Author + information
- Received July 1, 1996
- Revision received November 14, 1996
- Accepted November 26, 1996
- Published online March 1, 1997.
- ↵*Dr. Yuji Murakawa, The Second Department of Internal Medicine, University of Tokyo, 7-1-3 Hongo, Bunkyo-ku, Tokyo 113, Japan.
Objectives. We tested whether a new class III drug (MS-551) administered during ventricular fibrillation (VF) could decrease the defibrillation threshold (DFT) in anesthetized canine hearts.
Background. Pretreatment with class III antiarrhythmic agents is known to enhance electrical defibrillation efficacy.
Methods. In a preliminary study (n = 10), we ascertained the validity of DFT determination by a sequence of incremental defibrillation shocks in a single fibrillation/defibrillation episode. We then compared the DFTs after 130 s of VF with and without administration of MS-551 (2 mg/kg body weight) at 10 s after the onset of VF in 12 open chest dogs and 8 closed chest dogs.
Results. MS-551 decreased the DFT in both experimental models (open chest [mean ±SD]: from 416 ± 106 to 318 ± 92 V, p < 0.05; closed chest: from 714 ± 75 to 615 ± 112 V, p < 0.05). The change (Δ) in DFT in each heart was inversely correlated with the drug-induced prolongation of VF cycle length before the defibrillation attempt (ΔDFT vs. ΔVF cycle length 10 s before the first discharge: r = −0.58 and −0.81, p < 0.05).
Conclusions. MS-551 given after the induction of VF improved defibrillation efficacy. Class III antiarrhythmic agents deserve consideration when VF is resistant to electrical defibrillation during cardiopulmonary resuscitation.
(J Am Coll Cardiol 1997;29:688–92)
The recent advent of the automatic implantable cardioverter-defibrillator has raised interest in how antiarrhythmic agents influence electrical defibrillation efficacy . In general, sodium channel blockade decreases the defibrillation efficacy, whereas prolongation of the action potential duration increases it [2–12]. Specifically, long-term treatment with class III drugs, such as sotalol, is believed to facilitate defibrillation in patients with implanted cardioverter-defibrillators . However, it remains unknown whether these drugs can exert a favorable effect when administered during an episode of ventricular fibrillation (VF). In an attempt to extend the role of class III drugs in critical care medicine, we tested the effect on the defibrillation threshold (DFT) of a new nonspecific K+channel blocker, MS-551 [13, 14], when given after the induction of VF.
This study was carried out in four stages. The conventional DFT (conDFT) is determined by repeated fibrillation/defibrillation episodes that generally contain only one test shock plus one or more rescue shocks, if necessary . However, the effect of an acute intervention during VF needs to be estimated within a single episode. Therefore, we used a DFT defined by a sequence of incremental test shocks (seqDFT) in our main studies (studies 3 and 4). Study 1 was a preliminary study designed to validate this DFT by comparison with the conDFT and by inspecting its reproducibility. As an indication of the electrophysiologic action of MS-551, we analyzed the temporal profile of the VF cycle length (VF-CL) in studies 3 and 4. To confirm that the VF-CL reliably reflected drug-induced modulation of the action potential duration, we compared the changes of the monophasic action potential duration and VF-CL caused by MS-551 in study 2. Finally, effects of MS-551 given after the induction of VF on DFT were tested in open (study 3) and closed chest dogs (study 4).
1.1 Animal preparation
Thirty mongrel dogs weighing 7 to 15 kg were anesthetized with a bolus of intravenous sodium pentobarbital (30 mg/kg body weight), followed by continuous infusion at 1 to 5 mg/kg per h . Dogs were intubated with a cuffed endotracheal tube and ventilated with a mixture of room air and 100% oxygen through a Harvard respirator. Body temperature was maintained at about 37°C with an electric blanket. Arterial blood pressure was monitored using a left femoral artery cannula attached to a Statham pressure transducer (Amplifier AP621G, Nihon Kohden, Tokyo, Japan). Arterial blood gases and electrolytes were periodically determined by a portable clinical analyzer (i-STAT 200A, i-STAT Corporation), and abnormal pH, partial pressures of oxygen and carbon dioxide and K+values were corrected appropriately.
