Impact of Myocardial Ischemia and Reperfusion on Ventricular Defibrillation Patterns, Energy Requirements, and Detection of Recovery
Background— Shocks that have defibrillated spontaneous ventricular fibrillation (VF) during acute ischemia or reperfusion may seem to have failed if VF recurs before the ECG amplifier recovers after shock. This could explain why the defibrillation threshold (DFT) for spontaneous VF appears markedly higher than for electrically induced VF.
Methods and Results— The DFT for electrically induced VF (E-DFT) was determined in 15 pigs before ischemia, followed by left anterior ascending or left circumflex artery occlusion. VF was electrically induced 20 minutes after occlusion, followed 5 minutes later by reperfusion. Whether spontaneous or electrically induced, VF during occlusion or reperfusion was treated with up to 3 shocks at 1.5×E-DFT. If all 3 shocks failed, shock strength was increased. Thirty minutes after reperfusion, the other artery was occluded and the protocol was repeated. Defibrillation was considered successful if postshock sinus/idioventricular rhythm was present for ≥30 seconds. VF recurring within 30 seconds after the shock was considered immediate or delayed if the first postshock activation complex in a rapidly restored ECG recording was VF or sinus/idioventricular rhythm, respectively. Defibrillation efficacy at 1.5×E-DFT was significantly higher for electrically induced ischemic VF (76%) than for spontaneous VF (31%). The incidence of delayed recurrence after electrically induced nonischemic (3%) or ischemic (20%) VF was significantly lower than after spontaneous VF (75%). Mean VF recurrence time after spontaneous VF was 4.6±5.3 seconds.
Conclusions— Spontaneous VF can be halted by a shock but then quickly restart before a standard ECG amplifier has recovered from postshock saturation, making it appear that the shock failed.
Received September 7, 2001; revision received March 6, 2002; accepted March 6, 2002.
A major complication of coronary artery disease is sudden death caused by spontaneous VF secondary to acute ischemia.1 Although the defibrillation threshold (DFT) has been reported to be higher for spontaneous VF during acute ischemia or reperfusion than for electrically induced VF,2,3⇓ in most studies VF was electrically induced in the absence of regional ischemia.3 To determine the relative roles of acute ischemia and the mode of VF induction on elevation of the DFT, Ouyang et al2 compared the DFTs for spontaneous VF and electrically induced VF with or without acute ischemia. Their results suggested that the DFT was determined not only by the presence or absence of regional ischemia but also by whether VF occurred spontaneously or was induced electrically. Although the myriad electrophysiological changes induced by acute ischemia may explain why the DFT is elevated in the presence of regional ischemia independent of the mode of VF initiation,2,3⇓ it is less clear why, in the presence of ischemia, spontaneously occurring VF has a higher DFT than electrically induced VF. As one possible explanation, we hypothesized that the increased arrhythmogenic propensity of the ischemic tissue responsible for spontaneous initiation of VF remains present for some time after VF occurs so that after the original VF episode is halted by a shock, VF is rapidly reinitiated. Because of the large defibrillation shock, standard ECG amplifiers are saturated for many seconds after the shock. If VF is reinitiated during this interval, it could be falsely concluded that the defibrillation shock failed because it was too weak to halt the original spontaneous VF episode. To test this hypothesis, we determined the outcome of defibrillation shocks of the same strength in animals with acute ischemia during spontaneous and electrically induced VF episodes while recording cardiac electrical activity with amplifiers that resumed recording quickly after the shock.
The Animal Care and Use Committee at the University of Alabama at Birmingham approved this study. The guidelines in the Position of the American Heart Association on Research Animal Use4 were followed.
