Cardiac Memory Studies in Two Human Models

6 List of original papers 7 List of abbreviations 8 Introduction 9 Cardiac cellular electrophysiology 12 Recording cardiac electricity 16 Electrocardiography (ECG) 16 Vectorcardiography (VCG) 16 Pacemaker treatment 20 Sick Sinus Syndrome 21 Hypertrophic Obstructive Cardiomyopathy (HOCM) 21 Wolff-Parkinson-White Syndrome (WPW syndrome) 22 Aims 23 Material and Methods 24 Studies I and II 24 Study III 25 Study IV 26 Study V 27 VCG analysis (Studies I, II and V) 29 ECG analysis (Studies II and III) 29 Statistical analysis 29 Results 31 Time-dependent variations (Studies I, II and V) 31 Cardiac memory development (Studies I, II and III) 31 Presence of cardiac memory after WPW ablation (Studies IV and V) 35 Cardiac memory dissipation (Studies I, IV and V) 38 Cardiac memory – good, bad or neutral (Studies II and III) 40 Summary of results 42 General discussion 43 Definition of cardiac memory 43 Cardiac memory and ventricular pacing 43 Cardiac memory and HOCM 45 Cardiac memory and the WPW syndrome 46 Methodology aspects 48 Mechanisms of cardiac memory 49 Implications 50 Limitations 52 Cardiac memory – future speculation 52 Conclusions 54 Acknowledgements 55 References 57 Appendix 62 Papers I-V ABSTRACT 6 Liliane Wecke6 Liliane Wecke Background Cardiac memory is a form of electrical remodeling of the ventricles, where the T vector follows (“remembers”) a previously altered QRS vector. On the electrocardiogram (ECG), it presents as T-wave inversions. It has been observed after periods of ventricular pacing, ventricular tachycardia, intermittent bundle branch block and after periods of preexcitation in patients with the Wolff-Parkinson-White (WPW) syndrome. Aims To study the occurrence, development and dissipation of cardiac memory in two human models, ventricular pacing and WPW ablation. In addition to conventional ECG analysis, the spatial vectorcardiogram (VCG) was analyzed in 3 dimensions, including T-vector loop morphology. Studies I-II Twenty patients who received DDD-R pacemakers due to symptomatic sinus bradycardia were followed daily for 1 week (n=6) or weekly for 5-8 weeks (n=14). A baseline ECG and VCG were recorded prior to pacemaker implantation. The patients were all paced endocardially in the right ventricle and the pacemakers were programmed to a short AV delay to achieve maximum ventricular stimulation. At each follow-up, the pacemakers were temporarily programmed to AAI mode to re-establish normal ventricular activation and an ECG and VCG were recorded. After this protocol, the 14 patients who were studied weekly had their pacemakers reprogrammed to individually optimized AV delays and, after 4-5 weeks, they were seen at one last follow-up. Repolarization changes assessed by T-vector amplitude and morphology changes were already present within 24 hours after the initiation of pacing. Cardiac memory was fully developed within one week and the VCG revealed a backward and upward rotation of the maximum T vector. After the cessation of ventricular pacing, cardiac memory disappeared within 4 weeks. When ventricular pacing was continued, cardiac memory was preserved in proportion to the percentage of ventricular pacing. After the abrupt termination of ventricular pacing, the prolongation of repolarization time (QTc) and changes in T-loop morphology suggested unfavorable effects on repolarization. During continuous ventricular pacing, the repolarization time decreased. Study III The effects of ventricular pacing on repolarization and cardiac memory were compared in 15 patients with hypertrophic obstructive cardiomyopathy (HOCM) and the patients from Studies I and II. A baseline ECG was recorded prior to pacemaker implantation. The HOCM patients were evaluated after 3 months of right ventricular endocardial pacing with a short AV delay. An ECG was first recorded during DDD pacing and then during sinus rhythm after pacing was temporarily switched off. T-wave inversions consistent with cardiac memory were observed in all HOCM patients and similar to controls. In contrast to sick sinus patients, there was no change in repolarization time in the HOCM group. Baseline myocardial structure and function thus make a difference to the repolarization response related to cardiac memory. Studies IV-V The occurrence and dissipation of cardiac memory after an accessory pathway ablation in WPW patients were studied retrospectively (Study IV) and prospectively (Study V). In Study IV, T-wave inversions in leads II, aVF and III were evaluated on ECGs from 125 patients ablated due to a posteroseptal (PS) accessory pathway. Within one day after ablation, 123 (98%) of the patients showed cardiac memory (T-wave inversions). In most patients, cardiac memory disappeared within 3 months. In Study V, ECGs and VCGs were used to compare the occurrence of cardiac memory after the ablation of PS (n=11) and left lateral (LL) (n=6) accessory pathways. Cardiac memory in the form of T-wave inversions on ECG was only seen in the PS group. Evaluation by VCG, however, revealed T-vector changes in the transverse plane in the LL group and vertical T-vector changes in the PS group. In 82% of the patients, cardiac memory disappeared within 3-4 weeks. Conclusions In humans, cardiac memory developed and reached steady state within 1 week after the onset of right ventricular endocardial pacing. After WPW ablation, cardiac memory was present, independent of accessory pathway site. Cardiac memory is probably an adaptation mechanism of the heart, involving a gradual change in repolarization when the activation sequence (depolarization) is altered. VCG including T-vector loop morphology is a useful and sensitive method for evaluating repolarization changes.