Cardiac Memory in Patients With Wolff-Parkinson-White Syndrome
Noninvasive Imaging of Activation and Repolarization Before and After Catheter Ablation
Background— Cardiac memory refers to a change in ventricular repolarization induced by and persisting for minutes to months after cessation of a period of altered ventricular activation (eg, resulting from pacing or preexcitation in patients with Wolff-Parkinson-White syndrome). ECG imaging (ECGI) is a novel imaging modality for noninvasive electroanatomic mapping of epicardial activation and repolarization.
Methods and Results— Fourteen pediatric patients with Wolff-Parkinson-White syndrome and no other congenital disease, were imaged with ECGI a day before and 45 minutes, 1 week, and 1 month after successful catheter ablation. ECGI determined that preexcitation sites were consistent with sites of successful ablation in all cases to within a 1-hour arc of each atrioventricular annulus. In the preexcited rhythm, activation-recovery interval (ARI) was the longest (349±6 ms) in the area of preexcitation leading to high average base-to-apex ARI dispersion of 95±9 ms (normal is ≈40 ms). The ARI dispersion remained the same 45 minutes after ablation, although the activation sequence was restored to normal. ARI dispersion was still high (79±9 ms) 1 week later and returned to normal (45±6 ms) 1 month after ablation.
Conclusions— The study demonstrates that ECGI can noninvasively localize ventricular insertion sites of accessory pathways to guide ablation and evaluate its outcome in pediatric patients with Wolff-Parkinson-White syndrome. Wolff-Parkinson-White is associated with high ARI dispersion in the preexcited rhythm that persists after ablation and gradually returns to normal over a period of 1 month, demonstrating the presence of cardiac memory. The 1-month time course is consistent with transcriptional reprogramming and remodeling of ion channels.
Received March 21, 2008; accepted June 6, 2008.
Cardiac memory (CM)1 is the phenomenon of repolarization changes induced by long-standing altered cardiac activation (eg, pacing) that persist for a period of time after normal activation has been restored. The mechanism of CM is a remodeling process that alters the molecular determinants of action potential duration such as membrane density and kinetic properties of ion channels.2 As pacing therapy becomes increasingly prevalent in heart failure, bradycardia, heart block, and congenital heart defects, the effects of CM induced by a long-standing changed activation sequence are of important scientific and clinical interest. The Wolff-Parkinson-White (WPW) syndrome3 involves a long-standing altered activation sequence, different from normal sinus rhythm, as a result of ventricular preexcitation from the ventricular insertion site of an atrioventricular accessory pathway (AP). As such, it mimics long-standing ventricular pacing and provides a natural model for studying repolarization changes associated with long-term altered ventricular activation. Importantly, AP ablation provides a model of activation and repolarization changes and their time course after cessation of pacing and restoration of normal sinus rhythm.
Editorial p 899
Clinical Perspective p 915
Symptomatic Patients with WPW syndrome are treated by transvenous catheter ablation of the AP. APs are located mostly around the atrioventricular annuli, with the rare exception of Mahaim fibers.4 Noninvasive localization of AP around the atrioventricular annuli before catheterization can facilitate invasive catheter mapping. Current noninvasive techniques for localization of APs include algorithms based on the delta wave and QRS morphologies on the 12-lead ECG.5 Although easy and fast, a number of studies have shown inconsistencies and inaccuracies in these ECG algorithms, especially in the pediatric population.6 Moreover, in a smaller number of cases, multiple APs7 may exist, which makes localization from 12-lead ECG even more difficult.
