Patterns of spiral tip motion in cardiac tissues

CHAOS VOLUME 8, NUMBER 1 MARCH 1998

D

Patterns of spiral tip motion in cardiac tissuesDave T. Kim, Yvonne Kwan, John J. Lee, Takanori Ikeda, Takumi Uchida, KamyarKamjoo, Young-Hoon Kim, James J. C. Ong, Charles A. Athill, Tsu-Juey Wu, LawrenceCzer, Hrayr S. Karagueuzian, and Peng-Sheng ChenDivision of Cardiology, Department of Medicine, Cedars-Sinai Medical Center, and UCLA Schoolof Medicine, Los Angeles, California 90048

~Received 24 June 1997; accepted for publication 30 September 1997!

In support of the spiral wave theory of reentry, simulation studies and animal models have beenutilized to show various patterns of spiral wave tip motion such as meandering and drifting.However, the demonstration of these or any other patterns in cardiac tissues have been limited.Whether such patterns of spiral tip motion are commonly observed in fibrillating cardiac tissues isunknown, and whether such patterns form the basis of ventricular tachycardia or fibrillation remaindebatable. Using a computerized dynamic activation display, 108 episodes of atrial and ventriculartachycardia and fibrillation in isolated and intact canine cardiac tissues, as well asin vitro swine andmyopathic human cardiac tissues, were analyzed for patterns of nonstationary, spiral wave tipmotion. Among them, 46 episodes were from normal animal myocardium without pharmacologicalperturbations, 50 samples were from normal animal myocardium, either treated with drugs or hadchemical ablation of the subendocardium, and 12 samples were from diseased human hearts. Amongthe total episodes, 11 of them had obvious nonstationary spiral tip motion with a life span of.2cycles and with consecutive reentrant paths distinct from each other. Four patterns were observed:~1! meandering with an inward petal flower in 2;~2! meandering with outward petals in 5;~3!irregularly concentric in 3~core moving about a common center!; and ~4! drift in 1 ~linear coremovement!. The life span of a single nonstationary spiral wave lasted no more than 7 completecycles with a mean of 4.664.3, and a median of 4.5 cycles in our samples. Conclusion:~1! Patentlyevident nonstationary spiral waves with long life spans were uncommon in our sample of mostlynormal cardiac tissues, thus making a single meandering spiral wave an unlikely major mechanismof fibrillation in normal ventricular myocardium.~2! A tendency toward four patterns ofnonstationary spiral tip motion was observed. ©1998 American Institute of Physics.@S1054-1500~98!01601-2#

-

l

enirhas

estsing,o-

ngor-

ve

er-p-

nsorgcit-blein aed,

suethe

trola-du-

With the advent of novel action potential mapping tech-niques in cardiac tissues, patterns of spiral wave movement have been well illustrated. By incorporating a com-puterized dynamic activation display, the motion ofnonstationary spiral waves can be traced and character-ized to a degree of resolution better than ever before.Given the sheer number of activation recordings of vari-ous cardiac tissue preparations from animal models andpost-mortem human hearts available from prior experi-ments in our lab, we have exploited this valuable re-source to assess the prevalence of nonstationary spirawaves and its various patterns during VF. Our study hasrevealed that a nonstationary spiral wave is an uncom-mon occurrence with a short life span in our sample ofdiverse cardiac tissue preparations. In addition, tenden-cies toward four patterns of nonstationary spiral waveswere observed, and there was no uniform pattern thatwas consistently evident in cardiac tissue.

I. INTRODUCTION

Various mechanisms for cardiac fibrillation have beproposed in the past. However, the proposal of the spwave theory1–5 has sparked novel and exciting research tattempts to integrate and explain the complex mechanism

1371054-1500/98/8(1)/137/12/$15.00

ownloaded 07 Jan 2008 to 169.232.48.36. Redistribution subject to AIP lic

altof

fibrillation. There has been mounting evidence that suggthat spiral waves may be the basic mechanism underlyreentrant arrhythmias.3,4,6–8When a spiral wave is stationaryit results in regular and periodic excitation of cardiac mycardium, simulating atrial or ventricular tachycardia~VT!.9

However, when the spiral wave is nonstationary by driftior meandering, the electrocardiogram may show polymphic VT10 or even ventricular fibrillation~VF! through Dop-pler effects.9,11 Based on the latter observation, spiral wameandering,1,2,11,12or precession,13 plays a central role in thegeneration of cardiac fibrillation and sudden death. Meanding, which is governed by the recovery and excitability proerties used in computer simulation studies,12,14 describes apattern of spiral tip motion that results in flowery patterwith either inward or outward petals. It is also possible fthe spiral tip to drift in a linear fashion without forminflowers.4 Because cardiac tissues might have different exability and recovery properties than a noncardiac excitamedium, spiral waves in cardiac tissues may meandercertain preferential pattern. If these patterns can be identifithe excitability and recovery parameters of a cardiac tisduring spiral wave formation may be estimated based onsimulation results. Furthermore, it may be possible to conthe spiral wave movement by altering its physiological prameters, such as tissue excitability and action potential

© 1998 American Institute of Physics

ense or copyright; see http://chaos.aip.org/chaos/copyright.jsp

the

se

l

n

vdiethan

raep

l

eleb

e°C

oco-

dmomp

d0T

Fortive

to-n.ion

reenary

manro-har-andhemal

tzed

rstions

138 Chaos, Vol. 8, No. 1, 1998 Kim et al.

D

ration, resulting in the effective treatment of cardiac arrhymia. However, little information is available regarding thpath of spiral wave meandering. The purposes of the prestudy are as follows:~1! To determine the longevity~thenumber of rotations! of nonstationary spiral waves in atriaand ventricular myocardium, bothin vivo andin vitro. ~2! Todetermine the patterns of spiral wave meandering whespiral wave survives for.2 rotations.

