PACING/ICD/CRT
Impact of left ventricular epicardial and biventricular pacing on ventricular repolarization in normal-heart individuals and patients with congestive heart failure
1 Department of Internal Medicine/Cardiology, Tong-Ji Hospital, Tong-Ji Medical College, Huazhong University of Science and Technology, Wuhan 430030, People's Republic of China; 2 Department of Cardiology, National Heart Centre, Singapore 168752, Singapore
Manuscript submitted 3 November 2005. Accepted after revision 8 July 2006.
* Corresponding author. Tel/fax: +86 27 83632827. E-mail address: bairong74{at}yahoo.com.cn
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Aims Malignant ventricular arrhythmias can arise in a subset of congestive heart failure (CHF) patients after they undergo cardiac resynchronization therapy (CRT), thus counteracting the haemodynamic benefits typically associated with biventricular pacing. This study seeks to assess whether alteration of the ventricular transmural repolarization and conduction due to reversal of the depolarization sequence during epicardial or biventricular pacing facilitate the development of ventricular arrhythmias.
Methods and results ECGs and monophasic action potential (MAP) were recorded during programmed stimulation from right ventricle (RV) endocardium (RV-Endo), left ventricle (LV) epicardium (LV-Epi), or both (biventricular, Bi-V) in 15 individuals without structural heart diseases. In patients with severe CHF and CRT (n=21), ECGs were collected during RV-Endo, LV-Epi, and Bi-V pacing. MAP duration on intracardiac electrogram, the QT, JT, and Tpeak–Tend intervals on ECGs at different pacing sites were measured and compared. In subjects with or without structural heart disease, compared with RV-Endo pacing, LV-Epi and Bi-V pacing resulted in a longer JT (341.78±61.97 ms with LV-Epi, 325.86±59.69 ms with Bi-V vs. 286.14±38.68 ms with RV-Endo in CHF individuals, P<0.0001) or Tpeak–Tend interval (121.55±19.88 ms with LV-Epi, 117.71±42.63 ms with Bi-V vs. 102.28±12.62 ms with RV-Endo in normal-heart subjects, P<0.0001; 199.70±62.44 ms with LV-Epi, 184.89±74.08 ms with Bi-V vs. 146.41±31.06 ms with RV-Endo in CHF patients, P<0.0001), in addition to prolonged myocardial repolarization time and delayed endocardial activation. During follow-up, sudden death and arrhythmia storm occurred in two CHF patients after CRT.
Conclusion Epicardial and biventricular pacing prolong the time and increase the dispersion of myocardial repolarization and delay the transmural conduction. All of these should be considered as potential arrhythmogenic factors in CHF patients who receive CRT.
Key Words: Cardiac resynchronization therapy, Congestive heart failure, Epicardial pacing, Monophasic action potential, Tpeak–Tend interval, Transmural dispersion of repolarization
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Cardiac resynchronization therapy (CRT) can improve ventricular haemodynamics and the quality of life in patients with severe congestive heart failure (CHF); however, there is a risk that malignant ventricular arrhythmias can arise as a result of its therapeutic mechanism.1
–3 During CRT, activation of the left ventricle (LV) is from the epicardium to endocardium, whereas activation of the right ventricle (RV) is from the endocardium to epicardium. This non-physiological sequence may augment the transmural heterogeneity of repolarization, which is intrinsic to the ventricular myocardium, and thus subsequently facilitates the development of ventricular arrhythmias. The present study evaluates the feasibility of this hypothesis by assessing the effects of left ventricular epicardial pacing, right ventricular endocardial pacing, and biventricular pacing on ventricular repolarization properties and their potential arrhythmogenesis in CHF patients treated with CRT. |
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Patients without structural heart disease
Fifteen patients (aged 33±11 years, 10 males) who underwent successful radiofrequency catheter ablation (RFCA) for supraventricular tachycardia were recruited. The presence of structural heart disease was excluded in all subjects by evaluation of past medical history, physical examination, treadmill test, chest X-ray, and echocardiography examination. After RFCA, normal electrophysiological (EP) study results were established if an atrioventricular, intra-, or inter-ventricular block was not noted and no other tachycardia was induced. Written consent was obtained from all subjects for further EP studies after the RFCA procedure and the protocol was approved by the Ethics Committee of the hospital.