General anesthesia was induced in 15 healthy, mixed-breed pigs (20 to 44 kg; Animal Resource Program, University of Alabama at Birmingham) of either sex with intramuscular telazol (4.4 mg/kg), xylazine (2.2 mg/kg), and atropine (0.04 mg/kg). It was maintained by continuous administration of isoflurane (1% to 4%). An intravenous bolus of succinylcholine (0.25 to 0.50 mg/kg) was given several minutes before the first defibrillation attempt and repeated as needed to suppress muscular contraction in response to defibrillation shocks. The animal was placed in dorsal recumbency, intubated with a cuffed endotracheal tube, and ventilated with a pressure-controlled respirator (Model 2000, Hallowell EMC). The lateral aspects of the thorax were shaved for application of a pair of 115-cm2 self-adhesive defibrillation patches (Quick-Combo, Physio-Control Corp). The 2 external jugular veins, a femoral vein, a femoral artery, and a carotid artery were surgically isolated for placing hemostatic sheaths (8 to 10 F) and catheters. Under fluoroscopic guidance, 3 catheters containing stimulating or recording electrodes (EP Technologies) were introduced into the coronary sinus (CS), right atrium (RA), and right ventricle (RV). Pulmonary wedge and central venous pressures were monitored using a Swan-Ganz catheter engaged in a small distal pulmonary artery. A femoral arterial catheter was used for blood sampling and continuous arterial pressure monitoring. Blood gases were analyzed at least hourly, and corrections in Po2, Pco2, Na+, K+, Ca2+, and bicarbonate were maintained within normal ranges. Intravenous lactated Ringer’s solution was continuously administered (5 to 10 mL/kg per min). Body temperature was measured with an esophageal temperature probe and maintained at ≈37°C with a heating pad beneath the animal. The pig was euthanized by electrical induction of VF at the end of the experiment.
Determination of the DFT Before Acute Ischemia
Before coronary occlusion was performed, VF was induced by a 60-Hz current delivered through an electrode on the catheter in the RV apex. VF, indicated by a rapid drop of arterial pressure and disorganized ECGs, was allowed to continue for 15 seconds before defibrillation was attempted. A 200-μF external defibrillator (Lifepak 7b, Physio-Control) with energy outputs between 20 to 400 J was used to deliver a biphasic truncated exponential waveform to the defibrillation patches. The first-phase duration of the waveform was determined with the algorithm of Walcott et al5 using a time constant of 5 ms and the measured impedance. The second phase was two thirds of the first phase duration.
The DFT for electrically induced nonischemic VF (E-DFT) was determined using a multishock/episode step-up protocol.6 The initial shock strength was between 50 and 80 J. If this shock failed, shock energy was increased in 0.1-log steps until defibrillation succeeded. If successful defibrillation did not occur within 4 shocks, rescue shocks of 320 to 400 J were delivered, and the VF induction and defibrillation protocol was repeated with a beginning energy 0.1 log higher than the previous initial shock strength. If the initial shock succeeded, the protocol was repeated with a beginning energy 0.1 log lower than the previous initial shock strength until the first shock failed. The E-DFT was defined as the energy level of the first successful shock that was within 3 energy steps from that of an initially failed shock. Two E-DFTs were obtained and averaged. VF episodes were separated by at least 4 minutes.
Acute Coronary Balloon Occlusion
The left anterior descending artery (LAD) and left circumflex artery (LCX) were occluded sequentially, as described elsewhere.3 Briefly, an intravenous bolus of heparin (5000 to 10 000 IU) was administered just before coronary catheterization and repeated hourly (500 IU). With the help of a guidewire and guide catheter, an angioplasty balloon (3×20 mm) was inflated (4 to 6 atmospheres) immediately distal to the first diagonal branch of the LAD or ≈1 to 2 cm from the origin of the LCX. The choice of which artery to be occluded first was randomized for the first experiment. In later experiments, the artery first occluded was alternated. A coronary angiogram was performed to confirm both correct balloon position and complete occlusion to antegrade blood flow. After 20 minutes of occlusion, 60-Hz VF (electrically induced ischemic VF) was induced if the pig was then alive. Five minutes later, the balloon was deflated after being inflated for 25 minutes. Electrically induced ischemic VF and spontaneous VF during occlusion (occlusion VF) or reperfusion (reperfusion VF) was treated with up to 3 shocks at 1.5×E-DFT. If all 3 shocks failed, shock energy was increased in 0.1-log increments until defibrillation succeeded. Thirty minutes after reperfusion, the other artery was occluded and the protocol was repeated.
A computerized data acquisition system (Windaq 8, DATAQ Instruments) was used to simultaneously record electrograms from surface lead II, the defibrillation patches, and intracardiac electrodes in the RA, RV, and CS, plus femoral arterial pressure, central venous pressure, and pulmonary wedge pressure. Recording began immediately before electrical induction of nonischemic VF and lasted for 4 minutes or started immediately after balloon inflation and lasted for at least 30 minutes. Electrograms were sampled at 1000 Hz and amplified by Electronics for Medicine amplifiers (Honeywell Inc) that recovered from saturation about 300 ms after a shock. The low- and high-frequency filters were 1 and 500 Hz, respectively. To minimize signal saturation caused by electrical shocks, a relay was placed between the electrodes and the amplifiers to disconnect the recording system beginning 10 ms before and lasting until a few milliseconds after the end of a shock.
