in our series that is based on the earliest and latest occurrence of VF during the Ib phase. The fibrillation window in this series (19 to 30 minutes) corresponded with the previously published4 Ib phase between 12 and 30 minutes. VF did not occur in any experiment beyond 30 minutes of ischemia.
Ventricular arrhythmias other than VF also occurred during this series. The total number of spontaneous ventricular arrhythmias per minute, including PVCs, nonsustained ventricular tachycardia not exceeding six successive beats, and VF, is also given in Fig 1⇑. PVCs were by far the most common arrhythmia observed in this series. A biphasic pattern of arrhythmia during the first 30 minutes after the arrest of perfusion (as described by others4 6 7 8 ) was demonstrated; a third peak in the incidence of arrhythmia was seen between 50 and 60 minutes of ischemia.
Electrophysiological and Ionic Parameters During the Fibrillation Window
The time course of changes in tissue resistivity is shown in Fig 1⇑. A typical pattern from a single experiment is shown in the inset of Fig 3⇓. Resistivity is plotted as percent change from preischemic control; values obtained at the onset of ischemia averaged 187±120 Ω · cm. A triphasic pattern of resistivity is observed. There is a rapid and relatively small increase in resistivity during the first 3 minutes of ischemia to 130% to 150% of initial values. This increase has been shown in the isolated ischemic rabbit papillary muscle to be due to an increase in the extracellular component of whole-tissue resistance (ie, the longitudinal extracellular resistance).9 The initial rise is followed by a plateau that lasts about 10 to 12 minutes, after which a second, more prominent rise begins. This final rise in tissue resistivity is due to cell-to-cell electrical uncoupling. Fig 2⇓ clarifies the triphasic pattern of change seen in electrical resistivity during ischemia. Because the onset of the second rise in resistance occurs at different times during different experiments, the rapidity of uncoupling is not readily discernible in Fig 1⇑. The onset of cell-to-cell electrical uncoupling in the experiments (as previously defined) is indexed to time zero in Fig 2⇓, and mean resistance is plotted for all experiments during the 60-minute period of ischemia. Episodes of VF are indicated with an asterisk and indicate their close temporal association with the onset of cell-to-cell electrical uncoupling.
The second rise in mean tissue resistivity begins between 13 and 15 minutes of ischemia and ultimately progresses to a level of ≈300% of initial resistivity by the end of the 60-minute period of ischemia owing to cell-to-cell electrical uncoupling.9 During the period in which all episodes of VF occurred, tissue resistivity increased by an additional 45% (P=.0412).
Fig 3⇓ is a plot of the time to VF versus the time to the onset of the second rise in resistivity. Determination of this second rise time in individual experiments is illustrated in the inset. A linear relation is observed, with R2=.885. The correlation was significant, with P=.0052. In each case, the onset of VF occurred 3 to 4 minutes after the onset of the second rise in resistivity, as indicated with asterisks in Fig 2⇑.
Fig 4⇓ summarizes the changes in AT within the center of the ischemic zone. This figure shows the mean activation delay during ischemia and the percent of all electrodes that demonstrated local activation at each time point. The fibrillation window (19 to 30 minutes) is represented by shading. Like the changes in whole-tissue resistivity, there was a triphasic pattern to the time course of activation delay change. Initially, there was a rapid onset of delayed activation during ischemia, to a maximum of an 80±58-ms delay from baseline activation. This initial peak is concurrent with an initial phase of local activation block occurring in 10% to 15% of all electrodes. Between 6 and 14 minutes of ischemia, there is a brief period of spontaneous recovery to a mean of a 40±15-ms activation delay and a return of local activation to all but 2% of the regions that previously were blocked. After this phase of improvement in conduction, ATs again begin to rise, and the number of electrodes demonstrating local activation block increases progressively, associated with the rising tissue resistivity. After 40 minutes of acute ischemia, all electrodes in the ischemic zone demonstrate activation block. The third phase of activation delay occurred during the fibrillation window and closely correlated with the onset of cell-to-cell electrical uncoupling.
