The Pulmonary Veins in Pigs and Horses: Research Towards the Development of a New Treatment Strategy of Atrial Fibrillation in Human Patients and Horses

The pulmonary veins in pigs and horses: research towards the development of a new treatment strategy of atrial fibrillation in human patients and horses

Tim Vandecasteele

Dissertation submitted in fulfillment of the requirements for the degree of Doctor of Philosophy (PhD) in Veterinary Sciences

2018

Promoters: Prof. dr. P. Cornillie Prof. dr. G. van Loon Prof. dr. W. Van den Broeck dr. G. Van Langenhove

Department of Morphology Faculty of Veterinary Medicine Ghent University

Tim Vandecasteele 2018

The pulmonary veins in pigs and horses: research towards the development of a new treatment strategy of atrial fibrillation in human patients and horses

Front cover: heart image by Henry Vandyke Carter (Gray’s Anatomy)

Printed by University Press, Zelzate, Belgium www.universitypress.be

Success consists of going from failure to failure without loss of enthusiasm

-Winston Churchill-

Table of contents

List of abbreviations

Preface………………………………………………………………………………………………………………………………… 7

Chapter 1: General introduction………………………………………………………………………………………… 11

1.1 Heart and pulmonary veins (PVs)………………………………………….……………………………….. 13 1.2 Cardiac conduction………………………………………………………………………………………………… 20 1.3 Atrial fibrillation (AF)………………………………………………………………………………………….….. 25 1.4 PVs in horses……………………………………………………………………………………………………….…. 45 1.5 Pigs as cardiovascular model……………………………………………………………………….…………. 46

Chapter 2: Scientific aims………………………………………………………………………………………….…….... 73

Chapter 3: The pulmonary veins of the pig………………………………………………………………………… 77

Chapter 4: Presence of ganglia and telocytes in proximity to myocardial sleeve tissue in the porcine pulmonary veins wall…………………………………………………………………….……………………. 101

Chapter 5: A preclinical study of an implanted device in the pulmonary veins, intended for the treatment of atrial fibrillation in an ovine model……………………………………………………..… 123

Chapter 6: A preliminary study of pulmonary vein implant applicability and safety as a potential ablation platform in a follow-up study in pigs…………………………………………………… 139

Chapter 7: Immunohistochemical identification of stent-based ablation lesions in the superior vena cava and pulmonary veins……………………………………………………………………..…. 153

Chapter 8: Isolation of pulmonary veins using a thermo reactive implantable device with external energy transfer: Evaluation in a porcine model……………………………………………….... 169

Chapter 9: The pulmonary veins of the horse……………………………………………………..…………… 187

Chapter 10: 3D reconstruction of the porcine and equine pulmonary veins, supplemented with the identification of telocytes in the horse………………………………………………………………. 205

Chapter 11: Echocardiographic identification of atrial-related structures and vessels validated by CT images of equine hearts……………………………………………………………………....… 221

Chapter 12: General discussion…………………………………………………………………………………..…… 239

Summary…………………………………………………………………………………………………………………………. 263

Samenvatting………………………………………………………………………………………………………..……….. 269

Curriculum vitae……………………………………………………………………………………………………………… 275

Bibliography…………………………………………………………………………………………………………………….. 279

Dankwoord………………………………………………………………………………………………………………………. 285

List of abbreviations

AccV accessory pulmonary vein

ACT activated clotting time

AEF atriooesophageal fistula

AF atrial fibrillation

ALL accessory lung lobe

AN antrum

Ao aorta

AS atrial side

AV aortic valve

AVN atrioventricular node

BT brachiocephalic trunk

C temperature increase

Ca caudal

CaVC caudal vena cava

CDLL caudal lung lobe

CLL cranial lung lobe

CLL(cp) cranial part cranial lung lobe

CLL(cdp) caudal part cranial lung lobe

CP cardiac plexus

Cr cranial

CrVC cranial vena cava

CT computed tomography

D dorsal

DAB diaminobenzidine

ECG electrocardiography

EP electrophysiology

H&E hematoxylin & eosin

HPF high power field

HR heat ring

HRP horse radish peroxidase

ICV inferior cava vein

ILL intermediate lung lobe

IT intervenous tubercle

L lumen

LA left atrium

LCLL left cranial lung lobe

LCDLL left caudal lung lobe

LiCd bronchus left caudal lung lobe

LcdV1 left caudal pulmonary vein (cranial aspect)

LcdV2 left caudal pulmonary vein (intermediate aspect)

LcdV3 left caudal pulmonary vein (caudal aspect)

LcrV left cranial pulmonary vein

LiCrpCd bronchus left cranial lung lobe (caudal part)

LiCrpCr bronchus left cranial lung lobe (cranial part)

LIPV left inferior pulmonary vein

LLL lingual lung lobe

LPA left pulmonary artery

LSPV left superior pulmonary vein

LV left ventricle

MRI magnetic resonance imaging

MYBPC3 myosin binding protein C

Myo myocardial sleeve

NL nerves of left atrium

O oblique

OF oval fossa

PA pulmonary arteries pAF paroxysmal atrial fibrillation

PLLA-GP posterolateral left atrial ganglionated plexi

PMLA-GP posteromedial left atrial ganglionated plexi

Po podom

PS pulmonary side

PV pulmonary vein

PVO pulmonary vein ostia

PVs pulmonary veins

PVI pulmonary vein isolation

PVS pulmonary vein stenosis

PT pulmonary trunk

QCA quantitative coronary angiography

RA right atrium

RAA right atrial appendage

RCd bronchus right caudal lung lobe

RCDLL right caudal lung lobe

RcdV1 right caudal pulmonary vein (cranial aspect)

RcdV2 right caudal pulmonary vein (intermediate dorsal aspect)

RcdV3 right caudal pulmonary vein (intermediate ventral aspect)

RcdV4 right caudal pulmonary vein (caudal aspect)

RCLL right cranial lung lobe

RCr bronchus trachealis (right cranial lung lobe)

RcrV right cranial pulmonary vein

Rim bronchus right middle lung lobe

RLa bronchus accessory lung lobe

RPA right pulmonary artery

Rt right

RV right ventricle

SCV superior caval vein

SG stellate ganglion

SLA-GP superior left atrial ganglionated plexi

SN sinus node

TC terminal crest

Tcb telocyte cell body

TEM transmission electron microscopy

Tpr telocyte prolongation

V ventral

VN vagal nerves

VWT-peak peak temperature during ablation

VWT-pre vessel wall temperature prior to ablation

Preface

The research presented in this manuscript was originally initiated from a rather simple, straightforward question: “is the pig a suitable animal model to test a newly developed implantable device as a novel treatment option for atrial fibrillation in human patients?” Atrial fibrillation is in fact the most important cardiac arrhythmia in man, affecting an increasing number of people, especially at higher age. Although various treatment options have been developed, the method of the last resort, i.e. catheter ablation, has an unsatisfyingly high recurrence rate of at least 20-40% (Darby, 2016), requiring the patient to undergo this time-consuming procedure multiple times under general anesthesia. The intended procedure involving the newly developed ablation device is both aiming for a significantly lower recurrence rate as well as a much shorter intervention time. However, it still needed to be tested in an animal model. The pig was selected as cardiovascular model as pigs are frequently used as an animal model in different research domains and the porcine heart is comparable in size with that of humans. As the device was to be inplanted in the pulmonary veins, the main focus of the investigation resided in the comprehensive elucidation of the anatomical organization of these vessels draining into the left atrium of the heart. Although the cause and development of atrial fibrillation has still not been fully elucidated, in man, the role of the pulmonary vein walls as sites of ectopic foci triggering the arrhythmia have long been recognized. As such, many in-depth morphological and electrophysiological studies of the pulmonary vein walls in man are available in literature, which is in huge contrast with the availability of similar data for the pig. The first histological literature on the human pulmonary veins dates back to the 19th century, yet it is only since the 2000’s that the scientific interest in these structures re-emerged (Stieda, 1877; Nathan and Eliakim, 1966; Moubarak et al., 2000; Hassink et al., 2003). Therefore, as a first step of this investigation, the complete unravelment of the topographic organization of the porcine pulmonary veins was pursued. Secondly, the walls of these veins were histologically investigated to obtain an overview of all present structures and cell types that either might initiate or facilitate the propagation of ectopic pulses and/or might be affected by the intended ablation (heat scarring) in this region. These data made it possible to proceed towards the invasive procedures in anesthetized pigs. A prototype of the new device for atrial fibrillation treatment was developed, applied and tested on pigs and adjusted during different study phases.

Simultaneously with the examination of the porcine pulmonary veins, this research was extended towards the equine heart. In fact, also horses are quite prone to develop atrial fibrillation. However, in contrast to the situation in man, similar advances in the understanding of the underlying mechanisms of atrial fibrillation and the development of new treatment options in the equine patient are still too far out of reach. This is mainly because of our current inability to unambiguously visualize and identify the individual pulmonary veins in the living animal and consequently determine

Preface their involvement in impaired cardiac function. Echocardiography is the most important, if not the only feasible visualisation technique for the equine heart. However, due to the far more complex topography of drainage area of the equine pulmonary veins into the left atrium and the size of horses, most structures in this region are too difficult to identify on two-dimensional cross sectionial image. Especially, since detailed anatomical data on the three-dimensional organization of the pulmonary veins in horses are currently lacking. Therefore, a morphological study on the localization and variation of the equine pulmonary veins was designed to fill this gap and led to the description of an echocardiographic procedure to visualize these veins.

As such, the single primary research question eventually spawned a myriad of ensuing and parallel investigations, which are, despite their common origin, quite diverse in nature. Nevertheless, they all share a common basis, i.e. atrial fibrillation. Therefore, before elaborating on the various aspects of this research, first a concise overview of cardiac morphology and the problem of atrial fibrillation will be provided in the introductory chapter of this manuscript.

Chapter 1 General introduction

Chapter 1: General introduction

1.1 Heart and pulmonary veins (PVs)

The heart is a muscular organ containing four distinct cavities, i.e. the left and right atria and the left and right ventricles, that pump blood and, by action of the cardiac valves, direct the blood flow over two completely separated circulatory loops. The left atrium, which is the main focus of this thesis, receives oxygenated blood from the lungs through the pulmonary veins. This blood is transferred to the left ventricle, which pumps the blood via the aorta through the whole body to supply all organs with oxygen. On the other hand, deoxygenated blood is collected into the right atrium through the cranial and caudal vena cava. The right ventricle subsequently sends the blood via the pulmonary trunk towards the lungs, where it can be reoxygenated.

The cardiac wall consists of three layers indicated as endocardium, myocardium and epicardium. The myocardium is the part responsible for cardiac contraction. Myocardium consists mainly of muscle fibres which will contract during the depolarization phase. In contrast to skeletal muscle, cardiac muscle cells are characterized by a refractory period, preventing the heart from a tetanic contraction state. This ensures that for each contraction an electrical pulse is prerequisite (Patteson, 1996).

The pulmonary veins (PVs) are venous blood vessels which drain the oxygenated blood from the lungs into the left atrium. Normally, a venous vessel wall is built-up with connective and smooth muscle tissue. Besides this, PVs also contain cardiac muscle, part of the myocardium, which is non- uniformly organized in a sleeve pattern that fans out distally, further indicated as myocardial sleeve (Saito et al., 2000). This cardiac tissue consists of muscle fibers, intricately arranged in bundles of different thickness, which can change abruptly in direction (Hamabe et al., 2003; Ho et al., 2001). The structural and functional properties of these cardiomyocyte sleeves in the pulmonary vein (PV) wall play an essential role in creating an arrhythmogenic substrate for atrial fibrillation (Nattel, 2013). This links the morphological description of these venous structures with the cardiac disorder, atrial fibrillation, which will be described extensively further on.

1.1.1. Orientation of the heart in men versus animal

Although the heart of humans and pigs is comparable, some differences can be noticed. The position of the heart is clearly different in pigs and horses compared to humans. The orientation of the heart relative to the rib cage and sternum is also different between the pig and horse on the one hand and man on the other hand. These differences influence heart morphology. In addition, the thorax in pigs and horses is laterally compressed compared to the dorsoventrally compressed chest in men. As a

Chapter 1: General introduction result, the porcine heart demonstrates a shape which corresponds with the typical heart silhouette, whereas in men, this is more trapezoidal in silhouette (Crick et al., 1998). Moreover, the position of the heart in the thorax is not in accordance in men and most animals as the most ventral point of the porcine and equine heart, the cardiac apex, points more towards the sternum compared to the situation in men (Fig. 1). However, the porcine heart is directed more obliquely compared to the orientation of the equine heart. In addition, the heart in both animals has made a quarter turn along the long axis in counter-clockwise direction from a dorsal point of view. This results into the fact that human nomenclature, adopted in veterinary anatomy, describing the atria and ventricles does not match with the anatomical position of these structures in the thorax of pigs and horses. Respectively, the terms “superior”, “inferior”, “anterior” and “posterior” in human anatomy refer to terms “cranial”, “caudal”, “ventral” and “dorsal” in veterinary anatomy of the body. Because of the cardiac rotation around its long axis, the right and left half of the heart is oriented cranially and caudally, respectively.

Fig. 1. Left: schematic representation of the orientation of the heart inside the thorax in men (left image) and in the pig (right image) (Adapted from Crick et al., 1998). Right: image of the equine heart inside the thorax, seen from the left side of the horse (From P.F. Flood, 2017)

Chapter 1: General introduction

1.1.2. The pulmonary veins in humans

Fig. 2. Human heart with the pulmonary veins (PV) indicated. A: facies auricularis; B: facies atrialis; 1: left superior PV; 2: left inferior PV; 3: right superior PV; 4: right inferior PV. (Adapted from Barone, 1996)

Fig. 3. Medial view of human lungs with the different lung lobes and pulmonary veins (PV) indicated. A: left lung, B: right lung, CLL: cranial lung lobe, ILL: intermediate lung lobe, CDLL: caudal lung lobe, LLL: lingual lung lobe. (Adapted from Barone, 1976)

The major part of the human population possesses four PVs (Fig. 2, 3) whereas some individuals are seen having three or five veins. As mentioned before, the pulmonary veins drain the oxygenated blood from the lungs towards the left heart. Therefore, the drainage pattern of the PVs is mainly determined by the morphology of the lungs and its division into several lung lobes. The division pattern of the lungs is mainly the same for humans and animals but some species-specific variation can be noticed.

Normally, the right superior (cranial) PV drains the right cranial and intermediate lung lobes, while the left superior (cranial) PV drains the left cranial and lingual lobes. The right and left inferior (caudal) PVs drain the respective caudal lung lobes (Fig. 3). The superior PVs are slightly larger compared to the inferior ones (Barone, 1976; Perez-Lugones et al., 2003; Corradi et al., 2013). In a study by Ho et al. (2012), 74% of the cases demonstrated the normal arrangement of four PVs. In 17% and 9% of the cases, five PVs and a left or right common PV were observed respectively (Ho et

Chapter 1: General introduction al., 2012). In addition, also other PV arrangements are noticed such as the presence of a right middle PV, the existence of a right inferior PV with early branching and a so-called right top PV which originates from the left atrial roof (Tsao et al., 2001; Perez-Lugones et al., 2003; Lickfett et al., 2004; Corradi et al., 2011). In a study with atrial fibrillation (AF) patients, 40% of the cases demonstrated the normal pattern of two right and left PVs, around 30% had a common left trunk, 12.5% showed one right middle PV and in 1.5% of the cases two right middle PVs were noticed. Several other rare and complex patterns were noticed as well (Anselmino et al., 2011).

Both left and right PVs attach laterally and medially to the posterior and superior wall of the left atrium (Edwards, 1987). The right PVs move behind the right atrium and the superior vena cava and fuse with the left atrium near the interatrial septum. The inferior PVs move underneath the inferior border of the left and right bronchi and the superior PVs are positioned anterior to their bronchi (Edwards, 1987).

Besides the anatomical organization of the PVs, the histological buildup of its wall also contains valuable data in the context of a cardiac disorder. In humans, detailed histological literature about the several layers of the PV wall is available. Smooth muscle cells incorporated in a meshwork of elastic and fibrous tissue build up the middle layer (tunica media) of a pulmonary vein. Cardiac muscle tissue is located between the adventitia and the medial layer (Ho et al., 2012). This atrial myocardial tissue fans out into irregular myocardial sleeves over the atrio-venous junction into the PV wall (Fig. 4) (Stieda, 1877; Poirier and Charpy, 1911; Nathan and Eliakim, 1966; Spach et al., 1972; Corradi et al., 2013). The sleeves’ length ranges between 13 and 25 mm (Nathan and Eliakim, 1966) and the longest sleeves are mostly located in the upper PVs, especially in the left upper PV. These myocardial sleeves include interconnected bundles of cardiomyocytes which are oriented spirally or circularly (Ho et al., 2001; Roux et al., 2004). At the ostium of the PVs, which is the opening through which blood is drained into the left atrium, these bundles are organized into two layers of striated muscle tissue (Fig. 5). The external layer is oriented circularly and always present in the most proximal 5 mm of the PVs but does not always cover the entire circumference of the antrum of the PVs. The antrum is the part between the ostium and the spot where PVs fuse together. The internal layer, less organized into some packs or even into some isolated cells, demonstrates a longitudinal pattern (Roux et al., 2004). Ho et al. (2001) described the myocardial sleeve pattern as circularly organized myocardial bundles, often pierced by obliquely and longitudinally oriented strands. The sleeves thin out from the atrio-venous junction towards the lungs with a maximum thickness of around 1-1.5 mm (Ho et al., 2012). In addition, an irregularly oriented layer of elastic tissue becomes visible more distally and is often noticed distally to the second branch (Roux et al., 2004).

Chapter 1: General introduction

Regarding the role of the myocardial sleeve, it is hypothesized that these venous sphincter-like structures may be involved into an atrial systolic mechanical process of influencing the pulmonary circulation capacity (Almeida et al., 1975) or act as valve-like structures to prevent blood reflux during the atrial systolic phase (Burch and Romney, 1954; Little, 1960).

Fig. 4. A: Histological section of human pulmonary vein (PV) tissue with the myocardial sleeve (Myoc. Sleeve) fanning out from the left atrium (LA) and ending into a small tip (indicated by the box); B: higher magnification of the box in which the most distal myocardial cells are surrounded by fibrous tissue; PV: luminal side of the pulmonary vein; elastic van Gieson staining; bar = 5 mm. (From Saito et al., 2000)

Fig. 5. 3D-reconstruction of a part of a human pulmonary vein in which the myocardial sleeve, located between the tunica media (indicated by SM) and tunica adventitia (not shown), is divided into several layers each indicated by a different colour. All images demonstrate the same tissue block with three myocardial tissue layers, each with their own orientation, indicated in the left image and only one layer demonstrated in the right image. Endocardial or endothelial side (EN); smooth muscle tissue (SM); red: longitudinal myocardial fiber; yellow: oblique myocardial fiber; blue: transverse myocardial fiber. (Adapted from Saito et al., 2000)

Chapter 1: General introduction

1.1.2 The pulmonary veins in pigs and horses

Fig. 6. Heart of the pig with the pulmonary veins (PV) indicated. A: facies dorsalis; B: facies atrialis; C: facies auricularis. Red = oxygenated blood; blue = deoxygenated blood. (Adapted from Barone, 1996)

Fig. 7. Heart of the horse with the pulmonary veins (PV) indicated. A: facies auricularis; B: facies atrialis. Red = oxygenated blood; blue = deoxygenated blood. (Adapted from Barone, 1996)

Chapter 1: General introduction

Fig. 8. A: ventral view of the lungs of the horse, RCLL: right cranial lung lobe, LCLL: left cranial lung lobe, RCDLL: right caudal lung lobe, LCDLL: left caudal lung lobe, ALL: accessory lung lobe; B: left lung of the pig, CLL(cp): cranial part of the cranial lung lobe, CLL(cdp): caudal part of the cranial lung lobe, CDLL: caudal lung lobe; C: right lung of the pig, CLL: cranial lung lobe, ILL: intermediate lung lobe, CDLL: caudal lung lobe. (Adapted from Barone, 1976)

Fig. 9. A: left lung of the horse with the pulmonary veins (red), arteries (blue) and bronchi (white) visualised; VCL: veins of the cranial lobe, VCDL(cp): veins of the cranial part of the caudal lobe, VCDL(cdp): veins of the caudal part of the caudal lobe. (Adapted from Barone, 1976); B: three PV ostia (PVO) through which the blood from the lungs is drained into the left atrium (LA) of an equine heart, AO: aorta, LV: left ventricle. (Adapted from Constantinescu and Schaller, 2012)

According to Constantinescu and Schaller (2012), the PVs collect blood from the different lung lobes and drain into the left atrium. These veins can fuse together or may be found as plural separated entities when entering the left atrium. In general, each lung lobe, from the left and right cranial and caudal ones to the medial and accessory lobe, has its own pulmonary vein with the corresponding designation. Concerning the main pattern of the porcine PVs, the middle and cranial veins of each lung come together in a common vein before draining into the left atrium (Fig. 6, 8) (Barone, 1976). In the horse, Barone (1976) found that the PVs of the middle and cranial lung lobes drain separately into the left atrium whereas the PVs of both caudal lung lobes constitute one common vessel before debouching into the heart (Fig. 7, 8). However, literature is not conclusive on the number of equine PVs as one study describes seven or eight PVs debouching into the caudal and right part of the left

Chapter 1: General introduction atrium (Bright and Marr, 2010). Budras et al. (2009), on the other hand, depicts three PVs entering the equine left atrium. A third study concludes that the number of draining openings in the dorsal part of the left atrium is determined by the number of PV branches entering the heart, which may vary from three to eight (Fig. 9) (Gille and Salomon, 2015).

Some histological data is available, describing the composition of the equine and porcine PV wall. PV myocardial sleeves were investigated in a limited study (one horse and five pigs) (Nathan and Gloobe, 1970). It was indicated that usually in both species, one right PV and two left (superior and inferior) PVs were observed. However, in certain cases, a right superior and inferior pulmonary vein could be noticed. This study only intended to describe the myocardial sleeve length (around 10 mm) in a limited number of pigs and horses which indicates that reliable data is very scarce.

Further detailed information concerning the drainage pattern, position and variation of the different PVs in pigs and horses is lacking.

It is clear that there are differences concerning the PVs in humans on the one hand and in pigs and horses on the other hand. Therefore, comparisons and extrapolation of cardiovascular data between those animals and humans must be done with caution. In addition, it is of little significance to compare the equine heart and its large blood vessels with those of men, due to the difference in heart size. Because of the different orientation of the heart in the human chest compared with pigs and horses and the distinct position of the heart relative to the lungs, there are not only differences in nomenclature but the localization and orientation of the pulmonary veins is also different.

1.2 Cardiac conduction

1.2.1 Normal and abnormal cardiac conduction

1.2.1.1 Morphology

In the normal heart, the conduction is influenced by the sinoatrial node, the atrioventricular node and the His-Purkinje fibers. Histologically, a sinoatrial node is built-up by a mesh of connective tissue containing blood vessels, fat, collagen and nerve tissue and is surrounded by paranodal cells (Monfredi et al., 2010). The sinoatrial node acts as the pacemaker of a healthy heart which generates continuously electrical pulses, as action potentials, which are dispersed over the atria towards the atrioventricular node to determine the heart rhythm. Afterwards, the pulses are rapidly sent over the His bundle towards the Purkinje fibers and ventricular myocardium. Especially, the atrial conduction is of utmost importance in the context of this manuscript and mainly influenced by the build-up of the atrial wall. The architecture of the atrial myocardial tissue can be categorized into inter- and

Chapter 1: General introduction intra-atrial cardiac muscle bundles, organized into a complex pattern which is specific per heart (Corradi et al., 2011). Some structures demonstrate a constant pattern in the left atrium, with the most important indicated as the superficial Bachmann’s bundle. This bundle is located at the anterior wall of the left atrium, parallel to the atrioventricular groove and is the most important inter-atrial connection, playing a major role in inter-atrial conduction (De Ponti et al., 2002; Ariyarajah and Spodick, 2006). The sinoatrial node and the heart itself are also innervated by the autonomic nervous system as sympathetic and parasympathetic nerves conduct stimuli, changing the cardiac rate. Sympathetic stimuli will raise the speed of action potential formation and thus also the speed of cardiac contractions whereas parasympathetic influence will slow down cardiac rhythm (Bright and Marr, 2010; Corradi et al., 2011; Gordan et al., 2015).

Besides the influence of the central nervous system on the cardiac rhythm through sympathetic and parasympathetic post-ganglionic neurons, it was indicated that the cardiac control is also regulated by an intra-thoracic nervous system. This cardiac autonomic nervous system is composed by an intrinsic and extrinsic part. The intrinsic part exists of post-ganglionic neurons from the vagal nerve which constitute the so-called epicardial ganglionic plexi. The extrinsic part, also indicated as cardiac plexus, is connected with the vagal nerves and with the sympathetic stellate ganglion through post- ganglionic sympathetic nerves (Fig. 10A, B) (Lo et al., 2010; Barone, 2011). The stellate ganglion is located near the spinal cord. Parasympathetically, pre-ganglionic nerves from the brain stem synapse to post-ganglionic nerves in the cardiac plexus. Both sympathetic and parasympathetic autonomic nerves from the cardiac plexus may synapse with the intrinsic cardiac neurons (Hildreth et al., 2009).

The cardiac plexus of pigs resembles to the plexus of other domestic animals (McKibben and Ghoshal, 1975). Cardiac ganglia can be divided into superficial and deeper located ones. The latter are situated intramurally of which two are distinct, the sinoatrial and the atrioventricular ganglia. The other intramural ganglia are organized in small groups of ganglionic cells, located at the level of the main vascular divisions, distributing postganglionic fibers (Barone, 2011). In human hearts, it was indicated that aforementioned epicardial ganglionic plexi, located in epicardial fat pads at the left atrial-PV junction, send epicardial nerves between the sleeves’ myocardial cells and near the PV endothelium (Vaitkevicius et al., 2009; January et al., 2014). Regarding the function of this nerve tissue, 90% of the ganglia and almost 25% of the nerve fibers include cholinergic and adrenergic structures (Tan et al., 2006). The co-localization of the sympathetic and parasympathetic system is important in the context of treatment options, which will be discussed later on, as a selective approach during clinical interventions is impossible (Tan et al., 2007).

Chapter 1: General introduction

Armour et al. (1997) described ganglionated plexi (groups of ganglia containing several neurons and connected with each other via different nerves) at five ventricular and five atrial locations in the human heart. A couple of neurons may constitute smaller ganglia but also large ganglia of 0.5x1mm were observed (Fig. 11, 12) (Armour et al., 1997). The same research was performed in dogs by Yuan et al. (1994), describing epicardial ganglionated plexi in three ventricular and four atrial regions. At the level of fetal human PVs, epicardiac ganglia were indicated in consistent regions and interconnected through thin nerves with the largest density of ganglia situated at the roots of the veins. The distal parts of the PVs demonstrated only sporadically ganglia. An uneven distribution of ganglia in relation to the total ganglion area was noticed along the PV length (Fig. 13) (Tan et al., 2006; Vaitkevicius et al., 2008). The innervation of the human extrapulmonary parts of the PV was already described earlier but it was further mentioned that, specifically the left atrium is innervated by three epicardiac subplexi which send nerves towards the PVs. Overall, when comparing men with dogs, it was concluded that the distribution pattern of the intrinsic cardiac plexi is quite similar. However, regarding the subplexi, a significant difference was noticed (Pauza et al., 2000). Besides the differences present between different species, it was also indicated that the number of neurons in the epicardiac ganglia is lower in aged persons compared to infants (Pauza et al., 2000).

A B

Fig. 10. A: Schematic overview of the innervation of the heart. Black: parasympathetic efferent nerves; grey: sympathetic efferent nerves, dashed lines: afferent nerves, light grey: intrinsic cardiac neurons. (From Hildreth et al., 2009). B: Dorsal view of the equine heart with aorta (AO), pulmonary arteries (PA), pulmonary veins (PV), vagal nerves (VN), stellate ganglion (SG), cardiac plexus (CP) and nerves of the left atrium (NL) indicated. (Adapted from Barone, 2011)

Chapter 1: General introduction

Fig. 11. Human atrial ganglionated plexi dispersed over mainly the left and right atrium and around the cardiac vessels. A: posterior view; B: superior view; SLA-GP: superior left atrial ganglionated plexus; PLLA-GP: posterolateral left atrial ganglionated plexus; PMLA-GP: posteromedial left atrial ganglionated plexus. (From Armour et al., 1997; Corradi et al., 2013)

A Fig. 12. Schematic overview of the cardiac ganglia locations in men. A: four-chamber view of the heart; B: dorsal view of the heart. White circles: cardiac ganglia, Ao: aorta; RA: right atrium; LA: left atrium; PT: pulmonary trunk; PV: pulmonary veins; SCV: superior caval vein; RV: right ventricle; LV: left ventricle; SN: sinus node; AVN: atrioventricular node; ICV: inferior caval vein. (From Hildreth et al., 2009)

Chapter 1: General introduction

Fig. 13. Macrophotograph of the ganglionated nerves located on the human pulmonary veins. LD: left dorsal subplexus, LIPV: left inferior pulmonary vein, LSPV: left superior pulmonary vein, black arrowheads: epicardiac nerves and/or ganglia. (From Vaitkevicius et al., 2008)

1.2.1.2 Physiology

When an electrical signal reaches the heart or a stimulus is propagated from the sinus node, the cardiomyocytes are activated by an action potential. These action potentials coordinate the function of the heart, by regulating ion channel activity, to result into an efficient heart contraction (Stephenson, 2007). During rest stage, a rest membrane potential determines the amount of different electrolytes in and outside the cell. More specific, in this stage the cell membrane is permeable for potassium but not for calcium and sodium and a negative intracellular rest potential stimulates the inflow of potassium. As cardiac tissue is activated by stimuli and a depolarization wave is sent through the cardiomyocytes, the sodium-channels, which are voltage dependent, will open and permit the inflow of extracellular Na+. This process will cause a shift to a positively charged cytosol. Afterwards, as this pathway stops, the channels are inactivated and a repolarization process will start. However, in cardiac tissue, this repolarization is postponed (known as the refractory period) due to an extended depolarization phase in which the potassium flow is blocked and calcium- channels open to enable the inflow of Ca2+. Finally, the potassium-channels will open and K+ will leave the cell and the calcium-channels will close which causes a repolarization of the cell membrane to return to the negatively charged rest potential.

The atrial conduction is mainly determined by three major internodal tracts. After excitation of the sinoatrial node, the depolarization wave will spread throughout both atria. This wave can move directly from the sinoatrial node towards the adjacent myocardial cells and so be transferred to the other surrounding myocardial cells. However, the three anatomic conduction pathways from the sinoatrial node towards the atrioventricular node compose a short route between both nodes. From the anterior part of the sinoatrial node, the Bachmann’s bundle is a branch of the anterior pathway, delivering impulses towards the left atrium. The middle internodal tract (Wenckebach’s pathway) starts from the superior part of the sinoatrial node, travels posteriorly towards the superior vena cava and via the atrial septum towards the atrioventricular node. The third tract, called posterior

Chapter 1: General introduction tract, starts from the inferior part of the sinoatrial node, passes through the crista terminalis to end in the posterior part of the atrioventricular node (James, 1963).

Besides the autonomic nerve system, the cardiovascular function is also determined by several endocrine hormones. In this context, the adrenal gland is important which releases (nor)epinephrine (or (nor)adrenalin) and its precursor dopamine. Epinephrine is produced by a specific cell type (chromaffin cells) inside the adrenal gland medulla when stimulated through sympathetic nerve fibers. Both catecholamines can act as a hormone or a neurotransmitter which are involved in the elicitation of the “fight/flight-response”. (Nor)adrenalin can activate or deactivate the cardiac system via sympathetic receptors. As indicated, dopamine can convert into norepinephrine which may increase the heart rate and blood pressure (Gordan et al., 2015).

1.3 Atrial fibrillation (AF)

1.3.1 AF in man

In humans, atrial fibrillation is the most common cardiac arrhythmia with increased risk of morbidity and adverse effects as stroke and death (Go et al., 2001; Boriani and Pettorelli, 2016). During atrial fibrillation, the upper chambers of the heart do not work properly but demonstrate a rapid and irregular contraction due to an abnormal electrical conduction (Falk, 2001).

The cardiac depolarization process can be evaluated by use of electrocardiography. Electrocardiography, by which the cardiac rhythm is visualized and cardiac conduction is monitored, is an important tool to diagnose cardiac abnormalities or disorders such as AF in men, dogs and horses (McGurrin, 2015). AF is recognized electrocardiographically by irregular and rapid atrial depolarizations. In addition, no P waves are present but F-waves are seen instead and the RR interval is irregular (Xu et al., 2016). Due to this diagnostic method, AF can be distinguished from other abnormal cardiac rhythms that may sound similar during auscultation, such as sinus arrhythmia which evanishes during light exercise (McGurrin, 2015).

1.3.1.1 Epidemiology of AF

Overall, AF incidence in humans increases with age. Over the age of 55 years, the prevalence of AF doubles each decade in humans as an aging heart is the ideal basis for AF due to atrial dilatation and myocardial fibrosis. People at the age of 40, have an average risk of 25% to develop AF over the total lifespan (Lloyd-Jones et al., 2004). According to gender and age, the prevalence of AF ranges from 0.12% to 0.16% in the category below 49 years; 3.7% to 4.2% in the category between 60 and 70 years and over 9% in octogenarians (Benjamin et al., 1994; Boriani et al., 2006; Camm et al., 2010;

Chapter 1: General introduction

Piccini et al., 2012; Zoni-Berisso et al., 2014; Karamichalakis et al., 2015). Overall, 1-2% of the population is affected by AF (Ohsawa et al., 2005; Majeed et al., 2001; Andrade et al., 2014). However, these numbers may be underestimated as 10-40% of the patients with AF are asymptomatic (Boriani and Pettorelli, 2016).

Men demonstrate a 1.5 higher risk of developing AF compared to women based on data adjusted for various risk factors of AF. Regardless of gender, several abnormal conditions are significantly linked with AF risk such as diabetes, hypertension, heart valve disorders and congestive heart failure. Women with valve disease demonstrate a significant higher risk of developing AF compared to men with a valve disease (Benjamin et al., 1994). Excessive sports activity is also linked with a higher prevalence of AF (Andrade et al., 2014).

Regarding the effect on society, AF constitutes a major public health problem. It was estimated that this arrhythmia represented 1% of the National Health Service budget of the year 2000 in the UK (Stewart et al., 2004) and accounted for 16 to 26 billion dollar of annual US expenses in the years 2004-2006 (Lee et al., 2008; Kim et al., 2011). In the US, AF patients are hospitalized twice as often compared to patients without AF, AF is responsible for more than 99000 deaths per year and AF prevalence was estimated to increase from around 5 million cases in 2010 to around 12 million cases in 2030 (Kim et al., 2011; Colilla et al., 2013; Go et al., 2014).

1.3.1.2 Pathophysiology 1.3.1.2.1 AF eliciting factors and adverse effects

Clinically, AF is a very heterogeneous disorder. It may occur as a disease on its own, as recurrent paroxysmal lone AF or as an idiopathic condition in a healthy patient. Paroxysmal lone atrial fibrillation is a condition in which AF terminates spontaneously and no other underlying cardiac disorder could be identified. Other AF states, such as persistent and permanent AF, are explained in table 1. However, it may also be associated with or secondary to other diseases. AF can be present as a complication of another cardiac or extra-cardiac disease. Some examples of AF eliciting conditions are subclinical atherosclerosis, excessive exercise, smoking, drugs, obesity, coronary heart disease, dilated cardiomyopathy, hepatic disease and chronic obstructive pulmonary disease (Boriani et al., 2006; Camm et al., 2010; Boriani et al., 2011; Kirchhof et al., 2013; Kirchhof et al., 2016). Hyperthyroidism may also provide a higher risk of AF induction. An aberrant level of thyroid hormones can induce a shortening of the action potential duration and an increase of left atrial pressure and heart rate (Bielecka-Dabrowa et al., 2009).

Chapter 1: General introduction

The main adverse effect of AF is the increased risk of stroke. Due to the irregular or absent contractions of the atria, the blood in the atria becomes static, which can lead to blood clotting and embolism. As a result, blockage of the cerebral blood flow may lead to a fivefold increased risk of cardioembolic stroke (Kannel et al., 1998; Leary and Caplan, 2008; Copley and Hill, 2016). Besides AF, other risk factors for stroke are higher age, hypertension, diabetes and a prior stroke with the latter as the most important indicator for high stroke risk (average of 10%/year) (Hart et al., 2007). Overall, AF leads to a 1.5-1.9 fold increased risk of mortality (Kannel et al., 1998).

1.3.1.2.2 Mechanism of AF

Research in the past decades has unraveled that this supraventricular tachyarrhythmia originates from an interplay between ectopic electrical activity, abnormal atrial tissue and genetic predisposition. This leads to remodeling and deterioration of the substrate, cardiac tissue, creating abnormally propagating stimuli which results into an ineffective contraction of the atria (Camm et al., 2010; Fuster et al., 2011; Calkins et al., 2012; Heijman et al., 2014). Thus, abnormal impulses act as an ectopic activity disturbing the normal cardiac activation. Specifically for the PV, the “electrical automaticity of the PV” is deduced from its independent pulse generation capacity (Brunton and Fayer, 1876). Moreover, in terms of remodeling of cardiac tissue, Iwasaki et al. (2011) summarized that AF depends on structural, neural and electrical processes. Overall, AF demonstrates a progressive character, evolving from a paroxysmal to a persistent and finally to a chronic or permanent form. However, this pathway is not seen in all patients (Heijman et al., 2014).

Paroxysmal AF AF terminates spontaneously or with intervention within 7 days Episodes may occur with variable frequency. Persistent AF Continuous AF that is sustained more than 7 days Long lasting persistent AF Continuous AF that lasts longer than 12 months Permanent AF Term is used when the decision is made to stop further treatments Table 1. Overview of the different types of atrial fibrillation (AF). (Adapted from January et al., 2014a)

1.3.1.2.3 Anatomical and histological basis

PVs are known to be important in the elicitation and maintenance of AF, as around 94% of the AF initiating triggers are found inside the PVs and reach the left atrium through myocardial tissue, indicated as the myocardial sleeve (Haïssaguerre et al., 1998). This is generally not limited to one specific vein as paroxysmal AF patients are often diagnosed with different foci located in multiple veins in which ectopic pulses are generated distally (Fig. 14). In contrast, eliciting foci close to the PV ostium were found in patients, diagnosed with focal atrial tachycardia of which the disturbing pulses

Chapter 1: General introduction originated from within the PVs (Kistler et al., 2003). In men, the most important locations of non-PV ectopic triggers imply the posterior side of the left atrium, the coronary sinus, the ligament of Marshall, the superior vena cava, atrial ganglionated plexi and the region near the atrioventricular valves (Tsai et al., 2000; Calkins et al., 2012).

The PV myocardial sleeve architecture in men and dogs, as mentioned before, demonstrates fibers changing direction abruptly. This facilitates the development of focal ectopic activity (Hocini et al., 2002; Heijman et al., 2014). Till now, no node-like or conducting cells have been found which explain the arrhythmogenicity of the PVs. The superior PVs are found to be more arrhythmogenic compared to the inferior veins and this was attributed to the presence of longer myocardial sleeves in the former ones (Nathan and Eliakim, 1966; Haïssaguerre et al., 1998). In human AF patients, thicker PV sleeves are found with thickened areas characterized by ectopic foci (Guerra et al., 2003).

Several studies suggest that the PVs contain cells, which may act as pacemakers, based on immunohistochemical experiments with different antibodies (HNK-1, CCS LacZ, HCN4) related to cardiac conduction (Blom et al., 1999; Jongbloed et al., 2004; Harhun et al., 2004; Morel et al., 2008; Nguyen et al., 2009). This suggests that triggered activity and automaticity may be a major cause of PV arrhythmogenesis (Mahida et al., 2015). In this context, it was demonstrated that the myocytes of the canine PVs have different electrophysiological properties compared to the myocytes of the left atrium (Melnyk et al., 2005). Even specific cell types such as Cajal-like cells or telocytes were ultrastructurally identified at the level of the myocardial tissue of the PVs (Gerghiceanu et al., 2008; Cretoiu and Popescu, 2014). Concerning the role of these cell types, several paths are being investigated in literature but no unambiguous conclusion has yet been made.

Cardiac innervation can also play a role in AF pathology as this arrhythmia may be provoked through autonomic stimulation, more specific due to activation of sympathetic and/or parasympathetic pathways. Indeed, an increased activity of the latter was noticed before AF initiation in a few animal models and humans (Scherf et al., 1948; Moe and Abildskov, 1959; Choi et al., 2010; Arora, 2012; Calkins et al., 2012; He et al., 2012; Park et al., 2012; Shen et al., 2012; Heijman et al., 2014). Patterson et al. (2005) stipulated an interaction between the PVs and the atrial autonomic nerves which may be important in arrhythmogenesis. Specifically, a mechanism was suggested in which PV action potentials could be shortened by cholinergic stimulation. Vagal-induced-AF can be observed in some cardiac healthy patients during high parasympathetic activity as after having a meal or during sleep. AF can also be induced during exercise, caused by sympathetic activation which may create ectopic foci (Patton et al., 2005; January et al., 2014b). In animal studies, stimulation of the epicardial ganglionic plexi caused an increased atrial activity (Pappone et al., 2004b; Scanavacca et al., 2006).

Chapter 1: General introduction

Dog studies proved that stimulation (electrical or with acetylcholine) of cardiac autonomic ganglia (near the base of the PVs) provokes focal PV ectopy, leading to AF (Scherlag et al., 2005a; Scherlag et al., 2005b; Po et al., 2006).

Structural remodeling of myocardial tissue (Fig. 14) (changes of the heart wall structure) due to hypertrophy, inflammation and especially fibrosis is a major contributing factor in sustaining AF. Those adaptations, mostly caused by an underlying cardiac disease, can increase left atrial pressure and wall stress (Kistler et al., 2004; January et al., 2014). Fibrosis implies the replacement of cardiomyocytes with fibroblasts, which may slow the conduction of electrical signals. Also, many fibroblasts, linked electrically to cardiomyocytes, may promote re-entry or ectopic activity (Burstein et al., 2008; Nattel et al., 2008; Schotten et al., 2011; Yue et al., 2011; Wakili et al., 2011). Besides fibrosis, cellular ultrastructural adaptations such as glycogen accumulation, gap junction dysfunction or disturbance of the mitochondrial function, leading to hypocontractility of the atria, are characteristic for AF (Ausma et al., 1997; Ausma et al., 2003; Andrade et al., 2014).

