Augmented preexcitation assessed by scintigraphic phase analysis during atrial pacing

Augmented preexcibtion ames

scintigraphic phase analysis during atriai pming

We sought to characterize the effect of augmented preexcitation on the phase image pattern associated with scintigraphic acquisition during conduction via accessory arteriovenous connections. For this reason we assessed phase image scintigraphy, acquired in sinus rhythm and during rapid atrial pacing in 12 patients with documented right (five patients) or left (seven patients) lateral accessory pathways. Augmented preexcitation during atria1 pacing was documented at eiectrophyslologic study in all patients during atrial pacing at similar rates. Phase analysis was abnormal in only 8 patients during sinus rhythm but in all 12 patients during atria1 pacing. Atrial pacing brought a significant delay in both mean left and right ventricular phase angles, LVg and RVg, respectively. With atrial pacing, the site of earliest phase angle, interpreted to indicate the site of earliest excitation, shifted to the site of the accessory pathway. There was increased relative “prematurity” of the mean phase angle of the ipsiiateral ventricle and an absolute increase in the difference between mean and earliest left and right ventricular phase angles, A$ (LV-RV) and A& (LV-RV), respectively. In patients with right-sided pathways, Aq (LV-RV) increased from 9.5 * 12.8 degrees to 47.9 f 22.8 degrees, whereas A& (LV-RV) increased from 28.1 1. 18.0 degrees to 87.8 f 25.0 degrees (both p < 0.05). Patients with left-sided pathways demonstrated similar changes in which A$ (LV-RV) decreased from 2.9 ? 10.8 degrees to -28.5 + 9.0 degrees and A#J, (LV-RV) decreased from 3.4 + 14.2 degrees to -27.4 + 17.9 degrees (both p < 0.05). On the other hand, the site of latest phase angle, or fusion, shifted away from the bypass pathway and the preexcited ventricle. Scintigraphic phase analysis, as eiectrophysiologic study, reveals dynamic changes with augmented preexcitation during atria1 pacing. The parallel in image and electrophysiologic studies provides a physiologic basis for the scintigraphic method, which seeks to localize electrophysiologic foci on the basis of the mechanical sequence. This study makes no effort to assess clinical utility of the phase method. Yet it does demonstrate that phase analysis may be nondiagnostic or ambiguous in sinus rhythm, in the absence of maximum preexcitation. Scintigraphy may complement electrophysiologic catheter mapping in the evaluation of accessory pathways in patients with preexcitation. Atrial pacing may aid the scintigraphic assessment of preexcitation and locallration of the accessory pathway in some patients. (AM HEART J 1987;114:738.)

E. Botvinick, M.D.,* N. Schechtmann, M.D.,** M. Dae, M.D., M. Scheinman, M.D., J. Davis, M.D., J. Herre, M.D., T. Iskikian, M.D., and J. Abbott, M.D. San Francisco, Calif.

The phase image is one of several functional images derived from the equilibrium blood pool scinti-

From the Departments of Medicine, Cardiovascular Division, and Radiol- ogy, Section of Nuclear Medicine, and the Cardiovascular Research Institute at the University of California San Francisco.

Supported in part by a grant from the Fannie Rippel Foundation, Madison, N.J.

Received for publication March 6, 1987; accepted April 20, 1987. Reprint requests: Elias H. Botvinick, M.D., Departments of Medicine and Radiology, Nuclear Medicine Section/Cardiology Division, Room M 1166, University of California San Francisco, San Francisco, CA 94143.

*Performed while an Established Investigator of the American Heart Association. *Supported in part by a grant from the George D. Smith Fund, San Francisco, Calif.

