the preexcitation index: an aid in determining the localizing

DIAGNOSTIC METHODSPREEXCITATION SYNDROME

The preexcitation index: an aid in determining themechanism of supraventricular tachycardia andlocalizing accessory pathwaysWILLIAM M. MILES, M.D., RAYMOND YEE, M.D., GEORGE J. KLEIN, M.D., DOUGLAS P. ZIPES, M.D.,AND ERiC N. PRYSTOWSKY, M.D.

ABSTRACT In this study we sought to determine whether characteristics of ventricular-inducedatrial preexcitation during reciprocating tachycardia could help differentiate atrioventricular (AV)nodal reentry from orthodromic AV reentry using an accessory pathway and to identify the site ofaccessory pathways in patients with Wolff-Parkinson-White syndrome. Fifty-five patients with ortho-dromic AV reciprocating tachycardia and 22 patients with AV nodal reentrant tachycardia were studiedwith standard electrophysiologic techniques. There were 24 left free wall, 23 posterior septal, seven

anterior septal, and one right free wall accessory pathways. Progressively premature right ventricularcomplexes (V2) were introduced during reciprocating tachycardia (V1V1). The VIV, interval duringtachycardia minus the longest VIV2 at which atrial preexcitation occurred defined a preexcitation index(PI). Atrial preexcitation occurred in 49 of 55 (89%) patients with AV reentry compared with only threeof22 (14%) patients with AV nodal reentry (p < .001). In the three patients with AV nodal reentry whodemonstrated atrial preexcitation, the PI was distinct from that of the septal pathways and was in theupper range of values for left free wall pathways. The percentage of tachycardias demonstrating atrialpreexcitation was not different between the free wall and septal pathways, but His bundle activationwas visible at the time of atrial preexcitation in only six of 17 (35%) left free wall compared with 13 of16 (81%) posterior septal and seven of seven (100%) anterior septal pathways (p < .05 free wall vs

posterior or anterior septal). Mean PI was 88, 38, and 17 msec for left free wall, posterior septal, andanterior septal pathways, respectively; all values were significantly different from each other. PIdifferentiated septal from left free wall pathways in 37 of 48 (77%) patients; a PI of 75 msec or greateroccurred only with left free wall pathways and a PI under 45 msec only with septal pathways. Thuscharacteristics of ventricular-induced atrial preexcitation during reciprocating tachycardia help differ-entiate between AV nodal reentry and AV reentry and between left free wall and septal accessorypathways.Circulation 74, No. 3, 493-500, 1986.

OBJECTIVES of electrophysiologic study in patientswith supraventricular tachycardia include determina-tion of the tachycardia mechanism and localization ofthe accessory pathway if present, especially if surgicalor catheter ablative therapy is contemplated.lA Twocommon mechanisms of paroxysmal supraventriculartachycardia are atrioventricular (AV) nodal reentryand orthodromic AV reentry using an accessory path-

From the Krannert Institute of Cardiology, Department of Medicine,Indiana University School of Medicine, and the Richard L. RoudebushVeterans Administration Medical Center, Indianapolis; and UniversityHospital, London, Ontario, Canada.

Supported in part by the Herman C. Krannert Fund, by grants HL-06308 and HL-07182 from the NHLBI, by the American Heart Associ-ation, Indiana Affiliate, by the Veterans Administration, and by theOntario Heart Association, Toronto.

Address for correspondence: Eric N. Prystowsky, M.D., Duke Uni-versity Medical Center, Box 3816, Durham, NC 27710.

Vol. 74, No. 3, September 1986

way for retrograde conduction. Electrophysiologic cri-teria to differentiate these two tachycardia mechanismsinclude a minimum ventriculoatrial interval duringtachycardia of less than 61 msec or a ventricular tohigh right atrial (V-HRA) interval less than 95 msec,5 aretrograde atrial activation sequence during reciprocat-ing tachycardia,6,7 lengthening of the ventriculoatrialinterval or tachycardia cycle length with functionalbundle branch block,'11 and the ability to shorten theatrial cycle length during tachycardia (atrial preexcita-tion) by introducing a premature ventricular impulse ata time when the His bundle is known to be refrac-tory. 12-14 However, all of these criteria have limitationsand often are not present at all. For example, function-al bundle branch block, if it occurs, is useful onlywhen it is ipsilateral to the site of the accessory path-

493

by guest on April 6, 2018

http://circ.ahajournals.org/D

ownloaded from

http://circ.ahajournals.org/

MILES et al.

way, and a normal retrograde atrial activation se-quence does not distinguish between AV nodal reentryand AV reentry using a septal accessory pathway.