It should be noted from Fig 4⇑ that although mean activation delays are lower during the fibrillation window than in the early stages of ischemia, there also is progressively more conduction block. The mean conduction delay approaches 80 ms during early ischemia (Ia phase of arrhythmias); during the fibrillation window, the mean activation delay never exceeds 60 ms, but during this time a larger percentage of electrodes fails to record local activation. The relatively lower mean conduction delay (which might be interpreted as improved conduction) seen in the remaining electrodes during the fibrillation window occurs at the expense of conduction block in others. This observation suggests that loss of cell-to-cell interaction may contribute to the heterogeneity of conduction resulting from the loss of electrotonic interaction.
This correlation between the third phase of activation delay and the onset and progression of cell-to-cell electrical uncoupling is more evident on an individual experiment basis. In Fig 5⇓, potassium concentrations are plotted with corresponding activation delays from four discrete electrode sites. Superimposed are the resistivity measurements made during the same experiment (solid line). From this plot, one can see that the second phase of activation slowing occurs almost simultaneously with the onset of the second rise in resistivity, indicating cell-to-cell electrical uncoupling. In addition, there is little change in [K+]e during the period in which activation delay is increasing, indicating an effect predominantly of cell-to-cell electrical uncoupling. The plot also demonstrates the phenomenon described above: when the activation electrode in the top panel of Fig 5⇓ ceases to show local activation, mean activation delay for the four electrodes decreases (owing to the loss of values from an electrode with a relatively large delay), yet conduction is worsening in all electrodes at this time (evidenced by persistent block in the uppermost electrode and increasing delays in the others).
[K+]e and pHe
Fig 6⇓ shows the changes in mean [K+]e before the end of the fibrillation window; the inset in Fig 7⇓ shows typical changes from a single experiment. Ionic changes during ischemia in this series were the same as results previously reported from this laboratory.23 24 28 A rapid initial rise from the baseline mean of 3.4 mmol/L (range, 3.0 to 3.6±0.29 mmol/L) occurred during the first 8 minutes of ischemia, followed by a plateau phase during which [K+]e remained fairly constant. The plateau, at about three times the baseline concentrations (≈10±2 mmol/L), lasted ≈12 minutes. After the plateau, there was a second [K+]e rise, beginning gradually at ≈20 minutes of ischemia and rising more rapidly after 27 minutes.
The fibrillation window (shaded in Fig 6⇑) was characterized by a gradual [K+]e rise from 11.1±2.0 to 13.7±2.3 mmol/L (P=.0889). However, much of this [K+]e increase occurred during the last 2 to 3 minutes of the fibrillation window. In contrast, five of six episodes of VF occurred between 19 and 26 minutes, during which time [K+]e rose only from 11.1 to 11.7±2.0 mmol/L.
Cascio and colleagues37 showed that the second rise in [K+]e was closely related to and followed the onset of cell-to-cell electrical uncoupling. To test this relation, we plotted the second rise in [K+]e as a function of the second rise in Rt in Fig 7⇑. The method for determining the second rise in [K+]e is illustrated in the inset. The events correlate with R2=.834. There was little overall change in the mean [K+]e during the fibrillation window, but the second rise in [K+]e should correlate with VF because [K+]e correlates with the onset of uncoupling, which in turn correlates with VF. The mechanism relating the onset of the third phase of cellular K+ loss to cell-to-cell electrical uncoupling is unknown.
pHe was 7.40 at the onset of ischemia (7.28 to 7.45±0.06) and fell to a minimum plateau of 6.6±0.22 within the first 20 minutes of ischemia. pHe plateaued before the onset of the fibrillation window; relatively little change occurred during this period.
Association of the Reversal of TQ-Segment Depression to Cell-to-Cell Electrical Uncoupling
The injury current is manifest as TQ-segment depression. The amplitude of the TQ-segment depression that occurred during ischemia is plotted versus time in Fig 8⇓, again with the fibrillation window superimposed. Mean ΔTQ potential decreased to a minimum of −9.8±3.0 mV at 19 minutes. After reaching a minimum, the trend reversed; the mean potential increased to −5.0±3.5 mV over the next 11 minutes (P=.0174). After 30 minutes, mean ΔTQ changed little, although in three experiments, a continuous, slow drift toward zero potential was observed. This trend was not evident in the mean data.
Solid-angle theory predicts that the magnitude of the injury current is influenced by tissue conductance. A decrease in tissue conductance or, alternatively, an increase in tissue resistance secondary to cell-to-cell electrical uncoupling will decrease the magnitude of the injury current. To test this hypothesis, the change in the magnitude of the TQ-segment depression was measured during myocardial ischemia and related to the change in whole-tissue resistance.
