Effect of Field Stimulation on Cellular Repolarization in Rabbit Myocardium

(1)

Effect of

Field

Stimulation

on

Cellular

Repolarization in

Rabbit Myocardium

Implications for Reentry Induction

Stephen B. Knisley,William M. Smith, and Raymond E. Ideker

We have investigated the effects of electric field stimulation on membrane repolarization in rabbit

papillary muscles and assessed the consequences of these effects for the dispersion of intracellular

potentials and theproduction ofa propagationwavefrontorunidirectional block in relatively refractory

tissue. The stimuli studied had electric field strength of 0.25-14 V/cm, duration of 2 msec, and field

orientation alongor acrossthe myocardialfibers.Thefieldstrengthstoexcite the muscles in diastolewere

0.68 or 1.23V/cm for stimuli oriented alongor across the fibers, respectively (p<0.01, along versus

across).A2.5-V/cm stimulus givennearthe end of the actionpotential (AP)producedeithernoresponse

or, afterincreasing the stimulus delay only2-3 msec, afullresponsewith almostnoAP durations that

were intermediate. For stimulation along and across the fibers, respectively, given at 70% of the AP

duration,a4-V/cmstimulusproduced AP prolongation (measuredat90% repolarization) of209%, and 4%

(p<0.05), an 8-V/cm stimulus produced AP prolongation of 36% and 20% (p<O.O5), and a 14-V/cm

stimulusproducedAPprolongationof36%and30%o (p=NS).For either orientation, AP prolongation by

stimuliof 8V/cmor14V/cm increased graduallyasthestimulus delaywasincreased. The different effects

inrelatively refractory tissue ofstimuli of 2.5 V/cm compared with 8 V/cmcanexplain the propagation

wavefrontandblock thatoccurwithelectrically induced functionalreentryin the heart.After stimulation with fieldsbelowacriticalstrength (-5V/cm),alargeintracellular potential differencemayoccuramong

cells thatare sufficientlyrecovered tobecome excited by the stimulus and cells thatare notsufficiently recoveredtobecome excited, consistentwiththereportedpropagationwavefrontwhere thetwogroupsof

cells are closely opposed. After stimulation with fields above the critical strength, differences in

intracellular potentials among cells during repolarization may be decreased, and the intracellular

potentials maybeinarangeinwhich sodiumcurrentisinactivated, consistent withthereportedabsence

ofapropagationwavefront.Thus, the different effects of lowandhigh electric field strengthscanaccount for the "critical point" mechanism for unidirectional block and reentry. (Circulation Research

1992;70:707-715)

KEYWoRDs * myocardialstimulation * action potential * myocardialrepolarization * refractory period * gradedresponse * reentry * rotors * spiralwaves * vortices * excitable media

E lectrical initiation of reentry occurs in normal

myocardium around a point where a critical

electric shock field strength (5 V/cm for the particular waveform studied) intersects tissue that is

critically refractory(i.e., just comingoutof itsrefractory

period)."2

Under these conditions, extracellular

map-ping studies haveindicatedthatanactivation front first occurs after the shock where the critically refractory

tissue intersects a shock field weaker than the critical

From the Department of Biomedical Engineering and the EngineeringResearch Center inEmergingCardiovascular Tech-nologies, SchoolofEngineering,DukeUniversity,and the

Depart-ments of Medicine and Pathology, Duke University Medical Center,Durham,N.C.

Supported by National Institutes of Health research grants HL-28429,HL-33637, HL-44066, HL-42760,andHL-40092;

Na-tional Science Foundation Engineering Research Center grant

CDR-8622201; American Heart Association North Carolina Af-filiate grant NC-91-G-14; and a grant from the Lilly Research Laboratories.

Address forreprints: StephenB.Knisley,PhD,P.O. Box3140,

DukeUniversityMedicalCenter, Durham,NC 27710.

Received June 1,1991;acceptedNovember 25,1991.

electricfield strengthbut that an activation front does not occur where the shock field is stronger than the

criticalstrength. Activation thenpropagatesaround the "critical point" into the region of the stronger

field,

initiating reentry. The mechanisms for the

production

ofanactivation front in theregionwhere the shock field

strengthis <5V/cmand theunidirectionalblock in the

regionwhere thestrengthis >5V/cmareunknown. Itis

hypothesized that, in the region with the lower field

strength, an activation wave front propagates from the tissue that is

sufficiently

recovered to become

directly

excitedbythe shockinto thetissue that is

refractory

to

the shock.' Where a field

strength

>5 V/cm occurs in the relatively refractory tissue, it is

proposed

that a

graded response is produced that does not support a propagatedactivation front.3 Theelectric field

strengths

that are needed to produce such responses

intracellu-larlyhavenotbeenpreviouslydeterminedorcorrelated with the electric field

strengths

needed to

produce

reentry.

The effects of electric field stimulation

given

in the

refractoryperiod onthe

repolarization

ofthe

by guest on August 11, 2017

http://circres.ahajournals.org/

(2)

lular action potential (AP)mayultimately be explained by basic membrane mechanisms such as voltage- and time-dependent ioniccurrents.For example, changes in the transmembrane potentials are probably induced during the stimulus pulse; these changes, in turn, alter the ionic currents after the pulse. However, the direc-tions andmagnitudes of the changes in transmembrane potentials induced duringafield stimulus pulse andthe locationofthesechangesinthemyocardium have not as yetbeen reported. Furthermore, the effectsof induced transmembrane potential changes on the inward and outward membrane ionic currents that influence the timing of repolarization of theAP are notfully known.4 Most of what is known of the voltage dependence of ioniccurrentsappliestouniform transmembrane poten-tial changes, not the simultaneous hyperpolarization and depolarization expected in different parts of the membrane during field stimulation of the myocardium.5 Hence, the effects of electric field stimulation at strengths that initiate reentrant rotors in the heart on the intracellular AP cannot be reliably inferred from existing knowledge.

