Background While abnormalities of activation and repolarization play an important role in arrhythmogenesis, little information is available on the interaction between their spatial dispersions in the heart. This study examined the effects of activation spread on the spatial distribution of the repolarization properties during different depolarization patterns.
Methods and Results Left ventricular (LV) endocardial activation and repolarization patterns were mapped in 13 healthy pigs. LV local activation, repolarization, and activation-recovery interval (ARI) times were determined from the intracardiac unipolar electrograms, color-coded, and superimposed on a three-dimensional anatomic map of the ventricle generated with a nonfluoroscopic mapping system. ARI values correlated with the duration of monophasic activation potential recorded from onset of activation to time of 90% repolarization (r=.97, P<.01). Activation time range of the left ventricle was 42±5 ms (mean±SEM) during sinus rhythm and 54±5 ms during right ventricular septal pacing. ARI inversely correlated with the corresponding activation times during both sinus (r2=.76±.03) and paced (r2=.77±.02) rhythms. The longest ARIs were located at the sites of earliest activation and shortest at the latest activation areas, with gradual shortening between them.
Conclusions The spatial distribution of repolarization is dependent on the activation pattern. Repolarization dispersion in the healthy swine heart is relatively small as the result of tight coupling of the action potential duration to the activation process, assigning longer ARIs to sites activated earlier. This coupling reduces global and regional dispersion of repolarization and may serve as an important antiarrhythmic mechanism present in normal myocardium.
Changes in myocardial activation and repolarization may contribute to the genesis of ventricular arrhythmias.1–3 Ventricular tachycardia in patients with coronary artery disease and previous myocardial infarction is generally believed to be reentrant and accompanied mainly by abnormalities of activation.4 In contrast, ventricular arrhythmias in patients with long-QT syndrome, without structural heart disease, is thought to be more closely related to abnormalities of ventricular repolarization.5 It is generally believed that in many cases of ventricular arrhythmia both mechanisms are involved and that they tend to be mutually interdependent.3
Although endocardial activation patterns have been measured in numerous studies, relatively little information exists regarding the spatial dispersion of repolarization. Likewise, there is almost no information in the literature regarding the effect of endocardial activation sequence on the repolarization pattern. In several different models it has been shown that the known nonuniform anisotropic nature of myocardial structure affects activation spread by electrotonic currents.6–11 Furthermore, the effect of electrotonic currents related to the activation spread on the repolarization properties have been reported in some studies.12–16 These studies have demonstrated that activation sequence affects the spatial distribution of repolarization times in isolated tissue preparations and in a small or limited number of sites on the epicardium. Since both slow conduction and nonuniformity of repolarization are important in the genesis of reentry, such interactions may have important clinical implications. However, there has been no systematic study in the whole-animal model of the effects of activation sequence on the global distribution of repolarization times of the endocardium. The aim of this study was to examine the effects of the activation sequence on the spatial distribution of repolarization, both globally and regionally, in the normal heart.
A nonfluoroscopic, catheter-based endocardial mapping technique was used to generate three-dimensional electroanatomic maps of activation, repolarization, and ARI patterns. The system uses magnetic technology to establish the instantaneous location and orientation of a catheter tip and simultaneously records the local electrogram from the same site.17–19 By combining spatial and electrophysiological information from a plurality of endocardial sites, we were able to assess the three-dimensional patterns of activation and repolarization during different activation patterns.
Thirteen healthy male pigs (30 to 40 kg) were premedicated with ketamine HCl (20 mg/kg IM) and xylazine (2 mg/kg), anesthetized with pentobarbital (30 mg/kg), intubated, and ventilated with a Harvard large-animal mechanical respirator. Vascular access was obtained using a cut-down procedure for the jugular vein and carotid artery. All procedures were approved by the Ethics Committee of the Technion Faculty of Medicine, Haifa, Israel, and conform with American Physiological Society standards for use of animals in research.
LV Endocardial Mapping
An 8F locatable-tip bipolar electrophysiological catheter (NAVI-STAR, Cordis-Webster) was inserted into the RV apex or the coronary sinus to serve as a reference catheter. With the use of fluoroscopy, a similar mapping catheter was introduced retrogradely into the LV.
