The T wave: physiology and pathophysiology · description of the T wave of a normal pig heart in...

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

The T wave: physiology and pathophysiology

Meijborg, V.M.F.

Link to publication

Citation for published version (APA):Meijborg, V. M. F. (2015). The T wave: physiology and pathophysiology.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 17 Jan 2021

https://dare.uva.nl/personal/pure/en/publications/the-t-wave-physiology-and-pathophysiology(c11e7e29-c45c-46b3-b678-848ce7dcf183).html

The T wave: physiology and pathophysiologyVeronique M.F. Meijborg

Th

e T w

ave: ph

ysiology and

path

oph

ysiology

Vero

niq

ue M

.F. Meijb

org

The T wave: physiology and pathophysiology

Veronique M.F. Meijborg

Lay out, cover design and printing:Optima Grafische Communicatie, Rotterdam, the Netherlandswww.ogc.nl

© Veronique M.F. Meijborg, 2015All rights reserved. No part of this thesis may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, re-cording or otherwise without permission of the author or the copyright owning journal.

ISBN 978-94-6169-710-3

Financial support by the Dutch Heart Foundation for the publication of this thesis is gratefully acknowledged.

Additional generous financial support for printing this thesis was provided by: BioSemi B.V., BIOTRONIK Nederland B.V., ChipSoft B.V., NAMCO Healthcare Technology, St. Jude Medical Nederland B.V., and the Academic Medical Center – University of Amsterdam

THE T WAVE: PHYSIOLOGY AND PATHOPHYSIOLOGY

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctoraan de Universiteit van Amsterdamop gezag van de Rector Magnificus

prof. dr. D.C. van den Boomten overstaan van een door het College voor Promoties ingestelde commissie,

in het openbaar te verdedigen in de Agnietenkapelop donderdag 8 oktober 2015, te 14:00 uur

door

Veronique Marlinde Frederica Meijborggeboren te Zoetermeer

PROMOTIECOMMISSIE

Promotores: prof.dr. A.A.M. Wilde Universiteit van Amsterdam prof.dr.ir. J.M.T. de Bakker Universiteit van AmsterdamCopromotores: dr. R. Coronel Universiteit van Amsterdam dr. C.E. Conrath Universiteit van Amsterdam

Overige leden: prof.dr. P. Taggart University College London prof.dr. K. Zeppenfeld Universiteit Leiden prof.dr. F.W. Prinzen Universiteit Maastricht prof.dr. M.J. Janse Universiteit van Amsterdam dr. P.F.H.M. van Dessel Universiteit van Amsterdam dr. C.A. Remme Universiteit van Amsterdam

Faculteit der Geneeskunde

If you find a path with no obstacles,it probably doesn’t lead anywhere.

Frank A. Clark

Table of contents

1. Introduction and outline 9

2. Electrocardiographic T wave and its relation with ventricular repolarization along major anatomical axesCirculation: Arrhythmia and Electrophysiology 2014

17

3. Synchronization of repolarization by mechano-electrical coupling in the porcine heartCardiovascular Research 2015

39

4. The effect of increased left ventricular pressure on the STT segment in the surface electrocardiogramSubmitted for publication

55

5. Regional conduction slowing induces inferolateral J-wavesSubmitted for publication

69

6. Interventricular dispersion in repolarization causes bifid T waves in dogs with dofetilide-induced long QT syndromeHeart Rhythm 2015

91

7. QT-TQ dynamics in congenital long QT patients and controls during a supine-standing ECG testSubmitted for publication

115

8. Summary and future perspectives 131

9. Samenvatting en toekomstperspectieven 143

Appendices

List of publications 157

Contributing authors 159

Curriculum Vitae 163

Portfolio 161

Acknowledgements / Dankwoord 165

Chapter 1

Introduction and outline

Introduction and outline 11

1The electrocardiogram (ECG) is commonly used to evaluate the electrical activity of the heart. Due to its non-invasive character and diagnostic power,1 it has become one of the most essential tests that cardiologists use in their daily practice. The ECG results from the potential difference between the electrodes which is generated by the transmem-brane currents running in extracellular space. Implementation of the ECG in the clinical practice dates from the early 20th century when the Dutch physiologist Willem Einthoven modified the string galvanometer as an ECG recorder with a very high sensitivity.2 Prior to this, he also had contributed significantly to the description of the ECG.2,3 The 5 typical deflections in the ECG were named P-Q-R-S-T.

Before the first description of the ECG, investigators were already occupied with the electrical activity within the heart. In 1856, Kölliker and Müller developed an inventive method to investigate the cardiac electrical properties.4 They positioned the cut end of a frog sciatic nerve – still attached to the musculus gastrocnemius – on the surface of the ventricle of the heart. A contraction of the gastrocnemius was observed just before sys-tole of the heart, which is nowadays known as the depolarization phase and manifests as a QRS complex on the ECG.4 Kölliker and Müller were astonished to observe in some preparations also a second twitch of the frogs leg at the beginning of diastole. Today, we ascribe this second twitch to the repolarization phase of the heart and identify it as the T wave on the ECG.4,5

Currently, the physiology and pathophysiology underlying the QRS complex are well understood. Dirk Durrer provided in 1970 the first complete description of the activation in the human heart and related it to the QRS morphology.6 Since then, many studies were performed to scrutinize abnormal QRS morphologies in conducted and ectopic beats and their relation with the underlying activation.7,8 Knowledge of the T wave and the corresponding repolarization pattern is however less crystalized. T waves are especially difficult to interpret in ventricular rhythms or rhythms with aberrant ventricular conduc-tion. When the subject has an AV-conducted rhythm, we can merely denote the T wave as having a normal or abnormal morphology and indicate whether it is prolonged or shortened. Nevertheless, T waves hold valuable information on arrhythmogenesis, as T wave abnormalities have indisputably been associated with life-threatening arrhyth-mias.1,9,10

Although major efforts have been made to reveal the exact relation between the typical waveform of the T wave to the underlying repolarization pattern, a thorough description of the T wave is still lacking. A reason for this is that a detailed observation of the T wave – as was done with the QRS complex – is complicated. The QRS complex results from the electrical inhomogeneity during the depolarization phase. Depolariza-tion is a fast process that causes large transmembrane currents and potential differences (about 25 millivolt) over short distances (about 1 millimeter) between excitable and non-excitable cells.4 The T wave results from the electrical inhomogeneity during the

12 Chapter 1

repolarization phase. In contrast to depolarization, repolarization is a relatively slow process and leads to smaller potential differences over larger distances.4 In addition, the contribution of cellular coupling in reducing the potential differences is larger during repolarization than during depolarization.11 The QRS complex is caused by conduction, in which cellular coupling is essential. The T wave is, however, mainly caused by the intrinsic property of individual cardiomyocytes (action potential duration), which is modulated by the amount of cellular coupling, causing more or less ‘synchronization’ of repolarization.12 Consequently, repolarization results in a small and widespread dipole, which is more difficult to measure and to interpret. In recent years technology has im-proved and enabled recordings of multiple channels. This has facilitated more accurate and simultaneous analyses of T waves and repolarization.

Repolarization can be influenced by multiple factors, like temperature,13 ischemia,14 potassium level,14 stretch,15 sympathetic activation12 and variants in genes encoding cardiac ion channels.16,17 These factors may accordingly modulate the vulnerability for arrhythmias. Therefore, the T wave holds valuable information for diagnostics and risk stratification. Yet, knowledge about the genesis and modulation of the T wave is incom-plete. This thesis focuses on elucidating the physiological as well as pathophysiological aspects of the T wave and explores the explicit effects that diverse factors (pressure, current, autonomic nervous system) have on the T wave morphology. The thesis also touches on the translation of results from animal studies to the human ECG.

In order to understand the pathophysiology it is essential to first obtain a thorough understanding of the normal physiology. Therefore, this thesis starts with a detailed description of the T wave of a normal pig heart in relation to its underlying ventricular repolarization (Chapter 2). This chapter considers the differences in repolarization along the large anatomical axes within the heart (left-right, anterior-posterior, apico-basal and transmural) and gives an explanation for the origin of the T wave peak.

After obtaining insights in the genesis of the normal T wave, this thesis focuses on its modulation by ventricular pressure that induces stretch on the ventricular wall. For decades we are aware of the phenomenon that stretch – by mechano-electric coupling – impacts on the ventricular action potential and arrhythmogenesis.18,19 The time of application of stretch relative to the phase of an action potential (AP) has differential effects on AP duration, and thus on repolarization.15 Chapter 3 will deal with the ef-fect of the normal left ventricular (LV) pressure pulse on the repolarization in early- and late-activated myocardium. It discusses the relevance of physiological LV pressure in the repolarization process within the heart. As stretch is in many diseases also a contributor to arrhythmogenesis,20 it is relevant to know how the stretch-induced changes in repo-larization may be translated to the ECG. Therefore, the thesis continues in Chapter 4 with the exploration of the effects of LV pressure on the T wave morphology.


1In the clinical setting several ECG abnormalities are supposed to be related to ab-

normal repolarization. In the following chapters of this thesis the analyses of some of these ECG abnormalities will be described. First, in Chapter 5, inferolateral J-waves on the ECG will be considered. These inferolateral J-waves are also known as the early re-polarization pattern and their presence have been associated with an increased risk for life-threatening arrhythmias.21 There is, however, an ongoing debate as to its underlying mechanism. It was proposed that J-waves result from an prominent early repolarization phase in the epicardial cardiomyocytes compared to the endocardial cardiomyocytes.22 An alternative mechanism is based on regional conduction slowing. Chapter 5 presents the results of a study that tested whether repolarization abnormalities or depolarization abnormalities may induce inferolateral J-waves on the ECG.

The long QT syndrome (LQTS) is characterized by a prolonged QT interval and as-sociated with sudden cardiac death.23 Moreover, patients with LQTS often present with typical T wave morphologies.24 The congenital LQT syndromes are classified according to their associated genetic basis,25 resulting nowadays in 15 different types.26 Type 2 of the LQTS (LQT2) encompasses one of the largest groups of LQTS patients and typically manifests low amplitude bifid T waves.27,28 From the genetic perspective we learned that in LQT2 a loss-of-function of the fast activating component of the delayed rectifier chan-nels is involved.16 The mechanism by which the changes in repolarizing currents give rise to the bifid T waves in LQT2 patients is not known. In Chapter 6 results are shown of a dog model of dofetilide-induced LQT2, in which the relation between the repolarization changes and the bifid T waves is investigated.

In patients with LQTS, T wave abnormalities are not always present on the ECG and in absence of a marked QT prolongation on the resting ECG, the diagnosis of LQTS is less straightforward. It has been shown that LQTS patients have an impaired adaptation of the QT interval – i.e. repolarization duration – in response to a change in heart rate.29,30 Therefore, in the Academic Medical Center in Amsterdam, ECG tests are currently performed to obtain information about the dynamics in QT adaptation. Chapter 7 de-scribes the results on the evaluation of the beat-to-beat QT and TQ (i.e. diastole on ECG) responses to sudden standing in LQTS patients compared to healthy controls. These results indicate differences between the groups of individuals. In this chapter a new metric is proposed, which may possibly distinguish LQTS patients from healthy persons.

The last chapter of this thesis provides a comprehensive discussion dealing with all subjects mentioned above, and future perspectives will be presented.

14 Chapter 1

REFERENCES

1. Zipes DP, Camm a J, Borggrefe M, et al. ACC/AHA/ESC 2006 Guidelines for Management of Patients With Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death: a report of the American College of Cardiology/American Heart Association Task Force and the European Society of Cardiology Com. Circulation. 2006, pp. e385–e484.

2. AlGhatrif M, Lindsay J. A brief review: history to understand fundamentals of electrocardiography. J. Community Hosp. Intern. Med. Perspect. 2012,.

3. Einthoven W. Nieuwe methoden voor clinisch onderzoek [New methods for clinical investiga-tion]. Ned Tijdschr Geneeskd 1893; 37: 263–286.

4. Noble D, Cohen I. The interpretation of the T wave of the electrocardiogram. Cardiovasc Res 1978; 12: 13–27.

5. Burdon-Sanderson J, Page FJM. On the Electrical Phenomena of the Excitatory Process in the Heart of the Frog and of the Tortoise, as investigated Photographically. J Physiol 1884; 4: 327–338.

6. 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.

7. SippensGroenewegen a, Spekhorst H, van Hemel NM, Kingma JH, Hauer RN, Janse MJ, Dunning a J. Body surface mapping of ectopic left and right ventricular activation. QRS spectrum in patients without structural heart disease. Circulation 1990; 82: 879–896.

8. Van Dessel PF, van Hemel NM, de Bakker JM, Linnenbank T a, Potse M, Jessurun ER, Sippens-Groenewegen a, Wever EF. Relation between body surface mapping and endocardial spread of ventricular activation in postinfarction heart. J Cardiovasc Electrophysiol 2001; 12: 1232–1241.

9. Panikkath R, Reinier K, Uy-Evanado A, Teodorescu C, Hattenhauer J, Mariani R, Gunson K, Jui J, Chugh SS. Prolonged Tpeak-to-tend interval on the resting ECG is associated with increased risk of sudden cardiac death. Circ Arrhythm Electrophysiol 2011; 4: 441–447.

10. Rosenbaum DS, Jackson LE, Smith JM, Garan H, Ruskin JN, Cohen RJ. Electrical alternans and vulnerability to ventricular arrhythmias. N Engl J Med 1994; 330: 235–241.

11. Franz MR, Bargheer K, Rafflenbeul W, Haverich A, Lichtlen PR. Monophasic action potential map-ping in human subjects with normal electrocardiograms: direct evidence for the genesis of the T wave. Circulation 1987; 75: 379–386.

12. Conrath CE, Opthof T. Ventricular repolarization: an overview of (patho)physiology, sympathetic effects and genetic aspects. Prog Biophys Mol Biol 2006; 92: 269–307.

13. Coronel R, de Bakker JMT, Wilms-Schopman FJG, Opthof T, Linnenbank AC, Belterman CN, Janse MJ. Monophasic action potentials and activation recovery intervals as measures of ventricular action potential duration: experimental evidence to resolve some controversies. Heart Rhythm 2006; 3: 1043–1050.

14. Moréna H, Janse MJ, Fiolet JW, Krieger WJ, Crijns H, Durrer D. Comparison of the effects of regional ischemia, hypoxia, hyperkalemia, and acidosis on intracellular and extracellular potentials and metabolism in the isolated porcine heart. Circ Res 1980; 46: 634–646.

15. Zabel M, Koller BS, Sachs F, Franz MR. Stretch-induced voltage changes in the isolated beating heart: importance of the timing of stretch and implications for stretch-activated ion channels. Cardiovasc Res 1996; 32: 120–130.


1 16. Curran ME, Splawski I, Timothy KW, Vincent GM, Green ED, Keating MT. A molecular basis for

cardiac arrhythmia: HERG mutations cause long QT syndrome. Cell 1995; 80: 795–803.

17. Wang Q, Curran ME, Splawski I, et al. Positional cloning of a novel potassium channel gene: KVLQT1 mutations cause cardiac arrhythmias. Nat Genet 1996; 12: 17–23.

18. Lab MJ. Contraction-excitation feedback in myocardium. Physiological basis and clinical rel-evance. Circ Res 1982; 50: 757–766.

19. Franz MR. Mechano-electrical feedback in ventricular myocardium. Cardiovasc Res 1996; 32: 15–24.

20. Taggart P, Sutton PM. Cardiac mechano-electric feedback in man: clinical relevance. Prog Biophys Mol Biol 1999; 71: 139–154.

21. Wu SH, Lin XX, Cheng YJ, Qiang CC, Zhang J. Early repolarization pattern and risk for arrhythmia death: a meta-analysis. J Am Coll Cardiol 2013; 61: 645–650.

22. Yan GX, Antzelevitch C. Cellular Basis for the Electrocardiographic J Wave. Circulation 1996; 93: 372–379.

23. Priori SG, Schwartz PJ, Napolitano C, Bloise R, Ronchetti E, Grillo M, Vicentini A, Spazzolini C, Nastoli J, Bottelli G, Folli R, Cappelletti D. Risk stratification in the long-QT syndrome. N Engl J Med 2003; 348: 1866–1874.

24. Moss AJ, Zareba W, Benhorin J, Locati EH, Hall WJ, Robinson JL, Schwartz PJ, Towbin JA, Vincent GM, Lehmann MH. ECG T-wave patterns in genetically distinct forms of the hereditary long QT syndrome. Circulation 1995; 92: 2929–2934.

25. Goldenberg I, Moss AJ. Long QT syndrome. J Am Coll Cardiol 2008; 51: 2291–2300.

26. Mizusawa Y, Horie M, Wilde AA. Genetic and Clinical Advances in Congenital Long QT Syndrome. Circ J 2014; 78: 2827–2833.

27. Zhang L, Timothy KW, Vincent GM, et al. Spectrum of ST-T-Wave Patterns and Repolarization Parameters in Congenital Long-QT Syndrome: ECG Findings Identify Genotypes. Circulation 2000; 102: 2849–2855.

28. Kanters JK, Fanoe S, Larsen LA, Bloch Thomsen PE, Toft E, Christiansen M. T wave morphology analysis distinguishes between KvLQT1 and HERG mutations in long QT syndrome. Heart Rhythm 2004; 1: 285–292.

29. Viskin S, Postema PG, Bhuiyan ZA, et al. The response of the QT interval to the brief tachycardia provoked by standing: a bedside test for diagnosing long QT syndrome. J Am Coll Cardiol 2010; 55: 1955–1961.

30. Adler A, van der Werf C, Postema PG, et al. The phenomenon of “QT stunning”: the abnormal QT prolongation provoked by standing persists even as the heart rate returns to normal in patients with long QT syndrome. Heart Rhythm 2012; 9: 901–908.

Chapter 2

Electrocardiographic T wave and its relation with ventricular repolarization

along major anatomical axes

Veronique M.F. Meijborg, Chantal E. Conrath, Tobias Opthof, Charly N.W. Belterman, Jacques M.T. de Bakker, Ruben Coronel

Circulation: Arrhythmia and Electrophysiology 2014;7(3):524-31

18 Chapter 2

AbSTRACT

background: The genesis of the electrocardiographic T wave is incompletely under-stood and subject to controversy. We have correlated the ventricular repolarization sequence with simultaneously recorded T waves.

Methods and Results: Nine pig hearts were Langendorff perfused (atrial pacing, cycle length 650 ms). Local activation and repolarization times were derived from unipolar electrograms sampling the ventricular myocardium. Dispersion of repolarization time was determined along 4 anatomic axes: left ventricle (LV)–right ventricle (RV), LV:apico-basal, LV:anterior-posterior, and LV:transmural. The heart was immersed in a fluid-filled bucket containing 61 electrodes to determine Tp (Tpeak in lead of maximum integral), TpTe (Tp to Tend) and TpTe_total (first Tpeak in any lead to last Tend in any lead). Repolarization was nonlinearly distributed in time. RT25 (time at which 25% of sites were repolarized, 288±26 ms) concurred with Tp. TpTe was 38±8 ms and TpTe_total was 75±9 ms. TpTe_total correlated with dispersion of repolarization time in the entire heart (73±18 ms), but not with dis-persions of repolarization times along individual axes (LV-RV, 66±17 ms; LV:apico-basal, 51±18 ms; LV:anterior-posterior, 51±27 ms; mean LV:transmural, 14±7 ms, all n=9).

Conclusions: We provide a correlation between local repolarization and T wave in a pseudo-ECG. Repolarization differences along all anatomic axes contribute to the T wave. TpTe_total represents total dispersion of repolarization. At Tp, ≈25% of ventricular sites have been repolarized.

The T wave and Repolarization 19

2

INTRODuCTION

The T wave on the ECG (T-ECG) represents repolarization of the ventricular myocardium. Its morphology and duration are commonly used to diagnose pathology and assess risk of life-threatening ventricular arrhythmias. However, the physiological background of the T wave is incompletely understood thereby hampering reliable interpretation of the T wave.

Interpretation of the T wave morphology on body surface ECGs (T-ECG) is different from that on local electrograms (T-local). In a far-field recording (e.g. a 12-lead ECG), the T wave represents repolarization in the entire heart.1 In a near-field recording (e.g. a local electrogram), the T wave can be used to accurately determine local repolarization mo-ments.2,3 Unlike our understanding of T-local, knowledge of the T-ECG is still incomplete.

Previous studies have focused mainly on a single dominant repolarization gradient to explain the T-ECG morphology.4-6 It has become generally accepted that concordance of the T-ECG with the QRS complex is explained by opposing sequences of transmural activation and repolarization.5 Whereas activation progresses from subendocardial to subepicardial myocardium, repolarization was assumed to progress in the opposite direction.7 Additionally, T-ECG-peak and T-ECG-end were explained by the end of repo-larization of, respectively, the subepicardial and M cell layers.6 There is, however, con-troversy regarding the presence of a functional M cell layer in human hearts. M cells have been observed only in human wedge preparations,8,9 but were absent in vivo.10,11 In dogs, total rather than transmural dispersion of ventricular repolarization is more determinative for the peak-to-end interval of the T-ECG (TpTe).12 Other studies have suggested that an apico-basal gradient of repolarization is responsible for the T-ECG morphology.1,13 These observations have resulted in a persisting debate concerning the dominant repolarization differences giving rise to the T-ECG.14-16

We aimed to clarify the genesis of the T-ECG by correlating the ventricular repolariza-tion sequence with simultaneously recorded T-ECGs in isolated perfused pig hearts. Pig hearts most likely lack an M cell layer, like human hearts.17

METHODS

See the Data Supplement for detailed methods.

Experimental setup

The experimental protocol was approved by the local ethical committee on animal ex-perimentation. Male pigs (n=9; body weight, 34±12 kg) were anesthetized and a 12-lead ECG was recorded in supine position (in vivo ECG). The heart was excised and perfused

20 Chapter 2

according to Langendorff with a 1:1 blood-Tyrode’s mixture. The apex was fixed to avoid swinging of the heart, but to allow rotation and movement in the vertical plane.

Electrophysiological study and data acquisition

After removal of the sinus node area, the atrium was paced at a cycle length of 650 ms. Transmural multielectrode (0.5 mm diameter) needles were inserted in a predefined pattern to obtain transmural local unipolar electrograms throughout the heart. Exact locations of terminals were reconstructed after termination of the experiment. The heart was surrounded by a perfusion-fluid-filled bucket containing 61 regularly distributed electrodes (1 electrode at the bottom of the bucket) to obtain a pseudo body surface ECG (BS-ECG). The reference signal was the assembled average of all 61 electrodes.

In the BS-ECG and in vivo ECG, QRS duration, QRS vector, QT interval, QT interval cor-rected for heart rate via Fridericia18 (QTc_F), T-ECG duration (T-duration), T-ECG vector (T-vector), and TpTe interval were determined (see Figure 1). QRSonset and QRSend were defined as respectively the first deflection and last J-point of the QRS complex in any ECG lead. Integrals of the QRS complex and T-ECG (defined as area under the QRS and T curve) were calculated between manually set markers. These were used to identify main (in vivo) and detailed (BS-ECG) QRS and T vectors in 3 planes: frontal, transversal and sagittal (in vivo: lead I, V6 and V2 equal 0 degrees, and rotation to lead aVF, V2, and aVF denotes a positive angle; in BS-ECG, similar definitions were used).

In the ECG lead of maximum (positive or negative) T-integral, the onset (To), peak (Tp) and end (Te) of the T-ECG were manually determined relative to QRSonset using the tan-gent method (Figure 1).19 QT interval and TpTe were defined as interval between QRSonset and Te, and between Tp and Te in a single lead. We additionally determined TpTe_total in the BS-ECG as interval between earliest Tpeak in any lead and last Tend in any lead.

At each local recording site, local activation times and local repolarization times (RTs) were automatically determined as the interval between QRSonset to time of the minimum derivative of QRS complex and maximum derivative of T-local, respectively (Figure 1). All signals and time markers were visually validated. Recordings with ST-elevation or a flat T-local were excluded from analysis of RTs.

RTmin and RTmax were start and end of repolarization, respectively. The volume distribu-tion of repolarization was described with time points at which, respectively, n=5%, 25%, 50% (median RT), 75% and 95% of the recording sites were repolarized (RTn). Dispersion of repolarization (dRT) was defined as range of RTs along a specific anatomic axis: left ven-tricle (LV)–right ventricle (RV), LV:apico-basal, LV:anterior-posterior, and LV:transmural. Maximum and mean LV:transmural dRT (dRT_maxLVtransmural, dRT_meanLVtransmu-ral) were, respectively, maximum and mean of all dRTs per LV needle. The dRT_total was the maximum range of RTs across the entire heart (Figure 1). Signal analysis was per-formed offline using a custom-made data analysis program written in Matlab 2006b.20


2

Statistics

Continuous variables were given as mean±SD if normally distributed and in median (25th-75th percentile) if not normally distributed. Evaluation of normality was based

200 ms

4 m

V

200 ms

0.4

mV

dRT_maxLVtransmural

dRT_total & dRT_LV_RV

TpTe

T-duration

Needle 1:LV epi

LV endo

Needle 2:RV epi

RV endo

QRSendQRSonset

ATRT

QRS duration

QT interval

TeTo Tp

TpTe _total

largest T integral

BS-ECG

Local EG

Figure 1: Top: Local electrograms (EG) from 2 needles with 1 subepicardial (epi) and 1 subendocardial (endo) recording. Red dots denote activation times (ATs) and repolarization times (RTs). Maximum diff er-ence in RT along 1 needle (dRT_maxLVtransmural) and between all electrodes (dRT_total, in this example, equal to dRT_LV-RV) are shown.bottom: 4 leads of BS-ECG. Red dashed lines indicate interval over which T-integrals were determined. To, Tp, and Te are onset, peak, and end of T-ECG, respectively, in the lead with maximum T-ECG integral. T-duration indicates To to Te interval; TpTe, Tp to Te interval; QRS duration, interval between QRSonset (fi rst defl ection of QRS complex) and QRSend (last J-point); QT interval, QRSonset to Te interval; and TpTe_total, interval between fi rst peak of T-ECG (second lead) to last end of T-ECG (third lead).

22 Chapter 2

on Q-Q plots and correspondence of mean and median. Diff erences in characteristics between the in vivo ECG and BS-ECG, between RTn and T-ECG parameters, and between dRT values and TpTe or TpTe_total were tested using a paired t test. We determined the amount of similarity (combination of distance and correlation) between RTn and T-ECG parameters and between dRT values and TpTe or TpTe_total by calculating intraclass correlation coeffi cients (ICCs) using a 2-way mixed model of absolute agreement.21 Dif-

I

II

aVF

aVL

aVR

III

V1

V2

V6

V5

V4

V3

0.5 sec

5 m

V

A

B

posterior posterioranterior LVRV

central

apex

0.5 sec

1 m

V

QRS

T

bottom

base

Figure 2: A: Electrode placement of the 12-lead ECG of a pig (left) and a typical example of a 12-lead in vivo ECG recording (right). In the in vivo ECG, the T-ECG is concordant in all extremity leads and discordant in precordial leads V1 to V4. b: Typical example of an ex vivo BS-ECG recording of the same heart as in Figure 2A. It is a schematic view of an unfolded bucket with the electrograms arranged in the confi guration of the electrodes. Bottom refers to the recording derived at the bottom of the bucket. The QRS vector in this BS-ECG is directed to the left and posterior (arrow QRS), and the T-vector is directed to the left and anterior (arrow T). Most leads show a discordant T-ECG with the QRS complex, similar to the in vivo precordial leads.


2

ferences in RT50 between different regions along an anatomic axis were tested with a Friedman analysis. Differences and ICCs with a P value of ≤0.05 were considered statisti-cally significant.

RESuLTS

In vivo standard 12-lead ECG and ex vivo bS-ECG

Figure 2 shows a typical example of a porcine in vivo 12-lead ECG in supine position (Figure 2A) and the BS-ECG of the same heart in Langendorff perfusion (Figure 2B). In general, in the in vivo ECG the T-ECG was positive in precordial and inferior leads and flat, biphasic, or slightly negative in leads I, aVR, and aVL.

Overall, the T-ECG in BS-ECG was discordant with the QRS complex in most leads and positive in anterior leads. The Table features the main characteristics of the ex vivo BS-ECG and in vivo 12-lead ECG. In comparison with the in vivo ECG of the same animal, the QRS vector in the BS-ECG is rotated in each plane (frontal: 47±74°; P=n.s.; transversal: 56±60°; P=0.024; sagittal: 49±39°; P=0.005). There was no significant change in T-vector between the in vivo ECG and BS-ECG (frontal: −3±90°, transversal: −74±105°, sagittal: 56±93°; all P=n.s.) but the change in T-vector showed a larger SD than the change in

Table: Characteristics of in vivo ECG and the ex vivo BS-ECG

n=9

In vivo ECG bS-ECG

P valueMean ± SD Mean ± SD

RR, ms 594 ± 129 650 ± 0 0.227

QRS duration, ms 55 ± 7 60 ± 8 0.061

QRS vector, degrees

frontal −108 ± 73 −61 ± 43 0.092

transversal −104 ± 7 −49 ± 60 0.024

sagittal −169 ± 11 −120 ± 45 0.005

QT interval, ms 313 ± 16 321 ± 30 0.375

QTc_F interval, ms 375 ± 23 371 ± 35 0.750

T-duration, ms 118 ± 36 97 ± 12 0.137

TpTe, ms 41 ± 7 38 ± 8 0.642

T-vector, degrees

frontal 97 ± 33 94 ± 82 0.919

transversal 78 ± 18 4 ± 103 0.067

sagittal 26 ± 11 82 ± 94 0.108

P values for a paired t test between the in vivo ECG and BS-ECG (n=9). BS-ECG indicates body surface ECG; n, number of experiments; QTc_F, corrected QT interval via Fridericia; RR, interval between two successive QRS complexes; TpTe, Tp to Te interval in the lead with maximum T-ECG integral.

24 Chapter 2

QRS vector. QRS duration, QT interval, QTc_F, T-duration and TpTe in the BS-ECG were not statistically different from the in vivo ECG (Data Supplement).

Activation and Repolarization

Figure 3A and 3B shows typical examples of, respectively, activation and repolarization maps of 1 heart. In all hearts, activation times were obtained from 87±27 recording sites throughout the heart. Overall, activation occurred earliest at the left septum and lat-est at the RV basal free wall (as in Figure 3A). Transmural activation along needles was almost simultaneous. Activation was complete within 49±9 ms (see Data Supplement).

In all hearts, repolarization times were obtained from 61±21 recording sites per heart. Overall, repolarization started at the anterior or posterior RV. Repolarization in the LV progressed from basal to central (between apex and base) regions. Repolarization of

324329

324

321329

320

329

329336329

329

290275260 305 350 ms335320

317329324327

289 329

269

309 320

317327337

305308

329324306

281

324305

344340317

340337329

317 329

317 305

329

329

307289

327337341

317

344329

347343

337

344 341336329

343

329340

337 344 335329

337

317327

307289 329

327

289

327321324

329317327329329 324

321329329329 325

289 307317 329

329

321 329

LVRV

Posterior

Anterior

*

B

1118

15

1521

17

25

2428 20

27252231

243440

22

2324

25

3029

2925

31 3438 39

30

2733

32

25

32 32 37

29

2429

2626 24

22

30

121623

1211221030 1028

3717

32

11 10 7 14

3230 26

22 25

37

15 25

31

20100 30 60 ms5040

3851743

39 35

5247

4627

28 28 28 29

282626

2927

30

352221

27

22242126

27 27 33

36 34 35

22 25

28 29

37

39

263032

28

LVRV

Base

Central

Apex

Posterior

Anterior

*

A

Figure 3: Typical activation (A) and repolarization (b) map in transversal slices at basal (top), central (mid-dle) and apical (bottom) level. Red indicates early areas and blue indicates late areas. A: Activation starts at the septal basal and central region and progresses to the posterior central region and the anterior apex. The free walls of the ventricles activate latest (left ventricle [LV] preceding right ventricle [RV]). Activation of subendocardial (endo) and opposite subepicardial (epi) regions occurs almost simultaneously. The maxi-mum difference in this animal was only 18 ms (asterisk at the central anterior LV wall, with epi earlier than endo). b: Repolarization starts at the anterior base of RV and septum and at the posterior basal region. Lat-est repolarization occurs in the LV endo basal and central regions. Transmural repolarization was generally simultaneous although a region with a large transmural difference of 48 ms (asterisk at the basal posterior LV wall, endo repolarizing later than epi) was observed.


2

apical regions varied remarkably between hearts. Final repolarization was at the central free walls (RV before LV; see also Figure 3B). Transmural repolarization was simultaneous, although there were isolated sites with larger transmural differences (asterisk in Figure 3B, crowding of isochrones).

On average, RTmin was at 251±22 ms and RTmax at 324±33 ms (n=9 hearts). RT25 was at 288±26 ms (at midrepolarization), RT50 at 303±28 ms and RT75 at 309±30 ms. From the onset of repolarization, it took 52±13 ms to fully repolarize the first 50% of the recording sites. From that moment it took only 21±7 ms to complete repolarization.

T-ECG and the overall repolarization sequence

To occurred always earlier than RTmin (−27±15 ms; P<0.05; ICC=0.56; P<0.005). Te was not significantly different from RTmax (−3±10 ms; P=n.s.) and had a large ICC (0.95; P<0.001). We compared time of Tp on the BS-ECG (Tp= 283±28 ms) with the distribution of repolar-ization (RTmin, RT25, RT50, RT75, and RTmax). Figure 4 shows that RT25 had the smallest differ-ence with the time of Tp (6±12 ms; P=n.s.) and largest ICC (0.89; P<0.001). More precisely, at Tp the percentage of repolarized sites was 19.5±12.6%.

RTminRT50

RTmaxRT75

RT25

50

-50

-25

0

25

Di�

eren

ce b

etw

een

RT a

nd T

p (ms) ICC0.500.690.770.890.48

Figure 4: Tp versus repolarization time (RT) values. Bars represent the difference between the RT value (RTmin, RT25, RT50, RT75, RTmax) and Tp in ms. Error bars represent SDs. Intraclass correlation coefficients (ICCs) are given above the bars and were all significant (P<0.05). Note the small difference between RT25 and Tp and the large ICC.

T-ECG and the distribution of RT along 4 anatomic axes

RT distribution along the 4 anatomic axes was studied in relation to T-ECG morphology. Figure 5 shows an example of the time relation per predefined axis between the T-ECG in a BS-ECG lead (with maximum T-ECG integral, opposite the RV), and RT5, RT25, RT50, RT75, and RT95. The figure (top panel) demonstrates that half of RV (RT50) had already repolarized

26 Chapter 2

Freq

uenc

y

LV-RV-septum

LVapico-basal

LV antero-posterior

LV transmural

Entire Heart

LV (n=33)RV (n=11)Septum (n=4)

Base (n=12)Central (n=16)Apex (n=5)

Anterior (n=7)Lateral (n=15)Posterior (n=11)

Epi (n=7)Mid-Epi (n=10)Mid-Endo (n=9)Endo (n=7)

250 375

100 ms

0.2

mV

All (n=48)

RT5

RT25

RT50

RT75

RT95

0.022

293 (277-320)Base

305 (284-335)Central

299 (283-319)Apex

15 ± 7

19 ± 7

5 ± 3

0.092

304 (279-320)Anterior

305 (285-334)Lateral

296 (277-331)Posterior

12 ± 5

14 ± 3

13 ± 9

293 (277-326)Septum

288 (268-320)RV 0.097

302 (283-328)LV

12 ± 6

10 ± 3

39 ± 14

pnRT50

<0.001

293 (278-313)Epi

298 (284-321)Mid-epi

315 (284-338)Endo305 (285-335)Mid-endo

8 ± 3

12 ± 4

9 ± 4

10 ± 4

EXAMPLE OVERALL (n=9)

0

5

20

15

10

0 %

100 %

Figure 5: Left: Example of the T wave (of the BS-ECG lead with maximum T wave integral) and repolariza-tion time (RT) distribution (RT5 to RT95 with whiskers for RT25, RT50, and RT75) within a region per anatomic axis: LV-RV-septum (upper), LV:apico-basal (second), LV:anterior-posterior (third), LV:transmural (fourth), entire heart (lower). At the background of the lower section a histogram of the number of repolarized recording sites is depicted (grey bars; grey line indicates cumulative percentage). n is the number of sites per region.Right: Comparison of RT50 (median [25th-75th percentile]) values between the regions along an anatomic axis in 9 hearts. Differences in RT50 between regions along each axis were tested with a Friedman analysis (P value per axis is shown). n is number of sites per region (mean±SD).