The purposeofthis studywasto determine whether shocks having timings and electric field strengths that encompass the critical point produce effects on the intracellular AP that explain the critical point mecha-nism for the induction of reentry. Since myocardial electrical properties are anisotropic and hence the

effects ofelectricfields maydepend on the fiber

orien-tation,6 the effects were determined for electric fields

oriented along and across themyocardialfibers. Materials and Methods

Seven rabbits weighing 5-6 pounds were anesthe-tized with intravenous pentobarbital. The heart was

rapidly removed and placed in Tyrode's solution at

room temperature, and a right ventricular papillary

muscle 3.3+0.3 mm long and 0.8+0.3 mm wide was

carefully removed andplacedin an experimentalbath. The muscle was positioned near the center of the 4x4-cmbath andoriented parallelwiththe sidesofthe bath. The muscle was superfused at a rate of 3-3.5 ml/min, and -3mmofsuperfusingsolutionwaspresent above and below the muscle. The solution contained

(mM)glucose 11, CaCl2 1.8, NaCl 125, KCl 5.4, MgCl2

1.05, NaHCO324, andNaH2PO4 0.42,withapHof7.4,

andwas bubbled with a mixture of 95% 02-5%

Co2.

The temperature, monitored with a miniature probe (Physitemp, Clifton, N.J.) in the solution below the muscle, was held at 36.5-37°C. Four flat 1x3.8-cm chlorided silver electrodeswereattachedtothe sides of thebathtoperform field stimulation (stimulusS2).The 2-msecconstant-voltage S2pulse was applied to either of thepairsofelectrodesatoppositesides ofthe bathto

produceanS2electric field orientedalongoracrossthe

longitudinal axis of the muscle. The S2 voltage was adjustedtocontrol the electric fieldstrengthin thebath. The electric field strength (i.e., potential gradient) for each S2 orientationwasdetermined fromasquare array oftungsten-wire50-,m-diameter electrodespositioned

around the muscle. The differentialrecording amplifier

for the S2 potential gradient measurements had an

inputresistance of 1012 Q. Theinterelectrode distances

microscope and graduated reticle. The distance

mea-surementswereperformedatthe end of theexperiment

after the solution levelwas loweredto eliminate errors

dueto light diffraction where the recording electrodes entered the solution.

Anintracellular APwasrecorded withaglass

micro-electrodenearthecenterof the muscle and afine-wire

extracellular reference electrode near the

microelec-trode tip. The intracellular and extracellular potentials

were passed through awideband differential electrom-eter (model 773, World Precision Instruments, New Haven, Conn.).

The leads of the isolated S2source wereconnectedto

theamplifier ground through two30-kQ resistors. This minimized the potential change in thecenterof the bath and, hence, the common-mode potential at the ampli-fierinputs when S2was applied.

Themuscle was paced (stimulus S1) near the tendi-nous end at a rate of 0.5 Hz with an isolated bipolar

2-msecconstant-currentpulse of 1.5 times the diastolic threshold strength. After a 1-hour stabilization period,

the diastolic excitation thresholdswere determined for

S2 electric fields oriented along or across the muscle fibers. In each trial,asingle S2wasappliedatthe end of the diastolicinterval in the absenceofS1. The intracel-lular AP was monitored to determine whether the S2 produced excitation. For each preparation, the thresh-old determinationswere repeatable towithin -3%.

The S2 strength was then increased to produce electricfields of approximately 2.5, 4, 8, and 14 V/cm. These S2 strengths were chosen to include the range

andapproximate distribution of electric field strengths thatwerepreviously showntoproducerotor-type reen-try in the intact heart.1 At each S2 strength, S2 was

given atS1-S2 intervals thatscanned the relative

refrac-toryperiod. Foragiven S2 strength and S1-S2 interval,

trials were performed with S2 electric fields oriented

along or acrossthe fibers. Theorder of the S2 orienta-tions was alternated. Recordings for the two orienta-tionswereobtained from thesamecellularimpalement.

For each trial, an intracellular recordingwas obtained

foranS1 paced beat (control beat) and for the following

S1 paced beat during which the S2 shockwasgiven (test

beat). AP prolongation was determined by subtracting

the AP duration at 90% repolarization of the control beat from the total duration of theresponse to SI and S2 of the test beat.

Thepotentialsweredigitizedatasamplingrateof1,000 Hz with an NB-MIO16H data acquisition circuit board

(National Instruments, Austin, Tex.) and stored on a

Macintosh lIx computer (Apple Computer, Cupertino,

Calif.). The AP durationat90% repolarizationwas

mea-sured from the computer recordings using the Labview

system (National Instruments). APswerealsomonitored

during the experiments with adigitizing oscilloscope.

Results are given as mean+1 SD. Statistical

signifi-cance wasdeterminedby two-tailed pairedttestsexcept

where indicated otherwise in thetext.

Results

Action PotentialProlongation and Excitation: The Impact ofthe Shock Field Orientation

The intracellular recordings in Figure 1 show the of -5 mm in each direction were measured with a effect ofan S2electric field

having

a

potential gradient

(3)

o- Across

-l5OmV

Si 52g

8.1 V/cm 100ms

FIGURE 1. Superimposed recordings showing action poten-tial prolongation produced by a field stimulus having an electric field strength of 8.1 V/cm (S2) oriented along or across the myocardial fibers. The recordings were obtained from one cellular impalement. The stimulus interval S1-S2 was 200 msec. The repolarization of the control action potential, which did not receive an S2, is shown as the interruptedtracing. S2producedasmallrapid depolarization and aprolongation ofthe repolarization time ofthe action potential compared with the control. Theprolongation pro-duced by S2 oriented alongthefibers wasgreater than the prolongationproducedby S2 oriented across the fibers.

of 8.1 V/cm oriented along or across the myocardial

fibers.TheS1-S2 interval was 200 msec in the superim-posed recordings. The repolarization after S2 of either orientation occurred later than the repolarization of the control AP, shown as the interrupted tracing. The AP prolongation was greater for S2 oriented along the fibers compared with across the fibers. A greater effect of S2 oriented along the fibers compared with acrossthe

fibersalso occurred for S2 given in diastole and having strengths near the diastolic excitation threshold. When S2was oriented along the fibers, thediastolic excitation threshold was 0.68±0.16V/cm. When S2 was oriented across the fibers, the threshold was 1.23+±0.27 V/cm, which wassignificantlygreater than thethreshold for S2

along the fibers (p<0.01).