The mapping system (CARTO; Biosense) uses a magnetic field to accurately determine the location and orientation of the two catheter tips, each with 6 degrees of freedom (x, y, z, roll, yaw, and pitch). The cardiac cycle–gated location of the mapping catheter tip is recorded relative to the location and orientation of the intracardiac reference catheter, thus compensating for both animal and cardiac motion. Choosing the R wave as the gating fiducial point results in the acquisition of end-diastolic locations, allowing for reconstruction of the ventricular geometry at end diastole. The catheter is dragged over the endocardium, sequentially acquiring its tip location together with the local electrogram sensed while in contact with the endocardium. We acquired data from 50 to 70 distinct sites. The data were recorded, compiled, and displayed in real time. Sites were included in the map only if the distance between two consecutive end-diastolic locations of the mapping catheter was <2 mm.
Local Activation and Repolarization Measurements
The LAT at each site was determined as the time interval between the fiducial point on the body-surface ECG and the steepest negative intrinsic deflection (dV/dtmin) from the unipolar intracardiac recording (filtered at 0.5 to 400 Hz).
The LRT was determined from the local T wave with the method described by Wyatt et al.20,21 The fiducial point on the surface ECG was used as the reference time for all measured intervals. The local T wave measured from a unipolar electrogram demonstrated one of three different morphologies: positive, negative, and biphasic. The dV/dtmax was used to define local repolarization for negative and biphasic T waves (Fig 1⇓, a and b). For positive T waves, the dV/dtmin on the descending limb of the T wave (Fig 1c⇓) was used.22
ARI was calculated as the time difference between LAT and LRT at each site. ARI has previously been shown to correlate well with APD.20–23 The electrophysiological information was color-coded in real time and superimposed on the three-dimensional geometry of the chamber.
We mapped the hearts during SR (n=8), during RV midseptal pacing at different CLs (300 and 350 ms, n=13), and during atrial pacing (350 and 400 ms, n=6). The range of activation times, repolarization times, and ARIs were quantified. In addition, detailed three-dimensional electroanatomic maps of the patterns of activation, repolarization, and ARI were generated.
Correlation of ARIs With MAP Recording
In two pigs ARIs were correlated with MAP recordings with the use of a standard-curve MAP/pacing silver–silver chloride catheter (EPT). The electrode was positioned gently against the LV endocardium under fluoroscopic guidance until a stable MAP was recorded. The mapping catheter was positioned adjacent to the MAP catheter as judged by using multiplanar fluoroscopy. The time from the onset of activation to 90% repolarization (MAP90) was measured as an estimate of APD.24 ARIs were compared with the MAP90 at multiple sites and during different atrial and ventricular pacing rates. A total of 98 recordings at a variety of sites were compared.
In three animals, we also studied the spatial dispersion of RPs and compared it with the activation sequence of the LV during ventricular pacing (CL=350 ms). After eight S1 beats originating from the pacing catheter (positioned at the RV septum or right atrium), a single premature stimulus, S2, was given from the mapping catheter located in the left ventricle. Pacing was delivered at twice diastolic threshold. The shortest S1-S2 coupling interval still inducing a propagated response was measured. The RP at each measured site in the LV was taken as the shortest S1-S2 interval that induced a propagated response, minus the activation time at that site. Since the activation time at the test site was subtracted from S1-S2, RPs obtained were measures of local excitation of each site. By sequentially using this technique in an average of 20 sites in the LV, we were able to reconstruct three-dimensional maps of RPs during both atrial and ventricular pacing.
All data are reported as mean±SEM. Paired data comparing MAP90 to ARIs were analyzed by linear regression analysis. Correlations were examined for each of the three separate T-wave morphologies, as well as for the pooled data, using Pearson’s correlation test. The correlation between activation times and ARIs for each individual map was calculated using linear regression analysis.
Endocardial Activation Patterns
The LV activation pattern was studied during SR, right atrial pacing, and ventricular pacing from the RV septum.
Activation of the LV During SR and Atrial Pacing
Fig 2⇓a represents a typical activation map of the left ventricle during SR. The earliest activation site (represented by red) was located in the anterosuperior septal region in all maps. In some hearts, a second endocardial breakthrough was noted in the posterosuperior septum. Rapid conduction was noted from the earliest site of activation toward the apex. The activation then spread significantly slower to the rest of the endocardium, with the posterior basal wall, the atrioventricular ring, being activated last.