ECG imaging (ECGI)8,9 is a novel noninvasive imaging modality that reconstructs epicardial potentials (voltage maps), electrograms, activation (isochrones), and repolarization sequences from 250 simultaneously recorded body surface ECGs and an accurate heart-torso anatomy obtained from an ECG-gated thoracic computed tomography (CT) scan (Figure 1). ECGI has been extensively validated in canine experiments10–12 and in humans with intraoperative mapping data.13 It has been applied to image normal and abnormal cardiac electrophysiology (both activation and repolarization) in humans.8,9 ECGI could localize ventricular pacing sites in patients undergoing cardiac resynchronization therapy with an accuracy of 7±3 mm (using the CT coordinates of the pacing leads as the gold standard).12 These capabilities suggest its suitability for the noninvasive determination of the preexcitation region in patients with WPW syndrome before catheter mapping and ablation, thereby guiding the procedure and possibly shortening it and reducing its associated iatrogenic risks. The objective of the present study is 2-fold: to evaluate ECGI as a noninvasive preablation procedure for locating the region of preexcitation in patients with WPW syndrome and to image activation and repolarization before ablation and at several time points after ablation to study the time course of activation and repolarization changes in the context of CM.
The study population consisted of 14 WPW pediatric patients (6 male, 8 female patients; age, 12±3 years) who were referred to St Louis Children’s Hospital for ablation between January 2006 and May 2007; all of them had nonintermittent preexcitation (delta wave) on the surface 12-lead ECG, with transthoracic echocardiograms demonstrating structurally normal hearts. They were all symptomatic and subsequently underwent successful catheter ablation. A day before catheterization and ablation, each patient underwent an ECGI study with 200 conductive-carbon electrodes arranged in strips. Body-surface markers were adhered to the patient torso to mark the position of each strip for future application at the same locations during follow-up studies after ablation. In addition, digital photographs were taken of the patient’s torso to document the position of the strips with respect to anatomic markers like the sternal notch, nipples on the front, and any identifiable birthmarks on the back. Body-surface potentials were recorded in the baseline resting rhythm simultaneously from 200 electrodes, and the patient underwent a thoracic CT scan gated to 70% of the ECG R-R interval.
The body-surface potentials were combined with the heart-torso geometry from CT to reconstruct epicardial potentials and unipolar electrograms at 500 sites on the ventricular epicardium as previously described.8 Epicardial potential maps were generated at a temporal resolution of 0.5 ms (sampling rate, 2 kHz). The activation time at each epicardial site was computed by assigning local activation time as the time point when the negative time derivative of the corresponding electrogram reached a maximum(−dV/dtmax). Recovery time at each site was computed by assigning local recovery time as the time point when the time derivative of the T wave of the local electrogram reached a maximum (d[T-wave]/dtmax, activation-recovery interval [ARI] method).14 An alternative method that takes the time corresponding to the steepest downward slope (−d[T-wave]/dtmax) as the recovery time for positive T waves also was used. The second method gives a higher estimate of recovery time for positive T waves. Previous studies14 have shown recovery time as a reliable measure of local repolarization. Local ARI was computed as the difference between local recovery time and local activation time. ARI has been reported to correlate with local action potential duration14; it is an intrinsic index of repolarization that is independent of the activation sequence and local activation time. Epicardial base-to-apex ARI dispersion (ARId) was computed as the difference between the mean ARI of 100 epicardial sites on the annular basal area and the mean ARI of 100 sites on the epicardial apex. The preexcitation site was localized from noninvasive epicardial potential maps during activation and repolarization, as well as from the activation isochrone map. For localization, the epicardial aspect of each atrioventricular annulus was divided into twelve 1-hour clock divisions. The most anterior aspect of each annulus was designated a 12 o’clock position. The attachment of the mouth of the coronary sinus (CS) on the tricuspid annulus (TA) was assigned a 5 o’clock position. The different clock positions around the 2 annuli are shown in the left posterior view of the epicardium in Figure 2A. The ECGI reconstruction was done offline in ≈90 minutes, and results were communicated to the interventional electrophysiologist (E.K.R.) performing the procedure. On the next day (the day of ablation) before catheterization, electrode strips were applied to the patient’s torso in the same order and position, guided by the body-surface markers and photographs. This technique avoided repetition of CT and allowed use of subject-specific heart-torso geometry obtained during the first scan in this and subsequent follow-ups. The procedure was carried out with the patient in the same supine position as during the first ECGI study. Because the carbon electrodes are transparent to x-ray, they did not interfere with fluoroscopy during the catheter electrophysiology study. Body-surface potentials were recorded during supraventricular tachycardia (SVT)3 (induced during the electrophysiology study as part of the procedure). ECGI maps of anterograde ventricular activation and retrograde atrial activation (only when a clear retrograde P wave could be seen on the body-surface ECG) were reconstructed offline. After successful ablation (loss of delta wave and testing with adenosine), the clock position (schematic in Figure 2B) of the ablation catheter at the site of success was documented in the 60° left anterior oblique fluoroangiogram and was compared with the preablation ECGI localization. Another ECGI recording was made 45 minutes after successful ablation with the patient still in the same supine position and in normal sinus rhythm without preexcitation. Patients were followed up with ECGI using the same methods 1 week and 1 month after ablation with the body-surface potentials recorded at the baseline resting rhythm. The study protocol was reviewed and fully approved by the Human Research Protection Office at Washington University. Informed written consent was obtained from all subjects before the study.