Chapter 1: General introduction

Fig. 14. Scheme visualizing AF mechanisms.

In addition, atrial dilatation is often linked with a higher risk for AF and AF can be promoted by atrial stretch (Kalifa et al., 2003). Atrial stretch may affect the action potential duration due to stretch- sensitive channel adaptation causing spontaneous activity (Ravens, 2003). Summarizing, structural remodeling is often caused by AF itself which promotes development of a permanent AF state in persistent AF patients as structural adaptation of the myocardium can become irreversible (Burstein et al., 2007; Nattel et al., 2008; de Groot et al., 2010; Iwasaki et al., 2011).

Chapter 1: General introduction

1.3.1.2.4 Electrophysiological basis

Fundamental electrophysiological mechanisms in the development of AF are ectopic, triggered activity, enhanced automaticy and re-entry patterns (Fig. 14) (Comtois et al., 2005; Wakili et al., 2011; Atienza et al., 2012; Mahida et al., 2015). In these mechanisms, abnormal impulses are generated which disturb the sinus rhythm. However, AF is caused not only by eliciting impulses but this pathology is induced and maintained by the interaction between initiating triggers and an underlying atrial substrate. Both structural and electrical properties may contribute to increased susceptibility of the substrate to AF.

The PVs play a major role in AF induction and transfer of abnormal electrical pulses. Indeed, the number of AF paroxysms decreases after ablation of the PVs which indicates that the PVs are important in maintaining paroxysmal AF (Haïssaguerre et al., 2004a; Jais et al., 2006). (Ablation is discussed further in section 1.3.1.3). As such, the PVs play an important role as a trigger regarding persistent and permanent AF (Corradi et al., 2013). In permanent and persistent AF patients, structural remodeling of the myocardium is noticed and this is influenced by electrophysiological remodeling, which is further discussed.

Similar to the PVs, it has been demonstrated that the superior vena cava is the second major source of ectopic foci causing AF. Myocardial sleeve tissue links the right atrium and the vena cava, enabling electrical conduction towards the right atrium which may trigger an abnormal heart rhythm. Isolated canine myocardial sleeve myocytes from the superior vena cava display a specific ionic current profile and action potential which may possibly lead to arrhythmias (Chen et al., 2002).

Automaticity is the characteristic of cardiac cells to spontaneously induce action potentials. The sinoatrial node cells demonstrate the fastest intrinsic rate of impulse generation. Therefore, the sinoatrial node acts as the primary pacemaker of a normal heart. Under normal circumstances, atrial and ventricular myocardial cells do not show any automaticity or spontaneous depolarization. However, in abnormal conditions, myocardial cells can develop the property of repetitive impulse initiation (Antzelevitch and Burashnikov, 2011).

Re-entry can be described as the return of the same depolarization into a part of myocardium which was recently activated by the same depolarization wave. Re-entry is not a disorder of pulse induction but rather a condition of abnormal pulse propagation (Veenhuyzen et al., 2004). While a normal depolarization wave moves rather streamlined throughout the myocardium, the reentry wave shows a more circular or irregular conduction through the myocardium. This, potentially in combination

Chapter 1: General introduction with a shortened refractory period, leads to a chaotic tangle of electrical signals which maintain AF and may induce atrial contractile dysfunction (Calkins et al., 2012).

Re-entry usually starts with an ectopic impulse and a region of slow conduction. During re-entry, the electrical signals travel continuously around an anatomically defined spot. In anatomical re-entry, the boundary of the excitation pathway is a physical cardiac structure (Fig. 15) (Veenhuyzen et al., 2004). The atrial myocardial tissue becomes re-excitable before the re-entry pulse travels along which allows the impulse to restart depolarization (Comtois et al., 2005). In addition, re-entry may also be present around a functional substrate (Fig. 16). The boundary of a functional re-entry pathway is not an anatomical structure but rather variations in electrophysiologic characteristics of contiguous tissues (Veenhuyzen et al., 2004; Comtois et al., 2005; Aguilar and Nattel, 2015). Concerning the re- entry patterns, the action potential differences between PV cells and atrial myocytes suggest that micro-re-entry may be favored in the myocardial sleeves (Fig. 17). Since PV myocardial cells are more rapidly returning to an excitable state, micro re-entry patterns make the PVs particularly suitable for inducing abnormal stimulus formation (Ehrlich et al., 2003).

Chapter 1: General introduction

Fig. 15. Macro re-entry mechanism. a) A sinus impulse activates area A. b) A premature beat arising in area B fails to reach area A because the intervening tissue remains refractory from the preceding sinus beat. c) The premature stimulus travels slowly via an alternative route back to area A, allowing enough time for area A to recover and be excited. d) Area A re-excites area B and the cycle sustains itself. (From Veenhuyzen et al., 2004)

As mentioned, re-entry mechanisms may be important in sustaining AF. This mechanism can be facilitated by an increase of the potassium outflow or a decrease of the calcium inflow as this will speed up the repolarization process and shortens the atrial refractoriness and atrial action potential duration. Subsequently, abnormal electrical waves may result from this described pathway (Nattel, 2002; Iwasaki et al., 2011; Schotten et al., 2011; Wakili et al., 2011; Andrade et al., 2014). Specifically

Chapter 1: General introduction for calcium, each action potential is characterized by the inflow of calcium into the atrial cells. During AF, the high rate of depolarizations risks to lead to potentially dangerous intracellular calcium overload. Therefore, a downregulation of L-type-Ca channels is induced as a protection mechanism. But at the same time, this reduced calcium flow leads to a shortening of the action potential, which can stabilize re-entry patterns and enhance the AF sensitivity and durability (Goette et al., 1996; Allessie et al., 2001; Nattel et al., 2002; Pandit et al., 2005; Iwasaki et al., 2011).

Fig. 16. Abnormal pulses, originating within the PV wall, enter the left atrium which may lead to the formation of multiple micro re-entry circuits. These stimuli are filtered by the atrioventricular node and conducted towards the ventricles. (From Safaei et al., 2011)

Fig. 17. Anatomic and functional re-entry mechanisms. In anatomical re-entry, the pathway size is determined by anatomical structures (left image). In functional re-entry (middle image), the conduction velocity and refractory period determine the size of the pathway. If the circuit size is too small (right image), the pathway will extinguish itself as the wavefront will collide with refractory tissue. The circuit will be too small when the wavefront moves too fast or the refractory period is too long. (Adapted from Veenhuyzen et al., 2004)

Fast and repetitively firing foci can play the role of AF-inducing and supporting drivers (Fig. 18) (Heijman et al., 2016). These stimuli constitute a self-perpetuating process that is both the cause and

Chapter 1: General introduction consequence of the structural and electrophysiological changes of the atrial myocardium (Corradi et al., 2013). This can be specified as electrophysiological remodeling, indicating a change of cardiac ion channel function which may lead to AF development and even result into persistent AF. This type of remodeling is mostly caused by AF, meaning this is a self-reinforcing process (Allessie et al., 2001; Nattel, 2002; Nattel et al., 2008; Schotten et al., 2011; Wakili et al., 2011).

Fig. 18. A: ectopic foci from the PV disperse abnormal stimuli through the left (LA) and right (RA) atrium; B: one pattern of abnormal impulses indicated as “single-circuit re-entry” may originate inside the left atrium, propagating electrical stimuli inside the left atrium and towards the right atrium; C: several re-entry circuits can persist inside the left and right atrium; SVC: superior vena cava; IVC: inferior vena cava; PVs: pulmonary veins. (From Iwasaki et al., 2011)

Concerning abnormal impulse formation, ectopic foci consist mostly of delayed after-depolarisations but sometimes also of early after-depolarizations (Patterson et al., 2005; Andrade et al., 2014; Voigt et al., 2014). Abnormal calcium exchange is important in delayed but may also play a role in early after-depolarizations, especially in combination with a sympatho-vagal stimulation (Patterson et al., 2005).

The first hours of atrial tachycardia, which may result into AF, induce electrophysiological remodeling, mainly shortening of the refractory period. This is followed by structural remodeling which takes much more time and this mechanism will eventually lead to morphological adaptations of the atrial myocardium (Corradi et al., 2008; Grandi et al., 2011).

Besides structural and electrical remodeling, neural or autonomic remodeling also plays an important role in AF. As mentioned before, the autonomic nerve system may be involved in AF (Chou et al., 2009). Parasympathetic stimulation influences the effective refractory period and atrial action potential duration, and increases dispersion of atrial repolarization, constituting arrhythmogenic substrates for AF (Cao et al., 2018). It was noticed that in patients with persistent AF, sympathetic hyper-innervation of the atria can be present (Gould et al., 2006). The so-called autonomic remodeling may act as a part of the atrial substrate, promoting AF recurrence or persistence (Gould et al., 2006; Nishida et al., 2011).

Chapter 1: General introduction

1.3.1.2.5 Genetic predisposition

The genetic link between AF and the role of the PVs in this cardiac disorder is not completely elucidated (Mahida et al., 2015). First degree family members from people with AF have a two-fold increased risk for developing AF (Fox et al., 2004; Arnar et al., 2006; Christophersen et al., 2009). The most important gene that may be related to the AF mechanism is the PITX2 gene. In human studies and animal models different correlations between the PITX2 gene and AF were postulated, from its role in the pulmonary myocardium development to its altered expression leading to an increased arrhythmia susceptibility of the heart (Mommersteeg et al., 2007; Wang et al., 2010; Kirchhof et al., 2011; Scridon et al., 2015). Several studies demonstrated that mutations in ion channels were noticed in individuals with familial AF and this seemed to be a cause of arrhythmia (Chen et al., 2003; Ellinor et al., 2004; Yang et al., 2004).

1.3.1.3 AF treatment

Primarily, drugs and/or electrical cardioversion are used as treatment options to control heart rate and rhythm. Rate control is based on drugs such as beta-blockers and anticoagulantia while rhythm control is performed by use of drugs or electrical cardioversion to restore sinus rhythm. Rate control treats the side-effects of AF but the patient remains in AF. After rhythm control, however, in many cases AF recurrence is seen because of AF re-initiation by ectopic foci from the PVs. Isolation of the PVs by ablation has become the cornerstone in AF treatment as it was demonstrated that catheter ablation is superior to antiarrhythmic drugs to obtain an AF-free patient (Haïssaguerre et al., 1998; Wazni et al., 2005; Santangeli et al., 2012). Catheter ablation of the PVs is an intervention in which circular lesions are applied around the PVs ostia to isolate the latter anatomically and electrically from the left atrium (Fig. 19) (Chen et al., 1999; Wazni et al., 2005). An ablation procedure in which ablation lines are created in myocardial tissue can be performed by use of various energy sources such as cryoablation, radiofrequency or high-intensity focused ultrasound (Melby et al., 2005; Khargi et al., 2007; Calkins et al., 2012). The application of such an energy source will induce the formation of non-conductive fibrosis. In case of paroxysmal AF, isolation of the PVs due to an ablation procedure is frequently effective to maintain sinus rhythm. The long-term success rate of ablation procedures is lower, as Ganesan et al. (2013) reported success rates of respectively 54% and 79% for single- and multiple-interventions in paroxysmal AF patients during a follow-up study of more than three years. Patients with sustained AF require a combination treatment of medication or cardioversion and ablation to prevent AF recurrence. However, some paroxysmal (30%) and the major part (78%) of persistent/permanent AF patients retain this arrhythmia after ablation and after 5 months of follow-up (Oral et al., 2002), probably due to severe remodeling of myocardial tissue. A

Chapter 1: General introduction large study, following 445 patients for 10 years after 1 year of PVI success, demonstrated that the recurrence rate for paroxysmal AF patients was 3%, 11% and 27% at 2, 5 and 10 years respectively. In persistent AF patients, these percentages were clearly higher as a recurrence rate of respectively 13%, 29% and 62% was found for the same study period (Steinberg et al., 2014). Overall, the results of different ablation trials are difficult to compare as the population of AF patients is very heterogeneous, based on the duration and type of arrhythmia and other present pathologies. Moreover, the outcome or interpretation of the results of an ablation therapy may depend on the various definitions of success and different set-up of a follow up study. Also, an important remark is the fact that only a limited number of studies reported the success rate of ablation trials of persistent AF patients after more than 3 years (Tutuianu et al., 2015). The presence of non-PV triggers may be an important cause of treatment failure or recurrence of AF (Chen et al., 1999; Oral et al., 2002). In paroxysmal AF patients, isolation of the PVs prevents recurrence in 75%. However, complications such as the creation of an atrio-esophageal fistula is seen in 1% to 5% of the cases (Pappone et al., 2004a; Fang et al., 2007; Li et al., 2011). A worldwide survey indicated that more than 24% of the ablated patients needed a second ablation treatment and about 3% required a third procedure. Treatment failure can be caused by incomplete isolation of the PVs or reconnection of the PVs during the post-operative recovery phase. The ablation success rate of patients free of antiarrhythmic drugs was almost 53% in paroxysmal patients only and 48% in paroxysmal and persistent patients combined (Cappato et al., 2005).

Fig. 19. Ablation and mapping catheters are placed inside the left atrium through the interatrial septum. In the left atrium, circular ablation lines are created around the PV ostia. (From Safaei et al., 2011)

Circumferential ablation of the PVs in paroxysmal AF patients is a standard treatment, which is not the case in persistent AF patients as it was noticed that this type of AF is less dependent from the PVs (Smelley and Knight, 2009). In such patients, the re-entry spots and triggering foci can be located outside the PVs due to structural remodeling of the atria (Eckstein et al., 2008). Treatment of chronic

Chapter 1: General introduction

AF patients mostly requires ablation of the altered substrate which sustains the AF mechanism (Verma, 2011).

Some studies indicate an improved result when ablating the ganglionic plexi combined with PV isolation compared to PV isolation alone, due to suppression of autonomic signaling (Pappone et al., 2004a; Scanavacca et al., 2006; Nishida et al., 2011; Pokushalov et al., 2013; January et al., 2014b). However, some researchers question the possible benefit of ganglionic plexi ablation (Santangeli et al., 2012). Po et al. (2009) indicated a success rate of 88% and 86% to keep paroxysmal or persistent AF patients free of AF symptoms until 12 and 22 months, respectively, after a single combined ganglionic plexi and PV ablation procedure, stressing that this result was achieved because of autonomic neuron destruction, which is unable to regenerate. In addition, Platt et al. (2004) demonstrated that destruction of ganglionic plexi alone made 96% of the patients non-inducible to AF. In this context, Lemola et al. (2008) concluded that the ganglia are important structures during vagally-induced AF rather than the PVs itself. After percutaneous ablation and mapping technique development, which is much less radical and carries less risk compared to older techniques in which the thorax was opened, knowledge of the anatomy of the pericardial cavity became more important to perform epicardial ablation procedures. This proved to be a helpful technique to treat difficult persistent AF cases (Pak et al., 2007; Ernst et al., 2009; Lachman et al., 2010).

Nowadays, electrical isolation of the PVs is mostly achieved during an ablation procedure. However, the main reason for atrio-venous reconnection is the creation of non-transmural lesions during the intervention, as will be discussed further, and/or the presence of gaps in conduction block within the applied ablation line, which may allow PV triggers to enter the atrium and reinduce AF (Ouyang et al., 2005; Cheema et al., 2007; Kowalski et al., 2012; Ranjan et al., 2012). However, also a seemingly successful ablation procedure can lead to local conduction recovery afterwards. This phenomenon can be caused by an electrical uncoupling at the ablation spot, rather than the induction of cell death. This uncoupling and loss of electric pulses is a reversible result and hence a non-permanent conduction disturbance is established. In addition, the process of tissue heating itself can provide an acute conduction blockage which is transitory (Nath et al., 1994a,b; Wood and Fuller, 2002; Kumar and Michaud, 2016). Another reason for a seemingly successful electrical isolation of the PVs without the actual presence of a complete circular lesion around the PV ostia may be the fact that edema, which may establish a conduction block, can be located at the level of the applied lesion (Miller et al., 2012). In the weeks following after an ablation procedure, the affected tissue can recover by which conduction is reestablished through reappearing gaps in the ablation line (Ranjan et al., 2011). In recent years, major technical advancement took place in the development of catheter ablation and other ablation techniques. However, an ablation procedure remains a challenging, skill-demanding

Chapter 1: General introduction and a time-consuming intervention (Shin and Deneke, 2010). This emphasizes the need for the development of new ablation techniques with a high success rate which allow a faster retreatment and preferably without the need of a catheterization lab.

Overview of current AF treatment strategies in men

In recent years, many techniques were and are still being developed and adapted to enhance the success rate of ablation procedures. Besides the treatment options discussed above, in certain cases a surgical approach is used (Xu et al., 2016). As mentioned before, rate control includes the treatment of the side-effects of AF while AF itself remains, by using beta-blockers and anticoagulantia. Rhythm control, on the other hand, aims to treat AF with drugs and cardioversion. Ablation therapy also requires the application of anticoagulantia before and during an intervention to prevent clot formation.

A combination of an ablation procedure and drug therapy is of utmost importance in the successful management of this pathology. In fact, a major drawback in evaluating an ablation procedure may be the long post-operative observation time necessary to evaluate the effect of the intervention. After an incomplete ablation procedure, a conduction block may be observed during the first minutes, but may disappear after 30 to 60 minutes. This is caused by an initial depolarization of the resting membrane potential and sodium-channel inactivation which demonstrates inexcitability and conduction block (Ehrlich et al., 2003). Afterwards, a hyperpolarization can be noticed, leading to PV reconnection (Datino et al., 2010). These are called dormant PVs. Adenosine appears to be useful herein as adenosine may shorten the observation time to detect dormant conductive PVs. Adenosine facilitates membrane hyperpolarization which restores the excitability threshold (Gourraud et al., 2016). Isoproterenol, on the other hand, may be useful in paroxysmal and persistent AF patients as it may reveal non-PV triggering foci (Lee et al., 2005). Isoproterenol, a cardiac adrenoreceptor agonist, increases the intracellular calcium level and decreases the action potential duration and atrial refractory periods. It is used to “revive” a dormant focus. This product also facilitates abnormal automaticity and triggered activity (Liang et al., 1985; Chen et al., 2010). The organization of electrical foci in patients with persistent AF may be achieved by use of ibutilide and amiodarone but these products will not efficiently reduce AF recurrence. In persistent AF patients, the abnormal substrate is frequently located outside the antral region of the PVs. During AF, certain spots distribute signals with a very short cycle length or which are fractionated and/or with disturbance of the baseline. These spots demonstrate slow conduction, conduction block or are important for AF re- entry continuation. Ibutilide and amiodarone might facilitate the identification of these sites which

Chapter 1: General introduction are critical to AF preservation. After which, the firing focus may be retrieved and ablated (Gourraud et al., 2016).

Drugs are also used both in early anti-arrhythmic and early anticoagulant therapy (Xu et al., 2016). In general, stroke risk assessment is very important to determine the treatment strategy to improve in a most efficient way the outcome of the treatment and quality of an AF patient’s life. However, there are also specific conditions in which anti-arrhythmic pharmacological products are not advised as it may lead to ventricular fibrillation and higher morbidity. An example is Wolff-Parkinson-White syndrome, in which an extra connection between the atrium and ventricle can induce pre-excitation (Falk, 1992).

During rhythm control, the early anticoagulant therapy is necessary before cardioversion can be executed if the AF onset is not exactly known. AF patients are more susceptible to the formation of blood clots in the atria, especially inside the auricles. Cardioversion may lead to the detachment of those clots, posing the patient at a high risk for stroke (Laupacis et al., 1998; Copley and Hill, 2016). AF patients between 80 and 89 years old without anticoagulation therapy demonstrate an increased risk of clot formation of 23.5% (Wolf et al., 1991). Strict adherence to the medication schedule is of utmost importance, since even a single missed dose could lead to a significant increase of a thrombotic complication (January et al., 2014).

If AF persisted for more than 7 days, the chances are very low that this abnormal heart rate converts spontaneously into a normal sinus rhythm (Danias et al., 1998). Cardioversion can be performed to restore the normal sinus rhythm by applying a synchronized current on the heart, which depolarizes the cardiac cells simultaneously (Copley and Hill, 2016). As mentioned before, patients who need to be treated with cardioversion, require an adequate anticoagulation therapy (Falk, 1992; January et al., 2014). Patients are usually treated with the combination of an anti-arrhythmic drug and electrical cardioversion to increase the rate of restoring and preserving the sinus rhythm (Oral et al., 1999). In this context, it is important to mention that the drug should be well selected and appropriate for the individual patient as some products can cause unwanted side effects such as low cardiac function (Stambler et al., 1996).

As mentioned, catheter ablation is now the most important therapy in the treatment of AF patients. In addition, a pacemaker may also be used in patients who cannot be treated by ablation, to normalize the cardiac rhythm (Falk, 2001). Mainly two energy sources can be used to perform a catheter ablation procedure. Radiofrequency energy will heat up the target site while cryoablation cools down the targeted spot. The latter can be applied by use of a focal catheter or balloon catheter. Literature indicates that the safety and efficacy of both techniques is comparable (Luik et

Chapter 1: General introduction al., 2015). Cryoballoon ablation requires a shorter fluoroscopy time and the reproducibility is also higher compared to the other catheters (Wasserlauf et al., 2015; Providencia et al., 2017). When comparing cryoballoon with focal cryoablation procedures, the former technique provides a higher PVI durability as the use of a balloon catheter can deliver contiguous lesions (Miyazaki et al., 2016).

Surgical interventions are performed to specifically treat AF patients who require also concomitant heart surgery for another indication or for those patients who did not respond sufficiently to medical treatment and catheter ablation (Xu et al., 2016). The surgical therapy has been adjusted and improved several times since the 1980s and till date the Cox-Maze III technique is still the golden standard to restore sinus rhythm in AF patients. The Cox-Maze III treatment includes the application of several incisions in the atria to induce scar formation. These scars will prevent the formation of macro-reentry circuits which play a major role in the preservation of AF. During such an intervention, the PVs and posterior left atrium are isolated (Prasad et al., 2003). However, this technique is technically demanding and thus, several variations were developed. Alternative energy sources, named Cox-Maze IV technique, such as radiofrequency, cryo and high intensity focused ultrasonography were tested (Ninet et al., 2005; Gillinov and Saltman, 2007). Moreover, a minimal invasive thoracoscopic epicardial approach and even a hybrid ablation technique were evaluated for AF treatment. The hybrid ablation technique deserves also some explanation as it may be an interesting approach to apply complete transmural lesions and superior isolation durability compared to catheter procedures (Bugge et al., 2005). An important drawback is the fact that this procedure is very invasive as in one and the same intervention or asynchronous both endocardial catheter and epicardial thoracoscopic ablation lesions are applied (La Meir, 2013).

1.3.1.4 Transmurality of ablation lesions

Transmural lesions, which are created when the ablation procedure affects the total thickness of the myocardial tissue layer in the PV wall, are required to achieve a permanent conduction block (Melby et al., 2008). Many factors influence this, among which the contact force between catheter and tissue is of major importance. Enforced catheter-tissue contact reduces the recovery of the PV conduction, improves lesion efficacy and enhances a single-procedure efficacy (Kumar et al., 2012; Kumar et al., 2013; Marijon et al., 2014). In the context of post-mortem lesion evaluation, literature provides only a limited number of methods to assess the lesion transmurality during interventions. Concerning myocardial infarction assessment, several techniques are described for ex-vivo visualization and delineation of cardiac lesions (Redfors et al., 2012). Regular histological staining methods distinguish fibrous scar tissue from the original intact tissue. Consequently, such techniques require a time gap of at least several days after lesion induction in order to replace the applied lesion

Chapter 1: General introduction by fibrous tissue. This time-wise disadvantage, especially in an acute in-vivo animal trial, may be overcome by use of an alternative such as a supravital staining with for example triphenyltetrazolium chloride (Brown, 1953). Instead of staining fibrosis, triphenyltetrazolium chloride will stain the intact viable tissue red, while the denatured, non-viable lesion remains unstained. Tissue samples may be collected within a shorter period after lesion induction (Redfors et al., 2012). Notwithstanding its common use to detect myocardial infarction lesions, it also has its limitations: Kakimoto et al. (2013) stipulated that this supravital staining product works well on fresh tissue samples of maximum 1.5 days old, but tissue stainability declines in older samples. More important is the fact that its stainability efficiency is disturbed in case the time gap between infarction and death is less than 9 hours (Kakimoto et al., 2013). A very short time interval between ablation and euthanasia of subjects in an animal study could cause difficulties to visualize the applied ablation lesions.

During an intervention, the presence of transmural lesions post-ablation can be deduced from the loss of bipolar capture during an electrophysiological study. In case of a transmural lesion, electrical impulses will fail to cross the line of ablation. This method is superior to the use of electrocardiography to detect ablation line gaps (Steven et al., 2010; Kosmidou et al., 2013; Kumar and Michaud, 2016). Magnetic resonance imaging is also promising as diagnostic tool to evaluate the transmurality of ablation lesions (Ranjan et al., 2011; Arujuna et al., 2012; Ranjan et al., 2012).

1.3.1.5 Pulmonary vein stenosis (PVS)

In a small number of patients, PVS is seen as a complication after an ablation procedure of the PVs, probably due to lesions in the muscle tissue of the wall of the pulmonary veins (Natale et al., 2007). The incidence of stenosis varies between 0% and 40% depending on the experience of the cardiologist and the applied ablation technique. In the past, the method of choice to treat PVS was PV balloon angioplasty. However, restenosis occurred in 44-70% of the cases (Neumann et al., 2009). Another treatment option of PVS is the placement of a stent inside the PV at the level of the stenosis. In a study of Neumann et al. (2005), no restenosis was seen during the 12-month follow-up after using this technique. However, after 4 years of follow-up 23% of the patients demonstrated in-stent restenosis (Neumann et al., 2009). This is in accordance with Holmes et al. (2009), who state that even after stent implantation restenosis can be seen in 30% to 50% of the patients. In order to tackle these complications, covered stents, for instance with a coating of polytetrafluoroethylene, were used (Tehrani and Lipkin, 2013). In addition, Tehrani and Lipkin (2013) mention that it is important that the stent covers the entire ostium to minimize the occurrence of stenosis. A study with 34 PVS patients (55 stenotic veins) showed that of the patients treated with a dilation method, 72% required

Chapter 1: General introduction a second treatment for restenosis, whereas this was only 33% in case a stent was implanted (Prieto et al., 2008).

1.3.1 AF in horses

In horses, AF is the most important performance-restricting arrhythmia and is frequently linked with clinical signs (Deem and Fregin, 1982; Young, 2003; McGurrin, 2015). Lone AF, which means that no other underlying cardiac disorder is present, is most commonly observed in horses (Else and Holmes, 1971; Morris and Fregin, 1982; Reef et al., 1988; Ohmura et al., 2003). On the other hand, AF can be found secondary to congestive heart failure, mitral regurgitation and atrial dilatation (Miller and Holmes, 1985; Reef et al., 1998a; Davis et al., 2002). Sometimes, the paroxysmal form is seen, which means that AF demonstrates an intermittent phase of about 48 hours after which the arrhythmia spontaneously converts to sinus rhythm without any intervention (Amada and Kurita, 1975; McGurrin, 2015). Generally, paroxysmal AF is not associated with any other cardiac disorder. AF in horses occurs mostly as the persistent form, with the irregular heart rhythm persisting until treatment. In accordance to men, the term permanent is used when the abnormal rhythm is not treatable (McGurrin, 2015).

1.3.2.1 Risk factors and effect on cardiac function

Horses have large atria which increases AF susceptibility. Further atrial enlargement may originate from or be the cause of this arrhythmia besides other influencing structural cardiac alterations (Decloedt et al., 2013). The body weight was also indicated as a significant risk factor as horses with a weight of more than 550 kg were prone to develop AF compared to horses with a lower weight (Leroux et al., 2013). This probably reflected the effect of horse size, and therefore also atrial size. Not only a large atrial myocardial surface but also a high vagal tone was referred to as an important predisposing factor for AF (Loomis and Krop, 1955; Reef et al., 1995). Moreover, pathologies such as myocarditis, mitral regurgitation, disturbance of the electrolytes or the acid-base balance or even an autonomic nervous system disorder could induce AF in horses. The application of tranquilizers or anesthetics or other unknown factors may be the cause of AF (Reef and McGuirk, 2002). In certain racehorse breeds, AF was labeled as a genetic disorder (Reef et al., 2014). Some authors report increased atrial fibrosis in horses with AF, which may lead to atrial depolarisations and increased chance of AF initiation and maintenance (Allessie et al., 2001). On the other hand, atrial fibrosis is also found in horses with a normal heart rhythm (Else and Holmes, 1971).

Another important risk factor is the presence of tricuspid regurgitation. Concerning the different breeds, trotters and warmbloods were found to be more susceptible for AF (Deem and Fregin, 1982;

Chapter 1: General introduction

Reef et al., 1988; Davis et al., 2002; Leroux et al., 2013). However, it was indicated that the respective weight may play a major role in this seeming breed disposition (Leroux et al., 2013). In addition, strenuous exercise may trigger initiation of AF by inducing atrial myocardial stretch which can lead to atrial premature beats and electrophysiological adaptations (Decloedt et al., 2015).

AF, characterized by a chaotic irregular contraction of the atria, induces atrial contraction loss due to electrical chaos in the atria, but also due to altered calcium handling. All this leads to reduced ventricular filling and ends with a decrease in stroke volume (van Loon, 2001a), which may lead to reduced performance (Deem and Fregin, 1982; Verheyen et al., 2013). The adverse effects of cardiac function decrease are mostly seen during exercise as, at that moment, the atria are very important to get a sufficient ventricular filling. Due to the irregular atrial contractions affecting the ventricles, the ventricular filling is irregular which further disturb cardiac performance (Fregin, 1971; Verheyen et al., 2013; McGurrin, 2015). Symptoms become mostly visible during exercise, since then, the increased sympathetic tone causes tachycardia through atrioventricular conduction (van Loon, 2001a).

1.3.2.2 Prevalence and treatment

In horses, an AF prevalence of around 2.5%, ranging between 0.23 and 5.3%, depending on the population, has been reported (Holmes et al., 1969; Else and Holmes, 1971; Leroux et al., 2013). AF is noticed in horses of all ages, even found in neonatal foals (Yamamoto et al., 1992). Several authors point out that male horses would be more affected (Holmes et al., 1969; Else and Holmes, 1971).

First of all, AF is not a life threatening disease in horses at rest. If no decrease in performance is noticed, no underlying cardiac disease is present and the heart rate and rhythm are normal during normal activity, treatment may be unnecessary (Young and van Loon, 2014). However, as AF may pose safety problems, it was stated that in case a horse is diagnosed with persistent AF, it should be treated with cardioversion or be excluded from any competition. Especially, if the heart rate during maximal exercise exceeds 220 beats/min or if ventricular arrhythmias are noticed during exercise or after stimulation of the sympathetic nervous system (Reef et al., 2014). Moreover, persistent AF horses should only be ridden or driven by an adult who is sufficiently informed. The level of activity should be limited to an exercise level which is considered relatively safe based on an exercise electrocardiogram (Reef et al., 2014).

In sport horses without any underlying heart disease, therapy intends to restore the normal sinus rhythm (Young and van Loon, 2014). The treatment of horses which demonstrate AF for more than three months might be more difficult, due to electrophysiological, structural and contractile

Chapter 1: General introduction remodeling that stabilizes AF (Wijffels et al., 1995; Reef et al., 1988; van Loon et al., 2000; van Loon, 2001b; De Clercq et al., 2008).

A standard treatment of AF in horses is the administration of antiarrhythmic drugs such as quinidine sulfate. However, important cardiac and non-cardiac side-effects of quinidine sulfate are noticed (Reef et al., 1988; Reef et al., 1995; McGurrin et al., 2003; De Clercq et al., 2014). Amiodarone was also used but demonstrated a moderate success. Flecanaide showed a low success rate with major side effects, such as acute mortality due to severe ventricular arrhythmia. These results imply the need for other treatment options to restore sinus rhythm (van Loon et al., 2004; De Clercq et al., 2006; De Clercq et al., 2007; De Clercq et al., 2008; Dembeck et al., 2014). A possible alternative to achieve sinus rhythm is transvenous electrical cardioversion in which a direct current of electrical energy is applied to the atria to induce a depolarization of the myocardium, so to terminate the reentry in the atria (McGurrin, 2015). In horses, this procedure is performed with a success rate of 98% (McGurrin et al., 2005a; McGurrin et al., 2005b; Reef et al., 2014; McGurrin, 2015). The overall recurrence of AF, independent of treatment technique, is about 35-40% (Decloedt et al., 2015). In horses with a first AF episode, the factors associated with AF recurrence were a previous unsuccessful treatment and mild to moderate mitral regurgitation. In horses with a successful first AF episode treatment, the recurrence rate after cardioversion is as high as 39% after one year (Decloedt et al., 2015). It is important to mention that these techniques restore the sinus rhythm and have a good prognosis in horses with no other important cardiac disease (Deem and Fregin, 1982; Morris and Fregin, 1982; Bentz et al., 2002), but they do not solve the actual causing trigger of the arrhythmia. This emphasizes the need for further development of new AF treatment procedures.

1.4 PVs in horses

Detailed anatomical and histological information about the equine PVs is lacking. In regard to the AF mechanism, important data is still missing and frequently deduced from human literature. In analogy with humans, myocardial sleeve tissue is also expected in the equine PVs. However, this has never been studied in detail. It is suspected that these myocardial sleeves also play a role in the induction of AF in horses. In vivo research of equine PVs and their role in AF is hampered by the lack of imaging techniques since CT and MRI cannot be used. The only useful imaging technique is echocardiography but no studies on PV in horses have been done so far.

1.4.1 Equine echocardiography

In equine cardiology, ultrasound has become a standard diagnostic technique. The different forms of ultrasound as two-dimensional, M-mode and Doppler echocardiography enable to study cardiac

Chapter 1: General introduction structures, blood flow characteristics, heart chamber dimensions and cardiac function (Reef, 1998a; Reef, 1998b; Bonagura et al., 2010; Boon, 2011; Schwarzwald, 2014). However, the structures visible on ultrasound are limited because the acoustic window, where the heart makes directly contact with the chest wall, is relatively small, due to the air in the lungs and the sternum composed of bone (Patteson, 1999; Marr and Patteson, 2010). Equine echocardiography is therefore restricted to parasternal or paracostal views (Patteson, 1999).

Nowadays, echocardiography is indispensable for cardiovascular disease detection and should always be performed in horses with poor performance (Reef, 1998b). This technique enables the assessment and diagnosis of arrhythmias and cardiac murmurs in horses demonstrating a poor exercise performance (Bonagura and Blissitt, 1995). Many intra-cardiac structures are clearly recognized on ultrasound, with the exception of a few small structures. In literature, PV visualization has been mentioned by Reef (1998b) and Schwarzwald (2014), but not described in detail. Reef (1991, 1998b) stipulated that, besides other cardiac structures, the PVs are visualized echocardiographically by use of a segmental approach to cardiac anatomy. Further, it is described that two PVs are visible on ultrasound, entering the left atrium, caudal to and to the left side of the right atrium (Reef, 1998b).

1.5 Pigs as cardiovascular model

The pig is often used in research and due to the physiological, biochemical and anatomical resemblance with humans, pigs are more and more used as an animal model (Ibrahim et al., 2006; Swindle, 2007; Bode et al., 2010; Helke and Swindle, 2013). Although the porcine heart is not identical to that of men, it demonstrates a high level of similarity (Crick et al., 1998). For example, in the context of coronary dominance, pigs and humans demonstrate respectively in 80% and 90% of the cases right coronary dominance (Anderson and Becker, 1992; Crick et al., 1998). Moreover, in pigs and humans several major factors in cardiac function, namely stroke volume and cardiac output, but also heart rate and myocardial blood flow are indicated to be comparable (Thein and Hammer, 2004). In terms of cardiac growth, in men and pigs, heart size corresponds well with the height or length rather than with the age or weight (Nidorf et al., 1992). Although, arrhythmias are of little interest in pigs themselves, the differences between pigs and men were investigated, regarding the myocardial action potentials and the overall conduction system. Morphological differences between pigs and humans were noticed at the level of the Purkinje system and atrioventricular node (Verdouw and Hartog, 1986; Gardner and Johnson, 1988; Crick et al., 1999).

Several pig breeds are used in research but, according to Swindle (2007), Duroc, Yorkshire, Landrace and crossbred breeds are frequently used. Miniature breeds are also suitable of which the Hanford,

Chapter 1: General introduction

Göttingen and Yucatan miniature pig are very popular (McAnulty et al., 2012). Concerning malignant hyperthermia or porcine stress syndrome, it is indicated that Göttingen minipigs are easily anaesthetized for various hours without a major risk of complications due to the absence of the halothane gene (Alstrup, 2010).

Besides pigs, dogs are also often used as animal models, especially in biomedical research. However, various organs demonstrate a dimension and weight difference compared to humans. In addition, numerous breed differences need to be considered (Kararli, 1995; Bailey et al., 2013). And for that reason, dogs are not always the ideal model in surgical or interventional trials. Also, when including these animals in experimental procedures, ethical concerns may be raised.

To summarize, in the context of AF, substantial research was already performed in men. However, sufficient data from animal model studies is lacking which impairs the development of a new ablation treatment for AF in humans. Since in these treatments, the pulmonary veins are in scope, a detailed examination of these structures is a prerequisite, both in pigs as an animal model and in horses as target species.

Chapter 1: General introduction

References

Aguilar M. and Nattel S. (2015). The pioneering work of George Mines on cardiac arrhythmias: groundbreaking ideas that remain influential in contemporary cardiac electrophysiology. J Physiol.; doi:10.1113/JP270506.

Allessie M.A., Boyden P.A., Camm A.J., Kléber A.G., Lab M.J., Legato M.J., Rosen M.R., Schwartz P.J., Spooner P.M., Van Wagoner D.R. and Waldo A.L. (2001). Pathophysiology and prevention of atrial fibrillation. Circulation. 103:769–777.

Almeida O.P., Bohm C.M., de Paula Carvalho M. and Paes de Carvalho A. (1975). The cardiac muscle in the pulmonary vein of the rat: a morphological and electrophysiological study. J Morphol. 145:409–433.

Alstrup A.K.O. (2010). Anaesthesia and analgesia in Ellegaard Göttingen minipigs. PET centre, Aarhus University Hospital, Denmark.

Amada A. and Kurita H. (1975). Five cases of paroxysmal atrial fibrillation in the racehorse. Exp Rep Equine Health Lab. 12:89–100.

Anderson R.H. and Becker A.E. (1992). The heart: structure in health and disease. London: Gower Medical.

Andrade J., Khairy P., Dobrev D. and Nattel S. (2014). The clinical profile and pathophysiology of atrial fibrillation: relationships among clinical features, epidemiology, and mechanisms. Circ Res. 114:1453–1468.

Anselmino M., Blandino A., Beninati S., Rovera C., Boffano C., Beletti M., Caponi D., Scaglione M., Cesarani F. and Gaita F. (2011). Morphologic analysis of left atrial anatomy by magnetic resonance angiography in patients with atrial fibrillation: a large single center experience. J Cardiovasc Electrophysiol. 22:1–7.

Antzelevitch C. and Burashnikov A. (2011). Overview of Basic Mechanisms of Cardiac Arrhythmia. Card Electrophysiol Clin. 3:23–45.

Ariyarajah V. and Spodick D.H. (2006). The Bachmann Bundle and interatrial conduction. Cardiol Rev. 14:194-199.

Chapter 1: General introduction

Armour J.A., Murphy D.A., Yuan B.-X., Macdonald S. and Hopkins D.A. (1997). Gross and microscopic anatomy of the human intrinsic cardiac nervous system. Anat Rec. 247:289–298.

Arnar D.O., Thorvaldsson S., Manolio T.A., Thorgeirsson G., Kristjansson K., Hakonarson H. and Stefansson K. (2006) Familial aggregation of atrial fibrillation in Iceland. Eur Heart J. 27:708–712.

Arora R. (2012) Recent insights into the role of the autonomic nervous system in the creation of substrate for atrial fibrillation: implications for therapies targeting the atrial autonomic nervous system. Circ Arrhythm Electrophysiol. 5:850–9.

Arujuna A., Karim R., Caulfield D., Knowles B., Rhode K., Schaeffter T., Kato B., Rinaldi C.A., Cooklin M., Razavi R., O’Neill M.D. and Gill J. (2012). Acute pulmonary vein isolation is achieved by a combination of reversible and irreversible atrial injury after catheter ablation: evidence from magnetic resonance imaging. Circ Arrhythm Electrophysiol. 5:691–700.

Atienza F., Martins R.P. and Jalife J. (2012). Translational research in atrial fibrillation: a quest for mechanistically based diagnosis and therapy. Circ Arrhythm Electrophysiol. 5:1207–1215.

Ausma J., Wijffels M., Thoné F., Wouters L., Allessie M. and Borgers M. (1997). Structural changes of atrial myocardium due to sustained atrial fibrillation in the goat. Circulation. 96:3157–3163.

Ausma J., van der Velden H.M., Lenders M.H., van Ankeren E.P., Jongsma H.J., Ramaekers F.C., Borgers M., Allessie M.A. (2003). Reverse structural and gap-junctional remodeling after prolonged atrial fibrillation in the goat. Circulation. 107:2051–2058.

Barone R. (1976). Poumons. In: Anatomie compare des mammifères domestiques, tome 3, splanchnologie, appareil digestif - appareil respiratoire. Tixier & Fils, Lyon, pp.757-799.

Barone R. (1996). Coeur. In: Anatomie compare des mammifères domestiques, tome 5, angiologie. Éditions Vigot, Paris, pp.5–102.

Barone R. (2011). Vaisseaux et nerfs. In: Angiologie (Tome 5), Anatomie compare des mammifères domestiques. Éditions Vigot, Paris, pp.55–70.

