1146

Effects of Lidocaine on Relation BetweenDefibrillation Threshold and Upper Limit of

Vulnerability in Open-Chest DogsSteven L. Topham, BS; Yong-Mei Cha, MD; Barry B. Peters, MD; and Peng-Sheng Chen, MD

Background. The purpose of the present study was to test the effects of lidocaine on the relation betweenthe defibrillation threshold and the upper limit of vulnerability.

Methods and Results. The shock strength associated with a 50% probability of successful defibrillation(DFT50) and the shock strength associated with a 50%o probability of reaching the upper limit ofvulnerability (ULV,50) were determined in 11 open-chest dogs by using the delayed up-down method beforeand during lidocaine (seven dogs) or normal saline (four dogs) infusion. The ventricles were paced at acycle length of 300 msec. Shocks of various strengths were then given via a patch-patch electrodeconfiguration on the anterior and posterior surfaces of the ventricle to determine the ULV50. Onceventricular fibrillation was induced, shocks were given 15-20 seconds later via the same electrodeconfiguration to determine the DYT50. Lidocaine infusion resulted in a serum level of 15±4 jg/ml. Thiswas associated with a lengthening of the QT interval but not with the widening of the QRS complex. In alldogs, both the ULV50 and the DFY50 increased significantly when tested during lidocaine infusion. MeanULV50 during lidocaine infusion was 496±+70 V or 13.1+4.3 J, which were significantly higher than thebaseline values of 333+±67 V or 5.3+2.2 J (p<0.001 for both voltage and energy). Mean DFY50 duringlidocaine infusion was 407±41 V or 8.7+ 1.7 J, which were significantly higher than the baseline values of300+38 V and 4.4±1.1 J (p=0.004 for voltage and p=0.013 for energy). The r values between the ULV5Oand the DFYT50 were 0.79 (p=0.037) for voltage and 0.80 (p=0.030) for energy at baseline and 0.85(p=0.016) for voltage and 0.88 (p=0.009) for energy during the lidocaine infusion. However, theincrements of the ULV50 (163 ±88 V or 7.8+±4.6 J) were significantly greater than the increments of theDFE50 (107±51 V or 4.4± 1.9 J, p=0.035 for voltage and p =0.023 for energy). Normal saline infusion didnot alter DYT50 or ULV50.

Conclusions. Lidocaine infusion significantly increases both ULV50 and DFT50. These results arecompatible with the upper limit of vulnerability hypothesis of defibrillation. However, the greater increaseof the upper limit of vulnerability than the defibrillation threshold with lidocaine infusion indicates thatother factors may also need to be considered to explain the results. (Circulation 1992;85:1146-1151)KEY WoRDs * electrophysiology * defibrillation . cardioversion

T he upper limit of vulnerability hypothesis ofdefibrillation1-4 states that unsuccessful shocksthat are slightly weaker than is necessary for

defibrillation halt all activation fronts during ventricularfibrillation (VF) but stimulate regions of the myocar-dium during their vulnerable period, giving rise to newactivation fronts that reinitiate VF. To achieve success-

All editorial decisions for this article, including selection ofreviewers and the final decision, were made by a guest editor. Thisprocedure applies to all manuscripts with authors from the Uni-versity of California San Diego or UCSD Medical Center.From the Basic Arrhythmia Laboratory, Division of Cardiology,

Department of Medicine, University of California San DiegoMedical Center, and Veterans Affairs Medical Center, San Diego,Calif.

Supported in part by a Merit Review from the Department ofVeterans Affairs and a grant from the Whitaker Foundation(P.-S.C.). This work was done during the tenure of a Clinician-Scientist Award from the American Heart Association to P-S.C.Address for reprints: Peng-Sheng Chen, MD, 8411, Division of

Cardiology, Department of Medicine, UCSD Medical Center, 225Dickinson Street, San Diego, CA 92103-8411.