A 6 French (F) quadripolar endocardial mapping catheter (Biosense Webster, Diamond Bar, CA, USA) was passed into the coronary sinus (CS) via left subclavian vein approach. This catheter, later used for bipolar LV epicardial (LV-Epi) pacing, was advanced forward into the anterior or posterolateral branch of CS. Catheter position was confirmed by fluoroscopy (Angio Diagnost 5, Philips, Eindhoven, The Netherlands), intracardiac electrogram from the catheter, and QRS morphology on standard surface ECG when pacing from this catheter (Prucka CardioLab, GE, Piscataway, NJ, USA). Another 8 F steerable Franz MAP/pacing catheter (EP Technology, Natick, NJ, USA) was inserted from the right femoral vein and positioned at RV apex. This catheter can be used for simultaneous bipolar endocardial [RV endocardium (RV-Endo)] pacing and monophasic action potential (MAP) recording at the same area where the tip electrode contacts the RV-Endo.
The RV-Endo pacing and the LV-Epi bipolar pacing were delivered from the RV (MAP/pacing) catheter and LV (CS) catheter, respectively. For biventricular (Bi-V) bipolar pacing, the electrode tips of LV and RV catheters were connected and used as the cathode, whereas the proximal electrodes were connected and used as the anode. Bi-V pacing was confirmed by a narrower paced QRS complex on ECG.
S1S1 pacing was performed for 10–15 s at each pacing cycle length (CL), starting at 500 ms and decreasing by 50 ms until a minimum CL of 250 ms. S1S2 pacing was performed with a fixed driving CL of 500 ms (S1) for eight beats and one extrastimulus (S2) starting from 400 ms and decreasing by 10 ms until a minimum S1S2 coupling interval of 250 ms or until ventricular ectopic couplets or non-sustained ventricular tachycardia was induced. Both S1S1 and S1S2 pacing were delivered from RV-Endo, LV-Epi, and Bi-V.
Standard 12-lead ECG, intracardiac electrogram from LV and RV catheters, and MAP from the RV catheter were recorded simultaneously during pacing and saved onto a disk (Prucka CardioLab). All the ventricular arrhythmias induced by S1S2 stimulation were counted. These data were later retrieved and the following parameters were measured by two different electrophysiologists using the ‘measure tool’ (caliper) of the polygraph.
QT interval
The QT interval was defined as the time interval between the initial deflection of the QRS complex and the point at which a tangent can be drawn to the steepest portion of the terminal part of T-wave, which crosses the isoelectric line. The QT interval was measured on leads I, aVF, V2, and V5 of ECG.
Tpeak–Tend (Tp–e) interval
The Tp–e was defined as the interval between the peak and the end of T-wave. The Tp–e interval was measured on leads I, aVF, V2, and V5 on ECG, and biphasic, flat, or bifurcated T-waves were avoided.
Full duration of MAP
The duration of MAP (MAPD) was measured from the initial deflection point of phase 0 of MAP to the terminal point of phase 4 of MAP, where it crossed the isoelectric line (99% repolarization). The activation conduction time (ACT) was defined as the interval from the stimulation spike to the initial deflection point of MAP. The sum of MAPD and ACT was the so-called activation-recovery interval (ARI), which was measured from the stimulus signal to the terminal point of MAP (ARI=ACT+MAPD).
Patients with CHF
Twenty-one patients with either pacemaker or defibrillator CRT devices [Medtronic (Minneapolis, MN, USA) InSync Series or Guidant (St Paul, MN, USA) Contak Series, USA] implanted for severe CHF (New York Heart Association Class III or IV) were studied. The cardiomyopathy aetiology (ischaemic vs. non-ischaemic) was determined by coronary angiogram in conjunction with other standard evaluations. Prior to implantation, all patients had been on optimal drug therapy, which included ACE-inhibitors, diuretics, digoxin, and beta-blockers. Six (29%) of the patients were on the antiarrhythmic agent amiodarone before the procedure. All the patients had an AHA/ACC Class I or II indication for Bi-V permanent pacemaker (Bi-V PPM) or Bi-V implantable cardioverter-defibrillator (Bi-V ICD) implantation. The RV pacing or defibrillatory leads were fixed at RV apex or septum, whereas the LV lead was placed in an epicardial cardiac vein via retrograde CS access. Satisfactory testing parameters were obtained during the procedure in all subjects. Patients were followed-up for a mean period of 29.28±18.38 months and their clinical profiles are shown in Table 1. Immediately after the procedure or at the follow-up visit, patients' standard ECGs were recorded at both 25 and 50 mm/s paper speed after programming the device, respectively, as RV-Endo pacing alone, LV-Epi pacing alone and Bi-V pacing. At the time of ECG recording, the device was temporarily set at 80 bpm. The following parameters were measured manually on ECG leads I, aVF, V2, and V5 by two cardiologists using calipers:
- QT interval. The QT interval was defined as previously mentioned.