Fig 9⇓ is a plot of the time of the onset of the second rise in resistivity (see above) versus the time of TQ potential reversal (ie, the time during ischemia at which the last minimum TQ value occurred, shown in the inset) for each experiment. A linear relation was observed with R2=.886; this relation was found to be highly significant, with P=.0002.
Mechanisms of Ia and Ib Arrhythmias
The biphasic distribution of early ischemic arrhythmias suggests different mechanisms. The Ia and Ib arrhythmias have been recognized for years,5 but speculation persists as to their genesis. Type Ia arrhythmias have been shown to depend primarily on reentrant mechanisms. This fact was first derived from evidence obtained from fragmented local electrograms. Electrical activity in the form of low-amplitude, multicomponent electrogram deflections (hence, “fragmented”) observed throughout the cardiac cycle provided evidence that continuous activation through reentry may occur.38 39 Reentry was confirmed to be a source of VF during the Ia phase in studies using activation mapping with high spatial resolution.40 In contrast, less abnormal conduction and electrogram fragmentation have been reported during the Ib phase.4 41 42 Sensitivity of the Ib arrhythmias to β-blockade and depletion of catecholamines,43 which are released between 15 and 20 minutes of ischemia,44 imply that the sympathetic nervous system contributes to abnormal automaticity as the mechanism underlying Ib arrhythmias. The present study suggests that electrophysiological abnormalities associated with reentry may also contribute to ventricular arrhythmias during the Ib phase, in addition to their role in the Ia phase. Cell-to-cell electrical uncoupling as a facilitator of Ib arrhythmias has been suggested9 but never previously demonstrated in an in vivo model such as this one.
Cellular Electrical Uncoupling Contributes to the Onset of Ib VF
This study correlates the onset of cell-to-cell electrical uncoupling during ischemia with the occurrence of malignant ventricular arrhythmias, specifically those that occur between 12 and 30 minutes after the onset of acute ischemia, the Ib phase.4 7 8 In our series, we were able to define a discrete window of fibrillation between 19 and 30 minutes of ischemia during which VF occurred in 6 of 9 pigs studied beyond the Ia phase. Premature ventricular beats and episodes of nonsustained ventricular tachycardia were also prevalent during this interval.
The electrophysiological parameters monitored during our experiments demonstrated the greatest rate of change during the fibrillation window. Tissue resistivity increased by slightly <50% during this period, and the occurrence of arrhythmias was closely associated temporally with the second rise in tissue resistivity, which is a measure of cell-to-cell coupling.9 Assuming that extracellular resistance remains constant or changes little during this phase, the 50% increase in whole-tissue resistance corresponds to an approximate threefold change in the intracellular resistance, indicative of marked cell-to-cell electrical uncoupling. Cell-to-cell electrical uncoupling will add to altered active membrane properties to enhance microscopic resistive discontinuities and impair impulse propagation. As expected, local tissue activation deteriorated during the fibrillation window. At the onset of the window, >90% of electrodes demonstrated local activation within the cardiac cycle. Eleven minutes later, ≈30% still remained active, providing evidence of developing inhomogeneities in the myocardial conduction properties during this period. Although there is less conduction slowing during the fibrillation window than during the initial minutes of ischemia, the improved mean conduction times occur at the expense of more conduction block, highlighting the potential for marked heterogeneity of conduction during this time. Among those electrodes continuing to show local activation at the end of the fibrillation window (n=10), we found that their ATs had increased 53%, from 36 to 55 ms, during the fibrillation window. Conduction slowing is unequivocally taking place during the Ib phase in areas that are not blocked to activation. It is clear from Fig 4⇑ that conduction inhomogeneities with large areas of conduction block are generated during the period of fibrillation, which creates a myocardial environment suitable for reentry. Fig 5⇑ further illustrates the association between progressive conduction slowing and cell-to-cell electrical uncoupling. ATs from four individual electrodes are plotted with the mean tissue resistivity in the same experiment. The second phase of activation slowing corresponds with the period of cell-to-cell electrical uncoupling. This figure also presents the lack of association between the activation slowing and [K+]e changes during this time (see below).