Benjamin E.J., Levy D., Vazin S.M., D’Agostino R.B., Belanger A.J. and Wolf P.A. (1994). Independent risk factors for atrial fibrillation in a population-based cohort study. The Framingham Heart Study. JAMA. 271:840–844.

Bentz B.G., Erkert R.S. and Blaik M.A. (2002). Atrial fibrillation in horses: Treatment and prognosis. Compendium on Continuing Education for the Practicing Veterinarian 24:817–822.

Chapter 1: General introduction

Bielecka-Dabrowa A., Mikhailidis D.P., Rysz J. and Banach M. (2009). The mechanisms of atrial fibrillation in hyperthyroidism. Thyroid Res. 2:4.

Blom N.A., Gittenberger-de Groot A.C., DeRuiter M.C., Poelmann R.E., Mentink M.M. and Ottenkamp J. (1999). Development of the cardiac conduction tissue in human embryos using HNK-1 antigen expression: possible relevance for understanding of abnormal atrial automaticity. Circulation. 99:800–806.

Bode G., Clausing P., Gervais F., Loegsted J., Luft J., Nogues V. and Sims J. (2010). The utility of the minipig as an animal model in regulatory toxicology. J Pharmacol Toxicol. 62:196–220.

Bonagura J.D. and Blissitt K.J. (1995). Echocardiography. Equine Vet J Suppl. 19:5–17.

Bonagura J.D., Reef V.B. and Schwarzwald C.C. (2010). Cardiovascular disease. In: Equine Internal Medicine, 3rd edn (eds SM Reed, WM Bayly, DC Sellon). Saunders Elsevier, St. Louis, pp.372–487.

Boon JA (2011). Veterinary Echocardiography. Wiley Blackwell, Oxford.

Boriani G., Diemberger I., Martignani C., Biffi M. and Branzi A. (2006). The epidemiological burden of atrial fibrillation: a challenge for clinicians and healthcare systems. Eur Heart J. 27:893–894.

Boriani G., Cervi E., Diemberger I., Martignani C. and Biffi M. (2011). Clinical management of atrial fibrillation: need for a comprehensive patient-centered approach. Chest. 140:843–845.

Boriani G. and Pettorelli D. (2016). Atrial fibrillation burden and atrial fibrillation type: Clinical significance and impact on the risk of stroke and decision making for long-term anticoagulation. Vasc Pharmacol. 83:26–35.

Bright J.M. and Marr C.M. (2010). Introduction to cardiac anatomy and physiology. IN: Cardiology of the Horse, second edition, (eds: Marr C.M. and Bowen M.), Saunders Elsevier, Amsterdam, pp.3–21.

Brown FA. Jr. (1953). The use of triphenyl tetrazolium chloride as a supravital stain in chemotherapeutic trials with Ehrlich ascites tumor. Stanford Med. Bull. 11:139–141.

Brunton T.L. and Fayer J. (1876). Note on independent pulsation of the pulmonary veins and vena cava. Proc R Soc Lond. 25:174–176.

Budras K.-D., Sack W.O., Röck S., Horowitz A. and Berg R. (2009) (eds.). Thoracic cavity: heart and thymus. In: Anatomy of the horse. Schlütersche Verlagsgesellschaft mbH & Co. KG., Hannover, pp.59–63.

Chapter 1: General introduction

Burch G.E. and Romney R.B. (1954). Functional anatomy and ‘throttle valve’ action of the pulmonary veins. Am Heart J. 47:58–66.

Burstein B., Qi X.Y., Yeh Y.H., Calderone A. and Nattel S. (2007). Atrial cardiomyocyte tachycardia alters cardiac fibroblast function: a novel consideration in atrial remodeling. Cardiovasc Res. 76:442– 452.

Burstein B. and Nattel S. (2008). Atrial fibrosis: mechanisms and clinical relevance in atrial fibrillation. J Am Coll Cardiol. 51:802–809.

Calkins H., Kuck K.H., Cappato R., Brugada J., Camm A.J., Chen S.A., Crijns H.J., Damiano R.J. Jr., Davies D.W., DiMarco J., Edgerton J., Ellenbogen K., Ezekowitz M.D., Haines D.E., Haïssaguerre M., Hindricks G., Iesaka Y., Jackman W., Jalife J., Jais P., Kalman J., Keane D., Kim Y.H., Kirchhof P., Klein G., Kottkamp H., Kumagai K., Lindsay B.D., Mansour M., Marchlinski F.E., McCarthy P.M., Mont J.L., Morady F., Nademanee K., Nakagawa H., Natale A., Nattel S., Packer D.L., Pappone C., Prystowsky E., Raviele A., Reddy V., Ruskin J.N., Shemin R.J., Tsao H.M. and Wilber D. (2012). HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for patient selection, procedural techniques, patient management and follow-up, definitions, endpoints, and research trial design: a report of the Heart Rhythm Society (HRS) Task Force on Catheter and Surgical Ablation of Atrial Fibrillation. Developed in partnership with the European Heart Rhythm Association (EHRA), a registered branch of the European Society of Cardiology (ESC) and the European Cardiac Arrhythmia Society (ECAS); and in collaboration with the American College of Cardiology (ACC), American Heart Association (AHA), the Asia Pacific Heart Rhythm Society (APHRS), and the Society of Thoracic Surgeons (STS). Heart Rhythm. 9:632–696.

Camm A.J., Kirchof G.Y., Lip U., Schotten I., Savelieva S., Ernst S., Van Gelder I.C., Al-Attar N., Hindricks G., Prendergast B., Heidbuchel H., Alfieri O., Angelini A., Atar D., Colonna P., De Caterina R., De Sutter J., Goette A., Gorenek B., Heldal M., Hohloser S.H., Kolh P., Le Heuzey J.Y., Ponikowski P. and Rutten F.H. (2010). Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Europace. 12:1360– 1420.

Cao X., Aimoto M., Fukumoto M., Nagasawa Y., Tanaka H. and Takahara A. (2018). Influence of chronic volume overload-induced atrial remodeling on electrophysiological responses to cholinergic receptor stimulation in the isolated rat atria. J Pharmacol Sci. 136:73-78.

Chapter 1: General introduction

Cappato R., Calkins H., Chen S.-H., Davies W., Iesaka Y., Kalman J., Kim Y.-Ho, Klein G., Packer D. and Skanes A. (2005). Worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibrillation. Circulation. 111:1100–1105.

Cheema A., Dong J.,Dalal D., Marine J.E., Henrikson C.A., Spragg D., Cheng A., Nazarian S., Bilchick K., Sinha S., Scherr D., Almasry I., Halperin H., Berger R. and Calkins H. (2007). Incidence and time course or early recovery of pulmonary vein conduction after catheter ablation of atrial fibrillation. J Cardiovasc Electrophysiol. 18:387-391.

Chen S.A., Hsieh M.H., Tai C.T., Tsai C.F., Prakash V.S., Yu W.C., Hsu T.L., Ding Y.A. and Chang M.S. (1999). Initiation of atrial fibrillation by ectopic beats originating from the pulmonary veins: electrophysiological characteristics, pharmacological responses, and effects of radiofrequency ablation. Circulation. 100:1879–86.

Chen Y.J., Chen Y.C., Yeh H.I., Lin C.I. and Chen S.A. (2002). Electrophysiology and arrhythmogenic activity of single cardiomyocytes from canine superior vena cava. Circulation. 105:2679–85.

Chen YH, Xu SJ, Bendahhou S, Wang X.L., Wang Y., Xu W.Y., Jin H.W., Sun H., Su X.Y., Zhuang Q.N., Yang Y.Q., Li Y.B., Liu Y., Xu H.J., Li X.F., Ma N., Mou C.P., Chen Z., Barhanin J. and Huang W. (2003). KCNQ1 gain-of-function mutation in familial atrial fibrillation. Science. 299:251–4.

Choi E.K., Shen M.J., Han S., Kim D., Hwang S., Sayfo S., Piccirillo G., Frick K., Fishbein M.C., Hwang C., Lin S.-F. and Chen P.-S. (2010). Intrinsic cardiac nerve activity and paroxysmal atrial tachyarrhythmia in ambulatory dogs. Circulation. 121:2615–2623.

Chou C.C. and Chen P.S. (2009). New concepts in atrial fibrillation: neural mechanisms and calcium dynamics. Cardiol Clin. 27:35–43.

Christophersen I.E., Ravn L.S., Budtz-Joergensen E., Skytthe A., Haunsoe S., Svendsen J.H. and Christensen K. (2009). Familial aggregation of atrial fibrillation: a study in Danish twins. Circ Arrhythm Electrophysiol. 2:378–383.

Colilla S., Crow A., Petkun W., Singer D.E., Simon T. and Liu X. (2013). Estimates of current and future incidence and prevalence of atrial fibrillation in the U.S. adult population. Am J Cardiol. 112:1142– 1147.

Comtois P., Kneller J. and Nattel S. (2005). Of circles and spirals: bridging the gap between the leading circle and spiral wave concepts of cardiac reentry. Europace. 7:10–20.

Chapter 1: General introduction

Constantinescu G.M. and Schaller O. (eds.) (2012). Angiologia. In: Illustrated Veterinary Anatomical Nomenclature, Enke, Stuttgart, pp.234–413.

Copley D.J. and Hill K.M. (2016). Atrial Fibrillation: A Review of Treatments and Current Guidelines. AACN Adv Crit Care. 27:120–128.

Corradi D., Callegari S., Maestri R., Benussi S. and Alfieri O. (2008). Structural remodeling in atrial fibrillation. Nat Clin Pract Cardiovasc Med. 5:782–796.

Corradi D.,Maestri R., Macchi E. and Callegari S. (2011). The atria: from morphology to function. J Cardiovasc Electrophysiol. 22:223–235.

Corradi D., Callegari S., Gelsomino S., Lorusso R. and Macchi E. (2013). Morphology and pathophysiology of target anatomical sites for ablation procedures in patients with atrial fibrillation. Part II: Pulmonary veins, caval veins, ganglionated plexi, and ligament of Marshall. Int J of Cardiol. 168:1769–1778.

Cretoiu S. M. and Popescu L. M. (2014). Telocytes revisited. BioMol Concepts. 5:353–369.

Crick S.J., Sheppard M.N., Ho S.Y., Gebstein L. and Anderson R.H. (1998). Anatomy of the pig heart: comparisons with normal human cardiac structure. J Anat. 193:105.

Crick S.J., Sheppard M.N., Ho S.Y. and Anderson R.H. (1999). Localisation and quantitation of autonomic innervation in the porcine heart I: conduction system. J Anat. 195:341.

Darby A.E. (2016). Recurrent atrial fibrillation after catheter ablation: considerations for repeat ablation and strategies to optimize success. J Atr Fibrillation. 9: 1427.

Davis J. L., Gardner S. Y., Schwabenton B. and Breuhaus B. A. (2002). Congestive heart failure in horses: 14 cases (1984-2001). J Am Vet Med Assoc. 220:1512–1515.

De Clercq D., van Loon G., Baert K., Tavernier R., Croubels S., De Backer P. and Deprez P. (2006). Intravenous amiodarone treatment in horses with chronic atrial fibrillation. Vet J. 172:129–134.

De Clercq D., van Loon G., Baert K., Tavernier R., Croubels S., De Backer P. and Deprez P. (2007). Effects of an adapted intravenous amiodarone protocol in horses with atrial fibrillation. Equine Vet J. 39:344–349.

De Clercq D., van Loon G., Schauvliege S., Tavernier R., Baert K., Croubels S., De Backer P. and Deprez P. (2008). Transvenous electrical cardioversion of atrial fibrillation in six horses using custom made cardioversion catheters. Vet J. 177:198–204.

Chapter 1: General introduction

De Clercq D., Decloedt A., Sys S. U., Verheyen T., Van Der Vekens N. and van Loon G. (2014). Atrial fibrillation cycle length and atrial size in horses with and without recurrence of atrial fibrillation after electrical cardioversion. J Vet Intern Med. 28:624–629.

Decloedt A., Verheyen T., Van Der Vekens N., Sys S., De Clercq D. and van Loon G. (2013). Long-term follow-up of atrial function after cardioversion of atrial fibrillation in horses. Vet J. 197:583–588.

Decloedt A., Schwarzwald C. C., De Clercq D., Van Der Vekens N., Pardon B., Reef V. B. and van Loon G. (2015). Risk factors for recurrence of atrial fibrillation in horses after cardioversion to sinus rhythm. J Vet Intern Med. 29:946–953.

Deem D.A. and Fregin G.F. (1982) Atrial fibrillation in horses–a review of 106 clinical cases, with consideration of prevalence, clinical signs, and prognosis. J Am Vet Med Assoc. 180: 261–265.

De Groot N.M., Houben R.P., Smeets J.L., Boersma E., Schotten U., Schalij M.J., Crijns H. and Allessie M.A. (2010). Electropathological substrate of longstanding persistent atrial fibrillation in patients with structural heart disease: epicardial breakthrough. Circulation. 122:1674–1682.

Dembeck K.A., Hurcombe S.D.A., Schober K.E. and Toribio R.E. (2014). Sudden death of a horse with supraventricular tachycardia following oral administration of flecanaide acetate. J Vet Emerg Crit Car. 24:759–763.

De Ponti R., Ho S.Y., Salerno-Uriarte J.A., Tritto M. and Spadacini G. (2002). Electroanatomic analysis of sinus impulse propagation in normal human atria. J Cardiovasc Electrophysiol. 13:1–10.

Eckstein J., Verheule S., de Groot N.M., Allessie M. and Schotten U. (2008). Mechanisms of perpetuation of atrial fibrillation in chronically dilated atria. Prog Biophys Mol Biol. 97:435–451.

Edwards W.D. (1987). Applied anatomy of the heart. In: Brandenburg R.O., Fuster V., Giuliani E.R., McGoon D.C.(eds). Cardiology: Fundamentals and Practice, Book Medical Publishers, Chicago, pp.58.

Ehrlich J.R., Cha T.J., Zhang L., Chartier D., Melnyk P., Hohnloser S.H. and Nattel S. (2003). Cellular electrophysiology of canine pulmonary vein cardiomyocytes: action potential and ionic current properties. J Physiol. 551:801–813.

Ellinor P.T., Moore R.K., Patton K.K., Ruskin J.N., Pollak M.R. and Macrae C.A. (2004). Mutations in the long QT gene, KCNQ1, are an uncommon cause of atrial fibrillation. Heart. 90:1487–1488.

Else R.W. and Holmes J.R. (1971). Pathological changes in atrial fibrillation in the horse. Equine Vet J. 3:56–64.

Chapter 1: General introduction

Ernst S., Sanchez-Quintana D. and Ho S.Y. (2009). Anatomy of the pericardial space and mediastinum: relevance to epicardial mapping and ablation. In: Shivkumar K., Boyle N.G., (eds). Cardiac Electrophysiology Clinics: Epicardial Interventions in Electrophysiology. PA: Saunders, Philadelphia, pp.1–8.

Falk R.H. (2001). Atrial fibrillation. N Engl J Med. 344:1067–1078.

Fang M.C., Chen J. and Rich M.W. (2007). Atrial fibrillation in the elderly. Am J Med. 120: 481-487.

Fox C.S., Parise H., D’Agostino R.B. Sr., Lloyd-Jones D.M., Vasan R.S., Wang T.J., Levy D., Wolf P.A. and Benjamin E.J. (2004). Parental atrial fibrillation as a risk factor for atrial fibrillation in offspring. JAMA. 291:2851–2855.

Fregin G.F. (1971). Atrial fibrillation in the horse. Proc Am Ass Equine Practnrs. 16:383-388.

Fuster V., Ryden L.E., Cannom D.S., Crijns H.J., Curtis A.B., Ellenbogen K.A., Halperin J.L., Kay G.N., Le Huezey J.Y., Lowe J.E., Olsson S.B., Prystowsky E.N., Tamargo J.L. and Wann L.S. (2011). ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in partnership with the European Society of Cardiology and in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. J Am Coll Cardiol. 57:e101-98. doi: 10.1016/j.jacc.2010.09.013.

Ganesan A.N., Shipp N.J., Brooks A.G., Kuklik P., Lau D.H., Lim H.S., Sullivan T., Roberts-Thomson K.C. and Sanders P. (2013). Long-term outcomes of catheter ablation of atrial fibrillation: a systematic review and meta-analysis. J am heart Assoc. 2:e004549. doi: 10.1161/JAHA.112.004549.

Gardner T.J. and Johnson D.L. (1988). Cardiovascular system. In: Swindle M.M. and Adams R.J., (eds). Experimental Surgery and Physiology: Induced Animal Models of Human Disease. Williams and Wilkins, Baltimore, 74.

Gherghiceanu M., Hinescu M.E., Andrei F., Mandache E., Macarie C.E., Faussone-Pellegrini M.S. and Popescu L.M. (2008). Interstitial Cajal-like cells (ICLC) in myocardial sleeves of human pulmonary veins. J Cell Mol Med. 12:1777–1781.

Gille U. and Salomon F.-V. (2015). Herz-, Kreislauf- und Abwehrsystem, angiologia. In: Salomon F.-V., Geyer H. and Gille U., (eds). Anatomie für die Tiermedizin. Enke, Stuttgart, pp. 417-476.

Chapter 1: General introduction

Go A.S., Hylek E.M., Phillips K.A., Chang Y., Henault L.E., Selby J.V. and Singer D.E. (2001). Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA. 285:2370- 2375.

Go A.S., Mozaffarian D., Roger V.L., Benjamin E.J., Berry J.D., Blaha M.J., Dai S., Ford E.S., Fox C.S., Franco S., Fullerton H.J., Gillespie C., Hailpern S.M., Heit J.A., Howard V.J., Huffman M.D., Judd S.E., Kissela B.M., Kittner S.J., Lackland D.T., Lichtman J.H., Lisabeth L.D., Mackey R.H., Magid D.J., Marcus G.M., Marelli A., Matchar D.B., McGuire D.K., Mohler E.R., Moy C.S., Mussolino M.E., Neumar R.W., Nichol G., Pandey D.K., Paynter N.P., Reeves M.J., Sorlie P.D., Stein J., Towfighi A., Turan T.N., Virani S.S., Wong N.D., Woo D. and Turner M.B. (2014). Heart disease and stroke statistics–2014 update: a report from the American Heart Association. Circulation. 129:e28–e292.

Goette A., Honeycutt C. and Langberg J.J. (1996). Electrical remodeling in atrial fibrillation. Time Course and Mechanisms. Circulation. 94:2968–2974.

Gordan R., Gwathmey J.K. and Xie L.-H. (2015). Autonomic and endocrine control of cardiovascular function. World J Cardiol. 7:204–214.

Gould P.A., Yii M., McLean C., Finch S., Marshall T., Lambert G.W. and Kaye D.M. (2006). Evidence for increased atrial sympathetic innervation in persistent human atrial fibrillation. Pacing Clin Electrophysiol. 29:821–829.

Grandi E., Pandit S.V., Voigt N., Workman A.J., Dobrev D., Jalife J. and Bers D.M. (2011). Human atrial action potential and Ca2+ model: sinus rhythm and chronic atrial fibrillation. Circ Res. 109:1055– 1066.

Guerra P.G., Thibault B., Dubuc M., Talajic M., Crépeau J., Nattel S. and Tardif J.C. (2003). Identification of atrial tissue in pulmonary veins using intravascular ultrasound. J Am Soc Echocardiogr. 16:982–987.

Haïssaguerre M., Jaïs P., Shah D.C., Takahashi A., Hocini M., Quiniou G., Garrigue S., Le Mouroux A., Le Métayer P. and Clémenty J. (1998). Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med. 339:659–666.

Haïssaguerre M., Sanders P., Hocini M., Hsu L.-F., Shah D.C., Scavée C., Takahashi Y., Rotter M., Pasquié J.-L., Garrigue S., Clémenty J. and Jaïs P. (2004a). Changes in atrial fibrillation cycle length and inducibility during catheter ablation and their relation to outcome. Circulation. 109:3007–3013.

Chapter 1: General introduction

Hamabe A., Okuyama Y., Miyauchi Y., Zhou S., Pak H.-N., Karagueuzian H.S., Fishbein M.C. and Chen P.-S. (2003). Correlation between anatomy and electrical activation in canine pulmonary veins. Circulation. 107:1550–1555.

Harhun M.I., Gordienko D.V., Povstyan O.V., Moss R.F. and Bolton T.B. (2004). Function of interstitial cells of Cajal in the rabbit portal vein. Circ Res. 95:619–626.

Hart R.G., Pearce L.A., Albers G.W., Connolly S.J., Friday G.H., Gage B.F., Go A.S., Goldstein L.B., Gronseth G., Halperin J.L., Hylek E.M., Lip G.Y.H., Sherman D.G., Singer D.E. and van Walraven C. (2007). Independent predictors of stroke in patients with atrial fibrillation. Neurology. 69:546–554.

Hassink R.J., Aretz H.T., Ruskin J. and Keane D. (2003). Morphology of atrial myocardium in human pulmonary veins: A postmortem analysis in patients with and without atrial fibrillation. J Am Coll Cardiol. 42:1108-1114.

He B., Scherlag B.J., Nakagawa H., Lazzara R. and Po S.S. (2012). The intrinsic autonomic nervous system in atrial fibrillation: a review. ISRN Cardiol. 2012:490674. doi.org/10.5402/2012/490674.

Heijman J., Voigt N., Nattel S. and Dobrev D. (2014). Cellular and molecular electrophysiology of atrial fibrillation initiation, maintenance, and progression. Circ Res. 114:1483–1499.

Heijman J., Algalarrondo V., Voigt N., Melka J., Wehrens X. H. T., Dobrev D. and Nattel S. (2016). The value of basic research insights into atrial fibrillation mechanisms as a guide to therapeutic innovation: a critical analysis. Cardiovasc Res. 109:467–479.

Helke K.L. and Swindle MM. (2013). Animal models of toxicology testing: the role of pigs. Expert Opin Drug Metab Toxicol. 9:127–139.

Hildreth V., Anderson R.H. and Henderson D.J. (2009). Autonomic innervation of the developing heart: Origins and function. Clin Anat. 22:36–46.

Ho SY, Cabrera JA, Tran VH, Farre J, Anderson RH and Sanchez-Quintana D. (2001). Architecture of the pulmonary veins: relevance to radiofrequency ablation. Heart. 86:265–270.

Ho S.Y., Cabrera J.A. and Sanchez-Quintana D. (2012). Left atrial anatomy revisited. Circ Arrhythm Electrophysiol. 5:220–228.

Hocini M., Ho S. Y., Kawara T., Linnenbank A. C., Potse M., Shah D., Jais P., Janse M. J., Haïssaguerre M. and de Bakker J. M. T. (2002). Electrical conduction in canine pulmonary veins - Electrophysiological and anatomic correlation. Circulation. 105:2442–2448.

Chapter 1: General introduction

Holmes J.R., Darke P.G. and Else R.W. (1969). Atrial fibrillation in the horse. Equine Vet J. 1:211-222.

Holmes D.R.Jr., Monahan K.H. and Packer D. (2009). Pulmonary vein stenosis complicating ablation for atrial fibrillation: Clinical spectrum and interventional considerations. JACC Cardiovasc Interv. 2:267–276.

Ibrahim Z., Busch J., Awwad M., Wagner R., Wells K. and Cooper D.K.C. (2006). Selected physiologic compatibilities and incompatibilities between human and porcine organ systems. Xenotransplantation. 13:488-499.

Iwasaki Y.K., Nishida K., Kato T. and Nattel S. (2011). Atrial fibrillation pathophysiology: implications for management. Circulation. 124:2264-2274.

Jais P., Hocini M., Sanders P., Hsu L.F., Takahashi Y., Rotter M., Rostock T., Sacher F., Clementy J. and Haïssaguerre M. (2006). Long-term evaluation of atrial fibrillation ablation guided by noninducibility. Heart Rhythm. 3:140–145.

James T.N. (1963). The connecting pathways between the sinus node and AV node and between the right and the left atrium in the human heart. Am Heart J. 66:498-508.

January C.T., Wann L.S., Alpert J.S., Calkins H., Cigarroa J.E., Cleveland J.C., Conti J.B., Ellinor P.T., Ezekowitz M.D., Field M.E., Murray K.T., Sacco R.L., Stevenson W.G., Tchou P.J., Tracy C.M. and Yancy C.W. (2014a). AHA/ACC/HRS Guideline for the management of patients with atrial fibrillation: Executive summary. JACC. 64:2246-2280.

Jongbloed M.R., Schalij M.J., Poelmann R.E., Blom N.A., Fekkes M.L., Wang Z., Fishman G.I. and Gittenberger-De Groot A.C. (2004). Embryonic conduction tissue: a spatial correlation with adult arrhythmogenic areas. J Cardiovasc Electrophysiol. 15:349–355.

Kakimoto Y., Tsuruyama T., Miyao M., Abiru H., Sumiyoshi S., Kotani H., Haga H. and Tamaki K. (2013). The effectiveness and limitations of triphenyltetrazolium chloride to detect acute myocardial infarction at forensic autopsy. Am J Forensic Med Pathol. 34:242–247.

Kalifa J., Zaitsev A., Warren M., Berenfeld O., Nattel S. and Jalife J. (2003). High atrial pressure- induced stretch during atrial fibrillation reveals high frequency periodic sources in the pulmonary vein region of the sheep heart (Abstract). Pacing Clin Electrophysiol. 26:950.

Kannel W.B., Wolf P.A., Benjamin E.J. and Levy D. (1998). Prevalence, incidence, prognosis, and predisposing conditions for atrial fibrillation: population-based estimates. Am J Cardiol. 82:2–9.

Chapter 1: General introduction

Karamichalakis N., Letsas K.P., Vlachos K., Georgopoulos S., Bakalakos M.E. and Sideris A. (2015). Managing atrial fibrillation in the very elderly patient: challenges and solutions. Vasc Health Risk Manag. 11:555–562.

Khargi K., Keyhan-Falsafi A., Hutten B.A., Ramanna H., Lemke B. and Deneke T. (2007) Surgical treatment of atrial fibrillation: a systematic review. Herzschrittmacherther Elektrophysiol. 18:68–76.

Kim M.H., Johnston S.S., Chu B.C., Dalal M.R. and Schulman K.L. (2011). Estimation of total incremental health care costs in patients with atrial fibrillation in the United States. Circ Cardiovasc Qual Outcomes. 4:313–320.

McGurrin M.K.J. (2015). The diagnosis and management of atrial fibrillation in the horse. Vet Med: Res Rep. 6:83–90.

Kirchhof P., Kahr P.C., Kaese S., Piccini I., Vokshi I., Scheld H.H., Rotering H., Fortmueller L., Laakmann S., Verheule S., Schotten U., Fabritz L. and Brown NA. (2011). PITX2c is expressed in the adult left atrium, and reducing Pitx2c expression promotes atrial fibrillation inducibility and complex changes in gene expression. Circ Cardiovasc Genet. 4:123–133.

Kirchhof P., Breithardt G., Aliot E., Al Khatib S., Apostolakis S., Auricchio A., Bailleul C., Bax J., Benninger G., Blomstrom-Lundqvist C., Boersma L., Boriani G., Brandes A., Brown H., Brueckmann M., Calkins H., Casadei B., Clemens A., Crijns H., Derwand R., Dobrev D., Ezekowitz M., Fetsch T., Gerth A., Gillis A., Gulizia M., Hack G., Haegeli L., Hatem S., Häusler K.G., Heidbüchel H., Hernandez-Brichis J., Jais P., Kappenberger L., Kautzner J., Kim S., Kuck K.H., Lane D., Leute A., Lewalter T., Meyer R., Mont L., Moses G., Mueller M., Münzel F., Näbauer M., Nielsen J.C., Oeff M., Oto A., Pieske B., Pisters R., Potpara T., Rasmussen L., Ravens U., Reiffel J., Richard-Lordereau I., Schäfer H., Schotten U., Stegink W., Stein K., Steinbeck G., Szumowski L., Tavazzi L., Themistoclakis S., Thomitzek K., Van Gelder I.C., von Stritzky B., Vincent A., Werring D., Willems S., Lip G.Y. and Camm A.J. (2013). Personalized management of atrial fibrillation: Proceedings from the fourth Atrial Fibrillation competence NETwork/European Heart Rhythm Association consensus conference. Europace. 15:1540–1556.

Kirchhof P., Breithardt G., Bax J., Benninger G., Blomstrom-Lundqvist C., Boriani G., Brandes A., Brown H., Brueckmann M., Calkins H., Calvert M., Christoffels V., Crijns H., Dobrev D., Ellinor P., Fabritz L., Fetsch T., Freedman S.B., Gerth A., Goette A., Guasch E., Hack G., Haegeli L., Hatem S., Haeusler K.G., Heidbüchel H., Heinrich-Nols J., Hidden-Lucet F., Hindricks G., Juul-Möller S., Kääb S., Kappenberger L., Kespohl S., Kotecha D., Lane D.A., Leute A., Lewalter T., Meyer R., Mont L., Münzel F., Nabauer M., Nielsen J.C., Oeff M., Oldgren J., Oto A., Piccini J.P., PilmeyerA., Potpara T., Ravens U., Reinecke H., Rostock T., Rustige J., Savelieva I., Schnabel R., Schotten U., Schwichtenberg L., Sinner

Chapter 1: General introduction

M.F., Steinbeck G., Stoll M., Tavazzi L., Themistoclakis S., Tse H.F., Van Gelder I.C., Vardas P.E., Varpula T., Vincent A., Werring D., Willems S., Ziegler A., Lip G.Y. and Camm A.J. (2016). A roadmap to improve the quality of atrial fibrillation management: proceedings from the fifth Atrial Fibrillation Network/European Heart Rhythm Association consensus conference. Europace. 18:37–50.

Kistler P.M., Sanders P., Fynn S.P., Stevenson I.H., Hussin A., Vohra J.K., Sparks P.B. and Kalman J.M. (2003). Electrophysiological and electrocardiographic characteristics of focal atrial tachycardia originating from the pulmonary veins: acute and long-term outcomes of radiofrequency ablation. Circulation. 108:1968–1975.

Kistler P.M., Sanders P., Fynn S.P., Stevenson I.H., Spence S.J., Vohra J.K., Sparks P.B. and Kalman J.M. (2004). Electrophysiologic and electroanatomic changes in the human atrium associated with age. J Am Coll Cardiol. 44:109–116.

Kosmidou I., Houde-Walter H., Foley L. and Michaud G. (2013). Loss of pace capture after radiofrequency application predicts the formation of uniform transmural lesions. Europace. 15:601– 606.

Kowalski M., Grimes M.M., Perez F.J., Kenigsberg D.N., Koneru J., Kasirajan V., Wood M.A. and Ellenbogen A. (2012). Histopathologic characterization of chronic radiofrequency ablation lesions for pulmonary vein isolation. J Am Coll Cardiol. 59:930–938.

Kumar S., Morton J.B., Lee J., Halloran K., Spence S.J., Gorelik A., Hepworth G., Kistler P.M. and Kalman J.M. (2012). Prospective characterization of catheter-tissue contact force at different anatomic sites during antral pulmonary vein isolation. Circ Arrhythm Electrophysiol. 5:1124–1129.

Kumar S., Chan M., Lee J., Wong M.C.G., Yudi M., Morton J.B., Spence S.J., Halloran K., Kistler P.M. and Kalman J.M. (2013). Catheter-tissue contact force determines atrial electrogram characteristics before and lesion efficacy after antral pulmonary vein isolation in humans. J Cardiovasc Electrophysiol. 25:122–129.

Kumar S. and Michaud G.F. (2016). Pulmonary vein isolation in the treatment of atrial fibrillation. Res Rep Clin Cardiol. 7:47–60.

Lachman N., Syed F.F., Habib A., Kapa S., Bisco S.E., Venkatachalam K.L. and Asirvatham S.J. (2010). Correlative anatomy for the electrophysiologist, Part I: the pericardial space, oblique sinus, transverse sinus. J Cardiovasc Electrophysiol. 21:1421–1426.

Chapter 1: General introduction

Leary M.C. and Caplan L.R. (2008). Cardioembolic stroke: an update on etiology, diagnosis and management. Ann Indian Acad Neurol. 11:52–63.

Lee W.C., Lamas G.A., Balu S., Spalding J., Wang Q. and Pashos C.L. (2008). Direct treatment cost of atrial fibrillation in the elderly American population: a Medicare perspective. J Med Econ. 11:281– 298.

Lemola K., Chartier D., Yeh Y. H., Dubuc M., Cartier R., Armour A., Ting M., Sakabe M., Shiroshita- Takeshita A., Comtois P. and Nattel S. (2008). Pulmonary vein region ablation in experimental vagal atrial fibrillation: role of pulmonary veins versus autonomic ganglia. Circulation. 117:470–477.

Leroux A. A., Detilleux J., Sandersen C. F., Borde L., Houben R. M., Al Haidar A., Art T. and Amory H. (2014). Prevalence and risk factors for cardiac diseases in a hospital based population of 3434 horses (1994-2011). J Vet Intern Med. 27:1563–1570.

Li W.J., Bai Y.Y., Zhang H.Y., Tang R.B., Miao C.L., Sang C.H., Yin X.D., Dong J.Z. and Ma CS. (2011). Additional ablation of complex fractionated atrial electrograms after pulmonary vein isolation in patients with atrial fibrillation: a meta-analysis. Circ Arrhythm Electrophysiol. 4:143–148.

Lickfett L., Kato R., Tandri H., Jayam V., Vasamreddy C.R., Dickfeld T., Lewalter T., Luderitz B., Berger R., Halperin H. and Calkins H. (2004). Characterization of a new pulmonary vein variant using magnetic resonance angiography: incidence, imaging, and interventional implications of the “right top pulmonary vein”. J Cardiovasc Electrophysiol. 15:538–543.

Little R.C. (1960). Volume pressure relationships of the pulmonary left heart vascular segment: evidence for a ‘valve-like’ closure of the pulmonary veins. Ibid. 8:594.

Lloyd-Jones D.M., Wang T.J., Leip E.P., Larson M.G., Levy D., Vasan R.S., D'Agostino R.B., Massaro J.M., Beiser A., Wolf P.A. and Benjamin E.J. (2004). Lifetime risk for development of atrial fibrillation: the Framingham Heart Study. Circulation. 110:1042-1046.

Lo L.W. and Chen S.A. (2010). Superior vena cava–aorta ganglionated plexus: an arbitrator of the heart or just a passer-by? J Cardiovasc Electrophysiol. 21:1400–1402.

Loomis T.A. and Krop S. (1955). Auricular fibrillation induced and maintained in animals by acetylcholine or vagal stimulation. Circ Res. 3:390–396.

Mahida S., Sacher F., Derval N., Berte B., Yamashita S., Hooks D., Denis A., Amraoui S., Hocini M., Haïssaguerre M. and Jais P. (2015). Science Linking Pulmonary Veins and Atrial Fibrillation. Arrhythm Electrophysiol Rev. 4:40–43.

Chapter 1: General introduction

Majeed A., Moser K. and Carroll K. (2001). Trends in the prevalence and management of atrial fibrillation in general practice in England and Wales, 1994-1998: analysis of data from the general practice research database. Heart. 86:284–288.

Marijon E., Fazaa M., Narayanan K., Guy-Moyat B., Bouzeman A., Providencia R., Treguer F., Combes N., Bortone A., Boveda S., Combes S. and Albenque J.-P. (2014). Real-time contact force sensing for pulmonary vein isolation in the setting of paroxysmal atrial fibrillation: procedural and 1-year results. J Cardiovasc Electrophysiol. 25:130–137.

Marr C.M. and Patteson M. (2010). Echocardiography. IN: Cardiology of the Horse, second edition (eds.: Marr C.M. and Bowen M.), Saunders Elsevier, Amsterdam, pp.3–21.

McAnulty P.A., Dayan D.A., Ganderup N.C. and Hastings K.L. (2012). The minipig in biomedical research. CRC Press: Taylor & Francis, New York.

McGurrin K.M.J., Physick-Sheard P.W., Kenney D.G., Kerr C., Hanna W.J.B., Neto F.T. and Weese J.S. (2003). Transvenous Electrical Cardioversion in Equine Atrial Fibrillation: Technique and Successful Treatment of 3 Horses. J Vet Intern Med. 5:715–718.

McGurrin M.K., Physick-Sheard P.W. and Kenney D.G. (2005a). How to perform transvenous electrical cardioversion in horses with atrial fibrillation. J Vet Cardiol. 7:109–119.

McGurrin M.K., Physick-Sheard P.W., Kenney D.G., Kerr C. and Hanna W.J. (2005b). Transvenous electrical cardioversion of equine atrial fibrillation: technical considerations. J Vet Intern Med. 19:695–702.

McKibben J.S. and Ghoshal N.G. (1975). Autonomics. In: Sisson and Grossman’s The anatomy of the domestic animals, volume 2, Getty, R. (eds.). W.B. Saunders Company, Philadelphia, pp. 1397–1408.

Melby S., Lee A. and Damiano R. (2005). Advances in surgical ablation devices for atrial fibrillation. In: Wang P., Naccarelli G. and Rosen M. (eds). New arrhythmia technologies. Blackwell Publishing, MA, pp. 233–241.

Melby S.J., Lee A.M., Zierer A., Kaiser S.P., Livhits M.J., Boineau J.P., Schuessler R.B. and Damiano R.J. Jr. (2008). Atrial fibrillation propagates through gaps in the ablation lines: implications for ablative treatment of atrial fibrillation. Heart Rhythm. 5:1296–1301.

Melnyk P., Ehrlich J.R., Pourrier M., Villeneuve L., Cha T.J. and Nattel S. (2005). Comparison of ion channel distribution and expression in cardiomyocytes of canine pulmonary veins versus left atrium. Cardiovasc Res. 65:104–116.

Chapter 1: General introduction

Miller P.J. and Holmes J.R. (1985). Observations on seven cases of mitral insufficiency in the horse. Equine Vet J. 17:181–190.

Miller M.A., d’Avila A., Dukkipati S.R., Koruth J.S., Viles-Gonzalez J., Napolitano C., Eggert C., Fischer A., Gomes J.A. and Reddy V.Y. (2012). Acute electrical isolation is a necessary but insufficient endpoint for achieving durable PV isolation: the importance of closing the visual gap. Europace. 14:653–660.

Moe G.K. and Abildskov J.A. (1959). Atrial fibrillation as a self-sustaining arrhythmia independent of focal discharge. Am Heart J. 58:59–70.

Mommersteeg M.T., Brown N.A., Prall O.W., de Gier-de Vries C., Harvey R.P., Moorman A.F. and Christoffels VM. (2007). Pitx2c and Nkx2-5 are required for the formation and identity of the pulmonary myocardium. Circ Res. 101:902–902.

Monfredi O., Dobrzynski H., Mondal T., Boyett M.R. and Morris G.M. (2010). The anatomy and physiology of the sinoatrial node – a contemporary review. Pacing Clin Electrophysiol. 33:1392–1406.

Morel E., Meyronet D., Thivolet-Bejuy F. and Chevalier P. (2008). Identification and distribution of interstitial Cajal cells in human pulmonary veins. Heart Rhythm. 5:1063–1067.

Morris D.D. and Fregin G.F. (1982). Atrial fibrillation in horses: factors associated with response to quinidine sulfate in 77 clinical cases. Cornell Vet. 72:339–349.

Moubarak J.B., Rozwadowski J.V., Strzalka C.T., Buck W.R., Tan W.S., Kish G.F., Kisiel T., Fronc H.C. and Maloney J.D. (2000). Pulmonary veins-left atrial junction: anatomical and histological study. Pacing Clin Electrophysiol. 23:1836-1838.

Natale A., Raviele A., Arentz T., Calkins H., Chen S.A., Haïssaguerre M., Hindricks G., Ho Y., Kuck K.H., Marchlinski F., Napolitano C., Packer D., Pappone C., Prystowsky E.N., Schilling R., Shah D., Themistoclakis S. and Verma A. (2007). Venice Chart international consensus document on atrial fibrillation ablation. J Cardiovasc Electrophysiol. 18:560–580.

Nath S., Haines D.E., Kron I.L. and DiMarco J.P. (1994a). The long-term outcome of visually directed subendocardial resection in patients without inducible of mappable ventricular tachycardia at the time of surgery. J Cardiovasc Electrophysiol. 5:399–407.

Nath S., Redick J.A., Whayne J.G. and Haines D.E. (1994b). Ultrastructural observations in the myocardium beyond the region of acute coagulation necrosis following radiofrequency catheter ablation. J Cardiovasc Electrophysiol. 5:838–845.

Chapter 1: General introduction

Nathan H. and Eliakim M. (1966). The junction between the left atrium and the pulmonary veins. An anatomic study of human hearts. Circulation. 34:412–422.

Nathan H. and Gloobe H. (1970). Myocardial atrio-venous junctions and extensions (sleeves) over the pulmonary and caval veins. Anatomical observations in various mammals. Thorax. 25:317.

Nattel S. (2002). New ideas about atrial fibrillation 50 years on. Nature. 415:219–226.

Nattel S., Burstein B. and Dobrev D. (2008). Atrial remodeling and atrial fibrillation: mechanisms and implications. Circ Arrhythm Electrophysiol. 1:62–73.

Nattel S. (2013). Paroxysmal atrial fibrillation and pulmonary veins: relationships between clinical forms and automatic versus re-entrant mechanisms. Can J Cardiol. 29:1147–1149.

Neumann T., Sperzel J., Dill T., Kluge A., Erdogan A., Greis H., Hansel J., Berkowitsch A., Kurzidim K., Kuniss M., Hamm C.W. and Pitschner H.F. (2005). Percutaneous pulmonary vein stenting for the treatment of severe stenosis after pulmonary vein isolation. J Cardiovasc Electrophysiol. 16:1180– 1188.

Neumann T., Kuniss M., Conradi G., Sperzel J., Berkowitsch A., Zaltsberg S., Wojcik M., Erkapic D., Dill T., Hamm C.W. and Pitschner H.-F. (2009). Pulmonary Vein Stenting for the Treatment of Acquired Severe Pulmonary Vein Stenosis after Pulmonary Vein Isolation: Clinical Implications after Long-Term Follow-Up of 4 Years. J Cardiovasc Electrophysiol. 20:251-257.

