Pinacidil s Effects on Defibrillation Outcomes: Role of Increased Potassium Conductance Via the KATP Channel

Transcription

1 Pinacidil s Effects on Defibrillation Outcomes: Role of Increased Potassium Conductance Via the KATP Channel Allison P. Winecoff, PharmD,* J. Jason Sims, PharmD,* Michael L. Markel, MD, Michael R. Ujhelyi, PharmD* Background: It has been shown that the inhibition of potassium ion conductance decreases defibrillation threshold. We postulated that if potassium conductance is a primary mechanism affecting defibrillation threshold values, then increasing potassium ion conductance will increase defibrillation values. The primary objective of this study was to determine if the ATP-dependent potassium (KATP) channel opener pinacidil would increase defibrillation threshold values. The second objective was to prove that the observed changes were due to potassium conductance by using the KATP inhibitor, glyburide, to reverse the electrophysiologic actions of pinacidil. The third objective was to determine if the electrophysiology actions of pinacidil correlate with changes in defibrillation threshold value. Methods and Results: Domestic farm swine (n 14) = were anesthetized and intubated. Subsequently, they were instrumented with monophasic action potential catheters and epicardial defibrillation patches. Defibrillation threshold values, action potential duration, effective refractory period, and ventricular fibrillation cycle length were determined at baseline and during treatment phase 1 and treatment phase 2. Pigs were randomized into 2 groups: group 1 (n 6) received D5W in = treatment phase one followed by D5W in treatment phase 2 and group 2 (n 8) received pinacidil in = treatment phase one followed by the addition of glyburide in treatment phase two. DFT ED50 did not change at baseline, treatment phase one or treatment phase two for group 1 (10.5 ± 2, 11.1 ± 1.7, 10.5 ± 1.0 J) or for group 2 (10.1± 2.2, 11.4 ± 4.2, 11.4 ± 3.0 J). Electrophysiologic parameters (QRS, effective refractory period, action potential duration 90, and ventricular fibrillation cycle length) were not significantly changed from baseline in group 1. In contrast, effective refractory period, action potential duration 90, and ventricular fibrillation cycle length significantly decreased at all recorded sites after the administration of pinacidil in group 2 (range of 7-13%, 6-9%, and 12-17%, respectively). However, pinacidil did not change the basal level of dispersion in effective refractory period, action potential duration, and ventricular fibrillation cycle length during paced rhythm or ventricular fibrillation. Glyburide reversed pinacidil s electrophysiologic actions. Conclusions: Pinacidil does not alter defibrillation threshold, but it reduces effective refractory period, action potential duration, and ventricular fibrillation cycle length and does not increase electrical heterogeneity. Therefore, changes in potassium channel conductance as well as shortening repolarization are unlikely primary mechanisms for elevating defibrillation threshold. Key words: ventricular fibrillation, ventricular defibrillation, pinacidil, glyburide, cardiac electrophysiology, ion conductance. From the *University of Georgia College of Pharmacy, Medical College of Georgia College of Medicine, Augusta VA Medical Center, Augusta, Georgia; and the University of Cincinnati College of Medicine, Cincinnati, Ohio. This work was supported by a Grant-in-Aid from the American Heart Association Georgia Affiliate. The authors are grateful of the laboratory support provided by the Augusta Veterans Administration Medical Center. Support of this study in the form of defibrillation equipment was provided by Cardiac Pacemaker (St. Paul, MN). Reprint requests: Dr. Michael R. Ujhelyi, Medical College of Georgia, Room CJ-1020, Augusta, GA

2 172 Patients with implantable cardiac defribrillators are to reduce commonly treated with antiarrhythmic drugs the number of shocks that they receive. However, previous evidence has demonstrated that some antiarrhythmic agents modify cardiac electrophysiology such that energy requirements necessary to evoke successful defibrillation are altered (1-5). Although the electrophysiologic mechanisms for altering defibrillation thresholds (DFT) are unknown, a theoretical model has suggested that cation conductance is a primary mechanism regulating defibrillation outcomes (6). This model proposes that altering sodium and potassium conductance will produce a change in transmembrane potential. A change in transmembrane potential may affect the shock s ability to depolarize a critical mass of myocardium needed for successful defibrillation. Hence, inhibiting inward sodium conductance should raise DFT values and enhancing outward potassium conductance should also raise DFT values. However, only portions of this model have been tested. It has been well documented that inhibition of sodium conductance raises DFT values ( (-3), while the inhibition of outward potassium conductance decreases DFT values (3-5). On the other hand, recent data indicate that enhancing sodium conductance has no effect on DFT (7), whereas, the effect of enhanced potassium conductance on defibrillation energy requirements has not been evaluated. If inhibition of sodium or potassium conductance is the primary mechanism by which antiarrhythmic drugs affect DFT values, then