Figure2 shows graphs of theAP prolongation versus

the S1-S2 intervalfor S2 electric fields of 2.3, 4, 8.1, and 12.9 V/cm. Each data point represents a different S2 trial. The S1-S2 interval and the AP prolongation are given as a fraction of the control AP duration at 90% repolarization, which was 147 msec. Thus, an AP pro-longation of 1 would represent a response that had a duration as great as the duration of the control AP.

S2=2.3V/cm Along- . Across

-,-K

C 1.4- 1.2-c 1.0-0 * 0.8 o 0.6-A. CL 0.4. ( 0.2 0.0--0.2 0.2 0.4 0.6 0.8 1.0 1.2 S1-S2 0.0 0.2 0.4 0.6 S1-S2 s2=4.0V/cm D 1.4- 1.2-c 1.0 0 * 0.8-c o 0.6-CL 0.4-. 0.2- 0.0--0.29 -0.2 0.4 0.6 0.8 1.0 1.2 S2=12.9V/cm Along 0.0 0.2 0.4 0.6 QC1QIO 0.8 1.0 1.2 S1-S2 * '

FIGURE2. Graphs showingactionpotential (AP)prolongation versusthe stimulusinterval51-S2 forelectricfieldstimuli(S2) of2.3, 4,8.1,and 12.9V/cm. Each datapoint representsadifferent trial. The resultswereobtainedfromonecellularimpalement. TheS1-S2interval and theAPprolongationaregivenasfractions ofthe control APdurationat90%repolarization. Thecontrol

APduration,which hada meanvalueof147msec,wasconstant towithin4msec. Panel A: The2.3-V/cm S2givenaslateasthe timeof 90% repolarization oftheAPproduced onlyasmallAPprolongation. When S2ofthisstrengthwasgiven only2-3msec

later,anewAPwasproduced.Panel B: When the4-V/cm S2wasgivenasearlyasthemidpoint oftheAP,S2 hadnoeffect. When

given late in relationtotheAPrepolarization, S2produceda newAP. The4-V/cmS2producedAPprolongation when given at intermediateSJ-S2intervals. TheAPprolongationwasgreater forS2 orientedalongcomparedwithacrossthemyocardial fibers. Panel C: When theS2electricfield strengthwas8.1 V/cm, like the resultsforthe4-V/cmS2, therewasanintermediaterangeof S1-52intervalsin whichagreaterAP prolongationoccurredforS2 orientedalongthefibers comparedwithacrossthefibers. Panel

D: When theS2strength was12.9 V/cm,APprolongation occurredforS2givenasearlyasthemidpoint oftheAP. For this S2

strength, theAPprolongation wasnotmarkedlydifferent forS2alongversusacrossthefibers.

A 1.4 1.2 c 1.0 0 0 0.8 O 0.6 CL 0.4-0s 0.2-0.0o 0.0 S2=8.1V/cm Across 0.8 1.0 1.2 B 1.4- 1.2-c 1.0 0 0 0.8-C o 0.6-0 CL 0.4-a. c 0.2 0.0 -0.0 _n^g -n2iE by guest on August 11, 2017 http://circres.ahajournals.org/ Downloaded from

(4)

Data pointsat anS1-S2 interval of1would represent an S2 given at the time of 90% repolarization of the AP. Figure 2Ashows that the 2.3-V/cmS2givenevenaslate asthe time of 90% repolarization of the AP produced only a small AP prolongation. When the 2.3-V/cm S2 wasgivenonly2-3 mseclater,a newAPwasproduced.

The sudden production of a new AP indicates the all-or-none response foran S2 of this strength.

Figure 2B shows the effect of the4-V/cm S2 electric fieldappliedalongor acrossthe fibers. S2producedAP

prolongation when given at S1-S2 intervals of -0.8-1.

When givennearthe midpoint of the AP (S1-S2=0.5),

the 4-V/cm S2 did not prolong the AP for either S2 orientation. When S2was given late in relation tothe APrepolarization, a new AP wasproducedfor either S2 orientation. At intermediateS1-S2 intervals of-0.8,S2 along the fibers hada greatereffect than S2 across the fibers.

Figure 2C shows that when the S2 electric field strengthwas8.1V/cm,APprolongationoccurred for S2 givenduring the second half of theAP.Againtherewas an intermediate range of S1-S2 intervals in which a greater APprolongationoccurred forS2 orientedalong

the fibers comparedwith across thefibers.

Figure 2D shows that when the S2 electric field strength was increased to 12.9 V/cm, APprolongation

became noticeable for S2 given as early as an S1-S2

interval of 0.5. Unlike the results for the weakerS2,the APprolongationproducedbythe 12.9-V/cmS2was not markedly different for S2 oriented

along

the fibers

compared with across the fibers.

The control AP durationat90%

repolarization

inthe seven experiments was 151±32 msec. For S2 electric fields having a strength of 2.5±0.1 V/cm, which pro-duced all-or-none responses, the smallest S1-S2 interval that resulted in a new AP was 152±33 msec for S2

oriented along the fibers and

156±+

33 msec for S2

orientedacrossthe fibers

(p=NS).