Preexcitation of the atria during tachycardia by pre-mature ventricular complexes (ventricular-inducedatrial preexcitation) at a time when anterograde Hisbundle activation is identified demonstrates that anaccessory pathway capable of retrograde conduction ispresent, but a more generalized use of this technique todistinguish AV from AV nodal reentry or to localizeaccessory pathways has not been studied in detail.Since ventricular-induced atrial preexcitation duringright ventricular pacing should be more difficult toobtain in AV nodal reentry and AV reentry using leftfree wall as opposed to septal accessory pathways,14we tested the hypothesis that characteristics of ventric-ular-induced atrial preexcitation could differentiateAV from AV nodal reentry and could localize theposition of an accessory pathway.

MethodsFifty-five patients with Wolff-Parkinson-White syndrome

and orthodromic reciprocating tachycardia and 22 patients withAV nodal reentrant tachycardia underwent electrophysiologicstudy in the postabsorptive, nonsedated, drug-free state afterwritten informed consent had been obtained. Multipolar elec-trode catheters were introduced percutaneously into femoral,antecubital, and/or subclavian veins and positioned in the highright atrium, right ventricle, across the tricuspid valve in theregion of the His bundle, and the coronary sinus. Pacing wasperformed with 2 msec rectangular stimuli at twice late diastolicthreshold with a custom-built programmable stimulator. Intra-cardiac recordings filtered at 30 to 500 Hz and standard electro-cardiographic leads I, II, III, and V, filtered at 0. 1 to 20 Hz weredisplayed simultaneously on a multichannel oscilloscope andrecorded at paper speeds of 50 to 150 mm/sec. Measurementswere made either with hand-held calipers or an interactive com-puter program. 15

Patients with AV reentrant tachycardia had a V-HRA intervalgreater than 95 msec and satisfied at least one of the threefollowing criteria: (1) eccentric retrograde atrial activation dur-ing tachycardia, (2) a change in the ventriculoatrial interval withfunctional bundle branch block during tachycardia, and (3)atrial preexcitation with premature ventricular stimulation dur-ing tachycardia at a time when the His bundle was known to berefractory. Patients with AV nodal reentrant tachycardia had aconcentric retrograde atrial activation sequence and no evidenceof accessory pathway conduction in either anterograde or retro-grade directions as well as one or both of the following: (1)tachycardia reproducibly initiated after a critical AV nodal de-lay and (2) a V-HRA interval under 95 msec.

Accessory pathway location was determined at electrophys-iologic study by the following techniques: (1) orientation ofdelta wave morphology during maximal preexcitation,2 (2)atrial activation sequence mapped from multiple locations in theright atrium and within the coronary sinus during both supraven-tricular tachycardia and right ventricular pacing,7 (3) atrial pacing from right atrial and coronary sinus sites to determine thepacing site with the shortest stimulus to delta wave interval,6 (4)a change in ventriculoatrial interval during tachycardia when afunctional bundle branch block occurred ipsilateral to the loca-

tion of the pathway; an increase in ventriculoatrial interval of 30msec or greater was considered a free wall pathway and anincrease under 30 msec was considered a septal pathway.'1

Atrial preexcitation. During reciprocating tachycardia ineach patient, premature ventricular complexes were introducedat progressively shorter (decrements of 10 msec) coupling inter-vals beginning in late diastole and continuing until either theventricular effective refractory period was obtained or recipro-cating tachycardia terminated. Atrial preexcitation was definedas (1) a sudden decrease of 10 msec or more in the atrial cyclelength (AA) surrounding the premature ventricular complex and(2) a progressive further decrease in AA with earlier prematureventricular complexes (figure 1); reciprocating tachycardia ter-minated after the first AA decrease of 10 msec in a few in-stances. For this study, the definition of atrial preexcitation didnot require the identification of anterograde His bundle depolar-ization. The AA interval was measured from the first rapid atrialdeflection in the intracardiac lead having the earliest recogniz-able atrial activity during tachycardia. All patients had stabletachycardias without alternation of tachycardia cycle length.