II. MATERIALS AND METHODS

Although uncommon, nonstationary spiral waves habeen observed in cardiac tissue from prior mapping stuanalyzed by investigators in our lab. In order to increaseyield of detecting nonstationary spirals, we have includedmany episodes of epicardial bipolar recordings of atrial aventricular fibrillation in various animal and human prepations from prior experiments. These preparations includesodes from anin vitro canine right ventricle~RV!,15,16 in vivocanine RV,6 in vivo canine RV after subendocardiaablation,15 in vitro swine RV,17 in vivo Langendorff-perfusedhuman ventricles,18 in vivocanine right atrium~RA!,19 andinvitro canine RA with acetylcholine.20,21These episodes werrecorded using either a 317 or a 509 channel bipolar etrode array in which this plaque was carefully suturedavoiding arteries, ontoin vivo atrial or ventricular epicar-dium or placed beneathin vitro endocardial surfaces of thtissue superfused in oxygenated Tyrode’s solution at 36Care was taken not to damage the artery during the prdure. The Tyrode’s solution had the following ionic compsition in mM/L: NaCl 125, KCl 4.5, NaH2PO4 1.8, CaCl22.7, MgCl2 0.5, NaHCO3 24, and dextrose 5.5, in distilledeionized water.22 The interelectrode distance was 1.6 mand its interpolar distance was 0.5 mm. The signals frthese electrodes were registered by a computerized mapsystem~EMAP, Uniservices!.23 The electrogram was filterewith a high pass filter of 0.5 Hz and digitized at 100samples/s. The mapping system has a fixed gain of 10.


-

nt

a

esesd-i-

c-y

.e-

ing

he

sample-and-hold technique was performed sequentially.a 1 kHz sampling rate, the skew between two consecuchannels was 2ms. The maximal skew~between channels 1and 512! was 1 ms and was not corrected. The analog-digital conversions were calibrated with 18-bit resolutioTrue 16-bit conversions were achieved. The input resolutof the system was 13.7mV.

A total of 108 episodes from the above studies weselected for further analyses~Table I!. These episodes werselected because they were thought to contain nonstatioreentrant wave fronts~spiral waves! upon initial gross in-spection. Of these 108 episodes, 12 were from altered humyocardium in which they were diseased from either conary artery disease or cardiomyopathy. In 50 episodes pmacologic agents such as acetylcholine, cromakalim,procainamide were used to perturb the tissue. Some of tincluded chemically ablated subendocardium. ‘‘Normmyocardium’’ (N546) studied were fromin situ myocar-dium, and in vitro preps of excised myocardium withouchemical perturbations Table II. The data were then analyde novo according to methods published elsewhere.6 Thepoints of cell depolarization, or activation times, were fiselected by the computer that is programmed to seek reg

TABLE I. Total number of episodes evaluated. Ach5acetylcholine,crom5cromakalim, proc5procainamide.

In vivo In vitro Subtotal

Right 6 canine with Achatrium 9 canine 5 canine without Ach 26

1 swine5 human

Right 18 canine ablated 15 canineventricle 12 canine with proc 14 swine with crom 75

12 canine without proc 4 swine without crom

Left atrium 0 6 human 6

Left ventricle 0 1 human 1

Subtotal 51 57 Total 108

TABLE II. Distribution of altered and ‘‘normal’’ myocardium.

Number ofepisodes Atrium Ventricle

Nonstationaryspiral waves

Altered myocardium

Humandiseased

12 11 11 ventricle

~outward petalmeandering!

Chemicallyaltered

32 6 26 3 atrium~outward petal

meandering, 2 irregularlyconcentric!

Endocardiaablation

18 0 18 1 ventricle~inward petalmeandering!

‘‘normalmyocardium’’

In vitro 25 6 194 ventricle

~inward petal, 2 outwardpetal, drift!

In vivo 21 9 12 2 atrium~outward, irregularly

concentric!

Total number 108 32 76 11


tryn thees to yellow,

o. The red

ed

139Chaos, Vol. 8, No. 1, 1998 Kim et al.

D

FIG. 1. Tracing the tip of the spiral wave. The data were obtained from anin vitro canine right atrium superfused with acetylcholine. The reenspontaneously terminated following the fifth cycle. Panel~a! depicts a black and white version of the colored dynamic activation display as seen ocomputer screen at 20–75 ms intervals. Red dots, seen here as dark gray dots, represent the leading edge of the wave front. The color then changgreen, blue, and purple before reverting to the background color~seen in this figure as different shades of gray following the wave front!. Each color persistsfor 10 ms. The numbers on the upper edge of each picture shows the times of activation in ms, with the beginning of data acquisition as time zerdot that is nearest to the core of reentry is marked. Arrows indicate the direction of propagation of the spiral tip. Panel~b! illustrates the electrode array inpanel~a! reproduced on a graphics program. Colored arrows correspond to each numbered cycle. Panel~c! represents the approximate wave tip path derivfrom panel~b!. The outward petal flower pattern is evident in this final illustration.

ownloaded 07 Jan 2008 to 169.232.48.36. Redistribution subject to AIP license or copyright; see http://chaos.aip.org/chaos/copyright.jsp

with


Downloaded 07 Jan 2

TABLE III. Position of the core. One in vitro canine RA and one canine RV had obvious meanderingcycle lengths,2 and were excluded from further evaluation. cw5clockwise, ccw5counterclockwise,RA5right atrium, RV5right ventricle, LV5left ventricle.

Tissue Stationary Nonstationary Other

Invitro

Canine RA~5! 3 ccw 0 2 nonsustainedCanine RA w/ Ach~6! 1 figure 8 3 2 nonsustained

Swine RA ~1! 1 cw 0 0Human RA~5! 0 0 5 nonsustainedHuman RA~6! 0 0 6 nonsustainedCanine RV~15! 3 cw, 5 ccw 3 1 colliding/3 nonsustainedSwine RV ~4! 1 ccw, 1 figure 8 1 1 nonsustained

Swine RV with crom~14! 4 cw, 10 ccw 0 0Human LV ~1! 0 1 0

Canine RA~9! 0 2 4 colliding/3 nonsustainedIn Canine RV~12! 0 0 12 nonsustainedvivo Canine RV with proc~12! 1 ccw 0 11 colliding/nonsustained

Canine RV ablated~18! 1 cw, 2 ccw 1 14 colliding/nonsustained

Total 33 11 64

ioomipsorr

-1

reF

wzaion

teio

likb

-infted

esvatore

ahiwia

tatole

lyto

ryveofthld

teddesor-ithin

ve-six

alre-at-yromwere-

or

esheg-

ug-un-

of maximal slope on the electrocardiogram. The activattimes were then manually edited and displayed onto a cputer program, which shows activation patterns at each blar electrode via colored dots representing different phasethe activation. The dot initially illuminates as a red colthen proceeds to yellow, green, light blue, then finally dablue @Fig. 1~a!#. The total illumination time for each activation of one dot was manually selected to last 50 ms, thusms for each color. The 50 ms duration was selected toresent the refractory period of ventricular cells during Vwhich ranged from 48 to 77 ms.24 The refractory period wasapproximated at a constant~50 ms! and did not reflect thevariations of refractory periods in the cardiac tissue. Hoever, this shortcoming did not affect the accurate visualition of the wave front propagation on the dynamic activatdisplay.