In 22 dogs used for studies 1 to 3, the heart was exposed by a midsternal incision, and a pair of square titanium defibrillation electrodes with a 4-cm2surface area were sutured to the pericardium over the right and left ventricles. In the remaining eight dogs for study 4, a pair of external defibrillation pads (79 cm2and 112.5 cm2: PD-2200, Zoll Medical Corporation) was applied on the anterior and posterior thoracic walls after skin treatment. Defibrillation shocks were provided by a programmable two-capacitor (165 μF × 2) discharge unit that could delivery truncated exponential pulses at preset voltages up to 800 V (DPS-1200D, Diamedical, Tokyo, Japan). When calibrated with a 200-ohm load at 500 V, a 6-ms monophasic shock had a 16% tilt. To ensure appropriate shock delivery, the current waveform was monitored on a digitized oscilloscope (model 5020A, Kikusui, Tokyo, Japan). Two bipolar epicardial electrograms (studies 1 to 3: 50 to 1 kHz) or one bipolar endocardial electrogram (study 4) as well as two surface electrocardiograms were continuously monitored, and they were recorded on an ink-jet recorder (RIJ-2108, Nihon Kohden) together with the arterial blood pressure at a paper speed of 25 or 50 mm/s.
1.2 Study 1 (n = 10)
Ventricular fibrillation was induced by a 2-s train of 4-ms pulses (100 Hz) through a unipolar electrode on the right ventricle. After 10 s of VF, a 6-ms monophasic sock at a strength of 300 V was applied first. Then an iterative increment-decrement protocol was used, with the steps being ∼10% of the preset voltage. We defined the conDFT as the voltage of the successful shock, when the next lower setting failed to defibrillate the heart in the preceding or subsequent attempt. Only the result of the first test shock of each defibrillation attempt was used in determination of the DFT. Fibrillation/defibrillation episodes were separated by at least 4 min and were repeated until triplicate DFT values were determined. The mean of the triplicate values was then taken to represent the conDFT after 10 s of VF (conDFT − 10 s).
To determine the seqDFT after 10 s of VF (seqDFT − 10 s), 6-ms shocks were consecutively delivered at intervals of 4 to 6 s. The first shock was set at 60% of conDFT − 10 s, and the voltage of subsequent shocks was increased stepwise by 20% of conDFT − 10 s (i.e., 80%, 100%, 120%, 140%, 160%, 180% of conDFT − 10 s). The seqDFT − 10 s was defined by the voltage of the successful shock. The seqDFT − 70 s and seqDFT − 130 s were determined in a similar way. However, to simulate clinical resuscitation and to alleviate hemodynamic deterioration due to prolonged VF, manual direct cardiac massage was started after 10 s of VF and was continued until the first discharge. The rate of cardiac compression was 2/s or slightly lower, and the femoral systolic blood pressure was maintained at ∼50 mm Hg. The values of seqDFT − 10 s, seqDFT − 70 s and seqDFT − 130 s were determined twice in succession. Fibrillation/defibrillation episodes were separated by at least 15 min, and the order of measurement of seqDFT was not randomized.