Preexcitation index. Two preexcitation indexes were devel-oped to quantitate the degree of ventricular prematurity requiredfor the ventricular impulse to effect atrial preexcitation. Theseindexes and their method of determination are illustrated sche-matically in figure 1. The horizontal axis represents one tachy-cardia interval (V, to V,). Within this interval, premature ven-tricular complexes (V2) are introduced progressively earlierduring electrical diastole. The vertical axis represents the AA

AA

a 0 0 * 0 e

00

(RT TERMINOR VENT. E

Ptl = VlVi- v

VIV2

ViVi

Vj

00

ATION 1-RPJ)

f/V2 1L

p_I_ 1

'- V2 -~ VIFIGURE 1. Schematic illustrating the method of deterrnining preexci-tation indexes. The ventricular interval during tachycardia is represent-ed on the abscissa (V,VI). Progressively more premature ventricularcomplexes introduced during tachycardia are represented by V2. Theresultant atrial (AA) intervals surrounding the premature ventricularcomplex are plotted on the ordinate, and a hypothetical case is illustrat-ed. Premature ventricular complexes introduced relatively late in thetachycardia cycle cause no atrial preexcitation, but at a critical VIV2interval the AA interval begins to decrease (broken vertical line). Thedifference between the longest V V2 effecting preexcitation and thetachycardia cycle length (V1VI) defines PI,. P12 is defined by thelongest VIV2 interval effecting preexcitation divided by the tachycardiacycle length.

CIRCULATION

1 1

PI 2

494



ownloaded from


DIAGNOSTIC METHODS-PREEXCITATION SYNDROME

TABLE 1Comparison of AV reentrant and AV nodal reentrant tachycardias

AV AV nodalreentry reentry p value

n 55 22Cycle length (msec) 330 349 NS% TerminatedA 64 (35/55) 23 (5/22) p<.002% Preexcitation5B 89 (49/55) 14 (3/22) p<.00l% His +c 66 (27/41) 0 (0/3)

APercentage of tachycardias terminated with one premature ventricu-lar complex.

BPercentage of tachycardias demonstrating atrial preexcitation withone premature ventricular complex.

cPercentage of preexcited tachycardias with visible anterograde Hisat time of preexcitation. In eight patients the His was not consistentlyrecorded during tachycardia.

interval surrounding the premature ventricular complex. Rela-tively late premature ventricular impulses do not preexcite theatria and the AA interval remains the same as the V IV, interval.However, at a critical V1V2 interval the premature ventricularcomplex preexcites the atria and the AA interval shortens. Asthe premature ventricular impulse is introduced even earlier, theAA interval continues to shorten until either termination ofreciprocating tachycardia or ventricular refractoriness occurs.

For each patient, two indexes of preexcitation were derivedfrom this curve. Preexcitation index 1 (PI,) is expressed inmilliseconds and represents the difference between the tachy-cardia cycle length and the longest VIV2 interval at which atrialpreexcitation occurs. A second preexcitation index (P12) wasdeveloped to normalize preexcitation with respect to tachycar-dia cycle length (V1 V2V1VIV).

Statistical analysis. Tachycardia cycle lengths were com-pared between two groups with a t test. Termination and atrialpreexcitation of AV nodal compared with AV reentrant tachy-cardias (table 1) were compared by Fisher's exact test. Meanpreexcitation indexes (table 2) were compared with analysis ofvariance, and multiple comparisons were done with the Bonfer-roni correction. Characteristics of AV reentrant tachycardiaamong the three pathway locations (table 3) were compared bychi-square followed by Fisher's exact tests with the Bonferronicorrection.

ResultsPatients. Fifty-five patients had AV reentrant tachy-

cardia utilizing an accessory pathway for retrogradeconduction. Left free wall pathways included left later-al (n = 21) and left posterior (n = 3) pathways.-Posterior septal pathways consisted of left (n = 1 1)and right (n = 8) posterior paraseptal pathways andposteroseptal (n = 4) pathways. Anterior septal path-ways included one left anterior paraseptal and six rightanterior paraseptal pathways. Only one patient had a

right lateral pathway, and no patient had evidence ofmore than one bypass tract. Ventricular-induced atrialpreexcitation is illustrated in a patient with a septalaccessory pathway (figure 2) and in a patient with a leftfree wall pathway (figure 3).