Nguyen B.L., Fishbein M.C., Chen L.S., Chen P.S. and Masroor S. (2009). Histopathological substrate for chronic atrial fibrillation in humans. Heart Rhythm. 6:454–460.

Nidorf S.M., Picard M.H., Triulzi M.O., Thomas J.D., Newell J., King M.E. and Weyman A.E. (1992). New perspectives in the assessment of cardiac chamber dimensions during development and adulthood. J Am Coll Cardiol. 19: 983.

Nishida K., Maguy A., Sakabe M., Comtois P., Inoue H. and Nattel S. (2011). The role of pulmonary veins vs. autonomic ganglia in different experimental substrates of canine atrial fibrillation. Cardiovasc Res. 89:825–833.

Ohmura H., Hiraga A., Takahashi T., Kai M. and Jones J. (2003). H. Risk factors for atrial fibrillation during racing in slow-finishing horses. J Am Vet Med Assoc. 223:84–88.

Chapter 1: General introduction

Ohsawa M., Okayama A., Sakata K., Kato K., Itai K., Onoda T. and Ueshima H. (2005). Rapid increase in estimated number of persons with atrial fibrillation in Japan: an analysis from national surveys on cardiovascular diseases in 1980, 1990 and 2000. J Epidemiol. 15:194–196.

Oral H., Knight B.P., Tada H., Özaydın M., Chugh A., Hassan S., Scharf C., Lai S.W.K., Greenstein R., Pelosi F., Strickberger S.A. and Morady F. (2002). Pulmonary vein isolation for paroxysmal and persistent atrial fibrillation. Circulation. 105:1077–1081.

Ouyang F., Antz M., Ernst S., Hachiya H., Mavrakis H., Deger F.T., Schaumann A., Chun J., Falk P., Hennig D., Liu X., Bänsch D. and Kuck K.H. (2005). Recovered pulmonary vein conduction as a dominant factor for recurrent atrial tachyarrhythmias after complete circular isolation of the pulmonary veins: lessons from double Lasso technique. Circulation. 111:127–135.

Pak H.N., Hwang C., Lim H.E., Kim J.S. and Kim Y.H. (2007). Hybrid epicardial and endocardial ablation of persistent or permanent atrial fibrillation: a new approach for difficult cases. J Cardiovasc Electrophysiol. 18:917–923.

Pandit S.V., Berenfeld O., Anumonwo J.M., Zaritski R.M., Kneller J., Nattel S. and Jalife J. (2005). Ionic determinants of functional reentry in a 2-D model of human atrial cells during simulated chronic atrial fibrillation. Biophys J. 88:3806–3821.

Pappone C., Oral H., Santinelli V., Vicedomini G., Lang C.C., Manguso F., Torracca L., Benussi S., Alfieri O., Hong R., Lau W., Hirata K., Shikuma N., Hall B. and Morady F. (2004a). Atrio-esophageal fistula as a complication of percutaneous transcatheter ablation of atrial fibrillation. Circulation. 109:2724- 2726.

Pappone C., Santinelli V., Manguso F., Vicedomini G., Gugliotta F., Augello G., Mazzone P., Tortoriello V., Landoni G., Zangrillo A., Lang C., Tomita T., Mesas C., Mastella E. and Alfieri O. (2004b). Pulmonary vein denervation enhances long-term benefit after circumferential ablation for paroxysmal atrial fibrillation. Circulation. 109:327–334.

Park H.W., Shen M.J., Lin S.F., Fishbein M.C., Chen L.S. and Chen P.S. (2012). Neural mechanisms of atrial fibrillation. Curr Opin Cardiol. 27:24–8.

Patterson E., Po S.S., Scherlag B.J. and Lazzara R. (2005). Triggered firing in pulmonary veins initiated by in vitro autonomic nerve stimulation. Heart Rhythm. 2:624–631.

Patteson M. (1996). Cardiac anatomy and Physiology. In: Equine Cardiology. Blackwell Science, Oxford, pp. 1-25.

Chapter 1: General introduction

Patteson M. (1999). Two-dimensional and M-mode echocardiography. In: Cardiology of the horse. (eds. Marr C.M.). W.B. Saunders, London, pp. 93-116.

Patton K.K., Zacks E.S., Chang J.Y., Shea M.A., Ruskin J.N., Macrae C.A. and Ellinor P.T. (2005). Clinical subtypes of lone atrial fibrillation. Pacing Clin Electrophysiol. 28:630–638.

Pauza D.H., Skripka V., Pauziene N. and Stropus R. (2000). Morphology, distribution, and variability of the epicardiac neural ganglionated subplexuses in the human heart. Anat Rec. 259:353–382.

Pauza D.H., Skripka V. and Pauziene N. (2002). Morphology of the intrinsic cardiac nervous system in the dog: a whole-mount study employing histochemical staining with acetylcholinesterase. Cells Tissues Organs. 172:297–320.

Perez-Lugones A., Schvartzman P.R., Schweikert R., Tchou P.J., Saliba W., Marrouche N.F., Castle L.W., White R.D. and Natale A. (2003). Three-dimensional reconstruction of pulmonary veins in patients with atrial fibrillation and controls: morphological characteristics of different veins. Pacing Clin Electrophysiol. 26:8–15.

Piccini J.P., Hammill B.G., Sinner M.F., Jensen P.N., Hernandez A.F., Heckbert S.R., Benjamin E.J. and Curtis L.H. (2012). Incidence and prevalence of atrial fibrillation and associated mortality among Medicare beneficiaries, 1993-2007. Circ Cardiovasc Qual Outcomes. 5:85–93.

Platt M., Mandapati R., Scherlag B.J., Yamanashi W.S., Nakagawa H., Lazzara R. and Jackman W.M. (2004). Limiting the number and extent of radiofrequency applications to terminate atrial fibrillation and subsequently prevent its inducibility. Heart Rhythm. 1:S10.

Po S.S., Scherlag B.J., Yamanashi W.S., Edwards J., Zhou J., Wu R., Geng N., Lazzara R. and Jackman W.M. (2006). Experimental model for paroxysmal atrial fibrillation arising at the pulmonary vein- atrial junctions. Heart Rhythm. 3:201–208.

Po S.S., Nakagawa H. and Jackman W.M. (2009). Localization of left atrial ganglionated plexi in patients with atrial fibrillation. J Cardiovasc Electrophysiol. 20:1186–1189.

Poirier P. and Charpy A. (1911). Traite´ d’anatomie humaine, new edition, vol 2. Masson, Paris, pp. 601–603.

Pokushalov E., Romanov A., Katritsis D.G., Artyomenko S., Shirokova N., Karaskov A., Mittal S. and Steinberg J.S. (2013). Ganglionated plexus ablation vs linear ablation in patients undergoing pulmonary vein isolation for persistent/long-standing persistent atrial fibrillation: a randomized comparison. Heart Rhythm. 10:1280–1286.

Chapter 1: General introduction

Prieto L.R., Schoenhagen P., Arruda M.J., Natale A. and Worley S.E. (2008). Comparison of stent versus balloon angioplasty for pulmonary vein stenosis complicating pulmonary vein isolation. J Cardiovasc Electrophysiol. 19:673–678.

Ranjan R., Kato R., Zviman M.M., Dickfeld T.M., Roguin A., Berger R.D., Tomaselli G.F. and Halperin H.R. (2011). Gaps in the ablation line as a potential cause of recovery from electrical isolation and their visualization using MRI. Circ Arrhythm Electrophysiol. 4:279–286.

Ranjan R., Kholmovski E.G., Blauer J., Vijayakumar S., Volland N.A., Salama M.E., Parker D.L., MacLeod R. and Marrouche N.F. (2012). Identification and acute targeting of gaps in atrial fibrillation lesion sets using a real-time magnetic resonance imaging system. Circ arrhythm Electrophysiol. 5:1130– 1135.

Ravens U. (2003). Mechano-electric feedback and arrhythmias. Prog Biophys Mol Biol. 82:255–266.

Roux N., Havet E. and Mertl P. (2004). The myocardial sleeves of the pulmonary veins: potential implications for atrial fibrillation. Surg Radiol Anat. 26:285–289.

Redfors B., Shao Y. and Omerovic E. (2012). Myocardial infarct size and area at risk assessment in mice. Exp Clin Cardiol. 17: 268-272.

Reef V.B., Levitan C.W. and Spencer P.A. (1988). Factors Affecting Prognosis and Conversion in Equine Atrial-Fibrillation. J Vet Intern Med. 2:1-6.

Reef V.B. (1991). Echocardiographic findings in horses with congenital cardiac disease. Compend Contin Educ Pract Vet. 13:109-117.

Reef V.B., Reimer J.M. and Spencer P.A. (1995). Treatment of Atrial-Fibrillation in Horses - New Perspectives. J Vet Int Med. 9: 57-67.

Reef V.B., Bain F.T. and Spencer P.A. (1998a). Severe mitral regurgitation in horses: Clinical, echocardiographic and pathological findings. Equine Vet J. 30:18–27.

Reef V.B. (1998b). Cardiovascular ultrasonography. In: Equine Diagnostic Ultrasound, Reef V.B. (eds.) W.B. Saunders, Philadelphia, pp. 215-272.

Reef V.B. and McGuirk S.M. (2002). Diseases of the cardiovascular system. Atrial fibrillation. In: Large Animal Internal Medicine, 3rd ed. (eds. Smith B.P.), Missouri, pp. 471–474.

Chapter 1: General introduction

Reef V.B., Bonagura J., Buhl R., McGurrin M.K., Schwarzwald C.C., van Loon G. and Young L.E. (2014). Recommendations for management of equine athletes with cardiovascular abnormalities. J Vet Intern Med. 28:749–761.

Saito T., Waki K. and Becker A.E. (2000). Left atrial myocardial extension onto pulmonary veins in humans: anatomic observations relevant for atrial arrhythmias. J Cardiovasc Electrophysiol. 11:888– 894.

Santangeli P., Di Biase L., Burkhardt D.J., Horton R., Sanchez J., Bai R., Pump A., Perez M., Wang P.J., Natale A. and Al-Ahmad A. (2012). Catheter ablation of atrial fibrillation: state-of-the-art techniques and future perspectives. J Cardiovasc Med. 13:108–124.

Scanavacca M., Pisani C.F., Hachul D., Lara S., Hardy C., Darrieux F., Trombetta I., Negrão C.E. and Sosa E. (2006). Selective atrial vagal denervation guided by evoked vagal reflex to treat patients with paroxysmal atrial fibrillation. Circulation. 114:876–885.

Scherf D., Romano F.J. and Terranova R. (1948). Experimental studies on auricular flutter and auricular fibrillation. Am Heart J. 36:241–251.

Scherlag B.J., Yamanashi W., Patel U., Lazzara R. and Jackman W.M. (2005a). Autonomically induced conversion of pulmonary vein focal firing into atrial fibrillation. J Am Coll Cardiol. 45:1878–1886.

Scherlag B.J., Nakagawa H., Jackman W.M., Yamanashi W.S., Patterson E., Po S. and Lazzara R. (2005b). Electrical stimulation to identify neural elements on the heart: Their role in atrial fibrillation. J Interv Card Electrophysiol. 13:37–42.

Schotten U., Verheule S., Kirchhof P. and Goette A. (2011). Pathophysiological mechanisms of atrial fibrillation: a translational appraisal. Physiol Rev. 91:265–325.

Schwarzwald C.C. (2014). Ultrasonography of the Heart. In: Atlas of Equine Ultrasonography (eds. Kidd J.A., Lu K.G. and Frazer M.L.). John Wiley & Sons, Oxford, pp. 379-408.

Scridon A., Fouilloux-Meugnier E., Loizon E., Rome S., Julien C., Barrès C. and Chevalier P. (2015). Longstanding arterial hypertension is associated with Pitx2 down-regulation in a rat model of spontaneous atrial tachyarrhythmias. Europace. 17:160–165.

Shen M.J., Choi E.K., Tan A.Y., Lin S.F., Fishbein M.C., Chen L.S. and Chen P.S. (2012). Neural mechanisms of atrial arrhythmias. Nat Rev Cardiol. 9:30–39.

Chapter 1: General introduction

Shin D.-I. and Deneke T. (2010). Ablation catheters for atrial fibrillation: benefits and complications. Therapy. 7:133–138.

Smelley M.P. and Knight B.P. (2009). Approaches to catheter ablation of persistent atrial fibrillation. Heart Rhythm. 6:33–38.

Spach M.S., Barr R.C. and Jewett P.H. (1972). Spread of activation from the atrium into thoracic veins in human beings and dogs. Am J Cardiol. 30:844–854.

Steinberg J.S., Palekar R., Sichrovsky T., Arshad A., Preminger M., Musat D., Shaw R.E. and Mittal S. (2014). Very long-term outcome after initially successful catheter ablation of atrial fibrillation. Heart Rhythm. 11:771–776.

Stephenson R.B. (2007). Electrical activity of the heart. In: Textbook of veterinary physiology, 4th edition, Bradley G.K. and Cunningham J.G. (eds.), Saunders/Elsevier, Philadelphia, pp. 193–212.

Steven D., Reddy V.Y., Inada K., Roberts-Thomson K.C., Seiler J., Stevenson W.G. and Michaud G.F. (2010). Loss of pace capture on the ablation line: a new market for complete radiofrequency lesions to achieve pulmonary vein isolation. Heart Rhythm. 7:323–330.

Stewart S., Murphy N.F., Murphy N., Walker A., McGuire A. and McMurray J.J. (2004). Cost of an emerging epidemic: an economic analysis of atrial fibrillation in the UK. Heart. 90:286–292.

Stieda L. (1877). Ueber guergestreifle Muskelfasern in der Wand tier Lungvenen. Arch Mikr Anat. 14:243.

Swindle M.M. (2007). Swine in the laboratory surgery, anaesthesia, imaging and experimental techniques, 2nd edition, C.R.C Taylor & Francis, New York.

Tan A.Y., Li H., Wachsmann-Hogiu S., Chen L.S., Chen P.S. and Fishbein M.C. (2006). Autonomic innervation and segmental muscular disconnections at the human pulmonary vein-atrial junction: implications for catheter ablation of atrial-pulmonary vein junction. J Am Coll Cardiol. 48:132–143.

Tan A.Y., Chen P.S., Chen L.S., Fishbein M.C. (2007). Autonomic nerves in pulmonary veins. Heart Rhythm. 4:57–60.

Tehrani S. and Lipkin D. (2013). Angioplasty of acquired pulmonary vein stenosis using covered stent. Catheter Cardiovasc Interv. 82:617-620.

Thein E. and Hammer C. (2004). Physiologic barriers to xenotransplantation. Curr Opin Organ Transplant. 9:186.

Chapter 1: General introduction

Tsai C.F., Tai C.T., Hsieh M.H., Lin W.S., Yu W.C., Ueng K.C., Ding Y.A., Chang M.S. and Chen S.A. (2000). Initiation of atrial fibrillation by ectopic beats originating from the superior vena cava - Electrophysiological characteristics and results of radiofrequency ablation. Circulation. 102:67–74.

Tsao H.M., Wu M.H., Yu W.C., Tai C.T., Lin Y.K., Hsieh M.H., Ding Y.A., Chang M.S. and Chen SA. (2001). Role of right middle pulmonary vein in patients with paroxysmal atrial fibrillation. J Cardiovasc Electrophysiol. 12:1353–1357.

Tutuianu C., Szilagy J., Pap R. and Sághy L. 2(015), Very long-term results of atrial fibrillation confirm that this therapy is really effective. J Atr Fibrillation. 8:1226.

Vaitkevicius R., Saburkina I., Zaliunas R., Pauziene N., Vaitkeviciene I., Schauerte P. and Pauza D. H. (2008). Innervation of pulmonary veins: morphologic pattern and pathways of nerves in the human fetus. Ann Anat. 190:158–166.

Vaitkevicius R., Saburkina I., Rysevaite K., Vaitkeviciene I., Pauziene N., Zaliunas R., Schauerte P., Jalife J. and Pauza D.H. (2009). Nerve supply of the human pulmonary veins: an anatomical study. Heart Rhythm. 6:221–228. van Loon G., Tavernier R., Fonteyne W., Duytschaever M., Jordaens L. and Deprez P., (2000). Pacing induced long-term atrial fibrillation in horses. Europace. 2:84. van Loon G. (2001a). Atrial pacing and experimental atrial fibrillation in equines. Dissertation, Ghent University, Belgium. van Loon G. (2001b). Effect of experimental chronic atrial fibrillation in equines. In: Atrial Pacing and Experimental Atrial Fibrillation in Equines, (Chapter 7), pp. 162–206. van Loon G., Blissitt K.J., Keen J.A. and Young L.E. (2004). Use of intravenous flecainide in horses with naturally-occurring atrial fibrillation. Equine Vet J. 36:609–614.

Veenhuyzen G.D., Simpson C.S. and Abdollah H. (2004). Atrial fibrillation. Can Med Assoc J. 171:755- 760.

Verdouw P.D. and Hartog J.M. (1986). Provocation and suppression of ventricular arrhythmias in domestic swine. In: Swine in Cardiovascular Research, Vol. 2., Stanton H.C. and Mersmann H.J. (eds.), CRC Press, Boca Raton, pp. 121.

Verheyen T., Decloedt A., van der Vekens N., Sys S., De Clercq D. and van Loon, G. (2013). Ventricular response during lungeing exercise in horses with lone atrial fibrillation. Equine Vet J. 45:309-314.

Chapter 1: General introduction

Verma A. (2011). The techniques for catheter ablation of paroxysmal and persistent atrial fibrillation: a systematic review. Curr Opin Cardiol. 26:17–24.

Voigt N., Heijman J., Wang Q., Chiang D.Y., Li N., Karck M., Wehrens X.H., Nattel S. and Dobrev D. (2014). Cellular and molecular mechanisms of atrial arrhythmogenesis in patients with paroxysmal atrial fibrillation. Circulation. 129:145–156.

Wakili R., Voigt N., Kääb S., Dobrev D. and Nattel S. (2011). Recent advances in the molecular pathophysiology of atrial fibrillation. J Clin Invest. 121:2955–2968.

Wang J., Klysik E., Sood S., Johnson R.L., Wehrens X.H. and Martin J.F. (2010). Pitx2 prevents susceptibility to atrial arrhythmias by inhibiting left-sided pacemaker specification. Proc Natl Acad Sci USA. 107:9753–9758.

Wazni O.M., Marrouche N.F., Martin D.O., Verma A., Bhargava M., Saliba W., Bash D., Schweikert R., Brachmann J., Gunther J., Gutleben K., Pisano E., Potenza D., Fanelli R., Raviele A., Themistoclakis S., Rossillo A., Bonso A. and Natale A. (2005). Radiofrequency ablation vs antiarrhythmic drugs as first- line treatment of symptomatic atrial fibrillation: a randomized trial. JAMA. 293:2634-2640.

Wijffels M., Kirchhof J., Dorland R. and Allessie M. (1995). Atrial fibrillation begets atrial fibrillation. A study in awake chronically instrumented goats. Circulation. 92:1954–1968.

Wood M.A. and Fuller I.A. (2002). Acute and chronic electrophysiologic changes surrounding radiofrequency lesions. J Cardiovasc Electrophysiol. 13:56-61.

Xu J., Luc J.G.Y. and Phan K. (2016). Atrial fibrillation: a review of current treatment strategies. J Thorac Dis. 8:886-900.

Yamamoto K., Yasuda J. and Too K. (1992). Arrhythmias in newborn thoroughbred foals. Equine Vet J. 24:169-173.

Yang Y., Xia M., Jin Q., Bendahhou S., Shi J., Chen Y., Liang B., Lin J., Liu Y., Liu B., Zhou Q., Zhang D., Wang R., Ma N., Su X., Niu K., Pei Y., Xu W., Chen Z., Wan H., Cui J., Barhanin J. and Chen Y. (2004). Identification of a KCNE2 gain-of-function mutation in patients with familial atrial fibrillation. Am J Hum Genet. 75:899–905.

Young L.E. (2003). Equine athletes, the equine athlete’s heart and racing success. Exp Physiol. 88:659–663.

Chapter 1: General introduction

Young L.E. and van Loon G. (2014). Diseases of the heart and vessels. In: Equine sports and medicine & surgery, 2nd edition, Hinchcliff K.W., Kaneps A.J. and Geor R.J (eds.), Saunders Elsevier, London, pp. 695–744.

Yuan B.X., Ardell J.L., Hopkins D.A., Losier A.M. and Armour J.A. (1994). Gross and microscopic anatomy of the canine intrinsic cardiac nervous system. Anat Rec. 239:75–87.

Yue L., Xie J. and Nattel S. (2011). Molecular determinants of cardiac fibroblast electrical function and therapeutic implications for atrial fibrillation. Cardiovasc Res. 89:744–753.

Zoni-Berisso M., Lercari F., Carazza T. and Domenicucci S. (2014). Epidemiology of atrial fibrillation: European perspective. Clin Epidemiol. 6:213–220.

Chapter 2

Scientific aims

Chapter 2: Scientific aims

The first general aim of this thesis was to gather the required data to develop and test a new implantable ablation device in a pig model.

The specific scientific aims were:

1) To gather detailed information about the morphology of the PV in pigs (chapter 3) 2) To describe and visualize the organization of the myocardial sleeve in the PV of pigs (chapters 3 and 10) 3) To prove the presence of ganglia and telocytes at the level of the atrio-venous junction in pigs (chapter 4) 4) To develop a wireless ablation prototype in pigs which may lead to the development of a new ablation device in humans (chapters 5, 6 and 8) 5) To be able to visualize immunohistochemically post-mortem ablation lesion (chapter 7)

The second general aim was to describe the equine PV and identify those veins on ultrasound

The specific scientific aims were:

1) To gather detailed information about the anatomical organization of the PV in horses (chapter 9) 2) To describe and visualize the organization of the myocardial sleeve in the PV of horses (chapter 10) 3) To prove the presence of ganglia and telocytes at the level of the atrio-venous junction in horses (chapter 10) 4) To identify the equine PV on echocardiography to allow further development of interventions such as PV catheterization, mapping or even ablation (chapter 11)

Chapter 3 The pulmonary veins of the pig

Adapted from Vandecasteele T., Vandevelde K., Doom M., Van Mulken E., Simoens P. and Cornillie P. (2015). The pulmonary veins of the pig as an anatomical model for the development of a new treatment for atrial fibrillation. Anat Histol Embryol. 44:1-12.

Chapter 3: The pulmonary veins of the pig

Summary

The layout of the porcine atriopulmonary junction and immediately adjacent structures was investigated by gross anatomical and vascular corrosion casting studies to meet the need for more in- depth anatomical insights when using the pig as an animal model in the development of innovative approaches for surgical cardiac ablation in man. The veins from the right cranial and middle lung lobes drain through a common ostium in the left atrium, whereas a second ostium receives the blood returning from all other lung lobes, although limited variation to this pattern was observed. Surrounding anatomical structures that are most vulnerable to ablation damage as reported in man are located at a safer distance from the pulmonary veins in pigs, yet at certain locations, comparable risks are to be considered. Additionally, it was histologically confirmed that myocardial sleeves extend to over a centimetre in the wall of the pulmonary veins.

Chapter 3: The pulmonary veins of the pig

Introduction

Atrial fibrillation is the most commonly diagnosed cardiac arrhythmia responsible for substantial mortality and morbidity in man. The incidence increases with age and can be associated with several internal diseases or genetic disorders (Brugada et al., 1997; Kannel et al., 1998). Focal triggering sites of cardiac arrhythmia may be resided in various anatomical structures such as the crista terminalis, the ostium of the coronary sinus, the interatrial septum and the atrial free wall (Haïssaguerre et al., 1996, 1998; Chen et al., 1999; Hsieh et al., 1999; Pappone et al., 2000; Cappato et al., 2003; Gerstenfeld et al., 2003; Nanthakumar et al., 2004) as well as in myocardial sleeves located in the pulmonary veins. The latter structures are myocardial fibres spreading out from the left atrium into the wall of the pulmonary veins and are believed to be one of the most important origins of ectopic electric pulses (Nathan and Gloobe, 1970; Haïssaguerre et al., 1998; Chen et al., 1999; Saito et al., 2000; Roux et al., 2004). These myocardial fibres act as an ectopic pacemaker transmitting electrical pulses which interfere with the pulses originating from the sinoatrial node, disturbing the normal heart rhythm and inducing atrial fibrillation in humans (Haïssaguerre et al., 2000; Saito et al., 2000; Tagawa et al., 2001). The sleeves exhibit different arrangement patterns in humans and in various mammals. Patients suffering from atrial fibrillation are often not treated when their quality of life is not affected, or they receive only pharmacological treatment when a surgical intervention is not considered appropriate. When a pharmacological approach and electric cardioversion are unsuccessful and return to sinus rhythm is thought to be necessary, ablation procedures can be applied by creating local transmural lesions at the left atrial ostia draining the pulmonary veins to eliminate the influence of ectopic myocardial sleeve pacemakers (Haïssaguerre et al., 1998; Chen et al., 1999; Haïssaguerre et al., 2000; Pappone et al., 2000). Those procedures still have some disadvantages because they require difficult and prolonged interventions and are therefore more prone to failure, which may lead to conduction recurrence and major complications (Haïssaguerre et al., 1996; Robbins et al., 1998; Pappone et al., 1999; Mohr et al., 2002) including phrenic nerve injuries (Lee et al., 2004; Bunch et al., 2005; Okumura et al., 2008, 2009; Ahsan et al., 2010; Andrie et al., 2012), atrio-oesophageal fistulae (Gillinov et al., 2001; Sonmez et al., 2003; Pappone et al., 2004) and pulmonary vein stenosis (Robbins et al., 1998; Chen et al., 1999; Arentz et al., 2003).

To perform research into possible treatments for atrial fibrillation in humans, a suitable animal model is needed. The porcine model has multiple times been proven to be a good animal model for humans because of comparable body size and numerous anatomical, immunological, biochemical, physiological and genetic similarities between pigs and humans (Gregg et al., 1980; Shulman et al., 1988; Delange et al., 1992; Bermejo et al., 1993; Rowan et al., 1994; Jones et al., 1999; Paterson et

Chapter 3: The pulmonary veins of the pig al., 2002; Sommerer et al., 2004; Perry et al., 2005; Barker et al., 2006; Brunet et al., 2006; Ibrahim et al., 2006; Glenny et al., 2007; Hotchkiss et al., 2007; Wang et al., 2007; Rogers et al., 2008; Khatri et al., 2010). The porcine heart, of which the microanatomy is representative for the mammalian heart, has an average weight of approximately 300 g, which is similar to the average weight of the human heart (approximately 300 g in men and 250 g in women) (Barone, 1997).

The development of an appropriate animal model for ablation therapy studies requires in-depth data on the fine anatomical and histological architecture of the pulmonary veins and the myocardial sleeves and on the consequences of a heating process at the level of this myocardial tissue in pigs. This essential information is indispensable when approaching the ostia of the pulmonary veins from the left atrium to perform an intraluminal ablation at this level, but is lacking in current literature. Therefore, in the present study, the microanatomy of the porcine pulmonary veins was investigated to document the atriopulmonary junction in swine. This includes the number of pulmonary vein ostia, the different pulmonary veins draining into their specific ostia, the branching pattern of the various pulmonary veins and its variability, and the position of the pulmonary veins in relation to their surrounding structures such as the trachea, the oesophagus and the pulmonary arteries. These parameters, together with previous data provided by Vollmerhaus et al. (1999), who described the relation between the pulmonary arteries and the veins, between the pulmonary arteries and the trachea, and between the pulmonary veins and the trachea, are essential for the use of the porcine animal model, enabling an intraluminal ablation procedure of the pulmonary veins in swine.

Materials and Methods

Anatomical dissection

The anatomical examination was performed on cardiopulmonary sets of healthy pigs of 20–40 kg used for other studies and of pigs of approximately 100 kg obtained from the slaughterhouse (Table 1). The fresh cardiopulmonary sets were incised from the left ventricle into the left atrium to get an overview of the draining area of the pulmonary veins at the roof of the left atrium. Subsequently, the ostia and the pulmonary veins were incised longitudinally to provide an overview of the more distally located branches and to determine both the branching pattern of the pulmonary veins and their drainage area.

Silicone casting

Before casting the pulmonary veins with silicone, the left atrioventricular orifice was occluded by placing a clamp at the level of the coronary groove. Thereafter, both components of the silicone

Chapter 3: The pulmonary veins of the pig

(base and catalyst, ratio 1:1) were combined and coloured by adding the desired dye to indicate different structures. All pulmonary veins were casted with blue silicone [ZA22, 22 ShA silicone (type 1); Zhermack, Badia Polesine, Italy], unless indicated otherwise, whereas the pulmonary arteries and bronchi were, respectively, casted with red and white silicone [HT33, 30 ShA silicone (type 2); Zhermack]. Subsequently, the silicone was injected into the left atrium, and the pulmonary veins through a blunt needle inserted through the left auricular wall. Three cardiopulmonary sets (from two pigs of approximately 100 kg and one piglet of 20 kg) were casted with silicone (250 and 75 ml of both components for the large and smaller pigs, respectively) to visualize the pulmonary veins.

Number of cardiopulmonary sets Pig weight used Anatomical dissection 57 20-40 kg and 100 kg Silicone casting of the 3 100 kg and 20 kg pulmonary veins Silicone casting of the 2 35 kg pulmonary veins and arteries, trachea Silicone casting of the 1 30 kg pulmonary veins and arteries, trachea, oesophagus and aorta Silicone casting of complete 2 20 kg hearts Silicone casting of the 2 30 kg pulmonary veins in relation to the phrenic nerves Total number of used 67 cardiopulmonary sets Table 1. Number of cardiopulmonary sets used in the different techniques and for the various silicone casts.

The pulmonary trunk was transversely disconnected from the right ventricle by placing a clamp near the arterial cone to be able to cast the pulmonary arteries. Subsequently, silicone was injected through a blunt needle placed in the pulmonary trunk near the applied clamp. The trachea was casted by injection of silicone through a blunt needle placed in the trachea approximately 5 cm cranial to the origin of the tracheal bronchus, after the trachea was transversely clamped just cranial to the injection spot. To visualize the position of the oesophagus in relation to the pulmonary veins and the other adjacent structures, a plastic tube of 20 cm was inserted in the oesophagus. Cardiopulmonary sets of pigs of 35 kg were casted to demonstrate the pulmonary arteries (100 ml of both components), the pulmonary veins (100 ml of both components) and the trachea and bronchi (150 ml of both components). A silicone cast of a cardiopulmonary set of a pig of 30 kg was made to visualize the pulmonary veins (100 ml of both components), pulmonary arteries (100 ml of both components), trachea and bronchi (150 ml of both components) and the oesophagus and aorta. The

Chapter 3: The pulmonary veins of the pig cardiopulmonary sets of pigs of 30 kg were casted in situ with silicone (100 ml of both components) to visualize the ostia of the pulmonary veins in relation to the phrenic nerves after opening the thorax bilaterally. The position of both phrenic nerves was determined by removing the pericardium and the walls of the left auricle and the pulmonary veins surrounding the hardened silicone.

To localize the position of both ostia on the heart base and in relation to the other arterial and venous structures, cardiopulmonary sets of pigs of approximately 20 kg, suspended at the apex of the heart, were completely casted by pouring red silicone (type 2, 100 ml of both components) into the right ventricle and blue silicone (type 2, 100 ml of both components) into the left ventricle after clamping the aorta and the cranial and caudal vena cava.

After hardening overnight at room temperature, the casted cardiopulmonary sets were macerated in 25% potassium hydroxide during approximately 5 days and then rinsed in running tap water for 24 h.

Histology

To get an idea of the presence and the relative length of the myocardial sleeves, a reference point was defined by applying India ink on the luminal side of the wall of fresh pulmonary vein tissue at the level of the atrial-pulmonary junction of a pig of 100 kg, after which the myocardial sleeve length was measured on 5-µm-thick H&E-stained histological sections of 4% formalin-fixed (12 h) paraffin- embedded pulmonary vein samples. The distance was measured between the India ink spot and the most distal point of the myocardial sleeve on the histological sections.

Immunohistochemistry

The myocardial sleeve of the pulmonary vein draining the right middle lung lobe was stained immunohistochemically with the polyclonal myosin marker MYBPC3 (K-16: sc-50115; Santa Cruz Biotechnology, Santa Cruz, CA, USA) for the detection of the myosin-binding protein C (the cardiac type), which is reactive with porcine tissue and is present in the myofibrils of cardiac and other striated muscle tissue (Winegrad, 1999). The slides were immunostained using the Dako automated Autostainer Plus. No antigen retrieval was carried out. The sections were first incubated for 5 min with 3% hydrogen peroxide and 30 min with rabbit serum. After primary antibody incubation for 60 min, secondary antibody (rabbit/anti-goat, biotinylated, polyclonal, 1:500; Dako, Glostrup, Denmark) was applied for 30 min followed by streptavidin and horseradish peroxidase (streptavidin-HRP, 1:1500, Dako) for 30 min. Visualization was done with DAB (Dako) for 5 min.

To visualize the damage caused by heating myocardial tissue, mimicking ablation, immunohistochemical staining was performed, because this is almost impossible to demonstrate on

Chapter 3: The pulmonary veins of the pig general histology performed immediately after the procedure (own observations). Three-cm-thick samples of fresh myocardial tissue were unilaterally heated for 3 min on a hot plate on one side until the heated side was discoloured, subsequently fixated in formalin (4%) for 12 h and then embedded in paraffin. The histological sections were stained with the MYBPC3 marker (1:100). The same protocol was followed as described above, except for the application of a 1:500 streptavidin dilution as tertiary reagent.

Results

Pulmonary veins

The branching pattern

In all hearts examined, all pulmonary veins drained into the left atrium through two distinct ostia. An ostium is being defined as the common draining orifice of a number of pulmonary veins into the left atrium, while the common terminal part of the pertaining veins is referred to as the antrum of the respective ostium. The antrum therefore collects the venous blood of a set of pulmonary veins and discharges into the left atrium through the ostium.

One principal branching pattern (type I) was observed in 34 of 57 cases (Figs 1,3 and 7). In this type, the pulmonary veins from the right caudal and left caudal lung lobes (resp. V4 and V5) debouch together in a venous antrum which drains through ostium I. At their point of convergence, or slightly more peripheral in one of either veins, the orifice of the vein from the accessory lobe (V3) of the right lung can be located (highly variable pattern). The orifice of the pulmonary veins from the left cranial lobe (V7) is situated in the left wall of the antrum leading to ostium I. By dissection, it was seen that myocardial tissue extends into V7. The pulmonary veins draining the right cranial and right middle lung lobes (resp. V1 and V2) debouch together in a venous antrum which drains into the left atrium through ostium II.

This general branching pattern of the pulmonary veins presented some variations (Fig. 2). In a single case, V2 debouched together with V4 into the common venous antrum, together with V5, leading to ostium I (Fig. 1 type VI). In one other case, V1 and V2 drained separately into the left atrium at the level of ostium II (Fig. 1 type VII). The orifice of V3 was most variable as shown in Fig. 1 (types II (1 of 57 cases), III (14 of 57 cases), IV (3 of 57 cases) and V (2 of 57 cases)). The draining pattern of the left caudal lung lobe showed some variation as well. The different parts of the left caudal lung lobe can incidentally be drained not through a common vein but instead directly through two separate

Chapter 3: The pulmonary veins of the pig pulmonary veins (V5a and V5b) which opened together with V4, V6, V7 into the venous antrum leading to ostium I (type VIII, 1 of 57 cases). The principal branching pattern is illustrated in Fig. 3.

Topography of the pulmonary ostia

The lungs of the pig can be defined on the basis of the different lung lobes. The right lung is divided into a cranial, an intermediate and a caudal lung lobe. The left lung consists of a cranial and a caudal lung lobe. The position of the different pulmonary veins draining the corresponding lung lobes is visualized in Fig. 3.

The position of both pulmonary vein ostia in relation to their surrounding structures is visualized in Figs 4–7. Both ostia (I and II) are located ventral to the pulmonary arteries. From a dorsal view of the heart, the right pulmonary artery separates antrum I from antrum II (Fig. 4b). The larger pulmonary ostium, referred to as ostium I, is adjacent to and separated from the left auricle by the left azygos vein (Fig. 4b and e), while the caudal vena cava is located caudally and on its right side (Figs 4c and d). Ostium II is located more cranially towards the septal side of the left atrium (Fig. 4a), just above the oval fossa, at the level of the intervenous tubercle (Fig. 4b).

Fig. 1. Left image: schematic drawing of the branching pattern of the pulmonary veins (cranial view). Middle image: equivalent intra-atrial view of the left atrium. Right image: overview of the branching pattern of the pulmonary veins and the corresponding lung lobes (ventrocaudal view). For abbreviations, see Fig. 7.

Chapter 3: The pulmonary veins of the pig

Fig. 2. Different branching patterns of the pulmonary veins (cranial view). Type I: 32 of 57 cases (V7 single) and 2 of 57 cases (V7 double); types II, III, IV and VIII: respectively 1, 14, 3 and 1 cases of 57 cases (V7 single); types VI and VII: both 1 of 57 cases (V7 double); type V: 2 of 57 cases (V7 single or double). For abbreviations, see Fig. 7.

Fig. 3. Cranial view of a silicon cast of the porcine pulmonary veins (left image) and schematic drawing of the pulmonary veins (right image). For abbreviations, see Fig. 7.

Fig. 4. Different views on silicone casts of the porcine heart and the major vessels (a, b, c, d, e) and schematic representation (b). (a) Right lateral cranio-dorsal view. (b) Dorsal view. (c) Right lateral view. (d) Right caudo- lateral view. (e) Caudal view. (1) Right pulmonary artery; (1′) left pulmonary artery; (2) pulmonary veins draining into ostium I; (3) pulmonary veins draining into ostium II; (4) cranial vena cava; (5) aorta; (6) right auricle; (7) left auricle; (8) caudal vena cava; (9) pulmonary trunk; (10) left azygos vein.

Chapter 3: The pulmonary veins of the pig

The venous ostium I is located ventral to the left pulmonary artery and more distally, the antrum slightly deflects to the right side of the left pulmonary artery (Figs 5 and 6). The bifurcation of this antrum into V4 and V5 is situated ventral to the left principal bronchus. V6 is situated ventral to the left pulmonary artery, cranial to the bronchus of the left caudal lung lobe and caudal to the bronchi of the cranial and caudal parts of the left cranial lung lobe. V7, on the other hand, is located cranial to the latter bronchi (Fig. 7). The venous ostium II is situated ventral to the origin of the pulmonary artery of the cranial right lung lobe (A1) (Figs 5–7). Its corresponding venous antrum is located between this artery (A1) and the more caudal pulmonary artery irrigating the right middle (A2), accessory (A3) and caudal (A4) lung lobes (Figs 5–7).

After giving off the bronchus trachealis to the right cranial lung lobe, the trachea bifurcates into the primary bronchi at the level of the heart base, dorsal to the left atrium and to the right side of the median plane. The trachea is located dorsal to and on the right side of the pulmonary trunk. Both pulmonary veins ostia are located at the left side of the trachea, ostium II being situated closer to the trachea as compared to ostium I. Similar to the relative position of the trachea, the bronchi are also located dorsal to the pulmonary arteries, except in three places specified below (Fig. 7).

1. A branch of the right primary bronchus bends ventrally and crosses the right pulmonary artery ventrally, just cranial to the separation between the pulmonary arteries of the right accessory (A3) and middle (A2) lung lobes. Subsequently, the right pulmonary artery continues into A4.

2. The left primary bronchus divides into two large branches. The cranial branch enters the cranial part of the left cranial lobe, while the caudal branch ramifies in the caudal part of the left cranial lung lobe and the left caudal lung lobe. The caudal branch of the left bronchus is located ventral and between the pulmonary arteries, which irrigate the caudal part of the left cranial (A6) lung lobe and the left caudal (A5) lung lobe. This caudal bronchial branch is ventrally accompanied by V6.

3. The cranial branch of the left bronchus crosses the left pulmonary artery ventrally, just cranial to the origin of the pulmonary artery (A7) irrigating the cranial part of the cranial left lung lobe.

Chapter 3: The pulmonary veins of the pig

Fig. 5. (left image) Ventral view of a silicon cast of the lungs showing the pulmonary veins (blue), both ostia, pulmonary trunk (red) and trachea (white). Fig. 6. (right image) Ventral view of a silicone cast of the lungs showing both ostia (I, II) and right cranial (V1) and right middle (V2) pulmonary veins.

Fig. 7. Ventral view of the lung showing the pulmonary veins (blue), pulmonary trunk (red) and trachea (white) with abbreviations of the pulmonary veins, arteries and bronchi.

The position of the oesophagus in relation to the surrounding structures is demonstrated by the silicone cast presented in Fig. 8a and b. In the caudal part of the neck, the oesophagus is situated slightly to the left side of the trachea. More caudally, it is situated completely dorsal to the trachea in the mediastinum until the level of the tracheal bifurcation. At this point, the oesophagus is located to the right side of the aorta.

Chapter 3: The pulmonary veins of the pig

Both pulmonary veins ostia are separated from the oesophagus by the pulmonary trunk and the trachea. To the right of the venous ostium II, the most adjacent structure in ventral direction is the sinus venarum cavarum. While dissecting in dorsal direction, there can be passed between A1 and the pulmonary artery leading to A2, A3 and A4. Following this path, the trachea perfectly shields the oesophagus, which lies on the dorsal side of the trachea. To the right of the venous ostium I the caudal vena cava, while to the left side, the left azygos vein is reached. In dorsal direction, on the other hand, the passage is partially shielded by the left pulmonary artery on the right side, while the aorta and oesophagus are located more dorsally and the left primary bronchus is encountered more caudally. This indicates that intravascular ablation in both venous ostia in the pig is not likely to induce atriooesophageal fistulae, although ostium II may constitute more risks compared with ostium I (Figs 5, 6, 8a and b).

The position of the phrenic nerves in relation to the pericardium at the level of the left and right auricle is shown in Fig. 9. Both nerves lie dorsal to the auricles and the level of the ostia. The left phrenic nerve crosses ostium I, which might include a potential risk of causing phrenic nerve injuries during intra- and extra-ostial ablation procedures at this level (Fig. 9a–d). The right phrenic nerve lies lateral to the cranial and caudal vena cava (Fig. 9e and f). The right phrenic nerve is not lying adjacent to the pulmonary ostia, but the nerve crosses ostium I, ostium II, V1 and V2 ventrally and the right atrium dorsally, which might compromise an extra-ostial ablation procedure (Fig. 9g and h).