Total activation time of the LV during SR was 42±5 ms and during atrial pacing was 40±3 and 48±6 ms at CLs of 350 and 400 ms, respectively. Global activation patterns during SR and atrial pacing at 350 and 400 ms were similar.
Activation of the LV During Ventricular Pacing
Fig 3⇓a shows a right anterior oblique view of a typical three-dimensional activation map of the LV during pacing from the RV inferomedial septum. The map was reconstructed with the use of 50 sampled points. As can be seen, the earliest site of activation (represented by the area in red) was located in the inferior septum adjacent to the pacing site. Activation then spread throughout the ventricle, with the posterolateral areas (indicated by the blue and purple) being activated last. Intermediate LATs (green and yellow) were recorded from the remainder of the LV. An important feature that can be appreciated in these activation maps is conduction anisotropy, with more rapid conduction in the longitudinal direction. The total activation time of the ventricle depicted in Fig 3a⇓ was 61 ms. Mean activation time range of the LV during pacing was 56±3 and 50±4 ms, with CLs of 350 and 300 ms, respectively. The activation sequence during ventricular pacing at 300 or 350 ms was similar.
Endocardial Repolarization Patterns
T-Wave Morphology Distribution
Local T waves were classified as negative, positive, and biphasic (Fig 1⇑, a through c). These different morphologies were spatially arranged in clusters within the LV. During RV pacing, most of the T waves were negative, with the positive T waves being concentrated mainly in the area of earliest activation. During SR and atrial pacing, most sites exhibited positive T waves, while sites distant from the earliest sites of activation displayed negative T waves.
Correlation Between ARI and MAP90
A total of 98 MAP recordings and corresponding ARIs were compared during SR and pacing. The correlation between the MAP90 and ARIs (Fig 4⇓) was statistically significant for each of the three T-wave morphologies and for the pooled data. The overall correlation coefficient was .97 (P<.01), and the correlation coefficients for the positive, negative, and biphasic T waves were .98 (P<.01), .98 (P<.01), and .94 (P<.01), respectively.
Figs 2b⇑ and 3b⇑ illustrate typical ARI maps of the LV during SR and RV septal pacing, respectively. The ARI maps closely resembled the corresponding activation maps (Figs 2a⇑ and 3a⇑), with the longest ARIs located near the earliest sites of activation (red) and shorter ARIs observed as activation proceeds from this site. For example, the longest ARI (307 ms) during RV pacing (red) was located in the inferior septum (Fig 3⇑), whereas the shortest ARI (229 ms) was located in the posterior and inferolateral regions (purple area), with a gradient of ARIs between them. The range of the LV ARI was 65±5 ms [182±8 (mean shortest ARI) to 247±12 ms (mean longest ARI)] during pacing at 350-ms CL and 50±5 ms (164±9 to 208±9 ms) during pacing at 300 ms.
During SR and atrial pacing, the ARI maps also resembled the activation maps. The longest ARIs were located in the septal region, and the shorter ARIs were located on the lateral wall and close to the mitral annulus (Fig 2⇑). The range of LV ARI during SR was 63±7 ms (270±16 to 331±20 ms) and during atrial pacing was 36±3 ms (198±12 to 234±11 ms) and 48±10 ms (221±14 to 269±23 ms) at 350- and 400-ms CLs.
To quantify the relationship between the activation sequence and the spatial dispersion of ARI, detailed plots of the ARIs as a function of the activation times for each map were generated. Fig 5⇓ represents a typical plot of the ARI as a function of the activation time, in one animal, during SR and ventricular pacing. Note the inverse correlation (r2=.83, .79) for both rhythms. A similar inverse correlation was found in all animals during SR (r2=.76±.03), ventricular pacing (r2=.77±.02 and .76±.02 at CLs of 350 and 300 ms, respectively), and atrial pacing (r2=.70±.03 and .73±.03 at CLs of 400 and 350 ms, respectively).
Figs 2c⇑ and 3c⇑ display examples of repolarization maps of the LV generated during SR and RV pacing. The range of T-wave dispersion was small, averaging <40 ms. The total dispersion of repolarization during ventricular pacing was 37±3 and 30±3 ms at CLs of 350 and 300 ms, respectively. The dispersion during SR was 39±4 ms and during atrial pacing was 27±3 and 27±2 ms at CLs of 400 and 350 ms, respectively. Three patterns of repolarization were noted. In most maps, the repolarization sequence resembled that of activation; in others the sequence was reversed, whereas in some cases the earliest repolarization site was located between the earliest and latest activation sites.