Forstimulation with

S1-S2 intervals that were 60-80% of the control AP durationat90%repolarization

(an

S1-S2range in which thecellswere

highly

refractory

andhence the responses were not newAPs), S2 of 4.1±0.2 V/cm producedAP

prolongation of 20±12%

(p<0.05)

for S2 oriented

along the fibers and 4±4%

(p=NS)

for S2 oriented acrossthefibers; for the samerangeofS1-S2

intervals,

S2 of 8.3±0.3 V/cm

produced

AP

prolongation

of

36±8%

(p<0.05)

for S2 oriented

along

the fibers and

20±4%

(p<0.05)

forS2 orientedacross the

fibers,

and S2 of 14±1V/cm

produced

AP

prolongation

of36± 13%

(p<0.05) for S2 oriented

along

the fibers and 30±9%

(p<0.05) for S2 oriented across the fibers

(p

values were calculated by

comparison

withzeroAP

prolonga-tion, mean±SD for seven

preparations).

For the 4.1-and 8.3-V/cm S2, the AP

prolongations corresponding

to the two S2 orientations were

significantly

different

(p<0.05, alongversus

across).

Forthe 14-V/cm S2, the APprolongationscorrespondingtothetwoorientations were not significantlydifferent.

Assessment

of

theImpact

of

Shocks on the

Dispersion

ofRepolarization

Figure3illustratestheabilityofashockat anelectric fieldstrength of 8.4 V/cm(avalueprobablygreaterthan the critical

strength"2,7)

to decrease the

dispersion

of

0

-80

S1-S2 (ms)

170

100ms

FIGURE 3. Action potential recordings illustrating the de-crease in the dispersion ofrepolarization produced by an electric field stimulus (S2) having a strength of 8.4 V/cm orientedalong the fibers. The recordings, obtained from one cellular impalement, are aligned with the S2 time. StimulusS] wasapplied -3 msecbefore the phase-zero depolarization of each of the recordings. The longest and shortest S1-S2 intervals tested, 230 and 90 msec, are indicated with their respectivephase-zero depolarizations. In addition, the S1-S2 intervals foreachof the responses after S2 are indicated to the right of the recordings. The dispersion of repolarization for cells that receiveS2 at these different times during the action potential is indicated by the variation of times of repolariza-tion after S2. If S2 had no effect, the dispersion of repolar-izationfor the various trials would be 140 msec (the variation of times ofphase-zero depolarization). The responses after S2 indicate that S2decreased the dispersion of repolarization to 100 msec (the variationof times ofrepolarization). A window of S1-S2 intervals occurred in which the dispersion of repo-larization was negligible (S1-S2=130-170 msec). For S2 given either earlier or later than thiswindow, the repolariza-tiontimeafter S2 increased.

repolarization. The recordings are superimposed and aligned with the shock time. The decrease in the

dispersion of repolarizationcanbe assessed by

consid-eringthe time from the shocktorepolarizationafter the shock. Ifthe shock had notaffected repolarization, the variationinthe repolarization times oftherecordingsin Figure 3 would equal the 140-msec variation in the S1-S2 interval indicated by the dispersion of the AP

phase-zero depolarizations in the left part ofFigure 3. The variation in the repolarization timesinFigure3was only 100msec (i.e.,itwas decreased by 40msec). This decrease corresponds to adecrease in the dispersionof repolarization for cells that receive the shock at the various times during the APs shown. There was a window of timesduring theAPinwhichthevariationof repolarization time after the shockwas negligible

(S1-S2=130-170msec). This window occurred at-68% of the control AP duration at 90% repolarization, which was 221 msec in this experiment. Whenthe shock was applied at times earlier than the window

(S1-S2=90-110 msec), the repolarization time after the shock became greater. The repolarization time also became greaterwhen the shockwasappliedat timeslaterthan the window (S1-S2>180 msec). When the shock was applied sufficiently late in the AP, a new AP was produced (e.g., S1-S2=230 msec).

The effect ofashockat alow electric fieldstrengthon the cell'srepolarizationtime is shownin Figure4.The

(5)

1.6Vlcm, 2 ms

200-S1-S2 (ms) 222 -225 0 --8 -80 100 ms

FIGURE4. Recordings illustratingtheresponsetoanelectric

field stimulus (S2)atastrength of1.6V/cmorientedalongthe

fibers. ThetwoS2-alignedrecordingswereobtainedfromthe samecellasinFigure 3.TheS1-52stimulus intervalsfor each

of the responsesare indicatedto the right ofthe recordings.

Theresponsesweremarkedly different dependingon achange

in theS2timing ofonly3msec. S2givenatanS1-S2interval

of 222msecproduced onlyasmallresponse, whereas S2given

at an S1-S2 interval of225 msec produced a new action

potential.

two shock-aligned recordingsshowthemarkedly

differ-ent responses that occurred after the 1.6-V/cm shock with only a small change in the shock timing. This indicates the all-or-none responsethatoccurs withlow

field strengths.

Figure5 illustrates the effects ofshockshaving elec-tric fieldstrengthsfrom 1.4 to 14.6V/cm. The variation

in the time from the shock to 90% repolarization (S2-R90), onthevertical axisofFigure 5, corresponds

tothe dispersionofrepolarization after the shock. Ifa

shock of agiven strength had not affected repolariza-tion,theplotwould bealine withaslopeof -1. For the

8.4- or 14.6-V/cm shocks given over a 100-msec-wide

range of S1-S2 intervals, the variation in S2-R90was

only35msec,or onethird of what itwould have been if

the shock had no effect. The S1-S2 window for the

production of a constant S2-R90 was 60-90 msec, or

-54%of the controlAPdurationat90%repolarization, whichwas 140msecin thisexperiment.The absence of an effect of the 1.4- or 2.5-V/cm shocks for most times

duringtheAP is indicated where theslopeof the S2-R90

curveis-1. The sudden increase in the S2-R90interval,

duetotheproductionofa newAP,isseenfor theweak

shocksgiven late in the actionpotential.