2

before RT5 of LV, and that at RT95 in the RV, 50% of LV (RT50) still had to repolarize. We re-stricted analysis of RT distribution along other axes only to the LV, because it has the larg-est contributing mass and harbored more electrodes. In the figure, the dispersion of RT50 values within the LV:antero-posterior, LV:apico-basal or LV:transmural axes was not large enough to explain the T-duration. In this example, LV repolarization starts at the basal posterior midendocardial region and ends in the basal anterior subendocardial region, illustrating that repolarization along all axes plays a role in the time relation between the T-ECG and the repolarization process. The bottom panel of the figure shows the RT distribution of the entire heart along the T-ECG. The grey histogram of repolarized record-ing sites re-emphasizes that the majority of sites repolarized after the moment of Tp.

Figure 5 summarizes RT50 values per anatomic axis of all animals. RT50 values were significantly different between regions within the LV:apico-basal axis (central region was 17±9 ms longer than apical or basal region; n=7), and LV:transmural axis (the sub-endocardial or midendocardial region was 17±13 ms longer than the subepicardial or midepicardial region; n=9).

TpTe and dispersion in repolarization

TpTe has been associated with heterogeneity in RT, either in wedge preparations6 or in the whole heart.12,22 We, therefore, studied dispersion of repolarization moments (dRT) along each of the 4 axes and over the entire heart in relation to TpTe on the BS-ECG (Fig-ure 6). Locations of maximum and minimum RT across each single axis differed between animals, so the direction of the vector connecting the pair of values constituting the dRT_total was not similar among animals. TpTe_total on the BS-ECG was least different from dRT_total (Figure 6: mean difference of 1±13 ms; paired t test P=0.758), with an ICC of 0.626 (P=0.03). TpTe was least different from dRT_maxLVtransmural (Figure 6: mean difference of 4±20 ms; paired t test P=0.552). However, the ICC of 0.306 was low and statistically not significant. Other dRT values that were not statistically significant from TpTe_total or TpTe (Figure 6) had larger differences and lower ICC values with TpTe_total or TpTe.

DISCuSSION

The main findings of the present study are that (1) differences in repolarization moments along all axes contribute to the T-ECG morphology; (2) TpTe_total measured on the BS-ECG reflects total dispersion of repolarization within the heart; (3) at the moment of T-ECG peak, ≈25% of the recording sites are repolarized; and (4) onset of T-ECG precedes onset of repolarization. We provide the first complete correlation between local repolar-ization and the simultaneously recorded T-ECG on the surface ECG.

28 Chapter 2

In vivo ECG versus ex vivo ECG

The in vivo and pseudo-ECG closely resemble each other, although transversal and sagittal QRS vectors differ. It likely results from a different position of the heart in the thorax versus the bucket (Table and Data Supplement). Therefore, we assume that ac-tivation in the hearts was unchanged between in vivo and ex vivo. This is supported by the unchanged QRS duration. In addition, the in situ ventricular activation sequence of a dog heart recorded by Durrer et al23 did not change after isolation.

Repolarization and T-ECG

We showed significant differences in RT50 along LV:apico-basal and LV:transmural axes. However, differences (≈17 ms) were too small to explain an average T-ECG duration of 97 ms. It suggests that total dispersion of RT is reflected in the entire T-ECG. Indeed, we demonstrated that the interval between first peak to last end of all T-ECGs did cor-respond with total dispersion of RT, corroborating results from Xia et al.22 Although maximum transmural dRT was least different from TpTe, its ICC was low and not statisti-cally significant. The lack of correlation may be attributed to the fact that the maximum

80

0

20

40

60

100

dRT_total

dRT_LVtra

nsmural

dRT_LVanterio

r-poste

rior

dRT_LVapico

-basa

l

dRT_LV-R

V

ms

TpTe_total

TpTe

dRT_maxLVtra

nsmural

dRT_meanLVtra

nsmural

TpTe _total

TpTe

†† * † ** * †

75 ± 9 ms

38 ± 8 ms

73 ±

18

66 ±

17

51 ±

18

51 ±

27

65 ±

19

34 ±

22

14 ±

7

Figure 6: Dispersion of repolarization time (dRT) values along various anatomic axes (mean±SD; n=9) with corresponding TpTe and TpTe_total (solid line=mean; dashed line=SD; n=9). To exclude influences of other axes contributing to the left ventricle (LV):transmural axis, we also obtained dRT_maxLVtransmural and dRT_meanLVtransmural. Note that mean TpTe_total is least different from mean dRT_total (intraclass corre-lation coefficient [ICC]; P<0.05) and mean TpTe is least different from mean dRT_maxLVtransmural (ICC=ns). Paired t test: * P<0.05 dRT compared with TpTe, † P<0.05 dRT compared with TpTe_total.


2

difference represents a single outlier value, which may not affect T wave morphology. The mean transmural dRT more reliably represents transmural dispersion but did not correlate with TpTe, as demonstrated before.12

To precedes the first measured RT. This is unlikely the result of a sampling error, be-cause it occurred in all hearts. We suggest that regional differences in action potential (AP) morphology (phase 1 and 2) rather than differences in RTs contribute to genesis of the first part of the T-ECG. Te coincides with the last measured RT, consistent with the fact that RT in a unipolar electrogram relates to the steepest part of AP downstroke,2,3 which approximates APD90 (AP duration at 90% of repolarization).24

Fuller et al25 found a near equality of time of T wave peak and mean RT, which contrasts with our results that 75% of recording sites repolarizes after Tp. Methodological differ-ences may explain this discrepancy. We used a single surface ECG lead instead of a root mean square of epicardial recordings. Not only is the former clinically more relevant, the Data Supplement indicates that the root mean square of BS-ECGs shows similar results and does not explain the discrepancy. Furthermore, Fuller used only epicardial electro-grams, which may shift or transform RT distribution and underestimate RT dispersion. Differences in data presentation (medians and quartiles versus means) and species (pigs versus dogs) may cause additional discrepancy. We underscore that the finding that Tp reflects RT25 cannot indiscriminately be applied to a randomly chosen lead because each has a different view on the dominant vector. Nonetheless, we suppose it is applicable to the lead with maximum T wave area.

Our data show that the time during which the first 50% of sites repolarize is about twice as long as the time required for repolarization of the last 50% of sites. A possible physiological explanation for the nonlinear repolarization process is that repolarization accelerates exponentially as a result of radial propagation of the repolarization wave. This matches the repolarization course illustrated in Figure 5 (bottom panel). The highly skewed apparent velocity of the repolarization process has thus far not been described, although its reflection in an asymmetrical T wave is common knowledge (To to Tp interval of 59±7 versus TpTe interval of 38±8 ms).

We speculate that the first part of a normal T-ECG primarily depends on regional dif-ferences in repolarization time course (i.e. changes in AP morphology) preceding the moment of full repolarization (i.e. steepest part of AP downstroke). When full repolariza-tion has started, continuation may accelerate depending on the degree of electrotonic interaction, which may determine the last part of the T wave.

Pig versus human: In vivo ECG and repolarization sequence

The QRS configuration of the in vivo porcine ECG in our study is in agreement with another study in pigs.26 The in vivo porcine T-ECG in our study resembles a normal T-ECG in human27 as the polarity is positive in most leads and T-vectors have similar angles (if

30 Chapter 2

corrected for differences in position of the heart in the thorax).28 QRS duration in pig, however, is shorter than in man.23 This is because of a more extensive transmural Purkinje network29 and the smaller size of porcine hearts compared to human. Nevertheless, the general activation sequence observed in our study resembles the activation sequence described in humans23 and pigs.30,31

The porcine repolarization sequence in our study resembles that in earlier reports,22,31 although repolarization of the septal region occurred later in our study. Methodological differences (endocardial monophasic action potentials recorded sequentially at 50-70 sites instead of local transmural electrograms recorded simultaneously at 61±21 sites [our study]) may have caused the minor differences in repolarization sequences, although correlation between RT and monophasic APD90 is high.24

CONCLuSIONS

We provide the first complete correlation between local repolarization and the cor-responding electrocardiographic T wave. A main conclusion of this study is that total dispersion of repolarization along all anatomic axes is determinative for the T wave on the pseudo-ECG of the isolated perfused pig heart. TpTe_total (first peak to last end of all T-ECGs) across ECG leads represents total dispersion of repolarization. Moreover, at the T wave peak, only ≈25% of ventricular sites are repolarized. Therefore, the start of the T wave is likely the result of regional differences in early repolarization (phase 1 and 2 of the AP), and the nonlinear repolarization process may be reflected in the final part of the T wave.

ACkNOWLEDGEMENTS

We are grateful for technical support of Carel Kools and Dr. André Linnenbank.


2

REFERENCES

1. Noble D, Cohen I. The interpretation of the T wave of the electrocardiogram. Cardiovasc Res 1978; 12: 13–27.


3. 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.

4. Higuchi T, Nakaya Y. T wave polarity related to the repolarization process of epicardial and endo-cardial ventricular surfaces. Am Heart J 1984; 108: 290–295.

5. Franz MR, Bargheer K, Rafflenbeul W, Haverich A, Lichtlen PR. Monophasic action potential map-ping in human subjects with normal electrocardiograms: direct evidence for the genesis of the T wave. Circulation 1987; 75: 379–386.

6. Yan GX, Antzelevitch C. Cellular basis for the normal T wave and the electrocardiographic mani-festations of the long-QT syndrome. Circulation 1998; 98: 1928–1936.

7. Van Dam RT, Durrer D. The T Wave and Ventricular Repolarization. Am J Cardiol 1964; 14: 294–300.

8. Glukhov A V, Fedorov V V, Lou Q, Ravikumar VK, Kalish PW, Schuessler RB, Moazami N, Efimov IR. Transmural dispersion of repolarization in failing and nonfailing human ventricle. Circ Res 2010; 106: 981–991.

9. Drouin E, Charpentier F, Gauthier C, Laurent K, Le Marec H. Electrophysiologic characteristics of cells spanning the left ventricular wall of human heart: evidence for presence of M cells. J Am Coll Cardiol 1995; 26: 185–192.

10. Conrath CE, Wilders R, Coronel R, de Bakker JMT, Taggart P, de Groot JR, Opthof T. Intercellular coupling through gap junctions masks M cells in the human heart. Cardiovasc Res 2004; 62: 407–414.

11. Taggart P, Sutton PM, Opthof T, Coronel R, Trimlett R, Pugsley W, Kallis P. Transmural repolarisation in the left ventricle in humans during normoxia and ischaemia. Cardiovasc Res 2001; 50: 454–462.

12. Opthof T, Coronel R, Wilms-Schopman FJG, Plotnikov AN, Shlapakova IN, Danilo P, Rosen MR, Janse MJ. Dispersion of repolarization in canine ventricle and the electrocardiographic T wave: Tp-e interval does not reflect transmural dispersion. Hear Rhythm 2007; 4: 341–348.

13. Xue J, Chen Y, Han X, Gao W. Electrocardiographic morphology changes with different type of repolarization dispersions. J Electrocardiol 2010; 43: 553–559.

14. Patel C, Burke JF, Patel H, Gupta P, Kowey PR, Antzelevitch C, Yan G-X. Is there a significant trans-mural gradient in repolarization time in the intact heart? Cellular basis of the T wave: a century of controversy. Circ Arrhythm Electrophysiol 2009; 2: 80–88.

15. Opthof T, Coronel R, Janse MJ. Is there a significant transmural gradient in repolarization time in the intact heart?: Repolarization Gradients in the Intact Heart. Circ Arrhythm Electrophysiol 2009; 2: 89–96.

32 Chapter 2

16. Janse MJ, Coronel R, Opthof T, Sosunov E a, Anyukhovsky EP, Rosen MR. Repolarization gradients in the intact heart: transmural or apico-basal? Prog Biophys Mol Biol 2012; 109: 6–15.

17. Rodríguez-Sinovas a, Cinca J, Tapias a, Armadans L, Tresànchez M, Soler-Soler J. Lack of evidence of M-cells in porcine left ventricular myocardium. Cardiovasc Res 1997; 33: 307–313.

18. Funck-Brentano C, Jaillon P. Rate-corrected QT interval: techniques and limitations. Am J Cardiol 1993; 72: 17B – 22B.

19. Panicker GK, Karnad DR, Natekar M, Kothari S, Narula D, Lokhandwala Y. Intra- and interreader variability in QT interval measurement by tangent and threshold methods in a central electrocar-diogram laboratory. J Electrocardiol 2009; 42: 348–352.

20. Potse M, Linnenbank AC, Grimbergen CA. Software design for analysis of multichannel intracar-dial and body surface electrocardiograms. Comput Methods Programs Biomed 2002; 69: 225–236.

21. McGraw KO, Wong SP. Forming inferences about some intraclass correlation coefficients. Psychol Methods 1996; 1: 30–46.

22. Xia Y, Liang Y, Kongstad O. T peak -T end Interval as an Index of Global Dispersion of Ventricular Repolarization: Evaluations Using Monophasic Action Potential Mapping of the Epi- and Endocar-dium in Swine. 2005: 79–87.


24. Potse M, Vinet A, Opthof T, Coronel R. Validation of a simple model for the morphology of the T wave in unipolar electrograms. Am J Physiol Heart Circ Physiol 2009; 297: H792–H801.

25. Fuller MS, Sándor G, Punske B, Taccardi B, MacLeod RS, Ershler PR, Green LS, Lux RL. Estimates of repolarization and its dispersion from electrocardiographic measurements: direct epicardial assessment in the canine heart. J Electrocardiol 2000; 33: 171–180.

26. Hamlin L, The RL. The QRS electrocardiogram, and ventricular excitation of swine1.

27. Bristow JD. A study of the normal Frank vectorcardiogram. Am Heart J 1961; 61: 242–249.

28. Crick SJ, Sheppard MN, Ho SY, Gebstein L, Anderson RH. Anatomy of the pig heart: comparisons with normal human cardiac structure. J Anat 1998; 193(Pt 1): 105–119.

29. Bharati S. The conduction system of the swine heart. CHEST J 1991; 100: 207.

30. Gepstein L, Hayam G, Ben-Haim S a. Activation-repolarization coupling in the normal swine endocardium. Circulation 1997; 96: 4036–4043.

31. Yuan S, Kongstad O, Hertervig E, Holm M, Grins E, Olsson B. Global repolarization sequence of the ventricular endocardium: monophasic action potential mapping in swine and humans. Pacing Clin Electrophysiol 2001; 24: 1479–1488.


2

DATA SuPPLEMENT

Supplemental Methods

Experimental setupPigs were premedicated with ketamine (10-15 mg/kg, Nimatek, Eurovet Animal Health BV) and midazolam (1-2 mg/kg, Actavis, IJsland) intramuscularly and anesthetized with 15 mg/kg pentobarbital (Nembutal, Ceva Santé Animale) intravenously. The animals were intubated and ventilated with room air and oxygen plus isoflurane (1:1 + 1.5%). After a midsternal thoracotomy, heparin 5000 international units (Leo Pharma) was injected intravenously. Tyrode’s solution was infused, after which 4 to 5 L diluted blood was collected.

Electrophysiological study and data acquisitionTransmural multi-electrode (0.5 mm diameter) needles were inserted in a predefined pattern. In the left ventricle 16 to 26 needles (harboring 4 electrodes per needle) were inserted, and in the right ventricle 12 to 19 needles (harboring 2-4 electrodes per needle). Twenty-four to 35 of the terminals were located in the septum. A 256-channel amplifier (BioSemi, 24 bit dynamic range, 122.07 nV LSB, total noise 0.5 µV) was used. Signals were recorded with a sampling frequency of 2048 Hz (bandwidth (−3dB) DC −400 Hz). The active common mode electrode was positioned in the aortic root. Stimulation pulse amplitude was 1.5 times the stimulation threshold and pulse width was 1 ms.

Vector determination in ECGIn the in vivo ECG we were able to determine only the main vector due to a low number of leads. Integrals were automatically calculated by an algorithm in Matlab between manually set markers. The main vector (in vivo) was determined by calculating the angles using the integrals of all leads. In the BS-ECG exact vectors were determined. First, the time of the peak of the QRS-complex or T-ECG was determined in the BS-ECG lead with the absolute maximum integral (positive or negative) of the QRS-complex or T-ECG, respectively. We suppose that at this time the vector has the maximum amplitude and therefore we determined the vector angles at this moment. At this time the angle was calculated between the lead with the maximum and minimum voltage potential (see Figure S1). In the BS-ECG the three vector planes (frontal, transversal and sagittal) were assigned based on the position of the heart in the bucket and with similar direc-tions as in the in vivo ECG).

Dispersion of repolarizationMeasured RT in a unipolar electrogram relates to the steepest part of action potential (AP) downstroke,1,2 which is close to APD90 (AP duration at 90% of repolarization).3 Disper-

34 Chapter 2

sion of repolarization (dRT) was defi ned as the range of RTs along a specifi c anatomical axis: LV–RV, LV:apico-basal, LV:anterior-posterior and LV:transmural. Along each axis the recording sites were divided according to regions (all sites in dRT LV-RV and only LV sites in other axes), after which the maximum range between these regions was determined. E.g. along the LV:apico-basal axis all LV recording sites were subdivided into the apical, central or basal region after which the maximum diff erence in RT values between these regions was determined. These dispersion measures were not completely independent of dispersion along other axes. Especially dRT LV:transmural may be infl uenced by dis-persion along other axes, because each transmural region includes basal as well as api-cal recording sites. Therefore, we also obtained the maximum and mean LV:transmural dRT (dRT_maxLVtransmural, dRT_meanLVtransmural). Per LV needle we determined the dispersion of repolarization (defi ned as the range between any of at least 3 terminals on a needle). Subsequently, we determined the maximum and mean of all these dRTs.

200 ms

0.4

mV

largest T integral

largest QRS integral

A1: peak of QRS in leadwith largest QRS integral

Lead with minimum voltagepotential at the time of A1

B1: peak of T-wave in leadwith largest T integral

Lead with minimum voltage potential at the time of B1

Figure S1: Example of determination of the exact QRS and T vector in the BS-ECG. The lower recording is the lead with the largest QRS integral. In this lead the peak of QRS is determined. At this time the upper recording shows the lead with the minimum voltage potential at the time of QRS peak (A1). Between these two leads the vector angles were calculated. In this example the T vector was calculated between the upper lead (largest T integral at time of T wave peak (B1)) and lower lead (minimum voltage).


2

Root Mean Square (RMS)Because a single BS-ECG lead may be limited by the view from the dominant axis of the cardiac vector, the RMS (as global ECG measure) may provide more reliable information on the underlying repolarization process. However, a single BS-ECG lead is clinically more relevant than the RMS and we suppose that the BS-ECG lead with the largest T-integral is a close approximation of this RMS. Nevertheless, to validate the use of a single BS-ECG lead over the RMS we compared them. The RMS was composed by superimposing all BS-ECGs.

RMS = √ ∑61i=1 BSECGi

2(t)

61

where BSECG(t) denotes the BS-ECG signal at time t, of lead i. Subsequently, we deter-mined the peak of the T wave in the RMS (Tp_RMS) and compared this with Tp (of the maximum T-ECG integral) and the RT values (RTmin, RT25, RT50, RT75, RTmax).

Supplemental Results

In vivo standard 12-lead ECG and ex vivo BS-ECGIn general, in the in vivo ECG the amplitude of the QRS-complex was smallest in lead aVR (6/9 animals) and largest in lead V1 or V2 (8/9 animals). In most cases, the QRS-complexes of the extremity leads were biphasic and had no dominant polarity, causing a large varia-tion in the angle of their main QRS-vector in the frontal plane between animals. However, the QRS-vectors in the transversal and sagittal plane were rather constant (both had a smaller SD in QRS-vector compared to that in the frontal plane, (see Table 1: in vivo).

Comparison of the QRS-vectors and T-vectors between the in vivo ECG and BS-ECG showed a rotation of the QRS-vector in all planes and no change in T-vector, although the latter showed a larger standard deviation than the change in QRS-vector. After correc-tion for the change in QRS-vectors, the change in T-vectors remained not significant and the standard deviations were still large (frontal: −50±84 degrees, transversal: −49±127 degrees, sagittal: −33±82 degrees, all P= n.s.).

The differences in QRS-vectors (in transversal and sagittal plane) between the BS-ECG and the in vivo ECG may be expected as a result of a different position of the heart in the thorax versus the bucket. In the thorax the LV was situated towards posterior and the apex pointed towards anterior, whereas in the bucket the LV was rotated more to the front and the apex was directed to the bottom. This positional change corresponds with less negative transversal and sagittal QRS-vectors in the BS-ECG versus the in vivo ECG (see Table 1).

36 Chapter 2

Single BS-ECG lead versus the RMSIn Figure S2 an example of a comparison between the BS-ECG from the lead with the maximum T-integral (blue) and the RMS (red) is shown. In particular, comparison dem-onstrated that the time of the peak of the T wave in the RMS (Tp_RMS) is statistically not different from Tp (of max T-integral). The mean difference is −5 ± 10 ms (Tp – Tp_RMS, paired t test, P= n.s., n=9). Comparison of Tp_RMS with the RT values demonstrated that Tp_RMS was not significantly different from RT25 (Tp_RMS – RT25, mean of −1 ± 8 ms, paired t test, P=n.s., n=9), and this had also the largest ICC value (0.96). This result is similar to the comparison between Tp of the maximum T-integral and the RT values. As a consequence, about 24 ± 12% of the sites is repolarized at the moment of Tp_RMS, which is not significantly different form the percentage repolarized at the moment of Tp derived from the lead with maximum T-integral.

Freq

uenc

y

0

5

20

15

10

Entire Heart

250 375

100 ms

0.2

mV

All (n=48)

RT5

RT25

RT50

RT75

RT95

TpTo Te

RMS signalBS-ECG with max Tint0.

1 m

V

0 %

100 %

Figure S2: Copy of the bottom panel of Figure 5 with addition of the T wave of the root mean square (RMS) of all BS-ECG leads (red) for the comparison with the T-ECG from the lead with the maximum T-integral (blue). The onset (To), peak (Tp) and end (Te) of the T-ECG are marked with black dashed lines.

Activation and RepolarizationEarliest moment of activation was always a few milliseconds later than QRSonset on the BS-ECG (difference of 4±2 ms, one-sample t test P<0.001) and latest activation was always earlier than QRSend on the BS-ECG (difference of −11±4 ms, one-sample t test P<0.001).

Transmural activation and repolarizationTransmural activation along needles was virtually simultaneous. The mean difference between sub-endocardium (endo) and sub-epicardium (epi) along a needle was 0±2 ms (P= n.s. different from 0, Wilcoxon signed-rank test per animal with Bonferroni correc-tion, n=8, one animal has only one endo-epi needle and was excluded).


2

Transmural repolarization was also simultaneous. The averaged difference in RT be-tween epicardial and endocardial sites along a needle was 12±7 ms (P= n.s. different from 0, Wilcoxon signed-rank test per animal with Bonferroni correction, n=8).

Supplemental References 1. 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.


3. Potse M, Vinet A, Opthof T, Coronel R. Validation of a simple model for the morphology of the T wave in unipolar electrograms. Am J Physiol Heart Circ Physiol 2009; 297: H792–H801.

Chapter 3

Synchronization of repolarization by mechano-electrical coupling

in the porcine heart

Tobias Opthof *, Veronique M.F. Meijborg *, Charly N. Belterman, Ruben Coronel

* Both authors contributed equally

Cardiovascular Research 2015 May 1. [Epub ahead of print]

40 Chapter 3

AbSTRACT

Aims: The aim of this study was to evaluate the effect of increase in left ventricular (LV) pressure on repolarization and activation-recovery intervals.

Methods and results: Six pig hearts were Langendorff-perfused. A compliant liquid-filled balloon, connected with a pressure transducer, inserted through the mitral orifice, could be filled until the required LV systolic pressure was obtained. A grid of 121 elec-trodes (11x11; 5 mm interelectrode distance) was sutured on the LV free wall. Ventricular pacing at 600 ms and at 400 or 450 ms was either performed from the LV wall or from the ventricular septum. Under all these four conditions, the pressure wave occurred at the same moment relative to the onset of the QRS complex. Consequently, the time rela-tion between local repolarization and the pressure wave differed between the various pacing sites. Repolarization times (RTs) at a cycle length (CL) of 600 ms were prolonged by increased pressure. With stimulation from the LV, when the pressure wave coincides with the action potentials (APs) late in their phase (sites with relatively early repolariza-tion), an increase in pressure from 0 to 100 mmHg delayed repolarization more than with stimulation from the septum, when the pressure wave occurs at a relatively earlier phase of the AP (sites with relatively late repolarization). At pacing at CL 400/450 ms, an increase in pressure caused RT prolongation at the LV free wall during LV stimulation, but less RT prolongation or even shortening during septal stimulation.

Conclusion: The effect of increased LV pressure is synchronization of repolarization.

MEC synchronizes repolarization 41

3

INTRODuCTION

Heterogeneity of repolarization facilitates re-entrant arrhythmias.1,2 In the presence of a sufficiently large gradient of repolarization, a premature beat may cause re-entrant activation, depending on the activation delay and the mass of tissue with prolonged re-polarization.3-6 In normal hearts, dispersion of repolarization is mitigated by electrotonic interaction.7

Mechanical activity directly feeds back on the electrophysiologic properties of the heart (mechano-electrical coupling, MEC).4 The time of application of stretch relative to the phase of an action potential (AP) has differential effects on AP duration. Following a mechanical impulse received during the plateau phase, opening of stretch-activated channels leads to AP shortening, whereas application of stretch later in the repolariza-tion phase may lead to AP prolongation.8 Zabel et al8 have reported that mechano-elec-trical effects during a prolonged stretch pulse change in sign at a membrane potential of about −30 mV, i.e. shortening of AP is found at more positive potentials, whereas prolongation of AP is observed at more negative membrane potentials.

In view of the activation sequence of the heart, we hypothesized that the mechano-electrical effect of a normal left ventricular (LV) pressure pulse on repolarization differs between early- and late-activated sites. We therefore mapped the time of repolariza-tion (RT) and activation-recovery intervals (ARIs, an index of local AP duration9) in the absence and presence of an LV pressure pulse in early- and late-activated myocardium.

This study shows that the physiological LV pressure pulse causes a prolongation of the ARI that is larger at early repolarizing sites than at late repolarizing myocardium, leading to a decrease of dispersion in repolarization.

METHODS

Experiments were in accordance with the European Directive for the Protection of Ver-tebrate Animals Used for Experimental and Other Scientific Purposes (86/609/EU; http://ec.europa.eu/environment/chemicals/lab_animals/legislation_en.htm) and the Dutch Law on Animal Experimentation. The experimental protocol was approved by the local ethical committee on animal experimentation.

Experimental set-up

Male pigs (n=6; weight 30-35 kg) were premedicated with 10-15 mg/kg of ketamine (Nimatek, EUROVET Animal Health BV, Bladel, The Netherlands), 1-2 mg/kg of midazolam (Actavis Group, Iceland) and anaesthetized with 20 mg/kg of pentobarbital (Apotheek Erasmus MC, The Netherlands) intravenously. After intubation, the animals were venti-

42 Chapter 3

lated with room air and 1-1.5% isoflurane (Pharmachemie BV, Haarlem, The Netherlands). Heparin (5000 IU, LeoPharma, Breda, The Netherlands) was injected intravenously. After a midsternal thoracotomy, blood was collected and ventricular fibrillation was electri-cally induced. Then, the heart was isolated and immersed in ice-cold Tyrode’s solution.

The aorta was connected to a Langendorff perfusion set-up and retrogradely per-fused using a 1:1 mixture of blood and Tyrode’s solution (pH 7.35–7.45). The heart was defibrillated and atrioventricular block was created by crushing the AV nodal area.

A compliant balloon was inserted through the mitral orifice, filled with saline, and connected to tubing containing a water-air interface (Windkessel function). The tube was closed above this interface and a pressure transducer was connected to the balloon. A 50 mL syringe allowed filling the balloon until the required LV systolic pressure of 100 mmHg (P 100) was obtained. Unipolar cathodal stimulation at twice diastolic threshold intensity was performed through one of two stimulation sites (see diagram in Figure 1). One was located in the centre of the electrode grid overlying the LV free wall (LV stim), and the other was at the basal side of the interventricular septum (Sep stim). The anode was placed at the aortic root. Basic cycle length (CL) was alternated between 600 (n= 6) and 400 (n= 4) or 450 ms (n= 2). Observations made during pacing at CLs 400 and 450 ms were pooled. Sufficient time was allowed after a change of stimulation rate to reach a steady-state and then the intraventricular balloon was inflated to obtain a predeter-mined peak systolic pressure for ~30 s.10

Electrophysiologic and pressure recordings

A rectangular grid of 11×11 electrodes (5-mm interelectrode distance) was sutured on the LV free wall. Local unipolar electrograms were recorded against the reference electrode at the aortic root using a data acquisition system (Biosemi, Amsterdam, The Netherlands; sampling rate 2048 Hz, filtering DC 0.4 kHz [3dB]).

Analysis of the electrograms was performed offline using a custom-made analysis programme.11 Local activation times (ATs) were measured at the moment of the mini-mum dV/dt of the initial deflection and local RTs at the moment of the maximum dV/dt of the T wave relative to the stimulus artefact (t=0).9 Activation-recovery intervals (ARIs), as a measure of AP duration, were calculated by subtracting the local ATs from the local RTs. The number of sites with signals of sufficient quality varied from 82 in one heart to 117 in another from a maximum of 121 sites covering an area of 25 cm2.

In the pressure recording, we determined the moment of peak pressure. We also measured contraction time (CT) defined as the moment of half maximum pressure pre-ceding the peak pressure. Pressure recordings were fed into the same data acquisition system as used for the electrical recordings.


3

Data analysis

All measured variables were normally distributed and therefore data are presented as mean ± S.E.M. Data are provided with the number of sites (n) and the number of hearts (N). Multiple comparisons were made using analysis of variance with repeated measures. A P value <0.05 was considered statistically significant.

RESuLTS

Figure 1 shows an example of an LV pressure pulse superimposed on LV electrograms fol-lowing stimulation from the right side of the interventricular septum or from the LV free wall epicardium. The intraventricular balloon was filled (with saline) to obtain a systolic pressure of 100 mmHg (P 100), whereas diastolic pressure remained 0 mmHg (P 0). The large R-wave in the local QRS complex and the negative T wave indicate late activa-

40Ê mVÊ22Ê mmÊ HgÊ

200Ê msÊ

LVÊ stimÊ

SepÊ stimÊ

LVÊ

Figure 1: Increase in pressure and timing relative to phase of repolarization. Schematic drawing of the ex-perimental procedure. The grid of electrodes was at a fixed position at the LV lateral epicardial wall. Stimula-tion was either from the septum (top panel) or from the centre of the grid (bottom panel). The blue vertical line shows the moment of local repolarization (dV/dtmax in the local T wave) during stimulation from the septum (Sep stim); the red vertical line shows the moment of local repolarization during stimulation from the centre of the LV grid (LV stim). The black vertical lines indicate the moment of 50% of maximal pressure.

44 Chapter 3

tion and late repolarization at the lateral LV grid, respectively, during stimulation from the septum (top panel). Conversely, a deep Q-wave and a positive T wave in the lower panel indicate early activation and early repolarization during LV stimulation (bottom panel) at the same lateral LV grid. The dashed red lines indicate the moments of local repolarization. The dashed blue lines indicate the moment of half maximum pressure. Arrows point to the pacing artefacts (virtually absent in the top panel). The timing of the pressure pulse relative to the pacing artefacts is about the same under the two pacing conditions. The pressure wave occurs relatively later during the electrical systole (i.e. closer to the final repolarization phase of the local APs) following LV stimulation than following septal stimulation.

ATs and pressure wave

We tested whether ATs were affected by changes in pressure. First, we averaged the ATs per grid to arrive at a mean value for each of the six hearts under the eight conditions (two CLs, two pacing sites, and two pressures). The mean ATs before and after balloon inflation were compared at the two CLs and at the two pacing sites in each heart. No statistically significant differences in AT occurred between P 0 and P 100 with LV stim at either CL (Table 1). During Sep stim when ATs were 70-80 ms, a small increase in AT (by on average 5 ms; Table 1) was found between P 0 and P 100, which was significant at CL 600 ms, but not at 400/450 ms.

During high pressure, the moment of the peak pressure and the moment of 50% of the peak pressure did not differ between pacing CLs or between pacing sites (two-way repeated measures ANOVA). CT occurred at 159 ± 5.3 ms and the moment of the peak of the pressure wave was at 222 ± 3.7 ms after the stimulus artefact (24 conditions pooled

Table 1: Influence of pressure increase on ATs and CT (mean±S.E.M., n=6 experiments).

CL Stim P Mean AT S.E.M. P value Mean CT S.E.M.

600 Sep 0 73 5.7 - -

600 Sep 100 77 5.4 P<0.05 163 13.2

600 LV 0 40 2.7 - -

600 LV 100 40 2.6 ns 152 9.7

P value ns

400/450 Sep 0 79 6.8 - -

400/450 Sep 100 85 7.0 ns 166 12.8

400/450 LV 0 44 2.7 - -

400/450 LV 100 44 3.1 ns 156 7.2

P value ns

CL, cycle length; Stim, stimulation site; Sep, septum; LV, left ventricle; AT, activation time; CT, contraction time; significance: column P 0 vs P 100; rows: Sep stim vs. LV stim; ns, not significant.


3

(two CLs and two pacing sites) in 6 experiments). Table 1 summarizes the averaged data at the two CLs and the two pacing sites per experiment. Despite the absence of diff er-ences between the timing of the pressure wave in the six experiments, all RTs within one heart were compared with the CT of the same heart in further data analysis.

Repolarization times

Figure 2 (middle panels) shows two repolarization maps of the LV epicardium following LV pacing at CL 600 ms from an electrode in the centre of the electrode grid, during low and high systolic LV pressure (P 0 and P 100). Representative electrograms (top panels) recorded at the site indicated by the blue circle in the maps show a later RT at P 100 than at P 0 (blue circles projected on the electrograms). Data from the 114 electrograms recorded from the grid were averaged (Figure 2, bottom panel) and show an increase in RT from 296 ± 1.0 ms at P 0 to 306 ± 1.3 ms at P 100 (P<0.0005). The change varied from a

10 mm

280 280

270 Repolarization time (ms)

P 100

P 0 P 100

200 ms 30 m

V

LV stim (CL 600 ms)

P 0

340

290

295

300

305

310

P 0 P 100

RT (ms)

Figure 2: Eff ect of an increase in pressure on repolarization with stimulation from one site. Top panel: Local electrograms recorded at the blue dots in the mid panel. Stimulation from the mid of the LV grid at CL 600 ms. The circles indicate the local repolarization moments (dV/dtmax in the local T waves). Mid panel: Repo-larization map of the same LV lateral epicardial area at low and high pressure in 1 heart. bottom panel: Quantifi cation of the delay in repolarization. The increase was 10 ± 0.6 ms (P<0.001, n=114,N=1). See text for further numerical data.

46 Chapter 3

minimum of −1 ms to a maximum of +37 ms with +10 ±0.6 ms as average of the 114 sites in the grid.

Figure 3A shows the effect of a change in pacing site on the pressure-induced change in RT at CL 450 ms in another heart. During Sep stim, the increase in pressure induced an earlier RT at the large majority of sites (86 of 110 sites; left panels), whereas it caused overall an increase in RT during LV stim (right panels). Figure 3B shows that the increase in pressure caused on average earlier repolarization by −5 ± 0.6 ms (P<0.001) during Sep stim and on average slightly later repolarization by 2 ± 0.8 ms (P<0.05) during LV stim (each an averaged change of 110 electrograms). Obviously, the difference in RT during Sep stim and LV stim is smaller at P 100 mmHg than at P 0.

260

260

300

300

*338

Repolarization time (ms)

*331

10 ms classes

Sep stim CL 450 ms LV stim

P 0

P 100

10 mm

(A)

312

314

316

318

320

322

Sep stim P 0 Sep stim P 100270

272

274

276

278

280

LV stim P 0 LV stim P 100

RT (ms) RT (ms) (B)

Figure 3: Effect of an increase in pressure on repolarization with stimulation from different sites. A: Effects of a change in pacing site in combination with an increase in pressure on RTs at the left lateral epicardial wall in 1 heart (n=110, N=1). During stimulation (at 450 ms) from the septum (Sep stim), there was earlier repolarization in the presence of pressure. During stimulation from the LV (LV stim), there was slightly later repolarization at many sites (see the larger green and cyan areas). b: Quantification of the changes in RTs following an increase in pressure at the two pacing sites. During Sep stim, repolarization was advanced by the increase in pressure, and during LV stim, it was slightly delayed.


3

Activation-recovery intervals

Because the changes in RTs following an increase in pressure (Figures 2 and 3) do not result from changes in ATs (Table 1), they must result from changes in AP duration. ARIs were derived from 82 to 117 sites per heart.