Fig. 8. Overview of the casted porcine lungs. (a) Ventral view of the casted porcine lungs. (b) Dorsal view of the casted porcine lungs.

Chapter 3: The pulmonary veins of the pig

Fig. 9. Overview of the topography of the phrenic nerves in relation to the pulmonary veins. (a and b) Left view of the topography of the left phrenic nerve (indicated in yellow) before (left image) and after (right image) fenestration of the pericardium. (c and d) Left view of the topography of the left phrenic nerve (indicated in yellow) in relation to casted pulmonary veins with ostium I. (e, f, g and h) Right view of the topography of the right phrenic nerve (indicated in yellow) before (e) and after (f–h) fenestration of the pericardium in relation to the casted pulmonary veins (g and h).

Chapter 3: The pulmonary veins of the pig

Myocardial sleeve

Myocardial sleeves were localized in both antra. The myocardial sleeve in the wall of the venous antrum draining through ostium II is visualized in Fig. 10. The measurement indicates a relative length of the myocardial sleeve of approximately 13 mm at this sampling spot.

The immunohistochemical staining of myocardial sleeve tissue clearly distinguishes the cardiac muscle tissue from the surrounding smooth muscle tissue and connective tissue (Fig. 11a). The staining of the ex-vivo-heated myocardial tissue (referring to ablation procedures) with the myosin marker demonstrated the disappearance of the myosin fibres caused by the high temperatures at the heated side of the sample (Fig. 11b). In contrast, the non-heated tissue of the sample was still clearly stained, indicating that the myosin fibres were still present (Fig. 11c).

Fig. 10. Histologic section of the myocardial sleeve of a left pulmonary vein (HE staining).

Fig. 11. Immunohistochemical staining of myocardial sleeve tissue. (a) Myocardial sleeve (arrow) of a pulmonary vein. (b) Heated myocardial tissue. (c) Non-heated myocardial tissue. Discussion

A complication incidence after ablation procedures for atrial fibrillation of 4.5% has been reported by Anderson (2012) in 16309 patients studied between 2003 and 2006. Catheter manipulations can cause several complications, of which the majority is not directly caused by the delivery of radiofrequency energy (Angkeow and Calkins, 2001). Apart from the complications, such procedures

Chapter 3: The pulmonary veins of the pig can fail and cause conduction recurrence. Conduction recurrence may occur when the targeted pulmonary veins are not completely isolated, when the induced lesions are not completely transmural or when the pathological conduction is caused by stimuli which originate from ectopic foci located outside the pulmonary veins (Haïssaguerre et al., 1996, 1998; Chen et al., 1999; Hsieh et al., 1999; Pappone et al., 2000; Cappato et al., 2003; Gerstenfeld et al., 2003; Nanthakumar et al., 2004). An easily executable, fast ablation procedure with a high success rate, which can be repeated with low impact on the patient, if necessary, would be a big leap forward in terms of quality of the surgical technique and applicability in various patients. The continuous search for new ablation techniques requires appropriate animal models, in which the left atrium-pulmonary vein junction is documented in detail. In this investigation, we described all important characteristics of the pulmonary veins and their ostia, including the topography and the branching pattern of the pulmonary veins, the histological structure of the myocardial sleeves and their destruction as visualized by immunochemistry after heating.

Because of the adverse effects that may occur after an ablation procedure in man, this article has focused mainly on the risk for atrio-oesophageal fistulae and phrenic nerve injuries, because these are two of the most important complications. This emphasizes the need of having information about such possible risks in pigs.

While the human oesophagus covers the posterior side of the left atrium (Ho et al., 2012), the porcine oesophagus passes dorsal to the base of the heart, hereby separated from the pulmonary veins ostia by the pulmonary trunk and the trachea. While Barone (1997) already discussed in detail the topography of the porcine oesophagus and trachea, the present paper elaborates on their positions relative to the ostia of the pulmonary veins. In 40% of human cases, the distance between the oesophagus and the endocardium is <5 mm (Sanchez-Quintana et al., 2005) and the oesophagus and the left atrium are in close contact with each other, as the oesophagus courses from the left superior side to the right inferior side in 36% of the cases (Lemola et al., 2004). It must be taken into account that the oesophagus is mobile, which also applies to the pig, and its position can change from being close to the left posterior side to the right posterior side of the left atrium (Lemola et al., 2004; Kottkamp et al., 2005). Mönnig et al. (2005) indicated that the oesophagus was located nearest to the left superior pulmonary vein, while its proximity to the left superior pulmonary vein ostium is highly variable. It was concluded that in most cases, the oesophagus is positioned very close to the left pulmonary veins ostia. In the case of pigs, the present findings show that ostium II is located closer to the oesophagus and thus would constitute a greater risk as compared to ostium I when an ablation is performed.

Chapter 3: The pulmonary veins of the pig

The position of the phrenic nerves in relation to the heart is variable in man. The left phrenic nerve passes the left ventricle over the anterior (18% of the cases), lateral (59% of the cases) or posteroinferior (23% of the cases) side (Sanchez-Quintana et al., 2009). The distance between this nerve and the roof of the left auricle is <4 mm in 9% of the cases (Sanchez-Quintana et al., 2009). The right phrenic nerve is located anterior to the right pulmonary veins, where it is positioned nearer to the right superior pulmonary veins (Sanchez-Quintana et al., 2005). In pigs, both phrenic nerves are located dorsal to the auricles. In this context, ablation of ostium I involves a greater risk as the left phrenic nerve passes very close to the left atrio–venous junction.

Anatomical investigation and silicone casting

Based on anatomical dissection and by silicone casting of porcine cardiopulmonary organs, we were able to determine the different branching patterns of the pulmonary veins. It was found that the right cranial and right intermediate lung lobes are drained into a smaller ostium, which is located closer to the interatrial septum, whereas the other lung lobes are drained into a larger, more caudally situated ostium. On the silicone casts, three levels could be distinguished, namely the venous vasculature ventrally, the trachea and the primary bronchi dorsally, and the arterial vasculature in between. In three specific places, a different pattern is noted: the bronchus of the right intermediate lung lobe crosses the arterial branches of the same lung lobe ventrally. A similar pattern is seen at the point where the principal bronchus of the left lung crosses the pulmonary artery, which irrigates the left caudal lung lobe. The presence of two ostia in pigs, previously also reported by Crick et al. (1998), makes an ablation procedure simpler as compared to humans where the number of orifices amount to four (in 71% of the cases) or five (in 17% of the cases) (Ho et al., 2012). This feature together with the size of the heart and the pulmonary vein ostia, which are similar to their human counterparts, all support the importance of the pig as an animal model for medical research.

Histological structure of the myocardial sleeve

In humans, the incidence of myocardial sleeves in the pulmonary veins ranges between 68 and 88.8% (Kholova and Kautzner, 2003; Steiner et al., 2006). Nathan and Gloobe (1970) mention the presence of sleeves originating from the atrial myocardium inside the pulmonary vein walls in humans but also in a variety of mammals such as sheep, cattle, pigs and a horse. The presence of myocardial sleeves in pigs was confirmed and documented in the present study. The distance of a myocardial sleeve, measured on a histological section, is relative due to the shrinkage during tissue fixation. The length of the myocardial sleeve of both ostia was not compared with each other in this study. In humans,

Chapter 3: The pulmonary veins of the pig the left superior pulmonary vein contains a longer sleeve as compared to the left inferior pulmonary vein (Ho et al., 2012).

Immunohistochemical staining of the myocardial sleeve

Identifying the myocardial sleeve on histological sections with H&E staining is hampered due to the difficult distinction between smooth and cardiac muscle tissue. Little striation is seen when a myocardial sleeve is cut transversely on a longitudinal section of the pulmonary vein wall. Therefore, an immunohistochemical staining was performed to distinguish the smooth and cardiac muscle tissue and to prove the presence of myocardial sleeves in the pulmonary veins in pigs. The immunohistochemical staining was also used to prove the disappearance of myosin fibres during ex- vivo heating, as this cannot be visualized using H&E staining. After in-vivo ablation procedures, the penetration of the heat is easier to visualize if, as a part of the subsequent healing process, connective scar tissue is formed. In the present study, myosin destruction was determined because Tornberg (2005) suggested that during a heating process, this feature is the first transition in muscle tissue subjected to heating and occurs between about 50 and 60°C. (Martens and Vold, 1976; Wright et al., 1977).

Chapter 3: The pulmonary veins of the pig

References

Ahsan S.Y., Flett A.S., Lambiase P.D. and Segal OR (2010). First report of phrenic nerve injury during pulmonary vein isolation using the Ablation Frontiers pulmonary vein ablation catheter. J Interv Card Electrophysiol. 29:187–190.

Anderson K.P. (2012). Ablation for atrial fibrillation: what are acceptable levels of experience, efficacy, and complications? J Interv Card Electrophysiol. 34:125–128.

Andrie R.P., Schrickel J.W., Nickenig G. and Lickfett L. (2012). Left phrenic nerve injury during cryoballoon ablation of the left superior pulmonary vein. Pacing Clin Electrophysiol. 35:334–336.

Angkeow P. and Calkins H.G. (2001). Complications associated with radiofrequency catheter ablation of cardiac arrhythmias. Cardiol Rev. 9:121–130.

Arentz T., Jander N., von Rosenthal J., Blum T., Furmaier R., Gornandt L., Neumann F.J. and Kalusche D. (2003). Incidence of pulmonary vein stenosis 2 years after radiofrequency catheter ablation of refractory atrial fibrillation. Eur Heart J. 24:963–969.

Barker E., Haverson K., Stokes C.R., Birchall M. and Bailey M. (2006). The larynx as an immunological organ: immunological architecture in the pig as a large animal model. Clin Exp Immunol. 143:6–14.

Barone R. (1997). Tome 3, Splanchnologie I. Anatomie Comparée des Mammifères Domestiques. Paris: Editions Vigot, pp. 853.

Bermejo A., O. Gonzalez O., and Gonzalez J.M. (1993). The pig as an animal model for experimentation on the temporomandibular articular complex. Oral Surg Oral Med Oral Pathol. 75:18–23.

Brugada R., Tapscott T., Czernuszewicz G.Z., Marian A.J., Iglesias A., Mont L., Brugada J., Girona J., Domingo A., Bachinski L.L. and Roberts R. (1997). Identification of a genetic locus for familial atrial fibrillation. N Engl J Med. 336:905–911.

Brunet B., Doucet C., Venisse N., Hauet T., Hebrard W., Papet Y., Mauco G. and Mura P. (2006). Validation of Large White Pig as an animal model for the study of cannabinoids metabolism: application to the study of THC distribution in tissues. Forensic Sci Int. 161:169–174.

Chapter 3: The pulmonary veins of the pig

Bunch T.J., Bruce G.K., Mahapatra S., Johnson SB., Miller D.V., Sarabanda A.V., Milton M.A. and Packer D.L. (2005). Mechanisms of phrenic nerve injury during radiofrequency ablation at the pulmonary vein orifice. J Cardiovasc Electrophysiol. 16:1318–1325.

Cappato R., Negroni S., Pecora D., Bentivegna S., Lupo PP, Carolei A., Esposito C., Furlanello F. and De Ambroggi L. (2003). Prospective assessment of late conduction recurrence across radiofrequency lesions producing electrical disconnection at the pulmonary vein ostium in patients with atrial fibrillation. Circulation. 108:1599–1604.

Crick S.J., Sheppard M.N., Ho S.Y., Gebstein L. and Anderson R.H. (1998). Anatomy of the pig heart: comparisons with normal human cardiac structure. J Anat. 193:105–119.

Delange J., Vaneck P., Elliott G.R., Dekort W.L.A.M. and Wolthuis O.L. (1992). The Isolated Blood- Perfused Pig Ear – an Inexpensive and Animal-Saving Model for Skin Penetration Studies. J Pharmacol Toxicol Methods. 27:71–77.

Gerstenfeld E.P., Callans D.J., Dixit S., Zado E. and Marchlinski F.E. (2003). Incidence and location of focal atrial fibrillation triggers in patients undergoing repeat pulmonary vein isolation: implications for ablation strategies. J Cardiovasc Electrophysiol. 14:685–690.

Gillinov A.M., Pettersson G. and Rice T.W. (2001). Esophageal injury during radiofrequency ablation for atrial fibrillation. J Thorac Cardiovasc Surg. 122:1239–1240.

Glenny R.W., Bernard S.L., Luchtel D.L., Neradilek B. and Polissar N.L. (2007). The spatial-temporal redistribution of pulmonary blood flow with postnatal growth. J Appl Physiol. 102:1281–1288.

Gregg P.J., Walder D.N. and Rannie I. (1980). Caisson disease of bone: a study of the Gottingen minipig as an animal model. Br J Exp Pathol. 61:39–54.

Haïssaguerre M., Jais P., Shah D.C., Gencel L., Pradeau V., Garrigues S., Chouairi S., Hocini M., Le Metayer P., Roudaut R. and Clementy J. (1996). Right and left atrial radiofrequency catheter therapy of paroxysmal atrial fibrillation. J Cardiovasc Electrophysiol. 7:1132-1144.

Chapter 3: The pulmonary veins of the pig

Haïssaguerre M., Jais P., Shah D.C., Takahashi A., Hocini M., Quiniou G., Garrigue S., Le Mouroux A., Le Metayer P. and Clementy J. (1998). Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J of Med. 339:659-666.

Haïssaguerre M., Shah D.C., Jais P., Hocini M., Yamane T., Deisenhofer I., Chauvin M., Garrigue S. and Clementy J. (2000). Electrophysiological breakthroughs from the left atrium to the pulmonary veins. Circulation. 102:2463-2465.

Ho S.Y., Cabrera J.A. and Sanchez-Quintana D. (2012). Left atrial anatomy revisited. Circ Arrhythm Electrophysiol. 5:220-228.

Hotchkiss J.R., Sanders M.H., Clermont G. and Crooke P.S. (2007). Preventing “bored-lung disease” when treating patients with ventilatory failure. Crit Care Med. 35:1797-1799.

Hsieh M.H., Chen S.A., Tai C.T., Tsai C.F., Prakash V.S., Yu W.C., Liu C.C., Ding Y.A. and Chang M.S. (1999). Double multielectrode mapping catheters facilitate radiofrequency catheter ablation of focal atrial fibrillation originating from pulmonary veins. J Cardiovasc Electrophysiol. 10:136-144.

Jones R.D., Stuart B.P., Greufe N.P. and Landes A.M. (1999). Electrophysiology and pathology evaluation of the Yucatan pig as a non-rodent animal model for regulatory and mechanistic toxicology studies. Lab Anim. 33:356-365.

Kannel W.B., Wolf P.A., Benjamin E.J. and Levy D. (1998). Prevalence, incidence, prognosis, and predisposing conditions for atrial fibrillation: Population-based estimates. Am J Cardiol. 82:2N-9N.

Khatri M., Dwivedi V., Krakowka S., Manickam C., Ali A., Wang L., Qin Z., Renukaradhya G.J. and Lee C.W. (2010). Swine influenza H1N1 virus induces acute inflammatory immune responses in pig lungs: a potential animal model for human H1N1 influenza virus. J Virol. 84:11210-11218.

Kholova I. and Kautzner J. (2003). Anatomic characteristics of extensions of atrial myocardium into the pulmonary veins in subjects with and without atrial fibrillation. Pace. 26:1348-1355.

Chapter 3: The pulmonary veins of the pig

Kottkamp H., Piorkowski C., Tanner H., Kobza R., Dorszewski A., Schirdewahn P., Gerds-Li J.H. and Hindricks G. (2005). Topographic variability of the esophageal left atrial relation influencing ablation lines in patients with atrial fibrillation. J Cardiovasc Electrophysiol. 16:146-150.

Lee B.K., Choi K.J., Kim J., Rhee K.S., Nam G.B. and Kim Y.H. (2004). Right phrenic nerve injury following electrical disconnection of the right superior pulmonary vein. Pace. 27:1444-1446.

Lemola K., Sneider M., Desjardins B., Case I., Han J., Good E., Tamirisa K., Tsemo A., Chugh A., Bogun F., Pelosi F. Jr., Kazerooni E., Morady F. and Oral H. (2004). Computed tomographic analysis of the anatomy of the left atrium and the esophagus: implications for left atrial catheter ablation. Circulation. 110:3655-3660.

Martens H. and Vold E. (1976). DSC studies of muscle protein denaturation. In Proceedings of the 22nd European meeting of meat research workers, Malmö, Sweden (p. J 9.3).

Mohr F.W., Fabricius A.M., Falk V., Autschbach R., Doll N., Von Oppell U., Diegeler A., Kottkamp H. and Hindricks G. (2002). Curative treatment of atrial fibrillation with intraoperative radiofrequency ablation: short-term and midterm results. J Thorac Cardiovasc Surg. 123:919-927.

Mönnig G., Wessling J., Juergens K.U., Milberg P., Ribbing M., Fischbach R., Wiekowski J., Breithardt G. and Eckardt L. (2005). Further evidence of a close anatomical relation between the oesophagus and pulmonary veins. Europace. 7:540-545.

Nanthakumar K., Plumb V.J., Epstein A.E., Veenhuyzen G.D., Link D. and Kay G.N. (2004). Resumption of electrical conduction in previously isolated pulmonary veins - Rationale for a different strategy? Circulation. 109:1226-1229.

Nathan H. and Gloobe H. (1970). Myocardial atrio-venous junctions and extensions (sleeves) over the pulmonary and caval veins. Anatomical observations in various mammals. Thorax. 25:317-324.

Okumura Y., Henz B.D., Bunch T.J., Dalegrave C., Johnson S.B. and Packer D.L. (2009). Distortion of right superior pulmonary vein anatomy by balloon catheters as a contributor to phrenic nerve injury. J Cardiovasc Electrophysiol. 20:1151-1157.

Okumura Y., Kolasa M.W., Johnson S.B., Bunch T.J., Henz B.D., O'Brien C.J., Miller D.V. and Packer D.L. (2008). Mechanism of tissue heating during high intensity focused ultrasound pulmonary vein isolation: implications for atrial fibrillation ablation efficacy and phrenic nerve protection. J Cardiovasc Electrophysiol. 19:945-951.

Chapter 3: The pulmonary veins of the pig

Pappone C., Oral H., Santinelli V., Vicedomini G., Lang C.C., Manguso F., Torracca L., Benussi S., Alfieri O., Hong R., Lau W., Hirata K., Shikuma N., Hall B. and Morady F. (2004). Atrio-esophageal fistula as a complication of percutaneous transcatheter ablation of atrial fibrillation. Circulation. 109:2724-2726.

Pappone C., Oreto G., Lamberti F., Vicedomini G., Loricchio M.L., Shpun S., Rillo M., Calabro M.P., Conversano A., Ben-Haim S.A., Cappato R. and Chierchia S. (1999). Catheter ablation of paroxysmal atrial fibrillation using a 3D mapping system. Circulation. 100:1203-1208.

Pappone C., Rosanio S., Oreto G., Tocchi M., Gugliotta F., Vicedomini G., Salvati A., Dicandia C., Mazzone P., Santinelli V., Gulletta S. and Chierchia S. (2000). Circumferential radiofrequency ablation of pulmonary vein ostia: A new anatomic approach for curing atrial fibrillation. Circulation. 102:2619- 2628.

Paterson R.F., Lingeman J.E., Evan A.P., Connors B.A., Williams J.C.J.R. and Mcateer J.A. (2002). Percutaneous stone implantation in the pig kidney: a new animal model for lithotripsy research. J Endourol. 16:543–547.

Perry Y., Epperly M.W., Fernando H.C., Klein E., Finkelstein S., Greenberger J.S. and Luketich J.D. (2005). Photodynamic therapy induced esophageal stricture–an animal model: from mouse to pig. J Surg Res. 123,67–74.

Robbins I.M., Colvin E.V., Doyle T.P., Kemp W.E., Loyd J.E., McMahon W.S. and Kay G.N. (1998). Pulmonary vein stenosis after catheter ablation of atrial fibrillation. Circulation. 98:1769–1775.

Rogers C.S., Abraham W.M., K.A. Brogden K.A., Engelhardt J.F., Fisher J.T., Mccra P.B., Mclennan G., Meyerholz D.K., Namati E., Ostedgaard L.S., Prather R.S., Sabater J.R., Stoltz D.A., Zabner J. and Welsh M.J. (2008). The porcine lung as a potential model for cystic fibrosis. Am J Physiol Lung Cell Mol Physiol. 295:240–263.

Roux N., Havet E. and Mertl P. (2004). The myocardial sleeves of the pulmonary veins: potential implications for atrial fibrillation. Surg Radiol Anat. 26:285–289.

Rowan A.M., Moughan P.J., Wilson M.N., Maher K. and Tasmanjones C. (1994). Comparison of the Ileal and Fecal Digestibility of Dietary Amino-Acids in Adult Humans and Evaluation of the Pig as a Model Animal for Digestion Studies in Man. Br J Nutr. 71:29–42.

Chapter 3: The pulmonary veins of the pig

Sanchez-Quintana D., Cabrera J.A., Climent V., Farre J., Weiglein A. and Ho S.Y. (2005). How close are the phrenic nerves to cardiac structures? Implications for cardiac interventionalists. J Cardiovasc Electrophysiol. 16:309–313.

Sanchez-Quintana D., Ho S.Y., Climent V., Murillo M. and Cabrera J.A. (2009). Anatomic evaluation of the left phrenic nerve relevant to epicardial and endocardial catheter ablation: implications for phrenic nerve injury. Heart Rhythm. 6:764–768.

Shulman R.J., Henning S.J. and Nichols B.L. (1988). The miniature pig as an animal model for the study of intestinal enzyme development. Pediatr Res. 23:311–315.

Sommerer D., Suss R., Hammerschmidt S., Wirtz H., Arnold K. and Schiller J. (2004). Analysis of the phospholipid composition of bronchoalveolar lavage (BAL) fluid from man and minipig by MALDI-TOF mass spectrometry in combination with TLC. J Pharm Biomed Anal. 35:199–206.

Sonmez B., Demirsoy E., Yagan N., Unal M., Arbatli H., Sener D., Baran T. and Ilkova F. (2003). A fatal complication due to radiofrequency ablation for atrial fibrillation: atrio-esophageal fistula. Ann Thorac Surg. 76:281–283.

Steiner I., Hajkova P., Kvasnicka J. and Kholova I. (2006). Myocardial sleeves of pulmonary veins and atrial fibrillation: a postmortem histopathological study of 100 subjects. Virchows Arch. 449:88–95.

Tagawa M., Higuchi K., Chinushi M., Washizuka T., Ushiki T., Ishihara N. and Aizawa Y. (2001). Myocardium extending from the left atrium onto the pulmonary veins: a comparison between subjects with and without atrial fibrillation. Pace. 24:1459–1463.

Tornberg E. (2005). Effects of heat on meat proteins- Implications on structure and quality of meat products. Meat Sci. 70:493–508.

Vollmerhaus B., Roos H. and Troger U. (1999). Aa. and Vv. pulmonales of the miniature pig and a common view of their supportive function in the lung. Anat Histol Embryol. 28:143–149.

Wang S., Liu Y., Fang D. and Shi S. (2007). The miniature pig: a useful large animal model for dental and orofacial research. Oral Dis. 13:530–537.

Winegrad S. (1999). Cardiac myosin binding protein. C Circ Res. 84:1117–1126.

Wright D.J., Leach I.B. and Wilding P. (1977). Differential scanning calorimetric studies of muscle and its constituents. J Sci Food Agric. 28:557.

100

Chapter 4

Presence of ganglia and telocytes in proximity to myocardial sleeve tissue in the porcine pulmonary veins wall

Adapted from Vandecasteele T., Cornillie P., Vandevelde K., Logothetidou A., Couck L., van Loon G. and Van den Broeck W. (2017). Presence of Ganglia and Telocytes in Proximity to Myocardial Sleeve Tissue in the Porcine Pulmonary Veins Wall. Anat Histol Embryol. 46:325- 333.

101

102

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Summary

Ganglia and telocytes were identified inside the porcine pulmonary veins wall near myocardial sleeve tissue at the atrio-pulmonary junction. These structures are reported to play a role in the initiation of pulses from outside the heart, which potentially can cause cardiac conduction disorders such as atrial fibrillation (AF). In-depth knowledge on the fine structure of the pulmonary vein wall is a prerequisite to better understand the underlying pathophysiology of atrial fibrillation and the origin and conduction of ectopic pulses. The importance of pulmonary vein myocardial sleeves as triggering foci for atrial fibrillation has been shown in human patients. In this context, the fine structure of the pulmonary vein wall was investigated qualitatively by light and transmission electron microscopy in the pig, which is a frequently used animal model for development of new treatment strategies. Additionally, intra- and extramural ganglia, containing telocytes that create a network near the neuron cell bodies, were identified in pigs. Detailed illustration of the distribution and organization of tissues and cell types, potentially involved in the origin and propagation of ectopic stimuli originating from the pulmonary veins, might lead to a better insight on the actual composition of the tissues affected by ablation as studied in pigs.

103

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Introduction

Macroscopically, the draining pattern of the porcine pulmonary veins is well known (Vollmerhaus et al., 1999; Vandecasteele et al., 2015). Microscopically, on the other hand, little detailed information is available about the structure of the pulmonary veins wall of the pig, despite this species is commonly used as a model in cardiovascular research. The demand for detailed histological information on the pulmonary veins, more specifically on the atrio-pulmonary junction, is supported by the large numbers of clinical interventions that occur at this location, mostly in the context of atrial fibrillation (AF). These procedures are therapeutically performed in humans (Kneeland and Fang, 2009) and also in pigs and sheep for research purposes (Vandecasteele et al., 2016). To understand the consequences of such a procedure, it is prerequisite to identify the cell types and structures present inside the pulmonary veins wall.

In man, AF is the most common sustained cardiac arrhythmia with increasing prevalence at higher age and the most frequent arrhythmia in human clinical practice (Haïssaguerre et al., 1998; Haïssaguerre et al., 2007). Moreover, it is the major cause of cardioembolic cerebrovascular accidents (Haïssaguerre, 2010). The pathophysiology and causes of AF are very complex. Electrophysiological studies pointed out that the ectopic pulses that induced AF, originated from the myocardial tissue inside the pulmonary veins wall, indicated as myocardial sleeve. A myocardial sleeve is composed of myocardial fibers that spread out from the ostium of the pulmonary veins, distally into the wall of these veins, exhibiting different layers and patterns (Nathan and Gloobe, 1970; Haïssaguerre et al., 1998, 2006; Po et al., 2006; Katritsis et al., 2008). However, detailed morphological information of the pulmonary veins wall is mostly lacking, as previous research mainly focused on these myocardial sleeves, neglecting the presence of other cell types (Ho et al., 2001; Hamabe et al., 2003; Katritsis et al., 2008). Nowadays, AF is usually treated by ablation of the pulmonary veins (Xu et al., 2016). The consequent partial destruction of the pulmonary veins wall at the atrio-pulmonary junction causes degeneration of various cell types. In this context, the research question arises which other cell types, besides myocardial tissue, may be possibly affected due to this destruction process.

The presence of other cell types or histological structures, besides myocardial sleeve tissue was indicated before. Po et al., (2006) pointed out the presence of extramural autonomic ganglia located in fat pads at the base of the canine pulmonary veins. Besides these ganglia located at the base of the pulmonary veins, various authors proved the existence of a recently discovered cell type, called telocytes, in different tissues including the heart and the pulmonary veins, mainly in humans and rats

104

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

(Cretoiu et al., 2014). A telocyte is an interstitial stromal cell type found in different organs and tissues in a wide range of vertebrate species such as in the urinary tract of rats, in human placenta or in the eyes of mice (Popescu et al., 2010; Suciu et al., 2010; Zheng et al., 2012; Luesma et al., 2013; Cretoiu et al., 2014). Telocytes are distinguished from other interstitial cells by the presence of very long and slender prolongations called telopodes (Popescu et al., 2010). The presence of telopodes is a unique feature of telocytes in comparison with fibroblasts, which consist of thin and thicker sections, called podomers and podoms, respectively (Table 1). Popescu (2011) described shortly telocytes as ‘cells with telopodes’. Electron microscopy is the method of choice to identify telocytes and to differentiate them from other cell types, especially fibroblasts (Popescu and Nicolescu, 2013). These telocytes might influence the rhythmicity of the pulmonary veins and affect the conduction of stimuli towards the left atrium as well (Gherghiceanu et al., 2008; Morel et al., 2008) by intercellular signaling (Popescu and Faussone-Pellegrini, 2010). Telocytes would be involved in cell – cell connection and communication through their telopodes and secretion of microvesicles (Popescu and Faussone-Pellegrini, 2010).

105

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Celtypes

Cell organelles/ components Telocytes Fibroblasts Perineural cells

Cell body Small (< 2µm), spindle shaped Pleomorphic (phenotypic Very thin and flattened with

heterogenity) various shapes (rectangular to

hexagonal)

Cytoplasm Small amount Large amount

Nucleus One, oval/rod shaped One, oval Elongated nuclei

Chromatin 50% heterochromatin Typically euchromatin

Nucleolus Rarely visible 1-2 nucleoli

Membrane caveolae Several caveolae Few caveolae Numerous caveolae

Number of prolongations 2-5 telopodes Usually 2

Branching of the Dichotomic pattern, creating Randomly

threedimensional networks prolongations

Sprouting of the Thin Thick base, gradual thinning Thin and flat cytoplasmic

towards the tip processes prolongations out of the cell

body

Length of the prolongations Very long, tens of micrometers Usually some micrometers

Podomers Very thin (mostly below 0.2 µm) No No

Podoms Dilated parts of telopodes (0.5-1 No No

µm), containing caveolae,

mitochondria and ER (called Ca2+

uptake/release unit)

Table 1. Ultrastructural characteristics of telocytes and ﬁbroblasts. (Adapted from Popescu and Nicolescu (2013) and Inokuchi et al. (1991))

106

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Fig. 1. Schematic representation of different cell types such as mast cell, monocyte, lymphocyte, eosinophil, fibroblast, stem cell, telocyte with podomer and podom in relation to a blood vessel, nerve bundle and smooth muscle cell. (From Cretoiu et al., 2017).

The goal of the present research was to investigate qualitatively the presence of ganglia and telocytes inside the wall of the pulmonary veins base, on both light and electron microscopic level in pigs as an animal model. The research of the porcine pulmonary veins as an animal model is important for further development of atrial fibrillation treatment techniques (Vandecasteele et al., 2015) as possible ablation of these structures or cell types could lead to a variable outcome or recurrence rate.

Materials and Methods

Animals

For light microscopic analysis, four pigs (Landrace, body weight about 40 kg) were freshly sampled. Euthanasia was performed following the guidelines of the Ethical Committee of the Faculty of Veterinary Medicine, Ghent University, Belgium, but was not related to this study.

For transmission electron microscopic analysis, one pig (Landrace, 20 kg bodyweight), was freshly sampled after euthanasia following the guidelines of the Ethical Committee of the Faculty of Veterinary Medicine, Ghent University, Belgium, but was not related to this study.

107

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Sampling

For light microscopic analysis, the hearts were excised and the atriopulmonary junction was removed (Fig. 2). This junction was further reduced longitudinally in several samples which were fixed immediately by submersion in a 4% phosphate buffered formaldehyde solution for 48 hours.

For transmission electron microscopic analysis, the heart was removed and samples of 1cm2 were excised at the level of the atrio-pulmonary junction and fixed in Karnovsky’s fixative for 24 hours.

Fig. 2. Schematic drawing (cranioventral view) of the branching pattern of the pulmonary veins and the lung lobes. The samples were prelevated at the atrio- pulmonary junction, indicated by the oval marks.

Analysis

Light microscopy

After fixation, dehydration according to standard laboratory procedures and subsequent embedding in paraffin was performed using a Microm tissue processor STP 420D (Prosan, Merelbeke, Belgium) and Microm embedding station EC 350-1 (Prosan), respectively. Sections of 8 µm were cut longitudinally on a Microm H360 microtome (Prosan) and stained with haematoxylin and eosin (HE) according to Bancroft and Gamble (2002), with minor adaptations (6 minutes haematoxylin, 10 minutes running tap water, 20 times up/down in distilled water, 3 minutes eosin). Eventually, the sections were evaluated under the light microscope (Olympus BX61, Olympus DP73 camera, Olympus, Belgium).

108

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Transmission electron microscopy

After fixation, the samples were washed in 0.1M sodium cacodylate buffer after which the tissue was processed and embedded in paraffin, as ganglia are found more easily in paraffin sectioned tissue compared to semithin sections. Afterwards, tissue was sectioned at 5 µm and stained with HE. Once ganglia were found in the paraffin sections, the tissue was deparaffinized, postfixed with 1% OsO4 for 1 hour, dehydrated in series of alcohol and embedded in epon 812 (Leica EM TP, Leica Microsystems GmbH, Wetzlar, Germany). The epon samples were consequently trimmed according to the region of interest, containing ganglia. Semithin sections were cut (0.5-1 µm) on a Leica EMUC6 ultramicrotome (Germany), stained with toluidine blue and examined under the light microscope. Afterwards, ultrathin sections (80 nm) were cut, mounted on formvar coated grids, contrasted with 1% uranyl acetate and 1.33% lead citrate and observed under the Jeol JEM 1400 Plus transmission electron microscope (Jeol Ltd, Tokyo, Japan). The presence of telocytes was determined based on the characteristic properties as described by Popescu and Nicolescu (2013) and summarized in Table 1.

Results

Ganglia

HE staining of the pig pulmonary vein showed the presence of ganglia located in fat pads at the base of the vein (Fig. 3a) and intramurally within the venous wall which can either be partially surrounded by fat and connective tissue (Fig. 3b) or almost completely surrounded by cardiac muscle tissue (Fig. 3c). Myocardial, smooth muscle and connective tissue were surrounding the ganglia or were located nearby. The ganglia were located in close vicinity of the myocardial sleeve. The largest ganglia measured approximately 300 µm in length and 200 µm in diameter while the smaller ones measured approximately 200 µm in length and 100 µm in diameter. At certain locations, several ganglia were located near each other (Fig. 3a, b, c).

109

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

110

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Fig. 3. (a) fat pad near a myocardial sleeve in a porcine pulmonary vein (left side is towards the left atrium, and right side is towards the lungs), containing ganglia (indicated on the frame), HE staining; bar = 1000 µm, detail bar = 200 µm. 1: endothelium, 2: myocardial tissue, 3: fat tissue, 4: neurons. (b) Longitudinal section of a myocardial sleeve in a porcine pulmonary vein (bottom side is towards the left atrium, and top side is towards the lungs) with an intramural ganglion (indicated on the frame), HE staining; bar = 1000 µm, detail bar = 100 µm. 1: lumen, 2: endothelium, 3: myocardial tissue, 4: neurons. (c) Part of the roof of the porcine left atrium with a pulmonary vein (top side) containing ganglia (indicated on the frame), surrounded by fat and myocardial tissue, HE staining; bar = 1000 µm, detail bar = 100 µm. 1: endothelium, 2: myocardial tissue, 3: lumen, 4: neurons.

Telocytes

Telocytes were recognized as a cell type with very thin and long prolongations containing alternating thick and thin parts and a fine elongated cell body, lacking a basal lamina. These prolongations contain vesicles and caveolae, and prolongations of different telocytes may create a network with each other (Fig. 4). This cell type was found inside the base of the pig pulmonary vein wall. These cells were confirmed as telocytes in accordance with Popescu and Nicolescu (2013) (Table 1). Telocyte telopodes in close contact (< 20 nm) with the myocardial sleeve, in close vicinity with small veins or completely surrounded by collagen fibers were observed (Fig. 5, 6). This cell type was also discovered in two ganglia, located inside the wall of a pulmonary vein (Fig. 5, 7a-f), with the same characteristics as described by Popescu and Nicolescu (2013) (Table 1). Several telocytes telopodes located close to perineural cells were found in close vicinity with neuron cell bodies (Fig. 7a-f). Another cell type located inside the ganglion was characterized as fibroblasts (Fig. 4). The difference between telocytes and perineural cells, is visualized in figure 7a-f as telocytes lacked a basal lamina whereas perineural cells demonstrated a clear basal lamina.

111

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Fig. 4. (a) TEM image of a telocyte cell body (2) positioned with its telopode (1) close to cardiac muscle (3) of the myocardial sleeve of a pig pul-monary vein. Nucleus (6), bar = 1 µm. (b) detail of panel a: detailed image of a telocyte telopode (1) with vesicles (7) and caveolae (8). Nucleus of telocyte (6), myocardial sleeve tissue (3), bar = 200 nm. (c) TEM image of a telocyte cell body (2) with a telopode (1), which makes connection with a telopode of another cell (see box). Nucleus of telocyte (6), bar = 500 nm. (d) TEM image of a telocyte prolongation (1) inside a pig pulmonary vein wall, which is located very close to myocardial tissue (3), bar = 50 nm. The striation of the myocardial tissue is clearly visible, bar = 500 nm. Panel e: TEM image of a ﬁbroblast (10), with a large amount cytoplasm (9) surrounding the nucleus (6), located in a ganglion inside a pig pulmonary vein wall. Collagen (5), bar = 1 µm.

112

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Fig. 5. TEM image of two neurone cell bodies (1) inside a ganglion surrounded by Schwanncells (2) inside a pig pulmonary vein. Nerve tissue (3) containing Schwann cells and axons located close to the ganglion. Bar = 20 µm.

Fig. 6. (a) TEM image of two telocytes (3) with their telopodes (arrows) positioned in close vicinity to a vein (1). The insert demonstrates ﬁnger-like branches and enclosed vesicles (arrows) of a telopode, which are encompassed by an electron-dense telopode-like structure which closely follows the vein wall. Endothelial cell (4), pericyte (5), bar = 5 µm, bar insert = 1 µm. (b) TEM image of a telocyte cell body (3) with two long, thin telopodes (arrows), completely surrounded by collagen ﬁbres (2). Vein (1), bar = 1 µm.

113

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

114

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Fig. 7. (a) TEM image of perineural cells (2) and telocytes’ telopodes (3) next to a neuron cell body (1) in a ganglion located at the veno-atrial junction of a pig, bar = 1 µm. (b) Detail of (a) detailed image of perineural cells’ prolongations (4) with a basal lamina (5) and a telocytes’ prolongations (3) without a basal lamina. Neuron cell body (1), podom (6), bar = 500 nm. (c) TEM image of perineural cells (2) and telocytes’ telopodes (3) in the vicinity of a neuron cell body (1) and a blood vessel (9), bar = 1 µm. (d) Detail of (c) detailed image of perineural cells’ prolongations (4) with a basal lamina and a telocytes’ prolongations (3) without a basal lamina, bar = 500 nm. (e) TEM image of perineural cells (2) and telocytes’ telopodes (3) in the vicinity of a neuron cell body (1), bar = 2 µm. (f) Detail of (e) image of a telocyte cell body (7) with long telopodes (3) enclosing collagen (10), in close vicinity of perineural cells’ prolongations (11), which are separated from each other by collagen bundles. Telocyte prolongations (3) lack a basal lamina as the perineural prolongation (11) demonstrates a discontinuous basal lamina (12). Telocyte nucleus (8), bar = 1 µm. Discussion

In this study, we described the presence of myocardial sleeve tissue, ganglia and telocytes inside the porcine pulmonary veins wall. Moreover, the presence of telocytes in ganglia and the propinquity of telocytes and cardiomyocytes were demonstrated. Besides this, in literature the contact between telocytes and cardiomyocytes was shown in men (Gherghiceanu et al., 2012). Telocytes were already indicated in a human adult trigeminal ganglion (Rusu et al., 2016). The correlation between telocytes and arrhythmias and the elicitation of pulses by telocytes has already been suggested by Gherghiceanu et al. (2008) and Morel et al. (2008). Based on these data, we hypothesize that these three structures may interact with each other. Cell – cell communication attributed to telocytes, present in pulmonary veins and in close vicinity to cardiomyocytes of myocardial sleeve tissue and neuron cell bodies, leads to the deduction that this process may affect ectopic impulse formation in the pulmonary veins. Consequently, abnormal pulse conduction may be developed which, in certain pathological cases, could result in the development of cardiac arrhythmias such as atrial fibrillation.

115

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Atrial fibrillation is caused by abnormal electrical pulses generated by certain foci from the myocardial sleeve located in the pulmonary veins (Haïssaguerre et al., 1998). To better understand the pathophysiology of AF, further research should focus on the mechanisms of pulmonary vein impulse generation and impulse conduction towards the left atrium as there exists no specific conductive tissue (Haïssaguerre, 2010). It is assumed that the ganglia at the base of the pulmonary veins also play a role in the induction of AF. During treatment of AF by ablation, not only the myocardial sleeve but probably also other histological structures that might play a role in the induction of AF, are destroyed. Concerning the role of ganglia in the development of AF, Lemola et al. (2008) indicated that intact pulmonary veins were not essential structures to maintain cholinergic atrial fibrillation as vagal atrial fibrillation can be inhibited by ablation of the peri-pulmonary ganglionic plexi. It was stated that the pulmonary veins themselves are not indispensable in the induction of experimental atrial fibrillation due to atrial tachycardia remodeling or congestive heart failure (Nishida et al., 2010). Moreover, autonomic ganglia were found to be important in the induction of atrial fibrillation related to atrial tachycardia remodeling (Nishida et al., 2011). Also Scherlag et al. (2005) and Zhou et al. (2007) proved the role of ganglia in the fat pads at the base of a canine pulmonary vein in the induction of AF. This indicates that pulmonary veins are important structures in the elicitation of abnormal electrical pulses but opinions vary on the precise histological foci.

The interaction between telocytes and cardiomyocytes on the one side and between telocytes and nerve tissue on the other side was proven by Gherghiceanu et al. (2010), Popescu et al. (2010), Gherghiceanu and Popescu (2011) and Popescu and Nicolescu (2013). More specifically, telocytes were found to interact with cardiomyocytes through plasma membrane contacts and bridges of basal lamina-like components (Beltrami et al., 2003; Faussone-Pellegrini and Bani, 2010). Other communication mechanisms, for mechanical support during myocardial cell growth, were elucidated such as signal exchange through vesicular transfer and cell-cell connections (Deregibus et al., 2010). Until now, close contact between telocytes and ganglia, and the presence of telocytes inside ganglia of the pulmonary veins has never been indicated. Further research is needed to elucidate the cellular and electrophysiological interactions between telocytes, ganglia and myocardial tissue in the pulmonary veins. These data are indispensable to further unravel the mechanisms of AF.