enhancing conductance of these cations should affect DFT values in the opposite manner. Recent evidence suggest that this is not the case, where increased sodium channel conductance did not decrease DFT values (7). This study, however, was limited regarding the ability to actually prove that increased sodium conductance was achieved with hypertonic saline and resultant hypernatremia. A more feasible study would be to investigate the effects of enhanced potassium conductance since the potassium channel opener, pinacidil, has been shown to increase potassium conductance with electrophysiologic changes consistent with this ionic effect (8). Thus, the primary objective of this study was to determine if the adenosine triphosphate (ATP)-dependent potassium (KATP) channel agonist pinacidil would increase DFT values. Based on data with potassium channel blocking agents, we postulate that increasing potassium channel conduction should increase DFT values. The second objective was to prove that the observed changes were due to potassium conductance by using the KATP antagonist, glyburide, to reverse the electrophysiologic actions of pinacidil. The final objective was to determine if changes in electrophysiologic actions by pinacidil correlated with changes in DFT. Methods Animal Preparation and Surgical Instrumentation Domestic farm pigs weighing between kg were used in this investigation. All procedures were approved by the Medical College of Georgia and the Augusta VA Medical Center Animal Care and Use Committees prior to conducting this investigation. All animals were fasted overnight. On the morning of the investigation, the animals were premedicated with ketamine (15 mg/kg) administered intramuscularly. Subsequently, pentobarbital (25 mg/kg) was administered intravenously for initial anesthesia induction. Following intubation with a cuffed endotracheal tube, the animals were mechanically ventilated using a large animal Harvard pump ventilator. A level plane of anesthesia was subsequently maintained throughout the study period using pentobarbital (demonstrated not to affect defibrillation threshold), as a continuous infusion of mg/hour (9). The right external jugular vein, internal carotid artery, and the femoral artery were cannulated for catheterization, drug infusion, and blood collection. A combination pacing and contact monophasic action potential catheter (EP Technologies, Mountain View, CA) was placed into the right ventricular apex via the external jugular vein, and a second catheter was placed into the left ventricle against the left lateral wall via the internal carotid artery. Both catheter placements were guided by fluoroscopy. These catheters recorded monophasic APD from the right and left ventricular endocardium and paced the ventricles. A pigtail 5F Millar pressure sensing catheter was placed via the femoral artery for blood pressure monitoring. Surface electrocardiographic leads were placed on the four limbs for monitoring of leads II and AVF. The chest was opened using a mediastinotomy. One 14 cm - and one 28 CM2 titanium mesh patch electrodes (Models A and L 67, respectively, CPI-Guidant, St. Paul, MN) were sutured onto the surface of the pericardium. The large electrode was placed over the anterior and lateral wall of the right ventricle which was perpendicular to the small electrode placed over the lateralbposteriorbapical wall of the left ventricle. The electrodes were interfaced with an external defibrillator where the right ventricular patch served as the anode. The defibrillator was capable of delivering a monophasic truncated waveform at a 65% fixed tilt with a pulse duration

3 173 between 5-8 ms. The output of this device was determined by preset voltage adjustments (1-V increments) (Ventak ECG, CPI-Guidant, St. Paul, MN). An epicardial spring cantilever monophasic action potential probe and two platinum pacing wires were placed on the apex of the left ventricular epicardium. The chest was closed and chest tubes were placed into the pleural space for drainage via suction. Arterial blood gases, serum potassium, sodium, and glucose were measured every minutes throughout the duration of the protocol. Body temperature was monitored via a rectal probe and maintained at C using a surgical thermal blanket. Adequate hydration was maintained using lactated ringer solution 2-5 ml/kg/h. Study Design We chose to use the ATP-dependent potassium (KATP) channel agonist, pinacidil, to increase outward potassium conduction and thus decrease APD and refractoriness. To prove that the observed effects were secondary to potassium conductance, glyburide, a KATP antagonist, was used to reverse pinacidil s effect. The experiment consisted of three phases where defibrillation threshold and electrophysiologic parameters were measured during a baseline phase, which was followed by two treatment phases. Each pig was randomly assigned to a group: group I baseline followed by placebo (dextrose [D5W] followed by = placebo (D5W) plus placebo (D5W + DMSO) (n 6) = and group 2 baseline followed by pinacidil followed = by pinacidil plus glyburide (n 8). DMSO (10% v/v) = was used as the solvent for glyburide; therefore, it was included in the phase 2 placebo arm to control for its electrophysiologic effects. The baseline phase was started 30 minutes after completion of instrumentation. The first treatment phase began immediately