ARI and ARId numerical data are presented as mean±SD, with these measures taken over the entire patient population (n=14). The relationship between the epicardial activation maps at different time points after ablation and the epicardial ARI maps before and after ablation is determined by Pearson’s product-moment correlation coefficient (r).
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Epicardial Potentials and Activation
Figure 3A shows the preablation epicardial potential map in subject W1 during early activation in the preexcited rhythm. The earliest epicardial breakthrough, indicated by a local potential minimum (deep blue, asterisk), was observed in the left posterior area at 7 o’clock around the epicardial aspect of the mitral annulus (MA) 8 ms after onset of the delta wave on the body-surface ECG. The epicardial potential map during the initial phase of repolarization (Figure 3B; 181 ms after onset of delta wave) followed a pattern similar to that during early activation but with reversed polarity, with the minimum replaced by a potential maximum (red). The preablation ECGI isochrone map (Figure 3C) showed early activation (red, asterisk) initiating from the same area, confirming the presence of an activation source (AP) in this region. These results consistently indicated a left posterior insertion site of the AP, 7 o’clock around the MA. ECGI isochrone map (Figure 3D) of the patient in postablation rhythm (after successful ablation) showed late activation (light blue, +) in the left posterior annular aspect of the epicardium, with the posterolateral right ventricular (RV) base being the latest to activate (deep blue). The second row in Figure 3 shows another example (subject W2) of ECGI localization of a left-sided pathway with the preexcitation area (asterisk, Figure 3E, 3F, and 3G) located more laterally, localized at 4 o’clock around the MA. Subject W3 (Figure 3, third row) showed 2 discrete epicardial potential minima (Figure 3I, deep blue, black circles) during early activation in the preexcited rhythm at 3 o’clock and 5 o’clock around the MA. The early repolarization potential map exhibited 2 discrete potential maxima (red, asterisk; Figure 3J) replacing the 2 minima observed during early activation. The preablation isochrone map showed an elongated area of early activation (red; Figure 3K) in the left lateral and posterolateral areas. The presence of 2 potential minima during early activation (and 2 maxima during early repolarization) indicated the possibility of the presence of 2 discrete APs (later confirmed during the ablation procedure; the Table). Postablation isochrone maps showed the latest epicardial activation occurring in the left posterolateral and left posterior basal areas for subjects W2 and W3 (deep blue panels in Figure 3H and 3L, respectively). Figure 4 (first row; subject W4) shows an example of ECGI localization of a right posterior septal pathway. The ECGI epicardial potential and isochrone maps localized the preexcitation area to 5 o’clock around the epicardial aspect of the TA. It indicated a ventricular insertion site near the mouth of the CS. Subject W5 (second row) showed potential breakthrough (Figure 4E) and earliest activation (Figure 4G) in the midseptal area (4 o’clock around the TA or 8 o’clock around the MA). For midseptal pathways, ECGI could not predict laterality (right or left annulus) because ECGI reconstructions are limited only to the epicardial roof of the midseptal area. Preablation ECGI maps of subject W6 (third row, Figure 4I through 4K) indicated a left posterior pathway 7 o’clock around the MA. Postablation activation maps of subjects W4 through W6 showed late activation (blue, green) at the preablation region of preexcitation (+ in Figure 4D, 4H, and 4L). In general, the earliest annular area to activate after ablation was the superior anteroseptal aspect and the anterolateral RV base of the epicardium; the latest annular activation was usually in the left ventricular (LV) base in the lateral or posterolateral areas (8 of 14) or the posterolateral RV base (6 of 14). This is consistent with the pattern of normal human ventricular activation sequence reported by Durrer et al15 and with previous ECGI studies of ventricular activation in normal young adults.8,9
Retrograde Atrial and Anterograde Ventricular Activation During SVT