1.3 Study 2 (n = 5)
After completion of study 1, 5 of the 10 dogs were subjected to this study. The monophasic action potential was recorded from the anterior surface of the right ventricle via a suction electrode (interelectrode distance 3 mm) during right ventricular pacing at a cycle length of 350 ms. Monophasic action potential signals were amplified and filtered at a frequency of 0.05 and 1 kHz. The duration of monophasic action potential was determined at 90% repolarization (MAP90), after which VF was electrically induced. The VF-CL corresponding to the preceding MAP90 was calculated by averaging 4 to 10 VF-CL values from either the right or left ventricular bipolar electrogram after about 10 s of VF. After baseline measurement of the paired MAP90 and VF-CL, 0.1 mg/kg of MS-551 was intravenously administered as a bolus. MS-551 has been reported to block the delayed rectifier (Ik), the inward rectifier (Ik1) and the transient outward (Ito) K+channels . Fifteen minutes later, both variables were measured in the same manner as at baseline. Subsequently, MS-551 was cumulatively given at doses of 0.2, 0.4 and 0.8 mg/kg, with measurement of MAP90 and VF-CL being repeated at each dose.
1.4 Study 3 (n = 12)
Twelve dogs were subjected solely to this study. As a control, seqDFT − 130 s was obtained without MS-551 using the same method as in study 1. After at least 20 min, seqDFT − 130 s was reassessed after MS-551 (2 mg/kg) administration as a bolus through the femoral vein at 10 s after the onset of VF (Fig. 1). Blood samples for determination of the plasma drug concentration at 30, 60 and 120 s after MS-551 infusion were acquired through a long catheter introduced into the aortic arch. The local VF-CL was calculated every 10 s during fibrillation/defibrillation episodes with and without MS-551 using the same method as in study 2.
1.5 Study 4 (n = 8)
Eight closed chest dogs were subjected solely to this study. Because maximal voltage of our defibrillation unit was limited, smaller dogs (7 to 10 kg) were used. All the experimental procedures were as in study 3, except that VF was induced through the catheter electrode in the right ventricle, and transthoracic defibrillation shocks were applied to the heart.
1.6 Data analysis and statistical methods
Results are expressed as mean value ± SD. The significance of differences between paired values was tested by Student ttest. Comparison among three or more values was done by two-way analysis of variance. In the presence of a significant F value, further comparisons between each pair of variables were done by the Bonferroni method. Correlations between two variables were tested by linear regression analysis. Probability values <0.05 were considered to indicate significance.
2.1 Study 1
The conDFT − 10 s was 285 ± 107 V. The seqDFT − 10 s was 272 ± 109 V (p = NS vs. conDFT − 10 s) (Fig. 2), and there was a significant correlation between these two thresholds (r = 0.96, p < 0.001). The second measured value of each seqDFT was not significantly different from the initial measurement (seqDFT − 10 s: 4 ± 18%, seqDFT − 70 s: 3 ± 25%; seqDFT − 130 s: 16 ± 19%), and their correlation was always strong (r = 0.92, p < 0.001; r = 0.86, p < 0.005; r = 0.94, p < 0.001). The seqDFT − 130 s was higher than seqDFT − 10 s (p < 0.05) (Fig. 2).
2.2 Study 2
The baseline MAP90 was 209 ± 21 ms. It was uniformly prolonged as the cumulative dose of MS-551 increased (223 ± 25, 234 ± 30, 251 ± 33 and 256 ± 35 ms, p < 0.001). There was a dose-dependent increase in VF-CL from the baseline value of 124 ± 23 ms (148 ± 31, 166 ± 32, 176 ± 37 and 187 ± 43 ms, p < 0.001). Consequently, the linear relation between VF-CL and MAP90 was significant in all five dogs, as shown in Fig. 3. The individual correlation coefficients ranged from 0.92 (p < 0.05) to 0.99 (p < 0.005).
2.3 Study 3
The seqDFT − 130 s without MS-551 was 416 ± 106 V, and administration of MS-551 decreased this value in 8 of 12 dogs. Overall, the seqDFT − 130 s after MS-551 administration (318 ± 92 V) was significantly lower than before the drug was given (p < 0.05) (Fig. 4A).