Defibrillation was considered successful if the pig remained in organized (sinus/idioventricular) rhythm for ≥30 seconds.7 Recurrent VF after a failed shock was considered immediate or delayed if the first postshock activation complex was VF or organized rhythm, respectively. VF recurrence time was defined as the interval between the onset of the shock and the first postshock VF complex. A perfusing beat was defined as one that produced a femoral systolic arterial pressure >30 mm Hg before any cardiac massage was attempted. The earliest postshock perfusion time was defined as the interval between the onset of the shock and the first perfusing beat.
Quantitative data were expressed as mean±SD and compared by the Mann-Whitney U test. χ2 tests were used to compare categorical variables. Although counted in calculating defibrillation efficacy, 23 failed shocks for electrically induced nonischemic VF and 8 for occlusion VF episodes were excluded in analyzing the earliest postshock perfusion time and VF recurrence time and recurrence patterns, because the first electrogram complexes after these shocks could not be definitely interpreted as VF or sinus/idioventricular rhythm. P<0.05 was considered to be statistically significant.
Twenty-four occlusions were performed in 15 pigs. Six animals completed the entire experiment, 5 died during one of the occlusion episodes, 3 died of asystole after successful defibrillation, and 1 died of VF occurring during coronary catheterization before the balloon was inflated. There were 35 spontaneous VF episodes, 17 during occlusion and 18 during reperfusion. Of 17 electrically induced ischemic VF episodes, 4 were preceded by occlusion VF whereas the other 13 were not.
The mean E-DFT was 99±43 J. The mean shock energy that defibrillated electrically induced ischemic VF, occlusion VF, and reperfusion VF was 167±54, 215±113, and 200±108 J, respectively (P=NS among the three).
Defibrillation Efficacy for Electrically Induced VF During Ischemia and Spontaneous VF During Ischemia or Reperfusion
The percent of successful shocks at 1.5×E-DFT (Figure 1) was significantly higher for electrically induced ischemic VF (76%) than for occlusion (23%) or reperfusion VF (40%). There was no significant difference between the percent of shocks at 1.5×E-DFT that stopped occlusion VF or reperfusion VF. Most successful shocks (73%) defibrillated within 80 seconds after spontaneous VF initiated (Figure 2).
The fraction of electrically induced ischemic VF defibrillated by 1.5×E-DFT (16 of 17) was significantly higher than that of occlusion VF (8 of 17) but did not differ from that of reperfusion VF (14 of 18). Increasing shock energy >1.5×E-DFT eventually defibrillated all electrically induced ischemic VF and reperfusion VF but failed to halt 29% (5 of 17) of occlusion VF episodes (P<0.05).
VF Recurrence Patterns After Failed Defibrillation
Figure 3 illustrates electrogram examples of immediate and delayed VF recurrence patterns. Although the longest VF recurrence time (25 seconds) occurred after a failed shock for occlusion VF (Figure 4), the interval between successful defibrillation of a spontaneous VF episode and the onset of the next spontaneous VF during the same occlusion-reperfusion episode ranged from 54 to 960 seconds, supporting the use of a recurrence time of 30 seconds as a cutoff criterion for classifying successful and unsuccessful shocks.7
VF recurred significantly earlier for electrically induced nonischemic VF (0.5±0.2 seconds) than for electrically induced ischemic VF (3.3±5.0 seconds), occlusion VF (4.1±5.3 second), or reperfusion VF (5.3±5.3 seconds). The mean recurrence time did not differ significantly among electrically induced ischemic VF, occlusion VF, or reperfusion VF. The incidence of delayed VF recurrence was significantly lower for electrically induced ischemic or nonischemic VF than for spontaneous VF during occlusion or reperfusion (Figure 5). The incidence of delayed recurrence after failed shocks was similar for occlusion and reperfusion VF (P=NS).
There was no linear correlation between VF recurrence time and shock strength for any VF types (P=NS). Figure 6 illustrates a scatter plot of these 2 parameters for spontaneous VF.
Postshock Perfusion After Failed Defibrillation
The percent of recurrent VF episodes preceded by at least 1 perfusing beat after failed defibrillation of electrically induced nonischemic VF (0%) was significantly lower than that of occlusion VF (33%) or reperfusion VF (41%) but did not differ among electrically induced ischemic VF (10%), occlusion VF, or reperfusion VF (P=NS). The earliest postshock perfusion, however, occurred significantly later after failed shocks for occlusion VF (1.4±0.7 seconds) than for reperfusion VF (0.8±0.3 seconds).