ECGI maps of retrograde atrial activation were constructed in 5 subjects in whom a clear retrograde P wave was discerned after the QRS on the body-surface ECG during SVT. Examples of ECGI activation maps during SVT are shown in Figure 5 for a left-sided and a right-sided pathway. A retrograde atrial activation map for subject W1 (left posterior ventricular insertion) is shown in Figure 5A and 5B. Earliest activation (white, Figure 5B) was in the left posterior atrial wall. Activation spread out uniformly from this area, with the anterior lower right atrium activating last (deep blue, Figure 5A). Figure 5E and 5F shows the retrograde atrial activation map for subject W5 (right-sided midseptal AP). The sequence of retrograde atrial activation in this case was distinctly different from that of subject W1. The lower right atrium was the first to activate (white, light pink), from where activation propagated to the top of the right atrium and to the posterior side. The latest areas to activate (blue) were the left lateral atrial wall and the left atrial appendage. The earliest sites of retrograde atrial activation were consistent with the earliest site of ventricular activation before ablation. The anterograde ventricular activation showed early anterior RV epicardial activation (red, orange, Figure 5E and 5G) in apical (W1) or paraseptal (W5) areas. The latest area to activate was the posterolateral RV base (W1) or the LV basal lateral area (W5). The ECGI activation map of anterograde ventricular conduction during SVT was found to be similar (r=0.98) to the normal sinus rhythm activation map after ablation, indicating that in SVT, anterograde conduction involved the atrioventricular node and His-Purkinje system exclusively and was independent of the location of the AP.
Electrograms and ARI
ECGI epicardial ARI maps (computed by ARI method) before and 45 minutes, 1 week, and 1 month after successful AP ablation are shown in Figures 6 and 7⇓. The insets show the electrogram from the preablation preexcitation site (asterisk, Figures 6 and 7⇓). For all patients except W6, the epicardial electrogram at the preexcitation site showed an r-S activation morphology (insets in Figure 6A, 6E, 6I, 7A, and 7⇓E); the small positive r wave reflects the intramural activity as the activation wave front propagates from an endocardial site toward the epicardium. An r-S electrogram on the epicardial preexcitation site indicated endocardial AP insertion. For patient W6, complete absence of this small positive deflection and a pure QS-wave morphology of the electrogram at the preexcitation site (inset, Figure 7I) indicated that the activation source was epicardial or subepicardial. This, together with the proximity of the epicardial breakthrough to the CS, suggested a left posterior epicardial pathway associated with the coronary venous system (either in the CS or its branches). The T wave in the electrogram at the preexcitation site was inverted (insets, Figures 6A, 6E, 6I, 7A, 7E, and 7⇓I). Forty-five minutes after ablation, the electrogram morphology during activation changed from r-S (or QS) to q-R. In contrast, the T-wave inversion persisted and remained (though to a lesser degree) even a week after ablation (insets, Figures 6B, 6C, 6F, 6G, 6J, 6K, 7B, 7C, 7F, 7G, 7J, and 7⇓K). The T wave became flatter with a small positive deflection a month after ablation (insets, Figures 6D, 6H, 6L, 7D, 7H, and 7⇓L). In the preexcited rhythm, the basal annular area had a high ARI (red, Figures 6A, 6E, 6I, 7A, 7E, and 7⇓I) with the longest ARI at the preexcitation site (349±6 ms; same for both the ARI method and alternative method) with an ARId of 95±8 ms (ARI method) and 85±9 ms (alternative method). The ARI distribution remained unchanged 45 minutes after ablation (r=0.91) despite the major changes in the activation sequence (Figures 3 and 4⇑). The ARId decreased after a week (Figures 6C, 6G, 6K, 7C, 7G, and 7⇓K) but was still high(79±9 ms [ARI method] and 66±9 ms [alternative method]). ARId was restored to close to normal values (45±9 ms [ARI method] and 39±7 ms [alternative method]) a month after ablation (Figures 6D, 6H, 6L, 7D, 7H, and 7⇓L). Activation sequences a week and a month after ablation were the same as the one 45 minutes after ablation (normal sinus rhythm).