Concerning the discovery of telocytes, Popescu and Nicolescu (2013) warned for the mindset that “cells which are not clearly a fibroblast might be considered a priori fibroblast-like”. Many techniques can be used to pursue the ultimate goal of finding the incontestable prove of the existence of telocytes and the difference between telocytes and fibroblasts. Nevertheless, to date, not one specific marker for fibroblasts is found but different antibodies could be used depending on the

116

Chapter 4: Ganglia and telocytes in porcine pulmonary veins target tissue. The wide range of immunohistochemical properties of fibroblasts in different organs and tissues can be an indication for the existence of many subpopulations under the common heading of “fibroblasts”. But the enduring quest of many researchers to find this specific fibroblast marker is also the enigma of which researchers in the context of telocytes are faced with. So arises the question if telocytes are one cell type or a common heading of an entire cell population. It could be concluded that telocytes, as a cell type, are a large population of cells containing a wide range of cells with their own distinct morphological features and functions but with similar immunohistochemical properties as these cells may have a common origin. In contrast, in literature it was suggested that telocytes could dedifferentiate easier compared to other terminally differentiated cells (Popescu and Nicolescu, 2013). The extensive variety of different functions attributed to telocytes (Cretoiu and Popescu, 2014) could be regarded as unusual for a differentiated cell. However, when looking at mesenchymal stem cells as an undifferentiated cell type from which a range of differentiated cells with different functions but certain similar immunological characteristics arise during adult life (Gilbert, 2010), the question arises if telocytes can be categorized likewise or arise as a subpopulation of fibroblasts. Another interesting research path is the reprogramming of fibroblasts into a variety of cell types, such as neural stem cells, cardiomyocytes and hepatocytes (Huang et al., 2014, Zhu et al., 2015). In this context, research on the relationship between fibroblasts and telocytes may deliver useful insights. Regarding the comparison between fibroblasts and telocytes, through functional cellular differentiations, the original protein pattern can partially be preserved, especially if some similar functions are displayed, creating the problem of for example vimentin/Cd34/SMA/PDGF-R beta positive telocytes and fibroblasts.

Concerning the presence of a basal lamina in telocytes, fibroblasts and perineural cells, literature is not unambiguous. Popescu and Faussone-Pellegrini (2010) describe telocytes with a present basal lamina which is discontinuous while Kostin (2016) indicates that the discontinuous basal lamina can occasionally be observed in telopodes. Schurch et al. (1997) indicates the absence of a basal lamina in fibroblasts. Concerning perineural cells, Inokuchi et al. (1991) describes a discontinuous basal lamina on the endo- and epineurium side with in some cells, the presence of a one-sided continuous basal lamina or with a discontinuous basal lamina on the other side.

At the level of nerve tissue, it may be difficult to differentiate telocytes from perineural cells as both cell types have certain similar characteristics such as the presence of long prolongations. On the other hand, some differences can be observed as perineural cells are organized in a lamellar arrangement, consisting of several concentric layers with some collagen in between (Krstic, 1984). Moreover, the latter lacks the presence of thickenings on the prolongations.

117

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

In conclusion, this study has demonstrated qualitatively the occurrence of ganglia and telocytes inside the porcine pulmonary veins wall and the detailed histological characteristics of the myocardial sleeve in the pulmonary vein. To our knowledge, this is the first time that telocytes are described in the pulmonary veins of pigs, especially the presence of telocytes inside a ganglion located at the level of the pulmonary veins. In pigs, this information provides the morphological basis for an animal model (Vandecasteele et al., 2015) for further development of AF treatment options.

118

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

References

Bancroft J.D. and Gamble M. (2002). Theory and Practice of Histological techniques, 5th Edition. Amsterdam, The Netherlands: Elsevier Limited, pp 127-130.

Beltrami A.P., Barlucchi L., Torella D., Baker M., Limana F., Chimenti S., Kasahara H., Rota M., Musso E., Urbanek K., Leri A., Kajstura J., Nadal-Ginard B. and Anversa P. (2003). Adult cardiac stem cells are multipotent and support myocardial regeneration. Cell. 114:763-776.

Cretoiu S.M. and Popescu L.M. (2014). Telocytes revisited. Biomol Concepts. 5:353-369.

Cretoiu D., Radu B.M., Banciu A., Banciu D.D. and Cretoiu S.M. (2017). Telocytes heterogeneity: From cellular morphology to functional evidence. Sem Cell Develop Biol. 64:26-39.

Deregibus M.C., Tetta C. and Camussi G. (2010). The dynamic stem cell microenvironment is orchestrated by microvesicle-mediated transfer or genetic information. Histol Histopathol. 25:397- 404.

Faussone-Pellegrini M.-S. and Bani D. (2010). Relationships between telocytes and cardiomyocytes during pre- and post-natal life. J Cell Mol Med. 14:1061-1063.

Gilbert S.F. (2010). The stem cell concept. In: Developmental Biology, 9th edition. MA: Sinauer Associates, pp 330-331.

Gherghiceanu M., Manole C.G. and Popescu L.M. (2010). Telocytes in endocardium: electron microscope evidence. J Cell Mol Med. 14:2330-2334.

Gherghiceanu M. and Popescu L.M. (2011). Heterocellular communication in the heart: electron tomography of telocyte-myocyte junctions. J Cell Mol Med. 15:1005-1011.

Gherghiceanu M. and Popescu L.M. (2012). Cardiac telocytes - their junctions and functional implications. Cell Tissue Res. 348:265-279.

Haïssaguerre M., Jais P., Shah D.C., Takahashi A., Hocini M, Quiniou G., Garrigue S., Le Mouroux A., Le Metayer P. and Clementy J. (1998). Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. New Engl J Med. 339:659-666.

119

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Haïssaguerre M., Hocini M., Sanders P., Takahashi Y., Rotter M., Sacher F., Rostock T., Hsu L.F., Jonsson A., O'Neill M.D., Bordachar P., Reuter S., Roudaut R., Clementy J. and Jais P. (2006). Localized sources maintaining atrial fibrillation organized by prior ablation. Circulation. 113:616-625.

Haïssaguerre, M., Lim K.T., Jacquemet V., Rotter M., Dang L., Hocini M., Matsuo S., Knecht S., Jais P. and Virag N. (2007). Atrial fibrillatory cycle length: computer simulation and potential clinical importance. Europace. 9:64-70.

Haïssaguerre M. (2010). In search of the sources of cardiac fibrillation. EMBO Mol Med. 2:117-119.

Hamabe A., Okuyama Y., Miyauchi Y., Zhou S., Pak H.N., Karagueuzian H.S., Fishbein M.C. and Chen P.S. (2003). Correlation between anatomy and electrical activation in canine pulmonary veins. Circulation. 107:1550-1555.

Ho S.Y., Cabrera J.A., Tran V.H., Farre J., Anderson R.H. and Sanchez-Quintana D. (2001). Architecture of the pulmonary veins: relevance to radiofrequency ablation. Heart. 86:265-270.

Huang P., Zhang L., Gao Y., He Z., Yao D., Wu Z., Cen J., Chen X., Liu C., Hu Y., Lai D., Hu Z., Chen L., Zhang Y., Cheng X., Ma X., Pan G., Wang X. and Hui L. (2014). Direct reprogramming of human fibroblasts to functional and expandable hepatocytes. Cell Stem Cell. 14:370-384.

Inokuchi T., Yokoyama R., Higashi R., Takahashi Y. and Miyajima S. (1991). Ultrastructure of the Perineurial Cell of the Sciatic Nerve in Rats –A Transmission and Scanning Electron Microscopic Study-. Kurume Med J. 38:221-232.

Katritsis D., Giazitzoglou E., Sougiannis D., Goumas N., Paxinos G. and Camm A.J. (2008). Anatomic approach for ganglionic plexi ablation in patients with paroxysmal atrial fibrillation. Am J Cardiol. 102: 330-334.

Kneeland P.P. and Fang M.C. (2009). Trends in catheter ablation for atrial fibrillation in the United States. J Hosp Med. 4:E1-E5.

Kostin S. (2016). Cardiac telocytes in normal and diseasedhearts. Semin Cell Dev Biol. 55:22–30.

Krstic R.V. (1984). Illustrated Encyclopedia of Human Histology. New York, USA: Springer-Verlag, pp 324.

Lemola K., Chartier D. and Yeh Y.H. (2008). Pulmonary vein region ablation in experimental vagal atrial fibrillation: role of pulmonary veins versus autonomic ganglia. Circulation. 117:470-477.

120

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Luesma M.J., Gherghiceanu M. and Popescu L.M. (2013). Telocytes and stem cells in limbus and uvea of mouse eye. J Cell Mol Med. 17:1016-1024.

Morel E., Meyronet D. and Thivoet-Bejuy F. (2008). Identification and distribution of interstitial Cajal cells in human pulmonary veins. Heart Rhythm. 5:1063-1067.

Nathan H. and Gloobe H. (1970). Myocardial atrio-venous junctions and extensions (sleeves) over the pulmonary and caval veins. Anatomical observations in various mammals. Thorax. 25:317-324.

Nishida K., Sarrazin J.F., Fujiki A., Oral H., Inoue H., Morady F. and Nattel S. (2010). Roles of the left atrial roof and pulmonary veins in the anatomic substrate for persistent atrial fibrillation and ablation in a canine model. J Am Coll Cardiol. 56:1728-1736.

Popescu L.M., Manole C.G., Gherghiceanu M., Ardelean A., Nicolescu M.I., Hinescu M.E. and Kostin S. (2010). Telocytes in human epicardium. J Cell Mol Med. 14:2085-2093.

Popescu L.M. and Faussone-Pellegrini M.S. (2010). TELOCYTES - a case of serendipity: the winding way from Interstitial Cells of Cajal (ICC), via Interstitial Cajal-Like Cells (ICLC) to TELOCYTES. J Cell Mol Med. 14:729-740.

Popescu L.M. (2011). The tandem: telocytes-stem cells. Int J Biol Biomed Eng. 5:83–92.

Popescu L.M. and Nicolescu M.I. (2013). Telocytes and stem cells. In: dos Santos Goldenberg R.C., Campos de Carvalho A.C., editors. Resident stem cells and regenerative therapy. Oxford: Academic Press, pp 205–231.

Rusu M.C., Cretoiu D., Vrapciu A.D., Hostiuc S., Demengiu D., Manoiu V.S., Cretoiu S.M. and Mirancea N. (2016). Telocytes of the human adult trigeminal ganglion. Cell Biol Toxicol. 32:199-207.

Scherlag B.J., Yamanashi W., Patel U., Lazzara R. and Jackman W.M. (2005). Autonomically induced conversion of pulmonary vein focal firing into atrial fibrillation. J Am Coll Cardiol. 45:1878-1886.

121

Chapter 4: Ganglia and telocytes in porcine pulmonary veins

Schurch W., Seemayer T.A. and Gabbiani G. (1997). Myofibroblast. In: Sternberg S.S., editor. Histology for pathologists, 2nd edition. Philadelphia: Lippincott-Raven, pp 129-165.

Suciu L., Popescu L.M., Gherghiceanu M., Regalia T., Nicolescu M.I., Hinescu M.E. and Faussone- Pellegrini M.-S. (2010). Telocytes in human term placenta: morphology and phenotype. Cells Tissues Organs. 192:325-339.

Vandecasteele T., Boussy T., Philpott M., Clement E., Schauvliege S., Van den Broeck W., van Loon G., Cornillie P. and Van Langenhove G. (2016). A preclinical study of an implanted device in the pulmonary veins, intended for the treatment of atrial fibrillation in an ovine model. Pace. 39:822- 829.

Vollmerhaus B., Roos H. and Troger U. (1999). Aa. And Vv. Pulmonales of the miniature pig and a common view of their supportive function in the lung. Anat Histol Embryol. 28:143-149.

Xu J., Luc J.G.Y and Phan K. (2016). Atrial fibrillation: a review of current treatment strategies. J Thorac Dis. 8:886-900.

Zheng Y., Zhu T., Lin M., Wu D. and Wang X. (2012). Telocytes in the urinary system. J Transl Med. 10: 188.

Zhou J., Scherlag B.J., Edwards J., Jackman W.M., Lazzara R. and Po S.S. (2007). Gradients of atrial refractoriness and inducibility of atrial fibrillation due to stimulation of ganglionated plexi. J Cardiovasc Electrophysiol. 18:83-90.

Zhu S., Wang H. and Ding S. (2015). Reprogramming fibroblasts toward cardiomyocytes, neural stem cells and hepatocytes by cell activation and signaling-directed lineage conversion. Nat Protoc. 10: 959-973.

122

Chapter 5

A preclinical study of an implanted device in the pulmonary veins, intended for the treatment of atrial fibrillation in an ovine model

Adapted from Vandecasteele T., Philpott M., Boussy T., van Loon G., Cornillie P. and Van Langenhove G. (2016). A preclinical study of an implanted device in the pulmonary veins, intended for the treatment of atrial fibrillation in an ovine model. Pacing Clin Electrophysiol. 39:822-829.

123

124

Chapter 5: Stent implantation in ovine model

Summary

Atrial fibrillation is the most frequent arrhythmia in adults of which the interventional cure is hampered by high recurrence rates. Recurrence after ablation is due to an incomplete isolation of the pulmonary veins. A new ablation technique was performed, in the antra of ovine pulmonary veins, by device implantation, which was heated through a wireless heat-generating system. Implants were placed transatrially in the pulmonary veins of sheep. Using a wireless heating system, the energy was afterward transferred through wires to the implanted device according to a defined protocol. The position of the implant and the applied lesions were macroscopically evaluated. Samples of the ablated tissue of the atrio-pulmonary vein junction were histologically and immunohistochemically examined. Six ablation procedures in four sheep were successfully performed without adverse cardiac reactions. Implantation of the device and the wireless heat generation was feasible. Sufficient heat was produced at the level of the antra of the pulmonary veins to create ablation lesions, which were histologically and immunohistochemically confirmed.

125

Chapter 5: Stent implantation in ovine model

Introduction

Atrial fibrillation (AF) is the most common cardiac arrhythmia in adults associated with substantial mortality and morbidity. It is generally accepted that the main triggers of abnormal pulses, which induce AF, are localized at the level of the pulmonary veins base (Jais et al., 1997; Haïssaguerre et al., 1998; Chen et al., 1999). In the past decade, several catheter-based ablation techniques were tested and improved to isolate electrically the pulmonary veins from the left atrium (Haïssaguerre et al., 2000; Pappone et al., 2000; Shah et al., 2001; Jones et al., 2008; Gerstenfield and Michele, 2010). Nevertheless, ablation of the pulmonary veins is, still to date, a technically demanding technique and recurrence of AF is seen due to incomplete isolation of the pulmonary veins. During long-term follow-up, a recurrence rate of more than 50% was mentioned after a single procedure and without the use of antiarrhythmic drugs (Medi et al., 2011). Consequently, the need for a better treatment strategy which ensures a high success rate along with a relatively quick and simple procedure remains, ideally offering the possibility to repeat the procedure noninvasively.

In this context, an implantable device was developed, after positioning in an electromagnetic field, which can be wirelessly heated, creating the desired transmural ablation lesions at the level of the pulmonary vein-left atrial junction. In the first step, an in situ study examines wireless heat generation after which the heat is transferred to the implanted ablation device. This implant has a tubular shape through which a complete circular lesion can be applied. This procedure was performed in an ovine model by open heart surgery due to access of the pulmonary veins. The aim of this study was to evaluate this technique, directly after the ablation, by macroscopic, histological, and immunohistochemical investigation of the induced lesions in sheep.

Materials and Methods

Device

The implants (Fig. 1) contain a heating coil (Verhaert, Kruibeke, Belgium), which includes a temperature feedback system. Implant size ranged from 12 mm to 20 mm. An electromagnetic coil generating an electromagnetic field, in which a pick-up coil was placed according the same plane, constitutes the heat-generating system. The heat generated in the pick-up coil was transferred through wires to the heating coil.

126

Chapter 5: Stent implantation in ovine model

Fig. 1. Left image = Verhaert implant device with heater coil (1); right image = pick-up coil (2) and electromagnetic or applicator coil (3).

Procedure

The procedure was performed at IMM Recherche (Paris, France) under the Ethics committee approval number “RYTHMOLOGIE-12-12-FLUX MEDICAL 14–34.” Studies were performed on four sheep (1–1.5 years old).

Using 6 mg/kg propofol (Fresofol, Pharmatel Fresenius Kabi, Hornsby, Australia) induction and 2% isoflurane (Abbott Australasia, Kurnell, Australia) maintenance, sheep were anesthetized and placed in dorsal recumbency. Arterial and venous cannulae were inserted into the carotid artery and jugular vein using a cut-down technique for pressure monitoring. The chest was opened using conventional techniques. A 40/36-F two-staged venous cannula (Edwards Life Sciences, Irvine, CA, USA) was inserted into the right atrium and secured with a pursestring suture. A 22-F aortic cannula (Edwards Life Sciences) was inserted into the proximal part of the descending aorta. The extracorporeal circulation system consisted of a venous reservoir, roller pump (Cobe Cardiovascular, Arvada, CO, USA), and a Capiox RX 25 membrane oxygenator (Terumo Europe, Leuven, Belgium) connected by noncoated tubing. Cardiopulmonary bypass was established by priming with 1,000 mL crystalloid prime solution (lactated Ringer's solution 750 mL, 20% mannitol 100 mL, aprotinin 100 mL, 8.4% sodium bicarbonate 50 mL, and heparin 5,000 IU), which was allowed to circulate in the extracorporeal circulation system for 2 hours. Flow was maintained at >3 L/min with a perfusion pressure of 50–70 mm Hg. After opening of the left atrium, suction cannulas were placed inside the

127

Chapter 5: Stent implantation in ovine model pulmonary veins. Afterward, the most suitable-sized implant, according to the diameter of the pulmonary vein, was inserted into the antrum, which is the common venous space, of both ostia through which the pulmonary veins drain into the left atrium (for terminology see Vandecasteele et al., (2015)). Subsequently, the implanted device was connected through wires to a pick-up coil to ensure the heat transfer. Also, a temperature control device was put in place.

The electromagnetic field, created by the applicator coil, generates an alternating current in the pick- up coil, when placed in the same plane of the applicator coil, which is consequently transferred into heat. This heat causes the ablation process of the connected implant's heater coil. Before the ablation procedure was started, the temperature sensor was checked to ensure that the indicated temperature correlated with the body temperature.

At the start of the heating procedure, the temperature control software was set to 1°C above the body temperature after which the temperature was gradually increased from the initial temperature in 1°C steps every few seconds while trying to reach the required ablation temperature of 58°C during different time periods. This procedure was repeated for each ostium, one implant at a time.

Pathological Analysis

All sheep were euthanized directly after the ablation procedure. Immediately after euthanasia, the heart-lung packages were excised after which the pulmonary veins, containing the ablation device, were removed and examined macroscopically.

Histological Analysis

Four percent formaldehyde-fixed paraffin-embedded scarred tissue of the atrio-pulmonary vein junction of two sheep was examined on 5-μm-thick hematoxylin and eosin (H&E)-stained histological sections. The orientation in which the histological sections were cut is indicated in Figure 2.

128

Chapter 5: Stent implantation in ovine model

Fig. 2. Ablated atrio-pulmonary vein junction tissue samples of two sheep. AS = atrial side of the sample; PS = pulmonary side of the sample; the oval indicates the ablation region; red line = trimming line indicating the orientation of the histological sections.

Immunohistochemical Analysis

The myocardial sleeve of the ablated pulmonary vein was stained immunohistochemically with the myosin marker MYBPC3 (polyclonal, K-16: sc-50115, Santa Cruz Biotechnology, Santa Cruz, CA, USA) to detect the myosin-binding protein C (cardiac type), which is reactive with ovine tissue. Myosin- binding protein C is present in the myofibrils of cardiac tissue (Winegrad, 1999). The 5-μm-thick slides were immunostained using the Dako automated Autostainer Plus (Dako, Glostrup, Denmark). No antigen retrieval was performed. The sections were incubated for 5 minutes with 3% hydrogen peroxide and 30 minutes with rabbit serum. First, the primary antibody (1:100) was incubated for 60 minutes, afterward secondary antibody (rabbit/anti-goat, biotinylated, polyclonal, 1:500, Dako) was applied for 30 minutes followed by streptavidin and horseradish peroxidase (streptavidin-HRP, 1:500, Dako) for 30 minutes. The visualization was achieved with DAB (Dako) for 5 minutes.

Results

Ablation Data

The data of the ablation procedures performed with eight devices in four sheep (two devices per animal) are indicated in Figure 3 and Tables I and II. Not one of the following tests was aborted due to any cardiovascular disorder or abnormalities in rhythmicity.

129

Chapter 5: Stent implantation in ovine model

Fig. 3. Time/temperature table of the ablation procedure of the pulmonary veins during 250 seconds (panels A, B, and G), 400 seconds (panel E), 500 seconds (panel F), 600 seconds (panels C and D), 800 seconds (panel H). Tstent and Tenviro indicate implant temperature and body temperature, respectively, I refers to the used current, Tset indicates the set temperature.

130

Chapter 5: Stent implantation in ovine model

Implants Sheep 1 Sheep 2 Sheep 3 Sheep 4

Implant 1 Temperature Failure: as a To a higher Gradually Failure: temperature increase current temperature but faster surpassed the 60°C increase compared to the limit (demonstrated

around 100 test in sheep 1 by a large seconds temperature peak), induced no associated with the temperature production of a small increase amount of smoke (Fig.3a) after which the test was stopped. It was Set 50°C 55°C for 210 found that both the temperature seconds temperature and heater power wire Implant Followed nicely Reacted on were damaged (Fig.3g) temperature the set the temperature temperature causing the increase implant to ablate (only small at 50°C for 180 variations seconds (Fig.3c) were seen) (Fig.3e)

Implant 2 Temperature Gradually Of which only the Was heated Temperature raised increase peak similarly to gradually to two temperatures the previous different set reached briefly test temperatures set temperature

Set Increased from Was increased to Was First 52°C for 120 temperature 43.5°C to 44°C 55°C for 230 increased to seconds and for 180 seconds 55°C for 300 thereafter of 53°C for seconds seconds 480 seconds

Implant Followed set Was stable at an Followed set At each set temperature temperature average temperature temperature, the (Fig.3b) temperature of almost implant temperature 51°C (Fig.3d) perfectly was fairly constant with only with approximately some small 2°C variation variations compared to the set (Fig.3f) temperature (Fig.3h)

Table 1. Description of Temperature Data, Visualized in the Graphs of Figure 3, of the Devices Implanted in Different Sheep According to the Set Temperature.

131

Chapter 5: Stent implantation in ovine model

Implants Sheep 1 Sheep 2 Sheep 3 Sheep 4

Implant 1 Failure was Was heated to 50°C as this The required current Was stopped due to provides the required was relatively low due alleged malposition of ablation lesion according to compared to previous damage of the the previous tests. At maximum tests as the peaks temperature temperature set temperature, the surpassed 80 A and sensor or the sensor which required current peaked at the average levels heater coil was not in almost 120 A and averaged were around 70 A. The during the contact with around 100 A (Fig 3c) peak at the end was implantation the vessel wall due to system reset by (Fig. 3g) (Fig.3a) which the temperature dropped and consequently the software increased the current output accordingly (Fig. 3e)

Implant 2 The required Reached set temperature at Demonstrated a The power current certain moments which is required power peak peaked just peaked due to insufficient contact of just below 120 A under 120 A (almost 120 A) between the temperature and an average of and averaged at a maximum sensor and the vessel wall or around 80 A (Fig. 3f) around 60 A set between the heating coil and (Fig. 3h) temperature the vessel wall in the vicinity and was of the temperature sensor. averaged Due to this uncertainty, the around 60 A ablation time was reduced to (Fig. 3b) 230 seconds instead of the planned 300 seconds. The required power peaked and averaged just below 120 A and was increased gradually in the last half of the test to reach the set temperature (Fig. 3d)

Table 2. Additional Information about the Procedures Mentioned in Figure 3 and Table I.

132

Chapter 5: Stent implantation in ovine model

Pathological Analysis

Macroscopic Evaluation

Observation of the heart-lung packages at the level of the pulmonary veins revealed no damage to the surrounding structures. The endothelium of all pulmonary veins showed clear ablation lesions (Fig. 4) along the circumference of the device, although by comparing the samples some differences in degree of discoloration were noticed. At the level of the device, no thrombus formation was noticed in any of the sheep.

Fig. 4. Parts of pulmonary vein tissue at the level of the atrio-pulmonary vein junction. Left image: at the top side the white colored ablation site indicated by the arrows and at the bottom the normal pinkish-colored tissue. Right image: the removed device with the heater coil (green wire) of which an ablation mark (indicated by the frame) is seen on a piece of the pulmonary vein wall.

Microscopic Evaluation

The ablation lesion, investigated on two tissue samples, was recognized on the histological sections by fibrin deposition, visualized in Figure 5. Figure 6 demonstrates immunohistochemically the myosin staining of myocardial sleeve tissue of the ablated pulmonary vein, including the ablation site. At this latter spot, a lighter immunohistochemical reaction can be observed in comparison to the tissue proximal and distal to the site of the implanted stent.

133

Chapter 5: Stent implantation in ovine model

Fig. 5. Histological section of the ablated atrio-pulmonary vein junction (H&E staining). Left image: AS = atrial side; 1 = myocardium; 2 = fibrin deposits; 3 = interruption of the tissue. Scale bar is 250 μm. Right image: PS = pulmonary side; 2 = fibrin deposits; 4 = lung tissue; 5 = hemorrhagic necrosis; arrow = pulmonary vein wall; Scale bar is 1 mm.

Fig. 6. Immunohistochemical staining with myosin marker MYBPC3 of myocardial sleeve tissue inside pulmonary vein wall. The double arrow indicates the stent implantation site. AS = atrial side; PS = pulmonary side. Scale bar: top left and top right image is 200 μm; top middle image is 100 μm; bottom image is 2 mm.

134

Chapter 5: Stent implantation in ovine model

Discussion

This study was performed in sheep in an open heart surgery procedure in order to have a good accessibility to the pulmonary veins. In numerous cardiovascular studies, pigs are used as an animal model in which the localization of the draining openings or ostia of the pulmonary veins into the left atrium was studied in detail (Vandecasteele et al., 2015). However, anatomically, the number of ostia in pigs shows similarity with sheep or goats as at the level of the pulmonary vein-left atrial junction two vestibules of the left atrium were described in a clinical study in goats (Reddy et al., 2004).

Ablation Data

We demonstrated that ablation of the pulmonary veins through wireless energy transfer is feasible. Eight stents were implanted and tested in four living sheep of which two ablation procedures failed. One failure occurred during the initialization phase and one during the ablation phase itself. Cause of this failure was damage to the temperature sensor during implantation which gave wrong temperature data. Thus during system check of the implants, six of eight implants met the technical requirements.

The difference in extensiveness of the ablation lesions was due to the fact that no clear temperature protocol was predefined, and that different temperatures and different ablation times were applied. Histologically, on H&E-stained sections, the effects of heating at the level of the pulmonary veins could only be noticed by visualization of fibrin deposits in acute euthanized animals. Immunohistochemically, denaturation of myosin proteins of the myocardial sleeve at the level of the heated implanted stent proves the induction of a lesion in the pulmonary veins wall caused by the generated heat at the ablation site. After long-term studies, greater effects may be seen as fibrous tissue completely replaces the necrotic tissue.

The degree of contact between the implant and the vessel wall is also of great importance to create a perfect circular lesion. However, overstretching of the pulmonary vein wall is possible but within its safe range in terms of elasticity. In this study, the implant diameter ranged from 12 mm to 20 mm. The information deduced from these procedures will further improve stent design intended for future studies.

The proof of the efficacy of this system, by which the heat is picked up wirelessly and transferred from the pick-up coil to the pulmonary veins through a direct connection, will allow for development of an implantable device that can be heated by placing the animal in an electromagnetic field. The

135

Chapter 5: Stent implantation in ovine model fact that the implanted device can remain in place and may be endothelialized constitutes a major advantage as a second treatment can be performed without any surgery.

In conclusion, in this ablation study in a sheep model, all devices could be placed inside in one of both antra of the pulmonary veins and were heated through remote energy transfer. All sheep survived the implantation and heating of the ablation devices without any rhythm disturbances. Macroscopically, in all successfully treated animals ablation lesions were noticed in the pulmonary veins and histological and immunohistochemical confirmation was provided of the applied lesion. Immunohistochemical staining was used to indicate the disappearance of myosin fibers during the ablation process as this is difficult to visualize on H&E-stained tissue. Myosin destruction was examined as Tornberg (2005) indicated that during a heating procedure, this feature is the first transition occurring in muscle tissue (Martens and Vold, 1976; Wright et al., 1977).

The macroscopical difference between normal and ablated tissue was clearly demarcated. Heat production at the level of the pulmonary veins was proven histologically by the presence of fibrin deposits. No thrombus formation was seen on the device and no damage to the surrounding tissue was noted. Further research is needed to clear out which temperature and ablation time result in an optimal ablation. Moreover, self-expanding devices would provide an additional advantage in terms of implant-pulmonary vein contact. Future trials have to confirm the feasibility of the procedure through a transfemoral approach, the application of a transmural lesion, the long-term persistence of pulmonary vein isolation, the endothelialization of the implant, the absence of excessive neointima formation and the feasibility of noninvasively repeating the procedure.

136

Chapter 5: Stent implantation in ovine model

References

Gerstenfeld E.P. and Michele J. (2010). Pulmonary vein isolation using a compliant endoscopic laser balloon ablation system in a swine model. J Interv Card Electrophysiol. 29:1-9.

Jais P., Haïssaguerre M., Shah D.C., Chouairi S., Gencel L., Hocini M. and Clementy J. (1997). A focal source of atrial fibrillation treated by discrete radiofrequency ablation. Circulation. 95:572-576.

Jones D.L., Guiraudon G.M., Skanes A.C. and Guiraudon C.M. (2008). Anatomical pitfalls during encircling cryoablation of the left atrium for atrial fibrillation therapy in the pig. J Interv Card Electrophysiol. 21:187-193.

Martens H. and Vold E. (1976). DSC studies of muscle protein denaturation. In: Proceedings of the 22nd European Meeting of Meat Research Workers, Malmö, Sweden (p. J 9.3).

Medi C., Sparks P.B., Morton J.B., Kistler P.M., Halloran K., Rosso R., Vohra J.K., Kumar S. and Kalman J.M. (2011). Pulmonary vein antral isolation for paroxysmal atrial fibrillation: results from long-term follow-up. J Cardiovasc Electrophysiol. 22:137-141.

Pappone C., Rosanio S., Oreto G., Tocchi M., Gugliotta F., Vicedomini G., Salvati A., Dicandia C., Mazzone P., Santinelli V., Gulletta S. and Chierchia S. (2000). Circumferential radiofrequency ablation of 246 pulmonary vein ostia: A new anatomic approach for curing atrial fibrillation. Circulation. 102:2619-2628.

137

Chapter 5: Stent implantation in ovine model

Reddy V.Y., Houghtaling C., Fallon J., Fischer G., Farr N., Clarke J., Mcintyre J., Sinofsky E.D., Ruskin J.N. and Keane D. (2004). Use of a diode laser balloon ablation catheter to generate circumferential pulmonary venous lesions in an open-thoracotomy caprine model. PACE. 27:52-57.

Shah D.C., Haïssaguerre M., Jais P., Hocini M., Yamane T., Deisenhofer I., Garrigue S. and Clementy J. (2001). Curative catheter ablation of paroxysmal atrial fibrillation in 200 patients: strategy for presentations ranging from sustained atrial fibrillation to no arrhythmias. PACE. 24:1541-1558.

Tornberg E. (2005). Effects of heat on meat proteins—Implications on structure and quality of meat products. Meat Sci. 70:493–508.

Vandecasteele T., Vandevelde K., Doom M., Van Mulken E., Simoens P. and Cornillie P. (2015). The Pulmonary veins of the pig as an anatomical model for the development of a new treatment for atrial fibrillation. Anat Histol Embryol. 44:1-12.

Winegrad S. (1999). Cardiac myosin binding protein C. Circ Res. 84:1117–1126.

Wright D.J., Leach I.B. and Wilding P. (1977). Differential scanning calorimetric studies of muscle and its constituents. J Sci Food Agric. 28:557–564.

138

Chapter 6

A preliminary study of pulmonary vein implant applicability and safety as a potential ablation platform in a follow-up study in pigs

Adapted from Vandecasteele T., Schauvliege S., Philpott M., Clement E., van Loon G., Vera L., Boussy T., Van Bergen T., Van den Broeck W., Cornillie P. and Van Langenhove G. (2018). A Preliminary Study of Pulmonary Vein Implant Applicability and Safety as a Potential Ablation Platform in a Follow-up Study in Pigs. Pacing Clin Electrophysiol. 41:167-171.

139

140

Chapter 6: Implant applicability and safety in follow-up study in pigs

Summary

Recurrence of atrial fibrillation after an ablation procedure remains a major problem which emphasizes the need for improved pulmonary vein isolation techniques. The aim of this study was to describe an implantation procedure of a pulmonary vein-stent which may possibly serve as an ablation technique in the future and to examine stent safety in a follow-up study in pigs. Eight pigs were catheterized and nine self-expanding nitinol stents were implanted through a transfemoral or transatrial approach into the antra of the pulmonary veins. After three months follow-up, the animals were euthanized for further examination. During the follow-up phase, no complications were observed. Absence of thrombus formation or pulmonary vein wall dissection was noticed during anatomical and histological evaluation of the heart-lung packages. All implants were almost completely covered by neo-intima, of which thickness varied between 0.2 and 3.9 mm. Stents can safely be positioned and deployed into the antra of the pulmonary veins without any acute or long term (three months) adverse effects. In the future, these implants could function as a permanently implanted ablation device and provide new therapeutic strategies for pulmonary vein isolation in patients with atrial fibrillation.

141

Chapter 6: Implant applicability and safety in follow-up study in pigs

Introduction

Vascular stenting is a standard procedure in the context of coronary interventions to scaffold coronary arteries in men (Mennuni et al., 2016). However, stents can also be of value when implanted in venous structures (Vandecasteele et al., 2015). Pulmonary vein isolation (PVI) has become the cornerstone in the treatment of paroxysmal atrial fibrillation (AF), which is an intermittent phase of one of the most important public health problems in western countries (Zoni- Berisso et al., 2014; Kumar et al., 2016). In recent years, different catheter ablation methods were used at the level of the atrio-venous junction to interrupt pulmonary vein ectopic pulses from reaching the left atrium (Haïssaguerre et al., 2000; Pappone et al., 2000; Jones et al., 2008). However, AF recurrence, mainly due to pulmonary vein (PV) reconnection, remains a major drawback (Medi et al., 2011) and PVI by means of ablation is still a challenging technique. As such, there is still room for improvement of PVI techniques in order to create circumferential ostial lesions. An implantable stent with an incorporated ablation ring, which can be heated from distance by an electromagnetic field in order to perform ablation, would be useful to completely isolate the PV from the left atrium in a relatively quick and simple way. In addition, if reconnection would occur, the permanently implanted stent could be re-used to restore PVI without any invasive intervention. Ideally, the stent would be overgrown without excessive neo-intima formation.

A preclinical study was already conducted in sheep where a stent was implanted in the PV via open heart surgery (Vandecasteele et al., 2016). Subsequently, heat was wirelessly generated outside the animals via an electromagnetic field and transferred through wires to the stent which then acted as an ablation device, creating a circumferential lesion. In the present study, a custom-made self- expanding nitinol stent was delivered through catheterization at the antra of the pulmonary veins in pigs but was not heated yet. The aim of this preliminary study is to assess the applicability of a stent inside the pulmonary veins and to evaluate the safety and its long-term effect on the PV wall.

Materials and Methods

The study was approved by the ethical committee of IMMR (Institut Montsouris Recherche, Paris, France, study number: Fulgur Medical 16-22). The average weight of the pigs (17 weeks old) was 62 kg (+/- 4kg). One self-expanding nitinol implant, consisting of a heating and a supporting part, was positioned into the antrum of one of both PV ostia in seven pigs. In one pig two implants were delivered, one in the antrum of each ostium (pig n°5, Table 1, terms see Vandecasteele et al., (2015)). The stents were unpolished, untreated and semi-sterile.

142

Chapter 6: Implant applicability and safety in follow-up study in pigs

The procedure was performed under general anesthesia. All pigs were treated orally once a day with 300 mg aspirin and 75 mg clopidogrel, from one day before the procedure until the pigs were euthanized. In addition, during the procedure, heparin (15000 IU) was administered intravenously to maintain an activated clotting time (ACT) of more than 300 seconds.

Using either a transatrial or a transfemoral/transseptal approach, implants of 25 or 30 mm diameter were delivered at the level of the myocardial sleeve in the antrum of the PV, using a custom-made delivery system. Transatrial access was achieved by a left thoracic incision in lateral position, followed by a 15 F sheath insertion into the left atrium. Before stent delivery, contrast dye was injected to evaluate the position of the sheath and a validated QCA system (Siemens, Munich, Germany) was used to assess the diameter of the PV antrum (Table 1) and to delineate the implantation spot. Subsequently, implants were chosen so that they were at least 20% larger in diameter than the QCA measurement (Table 1). After delivery of the stent, with its heating part located proximally, in the antrum of the PV, contrast dye was injected to evaluate the implant position and blood flow through the device. Transfemoral/transseptal approach was achieved by an 18F sheath insertion into the femoral vein followed by transseptal puncture and transseptal sheath (TSX Fixed Curve Transseptal Sheath, Boston Scientific, Marlborough, MA, USA) insertion. A guidewire was used to engage the antrum of the PV with a proprietary dynamic curve guiding catheter. Contrast dye was injected, images were fixed and the stent was delivered as described above.

Three months after implantation, all pigs were catheterized to perform angiography for qualitative blood flow assessment at the PV antra. After euthanasia, heart-lung packages were excised, macroscopically examined and samples were fixed in 4% formaldehyde solution. The PV containing an implant was excised, trimmed and embedded in methylmetacrylate to allow 15 µm thickness sectioning of PV and stent as a whole. Subsequently, a proximal (to the atrium), mid (intermediate position) and distal (to the lungs) sample was paragon-stained (fuchsine and toluidine blue). The untreated PV were dissected, trimmed and embedded in paraffin, sectioned at 5 µm thickness and hematoxylin & eosin-stained according to standard protocols.

143

Chapter 6: Implant applicability and safety in follow-up study in pigs

Histopathologic evaluation of methylmetacrylate and paraffin sections was performed as follows (Table 1):

1) Inflammation of PV wall tissue: 0 = no inflammatory cells, 1 = mild inflammation (up to 10 cells per x10 high power field (HPF)), 2 = moderate inflammation (between 10 – 20 cells per x10 HPF), 3 = strong inflammation (more than 20 cells per x10 HPF)

2) Necrosis of PV wall tissue around the implant: yes/no

3) Percentage of implant is covered by PV wall tissue: %

4) Neo-intima thickness (measured with an ocular lineal): mm

Results

Implantation

All animals were successfully catheterized with an average intervention time, measured between puncture of the vessel till implant delivery, of 23 minutes (+/- 12 minutes) and all pigs survived for three months without major complications. Implantation time in case of transfemoral approach, excluding transseptal puncture, or in case of a transatrial access, excluding thoracotomy time, was 12 minutes (+/- 4 minutes). No abnormal periprocedural events were noticed and pulse rate and blood pressure remained stable during the procedure with an ACT value above 300 seconds. In seven out of eight pigs, the implant was delivered at the desired location inside the PV (Fig. 1). Both in the acute phase and after three months, no blood stasis was noticed at the level of the implant during angiography.

144

Chapter 6: Implant applicability and safety in follow-up study in pigs

Fig. 1. Panel A: View on the dorsal part of the left atrium with a stent implanted inside the posterior PV. The proximal heating struts are positioned at the level of the PV ostium. Panel B: Angiographic image of an implanted stent at the level of the PV antrum (AN). The injected contrast visualizes the transition, indicated by the dotted line, from the antrum to the left atrium (LA). Both full lines indicate the visible borders of the left atrium.

Gross Anatomy

The implants were found inside the wall of the antrum, overgrown by neo-intima. In one pig, the implant was embolized into the left atrium, therefore this pig was excluded from subsequent histopathological evaluation (Table 1, pig n°8). (Size of PV and implants see Table 1).

Histopathology

All PV tissue samples, including the implanted device, were histologically examined (Fig. 2). Average neo-intimal formation was 1.9 mm (range 0.2-3.9 mm) (Table 1). Compared to the implant size of 25 or 30 mm, this relates to an average in-stent stenosis of 7.6% and 6.4% respectively. A variable amount of inflammation at the level of the implants was noticed, ranging from low to high concentration of inflammatory cells (Table 1, Fig. 2). The extent of necrosis was highly variable and almost all samples demonstrated a near (90%) complete overgrown stent (Table1).

145

Chapter 6: Implant applicability and safety in follow-up study in pigs

Fig. 2. Histological images of a section of the PV wall. Panel A: Image of a mid-section of pig n°1, including a strut (1) of an implanted stent with visible neo-intima formation (arrow). High concentration of neutrophils are visible along the metal stent (2), paragon staining, bar = 0.2 mm. Panel B: Image of a mid-section of pig n°2, including a strut (1) of an implanted stent with visible neo-intima formation (arrow). No inflammation or necrosis is present, paragon staining, bar = 0.2 mm. Panel C: Image of a mid-section of pig n°7, including a strut (1) of an implanted stent with visible neo-intima formation (2). Concentration of inflammatory cells is visible near the strut (arrow). Lumen is located at the left side and the wall of the PV (3) is seen at the right side of the image, paragon staining, bar = 0.2 mm.