We plotted the diff erence between ARIs at low and high pressure at the same elec-trode site as a function of the diff erence in timing between local RT (at P 0) and CT (at P 100). Figure 4 shows an example of pressure-induced changes in ARIs (∆ARI) as a func-tion of the time period between local RTs (n=110) and CT at CL 450 ms. ∆ARI is plotted along the ordinate during LV stim (red symbols) and Sep stim (blue symbols). Figure 4 (black symbols) shows that the average pressure-induced ∆ARI is larger during LV stim than during Sep stim when the diff erence between RT and CT becomes larger. The S.E.M. values of the averages (black symbols) are so small (~1 ms) that they fall within them.

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

45

70 80 90 100110120130140150160170

∆AR

I (hi

gh p

ress

ure

min

us lo

w p

ress

ure)

(m

s)

RT-CT (ms)

LV stim

Sep stim

Averages

Later phase of Earlier phase of repolarization repolarization

Figure 4: Eff ect of an increase in pressure on changes in ARIs with stimulation from diff erent sites in 1 ex-periment. Relation between the time diff erence of local RT (n=110, N=1) and the moment of half maximal contraction (CT, n=1, N=1) at the abscissa and the eff ect of the change in local activation recovery intervals (∆ARIs) by an increase in pressure at the ordinate in 1 heart. The insets (upper left, upper right corners) show the two pacing modes (compare with Figure 1). The ∆ARIs are measured at the same 110 left lateral epicardial LV sites during stimulation from the LV and from the septum. The black symbols are the averaged X- and Y-values of the whole grid. The S.E.M. values in both the X- and Y-direction are so small (~1 ms) that they fall within the symbols. See text for further explanation.

48 Chapter 3

The physiological implication in this particular heart and at this CL is that a 35 ms later ‘arrival’ of the pressure wave (difference between the X-values of the black symbols) dur-ing an AP will reduce the difference in AP duration between early and late repolarizing sites by 8 ms (difference between the Y-values of the black symbols). This will reduce dispersion in RTs in the presence of pressure by 23% (8/35 x 100) compared with the condition without pressure. The degree of overlap between the red and blue data points varied between the six experiments and is explained by the distance between the two stimulation sites relative to the (fixed) recording site, but also to changes in heart size and to the exact position of the pacing electrodes.

Figure 5 shows the effect of pressure increase on all ∆ARI data as in Figure 4 in all six hearts, during each pacing mode and cycle length. The error data in the X- and Y-direc-tion now depict biological (interindividual) variability between the 6 hearts. Overall, an increased pressure causes more prolongation of the ARIs when RT minus CT is small (LV stim) than when RT minus CT is larger (Sep stim), consistent with Figure 4. Dividing the difference in Y-values by the difference in X-values in Figure 5 between LV stim and Sep

-5

0

5

10

15

20

100 120 140 160 180 200

∆AR

I by

incr

easi

ng p

ress

ure

(ms)

RT - CT (ms)

LV stim (600 ms)

Sep stim (600 ms)

LV stim (400/450 ms)

Sep stim (400/450 ms)

Later phase of repolarization Earlier phase of repolarization

Figure 5: Effect of an increase in pressure on changes in ARIs with stimulation from different sites in the average porcine heart. RT minus moment of contraction (RT-CT) vs. change in ARIs (∆ARIs) following an increase in pressure in all six hearts at two CLs and two pacing sites. The S.E.M. values in X- and Y-direction depict interindividual variability between the six hearts. The two data points at 400/450 ms are the average of black symbols as in Figure 4, in the six hearts. The same is pertinent to the two data points at 600 ms. The number of observations of each of the four data points is 599/6. See text for explanation.


3

stim both at CL 600 ms (filled symbols) and at CL 400/450 ms (open symbols), results in a ratio of ∆8 ms/37 ms= 0.22. This already yields a rough estimate of a 22% reduction in dispersion of RT as the effect of an increase in pressure. When the same procedure is applied to the 6 hearts (as in Figure 4) at the two CLs, these 12 ratios can be averaged to 18 ± 7.5% (P< 0.05).

In three of the six experiments, we assessed whether a dose-response effect could be established between the effect of pressure application and the effect on the ARIs. In all three experiments, pressure was increased to 50 mmHg as well as to 100 mmHg. The ∆ARIs of the increase from P 0 to P 100 were 92 ± 2.7% from that observed during an increase from P 0 mmHg to P 50 mmHg. In two of these three experiments the effect of an increase of pressure from 0 mmHg to 16 mmHg was tested as well. The result was 0 and 5% of the effect at P 50. In one of these three experiments the effect of an increase from 0 mmHg to 30 mmHg was tested and compared with the effect at P 50. The effect was 77% of that at P50.

DISCuSSION

Our data show that the repolarization of the heart is modulated by the LV pressure wave. The pressure wave induces a delay of RT at the longer CL. This delay is larger at sites with an early intrinsic repolarization (when the contraction coincides with the AP in a relatively late phase) than at sites with a relatively late intrinsic repolarization (when contraction occurs at an earlier phase of the AP). At a shorter CL, the same relation is observed, although here shortening may even be observed, when the pressure wave coincides with the AP at a relatively early phase, and local RT is relatively late.

In late repolarizing areas, ‘RT minus CT’ is long and the pressure wave arrives at a relatively early phase of repolarization. Our study shows that this leads to less ARI prolongation or even shortening following an increase in pressure. The physiological implication is that the presence of pressure leads to a decrease of dispersion in repolarization (Figures 4 and 5).

We should emphasize that the measurement of a single (central) intracavitary pressure wave is not informative on local stretch and strain values, which would require local-ized regional data on these parameters just as obtained for the electrophysiologic data. It is possible that local mechanical factors are different in the presence or absence of pressure as a kind of mechanical-mechanical coupling. This may complicate the relation-ship between pressure and ARIs/repolarization described in our study. In the pig, it has

50 Chapter 3

been reported that the activation pattern roughly coincides with the contractile pattern and is opposite to the repolarization pattern.12 In our recent study,13 on the genesis of the T wave in relation with repolarization patterns in the pig we could not confirm a base-apex or apex-base order of either activation or repolarization. Final repolarization occurred in the equatorial area, not in the apical or basal area, and RV-LV differences ap-peared important as well.13 In man, it has been described that the onset of contraction is far away from the site of earliest electrical activation. In fact, that site starts to contract relatively late.14 The strength of our experimental approach is that we record from the same sites during low and high pressure, as well as during early and late activation/repolarization. Local contractile characteristics are therefore the same, as well as the amount of preload and afterload. Although the epicardial grid may have somewhat prevented stretching of the recording sites, we cannot exclude that, during remote pac-ing, the recording sites were initially stretched, whereas this was not the case during proximal pacing. Nevertheless, we demonstrate that MEC can significantly modulate repolarization under physiological LV pressure, and that this tends to reduce repolariza-tion heterogeneity. This has consequences for various disease states associated with abnormal contractility.

MEC is thought to be mediated by activation of stretch-activated channels, although many other ion channels also have been demonstrated to be sensitive to stretch.15 Stretch-activated channels have a reversal potential of about −30 mV.4 The difference between the local membrane potential and the reversal potential of the stretch-activated channels determines the magnitude and direction of the effect on the AP. If the pressure pulse occurs relatively late in the electrical cycle, AP prolongation has been documented, whereas AP shortening occurs when the stretch is applied during the plateau phase of the AP.8

Our observation that the timing of a pressure pulse relative to the moment of repo-larization determines the magnitude and direction of its effect is in accordance with those of Zabel et al8 who applied artificial pressure pulses with various timing relative to the cardiac cycle in isolated perfused rabbit hearts. The novelty of our study is that a normal LV pressure was used, and that existing repolarization differences within the heart create timing differences between the pressure pulse and repolarization within a single heartbeat. Thus, the intervention did not involve variation of the timing of an artificial pressure wave through the different phases of the APs, but varying the phase of repolarization during a contraction at a fixed moment against increased pressure. If the myocardium under the electrode grid was activated late (and therefore repolarized late) as following septal pacing, the pressure wave fell relatively early in the local AP, and less prolongation (or even shortening) of the ARI occurred than when activation (and repo-larization) was early (as during LV stimulation). Our approach to measure ATs and RTs at


3

the same myocardial area while changing the site of stimulation has the advantage that the intrinsic properties of the tissue studied are unchanged.

We documented no difference in the time of the onset and peak of the pressure wave during the various pacing modalities (LV or septal pacing and long or short CL). Therefore, the time of the pressure wave relative to local repolarization depends on local repolarization differences only.

Our study describes an influence of MEC on physiology, which has consequences for pathophysiology. Our data lend mechanistic support for the observation that loss of synchronous contraction of the heart, as occurs in patients with myocardial infarction, is associated with an increased dispersion of repolarization.16 In our previous study, we demonstrated that the increased dispersion in repolarization was located preferentially in the non-infarcted tissue, indicating that MEC rather than electrophysiological remod-elling in the infarction zone contributed to this increased dispersion.16 Secondly, based on our present observations, we speculate that a premature beat with a short coupling interval has a more heterogeneous repolarization because of the absence of a relevant pressure pulse. Indeed, our data show that at a pressure pulse of <50 mmHg the effect of MEC starts to disappear, whereas it is abolished <20 mmHg. Thirdly, we speculate that, in patients with atrial fibrillation and the resulting variable LV filling and output, beat-by-beat variability of dispersion in repolarization exists that may predispose these patients for re-entry-based ventricular arrhythmias as well.

Atrial and ventricular dilatation has been shown to promote arrhythmias,17-20 whereas we infer that LV pressurization may contribute to reduced repolarization heterogeneity and therefore is antiarrhythmic. Contrary to most experimental studies, we used a nor-mal pressure pulse, without an increase in diastolic pressure. Whereas the early studies are relevant for arrhythmias following a sudden impact on the heart (commotio cordis)21 or (relief of ) acute dilatation,22 our data point to a protective mechanism during normal cardiac function. A further increase in LV pressure, diastolic, as well as systolic may be arrhythmogenic.

In patients with LQT syndrome, the repolarization phase of the APs will be shifted compared with the contractile wave, of which the onset will be dictated by the acti-vation process rather than by the repolarization process. It is therefore possible that a changed temporal relation between the pressure wave and the repolarization process will affect the effect of MEC. MEC reduces dispersion in repolarization in a normal heart (this study), but not necessarily in a heart with abnormal repolarization. Such a changed effect of MEC may be more important than the effects of the mutations.23,24 In the same way, our data may be relevant for resynchronization therapy as well. Such therapy aims primarily at obtaining a shorter QRS complex by pacing from multiple sites.25 In diseased hearts, this may result in increased dispersion of repolarization. Based on our data, an

52 Chapter 3

alternative approach would be to focus resynchronization therapy on optimizing the temporal relation between local repolarization and contractile function.

The relations between the pressure-induced alteration of the ARIs and the time of the repolarization relative to the pressure pulse differ when pacing is performed at CL 450 ms or at CL 600 ms. These differences cannot be explained by the difference in time between the intrinsic local repolarization and the pressure wave only, because larger prolongations of ARIs due to the increase in pressure would be anticipated at the shorter CL than at the longer CL. We speculate that the ratio between the densities of IKs and IKr current differs at different CLs leading to shorter APs at shorter CLs (restitution). It is conceivable that the influence of MEC is different under these different conditions.

Limitations

We measured (changes in) ARIs at a limited epicardial area of the LV myocardium, which precludes an assessment of dispersion in repolarization of the whole heart, including its transmural distribution. However, transmural needles would disturb the contraction process and are incompatible with the presence of an intraventricular balloon.

It may be anticipated that a change in initiation of the ventricular contraction wave would impact on its time course. The resulting difference in the morphology of the pres-sure wave may impact on repolarization in general. However, we show that the timing of the onset and of the peak of the pressure wave was not different between the conditions (altered pacing site and altered pacing CL). Indeed, Figure 1 shows a remarkable similar-ity of the morphology of pressure waves during pacing from different sites, whereas the timing relative to the local electrogram is prominent.

In conclusion, our study points to an antiarrhythmic contribution of MEC in the normal heart, by synchronization of RTs.

ACkNOWLEDGEMENTS

The authors wish to thank Eline van Groen for help with the experiments.


3

REFERENCES

1. Mines GR. On circulating excitations in heart muscles and their possible relation to tachycardia and fibrillation. Trans R Soc Can 1914; IV: 43–53.

2. 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.

3. Coronel R, Wilms-Schopman FJG, Opthof T, Janse MJ. Dispersion of repolarization and arrhythmo-genesis. Heart Rhythm 2009; 6: 537–543.


5. Hansen DE, Craig CS, Hondeghem LM. Stretch-induced arrhythmias in the isolated canine ven-tricle. Evidence for the importance of mechanoelectrical feedback. Circulation 1990; 81: 1094–1105.

6. Janse MJ, Coronel R, Opthof T. Counterpoint: M cells do not have a functional role in the ventricu-lar myocardium of the intact heart. Heart Rhythm 2011; 8: 934–937.

7. Moe GK. Oscillating concepts in arrhythmia research; a personal account. Int J Cardiol 1984; 5: 109–113.

8. Zabel M, Koller BS, Sachs F, Franz MR. Stretch-induced voltage changes in the isolated beating heart: importance of the timing of stretch and implications for stretch-activated ion channels. Cardiovasc Res 1996; 32: 120–130.


10. Janse MJ, Van Der Steen ABM, Van Dam RT, Durrer D. Refractory Period of the Dog’s Ventricular Myocardium Following Sudden Changes in Frequency. Circ Res 1969; 24: 251–262.


12. Sengupta PP, Khandheria BK, Korinek J, Wang J, Jahangir A, Seward JB, Belohlavek M. Apex-to-base dispersion in regional timing of left ventricular shortening and lengthening. J. Am. Coll. Cardiol. 2006, pp. 163–172.

13. Meijborg VMF, Conrath CE, Opthof T, Belterman CNW, De Bakker JMT, Coronel R. Electrocardio-graphic T wave and its relation with ventricular repolarization along major anatomical axes. Circ Arrhythmia Electrophysiol 2014; 7: 524–531.

14. Zwanenburg JJM, Götte MJW, Kuijer JPA, Heethaar RM, van Rossum AC, Marcus JT. Timing of cardiac contraction in humans mapped by high-temporal-resolution MRI tagging: early onset and late peak of shortening in lateral wall. Am J Physiol Heart Circ Physiol 2004; 286: H1872–H1880.

15. Morris CE. Voltage-Gated Channel Mechanosensitivity: Fact or Friction? Front Physiol 2011; 2: 25.

16. Opthof T, Sutton P, Coronel R, Wright S, Kallis P, Taggart P. The Association of Abnormal Ventricular Wall Motion and Increased Dispersion of Repolarization in Humans is Independent of the Pres-ence of Myocardial Infarction. Front Physiol 2012; 3: 235.

54 Chapter 3

17. White E, Benoist D, Bernus O. Regional variation in mechano-electrical coupling: the right ven-tricle. In Kohl P, Sachs F, Franz MR, eds: Cardiac mechano-electric coupling and arrhythmias. 2nd Edition. 2011, pp. 217–222.

18. Franz MR, Bode F. Mechano-electrical feedback underlying arrhythmias: The atrial fibrillation case. Prog Biophys Mol Biol 2003, pp. 163–174.

19. Taggart P, Lab M. Cardiac mechano-electric feedback and electrical restitution in humans. Prog Biophys Mol Biol 2008; 97: 452–460.

20. Reiter MJ, Synhorst DP, Mann DE. Electrophysiological effects of acute ventricular dilatation in the isolated rabbit heart. Circ Res 1988; 62: 554–562.

21. Nesbitt AD, Cooper PJ, Kohl P. Rediscovering commotio cordis. Lancet (London, England) 2001; 357: 1195–1197.

22. Coronel R, Langerveld J, Boersma LVA, Wever EFD, Bon L, Van Dessel PFHM, Linnenbank AC, Van Gilst WH, Ernst SMPG, Opthof T, Van Hemel NM. Left atrial pressure reduction for mitral stenosis reverses left atrial direction-dependent conduction abnormalities. Cardiovasc Res 2010; 85: 711–718.

23. Haugaa KH, Amlie JP, Berge KE, Leren TP, Smiseth O a, Edvardsen T. Transmural differences in myo-cardial contraction in long-QT syndrome: mechanical consequences of ion channel dysfunction. Circulation 2010; 122: 1355–1363.

24. Rosenbaum DS. Is long QT syndrome a disease of abnormal mechanical contraction? Circulation 2010; 122: 1353–1354.

25. Ploux S, Strik M, van Hunnik A, van Middendorp L, Kuiper M, Prinzen FW. Acute electrical and hemodynamic effects of multisite left ventricular pacing for cardiac resynchronization therapy in the dyssynchronous canine heart. Heart Rhythm 2014; 11: 119–125.

Chapter 4

The effect of left ventricular pressure on the STT segment in the surface

electrocardiogram

Veronique M.F. Meijborg, Ruben Coronel, Charly N.W. Belterman, Jacques M.T. de Bakker, Chantal E. Conrath

Submitted for publication

56 Chapter 4

AbSTRACT

Objective: Influence of LV pressure on the T wave is still not unraveled. We hypothesize that ventricular pressure influences the T wave.

Methods: Five Langendorff-perfused pig hearts (right atrial pacing at 650 ms cycle length) were submerged in a blood-filled container allowing recording of a 61-lead pseudo-ECG. A flexible balloon in the left ventricle (LV) enabled recording during low (P0= 0 mmHg) and high (P20= 20 mmHg) LV diastolic pressure. Integrals of the QRS interval and STT interval were calculated as area under the curve (QRSint and STTint). Within the integral maps we determined dipoles (vector angle and amplitude), cross-correlations (P0 versus P20), and absolute integral differences per lead (P20 minus P0). An epicardial multielectrode (9x12) sock was used to determine simultaneous local activation and repolarization times (ATs and RTs).

Results: Balloon inflation increased diastolic and systolic LV pressure, without changing the time of pressure pulse onset. With the increase in LV pressure patterns in the QRSint and STTint remain similar (unchanged vector angles and high cross-correlations). How-ever, integral map amplitudes (dipole amplitude and mean absolute amplitude) were increased, with significant larger changes in STTint than in QRSint. In the LV, ATs were slightly increased, whereas RTs show larger increases (3±3 versus 9±5 ms). RT dispersion in the heart was also significantly increased by 17±9ms and resulted from interventricu-lar differences.

Conclusions: Significant larger absolute integral changes in STT integrals between P0 and P20 than in QRSints indicates that mechano-electric feedback due to increased LV pressure influences the STT segment more than the QRS complex.

Pressure effects on the STT segment 57

4

INTRODuCTION

The T wave in the electrocardiogram (ECG) represents the repolarization of the cardiac ventricles. Repolarization abnormalities play a significant role in arrhythmogenesis.1,2 For example, an increased dispersion of repolarization has recurrently been associated with life-threatening arrhythmias.3 Multiple factors, like temperature,4 ischemia,5 potassium level,5 sympathetic activation,6 and mutations7,8 are known to influence the repolariza-tion and vulnerability for arrhythmias. Therefore, the T wave holds valuable information. Yet, knowledge about the genesis and modulation of the T wave is incomplete.

Heart failure is associated with sudden cardiac death and changes in repolariza-tion.9,10 Heart failure is also associated with ventricular overload and increased end diastolic left ventricular (LV) pressure.11 Both factors lead to stretch on the ventricular wall. It has been demonstrated that stretch – by mechano-electric coupling – has effect on the ventricular action potential and arrhythmogensis.12 It may prolong or shorten the repolarization depending on the timing of the pressure pulse with respect to the action potential phase.12 We recently demonstrated that loss of pressure in the LV causes heterogeneity of repolarization.13 However, the effect of pressure on the T wave and its relation with underlying repolarization changes is not clear.

Therefore, we experimentally modeled LV overload by increasing diastolic LV pressure and simultaneously examined the effects on a volume-conducted ECG and ventricular repolarization. We hypothesized that the increased LV pressure changes the T wave morphology rather than the QRS morphology, because the pressure pulse normally falls after the end of QRS and during the T wave.

METHODS

Five male pigs, weighing 55±8 kg (mean±SD) were premedicated, ventilated and an-aesthetized according to the protocol described before.14 The experimental protocol was approved by the local ethical committee on animal experimentation. The heart was excised and perfused according to Langendorff with a 1:1 blood–Tyrode’s mixture. The apex was fixed to avoid swinging of the heart, but to allow rotation and movement in the vertical plane.

Experimental setup

Hearts were paced from the right atrium at 650 ms cycle length. We positioned a flexible balloon in the LV, in which diastolic LV pressure was maintained at 0 mmHg (P0) or 20 mmHg (P20) and continuous LV pressure was recorded. A 9x12 electrode sock (interelec-trode distance of 9 mm vertically, and of 9 – 17 mm horizontally) was fixed on the heart

58 Chapter 4

overlapping the majority of the epicardial surface of the ventricles to obtain unipolar electrograms (see Figure 1). The heart was submerged in a perfusion-fl uid fi lled bucket incorporating 61 regularly distributed electrodes (1 at the bottom) to simultaneously record volume-conducted pseudo-ECGs with the assembled average of all 61 electrodes as reference. We performed experiments in nine hearts. In fi ve of these hearts we were able to increase the LV diastolic pressure from 0 to 20 mmHg and simultaneously obtain recordings of suffi cient quality without pressure-induced ST segment elevation/depres-sion.

Electrophysiologic recordings

In the pseudo-ECGs we determined the maximum QRS duration and maximum QT interval. QRSonset and QRSend were defi ned as, respectively, the fi rst defl ection and last J-point of the QRS complex in any ECG lead. The Tend was determined using the tangent method. Accordingly, in each lead QRS integrals and STT integrals were calcu-lated defi ned as area under the QRS and STT (=QRSend-to-Tend) curve. Integral maps were confi gured and dipoles (minimum to maximum integral) were determined. The QRS and STT dipoles were characterized by the vector amplitude (mV) and the vector angles (degrees) in the frontal and transversal plane (the left lateral lead equals 0 de-grees and rotation to the anterior/inferior lead denotes a positive angle). For the QRS and STT integral maps we calculated the cross-correlation and the mean of absolute integral diff erences between P0 and P20 (mVms). In each unipolar epicardial electro-

ECG lead

epicardial electrograms

QRS STT

RT

AT

onset of QRS

200 ms

200 ms

4 m

V0.

4 m

V

= RV

= LV

18 cm

7 cm

10 cm

9 mm

17 mm

9 mm

9x12 electrode grid

Figure 1: Schematic setup of electrophysiologic measurements (left): examples of 3 epicardial electro-grams and 1 pseudo-ECG. Red areas in the pseudo-ECG lead illustrate the integral determined for the QRS and STT interval.Experimental setup (right) shows a heart with a fl exible balloon positioned in the left ventricle (LV). An epicardial 9x12 electrode sock overlap the LV and large part of the right ventricle (RV). Electrodes assigned to the RV or LV regions were indicated by the red or blue area respectively. The block pulse indicate the position of the pacing electrode.

Pressure eff ects on the STT segment 59

4

gram we determined activation and repolarization times (ATs and RTs) as before4 (see Figure 1). The LV and right ventricular (RV) regions were determined by visual selection of electrodes that were in all hearts positioned at the LV and RV, respectively (see Figure 1). Dispersion in repolarization was determined within the entire heart, LV region and RV region as maximum RT minus minimum RT. The interventricular RT dispersion was determined as mean RT in the LV region minus mean RT in the RV region. From the LV pressure recordings we determined the systolic and diastolic pressure (Psys and Pdias). Also the moment of halfway peak pressure was determined as the time at 50% of peak pressure relative to the QRSonset as an indication of the start of the pressure pulse.

Statistics

Continuous variables were presented as mean ± SD. Diff erences between P0 and P20 and between QRS and STT integrals were tested with a paired t test. P values ≤ 0.05 were considered as statistically signifi cant.

RESuLTS

Pressure eff ect on pseudo-ECGs

Figure 2 shows an example of the eff ect of increased LV pressure on lead III of the pseudo-ECG. After balloon infl ation diastolic and systolic LV pressures were increased and caused major changes in the T wave without altering the QRS complex. Generally, a rise in diastolic LV pressure (from 1±3 to 20±2 mmHg, P<0.01) occurred and was associ-

0.5 1 1.5

-2

-1

0

0

10

20

30

40

50

time (sec)

mm

Hg

mV

LV pressure P0

LV pressure P20

ECG lead III P0ECG lead III P20

Figure 2: Example of left ventricular (LV) pressure changes with concomitant changes in the ECG lead III, after increasing the diastolic LV pressure from 0 mmHg (P0, blue) to 20 mmHg (P20, red).

60 Chapter 4

ated with an increase of the systolic pressure from 9±5 to 58±14 mmHg (P< 0.01). The moment of halfway peak pressure did not shift in time with respect to QRSonset (109±13 vs 112±16 ms, P= 0.77). QRS durations were prolonged, although not significant (62±12 ms vs 70±10 ms, P= 0.06). The QT intervals were also prolonged (322±18 ms vs 355±44 ms, P= 0.05). In 4 of 5 hearts, the increase in QT interval was at least 3 times larger than the increase in QRS duration.

Figure 3 shows examples of the QRS and STT integral maps including dipole vectors during low and high diastolic LV pressure. The figure demonstrates that when pressure is increased the pattern in the QRS and STT integral maps does not change (i.e. similar po-sitions of the maxima and minima and a similar direction of arrows). The difference maps of the integrals (P20-P0) show clearly that changes in amplitudes of the STT integral are larger (more red and more blue, representing T waves of larger amplitude) than that of the QRS integral. Table 1 demonstrates the mean characteristics of the integral maps of all hearts. Overall, after an increased LV pressure the patterns in the QRS and STT integral map did not change (i.e. dipole vectors did not differ in the frontal or transversal angles

0 500-1.6

-0.8

0

0.8

0 500-1.6

-0.8

0

0.8

posterior - RV - anterior - LV - posterior posterior - RV - anterior - LV - posterior

-30

35

30

-45

0

0

mV.

ms

mV.

ms

QR

Sint

STTi

nt

P0 P20

mV

mV

time (ms) time (ms)

V1 V2

V3

V4

V5 V6

V1 V2

V3

V4

V5 V6

V1 V2

V3

V4

V5 V6

V1 V2

V3

V4

V5 V6

* *

-20

20

0

mV.

ms

-20

20

0m

V.m

s

posterior - RV - anterior - LV - posteriorP20 - P0

-0.8

0

0.8

1.6

0 500

mV

time (ms)-0.8

0

0.8

1.6

0 500

mV

time (ms)

**

Figure 3: Typical example of QRS and STT integral maps with dipole vectors (white arrows) and isointegral lines (grey, 10-mVms distance) and two ECG lead (from the site indicated with the black box and the asterisk) during P0 (left) and P20 (middle). Difference maps (P20 – P0) of the QRS and STT integrals are shown on the right.


4

and cross-correlations were high). However, dipole amplitudes of the QRS integral map were significantly increased with 12±7 mVms (P=0.02). Dipole amplitudes of the STT inte-gral map showed also an increase of 17±14 mVms (P=0.05), but in 1 out of 5 hearts there was a minor decrease of 3 mVms in STT dipole amplitude. The mean absolute differences of the integral maps between high and low pressure (P20 – P0) were two times larger in the STT integral maps compared to the QRS integral maps (respectively 6.52±2.56 versus 3.26±1.36 mVms).

Pressure effect on repolarization

Figure 4 shows an example of the activation and repolarization map during P0 and P20 in the same heart. In all hearts, activation starts in the anterolateral region of the RV and is followed by the LV. In the example, activation is latest in the anterior RV base. In 4 other hearts, this area activated equally late as the LV. Activation times could be deter-mined on 73±9 electrodes in total, 43±4 electrodes in the LV region and 19±3 electrodes in the RV region. An increase in LV pressure showed a slight prolongation (3±3 ms) of the activation in the LV region, without an effect on AT in the rest of the heart (Table 2). Figure 4 also shows that repolarization is earliest at the RV base and latest in the LV apex. This pattern was similar in all 5 hearts, although in 1 heart also the RV apex showed late repolarization during high pressure. Also, the increase in LV pressure did not change the patterns in repolarization. Repolarization times could be determined on 59±7 electrodes in total, 33±5 electrodes in the LV region and 16±3 electrodes in the RV region. During LV

Table 1: Characteristics of the integral maps (mean ±SD).

P0 P20 P value

QRSint

A dipole, mV 65±15 76±12 0.021

Vf dipole, degree −27±40 −32±34 0.371

Vt dipole, degree 18±43 11±43 0.371

Xcorr map 0.99 (0.98 – 1.00)

∆int map, mVms 3.26±1.36

STTint

A dipole, mV 55±22 71±19 0.051

Vf dipole, degree −145±16 −127±49 0.311

Vt dipole, degree 173±40 138±106 0.341

Xcorr map 0.88 (0.55 – 0.98) 0.262

∆int map, mVms 6.52±2.56 0.042

QRSint= integral map of QRS interval; STTint= integral map of STT interval; A, Vf and Vt are, respectively, the amplitude and frontal and transversal vector angle of the dipole; Xcorr= cross-correlation of QRSint or STTint between P0 and P20 in mean (min – max); ∆int= mean of absolute differences (P20 – P0) of QRSint or STTint.P values are given for paired t test: 1P0 versus P20; 2QRSint versus STTint.

62 Chapter 4

pressure increase, repolarization in the LV prolonged significantly with 9±5 ms, without an effect on the RV repolarization (Table 2). These effects were clearly illustrated by the difference maps (P20 – P0) in panel C and F of Figure 4. Furthermore, the dispersion in re-polarization in the entire hearts was also increased with 17±9 ms (Table 2). This increased RT dispersion within the entire heart was mainly caused by an increased interventricular RT dispersion, which was augmented in 4 of the 5 hearts. In these hearts the minimum RT was in the RV and the maximum RT was in the LV.

Table 2: Characteristics of epicardial electrograms (mean±SD).

N=5 P0 P20 P value

AT max, ms 57±12 61±11 0.14

AT total, ms 32±8 35±6 0.06

AT_LV, ms 37±12 41±10 0.04

AT_RV, ms 21±3 21±4 0.75

RT total, ms 247±19 253±21 0.02

RT_LV, ms 258±21 267±24 0.02

RT_RV, ms 227±19 229±18 0.20

dRT total, ms 73±15 90±14 0.01

dRT_LV, ms 45±13 56±18 0.12

dRT_RV, ms 48±12 55±13 0.22

dRT_inter, ms 31±10 38±12 0.06

N= number of hearts, AT max= maximum AT; AT= average of mean AT per heart; RT= average of mean RT per heart; total, LV and RV indicate the sets of electrodes used to determine values; dRT= average of RTmax-RTmin per heart (for total, LV and RV). dRT_inter= average of mean RT_LV – mean RT_RV.P values are given for paired t tests.

ATR

T

P0 P20

200

260

320ms

Septum

0

50

25

ms

−10

10

3030msms

−10

10

30ms

P20 - P0A B C

D E F

Figure 4: Activation time (AT) maps during P0 (A) and P20 (b) and AT difference map P20 – P0 (C). Repolar-ization time (RT) maps during P0 (D) and P20 (E) and RT difference map P20 – P0 (F). Isochrones are at 10-ms intervals in the AT and RT maps, and at 5-ms intervals in the difference maps. Grey line indicates the septal border between right ventricle (left) and left ventricle (right).


4

DISCuSSION

Our study demonstrates that during an increase in LV pressure the pattern of the integral maps remains similar. However, significant increases occur in the integral amplitudes (dipole amplitude and mean absolute amplitudes of the map) in both the QRS and STT integral maps. Amplitude changes in the STT integrals maps are twice as large as that in the QRS integral maps. The LV pressure increase also prolongs the QT interval on the pseudo-ECG with 34 ms. Activation slightly prolongs in the LV, but prolongation of repolarization in the LV is considerably larger, although not as long as the prolongation of QT duration. The possible explanation for the latter remark will be discussed later.

Although activation in the LV was significantly prolonged, this was on average only 3 ms. In these experiments the diastolic LV pressure was elevated causing a continu-ous pressure change during the entire heart cycle. Stretch during diastole may cause a depolarization of the membrane,15 which may have caused slight changes in activation in our preparations. This also can explain the small increase in integral amplitudes of the QRS complex. The prolongation of repolarization was about three times larger and can therefore not be explained by the activation delay alone.

Accordingly, based on these data, we suggest that the increasing LV pressure causes prolongation of the repolarization via mechano-electric coupling in the LV and consequently leads to changes in the STT segments. The increased LV pressure induces stretch in the LV and may activate the stretch-activated channels which are involved in mechano-electric coupling.16 Alternatively, the balloon inflation itself causes a volume expansion of the heart reducing the distance between the ventricular walls and the electrodes on the bucket. This may have led to increased potential levels in the ECG and may therefore also have contributed directly to the increased amplitudes in QRS and STT integral maps. However, since repolarization was significantly prolonged, we suggest that the expansion of the heart due to balloon inflation plays a minor role in STT segment deviations. In addition, the absolute amplitude increase of the QRS integrals was smaller than that of the STT integrals.

Recently, we demonstrated that pressure in a normal heart contributes to synchroni-zation of the repolarization,13 which is in contrast with our results showing an increased interventricular dispersion in repolarization. The discrepancy may be explained by methodological differences. Firstly, the earlier study examined repolarization only in the LV, whereas also the RV was included in the present study. Moreover, in the earlier study only the systolic pressure was increased without changing the diastolic pressure and therefore inducing changes only during systole. Also, in our study systolic pressures were about twice as low.

Coulshed et al17 were able to examine action potential changes by separately control-ling the diastolic and systolic pressure and showed that diastolic pressure increases lead

64 Chapter 4

to action potential prolongation, which is consistent with our results. However, they also demonstrated that an increase in systolic pressure shortened the action potential (AP) duration, which just partly fit the results from our earlier report that showed also prolongation of the AP duration depending on the cycle length and timing of the pressure pulse.13 However, it was already suggested before in a review by Taggart and Sutton that the effects of mechano-electric coupling depend on timing, intensity and type of stretch.12 We speculate that an increase in systolic pressure would synchronize repolarization by prolonging and/or shortening of repolarization, whereas increasing diastolic pressures lead to prolongation of the repolarization. Moreover, Sedova et al18 has shown that load-induced changes in ventricular repolarization can be influenced by autonomic modulation. Therefore, our suggestions on the differential effects of systolic and diastolic pressure increase, may only apply to conditions without large autonomic modulations, which were absent in our isolated heart experiments.

Compared to the amplitude increases in STT integrals and the prolongation of the QT intervals, the observed prolongation of repolarization times was relatively small. Figure 5 shows three potential mechanisms of ST changes based on small RT changes. The first two situations illustrate the small shift in RT by action potential prolongation both with a steep repolarization phase. In the first situation heterogeneous and small pressure-induced RT prolongations without changing the AP morphology lead to a relatively

A B C

Figure 5: Illustrations of proposed hypotheses explaining STT segment changes on the ECG caused by ac-tion potential (AP) changes. The black lines (solid and striped) show baseline situation with short APs from the right ventricle (RV) and longer APs from the left ventricle (LV) resulting in a normal ECG. The red striped lines show the changes in AP in the LV and in the ECG. A: Small repolarization time (RT) changes, resulting in a large T wave amplitude with only slight QT prolongation. b: Small RT changes resulting in a large T wave amplitude with large QT prolongation. C: Small RT changes resulting in a small T wave amplitude with little QT prolongation.


4

large voltage gradient within the heart. This fits with the large changes in STT integral amplitudes, but the QT interval is slightly prolonged. In the second case – besides a het-erogeneous and small pressure-induced RT prolongations – AP morphology is changed at the end of repolarization. This will lead to larger changes in STT integral amplitudes and a large QT prolongation. The last situation supposes at baseline an AP morphology with a slower repolarization phase. Small pressure-induced RT prolongations will lead in this case to a much smaller voltage gradient per time unit, but will finally sum up to similar STT integral amplitudes as in the first case. Also, a rather small QT prolongation will be present. Thus, we hypothesize that the increased LV pressure might have led to a terminal AP morphology change involving only minor RT changes (Figure 5b). This is consistent with the AP prolongation resembling an afterdepolarization described by Franz et al.19

CONCLuSION

In summary, the increase in LV pressure causes prolongation of repolarization in the LV and not in the RV and consequently results in increasing amplitudes of the STT integrals. Although amplitudes of QRS integrals also increase, STT changes were more prominent. The effect in amplitude of the STT integral relates to the prolongation of the RTs, which may be the effect of mechano-electric feedback in the heart. Therefore, we conclude that changes in cardiac pressures lead to changes in the T wave morphology.