In three animals, RP maps of the LV were generated during ventricular pacing (CL=350 ms). Figs 6⇓a and 6b represent activation and RP maps of the LV during RV septal pacing. The RP at each site was taken as the shortest S1-S2 interval that induced a propagated response, minus the LAT (S1, pacing from the RV septum; S2, pacing from the mapping catheter). The RP maps resembled the activation maps, with longer RPs (red area in Fig 6b⇓) located near the earliest site of activation (red area in Fig 6a⇓) and shorter RPs located at sites activated later (purple areas).
Pattern of Ventricular Activation
The endocardial activation sequence of the swine LV was similar during SR and atrial pacing. The earliest activation site was at the anterosuperior septum, occasionally with a second endocardial depolarization noted more posteriorly. Activation then spread to the remainder of the ventricle, in the direction of the apex, atrioventricular ring, the posterolateral wall, and the outflow tract. The observed sequence of activation of the swine LV resembles the classic description of human LV activation given by Durrer et al.25 Conduction anisotropy was obvious, with faster conduction velocity in the longitudinal direction toward the apex and slower activation proceeding transversely. This anisotropy may represent the effects of endocardial activation after the activation of the left anterior and posterior fascicles of the left bundle.
During pacing from the RV inferior septum, LV activation was first noted from the inferior septum, adjacent to the pacing electrode. Activation then proceeded throughout the rest of the ventricle.
Local T-Wave Morphologies
Examination of the local electrograms revealed three types of T-wave morphologies, characterized by positive, negative, and biphasic wave forms. We found that there was a nonuniform distribution of the T wave, with clustering of sites with positive T-wave morphologies near the earliest activation site and negative or biphasic T waves at sites that were activated last. This observation is in agreement with previous animal and human studies performed both in vitro and in vivo.22,26–29 According to studies performed with isolated papillary muscle, the morphologies of local T waves were determined by the spatial relation that exists between the wave of excitation and the recording electrode.27 The T-wave polarity was positive when the recording electrode was near the origin of the activation and negative when it was opposite the origin of activation. Similarly, transmural recording showed that in normal beats there was transmural unidirectional gradient, with the inner wall being more negative and the outer wall more positive.28
Dispersion of ARIs During Both SR and Paced Rhythms
The spatial pattern of ARI was generally similar to the depolarization pattern. A uniform and homogenous gradient of ARI was observed, with the longest ARI positioned near the area of earliest activation, gradually shortening with increased distance from this site. To further quantify the relationship between the depolarization sequence and the pattern of ARI, detailed plots correlating the activation times with the time to ventricular repolarization (ARI) at each site were determined for each individual heart. A significant inverse correlation was found between the activation sequence and ARI for all rhythms (r2=.77±.02 and .76±.02 for ventricular pacing at CLs of 400 and 350 ms, respectively; r2=.76±.03 for SR and r2=.70±.03 and .73±.03 for atrial pacing at CLs of 400 and 350 ms, respectively).
Previous studies reported the electrotonic effects of propagation of activation on the dispersion of RP and APD. These studies included in vitro13,16,30,31 and in vivo12,32,33 studies and computer simulations.15,31,32 The shortening of ventricular APD as activation proceeds was noted previously in some of these studies. Specifically, Zubair et al33 studied the effect of the activation sequence of the spatial dispersion of RP at 36 sites within a 1-cm2 region of the epicardial surface of a canine pulmonary conus. They found that the spatial distribution of RPs was markedly affected by the activation spread. Similarly, Osaka et al13 examined the influence of the activation sequence on action potential configuration in the epicardial surface of isolated pieces (2.5 cm2) of canine ventricular muscle. They found that APD shortened gradually as the recording site was moved further from the stimulation site. The spatial gradient of APD was steeper in the transverse than in the longitudinal direction. They speculated that the changes in APD might be due to electrotonic interactions between neighboring cells.
This study is the first report on the effects of activation spread on the spatial distribution of in vivo global endocardial repolarization properties. The first finding in this study is that marked changes in the spatial distribution of ARI occurred when the activation pattern changed. A second novel finding of this study is the observed coupling between ARI and activation sequence, resulting in a relatively simultaneous termination of repolarization. This study demonstrated that the previously reported relationship between ARI and activation sequence is of global significance and not only of local importance.