following the completion of the baseline phase where the treatment (D5W or pinacidil) was administered as a 10-minute loading dose of pinacidil (1.5 mg/kg) followed by a continuous infusion (0.75 mg/kg/h). D5W served as placebo and was given in equal volume to the pinacidil infusion. The second treatment phase began after the completion of the first treatment phase where pinacidil or DSW continued to be infused and either glyburide (3 mg/kg load, 3 mg/kg/h infusion in DMSO 10% v/v) or vehicle (D5W + DMSO 10% v/v) was added as a 10-minute loading dose followed by a continuous infusion. Vehicle was given at the same infusion rate as if glyburide was being infused. Defibrillation threshold and other measurements were initiated 10 minutes after the end of the loading dose (20 minutes after initiation of loading dose) for both treatment phases so that testing began after the drug distribution phase. The dose of pinacidil was based on previous data where it was shown to markedly decrease the APD (8). However, we chose a slightly lower dose due to the profound hypotension reported by these investigators. Defibrillation Threshold Determination Ventricular fibrillation was induced by delivering a stimulus drive train.with a 100-ms cycle length for 2 seconds at a stimulus strength of 10 V (Grass S8800 Stimulator, Quincy, MA). Defibrillation shocks were applied using a truncated exponential waveform with a 65% fixed tilt using preset energy levels 10 seconds after the initiation of ventricular fibrillation. The time between defibrillation trials was at least 4 minutes but not until arterial blood pressure returned to within 10% of the preshock value. To quantitate defibrillation threshold, a step-down step-up method was used as previously described (9). Energy, impedance, pulse width and peak current delivered to the myocardium were measured by the defibrillator and subsequently printed. These values are accurate to within 10% of oscilloscopic measurements in our laboratory which is consistent with data on file with CPI-Guidant (St. Paul, MN). The defibrillation threshold response for each test was modeled based on the response at each energy level within a treatment phase using an iterative computer program (MERFFIT, CPI-Guidant, St. Paul, MN) ( 10). Electrophysiology Parameters The RR interval was obtained during normal sinus rhythm from surface electrocardiograph lead II and averaged from five consecutive beats. The QRS and JT intervals during right ventricular pacing at 300-ms cycle length were also obtained from surface electrocardiograph lead II, and averaged from 5 consecutive beats. Ventricular pacing was continued for 15 seconds prior to measuring these electrocardiographic parameters. A global assessment of ventricular conduction velocity was determined by QRS duration (lead II) during right ventricular pacing at 300-ms cycle length. Myocardial repolarization was assessed for right and left ventricular endocardium and left ventricular epicardium by simultaneous measurements of the monophasic APD at 90% of complete repolarization (APD90) at a pacing cycle length of 300 ms. Measurements were made by a blinded investigator using a digitizing pad interfaced with a computer program (Sigma Scan, Jandel Scientific,

4 174 Corte Madera, CA). The effective refractory period (ERP) for each myocardial site was determined by pacing the site (right ventricle, left ventricle, or left ventricular epicardium) for eight beats using a stimulus intensity twice the diastolic threshold at a cycle length of 300 ms followed by one premature extra stimulus. The drive train was repeated after a 3-second pause and the extra stimulus coupling interval was discriminated by 2 ms until ventricular capture failed on two consecutive attempts. Ventricular fibrillation cycle length (VFCL) was measured from monophasic action potentials at each recording site 7-8 seconds after induction of ventricular fibrillation where fibrillation has been shown to stabilize (11). VFCL was calculated as the average number of action potential upstrokes over a 2-second period. Action potentials with double potentials or those that were fractionated were counted as a single activation. Dispersion in myocardial electrophysiology (APD, ERP, or VFCL) was measured using the three discrete myocardial recording sites. Dispersion was calculated as the maximum less the minimum value between the three myocardial sites (12). All electrophysiologic measurements were obtained at the start of defibrillation threshold protocol and at the end of defibrillation threshold protocol for both baseline and drug treatment phases. These values were then averaged for each study phase. Fig. 1. Bar graph showing ED5v DFT values at baseline, treatment phase 1 (D5W or Pinacidil), and treatment phase 2 (D;W + D5W or Pinacidil + Glyburide) for each group. Data Analysis A two-way analysis of variance was used to test differences between parameters determined at baseline and the first and second treatment phases within a group (measurements using the animal as its own control). Between group comparisons were analyzed as the percent change from one study phase to another (which normalized the data to account for baseline of variance. differences) using one-way analysis Posthoc analysis for significant differences was determined using the Student-Knewman-Keuls test for both ANOVA tests. Data are presented as