66 Chapter 4

REFERENCES

1. Han J, Moe GK. Nonuniform Recovery of Excitability in Ventricular Muscle. Circ Res 1964; 14: 44–60.

2. Surawicz B. Ventricular fibrillation and dispersion of repolarization. J Cardiovasc Electrophysiol 1997; 8: 1009–1012.

3. Chauhan VS, Downar E, Nanthakumar K, Parker JD, Ross HJ, Chan W, Picton P. Increased ventricular repolarization heterogeneity in patients with ventricular arrhythmia vulnerability and cardiomy-opathy: a human in vivo study. Am J Physiol Heart Circ Physiol 2006; 290: H79–H86.


5. Moréna H, Janse MJ, Fiolet JW, Krieger WJ, Crijns H, Durrer D. Comparison of the effects of regional ischemia, hypoxia, hyperkalemia, and acidosis on intracellular and extracellular potentials and metabolism in the isolated porcine heart. Circ Res 1980; 46: 634–646.

6. Conrath CE, Opthof T. Ventricular repolarization: an overview of (patho)physiology, sympathetic effects and genetic aspects. Prog Biophys Mol Biol 2006; 92: 269–307.

7. Curran ME, Splawski I, Timothy KW, Vincent GM, Green ED, Keating MT. A molecular basis for cardiac arrhythmia: HERG mutations cause long QT syndrome. Cell 1995; 80: 795–803.


9. Coronel R, Wilders R, Verkerk AO, Wiegerinck RF, Benoist D, Bernus O. Electrophysiological changes in heart failure and their implications for arrhythmogenesis. Biochim Biophys Acta 2013; 1832: 2432–2441.

10. Jin H, Lyon AR, Akar FG. Arrhythmia mechanisms in the failing heart. Pacing Clin Electrophysiol 2008; 31: 1048–1056.

11. Shub C. Heart failure and abnormal ventricular function. Pathophysiology and clinical correlation (Part 1). Chest 1989; 96: 636–640.


13. Opthof T, Meijborg VMF, Belterman CNW, Coronel R. Synchronization of repolarization by mechano-electrical coupling in the porcine heart. Cardiovasc Res 2015.

14. Meijborg VMF, Conrath CE, Opthof T, Belterman CNW, De Bakker JMT, Coronel R. Electrocardio-graphic T wave and its relation with ventricular repolarization along major anatomical axes. Circ Arrhythmia Electrophysiol 2014; 7: 524–531.

15. Lab MJ. Contraction-excitation feedback in myocardium. Physiological basis and clinical rel-evance. Circ Res 1982; 50: 757–766.

16. Craelius W, Chen V, el-Sherif N. Stretch activated ion channels in ventricular myocytes. Biosci Rep 1988; 8: 407–414.

17. Coulshed DS, Cowan JC, Drinkhill MJ, Hainsworth R. The effects of ventricular end-diastolic and systolic pressures on action potential and duration in anaesthetized dogs. J Physiol 1992; 457: 75–91.


4

18. Sedova K a., Goshka SL, Vityazev V a., Shmakov DN, Azarov JE. Load-induced changes in ventricu-lar repolarization: evidence of autonomic modulation. Can J Physiol Pharmacol 2011; 89: 935–944.


Chapter 5

Regional conduction slowing induces inferolateral J-waves

Veronique M.F. Meijborg *, Mark Potse *, Chantal E. Conrath, Charly N.W. Belterman, Jacques M.T. de Bakker, Ruben Coronel,

* Both authors contributed equally


70 Chapter 5

AbSTRACT

background: J-waves in inferolateral leads are associated with a higher risk for arrhyth-mic death. We aimed to test potential mechanisms (depolarization or repolarization dependent) responsible for inferolateral J-waves.

Objective: We hypothesized that inferolateral J-waves are caused by delayed activation of myocardium that is activated late during normal conditions.

Methods: Computer simulations were performed to evaluate the effect on J-point eleva-tion of reduced sodium current conductivity (GNa), increased transient outward current conductivity (Gto) and cellular uncoupling in three predefined ventricular regions (lateral, anterior or septal). Two pig hearts were Langendorff-perfused with selective perfusion of lateral or anterior/septal regions. Volume-conducted pseudo-electrocardiograms (ECG) were recorded to detect the presence of J-waves. Epicardial unipolar electrograms were simultaneously recorded to obtain activation times (AT).

Results: Simulation data showed that conduction slowing in lateral, but not in other regions, induced inferolateral J-waves. An increase in transient outward potassium current or cellular uncoupling in the lateral zone did not induce J-waves, but elicited slight J-point elevations. Additional conduction slowing in the entire heart attenuated J-waves on the ECG. Experimental data confirmed that conduction slowing in the left lateral but not in the anterior/septal ventricular region induced inferolateral J-waves. J-waves coincided with the delayed activation.

Conclusion: Conduction slowing in the left lateral ventricular myocardium causes inferolateral J-waves on the ECG.

Conduction slowing induces J-waves 71

5

INTRODuCTION

J-waves in inferolateral leads of the surface electrocardiogram (ECG) – or early repolariza-tion (ER) pattern – are characterized as an elevation of the QRS-ST junction manifested as a notch or slur.1,2 The J-wave was considered a benign phenomenon3 until Haïssaguerre et al4 demonstrated an increased prevalence of J-waves in patients with idiopathic ven-tricular fibrillation. This association was confirmed by a meta-analysis of later studies.5

The mechanism underlying the ER pattern – or J-wave – is subject of an ongoing debate.6,7 Yan and colleagues8 proposed a cellular mechanism for J-waves based on ex-periments performed in canine arterially perfused ventricular wedge preparations. They postulated that J-waves are generated by a transmural voltage gradient resulting from a more prominent transient outward potassium current (Ito) in the subepicardium, leading to a spike-and-dome action potential (AP) morphology compared with endocardium. An alternative mechanism is based on regional conduction slowing. Late potentials at the terminal QRS complex in the ECG are related to delayed activation.9 The analogy between inferolateral J-waves and the ST segment elevations in the Brugada Syndrome (BrS) causes some investigators to posit a common mechanism for the two syndromes.10 However, sodium channel blockers cause a differential effect, as these are used to provoke ST segment elevations in the right precordial leads in BrS, but attenuate in-ferolateral J-waves.11 Also, the different location of the J-wave or ST segment elevations – right precordial12 or inferolateral leads4 – indicates involvement of different regions. The inferolateral location of J-waves suggests a substrate in the inferolateral area of the heart, which corresponds to a late activated region in the heart.13 We surmise that when this area undergoes additional conduction slowing the delayed action potential will generate a voltage gradient just enough to give rise to a J-wave in the inferolateral leads.

The aim of this study was to test 1) whether delayed depolarization and/or early repolarization can cause J-waves, 2) whether left lateral involvement is essential for J-wave appearance in inferolateral leads, and 3) a mechanism by which sodium channel blockers can reduce J-waves. For these purposes we used a computational approach. We employed a pig model to replicate the computational findings on regional sodium channel blockade. The repolarization hypothesis could not be tested in a pig model, because pig hearts lack Ito.14

METHODS

Computer simulations

A detailed description on the computer model is provided in the Data Supplement. Computer simulations were performed to test the possible contribution of three dif-

72 Chapter 5

ferent electrical properties in the genesis of J-waves. Within the modeled heart 3 areas were defi ned: lateral zone, anterior zone and septal zone (Figure 1A). Within each area we simulated the following interventions and evaluated their eff ects on the ECG. By reducing the sodium current conductivity (GNa) to 12.5% of baseline condition we tested the depolarization hypothesis, whereas by increasing the transient outward potassium current (Gto) 10-fold we tested the repolarization hypothesis. As an alternative test for the depolarization hypothesis we simulated diff use fi brosis with consequent conduction slowing by reducing the intracellular and extracellular conductivity to 12.5% of baseline condition (cellular uncoupling). On top of the simulations that induced the most and

left lateral view superior posterolateral view

canula inthe OM

epicardial electrogram

pseudo-ECG

200 ms

12 m

V2 m

V

200 ms

RT

AT

QRSbeginJ-wave onset

lateralanterior

septum

lateral

anterior

septum

B

A

QRSend

Figure 1: A: The three areas in which electrical properties were varied: lateral, anterior and septal zone. b: Schematic of experimental setup. The bucket wall contained 61 electrodes (grey dots). QRSbegin and QRSend mark the maximum QRS duration determined using all leads and may deviate from the QRS dura-tion in a single lead. The 11x11 electrode grid (green) overly the cannulated OM (red shaded area: selective-ly perfused tissue). See Supplemental Figure S1 for electrode confi guration of the 9x12 grid. AT= activation time. RT= repolarization time.


5

largest J-waves or J-point elevations, we tested the effects of sodium channels blockers by reducing GNa outside the affected zone to 40% of baseline.

In the calculated 12-lead ECGs we determined the total QRS duration (first QRS onset in any lead to last QRS end in any lead), J-point amplitude, and presence of J-point eleva-tions and J-waves. The J-point was defined as first deviation in the terminal limb of the QRS complex initiating a notch or slur. The J-point amplitude was determined at the top of the notch or at the point where slurring started. A J-point elevation was defined as a J-point amplitude of 0.05 mV or more in an inferolateral lead (I, II, III, aVF, aVL and V4-V6). A J-wave was a J-point elevation (notch or slur) of 0.1 mV or more. Difference ECGs were obtained by subtracting baseline ECGs from intervention ECGs.

Experimental setup

The experimental protocol was approved by the local Animal Experiments Committee (Academic Medical Center, University of Amsterdam) and carried out in accordance with national and institutional guidelines.

Pigs (n=2, male, 50-60 kg) were premedicated, intubated and ventilated. The heart was excised and perfused according to Langendorff with a (circa 1:1) blood-Tyrode’s mixture (pH=7.35 –7.45). The left obtuse marginal coronary artery (OM) or left anterior descending coronary artery (LAD) was separately cannulated. See Data Supplement for more details. The cannula was connected to a separate temperature-controlled perfu-sion system with the same blood-Tyrode’s mixture and with a side branch for the infusion of flecainide (Tambocor, 3M Nederland, Zoeterwoude, The Netherlands). After baseline electrophysiological recordings, we administered flecainide to the OM (60 µM) or LAD (6 µM). The high flecainide concentration was associated with spontaneous arrhythmias and therefore we lowered the concentration and awaited a similar activation delay in the myocardial regions. After baseline electrophysiologic recordings, we administered flecainide to the OM/LAD perfusion for, respectively, 2 and 5 minutes and obtained recordings.

Electrophysiologic recordings

The left atrium was paced at a cycle length of 450 ms. An 11x11 electrode grid (OM perfusion) or a 9x12 electrode grid (LAD perfusion) was fixed to the epicardial surface overlapping the entire LV (and with the 9x12 grid also the anterior RV) to obtain local unipolar epicardial electrograms. Supplemental Methods and Supplemental Figure S1 provide details on the electrode configurations. The Langendorff-perfused heart was submerged in a perfusion-fluid filled bucket, containing 61 electrodes to obtain pseudo electrocardiograms (pseudo-ECGs, Figure 1B). The reference signal was the average of all electrodes.

74 Chapter 5

In the pseudo-ECGs we determined the maximum QT interval and maximum QRS duration including the J-wave. Definitions for J-point and J-point amplitude were similar as described above. J-point elevations ≥ 0.2 mV were denoted as J-waves. Because the pseudo-ECG amplitudes were about twice that of real ECGs we adjusted the J-wave criteria accordingly. J-wave onset was defined as time of J-point – initiating a J-wave – relative to QRS onset. The inferolateral leads constituted the 6 columns of electrodes opposite the LV area (n=18) and 1 bottom electrode (grey boxes in Figure 5). J-wave interval was defined as J-wave onset to end of QRS. In the difference ECG – baseline ECG minus flecainide ECG – we determined the moment and amplitude of maximum peak difference (positive/negative). In each local electrogram we determined activation and repolarization times (ATs and RTs) as before (Figure 1B).15 Difference AT maps were calculated as baseline AT map minus flecainide AT map. Recordings with ST segment elevation or a flat T wave were excluded from analysis of RTs. Signal analysis was per-formed offline using software16 based on Matlab (The MathWorks, Inc., Natick, MA, USA).

Statistics

Continuous variables were presented as mean±SD if normally distributed and as median (25th-75th percentile) if not normally distributed. Differences in continuous variables between both hearts or between groups of leads were evaluated with an independent sample t test if normally distributed, and with a Mann-Whitney U test if not normally dis-tributed. Categorical variables were presented as frequencies and differences between hearts or between groups of leads were evaluated with a Fisher’s Exact test. P values ≤0.05 were considered statistically significant.

RESuLTS

Simulations

Table 1 summarizes the simulation data on QRS durations, J-wave occurrence, maximum activation time and activation delay for each zone of the various simulations. The ulti-mate activation time in the anterior and lateral zones were later than in the septal zone.

Simulations of regional GNa reduction

GNa reduction to 12.5% in the affected region caused a delay of the AP without affecting the AP morphology (Supplemental Figure S2A). Figure 2 shows 6 ECG leads at baseline and after GNa reduction in each zone.

GNa reduction in the lateral zone led to J-waves in the inferolateral leads (II, III, aVF and V6) with a maximum amplitude of 0.14 mV in lead II (Table 1). The extremity leads showed J-wave notches and lead V6 a J-wave slur. GNa reduction in the anterior zone


5

led to a notching J-wave in lead I only and to J-point depressions in the inferior leads (II,III,aVF). GNa reduction in the septal zone did not induce J-waves.

Simulations of regional Gto increase

A 10-fold Gto increase caused a deeper AP notch at the epicardium of the affected region, without influencing the endocardial AP notch (Supplemental Figure S2B). Figure 3 shows the ECG results of a 10-fold Gto increase in each zone. In all simulations of Gto increase J-waves were absent. There were minor J-point elevations, although in fewer leads and with lower amplitudes compared to GNa reduction (Table 1). Overall, Gto increase did prolong the QRS duration, but did not delay activation anywhere.

Simulations of regional cellular uncoupling

Cellular uncoupling in the affected region caused a delay of the AP without affecting the AP morphology (Supplemental Figure S2C). Figure 4 shows the ECG results of cellular uncoupling in each zone. By reducing intercellular coupling in the lateral or anterior region small J-point notching was induced, but no J-waves appeared. The difference

Table 1: Simulation data.

QRSms

AT maxms

AT delayms

J-wave Difference ECG

AmaxmV

# leadsJ≥ 0.1mV (J≥ 0.05mV)

ApeakmV

Tpeakms

baseline 81 - - 0 0 (0)

Lateral 61

Anterior 64

Septal 49

GNa reduction

Lateral 94 93 14 0.14 4 (5) 0.40 54

Anterior 88 88 13 0.10 1 (1) 0.35 58

Septal 81 65 11 0 0 (0) 0.33 38

Gto increase

Lateral 99 61 0 0.07 0 (4) 0.13 57

Anterior 113 64 0 0.09 0 (3) −0.15 61

Septal 99 51 1 0 0 (0) 0.21 47

uncoupling

Lateral 96 103 15 0.06 0 (2) −0.27 54

Anterior 94 98 14 0.09 0 (2) 0.20 51

Septal 82 83 0 0 0 (0) 0.23 45

QRS= QRS duration; AT max= maximum activation time in affected zone; AT delay= activation time delay in affected zone; A max= maximum J-point amplitude; # leads= number of leads with J-waves; Apeak= ampli-tude of maximum peak in difference ECG; Tpeak= timing of maximum peak in difference ECG.

76 Chapter 5

ECGs were of intermediate amplitude compared to GNa reduction and Gto increase. Tim-ing of the difference ECG peak was similar as with GNa reduction and Gto increase. Cellular uncoupling in the lateral and anterior region induced activation delays of about 15 ms and caused maximum activation in the affected zone that corresponds with end of QRS (Table 1).

Overall GNa reduction on top of J-waves and J-point elevations

Sodium channel blockers can attenuate inferolateral J-waves.11 Therefore, we simulated an overall GNa reduction on top of each intervention in the lateral zone (last column in Figure 2, 3 and 4). In the 3 simulations with overall GNa reduction, pre-existing J-waves or J-point elevations shrunk or disappeared and QRS duration increased with 27, 22 and 17 ms (GNa reduction, Gto increase and cellular uncoupling, respectively). Overall GNa reduc-tion delayed activation in all zones with latest activation occurring outside the 3 zones

Zone 1: lateral

Baseline ECG versus ECG after GNa reduction (12.5% of baseline) and difference ECG

Zone 2: anterior Zone 3: septal

I

II

III

V2

V4

V6

I

II

III

V2

V4

V6

200 ms

1 m

V

I

II

III

V2

V4

V6

*

* *

*

GNa reduction (12.5%)in zone 1: lateral

+ overall GNa reduction

200 ms

1 m

V

II

III

V4

V6

I V2

BA DC

Figure 2: ECG at baseline (blue) and resulting from GNa reduction to 12.5% of baseline (red) in 3 zones; A: lateral; b: anterior; C: septum. Grey: Difference ECG (GNa reduction minus baseline). Asterisks= J-waves. Boxed panel (D): additional overall GNa reduction (blue) causes disappearance of J-waves.


5

(i.e. 116 ms in the GNa reduction and Gto increase interventions) or in the lateral zone (i.e. 123 ms in the cellular uncoupling intervention).

Experiments

Figure 5 shows the unfolded bucket with electrodes and some examples of pseudo-ECGs at baseline and during OM flecainide infusion. In this heart, J-waves appeared on the inferolateral pseudo-ECG leads. One J-wave was observed just outside this area, with reciprocal J-point depressions in the other leads (Figure 5: pseudo-ECG at C). At baseline J-waves were present in 4 leads. After flecainide infusion the number of J-waves in inferolateral leads increased significantly, while in the other leads 2 of 3 J-waves disappeared (Table 2: OM perfusion). The amplitude of the difference ECG was significantly larger and positive in inferolateral leads compared to the other leads. After flecainide infusion the QRS duration was increased by 26 ms due to arising J-waves.

Baseline ECG versus ECG after Gto increase (10-fold of baseline)and difference ECG

Zone 1: lateral Zone 2: anterior Zone 3: septal

200 ms

1 m

V

I

II

III

V2

V4

V6

I

II

III

V2

V4

V6

I

II

III

V2

V4

V6

II

III

V4

V6

I V2

Gto increase (10 fold)in zone 1: lateral


200 ms

1 m

V

BA DC

Figure 3: ECG at baseline (blue) and resulting from a 10-fold Gto increase (red). No J-waves emerge after Gto increase. Organization as in Figure 2.

78 Chapter 5

In the heart with LAD perfusion (Figure 5 and Table 2), some J-waves were present at baseline, albeit mainly in the most superior leads. After flecainide infusion all J-waves disappeared in inferolateral leads and arose in 2 leads near the RV posterior wall. The amplitude of the difference ECG was negative and significantly larger in inferolateral leads than in the other leads. (Table 2: LAD perfusion) The peak of the difference ECG also occurred earlier in LAD perfusion (15 (12 – 27) ms) compared to OM perfusion (35 (28 – 41) ms, Mann-Whitney U test P<0.01).

Activation maps

Figure 6 demonstrates the activation maps at baseline and during flecainide infusion in the OM. At baseline, the OM region was activated latest (right side of the AT map) and showed the largest conduction delay after flecainide infusion. The J-wave interval overlaps this region of delayed activation (dotted surface in flecainide AT map) and even

Baseline ECG versus ECG after uncoupling (12.5% of baseline)and difference ECG

Zone 1: lateral Zone 2: anterior Zone 3: septal

200 ms

1 m

V

I

II

III

V2

V4

V6

I

II

III

V2

V4

V6

I

II

III

V2

V4

V6

II

III

V4

V6

I V2

Intercellular uncouplingin zone 1: lateral


200 ms

1 m

V

BA DC

Figure 4: ECG at baseline (blue) and resulting from cellular uncoupling to 12.5% of baseline (red). No J-waves emerge after uncoupling. Organization as in Figure 2.


5

Table 2: ECG characteristics at baseline and during fl ecainide.

OM perfusion LAD perfusion

Baseline Flecainide P Baseline Flecainide P

Total QRS, ms 87 113 77 76

Inferolateral leads, n=19

J-wave # leads, N/n 1/19ns 10/19* <0.05 4/19* 0/19ns 0.11

J-wave onset, ms 62 42 45 -

A_dECG, mV 0.22 (0.14 – 0.36)† −0.25 (−0.35 – −0.20)†

Other leads, n=42

J-wave # leads, N/n 3/42 1/42 0.62 1/42 2/42 1.00

J-wave onset, ms 55 55 48 34

A_dECG, mV −0.11 (−0.17 – −0.05) 0.17 (−0.16 – 0.23)

QT interval, ms 294 300 273 269

QRS= QRS duration; J-wave # leads= number of leads with J-wave (N) out of number of selected leads (n); J-wave onset= fi rst onset of J-wave; A_dECG= maximum or minimum amplitude of diff erence ECG (median [25th – 75th percentile]).P = P values for Fisher’s Exact test between baseline and fl ecainide.* P< 0.05 for Fisher’s Exact test between inferolateral & other.† P< 0.01 for Mann-Whitney U test between inferolateral & other.ns = not signifi cant.

500 ms

5 m

V

anterior LV posterior anteriorRV

base

apex

A

E

B

C

D

500 ms

5 m

V

anterior LV posterior anteriorRV

base

apex

A

E

B

C

D

BA

Figure 5: Unfolded bucket (middle boxes) with electrodes (black dots) and heart position. Shaded boxes indicate the inferolateral area. A: OM perfusion. b: LAD perfusion. Stars indicate leads showing J-waves. Surrounding are examples of pseudo-ECGs (blue= baseline, red= during fl ecainide). Green= diff erence ECG. A and E correspond with the pseudo-ECGs shown in Figure 6 and Figure 7. Flecainide infusion induced or exacerbated inferolateral J-waves in the OM perfusion but not in the LAD perfusion.

80 Chapter 5

outreaches the AT map. We quantifi ed the activation delay by selecting a column of elec-trodes inside and outside the OM region (white striped boxes) and summarized the data (Table 3). After fl ecainide infusion the delay in AT was larger in the OM region than the rest of the LV. In the non-perfused regions no signifi cant changes occurred, as expected.

5 c

m

5 cm

difference ECG

500 ms

2m

V

baseline

flecainide

= J-wave

Pseudo-ECG Epicardial EGAT map

*b

200 ms

12 m

V

2

1

2

1

0

50

100ms

2

1

*a

*b

*aA

E

A

E

A

E

0

25

50ms

difference AT map

= J-wave

0

50

100ms

2

1

200 ms

12 m

V

Figure 6: OM perfusion. Two pseudo-ECGs (left column); activation (AT) maps with two epicardial elec-trograms (EG, right column) at baseline (top panel) and after fl ecainide infusion (middle panel). AT map: thick black lines indicate isochrones of J-waves (*a and *b). Dotted area represents overlap with J-wave interval, which outreaches the AT map. Numbers 1 (‘perfused’ region) and 2 (‘not perfused’ region) record-ing sites of the corresponding EGs. Pseudo-ECG: arrows indicate the moment of last activation in the AT map. A and E correspond with pseudo-ECGs in Figure 5. bottom panel: shows the diff erence pseudo-ECGs and diff erence AT map. Dashed white ellipses indicate a region within and outside the area perfused with fl ecainide used to determine mean ATs. Note that the eff ect of fl ecainide extended beyond the QRS dura-tion before application and that the region subjected to conduction slowing (right side of AT map) was late activated before application of fl ecainide.


5

Figure 7 shows the activation maps at baseline and during fl ecainide infusion in the LAD region. This is an extended map with two-third of the map overlying the LV and one-third on the left overlying the RV anterior wall. At baseline, the J-wave interval overlaps that of late activation in the RV (dotted surface) and even outreaches the AT map. After 5 minutes of fl ecainide infusion, conduction was delayed in the LAD region and inferolateral J-waves were no longer present. Table 3 demonstrates that within the perfused region (lower left white striped box in Figure 7) activation at baseline was ear-lier in the activation sequence compared to the region outside the selectively perfused area (right upper box in Figure 7). After fl ecainide infusion activation was latest in the perfused region.

0

50

100ms

-10

0

25ms

0

50

100ms

500 ms

2m

V difference ECG

baseline

flecainide

*1 = J-wave

A

E

A

E

A

E

7 c

m

18 cm

*1

Pseudo-ECG AT map

difference AT map

10 cm

Figure 7: LAD perfusion. Description is similar to Figure 6. Note that the eff ect of fl ecainide was limited to the QRS duration before application and that the region subjected to conduction slowing was relatively early activated before application of fl ecainide.

82 Chapter 5

DISCuSSION

Our results show that J-waves can be induced in the inferolateral leads of the ECG as a result of regional conduction slowing only in the lateral region, but not in the anterior or septal region of the heart. A regional increase in transient outward potassium current or cellular uncoupling were less effective in inducing J-waves, irrespective of the region. Additionally, an overall sodium blockade attenuates J-waves on the ECG by masking the J-waves in the prolonged QRS.

The experimental data support the simulation data by showing that regional con-duction slowing resulting from sodium channel blockade in the lateral but not in the anterior/septal region, induces J-waves. The regional conduction slowing in the lateral zone causes local activation to occur beyond the duration of the baseline QRS complex. As a consequence, the J-wave interval coincides largely with the region of conduction delay (Figure 6).

J-wave mechanism

The mechanism of the inferolateral J-waves is a debated issue.6,7 The two prevailing hypotheses for inferolateral J-waves are focused either on depolarization17 or repolariza-tion, similar to Brugada Syndrome.8 We have shown that the amplitude of the J-point elevation is largest with GNa reduction compared to Gto increase and cellular uncoupling. Only GNa reduction in the lateral and anterior region caused J-point elevations that met the criteria of J-waves, and supports the depolarization hypothesis. Cellular uncoupling induces smaller J-point elevations (no J-waves). Since activation delays in GNa reduction

Table 3: Data of local electrograms.

Perfused Not perfused

Baseline Flecainide Baseline Flecainide

OM perfusion

No. electrodes 10 10 10 10

AT, ms 39±3 67±9 31±10 34±11

AT delay, ms - 28± 7 - 3±3

RT, ms 236±7 239±5 248±5 247±4

LAD perfusion

No. electrodes 8 7 8 8

AT, ms 26±10 42±16 31±5 28±5

AT delay, ms - 16±6 - −3±1

RT, ms 214±4 234±15 220±7 209±9

Values are given as mean±SD.‘Perfused’ and ‘not perfused’ refers to region that is or is not selectively perfused.AT= activation time; RT= repolarization time


5

and cellular uncoupling were similar (about 15 ms), we argue that their difference in J-point elevations may be explained by the reduced tissue conductivity following cellular uncoupling. In an area of reduced conductivity the current generated by the activation wave front is smaller, resulting in a smaller potential field on the torso. Alternatively, J-waves may be caused by early repolarization of the action potential since the increase in Gto resulted in minor J-point elevations which fits with an expected current flow from endocardium towards epicardium during phase 1 repolarization. However, although the simulated 10-fold increase of Gto in this study is very large (see also Supplemental Figure S2) it induces J-point elevations that are only half the size and do not reach the critical J-wave level of 0.1 mV. Also difference ECGs had smaller amplitudes. The latter hypothesis is therefore the less likely but cannot be excluded based on our data.

The amplitude of the difference ECG is largest with GNa reduction followed by cel-lular uncoupling and then Gto increase. Also, the difference ECG is mainly larger with intervention in the anterior region than in lateral region and smallest in septal region. However, the amplitude of the difference ECG is not directly reflected in the amplitude of a J-point elevation. This implies that the timing and spread of the difference ECG con-tributes importantly to the morphology of the change in QRS. Enhanced Gto causes more QRS widening compared to a decreased GNa or cellular uncoupling (Table 1), whereas no activation delay occurs. It indicates that early repolarization does contribute to QRS duration.

Our experimental and simulation data suggest that conduction delay in the region that is already late in the activation sequence at baseline (lateral region) predominantly contributes to the genesis of inferolateral J-waves, whereas conduction delay in earlier activating regions (anterior and/or septum) show minor or no J-point elevations and even J-point depression. We observed that although the AT delay in LAD perfusion was smaller than in OM perfusion it led to J-point depression rather than J-point elevations. Recently, it has been shown in a case report that J-waves in the inferior ECG leads can result from delayed activation in the basal inferior LV region, without shortening of the action potential duration.18 This confirms that the region of abnormality is important.

Remarkably, all different interventions – GNa reduction, Gto increase and cellular un-coupling – within the lateral region show similar results in polarity of J-point elevations and difference ECGs, although only GNa reduction led to J-waves. All interventions in the anterior region, led to negative difference ECGs in the inferior leads and J-point eleva-tion (no J-waves) were only present in lateral ECG leads. The reason that the effects of interventions are smaller in the anterior and septal zones – despite similar amplitudes of difference ECGs among zones per intervention – is therefore related to the extent of masking by the QRS complex.

84 Chapter 5

Modulation of J-waves

J-wave amplitude is modulated by heart rate,19 autonomic tone17 and drugs.11,20 It has been demonstrated in patients that sodium channel blockers like ajmaline, pilsicainide and flecainide attenuate J-waves in the inferolateral leads11,21,22 and broaden the QRS.11 Our simulation data explain these findings by demonstrating that overall GNa reduction widens the QRS complex and consequently mask the pre-existing inferolateral J-waves.

In patients with idiopathic ventricular fibrillation it has been shown that inferolateral J-waves augment after a pause and diminish at higher heart rates.4,21 It has been sup-posed that the attenuation of J-waves at increasing heart rate results from reduced tran-sient outward potassium current due to the relatively slow recovery from inactivation.23 We did not study the effect of heart rate increase, but we surmise that conduction may also play a role in this phenomenon. At high heart rates less sodium current is available, especially in diseased hearts.24 It may therefore have a similar effect as administration of ajmaline and we would expect a concomitant QRS prolongation.

CONCLuSION

Conduction slowing in the lateral region of the heart causes inferolateral J-waves on the ECG. The interval of J-waves coincides with the activation time of the region of delayed activation. The cardiac tissue in which J-waves are induced is characterized by a relatively late activation in the normal heart. Global conduction slowing attenuates J-waves due to masking by the prolonged QRS complex. Enhanced transient outward potassium current and cellular uncoupling have minor potency to elicit inferolateral J-waves. Although our study cannot exclude the role of repolarization abnormality, it predominantly affirms the depolarization hypothesis especially when tissue conduc-tivity is preserved. Our study also provides an explanation for J-wave attenuation by sodium channel inhibition.


5

REFERENCES

1. Sacher F, Lim HS, Haïssaguerre M. Sudden cardiac death associated with J wave elevation in the inferolateral leads: insights from a multicenter registry. J Electrocardiol 2013; 46: 456–460.

2. Mizusawa Y, Bezzina CR. Early repolarization pattern: its ECG characteristics, arrhythmogeneity and heritability. J Interv Card Electrophysiol 2014; 39: 185–192.

3. Shipley RA, Hallaran WR. The four-lead electrocardiogram in two hundred normal men and women. Am Heart J 1935; 11: 325–345.

4. Haïssaguerre M, Derval N, Sacher F, et al. Sudden cardiac arrest associated with early repolariza-tion. N Engl J Med 2008; 358: 2016–2023.

5. Wu SH, Lin XX, Cheng YJ, Qiang CC, Zhang J. Early repolarization pattern and risk for arrhythmia death: a meta-analysis. J Am Coll Cardiol 2013; 61: 645–650.

6. Hoogendijk MG, Potse M, Coronel R. Critical appraisal of the mechanism underlying J waves. J Electrocardiol 2013; 46: 390–394.

7. Wellens HJ. Early repolarization revisited. N Engl J Med 2008; 358: 2063–2065.

8. Yan GX, Antzelevitch C. Cellular Basis for the Electrocardiographic J Wave. Circulation 1996; 93: 372–379.

9. Simson MB, Untereker WJ, Spielman SR, Horowitz LN, Marcus NH, Falcone R a, Harken a H, Joseph-son ME. Relation between late potentials on the body surface and directly recorded fragmented electrograms in patients with ventricular tachycardia. Am J Cardiol 1983; 51: 105–112.

10. Antzelevitch C, Yan G-X. J wave syndromes. Heart Rhythm 2010; 7: 549–558.

11. Roten L, Derval N, Sacher F, et al. Ajmaline attenuates electrocardiogram characteristics of infero-lateral early repolarization. Heart Rhythm 2012; 9: 232–239.

12. Antzelevitch C, Brugada P, Borggrefe M, et al. Brugada syndrome: report of the second consensus conference. Hear Rhythm 2005; 2: 429–440.


14. Li GR, Du XL, Siow YL, Karmin O, Tse HF, Lau CP. Calcium-activated transient outward chloride current and phase 1 repolarization of swine ventricular action potential. Cardiovasc Res 2003; 58: 89–98.



17. Abe A, Ikeda T, Tsukada T, Ishiguro H, Miwa Y, Miyakoshi M, Mera H, Yusu S, Yoshino H. Circadian variation of late potentials in idiopathic ventricular fibrillation associated with J waves: insights into alternative pathophysiology and risk stratification. Heart Rhythm 2010; 7: 675–682.

18. Park C, Lehrmann H, Jadidi AS, Schiebeling-Römer J, Allgeier J, Arentz T, Weber R. Abstract PO04-209: Is this really a J-wave syndrome? Hear Rhythm 2014; 11: S411.

86 Chapter 5

19. Aizawa Y, Sato A, Watanabe H, et al. Dynamicity of the J-wave in idiopathic ventricular fibrillation with a special reference to pause-dependent augmentation of the J-wave. J Am Coll Cardiol 2012; 59: 1948–1953.

20. Haïssaguerre M, Sacher F, Nogami A, et al. Characteristics of Recurrent Ventricular Fibrillation Associated With Inferolateral Early Repolarization. Role of Drug Therapy. J Am Coll Cardiol 2009; 53: 612–619.

21. Nakagawa K, Nagase S, Morita H, Ito H. Left ventricular epicardial electrogram recordings in idiopathic ventricular fibrillation with inferior and lateral early repolarization. Heart Rhythm 2014; 11: 314–317.

22. Kawata H, Noda T, Yamada Y, Okamura H, Satomi K, Aiba T, Takaki H, Aihara N, Isobe M, Kamakura S, Shimizu W. Effect of sodium-channel blockade on early repolarization in inferior/lateral leads in patients with idiopathic ventricular fibrillation and Brugada syndrome. Heart Rhythm 2012; 9: 77–83.

23. Koncz I, Gurabi Z, Patocskai B, Panama BK, Szél T, Hu D, Barajas-Martínez H, Antzelevitch C. Mecha-nisms underlying the development of the electrocardiographic and arrhythmic manifestations of early repolarization syndrome. J Mol Cell Cardiol 2014; 68: 20–28.

24. Kawara T, Derksen R, de Groot JR, Coronel R, Tasseron S, Linnenbank a C, Hauer RN, Kirkels H, Janse MJ, de Bakker JM. Activation delay after premature stimulation in chronically diseased human myocardium relates to the architecture of interstitial fibrosis. Circulation 2001; 104: 3069–3075.


5

DATA SuPPLEMENT


Computational modelA whole-heart monodomain reaction-diffusion model at 0.2 mm spatial resolution (dx) and 10 to 50 µs temporal resolution (dt) was used to simulate propagating action potentials throughout the cardiac tissue.1 This model relates the transmembrane ionic current to the membrane potential on a macroscopic scale. Ionic currents within the heart were computed using a membrane model of the human ventricular myocyte including the differential characteristics of subendocardial (50%), midmyocardial (30%), and subepicardial myocytes (20%).2

The model anatomy (heart and torso) was based on MRI data of a subject with normal cardiac anatomy. From these data a three-dimensional finite-difference mesh was cre-ated. The transmural fiber rotation was also incorporated in the model.1 The heart model was embedded in an inhomogeneous human torso model (dx= 1 mm), including lungs and intracavitary blood volumes. The surface ECG was computed (dt= 1 ms) by solving the potential field in the heart and torso resulting from the computed transmembrane current densities.3

Experimental setupPigs (n=2, male, weight 50-60 kg) were premedicated with ketamine (10-15 mg/kg, Nimatek, Eurovet Animal Health BV) and midazolam (1-2 mg/kg, Actavis, Iceland) intramuscularly, and anesthetized with 15 mg/kg pentobarbital (Nembutal, Ceva Santé Animale) intravenously. The animals were intubated and ventilated with room air and oxygen plus isoflurane (1:1 + 1.5%). After a midsternal thoracotomy, 5000 international units of heparin (Leo Pharma) were injected intravenously. We infused Tyrode’s solution intravenously and simultaneously collected 4 to 5 L diluted blood. The heart was excised and perfused according to Langendorff. The left obtuse marginal coronary artery (OM) or left anterior descending coronary artery (LAD) was dissected over a distance of 5 mm and a ligature was passed underneath it. A cannula was introduced via a small incision into the OM or LAD and fixed by tying the ligature. The cannula was connected to a separate perfusion system with the same blood-Tyrode’s mixture.