Twenty-two patients met the criteria for AV nodalreentrant tachycardia. Twenty of these patients hadtypical (slow anterograde and fast retrograde conduc-tion) and two patients had atypical (fast anterogradeand slow retrograde conduction) AV nodal reentry.Four of the 20 typical AV nodal reentrant tachycardiashad V-HRA intervals greater than 95 msec (120 to 135msec). An AV nodal reentrant tachycardia that demon-strated ventricular-induced atrial preexcitation is illus-trated in figures 4 and 5.AV vs AV nodal reentry. Table 1 compares AV reentry

with AV nodal reentry. There was no significant dif-ference in mean tachycardia cycle length between thetwo groups. Atrioventricular reentry was terminatedby one premature ventricular complex (p < .002) anddemonstrated atrial preexcitation during tachycardia (p< .001) more frequently than AV nodal reentry. Ap-proximately two-thirds of AV reentrant tachycardiasthat demonstrated atrial preexcitation did so at a timewhen depolarization of the His bundle had visibly oc-curred anterogradely; no patient with AV nodal reentryshowed this finding. Only three of 22 (14%) patientswith AV nodal reentry demonstrated preexcitation dur-ing tachycardia. All three of these patients had thetypical slow-fast form ofAV nodal reentrant tachycar-dia with a V-HRA interval under 95 msec (75 to 80).The mean cycle length of the three AV nodal reentranttachycardias demonstrating atrial preexcitation was440 msec, compared with a mean cycle length of 335msec (p < .05) for the 19 tachycardias not demonstrat-ing preexcitation.

Table 2 shows the mean preexcitation indexes inthose patients with either AV or AV nodal reentranttachycardia who demonstrated atrial preexcitation.The PI, of the AV nodal reentrant tachycardias wasclosest to that of left free wall accessory pathways andwas significantly different from that of the septal path-ways. Figure 6 illustrates PI, values for each pathwaylocation and for the three patients with AV nodalreentry. The three patients with AV nodal reentry hadPI, values (.:-100 msec) greater than those in 14 of the

TABLE 2Mean preexcitation indexes

LEW PS AS AVNR(n =20) (n =21) (n =7) (n =3)

Pi, 88 38 17 108P12 0.75 0.88 0.95 0.75

LFW =left free wall; PS = posterior septal; AS = anterior septal;AVNR =AV nodal reentry.

p < .01 or smaller for LEW vs PS or AS, PS vs AS, AVNR vs PS orAS.Q

Vol. 74, No. 3, September 198649 495



ownloaded from


MILES et al.

TABLE 3Comparison of AV reentrant tachycardias

LFW PS AS RFW_ 1

24 23 7n-Cycle length (msec) 339 320 333 285% Terminated 37 (9/24)A 78 (18/23) 100 (7/7) 100 (1/1)% Preexcitation 83 (20/24) 91 (21/23) 100 (7/7) 100 (1/1)% His + 35 (6/l7)A 81 (13/16) 100 (7/7) 100 (l/1)

Abbreviations and definitions as in tables 1 and 2.Ap < .05 LFW vs PS or AS.

20 patients with left free wall pathways. Importantly,there was no overlap between the P1, of AV nodalreentry and the P1, of AV reentry using a septal acces-sory pathway.

Six patients with AV reentry and 19 patients withAV nodal reentry failed to demonstrate ventricular-induced atrial preexcitation. The longest VIV1 minusV1V2 intervals obtainable in these patients ranged from

848262 D

25 to 108 msec (mean 67) in those with AV reentry and45 to 160 msec (mean 105) in those with AV nodalreentry. In 11 of 19 patients with AV nodal reentry, thelongest obtainable V1V, minus V1V2 interval was un-der 100 msec. Thus preexcitation may not have oc-curred in some of these patients because obtainingsufficiently premature V1V2 intervals during tachycar-dia was limited by ventricular refractoriness or tachy-cardia termination.

802691 B11ll|lllll|!'l l|llill l illl!ll11 l. li, .1.. 31lll

11J111 1 v- 1VIIIlelmtCiVI

HBE

PCS

DCS

RVS

FIGURE 2. Ventricular-induced atrial preexcitation in a patient with a

septal accessory pathway. Surface leads I, II, III, and V1 are displayedalong with high right atrial (HRA), His bundle (HBE), proximal anddistal coronary sinus (PCS and DCS), and right ventricular (RV) elec-trograms. Reciprocating tachycardia is present with a cycle length of310 msec. A His bundle deflection (H) precedes each ventricular elec-trogram in the His bundle lead. During tachycardia, a premature ven-

tricular complex is introduced relatively late in diastole at a couplinginterval of 285 msec and, at a time when the His bundle is alreadyactivated anterogradely, the AA interval is preexcited to 295 msec. Theearliest atrial activation is within the ventricular electrogram in the Hisbundle lead followed by activation of the proximal coronary sinus,distal coronary sinus, and high right atrium. This retrograde atrial acti-vation sequence is indistinguishable from that of AV nodal reentry.