**Attending cardiologist at Ramos Mejia Hospital, Buenas Aires, Argenti- na; performed this work while a postdoctoral fellow at the University of California San Francisco.

gram.‘m6 A product of the first harmonic transform of the regional time vs radioactivity curve, the phase angle can be taken as the local estimate of the relative onset of sequential contraction. The phase image appears to relate to the sequential pattern of ventricular contraction and has been shown to corre- late with the extent of wall motion’-‘* and, indirectly, the ventricular conduction pattern.11,12 The se- quence of ventricular phase angle has been shown to accurately reflect the sequence of ventricular activa- tion in patients with right and left bundle branch block”, l2 and those with left anterior fascicular block.13 The site of earliest phase angle appears to correlate generally with the location of the pacing electrode in patients with artiCcial ventricular pace- makersi4*15 and with the site of earliest electrical

738

Volum* 114

Number 4, Part 1 Scintigraphic phase analysis in WPW 739

activation in patients with sustained ventricular tachycardia16-18 and preexcitation via accessory arta- riovenous (A-V) pathways.1g-23 These same studies document the relationship of scintigraphic findings with those of the surface ECG and electrophysiolog- ic findings in patients with ectopic sites of ventricu- lar activation.14-23 Preliminary animal studies con- firm the relationship between sequential phase angle and the pattern of atria1 activation in normal dogs at rest, with atria1 or ventricular pacing.24 Blood pool scintigraphy with phase analysis appears to be of increased clinical value to aid localization of the bypass pathway, especially in patients with preexcitation who are to undergo catheter or surgi- cal ablation.

In the electrophysiology laboratory, rapid atrial pacing, maximizing antegrade conduction and preexcitation via accessory A-V pathways, and rapid ventricular pacing, with mapping of retrograde atria1 activation, aid diagnosis and localization.25 We performed atrial pacing in association with blood pool scintigraphy in an effort to aid identification of the site of the bypass pathway in a patient not preexcited or only minimally preexcited in sinus rhythm. In this patient, reported in our initial study,lg the phase image changed dramatically with atria1 pacing. The image changes appeared to paral- lel those expected electrophysiologically. Such a correlation between dynamic image and electro- physiologic changes induced with atria1 pacing in patients with preexcitation via accessory A-V path- ways would provide evidence of a significant physio- logic basis of phase imaging. For this reason we sought to assess the ability of phase image analysis to demonstrate changes in the degree of preexcita- tion. We also sought the potential value of atria1 pacing to the scintigraphic assessment of preexcita- tion. To this end we objectively and blindly analyzed the phase images generated from blood pool studies, acquired in sinus rhythm and with rapid atria1 pacing, in 12 selected patients with Wolff-Parkin- son-white syndrome (WPW) and free wall path- ways. Each patient underwent imaging for clinical reasons in an effort to aid localization of the acces- sory A-V connection.

METHODS

General observations. We evaluated 12 patients with WPW among 40 who were studied by equilibrium blood pool scintigraphy and phase analysis to optimally localize the accessory pathway. The group was composed of select- ed patients with known left and right lateral accessory pathways who underwent imaging both in sinus rhythm and during rapid atria1 pacing in an effort to augment electrophysiologic findings related to preexcitation.

Patients demonstrated to have right or left lateral path- ways were specifically chosen because these are most apparent on phase analysis and would tend to maximize the image variation between the normal and preexcited pattern.

There were six men and six women with a mean age of 38 years. Each patient was assessed by electrode catheter mapping within 48 hours of imaging. Electrophysiologic date were acquired objectively and analyzed blindly before patient selection. Similarly, blood pool scintigrams and derived-phase image findings were independently and blindly assessed before selection with regard to their consistency with the documented ECG and electrophysio- logic findings in sinus rhythm and their alterations with atria1 pacing.

General relationships. Electrophysiologic studies were performed in all patients as clinically indicated for the evaluation of symptoms related to preexcitation. They were evaluated by observers different from those inter- preting phase image data, each blinded to the results of the other. In all cases scintigraphy was analyzed without any knowledge of the clinical history or electrophysiologic findings. A 12-lead ECG was always obtained at the time of imaging in sinus rhythm and also during atria1 pacing. These were compared with ECGs acquired during electro- physiologic evaluation to confirm identical conduction patterns during both studies. ECGs aided, as well, assess- ment of the effects of atria1 pacing on antegrade conduc- tion through the accessory pathway. The six standard limb leads and a modified V5 lead were monitored during imaging to document the presence of the conduction abnormality and the stability of cardiac rhythm and conduction during the acquisition period.

