- Email: [email protected]
Measuring spatial waves of repolarization in canine ventricles using high-resolution epicardial mapping
Journal of Electrocardiology Vol. 29 Supplement
Measuring Spatial Waves of Repolarization in Canine Ventricles Using High-resolution Epicardial Mapping
Robert
L. L u x , P h D , P h i l i p R . E r s h l e r , P h D , a n d B r u n o
Taccardi, MD, PhD
Abstract: The importance of the role of ventricular repolarization in arrhyth-
mogenesis and defibrillation prompted the exploration of n e w methods for observing and measuring repolarization. Specifically, the authors' goal was to establish independent procedures for assessing activation-recovery intervals. Canine epicardial electrograms from high-resolution arrays (2-ram spacing, 25 x 21 electrodes) were recorded during pacing from a variety of single or simultaneously paced epicardial locations in canine hearts. For each activation sequence, the activation and repolarization times were measured using timing of intrinsic QRS and T wave deflections (activation-recovery interval method) and timing of the peak magnitude of spatial derivatives (gradient method). Both methods should, theoretically, provide estimates of local activation and repolarization times, which reflect timing of local action potential upstrokes and downstrokes. Scattergrams comparing activation and recovery times for the two methods showed high correlation, slopes close to 1.0, and intercepts near the origin. For most activation sequences, observation of the potential and gradient distributions as dynamic, three-dimensional perspective displays, revealed a well-defined, rapidly propagating repolarization wave, superimposed on a slowly varying, high-amplitude distribution occurring during the T wave. These data suggest that repolarization times measured using temporal or spatial derivatives are consistent with theoretical predictions and reflect timing of local action potential downstrokes. They also suggest potential utility of combining spatial and temporal approaches for improving reliability i~ the measurements. K e y w o r d s : ventricular repolarization, activation-recovery interval, activation sequence.
occurrence, a likely mechanism for increased vulnerability to arrhytbmias lies in the subtle dynamic changes of local repolarization that create, momentarily, regions of localized refractoriness (block) around which excitation propagates but then reenters. To explore such a potential mechanism, it is important to have tools to measure repolarization, its spatial distribution, and its dynamics on a beat-to-beat basis. The classical measure of repolarization, the refractory period, does not meet the requirements for such observations in that it takes tens of beats to make the measurement from one site. Although promising, optical
The measurement of repolarization is of increasing importance in the study of the roles of repolarization, its inhomogeneity, and dynamics in cardiac arrhythmogenesis. In light of the unpredictable nature of arrhythmia
From the University of Utah, Salt Lake City, Ulah.
Supported in part by the Richard A. and Nora Eccles Harrison Fund for Cardiovascular Research, the Nora Eccles Treadwell Foundation, and NIH grant HE52338. Reprint requests: Robert L. Lux, Nora Eccles Harrison CVRTI, Nora Eccles Harrison Building, University of Utah, Salt Lake City, UT 84112.