- JT interval. The JT interval that excludes the QRS component was measured as the time from Point J to the end of T-wave.
- Tp–e interval. The Tp–e interval was defined as previously mentioned.
|
Statistical analysis
Statistical analysis was performed using SPSS (Chicago, IL, USA) 8.0 or SPSS 10.0 software. F-test was used in the multivariate analysis of variance (MANOVA) to test the between-subjects difference in ECG or intracardiac electrogram parameters which were under the effects of multiple variants (pacing site, pacing rate, ECG lead). Bonferroni was the post hoc test in the subsequent multiple comparison.
2 test was used to compare the number of ventricular arrhythmic events during S1S2 pacing at different sites. All data were expressed as mean±SD, and the criterion for statistical significance was P<0.05. |
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Results of patients without structural heart disease
At any pacing rate, the QRS duration of Bi-V pacing was significantly shorter than that of RV-Endo or LV-Epi pacing (117.47±11.92 ms with Bi-V pacing vs. 154.80±11.03 ms with RV-Endo pacing or 162.20±9.86 ms with LV-Epi pacing, P<0.01). LV-Epi pacing (377.19±39.43 ms) resulted in the longest QT interval, followed by those of Bi-V pacing (367.10±52.69 ms) and then RV-Endo pacing (349.44±32.30 ms, F=81.34, P<0.0001) regardless of the pacing CL. Compared with RV-Endo pacing (102.28±12.62 ms), LV-Epi pacing (121.55±19.88 ms) and Bi-V (117.71±42.63 ms) pacing resulted in a longer Tp–e interval (F=108.10, P<0.0001). There was no difference between ECG leads for the QT (F=0.089, P=0.966) and Tp–e intervals.
The MAPD of RV-Endo was significantly longer during LV-Epi pacing (294.12±49.65 ms) and Bi-V (273.01±25.16 ms) pacing compared with RV-Endo pacing (252.58±22.34 ms, F=48.25, P<0.0001).
Of the total 330 episodes of S1S2 stimulation in all subjects, ventricular arrhythmic events were more easily and frequently induced if the pacing was delivered from LV-Epi or Bi-V pacing (Table 2, Figure 1A–C).
|
|
Results of patients with CHF
ECG measurements
The QRS duration was broader during LV-Epi (194.07±25.27 ms) or RV-Endo (196.00±22.79 ms) pacing than that of intrinsic rhythm (173.60±32.91 ms), but became narrower during Bi-V pacing (137.74±18.18 ms, P<0.0001).
The QT interval was longer during LV-Epi pacing compared with RV-Endo pacing and Bi-V pacing (F=17.65, P<0.0001), but there was no difference between the latter two pacing sites (P=1.0). Significant pacing site-dependent difference was demonstrated in both JT interval (341.78±61.97 ms with LV-Epi pacing, 325.86±59.69 ms with Bi-V pacing vs. 286.14±38.68 ms with RV-Endo pacing, F=18.91, P<0.0001) and Tp–e interval (199.70±62.44 ms with LV-Epi pacing, 184.89±74.08 ms with Bi-V pacing vs. 146.41±31.06 ms with RV-Endo pacing, F=16.81, P<0.0001). Thus, there was an accordant relationship: LV-Epi pacing>Bi-V pacing>RV-Endo pacing (Figure 2).