Received May 10, 1991; revision accepted November 5, 1991.

ful defibrillation, a shock must reach the upper limit ofvulnerability so that VF cannot be reinitiated. Thishypothesis is supported by the significant correlationbetween the defibrillation threshold and the upper limitof vulnerability.2,56 One study2 tested the correlationbetween the upper limit of vulnerability and the de-fibrillation threshold when two different defibrillationpatch electrode configurations were used. The resultsshowed that when the defibrillation threshold increased,so did the upper limit of vulnerability. Significantcorrelations between the two values were found for bothdefibrillation electrode configurations. However, be-cause electrical induction and termination of VF de-pend on not only the shock field strength distribution78but also the electrophysiological state of the myocardi-um,S13,4,9 it is important to test this correlation when theelectrophysiological state is perturbed but the fieldstrength remains constant. Lidocaine was recentlyshown to significantly increase the defibrillation thresh-old10"l and is therefore an ideal agent with which toperturb the electrophysiological state of the myocar-dium and to test the correlation between the upper limit

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

Topham et al Defibrillation and Vulnerability 1147

of vulnerability and the defibrillation threshold. Thepurpose of the present study was to compare the upperlimit of vulnerability and the defibrillation thresholdbefore and during lidocaine administration withoutchanging the location of the defibrillation patch elec-trodes. The upper limit of vulnerability hypothesis ofdefibrillation will be supported if lidocaine infusionsignificantly increases both the upper limit of vulnera-bility and the defibrillation threshold, and there is asignificant correlation between the two values both atbaseline and during the administration of lidocaine.However, if the results show that lidocaine alters theupper limit of vulnerability and defibrillation thresholdto a different degree, then other mechanisms have to beconsidered to explain these findings.

MethodsSurgical Preparation

Adult mongrel dogs were anesthetized with 25-35mg/kg sodium pentobarbital,12,13 intubated, and venti-lated with room air with a Harvard respirator (HarvardApparatus, Millis, Mass.). An arterial line was insertedinto the femoral artery to monitor blood pressurecontinuously. Blood was periodically drawn to deter-mine pH, Po2, Pco2, base excess, and bicarbonateconcentrations. Esophageal temperature was monitoredand maintained at 36-37°C by heating the table withwarm circulating water. The surface ECG leads I, II, III,aVR, aVL, aVF, and V6 were recorded simultaneouslywith a computerized mapping system (BARD electro-physiology, Tewksbury, Mass.) and displayed on a multi-channel oscilloscope throughout the study. The chestwas opened through a medium sternotomy, and theheart was suspended in a pericardial cradle. Two patchdefibrillation electrodes with an active surface area of13.5 cm2 (CPI, St. Paul, Minn.) were sutured to theanterior and posterior surfaces of the ventricles. Aplatinum pacing electrode was attached to the epicar-dium of the left ventricular apex for unipolar cathodalS, stimulation, with the anode on the chest wall. Thesame pacing site was used for all animals studied. Theepicardium was kept moist with normal saline. After thestudy protocol (see below) was completed, the dogswere killed with an overdose of pentobarbital, and theheart was excised and weighed.An SMP-300 multichannel stimulator (Biologic In-

struments de Laboratories, Echirolles, France) wasused to drive constant-current stimulation isolationunits (Bloom, Reading, Pa.) to give 5-msec stimuli attwice cathodal diastolic threshold as the S. Anotherchannel of the SMP-300 stimulator was used to delivera premature stimulus (S2) at predetermined S1S2 cou-pling intervals. The S2 was used as an external signal totrigger the delivery of high-voltage truncated exponen-tial electric shocks with variable tilt via an HVS-02defibrillator (Ventritex, Sunnyvale, Calif.) to the epicar-dial patch electrodes for the induction of VF. Theleading edge voltage, delivered energy, and resistanceof the shock were displayed on the defibrillator imme-diately after each shock was delivered. The currentoutput was not measured. Once VF was induced, thedefibrillator was switched to the asynchronous mode toattempt defibrillation in 15-20 seconds. The same de-

Determination of ULV5o and DFT50Delayed up-down algorithm. The delayed up-down

algorithmA4 was used to determine the shock strengthassociated with a 50% probability of reaching the upperlimit of vulnerability (ULVso) and the shock strengthassociated with a 50% probability of successful defibril-lation (DFT50). The up-down algorithm began by givingthe first shock at a strength estimated to yield a 50%successful defibrillation. If this initial shock was unsuc-cessful, the shock strength was increased by a certain avalue for the next shock. If a shock was successful, theshock strength was decreased by the same S value forthe next shock. This process was continued until fourshocks were delivered and one shock strength wasprojected but not delivered. The five shock strengthswere then averaged for the threshold estimate.