Atrial preexcitation by a relatively late premature ventricular complexduring tachycardia was typical of septal accessory pathways and was

associated with a short PI, (25 msec in this case).

,i' IIjI jIIjIIjI'llllll!!tI !!fsj§in IjIII

FIGURE 3. Ventricular-induced atrial preexcitation in a patient with a

left free wall accessory pathway. The format is the same as in figure 2,except a high right atrial electrogram is not included. Reciprocatingtachycardia is present with a cycle length of 350 msec and a His bundledeflection precedes each ventricular electrogram in the His bundle lead.Atrial preexcitation did not occur in this patient until the couplinginterval of the premature ventricular complex was decreased to 260msec, at which time the AA interval decreased to 330 msec. The Hisbundle electrogram is obscured by the ventricular electrogram at thetime of preexcitation. During tachycardia and after preexcitation, theearliest retrograde atrial activation occurs in the distal coronary sinuselectrogram followed by the proximal coronary sinus and His bundleelectrograms, demonstrating retrograde atrial activation using the ac-

cessory pathway. In left free wall accessory pathways, atrial preexcita-tion could often be obtained but required relatively early ventricularextrasystoles (long PI ) and often occurred when the His bundle electro-gram was not visible.

CIRCULATION

11

Ill

Vi

HRA

HBEPCS

DCS

RV

l[

1

1

496



ownloaded from



V1

HRA J

65 160 3516

DCS A.

RV

FIGURE 4. Initiation of reciprocating tachycardia in a patient who demonstrated atrial preexcitation during AV nodal reentry.The format is the same as the preceding figures. The first two complexes represent sinus rhythm during which two prematureatrial extrastimuli are introduced. A modest increase in AH interval occurs after the first premature atrial stimulus (S2), but a

marked increase in AH interval occurs after the second premature atrial stimulus (S3) and tachycardia is initiated. Tachycardiainduction in this patient consistently required a large AH 'jump." The V-HRA interval is 70 msec, excluding retrogradeconduction via an accessory pathway. The earliest atrial activation is within the ventricular electrogram in the His bundle leadfollowed by the proximal and distal coronary sinus atrial electrograms.

Localization of accessory pathways. Table 3 comparescharacteristics of left free wall, posterior septal, anteri-or septal, and right free wall accessory pathways. Onlyone patient had a right free wall pathway and thus the

characteristics of this location cannot be adequatelycompared with those of the other locations. Tachycar-dia cycle length was similar among the accessory path-way locations. AV reentry involving left free wall ac-

FIGURE 5. Ventricular-induced atrial preexcitationin AV nodal reentry (same patient as in figure 4). Theformat is the same as that of figure 4. AV nodal

reentrant tachycardia is present with a cycle length of410 msec. A premature ventricular complex from the

right ventricular outflow tract (a stable apical site

could not be obtained) with a coupling interval of 300

msec is introduced during tachycardia and shortens

the AA interval to 395 msec. This was the longestVIV2 interval that effected atrial preexcitation in this

patient, although shorter V1V2 intervals could de-

crease the AA interval to as short as 355 msec.

Vol. 74, No. 3, September 1986

li

v1

HRA

HBE

PCS

DCS

497

111 1111. I, I. I. I_



ownloaded from


MILES et al.

PREEXC

U

E

0l-

20-

10-

100.

90-

80-

70-

60-

50-

4 0-

30.

2 0-

0O

N = 20

LFW

FIGURE 6. PI, is illustrated:(PS), anterior septal (AS), anway. The PI, values for the(AVNR) that demonstrated atihorizontal lines represent the n45 msec was seen only with scmsec or greater occurred only(dotted horizontal lines).

cessory pathways was ]one premature ventricudia involving septal paof which could be ternpulse. The percentage

'ITATION INDEX was overlap in 11 of 41 patients between posteriorseptal and left free wall accessory pathways. ThereforePI1 differentiated left free wall from septal accessorypathways in 77% (37 of 48) of patients with preexcita-tion. Within the posterior septal group, PI, did notdifferentiate left from right posterior paraseptal path-way locations (mean PI1 37 and 34 msec, respective-

_- _ _ly). However, all eight right posterior paraseptal path-ways were associated with atrial preexcitation at a timewhen the anterograde His bundle deflection was

* visible.* ~~~~~~~~Preexcitation data were available in two patients

.. who had both left bundle branch block and normalg QRS morphology tachycardias. Both of these patients

.. had left free wall accessory pathways and atrial preex-* *_____ citation could not be elicited when normal QRS mor-

phology was present. However, when functional left

2 1 7 3 bundle branch block occurred, atrial preexcitation oc-X 1 curred with premature ventricular complexes intro-

PS AS RFW AVNR duced late in diastole and the PI, values (16 and 26for left free wall (LFW), posterior septal msec) were in the range expected for septal accessory.d right free wall (RFW) accessory path-three AV nodal reentrant tachycardias pathways. Therefore bundle branch block ipsilateral to

rial preexcitation are included. The solid the accessory pathway may markedly alter preexcita-nean PI, for each group. A PI, of less than tion characteristics.eptal accessory pathways and a P1, of 75 PI1 vs P12. Values for PI2 are shown in table 2 andy with left free wall accessory pathways figure 7. The mean values of P12 for each of the three

pathway locations were significantly different (p <.001), and there was no overlap of the index between

less likely to be terminated with the anterior septal and the left free wall pathway loca-lar complex than was tachycar- tion. However, compared with PI, there was greaterithways (p < .05), the majority overlap between posterior septal and left free wall val-ainated with one premature im- ues, and PI2 differentiated left free wall from septalof patients demonstrating atrial accessory pathways in only 54% (26 of 48) of patients.

preexcitation did not differ between left free wall andseptal accessory pathways. However, the His bundledeflection was visible at the time of preexcitation inonly 35% of tachycardias involving left free wall ac-cessory pathways compared with 87% of septal path-ways (p < .05).

In those patients whose tachycardia demonstratedatrial preexcitation, the P1, was longest for left freewall pathways, shortest for anterior septal pathways,and intermediate for posterior septal pathways (table2). There was a significant difference in P1, betweenthe left free wall and septal accessory pathways (p <.001) and between the posterior septal and anteriorseptal accessory pathways (p < .001). Figure 6 illus-trates that a P1, of less than 45 msec occurred only withseptal accessory pathways and a PI, of 75 msec orgreater only with left free wall accessory pathways.The P1, differentiated all anterior septal (c25 msec)from all left free wall (-50 msec) pathways, but there

PREEXCITATION INDEX 2

a)(n

0-

.0 0

.90

.801

.7 0

.60 - ,, /IT

:ve

*0 v*- *-*-i-

N- 20 21 7 3U

L FW PS AS RFW AVNR

FIGURE 7. P12 is illustrated for left free wall (LFW), posterior septal(PS), anterior septal (AS), and right free wall (RFW) accessory path-ways as well as for AV nodal reentry (AVNR). The fornat is similar tothat of figure 6. The horizontal solid lines represent the mean P12 foreach group. The broken horizontal lines enclose the area of overlapbetween P12 of left free wall and posterior septal pathways.

CIRCULATION

(

0 0

0

:0:Q 0

-

0

S -

* 0

: 0

498



ownloaded from



The PI2 can also vary in the same patient if the tachy-cardia cycle length changes substantially. For exam-ple, one patient had dual AV nodal pathways and a leftposterior paraseptal accessory pathway. Only AVreentry was induced but anterograde conduction oc-curred over either the fast or slow AV nodal pathway.Thus two tachycardia cycle lengths (280 and 430msec) were present associated with distinctly differentAH intervals. This change in tachycardia cycle lengthdue to anterograde conduction characteristics had littleeffect on PI1 but P12 changed from 0.79 to 0.88.

DiscussionLocalization of accessory pathways is important be-

fore surgery or catheter ablative therapy, and it is criti-cal that AV nodal reentry not be mistaken for AVreentry using a septal accessory pathway for retrogradeconduction. Standard criteria for making these distinc-tions may sometimes be inconclusive. Thus we inves-tigated the ability of preexcitation indexes derivedfrom ventricular-induced atrial preexcitation during re-ciprocating tachycardia to differentiate AV reentryusing a septal accessory pathway from AV nodalreentry and to localize accessory pathways. The P1,distinguished all patients with septal accessory path-way from the three patients with atrial preexcitationduring AV nodal reentry. In addition, atrial preexcita-tion during tachycardia or termination of tachycardiawith one premature ventricular complex made AVreentry using a septal accessory pathway more likelythan AV nodal reentry. PI1 distinguished all anteriorseptal from left free wall pathways, and it differentiat-ed septal from left free wall pathways in 77% of pa-tients in whom preexcitation occurred; a P1, of 75 msecor greater occurred only with left free wall pathwaysand a P1, under 45 msec only with septal accessorypathways. Inability to preexcite the atrium duringtachycardia excluded anterior septal pathways and wasunusual (2/23) for posterior septal pathways.