Implications fortheInduction of Reentry

The intracellular potentials afterS2 can explain the

propagationwave front in a region of the heartwhere

the S2 electric field strength is less than a critical

electric field strength and the block where the S2

strengthisgreater thanthe criticalstrength.1,8 Figure6

shows intracellular potentials 10 msec after S2

mea-sured in repeated trials from a single cellular

impale-ment. The values are shown on a map ofintersecting linesofS2 electric field strengthsand states of cellular

refractorinesspatterned afterFigures2-4 of Frazier et

al.' In theregionwhere the S2 electric fieldstrengthwas

low,the cells thatweredirectlyactivatedbyS2areclose

(upper rightofFigure 6)tocells thatwere not directly activated; thus, a large intracellular potential gradient

new APs E O- ~ ~ ~4.I6.~ ~ 100 102 ~~

8.4

S. 12-U S S1.S

2ms)

FIGURE 5. Assessment of the effects of electric field stimuli (S2)at fieldstrengthsrangingfrom 1.4 to 14.6V/cmoriented alongthefiberson thedispersionofrepolarization. The time from the shock to 90% repolarization of the actionpotential (S2-R90) is shown on the vertical axis. The data points represent different trials performed with a single cellular impalement. Connecting lines are used to indicate trialswith the same S2strength (V/cm indicated). The control action potential(AP) duration at 90%repolarization was 140 msec inthisexperiment. For a given S2 electric field strength, the variation in S2-R90 for the various S1-52 intervals corre-sponds to the dispersion of repolarization after S2 for cells that receiveS2 at these different times during theAP. If a given S2shock hadnotaffected repolarization, the plot would be a linewith a slope of -1. For the8.4- or14.6-V/cm S2given over a 100-msec-wide range ofS1-S2intervals, the variation in S2-R90wasonly 35 msec, or one third of what itwould have beenifS2had no effect. Aconstant S2-R90, correspond-ing to the abolition ofdispersion of repolarization, occurred after the

8.4-

or

14.6-V/cm

S2 given in an

SJ-S2

window of

60-90msec. Theabsenceof an effect of the1.4-or2.5-V/cm

S2 formosttimesduring the AP isindicated where the slope of the S2-R90curve is -1. The sudden increase in the S2-R90 interval, which isdue to the production of a new AP, is seen fortheweakS2given late in theAP.

of -0.5 V/cm occurs that would initiate propagation from right to left in the upper part of the figure. The intracellularpotential gradient along a line of S2 elec-tric field strengthsof 15 V/cm is small (largestvalue is

only 0.017 V/cm for S1-S2 intervals of 210-220 msec) andtherefore much less likely to initiate a propagation wave front. The intracellular potentials in the region where the S2 electric field strength was -8-15 V/cm werefrom -23to+28mV. Inthisrange of intracellular

potentials, sodium channels are largely inactivated,10

furtherpreventing apropagationwave front.

Discussion

Action Potential

Prolongation

Producedby Field Stimulation: The Impact

of

the Field Orientation

The orientation of the stimulus electric field with respecttothemyocardialfibers isoneofthe factors that determine theAPprolongationby field stimulation. The

(6)

1.5 P 5 E u >I r1

4

8

15

140 160

S1

-

S2

(ms)

FIGURE 6. Intracellularpotentials after electricfieldstimulation (S2). The measurements were obtainedfrom intracellular

recordingswithonecellularimpalement forwhich variousS2electricfield strengthsorientedalongthefibersandS1-S2stimulus intervalsweretested. Theintracellularpotentials, givenin millivoltswithinthegraph, occurred10msecafterS2wasapplied. The intracellularpotentialsaregraphedwith theS2electricfield strengthon thevertical axis and thestateofrefractoriness, or 51-S2 interval, onthe horizontalaxistocorrespondwith thespatially dispersedstimulusstrengthsandstatesof refractorinessthatinduce

reentryin theheart.] The 1-cmcalibration bar isapproximateand is basedon anassumedconductionvelocity of65cm/secfor therepolarizationwaveofthepreceding S1beat.1Theisopotentialcontours,determinedbybivariatepolynomial interpolation9and

smoothing by hand, represent intracellularpotentials from -45 to 25 mV in 10-mV increments. Contoursfor intracellular

potentialsmorenegativethan -45 mVarenotshown becauseofuncertainty in theinterpolation.In theregionwhere theS2electric

field strengthis -1.5V/cm, anabrupt boundary(upper right)occursbetweenthetissuethat isdirectlyexcitedby52 and thetissue that isnotdirectlyexcited.Alargeintracellularpotential gradientattheboundary, -0.5 V/cm, canaccountfortheinitiationof propagation from righttoleftthathas been reportedintheregion havinga lowelectricfieldstrength.]8 Thelargeintracellular

potential gradientattheboundaryiscomparabletothatwhichoccursatapropagationwavefrontinmyocardialtissue with normal intercellular connections. In theregion havingahighS2electricfield strength, there isnoabrupt boundarybetweenhighandlow intracellularpotentials. Instead, on the lineofan S2 strength of15 V/cm, thelargestintracellularpotential gradient, ignoring

discontinuitiesofintracellularresistance, isonly0.017V/cm.Also, theintracellularpotentialsonthe15-V/cm linearein therange

inwhich sodiumcurrentisinactivated.'0 Theabsenceofalargeintracellularpotential gradientand thepresenceofintracellular

potentials thatinactivate sodiumcurrent, acurrentimportantfor propagation, explain theabsenceofpropagation afterashock where the electricfield strength duringtheshock ishigh."18

AP prolongation for an S2 of -4-8 V/cm was greater

whentheelectric fieldwasorientedalong themyocardial

fibers than across the fibers. When the S2strength was

increased to 14 V/cm, however, the AP prolongation becamesimilar for thetwoS2 orientations. Theseresults

can be explained on the basis of two stages of the myocardial response to field stimulation: stage 1, a

transmembrane componentof the applied stimulus cur-rentthat alters the transmembrane potential during the stimulus pulse but does not require changes in

mem-brane ionic conductances; and stage 2, a regenerative

transmembrane ioniccurrentduetothetransmembrane potential dependence of the membrane ionic

conduc-tances. In stage 1 the change in the transmembrane

potential is determined by linear passive properties of myocardium; hence, the change in the transmembrane potential is proportional to the stimulus strength, whereas in stage 2 it is limited by the availability of

transmembrane ionicdriving forces (e.g., the difference between the transmembrane potential and the equilib-riumpotential for each ion) and membrane ion channels. Diastolic excitation by electric field stimulation de-pends on the ability ofstage 1 to bring the

transmem-brane potential to a threshold potential for a

regener-ative response in stage 2 that produces an AP phase-zero depolarization. The lower electric field strength

needed for diastolic excitation by S2 oriented along fibers compared with across fibers suggests a greater

ability of S2 along fibers to produce a transmembrane

potential responsein stage 1.