Electrophysiologic recordingsWe used 2 different electrode grids for recordings of the local unipolar epicardial elec-trograms recorded against a reference electrode at the aortic root. In the heart with the OM perfusion we used an 11x11 electrode grid with an interelectrode distance of 5 mm. In the heart with the LAD perfusion we used a 9x12 electrode grid with an interelectrode

88 Chapter 5

distance of 9 mm vertically, and of 9 – 17 mm horizontally. Supplemental Figure S1 shows a schematic view of the two electrode grid confi gurations.


Baseline simulationsAt baseline, earliest activation was in the septal zone (29±8 ms), followed by the lateral zone (40±8 ms) and 3 ms later the anterior zone (43±7 ms). Total activation time was 77 ms and latest activation occurred outside the 3 predefi ned zones.

Action potential morphologies after interventionsSupplemental Figure S2 shows the resulting action potential (AP) morphologies trans-murally in the aff ected zone and the epicardial AP morphology before and after inter-vention. GNa reduction to 12.5% in de aff ected zone led to a delay in activation without aff ecting the AP morphology (Supplemental Figure S2A). A 10-fold Gto increase caused in the aff ected region a deeper AP notch at the epicardium, without infl uencing the endocardial AP notch. The endocardial layer did have minor Ito and therefore lacks a clear notch preceding the plateau phase (Supplemental Figure S2B). Cellular uncoupling in the aff ected region caused a delay of the AP without aff ecting the AP morphology (Supplemental Figure S2C).

5 cm

5 cm

18 cm

7 cm

10 cm

LAD

OM

5 mm 5 mm9 mm

17 mm

9 m

m

11x11 electrode grid 9x12 electrode gridBA

Figure S1: Schematic view of the electrode grid confi guration in both Langendorff -perfused hearts. A: 11x11 electrode grid in heart with OM perfusion, and b: 9x12 electrode grid in heart with LAD perfusion. Red shaded area indicates the selectively perfused tissue and the red arrows indicate the position of the cannula.


5Simulation of regional GNa reductionAfter GNa reduction in the lateral zone the difference ECG (lateral GNa reduction – baseline) had mainly positive amplitudes occurring during the late phase of the QRS complex. Af-ter GNa reduction in the anterior zone the difference ECG had positive amplitudes in the lateral leads (I, aVL and V4-V6) and negative amplitudes in the inferior leads (II,III,aVF), also occurring in the late phase of the QRS. After GNa reduction in the septal zone the difference ECG (septal GNa reduction – baseline) showed minor fractionated differences mainly of negative amplitude and halfway the QRS complex.

Simulation of regional Gto increaseGto increase in the lateral region induced J-point elevations in 4 of the 12 ECG leads (I,II,III and V6) with a maximum amplitude of 0.07 mV in lead V6. The difference ECG showed low-amplitude and generally positive differences during the late phase of the QRS. In the anterior region, Gto increase induced J-point elevations in 3 lateral leads (notch in I and V5 and slur in aVL) but led to J-point depression in the inferior leads. Difference ECGs had somewhat larger amplitudes (positive in lateral leads, negative in inferior leads) and larger QRS prolongation compared to the lateral Gto increase. Gto increase in the septal region did not induce J-point elevations. Also, difference ECGs were small, fractionated, and located halfway the QRS. Overall, Gto increase did prolong the QRS duration, but did not induce an activation delay in the affected zone or in the unaffected zones.

−100

−50

0

50

AP

(mV

)

0 100 200 300 400 500−100

−50

0

50

time (ms)

AP

(mV

)

epi baselineepi 12.5% GNa in lateral zone

epimidendo

transmurallyduring 12.5% GNa in lateral zone

epicardiumbaseline versus 12.5% GNa in lateral zone

−100

−50

0

50

AP

(mV

)

epimidendo

0 100 200 300 400 500−100

−50

0

50

time (ms)

AP

(mV

)

transmurallyduring 10-fold Gto in lateral zone

epicardiumbaseline versus 10-fold Gto in lateral zone

epi baselineepi 10x Gto in lateral zone

−100

−50

0

50

AP

(mV

)

0 100 200 300 400 500−100

−50

0

50

time (ms)

AP

(mV

)

epimidendo

transmurallyduring uncoupling in lateral zone

epicardiumbaseline versus uncoupling in lateral zone

epi baselineepi uncoupling in lateral zone

A B C

Figure S2: Simulated action potential (AP) morphologies in the affected zone. Top panels: APs transmu-rally during intervention. bottom panels: epicardial AP during intervention compared to baseline. A: after GNa reduction in the lateral zone (note the activation delay without affecting the AP morphologies) b: after 10-fold Gto increase in the lateral zone (note the deep notch of the epicardial action potential compared to the endocardial action potential) C: after cellular uncoupling in the lateral zone (note the activation delay without affecting the AP morphologies).

90 Chapter 5

Simulation of cellular uncouplingReduction of the intercellular coupling in the lateral zone cause small J-point notching in lead II and lead V6. Differences in ECGs occurred generally during the late phase of the QRS and the amplitudes were positive in lateral leads and negative in inferior leads. The amplitude of the difference ECGs were in between the amplitudes observed in GNa reduction and Gto increase. In the anterior zone uncoupling caused J-point notching in lead I and aVL with higher amplitudes than uncoupling in the lateral zone. The differ-ence ECGs show smaller, but wider, differences. Again, uncoupling in the septal zone did not induce J-point elevations and the difference ECGs were small and fractionated during the total QRS duration. Although – similar to GNa reduction – cellular uncoupling induced activation delays of about 15 ms and caused maximum activation in the affected zone that correspond with end of QRS, J-point elevations were smaller compared to GNa reduction.

Experiments: pseudo-ECGsThe timing of J-wave onset after flecainide infusion in the OM differed among pseudo-ECG leads and was earliest in inferior leads and later in superior leads (first moment of J-wave onset given in Table 2). The first moment of J-wave onset in the inferolateral leads was earlier after flecainide infusion than at baseline.

Supplemental References 1. Potse M, Dubé B, Richer J, Vinet A, Gulrajani RM. A comparison of monodomain and bidomain

reaction-diffusion models for action potential propagation in the human heart. IEEE Trans Biomed Eng 2006; 53: 2425–2435.

2. Ten Tusscher KHWJ, Noble D, Noble PJ, Panfilov a V. A model for human ventricular tissue. Am J Physiol Heart Circ Physiol 2004; 286: H1573–H1589.

3. Potse M, Dubé B, Vinet A. Cardiac anisotropy in boundary-element models for the electrocardio-gram. Med Biol Eng Comput 2009; 47: 719–729.

Chapter 6

Interventricular dispersion in repolarization causes bifid T waves

in dogs with dofetilide-induced LQT syndrome

Veronique M.F. Meijborg, Samuel Chauveau, Michiel J. Janse, Evgeny P. Anyukhovsky, Peter R. Danilo Jr., Michael R. Rosen,

Tobias Opthof, Ruben Coronel

Heart Rhythm 2015;12(6):1343-51

92 Chapter 6

AbSTRACT

background: Long QT2 (LQT2) syndrome is characterized by bifid (or notched) T waves, whose mechanism is not understood.

Objective: The purpose of this study was to test whether increased interventricular dispersion of repolarization induces bifid T waves.

Methods: We simultaneously recorded surface ECG and unipolar electrograms at base-line and after dofetilide in a canine model of dofetilide-induced LQT2 (6 male mongrel dogs). Standard ECG variables, T wave duration, and moments of peaks of bifid T waves (Tp1 and Tp2) were correlated with moments of local repolarization. Epicardial electro-grams were recorded over the left ventricular (LV) and right ventricular (RV) anterior walls (11x11 electrode grid, 5-mm interelectrode distance). In 5 of the 6 hearts we also recorded intramural unipolar electrograms (n= 4-7 needles per heart). In each unipolar recording, we determined activation time, repolarization time (RT) and activation-recovery interval. In addition, we studied RT response to heart rate changes.

Results: Dofetilide prolonged QT and QTc, induced bifid T waves in 4 of 6 animals, and prolonged RT heterogeneously in LV and RV, resulting in increased interventricular and LV intraventricular RT dispersion. Dofetilide did not induce a disparate response in activation-recovery interval across the transmural axis. Dofetilide-induced separation of RT across the RV-LV interface concurred with the moments of T wave peaks. Dofetilide-induced steepening of restitution slopes was larger in LV than RV.

Conclusion: Dofetilide-induced bifid T waves result from interventricular RT dispersion.

Bifid T waves in LQT2 93

6

INTRODuCTION

The long QT (LQT) syndrome is characterized by a prolonged QT interval and is associ-ated with sudden cardiac death.1 The congenital LQT syndromes are classified according to their associated gene mutations.2,3 Because genetic screening lacks sensitivity and QT prolongation is absent in approximately 25% of LQT patients with a mutation,4 other markers have been identified to improve diagnosis.5,6,7 LQT syndrome type 2 (LQT2) typi-cally manifests bifid (or notched) T waves,8,9 which change in morphology during abrupt heart rate increases provoked by sudden standing.10

LQT2 is associated with a mutation in the HERG gene11 (KCNH2), which encodes the alpha-subunit of the ion channel carrying the rapidly activating component of the de-layed rectifier potassium current IKr.12 Loss of function of IKr channels leads to prolonged action potentials and QT interval.13 The mechanism of bifid (or notched) T waves of LQT2 patients is not known. The relation with sudden standing suggests that altered restitu-tion may be involved in the genesis of the bifid T wave, because restitution determines the amount of heart rate-induced changes in repolarization, which is the main determi-nant for T wave morphology.

Volders et al14 showed a gain of function of the slowly activating component of the delayed rectifier potassium current IKs in canine cardiomyocytes during IKr block, and higher IKs densities in RV cardiomyocytes than LV cardiomyocytes.15 Because IKs is primar-ily responsible for adaptation of action potential duration to altered cycle length, spatial differences in IKs are accentuated and alter restitution properties during IKr block.16 Therefore, we postulated that during IKr block interventricular differences in repolariza-tion become more prominent and may lead to a bifid T wave. We tested this hypothesis in a canine LQT2 model with IKr channel blocker dofetilide infusion and sudden heart rate changes. Our data shed light on the mechanism responsible for bifid (or notched) T waves in LQT2 patients.

METHODS

The experimental protocol complied with the Guide for the Care and Use of Laboratory Animals (US National Institutes of Health Publication 85-23, revised 1996) and was ap-proved by the Stony Brook University Animal Care and Use Committee.

Surgical preparation

Six male mongrel dogs (weight 23-26 kg) were preanesthetized with propofol (6-8 mg/kg IV), intubated, artificially ventilated, and anesthetized with a mixture of isoflurane (2%-3.5%) and oxygen. A left thoracotomy was performed at the fifth intercostal space,

94 Chapter 6

and the heart was suspended in a pericardial cradle (see Supplementary Material for more details).

Drug administration protocol

After the baseline recordings, dofetilide was infused intravenously, and rate-corrected QT (QTc, Bazett formula17) was continuously monitored. When QTc prolongation was at least 25%, we stopped the infusion and identified this dofetilide dose as the bolus. We then commenced dofetilide infusion at a rate= bolus/hour. The total dofetilide dose varied among dogs from a 12-18 µg/kg bolus followed by a 12-18 µg/kg/hour infusion. This protocol induced stable QTc prolongation, that is, QTc before pacing was similar to that measured after completion of the pacing protocols (580±34 ms vs 564±28 ms [mean±SEM], P>0.05, paired t test).

Electrophysiologic study

Electrophysiologic study was performed at baseline and during dofetilide infusion. A grid with 11x11 electrodes (5-mm interelectrode distance) was sutured to the left ventricular (LV) and right ventricular (RV) anterior walls to obtain local epicardial electrograms (see Supplemental Figure S1). In addition, 4-7 intramural plunge needles (each containing 4 electrode terminals, 4-mm interelectrode distance) were inserted in the LV and RV wall in each heart of 5 dogs. Simultaneously, 6 surface ECG leads (I, II, III, aVR, aVL, aVF) were recorded.

Recordings were performed during atrial pacing (A-pace; left atrial appendage) and ventricular pacing (V-pace, anterobasal LV; see Supplemental Methods for details on pacing protocols). In brief, during V-pacing the stimulation protocol incorporated suc-cessive sequences of at least 300 beats: first at basic cycle length (S1), then at a shorter cycle length (S2), and then again at S1 (S1-S2-S1). S2 was shortened in 30-ms steps until S2 approached the T wave end. During A-pacing, only the S1-S2-S1 sequence with the shortest possible S2 cycle length was used.

The following surface ECG measurements were made: RR interval, QRS duration, QT interval and Bazett-corrected QT interval (QTc). We also determined moments of T wave peaks (Tp1= first peak; and Tp2= second peak for bifid T waves) relative to QRS onset. T wave duration (T-width) was determined using the tangent method and averaged over all leads.

Activation time (AT) and repolarization time (RT) were determined at each recording site as the interval from QRS onset or ventricular pacing spike on ECG to the minimum derivative of the local QRS and maximum derivative of the local T wave,18 respectively. Local activation-recovery intervals (ARI) – a measure of local action potential dura-tion – was defined as RT – AT.18,19 Diastolic interval (DI) was calculated as preceding RR interval – RT. Dispersion of repolarization time (dRT) within a region was determined as


6

maximum – minimum RT. Interventricular dRT was defined as mean LV RT – mean RV RT, and LV apicobasal dRT was defined as mean LV apex RT – mean LV base RT. Signal analysis was performed offline using a custom-made data analysis program written in Matlab 2006b (The MathWorks Inc., Natick, MA).20

We evaluated beat-to-beat RT responses to RR transition and constructed steady-state ARI-DI restitution lines (last complex of S2 sequence) from V-pacing recordings. Details of acquisition and analysis are provided in the Supplemental Methods.

Statistics

Continuous variables are presented as mean±SEM. A paired t test was used to compare variables between baseline and dofetilide, between LV and RV, and between LV base and LV apex. Correlations between T-width and RT dispersion were tested with Pearson r. Transmural dispersion and responses in ARI were tested by paired t tests with Bonfer-roni correction. P≤0.05 was considered statistically significant.

RESuLTS

Surface ECG

Figure 1 shows the typically negative T waves in canine ECG leads II, III, and AVF21 at baseline and during dofetilide infusion (atrial pacing). During dofetilide infusion, the T waves in leads II, III, aVL, and aVF were bifid, not unlike those in the limb leads and mainly precordial leads in the LQT2 patient.8 Table 1 lists the ECG characteristics dur-ing sinus rhythm and slow A-pacing. In all dogs, dofetilide significantly increased RR intervals during sinus rhythm and prolonged QT and QTc intervals during sinus rhythm and atrial pacing. Bifid T waves appeared in 4 of 6 dogs. In the other 2 dogs, steady-state QTc prolongation did not reach 25% and T-width widening was smallest.

Table 1: ECG variables during sinus rhythm and A-pacing.

Sinus Rhythm A-pacing

baseline Dofetilide P value baseline Dofetilide P value

N 6 6 6 6

RR, ms 519 ± 26 739 ± 41 < 0.05 505 ± 24 617 ± 53 0.10

QRS, ms 67 ± 3 66 ± 3 0.87 76 ± 1 76 ± 2 0.81

QT, ms 314 ± 6 536 ± 44 < 0.05 266 ± 4 421 ± 42 < 0.05

QTc 359 ± 6 573 ± 32 < 0.05 376 ± 6 532 ± 31 < 0.05

Values are given as mean ± SEM.

96 Chapter 6

Repolarization maps

AT and RT were obtained from 118-120 epicardial recording sites per heart. Figures 2A and 2C show typical repolarization maps at baseline and during dofetilide (slow A-pacing). At baseline, repolarization starts at the RV apex and ends at the LV apex. During dofetilide, repolarization still proceeds from RV apex to LV apex, although in-terventricular RT dispersion is larger compared to baseline (note crowded isochrones). With dofetilide, RV repolarization occurs substantially earlier than LV repolarization. Dispersion is smaller within RV than LV regions. Increased interventricular RT dispersion and smaller RT dispersion within RV compared to LV occurred during dofetilide in 4 of 6 hearts and was smaller or absent in animals without bifid T waves after dofetilide infusion. Activation maps were similar between baseline and dofetilide (see Supple-mental Figure S2). Table 2 summarizes data on activation and repolarization during slow A-pacing for the entire heart, the RV, and the LV. RT and dispersion in RT increased during dofetilide. The prolonging effect on RTs was about twice as large in LV as in RV. RT dispersion increased only in LV, not in RV. Interventricular and LV apicobasal RT dispersion was also increased.

At faster pacing rates (Figures 2B and 2D), the repolarization pattern was similar to that at slower rates, although RTs were shorter because of physiologic adaptation (resti-tution) to higher heart rate. During V-pacing, similar changes in repolarization occurred,

1.0 sec

1 m

V

II

I

III

aVR

aVL

aVF

BASELINE DOFETILIDERR= 500 ms RR= 650 ms

Figure 1: T waves on an ECG of a dog at baseline and dofetilide.

Bifi d T waves in LQT2 97

6

but with smaller eff ects in interventricular RT dispersion, probably because of a slower and opposite activation sequence (see Supplemental Figure S3 and Supplemental Table).

Larger repolarization dispersion during dofetilide led to biphasic T waves in local epicardial recordings at the LV-RV border zone. Figure 3 shows an example of the transi-tion of T wave morphology in epicardial electrograms recorded from RV to LV across the anterior wall, in which biphasic T waves are obvious (asterisks). As a result, transition from RV to LV (recordings 5 and 6) manifests a clear jump in RT. This phenomenon oc-curred in 5 of 6 hearts.

The fi rst T wave peak (Tp1) was signifi cantly later during dofetilide than at baseline (Table 2). In 4 of 6 dogs, a second T wave peak (Tp2) was observed. There was a high

RR 500 ms

RR 650 ms

RR 300 ms

RR 550 ms

isochrones at 10 ms distances


BASELINE

DOFETILIDE

SLOW HR FAST HR

150

325

500

ms

150

325

500

ms

A B

C D

Figure 2: Repolarization maps from a dog during atrial pacing at diff erent cycle lengths, at baseline and during dofetilide. A: Baseline. Last beat of slow pacing rate sequence (RR=500 ms). b: Baseline. Beat 3 of fast pacing rate sequence (RR=300 ms). C: Dofetilide. Last beat of slow pacing rate sequence (RR=650 ms). D: Dofetilide. Beat 3 of fast pacing rate sequence (RR=550 ms). See text for discussion. HR= heart rate.

98 Chapter 6

Table 2: Electrophysiologic variables during slow A-pacing.

baseline Dofetilide P value

No. of animals 6 6

AT duration, ms 43 ± 3 39 ± 2 0.09

RT, ms 215 ± 5 313 ± 24 0.01

RT_RV, ms 201 ± 5 * 270 ± 16 * 0.01

RT_LV, ms 226 ± 5 348 ± 32 0.01

dRT, ms 48 ± 5 128 ± 25 0.02

dRT_RV, ms 24 ± 4 23 ± 4 * 0.75

dRT_LV, ms 23 ± 1 74 ± 10 0.02

dRT_inter, ms 25 ± 4 79 ± 19 0.03

dRT_apicobasal, ms 15 ± 2 43 ± 10 0.03

Tp1, ms 214 ± 7† 287 ± 18‡ 0.03

Tp2, ms 354 ± 23§ (n=4)

T-width, ms 99 ± 6 172 ± 25 0.03

Values are given as mean ± SEM.AT duration= maximum activation time; RT= repolarization time; dRT= dispersion of repolarization time; dRT_apicobasal= left ventricular apicobasal dispersion of repolarization time; dRT_inter= interventricular dispersion of repolarization time; LV= left ventricle; RV= right ventricle; T-width = T wave duration; Tp1= fi rst T wave peak; Tp2= second T wave peak.* P<0.05 for RV vs LV† P= NS for correlation between RT_RV and Tp1 at baseline (R2=0.30)‡ P< 0.01 for correlation between RT_RV and Tp1 during dofetilide (R2=0.98)§ P< 0.01 for correlation between RT_LV and Tp2 during dofetilide (R2=0.98)

1 2 35 6 7 8 9 10 11

4

5 mm

RV

LV

12

1110987

56

34

200 ms

10 m

V

**

Figure 3: Example of transition of epicardial electrograms from right ventricle (RV) to left ventricle (LV) dur-ing atrial pacing (RR=550 ms) and dofetilide infusion. The repolarization time (RT) map (bottom) is copied from Figure 2D. Numbers of recordings correspond to electrode positions on RT map. Red dots indicate time of local repolarization (dV/dt max). Note that T waves in LV and RV are large and opposite in direction, and that at intermediate recording sites T waves are biphasic, with a second maximum dV/dt (red unfi lled circles, remote repolarization).


6

correlation between Tp1 and Tp2 with mean RV and LV repolarization, respectively (Table 2), although there was no correlation of Tp1 with either mean RV or LV repolarization at baseline. Figure 4A shows a typical repolarization map at baseline and during dofetilide with moments of T wave peaks (black lines). In all hearts with bifi d T waves, both Tp isochrones coincided with the RV-LV border. Figure 4B shows a highly signifi cant cor-relation between T-width and interventricular RT dispersion during dofetilide. Bifi d T waves (asterisks) occurred at the largest interventricular RT dispersions. The correla-tion between T-width and LV apicobasal RT dispersion during dofetilide was low and not signifi cant. When heart rate increased, interventricular RT dispersion reduced and bifi d T wave morphology attenuated (see Supplemental Figure S4).

0 50 100 1500

50

100

150

200

250

300

dRT (ms)

T w

idth

(ms)

y = 1.37 * x + 64.03R2 = 0.92 , p < 0.01

*

**

*

= bifid T-wave*

= dRT_inter

= dRT_apicobasal

*

**

*

y = 1.95 * x + 87.31R2 = 0.46 , p = 0.14

RR 550 ms RR 700 ms


BASELINE DOFETILIDE

150

325

500

ms

Tp1= 197 ms Tp1= 270 msTp2= 329 ms

A

B

Figure 4: A: Repolarization maps together with moments of T wave peaks (black lines, Tp1 and Tp2) of a heart at baseline (left) and during dofetilide (right), during slow atrial pacing. b: Relation between T wave width (T-width) and mean interventricular or left ventricular apicobasal repolarization time dispersion (dRT_inter, dRT_apicobasal, respectively) in all hearts during dofetilide and atrial pacing. Note that the greater the interventricular repolarization time dispersion, the wider the T wave.

100 Chapter 6

Regional restitution lines

Supplemental Figure S5 shows beat-to-beat responses of RT after a change in pacing rate. As a result of fast pacing RTs in LV shortened in an alternating fashion, whereas little or no adaption of RTs occurred in RV. To quantify the heart rate-induced response in re-polarization, we reconstructed restitution lines per region (RV, LV, LV apex, and LV base) at baseline and dofetilide. Figure 5A shows typical regional restitution data in 1 heart at baseline and during dofetilide (lower and upper boxes, respectively). Restitution lines were constructed as steady-state ARI vs DI. Solid lines represent the averaged restitution lines of electrodes within a region. At baseline, restitution slopes were diff erent between regions (LV vs RV, and LVbase vs LVapex), although diff erences were small. During dofetilide, restitution slopes were larger, also between regions, particularly between LV and RV. Figure 5B summarizes restitution data for all experiments (n=6). At baseline, RV restitution slopes were signifi cantly smaller than LV slopes, and LV slopes were smaller at the base than at the apex. Diff erences in slopes between LV and RV and between LV apex and base were signifi cantly larger during dofetilide than baseline.

DOFETILIDERV ARI = 0.13 * DI + 210.81LV ARI = 0.69 * DI + 194.33LVbase ARI = 0.57 * DI + 174.51LVapex ARI = 0.76 * DI + 218.04

BASELINERV ARI = 0.09 * DI + 132.39LV ARI = 0.21 * DI + 140.54LVbase ARI = 0.17 * DI + 141.28LVapex ARI = 0.23 * DI + 137.22

0.0

0.2

0.4

0.6

0.8

baseline dofetilide

*

* *

*

Restitution slopesS

lope

100 150 200 250 300 350 400100

150

200

250

300

350

400

450Example: restitution lines ARI − DI

DI (ms)

AR

I (m

s)

DOFETILIDE

BASELINE

Slope differences baseline dofetilideLV-RV 0.06 ± 0.01LVapex-LVbase 0.04 ± 0.01

0.30 ± 0.08 *0.23 ± 0.06 *

A B

Figure 5: A: Mean restitution lines of diff erent areas in a heart at baseline (bottom) and during dofetilide (top). There is a major shift in the lines after dofetilide, with a much steeper slope of left ventricle (LV) com-pared to right ventricle (RV). ARI= activation-recovery interval; DI= diastolic interval. b: Mean restitution slopes were averaged for all 6 animals, and diff erences between LV and RV and between LVapex and LVbase were tested (paired t test, * P<0.05, top). Regional slope diff erences (LV-RV and LVapex-LVbase) between baseline and dofetilide were tested as well (paired t test, * P<0.05, bottom).


6

Transmural repolarization

It has been suggested that IKr block leads to a disparate response in ARI across the LV wall.22,23 This may contribute to a change in T wave morphology. Therefore, we deter-mined whether the effect of dofetilide was different across the LV wall. As shown in the Supplemental Results and Supplemental Figure S6 ARIs were shorter at the epicardium than at the endocardium both at baseline and during dofetilide without midmural maxima. Furthermore, dofetilide-induced changes in ARI were homogeneous across the LV wall.

DISCuSSION

The appearance of bifid (or notched) T waves in dogs with dofetilide-induced LQT2 is related to increased interventricular dispersion of repolarization and steepening of the restitution slope that is larger in LV than RV. The dofetilide effect on ARIs across the ventricular walls is homogeneous.

Dofetilide effect

The dofetilide-induced increase in QT and QTc intervals, RTs, and dispersion of repo-larization is consistent with earlier reports.24 The repolarization-prolonging effect of dofetilide was larger in LV than RV. Intraventricular RT dispersion increased in LV alone. This may be explained by the higher IKs density in RV than LV,15 which plays a dominant role in repolarization after IKr blockade.14 As a consequence, less repolarizing current (IKs) will be available in LV, resulting in greater prolongation of LV than RV repolarization.

After dofetilide, we observed a steeper restitution slope in LV than in RV, which is inconsistent with the anticipation that a higher IKs density in RV encompasses a steeper restitution slope than in LV. However, DIs in RV were approximately 50 ms larger than in LV and therefore may be nearer to the plateau of its restitution curve.25 Addition-ally, methodologic considerations (use of linear instead of a higher-order function and not pacing at minimum possible DI) may have led to underestimation of the maximum steepness of the slope.

When heart rate was increased by pacing during dofetilide, an alternation in RT oc-curred during the first beats of the rapid rate in LV only (see Supplemental Figure S4).26 Our data suggest that this change is due to the steeper restitution slope and the shorter DIs in LV compared to RV.

bifid T waves

We observed bifid T waves in 4 of 6 dogs after dofetilide infusion. This is consistent with results of another study in dofetilide-treated dogs.27 The T waves had a negative polarity

102 Chapter 6

in most leads, typical of canine hearts,21 and in contrast with positive T waves in most leads in the human ECG. This discrepancy is likely explained by the different thoracic heart position and thorax shape of the dog. Additionally, there may be a species differ-ence in the repolarization sequence. Furthermore, in LQT2 patients bifid (or notched) T waves have a low amplitude,8 in contrast with our observations. However, Zhang et al8 showed that the amplitude of bifid T waves was variable (SD 0.25 mV) and that its distribution was skewed and therefore must have included some high voltage T waves. Although T wave morphology is recognizable for both congenital8 and acquired LQT,28 bifid T waves have a low incidence in acquired LQT patients. However, this low incidence may be associated with smaller QTc prolongation compared to our study, due to a more prudent drug challenge in patients.

Clinically, the first peak of the T wave usually has a higher amplitude in right pre-cordial leads, whereas the second peak of the T wave has a higher amplitude in left precordial leads (Figure 2 in Zhang et al8). Therefore, we reasoned that the first T wave peak is spatially associated with RV repolarization and that the second T wave peak with LV repolarization. Our data confirm that RV repolarization is early, whereas LV repolariza-tion is late (Figures 2-4). The disparate repolarization of RV and LV during dofetilide may result in 2 separated T waves, which, by superimposition, may lead to a bifid morphol-ogy. Accordingly, we demonstrated that isochrones of the T wave peaks (Tp1 and Tp2) did concur with the end of RV repolarization and the first LV repolarization (Figure 4A). Moreover, T-width did correlate with mean interventricular dispersion of repolarization (Figure 4B). This indicates that interventricular dispersion of repolarization play a role in generating both components of the bifid T wave. LV apicobasal RT dispersion and difference in ARI restitution slopes also were increased and may contribute to the bifid T wave as well. However, interventricular differences were approximately 1.5–2 times larger. Also, the relationship between LV apicobasal RT dispersion and T-width was weaker. Therefore, we suggest that LV apicobasal RT dispersion has no prominent role in bifid T wave genesis.

In children without detectable heart disease, the incidence of bifid T waves in leads V2 and V3 is high (18.3%)29 and usually concurs with normal QTc intervals, likely exclud-ing repolarization pathology. Calabrò et al29 suggested that this phenomenon depends on the hypothetical figure-of-eight shape T-loop on the horizontal plane, which may disappear with aging because of changes in the ventricular repolarization process. Thus, an age-dependent shift in the balance between IKr and IKs may explain the bifid T wave morphology in the young. Indeed, our voltage-clamp study demonstrated that in the majority of LV cardiomyocytes from young dogs, only IKr is functionally expressed, whereas IKr and IKs are both present in adult myocardium.30

Bifid (or notched) T waves can also be observed in patients with alcoholic cardiomy-opathy31 or in those with central nervous system disease.32 Interestingly, the latter study


6

suggested that notched T waves may be related to divergence of RT of 2 ventricular cell populations, due to alterations in sympathetic tone. This is in line with the disparate repolarization of the RV and LV that we observed.

Nevertheless, some precaution should be taken when comparing our model to LQT2 patients. Our model is essentially an “acquired” LQT2 model (i.e., acute IKr block) and may differ from “congenital” LQT2 patients (i.e., chronic IKr absence), in which remodeling of the balance in IKr and IKs may have played a role in the mechanism. Also, remodeling of other repolarizing currents may influence the mechanism. In addition, contrary to man, the T wave in dogs typically is negative in the “standard” leads. Therefore, extensive electrophysiologic mapping in LQT2 patients is needed to fully corroborate that a similar mechanism holds for the human bifid T wave.

Viskin et al10 showed that bifid (or notched) T waves exaggerate during sudden stand-ing and increased heart rates. In contrast, in our study, bifid T waves attenuated at faster rates. However, an exaggeration or attenuation of bifid T waves may depend on how fast the heart rate is before standing (Figure 6, left or right to intersection of LV and RV slopes). Also, we did not incorporate autonomic nerve modulations occurring during sudden standing, and these may also play a role in bifid T waves.

LV dofetilide

RV dofetilide

ARI

RR

LV baseline

RV baseline

Figure 6: Schematic restitution lines (activation-recovery interval [ARI] versus RR) at baseline and during dofetilide for 2 regions within the heart. Lines were based on our data (solid lines) and extrapolation using literature (dashed lines). LV= left ventricle; RV= right ventricle.

Transmural dispersion of repolarization

Antzelevitch et al22,23 have demonstrated a disparate response of an IKr blocker on action potential duration across the LV wall in a canine wedge preparation. Our study in open-chested dogs showed that dofetilide-induced changes in ARI were homogeneous across the LV wall. The discrepancy in transmural response may be due to the use of a different pharmacological agent and, probably more importantly, the use of different prepara-tions. It has been suggested that the sodium-blocking effect of propofol may reduce

104 Chapter 6

heterogeneity of repolarization.23 However, propofol was used only to induce anesthesia (see Supplementary Methods), so it is unlikely that it influenced repolarization at the time of recording. Furthermore, we did observe a large increase in interventricular RT dispersion, and there is no compelling reason to assume that transmural RT dispersion would be selectively decreased.

Implications for arrhythmogenesis

Restitution of the RV had a flatter slope in our experiments than that of the LV, par-ticularly after dofetilide infusion. This may have consequences for arrhythmogenesis. Dofetilide alters the repolarization properties of myocardial tissue such that RV and LV restitution shift away from each other (schematically illustrated in Figure 6). We could only construct restitution lines over a limited range of cycle lengths, due to LV pacing (solid lines). The rest of the line was extrapolated using theory from the literature (bro-ken lines).25 During dofetilide infusion, interventricular RT dispersion is largest during long pacing cycle lengths. In this condition, a short coupled premature beat occurring in the RV will be able to conduct through the RV but will find activation block at the LV due to refractoriness (Figure 6), especially after a preceding pause. This is in line with the reported long-short sequence preceding torsade de pointes.33

The schematic restitution lines in Figure 6 may also explain alterations of bifid (or notched) T wave morphology during heart rate changes. For example, a transition in heart rate will cause a shift on the x-axis resulting in a change in interventricular disper-sion in ARI (i.e., change in interventricular RT dispersion), which in turn may determine the presence and morphology of bifid T waves.

CONCLuSION

The dofetilide-induced bifid (or notched) T wave is correlated with the large dispersion of repolarization between RV and LV.

ACkNOWLEDGEMENTS

We thank Ya-Ping Jiang, MD, Tania Rahim, MS, and Ira S. Cohen, MD, PhD, for technical support during the experiments.


6

REFERENCES

1. Priori SG, Schwartz PJ, Napolitano C, Bloise R, Ronchetti E, Grillo M, Vicentini A, Spazzolini C, Nastoli J, Bottelli G, Folli R, Cappelletti D. Risk stratification in the long-QT syndrome. N Engl J Med 2003; 348: 1866–1874.

2. Goldenberg I, Moss AJ. Long QT syndrome. J Am Coll Cardiol 2008; 51: 2291–2300.

3. Bhuiyan ZA, Al-Shahrani S, Al-Aama J, Wilde AAM, Momenah TS. Congenital Long QT Syndrome: An Update and Present Perspective in Saudi Arabia. Front Pediatr 2013; 1: 39.

4. Tester DJ, Will ML, Haglund CM, Ackerman MJ. Effect of clinical phenotype on yield of long QT syndrome genetic testing. J Am Coll Cardiol 2006; 47: 764–768.

5. Horigome H, Ishikawa Y, Shiono J, Iwamoto M, Sumitomo N, Yoshinaga M. Detection of extra components of T wave by independent component analysis in congenital long-QT syndrome. Circ Arrhythm Electrophysiol 2011; 4: 456–464.

6. Struijk JJ, Kanters JK, Andersen MP, Hardahl T, Graff C, Christiansen M, Toft E. Classification of the long-QT syndrome based on discriminant analysis of T-wave morphology. Med Biol Eng Comput 2006; 44: 543–549.



9. Kanters JK, Fanoe S, Larsen LA, Bloch Thomsen PE, Toft E, Christiansen M. T wave morphology analysis distinguishes between KvLQT1 and HERG mutations in long QT syndrome. Heart Rhythm 2004; 1: 285–292.



12. Sanguinetti MC, Jiang C, Curran ME, Keating MT. A mechanistic link between an inherited and an acquired cardiac arrhythmia: HERG encodes the IKr potassium channel. Cell 1995; 81: 299–307.

13. Sanguinetti MC. Dysfunction of delayed rectifier potassium channels in an inherited cardiac ar-rhythmia. Ann N Y Acad Sci 1999; 868: 406–413.

14. Volders PGA, Stengl M, van Opstal JM, Gerlach U, Spätjens RLHMG, Beekman JDM, Sipido KR, Vos MA. Probing the contribution of IKs to canine ventricular repolarization: key role for beta-adrenergic receptor stimulation. Circulation 2003; 107: 2753–2760.

15. Volders PGA, Sipido KR, Carmeliet E, Spatjens RLHMG, Wellens HJJ, Vos MA. Repolarizing K+ Cur-rents ITO1 and IKs Are Larger in Right Than Left Canine Ventricular Midmyocardium. Circulation 1999; 99: 206–210.

16. Jost N, Papp JG, Varró A. Slow delayed rectifier potassium current (IKs) and the repolarization reserve. Ann Noninvasive Electrocardiol 2007; 12: 64–78.

106 Chapter 6

17. Bazett H. An analysis of the time-relations of electrocardiograms. Heart 1920; 7: 353–370.


19. 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.