mean ± SD and the significance level was set at P <.05. Results Defibrillation Threshold Baseline mean DFT values between the two groups were not significantly different (Fig. 1 ). DFT values for group I during each treatment phase (DSW, and DSW plus DSW) did not differ significantly from baseline ( 10.5 ± 2.0 to 11.1 ± 1.7 to 10.5 ± 1.0 J), illustrating the consistency of our data over time. Moreover, the DFT values for group 2 (pinacidil and pinacidil plus glyburide) did not change significantly after the administration of pinacidil (treatment phase 1) (10.1 ± 2.2 J to 1 l.4 ± 4.2 J) or after the addition of glyburide (treatment phase 2) (11.4 ± 4.2 J to 10.6 ± 3.0 J). Consequently, the values for treatment group 2 did not significantly differ from those of treatment group I (Fig. 1 ). Electrophysiologic, Hemodynamic, and Metabolic Parameters The mean and standard deviation of electrophysiologic values for both groups are reported in Tables I and 2. In treatment group 1, no changes were observed in any of the electrophysiologic parameters (paced QRS, paced JT, RR interval, APD9o, ERP, VFCL) measured at each phase. Pinacidil administration markedly decreased APD90, ERP, and VFCL for all three sites (Table 1). The magnitude of changes range from 6-9%, 7-13%, and 12-17% for APD9o, ERP, and VFCL, respectively. Representative tracings of monophasic action potential and VFCL shortening with pinacidil are shown in Figs. 2 and 3, respectively. Accordingly, the paced JT interval (Table 2) was decreased to a similar magnitude (12%) as the other measurements of repolarization. In addition, the RR interval was significantly decreased (442 ± 49 ms to 348 ± 42 ms) with the administration of pinacidil. In contrast, pinacidil did not affect ventricular conduc-

5 175 Table 1. Intracardiac Electrophysiologic Parameters * Baseline versus Pinacidil and Pinacidil + Glyburide, P <.05. Pinacidil versus Pinacidil + Glyburide, P <.05. t D5W versus Pinacidil and Pinacidil + Glyburide, P <.05. APD9c, action potential duration at 90% repolarization; Epi, left ventricular epicardium; ERP, effective refractory period; Lido, lidocaine; LV, left ventricle; RV, right ventricle; VFCL, ventricular fibrillation cycle length. tion velocity as demonstrated by similar paced QRS intervals at baseline and after pinacidil (84 ± 1 ms to 84 ± 1 ms). The addition of glyburide to pinacidil reversed the RV APD90 and EPI APD90 to values similar to baseline, however, LV APD90 showed a trend toward reversal but did not significantly change with the addition of glyburide. In addition, glyburide reversed pinacidil s effect on the ERP and JT interval, increasing these values back to baseline levels. Conversely, neither the VFCL nor the RR interval changed from phase I values with the addition of glyburide. Finally, paced QRS remained unchanged with the addition of glyburide. The mean and standard deviation for mean arterial pressure in both groups are reported in Table 2. In treatment group I there were no changes observed in mean arterial pressure for any of the treatment phases. In contrast, mean arterial pressure was significantly decreased from baseline values with the administration of pinacidil (94 ± 11 mmhg to 66 ± 7 mmhg). Partial reversal of mean arterial pressure was noted with the to 77 ± 13 addition of glyburide (66 ± 7 mmhg mmhg), however, this was not statistically significant. Metabolic indices were measured every minutes throughout the protocol. Arterial ph was maintained between 7.37 and 7.45, Pa02 between 80 and 120 mmhg, and PaC02 between 35 and 45 mmhg. Abnormal values were corrected via ventilator changes. Serum sodium and potassium were maintained between 135 and 145 meq/l and 3.5 and 4.5 meq/l, respectively. Abnormal concentrations were treated with intravenous supplementation. Finally, serum glucose was measured and remained between mg/dl throughout each phase of the experiment without supplementation. Electrical Heterogeneity The difference in APDQo, ERP, and VFCL between each site was measured during each phase in order to determine if pharmacologic intervention changed dispersion in myocardial electrophysiology during paced rhythm (300 ms) and fibrillation. Figure 4 represents the dispersion in APD9(), ERP, and VFCL. In the control group, the basal level of dispersion in APD90, VFCL, and ERP did not change throughout the experiment. Similarly, pinacidil did not affect the basal level of dispersion in APD90, ERP, or VFCL. The addition of glyburide did increase the level of APD9o dispersion without affecting ERP or VFCL. This reflects the fact that glyburide reversed APD shortening of the right ventricular and epicardial sites, but not that of the left ventricle. This is of unlikely significance since dispersion in refractoriness was unaffected. Discussion Decreasing potassium channel conductance, thereby, increasing the ERP, decreases energy requirements necessary for successful defibrillation (3-5). However, the opposite, (increasing potassium conductance, therefore, decreasing ERP) has never been evaluated. If potassium ion conduction is primarily responsible for affecting DFT values, then increasing outward potassium conductance should increase DFT. The current study addressed this hypothesis and showed that the K,e,TP channel agonist, pinacidil, did not alter defibrillation outcomes. Electrophysiologic data were consistent with.increased outward potassium conductance since there was a significant decrease in monophasic