All electrophysiologic studies were performed after discontinuation of antiarrythmic medications for at least five half-lives. Four quadripolar electrode catheters were inserted into the femoral and subclavian veins and posi- tioned against the high lateral right atrium, across the tricuspid valve, at the right ventricular apex and in the coronary sinus. Standard surface leads Vl, I, and III and intracardiac electrograms were displayed simultaneously and recorded on an Electronics for Medicine VR-12 or VR-16 recorder (Pleasantville, N.Y.) at paper speeds of 100 to 200 mm/set.

Programmed electrical stimulation was performed at twice the current required to capture the atrium or ventricle in late diastole with a pulse duration of 2 msec using a DTU-101 Bloom and Associates stimulator (Red- ding, Pa.). Incremental right atrial, left atria1 (coronary sinus), and right ventricular pacing were performed between cycle lengths just below the normal sinus R-R interval and a paced cycle length of 275 msec to assess antegrade and retrograde conduction properties. Single atria1 or ventricular extrastimuli were introduced during pacing at a cycle length of 500 msec and decremented in 10 msec intervals throughout diastole until refractoriness was achieved, permitting determination of refractory peri- ods of the atria, A-V node, accessory pathway, and right ventricle. Atria1 and ventricular overdrive pacing and

740 Botvinick et al. October 1987

Amerkcm Heart Journal

Fig. 1. Effects of atria1 pacing on the ECG. Shown are brief excerpts from a single limb (one or two) or precordial (V) lead monitored at rest during sinus rhythm (REST) and during atria1 pacing (PACE). Although a short PR interval, delta wave, and broad QRS interval are seen in some tracings in sinus rhythm, a broadened QRS complex with evidence of a delta wave were obvious in all cases of atria1 pacing. Enhanced preexcitation during atria1 pacing at similar rates was confirmed at electrophysiologic study.

scanning with progressively more premature atria1 or ventricular extrastimuli were performed during induced orthodromic atrioventricular reentrant tachycardia, per- mitting further characterization of cardiac refractory peri- ods and A-V conduction intervals. Additionally, the site of early atria1 activation during orthodromic tachycardia or right ventricular pacing was determined by evaluation of the retrograde atria1 electrogram. Left atria1 sites were mapped by moving a 6F quadripolar USC1 catheter to multiple sites within the coronary sinus, and the right atrium and tricuspid annulus were mapped using a 5F bipolar stearable Brockenbrough catheter. The atria1 insertion of the accessory pathway was taken to be the site with the shortest ventriculoatrial interval.

Scintigraphy. Phase image analysis was performed on 28 frame equilibrium multiple-gated blood pool scinti- grams acquired using as standard 37 phototube, Searle Pho Gamma 5 scintillation camera, portable Ohio Nuclear Series 120, or Siemens LEM (Siemens Corp., Iselin, N.J.) cameras using a linear, all-purpose, 20-degree slant-hole collimator. Studies were processed on a Digital Equip- ment PDP-11/40 minicomputer (Marlboro, Mass.). All patients underwent imaging during sinus rhythm at rates

varying from 47 to 96 and during atria1 pacing at the highest conducted response revealing maximal preexcita- tion and compatible with hemodynamic stability and patient comfort. Pacing rates varied from 100 to 170 bpm. The timing of the gating signal was noted by a mark made on the surface ECG at the onset of each scintigraphic cycle. In 10 patients, imaging was conducted according to our standard methodology,‘2~14~26~27 in the anterior, “best septal,” and ‘IO-degree left anterior oblique projections, whereas in two patients the 78degree projection was omitted because of time constraints. These projections permitted the triangulation of the pattern of scintigraphic findings. Data were displayed in a 128 by 128 format, using 256 gray shades, as described previously.1g

Regional left ventricular wall motion was blindly and objectively assessed from blood pool images in all projec- tions according to a semiquantitative method based on outlines generated from end-diastolic and end-systolic images reported previously.26a27 Normal left and right ventricular ejection fractions, calculated according to standard methods, were 55 % or greater or 45 % or greater, respectively.26s 27