This up-down algorithm is accurate only when the apriori estimates are good. The accuracy is significantlyreduced for poor a priori estimates of the DFT5o. Thisproblem can be overcome by the delayed four-episode,up-down algorithm. This algorithm does not start count-ing the four required observations until the first reversalin response. With this procedure, the accuracy of deter-mining the DFT50 is greatly improved because it doesnot depend on the accuracy of an a priori estimate ofthe initial shock strength.VF induction and ULV50 determination. The ULV50

was determined by the delayed up-down algorithm. Theventricles were paced at a 300-msec cycle length for 11beats. The intervals between the last S, pacing artifactand the onset, peak, and end of the T waves on ECGlead II were determined by the mapping system. Toinduce VF and test the ULV5,, a shock with a strengthof 300 V was given at the onset of the T wave. The S1S2intervals were then progressively increased at 10-msecincrements until a total of 15 shocks were given or VFwas induced. With this method, the entire T wave wasscanned. All shocks that did not induce VF wereseparated by 1 minute. Depending on whether VF wasinduced, subsequent shocks were given with a shockstrength of 50 V higher or lower, respectively, than theprevious shock energy until the opposite result wasobserved. The shock strength immediately precedingthe opposite result was the first data point. The shockstrength associated with the opposite result was thesecond data point. The up-down algorithm was thencontinued until a total of four data points were obtainedand one was predicted.14 The ULV50 was determined byaveraging the five shock strengths. The shocks givenbefore the first data point were excluded from analysis.VF termination and DFT50 determination. VF was

induced as part of the upper limit of vulnerabilitytesting described above. If the determination of DFTsohad not been completed at the end of the ULV50 testing,additional episodes of fibrillation-defibrillation wereperformed. The VF during these episodes was inducedby giving a 100-V shock during the vulnerable period.Once VF was induced, the DFTso was determined with6-msec monophasic shocks using the delayed up-downalgorithm. The first defibrillation shock was 300 V foreach dog studied. In subsequent episodes, the shockenergy was increased by 50 V after failures or decreasedby 50 V after successes until the opposite results wereobserved. The shock strength immediately precedingfibrillation waveform was used throughout the study.

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

1148 Circulation Vol 85, No 3 March 1992

the opposite result was the first data point. The shockstrength associated with the opposite result was the sec-ond data point. The up-down algorithm was then contin-ued until a total of four data points were obtained and onewas predicted.14 The DFT50 was determined by averagingthe five shock strengths. The shocks given before the firstdata point were excluded from analysis. All fibrillation-defibrillation episodes were separated by at least 5 min-utes. Salvage shocks were given immediately after anunsuccessful defibrillation shock. The shock strength ofthe salvage shock was not included in the data analysis.

Protocol 1: ULV50 and DFT50 Before and DuringLidocaine Infusion

Seven dogs were used in this protocol. After baselinetesting of the ULV50 and the DFT5O, a high dose oflidocainell was administered (9.2-mg/kg load over 10minutes followed by 285 ,Lg/kg/min maintenance). Thisdosage of lidocaine was shown to result in a stableplasma concentration during the measurement (pseu-do-steady state) and to increase the DFT50. The ULV50and the DFT50 were redetermined during the mainte-nance infusion 1 hour after the loading dose. Thelidocaine concentration was sampled at the beginningand end of testing.

Protocol 2: Control StudiesFour dogs were used in this protocol. After baseline

testing of the ULV5o and the DFT50, normal saline wasadministered (10-ml load over 10 minutes followed by 1ml/min maintenance). The ULV50 and the DFT50 wereredetermined during stable maintenance infusion 1 hourlater. The DFT50 was then determined for the third timewhen the VF was induced with rapid ventricular pacing.

Data AnalysisThe QRS widths and the QT intervals of ECG lead II

were measured during sinus rhythm. The QT, wascalculated by dividing the QT interval by the squareroot of the preceding RR interval (seconds). The inter-vals between the stimulus and the onset, peak, and endof the T wave of the last S, beat were also measured.These data were compared with the SIS2 interval todetermine where on the T wave a shock was given.