Comparison With Sites of Successful Ablation
After successful ablation, the clock position of the ablation catheter was documented in the left anterior oblique projection of the fluoroangiogram using the clock diagram shown in Figure 2B. The Table compares the ECGI-localized site of preexcitation, the predicted location of the AP from the 12-lead ECG–based Arruda algorithm,5 and the site of successful ablation. ECGI-localized preexcitation sites were consistent with the successful ablation site to an accuracy of an hour’s arc. In 3 cases (subjects W3, W4, and W6), ECGI provided a clinically meaningful improvement in the pathway localization. It identified multiple pathways in subject W3. It pinpointed 1 pathway to the right posterior septum in subject W4 that could have been on either side of the septum according to the Arruda algorithm. It also correctly identified 1 left posterior epicardial pathway (subject W6) that appeared to be a right posteroseptal insertion from the ECG-based algorithm.
Noninvasive Localization of the AP
Although the conventional ECG-based Arruda algorithm (valid only for single APs) provides a broad subjective idea of the AP location, the ECGI prediction is more focused and corresponds better with the sites of successful ablation. ECGI is able to distinguish between endocardial and epicardial APs (from electrogram morphology). It also is able to determine the presence and locations of dual pathways before ablation. The spatial pattern of epicardial potentials during early activation (5 to 10 ms after the onset of the delta wave) is mirrored with reverse polarity in the initial repolarization phase, implying that there is no or minimal involvement of conduction system in the propagation of the preexcitation wave front.8 Hence, even for endocardial APs (12 of 14), the insertion site corresponds closely to the site of earliest epicardial activation (epicardial breakthrough). The study shows that ECGI can be used successfully for noninvasive localization of target ablation sites before catheter ablation in patients with WPW syndrome. It also shows that ECGI can be used as a follow-up tool after ablation to determine whether normal activation had been restored. This concept of preablation localization of target areas and postablation follow-up with ECGI can be extended to other arrhythmias or substrate-based ablation procedures in the future.
CM and Repolarization Changes
One important result of the study is the observation of abnormal repolarization in patients with WPW syndrome, resulting in altered epicardial apicobasal ARId. Epicardial apicobasal recovery in normal LV is characterized by an epicardial ARId of 40 to 50 ms, the posterolateral base having a longer ARI compared with the apex.9 ECGI repolarization images in normal adults9 have shown an LV epicardial basal ARI of 280 to 300 ms during resting normal sinus rhythm. The LV epicardial basal ARI in the resting WPW left-sided and posteroseptal preexcited rhythm is significantly longer at 337 to 355 ms, leading to an abnormally large ARId of 86 to 112 ms. A previous study16 provided direct evidence for endocardial ARI prolongation over the preexcited area in patients with manifest WPW, which persisted without any significant change minutes after ablation. Our study shows that in contrast to the fast return of activation to sinus rhythm after AP ablation, the repolarization pattern remains abnormal and unaltered. One week after ablation, the epicardial apicobasal recovery gradients decrease but are still abnormally high; they decrease to normal values only after 1 month. This time course is consistent with “long-term” CM. Suggested mechanisms for long-term CM2 include attenuation of the expression of the transient outward current (Ito) as a result of altered gene transcription by continued downregulation of the cAMP response element binding protein, altered kinetics of L-type calcium channel (ICa(L)), and changes in the density and kinetics of the rapid delayed rectifier potassium current (IKr). Experiments in canine models have shown that prolonged ventricular pacing downregulates nuclear cAMP response element binding protein, with the change being maximal at the pacing site and diminishing with distance.2 Consistent with this finding, our ECGI ARI maps (Figures 6 and 7⇑, first and fourth columns) show a maximal ARI difference between preablation and postablation rhythms at the preexcitation region (27%) with only an 8% change at remote apical locations. This spatial pattern and the long time course of resolution after ablation (1 month) suggest that CM in the human heart may be associated with transcriptional reprogramming.