Phase image analysis. In each study, phase image analysis was performed using the fundamental Fourier harmonic applied to the first 25 frames of the blood pool study according to our standard method,‘2,‘4~1g which minimizes the effects of varying relative diastolic filling periods and varying heart rates on calculated phase angle.28 The phase image was displayed and analyzed in steps as reported previously,1g in which the phase angle was gray scale coded from black to white. The gray scale was rotated initially through -54 degrees (0.3 radians) and subsequently through -90 degrees (0.5 radians) to prevent a black-white interface in the ventricular region of interest when imperfect curve fitting methods resulted in ventricular phase angles of less than zero degrees. When gating was triggered by the R wave spike the ventric- ular regions appeared in black and dark gray and atria1 regions appeared in light shades, whereas uncoordi- nated regions of background appeared in varying gray shades.

Phase histograms. To gain temporal resolution intrin- sic to the method we constructed a phase histogram of each ventricle, relating phase angle on the abscissa to the number of pixels at each phase angle on the ordinate, using data obtained from the “best septal” left anterior oblique image. The end-diastolic outline was superim- posed on the composite phase image and used as an objective delineation of ventricular regions of interest. Movable histogram cursors were used to highlight pixels of any selected interval of phase angles as small as 2.8 degrees and permitted localization of the site of earliest phase angle and its progression within each ventricle. Within the ventricular regions of interest, mean left and right ventricular phase angles and their difference were calculated. As applied previously,‘g ventricular regions of interest were redrawn to exclude from analysis all pixels with phase angles larger than the angle at which phase histogram frequency for both ventricles fell below 5% of the peak value. Similar ventricular regions of interest were

Volume 114

Number 4. Part 1 Scintigraphic phase analysis in WPW 741

Fig. 2. Effects of rapid atria1 pacing on phase analysis-left accessory A-V connection. A, Sinus rhythm shows the phase image acquired at rest in the “best septal” left anterior oblique projection in a patient with a documented left lateral A-V connection (LT WPW-REST). At left the homogeneous dark gray shade of the ventricles, consistent with early symmetric ventricular contraction and conduction, is apparent. Below, phase histograms plot phase angle on the abscissa vs incidence on the ordinate for the left ventricle (white) and right ventricle (black). In this projection the right ventricle reveals phase angle slightly preceding that of the left with a slight right-sided delay as well. Progressive white highlights applied to the phase image from left to right, parallel the phase angle interval indicated as a gray panel superimposed on the histograms below each image and document the spatial progression of phase angles. It appears that the site of earliest phase angle is in the normal septal region and progresses symmetrically to both ventricles. B, Atria1 pacing. Shown according to the same format are phase images and histograms in the same patient studied during atria1 pacing (LT WPW-PACED). Compared with the resting study both ventricles demonstrate a lighter gray shade and absolute increased mean phase angle, likely as a result of the effects of increased rate on the fitted time vs radioactivity curve. However, there is increased relative prematurity of the ipsilateral ventricle, because the mean left ventricular phase angle and left ventricular histogram greatly precedes that of the right. The site of earliest phase angle in relation to the ECG R wave gating signal is now clearly localized to the left ventricular lateral wall, the site of insertion of the bypass pathway.

approximated in other projections to assess the direction of phase angle progression.

The sights of earliest left ventricular phase angle and the pattern of sequential phase angle progression were evaluated in the anterior projection in anterior and apical regions, whereas basal, posterolateral, apicoinferior, and septal regions were assessed in the left anterior oblique projection. Sequential right ventricular phase variation was assessed in the left anterior oblique projections in septal, inferoapical, anterolateral, and basal regions. Although overlap of the ventricles prohibits complete phase analysis in anterior and 70-degree left anterior oblique projections, knowledge of the relations between regions of the ventricular perimeter permit triangulation of regions based on a comparison of phase angles. The anterior and 70-degree left anterior oblique projections

added localizing information along the ventricular long axis, complementing data in the best septal projection along the ventricular short axis.

Statistical analysis. Because the distribution of phase angles in ventricular regions of interest is nonparametric,‘* phase angle data were analyzed nonparametrically, and intrapatient comparisons of ventricular phase angles were performed by the Wilcoxon paired-sample test. Numeric values are presented plus or minus standard deviations. Intragroup comparisons were made by the unpaired t test.