130
New Methods for Measuring Repolarization methods suffer from poor signal-to-noise ratio, m o d o n artifact, and ability to sample visible sites only (1). Pressure potentials, as used in monophasic action potential recordings, are useful for measuring dynamic changes of repolarization but are ill suited for mapping distributions of repolarization from large numbers of sites (2,3). The activation-recovery interval (ARI) permits m e a s u r e m e n t of repolarization from unipolar electrograms from m a n y sites and across sequences of m a n y consecutive beats (4). Moreover, it has been documented to correlate well with refractory period measurements as well as with measurements of action potential duration obtained from floating microelectrodes located close to the unipolar recording site (5). However, ARIs have not received widespread acceptance, are often equivocal, and have not been confirmed using other techniques. A crucial element of repolarization measurement revolves around the nature of repolarization itself. In the case of cardiac depolarization, the propagation of a wavefront through the myocardium is observed as a moving potential "jump" with a trailing, large negative region and a preceding positive region. The potential waveform transcribed by the passage of the wave is classically described by an initial positivity followed by an intrinsic deflection (rapid negative going deflection) and return toward the reference potential. This scenario is modified, depending on w h e t h e r the wave propagates in the longitudinal or transverse direction, although the intrinsic deflection is always well defined and easily detected in normal tissue. The timing of the fastest part of the downstroke has been well documented to reflect the timing of the passage of the wave that separates the resting or excitable region in advance of the wave from the region behind the wave that has just been depolarized. Moreover, the timing of this event is synonymous with the timing of the greatest action potential upstroke velocity in cells proximal to the electrode. In the case of repolarization, it is not clear that the process can be considered to propagate in a classical sense. Nevertheless, it is well k n o w n that there is a definite sequence of repolarization that is dictated by the actuation sequence, the distribution of intrinsic action potential durations, and electrotonic interaction between cells. As such, repolarization mimics the characteristics of a propagating wave. If, in fact, it behaves as a propagating wave, then it should be measurable using traditional techniques that are sensitive to detecting waves. It is the purpose of this study to present n e w data that compare temporal and spatial methods for measuring both depolarization and repolarization observed from high-resolution arrays of epicardial electrograms. These data document the presence of repolarization waves and confirm, but do not prove, the theory behind the temporal methods for measuring repolarization (ARIs).
Materials and Methods Intact, canine hearts were exposed using conventional surgical techniques. A 25 x 21 array of insulated,
•
Luxet al.
131
5-mil-diameter silver wires sewn into a nylon mesh was applied to the right ventricular epicardial surface. Using a 256-channel amplification and multiplexing system, we recorded two banks of electrograms, each of which included five electrograms c o m m o n to each bank and used for time alignment. Electrograms were recorded for a variety of activation sequences, including those originating from supraventricular pacing as well as those from single or multiple ventricular site epicardial pacing. For each activation sequence, potential distributions were displayed as isopotential contour maps, gradient magnitude maps, and dynamic, threedimensional perspective displays. In addition, depolarization and repolarization times were calculated using two methods: (1) conventional ARIs were m e a s u r e d in which depolarization times were defined as time of m i n i m u m dV/dT of electrograms during the QRS and time of m a x i m u m dV/dT near the m a x i m u m T wave amplitude; and (2) the times of m a x i m u m spatial gradient magnitude during the QRS and T wave were taken as depolarization and repolarization times, respectively. Temporal and spatial measures of activation and repolarization times were compared using linear correlation. Dynamics of potential and gradient distributions were observed using three-dimensional visualization software written for a Silicon Graphics w o r k station (Mountain View, CA).
Results Isopotential distributions and their dynamics during depolarization have been well documented and described. Observation of potential distributions of our data replicate the previous findings, including elongation of isopotential ellipsoids along the fiber direction and rotation of the negative potential ellipsoid and its associated positive extrema consistent with rotation of the fiber orientation with depth (6). During repolarization, the dominant reversal of potential polarity from that during the QRS is reflected in a slowly rising and falling positive potential (upright T waves). Superimposed on this background distribution was a rapidly moving, low-amplitude ripple that emanated from and propagated away from the region of pacing. This observation, to our knowledge, has not been reported previously, and we investigated the ability of activation and repolarization timing measurements based on the timing of the m a x i m u m spatial gradient magnitude during the QRS and T wave to reflect the activation and recovery time measurements from electrogram derivatives. Figure 1 shows the distributions of epicardial isochrones of activation times (top panels) and repolarization times (bottom panels) obtained from the times of m i n i m u m dV/dT (left panels) and the m a x i m u m spatial gradient magnitude (right panels). These distributions reflect measurements from electrograms obtained during pacing from a single epicardial site. Figure 2 shows examples of electrograms, their temporal derivatives, and the magnitude of spatial potential gradients at those
132
Journal of Electrocardiology Vol. 29 Supplement
Temporal Method
Spatial Method
C
.9 O
<
•:
i
E c-
O
Fig. 1. The upper left panel shows isochrone lines determined from the times of m i n i m u m dV/dT in the QRS for epicardial pacing. The upper right panel shows isochrone lines determined as the times of m a x i m u m spatial gradient during the ORS. The lower left panel shows isochrones of repolarization determined as the time of m a x i m u m dV/dT near the peak of the T wave. The lower right panel shows isochrones of repolarization determined as the m a x i m u m times of the spatial gradient during the T wave. All isochrones are drawn at 4-ms increments. The shaded area in the lower right panel indicates regions where the spatial gradient was insufficient to measure a well-defined peak or there were multiple, low-amplitude peaks.