All statistical analyses were performed with SYSTAT.15The Pearson correlation coefficient analysis, linear re-gression analysis, and Newman-Keuls test were used tocompare the ULV50 and the DFT50 before and duringlidocaine infusion. The t tests and Pearson correlationcoefficient analyses were also used to compare the QRSwidths, QT, intervals, and increments of the ULV50 andthe DFT50 before and during lidocaine infusion. A valueofpc0.05 was considered significant.

ResultsProtocol 1The mean+SD body weight of the seven dogs studied

was 21+3 kg, and the mean±SD heart weight was162±25 g. The serum lidocaine concentration was18±13 ,gg/ml 1 hour after the loading dose and 15±4,ug/ml at the end of the study (p=NS). The duration ofthe experimental protocol (from the time of the firstshock to the end of the study) averaged 275+±22 min-

before (67±+15 msec) and during (73±16 msec)lidocaine infusion. However, the QTc intervals in-creased from 335±59 msec at baseline to 397±32 msecduring the lidocaine infusion (p=0.045). Lidocaineinfusion was also associated with a significantly de-creased heart rate. The RR intervals at baseline(420±14 msec) were significantly less than those duringlidocaine infusion (547±103 msec).

ULV50 and DFl50 before and during lidocaine infusion.In all dogs, both the ULV50 and the DFT50 increasedsignificantly when tested during lidocaine infusion. Themean ULV50 during lidocaine infusion was 496±70 V or13.1±4.3 J, which were significantly higher than thebaseline values of 333±67 V or 5.3 ±2.2 J (p<0.001 forboth voltage and energy). Mean DFT50 during lidocaineinfusion was 407±41 V or 8.7±1.7 J, which were signif-icantly higher than the baseline values of 300±38 V and4.4±1.1 J (p=0.004 for voltage and p=0.013 for ener-gy). The correlation between the ULV50 and the DFT50was significant both before and during the lidocaineinfusion (Figure 1).The increments of the ULV50 and the DFT50 were

highly correlated (Figure 2), indicating that dogs with agreater increase of the ULV50 also had a greater in-crease of the DFT50. However, the increments of theULV50 (163+88 V or 7.8±4.6 J) were significantlygreater than the increments of the DFT50 (107±51 V or4.4±1.9 J,p=0.035 for voltage andp=0.023 for energy).As a result, although there was no difference betweenthe ULV50 and the DFT50 at baseline, the ULV50 becamesignificantly greater than the DFT5o during lidocaineinfusion (p =0.007 for voltage andp =0.004 for energy).

Timing of shock and induction of VF. Lidocaine infu-sion significantly increased the intervals from the last S,stimulus to the beginning, peak, and end of the T waveson ECG lead II. The intervals from the last S, to thebeginning, peak, and end of the T waves at baselinewere 135±8, 244+24, and 291+20 msec respectively.These intervals were significantly shorter than the197±+-64 (p=O0.048), 310± 84 (p=0.050), and 383±+ 85msec (p=0.019) observed during lidocaine infusion.

For shocks that induced VF during the upper limitof vulnerability testing, the interval from the last S, tothe time of the shock was 191±26 msec (n=21) atbaseline and 225±80 msec (n=26) during lidocaineinfusion (p=0.05). In all except one of the 47 epi-sodes, VF was induced when the shock was given onthe upslope of the T wave.

Protocol 2The body weight of the four dogs in the study

averaged 17.9±5.4 kg, and the heart weight averaged157+36 g. There was no difference between the QRSwidth before (51±17 msec) and during (46±10 msec)normal saline infusion. The QT, intervals before(326±58 msec) and during (337±51 msec) normal sa-line infusion were also not statistically different. Theinfusion of normal saline did not change the heart rate.The RR intervals at baseline (431±86 msec) were notsignificantly different from those recorded during nor-mal saline infusion (443±10 msec).The ULV5o determined at the baseline (350±127 V or

6.1±4.7 J) was not significantly different from thatutes. There was no difference between the QRS width recorded during normal saline infusion (332-t-167 V

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

Topham et al Defibrillation and Vulnerability

,00

DFT, = 152 + 0.45 x ULVs,6 R=0. 79. P=0. 037

3000 0

0

200200 300 400 500 600 700

ULV50 BEFORE LIDOCAINE (V)

S:

z4

00

I-4t0c

600

500 .