RESULTS

Wall motion and right and left ventricular ejec- tion fraction were normal in all patients. Resting

742 Botvinick et al. October 1987

Amarlcan Haart Journal

Fig. 3. Effects of rapid atrial pacing on phase analysis-right accessory A-V connection. A, Sinus rhythm shows the phase image acquired at rest in the “best septal” left anterior oblique projection in a patient with a documented right lateral A-V connection (RT WPW-REST). At left the homogeneous dark gray shade of the ventricles, consistent with early symmetric contraction and conduction, is apparent. Below, phase histograms plot phase angle on the abscissa vs incidence on the ordinate for the’left ventricle (white) and right ventricle (black). In this projection the right ventricle reveals phase angle slightly preceding that of the left. Progressive white highlights applied to the phase image from left to right, parallel the phase angle interval indicated as a gray panel superimposed on the histogram below each image and document the spatial progression of phase angles. It appears that the site of the earliest phase angle is in the normal septal region and progresses symmetrically to both ventricles. B, Atria1 pacing. Shown according to the same format are images and histograms in the same patient studied during atria1 pacing (RT WPW-PACED). Compared with the resting study, and similar to the image effects of pacing in association with the left pathway shown above, both ventricles demonstrate a lighter gray shade and absolute increased mean phase angle, likely as a result of the effects of increased rate on the fitted time vs radioactivity curve. However, there is increased relative prematurity of the ipsilateral ventricle, because the mean right ventricular phase angle and right ventricular histogram greatly precedes that of the left. The site of earliest phase angle is now localized to the right ventricular lateral wall, the site of insertion of the bypass pathway.

heart rate at the time of study ranged from 47 to 96 bpm and A-V conduction was always 1:l during atrial pacing, which was maintained at rates of 100 to 170/minute. In each case there was complete agreement by two blinded observers of the sight of earliest phase angle and its variation in the same patient with atrial pacing. Previous analysis con- ducted by two blinded observers on two separate occasions demonstrated an interobserver variation in left ventricular mean phase angle of 3 degrees.is

There were five right and seven left lateral con- nections. Eleven patients revealed ECG and electro- physiologic evidence of preexcitation in normal sinus rhythm. Rapid atria1 pacing increased and tended to maximize the degree of preexcitation as the mean ECG QRS duration increased from 100 to 150 msec (Fig. l), associated at electrophysiologic

study with an obvious reduction in the HV interval. Similarly, pacing increased the relative prematurity of the preexcitation focus.

In sinus rhythm the phase image of each ventricle appeared homogeneous, but preexcitation was apparent because aspects of the ipsilateral ventricle were darker than those of the contralateral ventricle in eight patients. In these eight patients the sight of earliest phase angle always involved the appropriate region of the lateral aspect of the ipsilateral ventri- cle, often in obvious contrast to the light gray shade of the contralateral ventricle (Figs. 2 and 3). The site of the earliest phase angle was localized laterally in all cases to the site of insertion of the preexcitation pathway. Associated with this visual impression there was an absolute increase in the differences between mean and earliest left and right ventricular

Volume 114

Number 4, Part 1 Scintigraphic phase analysis in WPW 743

Q

t

R!3 LEFT

220

140 1

120 t

100

80

60

40 i

-60 t l \ -\\

ATRIAL PACED RlGHT LEFT

//’ J

/ i’

, n=5 , , n=77 ,n=5 , , n=7,

LV$ RV$ LV$ RV~ LV$ RV% LV6 RV$

Fig. 4. Alteration of mean right and left ventricular phase angles with atrial pacing. Shown are the mean left and right ventricular phase angles LV $ and RV $, respectively, at rest and with atria1 pacing (ATRIAL PACED) in patients with right and left A-V connections. Although mean values generally increase with atria1 pac- ing, regardless of the location of the bypass pathway, the large absolute difference in mean ventricular phase angles reflects the increased level of preexcitation.

phase angles, Ai$ (LV-RV) and & (LV-RV), respec- tively. In patients with right-sided pathways, A$ (LV-RV) increased from 9.5 f 12.6 degrees to 47.9 & 22.8 degrees, whereas A& (LV-RV) increased from 28.1 & 18.0 degrees to 67.6 f 25.0 degrees (both p < 0.05). Patients with left-sided pathways demonstrated similar changes in which A$ (LV-RV) decreased from 2.9 rl: 10.8 degrees to -26.5 + 9.0 degrees and A0 (LV-RV) decreased from 3.4 -t 14.2 degrees to -27.4 * 17.9 degrees (both p < 0.05) (Figs. 4 and 5). On the other hand, the site of the latest phase angle, or fusion, shifted away from the preexcited ventricle (Figs. 2 and 3).