sites. The site n u m b e r s correspond to the n u m b e r e d locations in the isochrone distribution of Figure i (upper left panel) and give examples of waveforms from regions of longitudinal and transverse propagations. From these signals, it is apparent that there is good correspondence between temporal and spatial timings of depolarization and repolarization. Figure 3 shows scattergrams of depolarization times (Ieft panel) and repolarization times (right panel) with the temporal mea-
surement along the ordinate and the spatial (gradient) measurement along the abscissa. Also shown are the linear regression curve, its equation, and the estimated linear correlation coefficient. The characteristics of these scattergrams are similar for all of the activation sequences studied. Typically, for activation timing, the intercept of the regression line was near zero and the slope was near 1 with correlation coefficients higher than .98. For repolarization, the regression line reflected a slight bias (pos-
New Methodsfor Measuring Repolarization
Site 2
Site 3
•
L u x e t al.
133
Site 5 I I
EG
dV/dt
~I~~'~~ ~ ~
-~--~
]i
tvvt
Fig, 2. The n u m b e r e d sites correspond to data at the n u m b e r e d sites in the upper left panel of Figure 1: two in regions of longitudinal propagation and three in regions of transverse propagation. For each site, the electrogram (EG) is shown at the top, its temporal derivative (dV/dT) is shown in the middle, and the magnitude of the spatial gradient (VV) is shown at the bottom. To emphasize the alignment of the spatial and temporal derivatives, a vertical line has been drawn at the time of repolarization in each lead.
~o
6o
210
3E
._~
50
ATv=.38 + .98 A T r=.987
¢~ 40
_~ =~:"
~
|•1
200 190
RT =13.0 + .94 RT W
r=.gS
•
I-
t-i,.•¢ ••
j
I I,,-
i
! "~.~7~ "
..,I 'lIW."
180
30
=~ 20
t
.
170
.t. =.!~! • 160
>~ 0
I
i
10
20
Activation
Time
[
i
30 40 (ms) - T e m p o r a l
I
i
50
60
Method
ill
150 150
I
160
i
I
I
170 180 190 R e c o v e r y Times (ms) . T e m p o r a l
t
I
200
210
Method
Fig. 3. Scattergrams of activation (left) and repolarization (right) cornparing temporal and spatial measurement methods. The regression lines and correlation coefficients for each are shown. AT, activation time; RT, recovery time.
itive offset of several milliseconds) but the slope was very close to 1 and correlation coefficients were higher than .94. Somewhat lower correlation and the spread of points about the regression Iine reflected poorer agreem e n t between the two methods in comparison to that obtained for measuring activation time. This is likely a consequence of a poorer signal-to-noise ratio of T waves relative to QRS waves, which reflects the lower frequency content and smaller currents that produce the observed potentials.
Observation of the depolarization and repolarization waves using dynamic, three-dimensional perspective plots documented the usual propagation of excitation jumps during depolarization, but also exhibited a rapidly moving, low-amplitude wave during repolarization. These waves propagated away from the stimulus site, regardless of the site location and also replicated "collision" w h e n multiple sites were simultaneously stimuiated. For obvious reasons, it is not possible to illustrate these waves easily.