During ischemia, conduction is impaired and reentry occurs, which may lead to malignant arrhythmias.45 46 Those arrhythmias are logically facilitated by cell-to-cell electrical uncoupling.19 21 By creating pathways of nonuniform electrically uncoupled cells in which activation block occurs in one limb while conduction is preserved in another, a scenario suitable for reentry may be established. These conditions are sufficient to create reentrant circuits on the basis of spatial nonuniformities and to cause the malignant arrhythmias observed in our experimental series. The present study confirms the relation of the second increase in tissue resistance during ischemia and the occurrence of Ib arrhythmias in a whole-animal preparation. The occurrence of these arrhythmias in the setting of conduction slowing and block and cell-to-cell electrical uncoupling suggests that reentry may contribute to VF during the Ib phase. In addition, it appears that the mechanism of VF is not directly related to (although it may be dependent on) changes in [K+]e and [H+] that occur simultaneously (see below).
In contrast to the electrophysiological parameters, the ionic parameters showed relatively little change during the period of VF. The mean [K+]e rose from 11.1 to 13.7 mmol/L during the fibrillation window, and the onset of the second rise for the mean data occurred late within the arrhythmogenic phase. [K+]e rose only 0.6 mmol/L in the period between 19 and 26 minutes, during which all but a single episode of VF occurred. Furthermore, Fig 5⇑ demonstrates the lack of association between [K+]e changes and conduction slowing during the Ib phase, emphasizing that this effect is most likely due to simultaneous cell-to-cell electrical uncoupling. Although there seemed to be a lack of any association between the occurrence of VF and absolute or rate of change in [K+]e, there was an association between the time of the third phase of cellular K+ loss and the onset of cell-to-cell electrical uncoupling, as evidenced in Fig 7⇑. This relation is not unexpected because cellular uncoupling has been related temporally to the third phase of cellular K+ loss.37 This correlation would imply a relation between VF and the second rise in [K+]e, but because it occurs after uncoupling, its significance is unknown.
Like [K+]e, pHe had essentially plateaued during the period of arrhythmias, exhibiting less than 1/10th of a pH unit decrease during that time. These results strongly suggest that electrophysiological changes, namely cell-to-cell electrical uncoupling and increasing conduction delay and regions of conduction block superimposed on altered tissue excitability secondary to high K+, low pH, and the accumulation of other metabolic by-products, play an important role in the genesis of type Ib ventricular arrhythmias.
During the current series, 1 pig fibrillated during the Ia phase and 3 pigs did not fibrillate during the entire course of ischemia. The Ia phase VF occurred <2 minutes after the arrest of perfusion. Tissue resistivity had increased minimally (<30%) at this time. Mean [K+]e had reached 6.6 mmol/L and pHe was 7.26 in this experiment. The absence of uncoupling and the occurrence of VF with ongoing K+ and pH changes suggest that the type Ia arrhythmias are due to changes of active membrane properties alone or in combination with inherent microscopic resistive discontinuities. In the 3 pigs that had no VF, cellular electrical uncoupling occurred earlier (16, 17, and 20 minutes after no-flow ischemia) relative to the majority of the other experiments. This suggests that while cell-to-cell uncoupling is important for creating an arrhythmogenic myocardium, it is insufficient alone. Other factors, such as altered tissue excitability (caused by high K+, low pH, and lipid accumulation), are still involved; these factors appear to facilitate arrhythmias when uncoupling occurs during acute ischemia. The hypothesis that type Ib arrhythmias are dependent on both the timing and the presence of cell-to-cell electrical uncoupling is consistent with the fact that they may be suppressed by β-blockade.43 Application of β-receptor agonists has been shown to markedly increase intracellular resistance in cat hypoxic cardiac muscle.47 The presence of catecholamines or adrenergic blockade may therefore play an important role in the timing of cellular uncoupling or the initiation of a premature beat that reenters in the arrhythmogenic substrate created by cell-to-cell uncoupling.