The marked changes at the spatial distribution of the time to ventricular repolarization (ARI) as a function of the depolarization sequence did not require long-standing alteration of ventricular activation sequence (repolarization memory). In 1982, Rosenbaum et al34 reported that rapid atrial or ventricular pacing in humans could induce T-wave changes that develop to a maximum in about 24 hours of pacing. As reviewed by Katz,35 these changes, referred to as cardiac “memory,” develop more slowly and persist longer than the transient changes mediated by cellular metabolism but are still reversible, are not associated with organ damage, and may arise from structural changes in the membrane channel proteins.
The relatively short pacing periods and the lack of changes in T-wave morphologies in the body-surface ECG in the postpacing sinus rhythm interval decrease the likelihood that this mechanism was involved in the changes in the repolarization pattern observed in the current study. Furthermore, the fact that the gradient of ARI was always inversely correlated with the activation sequence may further support the role of electrotonic interactions between cells as the major mechanism of the findings of this study.
Significance of Findings
Dispersion of repolarization usually parallels the dispersion of refractoriness. The dispersion of repolarization is determined by the regional differences in activation times and APD. Augmented dispersion of repolarization has been shown to increase the propensity of ventricular arrhythmias, both clinically36–39 and in experimental models.40–42 Examples of experimental models in which dispersion of repolarization appears to play an important role include the circus-movement tachycardia induced by Allessie et al40 in isolated segments of rabbit atrial tissue and the late ventricular arrhythmias elicited after the ligation of left anterior descending coronary artery in dogs.3
In humans, the mechanism of dispersion of repolarization varies with the disease process. Vassallo et al36 measured recovery of excitability at multiple endocardial sites in the LV in three groups of patients. In normal subjects the mean dispersion of full recovery time was 52 ms. In patients with a history of myocardial infarction and sustained ventricular tachycardia, the dispersion of recovery time was 90 ms, due mainly to prolonged activation time. In contrast, patients with long-QT syndrome demonstrate prolonged dispersion of recovery times (114 ms), primarily because of variable refractoriness.
The current study has shown that in the normal swine heart, dispersion of repolarization in the LV is relatively homogeneous and in most cases does not exceed 40 to 50 ms. This study has reported for the first time the close dependency of local repolarization on the activation process of the entire LV endocardium. We have demonstrated that in the normal swine LV endocardium, the dispersion of repolarization is functionally dependent on the location of earliest activation and the direction of the wave front of propagation. Sites closer to the earliest activation have longer ARIs, whereas sites further away from the activation source have significantly shorter ARIs. Shortening of the APD along the propagation route seems to partially compensate for the delay in arrival of activation to remote sites. Therefore, the repolarization process has little or no dispersion relative to that of the activation process. Furthermore, this tight coupling may also be of regional importance because it compensates for activation time differences between adjacent anatomic sites, thereby diminishing regional spatial dispersion of repolarization. This would tend to reduce the propensity for reentrant arrhythmias.
Use of Unipolar Electrograms to Estimate LRTs
The accuracy of repolarization and ARI maps depends on analysis of the local T wave for calculations of LRT. Unlike the unipolar recording of depolarization, the local T wave lacks sharp intrinsic deflection, so that determining the LRT is more tedious and at times of questionable value. Wyatt et al20,21 reported a method for determining the LRT by which the timing of the maximum rate of rise in voltage (dV/dtmax) in the local unipolar wave form is used as a surrogate for the recovery from depolarization of the intracellular action potential. Using this method, experimental studies in humans and dogs have shown good correlation between the ARI and local APD, measured by transmembrane action potential or by single premature stimuli. Millar et al23 and Chen et al22 reported that for positive T-wave forms, underestimation of the APD was found by use of the Wyatt method. They proposed taking the dV/dtmin on the descending limb of the T wave, reporting better correlation with the APD. We used this modified approach. For positive T waves we used the Wyatt approach, whereas for negative T waves we used the Millar approach. With this method, we found tight correlation with MAP90 as measured by a MAP catheter with ARI intervals.
Limitations of the Study
The results of this study depend on several technical considerations. In contrast to the rapid downstroke of the local action potential, the low amplitude and broad nature of the local T wave limit the accuracy of measurements of local repolarization because the timing of the maximum change in slope is more variable. Furthermore, the multiple T-wave morphologies preclude the use of a uniform method for measuring repolarization times.