6 176 Fig. 2. Tracing of monophasic action potential at three myocardial sites: left ventricular (LV) epicardium (upper panel), right ventricular (RV) endocardium (middle panel) and left ventricular endocardium (lower panel) in a single animal during baseline phase and after receiving pinacidil. APD, ERP, and VFCL. Further proof that pinacidil increased outward potassium conductance was established when glyburide, the KATP channel antagonist, reversed pinacidil s electrophysiologic effects. Finally, pinacidil did not affect myocardial electrical heterogeneity since the basal level of myocardial dispersion in refractoriness, monophasic APD, and VFCL was unaffected by pinacidil. These data are important because they eliminate enhanced potassium conductance and shortening of repolarization as possible mechanisms by which drugs may increase defibrillation threshold values. Changes in cation conductance have been postulated as the primary mechanism by which antiarrhythmic drugs alter energy requirements for defibrillation (6). Previous data have shown that agents that inhibit sodium conductance either increase or do not affect DFT values, whereas, agents that inhibit potassium conductance, lower DFT (1-5,13). The current data illustrate that enhanced outward potassium conductance via K~P does not affect DFT values. It is unlikely that enhancing outward potassium conductance via a different channel would yield different results, since both specific and nonspecific blockers of Ik1 and Ikr,s have been shown to decrease DFT values by equal magnitude (3-4,14-15). Thus, the current data indicate that pharmacologic changes in ion conductance are not the primary mechanism by which antiarrhythmic agents affect defibrillation. It is more likely that antiarrhythmic agents alter DFT values by changing myocardial electrophysiology, a result of their effects on ion conductance (3,14). It has been proposed that drugs that shorten APD and perhaps impulse wavelength increase DFT values (3,16,17). In the current study pinacidil decreased APD and ERP but did not affect conduction velocity. Thus, pinacidil decreased the impulse wavelength. A reduction in the impulse wavelength can increase the excitable gap of a circuit, which may facilitate reentry and thus the propagation of postshock fibrillation wave fronts. Hence, higher energy levels may be needed for successful defibrillation. However, these electrophysiologic changes were not associated with changes in DFT values. These results are consistent with previous findings where electrophysiologic parameters measured during rapid pacing did not correlate with changes in DFT values during antiarrhythmic drug treatment (16). Therefore, these electrophysiologic alterations, measured during rapid pacing, are unlikely mechanisms for drug induced changes in DFT values. The above data, however, do not rule out the possibility that changes in myocardial electrophysiology are a mechanism by which drugs alter DFT values. Rather, measurements of cardiac electrophysiology during rapid pacing are not reflective of the electrophysiologic state during ventricular fibrillation at the time of the shock or immediately after. Therefore, the nature by which drugs alter myocardial electrophysiology during fibrillation or immediately postshock may predict and perhaps cause DFT values to change. Decreasing refractoriness during fibrillation (18), illustrated by a decrease in the VFCL, has been shown to

7 177 Table 2. Blood Pressure and Electrocardiographic Parameters * Baseline versus Pinacidil and Pinacidil + Glyburide, P <.05. Pinacidil versus Pinacidil + Glyburide, P <.05. D5W versus Pinacidil and Pinacidil + Glyburide, P <.05. MAP, mean arterial pressure; RR, heart rate. predict increased DFT values (17). Data from the present study, however, indicate that decreasing VFCL does not predict changes in DFT. Our data are consistent with those of others who have also shown that changes in VFCL cannot accurately predict how a drug will affect DFT values (3,19). On the other hand, recent evidence suggests that any intervention that makes fibrillation activations more refractory and organized trends to decrease DFT values (19-21). These findings are consistent with optical mapping studies, which have shown that the propagation of postshock arrhythmic activations occurs when refractoriness of the fibrillating myocardium is below a critical level (22). Antiarrhythmic drugs may decrease DFT values by increasing the refractoriness of the fibrillating myocardium above this critical level. However, the present data indicate that the mechanisms by which drugs increase DFT values can not be explained by the opposite action, a reduction in fibrillation refractoriness. This may occur because refractoriness remains below the critical level and reducing refractoriness further has no impact on defibrillation outcomes. Data from the current study lead us to believe that the mechanism by which drugs increase DFT values are a result of their electrophysiologic actions post- Fig. 3. Tracing of ventricular fibrillation at three myocardial sites: left ventricular (LV) epicardium (upper panel), right ventricular (RV) endocardium (middle panel), and left ventricular endocardium (lower panel) in a single animal during baseline phase and after receiving pinacidil.