134
Journal of Electrocardiology Vol. 29 Supplement
Conclusion
References
The findings presented here support the interpretation of spatial and temporal derivatives of cardiac potentials as a means to observe not only the classical wavefronts of depolarization, but also those of repolarization. In this study, there was very close a g r e e m e n t b e t w e e n the temporal and gradient methods w h e n applied to depolarization, a finding that confirmed that the traditional m e t h o d for detecting propagating waves was applicable in this setting. The fact that the close correspondence b e t w e e n the two m e a s u r e m e n t s was m a i n t a i n e d w h e n applied to repolarization, albeit not to the same level of agreement, suggests that repolarization can, in practice, be considered a propagating p h e n o m e n o n , Of practical importance, the fact that the spatially d e t e r m i n e d m e a s u r e m e n t s of both depolarization and repolarization closely tracked the temporally d e t e r m i n e d m e a s u r e m e n t s supports previous claims that the ARI technique provides local estimates of activation and repolarization times. Although these studies p r e s e n t e d only data from epicardial surfaces, further studies in the thickness of the ventricular wall are planned.
1. Dillon SM: Optical recordings in the rabbit heart show that defibrillation strength shocks prolong the duration of depolarization and the refractory period. Circ Res 69:842, 1991 2. France MR, Bargheer K, Rafflenbeul W e t al: Monophasic action potential mapping in h u m a n subjects with normal electrocardiograms: direct evidence for the genesis of the T wave. Circulation 75:39, 1987 3. Franz MR: Method and theory of monophasic action potential recording. Prog Cardiovasc Dis 33:347, 1991 4. Millar CK, Kralios FA, L u x RL: Correlation between refractory periods and activation-recovery intervals from electrograms: effects of rate and adrenergic interventions. Circulation 72:1372, 1985 5. Haws CW, Lu× RL: Correlation between in vivo transmembrane action potential durations and activationrecovery intervals from electrograms: effects of interventions which alter repolarization times. Circulation 81:281, 1990 6. Taccardi B, Macchi E, Lux RL et al: Effects of myocardial fiber direction on epicardiaI potentials. CircuIation 90:3076, 1994
Measuring Spatial Waves of Repolarization in Canine Ventricles Using High-resolution Epicardial Mapping
Robert
L. L u x , P h D , P h i l i p R . E r s h l e r , P h D , a n d B r u n o
Taccardi, MD, PhD
Abstract: The importance of the role of ventricular repolarization in arrhyth-
mogenesis and defibrillation prompted the exploration of n e w methods for observing and measuring repolarization. Specifically, the authors' goal was to establish independent procedures for assessing activation-recovery intervals. Canine epicardial electrograms from high-resolution arrays (2-ram spacing, 25 x 21 electrodes) were recorded during pacing from a variety of single or simultaneously paced epicardial locations in canine hearts. For each activation sequence, the activation and repolarization times were measured using timing of intrinsic QRS and T wave deflections (activation-recovery interval method) and timing of the peak magnitude of spatial derivatives (gradient method). Both methods should, theoretically, provide estimates of local activation and repolarization times, which reflect timing of local action potential upstrokes and downstrokes. Scattergrams comparing activation and recovery times for the two methods showed high correlation, slopes close to 1.0, and intercepts near the origin. For most activation sequences, observation of the potential and gradient distributions as dynamic, three-dimensional perspective displays, revealed a well-defined, rapidly propagating repolarization wave, superimposed on a slowly varying, high-amplitude distribution occurring during the T wave. These data suggest that repolarization times measured using temporal or spatial derivatives are consistent with theoretical predictions and reflect timing of local action potential downstrokes. They also suggest potential utility of combining spatial and temporal approaches for improving reliability i~ the measurements. K e y w o r d s : ventricular repolarization, activation-recovery interval, activation sequence.
occurrence, a likely mechanism for increased vulnerability to arrhytbmias lies in the subtle dynamic changes of local repolarization that create, momentarily, regions of localized refractoriness (block) around which excitation propagates but then reenters. To explore such a potential mechanism, it is important to have tools to measure repolarization, its spatial distribution, and its dynamics on a beat-to-beat basis. The classical measure of repolarization, the refractory period, does not meet the requirements for such observations in that it takes tens of beats to make the measurement from one site. Although promising, optical
The measurement of repolarization is of increasing importance in the study of the roles of repolarization, its inhomogeneity, and dynamics in cardiac arrhythmogenesis. In light of the unpredictable nature of arrhythmia
From the University of Utah, Salt Lake City, Ulah.