DFT,0 = 164 * 0.49 x ULYV

R=0.85. P=-0.016

4. .

300

200200 300 400 500 600 700

ULY50 AFTER LIDOCAINE M

D

2

DFTs,= 2.26 + 0.39 x ULY,

L a=0.o0.8 =0.030z4U

oW

4

0

22

17

12

7 12 17 22

OFT,= 4.01 * 0.36 x ULV,Y

R=0.88, P0.009

FIGURE 1. Scatterplots of correlation be-tween a 50% probability of successful de-fibrillation (DFT50) and a 50% probabilityof reaching the upper limit of vulnerability(ULV50) before and during lidocaine infu-sion. Panels A and B: Correlation betweenDFT50 and ULV50 before and duringlidocaine infusion, respectively, measured involts. Panels C and D: Same correlationsmeasured in joules. Although values ofULV50 and DFT50 both increased signifi-cantly with lidocaine infusion, correlationbetween DFT50 and ULV50 remains signifi-cant. o, Data obtained at baseline; *, dataobtained during lidocaine infusion.

7 12 17 22

ULY50 BEFORE LIDOCAINE (3)

or 6.3 + 6.7 J). The DFT50 determined at baseline(312+ 126 V or 5.3 ±4.4 J), during normal saline infusion(305±121 V or 5.1±4.4 J), and with VF induced byrapid ventricular pacing (305+117 V or 5.1±4.4 J) werenot significantly different.

DiscussionEffects of Lidocaine on Correlation Between UpperLimit of Vulnerability and Defibrillation ThresholdThe presence of an upper limit of vulnerability has

been demonstrated since the discovery of a vulnerableperiod of the cardiac cycle.16 By definition, the upperlimit of vulnerability is the stimulus strength above

ULVSO AFTER LIDOCAINE (J)

which VF cannot be induced even if the stimulus weregiven during the vulnerable period of the cardiac cycle.A similar phenomenon has also been observed in theatrium.17 Subsequently, investigators demonstrated thatthe upper limit of vulnerability correlated well with thedefibrillation threshold.25'6 By changing the location ofthe anodal defibrillation patch electrode from the rightto the left atrium, with the same cathodal electrode onthe left ventricular apex, alterations of the electricalfield distribution resulted in the increase of not only thedefibrillation threshold but also the upper limit ofvulnerability as well.2 The correlation between the twovalues was significant for either electrode combination.

B17.0 r

ADFT50= 29 + 0.48 x AULVY,

13.6 1

aDFT,0= 1.49 * 0.37 x ULVY,

R=0.89, P=0.007

10.2 1I-

Ln.-0

6.8

3.4

0.00.0 3.4 6.8 10.2 13.6 17.0

ALYV50 (V)

FIGURE 2. Scatterplots of corre-lation between increments of a50% probability of successful de-fibrillation (DFT50) and a 50%probability of reaching the upperlimit of vulnerability (ULV50) ofeach dog during lidocaine infusion.Panel A: Correlation measured involts. Panel B: Correlation mea-sured in joules. Slopes of regressionline indicate that increments ofULV50 are greater than incrementsof DFT50. ADFT50 is increment ofDFT50 from baseline; AULV50 isincrement of UL V50 from baseline.

A700

B

500S

S3W

z-cU0

:3

1

CX0

0

,400

1149

C22

17z

0a

I-so

MO.

0

12 F

2

A350

300 k R=0. 82, p=0. 024

250

I-

LA- 350

100so

50

50

0 50 100 150 200 250 300 350

oh

a

0

i

0

dh

7

* 0

AUSV0 (J)

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

1150 Circulation Vol 85, No 3 March 1992

Furthermore, more recent studies'8 showed that like thedefibrillation threshold,19 the upper limit of vulnerabil-ity is a probability function, and the probability curvesof these two tests in the same animal parallel eachother. In the present study, we demonstrated that theupper limit of vulnerability and the defibrillationthreshold can both be increased by the administration oflidocaine without changing the location of the defibril-lation patch electrodes. The two values were highlycorrelated, both at baseline and during the lidocaineinfusion. The increments of these two values were alsohighly correlated. In contrast, normal saline infusion didnot alter either the ULV50 or the DFT50. These datashowed that with the perturbation of either the fieldstrength distribution2 or the electrophysiological state ofthe myocardium, the defibrillation threshold and theupper limit of vulnerability remain closely correlated.Although this close correlation supports the upper limitof vulnerability hypothesis of defibrillation,1-4 a greaterincrease of the upper limit of vulnerability than of thedefibrillation threshold with the administration oflidocaine indicates that other factors may also need tobe considered to explain these results.