In the absence of MS-551, the VF-CL was largely constant throughout VF (Fig. 5A). In contrast, MS-551 caused a time-dependent prolongation of VF-CL. The difference in the VF-CL values in the presence and absence of MS-551 became significant at 40 s after drug infusion. The drug-induced change in VF-CL at 120 s of VF (i.e., 10 s before first discharge) was inversely correlated with the changes in seqDFT − 130 s (r = −0.58, p < 0.05).
The plasma MS-551 concentration at 30, 60 and 120 s after drug infusion was 25 ± 27, 43 ± 34 and 24 ± 12 μg/ml, respectively. MS-551-induced changes of seqDFT − 130 s did not show any definite relation to the plasma drug level at any of these three times.
2.4 Study 4
The seqDFT − 130 s after MS-551 administration (615 ± 112 V) was lower than the baseline value (714 ± 75 V, p < 0.05) (Fig. 4B). The difference in the VF-CL values in the presence and absence of MS-551 became significant at 70 s after drug infusion. The drug-induced change in VF-CL at 120 s of VF was inversely correlated with the changes in seqDFT − 130 s (r = −0.81, p < 0.05).
The plasma MS-551 concentration at 30, 60 and 120 s after drug infusion was 41 ± 29, 21 ± 8 and 17 ± 8 μg/ml, respectively. MS-551-induced changes in seqDFT − 130 s did not show any significant relation to the plasma drug level.
The chief findings of the present study were as follows: 1) the DFT obtained with sequential multiple shocks showed excellent reproducibility and time-dependent increases; 2) defibrillation efficacy was significantly improved when MS-551 was given after the induction of VF in both open and closed chest dogs.
3.1 Effect of antiarrhythmic agents on DFT
With the exception of a few reports [17, 18], most studies have shown that antiarrhythmic agents with a class Ib or Ic action reduce defibrilllation efficacy in both animal and human hearts [2–9]. Most drugs with a class III action [2, 11, 12], but probably not amiodarone [19, 20], decrease the DFT. Accordingly, drugs with class Ia action have a small, if any, effect on the DFT, probably because their adverse and favorable effects cancel each other [2, 9, 10].
However, earlier experimental and clinical studies have focused on how pretreatment with antiarrhythmic drugs influenced defibrillation efficacy. To our knowledge, no systematic investigation has been performed concerning the effect of antiarrhythmic agents given during VF. To show prompt electrophysiologic action, drugs need to reach the myocardium and block certain ion channels without delay. In addition, the reverse-frequency dependence that is characteristic of some class III drugs diminishes their pharmacologic effect. Thus, it has been unclear whether class III drugs given during VF with its hemodynamic deterioration can favorably affect the defibrillation efficacy. However, the present study indicated that the defibrillation efficacy was enhanced when MS-551 was given after the onset of VF. This observation suggests the possible usefulness of class III antiarrhythmic drugs in patients who are refractory to electrical defibrillation.
3.2 DFT measurement with sequential shocks
We quantified the defibrillation efficacy using multiple incremental shocks. This procedure has the advantage that only a single fibrillation/defibrillation sequence can determine the DFT. However, its demerits are as follows: 1) Very few earlier data are available for comparison . 2) The time required for multiple shocks and the sequelae of subthreshold shocks are not taken into consideration. Time-dependent increase of DFT should be important to estimate the actual merit of pharmacologic modification of defibrillation efficacy. 3) The precision of the measurements is affected by the arrangement of shock strengths. Despite these potential problems, study 1 showed that seqDFT − 10 s was similar to conDFT − 10 s, and the small variability of seqDFT values between measurements was supportive of their excellent reproducibility. On the basis of these observations, we considered that seqDFT was an appropriate variable for the purposes of the present study.
3.3 Time course of electrophysiologic manifestations of MS-551.
Although the interval of local activation during VF varies from beat to beat, some studies have utilized VF-CL as an indicator of the refractoriness of regional myocardial tissue [23, 24]. Our study 2 revealed a distinct association between MAP90 and VF-CL. Although a quantitative relation of MAP90 and VF-CL may not be common, these results confirmed that VF-CL is a sensitive indicator of regional myocardial refractoriness.