DISCUSSION

Phase analysis is based on a mathematic approxi- mation of the time vs radioactivity curve. When applied to the identification of abnormalities of

100

80

60

40

20

0

-20

-40

-60

L-

RIGHT LEFT RIGHT LEFT

:?

n=5 n=7 n=5 n=7 1

R AP R AP R AP R AP

A$ (LV-RV) A* ONSET (LV- R’4

Fig. 5. Alteration in mean phase angle differences with atria1 pacing. Plotted are the differences in mean and earliest left and right ventricular phase angle A$ (LV-RV) and A& (LV-RV), respectively, at rest (R) and with atrial pacing (AP) in patients with documented right or left lateral A-V connections. The increasing difference between right and left ventricular values for these param- eters again provides objective phase image evidence of augmented preexcitation with atrial pacing.

ventricular excitation and conduction, the method is based entirely on a mathematic estimate of sequen- tial mechanical alterations during the cardiac cycle. Yet the method has already demonstrated an ability to characterize a number of conduction patterns.11-24 Influenced as well by abnormalities of contraction, difficulties may be expected when the method is applied to the delineation of conduction patterns in the presence of severe contraction abnormalities. An added theoretical difficulty relates to potential errors in phase analysis as a result of electromechan- ical dissociation and the possibility that an electrical focus may originate from a mechanically inactive region. Yet, even in the setting of ventricular tachy- cardia, a conduction abnormality most likely related to such dissociation, phase analysis has demon- strated a good correlation with electrophysiologic localization.*6-1* In this study the phase method was applied to the localization of the initial focus of electrical activation in the absence of severe contrac- tion abnormalities or evidence of electromechanical dissociation.

The phase analytic method used here minimizes

744 Botuinick et al. October 1087

Amdcan Ha& Journal

errors introduced by curve asymmetry and irregu- larities of heart rate and was the same applied in previous reports. It was applied here to permit comparison with prior data analyzed in our labora- tory. However, the most logical analytic method is based on selective analysis of cardiac cycles of similar duration permitting inclusion of all 28 frames.2g’30 Although we currently employ this latter method, we see no substantive difference between the findings presented and those resulting from such alternate analysis.

We did not seek to reevaluate scintigraphic accu- racy for bypass localization. Rather it was our purpose to simply and objectively evaluate the dynamic effects of rapid atria1 pacing on the scinti- graphic phase image in patients with preexcitation via A-V accessory pathways and correlate them with electrophysiologic findings. Although prior reports by us and others revealed group differences in phase image characteristics between normally conducting and preexcited populations, the patterns noted in individual cases varied widely, likely a function of the degree of preexcitation. Chan et aL31 reported the accuracy of phase analysis for localization of accessory pathways in patients studied during esophageal pacing, and we had reported a patient studied with atrial pacing.lg However, no compari- son was made of image findings in sinus rhythm and with atria1 pacing. We selected patients with known right or left ventricular lateral pathways. These are generally prominent on phase analysis and would, because of their distance from the septum, provide the most dramatic shifts in the related phase image and best permit visualization of pacing-induced image changes. Although patients were selected retrospectively on the basis of known electrophysio- logic findings, all scintigraphic and electrophysiolog- ic evaluation had been previously performed objec- tively and blindly by different observers without knowledge of other study results.

In these patients with A-V connections, as in those reported previously,1g-23 the sight of earliest phase angle and the pattern of phase progression correlated well with the focus of preexcitation and the pattern of ventricular activation documented on electrophysiologic study. However, among those studied here, not all were preexcited at rest and none were preexcited maximally; the preexcitation focus was most sensitively identified by scintigraph- ic phase analysis only during atrial pacing. This is itself not surprising. The principle of rapid atria1 pacing is used in the electrophysiology laboratory to encourage conduction through the accessory path-

way, augmenting preexcitation as a result of the relatively greater refractoriness of the normal con- duction pathway.