A greater stage 1 response for S2 electric fields

oriented along fibers compared withacrossfiberscould

explain thegreaterAPprolongation by 4-or8-V/cm S2

along fibers compared with acrossfibers, assuming that

the stage 2 response to S2 of these strengths given to

relatively refractorymyocardium isgreaterforagreater

12 10 14' 25 28 = lci 180 200 220 l

240

a 5 1 a :m by guest on August 11, 2017 http://circres.ahajournals.org/ Downloaded from

(7)

stage 1 response. This assumption is supported by the greater APprolongation for a given S2 orientation when the S2 electric field strength was increased from 4 to 8 V/cm. However, the 14-V/cm S2 did not produce a

significantlygreater AP prolongation for S2 along fibers

compared with across fibers even though, like the weaker S2, the stage 1 response to the 14-V/cm S2 would be greater for S2 along fibers compared with acrossfibers. Thesimilarity of the APprolongationfor

14-V/cm S2 along fibers compared with across fibers could be due to limited amounts of transmembrane ionic driving forcesormembrane ion channels in stage 2. If true, agreater stage 1 response (e.g.,for 14-V/cm S2 alongfiberscomparedwith acrossfibers) wouldnot further increase the stage 2 ionic current. Limited amounts of stage 2ionicdriving forces orchannelscan also explain the similarity of AP prolongation by the 8-V/cm S2along fibers comparedwith the 14-V/cm S2 along fibers. Under this interpretation of the results,

limitation ofthe stage 2 response occurswithanelectric field strength of -8 V/cm oriented along the fibers,

whereas itoccurs witha higher field strengthfor fields

oriented across thefibers.

A greater stage 1 response for S2 along the fibers compared with S2 across the fibers would be consistent withthe greaterlengthof thecells in the directionalong themyocardial

fibers.1'

Also, the myocardialresistance is lower in the directionalong themyocardial fibers.6 An electric field ofagivenstrength would produce a greater

myocardial current density when the field is oriented along the fibers compared with across the fibers.

In-creased current density might increase the ability to stimulate the cells.

The Importance ofActionPotential

Prolongation for

Electrical

Defibrillation

orCardioversion

The distribution of repolarization in the heart is an important factor inmosthypothesesofarrhythmiasand

fibrillation."1'2-'4

Thechanges in cellular repolarization

that are produced by an electric shock314-17 may be important fordefibrillation. For example, the repolar-izationdelayproduced byasufficientlystrongelectrical stimulus during the relative refractoryperiod can pre-ventpostshock

activation18'19

and hencemayprevent the cells frombecomingexcited by a reentrant wavefront. The assessment of the effect ofa shock on the

disper-sion ofrepolarization indicates that the electric fields havingstrengthsof 8-14V/cm decrease the variation of

repolarization times by -30% (Figure 3) or more (Figure 5). This suggests that theamountof tissue that is inagivenstateofrefractorinessat aninstant after the stimulus increases at least

40%.

Thecontinuation of a reentrantarrhythmiarequires that tissue somewhere in the reentrant pathway is sufficiently recovered to be-comeexcited. Excitable gapsduringreentrant ventricu-lartachycardiainendocardiallyfrozenrabbit hearts are only 23%of the pathwayand, inleadingcirclereentry, canbe much less.20 Given that the amount of tissue is

fixed, a 40% increase in the amount of tissue that is refractorycan besufficienttoabolish the excitable gap and henceinterrupt reentrant arrhythmias.

Incontrastwith thepossibilitythat theAP

prolonga-tion mayinterruptorterminate areentrantarrhythmia,

the prolongation may contribute to the initiation of

reentry. Such electrically induced reentry may cause defibrillation to fail by restarting fibrillation.8

The Importance of Action Potential Prolongation for RotorInitiationAround a Critical Point

In the critical point mechanism,1,2 reentry occurs where contours ofspatially dispersed states of refracto-riness,isorefractoriness lines, intersect contours of spa-tiallydispersed stimulus electric field strengths, isostim-ulus lines. The contour lines need not bestraight lines. The isorefractoriness lines correspond to states of the recovery process that have been described in excitable media

theory.7'21'22

The isostimulus lines, along which the stimulusstrength is constant,correspondtostatesof the excitatory process. The excitatory state increases rapidly when the medium is stimulated, which

corre-sponds to the phase-zero depolarization of the cell membrane in themyocardium. The recovery state cor-responds to the recovery of excitability during the

repolarizationofthecell membrane. The recovery state is reset by an increase in the excitatory state and

changes slowly thereafter. It is postulated that critical statesof each of the excitatory and recovery processes occur when a stimulus of an appropriate strength is given at anappropriate time during the recovery. When a line corresponding to one of the critical excitatory states intersects a line corresponding to one of the criticalrecovery states,acriticalpoint isproducedinan excitable medium. An excitation wave front occurs on oneside of the criticalpointand thenpivotsaround the critical point, provided that a sufficient amount of the mediumexistson all sides of the criticalpointand that themedium on the respective sides contains excitatory and recoverystatesthatinclude values greater than and less than the critical states. The theoretical framework

involving the critical point has been used to explain

vortices in several chemical and biological excitable media and in computer simulations (for references see Reference 21).