This study confirmed the previous findings that spontaneous VF during acute ischemia or reperfusion requires significantly higher defibrillation energy than electrical VF induced in the absence of ischemia and that 1.5×E-DFT is not sufficient shock energy to defibrillate most spontaneous VF.3 In addition, this study demonstrates that the percent defibrillation success for 1.5×E-DFT energy shocks is significantly higher for electrically induced ischemic VF than for occlusion or reperfusion VF. This finding extends a previous observation by Ouyang et al2 that the canine DFT during 10-minute acute coronary occlusion or the ensuing reperfusion is lower for electrically induced ischemic VF than for spontaneous VF. These findings are partially explained by the new finding in this study that 2 markedly different VF recurrence patterns occur after failed defibrillation shocks. After failed shocks for most electrically induced VF, whether nonischemic or ischemic, and for a small percent of occlusion or reperfusion VF, VF is recorded as soon as the amplifiers have recovered (≈300 ms) from shock-induced saturation without any intervening organized electrical or mechanical activity. In contrast, after most failed defibrillation shocks for occlusion or reperfusion VF, postshock organized beats, many of which generate a pulse pressure, occur during a postshock interval of 4.6±5.3 seconds before VF recurs.
The presence of the first of these 2 VF patterns, immediate restoration of VF on recovery of the amplifiers, suggests 2 possible mechanisms. First, the shock may have been too weak to halt the VF wavefronts. Second, VF may have been stopped, but a new VF episode could have been initiated by the shock itself via a critical point or focal mechanism.8–10⇓⇓ No matter which of the 2 mechanisms is operative, insufficient shock strength may explain the immediate VF recurrence pattern. Insufficient shock strength according to either mechanism, however, cannot explain the delayed VF recurrence pattern. The postshock organized rhythm, the prolonged time until VF recurrence, and, particularly, the high incidence of postshock perfusing beats all strongly suggest that shocks ≥1.5×E-DFT energy stopped most spontaneous VF episodes but that a new postshock VF episode was reinitiated.
Acute ischemia could be one factor responsible for this reinitiation, because it can cause many detrimental changes, such as increased sympathetic activity,11,12⇓ hyperkalemia,13 and acidosis,13 which have been shown to increase vulnerability to VF. These ischemia-induced changes may have caused the reinitiation of VF within the 30 seconds after the defibrillation of spontaneous VF. Yet ischemia-induced alterations should also have existed after defibrillation of electrically induced ischemic VF. Then, why is spontaneous VF more difficult to defibrillate than electrical VF induced during occlusion? One possible reason is that ischemia was more severe when VF occurred spontaneously than when it was electrically induced. However, in a study in which the LAD and LCX were alternatively ligated with a protocol similar to ours,2 electrically induced ischemic VF was induced only when no spontaneous VF occurred throughout the ligation. Yet there was no significant difference in the ischemic mass among dogs with electrical VF induced during occlusion or spontaneous VF during occlusion or reperfusion. Thus, the ischemic area had little, if any, influence over the DFTs in that study.
Strong shocks may cause conduction block and prolonged depolarization,14 which in turn can cause VF. It is also possible that ischemic tissue is more vulnerable to the effects of strong shocks.15 If so, strong defibrillation shocks delivered during acute ischemia or reperfusion may contribute to the reinitiation of VF after an interval of postshock organized rhythm, leading to defibrillation failure and delayed VF recurrence. However, the lack of correlation between shock strength and VF recurrence time (Figure 6) and the substantial fractions of occlusion (4 of 12) and reperfusion VF episodes (4 of 18) defibrillated by shocks >1.5×E-DFT energy level do not support this scenario. Instead, they are consistent with another hypothesis: defibrillation outcome during spontaneous VF is determined not only by shock energy delivered to the heart but also by arrhythmogenic propensity of the acutely ischemic or reperfused tissue, particularly the original trigger responsible for the spontaneous initiation of VF.