Plotnikov et al17 showed that inducibility of CM in canine pacing experiments increases with age and correlates directly with the accentuation of the epicardial action potential notch attributed to the age-related evolution of Ito. The interpatient consistencies in the preablation ARI value at the preexcitation region and the time course of resolution after ablation do not indicate age dependence of inducibility or resolution of CM within the limited age spread of the patients included in this study. We could not find any data in the literature on age-related evolution of Ito in the human ventricular myocyte. However, data from atrial myocytes18 suggest that Ito is expressed in the human atria from early infancy and does not exhibit significant developmental changes.
Memory-related body-surface T-wave changes in patients with WPW syndrome were studied19 with vectorcardiography. Vectorcardiography also was used to study the development and resolution of CM induced by a week of maximal RV endocardial pacing in humans,20 reporting a resolution time of 1 month. Although vectorcardiography records the manifestation of CM in the body-surface T wave, ECGI maps cardiac activation and repolarization changes caused by the myocardial remodeling processes that underlie CM. The present study shows persistent ARI prolongation over the preablation epicardial region of preexcitation both before and after ablation. Soon after ablation, the activation sequence has reverted to normal as depicted by the ECGI isochrone maps (fourth column, Figures 3 and 4⇑). However, the deeper T wave of the epicardial electrogram (insets, second column, Figures 6 and 7⇑) at the preablation preexcitation site still tracks the preablation inverted r-S wave (insets, first column, Figures 6 and 7⇑), and the epicardial ARIs remain unchanged from their preablation distribution, returning to normal gradually over a time course of 1 month, consistent with long-term CM.
The demonstrated ability of ECGI to locate the region of WPW preexcitation more accurately than the ECG-based Arruda algorithm suggests its utility in guiding ablation of AP and evaluating its outcome noninvasively. The study shows that the altered activation sequence of the preexcitation rhythm is accompanied by high dispersion of repolarization, which resolves over a period of 1 month after conversion to normal sinus rhythm. It is well established21 that high dispersion of repolarization provides a substrate for unidirectional conduction block and reentrant arrhythmias. The results of the present study may apply to other clinical situations such as cardiac pacing in which a high dispersion of repolarization in a structurally abnormal heart might provide a possible mechanistic basis for pacing-induced proarrhythmia.22,23
The patient population (n=14) included in the study is small, and ECGI AP localization is compared with the predicted pathway location from only 1 ECG algorithm. Within these limitations, ECGI provides a clinically relevant improvement in pathway localization in 3 of 14 cases (21%). The present study also lacks quantitative measures of the accuracy of ECGI localization relative to the site of successful catheter ablation. A quantitative measure of localization error requires an exact and accurate method of registration between the ECGI CT anatomy and electrophysiological fluoroangiograms that can be implemented in the clinical setting of the electrophysiology laboratory. Moreover, ECGI localized the preexcitation site on the epicardial aspect of the atrioventricular annulus, whereas in most cases, the ablation site was endocardial, beneath this epicardial site.
Sources of Funding
The study was supported by Merit Award R37-HL-33343 and grant R01-HL-49054 from the National Heart, Lung, and Blood Institute to Dr Rudy. Dr Rudy is the Fred Saigh Distinguished Professor at Washington University in St Louis, Mo.
Dr Rudy chairs the scientific advisory board and holds equity in CardioInsight Technologies (CIT). CIT does not support any research conducted by Dr Rudy, including that presented here.