It is well known that the sodium channel-blockingeffect of lidocaine is dependent on the heart rate (usedependence) and the transmembrane potential.20-23Ventricular fibrillation is associated with a cycle lengthof approximately 100 msec1 and transmembrane actionpotentials of extremely short duration and low ampli-tude.24,25 Although the short cycle length during VF mayfacilitate sodium channel block,20-23 the brief durationof the action potential is unfavorable for the blockingaction of lidocaine.2627 Because of the differences incycle lengths and the transmembrane action potentialsbetween the paced rhythm and VF, the sodium chan-nel-blocking effects of lidocaine could also differ. Thisdifference may account for the discrepancies betweenthe increments of the upper limit of vulnerability andthe defibrillation threshold during lidocaine infusion.The second possibility is the different magnitude of

action potential and refractory period extension pro-duced by shocks during VF and during paced rhythm.The upper limit of vulnerability hypothesis proposedthat after an unsuccessful defibrillation shock, newactivation wave fronts arise as a result of a complexinteraction between the shock's electric field and tissuerefractoriness.4 Two results of such an interaction arethe time- and energy-dependent action potential28'29and refractory period30-32 extension. The magnitude ofthe action potential and the refractory period extensionmay be important in determining vulnerability anddefibrillation. Because of the faster excitation rate dur-ing VF, the action potential was much shorter duringVF than it was during paced rhythm. It is highlyprobable that the extents of the action potential and therefractory period extension after a shock during VFdiffer significantly from that found during pacedrhythm, even though the strength of the shock is thesame. The effects of such a difference may be aggra-vated by lidocaine and account for the differencesbetween the ULV50 and the DFT50 during lidocaineinfusion.The third possible explanation is that the upper limit

of vulnerability hypothesis is incorrect. Shibata et a19

by shocks delivered during the vulnerable period wasusually associated with large reentrant circuits. How-ever, the new activation fronts observed at the earlysites after an unsuccessful shock often arise from a focal(or perhaps microreentrant) pattern.3 Thus, differentmechanisms may be responsible for vulnerability anddefibrillation. Lidocaine infusion perturbed these twomechanisms to different degrees and therefore results ingreater increase of the upper limit of vulnerability thanthe defibrillation threshold.

Upper Limit of Vulnerability and Midupslope ofT WaveWe33 previously demonstrated that the upper limit of

vulnerability determined at the midupslope of the Twave can be used to accurately predict the defibrillationthreshold. In contrast, the upper limit of vulnerabilitydetermined at the peak and the middownslope of the Twave had a poorer correlation with, and was signifi-cantly lower than, the defibrillation threshold. We con-cluded that the defibrillation threshold could be esti-mated by shocks given only at the midupslope of the Twave. In the present study, we scanned the T wavestarting with short S1S2 intervals, and then progressivelyincreased the SIS2 coupling interval if the previousshock failed to induce VF. We found that both beforeand during lidocaine infusion, the shock that inducedVF also occurred at the upslope of the T wave. Furtherstudies will be needed to determine whether the mid-upslope is also the best time to test the upper limit ofvulnerability during lidocaine infusion.

Accuracy of DFT50 DeterminationThere is no consensus on which method best estimates

the defibrillation efficacy. Two methods have been usedthe most. The first is the "defibrillation threshold" meth-od,34 and the second is the "dose-response curve"method.'9 Jones et a135 recently compared these twomethods using animal experiments. To determine thedefibrillation threshold, they used an iterative increment-decrement protocol similar to the delayed up-downmethods14 used in the present study. The results werecompared with the DFT50 determined by the dose-response curve method. The results showed that therewas no statistical difference between the DIFI50 as esti-mated by either method. These experimental results arecompatible with the results obtained by mathematicalmodeling study14 and showed that the DFT50 can beaccurately estimated with an up-down algorithm withoutconstructing the entire dose-response curve of defibril-lation. Based on the results of these studies,'4'35 webelieve that the methods used in the present studyproduce accurate estimates of the DFT50.