|
Clinical outcome
Following implantation, all 21 patients reported improvement in symptoms during the follow-up period. Of those six patients who were on antiarrhythmic agents prior to the procedure, amiodarone was maintained in the same dosage in five patients but changed to sotalol in one. Three (14%) patients were newly prescribed amiodarone after Bi-V PPM insertion because non-sustained VT was observed on Holter or telemetried tracing from the pacemaker memory. Four patients (19%) died during the follow-up: the reason for death was ‘severe liver and renal dysfunction’ in one and ‘complications of heart transplantation’ in the other. The third patient with Bi-V PPM died suddenly outside hospital but unfortunately the device-memory data were not retrieved. Another patient with previous myocardial infarction who underwent Bi-V ICD implantation because of inducible monomorphic VT during EP study suffered an arrhythmia storm (incessant monomorphic and polymorphic VT, Figure 3) after the procedure. He was re-admitted several times for multiple ICD therapies even with large doses of antiarrhythmic drugs. Finally, this patient died from ventricular tachyarrhythmia. The ECG measurements in the fourth patient during LV-Epi, Bi-V, and RV-Endo pacing were 682.50±22.55, 546.25±33.00, and 577.50±28.43 ms for QT interval; 468.75±45.53, 370.00±31.88, and 353.75±22.87 ms for JT interval; 257.50±54.39, 178.75±31.19, and 163.75±26.26 ms for Tp–e interval.
|
4:40) on eighth day after Bi-V ICD implantation. ‘P’ and ‘V’ refer to paced and spontaneous ventricular beats, respectively. The paced QRS duration was 120–140 ms, which indicated biventricular pacing. The P–V coupling interval between the first QRS complex of every run of ventricular tachycardia and the preceding paced beat was constant, which implied a possible re-entry or trigger mechanism of these ventricular arrhythmias. |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Main findings
Under normal physiological conditions, depolarization of the ventricular myocardium starts from subendocardium via midmyocardium and spreads across the ventricular walls to subepicardium, whereas repolarization occurs in the reverse direction. It is not clear how transmural repolarization and conduction might be affected if the depolarization sequence is altered. During CRT, the RV-Endo and LV epicardium are activated simultaneously, which might change the sequence of intra- and inter-ventricular activation in such a way as to prolong the transmural repolarization time (RT) or change the direction and velocity of transmural conduction.4
All of these factors may contribute to the development of malignant arrhythmias. The present study evaluates the feasibility of this hypothesis by assessing the effects of different pacing sites on ventricular repolarization properties, especially the changes in some ECG measurements.
The QT interval is classically used to measure myocardial RT on ECG, but it actually includes both ventricular depolarization time (DT) and RT. On ECG, the QT interval calculates the sum of ventricular transmembrane DT and maximum RT (QT=DT+RTmax).5,6 Therefore, for any changes in RT to be reflected in the QT interval, the DT (QRS duration) must remain constant. When the normal direction of ventricular activation is reversed, as during LV-Epi or Bi-V pacing, both transmural conduction time and DT are altered. The QT interval may no longer be used to represent myocardial RT. Even though the present study demonstrated pacing site-dependent prolongation of QT intervals in individuals without structural heart disease, the results were not consistent with those in CHF patients and our previous animal experiments.7 In these special situations with pacing-induced variable intra- or inter-ventricular conduction time, new ECG parameters, which more accurately measure myocardial RT and describe its properties, should be introduced.8
Point J on ECG always represents the beginning of subepicardial repolarization, and JT interval measures the whole transmural RT. Our study verified that the LV-Epi or Bi-V pacing not only significantly prolong the JT interval in patients with CHF reflecting augmentation of overall transmural RT, but also amplify the MAPD of the subendocardium in normal-heart individuals, which represents an increase in local RT. Furthermore, LV-Epi pacing results in a remarkable delay in subendocardial activation, which is demonstrated by significantly longer ACT on the intracardiac electrogram (Figure 4A and B). This epicardial-to-endocardial conduction time is much longer than the normal endocardial-to-epicardial propagation time. These findings were consistent with the latest results from both Fish et al.9 and Poelzing et al.10 who have demonstrated in canine LV wedge preparations that the additional conduction delay during epicardial pacing occurs at the border region from subepicardium to midmyocardium and is due to the increased tissue resistivity in the deep subepicardium. The fact that pacing from LV epicardium leading to delayed activation and RT prolongation of RV-Endo may worsen the heterogeneity in different myocardial layers and contribute to the amplification of transmural dispersion of repolarization (TDR). The latter has been acknowledged as one of the substrates and plays a role in the development of malignant ventricular arrhythmias.11,12 During Bi-V pacing, the status is more complex because the myocardial depolarization, repolarization, and propagation of both ventricles come from or go in different directions.
|
In this study, the longest Tp–e interval presents during LV-Epi pacing, followed by Bi-V pacing, both of which are much longer than that during RV-Endo pacing. This prominent increase in Tp–e interval, as a consequence of reversal of direction of transmural activation, is a risk factor for the development of torsade de pointes (TdP).1
,4,13 One may also note that the shifting of pacing site prolongs the Tp–e interval more significantly in normal-heart individuals (F=108.10) than in patients with CHF (F=16.81). But this would not predicate safety of epicardial or Bi-V pacing in CHF patients because other confounding intrinsic factors, such as regional repolarization dispersion and spontaneous afterdepolarization, also provide arrhythmogenic substrates or triggers and facilitate the development of VTs in the diseased heart.