8 178 increases DFT values, also does not change the basal level of dispersion in APD, ERP, or VFCL (27,28), but may increase electrical heterogeneity postshock (29). Thus, we cannot rule out the possibility that this is a primary mechanism by which drugs increase DFT values. Fig. 4. Bar graph showing dispersion of action potential duration at 90% of repolarization (APD90) (upper panel), effective refractory period (ERP) (middle panel) and ventricular fibrillation cycle length (VFCL) (lower panel) at baseline, treatment phase one (DSW or Pinacidil), and treatment phase 2 (D5W + D5W or Pinacidil + Glyburide) for each group. *Pinacidil vs. Pinacidil + Glyburide shock. It has been suggested that increased electrical causes defibrillation to fail heterogeneity postshock and that higher energy levels are needed to limit postshock dispersion (23-25). In the present study, pinacidil did not change the basal level of dispersion in APD and ERP determined at rapid pacing rates (pinacidil has been shown to increase APD dispersion at slow cycle lengths >500 ms). Similarly, pinacidil did not affect the dispersion in VFCL (26). However, this does not imply that pinacidil has no affect on electrical heterogeneity postshock. Lidocaine, which Reversal With Glyburide Glyburide reversed pinacidil s effects on repolarization and refractoriness during pacing with the exception of left ventricular APD90. This indicates that pinacidil s electrophysiologic effects are the result of increased potassium channel conductance rather than sympathetic activation. Moreover, it is probable that endogenous sympathetic outflow is maximal during rapid ventricular pacing and ventricular fibrillation regardless of pinacidil because both scenarios produce hemodynamic collapse. The former parameter (left ventricular APD9o) was expected to increase given that the left ventricular effective refractory period was reversed by glyburide. It may be that the variability in this measurement was greater than the sensitivity to detect a change. Conversely, VFCL, RR interval, and mean arterial pressure were unaffected by glyburide. It is likely that shortening of VFCL by pinacidil was secondary to potassium conductance, but that because of ischemia during fibrillation, glyburide lost its ability to antagonize pinacidil s effects. This has been reported by others where repeated bouts of ischemia inhibited glyburide s ability to counteract pinacidil s effects on potassium conductance (30). These investigators suggest that more KATP channels are recruited by pinacidil during repeated ischemia, thus, negating glyburide s effect. It is also possible that increased adenosine diphosphate during repeated ischemic episodes decreases glyburide s affects on the KATP channel. Regardless, inducing ventricular fibrillation causes hemodynamic collapse thereby inducing repeated short bouts of ischemia, thus, the inability of glyburide to reverse VFCL may be explained with these findings. Finally, mean arterial pressure and heart rate did not reverse with the administration of glyburide. Since pinacidil s effect on vascular smooth muscle is more sensitive than its effects on cardiac electrophysiology, it is likely that a greater concentration of glyburide would be needed to reverse its effect on blood pressure (31 ). Study Limitations We used a swine model with a healthy cardiovascular system and a monophasic defibrillation system. It is

9 179 unknown if these findings can be directly extrapolated to other defibrillation systems or diseased (infarcted or myopathic) hearts. However, the results from animal studies of antiarrhythmic drugs and monophasic defibrillation have correlated with experience in humans (32-34). It has been shown that antiarrhythmic agents that increase DFT values in whole animal models increase DFT and/or cause ventricular fibrillation to be refractory in humans with cardiovascular disease (32-34). A recent meta-analysis has confirmed these observations, showings a very close correlation between animal models of defibrillation and clinical data observed in humans (35). Moreover, a retrospective analysis indicates that antiarrhythmic agents are a responsible factor in producing high DFT values in patients with