The different effects of the weak versus strong S2 electric field stimuli can explain the initiation of rotors of reentry in the heart by an appropriately timed

stimulus.12'7

On a line of low S2 electric field strength (e.g.,1or2V/cm),there isastateofrefractorinessnear the end of an AP where some cells have very little response(i.e., arenotexcited),whereas other cells that areonlyafew millisecondsmore recovered haveavery

large depolarization (i.e., are directly excited) (Figure 6). Intracellular current from the cells that have the

large depolarization to the cells that are not excited shouldinitiateapropagationwavefrontattheborder of the excited and nonexcited cells. Thewave front then propagates along the line of low S2 electric field strength into theregionthat was notexcitedbythe S2. OnalineofhighS2 electric fieldstrength(e.g., -8or14

V/cm), there is no abrupt boundary between excited andnonexcitedregions.Thecells thataretoorefractory to become directly excited by the shockundergo

pro-longationof theAP.Theprolongationisgreatest for the cells thatare mostrecovered andgraduallybecomes less forcells that arelessrecovered (Figure

2D).

Withouta

region of excited cellsadjacentto aregionof nonexcited cells, intracellular current is small; hence, apropagation wave front does not occur immediately after the S2

(8)

where the S2 electric field strength is high. Such a propagation wave front is further prevented by the prolonged period ofvoltage-dependent inactivation of sodium current that is due to the prolongation of the AP. At a later time, the region in which the AP was prolonged by the strong S2 electric field recovers enoughtobecome excited again bya propagationwave front. The wave front that was initiated where the S2 electricfieldwas weak then propagates into the recov-eredregion, whichproducespivotingof thepropagation wavefrontaround the critical point (i.e., counterclock-wise rotation in Figure 6). When the region that was directly excited by the S2 recovers, the wave front continues to pivot into that region and eventually reachesits origin.Thus,bypropagatingaroundacritical point, thewavefrontcompletesthefirstcycleofreentry. Extracellular mapping studies in the intact heart have indicated that thecenterof the reentrantcircuitoccurs at a critical electric field strength of -5 V/cm for a

3-msec truncated exponential waveform,' a strength greater than the values that consistently produced all-or-none responses (-2 V/cm) and less than the values thatconsistently produced APprolongation (-8V/cm)

in the present results. Limitationsof the Study

The interpretation of the findings in terms of the spatial dispersion of repolarization and intracellular potential is based on the previously observed consist-encyofpropagation velocityoftherepolarizationwave during which the shock is applied' and an assumption

that theintercellularconnections in themyocardiumdo not prevent intracellular potential gradients such as those shown inFigure 6. This assumption is supported by experimental evidence. Large intracellular potential differences over small distances occur during propaga-tion in normally coupled rabbit papillary muscles

(au-thors' unpublishedobservations)andduring repolariza-tion afterelectrical stimulation in myocardialfibers.3"15,23

Therefore, it can be assumed that the intracellular po-tential differences described in Figure 6 are not pre-vented by intercellular connections. Since the connec-tions maydecrease the intracellular potential gradients, the differences shown indicate the upper limit of the differences thatoccurinthe heart.

Theexperimentswereperformedat alongerS1cycle length than occursin vivo. Experimental evidence sug-gests that results qualitatively similar to the present resultscould be obtained withashorterS1cyclelength. Repolarization prolongation by shocks occurs during basic pacing at a cycle length of 350 msec in dog

hearts.'6

Also, refractory period extension by shocks, which is related to AP prolongation,24 occurs at cycle lengths shorter than those used here.

Conclusion

The effects of electric field stimulation on the myo-cardial intracellular AP reported here can explain a mechanism for theinductionof reentry by apremature electrical stimulus. For tissue in which the refractory state is distributed, stimulus electric field strengths above or below a critical value of -5 V/cm produce either graded prolongation ofthe repolarization of the

tively. Althoughtheexistence of the basic all-or-noneor

graded responses in relatively refractory myocardium was known from previous studies,3 5.623 the stimulus

electric field strengths required to produce these

re-sponses werenotpreviouslyknown. Since such

intracel-lularresponseshave beenhypothesizedtobeimportant for the initiation of reentrant rotors by electric field stimulation,'18 this study determined whether the

re-sponses areproduced bythe electric field strengthsthat

are known to initiate reentrant rotors. The results indicate forthe first timetheclose agreement between theelectricfieldstrengths thatproduce graded

prolon-gation ofrepolarization and thestrengthsthatproduce block in critically refractory tissue in the heart.' The graded prolongation, which is shown here to decrease intracellular potential differences and depolarize the cellstointracellular potentialsatwhich sodiumcurrent

isinactivated, accounts forthe block in a region ofthe

heart that receives electric fields stronger than the critical strength. The all-or-noneresponse, shownhere

to occur for electric fields weaker than the critical

strength, introduces a region of excited cells in close

proximity to aregion of cells that are not excited and

arenearly repolarized and hence increases the

intracel-lular potential difference atthe boundary between the regions. Thisaccountsfor theinitiation ofthe

propaga-tionwave front seen where theelectric field is weaker

than the critical strength.' Thus, the magnitude of the critical electric field strength above or below which

thesemarkedly differenttypes ofeffects (gradedversus

all-or-none) occur agrees qualitatively with the

--5-V/cm value of the electric field reported for the center

of the stimulus-induced reentrantrotor.1

The measurements of the effects of electric fields

weaker than 14 V/cm on relatively refractory tissue

indicate for the first time that electric fields oriented along the myocardial fibersaremoreabletoprolongthe repolarization thanarefieldsorientedacrossthefibers.

Thus, fiber orientation, important for myocardial char-acteristics such as resistance and conduction velocity6

and the occurrence of conduction block,25,26 is also importantforexcitation andprolongation of repolariza-tion produced by electrical stimulation. The greater

effectiveness observed forelectricfields oriented along the fibers compared with across the fibersimplies that

block andreentrymayoccurwithaweaker electricfield

when the field is orientedalong the fibers.