Interestingly, the results from present study conflict with some prior ones. Berger et al.8 measured Tp–e integral and QT interval from 65-lead body surface mapping and stated that LV-Epi pacing alone may have detrimental effects on myocardial repolarization but Bi-V pacing reduced TDR. Importantly, Tp–e integral obtained from body surface mapping may involve more ‘regional’ components rather than ‘transmural’ aspects of myocardial repolarization, compared with Tp–e interval obtained from standard ECG. Whether this Tp–e integral parameter is closely associated with clinical arrhythmia events as much as Tp–e interval remains unclear. On the other hand, previous experimental findings by Medina-Ravell et al.1 and Fish et al.9 implied that an increase in TDR during LV-Epi pacing and Bi-V pacing is the consequence of altered ventricular activation sequence independent of action potential duration. This is probably due to a difference in methodology of recording action potential duration. In their studies, intracellular action potentials across the ventricular wall (transmembrane action potential, TAP) were recorded, whereas in our in vivo study, MAPs were measured. As we know, MAP is a composite signal that reflects changes in membrane potentials of numerous myocytes in ‘local’ myocardium. Therefore, the MAPD may represent not only ventricular repolarization per se but also a repolarization gradient locally, and it may not be directly comparable with TAP measurements.
Recent reports14,15 in relatively small series of patients with CRT have indicated improvement in ventricular heterogeneity of repolarization. These results seem to be in contrast with those presented here but may possibly be explained by differences in patient selection.
The follow-up data of the CHF patients in our study were also alarming. Isolated cases should not be ignored, as they may be those at high risk, which has been previously underestimated. On the other hand, to distinguish whether sudden death or an arrhythmic event was due to the heart failure or to the CRT device remains difficult, as so far none of the clinical trials has clarified the exact difference in sudden death and arrhythmic events between pre- and post-CRT in CHF patients. The AHA/ACC guidelines state that all CRT device implanters should be aware of the potential risk of ventricular pro-arrhythmia after biventricular pacing.16 In the population of the Cardiac Resynchronization Heart Failure (CARE-HF) Study, which was the first to declare a mortality benefit of CRT,17 the incidence of sudden cardiac death remains high or even higher than that in the general CHF cohort.18
Study limitations
Generalization of the results is limited by the small patient sample and the non-controlled, non-randomized design of this study. The data from patients of two groups were not directly comparable and the follow-up of the CHF patients was not prospective. Evidence is insufficient to demonstrate directly that ventricular arrhythmias which occurred in CHF patients were CRT related.
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Epicardial and biventricular pacing prolong the time and increase the dispersion of myocardial repolarization and delay the transmural conduction through the ventricular walls. All of these risk factors should be considered as potential arrhythmogenic substrates in CHF patients who receive CRT. Further randomized, controlled clinical investigations are necessary to focus on the incidence of CRT-related arrhythmic events and sudden death and to analyse the precise causes of mortality after CRT in CHF patients.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
The authors express their gratitude to Mrs Yang You Nian from Audit Office of Singapore National Eye Centre and Dr Wei Shen from Institute of Public Health of Tong-Ji Medical College for their help in statistical analyses.
Technical support from our colleagues Mrs Wang Chen, Mrs Luo Hui Zhen (Tong-Ji Hospital, Wuhan, China), Mrs Daw Mya Mya, and Miss Kuik Hwee Bing (Singapore National Heart Centre) are important for the completion of this study. We are grateful to Dr Dimpi Patel from Cleveland Clinic Foundation (USA) for revising the manuscript.
This study was partially supported by Scientific Grant of Tong-Ji Hospital (Wuhan, China) and The National Natural Science Foundation of China (NSFC-30370573 and NSFC- 30470714).
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
[1] Medina-Ravell VA, Lankipalli RS, Yan GX, Antzelevitch C, Medina-Malpica NA, Medina-Malpica OA, et al. Effect of epicardial or biventricular pacing to prolong QT interval and increase transmural dispersion of repolarization: does resynchronization therapy pose a risk for patients predisposed to long QT or torsades de pointes? Circulation 2003; 107: 740–6.