However, for four reasons we believe that our method for measuring repolarization is accurate. First, we observed reproducible responses within animals and from animal to animal. Second, our results are consistent with published reports demonstrating excellent correlation between ARI and MAP duration, as well as direct measurement of RP. Third, our methods for measuring repolarization times are based on previously validated studies that took into account the variable morphology of the T wave.6–9 Fourth, maps of ARIs and RPs showed similar results.
The association between increased dispersion of repolarization and arrhythmogenesis is mentioned above. The current study documents a tight association between activation, propagation, and dispersion of repolarization. These dependencies may be caused by electrotonic currents flowing between adjacent cells. In the presence of a pathology that decreases cell-to-cell conductivity, the effects of these electrotonic currents may decrease. This, in turn, may cause uncoupling of the activation and repolarization process and may account for the observed increased dispersion of repolarization in diseased myocardium simultaneous with the increased propensity to arrhythmogenesis. Studying endocardial activation-repolarization patterns during disease states may therefore improve our basic understanding of arrhythmogenesis.
Selected Abbreviations and Acronyms
|APD||=||action potential duration|
|LAT||=||local activation time|
|LRT||=||local repolarization time|
|LV||=||left ventricle (ventricular)|
|MAp||=||monophasic action potential|
|RV||=||right ventricle (ventricular)|
We are grateful to David D. Gutterman, MD, for critical review of the manuscript and to Ruth Singer for editorial assistance.
- Received June 12, 1997.
- Revision received July 29, 1997.
- Accepted August 25, 1997.
- Copyright © 1997 by American Heart Association
Han J, Moe GH. Nonuniform recovery of excitability in ventricular muscle. Circ Res. 1964;14:44–60.
Han J, Garcia de Jalon P, Moe GH. Adrenergic effect on ventricular vulnerability. Circ Res. 1964;14:516–524.
Gough WB, Mehra R, Restivo M, Zeiler RH, el-Sherif N. Reentrant ventricular arrhythmias in the late myocardial infarction period in the dog: correlation of activation and refractory maps. Circ Res. 1985;57:432–442.
Schwartz PJ, Locati EH, Napolitano C, Priori SG. The long QT syndrome. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. 2nd ed. Philadelphia, Pa: WB Saunders; 1995:788–811.
Allessie MA, Schalij MJ, Kirchhof CJHJ, Boersma L, Huybers M, Hollen J. Experimental electrophysiology and arrhythmogenicity: anisotropy and ventricular tachycardia. Eur Heart J. 1989;10(suppl E):2–8.
Balke CW, Lesh MD, Spear JF, Kadish A, Levine JH, Moore EN. Effects of cellular uncoupling on conduction in anisotropic canine ventricular myocardium. Circ Res. 1988;63:879–892.
Dillon SM, Allessie MA, Ursell PC, Wit AL. Influences of anisotropic tissue structure on reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res. 1988;63:182–206.
Spach MS, Miller WT III, Geselowitz DB, Barr RC, Kootsey JM, Johnson EA. The discontinuous nature of propagation in normal canine myocardium: evidence for recurrent discontinuities of intracellular resistance that affect the membrane currents. Circ Res. 1981;48:39–54.
Spach MS, Miller WT III, Dolber PC, Kootsey JM, Sommer JR, Mosher CE Jr. The functional role of structural complexities in the propagation of depolarization in the atrium of the dog: cardiac conduction disturbances due to discontinuities of effective axial resistivity. Circ Res. 1982;50:175–191.
Osaka T, Kodama I, Tsuboi N, Toyama J, Yamada K. Effects of activation sequence and anisotropic cellular geometry on the repolarization phase of action potential of dog ventricular muscles. Circulation. 1987;76:226–236.
Toyoshima H, Burgess MJ. Electrotonic interaction during canine ventricular repolarization. Circ Res. 1978;43:348–356.
Lesh MD, Pring M, Spear JF. Cellular uncoupling can unmask dispersion of action potential duration in ventricular myocardium: a computer modeling study. Circ Res. 1989;65:1426–1440.
Spach MS, Dolber PC, Heidlage JF. Interaction of inhomogeneities of repolarization with anisotropic propagation in dog atria: a mechanism for both preventing and initiating reentry. Circ Res. 1989;65:1612–1631.