Patterns of nonstationary spiral waves were illustraby tracing the innermost red dot on the dynamic activatdisplay during various time intervals as in Fig. 1~a!. The grayarrow represents the direction of wave propagation. Unstationary spiral waves where a fixed core could be tracedfollowing the wave tip, the position of ‘‘the core’’ in nonstationary spiral waves could only be approximated by definit as the area enclosed by the spiral wave tip tracing aeach cycle of reentrance excitation. The same method tofine the position of the core was used by Leeet al. in aprevious study.6 By this manner, the core of reentrant wavwere traced on the computer screen at 10–80 ms inter@Fig. 1~a!#. If the subsequent spiral wave tip was foundgrossly deviate from its initial wave tip path, i.e., by mothan one interelectrode distance~1.6 mm! in greater thanroughly 75% of its path, then this reentrant wave front wconsidered to be nonstationary. Then a computer grapprogram was used to code each cycle of the spiral wavea different color onto a cartoon electrode array to delinethe wave tip pathway.

The period of rotation, or the cycle length, for each stionary spiral was obtained by calculating the time it tookreturn to its initial position, i.e., to make one complete circ

008 to 169.232.48.36. Redistribution subject to AIP lic

n-

o-of

k

0p-,

--

dn

ey

gre-

ls

scsthte

-

.

However, with nonstationary spirals, the period could onbe roughly approximated since the wave tip did not returnits point of origin. The period in our sample of nonstationaspirals were determined by calculating the time for the watip to return to its angular position in relation to the center‘‘the core.’’ In this manner, although the spiral wave tip pawas displaced, its return to the initial angular position coube used to determine the period.

III. RESULTS

Table I shows the distribution of the episodes evaluain this study. Table II outlines the distribution of altereversus ‘‘normal’’ myocardium. Except for the 12 episodfrom human atria and ventricles, all tissues were from nmal healthy animals, albeit 50 episodes were perturbed wdrugs or chemical ablation. Most of the episodes studiedthe altered myocardium were from the ventricles, and fiepisodes showed clear nonstationary spiral waves. Fortyepisodes of ‘‘normal myocardium’’ in which healthy animmyocardium were studied without any perturbing factorsvealed four ventricles with obvious meandering and drift pterns, and twoin vivo atria with meandering and irregularlconcentric patterns. The human tissues were obtained fthe native hearts of transplant recipients as soon as theyexcised from the patient’s chest.18,25 These hearts were invariably diseased due to either coronary artery diseaseidiopathic dilated cardiomyopathy.

A. Absence of persistent single meandering spiralwaves

Table III shows that the vast majority of the spiral wavcould not be used for the analysis of spiral tip motion for tfollowing reasons:~1! The spiral waves did not meander sinificantly from one place to another (N533). ~2! The spiralwaves were nonsustained~,2 rotations! (N564). Althoughthe entire epicardium was not analyzed, these findings sgest that persistent single meandering spiral wave is ancommon occurrence in the present study.


thex

on

om

le,inger.

d tom

lesingter

ycle

att--

piralpo-hoth

hrduldheng

not

ourf

tht aorr


D

B. The direction of spiral tip motion

Only 11 episodes were nonstationary, had a life span.2 cycles, and remained in the mapped region duringentire period from the beginning to end of the reentrantcitation to be included in this part of the analyses~Table III!.By tracing the spiral wave tip, various patterns of nonstatiary spiral waves have been observed. Figure 1~a! illustratesthe actual dynamic activation display as seen on the c

FIG. 2. Meandering flower patterns resulting from the permutation ofdirection of movement of the spiral wave tip and ‘‘the core.’’ Note thaflower pattern with inward petals result when the spiral wave tip and cmovement directions are the same~a,d!. On the other hand, when theidirections are opposite, a flower with outward petals result~b,c!.


ofe-

-

-

puter screen with the wave tip tracing. Unlike this exampother wave tip tracings that were more closely overlappeach other, were difficult to distinguish from one anothThe electrode array underlying Fig. 1~a! is represented usinga computer graphics program@Fig. 1~b!#. Unfilled circles inthe background represent bipolar electrodes that were useplot the nearest location of the spiral wave tip derived frothe computer display. After connecting these filled circwith a straight line, its path was grossly approximated usvarious colored, curved lines generated by the compugraphics program. Each colored arrow represents one c@Fig. 1~c!#.

C. Patterns of flowers with inward and outward petals

The combination of the direction of the wave tip and thof the core yields either flowers with inward petals or ouward petals~Fig. 2!. Flowers with inward petals can be generated by a core that rotates in the same direction as its swave tip. On the other hand, a core that rotates in the opsite direction as that of the wave tip will yield a flower witoutward petals. For example, a wave tip and core where brotate clockwise~CW! or counterclockwise~CCW! willyield an inward petal flower. A wave tip and core in whicone rotates CW and the other CCW will yield an outwapetal flower. In most cases, a clear cut flower pattern conot be illustrated as in the theorized depiction in Fig. 2. Tshort life spans of the spiral waves was a major limitifactor, however atendencytoward an inward or outwardpattern could be elucidated when the spiral wave tip diddrift or waver about a common center~as described below!.Seven episodes with such tendencies were uncovered insample. Inward petal flowers were found in one episode oin

e

e

ajor fiber
FIG. 3. Flower patterns with a tendency toward inward petals. The black arrow on the upper left-hand corner represents the direction of morientation. The long axis of the core is parallel to the fiber orientation in these meandering spirals in anin vitro canine right ventricle@Panel~a!# and in anin vivo canine right ventricle@panel~b!#. The wave front spontaneously terminated in both episodes.

n

.