It is possible that whatever initiates the original spontaneous VF episode is still present to induce VF recurrence during the 30 seconds after the shock. Although the exact duration of the increased arrhythmogenic propensity responsible for the initiation of spontaneous VF episodes is not known, evidence from this study suggests that it frequently has decreased or disappeared within 80 seconds after the onset of the original VF episodes (Figure 2). The possibility that the duration and degree of the increased arrhythmogenic propensity could differ in occlusion or reperfusion episodes or during different portions of the same episode may help explain the large standard deviation in VF recurrence time for delayed recurrent spontaneous VF. The duration of increased arrhythmogenic potential may be longer during occlusion than during reperfusion, as supported by the following evidence: of the 17 occlusion VF episodes, 5 were never halted with shocks as great as 400 J, yet increasing shock strength up to 400 J eventually defibrillated 100% of reperfusion VF episodes, suggesting that the arrhythmogenic mechanism during reperfusion, which is thought to be caused by sudden release of large amounts of accumulated ischemic metabolites,11,12⇓ is probably different from that during occlusion. The eventual successful defibrillation of reperfusion VF episodes may not be a direct result of multiple strong shocks; rather, it is possible that delivery of multiple shocks allowed time for the condition responsible for the spontaneous initiation of reperfusion VF to dissipate. Although the incidence of postshock perfusing beats is comparable after failed shocks for reperfusion and occlusion VF episodes, these perfusing beats may allow washout of arrhythmogenic substances during reperfusion to a higher extent than during occlusion because the artery perfusing the ischemic tissue is open after reperfusion but is still occluded during occlusion VF.
These findings raise the possibility that although increasing shock strength may defibrillate a higher percentage of immediately recurrent VF episodes, multiple shocks of escalating strength may not improve defibrillation efficacy of persistently recurring VF during acute ischemia or reperfusion in all patients. Instead, a defibrillation shock of the same appropriate strength repeated after a period of cardiac massage, during which the original arrhythmogenic conditions may decrease, might be more efficacious than a shock given immediately after the recurrence of the spontaneous VF episode. Although this is an external defibrillation study, these findings probably are also applicable to selection of shock energy levels for implantable cardioverter-defibrillators (ICDs) in patients with acute ischemia or reperfusion. For example, programming ICDs to maximum output to prevent the elevated DFT during ischemia may not always be helpful or may even be harmful if the mechanisms of defibrillation are not related to insufficient shock energy. Antiarrhythmic drugs, such as intravenous sotolol,16 which has been shown to prevent early recurrence of atrial fibrillation after successful cardioversion and intravenous amiodarone during cardiopulmonary resuscitation,17 may also reduce the incidence of VF recurrence. All of these ideas require comprehensive testing before being used clinically, and it is extremely important to exercise caution when extrapolating the findings in this animal study to potential changes in the clinical protocol for shock-energy sequences.
Acute ischemia was induced by an inflated balloon in a structurally normal animal, so the arrhythmogenic effects of atherosclerotic lesions and acute thrombi were not present in the study. Defibrillation was attempted within <1 minute after the onset of a spontaneous VF episode; thus, results from this study may not hold for longer durations of VF before the shock. It is possible that some VF episodes during occlusion or reperfusion could have been halted by shocks weaker than 1.5×E-DFT, but such lower shock strengths were not tested.
This study was supported in part by the National Institutes of Health Research Grants HL-28429, HL-42760, and HL-63775 and by Medtronic Physio-Control Corp. The authors thank Frank L. Vance, Tracy L. Gamblin, and Windy H. Jones for their excellent technical assistance during the study.
- ↵AHA Special Report. Position of the American Heart Association on research animal use. Circulation. 1985; 71: 849A–850A.
- ↵KenKnight BH, Hahn SJ, Heil JE, et al. A high resolution method for comparing defibrillation efficacy of experimental treatments in pigs.In: Szeto AYJ, Rangayyan RM, eds. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway, NJ: Institute of Electrical and Electronics Engineers; 1993: 875–876.
- ↵Chen P-S, Shibata N, Dixon EG, et al. 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.
- ↵Efimov IR, Cheng Y, Van Wagoner DR, et al. Virtual electrode-induced phase singularity: a basic mechanism of defibrillation failure. Circ Res. 1998; 82: 918–925.
- ↵Shaw RM, Rudy Y. Electrophysiologic effects of acute myocardial ischemia: a mechanistic investigation of action potential conduction and conduction failure. Circ Res. 1997; 80: 124–138.
- ↵Yabe S, Smith WM, Daubert JP, et al. Conduction disturbances caused by high current density electric fields. Circ Res. 1990; 66: 1190–1203.
- ↵Tse HF, Lau CP, Ayers GM. Incidence and modes of onset of early reinitiation of atrial fibrillation after successful internal cardioversion, and its prevention by intravenous sotalol. Heart. 1999; 82: 319–324.