Foranelectric field strength of 14 V/cm,orienting the

fieldalong rather thanacrossthemyocardial fibersdoes notsignificantly increase the prolongation of repolariza-tion. Furthermore, when the electric field is oriented along thefibers,increasing the field strength from 8.3to

14V/cm doesnot increase the prolongation of repolar-ization. Thissuggeststhatanupperlimitoftheamount

of prolongation by these stimuli exists and that, for electric fields oriented along the fibers, the limit is reached withan electricfield of -8 V/cm.

Noteadded inproof. Adescription of cellular effects ofelectric shocks in rabbit heart27 was published after

this articlewas submitted.

References

1. Frazier DW, Wolf PD, Wharton JM, Tang ASL, Smith WM,

intracellular AP or an all-or-none response, respec- IdekerRE: Stimulus-induced criticalpoint:Mechanism for

(9)

trical initiation ofreentry in normal canine myocardium. J Clin Invest1989;83:1039-1052

2. Winfree AT:Sudden cardiacdeath: A problem in topology. SciAm

1983;248:144-161

3. KaoCY,HoffmanBF:Graded and decremental response in heart musclefibers.Am JPhysiol1958;194:187-196

4. Noble D: Ionicmechanisms controllingtheaction potential dura-tion and thetiming of repolarization.Jpn Heart J1986;27:3-19 5. KrassowskaW, FrazierDW,Pilkington TC, IdekerRE:Potential

distribution in three-dimensional periodic myocardium: Part II.

Applicationtoextracellularstimulation.IEEE TransBiomedEng

1990;37:267-284

6. Clerc L: Directional differences of impulse spread in trabecular

muscle frommammalian heart.JPhysiol (Lond) 1976;255:335-346

7. Winfree AT: When Time Breaks Down: The Three-dimensional

DynamicsofElectrochemical Wavesand CardiacArrhythmias.

Prince-ton,NJ,Princeton University Press,1987

8. ChenP-S, Wolf PD,DixonEG,DanieleyND, Frazier DW,Smith

WM, IdekerRE: Mechanism of ventricularvulnerabilitytosingle

prematurestimuli inopen-chest dogs.Circ Res1988;62:1191-1209

9. SAS/GRAPHSoftware: Reference, Version6, ed 1.Cary, NC,SAS

Institute Inc, 1990, vol 2, p 1320

10. FozzardHA,ArnsdorfMF:Cardiacelectrophysiology,in Fozzard

HA, Haber E, Jennings RB, Katz AM, Morgan HE (eds): The Heart andCardiovascularSystem. NewYork,RavenPress, Publish-ers, 1986, pp 1-30

11. KleeM,PlonseyR:Stimulation ofspheroidalcells: The role of cell

shape.IEEE TransBiomedEng1976;23:347-354

12. Surawicz B, Knoebel SB: LongQT: Good, badorindifferent?JAm

Coll Cardiol1984;4:398-413

13. KuoC-S,Munakata K,Reddy CP, SurawiczB:Characteristics and

possible mechanismofventricular arrhythmia dependent onthe

dispersion of action potential durations. Circulation 1983;67:

1356-1367

14. DillonSM:Defibrillatingshockscausenearlyconstant

repolariza-tion time in rabbit hearts (abstract). Circulation 1990;82(suppl III):III-640

15. Knisley SB, Afework Y, Li J, Smith WM, Ideker RE: Dispersion of repolarization induced by a nonuniform shock field. PACE 1991; 14:1148-1157

16. ZhouX,Knisley SB, WolfPD,RollinsDL,SmithWM,Ideker RE:

Prolongation of repolarization time by electric field stimulation with monophasicand biphasic shocks in open-chest dogs. Circ Res

1991;68:1761-1767

17. Swartz J, Jones JL, Fletcher R: Symmetrical biphasic defibrillator waveforms enhance refractory period stimulation in the human heart(abstract).JAmCoil Cardiol1991;17:335A

18. Sweeney RJ, Gill RM, Steinberg MI, Reid PR:Ventricular refrac-toryperiod extension caused bydefibrillationshocks.Circulation 1990;82:965-972

19. Knisley SB, ZhouX,RollinsDL, SmithWM,IdekerRE:Shock

strength and timing to prolong ventricular repolarization and

refractoriness (abstract).JAmColl Cardiol 1991;17:335A

20. Allessie MA, Schalij MJ,KirchhofCJHJ, Boersma L,Huybers M,

HollenJ:Experimental electrophysiologyandarrhythmogenicity:

Anisotropyandventriculartachycardia.Eur Heart J1989;10(suppl

E):2-8

21. WinfreeAT: Vortexaction potentials in normal ventricular

mus-cle. AnnNYAcadSci 1990;591:190-207

22. Keener JP: A geometrical theory for spiral waves in excitable media.SIAMJAppl Math 1986;46:1039-1056

23. CranefieldPF, HoffmanBF: Propagatedrepolarization in heart

muscle. JGenPhysiol1958;41:633-649

24. Sweeney RJ, Gill RM, Reid PR:Characterization of refractory

period extension by transcardiac shock. Circulation 1991;83:

2057-2066

25. Janse MJ: Reentryrhythms, in Fozzard HA,Haber E,Jennings RB, KatzAM,Morgan HE(eds): TheHeartand Cardiovascular

System. NewYork,RavenPress, Publishers, 1986,pp1203-1238

26. Spach MS: Anisotropicstructural complexities in the genesis of

reentrant arrhythmias. Circulation 1991;84:1447-1450

27. Dillon SM: Optical recordings in the rabbit heart show that

defibrillation strengthshocksprolong the duration of

depolariza-tionand therefractoryperiod.Circ Res1991;69:842-856

(10)

S B Knisley, W M Smith and R E Ideker

for reentry induction.

Print ISSN: 0009-7330. Online ISSN: 1524-4571

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Circulation Research

doi: 10.1161/01.RES.70.4.707 1992;70:707-715

Circ Res.

http://circres.ahajournals.org/content/70/4/707

World Wide Web at:

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

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research

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

about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further information Editorial 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 the Circulation Research

in

Requests for permissions to reproduce figures, tables, or portions of articles originally published

Permissions: