non-invasivethree-dimensional wolff-parkinson-white · in ninepatients with wolff-parkinson-white...

Br Heart J 1993;69:201-210

ORIGINALS

Non-invasive three-dimensional localisation ofarrhythmogenic foci in Wolff-Parkinson-Whitesyndrome and in ventricular tachycardia byradionuclide ventriculography: phase analysis ofdouble-angulated integrated single photonemission computed tomography (SPECT)

P Weismuller, M Clausen, R Weller, P Richter, J Steinmann, E Henze, I Dormehl,M Kochs, W E Adam, V Hombach

Department ofCardiology,Angiology, andPneumology,University ofUlm,GermanyP WeismullerP RichterM KochsV HombachDepartment ofNuclear Medicine,University of UlIm,Federal Republic ofGermanyM ClausenR WellerJ SteinmannE HenzeW E AdamAEC-Institute of LifeScience, Faculty ofMedicine, Universiteitvan Pretoria, SouthAfricaI DormehlCorrespondence to:Dr Peter Weismuller,Department of Cardiology,Angiology and Pneumology,University of Ulm, 8Robert-Koch Street, 7900Ulm, Germany.Accepted for publication15 September 1992

AbstractA new tomographic technique combinedwith phase analysis was used to detectpremature and ectopic ventricular con-traction patterns in 15 patients withWolff-Parkinson-White syndrome andduring ventricular tachycardia in sevenpatients. Data generated by gated single-photon emission computed tomography(SPECT) were analysed by backprojec-tion of the Fourier coefficients, double-angulation, and integration to thick slicescontaining the ventricles, thus allowingvisualisation of the contraction patternsin three perpendicular views. The resultswere compared with those of cathetermapping.

In nine patients with Wolff-Parkinson-White syndrome the site of initial con-traction detected was identical with thesite of the accessory pathway found bycatheter mapping. The sites of origin ofthe ventricular tachycardias determinedby catheter mapping were within 3 cm ofthe sites detected by the new technique.This new technique seems to be a

promising non-invasive method for local-ising ectopic ventricular activity that willconsiderably shorten the time requiredfor subsequent invasive procedures.

(Br HeartJ 1993;69:201-210)

The exact site of the accessory pathway mustbe identified before patients with Wolff-Parkinson-White syndrome (WPW syn-drome) can be treated with catheter ablation.Similarly, the detection of the site of origin oftachycardia is essential for intraoperativel orcatheter ablation2 in patients with sustainedventricular tachycardia. Invasive cathetermapping is the standard method used tolocalise accessory pathways in WPW syn-drome' and identify the site of origin of ven-tricular tachycardias4 before interventionalprocedures.

It is difficult to localise anterograde con-

ducting accessory pathways from the surfaceelectrocardiogram.56 Also the determinationof the origin of ventricular tachycardias fromthe configuration of the surface electrocardio-gram is inaccurate.78 We and others haveshown that phase analysis of radionuclideventriculography is able to show the localcontraction pattern and thus the site of initialcontraction.9-1" If mechanical contraction fol-lows the electrical excitation, localisation ofectopic ventricular depolarisation is possible.When phase analysis of planar radionuclideventriculography was used to localise ectopicventricular excitation in WPW patients'4'6and in patients with sustained ventriculartachycardia'7-19 the results corresponded wellwith the results of invasive mapping. In con-trast, experimental data showed that thismethod is only able to detect an extendedarea of initial contraction20 because the ventri-cles are displayed separately in only one pro-jection (left anterior oblique). Even in thisprojection other cardiac structures are super-imposed on the ventricles. In a previous studywe reported the preliminary results of single-photon emission computed tomography(SPECT) and integrated slices that providethree dimensional cross sectional images ofthe heart derived from phase analysis.2122 Inaddition to the left anterior oblique view ofconventional planar phase analysis, an axialview of the heart was generated, which wasperpendicular to the long axis of the body.

This study reports on the first clinical dataof phase analysis of radionuclide ventriculog-raphy obtained with a newly developed com-puter algorithm for analysing gated SPECTdata. The method allows images to be gener-ated in any projection of the cardiac bloodpool with isolated display of the ventricularchambers.

Patients and methodsCONTROL GROUPForty patients with normal atrioventricularconduction in the surface electrocardiogram

201

on April 2, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.69.3.201 on 1 March 1993. D

ownloaded from

http://heart.bmj.com/

Weismiuller, Clausen, Weller, Richter, Steinmann, Henze, Dormehl, Kochs, Adam, Hombach

made up the control group. Double-angulated integrated SPECT was performedduring sinus rhythm to assess left ventricularfunction.

PATIENTS WITH WPW SYNDROMEWe studied four women and 11 men (41.9(13-9) 25-71 years). All of them had a deltawave in the surface electrocardiogram. Sevenof them were studied during preexcitation ofthe ventricles caused by depolarisationthrough the accessory pathway and also dur-ing normal atrioventricular conduction with a

narrow QRS complex: six patients afteradministration of ajmaline (50 mg intra-venously) and one after successful catheterablation.

In 14 patients catheter mapping was per-

formed during an electrophysiological studyand the atrial insertion of the accessory path-way was determined by the following criteria:shortest P-delta interval measured duringstimulation with the mapping electrode,shortest VA interval during atrioventricularreentrant tachycardia and shortest VA inter-val during ventricular pacing. In all patientsthe whole atrioventricular ring was scannedby the mapping electrode to detect multiplebypass tracts. In one of these patients,catheter mapping could not be performedduring the first mapping procedure becauserepeated atrial fibrillation episodes were

induced mechanically by the mappingcatheter. In two patients (case 19 and case

16) successful radiofrequency catheter abla-tion of the atrial bypass insertion was per-

formed.2"

PATIENTS WITH VENTRICULAR TACHYCARDIAOnly patients with haemodynamically welltolerated ventricular tachycardias were inves-tigated. The mean (SD) age of the patientswas 56-3 (12-2) years (range 32-70). Six hadcoronary artery disease and one had rightventricular dysplasia with areas of fattydegeneration in the left ventricle and in theseptum detected by magnetic resonance

imaging. In all but one patient two radionu-clide studies were performed-during clinicalventricular tachycardia and then during sinusrhythm. Table 1 shows the details of these

patients. The mean ejection fraction mea-

sured from left ventricular angiography was

39.3 (14-8)% (24-68%). The mean ejectionfraction during tachycardia was 27-3 (7.7)%(17-38%) as determined by planar radionu-clide ventriculography. The mean rate of theventricular tachycardias was 152 (19) min(120-175/min). In all patients the origin ofthe tachycardia could be detected invasively,either by catheter mapping4 or by intraopera-tive mapping (case 4). All patients presentedwith one clinical ventricular tachycardia. Twopatients showed another configuration of a

non-clinical ventricular tachycardia. The siteof origin of the clinical tachycardia was deter-mined by the detection of a mid diastolicpotential and by pace-mapping.4 A diastolicpotential during tachycardia was found in sixpatients, and preceded the QRS complex by60 ms (15-100 ms) on average. The criterionof pace mapping was fulfilled when the QRScomplex paced by the mapping catheter was

identical in more than 10 of the 12 surfaceleads with the configuration of clinical tachy-cardia on the electrocardiogram. This occur-

red in three of the patients. In three patients(cases 2, 3, and 7) catheter ablation was per-

formed, in two of them (cases 3 and 7) byradiofrequency catheter ablation24 and in theother (case 2) by direct current.'5 In one

patient (case 4) the tachycardia could betolerated haemodynamically for 20 min onlyafter long-term administration of amio-darone. In this patient the site of origin ofthe clinical tachycardia was determined intra-operatively during surgical ablation.

MODE OF INDUCTION OF THE VENTRICULAR

TACHYCARDIA

In all patients with ventricular tachycardia thetachycardia was induced directly after a basicor control electrophysiological study and thestimulation was delivered through a catheterplaced in the apex of the right ventricle.Single or double premature stimuli were

used.

DOUBLE-ANGULATED INTEGRATED SPECT

(DA-ISPECT)After in vivo labelling of the patients' ery-throcytes with 740 MBq (20 mCi) tech-

Table 1 Details ofpatients with venticular tachycardia

Case Age EF (SR) EF (VT) VT rate VT ORS AxisNo Sex (yr) Disease (%) (%) (/min) configuration (ms) ()

1 M 55 CAD 37 29 150 LBBB 180 + 120PW-An

2 M 70 CAD 36 21 171 LBBB 180 + 110AW-AnPW-An

3 M 32 RVD 68 38 150 RBBB 130 - 904 M 56 CAD 24 - 158 LBBB 180 0

AW-An5 F 58 CAD 37 26 120 RBBB 180 - 60

AW-An6 M 67 CAD 47 33 175 RBBB 130 - 60

PW-An7 M 56 CAD 26 17 142 RBBB 170 + 100

AW-An

The mean (2SD) ejection fraction measured by angiography during sinus rhythm is 39 3 (14-8)%. Mean ejection fraction duringtachycardia (planar radionuclide ventriculography) is 27-3 (7 7)%. In patient 4 planar radionuclide ventriculography could not beused to determine the ejection fraction during ventricular tachycardia because of poor haemodynamic tolerance.AW-An, anterior wall aneurysm; CAD, coronary artery disease; PW-An, posterior wall aneurysm; RVD, right ventricular dys-plasia.

202



j.com/

Br H


ownloaded from


Localisation of arrhythmogenic foci by SPECT

netium 99m pertechnetate we performedECG-triggered 1800 SPECT with a 64matrix, 32 angles, 15 s acquisition per stepand 8 frames per cycle (single-headed cam-era: Orbiter, Siemens). Fourteen minutes ofactive data acquisition was needed and sixminutes for the deposition of the data. Theparameters mean (zero harmonic), sine, andcosine coefficients of the first harmonic wereextracted by Fourier analysis and back-projected. All pixels were then re-alignedalong the long axis of the left ventricle (hencedouble-angulation). Four defined projections(those chosen for angiography and echocar-diography) were calculated. Mean, sine, andcosine coefficients were added (integrated) toform thick slices that contained the completeventricle under investigation. The integrationwas performed for each of the four chosenviews (fig 1). This view dependent addition ofvoxels means that the images represent a pro-jection of the corresponding view rather thana single slice. Finally, we calculated the phasevalues from each pixel (fig 2). In summary,these procedures result in planar scans of theheart in three projections. In this way, thecontraction patterns and initial inwardmotion of each cardiac chamber was assessedseparately without significant overlap of othercardiac structures.26We performed conventional radionuclide

ventriculography to calculate the ejectionfraction after each SPECT investigation (seetable 1). Two investigators who were un-aware of the results of the catheter mappingprocedure independently evaluated theframes of double-angulated integratedSPECT.

LOCALISATION OF ACCESSORY PATHWAYS BYDA-ISPECTFigure 3 shows the eight different areas alongthe atrioventricular ring that were used tolocalise the accessory pathways. The resultswere compared with those from cathetermapping. When the sites were identical thecorrelation was 2, when the sites were adja-cent the correlation was 1 and in all othercases the correlation was 0.

LOCALISATION OF THE ORIGIN OFVENTRICULAR TACHYCARDIAS BYDA-ISPECTWe used a detailed diagram suggested byKuchar 19898 that divides the left ventricleinto 24 different areas (fig 4) to localise theorigin of the ventricular tachycardia withinthe left ventricle. The results from cathetermapping were compared with those fromDA-ISPECT. We attempted to measure thedistance between the origin of the tachycar-dias as determined by catheter mapping and

4 4

,, ~~~~~~~~~~~~~~~.~~~~~~~~~~~~~~~~~....... ................

Figure I Double-angulated integrated SPECT (DA-ISPECT). Schematic display of the zones of integration (addition) after double angulationaccording to the long axis of the left ventricle. These zones are depicted non-darkened in the upper part of the figure and are used to calculate the "thickslices" shown in the lower part of the figure. The projections are short axis view (SA), vertical long axis view for the right (RV-vLA) and the left ventricle(LV-vLA), andfour chamber view (4CH).

4

SA:

anteriorR Lposterior

RV-vLA:anterior

basal apicalinferior

LV-vLA:anterior

basal apicalinferior

4H:14CH

basalR L

apical

.............. ...................

..............................................................................................

...............................

:: :::: ::: :: -* :: :: : :: :: -: :- ::-:: .:::-::::

203



j.com/

Br H


ownloaded from


Weismiiuer, Clausen, WeMler, Richter, Steinmann, Henze, Dormehl, Kochs, Adam, Hombach

Figure 2 Fourier phaseimages from DA-ISPECT.The projections arearranged as in fig 1. Acolour code is shown on theright hand side. In thephase image the purpleencodes the initialcontraction and red the latecontraction. The upperpanel shows the contractionpattern in a healthyindividual. There is earlycontraction (blue) at theseptum (circle in the shortaxis view) and at the rightfree wall (circle in the longaxis view of the rightventricle andfour chamberview). The lower panelshows the contractionpattern in a WPWpatientwith an accessory pathwaylocated anteroseptally(circle in short axis viewandfour chamber view).For the surfaceelectrocardiogram see fig 8.

V$ I

I A.

L-:- --'

by DA-ISPECT.All patients had given informed c4

The study complies with the declaraHelsinki.

ResultsCONTRACTION PATTERNS IN THE CONTRGROUPSixteen (40%) of the 40 patients dhave a recognisable contraction patteireflected normal physiology. In 24 p(60%) typical contraction patterns werIn five patients the initial contracticseen at the anterior part of the right vclose to the apex (upper panel in fig

AS

LAL

LL

Figure 3 Diagram of the atrioventricular ringfor the evaluation of the site of theaccessory pathway. There are eight areas: AS, anteroseptal; RAL, right anterolatright lateral; RPL, right posterolateral; PS, posteroseptal; LPL, left posterolateral;lateral; LAL, left anterolateral.

onsent.tion of

tOL

[id notrn that

four other patients the initial contraction wasdetected paraseptally at the right and left baseof the ventricles. The remaining 15 patientsshowed a combination of these two patterns.This accords with the electrophysiologicalfindings of electrical excitation during sinusrhythm.27 The early contraction in the rightventricle may be explained by a rapid propa-gation of the electrophysiological impulsethrough the moderator band.

patients CONTRACTION PATTERNS IN PATIENTS WITHe seen. ACCESSORY PATHWAYSrn was Table 2 shows the results in the WPWentricle patients. Additionally, Kent bundle localisa-2). In tion from surface electrocardiogram was

assessed by the algorithms of Milstein et al6and Gallagher et al.5

Figure 5 shows the DA-ISPECT image fora patient (case 22) with a left lateral accessorypathway confirmed by catheter mapping.Because of a long effective refractory period,atrioventricular conduction along the acces-sory pathway could be blocked by intra-venous ajmaline (50 mg), and a normalcontraction pattern ensued (fig 5). Figure 6shows the results of a DA-ISPECT studyfrom a patient in whom catheter mappingidentified a left posterolateral accessory path-way 3 cm distal to the ostium of the coronarysinus.

Four of the 15 patients had two accessoryst pathways. In one patient (case 17) no con-

R~LJ duction could be detected from the ventriclesto the atria through the accessory pathways.This patient had recurrent atrial fibrillation,resulting in rapid ventricular excitationthrough the bypass tract with rates of250/min. The preoperative DA-ISPECTstudy did not show a distinct and consistentearly contraction site (fig 7). During preoper-ative catheter mapping one accessory pathwaywas found posteroseptally 2 cm distal to the

1r; RL coronary sinus with an effective refractoryrLL, left period of 200 ms. During intraoperative map-ping the atrial and ventricular insertion of the

204

i. -

Z,.,E I ii

2.c

-4d4 z

.., .-4%,I

.,L. I ll.-

9i2.2..- .1

.i.;,- -

.-

7

I .L,

.I kit.. " "",.:..

I !LI -'.



j.com/

Br H


ownloaded from



I

I

9

Figure 4 Grid showing 24 different areas of the left ventricle definedfor the evaluation ofthe site of origin of ventricular tachycardias from surface electrocardiogram, cathetermapping, and DA-ISPECT. The left side of thefigure shows the right anterior oblique 30°projection and the right side shows the left anterior oblique 600 projection. 1, 2, 3 = apical,mid-ventricular, basal; A, anterior; M, middle; I, inferior; S, septal; C, central;I, inferior. Adaptedfrom Kuchar et al.8

accessory pathway was determined postero-septally 1 cm anterior to the crux cordis. Nofurther bundle was found. After operationintermittent atrioventricular conduction was

seen with wide QRS complexes. The surfaceelectrocardiogram indicated a second acces-

sory pathway in the right lateral atrioventricu-lar groove.4 The second DA-ISPECT study

showed an initial contraction right postero-laterally during extensive pre-excitation (fig7). The postoperative electrophysiologicalstudy showed a second accessory pathway atthis site with an effective refractory period of250 ms. However, this bundle was operatingonly intermittently, which is why it was notfound during the preoperative and intra-operative electrophysiological examination.When the preoperative DA-ISPECT studywas retrospectively re-evaluated two sites ofearly contraction caused by two accessorypathways became apparent (fig 7).One patient with WPW syndrome (case

16) had atrioventricular reentrant tachycar-

dias (260/min). Figure 8 shows the surfaceelectrocardiogram. The surface electrocardio-gram did not show a distinct site of an

accessory pathway.56 The electrophysiologicalstudy showed an accessory pathway witha very short effective refractory period(180 ms). Localisation of the bypass tract wasnot possible because repeated mechanicalinduction of atrial fibrillation by the mappingcatheter resulted in ventricular rates of250/min. On one occasion induction of an

atrioventricular reentrant tachycardia pro-

duced a left bundle branch block configura-tion with the same rate of 260/min, indicatingan accessory pathway located on the rightside or within the septum.28 A surface electro-cardiogram recorded during maximum pre-

excitation (rapid atrial stimulation) indicateda right lateral accessory pathway.6 The DA-ISPECT study showed clear evidence of earlyanteroseptal contraction (fig 2). Later, afteroral administration of flecainide (300 mg/day), successful radiofrequency catheter abla-tion confirmed the presence of an accessory

pathway located 1 cm anterior to the bundleof His. In the DA-ISPECT study aftercatheter ablation without preexcitation, thebeginning of the contraction could belocalised posteriorly only-a small distancefrom the site detected by the first DA-ISPECT study. In patient 19, another suc-

cessful radiofrequency catheter ablation23confirmed the exact atrial insertion site of theaccessory pathway. In patient 12 with twobypass tracts, the site of the second pathwaywas localised from DA-ISPECT only byretrospective re-evaluation.

Table 2 summarises the results. In 10 ofthe 14 patients in whom catheter mappingwas performed the DA-ISPECT data corre-

lated well with the catheter mapping data (fig3). In three patients DA-ISPECT showed an

area of early contraction next to the area

found by catheter mapping. In one patient nocorrelation could be found between the twomethods. In the four patients with two acces-

sory bundles, only one pathway could be

Table 2 Localisation of initial excitation and contraction in patients with Wolff-Parkinson-White syndrome

Case Age QRS ECG Catheter DA-ISPECT Correlation*No Sex (yr) duration (ms) Milsgein6 ((fig 3) (fig 3) (21110)

8 M 25 130 AS AS AS 29 M 41 110 AS LPL LPL 210 F 42 120 PS PS PS 211 M 71 140 RL RL 1RL 2

2 pathways LL12 M 31 130 LL I LL (LL) (2)

2 pathways 2 PS PS 213 M 33 140 LL LL LL 214 M 40 120 LL PS PS 215 M 62 140 UD LL LL 216 M 36 120 PS AS ASt 217 M 36 150 1 RLpreop LPL

2 pathways 130 2 RL postop RPL RPL 218 M 67 140 PS I PS RPL 1

2 pathways 2 AS19 M 28 130 RL RPL PSt 120 F 44 120 PS PS LPL 121 F 39 140 AS AS RPL 022 M 34 160 AS - RL -

Site of the accessory pathway determined from surface electrocardiogram according to Milstein et al 6 AS, anteroseptal; LL, leftlateral; RL, right lateral; PS, posteroseptal; UD, undetermined. See fig 3 for designation of the site of the accessory pathwaysdetermined from catheter mapping and DA-ISPECT.*Correlation: 2, same area detected by catheter mapping and DA-ISPECT; 1, area detected by catheter mapping was adjacent tothe area detected by DA-ISPECT; 0, different areas detected by catheter mapping and DA-ISPECT.tSuccessful radiofrequency catheter ablation.

205



j.com/

Br H


ownloaded from


Weismuller, Clausen, Weller, Richter, Steinmann, Henze, Dormehl, Kochs, Adam, Hombach

Figure 5 Founier phase images from DA-ISPECT of a patient (case 4) with a left l1accessory pathway. The upper panel of the figure shows the long axis view of the leftventricle and the four chamber view during anterograde conduction along the accessor

pathway. The lower panel show the same views after administration of 50 mg ajmalinwhich led to normal conduction. The circle indicates early contraction in the presence cdelta wave in contrast to the contraction pattern during normal conduction.

WI

Figure 6 Fourier phase images from DA-ISPECT of a patient (case 9) with a left

posterolateral accessory pathway found 3 cm distal from the ostium of the coronary sin

during catheter mapping. The figure is arranged like fig 5.

detected by DA-ISPECT Localisatior

accessory pathways from standard 12

surface electrocardiogram according to

cntena proposed by Gallagher et a15

Milstein et a16 was less accurate than

ISPECT.

In two of the three patients with a

tract in the septum (cases 16, 20 and

DA ISPECT was not able to detect thefttion of the accessory pathway.

CONTRACTION PATTERNS IN THE PATIENTSWITH VENTRICULAR TACHYCARDIAS

Ai! Table 3 shows the results in the patient groupstudied by DA-ISPECT during ventriculartachycardia. In two patients (cases 2 and 7)catheter ablation was not completely success-ful, though subsequent induction of the ven-tricular tachycardia was more difficult. Thepatients have not had tachycardia for 12 and13 months respectively.

Figure 9 shows the results of a DA-ISPECT study in patient 2, who had ananterior wall aneurysm. The contractionpatterns from phase analysis during ventricu-lar tachycardia and during sinus rhythm werecompletely different. During ventriculartachycardia the contraction started in themid-ventricle at the anterior part of the sep-tum. This coincided with the origin of thetachycardia found by catheter mapping andwas confirmed by catheter ablation. Figure 10shows a DA-ISPECT study from a patient(case 6) with a posterior wall aneurysm andan ejection fraction of 33% during ventricular

ateral tachycardia. The initial contraction was^, located at the lateral side of the base of the?e, left ventricle. Catheter mapping showed theofthe site of origin of the ventricular tachycardia

with a diastolic potential 100 ms before theQRS complex at the base of the ventricle, butthis lay posteriorly, about 4 cm from the siteof the initial contraction. In another patient(case 7) with a large anterior wall aneurysm(ejection fraction during ventricular tachycar-dia 17%) the site of origin of the ventriculartachycardia determined by catheter mappingwas about 5 cm from the site detected byDA-ISPECT (table 3). The interpretation ofthe DA-ISPECT image was difficult becauselarge areas showed early contraction in thispatient. This was also true of patient 4, whoalso showed poor left ventricular perfor-mance. A second early contracting area(fig 4) was identified from the DA-ISPECTimage.

In five of the patients the site of initial con-traction accorded with the origin of the ven-tricular tachycardia (table 3). The differencesseen in two of the patients will be discussed.The mean difference between the localisationof the focus by catheter mapping and by DA-ISPECT was 2-4 cm (range 1-5 cm).

DiscussionThese are the first clinical experiences to be

1US reported with the new method of double-angulated integrated SPECT. This newmethod of evaluating SPECT data has theadvantage of showing ventricles in any chosen

n of view. Difficulties, limitations, and possiblelead clinical applications will be discussed for eachthe patient group.andDA- CONTRACTION PATTERN IN THE CONTROL

GROUPypass Only 60% of the control patients had a

21), typical contraction pattern. This pattern ofloca- contraction corresponds to physiological

excitation of ventricular myocardium via the

206



j.com/

Br H


ownloaded from



Figure 7 Fourier phaseimages from DA-ISPECTof a patient (case 17) withtwo accessory pathways.The upper set of thepreoperative phase imagesshows no consistent site ofearly contraction (earlycontraction marked bycircles at two differentplaces-left posterolateraland right posterolateral).In the lower part of thefigure the phase imageclearly shows initialcontraction in the rightposterolateral area(indicated by the circle),which can now also beidentified in thepreoperative image (seeshort axis view and longaxis view of the rightchamber). The phaseimages are arranged as infig 1.

His-Purkinje system, as reported for thehuman heart by Durrer et al.27 In the remain-ing control patients no distinct contractionpattern was identified. Even in 13 healthydogs previously studied by DA-ISPECT, twoanimals showed contraction patterns thatwere different from physiological expecta-tions. The contraction patterns in healthyindividuals therefore seem to vary probablybecause of interindividual temporal variationin the electrical excitation of ventricular

Figure 8 Twelve leadsurface electrocardiogramfrom a WPWpatient (case16), whose accessorypathway initially could notbe localised by invasivemeans because of repeatedmechanically inducedatrialfibrillation.Extr, limb leads.

is6

myocardium. This makes it more difficult toevaluate the results from the DA-ISPECT inWPW patients.

CONTRACTION PATTERNS IN WPW SYNDROMEBecause there is no standard contractionpattern in controls, the investigator perform-ing DA-ISPECT needs to know whether thepatient being studied has a delta wave. In theclinical setting this information is alwaysavailable. Each beat in a WPW patient with adelta wave is a combination of normal excita-tion along the His-Purkinje system and ofexcitation along the accessory pathway. Thepropagation of the earliest electrical excitationand ensuing contraction from the ventricularinsertion site of the accessory pathway isslower than the electrical excitation along thephysiological route (His-Purkinje system)which begins later. This makes it difficult toanalyse the contraction patterns of suchfusion beats. In seven of the patients thenormal contraction pattern was seen whenconduction through the accessory pathwaywas blocked (figs 5 and 6). This facilitatedthe analysis. In the WPW patients preexcita-tion was not increased by either pacing theatria or by drugs to broaden the delta wave.Pacing is an invasive procedure. Nevertheless,the effect of atrial pacing should be investigat-ed to assess the diagnostic potential of DA-ISPECT in WPW patients paced from theatria.

CONTRACTION PATTERNS IN VENTRICULARTACHYCARDIACompared with WPW patients preexcitationwas always present during ventricular tachy-cardia and ventricular pacing (which wasperformed in two patients). This facilitatesthe evaluation of the DA-ISPECT results inthe patients with ventricular tachycardia. Wefound a good correlation between the resultsof DA-ISPECT and catheter mappingregarding the site of origin of the ventriculartachycardia (table 3).

During the procedure of catheter mappingthe pathway of slow conduction is localised

207

..........

fil



j.com/

Br H


ownloaded from


Weismiiuler, Clausen, Weller, Richter, Steinmann, Henze, Dormehl, Kochs, Adam, Hombach

Table 3 Correlations between surface electrocardiogram, catheter mapping, and DA-ISPECT during ventnicular tachycardia

Catheter SpatialCase ECG DA-ISPECT mapping Correlation difference (cm)t

1 2/I/S 3/I/S 3/I/S 2 12 2/A/S 2/A/S 2/A/S 2 13 2/I/S 2/I/S 2/I/S 2 14 2/M/S 2/MIS 2/MIS 2 25 2/1/S 21I/S 2/I/C 2 36 2/I/L 2/M/L 2/I/C 0 47 2/AlL 2/M/L 3/MIS 0 5

The abbreviations for area designation are given in fig 4.*Correlation: 2, same area detected by catheter mapping and DA-ISPECT; 1, area detected bycatheter mapping in directly adjacent to the area detected by DA-ISPECT; 0, different areasdetected by catheter mapping and DA-ISPECT.tSpatial difference between sites determined by catheter mapping and by DA-ISPECT.

by recording diastolic potentials and by othercriteria.4 This small amount of electro-physiological activity in severely damagedmyocardium does not lead to contraction.It is assumed that DA-ISPECT identifieshealthy-that is-contractile myocardiumthat shows mechanical activity at the start ofcontraction. This may correspond to the siteat the end of the slow conducting pathway29 afact that may explain the difference betweencatheter mapping and DA-ISPECT results inpatient 6. At present only invasive meanssuch as catheter mapping (feasible in 16%)30or mapping during surgery3' are able to detectthe complete pathway of slow conduction.

In patient 4, who had an ejection fractionof only 17% during tachycardia, large areas ofearly contraction were identified. These mayreflect areas of reduced contractility andcould be the cause of the poor result of DA-ISPECT in this case.

Animal studies are needed to clarify thespecificity ofDA-ISPECT for identifying sitesof excitation and subsequent contraction.

ADVANTAGES AND DISADVANTAGES OFDA-ISPECTAdvantages of the DA-ISPECT in WPWpatients* The results indicate that DA-ISPECT ismore Xexact and reliable than the surfaceelectrocardiogram in localising accessorypathways.* It is non-invasive and though subsequentinvasive procedures cannot be replaced byDA-ISPECT they are probably shortened byreducing the time required for fluoroscopy.* Unlike DA-ISPECT catheter mappingusually identifies the atrial insertion of theaccessory bundle. For catheter ablation ofleft-sided accessory pathways it is imperativeto localise the ventricular insertion sitebecause the atrial and ventricular insertions ofthe bypass tract may differ.'2* DA-ISPECT can provide information inthose rare patients in whom catheter mappingis not suitable.* With DA-ISPECT there is only a littleradiation exposure from the radionuclide.

Advantages of DA-ISPECT in patients withventricular tachycardia* DA-ISPECT takes less time than cathetermapping, it does not need an arterial punc-ture, and it is better tolerated by the patients.* The resolution of the site of origin of theventricular tachycardia is good enough forpreoperative evaluation.* Catheter mapping is contraindicated whena left ventricular thrombus is present.

Disadvantages of DA-ISPECT in WPWpatients* The success of DA-ISPECT is limitedwhen the preexcitation is small (narrow deltawave).

Figure 9 Phase imagesfrom DA-ISPECT of apatient with ventriculartachycardia (case 2). Theupper row shows thecontraction pattern duringsinus rhythm. There isslight contraction in theapex of the left ventricle(anterior wall aneurysm).The lower panel showsphase images duringventricular tachycardia.The contraction startedseptally in the anterior wallof the mid-ventricular area(see fig 4), as indicated bythe circle. The projectionsare arranged as in fig 1.

208

..........

li.. :..

I ..



j.com/

Br H


ownloaded from



Figure 10 Phase imagesfrom DA-ISPECT of apatient with an inferioraneurysm (case 6) duringsinus rhythm (upper row)and during ventriculartachycardia (lower row).In contrast to thecontraction pattern duringsinus rhythm, the initialcontraction during theventricular tachycardia islocated in the middle of theleft ventricle, laterally andin the mid-ventricular area(see fig 3), as indicated bythe circle.

* In subsequent investigations preexcitationcould be increased by pacing or drugs toimprove the results of DA-ISPECT.* The presence and location of multipleaccessory pathways cannot be detected.* Concealed bypass tracts are not detected.33* The functional properties of. accessorypathways cannot be evaluated.

Disadvantages of DA-ISPECT in patients withventricular tachycardias* For data acquisition tachycardia must beinduced by a stimulation catheter.* The tachycardia must be tolerated for20 min. This limits the number of patients inwhom this method is suitable. Attempts toshorten the total time of data acquisition arein progress. The tachycardia rate can bereduced by drugs. This results in a betterhaemodynamic tolerance, as it does inpatients who require catheter ablation.* DA-ISPECT relies on mechanical contrac-tion for the evaluation of excitation. But elec-trical excitation, which is recorded bycatheter mapping, does not always lead tocontraction.* The success of DA-ISPECT is limited inpatients with poor left ventricular perfor-mance.

ConclusionsCatheter mapping and intraoperative map-ping remain the standard methods of localis-ing accessory pathways and the origin ofventricular tachycardias. The results of DA-ISPECT are compared with the results ofthese methods. But in catheter mapping andeven intraoperative mapping spatial discrimi-nation is limited.' The most exact method oflocalisation seems to be successful radiofre-quency ablation,2' because it destroys smallareas.24DA-ISPECT is a new method of eval-

uating cardiac blood pool SPECT. It haslimitations and disadvantages in the evalua-

tion of patients with arrhythmias. Never-theless it is an additional tool to assess thesepatients before invasive procedures. In a fewpatients DA-ISPECT provides additionalinformation that influences treatment (forexample, in patients in whom atrial fibrilla-tion is easily mechanically induced and inpatients with thrombi). To date only a fewpatients have been investigated by DA-ISPECT. As the software becomes availableto other centres, the method could be evalu-ated in more patients and its clinical useful-ness could be determined.

1 Josephson ME, Harken AH, Horowitz LN. Endocardialexcision. A new surgical technique for treatment ofrecurrent ventricular tachycardia. Circulation 1979;60:1430-9.

2 Evans GT, Scheinmann MM. Catheter ablation for con-trol of ventricular tachycardia: A report of the percuta-neous cardiac mapping and ablation registry. PACE1986;9:1391-4.

3 Gallagher Jj. Localization of accessory atrioventricularpathways: what's the "gold standard"? PACE 1987;10:583-4.

4 Josephson ME, Horowitz LN, Waxman HL, SpielmanSR, Untereker WJ, Marchlinksi FE. Role of cathetermapping in evaluation of ventricular tachycardia. In:Josephson ME, ed. Ventricular tachycardia. Mech-anisms and management, Mount Kisco, NY: FuturaPublishing Company, Inc, 1982;309-30.

5 Gallagher JJ, Prittchett ELC, Sealy WC, Kasell J, WallaceAG. The preexcitation syndromes. Prog Cardiovasc Dis1978;20:285-327.

6 Milstein S, Sharma AD, Guiraudon GM, Klein GJ. Analgorithm for the electrocardiographic localization ofaccessory pathways in the Wolff-Parkinson-White syn-drome. PACE 1987;10:555-63.

7 Josephson ME, Horowitz LN, Waxman HL, et al.Sustained ventricular tachycardia of the 12-lead electro-cardiogram in localizing site of origin. Circulation 1981;64:257-72.

8 Kuchar DL, Ruskin JN, Garan H. Electrocardiographiclocalization of the site of origin of ventricular tachycar-dia in patients with prior myocardial infarction. J AmCoil Cardiol 1989;13:893-900.

9 Geffers H, Adam WE, Bitter F, Sigel H, Stauch M.Radionuclide ventriculography I. Fundamentals andmethods. Nucl Med 1978;17:206-10.

10 Adam WE, Tarkowska A, Bitter F, Stauch M, Geffers H.Equilibrium (gated) radionuclide ventriculography.Cardiovasc Radiol 1979;2:161-73.

11 Adam WE, Clausen M, Hellwig D, Henze E, Bitter F.Radionuclide ventriculography equilibrium gated bloodpool scanning-its present clinical position and recentdevelopments. EurJ Nucl Med 1988;13:637-47.

12 Botvinick EH, Frais MA, Shosa DW, et al. An accuratemeans of detecting and characterizing abnormal pat-terns of ventricular activation by phase image analysis.AmJ Cardiol 1982;50:289-98.

209

,P;"!

,--7. k.-1-71%

...L:.- I-;'

i... II ,c,



j.com/

Br H


ownloaded from


Weismuller, Clausen, Weller, Richter, Steinmann, Henze, Dormehl, Kochs, Adam, Hombach

13 Pavel DG, Byrom E, Lam W, Mayer-Pavel C, Swyrin S,Pietras R. Detection and quantification of regional wallmotion abnormalities using phase analysis of equili-brium gated cardiac studies. Clin Nucl Med 1983;8:315-21.

14 Chan W, Kalff V, Mach Donald D, et al. Topography ofpreemptying ventricular segments in patients withWolff-Parkinson-White syndrome using scintigraphicphase mapping and esophageal pacing. Circulation 1983;67:1139-46.

15 Nakajima K, Bunko H, Tada A, et al. Phase analysis in theWolff-Parkinson-White syndrome with surgical provenaccessory conduction pathways: Concise communica-tion. JNuclMed 1984;25:7-13.

16 Dormehl I, Bitter F, Henze E, Adam WE, Weismuller P.An evaluation of the diagnostic efficacy of phase analysisof data from radionuclide ventriculograms in patientswith Wolff-Parkinson-White syndrome. Eur _7 Nucl Med1985;11:150-5.

17 Swiryn S, Pavel D, Byrom E, et al. Sequential regionalphase mapping of radionuclide gated biventriculogramsin patients with sustained ventricular tachycardia: Closecorrelation with electrophysiologic characteristics. AmHeartg7 1982;103:319-32.

18 McCarthy DM, Makler PT, Waxman HL, et al. Fourierphase image during ventricular tachycardia: Relation-ship to endocardial site of origin [abstr]. _7 Am CollCardiol 1983;1:712.

19 Le Guludec D, Sebag C, Bourguignon M, et al. Regionalphase mapping of radionuclide gated biventriculogramsin patients with sustained ventricular tachycardia. In:Aliot E, Lazarra R, eds. Ventricular tachycardias-frommechanism to therapy. Dordrecht/Boston Lancaster:Martinus Nijhoff Publishers, 1987;231-45.

20 Weismuller P, Henze E, Adam WE, Roth J, Bitter F,Stauch M. Parametric imaging of experimentally simu-lated Wolff-Parkinson-White syndrome conductionabnormalities in dogs. Am I Physiol Imaging 1986;1:208-13.

21 Weismuller P, Clausen M, Henze E, et al. Localization ofpremature and of ectopic ventricular depolarization bymeans of a new tomographic radionuclide technique. ZKardiol 1990;79:529-34.

22 Clausen M, Henze E, Weller R, et al. Tomographicradionuclide ventriculography (RNV) with reconstruc-

ted planar views perpendicular to the body axis(ISPECT). Nucl Compact 1990;21:95-9.

23 Jackmann WM, Wang X, Friday K, et al. Catheter abla-tion of accessory atrioventricular pathways (Wolff-Parkinson-White syndrome) by radiofrequency current.NEnglJfMed 1991;324:1605-11.

24 Borggrefe M, Hindricks G,; Haverkamp W, Budde Th,Breithardt G. Radiofrequency ablation. In: Zipes DP,Jalife J, eds. Cardiac Electrophysiology. From cell tobedside. Philadelphia: WB Saunders, 1990;997-1004.

25 Morady F, Scheinman MM, Di Carlo LA, et al. Catheterablation of ventricular tachycardia with intracardiacshocks: results in 33 patients. Circulation 1987;75:1037-49.

26 Clausen M, Weismuller P, Weller R, et al. Double-angulated, integral emission-computed tomography(DA-ISPECT) in radionuclide ventriculography (RNV)with reconstructed planar views oriented according tothe long axis of the heart. Nucd Compact 1990;21:167-70.

27 Durrer D, van Dam RT, Freud GE, Janse MJ, MeijlerFL, Arzbaecher RC. Total excitation of the isolatedhuman heart. Circulation 1970;41:899-912.

28 Coumel P, Attuel P. Reciprocating tachycardia in overtand latent preexcitation. Influence of functional bundlebranch block on the heart rate of the tachycardia. Eur_7Cardiol 1974;1:423-36.

29 El-Sherif N, Mehra R, Gough WB, Zeiler RH. Reentrantventricular arrhythmias in the late myocardial infarctionperiod. Circulation 1983;68:644-56.

30 Kuck K-H, Schluter M, Geiger M, Siebels J.Vollstandiges "Mapping" des Areals langsamer Leitungbei ventrikularen Tachykardien [abstr]. Z Kardiol 1991;80:49.

31 Harris L, Downar E, Micklebourough L, Shaikh N,Parson I. Activation sequence of ventricular tachycardia:endocardial and epicardial mapping studies in thehuman ventricle. JAm Coll Cardiol 1987;10: 1040-7.

32 Jackman WM, Kuck KH, Friday KJ, Lazarra R. Catheterrecordings of atrioventricular pathway activation. In:Zipes DP, Jalife J, eds. Cardiac electrophysiology. Fromcell to bedside. Philadelphia: WB Saunders, 1990;491-502.

33 Neuss H, Schlepper M, Thormann J. Analysis of re-entrymechanisms in three patients with concealed Wolff-Parkinson-White syndrome. Circulation 1975;51 :75-81.

210



j.com/

Br H


ownloaded from


non-invasivethree-dimensional wolff-parkinson-white · in ninepatients with wolff-parkinson-white...

Documents