21. Janse MJ, Sosunov EA, Coronel R, Opthof T, Anyukhovsky EP, de Bakker JMT, Plotnikov AN, Sh-lapakova IN, Danilo P, Tijssen JGP, Rosen MR. Repolarization gradients in the canine left ventricle before and after induction of short-term cardiac memory. Circulation 2005; 112: 1711–1718.

22. Yan GX, Antzelevitch C. Cellular basis for the normal T wave and the electrocardiographic mani-festations of the long-QT syndrome. Circulation 1998; 98: 1928–1936.

23. Shimizu W, Antzelevitch C. Sodium channel block with mexiletine is effective in reducing dis-persion of repolarization and preventing torsade des pointes in LQT2 and LQT3 models of the long-QT syndrome. Circulation 1997; 96: 2038–2047.

24. Mounsey JP, DiMarco JP. Dofetilide. Circulation 2000; 102: 2665–2670.

25. Franz MR. The Electrical Restitution Curve Revisited: . Steep or Flat Slope-Which is Better? J Cardio-vasc Electrophysiol 2003; 14: S140–S147.

26. Janse MJ, Van Der Steen ABM, Van Dam RT, Durrer D. Refractory Period of the Dog’s Ventricular Myocardium Following Sudden Changes in Frequency. Circ Res 1969; 24: 251–262.

27. Haushalter TM, Friedrichs GS, Reynolds DL, Barecki-Roach M, Pastino G, Hayes R, Bass AS. The cardiovascular and pharmacokinetic profile of dofetilide in conscious telemetered beagle dogs and cynomolgus monkeys. Br J Pharmacol 2008; 154: 1457–1464.

28. Jackman WM, Friday KJ, Anderson JL, Aliot EM, Clark M, Lazzara R. The long QT syndromes: a critical review, new clinical observations and a unifying hypothesis. Prog Cardiovasc Dis 1988; 31: 115–172.

29. Calabrò MP, Barberi I, La Mazza A, Todaro MC, De Luca FL, Oreto L, Russo MS, Cerrito M, Bruno L, Oreto G. Bifid T waves in leads V2 and V3 in children: a normal variant. Ital J Pediatr 2009; 35: 17.

30. Obreztchikova MN, Sosunov EA, Plotnikov A, Anyukhovsky EP, Gainullin RZ, Danilo P, Yeom Z-H, Robinson RB, Rosen MR. Developmental changes in IKr and IKs contribute to age-related expres-sion of dofetilide effects on repolarization and proarrhythmia. Cardiovasc Res 2003; 59: 339–350.

31. Evans W. The electrocardiogram of alcoholic cardiomyopathy. Br Heart J 1959; 21: 445–456.

32. Millar K, Abildskov JA. Notched T Waves in Young Persons with Central Nervous System Lesions. Circulation 1968; 37: 597–603.

33. Kay GN, Plumb VJ, Arciniegas JG, Henthorn RW, Waldo AL. Torsade de pointes: The long-short initiating sequence and other clinical features: observations in 32 patients. J Am Coll Cardiol 1983; 2: 806–817.


6

DATA SuPPLEMENT


Surgical preparationSix male mongrel dogs weighing 23-26 kg were preanesthetized with propofol (6-8 mg/kg i.v., 60-90 minutes before measurements were made), intubated, artifi cially ventilated and anesthetized with a mixture of isofl urane (2% - 3.5%) and oxygen. Two peripheral intravenous catheters were inserted for administration of fl uids and drugs. A femoral artery catheter was used to monitor blood pressure continuously. A left thoracotomy was performed at the fi fth intercostal space, and the heart was suspended in a pericar-dial cradle. To provide atrial and ventricular stimulation, bipolar silver electrodes were sewn to the left atrial appendage and the anterobasal left ventricle, respectively. Body temperature was monitored via a thermistor probe placed deep within the thorax and maintained at 36-37°C via a heating pad and heating lamp.

Experimental setupSupplemental Figure S1 show a schematic of the experimental setup. A grid with 11x11 electrodes (5-mm interelectrode distance) and 9-10 intramural plunge needles were used to obtain local electrograms. Simultaneously, surface ECGs were obtained.

= RV region

= LV region

= LV base

= LV apex

Figure S1: Schematic of the experimental setup. Block pulses at left atrium and left ventricular base indicate the location of stimulating electrodes. An 11x11 electrode grid was positioned over the anterior left ven-tricle (LV) and right ventricle (RV). Plunge needles were inserted at several locations surrounding the 11x11 grid. To determine electrophysiologic variables for diff erent regions within the heart, we selected groups of electrodes of the 11x11 grid that represent these regions. In the fi gure the selections are indicated by the colored boxes.

108 Chapter 6

Pacing protocolThe pacing protocol incorporated pacing at the atrium (A-pacing) or at the ventricle (V-pacing). During V-pacing each ventricular stimulus was preceded by a stimulus at the left atrial appendage with an interval of 20 to 30 ms, to avoid influences of decre-mental AV conduction on ventricular rhythms and prevent coincidence of the remote P-wave with T waves in local unipolar electrograms. The V-pacing stimulation protocol incorporated a S1-S2-S1 sequence with at least 300 beats per cycle length. The S1 was the basic cycle length (just faster than intrinsic sinus rate) and S2 was at a shorter cycle length than S1. S2 was sequentially shortened by 30-ms until S2 approached the end of the T wave. Following we performed the A-pacing stimulation protocol, which was a repetition of the S1-S2-S1 sequence with the shortest possible S2 cycle length performed during V-pacing.

Steady-state restitution linesSteady-state restitution lines (last complex of S2 sequence) were constructed from ARI and DI data obtained during V-pacing. First we calculated a linear regression line (ARI= slope × DI + intercept) per epicardial electrode. For each region (RV, LV, LV base, and LV apex) we determined a mean restitution line as the average slope and intercept of all lines within a group of epicardial electrodes (Supplemental Figure S1: 33 electrodes for RV, 44 for LV of which 16 for LV base and 16 for LV apex. In one dog, RV area was small, permitting use of only 22 electrodes).

AcquisitionLocal unipolar electrograms were recorded using a 256-channel amplifier (BioSemi, 24 bit dynamic range, 122.07 nV bit step, total noise 0.5 µV); sampling frequency of 2048 Hz (bandwidth [−3dB] DC-400 Hz). The active common mode electrode was positioned in the subcutaneous fat at the sternal incision site. Stimulation pulse amplitude was 1.5 x stimulation threshold and pulse width =2 ms. We used an EMKA system (version 1.8) to simultaneously record 6-lead ECGs throughout each experiment. Sampling frequency was 1000 Hz with filtering with a high-pass 0.05Hz and a low-pass 200Hz.


Activation: A-pacing versus V-pacingThe activation during A-pacing started at the RV apical region and ended at the LV or RV basal region. During V-pacing activation was in the opposite direction, starting always at the LV basal region and ended at the RV apical region. An example of activation maps during A-pacing and V-pacing is demonstrated in Supplemental Figure S2. This figure demonstrates that the activation sequence is opposite in direction between the two


6

pacing modes. It demonstrates also that activation is slower and occurs later after the start of activation within the heart (defi ned as the beginning of the QRS complex or pacing artefact). Activation did not diff er between baseline and dofetilide.

RR 500 ms

RR 650 ms

RR 500 ms

RR 650 ms



BASELINE

DOFETILIDE

A-pace V-pace

0

60

120

ms

0

60

120

ms

A B

C D

Figure S2: Example of activation maps during atrial pacing (A-pace) and ventricular pacing (V-pace) at baseline (A and b) and during dofetilide (C and D) in the same animal. Each activation map is relative to the beginning of the QRS complex on the ECG. Activation during V-pacing is opposite in direction and is later in total activation sequence (more blue instead of red). Activation did not diff er between baseline and dofetilide.

110 Chapter 6

Repolarization during V-pacingSupplemental Figure S3A and S3C show typical examples (same dog as in Figure 3) of repolarization maps at baseline and during dofetilide during V-pacing at a slow pacing rate (RR 500 ms and RR 650 ms, respectively). At baseline, repolarization started in the LV basal region and ended in 3 animals in the LV apical region (as in the fi gure) and in

RR 500 ms

RR 650 ms

RR 300 ms

RR 550 ms

200

350

500

ms

200

350

500

ms



BASELINE

DOFETILIDE

SLOW HR FAST HR

A B

C D

Figure S3: Examples of repolarization maps of the dog in Figure 2 during ventricular pacing (V-pace) at diff erent cycle lengths, at baseline and during dofetilide. A: Baseline. Last beat of the slow pacing rate se-quence (RR= 500 ms). b: Baseline. Beat 10 of the fast pacing rate sequence (RR= 300 ms). C: Dofetilide. Last beat of the slow pacing rate sequence (RR= 650 ms). D: Dofetilide. Beat 10 of the fast pacing rate sequence (RR= 550 ms).


6

3 animals in the RV apical region. During dofetilide infusion, interventricular RT disper-sion is increased and repolarization starts at the RV basal region and ends in all animals at the LV apex. However, due to stimulation from the LV base repolarization at LV base remains substantially early, leading to smaller interventricular dispersion in repolariza-tion compared to repolarization during A-pacing after dofetilide (see Figure 3C and 3D).

Supplemental Table summarizes the data on activation and repolarization times and dispersion in repolarization during V-pacing at a slow rate for the entire heart, the RV and the LV. It shows similar results as during A-pacing (Table 2), although differences in repolarization between the LV and RV were less prominent.

Supplemental Table: Electrophysiologic variables during ventricular pacing (V-pacing) at slow heart rate.

baseline Dofetilide P value

No. of animals 6 6

AT duration, ms 108 ± 2 105 ± 3 0.25

RT, ms 265 ± 5 360 ± 20 0.01

RT_RV, ms 266 ± 6 336 ± 12 0.01

RT_LV, ms 263 ± 5 377 ± 28 0.01

dRT, ms 40 ± 2 102 ± 19 0.01

dRT_RV, ms 20 ± 3 * 22 ± 6 * 0.50

dRT_LV, ms 39 ± 3 88 ± 16 0.01

dRT_inter, ms 6 2 45 15 0.03

Values are given as mean ± SEM.AT duration= maximum – minimum activation time; RT= repolarization time; dRT= dispersion in repolariza-tion time; dRT_inter= interventricular dispersion of repolarization time; LV= left ventricle; RV= right ven-tricle.* P<0.05 for RV vs LV

Response in bifid T wave morphology to heart rate changesSupplemental Figure S4 shows a typical response in bifid T wave morphology with corresponding repolarization times to a change in heart rate during atrial pacing and dofetilide. The first beat at slow heart rate (RR650) show a clear bifid T wave on the ECG. During the following beats at a faster heart rate (RR550), the RT dispersion between the LV and RV decreased (bottom panel) and led to a less pronounced morphology of the bifid T wave (top panel).

RT response to abrupt heart rate changesWe assessed the response in repolarization to changes in heart rate. Supplemental Figure S5A and S5B show typical RT responses in 5 selected epicardial recordings (2 at RV epicardium, base and apex; and 3 at LV epicardium, base, central, and apex) during transitions of cycle length (long [1 complex] – short [11 complexes] – long [11 complexes]) in absence and presence of dofetilide. At baseline (Supplemental Figure S5A), the transi-

112 Chapter 6

tion in pacing rate (RR= 500 ms to 300 ms) led to shortening of repolarization times of approximately 30 ms in LV and 20 ms in RV. RTs prolonged after the return to slow pacing. However, during dofetilide infusion (Supplemental Figure S5B), the response to transitions in pacing rate (RR= 650 to 550 ms) differed substantially between LV and RV. In the LV, RTs shortened as a result of fast pacing and this occurred in an alternating fashion. In contrast, little or no adaption of RTs occurred in RV in response to a change in pacing rate. The difference in adaptation is emphasized in the epicardial electrograms of the first 6 complexes (last slow and first 5 fast) in Supplemental Figure S5C for one LV and one RV recording site (asterisks in Supplemental Figure S5B).

Transmural dispersion of repolarizationIt has been suggested that IKr block leads to a disparate response in ARI across the LV wall.1,2 This may contribute to a change in T wave morphology. Therefore, we determined whether the effect of dofetilide was different across the LV wall. For testing transmural ARI dispersion and ARI responses we selected needles without missing data (i.e., 4 elec-trograms per needle) both at baseline and during dofetilide. Supplemental Figure S6A shows 11 transmural ARI profiles obtained in 4 hearts. ARIs were shorter at the epicar-

RR650 RR550 RR550 RR550 RR550250

300

350

400

450

Complex

RT

(ms)

RVapexRVbase

LVapexLVmidLVbase

aVF

500 ms

1 m

V

Figure S4: Typical response of bifid T wave morphology with corresponding repolarization times (RTs) to a change in heart rate (A-pacing during dofetilide). Top panel: ECG lead aVF of successive complexes with preceding RR-interval RR650 or RR550 (shown on the x-axis of bottom panel). bottom panel: RT responses in 5 epicardial recordings (2 at RV epicardium [red] and 3 at LV epicardium [blue]) corresponding with the bifid T waves shown above.


6

dium than at the endocardium both at baseline and during dofetilide without midmural maxima. We also calculated for each transmural recording site the dofetilide-induced change in ARI (ARI dofetilide – ARI baseline). The changes in ARIs were homogeneous

last 1 2 3 4 5 6 7 8 9 10 last 1 2 3 4 5 6 7 8 9 10 last200

220

240

260

280

300

Complex no.

RT

(ms)

last 1 2 3 4 5 6 7 8 9 10 last 1 2 3 4 5 6 7 8 9 10 last320

340

360

380

400

420

440

460

480

500

520

RT

(ms)

Complex no.

BASELINE V-pacing

DOFETILIDE V-pacing

RR300 RR500RR500

RR550 RR650RR650

RVapexRVbase

LVapexLVmidLVbase

100 ms

10 m

V

RV

LV

last RR650

1

2

3-5

last RR650

1-5

*

*

*

*

A

B

C

Figure S5: A typical response of repolarization time (RT) to a change in heart rate (V-pacing: slow-fast and fast-slow) at baseline (A) and dofetilide (b). The RT responses in 5 epicardial recordings (2 at RV epicardium [red] and 3 at LV epicardium [blue]) are shown. The x-axes show the following RT complex numbers: last of slow pacing sequence – first 10 and last of fast pacing sequence – first 10 and last of returning slow pacing sequence. The y-axes show RT. C: Epicardial recordings from LV (blue) and RV (red) of sequential complexes (last of slow and first 5 of fast pacing rate), aligned on the QRS complex. Note the large changes in LV repo-larization (ascending slope of the T wave) compared to minimal changes in RV.

114 Chapter 6

across the LV wall, with minor difference (10 ms) between epicardium and endocardium. As a consequence, the RT profiles were almost flat between epicardium and endocardi-um at baseline and during dofetilide (Supplemental Figure S6B). Obviously, differences in RT between endocardium and any transmural site, including the epicardium, cannot explain the occurrence of bifid T waves in the presence of dofetilide.

Supplemental References 1. Yan GX, Antzelevitch C. Cellular basis for the normal T wave and the electrocardiographic mani-

festations of the long-QT syndrome. Circulation 1998; 98: 1928–1936.

2. Shimizu W, Antzelevitch C. Sodium channel block with mexiletine is effective in reducing dis-persion of repolarization and preventing torsade des pointes in LQT2 and LQT3 models of the long-QT syndrome. Circulation 1997; 96: 2038–2047.

mean ± SEM for all 11 needles

epimid−epimid−endoendo180

200

220

240

260

280

300

320

0

20

40

60

80

100

120

140

AR

I (m

s) o

r RT

(ms)

∆ A

RI (

ms)

ARI b.l. ARI dof ∆ ARIRT b.l. RT dof

endo mid-endo mid-epi epi150

200

250

300

350

dog 2 (RR=480ms)dog 3 (RR=600ms)dog 4 (RR=450ms)dog 5 (RR=500ms)

BASELINE

dog 2 (RR=500ms)dog 3 (RR=600ms)dog 4 (RR=450ms)dog 5 (RR=650ms)

DOFETILIDE

ARI per needleA

RI (

ms)

A B

Figure S6: A: Activation-recovery intervals (ARI, y-axis) at different transmural depths (x-axis) per full nee-dle (n=11 in 4 hearts) at baseline (solid lines) and during dofetilide (dashed lines) during atrial pacing. Little to no ARI dispersion occurred along the transmural axis. b: Average ARIs and RTs for all 11 full needles at baseline and during dofetilide and dofetilide-induced changes in ARI (∆ARI). Baseline = b.l.; dofetilide = dof.

Chapter 7

QT-TQ dynamics in congenital long QT patients and controls

during a supine-standing ECG test

Veronique M.F. Meijborg, Pieter G. Postema, Henk J. Ritsema van Eck, Arthur A.M. Wilde, Ruben Coronel


116 Chapter 7

AbSTRACT

background: The diagnosis of long QT syndrome (LQTS) is not straightforward in the absence of clear QT prolongation in the resting ECG. Detailed analysis of the dynamics of QT and TQ (=RR-QT) during supine-standing tests may improve the diagnostic pos-sibilities of LQTS.

Objective: To evaluate QT-TQ dynamics in LQTS patients.

Methods: Age (42±5 years, mean±SD) and gender (63% male) matched subjects with LQTS1, LQTS2 and healthy relatives (controls) were studied (each group n=8) off beta-blocker therapy. Continuous 12-lead ECGs were made during 2 minutes in supine posi-tion (baseline) followed by 3 minutes of standing. A custom-made program applying fiducial segment averaging allowed beat-to-beat analysis of RR, TQ, QT and QTc and QT/TQ-ratios at baseline, during stand-up (first 30s of standing) and during standing (1min following stand-up). QT/TQ-crossover was defined as a change in QT/TQ-ratio from <1 at baseline to >1 during standing.

Results: In all groups TQ significantly decreased during stand-up, whereas QT hardly decreased. It resulted in a significant QTc-increase during stand-up (‘QTc-stretching’) in control and LQT2, but not in LQT1. Restitution slopes constructed by beat-to-beat chang-es differed among groups, with steeper slopes in LQT1 and slightly higher intercepts in LQT2. In 6/8 controls QTc of several beats exceeded the previously established critical level of 490 during stand-up. All four LQTS patients with a normal baseline QTc<465, and only 1/5 controls, demonstrated a QT/TQ-crossover during standing (P<0.05).

Conclusion: QT/TQ-crossover during standing may add to diagnosis of LQTS, and identi-fies LQTS patients with otherwise borderline baseline QTc.

QT-TQ dynamics in LQTS 117

7

INTRODuCTION

The congenital long QT syndrome (LQTS) is associated with syncope and sudden cardiac death resulting from Torsade de Pointes, commonly triggered by emotional or physical stress.1,2 The diagnosis of LQTS is based on a marked QT prolongation on the ECG with or without a clinical history of unexplained syncope.3 Although clinical criteria have been developed to improve diagnosis and risk stratification,4 the diagnosis is less straightfor-ward in the absence of a clear QT-prolongation in the resting ECG.

Due to the typical autonomic triggers for arrhythmias, the syndrome was originally attributed to imbalance of the autonomic system in these patients.5 Cardiac sympa-thetic denervation is likewise being deployed as effective therapy for patients with therapy-refractory events.6 Research has increasingly focused on genetics following the discovery of the first LQTS associated gene in 1991.7 Since, 15 different types of LQTS have been classified based on their genetic profile.8 Types 1 and 2 of LQTS (LQT1 and LQT2, respectively) encompass the largest group of LQTS patients, in whom a loss-of-function of, respectively, the slow or fast activating component of the delayed rectifier channels is involved.9,10 However, a diagnosis based on genetics alone causes misdiagnosis of 15-20% of the phenotypical LQTS patients in whom genetics remains elusive.11 Furthermore, a diagnosis based on the QT interval alone can be difficult, because the QTc range in LQTS patients overlaps that of the healthy population.12,13 Therefore, several studies aimed to identify other markers that improve diagnosis.14,15,16 New measures are short-term QT variability,17 QT-stretching18 and QT-stunning.19 The latter measures involve the direct and short-term QT interval adaptation to heart rate changes induced by brisk standing. In collaboration with our group, Viskin et al showed that QTc during QT-stretching – i.e. at shortest interval between T wave and next P wave – was higher among LQTS patients compared to healthy controls.18 It was posited that a cut-off of QTc= 490 during QT-stretching would be a valuable qualitative diagnostic in individuals without obvious baseline QT prolongation.19 In addition, we demonstrated that QT during QT-stunning (i.e. upon return to baseline heart rate) lengthened in LQTS patients, whereas it short-ened in control subjects. These studies imply that the dynamicity of heart rate and QT intervals is different in LQTS patient and may require a more profound analysis. The aim of this study was to evaluate the beat-to-beat QT-RR responses to sudden standing in LQTS patients compared to healthy controls.

118 Chapter 7

METHODS

Population

We retrospectively collected ECG recordings of age and gender matched groups of patients with a LQT1 mutation (LQT1), patients with a LQT2 mutation (LQT2), and as-ymptomatic relatives without a mutation (controls). We selected individuals that visited our cardiogenetics clinic at the Academic Medical Center either for evaluation of their symptoms or for family screening. All subjects underwent an ECG recording during a supine-standing test, before any beta-blocker therapy was started. We obtained a waiver from the local ethical committee for ethical approval for the conduct of this study.

Acquisition and signal analysis

Continuous standard 12-leads electrocardiogram (ECG) recordings were made during 2 minutes in supine position (baseline) and after the instruction to stand up quickly and stand still for 3 minutes. Recordings were made using the Cardio Perfect system (Welch-Allyn Cardio Control, Delft, the Netherlands) at a sampling rate of 600 Hz without filtering, and offline up-sampled to 1200 Hz for a more accurate alignment of complexes.

ECG recordings were divided into fragments to allow data analysis: 1) first supine: fragment of 60 seconds; 2) second supine (baseline): fragment until stand-up (about 60 seconds); 3) first 30 sec of standing (stand-up): first 5-10 seconds missing due to skeletal muscle artefacts; 4) second standing (standing): fragment of following 60 seconds; 5) third standing: last fragment of the test (minor cases). The number of measured frag-ments per subject varied and therefore we mainly focused on the fragment before, during and after stand-up (baseline, stand-up and standing). For analysis we used 9 of the 12 leads (I, II, III, V1 – V6).

Beat-to-beat signal analysis was performed offline using a custom-made program applying fiducial segment averaging (Intraval™).20 In short, the program identifies in-dividual QRS complexes by using a matching filter technique. Each individual complex is then in turn cross-correlated with the average segment of the remaining complexes and subsequently shifted in time until maximum correlation is obtained. This process is repeated until no further improvement is realized. This iterative process is subsequently repeated for all default fiducial segments, i.e. the segments around the fiducial points: P wave onset, QRS onset, QRS end, T wave peak, and T wave end, and resulted in optimally aligned segments of similarly looking deflections (e.g. all T wave peaks aligned at their apex). In this way we obtained beat-to-beat measurements of RR and QT, from which QTc (heart rate corrected according to Bazett) was calculated.

The diastolic intervals (TQ) rather than RR intervals are determinative for the response in QT21 (QT-restitution). Therefore, TQ intervals were calculated as the interval from the preceding end of T wave to QRS onset (RR minus QT). Beat-to-beat QT/TQ-ratio is ob-


7

tained as: QT interval / preceding TQ interval. Next, we determined beat-to-beat QT-TQ responses to evaluate QT-restitution. The period of stand-up is excluded because we considered it too dynamic – i.e. instable with large RR variation – as it may interfere with reliable restitution determination. QT-restitution was constructed first by determin-ing linear regression lines (QT= slope x TQ+intercept) per fragment, after which slopes and intercepts with y-axis were averaged to obtain a mean restitution line per subject (short-term restitution, i.e. beat-to-beat). An alternative construction incorporated one linear regression line per subject irrespective of the fragment (long-term restitution, i.e. approaching steady-state). Slopes and intercepts were tested between groups.

Statistics

Continuous variables were presented as mean±SD and categorical variables as frequency (percentage). Comparisons between groups were performed using a one-way ANOVA followed by Bonferroni post-hoc testing. Comparisons within groups between baseline and stand-up were performed using a paired t test. For comparisons of categorical vari-ables we used a Fisher’s Exact test. P values ≤0.05 were considered statistically significant.

RESuLTS

We collected data from gender and age matched controls, LQT1 patients and LQT2 patients (n=8 each group). The total number of beats analyzed per subject was 347±36. Table 1 shows the baseline characteristics (i.e. supine fragment). The mean RR intervals

Table 1: Baseline characteristics of the three groups (control, LQT1, LQT2).

Control LQT1 LQT2 P value

N 8 8 8

Male, n (%) 5(63) 5(63) 5(63)

Age, years 42±13 43±14 42±14 0.99

baseline means

RR, ms 859±169 904±183 910±89 0.77

TQ, ms 452±133 435±125 458±85 0.92

QT, ms 407±38 *1 469±64 452±23 0.03

QTc 442±13 *1,*2 495±27 476±29 <0.01

QT/TQ-ratio 0.99±0.34 1.14±0.22 1.04±0.25 0.57

<1 (n (%)) 5(63) 3(38) 5(63)

Continuous variables are given as mean±SD. Categorical variables are given as frequency (percentage).N= number of subjects; RR= RR-interval; TQ= TQ interval; QT= QT interval; QTc= Bazett corrected QT interval.P values were given for one-way ANOVA on continuous variables. * P<0.05 for Bonferroni post-hoc testing after one-way ANOVA; c= compared to control; 1= compared to LQT1; 2= compared to LQT2.

120 Chapter 7

were similar among groups. The mean QTc intervals were significantly longer in both LQT groups compared to controls. The mean QT intervals were shorter in controls than in the LQT1 group (significantly) and LQT2 group (tendency). Accordingly, the TQ intervals tended to be shorter in LQT1 patients than in controls and LQT2 patients, although dif-ferences were not statistically significant.

beat-to-beat response of TQ and QT

Figure 1 shows three typical beat-to-beat responses of QT and TQ for age and gender matched subjects from the control, LQT1 and LQT2 group. Generally, during stand-up there was a transient fast and large TQ decrease in all subjects, followed by a return (or sometimes overshoot) to baseline TQ values within 30 seconds. In about half of the cases this was followed by a slight decrease in TQ. Conversely, the QT intervals showed a slower and less pronounced response to stand-up. Figure 1 also demonstrates that at baseline the TQ intervals were generally longer than the QT intervals resulting in a QT/TQ-ratio <1. This indicates that the QT interval did cover less than half of the RR interval, which is a rough measure indicating a normal QT interval. In a number of cases (3/8 controls, 5/8 LQT1 patients, 3/8 LQT2 patients) the TQ intervals were shorter than QT (QT/TQ-ratio >1). During stand-up in all subjects the QT/TQ-ratio was larger than 1, which can be considered as QT-stretching. Following stand-up, this QT/TQ-ratio did return to values <1 in many controls (n=4), but remained >1 in most LQT patients (n=15). Therefore,

Control LQT2LQT1

1st su

pine

supine

stand-u

p

standing

2nd st

anding

3rd st

anding0

200

400

600

800

1000

1200

RR (m

s), Q

T (m

s), T

Q (m

s), Q

Tc (-

)

0

200

400

600

800

1000

1200

1st su

pine

supine

stand-u

p

standing

2nd st

anding

3rd st

anding

RR (m

s), Q

T (m

s), T

Q (m

s), Q

Tc (-

)

0

200

400

600

800

1000

1200

1st su

pine

supine

stand-u

p

standing

2nd st

anding

3rd st

anding

RR (m

s), Q

T (m

s), T

Q (m

s), Q

Tc (-

)

0 50 100 150 200 250 300 3500

200

400

600

800

1000

complex no

QT

(ms)

, TQ

(ms)

, QTc

(-)

0 50 100 150 200 250 3000

200

400

600

800

1000

complex no

QT

(ms)

, TQ

(ms)

, QTc

(-)

0 50 100 150 200 250 300 3500

200

400

600

800

1000

complex no

QT

(ms)

, TQ

(ms)

, QTc

(-)

QTTQQTc

RR (mean ± sd)TQ (mean ± sd)QT (mean ± sd)QTc (mean)

A B C

D E F

Figure 1: Top panels: Typical beat-to-beat responses in TQ (green), QT (red) and QTc (grey) for age and gen-der matched subjects from controls (A), LQT1 (b) and LQT2 (C). bottom panels: Mean±SD of RR (blue), TQ (green), QT (red) and QTc (black striped) values per fragment for controls (D), LQT1 (E) and LQT2 (F).


7we defined QT/TQ-crossover as a change in QT/TQ-ratio from <1 at baseline to >1 during standing (i.e. following stand-up). In Figure 1, crossover is observed in panels B and C, but not in panel A. Crossover is also evident when data from each segment are averaged (bottom panels).

Figure 2 shows, per group of subjects, the mean RR, TQ, QT and QTc intervals and the QT/TQ-ratio of the successive fragments. In 1 LQT2 patient data during the first 30 seconds of standing were of insufficient quality. The mean RR and TQ intervals were similar among groups for all fragments. The mean QT intervals, however, significantly differed between controls and LQT1 patients for the fragments at baseline, standing and subsequent fragments. Mean QTc intervals differed between controls and both LQT groups in almost all fragments, but not during stand-up. In all three groups, the mean RR and TQ intervals significantly decreased during stand-up, without significant differ-ence between groups. Data on the changes (baseline minus stand-up) are presented in Table 2. Changes in RR and TQ tended to be largest in LQT2 patients and smallest in LQT1 patients, although these differences between groups were not statistically significant.

COMPUTE FragNr_new

4e fragment staan

3e fragment staan

2e fragment staan

1e fragment staan

1fragmentenvoor staan


Mea

n TQ

inte

rval

(ms) 600

400

200

0

**

*

COMPUTE FragNr_new

4e fragment staan

3e fragment staan

2e fragment staan

1e fragment staan



Mea

n RR

inte

rval

(ms)

1200

1100

1000

900

800

700

600

500

**

*

Di�erence between fragments* = p< 0.05 supine vs stand-up or standing

Di�erences between groups† = p<0.05 ctrl vs LQT1‡ = p<0.05 ctrl vs LQT2# = p<0.05 LQT1 vs LQT2

COMPUTE FragNr_new

4e fragment staan

3e fragment staan

2e fragment staan

1e fragment staan



Mea

n Q

T in

terv

al (m

s)

600

550

500

450

400

350

nsns

ns

† † † †

COMPUTE FragNr_new

4e fragment staan

3e fragment staan

2e fragment staan

1e fragment staan



Mea

n Q

Tc in

terv

al (m

s)

600

550

500

450

400

350

300

* ns

*† †

‡†‡

†‡#

†‡

COMPUTE FragNr_new2e fragment staan

1e fragment staan1 fragmenten voor staan

Mea

n Q

T/TQ

-rat

io (m

s)

3.0

2.5

2.0

1.5

1.0

0.5

*ns

ns

†

baselin

e

stand-u

p

standing

= QT/TQ-ratio = 1

baselin

e

stand-u

p

standing

1st base

line

2nd st

anding

3rd st

anding

baselin

e

stand-u

p

standing

1st base

line

2nd st

anding

3rd st

anding

baselin

e

stand-u

p

standing

1st base

line

2nd st

anding

3rd st

anding

baselin

e

stand-u

p

standing

1st base

line

2nd st

anding

3rd st

anding

LQT2LQT1Healthy relative

A B

C D E

Figure 2: bar graphs: per group mean±SD of RR (A), TQ (b), QT (C), QTc (D) and QT/TQ-ratio (E) per frag-ment during the ECG recording. Differences between baseline and first period of standing (stand-up) or first fragment after standing-up (standing) were tested with a paired t test (*). Differences between groups were tested with a one-way ANOVA test with post-hoc Bonferroni testing (†, ‡, #, difference values given in Table 1).

122 Chapter 7

No significant changes occurred in mean QT intervals when subjects stood up (Figure 2), although LQT2 patients had a tendency for increased QT intervals (i.e. negative ∆QT values for LQT2 contrary to controls and LQT1, Table 2). Consequently, all three groups showed QTc-stretching (i.e. deducible from negative ∆QTc values in Table 2). In LQT2 patients and controls – but not in LQT1 patients – this QTc increase was statistically significant.

Since all observations above were based on average changes (i.e. differences between means, see methods) – which may blur the true dynamics in RR, TQ, QT and QTc – we

Table 2: Dynamic characteristics: mean and maximum changes.


Differences of means (baseline – stand-up)

N 8 8 7

∆RR, ms 165±81 133±97 189±113 0.55

∆TQ, ms 143±73 121±125 188±114 0.48

∆QT, ms 16±20 9±34 −3±31 0.44

∆QTc −32±27 −29±61 −61±46 0.37

∆QT/TQ-ratio(baseline-standing)

−0.09±0.10 −0.16±0.19 −0.17±0.20 0.39

Maximum differences (last beat baseline – max standing)

N 8 8 7

Max ∆RR, ms 250±141 222±150 286±104 0.66

Max ∆TQ, ms 229±132 205±158 289±96 0.47

Max ∆QT, ms 30±26 25±20 18±21 0.56

Max ∆QTc 60±33 68±63 105±36 0.18

Min RR in stand-up, ms 602±83 683±67 674±71 0.08

Min TQ in stand-up, ms 217±80 230±55 215±47 0.88

Min QT in stand-up, ms 373±25 433±68 429±30 0.03

Max QTc in stand-up 501±38 551±79 561±36 0.11

n of max QTc >490 6/8 6/8 7/7

QT at min RR, ms 386±23 453±80 454±39 0.03

∆QT at min RR, ms 17±21 5±36 −8±28 0.27

QT at min TQ, ms 385±23 452±81 456±40 0.03

∆QT at min TQ, ms 18±22 6±37 −9±26 0.22

Values are given as mean±SD.N= number of subjects; n of max QTc >490= number of patients with a QTc >490 during stand-up over the total number of patients.P values were given for one-way ANOVA on continuous variables.* P<0.05 for Bonferroni post-hoc testing after one-way ANOVA, c= compared to control; 1= compared to LQT1; 2= compared to LQT2.# P<0.05 for Fisher’s Exact test on categorical variables


7

also determined maximum changes in RR, TQ, QT and QTc (Table 2) based on our beat-to-beat analysis. Generally, we observed similar patterns as with the average changes. Despite the dissimilar dynamics between LQT1 and LQT2, the maximum QTc during stand-up was in both LQT groups larger compared to controls (not signifi cant). However, in 6/8 controls, the QTc of several beats exceeded the previously established critical level of 490.19 In contrast, the QT/TQ-ratio was more discriminative. In subjects showing at baseline a QT/TQ-ratio<1 (‘normal QTc’), we observed a QT/TQ-crossover in 3/3 LQT1 and 4/5 LQT2, but only in 1/5 controls (Figure 2) with average changes in QT/TQ-ratio of 0.33 ± 0.23, 0.26 ± 0.21 and 0.03, respectively. The QT/TQ-crossover was also discriminative in subjects with a borderline baseline QTc (<465), as all four LQTS patients – and only 1 of 5 controls – with a borderline baseline QTc (< 465) demonstrated a QT/TQ-crossover during standing (P<0.05).

QT-TQ relations

To further evaluate the beat-to-beat response in QT we constructed QT-TQ scatterplots, from which restitution lines were obtained. Figure 3A shows a typical QT-TQ plot for age and gender matched subjects from the control, LQT1 and LQT2 group with diff er-ent colors for diff erent fragments. In all three examples (same subjects as in Figure 1), variation (per fragment) in TQ is larger than that in QT. As a result of standing up there

0 200 400 600 800 1000300

400

500

600

700QT−TQ dataQTc 450 lineQTc 490 linegroup-based restitution

QT = 0.24 x TQ + 290

TQ (ms)

QT

(ms)

0 200 400 600 800 1000300

400

500

600

700

TQ (ms)

QT

(ms)

QT = 0.42 x TQ + 285

0 200 400 600 800 1000300

400

500

600

700

TQ (ms)

QT

(ms)

QT = 0.14 x TQ + 375

Control LQT2LQT1

0 200 400 600 800 1000300

400

500

600

700

TQ (ms)

QT

(ms)

0 200 400 600 800 1000300

400

500

600

700

TQ (ms)

QT

(ms)

0 200 400 600 800 1000300

400

500

600

700

TQ (ms)

QT

(ms)

short-term restitutionlong-term restitution

QT/TQ= 1

supine1supine2standing1standing2standing3standing4

Control LQT2LQT1

A

B

Figure 3: A: Typical QT-TQ graphs for age and gender matched subjects from the control, LQT1 and LQT2 group. Short-term and long-term restitution lines are shown. Also the line for QT/TQ-ratio=1 is shown. b: QT-TQ graphs per group with all subjects pooled with determined group-based restitution line, QTc=450 line and QTc=490 line. Each color indicates a diff erent subject.