D

FIG. 4. Flower patterns with a tendency toward outward petals are illustrated in these panels, which represent activation wave fronts from ain vitromyopathic human left ventricle@panel~a!#, an in vitro canine right ventricle@Panels~b! and ~c!#, an in vivo canine right atrium@Panel~d!#. In panel~b!, abreakthrough wave~four red dots! displaced the spiral wave front upward during cycle #4. The spiral wave in panel~d! meandered off the top of the plaque

rd

so

or

byipthee

vitro canine RV@Fig. 3~a!# and another in ablatedin vivocanine RV@Fig. 3~b!#. Five episodes with a tendency towaoutward petal flowers were detected inin vitro canine RAwith acetylcholine~Fig. 1!, in vitro canine RV@Figs. 4~b!,and 4~c!#, in vitro myopathic human LV@Fig. 4~a!#, and invivo canine RA@Fig. 4~d!#. Of note, every combination waobserved except for the core and the spiral wave tip brotating in the CCW direction~Table IV!.


th

D. Drifting or irregular patterns

Not all episodes could be classified into either inwardoutward flower petals. One episode ofin vitro swine RVdemonstrated a drift pattern, as previously describedPertsovet al.7 Figure 5 illustrates the CCW rotating wave tas the core moves in a linear fashion perpendicular todirection of fiber orientation, i.e., the longitudinal axis of th


otal


Downloaded 07 Jan 2

TABLE IV. Meandering episodes (N511). Note the variations in cycle lengths due to meandering. Tobvious meandering episodes with period~cycle length! .2511.

Type of tissueNo. ofcycles

Period~ms!

Directionof wave tip

Directionof core

Nonstationarypattern

4.5 115, 120, 150150

ccw cw Outwardpetal

In vitrocanine RAwith Ach

7.5 140, 190, 145135, 180, 115

135

cw Erratic Irregularlyconcentric

5.5 145, 135, 165165, 160

cw Erratic Irregularlyconcentric

2.5 120, 115 cw cw Inwardpetal

In vitrocanine RV

5 140, 100, 120335, 160

cw ccwa Displacedoutward

petal3.5 130, 140, 130 cw ccw Outward

petal

In vitroswine RV

7 115, 130, 115120, 100, 110

65

ccw Linear Drift

Myopathichuman LV

3 120, 89, 80 ccw cw Outwardpetal

In vivo3 70, 115, 115 cw ccw Outward

petalcanine RA 5 120, 130, 135

150, 130cw Erratic Irregularly

concentric

In vivocanine RVablated

4 105, 115, 95115

cw cw Inwardpetal

aWave tip displaced by a breakthrough wave in anin vitro canine RV@Fig. 4~b!#.

um

gRce

d

on-icu-

l

myocytes. The direction of fiber orientation for each tisspreparation were documented from prior histological exanation of the tissue.

Last, a CW rotating wave tip with its core waverinerratically about a common center was observed only insamples. Such a pattern is described as irregularly contric. This pattern was observed in twoin vitro canine RAwith acetylcholine@Figs. 6~a!, and 6~b!# and in onein vivocanine RA @Fig. 6~c!#. All three episodes were associatewith a CW rotating wave tip.

TABLE V. Patterns of meandering.

Nonstationarypatterns Tissue type Number Tota

Flower with In vitro canine RV 1inward petals In vivo canine RV 1 2

ablated

In vitro canine RA 1Flower with with Ach

outward petals In vitro canine RV 2 5In vivo canine RA 1

Myopathic human LV 1Drift In vitro swine RV 1 1

Irregularly In vitro canine RA 2concentric with Ach 3

In vivo canine RA 1

008 to 169.232.48.36. Redistribution subject to AIP lic

ei-

An-

FIG. 5. Drift pattern with the long axis of the core parallel to the directiof fiber orientation in anin vitro swine left ventricle. The wave front meanders off to the bottom right. The direction of meandering was perpendlar to the major fiber orientation.


regularly


D

FIG. 6. This figure shows clockwise rotating wave tips with their cores wavering erratically about an approximate common center, termed irconcentric meandering. This pattern was observed in twoin vitro canine right atrium superfused with acetylcholine@panels~a! and ~b!# and in onein vivocanine right atrium@panel~c!#. The wave front propagates off the plaque in~a! and ~b!, and spontaneously terminated in~c!.

r

tbe

nycle

ife

E. Effects of fiber orientation

The direction of fiber orientation in the ventriculasamples have been marked along its respective [email protected]~a!, 3~b!, 4~a!, 4~b!, 4~c!, and 5#. In all of these samples, iis evident that the long axis of the core lies along the fidirection.


r

F. The cycle length and life span of the spiral wavesanalyzed

The cycle length~period! was variable and the life spaof the meandering spiral waves were short. The mean clengths were calculated from the approximated period~TableIV !. The mean period varied from 96 to 154 ms. The l


.5

ep

le

at

ira

a

tvei

ithng

atero

nesnsonoab

ou

siut

nd-a

ethorettnnrab

ctr

aroa

-

ace-

ring

lsoicaln-

msedi-25arte IId in

al-ion-ur-re,

forcu-

Gla-y ain-toce

re-yrr oftheirals

alrayro-as

u-

iv-ets,


D

span of the meandering spiral waves ranged between 27.5 cycles, with a mean of 4.664.3 and a median of 4.5cycles. Spontaneous termination was observed in sixsodes@Figs. 1, 3~a!, 3~b!, 4~a!, 4~c!, and 6~c!#. In four epi-sodes the meandering spiral wave propagated off the etrode plaque@Figs. 4~d!, 5, 6~a!, and 6~b!#. Lastly, Fig. 4~b!illustrates an episode ofin vitro canine RV in which thepropagating wave front becomes displaced superiorly byother wave front. This interaction prolongs the cycle lengduring its fourth cycle. In this episode, the meandering spwave subsequently becomes stationary.