146

Chapter 6: Implant applicability and safety in follow-up study in pigs

Pig n° Age Weight Approach PV antral size Implant size Cross section position Inflammation Necrosis Tissue integration in cross Neointima thickness Additional (months) (kg) (diameter, mm) (mm) score section (%) (mm) remarks 1 3 60 transatrial 14 25 prox 3 Yes 90 0.7 Severe edema mid 2 Yes 90 0.7 Severe edema dist 2 Yes 90 0.9 Severe edema 2 3 62 transfemoral 18 25 prox 0 No 90 0.5 - mid 0 No 90 0.5 - dist 0 No 70 0.7 - 3 3 60 transfemoral 21 30 prox 2 Yes - 2.1 Many artifacts mid 2 Yes - 1.9 Many artifacts dist 2 Yes - 1.9 Many artifacts 4 3 63 transfemoral 22 25 prox 0 No 90 0.3 - mid 0 No 90 0.2 - dist 0 No 90 0.2 - 5 3 65 transatrial 14 25 prox superior 0 No 90 3.7 - mid superior 0 No 90 3.6 - dist superior 0 No 60 3.5 -

15 25 prox inferior 3 Yes 90 3.9 - mid inferior 3 Yes 90 3.8 -

dist inferior 2 Yes 90 3.2 -

6 3 65 transfemoral 27 30 Prox 1 No 90 1.5 Few multinucleated giant cells mid 1 No 90 1.6 - dist 0 No 90 1.1 - 7 3 61 transfemoral 23 25 prox 2 Yes 90 3.5 - mid 2 Yes 90 2.8 - dist 2 Yes 90 2.5 - 8 3 61 transatrial 23.5 30 / / / / / / Table 1. Overview of animal data, PV and implant size and histopathological results after three-months follow-up.

147

Chapter 6: Implant applicability and safety in follow-up study in pigs

Discussion

This pig study demonstrated safe delivery of a stent into the pulmonary veins antrum without complications during the three-months follow up. Therefore, an ablation approach using an implantable device that allows wireless heating seems attractive and should be explored further to assess efficacy and safety of the heating process and the long term outcome in a larger study group. Besides thermal ablation, also toxin or drug eluting stents, using botulin toxin, alcohol, rapamycin or beta-carboline derivates could be alternatives to achieve PVI.

Histopathology showed mild neo-intimal growth but with high variation between the different samples. Compared to Ishiwata et al. (2000) and Banai et al. (2004), the term “mild” used in this research is justifiable compared to approximately 50% of stenosis due to neo-intimal formation in porcine coronary arteries described in those papers. In addition, the larger size of our stents compared to the antrum resulted in mild dilatation so that the 1.9 mm neo-intimal growth was unlikely to result in stenosis. The different implants demonstrated no clear correlation between inflammation degree and neo-intimal formation. This is in contrast with coronary artery stenting in humans, in which increased neo-intimal growth is correlated with arterial damage and increased inflammation (Farb et al., 2002). The main reason for the variation in inflammation and neo-intimal formation was probably related to the fact that the stents were unpolished, untreated and only semi- sterilized with chlorhexidine before implantation. Use of sterile, polished and treated implants may diminish neo-intimal formation due to reduced inflammation and damage.

Blood flow at the level of the implant was assessed qualitatively using angiography. Quantitative determination of the blood flow, as performed in coronary arteries, is not appropriate or feasible in pulmonary veins.

In one pig dislodgement of the stent was seen which results into a complication rate of 12.5%. Further optimization of the technique with better stent-pulmonary vein matching and extensive follow-up in a large group of experimental animals is needed. Also the effect of stent-induced stretch on inflammation and neo-intimal growth and risk for stenosis should be investigated. However, the use of these custom-made self-expanding stent offers a great advantage of antral lumen enlargement which may provide a substantial buffer against restenosis.

In conclusion, self-expanding nitinol implants were safely delivered to the ostia and antra of pig PV, except in one case. Three months after implantation a variable but mild neo-intimal formation was noticed which did not hamper PV blood flow. Results from this preliminary study open perspectives

148

Chapter 6: Implant applicability and safety in follow-up study in pigs

for future development of stent-based atrial fibrillation treatment strategies in order to achieve an instant circumferential lesion for PVI isolation.

149

Chapter 6: Implant applicability and safety in follow-up study in pigs

References

Banai S., Gertz S.D., Gavish L., Chorny M., Perez L.S., Lazarovischi G., Ianculuvich M., Hoffmann M., Orlowski M., Golomb G. and Levitzki A. (2004). Tyrphostin AGL-2043 eluting stent reduces neointima formation in porcine coronary arteries. Cardiovasc Res. 64:165-171.

Farb A., Weber D.K., Kolodgie F.D., Burke A.P. and Virmani R. (2002). Morphological predictors of restenosis after coronary stenting in humans. Circulation. 105:2974-2980.

Haïssaguerre M., Shah D.C., Jais P., Hocini M., Yamane T., Deisenhofer I., Chauvin M., Garrigue S. and Clémenty J. (2000). Electrophysiological breakthroughs from the left atrium to the pulmonary veins. Circulation. 102:2463-2465.

Ishiwata S., Verheye S., Keith A.R., Mahomed Y.S., de Leon H., King S.B. and Chronos N.A.F. (2000). Inhibition of neointima formation by tranilast in pig coronary arteries after balloon angioplasty and stent implantation. J Am Coll Cardiol. 35:1331-1337.

Kumar S. and Michaud G.F. (2016). Pulmonary vein isolation in the treatment of atrial fibrillation. Res Rep Clin Cardiol. 7:47-60.

Mennuni M.G., Pagnotta P.A. and Stefanini G.G. (2016). Coronary Stents: The Impact of Technological Advances on Clinical Outcomes. Ann Biomed Eng. 44:488.

150

Chapter 6: Implant applicability and safety in follow-up study in pigs

Vandecasteele T., Boussy T., Philpott M., Clement E., Schauvliege S., Van den Broeck W., van Loon G., Cornillie P. and Van Langenhove G. (2016). A Preclinical Study of an Implanted Device in the Pulmonary Veins, Intended for the Treatment of Atrial Fibrillation in an Ovine Model. PACE. 39:822- 829.

Zoni-Berisso M., Fabrizio L., Tiziana C. and Domenicucci S. (2014). Epidemiology of atrial fibrillation: European perspective. Clin Epidemiol. 6:213–220.

151

152

Chapter 7

Immunohistochemical identification of stent-based

ablation lesions in the superior vena cava and pulmonary

veins

Adapted from Vandecasteele T., Schauvliege S., Boussy T., Philpott M, Clement E., Vera L., Cornillie P., De Spiegelaere W., Van Langenhove G., van Loon G. and Van den Broeck W. (2017). Immunohistochemical identification of stent-based ablation lesions in the superior vena cava and pulmonary veins. J Histol Histopathol. 4:14.

153

154

Chapter 7: Immunohistochemical identification of ablation lesions

Summary

Ablation procedures, in the context of atrial fibrillation treatment, induce lesions which are difficult to visualize in an acute phase. This study sought to develop an immunohistochemical technique to evaluate and visualize microscopically the applied acute ablation lesions, which may enable an objective assessment of the lesions as this is crucial to determine lesion transmurality. In the present study, an ablation procedure was performed in vivo in six pigs at the level of the cranial vena cava and in-vitro at the level of the ostia of the pulmonary veins using the same custom-made stents and delivery system. The antibody myosin MYBPC3 was used to visualize indirectly the denaturation process induced by the heating procedure. Subsequently, the histological images were processed by using Fiji software to make standardized sections which are objectively comparable. In all samples, the ablation lesions were visualized immunohistochemically and clearly identified to enable lesion measurement. The absence of transmural lesions was noticed in the samples which were exposed to an insufficient or an extreme high energy level or in the case the myocardial layer was divided by an intermediate fat layer. During postmortem histopathologic evaluation of ablated tissue, this technique offers a method to specifically determine the grade of muscle protein denaturation and to assess the degree of transmurality. Moreover, this study highlights the importance of myocardial tissue shielded by connective, fat and/or charred tissue, from the penetrating ablation heat in the context of recurrence or persistence of atrial fibrillation.

155

Chapter 7: Immunohistochemical identification of ablation lesions

Introduction

In men, atrial fibrillation (AF) is the most important cardiac arrhythmia with an increasing prevalence due to an ageing population (Roberts et al., 1993; Iwasaki et al., 2011). Besides a medical treatment of AF, ablation of myocardial sleeve tissue is increasingly used (Calkins et al., 2012). AF is mainly caused by the elicitation of ectopic pulses from myocardial sleeve tissue and their conductance towards the heart (Haïssaguerre et al., 1998). Ablation includes the interruption of the transfer of these pulses by using a heat or cold source. During an ablation procedure, lesions are applied to the myocardial tissue, especially at the level of the pulmonary veins wall or less commonly at the wall of the superior vena cava, coronary sinus or interatrial septum (Oral et al., 2002). During evaluation of preclinical or clinical ablation trials, a distinct histological determination of the applied lesions, in an acute stage without the presence of repair connective tissue, remains difficult as regular histological staining of ablated tissue fails to distinguish clearly denatured cells from intact cells. Consequently, an immunohistochemical technique might be valuable to visualize and delineate precisely and objectively the ablated lesions and to evaluate the grade of lesion transmurality. Transmural lesions can be the cornerstone for a successful ablation procedure as the use of the appropriate energy level will define the result of creating the required local lesion (Kowalski et al., 2012).

However, during chronic follow-up studies based on biopsies or postmortem evaluation, ingrown connective tissue can be stained histologically to differentiate damaged from intact myocardial tissue. Different histological staining techniques for chronic lesions have been described, such as masson trichrome, hematoxylin and eosin, and elastic-van Gieson stains (Accord et al., 2005; Deneke et al., 2005; Ranjan et al., 2011; Kowalski et al., 2012).

In this study, a technique is described to visualize and delineate immunohistochemically acute porcine ablation lesions at the level of the cranial vena cava (equivalent of the superior vena cava in humans) and at the level of the ostia of the pulmonary veins.

156

Chapter 7: Immunohistochemical identification of ablation lesions

Materials and Methods

Stent delivery

Superior vena cava

Six 1.5 month old pigs (pigs 1-6, Table 1), weighing between 40 and 45 kg were used in the animal trials performed at the IMMR institute, Paris, France according to the Ethical Committee approval number “Fulgur Medical 16-22”. Pigs were anesthetized using 6 mg/kg propofol and 2 % isoflurane. In each pig, one custom-made stent was placed inside the proximal part of the cranial vena cava. More specific, by using angiography the ablation ring of the stent was positioned to target myocardial tissue. Delivery of the stent through the right auricle was performed using a custom- made delivery catheter, followed by angiographic evaluation of the stent position.

Pulmonary veins

Fresh cardiopulmonary sets of nine pigs of 60 kg (pigs 7-15, Table 1) were used in the in-vitro trials performed at the Faculty of Veterinary Medicine, Ghent, Belgium. In each set, one custom-made stent was placed inside an ostium of the pulmonary veins and afterwards the stent and the surrounding tissue was excised from the cardiopulmonary set.

Ablation procedure

Superior vena cava

The stent was heated wirelessly by placing the pig in a coil to induce a magnetic field between the stent and the coil as described by Vandecasteele et al. (2016). Each pig was treated with a different combination of stent diameter, ablation time and power (Table 1).

157

Chapter 7: Immunohistochemical identification of ablation lesions

First treatment Second treatment Third treatment Implant size Remarks (s - W/mm2) (s - W/mm2) (s - W/mm2) (mm)

Pig 1 150 - 0.190 280 - 0.210 150 - 0.225 21 No circumferential wall contact

Pig 2 210 - 0.195 150 - 0.210 21 No circumferential wall contact

Pig 3 220 - 0.220 150 - 0.254 24

Pig 4 60 - 0.282 30

Pig 5 60 - 0.195 30

Pig 6 210 - 0.207 60 - 0.254 24

W/mm2 Duration (s) Flowrate (l/min) Implant size (mm)

Pig 7 0.209 90 0 18

Pig 8 0.205 180 0 18

Pig 9 0.205 180 0 18

Pig 10 0.313 90 1.25 21

Pig 11 0.313 90 1.25 21

Pig 12 0.217 90 2 21

Pig 13 0.313 180 2 21

Pig 14 / / 0 18

Pig 15 / / 2 21

Table 1. Overview of the treatment strategies of the different pigs with indication of the ablation time (s), the applied power (W/mm2), flowrate (l/min) and stent size (mm).

158

Chapter 7: Immunohistochemical identification of ablation lesions

Pulmonary veins

The sample was placed inside a closed water-perfused circuit and heated wirelessly, by placing a coil around the stent and the surrounding ostium, for different time periods, flow rate and with different power settings.

Histology

Superior vena cava

All pigs were euthanized directly after the ablation procedure. After removal of the heart-lung package, a part of the right auricle and the cranial vena cava, including the stent, were dissected. Hereafter, the stent was removed from the longitudinally opened blood vessel. The part of the vena cava demonstrating the burned marking of the stent was cut out and fixed in 4% formaldehyde solution for 24 hours. Subsequently, the samples were stored in alcohol 70%. Prior to the dehydration steps, the endothelial scar markings were stained with tissue ink (Fig. 1). The dehydration procedure occurred according to standard laboratory procedures and, subsequently, the samples were embedded in paraffin using a Microm tissue processor STP 420D (Prosan, Merelbeke, Belgium) and Microm embedding station EC 350-1 (Prosan), respectively. Sections of 5 μm were cut longitudinally on a Microm H360 microtome (Prosan).

Fig. 1. Ablated superior vena cava sample of pig 1 with an applied endothelial ablation lesion, before (panel a) and after applying the blue tissue ink (panel b), scalebar = 0.5 cm.

Pulmonary veins

The ablated tissue was removed from the stent and fixed, dehydrated, embedded and cut as described above.

Immunohistochemistry

The myocardial sleeve tissue of the cranial vena cava and the pulmonary veins of the different pigs was stained immunohistochemically with the polyclonal myosin marker MYBPC3 (K-16: sc-50115;

159

Chapter 7: Immunohistochemical identification of ablation lesions

Santa Cruz Biotechnology, Santa Cruz, CA, USA) to detect myosin-binding protein C (cardiac type), which is reactive with pig tissue and present in the myofibrils of cardiac and striated muscle tissue (Winegrad, 1999). Immunostaining was performed using the Dako automated Autostainer Plus. No antigen retrieval was needed. Firstly, the sections were incubated for 5 min with 3% hydrogen peroxide and 30 min with rabbit serum. After primary antibody incubation for 60 min, secondary antibody (rabbit/anti-goat, biotinylated, polyclonal, 1:500; Dako, Glostrup, Denmark) was used for 30 min followed by streptavidin and horseradish peroxidase (streptavidin-HRP, 1:1500, Dako) for 30 min. Visualization was performed with DAB (Dako) for 5 min. The immunohistochemical sections were observed under the light microscope (Olympus BX61, Olympus DP73 camera, Olympus, Belgium).

Indirect lesion identification is done as follows. The antibody MYBPC3 positive regions (dark staining) will refer to intact myocardial tissue whereas the ablated parts (lighter staining) will be negative for the immunohistochemical staining.

Software processing

Free software Fiji was used to process and to standardize the images of the immunohistochemical sections by using the same threshold for each section (Fig. 2) (Schindelin et al., 2012). After color deconvolution (for haematoxylin and DAB) and duplicating the image, the threshold was set at zero and 90. Afterwards, the adapted image was overlaid with the original image at 50% opacity and flattened together. The scale bar was set and the images were measured.

Fig. 2. Immunohistological image of an ablated superior vena cava of pig 1 with a transmural lesion before (left) and after (right) Fiji software processing. The ablation lesion is contoured in both images, scale bar = 250 μm.

160

Chapter 7: Immunohistochemical identification of ablation lesions

Results

The Fiji processed images showed the lesions more clearly compared to the original images as demonstrated in Fig. 2. A gradual change of brown shades was observed, depending on the position of the observed spot. If this spot is located close to the position of the applied heat source, a lighter zone is noticed compared to a spot located further away from the heat source (Fig. 3).

All immunohistochemical samples of the different pigs demonstrated clear ablation lesions, except of the two control samples (pigs 14 and 15, Table 1), which are visible on the sections as lighter zones compared to the non-ablated regions. These lighter zones also correspond with the tissue ink present at the endothelial side of the ablated region of the caudal vena cava sections. Some samples displayed transmural lesions (pigs 1, 2, 3, 4, 6, 8 and 13; Fig. 2-4 and 6) whereas other samples demonstrated no transmural lesions with lighter stained myosin-denaturation-patterns at different depths, measured from the endothelium at the level of the applied ink. All samples included, the ablation depth ranged from 0.45 mm to 2.38 mm.

Fig. 3. Panel a: Fiji-software-processed immuohistological image of an ablated superior vena cava of pig 6 with tissue ink applied (insert) at the endothelial side at the level of the heating spot (between two arrows), with the latter identifiable as the myosin proteins are almost all denatured at this level, scalebar=500μm. Panel b: Detail of panel a with the insert indicating the applied ink, scalebar = 100 μm. Panel c: Higher magnification of the insert indicated in panel b with the arrows indicating the applied blue tissue ink, scale bar = 100μm.

161

Chapter 7: Immunohistochemical identification of ablation lesions

If superficially induced charring and/or physiologic connective tissue was present between the myocardial tissue and the heat source, little or no myosin denaturation was observed (Fig. 4). Often, pulmonary vein samples demonstrated several layers of myocardial tissue separated from each other by fat and connective tissue. In that case, no myosin denaturation was noticed in the myocardial sleeve layer located at the adventitial side of the fat whereas at the same level, a transmural lesion could be observed in the myocardial tissue located at the endothelial side.

Fig. 4. Panel a: Fiji-software-processed immunohistological image of an ablated superior vena cava of pig 4 with tissue ink applied at the endothelial side at the level of the heating spot, indicated by the arrows in panel b, scale bar of panel a = 1mm. Panel b: Higher magnification of the delimited area in panel a, indicating both a non-ablated region (1) and ablated region (2), scale bar=100μm. Panel c: Highly enlarged detail of panel a, demonstrating an insulating barrier consisting of charred (1) and connective tissue (2) between the heating area (indicated by the blue ink) and the intact myocardial tissue (3), scale bar = 100μm.

Two samples of pulmonary vein tissue were ablated under the same conditions and with the same power setting. The sample in which no fat was present demonstrated an ablation lesion, more than twice the depth compared to the lesion observed in the sample in which the myocardial sleeve layer was divided in two by an intermediate layer of fat (Fig. 5).

162

Chapter 7: Immunohistochemical identification of ablation lesions

Fig. 5. Both images (a and c) are derived from ablation trials with the same setting and conditions. Panel a: Fiji- softwareprocessed immunohistological image of an ablated pulmonary vein of pig 10. The arrow indicates an ablation lesion of approximately 2.38 mm, L = lumen, scalebar = 1 mm. Panel b: detail of panel a, indicated by the box, in which the loss of staining is visualized, L = lumen, scalebar = 0.5 mm. Panel c: Fiji-software- processed immunohistological image of an ablated pulmonary vein of pig 11. The black arrows indicate two ablation spots (0.6 mm and 1 mm) in myocardial layer 1. The white arrows indicate the fat layer between myocardial layer 1 and 2, L = lumen, scalebar = 1 mm. Panel d: detail of panel c, indicated by the box, in which the lighter staining visualizes the transition of normal tissue to ablated tissue, L = lumen, scalebar = 0.5 mm.

Several immunohistochemical sections of the pulmonary veins provided an ablation pattern which is in accordance with the structure of the stent. The stent is build-up by a number of struts and the diamond-like shape of the struts was recognized in the ablated tissue (Fig. 6).

163

Chapter 7: Immunohistochemical identification of ablation lesions

Fig. 6. Panel a: Fiji-software-processed immunohistological image of an ablated pulmonary vein of pig 12. The section was cut parallel to and 10 μm from the endothelium. The arrows indicate the diamond-shaped ablation lesion which matches perfectly the organization of the struts, scale bar = 2mm. Panel b: detail of panel a, indicated by the box, in which the loss of staining is indicated in higher magnification by the double arrow, scale bar = 0.4mm. Panel c: Fiji-software-processed immunohistological image of an ablated pulmonary vein of pig 13. The section was cut longitudinally and the arrows indicate several transmural ablation lesions at regular distances, due to the structure of the stent. Scale bar = 2mm. Discussion

The standard histological staining methods fail to make a clear distinction between destructed tissue and intact tissue. Moreover, with these techniques it is sometimes hard to visualize objectively the gradual reduction of heat intensity when sampling away from the ablation point. A misinterpretation of the grade of heat penetration, based on a regular histological staining, could be caused by an unequal fixation of the tissue, due to a variable tissue thickness as no control is available. Further, regular histological staining methods are not directly linked with the process occurring in ablated, destructed or denatured tissue but offer an overview of the present tissue types without discriminating between heated and non-heated tissue. The supra-vital staining with tetrazolium is frequently used to distinguish non-vital from vital tissue (Calcutt et al., 1952). However, in the case of determining the grade of transmurality, more detail or histological images on higher magnification are required which is not possible with this supra-vital staining. On the other hand,

164

Chapter 7: Immunohistochemical identification of ablation lesions

immunohistochemical staining with antibody myosin MYBPC3 visualizes exactly the gradual denaturation of the muscle protein myosin, which is, according to Tornberg (2005), the first characteristic of muscle tissue that will be affected during a heating process.

The evaluation of the degree of chronic ablation lesions will offer less difficulties as ingrown connective tissue will replace destructed cardiac tissue. However, in chronic samples, the described immunohistochemical technique will also be valuable as it detects viable myocardial tissue.

The use of a too high or too low temperature during an ablation procedure will fail to achieve a transmural lesion. The application of extreme temperatures or distribution of energy on a small area will lead to superficial endothelial charring, which reduces energy transfer to the underlying cardiac muscle. On the other hand, myocardial sleeve tissue fans out distally, thus heart muscle fibers, covered by fat and connective tissue, could get shielded from the ablating heat source. Some samples, investigated in this study, demonstrated that heat penetration into the venous wall was limited due to the insulating effect of connective, fat and/or charred tissue.

This could be a reason why recurrence of ectopic electrical conductance is frequently seen after ablation procedures. The constitution of the venous wall needs to be considered in case of ablation failure, AF recurrence is noticed or electrical conductance persists.

In conclusion, this study demonstrates an immunohistochemical staining technique with the MYBPC3 antibody to visualize and to assess objectively acute ablation lesions. Moreover, the importance of the insulating role on heat transfer of endothelial charring, fat or connective tissue is emphasized.

165

Chapter 7: Immunohistochemical identification of ablation lesions

References

Accord R.E., van Suylen R.J., van Brakel T.J. and Maessen J.G. (2005). Post-mortem histologic evaluation of microwave lesions after epicardial pulmonary vein isolation for atrial fibrillation. Ann Thorac Surg. 80:881-887.

Calcutt G. (1952). Supra-vital staining of striated muscle with tetrazolium compounds. Nature. 170:42.

Deneke T., Khargi K., Muller K.M., Lemke B., Mugge A., Laczkovics A., Becker A.E. and Grewe P.H. (2005). Histopathology of intraoperatively induced linear radiofrequency ablation lesions in patients with chronic atrial fibrillation. Eur Heart J. 26:1797-1803.

Iwasaki Y.K., Nishida K., Kato T. and Nattel S. (2011). Atrial fibrillation pathophysiology: implications for management. Circulation. 124:2264-2274.

Kowalski M., Grimes M.M., Perez F.J., Kenigsberg D.N., Koneru J., Kasirajan V., Wood M.A. and Ellenbogen K.A. (2012). Histopathologic characterization of chronic radiofrequency ablation lesions for pulmonary vein isolation. J Am Coll Cardiol. 59:930-938.

Oral H., Knight B.P., Tada H., Ozaydin M., Chugh A., Hassan S., Scharf C., Lai S.W., Greenstein R., Pelosi F.Jr., Strickberger S.A. and Morady F. (2002). Pulmonary vein isolation for paroxysmal and persistent atrial fibrillation. Circulation. 105:1077-1081.

166

Chapter 7: Immunohistochemical identification of ablation lesions

Roberts S.A., Diaz C., Nolan P.E., Salerno D.M., Stapczynski J.S., Zbrozek A.S., Ritz E.G., Bauman J.L. and Vlasses P.H. (1993). Effectiveness and costs of digoxin treatment for atrial fibrillation and flutter. Am J Cardiol. 72:567-573.

Schindelin J., Arganda-Carreras I., Frise E., Kaynig V., Longair M., Pietzsch T., Preibisch S., Rueden C., Saalfeld S., Schmid B., Tinevez J.Y., White D.J., Hartenstein V., Eliceiri K., Tomancak P. and Cardona A. (2012). Fiji: an open-source platform for biological-image analysis. Nat Methods. 9:676-682.

Tornberg E. (2005). Effects of heat on meat proteins - Implications on structure and quality of meat products. Meat Sci. 70:493-508.

Vandecasteele T., Boussy T., Philpott M., Clement E., Schauvliege S., Van den Broeck W. , van Loon G., Cornillie P. and Van Langenhove G. (2016). A Preclinical Study of an Implanted Device in the Pulmonary Veins, Intended for the Treatment of Atrial Fibrillation in an Ovine Model. PACE. 39:822- 829.

Winegrad S. Cardiac myosin binding protein C. Circ Res. 1999; 84:1117-1126.

167

168

Chapter 8

Isolation of pulmonary veins using a thermo reactive implantable device with external energy transfer: Evaluation in a porcine model

Adapted from Boussy T., Vandecasteele T., Vera L., Schauvliege S., Philpott M., Clement E., van Loon G., Willenz U., Granada J.F., Stone G.W., Reddy V.Y. and Van Langenhove G. (2018). Isolation of Pulmonary Veins using a Thermo Reactive Implantable Device with External Energy Transfer: Evaluation in a Porcine Model. Pacing Clin Electrophysiol. DOI: 10.1111/pace. 13345.

169

170

Chapter 8: Pulmonary vein isolation in porcine model

Summary

Pulmonary vein isolation (PVI) is a well-established method for the treatment of symptomatic paroxysmal atrial fibrillation (pAF), but is only partly successful with a high rate of electrical reconnection. We introduce a novel technique in which PVI is acomplished by non-invasive heating of a dedicated thermo-response implant inserted into the pulmonary veins (PV), demonstrated in a porcine model. A self-expanding nitinol-based implant was positioned in the common inferior PV of 11 pigs, using a fluoroscopy-guided trans-atrial appendage approach. Ablation was performed through contactless energy transfer from a primary extracorporal coil to a secondary heat ring (HR) embedded in the proximal part of the implant. Electrophysiological conduction was assessed prior to and post ablation, and at 3 months. Histological samples were obtained acutely (n=4) and after 3 months (n=7). In total, thirteen PV implants were succesfully positioned in the inferior PVs of 11 animals. Ablation was performed without injury of adjacent structures. PVI and bidirectional block was electrophysiologically confirmed in all cases immediately at the time of implantation and 3 months later in 7 chronic animals in whom testing was repeated. Marked evidence of ablation around the proximal HR was evident at 3 months post-procedure, with scar tissue formation and only mild neointimal proliferation. Successful PVI can be obtained by external electromagnetic heat transfer to a novel pulmonary vein implant.

171

Chapter 8: Pulmonary vein isolation in porcine model

Introduction

Atrial fibrillation (AF) is the most common arrhythmia in human adults. The overall prevalence of AF is estimated between 1.5% and 2% of the general population (Camm et al., 2012). Furthermore, since AF can also be asymptomatic, the true prevalence of this disease is likely to be underestimated. Because AF is associated with a five-fold increased risk of stroke and a three-fold increased incidence of heart failure, the impact of AF on patient well-being and societal health care expenditures is enormous (Banerjee et al., 2012; Healey et al., 2012). In the last two decades, pulmonary vein isolation (PVI) has become the preferred treatment of patients with pAF. Current PVI techniques show 1-year success rates of 64% to 70%. Success rates appear to be even lower when analyzing long-term results (44% to 54%) (Camm et al., 2012; Banerjee et al., 2012; Healey et al., 2012; Ganesan et al., 2013). In this article we introduce a novel non invasive ablation technique using electromagnetic energy transfer from an extracorporal coil to a self expanding, heatable device permanently positioned inside the pulmonary veins.

Materials and Methods

Implant

We developed a self-expanding nitinol-based implant that ensures permanent fully circumferential contact with the PV wall. Three different regions are distinguishable: the active heating section or heat ring (HR), the connection part and the stabilizing support part (Figure 1). As described below, the implant was placed into the proximal segment of the PV via a fluoroscopy-guided transatrial approach.

Principles of electromagnetical energy transfer (inductive power transmission)

The heating of the implant (and consequently tissue ablation) in the PV is achieved through contactless transfer of energy from a primary transmitter coil (L1 in Figure 1) to a secondary receiver coil (L2 in Figure 1). The energy delivered to the primary coil originates from a 400 V 3-phase mains supply. This supply is transformed by the converter (Easyheat 8310 LI, Ambrell, Scottsville, NY, USA) through the primary coil to generate the required electromagnetic field, which induces a current in the secondary coil. This current then, depending on the electrical properties of the coil material, results in heating of the secondary coil (the implant). Using the converter, the amount of electromagnetic energy generated can be adjusted. The coupling factor (the energy transfer efficiency between coils) is approximately 2.1% at an operating frequency of 370kHz. The coupling

172

Chapter 8: Pulmonary vein isolation in porcine model

factor decreases with the cosine (cos) of the angle between the transmitter and receiver coil. Figure 1 shows the custom-made primary coil.

Fig. 1. Panel 1: Schematic drawing of an inductive coupled power transfer system. L1 is the primary (transmitter) coil with diameter D, L2 is the secondary (receiver) coil with diameter D2. The efficiency of the power transfer depends on the coupling (k) between the coils and their quality (Q). The coupling is determined by the distance between the transmitter and receiver coil (z) and the ratio of D2 /D. The coupling is further determined by the shape of the coils, the angle between them and the used frequency. Panel 2: The custom- made coil (Verhaert NV, Kruibeke, Belgium). The coil has 2 water-cooled copper windings that are electrically insulated. Panel 3: X-ray image of the implant with delivery catheter (A), guidewire (B), heating segment with diamond-like struts that create resistance which allow current to be transformed into heat (C), connecting segment (D) and stabilizing support structure (E). Panel 4: Schematic figure of the temperature probe.

Non invasive temperature-controlled ablation

As the extent of thermal injury is dependent upon the uniformity of contact between the HR and PV tissue, the local temperature achieved and duration of the ablation, we assessed the temperature achieved through temperature sensors located between the heat ring (HR) and the venous tissue (which approximates the actual vein tissue surface temperature). We used the OpSens OTG-M170 fiberoptic sensor (OpSens, Quebec, Canada) for temperature measurement. A two-lumen concept was utilized with the temperature sensor able to be freely re-positioned to localize the optimal position for temperature feedback (Fig. 1). When activating the electromagnet, using a test dose, small temperature increases were sought to determine the optimal position of the temperature

173

Chapter 8: Pulmonary vein isolation in porcine model

sensor with respect to the implant. Using Comsol simulations, we calculated the temperature profile along an axial line between the implant and tissue.

Animal model

The porcine model was used for catheterization of the PVs. The breed chosen was a mixture between Land race and Large white, and the animals were between 3 and 4 months of age, weighing between 60 and 65 kg. This age and size of the pigs was chosen so that they would fit inside the transmitter coil with an inner diameter of 30 cm. Eleven animals were used. All animals were orally preloaded with oral aspirin 300 mg and clopidogrel 300 mg at least 24 hours pre-procedure, and were continued on aspirin 75 mg and clopidogrel 75 mg daily until termination of the trial.

Procedure

Experimental setting

Under general anesthesia, a left thoracic incision in lateral position was made and direct puncture of the left atrial appendage was performed. A sheath was placed over a guidewire. A second guidewire was placed through the sheath. The sheath was taken out of the left atrium and put in place again, but only over one of the two guidewires, thus providing direct sheath access to the left atrium (for placement of the temperature control system) and a guidewire for delivery of the implant. A guiding catheter was brought into the left atrium, and contrast dye was injected to visualize the antrum and pulmonary vein ostia. Both guidewires were positioned distally into the common inferior vein. These images were stored and used as a guide. The outline of the PVs was drawn on the monitor, also as a guide. The PV diameter was then measured angiographically using a validated on-line quantitative system (Siemens, Munich, Germany), and the appropriate size of the implant was chosen. Implants were chosen to be ≥10% larger in diameter than the angiographic measurement.

Implant positioning

The implant was introduced into the PV and positioned at the atrio-pulmonary junction, specifically in the antrum of the PV ostium (at the transition between the ostium and left atrium, and the non- muscular parts of the PV). Positioning and deploying of the implant was performed using angiography. After the implantation of the device, the transmitter coil was positioned in exactly the same plane as the heating segment of the implant.

174

Chapter 8: Pulmonary vein isolation in porcine model

Non invasive ablation

Correct temperature probe positioning was performed using an ablation test dose. When the correct position was confirmed, the temperature probe was fixed, and the calculated ablation power was delivered over three minutes. The power needed was calculated using a proprietary algorithm, which incorporates the size of the implant, the measured expansion, and the calculated deviation from the perfect alignment of device and ablation coil in the same plane. The temperature probe was removed after two minutes, to also ensure ablation of the small space where the temperature probe was positioned between the implant and the PV wall.

Verification

After the ablation, the position of the implant was again checked using angiography, to ensure the device had remained in the same position. Prior to and following the ablation, a 10-pole steerable diagnostic EP catheter (Viacath, Biotronik, Berlin, Germany) was introduced into the pulmonary veins. Distal pacing was performed to check for exit block.

After removal of all catheters, 4 animals were sacrificed immediately, and 7 at three months. The animals that were survived were again catheterized at three months, and conduction block patency checked. Heart and lungs were carefully inspected macroscopically, and specimens of the PVs were histologically evaluated.

Results

Animal characteristics

In total, 11 pigs were treated. In all animals, the antrum of the inferior PV was chosen as ablation target. No major periprocedural complications occurred. A single implant was placed in 10 animals while two implants were placed in one animal. The sequence of the two implants being placed into the two PVs (i.e. left and right) from upper left to lower right is shown in Fig. 2.

175

Chapter 8: Pulmonary vein isolation in porcine model

Fig. 2. Panel1: Implantation sequence with two consecutive implants being placed into both pulmonary veins ostia (left and right) from upper left to lower right. Panel 2: Angiographic measurements of the antrum of the inferior PV complex. (A) shows the reference measurement taken with the injection sheath (7 French) taken as actual reference, (B) shows the actual measurements and (C) the corresponding sites where the measurements were taken.

Implants

Fig. 2 shows an example of angiographic measurements prior to the implantation. Table 1 shows the procedural data. All implants were successfully positioned. In pig number 7, two implants in two different PVs were implanted. In pig number 11, one implant dislocated from the PV due to undersizing of the device, which embolized to the left atrium. Therefore, this animal was sacrificed immediately. Average duration of the procedure was 81 ± 22 minutes including surgery, catheterization, implant delivery and ablation. The implantation and ablation process took 17 ± 6 minutes. After device implantation the ablation coil was positioned over the animals. The coil was positioned such that the ablation coil and the implant were in the same plane. Deviations of the plane were fed into the algorithm described above and power/current delivered was adjusted accordingly. Temperature measurements were performed following a fixed protocol. Table 2 shows the current provided to the ablation coil and the temperatures reached during the procedures. The

176

Chapter 8: Pulmonary vein isolation in porcine model

average temperature increase generated in the vessel wall was 8.9 ± 6.8 °C. Duration of ablation was 200 ± 201 seconds.

Pig Age Size Common PV size Implant size Successful Total numbe (months) (kg) (diameter, mm) (mm) placement of procedural time r implant (min) 1 3 60 14 25 Yes 80

2 3 62 18 25 Yes 75

3 3 60 21 30 Yes 55

4 3 65 27 30 Yes 90

5 3 63 22 25 Yes 80

6 3 61 23 25 Yes 60

7 (*) 3 65 14 25 Yes 115

15 25 Yes

8 3 62 21 25 Yes 85

9 3 63 13 25 Yes 125

10 3 63 22 25 Yes 70

11 (**) 3 61 23.5 (average of 30 Yes (**) 60 20 (distal) and 27 (proximal))

Table 1. Procedural data of the 11 treated pigs. (*) In pig number 7 two implants were positioned into two different pulmonary veins. (**) First implant in this animal dislocated from the PV into the left atrium.

177

Chapter 8: Pulmonary vein isolation in porcine model

Pig Core body Positioning Max Duration number temperature VWT test dose VWT Δ °C Current of (°C) pre(°C) current peak(°C) (amp) ablation (amp) (sec) 1 33.4 39.0 148.8 58 25.6 284.0 180

2 33.8 None 150.4 40 6.2 302.4 180

3 34.3 35.7 165.9 40 5.7 275.1 180

4 33.5 35.2 243.6 38 4.5 390.0 180

5 35.9 Dislocated NA NA NA 396.0 180 temp. sensor

6 35.1 36.0 207.9 39.0 3.9 401.1 180

7 35.1 36.5 310.8 40.0 4.9 399.0 180

35.1 36.0 310.8 37.0 1.9 399.0 180

8 35.5 35.8 140.8 49 13.5 300.8 300

9 36.0 42.0 201.6 49 13.0 270.4 300

10 35.4 39.0 150.4 52 13.0 284.0 180

11 33.7 35.0 133.3 40 6.3 354.1 180

Table 2. Temperature and data during testing and ablation. Core body temperature= anal measurement of pig body temperature. VWTpre: vessel wall temperature prior to ablation. VWT-peak: peak temperature during ablation. C: Temperature increase.

Electrophysiological testing

Prior to ablation, PV signals were detected distally from the HR in all animals, demonstrating intact conduction. After ablation, the same location was checked for entry- and exit block. Bidirectional block was confirmed from the 12 tested PVs in 11 animals. Electrophysiologic (EP) characteristics are shown in Table 3.

178

Chapter 8: Pulmonary vein isolation in porcine model

Pig number Signals Conduction Conduction Long term recorded of paced of paced persistence of pre- signals pre- signals post- conduction ablation ablation ablation block (3 months) 1 + + - +

2 + + - +

3 + + - +

4 + + - +

5 + + - +

6 + + - +

7 + + - +

+ + - +

8 + + - NA

9 + + - NA

10 + + - NA

11 + + - NA

Table 3. Results of sensing and pacing from a distal focus in the pulmonary vein before and after implantation and heating of the device.

Acute tissue injury evaluation

Four animals were sacrificed immediately, and seven were kept alive for 3 months. In the animals that were sacrificed immediately, the ablation zone was evaluated after removal of the implants for gross anatomical changes. A typical ablation zone is shown in Fig. 3. No damage to the surrounding structures was observed in any of the 4 acute animals, nor in the 7 chronic ones.

Histology

Histological data of the 7 animals at 3 months post procedure showed marked evidence of ablation around the proximal HR with loss of tissue architecture and scar tissue formation. There was no inflammation or necrosis adjacent to the mid and distal section of the implant. A mild degree (0.4 ± 0.7 mm) of neointimal proliferation was noted (Fig. 3). Fig. 3 demonstrates Mallory-Azan-stained ablated myocardial sleeve tissue, in which ingrown fibrous connective tissue, due to the ablation procedure, is indicated by blue staining.

179

Chapter 8: Pulmonary vein isolation in porcine model

Fig. 3. Panel 1: View from the left atrium after sacrifice of the animal. (A) shows the left atrium with the antrum of and entrance to the inferior pulmonary vein; (B) shows the pulmonary vein implant in situ. Panel 2: Left atrium opened from the anterior side. Mitral valve (A); left atrium (B); original position of the implant with the diamond-shaped struts of the ablation ring (C); ablation zone in the pulmonary vein revealing the same diamond-like shape (D); distal side of the pulmonary vein (E). Panel 3: Imprint of the implant after ablation and removal of the implant. A continuous circumferential ablation zone in the transition zone between pulmonary vein and left atrium is seen. The diamond-like shape of the ablation zone is clearly visible. Panel 4: Detailed section of the ablated pulmonary vein of pig number 2. Clear demarcation of the ablation effect of a single strut of the implant is visualized by white arrows. Panel 5: Representative histologic images at 3 months. A: Histological samples alongside the proximal, mid and distal region of the implant. B: Pulmonary vein, Proximal section showing loss of tissue architecture. C: Pulmonary vein, mid section, revealing no inflammation or necrosis and mild neointimal formation. D: Pulmonary vein, distal section, without inflammation or necrosis and mild neointimal formation. Panel 6: Histologic image of myocardial sleeve tissue of the inferior PV. Ingrown fibrous connective tissue (stained blue) was present as reaction on the ablation procedure, myocardial tissue is stained orange to red. Top side is the lumen, Mallory-Azan staining, scale bar is 500µm.

180

Chapter 8: Pulmonary vein isolation in porcine model

Discussion

In this report, we have shown for the first time that complete PVI using an externally applied energy source is safe and feasible via a novel catheter-delivered implant. In this experiment, 11 animals received nitinol self-expanding implants into the antrum of the inferior pulmonary vein without periprocedural complications. In all animals, the current test allowed establishing an optimal temperature-sensing zone, after which a full ablation current was applied. All treated PVs were electrically isolated without vascular necrosis or other evidence of damage to adjacent structures. No PV electrical reconnections occurred 3 months post ablation in 7 studied animals.

Ablation Target

In contrast to current ablation modalities, this new approach targets the PVs via a permanent implant. PVI is achieved by uniform heating of the implant’s proximal ring. Preliminary histological results show only minimal intimal proliferation, suggesting that PV stenosis as a potential complication may be infrequent (Dill et al., 2003; Lu et al., 2015). As a result of expansion of the vein during implant deployment a slight increase in PV size was seen, despite scar formation on the heating ring (HR) 3 months post ablation.

Precise positioning of the HR potentially enables exact ostial isolation of the PVs in a safe manner. Current (antral) ablation techniques are often challenged by considerable differences in tissue thickness (posterior wall versus LAA ridge) (McLellan et al., 2015). If the perivenous myocardial tissue around the PV itself can be targeted, application of a short uniform circumferential heating protocol should be sufficient to achieve PVI, as observed herein.