implanted defibrillators (36). We did not assess the effect of glyburide alone on DFT. This is because the KATP channel contributes little to outward currents in comparison to the inward and delayed rectifying channels. Thus, KATP antagonists do not exert any measurable electrophysiologic effects that are consistent with other potassium channel antagonists. This has been suggested where intracoronary glyburide administration did not affect any electrophysiologic measurements of repolarization (37). Therefore, if the electrophysiologic milieu remains unchanged, we would not expect DFT outcomes to be altered, as recently described in abstract form (38). Clinical Implications Antiarrhythmic agents have been shown to affect DFT and are used concomitantly in 50-70% of patients with implantable defibrillators (39). If the mechanisms by which antiarrhythmic agents alter DFT can be elucidated, then clinicians will be able to predict how drug regimens, based on electrophysiologic properties, will affect defibrillation outcomes. Previous studies have implicated that a drug s effect on repolarization dictates its effect on DFT (3,6,12). However, our findings suggest that alterations in repolarization or changing cation conduction does not define drug s effect of DFT since shortening refractoriness by increasing potassium conductance with pinacidil did not alter DFT values. These findings imply that a drug s ability to cause electrical heterogeneity may be a more important factor to predict its adverse effect on DFT. Summary Agents that increase the refractory period (potassium channel antagonists) lower DFT values. This is the first study to address if the converse, shortening refractoriness via stimulation of potassium efflux, increases DFT values. The findings from this study illustrate that pinacidil, a KATP agonist, markedly shortens both the ERP and APD without causing a dispersion in refractoriness or altering defibrillation efficacy. These findings imply that shortening repolarization without a concomitant dispersion in refractoriness is not a mechanism by which antiarrhythmic agents increase DFT. References 1. Ujhelyi MR, O Rangers EA, Chow MSS, et al. Defibrillation energy requirements during moricizine and combination antiarrhythmic therapy. J Cardiovasc Pharmacol 20:932, Fain ES, Dorian P, Davy JM, et al. Effects of encainide and its metabolites on energy requirements for defibrillation. Circulation 7:397, Echt DS, Black JN, Barbey JT, et al. Evaluation of antiarrhythmic drugs on defibrillation energy requirements in dogs: sodium channel block and action potential prolongation. Circulation 79:1 106, Wang M, Dorian P. DL: D sotalol decrease defibrillation energy requirements. PACE 12:1522, Jones DL, Kim YH, Natale A, et al: Bretylium decreases and verapamil increases defibrillation threshold in pigs. PACE 17:1380, Babbs CF. Alteration of defibrillation threshold by antiarrhythmic drugs: a theoretical framework. Crit Care Med 9:362, Ujhelyi MR, Winecoff AP, Schur M, et al. Influence of hypertonic saline infusion on defibrillation efficacy. Chest 110:784, D Alonzo AJ, Hess TA, Darbenzio RB, et al. Effects of cromakalim or pinacidil on pacing-and ischemia-induced ventricular fibrillation in the anesthetized pig. Basic Res Cardiol 89:163, Babbs CF. Effect of pentobarbital anesthesia on ventricular defibrillation threshold in dogs. Am Heart J 95:331, Ujhelyi MR, Schur M, Frede T, et al. Differential effects of lidocaine on defibrillation threshold with monophasic versus biphasic shock waveforms. Circulation 92:1644, Zhou X, Guse P, Wolf PD, Rollins DL, Smith WM, Ideker RE. Existence of both fast and slow channel activity during the early stages of ventricular fibrillation. Circ Res 71:773, Kowey PR, Friehling TD, Sewter J, et al. Electrophysiological effects of left ventricular hypertrophy: effect of calcium and potassium channel blockade. Circulation 83: 2067, Szabo TS, Jones DL, McQuinn RL, et al. Flecainide acetate does not alter requirements for direct ventricular defibrillation using sequential pulse defibrillation in pigs. J Cardiovasc Pharmacol 12:377, Dorian P, Wang M, David I, et al. Oral clofilium produces sustained lowering of defibrillation energy requirements in a canine model. Circulation 83:614, 1991