124 Chapter 7

is a decrease in TQ (a leftward shift in the plot) followed by a gradual decrease in QT (a downward shift in the plot). All three examples showed QT-stretching during stand-up, which manifests as a little cloud (light blue) outreaching the main cloud toward the left side of the plot.

In the plots we also exemplify the two different ways of restitution determination (short-term and long-term). The short-term restitution lines (thin black solid lines) were rather flat as a result of minor QT changes in response to large TQ changes (Table 3). Slopes of these restitution lines did not differ between groups. The intercept however, was smaller among controls compared to LQT1 patients.

Long-term restitution lines (thick black solid lines) were steeper than short-term restitution lines, because in time larger changes in QT were achieved as response to shortening of the TQ. The LQT1 patients had significantly steeper restitution slopes compared to LQT2 patients (Table 3). Control subjects had slopes that were intermediate between LQT1 and LQT2. The range of TQ intervals varied between subjects and therefore we also calculated a group-based restitution line in which data of all subjects within a group were pooled (Figure 3B). In these plots we observed clear differences is slope and intercept between the groups. The LQT1 patients appeared to have a steeper restitution slope than LQT2 patients and controls, whereas LQT2 patients had a higher restitution intercept than both other groups. These group-based restitution results corresponded with the results on the long-term restitution lines (Table 3 and Figure 3A).

Heart rate corrected QT

Figure 4 shows the reconstructed restitution lines resulting from the group-based de-termination, and the lines of heart rate corrected QT intervals. The lines representing QTc=450 and QTc=490 were recalculated for TQ based on Bazett’s formula. The resti-tution lines demonstrate that for each group of subjects a different type of heart rate correction of QT interval is applicable. In the control group (blue) we observed that for

Table 3: Restitution parameters of QT-TQ plots.


short-term restitution

slope 0.05±0.01 0.06±0.02 0.05±0.01 0.09

intercept 371±27*1 426±56 414±21 0.02

long-term restitution

slope 0.21±0.12 0.32±0.14*2 0.11±0.17 0.04

intercept 303±50 333±71 380±60 0.06

Values are given as mean±SD. Short-term and long-term restitution, see text for further explanation.P values were given for one-way ANOVA on continuous variables.* P<0.05 for Bonferroni post-hoc after one-way ANOVA on continuous variables; c= compared to control; 1= compared to LQT1; 2= compared to LQT2.


7

the main part of TQ range, the restitution line remains below the QTc=450 line. The LQT2 line (green) extends outside the critical margin at high heart rates (small TQ), the LQT1 line (red) exceeds the limits at lower heart rates.

DISCuSSION

This observational study shows that the beat-to-beat response of TQ and QT intervals to a rapid stand-up differ between controls, LQT1 patients and LQT2 patients. The response in QT intervals is more gradual than the large and transient TQ shortening, consequently leading to QTc stretching in the majority of cases of all three groups. We propose that the use of QT/TQ-crossover metric is of discriminative value for the identification of LQT patients without a prolonged QTc at rest, as these patients fail to return their QT/TQ-ratio below 1 after standing. The slow adaption of QT to TQ shortening is underlined by the flat short-term restitution lines observed in all three groups. Over time, QT adaptation is more prominent and leads to steeper restitution lines demonstrating the differences between the groups. LQT1 patients show steeper restitution lines whereas LQT2 patients show a larger restitution intercept.

Value of supine-standing test

Rapid stand-up was accompanied by a transient and significant shortening of the TQ and RR intervals. The corresponding response in QT is however more gradual. This behavior of TQ, RR and QT following standing is consistent with earlier reports,18,19 although the baseline QT and QTc values were ±20 ms longer in our study. However, changes in QT

0 200 400 600 800 1000300

350

400

450

500

550

600

650

700

TQ (ms)

QT

(ms)

ControlLQT1LQT2QTc=450QTc=490

QT = 0.24 x TQ + 290QT = 0.42 x TQ + 285QT = 0.14 x TQ + 375

Figure 4: Reconstructed group-based restitution lines together with QTc=450 line and QTc=490 line.

126 Chapter 7

intervals upon standing (shortening in controls and LQT1 and prolongation in LQT2) were consistent with these studies.18 The increases in QTc were smaller in our study. Discrepancies in results between our study and our earlier studies18,19 may be explained by some methodological differences: i) for QT interval determination we used signal averaging and all 9 leads together instead of using single selected complexes in a single lead, ii) we obtained beat-to-beat data during the entire recording split into fragments instead of selecting a couple of single complexes throughout the recording, iii) we had a lower percentage of females in our LQT groups and iv) patient selection was not based on a borderline QTc value.

The different dynamics of the TQ and QT intervals following a rapid stand-up may lead to distinction between controls, LQT1 patients and LQT2 patients. All three groups have a similar minimum TQ during stand-up, indicating an equal ‘QT adapting’ challenge among the groups. During this challenge, QT interval in controls and LQT1 patients shorten, whereas LQT2 patients even show a small prolongation of the QT interval. As a result of the gradual QT changes in response to a large and immediate TQ decrease – represent-ing an increase in heart rate – QTc values are prolonged and result in QTc-stretching in all three groups. Although this may be interpreted as a faulty adaptation of QT to heart rate changes, QTc-stretching was also observed in control subjects, as expected.18 It should be kept in mind, however, that within each episodes of TQ change, hardly any adapta-tion of QT can be observed (Figure 1 and 3A). QTc-stretching is therefore the normal expression of the dynamicity of heart rate adaptation (in the absence of QT adaptation) following standing up,21 but tends to be augmented in LQT2 patients (Table 2).

LQT2 patients showed the largest QTc-stretching during stand-up. However, in many controls maximum QTc during stand-up also exceeded the cut-off of 490 defined by Adler and colleagues.19 This may lead to a misdiagnosis of healthy relatives. Therefore, we suggest that caution should be exerted when using single QTc values during such a dynamic condition, and that an average of several successive QTc values may be more revealing.

We demonstrated that the use of QT/TQ-crossover might be of added discriminative value for the detection of LQT patients with a borderline QTc at rest (Figure 2). Our study is not designed, nor powered to obtain a positive or negative predictive value but has identified a potentially valuable metric based on the detailed beat-to-beat analysis of the QT response to heart rate changes. The selection of patients in our study was not based on the identification of patients with a normal QTc at rest.

Restitution characteristics

The short-term restitution lines, representing the beat-to-beat dynamic character, have flat slopes that are similar among groups. It indicates that variation in TQ is much larger than the variation in QT. The long-term restitution lines, by contrast, represent the


7

‘steady-state’ adaptation in QT and TQ. We observed that long-term restitution slopes were steeper in LQT1 compared to control (not significant) and LQT2 (significant). This difference is probably related to the loss of IKs in LQT1 patients.9 IKs accumulates during high heart rates22 and is therefore primarily responsible for adaptation to heart rate changes (i.e. restitution slope). Therefore we anticipated that a change in the restitu-tion slope in LQT1 patients is plausible. Still, the observed steeper slope in LQT1 patients would imply better adaptation of QT to TQ, which is opposite to what we expected to observe during a loss of IKs (i.e. a loss of restitution). Longer baseline QT interval values might have played a role in this discrepancy.

LQT2 has been associated with a loss of function10 of the rapidly activating compo-nent of the delayed rectifier potassium current (IKr).23 Due to loss of repolarizing current through IKr, which is relatively heart rate independent, the restitution line is expected to show an upward shift leading to a larger restitution intercept. Indeed, LQT2 patients have the tendency of a higher restitution intercept (Table 3).

The differences in restitution lines between the groups suggest that a group-specific heart rate correction formula should be used to obtain correct QTc values and that the use of the Bazett formula in such dynamic conditions may lead to overestimation or underestimation of the QTc value (see also Figure 4). In the LQT1 group the restitution mainly deviates from normal at low heart rates (long TQ intervals), whereas in the LQT2 group it mainly deviates from normal at high heart rates (short TQ intervals). The latter finding fits in the results of earlier studies that showed that QT-stretching (i.e. at high heart rates) is particularly abnormal in LQT2 patients.18 When using the Bazett formula in these cases, true QTc may be underestimated at slow heart rates and overestimated at high heart rates. Batchvarov and co-workers showed that even between subjects QT-RR relationships may differ substantially.24 Rather than presenting QTc values during rapidly changing conditions, we argue that it is preferable to give several absolute QT and TQ or RR values at different stages.

Methodological considerations

A major drawback of this study is that selection of patients was not based on a normal baseline QTc value at rest. Therefore, the QT/TQ-crossover metric needs further explora-tion before its definite diagnostic value could be ascertained. Another methodological consideration is that the use of a linear regression model to determinate QT-restitution may not completely match the physiologic restitution.25 However, Batchvarov and col-leagues have shown that a linear model have the most optimal fit in QT-RR data when compared to a parabolic or hyperbolic model.24 In addition, the linear regression lines allow easy interpretation of the slopes and intercepts.

128 Chapter 7

CONCLuSION

Following rapid stand-up the TQ interval shows a transient and large decrease as a result of the increase in heart rate. The response in QT is much smaller, more gradual and dif-fers between controls, LQT1 patients and LQT2 patients. As a direct consequence, QTc in-tervals become suddenly longer (QT-stretching). Restitution slopes differ among groups, with steeper slopes in LQT1 patients and slightly higher intercepts in LQT2 patients. In order to classify an individual within a group accurately, taking the dynamicity during the supine-standing test into account is important. Comparison of a single QTc value with a predefined level may therefore not be accurate enough. Determination of the QT/TQ-crossover or several QT and TQ values during the supine-standing test is potentially of added discriminative value, especially in patients with a normal QTc at rest.

ACkNOWLEDGEMENTS

We thank Dr. C. van der Werf, Dr. L.A. Olde Nordkamp and Drs. S.P.J. Krul for performing continuous ECG recordings during the supine-standing tests.


7

REFERENCES

1. Moss AJ, Schwartz PJ, Crampton RS, Tzivoni D, Locati EH, MacCluer J, Hall WJ, Weitkamp L, Vincent GM, Garson A. The long QT syndrome. Prospective longitudinal study of 328 families. Circulation 1991; 84: 1136–1144.

2. Schwartz PJ, Periti M, Malliani A. The long Q-T syndrome. Am Heart J 1975; 89: 378–390.

3. Priori SG, Wilde A a., Horie M, et al. HRS/EHRA/APHRS Expert Consensus Statement on the Di-agnosis and Management of Patients with Inherited Primary Arrhythmia Syndromes: Document endorsed by HRS, EHRA, and APHRS in May 2013 and by ACCF, AHA, PACES, and AEPC in June 2013. Hear Rhythm 2013; 10: 1932–1963.

4. Schwartz PJ, Ackerman MJ. The long QT syndrome: A transatlantic clinical approach to diagnosis and therapy. Eur Heart J 2013; 34: 3109–3116.

5. Vincent GM, Abildskov JA, Burgess MJ. Q-T interval syndromes. Prog Cardiovasc Dis 1974; 16: 523–530.

6. Schwartz PJ, Priori SG, Cerrone M, et al. Left cardiac sympathetic denervation in the management of high-risk patients affected by the long-QT syndrome. Circulation 2004; 109: 1826–1833.

7. Keating M, Atkinson D, Dunn C, Timothy K, Vincent GM, Leppert M. Linkage of a cardiac arrhyth-mia, the long QT syndrome, and the Harvey ras-1 gene. Science 1991; 252: 704–706.

8. Mizusawa Y, Horie M, Wilde AA. Genetic and Clinical Advances in Congenital Long QT Syndrome. Circ J 2014; 78: 2827–2833.



11. Ackerman MJ, Priori SG, Willems S, et al. HRS/EHRA expert consensus statement on the state of genetic testing for the channelopathies and cardiomyopathies. Europace 2011; 13: 1077–1109.

12. Taggart NW, Haglund CM, Tester DJ, Ackerman MJ. Diagnostic miscues in congenital long-QT syndrome. Circulation 2007; 115: 2613–2620.

13. Hofman N, Wilde A a M, Kääb S, van Langen IM, Tanck MWT, Mannens MM a M, Hinterseer M, Beck-mann B-M, Tan HL. Diagnostic criteria for congenital long QT syndrome in the era of molecular genetics: do we need a scoring system? Eur Heart J 2007; 28: 575–580.

14. Horigome H, Ishikawa Y, Shiono J, Iwamoto M, Sumitomo N, Yoshinaga M. Detection of extra components of T wave by independent component analysis in congenital long-QT syndrome. Circ Arrhythm Electrophysiol 2011; 4: 456–464.

15. Struijk JJ, Kanters JK, Andersen MP, Hardahl T, Graff C, Christiansen M, Toft E. Classification of the long-QT syndrome based on discriminant analysis of T-wave morphology. Med Biol Eng Comput 2006; 44: 543–549.


130 Chapter 7

17. Hinterseer M, Thomsen MB, Beckmann B-M, Pfeufer A, Schimpf R, Wichmann H-E, Steinbeck G, Vos M a, Kaab S. Beat-to-beat variability of QT intervals is increased in patients with drug-induced long-QT syndrome: a case control pilot study. Eur Heart J 2008; 29: 185–190.


19. Adler A, van der Werf C, Postema PG, et al. The phenomenon of “QT stunning”: the abnormal QT prolongation provoked by standing persists even as the heart rate returns to normal in patients with long QT syndrome. Heart Rhythm 2012; 9: 901–908.

20. Ritsema van Eck HJ. Fiducial segment averaging to improve cardiac time interval estimates. J Electrocardiol 2002; 35 Suppl: 89–93.

21. Fossa A a, Zhou M. Assessing QT prolongation and electrocardiography restitution using a beat-to-beat method. Cardiol J 2010; 17: 230–243.

22. Jost N, Papp JG, Varró A. Slow delayed rectifier potassium current (IKs) and the repolarization reserve. Ann Noninvasive Electrocardiol 2007; 12: 64–78.

23. Sanguinetti MC, Jiang C, Curran ME, Keating MT. A mechanistic link between an inherited and an acquired cardiac arrhythmia: HERG encodes the IKr potassium channel. Cell 1995; 81: 299–307.

24. Batchvarov VN, Ghuran A, Smetana P, Hnatkova K, Harries M, Dilaveris P, Camm a J, Malik M. QT-RR relationship in healthy subjects exhibits substantial intersubject variability and high intrasubject stability. Am J Physiol Heart Circ Physiol 2002; 282: H2356–H2363.

25. Franz MR. The Electrical Restitution Curve Revisited: . Steep or Flat Slope-Which is Better? J Cardio-vasc Electrophysiol 2003; 14: S140–S147.

Chapter 8

Summary and future perspectives

Summary and future 133

8

THE T WAVE: PHYSIOLOGY AND PATHOPHYSIOLOGY

The T wave on the electrocardiogram (ECG) represent the repolarization of the ventricu-lar myocardium. Abnormalities in repolarization and abnormal T waves are related with a higher risk for life-threatening arrhythmias.1,2,3,4 However, a thorough understanding of the T wave in relation with the underlying repolarization pattern is lacking. The aim of this thesis was to elucidate some aspects of the physiology and pathophysiology under-lying the T wave and to explore the effects of various factors (pressure, drugs, autonomic nervous system, positional changes) on the morphology of the T wave.

PHYSIOLOGY uNDERLYING THE T WAVE

We show in Chapter 2 that, even for the normal T wave, there is a hiatus in the relation between T wave morphology and the corresponding repolarization pattern. We provide the first complete correlation between local repolarization and the corresponding elec-trocardiographic T wave. Repolarization heterogeneity in the entire heart, rather than the heterogeneity along a single anatomic axis is determinative for the T wave.

The novelty in this chapter is that it demonstrates that repolarization is a nonlinear process in which at the moment of the peak of the T wave only 25% of ventricular sites is repolarized. From these results we infer that the first part of the normal T wave primarily depends on differences in action potential morphology (i.e. due to phase 1 and 2 of the action potential). Accordingly, the last part of the T wave is mainly explained by differences in the steepest part of repolarization (phase 3). We speculate that abnor-malities in the part of the T wave following the peak may indicate relevant changes in heterogeneity in the end of repolarization, which have implication on the vulnerability for arrhythmias. Indeed a prolonged Tpeak-to-Tend interval has been associated with sudden cardiac death.3 On the other hand, we realize that repolarization heterogeneity is not the only prerequisite to initiate re-entry.5

The normal T wave has a longer onset-to-peak duration than peak-to-end duration. Based on our results we infer that it represents a longer interval of differences in the action potential (AP) plateau phase than of differences in full repolarization moments. The symmetry of the T wave may therefore reveal information about the course of the AP plateau. We speculate that when the AP morphology starts to resemble a block pulse morphology, the interval in which AP plateau differences occur becomes shorter, where-as the interval in which differences in full repolarization occur will remain unchanged. This may result in a symmetric T wave. Indeed during hyperkalemia phase 3 of the AP is steeper6 and symmetric T waves are present.7

134 Chapter 8

Although the results were obtained from pig hearts, the activation and repolarization patterns and the T wave resemble those described in humans considerably.8,9,10 Never-theless, in order to affirm that the mechanism proposed in this thesis holds for T waves in humans, experiments needs to be repeated in human hearts.

MODuLATION bY LEFT VENTRICuLAR PRESSuRE

Stretch influences the action potential duration and plays a relevant role in arrhythmo-genesis.11 This modulation is known as mechano-electric coupling (MEC) and results from the activation of stretch-activated channels in the cell membrane.12 Since many studies have focused on the stretch effect in diseased conditions, we were interested in the MEC effect of physiologic pressures within the normal heart. Chapter 3 presents the results of the influence of physiologic left ventricular (LV) pressure on repolarization and action potential duration. The novelty of this study is that we stimulated from either the LV free wall or septum in order to change the timing of activation (and thus repolarization) with respect to the pressure pulse, without altering the myocardial substrate (i.e. electrical and mechanical properties). Results in this chapter demonstrates that the presence of LV pressure synchronizes repolarization in the heart, by prolonging the repolarization of intrinsically early repolarizing cardiomyocytes more than the intrinsically late repolarizing cardiomyocytes. The latter may even show shortening of the repolarization when the cycle length is short. These results implicate that systolic pressures – within the physiological range – reduce heterogeneity in repolarization via mechano-electric coupling. We sug-gest that this will narrow the T wave (see Figure 1) and may protect the heart against arrhythmias. In the absence of LV pressure, repolarization is shorter and dispersion in repolarization is larger, which are both critical conditions for the initiation of re-entry.13 Systolic pressure is commonly low during a short-coupled premature beat and may there-fore – via mechano-electric coupling – broaden the T wave and increase the vulnerability for arrhythmias, especially in the already compromised heart. This implies that a second premature beat is more arrhythmogenic than a first premature beat. Janse et al have tested the assumption that a second premature beat is associated with more heterogene-ity in refractory periods.14 This was not the case, but these experiments were executed in Langendorff perfused porcine hearts, where left ventricular pressure is absent. Timing of the pressure pulse with regards to the moment of repolarization is also important in the extent of prolongation of repolarization. This may be relevant in diseases with a potentially shifted timing of the pressure pulse with respect to the repolarization moment.

The majority of studies in literature only describes MEC effects on action potential dura-tion and repolarization, but it is also relevant to know how the stretch-induced changes


8

in repolarization are translated to the ECG. Chapter 4 describes the impact of LV pres-sure on the STT segment of the ECG and on the underlying repolarization. The results demonstrated that increase in LV pressure causes little prolongation of repolarization in the LV, but no prolongation in the RV, resulting in larger interventricular dispersion in RT. This led to large increases in the amplitudes of the STT integrals and QT intervals with-out changing the pattern of the STT integral maps. Prolongation of the QT interval was larger than prolongation of the repolarization. This discrepancy implies that the increase in pressure changes primarily the morphology at the end of the action potential, which is consistent with the AP prolongation resembling an afterdepolarization as described by Franz et al.15 We conclude that increased diastolic pressure leads to concomitant rise in systolic pressure and results in a small prolongation of repolarization times. In addi-tion, a slightly larger prolongation of the QT intervals and increases in the amplitude of the STT segment on the ECG occurs (see Figure 1).

∆RT∆RT &

∆APplateau

Pdias LV

Psys LV dRTinterLV lateral

GNa

Control

loss of IKs

LQT1 loss of IKr

LQT2

HR

IKr

blockade

TQ

= LV= RV= RT

preceding QT interval

Figure 1: Proposed physiology and pathophysiology of the T wave. LV= left ventricle; RV= right ventricle; RT= repolarization time; ∆APplateau= differences in action potential (AP) plateau; ∆RT = differences in RT (indicated by the dots in the AP); GNa= sodium current conductivity; Psys= systolic pressure; Pdias= diastolic pressure; dRTinter= interventricular dispersion in RT; IKr= rapidly activating component of the delayed recti-fier potassium current; IKs= slowly activating component of the delayed rectifier potassium current; HR= heart rate

136 Chapter 8

Considering both studies on LV pressure effects, I suggest that the T wave morphology may be of added value in cardiac resynchronization therapy (CRT). In patients with heart failure, CRT is focused on synchronizing the contraction of the heart in order to increase the ejection fraction.16 The synchronization may, however, result in extreme strain pat-terns within the heart that unintentionally increase the vulnerability for arrhythmias. The T wave morphology may be indicative for these extreme strain patterns and conse-quently predicting the risk for arrhythmias. CRT has already been shown to induce ar-rhythmias.17 Also, an increased Tpeak-to-Tend interval was shown to be an independent predictor of appropriate ICD therapy.18 A detailed analysis of the relation between T wave morphology and strain patterns within heart failure patients is needed to determine its predictive value for arrhythmias. In this exploration the activation sequence should also be incorporated as this influences the T wave morphology considerably.

PATHOPHYSIOLOGY uNDERLYING THE T WAVE

In this thesis we have discussed some ECG abnormalities related to abnormal repolariza-tion. In Chapter 5 we try to unravel the mechanism underlying inferolateral J-waves. In the literature there is controversy whether the mechanism underlying the genesis of inferolateral J-waves is based on depolarization or repolarization. The results of this chapter provide important support for the depolarization hypothesis, because we demonstrate that conduction slowing in the LV lateral region can induce J-waves on the ECG. In the same model a regional increase in transient outward current is less provoca-tive. However, our data do not completely exclude the role of early repolarization or a combination of conduction and repolarization changes. We have also tested tissue con-ductivity as a contributor to J-waves. A reduced tissue conductivity – e.g. due to cellular uncoupling – results in a lower current generated by the activation wave front and will cause a smaller potential field on the torso. Diffuse fibrosis causes conduction slowing and should therefore be able to induce J-waves. However, the concomitant reduction in tissue conductivity reduces the likeliness of inducing J-waves. Based on these results we suggest that conduction slowing can induce inferolateral J-waves and that its genesis may depend on tissue conductivity, size and location of the region within the heart that is involved. The model did not provide an explanation for arrhythmogenesis in the presence of J-waves caused by conduction slowing. Conduction slowing is one of the requirements for the initiation of re-entry.19 Therefore it is not unlikely that the regional conduction slowing may facilitate re-entry.

The thesis also provides insights in the electrocardiographic abnormalities in long QT syndrome (LQTS). Chapter 6 presents an explanation for the bifid morphology of the


8

T wave in type 2 LQTS. An increased interventricular dispersion of repolarization induces bifid T waves on the ECG, with earlier repolarization in the right ventricle and later repo-larization in the left ventricle. Although the data were obtained from a dog model with a drug-induced LQT2 syndrome, we have arguments that suggest a similar mechanism in human. Firstly, the bifid T wave morphology is typical for both congenital20 and ac-quired21 LQTS in human. Besides, the configuration of the bifid T wave peaks observed in humans20 (i.e. larger first T wave peak in right precordial leads and larger second T wave peak in left precordial leads) corresponds with our suggestion that the first peak of the bifid T wave reflects (earlier) repolarization of the RV and that the second peak of the bifid T wave reflects (later) repolarization of the LV. We also showed in this chapter that the width of the T wave – as surrogate for the time interval between bifid T wave peaks – is highly correlated with the interventricular dispersion in repolarization. It would be of great interest to investigate whether the time interval between the bifid T wave peaks may also be correlated with the risk for arrhythmias in a clinical study.

Chapter 7 describes the beat-to-beat response of the QT and TQ (= diastolic interval on ECG) intervals to rapid stand-up in LQT1 and LQT2 patients and controls. We dem-onstrate a transient and large TQ shortening in response to an increase in heart rate, which was not different between LQT patients and controls. The gradual and smaller response in QT differs between groups, with QT shortening in the controls more than in LQT1 patients and with minor QT shortening or even prolongation in LQT2 patients. As a direct consequence, QTc intervals become suddenly longer (QT-stretching) in all groups in the first period after standing. I argue that QT-stretching is a poor denomination of the physiological consequence of standing up from a recumbent position, because it represents a lack of immediate adaptation of the QT interval to the change in heart rate rather than an abnormal prolongation of the QT interval. Viskin et al22 also commented on this point in an earlier study.

We describe that the period following the acute TQ shortening may be more discrimi-native between groups. Based on these results we propose that the QT/TQ-crossover metric, or several QT and TQ values, during a supine-standing test may be of potential added discriminative value for the diagnosis of LQTS, especially in patients with a normal QTc at rest. We should, however, be cautious in applying this metric directly to the clinic, because individuals in our study groups were not selected on the basis of a normal QTc value at baseline. With this study we would like to express our concerns about the use of QTc values in dynamic conditions, as the QTc is typically based on steady-state condi-tions.23

Chapter 7 also presents differences in restitution lines between LQT patients and controls. We conclude that the restitution slopes are steeper in LQT1 patients compared to LQT2 patients and controls. A restitution slope larger than 1 has been associated with

138 Chapter 8

an increased risk for ventricular arrhythmias.24 Coronel et al showed, however, that the critical value of 1 for the restitution slope is not required to induce re-entry.5 The combi-nation of conduction delay and the restitution characteristics of the earlier repolarizing tissue is critical for the initiation of re-entry.5 This implies that also for LQT patients an increased restitution slope in itself would only be arrhythmogenic when also a certain activation delay occurs.

TRANSMuRALITY AND THE T WAVE

Many cardiology textbooks describe the T wave as the result of transmural dispersion of repolarization with the epicardial layers repolarizing earlier than the endocardial layer. In this thesis we demonstrate that in an intact heart transmural dispersion of repolariza-tion is small and only plays a minor role in the genesis of the T wave. Previously, others have examined the role of transmurality, and also concluded that transmural disper-sion in repolarization and action potential duration in the intact heart is small.25,26,27,28,29 Accordingly, they suggested that it would contribute little to the origin of the T wave. We recorded repolarization gradients throughout the entire heart simultaneously with the T wave on an ECG and the results confirm this suggestion. Therefore, I propose that textbooks need to be reviewed concerning the description of the T wave genesis and it should be considered to incorporate insights inferred from this thesis.

IMPLICATIONS FOR THE T WAVE MORPHOLOGY

Figure 1 presents a summarizing representation of the physiology and pathophysiol-ogy underlying the T wave derived from the studies presented in this thesis. The figure shows a normal T wave (solid black line), with a longer onset-to-peak interval than a peak-to-end interval. The former interval (light grey area) is mainly created by differ-ences in action potential shape (∆AP) and in less extent by differences in full repolariza-tion moments (∆RT). The Tpeak-to-Tend interval (dark grey area) is mainly the result of differences in the full repolarization moments of the majority (about 75%) of the heart.

Pressure can modulate the T wave. Loss of a relevant systolic LV pressure (Psys LV) will generate more heterogeneous repolarization within the heart resulting in a wider T wave (light green area). An increase in diastolic LV pressure (Pdias LV) will mainly influ-ence the end of the action potential morphology leading to little prolongation of the repolarization. This will result in longer QT intervals (due to a longer ‘tail’) and a larger T wave amplitude (dark green area).


8

Conduction slowing – e.g. as result of a reduced sodium conductivity (GNa) – in the LV lateral region of the heart may induce J-waves (notch or slur) in the inferolateral leads of an ECG (blue line). The J-wave amplitude will depend on the tissue conductivity within the heart, in which a lower tissue conductivity may reduce the amplitude.

Interventricular dispersion in repolarization in the heart contributes to a bifid morphology of the T wave. A loss of IKr – in LQT2 patients – may lead to an increased interventricular dispersion in repolarization (dRTinter), which results in a bifid T wave (red line). The first peak of the bifid T wave will be generated by repolarization in the RV and the second peak will be generated by repolarization in the LV.

In LQT patients the dynamics in QT in response to an increase in heart rate is impaired. In the normal heart (black striped line) shortening of the TQ interval (i.e. diastolic interval on the ECG) will cause an appropriate QT shortening. With a similar shortening of the TQ interval, LQT1 patients have a moderate QT shortening (black dot-striped line) and LQT2 patients have a minor QT shortening and may even show minor QT prolongation (black dotted line). The differences in T wave morphology between LQT1 and LQT2 were not incorporated in the figure.

140 Chapter 8

REFERENCES






6. Yan G-X, Lankipalli RS, Burke JF, Musco S, Kowey PR. Ventricular repolarization components on the electrocardiogram. J Am Coll Cardiol 2003; 42: 401–409.

7. Tarail R. Relation of abnormalities in concentration of serum potassium to electrocardiographic disturbances. Am J Med 1948; 5: 828–837.






13. Robert E, Aya AG, de la Coussaye JE, Péray P, Juan JM, Brugada J, Davy JM, Eledjam JJ. Dispersion-based reentry: mechanism of initiation of ventricular tachycardia in isolated rabbit hearts. Am J Physiol 1999; 276: H413–H423.

14. Janse MJ, Capucci A, Coronel R, Fabius MAW. Variability of Recovery of Excitability in the Normal Canine and the Ischaemic Porcine Heart. Eur Heart J 1985; 6: 41–52.


16. Brignole M, Auricchio A, Baron-Esquivias G, et al. 2013 ESC Guidelines on cardiac pacing and car-diac resynchronization therapy: the Task Force on cardiac pacing and resynchronization therapy of the European Society of Cardiology (ESC). Developed in collaboration with the European Heart Rhythm Association. Eur Heart J 2013; 34: 2281–2329.


8

17. Roque C, Trevisi N, Silberbauer J, et al. Electrical storm induced by cardiac resynchronization therapy is determined by pacing on epicardial scar and can be successfully managed by catheter ablation. Circ Arrhythm Electrophysiol 2014; 7: 1064–1069.

18. Lellouche N, De Diego C, Akopyan G, Boyle NG, Mahajan A, Cesario D a, Wiener I, Shivkumar K. Changes and predictive value of dispersion of repolarization parameters for appropriate therapy in patients with biventricular implantable cardioverter-defibrillators. Heart Rhythm 2007; 4: 1274–1283.

19. Mines GR. On dynamic equilibrium in the heart. J Physiol 1913; 46: 349–383.





24. Riccio ML, Koller ML, Gilmour RF. Electrical restitution and spatiotemporal organization during ventricular fibrillation. Circ Res 1999; 84: 955–963.






Chapter 9

Samenvatting en toekomstperspectieven

Samenvatting en toekomst 145

9

DE T-GOLF: FYSIOLOGIE EN PATHOFYSIOLOGIE

De T-golf op het elektrocardiogram (ECG) is een weergave van de repolarisatie van de ventrikels. Afwijkingen in de repolarisatie en abnormale T-golven zijn geassocieerd met een hoger risico op levensbedreigende hartritmestoornissen.1,2,3,4 Er ontbreekt echter gedetailleerde kennis over de T-golf in relatie met het onderliggende repolarisatiepa-troon. Dit proefschrift heeft als doel om sommige aspecten van de fysiologie en pa-thofysiologie die ten grondslag liggen aan de T-golf op te helderen en te onderzoeken wat de effecten van verscheidene factoren (druk, medicijnen, autonome zenuwstelsel, positionele wijzigingen) op de T-golf morfologie zijn.

FYSIOLOGIE ONDERLIGGEND AAN DE T-GOLF

In Hoofdstuk 2 laten we zien dat, zelfs voor de normale T-golf, er een hiaat bestaat in de relatie tussen de T-golf morfologie en het onderliggende repolarisatiepatroon. Wij beschrijven de eerste volledige correlatie tussen lokale repolarisatie en de bijbehorende elektrocardiografische T-golf. Repolarisatieheterogeniteit in het gehele hart, en niet zozeer de heterogeniteit langs een enkele anatomische as, is bepalend voor de T-golf.

De noviteit in dit hoofdstuk is dat het laat zien dat repolarisatie een niet-lineair proces is waarin op het moment van de top van de T-golf slechts 25% van het ventrikel is gerepo-lariseerd. Vanuit deze resultaten concluderen wij dat het eerste deel van de normale T-golf voornamelijk afhankelijk is van verschillen in de actiepotentiaalmorfologie (d.w.z. als re-sultaat van fase 1 en 2 van de actiepotentiaal). Het laatste deel van de T-golf wordt daarom voornamelijk verklaard door verschillen in het steilste deel van de repolarisatie (fase 3). Wij veronderstellen dat afwijkingen in het gedeelte na de top van de T-golf kunnen wijzen op relevante veranderingen in de heterogeniteit in fase 3 repolarisatie. Deze veranderingen hebben invloed op de gevoeligheid voor ritmestoornissen. Een langer Ttop-tot-Teind-interval is inderdaad geassocieerd met plotse hartdood.3 Aan de andere kant, beseffen we dat repolarisatieheterogeniteit niet de enige voorwaarde is voor het initiëren van re-entry.5

De normale T-golf heeft een langere begin-tot-piek duur dan een piek-tot-eind duur. Op basis van onze resultaten leiden we af dat het interval van verschillen in de actiepotentiaal(AP)-plateaufase langer is dan die van de verschillen in volledige repo-larisatiemomenten. De symmetrie van de T-golf kan daarom informatie geven over het verloop van het AP-plateau. Wij veronderstellen dat als de morfologie van de AP meer op een blokgolf lijkt, het interval van AP-plateauverschillen zal verkorten, terwijl het interval waarin verschillen in volledige repolarisatie optreedt onveranderd zal blijven. Dit kan resulteren in een symmetrische T-golf. Gedurende een hyperkaliëmie is fase 3 van de AP inderdaad steiler6 en zijn er symmetrische T-golven aanwezig.7

146 Chapter 9

Hoewel de resultaten werden verkregen uit varkensharten, lijken de activatie- en repolarisatiepatronen en de T-golf behoorlijk op hetgeen beschreven is bij de mens.8,9,10 Echter, om te bevestigen dat het mechanisme dat beschreven is in dit proefschrift ook geldt voor T-golven bij de mens, is het nodig om de experimenten in menselijke harten te herhalen.

MODuLATIE DOOR LINkERVENTRIkELDRuk

Rek beïnvloedt de actiepotentiaalduur en speelt een belangrijke rol in de ontwikke-ling van ritmestoornissen.11 Deze modulatie is bekend als mechano-elektrische kop-peling (MEC) en ontstaat door rek-geactiveerde kanalen die zich in de celmembraan bevinden.12 Omdat veel studies zich richtten op de rekeffecten in zieke harten, waren wij geïnteresseerd in het MEC effect van fysiologische druk binnen het normale hart. Hoofdstuk 3 toont de resultaten van de invloed van fysiologische linkerventrikel (LV) druk op repolarisatie en actiepotentiaalduur. De nieuwigheid van deze studie is dat wij de LV vrije wand of het septum stimuleerden opdat de timing van de activatie (en tevens repolarisatie) ten opzichte van de drukgolf veranderde, waarbij het myocardiale sub-straat (d.w.z. elektrische en mechanische eigenschappen) onveranderd bleef. Resultaten in dit hoofdstuk tonen aan dat de aanwezigheid van LVdruk repolarisatie in het hart synchroniseert. Deze synchronisatie vindt plaats doordat de repolarisatie van de intrin-sieke vroeg-repolariserende cardiomyocyten meer verlengt wordt dan de repolarisatie van de intrinsieke laat-repolariserende cardiomyocyten. Bij een korte cycluslengte kan dit in de intrinsieke laat-repolariserende cardiomyocyten zelfs tot een verkorting van de repolarisatie leiden. Deze resultaten impliceren dat systolische druk, binnen fysiolo-gische grenzen, heterogeniteit in repolarisatie reduceert middels mechano-elektrische koppeling. Wij suggereren dat dit de T-golf zal versmallen (zie Figuur 1) en het hart kan beschermen tegen ritmestoornissen. Het ontbreken van een LVdruk verkort de repolari-satie en vergroot de dispersie in repolarisatie, welke beiden kritische voorwaarden zijn voor de initiatie van re-entry.13 De systolische druk is vaak laag bij een kortgekoppelde premature slag en kan derhalve, middels mechano-elektrische koppeling, de T-golf verbreden en de gevoeligheid voor ritmestoornissen verhogen. Dit vormt vooral een risico in een reeds aangedaan hart. Het impliceert dat een tweede premature slag meer aritmogeen is dan een eerste premature slag. Janse et al hebben getoetst of een tweede premature slag geassocieerd is met meer heterogeniteit in de refractaire perioden.14 Dit bleek niet het geval te zijn, maar dat valt te verklaren door het gegeven dat deze experi-menten werden uitgevoerd in Langendorff geperfundeerde varkensharten, waarbij een linkerventrikeldruk afwezig is. De timing van de drukgolf ten opzichte van het moment van repolarisatie is ook belangrijk voor de mate van verlenging van repolarisatie. Dit kan


9

relevant zijn in ziektes waarbij de timing van de drukgolf mogelijk verschoven is ten opzichte van de repolarisatiemomenten.