IV. DISCUSSION

There are two major findings in this study. One is ththe functional reentrant wave fronts~spiral waves! in cardiactissue are, by and large, short-lived. The second is thatpatterns of spiral tip motion are not uniform. The spiral watip tracings revealed nonstationary spiral wave patterns wtendencies toward flowers with inward petals, flowers woutward petals, and drifting more or less linearly, or movierratically about a common center.~See Table V.!

A. Effects of subendocardial ablation and/orpharmacological treatment

Fifty episodes in our sample represent tissues that hhad subendocardial ablation with Lugol solution, or treapharmacologically with acetylcholine, procainamide, or cmakalim for various other experiments~Table I!. Leeet al.6

have shown that the generation and maintenance of reedo not depend on intact subendocardium, as their expments have shown that subendocardial ablation had nonificant effects on incidence, life span, cycle length, amean core size of reentrant waves. Thus, it would be reaable to conclude that meandering spiral waves wouldhave been suppressed in some way by subendocardialtion.

Samples treated with procainamide showed no obvimeandering spiral waves. Studies by Starmeret al.13 sug-gested that reducing repolarizing currents such as potasconductance destabilize the unexcited core by increasingwavelength~the product of the action potential duration athe velocity of propagation! above a critical value. The results are a nonstationary spiral and a polymorphic ECGthose hypothesized initially by Zykov2 and elaborated on byWinfree.12 Their group hypothesized that sodium channblockade reduces cellular excitability, thereby increasingsize of the core and reducing the wavelength. A spiral cdrift can occur when the wavelength exceeds the perimof the unexcited core and rendering it unstable. In contrasreducing potassium channel conductance, sodium chablockade, as with procainamide, will yield a stationary spiwave or a monomorphic ECG. Our results and a studyBuxtonet al.26 in which patients with inducible polymorphitachyarrhythmias at the baseline had only monomorphicchycardia inducible after procainamide, support Starmehypothesis.

The effects of cromakalim on reentrant wave frontsvariable. As shown in this study, it results in stabilizationthe rotor ~no meandering!. Cromakalim could also cause


to

i-

c-

n-hl

t

he

th

ved-

tryri-ig-dn-tla-

s

mhe

s

leeertoelly

a-’s

ef

breakdown of the spiral wave,27 thereby converting ventricular tachycardia to fibrillation.28 In the latter situation, a briefperiod of meandering was observed. On the other hand,tylcholine, from our previous publication21 and from thepresent study has shown a tendency to induce meande~three of six episodes inin vitro canine RA, Table I!. How-ever, acetylcholine was used only in atrial tissues, which ahave a complicated endocardial structure. The anatomcomplexity of the atrial endocardium could be an indepedent factor that influences spiral wave activity.10,25,29,30

B. The prevalence of nonstationary spiral waves inaltered versus ‘‘normal’’ myocardium

Although 96 out of 108 episodes studied were frohealthy animal myocardium, 50 episodes, as discusabove, were not under ‘‘normal’’ conditions. Forty-six epsodes were considered ‘‘normal’’ myocardium, althoughepisodes were recordings from in vitro studies in which pof the myocardium may have been poorly perfused. Tablshows that five nonstationary spiral waves were observealtered myocardium, not including thein vitro, otherwisenormal myocardium. If these episodes were included astered myocardium, then the number of observed nonstatary spiral waves would be even lower, namely 2, and fthermore, they were only observed in atrial tissue. Therefoif the in vitro preparations were excluded, the argumentthe scarcity of nonstationary spiral waves in normal ventrilar myocardium would be made stronger.

C. The life span of the spiral waves

Grayet al.10,11has demonstrated that characteristic ECpatterns of polymorphic ventricular tachycardia and fibriltion can be demonstrated in the isolated rabbit heart bsingle nonstationary vortex-like reentrant activity. To matain the arrhythmia, this reentrant activity must be ablesustain itself for the duration of the arrhythmia and produirregular activity through Doppler effects.11 In normalhealthy ventricles, however, the organized reentry is infquently observed during VF.6 Whether or not nonstationarreentry~vortex! is a major mechanism of polymorphic VT oVF is unclear. Further experiments with a greater numbesamples will have to be studied to increase the power ofstudy and to ascertain the prevalence of nonstationary spwith greater confidence.

We have been studying the lifespan of the functionreentrant activity in models different than that used by Get al. Their group used an isolated rabbit’s heart whose ptocol involved cooling the heart to a temperature as low30 °C in order to induce VF.11 In the intact normal canineright ventricle, reentrant activity induced by electrical stimlation only persists for a short time. In one study,31 the reen-try persisted for an average of only 1.36 s~range 0.15–2.75s! lasting 9.6 cycles~range 2–16 cycles!. In another studyusing the same animal species,24 the reentrant activity termi-nated an average of 3.261.9 rotations after induction by anelectric shock. During Wiggers’ stage II VF, reentrant actity was spontaneously generated by the collision of wavelthen terminated after an average of 3.461.1 rotations.6 The


tto-

dv

assusureds

re

ulaaras

ra

iF

glbeamte

ha

eres-it

ine’trinr

e

llyuiarncosu

notto

y to-by

erriftherlsol to

at-ies.anlesinis

lec-es-

m-o-

iac

x-icshermu-sisand

ural

ralso-is-es

pi-

iacm-aveob-of

oreth-

vi-ple

eh a


D

results of the present study seems to further strengthenargument that a single meandering reentrant wave fronunlikely to be a common mechanism for VF in normal mycardium.

In contrast to the studies performed by Grayet al.,10,11

where the entire epicardial surface was mapped, our stumapped only a portion of the ventricles. Reentrant wafronts that meandered outside of the roughly 334 cm2

mapped region could not be studied. Moreover, Grayet al.used an isolated perfused heart where perfusion was mtained during VF, whereas our samples were only superfuin Tyrode’s solution. It is possible that in certain instancesingle meandering functional reentrant wave front could svive long enough to cause a persistent tachycardia andsequently result in ischemia that could facilitate the genetion and maintenance of VF. However, in a well-perfusand nonischemic preparation,32 nonstationary spiral waveduring VF in isolated swine RV tissue is also short-lived.