Gepstein L, Hayam G, Ben-Haim SA. A novel method for nonfluoroscopic, catheter-based electroanatomical mapping of the heart: in vitro and in vivo accuracy results. Circulation. 1997;95:1611–1622.
Shpun S, Gepstein L, Hayam G, Ben-Haim SA. Guidance of radiofrequency endocardial ablation with real-time three-dimensional magnetic navigation system. Circulation.. 1997;96:2016–2021.
Wyatt RF. Comparison of estimates of activation and recovery times from bipolar and unipolar electrograms to in vivo transmembrane action potential duration. Proc IEEE/Engineering in Medicine and Biology Society, 2nd annual conference, Washington DC; September 1980:22–25.
Wyatt RF, Burgess MJ, Evans AK, Lux RL, Abildskov JA, Tsutsumi T. Estimation of ventricular transmembrane action potential duration and repolarization times from unipolar electrograms. Am J Cardiol. 1981;47:488. Abstract.
Chen PS, Moser KM, Dembitsky WP, Auger WR, Daily PO, Calisi CM, Jamieson SW, Feld GK. Epicardial activation and repolarization patterns in patients with right ventricular hypertrophy. Circulation. 1991;83:104–118.
Millar CK, Kralios FA, Lux RL. Correlation between refractory periods and activation-recovery intervals from electrograms: effects of rate and adrenergic interventions. Circulation. 1985;72:1372–1379.
Franz MR. Long-term recording of monophasic action potential from human endocardium. Am J Cardiol. 1963;51:1629–1634.
Durrer D, van Dam RT, Freud GE, Janse MJ, Meijler FL, Arzbaecher RC. Total excitation of the isolated human heart. Circulation. 1970;41:899–912.
Spach MS, Barr RC. Origin of epicardial ST-T wave potentials in the intact dog. Circ Res. 1976;39:475–487.
Hoffman B, Suckling EE. Effect of heart rate on cardiac membrane potentials and the unipolar electrogram. Am J Physiol. 1954;179:123–130.
Spach MS, Barr RC. Ventricular intramural and epicardial potential distributions during ventricular activation and repolarization in the intact dog. Circ Res. 1975;37:243–257.
Spach MS, Barr RC. Analysis of ventricular activation and repolarization from intramural and epicardial potential distributions for ectopic beats in the intact dog. Circ Res. 1975;37:830–843.
Joyner R. Modulation of repolarization by electrotonic interactions. Jpn Heart J. 1986;17(suppl 1):167. Abstract.
Burgess MJ, Steinhaus BM, Spitzer KW, Ershler PR. Nonuniform epicardial activation and repolarization properties of in vivo canine pulmonary conus. Circ Res. 1988;62:233–246.
Vassallo JA, Cassidy DM, Kindwall KE, Marchlinski FE, Josephson ME. Nonuniform recovery of excitability in the left ventricle. Circulation. 1988;78:1365–1372.
Surawicz B. Electrophysiologic Basis of ECG and Cardiac Arrhythmias. Baltimore, Md: Williams & Wilkins; 1995;242–246.
Misier AR, Opthof T, van Hemel NM, Vermeulen JT, de Bakker JM, Defauw JJ, van Capelle FJ, Janse MJ. Dispersion of ‘refractoriness’ in noninfarcted myocardium of patients with ventricular tachycardia or ventricular fibrillation after myocardial infarction. Circulation. 1995;91:2566–2572.
Allessie MA, Bonke FI, Schopman FJ. Circus movement in rabbit atrial muscle as a mechanism of tachycardia, II: the role of nonuniform recovery of excitability in the occurrence of unidirectional block, as studied with multiple microelectrodes. Circ Res. 1976;39:168–177.
Ogawa S, Furuno I, Satoh Y, Yoh S, Saeki K, Sadanaga T, Katoh H, Nakamura Y. Quantitative indices of dispersion of refractoriness for identification of propensity to re-entrant ventricular tachycardia in a canine model of myocardial infarction. Cardiovasc Res. 1991;25:378–383.
Kuo CS, Munakata K, Reddy CP, Surawicz B. Characteristics and possible mechanism of ventricular arrhythmia dependent on the dispersion of action potential durations. Circulation. 1983;67:1356–1367.