10 Dorian P, Witkowski FX, Penkoske PA, et al. Barium decreases defibrillation energy requirements. J Cardiovasc Pharmacol 23:107, Ujhelyi MR, Schur M, Frede T, et al. Mechanisms of antiarrhythmic drug-induced changes in defibrillation threshold: role of potassium and sodium channel conductance. J Amer Coll Cardiol 27:1534, Schrader R, Brooks M, Echt DS. Effects of verapamil and bay K 8644 on defibrillation energy requirements in dogs. J Cardiovasc Pharmacol 19:839, Opthof T, Ramdat Misier AR., Coronel R, et al. Dispersion of refractoriness in canine ventricular myocardium: effects of sympathetic stimulation. Circ Res 68:1204, Dorian P, Newan D, Sheahan R, et al. d-sotalol decreases defibrillation energy requirements in humans: a novel indication for drug therapy. J Cardiovasc Electrophysiol 7:952, Dorian P, Penkoske PA, Witkowski FX. Order in disorder: effect of barium on ventricular fibrillation. Can J Cardiol 12:399, Ujhelyi MR, McSwain R, Carter S, Sims JJ, Oskouei S, Winecoff AP. Sodium channel blockade is not the primary mechanism by which lidocaine raises defibrillation threshold values. Pace 19:736 (abstract), Kwaku KF, Dillon SM. Shock-induced depolarization of refractory myocardium prevents wave-front propagation in defibrillation. Circ Res 79:957, Dillon SM. Synchronized repolarization after defibrillation shocks: a possible component of the defibrillation process demonstrated by optical recordings in rabbit heart. Circulation 85:1865, Tovar OH, Jones JL. Biphasic defibrillation waveforms reduce shock-induced response duration dispersion between low and high shock intensities. Circ Res 77:430, Fabritz CL, Kirchhof PF, Behrens S, et al. Myocardial venerability to Twave shocks: relation to shock strength. shock coupling interval, and dispersion of ventricular repolarization. J Cardiovasc Electrophysiol 7:231, Di Diego JM, Antzelevitch C. Pinacidil-induced electrical heterogeneity and extrasystolic activity in canine ventricular tissues: does activation of ATP-regulated potassium current promote phase 2 re-entry? Circulation 88:1177, Ujhelyi MR, McSwain R, Sims JJ, Li S, Winecoff AP. Lidocaine does not affect dispersion in myocardial repolarization or refractoriness: implications for low pro-arrhythmic actions. Pharmacotherapy 16:492 (abstract), Anderson KP, Walker R, Lux RL, et al. Conduction velocity depression and drug-induced ventricular tachyarrhythmias : effects of lidocaine in the intact canine heart. Circulation 81:1024, Ujhelyi MR, Winecoff AP, Oskouei S, Carter S, Markel ML. Dispersion in post-shock activations predicts defibrillation success during lidocaine. Circulation 94:101 (abstract), Yao Z, Cavero I, Gross G. Activation of cardiac KATP channels: an endogenous protective mechanism during repetitive ischemia. Am J Physiol 264:H495, Steinberg MI, Ertel P, Smallwood JK, et al. The relation between vascular relaxant and cardiac electrophysiologic effects of pinacidil. J Cardiovasc Pharmacol 12 (suppl 2):S30, Frame LH. The effect of chronic oral and acute intravenous amiodarone administration on ventricular defibrillation threshold using implanted electrodes in dogs. PACE 12:339, Tworek DA. Nazari J, Ezri M, et al. Interference by antiarrhythmic agents with function of electrical cardiac devices. Clin Pharm 11:48, Jung W, Manz M, Pfeiffer D, et al. Effects of antiarrhythmic drugs on epicardial defibrillation energy-requirements and the rate of defibrillator discharges. PACE 16:213, Kroll MW, Anderson KM, Supino CG, et al. Decline in defibrillation thresholds. PACE 16:213, Epstein AE, Ellenbogen KA, Kirk KA, et al. Clinical characteristics and outcome of patients with high defibrillation thresholds: a multicenter study. Circulation 86: 1206, D Alonzo AJ. Hess TA, Darbenzio RB, et al. Effects of intracoronary glyburide on cesium-induced arrhythmias in anesthetized dogs. J Cardiovasc Pharmacol 23:446, Harbinson M, Allen JD, Adgey AAJ. The effects of potassium-atp channel modulation on ventricular fibrillation and defibrillation. PACE 19:565 (abstract), Pinski SL, Vanerio G, Castle LW, et al. Patients with a high defibrillation threshold: clinical characteristics, management, and outcome. Am Heart J 122:89, 1991