A second difference is that normal myocardium weprimarily studied, as well as in several previous studies.6,24,31

The situation associated with abnormal myocardium cobe entirely different. For example, reentrant wave fronts manchor around the infarcted tissue in patients with coronartery disease, resulting in a longer life span of the reentexcitation.33 An anatomical obstacle created by punch biopcould also serve as the anchoring point of the reentactivity.34 If the hole is small enough (,6 mm), the reen-trant waves frequently detached from the hole, resultingirregular electrograms. Most importantly, in a chronic Amodel created by rapid pacing, only a long-lasting sinmeandering reentrant wave front was detectablemapping.35 It is therefore possible that, although single mandering reentrant wave fronts are difficult to find in normmyocardium, they may be found in abnormal myocardiuand contribute significantly to the generation and mainnance of VF.

D. The direction of meandering

Predictions of spiral wave behavior in cardiac tissuebeen made using mathematical models.2,36,37Winfree, usinga mathematical model, derives various meandering pattdependent on the excitation threshold and ratio of recovto the excitation rate.12 Two patterns of meandering are posible, namely, flowers with outward petals and those winward petals. The former have been mostly observednoncardiac medium. It is interesting to note that in Winfrediscussion of spiral meandering,12 one can conclude thadrifting could be considered another pattern of meandein which the core follows a virtually linear path in a flowethat is infinitesimally large.

In this study, the patterns of meandering spiral wavwere analyzed to investigate a possible uniform patterncardiac tissue as skeptically speculated by Winfree.38 ~com-munications!. As outward petal flower patterns generahave been observed in noncardiac excitable media, it wobe interesting to note if such a consistency exists in cardtissue. Although various meandering spiral wave pattehave been observed and characterized in this study, nosistent patterns could be derived from the myriad tis


heis

iese

in-edar-b-

a-

dyryntynt

n

ey-l

-

s

nsry

ha

s

g

sin

ldcsn-e

preparations. Because a complete flower pattern isformed due to its short life span, it would be reasonablestate that the flower patterns observed have a tendencward the cycloid meandering patterns describedWinfree.12

Drifting was observed in one sample ofin vitro swineLV ~Fig. 5!. As described previously by Pertsovet al.,7,39

The long axis of the core was along the direction of fiborientation and the horizontal component of spiral wave dwas along the direction of the spiral wave front. Althougthe direction of drift was noted to be perpendicular to fiborientation in a sample of Pertsov’s experiment, as is atrue in our sample, the number of episodes is too smalconclude that this is true in all cases.

Irregularly concentric meandering spiral waves are pterns that were not predicted by computer simulation studThe spiral wave tip can definitely be seen to hover aroundapproximate common center. Despite only three sampshowing this pattern, it is interesting that this is only seenatrial samples and not in ventricles. Whether or not thisdue to the differences in anisotropy, anatomy, or other etrophysiological properties of atria and ventricles is a qution that needs to further investigated.

E. Limitations of the study

As in all epicardial mapping techniques to date, the coputerized dynamic activation display is only a twdimensional depiction~spiral wave! of an activation patternthat is occurring in a three-dimensional structure, i.e., cardtissue. A three-dimensional propagating wave front40,41 suchas a scroll wave8,36 may be one mechanism that could eplain such an activation pattern; however, the true dynamof reentrant wave fronts in cardiac tissue need to be furtelucidated and, for now, can only be speculated and silated on computer programs. A three-dimensional analycould explain such phenomena as breakthrough wavesspontaneous terminations by demonstrating transmpropagation.

Only grossly obvious patterns of nonstationary spiwaves were considered in our episodes, limited by the relution of the epicardial mapping technique and computer dplay. Improved resolution of cardiac mapping techniquwill help to further elucidate the extent of nonstationary sral waves and their role in cardiac tachyarrhythmias.

Unlike numerical systems depicted by Winfree, cardtissue is a complicated excitable medium with such coplexities as ionic currents and tissue anisotropy. We hattempted to incorporate a theory to actual experimentalservations that will aid in enhancing our understandingsuch complex excitable media and possibly making maccurate predictions of spiral wave behavior using maematical models in the future.

V. CONCLUSION

Although few in number, various patterns of grossly edent nonstationary spiral waves were observed in our samof 108 recordings of cardiac fibrillation. Only 10% of thsamples revealed single nonstationary spiral waves wit


ad

raglisn-

ina.Iolap-

b-

ar

fananrt,e

in

7

life,’’

lifeac

ara

Cras,R

Jed

Ni

lar,

kohart,’

v-ril

’s

binm

s,’’

theJ.

a-ve-c-

.tiontryre-

.illa-

A.hen,ularni-

.-S.n-y,’’

S.ntry

el,ontsol.

el,rmi-for

S.rical,’’

-S.lar

C.tiontis-

r,la-

en,

n

, S.a,ril-

rg,dial

tion

nde

.le

.


D

median life span of.2 cycles. Characteristics of these spirwaves have been demonstrated using a computerizednamic activation display of various cardiac tissue prepations. However, the paucity and short life span of a sinnonstationary spiral wave in our samples suggest that thunlikely to be a major cause of fibrillation in normal vetricular myocardium.

ACKNOWLEDGMENTS

This work was done during the tenure of a Cedars-SBurns and Allen Research Institute Fellowship Grant to Ta Research Fellowship Grant from the Division of Cardiogy, College of Medicine, Korea University to Y.-H.K.,National Institutes of Health Clinician Scientist Develoment Award to J.J.C.O.~James J. C. Ong!, an ECHO Foun-dation Award to H.S.K., and an AHA/Wyeth-Ayerst Estalished Investigatorship Award to P.-S.C.~93002670!, andwas supported in part by a Specialized Center of Rese~SCOR! Grant in Sudden Death~P50HL 52319!, and aFIRST Award ~HL 50259! from the National Institutes oHealth, the Electrocardiographic Heartbeat Organization,the Ralph M. Parsons Foundation. The authors wish to thDr. Arthur T. Winfree for his invaluable input and suppoAvile McCullen and Meiling Yuan for technical assistancand Elaine Lebowitz for secretarial assistance.