Previous studies have demonstrated that the locationof ventricular stimulation in relation to the site of theaccessory pathway is an important determinant of themaximal amount of atrial prematurity obtainable byventricular-induced atrial preexcitation" and the ven-triculoatrial interval following the premature ventricu-lar complex. 16 For example, Sellers et al. 14 showed thatpremature ventricular impulses introduced from theright ventricle resulted in a greater degree of atrialpreexcitation in patients with septal accessory path-ways than in those with left free wall pathways. Inaddition, more atrial preexcitation was obtained withleft than with right ventricular extrasystoles in patientsVol. 74, No. 3, September 1986

with left free wall accessory pathways. Thus the further the pacing electrode is from the anatomic locatiorof the tachycardia circuit, the more premature the ven-tricular extrasystole must be to enter the accessorpathway and preexcite the atria. This also explainwhy right ventricular-induced atrial preexcitation ismore readily accomplished with a left-sided accessoropathway when functional left bundle branch block oc-curs during tachycardia, 14 presumably because theright bundle branch and right ventricle become part olthe tachycardia circuit during left bundle branch bloclaberrancy. Indeed, the preexcitation index in two olour patients with left free wall accessory pathwayssuggested a septal accessory pathway location wherleft bundle branch block aberrancy occurred duringtachycardia.The total amount of atrial preexcitation obtainable iP

determined by both the degree of prematurity neces-sary to produce preexcitation as well as either the ven-tricular or accessory pathway retrograde effective re-fractory period during reciprocating tachycardia. Wcanalyzed only the latest premature ventricular comple)that produced atrial preexcitation as a determinant o1pathway location to avoid potentially misleading dautoccurring as a result of differences in ventricular ancaccessory pathway refractoriness among patients. Theoverlap of PI1 between posterior septal and left freewall pathways is not surprising because the location o0accessory pathways is a continuum. Pathways locatecnear the junction of the posterior septal area and theposterior left free wall would be expected to exhibisimilar preexcitation properties.AV reentry using septal accessory pathways may b

particularly difficult to differentiate at electrophysiologic study from AV nodal reentry when the V-HRAinterval is greater than 95 msec and dual-pathwa3physiology is not observed with initiation of tachycardia.5 Both arrhythmias show a similar retrograde atriaactivation sequence during tachycardia and little or n(change in the ventriculoatrial interval with functionabundle branch block. Our data demonstrate that atriapreexcitation in AV nodal reentry occurs only witlvery early ventricular extrasystoles (mean PI1 of l0Mmsec). The overlap of P1, between AV nodal reentrand left free wall pathways is only at the upper extreme, and P1, values below 100 msec did not occuwith AV nodal reentry. Although preexcitation indexes may not distinguish all left free wall accessory pathways from AV nodal reentry, they distinguish all septal accessory pathways from AV nodal reentry even ithe His deflection is not noted during preexcitation.

Normalization of the preexcitation index for tachy499



ownloaded from


MILES et al.

cardia cycle length (P12) was not as useful in distin-guishing left free wall from septal accessory pathwaysas was the actual degree of prematurity required toeffect atrial preexcitation (P1,). Tachycardia cyclelength can vary because of changes in anterogradeconduction characteristics, such as the AH interval,without affecting PI1.The use of preexcitation indexes has limitations.

Reciprocating tachycardia must be of sufficient dura-tion to allow introduction of progressively prematureventricular extrasystoles, although this is usually not aproblem. Determination of the longest V1V2 intervalresulting in atrial preexcitation may be difficult in pa-tients with alternation in tachycardia cycle length.There is some overlap of preexcitation indexes be-tween patients with left free wall and posterior septalpathways. In addition, preexcitation indexes derivedduring bundle branch block aberrancy may yield mis-leading data if the bundle branch block is ipsilateral tothe location of the accessory pathway. Lastly, in somepatients preexcitation can occur during AV nodal reen-trant tachycardia and the indexes overlap with the outerlimits of the indexes of left free wall accessory path-ways, thus precluding differentiation of AV nodalreentry from AV reentry in this situation.

In conclusion, our findings show that the preexcita-tion index differentiates patients with AV nodal reen-trant tachycardia from those with septal accessorypathways. In addition, characteristics of atrial preexci-tation during reciprocating tachycardia help to differ-entiate left free wall from septal accessory pathwaylocations. Preexcitation characteristics are useful inconjunction with other criteria to establish the mecha-nism of supraventricular tachycardia and to localizeaccessory pathways.