AcknowledgmentsThe authors wish to thank Mr. Richard Pavelec and Ms.

Amy Bloom for their technical assistance and CPI, St. Paul,Minn., for donating the HVS-02 external defibrillator.

References1. Chen P-S, Shibata N, Wolf P, Dixon EG, Danieley ND, Sweeney

MB, Smith WM, Ideker RE: Activation during successful andunsuccessful ventricular defibrillation in open chest dogs: Evidenceof complete cessation and regeneration of ventricular fibrillationdemonstrated that with a few exceptions, VF initiated after unsuccessful shocks. J Clin Invest 1986;77:810-823

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

Topham et al Defibrillation and Vulnerability 1151

2. Chen P-S, Shibata N, Dixon EG, Martin RO, Ideker RE: Com-parison of defibrillation threshold and the upper limit of ventric-ular vulnerability. Circulation 1986;73:1022-1028

3. Chen P-S, Wolf PD, Melnick SD, Danieley ND, Smith WM, IdekerRE: Comparison of activation during ventricular fibrillation andfollowing unsuccessful defibrillation shocks in open chest dogs.Circ Res 1990;66:1544-1560

4. Chen P-S, Wolf PD, Ideker RE: The mechanism of cardiacdefibrillation: A different point of view. Circulation 1991;84:913-919

5. Fabiato PA, Coumel P, Gourgon R, Saumont R: Le seuil der6ponse synchrone des fibres myocardiques: Application a lacomparaison experimentale de l'efficacit6 des diff6rentes formesde chocs 6lectriques de defibrillation. Arch Mal Coeur 1967;60:527-544

6. Lesigne C, Levy B, Saumont R, Birkui P, Bardou A, Rubin B: Anenergy-time analysis of ventricular fibrillation and defibrillationthresholds with internal electrodes. Med Biol Eng 1976;14:617-622

7. Chen P-S, Wolf DP, Claydon FJ, Dixon EG, Vidaillet H Jr,Danieley ND, Pilkington TC, Ideker RE: The potential gradientfield created by epicardial defibrillation electrodes in dogs. Circu-lation 1986;74:626-636

8. Witkowski FX, Penkoske PA, Plonsey R: Mechanism of cardiacdefibrillation in open-chest dogs using unipolar DC-coupled simul-taneous activation and shock potential recordings. Circulation1990;82:244-260

9. Shibata N, Chen P-S, Dixon EG, Wolf PD, Danieley ND, SmithWM, Ideker RE: Influence of epicardial shock strength and timingon the induction of ventricular arrhythmias in dogs. Am J Physiol1988;255:H891-H901

10. Dorian P, Fain ES, Davy J-M, Winkle RA: Lidocaine causes areversible, concentration-dependent increase in defibrillationenergy requirements. JAm Coll Cardiol 1986;8:327-332

11. Echt DS, Black JN, Barbey JT, Coxe DR, Cato E: Evaluation ofantiarrhythmic drugs on defibrillation energy requirements indogs: Sodium channel block and action potential prolongation.Circulation 1989;79:1106-1117

12. Babbs CF: Effect of pentobarbital anesthesia on ventriculardefibrillation threshold in dogs. Am Heart J 1978;95:331-337

13. Amlie JP, Owren T: The effect of prolonged pentobarbital anaes-thesia on cardiac electrophysiology and inotropy of the dog heartin situ. Acta Pharmacol Toxicol 1979;44:264-271

14. McDaniel WC, Schuder JC: An up-down algorithm for estimationof cardiac ventricular defibrillation threshold. Med Instru 1988;22:286-292

15. Wilkinson L: SYSTAT: The System for Statistics. Evanston, Ill,SYSTAT, Inc., 1989

16. King BG: The effect of electric shock on heart action with specialreference to varying susceptibility in different parts of the cardiaccycle (dissertation). New York, Columbia University, 1934

17. Orias 0, Gilbert J, Siebens AA, Suckling EE, Brooks CM:Effectiveness of single rectangular electrical pulses of knownduration and strength in evoking auricular fibrillation.Am JPhysiol1950;162:219-225