Supported in part by the Richard A. and Nora Eccles Harrison Fund for Cardiovascular Research, the Nora Eccles Treadwell Foundation, and NIH grant HE52338. Reprint requests: Robert L. Lux, Nora Eccles Harrison CVRTI, Nora Eccles Harrison Building, University of Utah, Salt Lake City, UT 84112.
130
New Methods for Measuring Repolarization methods suffer from poor signal-to-noise ratio, m o d o n artifact, and ability to sample visible sites only (1). Pressure potentials, as used in monophasic action potential recordings, are useful for measuring dynamic changes of repolarization but are ill suited for mapping distributions of repolarization from large numbers of sites (2,3). The activation-recovery interval (ARI) permits m e a s u r e m e n t of repolarization from unipolar electrograms from m a n y sites and across sequences of m a n y consecutive beats (4). Moreover, it has been documented to correlate well with refractory period measurements as well as with measurements of action potential duration obtained from floating microelectrodes located close to the unipolar recording site (5). However, ARIs have not received widespread acceptance, are often equivocal, and have not been confirmed using other techniques. A crucial element of repolarization measurement revolves around the nature of repolarization itself. In the case of cardiac depolarization, the propagation of a wavefront through the myocardium is observed as a moving potential "jump" with a trailing, large negative region and a preceding positive region. The potential waveform transcribed by the passage of the wave is classically described by an initial positivity followed by an intrinsic deflection (rapid negative going deflection) and return toward the reference potential. This scenario is modified, depending on w h e t h e r the wave propagates in the longitudinal or transverse direction, although the intrinsic deflection is always well defined and easily detected in normal tissue. The timing of the fastest part of the downstroke has been well documented to reflect the timing of the passage of the wave that separates the resting or excitable region in advance of the wave from the region behind the wave that has just been depolarized. Moreover, the timing of this event is synonymous with the timing of the greatest action potential upstroke velocity in cells proximal to the electrode. In the case of repolarization, it is not clear that the process can be considered to propagate in a classical sense. Nevertheless, it is well k n o w n that there is a definite sequence of repolarization that is dictated by the actuation sequence, the distribution of intrinsic action potential durations, and electrotonic interaction between cells. As such, repolarization mimics the characteristics of a propagating wave. If, in fact, it behaves as a propagating wave, then it should be measurable using traditional techniques that are sensitive to detecting waves. It is the purpose of this study to present n e w data that compare temporal and spatial methods for measuring both depolarization and repolarization observed from high-resolution arrays of epicardial electrograms. These data document the presence of repolarization waves and confirm, but do not prove, the theory behind the temporal methods for measuring repolarization (ARIs).
Materials and Methods Intact, canine hearts were exposed using conventional surgical techniques. A 25 x 21 array of insulated,
•
Luxet al.
131
5-mil-diameter silver wires sewn into a nylon mesh was applied to the right ventricular epicardial surface. Using a 256-channel amplification and multiplexing system, we recorded two banks of electrograms, each of which included five electrograms c o m m o n to each bank and used for time alignment. Electrograms were recorded for a variety of activation sequences, including those originating from supraventricular pacing as well as those from single or multiple ventricular site epicardial pacing. For each activation sequence, potential distributions were displayed as isopotential contour maps, gradient magnitude maps, and dynamic, threedimensional perspective displays. In addition, depolarization and repolarization times were calculated using two methods: (1) conventional ARIs were m e a s u r e d in which depolarization times were defined as time of m i n i m u m dV/dT of electrograms during the QRS and time of m a x i m u m dV/dT near the m a x i m u m T wave amplitude; and (2) the times of m a x i m u m spatial gradient magnitude during the QRS and T wave were taken as depolarization and repolarization times, respectively. Temporal and spatial measures of activation and repolarization times were compared using linear correlation. Dynamics of potential and gradient distributions were observed using three-dimensional visualization software written for a Silicon Graphics w o r k station (Mountain View, CA).