We thank Dr. Naomi Fineberg for performing the statisticalanalysis, Elizabeth Darling and Sue Faust for assisting with thestudies, and Sue Hennigar and Debbie Carter for preparing themanuscript.

References1. Prystowsky EN, Miles WM, Heger JJ, Zipes DP: Preexcitation

syndromes: mechanisms and management. Med Clin North Am 68:831, 1984

2. Gallagher JJ, Pritchett ELC, Sealy WC, Kasell J, Wallace AG: Thepreexcitation syndromes. Prog Cardiovasc Dis 20: 285, 1978

3. Klein GJ, Guiraudon GM, Perkins DG, Jones DL, Yee R, Jarvis E:Surgical correction of the Wolff-Parkinson-White syndrome in theclosed heart using cryosurgery: a simplified approach. J Am CollCardiol 3: 405, 1984

4. Morady F, Scheinman MM, Winston SA, DiCarlo LA, Davis JC,Griffin JC, Ruder M, Abbott JA, Eldar M: Efficacy and safety oftranscatheter ablation of posteroseptal accessory pathways. Circu-lation 72: 170, 1985

5. Benditt DG, Pritchett ELC, Smith WM, Gallagher JJ: Ventriculo-atrial intervals: diagnostic use in paroxysmal supraventriculartachycardia. Ann Intern Med 91: 161, 1979

6. Wellens HJJ, Schuilenberg RM, Durrer D: Electrical stimulation ofthe heart in patients with Wolff-Parkinson-White syndrome, typeA. Circulation 43: 99, 1971

7. Gallagher JJ, Pritchett ELC, Benditt DG, Tonkin AM, CampbellRWF, Dugan FA, Bashore TM, Tower A, Wallace AG: Newcatheter techniques for analysis of the sequence of retrograde atrialactivation in man. Eur J Cardiol 6: 1, 1977

8. Motte G, Bellanger P, Vogel M, Welti J: Disappearance of abundle branch block with acceleration of reciprocal tachycardia inWolff-Parkinson-White syndrome. Ann Cardiol Angeiol 22: 343,1973

9. Coumel P, Attuel P: Reciprocating tachycardia in overt and latentpreexcitation: influence of bundle branch block on the rate of thetachycardia. Eur J Cardiol 1: 423, 1974

10. Pritchett ELC, Tonkin AM, Dugan FA, Wallace AG, Gallagher JJ:Ventriculo-atrial conduction time during reciprocating tachycardiawith intermittent bundle branch block in the Wolff-Parkinson-White syndrome. Br Heart J 38: 1058, 1976

11. Kerr CR, Gallagher JJ, German LD: Changes in ventriculoatrialintervals with bundle branch block aberration during reciprocatingtachycardia in patients with accessory atrioventricular pathways.Circulation 66: 196, 1982

12. Zipes DP, DeJoseph RL, Rothbaum D: Unusual properties of ac-cessory pathways. Circulation 49: 1200, 1974

13. Neuss H, Schlepper M, Thormann J: Analysis of reentry mecha-nisms in three patients with concealed Wolff-Parkinson-White syn-drome. Circulation 51: 75, 1975

14. Sellers TD, Gallagher JJ, Cope GD, Tonkin AM, Wallace AG:Retrograde atrial preexcitation following premature ventricularbeats during reciprocating tachycardia in the Wolff-Parkinson-White syndrome. Eur J Cardiol 4: 283, 1976

15. Smith WM, Pritchett ELC, Campbell RWS, Gallagher JJ: An inter-active computer program for measurement of data obtained usingthe extrastimulus technique during electrode catheter studies. Com-put Biomed Res 10: 595, 1977

16. Weiss J, Brugada P, Roy D, Bar FWHM, Wellens HJJ: Localiza-tion of the accessory pathway in the Wolff-Parkinson-White syn-drome from the ventriculoatrial conduction time of right ventricularapical extrasystoles. PACE 6: 260, 1983

CIRCULATION500



ownloaded from


W M Miles, R Yee, G J Klein, D P Zipes and E N Prystowskytachycardia and localizing accessory pathways.

The preexcitation index: an aid in determining the mechanism of supraventricular

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1986 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.74.3.493

1986;74:493-500Circulation.

http://circ.ahajournals.org/content/74/3/493the World Wide Web at:

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

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

is online at: Circulation Information about subscribing to Subscriptions:

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

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

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

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



ownloaded from

http://circ.ahajournals.org/content/74/3/493

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

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


the preexcitation index: an aid in determining the localizing

Documents