The timing of uncoupling may also be critical in providing a substrate for reentry at a time when initiating beats may be induced by flow of injury current. The TQ potential difference between the ischemic and normal zones is the source of an injury current that may affect tissue excitability, especially in the border zones.22 40 48 49 As tissue resistance rises secondary to cell-to-cell electrical uncoupling, the injury current density increases. The resulting intensified current may potentially bring ischemic cells to the threshold and initiate premature ventricular beats. Premature beats may be induced by injury current either by reexcitation of cells at the end of the refractory period or by an increase in the rate of phase 4 depolarization.40 In our experiments, the magnitude of the TQ potential tended to be greater in experiments in which VF occurred (−10.3±3.0 [n=6] versus −7.3±2.6 mV [n=3], P=.098). In those experiments in which cell-to-cell electrical uncoupling occurred early, the TQ potential may not have reached a magnitude sufficient to initiate arrhythmias. We speculate that the critical time period during which the myocardium was susceptible to reentry owing to cellular uncoupling occurred before the peak period of premature beat generation resulting from injury current. Thus, these pigs escaped VF.
Cellular electrical uncoupling, with its attendant effects on cellular activation, may be a primary mechanism contributing to the genesis of arrhythmias during the Ib phase of ischemia, yet uncoupling ultimately provides a survival method for normal cells by establishing an ionic and electrical barrier between those cells and ischemic tissue.21 Electrophysiological heterogeneities that occur transiently at the onset of uncoupling appear to predispose the ventricle to arrhythmias that would adversely affect survival of the individual. This notion is suggested by our results. VF occurs early after the onset of cell-to-cell electrical uncoupling. Complete uncoupling abolishes the occurrence of VF and appears to abolish other arrhythmias during the Ib phase. The appearance of a third phase of ventricular arrhythmia between 50 and 60 minutes after the arrest of perfusion was unexpected. These arrhythmias occur under conditions when the ischemic zone is unexcitable and uncoupled. Most likely, they originate in the border zone.
Injury Potential Decreases in Proportion to Increasing Tissue Resistance
This series of experiments provides further insight into a phenomenon first described by Kléber29 : the decrease in the magnitude of TQ injury potentials occurring between 15 and 25 minutes after the onset of ischemia is due to cell-to-cell electrical uncoupling. TQ-segment shifts (manifest on the surface ECG as ST-segment elevations) occur during acute ischemia as a result of differences in the resting membrane potential between the ischemic and nonischemic zones; the voltage difference drives a current that is measured on the ECG. Factors such as the variation in the resting membrane potential, changes in tissue resistivity, and the spatial relation of the measuring electrode relative to the ischemic border may affect the measured potential. The time course of TQ-segment depression and “recovery,” ie, a return toward baseline values after a maximum depression is reached during early ischemia, was described previously.22 29 50 As Fig 10⇓ shows, changes in ΔTQ reported by others22 51 fall within 1 SD of the mean reported in this article.
A close association between cellular electrical uncoupling, evidenced by a tissue resistivity rise, and the time of the onset of TQ recovery was seen in these experiments (Fig 8⇑). Current experimental evidence confirms the presumption that the recovery of the injury potential is a function of increasing tissue resistivity. In addition, it serves as an in vivo validation of the method that utilizes the onset of the second rise in resistivity to indicate the onset of cellular electrical uncoupling. At the onset of uncoupling, resistivity rises rapidly in the ischemic zone; the current flowing as a result of the diastolic membrane voltage difference from normal tissue decreases proportionally. Mechanisms such as an actual recovery of the cell membrane in the ischemic zone are unlikely and probably do not contribute to the measured TQ recovery.
The observation that diminishing TQ-segment depression during ischemia occurs as a result of a loss of cell-to-cell electrical interaction has important clinical implications. Early reversal of ST-segment elevation has been believed to represent an expression of the effective treatment (and resulting reperfusion) of an acute myocardial infarction.52 53 The current data suggest that in some cases improvement in ST-segment changes may actually represent the evolution of the ischemic process. In addition, the transient recovery of ST-segment elevation during an infarction may contribute to false-negative ECG interpretations and provides a possible explanation for proven infarctions in which there was no ST-segment elevation apparent at the time of the ECG.54 55 Our experiments demonstrate that a diminishing current of injury does not exclusively indicate myocardial improvement.
It may be concluded from this series of experiments that the onset of cell-to-cell electrical uncoupling, evidenced by a rise in whole-tissue resistivity, is temporally associated with the onset of VF during the Ib phase. Tissue electrophysiological changes during this period, such as progressive conduction block and persistent activation delay, may create a myocardial environment suitable for reentry. This series of experiments provides evidence suggesting that deteriorating myocardial conduction properties resulting from cell-to-cell electrical uncoupling may enhance microscopic resistive discontinuities, create unidirectional block, and thus initiate malignant arrhythmias. This study also provides evidence that cell-to-cell electrical uncoupling causes reversal of the injury current during early ischemia as measured by TQ-segment depression.