Two imaging studies were unavoidably gated off the atria1 pacing spike. Although this undoubtedly increased values for mean ventricular phase angle, there is no reason to believe that it influenced the sequence of phase progression or interventricular or intraventricular phase differences. Further, the phase pattern in these patients correlated well with electrophysiologic data and was not different from the group as a whole. Although it may not always be possible to avoid triggering the gating signal by the atria1 pacing spike, it can be discouraged by select- ing a lead for gating that minimizes the pacing spike and maximizes the R wave. In any case, the phase sequence will be reliable regardless of the gating signal, providing that signal is consistently sensed.

Scintigraphic alterations recorded during atrial pacing appeared to parallel those recorded electro- cardiographically and electrophysiologically as the result of the increased level of preexcitation. While a rate increase may increase phase values, it should not greatly increase resting ventricular phase differ- ences. Rate changes alone could not produce phase patterns that mirror electrical patterns, nor could they shift the site of earliest phase angle or the pattern of phase progression. This serves as an important example of the ability of the phase meth- od to reflect altered patterns of myocardial conduc- tion. It provides evidence of the ability of the imaging method to accurately reflect the shifting pattern of electrical conduction. The results also support the presence of electromechanical associa- tion in these cases. Although these observations provide a strong physiologic basis for the phase imaging method, it cannot, of course, be extrapo- lated to cases with significant ventricular dysfunc- tion and possible electromechanical dissociation, as patients studied by phase imaging in ventricular tachycardia.

There are additional practical clinical implica- tions. In a significant proportion of patients with accessory A-V connections who may be minimally or not at all preexcited, phase analysis at rest may be nondiagnostic or ambiguous. Here pacing interven- tion appears necessary for both the scintigraphic and electrophysiologic identification of the site of the preexcitation pathway. Such noninvasive scinti- graphic evaluation and localization of the bypass pathway may be a useful complement to electro- physiologic study, especially in patients with preex- citation being considered for surgical resection or

Volume 114

Number 4, Part 1

catheter ablation. The clinical utility of phase imag- ing in relation to the evaluation of preexcitation or other conduction abnormalities has not yet been established. The need for atrial pacing to optimize image findings in some of these patients may affect clinical utility.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

Verba JW, Bornstein I, Alazraki NP, et al. Onset and progression of mechanical systole derived from gated radio- nuclide techniques and displayed in tine format. J Nucl Med 1980;20:625. Pave1 D, Sweryn S, Lam W, et al. Ventricular phase analysis of radionuclide gated studies. Am J Cardiol 1980;45:398.- Maddox DE. Holman BL. Wvnne J. et al. Eiection fraction image: a nodinvasive index o> regional left Ventricular wall motion. Am J Cardiol 1978;41:1230-8. Holman BL, Wynne J, Idoine J, et al. The paradox image: noninvasive index of regional left ventricular dyskinesis. J Nucl Med 1979;20:661. Gandsman EJ, North DL, Shulman RS, et al. Measurement of the ventricular stroke volume ratio by gated radionuclide angiography. Radiology 1981;138:161-5. Dae MW, Engelstad BL, Botvinick EH, et al. Parametric imaging in cardiovascular nuclear medicine. Cardiovasc Rev Rep 1985$X6-27. Botvinick E, Dunn R, Frais M, et al. The phase image: its relationship to patterns of contraction and conduction. Circu- lation 1982;65:551-60. Turner DA, Shima MA, Ruggie N, et al. Coronary artery disease: detection of phase analysis of rest/exercise radionu- elide angiograms. Radiology 1983;148:539-45. Norris SL, Slutsky RA, Gerber KH, et al. Sensitivity and specificity of nuclear phase analysis versus ejection fraction in coronary artery disease. Am J Cardiol 1984,53:1547-52. Schwaiger M, Ratib 0, Henze E, et al. Limitations of quantitative phase analysis of radionuclide angiograms for detecting coronary artery disease in patients with impaired left ventricular function. AM HEART J 1984;108:942-9. Swiryn S, Pave1 D, Byrom E, et al. Sequential regional phase mapping of radionuclide gated biventriculograms in patients with left bundle branch block. AM HEART J 1981;102:1000- 10. Frais MA, Botvinick EH, Shosa DW, et al. Phase image characterization of ventricular contraction in left and right