Results Isopotential distributions and their dynamics during depolarization have been well documented and described. Observation of potential distributions of our data replicate the previous findings, including elongation of isopotential ellipsoids along the fiber direction and rotation of the negative potential ellipsoid and its associated positive extrema consistent with rotation of the fiber orientation with depth (6). During repolarization, the dominant reversal of potential polarity from that during the QRS is reflected in a slowly rising and falling positive potential (upright T waves). Superimposed on this background distribution was a rapidly moving, low-amplitude ripple that emanated from and propagated away from the region of pacing. This observation, to our knowledge, has not been reported previously, and we investigated the ability of activation and repolarization timing measurements based on the timing of the m a x i m u m spatial gradient magnitude during the QRS and T wave to reflect the activation and recovery time measurements from electrogram derivatives. Figure 1 shows the distributions of epicardial isochrones of activation times (top panels) and repolarization times (bottom panels) obtained from the times of m i n i m u m dV/dT (left panels) and the m a x i m u m spatial gradient magnitude (right panels). These distributions reflect measurements from electrograms obtained during pacing from a single epicardial site. Figure 2 shows examples of electrograms, their temporal derivatives, and the magnitude of spatial potential gradients at those
132
Journal of Electrocardiology Vol. 29 Supplement
Temporal Method
Spatial Method
C
.9 O
<
•:
i
E c-
O
Fig. 1. The upper left panel shows isochrone lines determined from the times of m i n i m u m dV/dT in the QRS for epicardial pacing. The upper right panel shows isochrone lines determined as the times of m a x i m u m spatial gradient during the ORS. The lower left panel shows isochrones of repolarization determined as the time of m a x i m u m dV/dT near the peak of the T wave. The lower right panel shows isochrones of repolarization determined as the m a x i m u m times of the spatial gradient during the T wave. All isochrones are drawn at 4-ms increments. The shaded area in the lower right panel indicates regions where the spatial gradient was insufficient to measure a well-defined peak or there were multiple, low-amplitude peaks.
sites. The site n u m b e r s correspond to the n u m b e r e d locations in the isochrone distribution of Figure i (upper left panel) and give examples of waveforms from regions of longitudinal and transverse propagations. From these signals, it is apparent that there is good correspondence between temporal and spatial timings of depolarization and repolarization. Figure 3 shows scattergrams of depolarization times (Ieft panel) and repolarization times (right panel) with the temporal mea-
surement along the ordinate and the spatial (gradient) measurement along the abscissa. Also shown are the linear regression curve, its equation, and the estimated linear correlation coefficient. The characteristics of these scattergrams are similar for all of the activation sequences studied. Typically, for activation timing, the intercept of the regression line was near zero and the slope was near 1 with correlation coefficients higher than .98. For repolarization, the regression line reflected a slight bias (pos-
New Methodsfor Measuring Repolarization
Site 2
Site 3
•
L u x e t al.
133
Site 5 I I
EG
dV/dt
~I~~'~~ ~ ~
-~--~
]i
tvvt
Fig, 2. The n u m b e r e d sites correspond to data at the n u m b e r e d sites in the upper left panel of Figure 1: two in regions of longitudinal propagation and three in regions of transverse propagation. For each site, the electrogram (EG) is shown at the top, its temporal derivative (dV/dT) is shown in the middle, and the magnitude of the spatial gradient (VV) is shown at the bottom. To emphasize the alignment of the spatial and temporal derivatives, a vertical line has been drawn at the time of repolarization in each lead.