Selected Abbreviations and Acronyms
|aK+||=||extracellular [K+]e activity|
|bpm||=||beats per minute|
|[K+]e||=||extracellular K+ concentration|
|LAD||=||left anterior descending coronary artery|
|PVC||=||premature ventricular contraction|
|ΔTQ||=||change of TQ potential|
This work was supported by grant PO1 HL-27430 from NHLBI, Bethesda, Md. William T. Smith IV is a Howard Hughes Medical Institute Medical Student Research Training Fellow. We wish to acknowledge the excellent technical assistance provided by Lisa Tomasko and Gary Koch, the Department of Biostatistics, University of North Carolina, Chapel Hill. We would also like to express our gratitude to Dr Leonard Gettes for his insightful comments and criticisms that helped us prepare this manuscript and to Cynthia Womack, David Martin, Weili Sun, Darrell Sandiford, and Zoe Sherman for their expertise.
- Received March 20, 1995.
- Revision received June 25, 1995.
- Accepted June 25, 1995.
- Copyright © 1995 by American Heart Association
Adgey AAJ. Sudden death, ventricular fibrillation, ventricular defibrillation: historical review and recent advances. In: Adgey AAJ, ed. Acute Phase of Ischemic Heart Disease and Myocardial Infarction. Boston, Mass: Martinus-Nijhoff; 1982:1-2.
Kaplinsky E, Ogawa S, Balke CW, Dreifus LS. Two periods of early ventricular arrhythmia in the canine acute myocardial infarction model. Circulation. 1979;60:397-403.
Haase M, Schiller U. Zur zeitlichen parallelität zwishen der activität ectopischer schrittmacher und dem eintritt von kammerflimmern nach ligatur eines hauptkoronarastes beim hund. Acta Biol Med Germ. 1969;23:413-422.
Meesman W, Wiegand V, Menken U, Komhard W, Rehwald U. Early mortality due to ventricular fibrillation and the vulnerability of the heart following acute experimental coronary occlusion: possible mechanisms and pharmacological prophylaxis. In: Bauer RD, Busse R, eds. The Arterial System: Dynamics, Control Theory and Regulation. New York, NY: Springer Publishing Co Inc; 1978:275-284.
Kléber AG, Riegger CB, Janse MJ. Electrical uncoupling and increase of extracellular resistance after induction of ischemia in isolated, arterially perfused rabbit papillary muscle. Circ Res. 1987;61:271-279.
Gettes LS, Cascio WE. Effect of acute ischemia on cardiac electrophysiology. In: Fozzard HA, Jennings RB, Haber E, Katz AM, Morgan HE, eds. The Heart and Cardiovascular System. 2nd ed. New York, NY: Raven Press Publishers; 1992:2021-2054.
DeMello WC. Influence of intracellular injection of H+ on the electrical coupling in cardiac Purkinje fibers. Cell Biol Int. 1980;4:51-57.
DaTorre S, Creer MH, Pogwizd SM, Corr PB. Amphipathic lipid metabolites and their relation to arrhythmogenesis in the ischemic heart. J Mol Cell Cardiol. 1991;23(suppl 1):11-22.
Sugiura H, Toyama J, Tsuboi N, Kamiya K, Kodama I. ATP directly affects junctional conductance between paired ventricular myocytes isolated from guinea pig heart. Circ Res. 1990;66:1095-1102.
Janse MJ, Kléber AG. Electrophysiological changes and ventricular arrhythmias in the early phase of regional myocardial ischemia. Circ Res. 1981;49:1069-1081.
Spach MS, Dolber PC, Heidlage JF. Influence of the passive anisotropic properties on directional differences in propagation following modification of the sodium conductance in human atrial muscle: a model of reentry based on anisotropic discontinuous propagation. Circ Res. 1988;62:811-832.
DeMello WC. Intercellular communication in cardiac muscle. Circ Res. 1982;50:2-35.
Kléber AG, Janse MJ, van Capelle FLJ, Durrer D. Mechanism and time course of ST and TQ changes during acute regional myocardial ischemia in the pig heart determined by extracellular and intracellular recordings. Circ Res. 1978;42:603-613.