VF-CL began to increase ∼40 s after injection of MS-551 (Fig. 5). This time-dependent prolongation of VF-CL presumably represented arrival of the drug in the heart and its pharmacologic action. Although the transport and diffusion of a drug are dependent on several factors, such as cardiac output and site of infusion, the electrophysiologic action of MS-551 seemed to be detectable soon after the drug reached the heart. In contrast, we failed to demonstrate a clear relation between the plasma drug level and changes in defibrillation efficacy. This may be explained by the fact that the plasma level did not necessarily correspond to the tissue level of MS-551 because of variable circulation during cardiopulmonary resuscitation or because its modulation of K+channels differed in individual hearts.
In the clinical setting, physicians prefer early defibrillation. Because we only assessed the influence of MS-551 at 120 s after injection, information about when the drug begins to improve defibrillation efficacy is lacking. On the basis of the inverse relation between the prolongation of VF-CL and changes in seqDFT − 130 s, MS-551 may well have influenced defibrillation efficacy from 40 to 60 s after infusion. However, the magnitude of its immediate effect (<1 min after infusion) would probably be too small to detect and may have limited clinical significance.
3.4 Study limitations
Extrapolation of the results obtained in intact hearts to diseased hearts is limited. Also, it is not clear whether external cardiac massage in the clinical setting can effectively transport the drug from peripheral vessels to the heart, as it did in our experimental models [25, 26]. Furthermore, we gave MS-551 only 10 s after the onset of VF, whereas several unsuccessful defibrillation attempts are usually made before initiation of interventions to facilitate defibrillation in the clinical setting. Therefore, metabolic abnormalities caused by prolonged VF before treatment with a class III drug might attenuate its effect.
When MS-551 was infused during VF, it successfully enhanced defibrillation efficacy in canine hearts. This finding suggests that class III antiarrhythmic agents deserve consideration when VF is resistant to electrical defibrillation or when an unacceptably high DFT is anticipated.
☆ This study was supported by the Fugaku Trust for Medical Research, Tokyo, Japan.
- defibrillation threshold determined by conventional method
- conDFT − 10s, seqDFT − 10s, seqDFT − 70s, seqDFT − 130s
- conDFT determined at 10-s VF and seqDFTs determined by sequential shocks applied at 10, 70 or 130 s of VF, respectively
- defibrillation threshold
- duration of monophasic action potential determined at 90% repolarization
- defibrillation threshold determined by sequence of incremental shocks
- ventricular fibrillation
- cycle length ventricular fibrillation
- Received July 1, 1996.
- Revision received November 14, 1996.
- Accepted November 26, 1996.
- The American College of Cardiology
- Naccarelli GV,
- Doughterty AH,
- Wolbrette D
- Echt DS,
- Black JN,
- Barbey JT,
- Coxe DR,
- Cato E
- Dorian P,
- Fain E,
- Davy JM,
- Winkle RA
- Fain ES,
- Dorian R,
- Davy JK,
- Kates RE,
- Winkle RA
- Kerber PE,
- Pandian NG,
- Jensen SR,
- et al.
- Frame LH,
- Sheldon JH
- Hernandez R,
- Mann DE,
- Breckinridge S,
- Williams GR,
- Reiter MJ
- Dorian P,
- Wang M,
- David I,
- Feindel C
- Sato R,
- Koumi S,
- Hisatome I,
- et al.
- Jones DL
- Epstein AE,
- Ellenbogen KA,
- Kirk KA,
- Kay GN,
- Dailey SM,
- Plumb VJ
- Colatsky TJ,
- Follmer CH,
- Starmer CF
- Opthof T,
- Misier R,
- Coronel R,
- et al.
- Misier ARR,
- Opthof T,
- van Hemel NM,
- et al.