18. Kavanagh KM, Harrison JH, Dixon EG, Guse P, Smith WM,Wharton J, Ideker RE: Correlation of the probability of successcurves for defibrillation and for the upper limit of vulnerability(abstract). PACE 1990;13:536

19. Davy JM, Fain ES, Dorian P, Winkle RA: The relationshipbetween successful defibrillation and delivered energy in open-chest dogs: Reappraisal of the "defibrillation threshold" concept.Am Heart J 1987;113:77-84

20. Grant AO, Starmer CF, Strauss HC: Antiarrhythmic drug action:Blockade of the inward sodium current. Circ Res 1984;55:427-439

21. Davis J, Matsubara T, Scheinman MM, Katzung B, HondeghemLH: Use-dependent effects of lidocaine on conduction in caninemyocardium: Application of the modulated receptor hypothesis invivo. Circulation 1986;74:205-214

22. Nattel S: Interval-dependent effects of lidocaine on conduction incanine cardiac Purkinje fibers: Experimental observations andtheoretical analysis. J Pharmacol Exp Ther 1987;241:275-281

23. Gilliam FR III, Starmer CF, Grant AO: Blockade of rabbit atrialsodium channels by lidocaine: Characterization of continuous andfrequency-dependent blocking. Circ Res 1989;65:723-739

24. West TC, Frederickson EL, Amory DW: Single fiber recording ofthe ventricular response to induced hypothermia in the anesthe-tized dog: Correlation with multicellular parameters. Circ Res1959;7:880-888

25. Downar E, Janse MJ, Durrer D: The effect of acute coronaryartery occlusion on subepicardial transmembrane potentials in theintact porcine heart. Circulation 1977;56:217-224

26. Hondeghem LM, Katzung BG: Antiarrhythmic agents: The mod-ulated receptor mechanism of action of sodium and calciumchannel-blocking drugs. Ann Rev Pharmacol Toxicol 1984;24:387-423

27. Kojima M, Ban T: Nicorandil shortens action potential durationand antagonists the reduction of Vm. by lidocaine but not disopy-ramide in guinea pig papillary muscles. Naunyn SchmiedebergsArch Pharmacol 1988;337:203-212

28. Zhou X, Knisley SB, Wolf PD, Rollins DL, Smith WM, Ideker RE:Prolongation of repolarization time by electric field stimulationwith monophasic and biphasic shocks in open-chest dogs. Circ Res1991;68:1761-1767

29. Dillon SM: Optical recordings in the rabbit heart show thatdefibrillation strength shocks prolong the duration of depolariza-tion and the refractory period. Circ Res 1991;69:842-856

30. Sweeney RJ, Gill RM, Steinberg MI, Reid PR: Ventricular refrac-tory period extension caused by defibrillation shocks. Circulation1990;82:965-972

31. Sweeney RJ, Gill RM, Reid PR: Characterization of refractoryperiod extension by transcardiac shock. Circulation 1991;83:2057-2066

32. Swartz JF, Jones JL, Jones RE, Fletcher R: Conditioning prepulseof biphasic defibrillator waveforms enhances refractoriness tofibrillation wavefronts. Circ Res 1991;68:438-449

33. Chen P-S, Feld GK, Mower MM, Peters BB: Effects of pacing rateand timing of defibrillation shock on the relationship between thedefibrillation threshold and the upper limit of vulnerability inopen-chest dogs. JAm Coll Cardiol 1991;18:1555-1563

34. Babbs CF, Whistler SJ, Yim GKW: Temporal stability and preci-sion of ventricular defibrillation threshold data. Am J Physiol1978;H553-H558

35. Jones DL, Irish WD, Klein GJ: Defibrillation efficacy: Comparisonof defibrillation threshold versus dose-response curve determina-tion. Circ Res 1991;69:45-51

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

S L Topham, Y M Cha, B B Peters and P S Chenvulnerability in open-chest dogs.

Effects of lidocaine on relation between defibrillation threshold and upper limit of

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1992 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.85.3.1146

1992;85:1146-1151Circulation. 

http://circ.ahajournals.org/content/85/3/1146the World Wide Web at:

The online version of this article, along with updated information and services, is located on

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from