Fleet WF, Johnson TA, Graebner CA, Gettes LS. Effect of serial brief ischemic episodes on extracellular K+, pH, and activation in the pig. Circulation. 1985;72:922-932.
Johnson TA, Engle CL, Boyd LM, Kock GG, Gwinn M, Gettes LS. Magnitude and time course of extracellular potassium inhomogeneities during acute ischemia in pigs. Circulation. 1991;83:622-634.
Hill JL, Gettes LS, Lynch MR, Hebert NC. Flexible valinomycin electrodes for on-line determination of intravascular and myocardial K+. Am J Physiol. 1978;235(Heart Circ Physiol 4):H455-H459.
Johnson TA, Engle CL, Kusy RP, Knisley SB, Graebner CA, Gettes LS. Fabrication, evaluation and use of extracellular K+ and H+ ion-sensitive electrodes. Am J Physiol. 1990;258(Heart Circ Physiol. 4):H1224-H1231.
LeBlanc OH, Brown JF, Klebe JF, Niedrach LW, Slusarczuk GMJ, Stoddard WH. Polymer/membrane sensors for continuous intravascular monitoring of blood pH. J Appl Physiol. 1976;40:644.
Hill JL, Gettes LS. Effect of acute coronary artery occlusion on local myocardial extracellular K+ activity in swine. Circulation. 1980;61:770-778.
Kléber AG. Resting membrane potential, extracellular potassium activity, and intracellular potassium activity during acute global ischemia in isolated perfused guinea pig hearts. Circ Res. 1983;52:442-450.
Kohlrausch F, Holborn L. Das leitvermgen der elektrolyte, insbesondere der lsungen. Methoden, resultate und chemische anwendungen. Leipzig, Germany: Teubner; 1898:74.
Lide DR, ed. CRC Handbook of Chemistry and Physics. 75th ed. Boca Raton, Fla: CRC Press Inc; 1994:5-89.
Cascio WE, Yan GX, Kléber AG. Passive electrical properties, mechanical activity and extracellular potassium in arterially perfused and ischemic rabbit ventricular muscle: effects of calcium entry blockade or hypocalcemia. Circ Res. 1990;66:1461-1473.
Scherlag BJ, El Sherif N, Hope RR, Lazarra R. Characterization and localization of ventricular arrhythmias resulting from myocardial ischemia and infarction. Circ Res. 1974;35:372-383.
Janse MJ, van Capelle FJL, Morsink H, Kléber AG, Wilms-Schopman F, Cardinal R, d’Alnoncourt CN, Durrer D. Flow of ‘injury’ current and patterns of excitation during early ventricular arrhythmias in acute regional myocardial ischemia in isolated porcine and canine hearts: evidence for two different arrhythmogenic mechanisms. Circ Res. 1980;47:151-165.
Penny WJ. The deleterious effects of myocardial catecholamines on cellular electrophysiology and arrhythmias during ischemia and reperfusion. Eur Heart J. 198;45:960-973.
Schoemig A, Dart AM, Dietz R, Mayer E, Kuebler W. Release of endogenous catecholamines in the ischemic myocardium of the rat. Circ Res. 1984;55:689-701.
Janse MJ, Wit AL. Electrophysiological mechanisms of ventricular arrhythmias resulting from myocardial ischemia and infarction. Physiol Rev. 1989;69:1049-1069.
Coronel R, Wilms-Schopman FJG, Opthof T, van Capelle FJL, Janse MJ. Injury current and gradients of diastolic stimulation threshold, TQ potential, and extracellular potassium concentration during acute regional ischemia in the isolated perfused pig heart. Circ Res. 1991;68:1241-1249.
Nielsen BL. ST-segment elevation in acute myocardial infarction. Circulation. 1973;48:338-345.
Poliwoda H. The thrombolytic therapy of acute myocardial infarction. Angiology. 1966;17:528-540.
Rude RE, Poole WK, Muller JE, Turi Z, Rutherford J, Parker C, Roberts R, Raabe DS, Gold HK, Stone PH, Willerson JT, Braunwald E, for the MILIS Study Group. Electrocardiographic and clinical criteria for recognition of acute myocardial infarction based on analysis of 3697 patients. Am J Cardiol. 1983;52:936-942.