1A. T. Winfree, ‘‘Spiral waves of chemical activity,’’ Science175, 634~1972!.

2V. S. Zykov, ‘‘Simulation of the process of circulation of an excitationheart muscle,’’ inSimulation of Wave Processes in Excitable Media, ed-ited by A. T. Winfree~Manchester University Press, Manchester, 198!,pp. 93–112.

3J. M. Davidenko, A. V. Persow, R. Salomonza, W. Baxter, and J. Ja‘‘Sustained vortex-like waves in normal isolated ventricular muscleProc. Natl. Acad. Sci. USA355, 349 ~1990!.

4J. M. Davidenko, A. M. Pertsov, R. Salomonsz, W. Baxter, and J. Ja‘‘Stationary and drifting spiral waves of excitation in isolated carditissue,’’ Nature~London! 355, 349 ~1991!.

5C. F. Starmer, A. A. Lastra, V. V. Nesterenko, and A. O. Grant, ‘‘Prorhythmic response to soduim channel blockade. Theoretical modelnumerical experiments,’’ Circulation84, 1364~1991!.

6J. J. Lee, K. Kamjoo, D. Hough, C. Hwang, W. Fan, M. C. Fishbein,Bonometti, T. Ikeda, H. S. Karagueuzian, and P.-S. Chen, ‘‘Reentwave fronts in Wiggers’ stage II ventricular fibrillation: Characteristicand mechanisms of termination and spontaneous regeneration,’’ Circ.78, 660 ~1996!.

7A. M. Pertsov, J. M. Davidenko, R. Salomonsz, W. T. Baxter, andJalife, ‘‘Spiral waves of excitation underlie reentrant activity in isolatcardiac muscle,’’ Circ. Res.72, 631 ~1993!.

8C. Hwang, H. S. Karagueuzian, M. D. Fishbein, L. Czer, A. Trento, J.Weiss, and P.-S. Chen, ‘‘Demonstration of spiral waves of excitationhuman ventricular tissue,’’ Circulation90, I-466 ~1994! ~abstract!.

9A. T. Winfree, ‘‘How does ventricular tachycardia decay into ventricufibrillation?,’’ in Cardiac Mapping, edited by M. Shenasa, M. Borggrefeand G. Breithardt~Futura Publishing Co., Mount Kisco, NY, 1993!, pp.655–682.

10R. A. Gray, J. Jalife, A. Panfilov, W. T. Baxter, C. Cabo, J. M. Davidenand A. M. Pertsov, ‘‘Nonstationary vortexlike reentrant activity as mecnism of polymorphic ventricular tachycardia in the isolated rabbit heaCirculation91, 2454~1995!.

11R. A. Gray, J. Jalife, A. V. Panfilov, W. T. Baxter, C. Cabo, J. M. Daidenko, A. M. Pertsov, and P. Hogeweg, ‘‘Mechanisms of cardiac fiblation,’’ Science270, 1222~1995!.

12A. T. Winfree, ‘‘Varieties of spiral wave behavior: An experimentalistapproach to the theory of excitable media,’’ Chaos1, 303 ~1991!.

13C. F. Starmer, D. N. Romashko, R. S. Reddy, Y. I. Zilberter, J. StaroA. O. Grant, and V. I. Krinsky, ‘‘Proarrhythmic response to potassiu


ly--

eis

i.,-

ch

dk

,

,

,

-nd

.nt

es.

.

.n

,-’

-

,

channel blockade. Numerical studies of polymorphic tachyarrhythmiaCirculation92, 595 ~1995!.

14C. F. Starmer and J. Starobin, ‘‘Spiral tip movement: The role ofaction potential wavelength in polymorphic cardiac arrhythmias,’’ Int.Chaos Bifurc.~in Press!.

15Y. Y. Kwan, W. Fan, D. Hough, J. J. Lee, M. C. Fishbein, H. S. Kargueuzian, and P.-S. Chen, ‘‘Procainamide induced prevention of wafront breakup—A novel mechanism of antiarrhythmic drug action,’’ Paing Cardiac Electrophysiol.20, 1121~1997!.

16K. Kamjoo, T. Uchida, T. Ikeda, M. C. Fishbein, A. Garfinkel, J. NWeiss, H. S. Karagueuzian, and P.-S. Chen, ‘‘The importance of locaand timing of electrical stimuli in terminating sustained functional reenin isolated swine ventricular tissues—Evidence in support of a smallentrant circuit,’’ Circulation~in press!.

17Y.-H. Kim, T. Ikeda, T.-J. Wu, C. A. Athill, A. Garfinkel, J. N. Weiss, HS. Karagueuzian, and P.-S. Chen, ‘‘Route to chaos in ventricular fibrtion revealed by tissue size reduction,’’ J. Am. Coll. Cardiol.29, 2A~1997! ~abstract!.

18T.-J. Wu, J. J. C. Ong, C. Hwang, J. J. Lee, M. C. Fishbein, L. Czer,Trento, C. Blanche, W. J. Mandel, H. S. Karagueuzian, and P.-S. C‘‘Generation and maintenance of reentrant wave fronts during ventricfibrillation in human hearts with dilated cardiomyopathy: Role of nonuform anisotropy,’’ J. Am. Coll. Cardiol.29, 4A ~1997! ~abstract!.

19J. J. C. Ong, J. J. Lee, M. C. Fishbein, H. S. Karagueuzian, and PChen, ‘‘Initiation and maintenance of atrial fibrillation by focal and reetrant mechanisms: The role of complex 3-dimensional atrial anatomCirculation95, I351 ~1996! ~abstract!.

20C. A. Athill, T. Ikeda, Y.-H. Kim, T.-J. Wu, H. S. Karagueuzian, and P.-Chen, ‘‘Transmembrane potential properties of cells at the core of reein isolated canine right atria,’’ J. Am. Coll. Cardiol.29, 255A ~1997!~abstract!.