~o
6o
210
3E
._~
50
ATv=.38 + .98 A T r=.987
¢~ 40
_~ =~:"
~
|•1
200 190
RT =13.0 + .94 RT W
r=.gS
•
I-
t-i,.•¢ ••
j
I I,,-
i
! "~.~7~ "
..,I 'lIW."
180
30
=~ 20
t
.
170
.t. =.!~! • 160
>~ 0
I
i
10
20
Activation
Time
[
i
30 40 (ms) - T e m p o r a l
I
i
50
60
Method
ill
150 150
I
160
i
I
I
170 180 190 R e c o v e r y Times (ms) . T e m p o r a l
t
I
200
210
Method
Fig. 3. Scattergrams of activation (left) and repolarization (right) cornparing temporal and spatial measurement methods. The regression lines and correlation coefficients for each are shown. AT, activation time; RT, recovery time.
itive offset of several milliseconds) but the slope was very close to 1 and correlation coefficients were higher than .94. Somewhat lower correlation and the spread of points about the regression Iine reflected poorer agreem e n t between the two methods in comparison to that obtained for measuring activation time. This is likely a consequence of a poorer signal-to-noise ratio of T waves relative to QRS waves, which reflects the lower frequency content and smaller currents that produce the observed potentials.
Observation of the depolarization and repolarization waves using dynamic, three-dimensional perspective plots documented the usual propagation of excitation jumps during depolarization, but also exhibited a rapidly moving, low-amplitude wave during repolarization. These waves propagated away from the stimulus site, regardless of the site location and also replicated "collision" w h e n multiple sites were simultaneously stimuiated. For obvious reasons, it is not possible to illustrate these waves easily.
134
Journal of Electrocardiology Vol. 29 Supplement
Conclusion
References
The findings presented here support the interpretation of spatial and temporal derivatives of cardiac potentials as a means to observe not only the classical wavefronts of depolarization, but also those of repolarization. In this study, there was very close a g r e e m e n t b e t w e e n the temporal and gradient methods w h e n applied to depolarization, a finding that confirmed that the traditional m e t h o d for detecting propagating waves was applicable in this setting. The fact that the close correspondence b e t w e e n the two m e a s u r e m e n t s was m a i n t a i n e d w h e n applied to repolarization, albeit not to the same level of agreement, suggests that repolarization can, in practice, be considered a propagating p h e n o m e n o n , Of practical importance, the fact that the spatially d e t e r m i n e d m e a s u r e m e n t s of both depolarization and repolarization closely tracked the temporally d e t e r m i n e d m e a s u r e m e n t s supports previous claims that the ARI technique provides local estimates of activation and repolarization times. Although these studies p r e s e n t e d only data from epicardial surfaces, further studies in the thickness of the ventricular wall are planned.
1. Dillon SM: Optical recordings in the rabbit heart show that defibrillation strength shocks prolong the duration of depolarization and the refractory period. Circ Res 69:842, 1991 2. France MR, Bargheer K, Rafflenbeul W e t al: Monophasic action potential mapping in h u m a n subjects with normal electrocardiograms: direct evidence for the genesis of the T wave. Circulation 75:39, 1987 3. Franz MR: Method and theory of monophasic action potential recording. Prog Cardiovasc Dis 33:347, 1991 4. Millar CK, Kralios FA, L u x RL: Correlation between refractory periods and activation-recovery intervals from electrograms: effects of rate and adrenergic interventions. Circulation 72:1372, 1985 5. Haws CW, Lu× RL: Correlation between in vivo transmembrane action potential durations and activationrecovery intervals from electrograms: effects of interventions which alter repolarization times. Circulation 81:281, 1990 6. Taccardi B, Macchi E, Lux RL et al: Effects of myocardial fiber direction on epicardiaI potentials. CircuIation 90:3076, 1994