- bundle branch block. Am J Cardiol 1982;50:95-105. Dae M. Wen YM. Botvinick E. et al. Assessment of left anterior fascicular block by &&graphic phase analysis. J Am Co11 Cardiol 1984;3:591. - - Botvinick EH. Frais MA. Shosa DW. et al. An accurate means of detecting and characterizing abnormal patterns of ventric- ular activation by phase image analysis. Am J Cardiol 1982;50:289-98.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

Scintigraphic phase analysis in WPW 745

Bashore TM, Stine RA, Schafl’er PB, et al. The noninvasive localization of ventricular pacing sites by radionuclide phase imaging. Circulation 1984;?0:68I-94. Botvinick E. Schechtmann N. Dae M. et al. Scintimanhv provides a thorough evaluation of “electrical” and m&ham- cal events during ventricular tachycardia. J Am Co11 Cardiol 1986;7:235A. McCarthy DM, Makler PT Jr, Waxman HL. Fourier phase image during ventricular tachycardia: relationship to endo- cardial site of origin. J Am Cob Cardiol 1983;1:712. Swirvn S. Pave1 D. Bvrom E. et al. Seauential regional phase mappingof radionuciide gated bivent&ulogra& in patients with sustained ventricular tachycardia: close correlation with electrophysiologic characteristics. AM HEART J 1982,103:319- 32. Botvinick E, Frais M, O’Connell W, et al. Phase image evaluation of patients with ventricular preexcitation syn- dromes. J Am Coll Cardiol 1984,3:799-814. Nishimura Y, Yamamuro M, Kooishi T, et al. Visualization of Kent bundle position in WPW syndrome using phase analy- sis. J Nucl Med 1985;26:58. Nakajima K, Bunko H, Tada A, et al. Nuclear tomographic phase analysis: localization of accessory conduction pathways in patients with Wolff-Parkinson-White syndrome. AM HEART J 1985,109:809-15. Nakajima K, Bunko H, Tada A, et al. Phase analysis in the Wolff-Parkinson-White svndrome with sureicallv oroven accessory conduction pathways: concise co&nu&ation. J Nucl Med 1984;25:7-13. Dae M, Botvinick E, Schechtmann N, et al. Correlation of scintigraphic phase maps with intraoperative epicardial/ endocardial maps and surgical results in patients with activa- tion disturbances. Circulation 1984; 70 (suppl II): 11-371. Dae M, David J, Botvinick E, et al. Comparison of scinti- graphic phase maps to epicardial activation maps in dogs. J Nucl Med 1985;26:57. Gallagher JJ, Pritchett ELC, Sealy WC, et al. The pre- excitation svndromes. Pros Cardiovasc Dis 1978:20:285-327. Rahimtoola”S, Botvinick E, Perez-Gonzalez J, et al. Biplane geometric ventricular volumes from combined first-pass equi- librium radionuclide ventriculography. Circulation 1979; 59, 60 (suppl II): 11-168.

27. Greenberg BH, Drew D, Botvinick E, et al. Evaluation of left ventricular performance by gated radionuclide angiography. Clin Nucl Med 1980;60:581-7.

28. O’Connell W, Shosa D, Frais M, et al. Symmetry violation errors in first Fourier harmonic phase analysis. Am J Cardiol 1982;49:1045.

29. Pave1 DG, Briandet PA. Quo vadis phase analysis. Clin Nucl Med 1983;8:564-9.

30. Goris ML. Function or parametric images. J Nucl Med 1982;23:360.

31. Chan WWC, Kalff V, MacDonald D II, et al. Topography of preemptying ventricular segments in patients with Wolff- Parkinson-White syndrome using scintigraphic phase map- ping and esophageal pacing. Circulation 1983;67:1139-46.