[2] Guerra JM, Wu J, Miller JM, Groh WJ. Increase in ventricular tachycardia frequency after biventricular implantable cardioverter defibrillator upgrade. J Cardiovasc Electrophysiol 2003; 14: 1245–7.[CrossRef][Web of Science][Medline]
[3] Shukla G, Chaudhry GM, Orlov M, Hoffmeister P, Haffajee C. Potential proarrhythmic effect of biventricular pacing: fact or myth? Heart Rhythm 2005; 2: 951–6.[CrossRef][Web of Science][Medline]
[4] Di Diego JM, Belardinelli L, Antzelevitch C. Cisapride-induced transmural dispersion of repolarization and torsade de pointes in the canine left ventricular wedge preparation during epicardial stimulation. Circulation 2003; 108: 1027–33.
[5] Yan GX and Antzeleivtch C. Cellular basis for the normal T wave and the electrocardiographic manifestations of the Long-QT syndrome. Circulation 1998; 98: 1928–36.
[6] Franz MR, Bargheer K, Rafflenbeul W, Haverich A, Lichtlen PR. Monophasic action potential mapping in human subjects with normal electrocardiograms: direct evidence for the genesis of the T wave. Circulation 1987; 75: 379–86.
[7] Bai R, Pu J, Liu N, Lu JG, Zhou Q, Ruan YF, et al. Influence of pacing site on myocardial transmural dispersion of repolarization in intact normal and dilated cardiomyopathy dogs (in Chinese). Sheng Li Xue Bao 2003; 55: 722–30.[Medline]
[8] Berger T, Hanser F, Hintringer F, Poelzl G, Fischer G, Modre R, et al. Effects of cardiac resynchronization therapy on ventricular repolarization in patients with congestive heart failure. J Cardiovasc Electrophysiol 2005; 16: 611–17.[CrossRef][Web of Science][Medline]
[9] Fish JM, Di Diego JM, Nesterenko V, Antzelevitch C. Epicardial activation of left ventricular wall prolongs QT interval and transmural dispersion of repolarization: implications for biventricular pacing. Circulation 2004; 109: 2136–42.
[10] Poelzing S, Dikshteyn M, Rosenbaum DS. Transmural conduction is not a two-way street. J Cardiovasc Electrophysiol 2005; 16: 455.[Web of Science][Medline]
[11] Akar FG and Rosenbaum DS. Transmural electrophysiological heterogeneities underlying arrhythmogenesis in heart failure. Circ Res 2003; 93: 638–45.
[12] Yan GX, Lankipalli RS, Burke JF, Musco S, Kowey PR. Ventricular repolarization components on the electrocardiogram: cellular basis and clinical significance. J Am Coll Cardiol 2003; 42: 401–9.
[13] Antzelevitch C. Transmural dispersion of repolarization and T wave. Cardiovasc Res 2001; 50: 426–31.
[14] van Huysduynen B, Swenne CA, Bax JJ, Bleeker GB, Draisma HHM, van Erven L, et al. Dispersion of repolarization in cardiac resynchronisation therapy. Heart Rhythm 2005; 2: 1286–93.[CrossRef][Web of Science][Medline]
[15] Santangelo L, Ammendola E, Russo V, Cavallaro C, Vecchione F, Garofalo S, et al. Influence of biventricular pacing on myocardial dispersion of repolarization in dilated cardiomyopathy patients. Europace 2006; 8: 502–5.
[16] Strickberger SA, Conti J, Daoud EG, Havranek E, Mehra MR, Pina IL, et al. Council on Clinical Cardiology Subcommittee on Electrocardiography and Arrhythmias and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Heart Rhythm Society: Patient selection for cardiac resynchronization therapy: from the Council on Clinical Cardiology Subcommittee on Electrocardiography and Arrhythmias and the Quality of Care and Outcomes Research Interdisciplinary Working Group, in collaboration with the Heart Rhythm Society. Circulation 2005; 111: 2146–50.
[17] Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, et al. Cardiac Resynchronization-Heart Failure (CARE-HF) Study Investigators. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005; 352: 1539–49.
[18] Cannom DS. ACC Annual Scientific Session Interview. http://www.cardiosource.com/rapidnewssummaries/index.asp?EID=22&DoW=Sun&SumID=153(4 April 2006).
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||