Pappone C, Santinelli V, Manguso F, Augello G, Santinelli O, Vicedomini G, Gulletta S, Mazzone P, Tortoriello V, Pappone A, Dicandia C, Rosanio S. A randomized study of prophylactic catheter ablation in asymptomatic patients with the Wolff-Parkinson-White syndrome. N Engl J Med. 2003; 349: 1803–1811.
Gallagher JJ, Smith WM, Kasell JH, Benson DW Jr, Sterba R, Grant AO. Role of Mahaim fibers in cardiac arrhythmias in man. Circulation. 1981; 64: 176–189.
Castellanos A, Agha AS, Befeler B, Myerburg RJ. Double accessory pathways in Wolff-Parkinson-White syndrome. Circulation. 1975; 51: 1020–1025.
Ramanathan C, Jia P, Ghanem RN, Ryu K, Rudy Y. Activation and repolarization of the normal human heart under complete physiological conditions. Proc Nat Acad Sci U S A. 2006; 103: 6309–6314.
Oster HS, Taccardi B, Lux RL, Ershler PR, Rudy Y. Noninvasive electrocardiographic imaging: reconstruction of epicardial potentials, electrograms and isochrones, and localization of single and multiple electrocardiac events. Circulation. 1997; 96: 1012–1024.
Ghanem RN, Burnes JE, Waldo AL, Rudy Y. Imaging dispersion of myocardial repolarization, II: noninvasive reconstruction of epicardial measures. Circulation. 2001; 104: 1306–1312.
Haws CW, Lux RL. Correlation between in vivo transmembrane action potential durations and activation-recovery intervals from electrograms: effects of interventions that alter repolarization time. Circulation. 1990; 81: 281–288.
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.
Inden Y, Hirai N, Takada Y, Shimizu A, Shimokata K, Yoshida Y, Akahoshi M, Kondo T, Saito H. Prolongation of activation-recovery interval over a preexcited region before and after catheter ablation in patients with Wolff-Parkinson-White syndrome. J Cardiovasc Electrophysiol. 2001; 12: 939–945.
Plotkinov AN, Sosunov EA, Patberg KW, Anyukhovsky EP, Gainullin RZ, Shlapakova IN, Krishnamurthy G, Danilo P Jr, Rosen MR. Cardiac memory evolves with age in association with development of transient outward current. Circulation. 2004; 110: 489–495.
Gross GJ, Burke RP, Castle NA. Characterisation of transient outward current in young human atrial myocytes. Cardiovasc Res. 1995; 29: 112–117.
Baker LC, London B, Choi B-R, Koren G, Salama G. Enhanced dispersion of repolarization and refractoriness in transgenic mouse hearts promotes reentrant ventricular tachycardia. Circ Res. 2000; 86: 396–407.
Himmrich E, Przibille O, Zellerhoff C, Liebrich A, Rosocha S, Andreas K, Nebeling D, Omogbehin B, Meyer J. Proarrhythmic effect of pacemaker stimulation in patients with implanted cardioverter-defibrillators. Circulation. 2003; 108: 192–197.
ECG imaging, a noninvasive imaging modality for cardiac electrophysiology, determines the locations of Wolff-Parkinson-White accessory pathways with greater accuracy than the conventional ECG-based Arruda algorithm. This capability makes ECG imaging an effective tool for guiding accessory pathway ablation procedures, possibly shortening their time and reducing the associated iatrogenic risks. ECG imaging also can be used to evaluate the outcome of ablation. The study shows that persistent preexcitation, which constitutes a natural model for prolonged ventricular pacing from the accessory pathway insertion site, leads to high dispersion of repolarization and a prolonged activation-recovery interval (action potential duration) at the preexcited (paced) region. The abnormal repolarization gradients resolve over a period of 1 month after a return to normal sinus rhythm by successful ablation. This time course is consistent with long-term cardiac memory, possibly involving transcriptional changes that trigger remodeling processes that alter the molecular determinants of action potential duration. The results of this study may apply to other clinical situations like cardiac pacing in which a high dispersion of repolarization in a structurally abnormal heart might provide a possible mechanistic basis of pacing-induced proarrhythmia reported in the literature.