De meerderheid van de studies beschrijft enkel MEC effecten op de actiepotentiaalduur en repolarisatie, maar het is ook relevant om te weten hoe de door rek-geïnduceerde veranderingen in repolarisatie zijn te vertalen naar het ECG. Hoofdstuk 4 beschrijft de invloed van LVdruk op zowel het STT segment van het ECG als ook op de onderlig-gende repolarisatie. De resultaten laten zien dat toename in LV druk resulteert in een geringe verlening van de repolarisatie in de LV, maar niet tot een verlenging in de RV, wat leidt tot grotere interventriculaire dispersie in RT. Dit leidde tot grote toenames in de amplitude van de STT integralen en QT intervallen, zonder dat het patroon van de STT integraalkaarten veranderde. Verlenging van het QT interval was groter dan de verlenging van de repolarisatie. Deze discrepantie impliceert dat de toename van de druk hoofdzakelijk de morfologie aan het einde van de actiepotentiaal verandert. Dit

∆RT∆RT &

∆APplateau

Pdias LV

Psys LV dRTinterLV lateraal

GNa

Controle

verlies van IKs

LQT1 verlies van IKr

LQT2

HF

IKr

blokkade

TQ

= LV= RV= RT

voorgaand QT interval

Figuur 1: Voorgestelde fysiologie en pathofysiologie van de T-golf. LV= linkerventrikel; RV= rechterventri-kel; RT= repolarisatietijd; ∆APplateau = verschillen in actiepotentiaal(AP)-plateau; ∆RT = verschillen in re-polarisatiemomenten (aangeduid met de stippen in de AP); GNa= geleidbaarheid van natriumstroom; Psys= systolische druk; Pdias= diastolische druk; dRTinter= interventriculaire dispersie in RT; IKr= snel activerende component van de ‘delayed rectifier’ kaliumstroom; IKs= traag activerende component van de ‘delayed rec-tifier’ kaliumstroom; HF= hartfrequentie

148 Chapter 9

komt overeen met de AP-verlenging die lijkt op een afterdepolarisatie zoals beschreven door Franz et al.15 Wij concluderen dat een verhoogde diastolische druk eveneens de systolische druk laat stijgen en een kleine verlenging van repolarisatiemomenten tot gevolg heeft. Bovendien resulteert dit in een iets grotere verlenging van de QT interval-len en een toename van de amplitude van het STT segment op het ECG (zie Figuur 1).

Beide studies over de LVdruk effecten in overweging nemende, suggereren wij dat de T-golf morfologie van toegevoegde waarde kan zijn binnen cardiale resynchronisatie-therapie (CRT). Bij patiënten met hartfalen is CRT gericht op het synchroniseren van de hartcontractie om de ejectiefractie te verhogen.16 Synchronisatie kan echter resulteren in extreme spanningspatronen in het hart die het hart onbedoeld gevoeliger kan ma-ken voor ritmestoornissen. De T-golf morfologie kan mogelijk indicatief zijn voor deze extreme spanningspatronen om vervolgens het risico op ritmestoornissen voorspellen. Het is reeds aangetoond dat CRT ritmestoornissen kan opwekken.17 Tevens is gebleken dat een verlengd Tpiek-tot-Teind-interval een onafhankelijke voorspeller is voor terecht gegeven ICD-therapie.18 Een gedetailleerde analyse van de relatie tussen T-golf morfo-logie en spanningspatronen binnen hartfalen patiënten is nodig om de voorspellende waarde ervan op ritmestoornissen te bepalen. Omdat het activatiepatroon de T-golf morfologie aanzienlijk beïnvloed, moet deze ook in beschouwing worden genomen.

PATHOFYSIOLOGIE ONDERLIGGEND AAN DE T-GOLF

In dit proefschrift beschrijven we enkele ECG afwijkingen die gerelateerd zijn met abnor-male repolarisatie. In Hoofdstuk 5 trachten wij het mechanisme achter de inferolateral J-golven te doorgronden. In de literatuur bestaat er controverse of het mechanisme dat ten grondslag ligt aan inferolaterale J-golven gebaseerd is op depolarisatie dan wel repolarisatie. De resultaten van dit hoofdstuk ondersteunen de depolarisatiehypothese, aangezien we laten zien dat een geleidingsvertraging in de laterale regio van de LV J-golven op het ECG kan veroorzaken. In ditzelfde model leidt een regionale toename van de voorbijgaande buitenwaartse kaliumstroom (Ito) tot minder grote J-punt eleva-ties. Echter, op basis van onze resultaten kunnen wij de rol van vroege repolarisatie of een combinatie van geleidings- en repolarisatieveranderingen niet volledig uitsluiten. Tevens hebben wij de rol van weefselgeleidbaarheid in het veroorzaken van J-golven getest. Een verminderde weefselgeleidbaarheid, bijvoorbeeld als gevolg van cellulaire ontkoppeling, resulteert in een lagere stroom die gegenereerd wordt door het activatie-golffront en dit zal leiden tot een kleiner potentiaalvlak op het lichaamsoppervlak. Dif-fuse fibrose leidt tot geleidingsvertraging en zou daarom J-golven kunnen induceren. Echter, zal de daarmee gepaard gaande verminderde weefselgeleidbaarheid de inductie


9

van J-golven minder waarschijnlijk maken. Op basis van deze resultaten suggereren wij dat geleidingsvertraging inferolateral J-golven kan veroorzaken en dat het ontstaan van J-golven kan afhangen van de weefselgeleidbaarheid, de grootte en de locatie van het aangedane gebied in het hart. Het model geeft geen verklaring voor de aritmogenese bij de door geleidingsvertraging veroorzaakte J-golven. Geleidingsvertraging is een van de voorwaarden voor het opstarten van re-entry.19 Daarom is het aannemelijk dat de regionale geleidingsvertraging het ontstaan van re-entry kan bevorderen.

Dit proefschrift biedt tevens inzichten in de elektrocardiografische afwijkingen in het lange QT syndroom (LQTS). Hoofdstuk 6 geeft een verklaring voor de ‘bifide’ (gespleten) morfologie van de T-golf in de patiënten met type 2 LQTS. Een verhoogde interventricu-laire repolarisatiedispersie leidt tot bifide T-golven op het ECG, met vroegere repolarisa-tie in het rechterventrikel en latere repolarisatie in het linkerventrikel. Alhoewel de data afkomstig zijn uit een model van een hond met een door medicijnen geïnduceerd LQT2 syndroom, zijn er argumenten die op een soortgelijk mechanisme voor de mens wijzen. Ten eerste, de bifide morfologie van de T-golf is kenmerkend voor zowel aangeboren20 als verworven21 LQTS in de mens. Bovendien, is er bij mensen met een bifide T-golf waargenomen dat de eerste T-golf piek groter is in de rechter precordiale afleidingen en de tweede T-golf piek groter is in de linker precordiale afleidingen.20 Deze configuratie van bifide T-golf toppen is overeenkomstig met onze suggestie dat de eerste top van de bifide T-golf de vroegere repolarisatie van de RV weerspiegelt en dat de tweede top van de bifide T-golf de latere repolarisatie van de LV weerspiegelt. Wij tonen ook aan dat de breedte van de T-golf, als surrogaat voor het tijdsinterval tussen de toppen van de bifide T-golf, sterk gecorreleerd is met de interventriculaire dispersie in repolarisatie. Het zou erg interessant zijn om in een klinische studie te onderzoeken of het tijdsinterval tussen de toppen van de bifide T-golf ook kan worden gecorreleerd met het risico voor ritmestoornissen.

Hoofdstuk 7 beschrijft de slag-op-slag aanpassing van de QT en TQ (= diastolische interval op ECG) intervallen als reactie op snel opstaan in LQT1 en LQT2 patiënten en in controles. De resultaten laten een grote en voorbijgaande verkorting van TQ intervallen zien als reactie op een versnelling van de hartslag. Deze TQ-verkorting was niet verschil-lend tussen LQT patiënten en controles. De geleidelijke en kleinere aanpassing in QT intervallen verschilt wel tussen de groepen, waarbij de QT-verkorting sterker was bij controles dan bij LQT1 patiënten en met geringe QT-verkorting of zelfs QT-verlenging in LQT2 patiënten. Als direct gevolg hiervan worden in alle groepen de QTc intervallen plotseling langer (‘QT-stretching’) in de eerste periode na het opstaan. QT-stretching is onzes inziens een onjuist benaming voor het fysiologische gevolg van opstaan vanuit een liggende positie. Het effect van opstaan is juist een weergave van een gebrek aan

150 Chapter 9

onmiddellijke aanpassing van het QT interval aan een verandering in hartslag en niet een weergave van een abnormale verlenging van het QT interval. Viskin et al22 gaf in een eerdere studie ook al eens commentaar op dit punt.

We beschrijven dat de periode die volgt op de acute TQ-verkorting mogelijk beter on-derscheid maakt tussen groepen. Op basis van deze resultaten maken wij aannemelijk dat de QT/TQ-crossover maat, of verschillende QT- en TQ-waarden, tijdens een liggen-staan test mogelijk van aanvullende onderscheidende waarde kan zijn voor de diagnose van LQTS, vooral bij patiënten met een normale QTc in rust. Enige voorzichtigheid is echter geboden bij de directe toepassing van deze maat in de kliniek, omdat personen in onze studiegroepen niet geselecteerd zijn op basis van een normale QTc-waarde in rust. Middels deze studie willen wij onze zorgen uiten over het gebruik van QTc-waarden in dynamische condities, aangezien de QTc normaliter gebaseerd is op steady-state condities.23

Hoofdstuk 7 beschrijft tevens de verschillen in restitutielijnen tussen LQT patiënten en controles. We concluderen dat de restitutiehellingen steiler zijn in LQT1 patiënten vergeleken met LQT2 patiënten en controles. Een restitutiehelling groter dan 1 is geas-socieerd met een verhoogd risico voor ventriculaire ritmestoornissen.24 Coronel et al toonde echter aan dat de kritieke waarde van 1 voor een restitutiehelling niet noodza-kelijk is voor het initiëren van re-entry.5 Juist de combinatie van geleidingsvertraging en restitutie-eigenschappen van het vroeg-repolariserende weefsel is cruciaal voor het ontstaan van re-entry.5 Dit impliceert dat ook voor LQT patiënten een steilere restitutie-helling pas aritmogeen zou zijn wanneer ook een bepaalde activatievertraging optreedt.

TRANSMuRALITEIT EN DE T-GOLF

Vele cardiologische leerboeken beschrijven de T-golf als een resultante van transmu-rale dispersie van repolarisatie waarbij de epicardiale laag eerder repolariseert dan de endocardiale laag. In dit proefschrift laten we zien dat in een intact hart transmurale dispersie van repolarisatie klein is en slechts een marginale rol speelt in het ontstaan van de T-golf. Al eerder hebben anderen de rol van transmuraliteit onderzocht, en zij conclu-deerden ook dat de transmurale dispersie in repolarisatie en actiepotentiaalduur in het intacte hart klein is.25,26,27,28,29 Vervolgens suggereerden zij dat dit een geringe bijdrage zou leveren aan de genese van de T-golf. Wij hebben de repolarisatiegradiënten in het gehele hart en de T-golf op het ECG tegelijkertijd gemeten en de resultaten bevestigen de bovengenoemde suggestie. Ik stel daarom voor om de leerboeken betreffende de beschrijving van de ontstaanswijze van de T-golf te herzien, waarbij de inzichten die uit dit proefschrift voortvloeien in beschouwing moeten worden genomen.


9

IMPLICATIE VOOR DE MORFOLOGIE VAN DE T-GOLF

Figuur 1 is een samenvattende weergave van de fysiologie en pathofysiologie on-derliggend aan de T-golf welke is gebaseerd op de studies die in dit proefschrift zijn beschreven. De afbeelding laat een normale T-golf (zwarte lijn) zien, met een langer begin-tot-piek-interval dan een piek-tot-eind-interval. Het eerstgenoemde interval (licht grijze gebied) wordt voornamelijk veroorzaakt door verschillen in de actiepotenti-aalvorm (∆AP) en in mindere mate door verschillen in volledige repolarisatiemomenten (∆RT). Het Tpiek-tot-Teind-interval (donker grijze gebied) is voornamelijk het gevolg van verschillen in de volledige repolarisatiemomenten van de meerderheid (ongeveer 75%) van het hart.

Druk kan de T-golf moduleren. Het verlies van een relevante systolische LVdruk (Psys van LV) zal een meer heterogene repolarisatie in het hart genereren wat vervolgens resulteert in een bredere T-golf (lichtgroen gebied). Een toename van de diastolische LVdruk (Pdias van LV) zal voornamelijk invloed hebben op het einde van de AP morfologie en leidt tot een kleine verlenging van de repolarisatie. Dit zal leiden tot langere QT intervallen (als gevolg van een langere ‘staart’) en een grotere amplitude van de T-golf (donkergroen gebied).

Geleidingsvertraging, bijvoorbeeld als gevolg van een verminderde natriumgeleid-baarheid (GNa), in de laterale LV-regio van het hart veroorzaakt J-golven (‘notch’ of ‘slur’) in de inferolateral afleidingen van het ECG (blauwe lijn). De J-golf amplitude zal afhankelijk zijn van de weefselgeleidbaarheid in het hart, waarbij een lagere weefselgeleidbaarheid de amplitude kan verminderen.

Interventriculaire dispersie in repolarisatie draagt bij aan een bifide morfologie van de T-golf. Een verlies aan IKr, in LQT2 patiënten, kan leiden tot een verhoogde interven-triculaire dispersie in repolarisatie (dRTinter), wat resulteert in een bifide T-golf (rode lijn). De eerste top van de bifide T-golf zal worden gegenereerd door repolarisatie in de rechterventrikel en de tweede top zal worden gegenereerd door repolarisatie in de LV.

LQT patiënten hebben een verminderde dynamiek in QT als reactie op een versnel-ling van de hartslag. In het normale hart (zwarte gestreepte lijn) zal een bepaalde ver-korting van het TQ interval (d.w.z. diastolische interval op het ECG) leiden tot voldoende verkorting van het QT interval. Met eenzelfde TQ-verkorting hebben LQT1 patiënten een gematigde QT-verkorting (zwarte stip-gestreepte lijn) en LQT2 patiënten een kleine QT-verkorting en zelfs een kleine QT-verlenging (zwarte stippellijn). De verschillen in T-golf morfologie tussen LQT1 en LQT2 zijn in de figuur buiten beschouwing gelaten.

152 Chapter 9

REFERENTIES






6. Yan G-X, Lankipalli RS, Burke JF, Musco S, Kowey PR. Ventricular repolarization components on the electrocardiogram. J Am Coll Cardiol 2003; 42: 401–409.

7. Tarail R. Relation of abnormalities in concentration of serum potassium to electrocardiographic disturbances. Am J Med 1948; 5: 828–837.






13. Robert E, Aya AG, de la Coussaye JE, Péray P, Juan JM, Brugada J, Davy JM, Eledjam JJ. Dispersion-based reentry: mechanism of initiation of ventricular tachycardia in isolated rabbit hearts. Am J Physiol 1999; 276: H413–H423.

14. Janse MJ, Capucci A, Coronel R, Fabius MAW. Variability of Recovery of Excitability in the Normal Canine and the Ischaemic Porcine Heart. Eur Heart J 1985; 6: 41–52.


16. Brignole M, Auricchio A, Baron-Esquivias G, et al. 2013 ESC Guidelines on cardiac pacing and car-diac resynchronization therapy: the Task Force on cardiac pacing and resynchronization therapy of the European Society of Cardiology (ESC). Developed in collaboration with the European Heart Rhythm Association. Eur Heart J 2013; 34: 2281–2329.


9

17. Roque C, Trevisi N, Silberbauer J, et al. Electrical storm induced by cardiac resynchronization therapy is determined by pacing on epicardial scar and can be successfully managed by catheter ablation. Circ Arrhythm Electrophysiol 2014; 7: 1064–1069.

18. Lellouche N, De Diego C, Akopyan G, Boyle NG, Mahajan A, Cesario D a, Wiener I, Shivkumar K. Changes and predictive value of dispersion of repolarization parameters for appropriate therapy in patients with biventricular implantable cardioverter-defibrillators. Heart Rhythm 2007; 4: 1274–1283.

19. Mines GR. On dynamic equilibrium in the heart. J Physiol 1913; 46: 349–383.





24. Riccio ML, Koller ML, Gilmour RF. Electrical restitution and spatiotemporal organization during ventricular fibrillation. Circ Res 1999; 84: 955–963.






Appendices

List of publications 157

A

LIST OF PubLICATIONS

Manuscripts

Opthof T, Meijborg VM (shared 1st author), Belterman CN, Coronel R. Synchronization of repolarization by mechano-electrical coupling in the porcine heart.Cardiovasc Res. 2015, [Epub ahead of print].

Meijborg VM, Chauveau S, Janse MJ, Anyukhovsky EP, Danilo PR Jr, Rosen MR, Opthof T, Coronel R. Interventricular dispersion in repolarization causes bifid T waves in dogs with dofetilide-induced long QT syndrome.HR 2015, 12(6):1343-51.

Krul SP, Meijborg VM, Berger WR, Linnenbank AC, Driessen AH, van Boven WJ, Wilde AA, de Bakker JM, Coronel R, de Groot JR. Disparate response of high-frequency ganglionic plexus stimulation on sinus node function and atrial propagation in patients with atrial fibrillation.HR 2014, 11(10):1743-51.

Meijborg VM, Conrath CE, Opthof T, Belterman CN, De Bakker JM, Coronel R. Electrocar-diographic T wave and its relation with ventricular repolarization along major anatomi-cal axes.Circ. Arrhythm. Electrophysiol. 2014, 7(3):524-31.

Contributing Authors 159

A

CONTRIbuTING AuTHORS

Anyukhovsky EPDepartment of Physiology and Biophysics, Institute for Molecular Cardiology, Stony Brook University, Stony Brook, New York, USA

de bakker JMTDepartment of Clinical and Experimental Cardiology, Academic Medical Center, Amster-dam, The NetherlandsDepartment of Medical Physiology, University of Utrecht, Utrecht, The NetherlandsInteruniversity Cardiology Institute of the Netherlands, Utrecht, The Netherlands

belterman CNWDepartment of Clinical and Experimental Cardiology, Academic Medical Center, Amster-dam, The NetherlandsL’Institut de RYthmologie et de modélisation Cardiaque (LIRYC), Université Bordeaux Segalen, Bordeaux, France

Chauveau SDepartment of Physiology and Biophysics, Institute for Molecular Cardiology, Stony Brook University, Stony Brook, New York, USA

Conrath CEDepartment of Clinical and Experimental Cardiology, Academic Medical Center, Amster-dam, The Netherlands

Coronel RDepartment of Clinical and Experimental Cardiology, Academic Medical Center, Amster-dam, The NetherlandsL’Institut de RYthmologie et de modélisation Cardiaque (LIRYC), Université Bordeaux Segalen, Bordeaux, France

Danilo Jr PRDepartment of Physiology and Biophysics, Institute for Molecular Cardiology, Stony Brook University, Stony Brook, New York, USA

Janse MJDepartment of Clinical and Experimental Cardiology, Academic Medical Center, Amster-dam, The Netherlands

160 Appendices

Opthof TDepartment of Clinical and Experimental Cardiology, Academic Medical Center, Amster-dam, The NetherlandsDepartment of Medical Physiology, University Medical Center Utrecht, The Netherlands

Postema PGDepartment of Clinical and Experimental Cardiology, Academic Medical Center, Amster-dam, The Netherlands

Potse ML’Institut de RYthmologie et de modélisation Cardiaque (LIRYC), Université Bordeaux Segalen, Bordeaux, FranceCARMEN team, Inria Bordeaux Sud-Ouest, Bordeaux, FranceUniversità della Svizzera italiana, Institute of Computational Science, Center for Compu-tational Medicine in Cardiology, Lugano, Switzerland.

Ritsema van Eck HJDepartment of Medical Informatics, Erasmus Medical Center, Rotterdam, The Nether-lands

Rosen MRDepartments of Pharmacology and Pediatrics, Columbia University, New York, USA

Wilde AAMDepartment of Clinical and Experimental Cardiology, Academic Medical Center, Amster-dam, The NetherlandsPrincess Al-Jawhara Al-Brahim Centre of Excellence in Research of Hereditary Disorders, Jeddah, Kingdom of Saudi Arabia

Portfolio 161

A

PORTFOLIO

PhD student: V.M.F. MeijborgPhD period: December 2009 – June 2015Supervisors: prof.dr. A.A.M. Wilde and prof.dr.ir J.M.T. de BakkerCo-supervisors: dr. R. Coronel and dr. C.E. Conrath

PhD training

Year ECTS

General courses

Project Management 2012 0.7

Clinical Data Management 2011 0.5

Practical Biostatistics 2011 1.5

Scientific writing in English 2011 0.9

ICH – Good Clinical Practice (BROK course) 2010 0.9

AMC World of Science (Introduction course) 2010 0.7

Specific courses

Radiation Expertise (level 5A) 2014 0.9

Electrophysiology curriculum 2014 0.2

ICD Essentials Course 2012 0.7

CVOI course: Electrocardiography and Electrophysiology 2010 0.2

Presentations, Conferences and Workshops

Heart Rhythm Society Meeting, Boston, USA (poster presentation) 2015 1.5

The imaging and electrical technologies (MALT) Meeting, Lugano, Switzerland (invited oral presentation)

2015 0.7

Gordon Research Seminar and Conference – ‘Cardiac Arrhythmia Mechanisms’, Barga, Italy (poster presentation)

2015 2.0

International Stem Cell Forum (ISCF) Symposium, Amsterdam, The Netherlands 2014 0.2

Netherlands Heart Rhythm Association – ‘Brady & Tachy’, Ermelo, The Netherlands (oral presentation)

2014 0.7

European Heart Rhythm Association Cardiostim Congress, Nice, France 2014 1.5

Heart Rhythm Society Meeting, San Francisco, USA (poster presentation) 2014 1.5

L’Institut de RYthmologie et Modélisation Cardiaque (LIRYC) workshop, Bordeaux, France

2013 0.5

6th International Workshop on Cardiac MEC and Arrhythmias, Oxford, England (poster presentation)

2013 1.0

European Society of Cardiology Congress, Amsterdam, The Netherlands (poster presentation)

2013 1.5

Denis Escande Symposium, Amsterdam, The Netherlands 2013 0.5

162 Appendices

Dutch Society of Technical Physicians (NVvTG) Congress, Arnhem, The Netherlands (poster presentation)

2012 0.3

Netherlands Heart Rhythm Association – ‘Brady & Tachy’, Ermelo, The Netherlands (oral presentation)

2012 0.7

European Society of Cardiology Congress, München, Germany (poster presentation) 2012 1.5

Workshop Frontiers in Computational Electrocardiology, Maastricht, The Netherlands 2010 0.3

5th International Workshop on Cardiac MEC and Arrhythmias, Oxford, England 2010 1.0

European Society of Cardiology Congress, Stockholm, Sweden 2010 1.5

Teaching

Year ECTS

Lecturing

Lecturer electrophysiology workshop for medical students 2010-2014 2.0

Lecturer electrophysiology & research for FNWI master students 2012-2014 0.6

Mentoring / supervising

2 medical students for bachelor thesis 2014-2015 4.0

Other

Co-organizer of workshop Careers in Science during 6th International Workshop on Cardiac MEC and Arrhythmias, Oxford, England

2013 0.3

Parameters of Esteem

Year

Awards and prizes

Young Investigators Award – MALT-meeting 2015

Curriculum Vitae 163

A

CuRRICuLuM VITAE

Veronique Marlinde Frederica Meijborg was born on May 8th 1985 in Zoetermeer, the Netherlands. She grew up being the middle child with three brothers and a sister. In 2003, she graduated from ’t Hooghe Landt College (high school) with a focus on biology and technical sciences. Using this background Veronique followed the new programme ‘Technical Medicine’ at the University of Twente in Enschede. This is also where she lived during her study until 2007 when she did a number of internships in hospitals throughout the Netherlands. In 2009, Veronique obtained her master’s degree in Techni-cal Medicine, with a specialization in Medical Signaling. Her master thesis on analysis of ventricle fibrillation signals already showed her interest in cardiac electrophysiology, which preluded her career in this domain.

After her graduation Veronique started her PhD project at the Department of Cardiology (AMC) under the supervision of prof.dr.ir. Jacques de Bakker and prof.dr. Arthur Wilde. This project involved both animal and human studies. During her PhD, Veronique also served as treasurer of the professional Dutch Association for Technical Physicians (the NVvTG). Currently, she is employed as a post-doc on stem cell therapy after myocardial infarction within the AMC. Veronique’s passion lies with the imple-mentation and validation of new technologies within clinical practice, which she would like to focus on in the future.

Dankwoord 165

A

DANkWOORD

Ter afsluiting van dit proefschrift wil ik mijn dank uiten aan velen waarmee ik heb sa-mengewerkt en die mij op eniger wijze gesteund hebben tijdens mijn promotietraject. Omdat er zo veel zijn die ik wil bedanken is het de kunst om niemand te vergeten, al weet ik dat ik hierin waarschijnlijk niet slaag.

Allereerst wil ik mijn copromotores Dr. Coronel en Dr. Conrath bedanken, aangezien zij degene zijn geweest die mij in velerlei opzichten hebben bijgestaan en bijgestuurd. Beste Ruben, ondanks dat de rollen in het begin niet helemaal duidelijk waren, ben jij uitgegroeid tot een belangrijke supervisor, die mij altijd weer op weg hielp. Ik wil je danken voor alle discussies zowel op wetenschappelijk als persoonlijk vlak. Ik heb veel geleerd en profijt gehad van jouw onmetelijke kennis van de elektrofysiologie. Ik ben blij dat onze samenwerking nog even mag voortduren.

Beste Chantal, in een van onze eerste gesprekken vertelde je me dat een promotie-traject heel anders kan lopen dan vooraf gedacht. Binnen mijn promotietraject heb ik uiteindelijk ook paden bewandeld die ik vanaf het begin niet voor mogelijk hield. Ik bewonder jouw scherpe en klinische blik, die mij steeds weer uitdaagde om het ‘onder-ste uit de kan’ te halen. Ik zie jou als een uitmuntende elektrofysiologe die zich uitermate inspant voor haar patiënten zodat zij de beste zorg krijgen. Ik hoop in de toekomst nog veel van jou te leren op het vlak van de klinische elektrofysiologie.

Tevens wil ik mijn promotores prof.dr. Wilde en prof.dr.ir. de Bakker bedanken voor hun supervisie. Beste Arthur, ik waardeer het zeer dat je ondanks een overvolle agenda toch altijd ruimte hebt gemaakt om de voortgang van de promotie te bespreken. Na deze momenten stonden alle neuzen altijd weer dezelfde kant op.

Beste Jacques, onze samenwerking reikt de laatste tijd ook tot buiten het AMC waarbij we uitstapjes maken naar Zwolle en Utrecht om elektrische mappings te doen bij zowel patiënten als ook dieren. Ik ben je zeer dankbaar dat je mij betrokken hebt bij deze verschillende projecten. Aangezien binnen mijn promotietraject het focus voornamelijk op de T-golf en repolarisatie lag, leerde ik op deze momenten juist meer bij betreffende de activatie van het hart. Inmiddels ben je met emeritaat, maar ik hoop dat we onze samenwerking nog even kunnen voortzetten.

I would also like to thank my thesis committee, prof.dr. Taggart, prof.dr. Zeppenfeld, prof.dr. Prinzen, prof.dr. Janse, dr. van Dessel and dr. Remme. I am honoured that you were willing to give your expert opinion on my work.

Beste Giel, ik vond het leuk om met je samen te werken en ik heb veel kunnen leren van jouw aanpak. Vanuit een goed overzicht blijf je focus houden op het einddoel en de

166 Appendices

stappen die nodig zijn om dat doel te bereiken. Wat een eer dat ik in Bordeaux jouw plek in de helikopter mocht innemen, dat was een mooie (wijnvolle) ervaring.

Beste Pascal, jouw input en steun vanaf de zijlijn heb ik altijd erg gewaardeerd. Ik hecht veel waarde aan jouw klinisch en technische kennis waarvan ik hoop nog vaak dankbaar gebruik te kunnen maken.

Vervolgens wil ik mijn grote dank uiten aan mijn lieve paranimfen en kamergenoten, Judith en Suzanne. Lieve Juud, al vanaf het begin hebben wij ons lief en leed betref-fende onze promoties kunnen delen. Ik wil je dan ook bedanken voor alle steun die je me door de jaren heen hebt gegeven. De vele momenten die wij samen hebben mogen meemaken zijn mij dierbaar. Ik denk nog vaak aan de eindeloze celisolaties: grote vs. kleine boon, oplossing A vs. V, schudden vs. roeren, pfff. Helaas resulteerde dit niet in hoogstaande wetenschap. Toch zijn dit momenten geweest waarin we veel hebben gediscussieerd, maar ook gelachen. Naast een goed collega ben je ook een sportmaatje, voor de ontspanning naast het werk. Inmiddels is onze samenwerking op een lager pitje komen te staan, maar ik hoop dat dit in de toekomst wel weer gaat oplaaien. Veel succes met de opleiding. Je bent een topper!

Lieve Suuz, onze tijd samen is nog maar van korte duur geweest. Toch voelt het alsof wij elkaar al jarenlang kennen. Ik ben trots op onze EP club, die dankzij jouw toedoen in het leven is geroepen. Het draagt bij aan de wetenschappelijke discussies op de kamer. Ik bewonder ook jouw doorzettingsvermogen en positieve instelling, ondanks de moei-lijke thuissituatie. Ik dank je voor jouw enthousiasme en steun gedurende mijn laatste loodjes. Ik verheug me nu al op de uitvoer van onze onderzoeksplannen.

Graag wil ik mijn andere (oud)kamergenoten bedanken voor de prettige werksfeer. Ik voelde mij er altijd thuis. Sébas, bedankt voor de opbeurende woorden wanneer ik even vast zat met mijn onderzoek/analyses. Louise, bedankt voor de hulp bij de laatste loodjes met mijn proefschrift. Madelon, dank voor jouw gezelligheid en openheid, die bijdraagt aan de prettige sfeer op de kamer. Chris, onze samenwerking was van korte duur. Ik ken je vooral als serieuze en harde werker en ik bewonder jouw discipline. Yuka, ik dank je voor je hulp. Veel klinische EP vragen kon ik ook bij jou kwijt. Het ga je goed, veel geluk in Italië! Wouter, jij bent een echte goedzak en helpt waar mogelijk en dat waardeer ik zeer. Naast het harde werken maak jij ook zeker tijd voor de vrijdagmid-dagborrel. Gezellig! Tom, Jim en Nicoline, jullie zijn de jongste aanwinsten op de kamer, maar een ieder van jullie levert een positieve bijdrage aan de sfeer en discussies die wij regelmatig voeren. Dank.

Tevens gaat mijn dank ook uit naar collegae van de experimentele kant van de cardio. Tobias, bedankt voor de vele discussies zowel op het wetenschappelijke als persoonlijke

Dankwoord 167

A

vlak. Ik ervoer dat als een steuntje in de rug. En vergeet vooral niet: ‘je bent zo oud als je je voelt…’ Charly, de experimenten waren geen succes geworden zonder jouw kennis en kunde. Eigenlijk was er altijd van alles mogelijk en daar ben ik je dankbaar voor. Carel, zonder jouw gouden handen en creatieve geest waren de experimenten onmogelijk. Ook dank voor je openhartigheid. En Wim zonder jou geen naaldexperimenten, vooral van de laatste set hebben we nog veel onderzoeksplezier. André, bedankt voor jouw behulpzaamheid. Bij jou kon ik altijd terecht met technische/Matlab vragen. Marieke, bedankt dat je mij hebt leren patch-clampen. Jammer genoeg wilde de celisolaties niet echt meewerken. Cees, dank voor de hulp en humor bij het optimaliseren van de celiso-laties. Nicoline, onze samenwerking zal vooral het komende jaar intensiever worden. Je bent een vrolijke noot op de afdeling en een harde werker en ik hoop dat we samen nog mooie proeven mogen doen. Mark Potse, jou dank ik voor je constructieve samenwer-king en ik waardeer jouw doortastendheid. Ook bedank ik Dénise, Bas, Mathilde, Shirley, Ton, Jan, Antoni en Eelco voor alle discussies en gezelligheid.

In het algemeen wil ik nog andere collegae bedanken voor hun steun en gezelligheid. Ik bedank de ‘oude garde’ Peter, Anja, Pier, Ze-Yie, Maik, Tim, Wichert, Loes, Bimmer, Ronak, Kirsten B, Ralf, Wouter K, Marcel, Krischan en Carlijne. En in overloop naar de ‘nieuwere garde’ dank ik ook Susan, Ties, Michiel, Fleur, Martijn vL, Joëlle, Ling, Esther, Robin, Mariëlla, Gilbert, Martijn vM, Dagmar, Ivo, Debby, Martina, Niels, Juliëtte, Lonneke en Sangeeta. Van de experimentele cardio dank ik ook Krystien, Yolan, Joeri, Christiaan, Maaike, Najim, Geert en Erik. Van de congenitals dank ik Zeliha, Paul, Alex, Romy, An-nelieke, Piet, Teun, Michiel, Jeroen, Maayke, Mark, Jouke, Joey, Ilja, Hayang en Josephine. En ook: Piety, Anita, Gerrie, Rena, Jonas, Freek, Vokko, Kirsten K, Vincent, Mark en Kim. Ik dank ook Sarvi en Monique voor de steun en energie-gevende koffiemomentjes.

Dank aan Joris, voor onze discussies over de toekomst; Reinoud, voor de gezellige kof-fiemomenten op de kamer; Regina, voor je hulp en persoonlijke gesprekken; Amin, voor de gezelligheid en goede discussies in Nice; en Pieter, dank dat je zo nu en dan ff kwam langslopen om te checken hoe het ging.

En dan mijn vrienden. Maryse, jouw dank ik voor jouw steun door de jaren heen. Ik hecht veel waarde aan onze vriendschap die al zo’n 15 jaar voortduurt. Linda, Bernardine en Mijke, ik dank jullie voor alle mooie momenten die we samen hebben meegemaakt. Jul-lie steun en motiverende woorden betekenden veel voor mij. Joost, Wendy en Richard, jullie dank ik voor de gezellige momenten en wintersportvakanties waaraan ik altijd veel plezier heb beleefd.

168 Appendices

En tot slot wil ik ook mijn dank uiten aan mijn familie. Eerst mijn broers, zusje en aan-hang. Rolf, Joyce, Oedger, Juliët, Marthijs, Marlinde en Andries, ik dank jullie voor de on-voorwaardelijke liefde en steun. Jullie interesse voor waar ik mee bezig was gaf mij altijd energie. Lin, het volgende experiment is voor jou! Oma Rinsma wil ik speciaal bedanken voor haar grote enthousiasme en trots. Ik vind het erg jammer dat u niet meer bij de verdediging kan zijn. Ook mijn schoonfamilie (Leen, Ria, oma van Bruinessen, Eelco en Anne) wil ik danken voor hun steun en vertrouwen.

En dan mijn ouders. Pap en mam, grote dank voor jullie onuitputtelijke liefde, geduld en vertrouwen. Zonder dat was ik nooit zover gekomen.

Tot slot dank ik ook Corné. Schat, jij bent mijn steun en toeverlaat. Jouw liefde, ver-trouwen en geduld (ja, dat heb je best ;-) ) helpen mij naar een hoger niveau.

The T wave: physiology and pathophysiologyVeronique M.F. Meijborg

Th

e T w

ave: ph

ysiology and

path

oph

ysiology

Vero

niq

ue M

.F. Meijb

org

The T wave: physiology and pathophysiology · description of the T wave of a normal pig heart in...

Documents

Transcript of The T wave: physiology and pathophysiology · description of the T wave of a normal pig heart in...