21T. Ikeda, T.-J. Wu, T. Uchida, D. Hough, M. C. Fishbein, W. J. MandP.-S. Chen, and H. S. Karagueuzian, ‘‘Unstable reentrant wave frinduced by acetylcholine in isolated canine right atrium,’’ Am. J. Physi~in press!.

22T. Ikeda, T. Uchida, D. Hough, J. J. Lee, M. C. Fishbein, W. J. MandP.-S. Chen, and H. S. Karagueuzian, ‘‘Mechanism of spontaneous tenation of functional reentry in isolated canine right atrium: Evidencethe presence of an excitable but nonexcited core,’’ Circulation94, 1962~1996!.

23C. Bonometti, C. Hwang, D. Hough, J. J. Lee, M. C. Fishbein, H.Karagueuzian, and P.-S. Chen, ‘‘Interaction between strong electstimulation and reentrant wavefronts in canine ventricular fibrillationCirc. Res.77, 407 ~1995!.

24Y.-M. Cha, U. Birgersdotter-Green, P. L. Wolf, B. B. Peters, and P.Chen, ‘‘The mechanisms of termination of reentrant activity in ventricufibrillation,’’ Circ. Res.74, 495 ~1994!.

25T. Ikeda, L. Czer, A. Trento, C. Hwang, J. J. C. Ong, D. Hough, M.Fishbein, W. J. Mandel, H. S. Karagueuzian, and P.-S. Chen, ‘‘Inducof meandering functional reentrant wavefront in isolated human atrialsues,’’ Circulation~in press!.

26A. E. Buxton, M. E. Josephson, F. E. Marchlinski, and J. M. Mille‘‘Polymorphic ventricular tachycardia induced by programmed stimution: Response to procainamide,’’ J. Am. Coll. Cardiol.21, 90 ~1993!.

27H. S. Karagueuzian, M. Gotoh, T. Uchida, W. J. Mandel, and P.-S. Ch‘‘Transition from single to multiple functional reentrant wave frontsinvitro induced by Cromakalim, a K-ATP channel activator,’’ Circulatio90, I-40 ~1994!.

28A. Garfinkel, P.-S. Chen, D. O. Walter, H. S. Karagueuzian, B. KoganJ. Evans, M. Karpoukhin, C. Hwang, T. Uchida, M. Gotoh, O. NwasokwP. Sager, and J. N. Weiss, ‘‘Quasiperiodicity and chaos in cardiac fiblation,’’ J. Clin. Invest.99, 305 ~1997!.

29R. B. Schuessler, T. Kawamoto, D. E. Hand, M. Mitsuno, B. I. BrombeJ. L. Cox, and J. P. Boineau, ‘‘Simultaneous epicardial and endocaractivation sequence mapping in the isolated canine atrium,’’ Circula88, 250 ~1993!.

30M. S. Spach, W. T. Miller, P. C. Dolber, J. M. Kootsey, J. R. Sommer, aC. E. Mosher, ‘‘The functional role of structural complexities in thpropagation of depolarization in the atrium of the dog,’’ Circ. Res.50, 175~1982!.

31P.-S. Chen, P. Wolf, E. G. Dixon, N. D. Danieley, D. W. Frazier, W. MSmith, and R. E. Ideker, ‘‘Mechanism of ventricular vulnerability to singpremature stimuli in open chest dogs,’’ Circ. Res.62, 1191~1988!.

32Y.-H. Kim, A. Garfinkel, T. Ikeda, T.-J. Wu, C. A. Athill, J. N. Weiss, H


eninhy

. Ti

c

iaa

ol.

gsi

.

al

al

ddrc.

-le,’’


D

S. Karagueuzian, and P.-S. Chen, ‘‘Spatiotemporal complexity of vtricular fibrillation revealed by tissue mass reduction in isolated swright ventricle: Further evidence for the quasiperiodic route to chaospothesis,’’ J. Clin. Invest.~in press!.

33C. Hwang, D. J. Martin, J. S. Goodman, E. S. Gang, W. J. Mandel, CPeter, and P.-S. Chen, ‘‘Demonstration of reentrant wavefrontsLangendorff-perfused fibrillating human ventricle,’’ Pacing Cardiac Eletrophysiol.19, 595 ~1996! ~abstract!.

34T. Ikeda, T. Uchida, W. J. Mandel, P.-S. Chen, and H. S. Karagueuz‘‘Conversion of non-stationary to stationary reentrant wave front withcritically-sized anatomical obstacle in the atrium,’’ J. Am. Coll. Cardi27, 60A ~1996! ~abstract!.

35J. J. C. Ong, T.-J. Wu, H. S. Karagueuzian, and P.-S. Chen, ‘‘Sinmeandering spiral wave of excitation underlies acute and chronic incanine atrial fibrillation,’’ Pacing Cardiac Electrophysiol.20-II , 1081~1997!.


-e-

.n-

n,

letu

36A. T. Winfree, ‘‘Rotors in normal ventricular myocardium,’’ Proc. KNed. Akad. Wet.93, 513 ~1990!.

37A. M. Pertsov, E. A. Ermakova, and A. V. Panfilov, ‘‘Rotating spirwaves in a modified Fitzhugh–Nagumo model,’’ Physica D14, 117~1984!.

38A. T. Winfree ~personal communication!.39A. M. Pertsov and Ye. A. Yermakova, ‘‘Mechanism of drift of a helic

wave in an inhomogeneour medium,’’ Biophys. J.33, 364 ~1988!.40P.-S. Chen, P. D. Wolf, S. D. Melnick, N. D. Danieley, W. M. Smith, an

R. E. Ideker, ‘‘Comparison of activation during ventricular fibrillation anfollowing unsuccessful defibrillation shocks in open chest dogs,’’ CiRes.66, 1544~1990!.

41W. Baxter, A. M. Pertsov, O. Berenfeld, S. Mironov, and J. Jalife, ‘‘Demonstration of three-dimensional reentry in isolated sheep right ventricPacing Cardiac Electrophysiol.20, 1080~1997! ~abstract!.


Date post:	21-Nov-2023
Category:	Documents
Upload:	csmc
View:	0 times
Download:	0 times

Patterns of spiral tip motion in cardiac tissues

Documents