Potential advantage of a permanent implant

Currently, AF recurrence following PVI is thought to be mainly due to gap formation in the ablation zones or tissue recovery (Santangeli et al., 2015). Contributing factors are both patient related (PV anatomy, individual scar formation) and technique related (radiofrequency vs cryoablation, operator skill, etc.) (Kuck et al., 2016). In the present study, we sought to evaluate the PV reconnection rate in the presence of permanent implants, positioned slightly more to the ostial side of the traditional RF ablation zone. Theoretically, the chronic presence of the device would facilitate repeated electromagnetic heating should AF recur.

181

Chapter 8: Pulmonary vein isolation in porcine model

Externally delivered ablative energy

For the first time, we have demonstrated a method wherein the energy required for ablation is delivered from an external source to an indwelling implant, in contrast to current clinical strategies were radiofrequency or cryoablative energy is delivered locally from an indwelling catheter. The potential advantages of this new technique are multiple and include: a) The amount of energy delivered can be tailored individually, depending on patient characteristics, size and orientation of the implants. b) The adjustment and calculation of the magnetic field results in focused heating of the implant without damaging surrounding structures. c) A short ablation time (±200 sec) is needed for effective lesion formation. d) Ablation may be performed offline. In case of recurrence, repeat ablation could be performed in an ambulatory setting. e) Since the ablation target is located more distally in comparison to conventional techniques, formation of atrioesophageal fistulas (AEF) is less likely to occur (Chavez et al., 2015).

Limitations

This article describes the feasibility of a new technique with the potential to disrupt the way PVI procedures are performed. Especially the potential to simplify the procedure minimizing the use of complex mapping and ablation techniques are appealing. However, there are several limitations that deserve to be highlighted: a) Separate energy delivery is needed for each implant, since setting of the magnetic field and positioning of the external coil need to be adjusted to each implant’s position and orientation. This could be completed in one procedure, however, as each ablation session is short. b) Although the ablation itself is not invasive, positioning of multiple devices is. Nevertheless, shorter procedure times are possible than for current techniques. c) This technique may not be sufficient for patients with persistent or long-standing persistent AF, since the underlying mechanism of AF may not be confined to the PVs (Climent et al., 2015). d) Although all 7 chronic animals were in apparent good health before sacrifice, we did not actively investigate the possibility of phrenic nerve injury.

182

Chapter 8: Pulmonary vein isolation in porcine model

e) AF ablation is not limited to PVI but this poses a limitation for this technology as no additional ablation lines can be applied with this technique. f) In our study, we cannot conclude whether the stent implantation or ablation procedure was associated with thrombus formation. Therefore, prior to human trial initiation, further investigations using brain MRI are needed to assess the risk of thrombo-embolization and potential brain injury.

Future Development

Several improvements in this technology are under investigation and need to be implemented in future studies. First, in this study, the pulmonary veins were approached through the transatrial approach. This will need to be adapted to a transseptal approach (Bollmann et al., 2015). Secondly, we implanted one single device per animal, while in humans up to 4 implants will be needed to achieve full PVI. However, in contrast to arterial stents that contain a considerable amount of metal to achieve sufficient radial support (and provoke substantial vascular responses), the present implants target expansion of venous structures. Therefore, these implants imply a much lower metallic burden and radial force and neointimal hyperplasia has been minimal.

In our experiment, the temperature sensor was placed adjacent to the implant, and was removed during the procedure. This may prove to be impractical when implants need to be placed into several PVs during the same procedure (as the temperature probe needs to be positioned through a separate guiding catheter for each PV). Other means for assessing the amount of energy delivered locally to the implant (and the temperature generated) are currently being developed. Flexible micro- electronics could be added to the implant, allowing real-time in situ temperature feedback during ablation, electrical signal recognition and even detection and monitoring of acute and chronic intracardiac electrocardiographic signals (to derive a 12-lead ECG), to monitor for AF recurrence, etc. Position and orientation of the implant with respect to the transmitter coil is of high importance. Currently, it is difficult to find the optimal position of the transmitter coil for efficient and homogenous heating of the implant. Three dimensional mapping or other visualization techniques should improve or overcome this limitation.

The entire procedure took on average 81 minutes, including positioning of the ablation coil around the animal, which proved cumbersome and often required 30-40 minutes. Significantly shorter procedure times are expected in the future. However, complete electrical isolation would require an implant in each PV. In our study, conduction block acutely and at 3 months was demonstrated via

183

Chapter 8: Pulmonary vein isolation in porcine model

pacing on both sides of the ablation ring. Long term validation is needed using a 3D electromagnetic mapping system to confirm a closed circumferential ablation line as well as bidirectional block.

Conclusion

In the present study, we have demonstrated that implantation of a self-expanding nitinol-based device in the PV is safe and feasible, and that energy can be delivered to the implant in a noninvasive, controlled manner, achieving bidirectional conduction block without complication. Compared to contemporary techniques to achieve AF ablation, this device may provide advantages including ease of use, consistency of effect and repeatability, if necessary. Further device iterations and technique modifications are ongoing to improve the utility of this novel approach.

184

Chapter 8: Pulmonary vein isolation in porcine model

References

Banerjee A., Taillandier S., Olesen J.B., Lane D.A., Lallemand B., Lip G.Y. and Fauchier L. (2012). Ejection fraction and outcomes in patients with atrial fibrillation and heart failure: the Loire Valley Atrial Fibrillation Project. Eur J Heart Fail. 14:295 –301.

Bollmann A., Kosiuk J., Hilbert S., John S. and Hindricks G. (2015). Percutaneous transapical access for pulmonary vein mapping and ablation in a porcine model with a new high-density electroanatomical mapping system. Int J Clin Exp Med. 8:12631-12636.

Camm A.J., Lip G.Y., De Caterina R., Sa`velieva I., Atar D., Hohnloser S.H., Hindricks G. and Kirchhof P. (2012). Focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation. Developed with the special contribution of the European Heart Rhythm Association. Eur Heart J. 33:2719-2747.

Chavez P., Messerli F.H., Casso D.A., Aziz E.F., Sichrovsky T., Garcia D., Barrett C.D. and Danik S. (2015). Atrioesophageal fistula following ablation procedures for atrial fibrillation: systematic review of case reports. Open Heart. 2:doi:10.1136/openhrt-2015-000257.

Climent A.M., Guillem M.S., Atienza F. and Fernández-Avilés F. (2015). Electrophysiological characteristics of permanent atrial fibrillation: insights from research models of cardiac remodeling. Rev Cardiovasc Ther. 13:1-3.

Dill T., Neumann T., Ekinci O., Breidenbach C., John A., Erdogan A., Bachmann G., Hamm C.W. and Pitschner H.F. (2003). Pulmonary vein diameter reduction after radiofrequency catheter ablation for paroxysmal atrial fibrillation evaluated by contrast-enhanced three-dimensional magnetic resonance imaging. Circulation. 107:845-850.

Ganesan A.N., Shipp N.J., Brooks A.G., Kuklik P., Lau D.H., Lim H.S., Sullivan T., Roberts-Thomson K.C. and Sanders P. (2013). Long-term outcomes of catheter ablation of atrial fibrillation: A Systematic Review and Meta-analysis. J Am Heart Assoc. 2:doi:10.1161/JAHA.112.004549.

Healey J.S., Connolly S.J., Gold M.R., Israel C.W., Van Gelder I.C., Capucci A., Lau C.P., Fain E., Yang S., Bailleul C., Morillo C.A., Carlson M., Themeles E., Kaufman E.S. and Hohnloser S.H. (2012). ASSERT Investigators. Subclinical atrial fibrillation and the risk of stroke. N Engl J Med. 366:120–129.

Kuck K.H., Hoffmann B.A., Ernst S., Wegscheider K., Treszl A., Metzner A., Eckardt L., Lewalter T., Breithardt G. and Willems S. (2016). Impact of complete versus incomplete circumferential lines

185

Chapter 8: Pulmonary vein isolation in porcine model

around the pulmonary veins during catheter ablation of paroxysmal atrial fibrillation: Results From the gap-atrial fibrillation-german atrial fibrillation competence network 1 Trial. Circ Arrhythm Electrophysiol. 9:doi:10.1161/CIRCEP.115.003337.

Lu H.W., Wei P., Jiang S., Gu S.Y., Fan L.C., Liang S., Ji X., Rajbanshi B. and Xu J.F. (2015). Pulmonary vein stenosis complicating radiofrequency catheter ablation: five case reports and literature review. Medicine. 94:doi: 10.1097/MD.0000000000001346.

McLellan A.J., Ling L.H., Azzopardi S., Lee G.A., Lee G., Kumar S., Wong M.C., Walters T.E., Lee J.M., Looi K.L., Halloran K., Stiles M.K., Lever N.A., Fynn S.P., Heck P.M., Sanders P., Morton J.B., Kalman J.M. and Kistler P.M. (2015). A minimal or maximal ablation strategy to achieve pulmonary vein isolation for paroxysmal atrial fibrillation: a prospective multi-centre randomized controlled trial (the Minimax study). Eur Heart J. 36:1812-1821.

Santangeli P. and Lin D. (2015). Catheter ablation of paroxysmal atrial fibrillation: Have we achieved cure with pulmonary vein isolation? Methodist Debakey Cardiovasc J. 11:71-75.

186

Chapter 9

The pulmonary veins of the horse

Adapted from Vandecasteele T., van Loon G., Vandevelde K., De Pauw B., Simoens P. and Cornillie P. (2016). Topography and ultrasonographic identification of the equine pulmonary vein draining pattern. Vet J. 210:17-23.

187

188

Chapter 9: The pulmonary veins of the horse

Summary

In human echocardiography, pulmonary venous flow is monitored to assess different pathologies such as mitral valve regurgitation. Hardly any information on ultrasound examination of equine pulmonary veins is available due to a lack of in-depth anatomical information, which is currently mainly restricted to the knowledge that each pulmonary vein drains a specific lung lobe region, after which those different veins merge together into a collecting antrum, before opening up into the left atrium through the respective ostium. The aim of this study was, by using anatomical dissection and silicone casting of equine cardiopulmonary sets, to study the venous drainage of both lungs and the position of the different ostia and to investigate whether the different ostia can be identified on ultrasound. Three out of the four ostia can be observed echocardiographically in the standing horse. The ostium draining the most caudal parts of both lungs showed little variability, while the ostium draining the rest of the right lung can be used as an easily recognisable landmark since it is located adjacent to the interatrial septum. The identification of the different equine pulmonary vein ostia on ultrasound might allow further determination of size and flow patterns in the assessment of cardiovascular disease.

189

Chapter 9: The pulmonary veins of the horse

Introduction

In horses, echocardiography is routinely performed in cases of cardiac disease (Marr, 1994). In human and small animal medicine ultrasonographic measurement of pulmonary vein diameter and flow has been used for the diagnosis of ventricular diastolic dysfunction and mitral valve regurgitation (Chiang et al., 1998; Rusconi et al., 2001; Weidemann et al., 2013). A similar assessment in horses is hampered by limited anatomical and echocardiographic information available about the equine pulmonary veins (Nathan, 1970; Barone, 1997). Anatomical characterisation of the pulmonary vein layout is needed to allow echocardiographic identification and examination of the pulmonary veins.

The first goal of the cadaver study was to portray the terminal portions of the equine pulmonary veins and the orifices through which they drain into the left atrium so they could be used as anatomical landmarks to guide the ultrasonographic exploration. Typically, different pulmonary veins merge together before draining through a common orifice or ostium in the left atrium. Before emptying through its ostium into the left atrium, the set of pulmonary veins associated with each ostium generally coalesces slightly proximal to this common orifice, creating a terminal common venous space, indicated as the antrum. As such, the initial focus in this study was to describe the common antra and ostia.

In a second step, the drainage pattern of each pulmonary vein, extending from the respective lung lobes to the ostium through which it drains, was assessed to offer anatomical insights in the flow distribution pattern in the pulmonary venous tree. This data, along with the topographical insights obtained from the first part of this study, were finally used to develop and validate a standard approach for the ultrasonographic visualisation of these structures in the standing horse.

Materials and Methods

Number of dissected and casted cardiopulmonary sets

Thirty-five horse cadavers, euthanised for various reasons not related to cardiovascular or pulmonary disorders, and donated by the owners for educational and scientific purposes, were used. Eighteen cardiopulmonary sets were anatomically dissected and seventeen sets (Table 1) were casted. Fifteen casts were made with silicone (HT33, Zhermack), of which three were compared with the in-vivo ultrasound recording. Two cardiopulmonary sets of ponies were casted in-situ with Technovit (Technovit 7001, Heraeus Kulzer) to preserve the 3-D anatomy, of which one was compared with the

190

Chapter 9: The pulmonary veins of the horse

post-mortem echocardiography. The current study was performed following the guidelines of the Ethical Committee of the Faculty of Veterinary Medicine, Ghent University, Belgium.

Anatomical dissection of the left atrium and the pulmonary veins

Cardiopulmonary sets from 18 horses were examined by anatomical dissection. First, the wall of the left ventricle along with the mitral valve were excised in order to get an overview of the dorsal wall of the left atrium. The different pulmonary vein ostia were probed with a forceps to localise the draining area of the respective veins.

Casting of the pulmonary veins and arteries

To determine the number of pulmonary vein ostia and antra and to record the variation in the branching pattern of the veins, silicone was used as described by Vandecasteele et al. (2015).

Two-components white silicone (base and catalyst, 1:1) was coloured by adding a blue dye to inject the pulmonary veins and a red dye to demonstrate the pulmonary arteries or the aorta. In situ casting with Technovit was performed to determine the exact 3-D position of the pulmonary veins (Table 1).

For casting the pulmonary veins, the left ventricle was opened and rinsed with tap water after removing the blood clots. After draining of the water, the heart was suspended upside-down and the casting material was poured through the opening in the left ventricle into the left atrium.

For the in situ casting, left thoracotomy was performed to gain access to the heart. After incision of the pericardium, the heart was opened, rinsed and subsequently fixated to the sternum with wire. Afterwards, the ponies were turned into dorsal recumbency to be able to pour the Technovit into the heart similar to the procedure as described for the silicon casts.

191

Chapter 9: The pulmonary veins of the horse

Animal Age Breed Sex Bodyweight Cast Amount (ml) (years) (kg) structures of silicone or Technovitc Silicone casts Cast 1 Foal 0.5 / / 200 PV 600 Cast 2 Pony / / / 150 PV 400 Cast 3 Horse 10 WB / / PV, PA 1400 Cast 4 Horse 17 WB Female 570 PV 1000 Cast 5 Horse 25 WB Male 580 PV 1200 Cast 6 Horse 19 / Female 406 PV 1500 Cast 7 Horse 6 TB Male 475 PV 1300 Cast 8 Horse 17 WB Female 640 PV 1400 Cast 9 Pony 15 / Female 520 PV 2100 Cast 10 Horse 25 WB Male 668 PV 1100 Cast 11 Horse 10 WB Female 620 PV 1600 Cast 12 Horse 18 WB Male 536 PV, PA, Ao, 5400 CR, CD Cast 13 Horseb / / Male 530 PV, PA 1400 Cast 14 Horseb / WB Female 410 PV, A 1000 Cast 15 Horseb / WB Male 494 PV, A 1500 Technovit casts

Cast 16 Pony / / Male 150 PV 550 Cast 17 Ponya,b 22 Shetland Female 206 PV 475

Table 1. Animal and product data for the cast structures. WB, Warmblood horse; TB, Thoroughbred; PV, pulmonary veins; PA, pulmonary arteries; Ao, aorta; CR, cranial vena cava; CD, caudal vena cava; /, unknown. a Fixed in 4% formaldehyde, b Examined echocardiographically, c Amount of silicone (Casts 1–15) or Technovit (Casts 16 and 17).

Echocardiography

Of four of the horses, echocardiographic recordings from a left and right parasternal window, obtained earlier during clinical examination of the living animal, were available from the database of the Clinical of Large Animal Internal Medicine (Ghent University) and compared with the corresponding casts afterwards (table 1).

The cadaver of one of those four horses was fixated by perfusion with a 4% formaldehyde solution to perform a post-mortem in-situ ultrasound recording directly on the heart. After the fixated cadaver

192

Chapter 9: The pulmonary veins of the horse

was hoisted in upright position, the left front leg and the left wall of the thorax were removed. An isotonic saline solution was pumped through the jugular veins to create a flow through the heart during ultrasonographic examination. Echocardiography was performed by placing the transducer directly on the heart (GE Vivid 7 Dimension ultrasound with 3S phased array transducer at 1.7/3.4 MHz, GE Healthcare).

With the information of the casting study, the left atrium was scanned in multiple planes in order to visualise different pulmonary veins, antra and ostia. Recorded images were compared to the casts of the same horse for correct identification of the anatomical structures. From these data, optimal probe position, rotation and angulation were determined.

Results

Anatomical description of the pulmonary veins (see Fig. 1 and Appendix)

Position of the pulmonary vein ostia

Typically, four major ostia could be discerned in the dorsal region of the left atrium between the left auricle and the interatrial septum. In a single cast (cast 1, Table 1), only three ostia were observed. Separate orifices draining the vein coming from the accessory lung lobe were not denoted as true ostia.

A first ostium (ostium I) is located most caudally and close to the left auricle, normally draining the cranial parts of the left caudal lung lobe (Fig. 2). In case ostium I occurs as an unsegmented single entity (n = 15), this ostium is the orifice of a caudodorsally orientated antrum (Figs. 1c and 2). In one horse ostium I was split up, and therefore represented by two separate openings, with the orifice of LcdV1 (left caudal pulmonary vein (cranial part)) (Fig. 3) closest to ostium IV and the orifice of LcdV2 (left caudal pulmonary vein (intermediate part)) located closer to the left auricle (Figs. 1a and 3). In cast 1, as an exception, LcdV1 and LcdV2 drained into the second ostium (ostium II) without forming a separate ostium I.

Ostium II is the larger opening and drains the caudal segments of both lungs. Ostium II is positioned to the right side of ostium I (Figs. 1 and 2). Through this ostium, the antrum of LcdV3 (left caudal pulmonary vein (caudal part)) and RcdV4 (right caudal pulmonary vein (caudal part)) debouches into the left atrium (Fig. 3). In most cases, ostium II is seen as a single orifice because the point of confluence of the two draining branches is located more distally. The depth of this antrum varied considerably according to the point where LcdV3 and RcdV4 fused. When ostium II is split up into

193

Chapter 9: The pulmonary veins of the horse

two almost equal outlet openings, the orifice of LcdV3 is located near ostium I and the orifice of RcdV4 is situated near ostium III (Fig. 1).

The third ostium (ostium III), draining the cranial and middle segments of the right lung, is positioned rightmost in the left atrial roof, cranial and to the right of ostium II (Figs. 2 and 4). The right border of ostium III fuses with the interatrial septum (Figs. 1 and 2). Three (n = 9) or four (n = 8) pulmonary veins discharge into the antrum of ostium III (Figs. 1d and 2). The distance between their debouchment into the antrum and the ostium is variable (Fig. 1a, c and d). RcdV3 (right caudal pulmonary vein (intermediate ventral part)), draining the ventral middle part of the right caudal lung lobe, follows the direction of RcdV4 and debouches into the largest orifice of antrum III located closest to ostium II.

Fig. 1. (a) Ventral view of the ostia of the pulmonary veins, as seen from within the left ventricle with a view of the roof of the left atrium. The top of the picture is oriented caudally, the bottom is oriented cranially. The interatrial septum is located on the right; the left auricle is on the left. The accessory pulmonary vein (AccV) emerges as separate openings. (b) Ventral view of ostium 1 with the openings of the AccV. (c) Ventral view of ostia 1, 2 and 3 with the openings of AccV located inside the antrum of ostium 2. (d) Ventral view of ostium 3 with right caudal pulmonary vein (intermediate ventral aspect, RcdV3; intermediate dorsal aspect, RcdV2; cranial aspect, RcdV1) and right cranial pulmonary vein, RcrV.

The fourth ostium (ostium IV), of which the antrum is orientated craniodorsally, is positioned on the left side of the left atrium close to the left auricle and cranial to ostium I (Figs. 1 and 2).

194

Chapter 9: The pulmonary veins of the horse

Fig. 2. (a) Dorsal view of the heart. (b) Dorsal view of the ostia and pulmonary veins (blue) and the pulmonary arteries (red). Left (c) and caudal (d) view of the heart: 1, cranial vena cava; 2, caudal vena cava; 3, aorta; 4, pulmonary trunk; 5, right auricle; 6, left auricle; I, ostium 1; II, ostium 2; III, ostium 3; IV, ostium 4; AccV, accessory pulmonary vein.

Drainage of the different lung lobes

The left cranial lung lobe is drained by LcrV (left cranial pulmonary vein) (Fig. 3) which consists of three or four branches discharging into the antrum of ostium IV. In certain cases a two-, three- or four-fold antral division could be noticed close to the atrium, but usually the various branches converged more distally (Fig. 1, 3). Nevertheless, an antrum was observed in all casts, formed by the convergence of one large branch and three (n = 8) or two (n = 4) smaller branches or another, single large branch (n = 3). Two casts could not be assessed because ostium IV was not sufficiently filled with the casting medium.

195

Chapter 9: The pulmonary veins of the horse

The drainage pattern of the left caudal lung lobe can be divided into three areas, being a cranial, middle and caudal part (Fig. 4). The cranial and middle parts of this lobe were drained by LcdV1 and LcdV2 respectively. Usually, both veins merge at some distance from or quite close to ostium I, resulting in a larger or smaller antrum, respectively (n = 10) (Figs. 3 and 4). Typically, the debouchment of LcdV1 is located close to the roof of the left atrium. Less frequently, no antrum is formed as LcdV1 remains separate from LcdV2, resulting in two distinct ostia (n = 7) (Fig. 1). The caudal part of the left caudal lung lobe is drained by LcdV3. The latter fuses with RcdV4, which drains the caudal part of the right lung lobe, and both are collected in the antrum of ostium II. Their point of confluence can be either at some distance from or very close to ostium II, resulting in a larger (n = 12) or an almost absent (n = 5) antrum, respectively. In the latter case, ostium II is divided into two separate openings.

The cranial part of the right caudal lung lobe is drained by RcdV1 (right caudal pulmonary vein (cranial part)) and the middle part of the right caudal lung lobe is dorsally drained by RcdV2 (right caudal pulmonary vein (intermediate dorsal part)) and ventrally by RcdV3. These three branches fuse with RcrV (right cranial pulmonary vein) draining the right cranial lung lobe, to create the antrum of ostium III (Fig. 4). RcdV3 represents a large orifice in this antrum, located adjacent to ostium II, and is the last vein to fuse with the other three veins (RcrV, RcdV1 and RcdV2) which debouch through a common part into the antrum of ostium III (Fig. 1d). RcdV2, draining the dorsal part of the caudal lung lobe, is found nearest to RcdV3 and its orifice is located most caudally. RcdV1 is situated in between RcdV2 and RcrV, the orifice of the latter being positioned closest to the interatrial septum (Figs. 1 and 3). In nearly all casts, the common part collecting the three smaller branches (RcdV1, RcdV2 and RcrV) was very short, whilst only a single cast demonstrated a longer and large common branch alongside an equally sized RcdV3.

The accessory lung lobe demonstrated a highly variable drainage pattern (Figs. 1 and 4). This lobe is drained by AccV (accessory pulmonary vein), which consists of several branches either opening through separate orifices directly into the left atrium or through ostium I and/or II. Most often, the orifices of AccV branches were located separately in the roof of the left atrium between and caudal to ostium I and II (n = 8) or caudal to ostium II (n = 3). In a few cases, the accessory lung lobe was drained either through ostium I (n = 2) or ostium II (n = 1) or through both ostia (n = 1). Two casts demonstrated a draining pattern which combined all previous alternatives.

196

Chapter 9: The pulmonary veins of the horse

Fig. 3. Ventral view of the ostia and pulmonary veins (blue) and the pulmonary arteries (red). LcrV, left cranial pulmonary vein; LcdV1, left caudal pulmonary vein (cranial aspect); LcdV2, left caudal pulmonary vein (intermediate aspect); LcdV3, left caudal pulmonary vein (caudal aspect); I, ostium 1; II, ostium 2; III, ostium 3; IV, ostium 4 (located dorsally); RcrV, right cranial pulmonary vein; RcdV1, right caudal pulmonary vein (cranial aspect); RcdV2, right caudal pulmonary vein (intermediate dorsal aspect); RcdV3, right caudal pulmonary vein (intermediate ventral aspect); RcdV4, right caudal pulmonary vein (caudal aspect); AccV, accessory pulmonary vein.

Fig. 4. Dorsal view of the lung lobes. The letters indicate the draining pattern of each of the lung lobes (a, ostium 1; b, ostium 2; c, ostium 3, d, ostium 4; e, separate orifices draining the accessory lung lobe, ostium 1, ostium 2, or ostia 1 and 2).

197

Chapter 9: The pulmonary veins of the horse

Anatomical comparison between the arterial and venous pulmonary circulation (Fig. 2)

The proximal part of the pulmonary arterial circulation is located dorsal to the pulmonary veins. The division of the pulmonary trunk into the right and left pulmonary arteries is located dorsal to the angle between RcdV4 and RcdV3, dorsal to ostium II and III.

The right pulmonary artery continues dorsal to and together with RcdV3 in distal direction, covering antrum III dorsally. The left pulmonary artery proceeds caudodorsally, covering antrum II or the separated LcdV3 and RcdV4 dorsally. Further caudally, the left pulmonary artery continues dorsally along with LcdV2.

Ostium IV is located adjacent to the left pulmonary artery at the level of the bifurcation of the pulmonary trunk and the corresponding antrum is crossed perpendicularly by the pulmonary trunk.

Echocardiographic identification of the pulmonary veins

From a right parasternal approach, the four chamber view allowed visualisation of ostium III and its antrum adjacent to the interatrial septum with minimal cranial probe angulation (Fig. 5A). This view also allowed to identify ostium II. Best images were obtained from the left third intercostal space. Starting from the left atrial and left ventricular long axis view, angulating the probe slightly cranially, ostium III was clearly visible (Fig. 5B). Rotating the probe to the 1 or 2 o’clock position and angulating it slightly more caudally allowed to identify ostium II (Fig. 5C). On a short axis, rotating the probe to the 2 o’clock position and angulating it dorsally, ostium III and ostium II were visualised adjacent to each other (Fig. 5D). Ostium IV was visible on a left short axis view rotating the probe to the 2 or 3 o’clock position and with marked dorsal probe angulation. On this image, ostium IV was found to the left of the left pulmonary artery (Fig. 5E). Ostium I was shown on echocardiographic images obtained during postmortem in situ ultrasound directly on the heart. Investigation of the position of the pulmonary vein ostia in ventro-dorsal direction reveals the most ventral position for ostium III, followed by ostium II and I with ostium IV located most dorsally.

198

Chapter 9: The pulmonary veins of the horse

Fig. 5. (a) Ostium 3 adjacent to the interatrial septum on the right parasternal 4-chamber view. Ostium 2 is also visible. (b) Ostium 3 on the left parasternal long-axis view. (c) Ostium 2 on the left long-axis view. Compared to the left long-axis view, the probe is angulated slightly more cranially and rotated to the 1 o’clock position. (d) On the left short-axis view, with dorsal probe angulation and rotation to the 2 o’clock position, ostium 3 and ostium 2 were visualised adjacent to each other. (e) Ostium 4 was visualised from the left short-axis view with probe rotation to the 2 or 3 o’clock position and with marked dorsal probe angulation. LPA, left pulmonary artery; PA, pulmonary artery; IV, ostium 4; LA, left atrium.

199

Chapter 9: The pulmonary veins of the horse

Discussion

In the present study, the drainage pattern of the equine pulmonary veins from the different lung lobes to the respective opening in the left atrium is demonstrated. This overview provides the anatomical basis for further research regarding pulmonary vein size and flow pattern and development of new treatment strategies such as ablation of tachyarrhythmia and mapping techniques. Clinically, atrial fibrillation is the most important arrhythmia in horses (Reef et al., 1995; De Clercq et al., 2008a). The anatomical and histological basis for developing AF has not completely been elucidated yet, although many predisposing factors have been identified (Kiryu et al., 1999; Gelzer et al., 2000; van Loon et al., 2002; Declercq et al., 2008b; Mont et al., 2008; Leroux et al., 2013). In man, the role of myocardial sleeves, extending into the wall of the pulmonary veins and acting as sources of ectopic electrical pulses has since long been recognised (Haïssaguerre et al., 1998, Chen et al., 1999, Saito et al., 2000, Roux et al., 2004).

In the context of cardiac ultrasound imaging, the left pulmonary artery can be used as an important reference point. While scanning in dorsal direction, LcrV is seen between the left auricle and the left pulmonary artery. Hence, ostium IV, in which LcrV drains, can be easily recognised. Ostium III is located adjacent to the interatrial septum which constitutes a good basis during echocardiography to recognise ostium III, and subsequently ostium II (Fig. 1, 2, 8). Notwithstanding, ultrasound visualisation of the ostia might be hampered by high body condition score. The pulmonary veins are difficult to visualise using ultrasound since there is only a very small acoustic window available between the lungs and the left atrium. This makes ultrasound recording of this region technically challenging.

In dogs, radiographic comparison of right caudal pulmonary vein diameter relative to its accompanying artery indicated the highest specificity and sensitivity in mitral regurgitation prediction. Other reference parameters include the diameter of the ninth rib (Oui et al., 2015). In our study, ostium II, III and IV and their antra seemed most suited for diameter measurements using echocardiography. The pulmonary vein diameter could then be compared to aortic and pulmonary artery diameter.

Ultrasonographic measurement of pulmonary venous flow is commonly applied in human and small animal medicine for assessing left ventricular filling pressures, left ventricular diastolic function, left atrial function and mitral valve regurgitation (Rossvoll and Hatle, 1993; Nakatani et al., 1994; Rusconi et al., 2001; Merveille et al., 2013; Weidemann et al., 2013; Roels et al., 2014). For flow measurement, best results are obtained when flow is parallel with the ultrasound beam. In horses,

200

Chapter 9: The pulmonary veins of the horse

left parasternal long axis and short axis views from ostium II and III seem most appropriate to measure flow patterns as this flow is along the ultrasound beam direction. Ultrasonographic assessment of flow from ostium IV would therefore not be suited.

In conclusion, Ostia II, III and IV, but not ostium I can be clearly observed during an ultrasound recording of the equine pulmonary veins. The branches opening through ostium III show more variation in merging pattern, while ostium II showed less variability. In case ostium II has a large antrum, and consequently the division of the branches is located more distally, the recognition of this antrum can be difficult due to the small acoustic window, visualising only a small proximal part of this antrum. The parallel orientation of the antrum of ostium II and III ensures that these ostia are most suitable for ultrasound flow measurements according to a defined protocol. The variation in the distal fusion pattern of the different branches draining through ostium III do not constitute a disadvantage in order to measure the respective antrum which is located proximally. Ostium IV of which the antrum is orientated transversely to the scanning plane is less suitable for flow measurements.

201

Chapter 9: The pulmonary veins of the horse

References

Barone R. (1997). Appareil respiratoire. In: Barone, R. Anatomie Comparée des mammifères domestiques, tome troisième, Splanchnologie, foetus et ses annexes, fascicule premier, appareil digestif - appareil respiratoire, Éditions Vigot, Paris, pp. 768-769.

Chiang C.-H., Hagio M., Yoshida H. and Okano S. (1998). Pulmonary Venous Flow in Normal Dogs Recorded by Transthoracic Echocardiography: Techniques, Anatomic Validations and Flow Characteristics. J Vet Med Sc. 60:333-339.

De Clercq D., van Loon G., Schauvliege S., Tavernier R., Baert K., Croubels S., De Backer P. and Deprez P. (2008a). Transvenous electrical cardioversion of atrial fibrillation in six horses using custom made cardioversion catheters. Vet J. 177:198-204.

De Clercq D., van Loon G., Tavernier R., Duchateau L. and Deprez P. (2008b). Atrial and ventricular electrical and contractile remodeling and reverse remodeling owing to short-term pacing-induced atrial fibrillation in horses. J Vet Int Med. 22:1353–1359.

Gelzer A.R.M., Moise N.S., Vaidya D., Wagner K.A. and Jalife J. (2000). Temporal organization of atrial activity and irregular ventricular rhythm during spontaneous atrial fibrillation: An in vivo study in the horse. J Cardiovasc Electrophysiol. 11:773-784.

Kiryu K., Machida N., Kashida Y., Yoshihara T., Amada A. and Yamamoto T. (1999). Pathologic and electrocardiographic findings in sudden cardiac death in racehorses. J Vet Med Sc. 61:921-928.

Leroux A.A., Detilleux J. C., Sandersen F., Borde L., Houben R.M., Al Haidar A., Art T. and Amory H. (2013). Prevalence and risk factors for cardiac diseases in a hospital-based population of 3,434 horses (1994-2011). J Vet Int Med. 27:1563-1570.

202

Chapter 9: The pulmonary veins of the horse

Marr G.M. (1994). Equine echocardiography - sound advice at the heart of the matter. Br Vet J. 150:527-545.

Merveille A.C., Cabrero M., Krafft E., Etienne A.L., Jespers P., Gomart S., Bolen G., Clercx C. and Mc Entee K. (2013). Pulmonary vein to pulmonary artery ratio is an echocardiographic index of left congestive heart failure in dogs with myxomatous mitral valve disease. Proceedings of the 23th ECVIM Meeting, Liverpool, UK.

Mont L., Tamborero D., Elosua R., Molina I., Coll-Vinent B., Sitges M., Vidal B., Scalise A., Tejeira A., Berruezo A. and Brugada J. (2008). Physical activity, height, and left atrial size are independent risk factors for lone atrial fibrillation in middle-aged healthy individuals. Europace. 10:15-20.

Nakatani S., Yoshitomi H., Wada K., Beppu S., Nagata S. and Miyatake K. (1994). Noninvasive estimation of left ventricular end-diastolic pressure using transthoracic Doppler-determined pulmonary venous atrial flow reversal. Am J Cardiol. 73:1017-1018.

Nathan H.G. (1970). Myocardial atrio-venous junctions and extensions (sleeves) over the pulmonary and caval veins. Anatomical observations in various mammals. Thorax. 25:317-324.

Oui H., Oh J., Keh S., Lee G., Jeon S., Kim H., Yoon J. and Choi J. (2015). Measurements of the pulmonary vasculature on thoracic radiographs in healthy dogs compared to dogs with mitral regurgitation. Vet Radiol Ultrasound. 56:251-256.

Reef V.B., Reimer J.M. and Spencer P.A. (1995). Treatment of atrial fibrillation in horses: new perspectives. J Vet Int Med. 9:57-67.

Roels E., Merveille A.C., Krafft E., Farnir F., Gomart S., Clercx C. and Mc Entee K. (2014). Right pulmonary vein to pulmonary artery ratio: a new echocardiographic index of pulmonary hypertension in west highland white terriers with idiopathic pulmonary fibrosis. Proceedings of the 24th ECVIM Meeting, Mainz, Germany.

Rossvoll O. and Hatle L.K. (1993). Pulmonary venous flow velocities recorded by transthoracic Doppler ultrasound: relation to left ventricular diastolic pressures. J Am Coll Cardiol. 21:1687-1696.

Roux N., Havet E. and Mertl P. (2004). The myocardial sleeves of the pulmonary veins: potential implications for atrial fibrillation. Surg Radiol Anat. 26:285-289.

203

Chapter 9: The pulmonary veins of the horse

Rusconi C., Sabatini T., Faggiano P., Ghizzoni G., Oneglia C., Simoncelli U., Gualeni A., Sorgato A. and Marchetti A. (2001). Prevalence of isolated left ventricular diastolic dysfunction in hypertension as assessed by combined transmitral and pulmonary vein flow Doppler study. Am J Cardiol. 87:357-360.

Saito T., Waki K. and Becker A.E. (2000). Left atrial myocardial extension onto pulmonary veins in humans: Anatomic observations relevant for atrial arrhythmias. J Cardiovasc Electrophysiol. 11:888- 894.

Vandecasteele T., Vandevelde K., Doom M., Van Mulken E., Simoens P. and Cornillie P. (2015). The pulmonary veins of the pig as an anatomical model for the development of a new treatment for atrial fibrillation. Anat Histol Embryol. 44:1-12. van Loon G., Duytschaever M., Tavernier R., Fonteyne W., Jordaens L. and Deprez P. (2002). An equine model of chronic atrial fibrillation. Methodology. Vet J. 164:142-150.

Weidemann F., Niemann M., Herrmann S., Ertl G. and Störk S., (2013). Assessment of diastolic heart failure. Current role of echocardiography. Herz. 38:18-25.

204

Chapter 10

3D reconstruction of the porcine and equine pulmonary veins, supplemented with the identification of telocytes in the horse

Adapted from Vandecasteele T., Van den Broeck W., Tay H., Couck L., van Loon G. and Cornillie P. (2018). 3D reconstruction of the porcine and equine pulmonary veins, supplemented with the identification of telocytes in the horse. Anat Histol Embryol. 47:145- 152.

205

206

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Summary

The myocardial sleeve of the porcine and equine pulmonary veins were histologically investigated and reconstructed three-dimensionally. Moreover, the localization of neuron cell bodies at the veno- atrial junction and alongside the myocardial sleeve was light microscopically visualized to depict the organization of nerve, myocardial and fat tissue. Finally, the presence of telocytes inside the equine pulmonary veins was demonstrated by use of transmission electron microscopy. These structures are thought to play a role in the induction of atrial fibrillation, which is frequently seen in horses, while pigs are often used as a cardiovascular model in this context. This data fills in remaining gaps in literature concerning the histological build-up of the pulmonary veins wall in pigs and horses. In- depth knowledge on the myocardial sleeve and its surrounding cell types is important to understand the possible outcome of an ablation therapy as an atrial fibrillation treatment. In pigs and horses, the layout of the pulmonary veins wall concerning these structures is comparable to humans. However neuron cell bodies were recovered at the veno-atrial junction in both species but not alongside the myocardial sleeve in horses.

207

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Introduction

In man, atrial fibrillation (AF) is the most common sustained cardiac arrhythmia with increasing prevalence at higher age and the most frequent arrhythmia in clinical practice (Haïssaguerre et al., 1998, Haïssaguerre et al., 2007). Moreover, it is the major cause of cardioembolic cerebrovascular accidents (Haïssaguerre, 2010). In the last decades, a lot of research has been performed on the human pulmonary veins in the context of AF as these are important spots for the induction and conduction of abnormal pulses towards the heart (Haïssaguerre et al., 1998). Specifically, myocardial tissue fanning out inside the pulmonary veins wall was indicated as the origin of ectopic electrical signals conducting towards the left atrium (Haïssaguerre et al., 1998; Saito et al., 2000; Roux et al., 2004).

In the context of AF, horses and pigs are interesting animals as horses are very susceptible to develop AF and pigs are frequently used as cardiac models (van Loon et al., 2002; Young, 2003; Peng et al., 2004; De Clercq et al., 2008; Reef et al., 2014; Bode et al., 2015; Decloedt et al., 2015). However, much more research has been conducted in men, compared to horses and pigs, to describe the anatomical and histological organization of the pulmonary veins. Besides the detailed description of the anatomical organization of the human pulmonary veins, several structures inside the pulmonary veins wall were researched in men. The presence and the arrangement of the myocardial sleeve which was demonstrated partially in a 3-D reconstruction, the separation of myocardial tissue by fibrofatty tissue, the localization of telocytes and ganglia inside the pulmonary veins were pointed out in human studies concerning AF initiation (Saito et al., 2000; Ho et al., 2001; Tan et al., 2006; Gherghiceanu et al., 2008; Morel et al., 2008; Vaitkevicius et al., 2008). Telocytes are an interstitial cell type, typically with very long prolongations, which are thought to play a role in AF induction through interactions with myocardial sleeve tissue (Gherghiceanu et al., 2008; Morel et al., 2008). Concerning the electrical coupling between the pulmonary veins and the left atrium, it was indicated that the human left atrial wall includes regions with ganglionated plexi sending nerves towards the pulmonary veins (Armour et al., 1997; Pauza et al., 2000).

In horses and pigs, histological data regarding pulmonary veins is limited. We already described the anatomical organization of the equine and porcine pulmonary veins (Vandecasteele et al., 2015; 2016) but three-dimensional information was not available. AF is the most important cardiac arrhythmia in horses. Therefore, detailed insight on the type and relative position of the different tissue layers of pulmonary veins is indispensable in order to fully understand AF pathophysiology and the efficacy or failure of an ablation therapy.

208

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Previous studies mentioned the presence of myocardial sleeve tissue in pigs and Vandecasteele et al. (2017) proved the localization of telocytes and ganglia near myocardial tissue inside the porcine pulmonary veins wall. However, these structures were not investigated in detail in horses. In addition, the localization of fat and connective tissue in relation to myocardial sleeve tissue in pigs and horses has not been reported in literature.

As pigs are frequently used as a cardiovascular model to develop new treatment strategies for AF and because AF is clinically the most important arrhythmia in horses, this study aims to fill in remaining gaps in literature regarding the types and relative 3D position of tissues that build up the pulmonary veins.

Materials and Methods

Animals and samples

Horse

Heart samples were obtained from 17 horses immediately after euthanasia, which was not related to this study. Tissue samples of 15 horses were used for light microscopic studies of which of 11 horse cadavers of different breeds one tissue sample per heart was collected of the left veno-atrial junction to investigate the myocardial sleeve. Veno-atrial tissue samples of the remaining four horses were used to detect and visualize ganglia.

One sample of 5 cm pulmonary vein tissue together with a part of the atrial wall was excised at pulmonary vein ostium IV, which drains the left cranial lung lobe and is situated on the left side of the left atrium, close to the left auricle (Vandecasteele et al., 2016). This sample was used for three- dimensional reconstruction, built up from serial transverse histological sections, to visualize the myocardial sleeve. Due to the large sample size, the sample was divided in three equal parts, one containing the ostium, and two parts containing the rest of the pulmonary vein.

Tissue samples of two horses were used for transmission electron microscopic analysis. Samples of 1 cm2 were prelevated at the level of the veno-atrial junction to visualize the presence of telocytes.

Pig

Both pulmonary vein ostia I and II (Vandecasteele et al., 2015) from a pig of 60 kg bodyweight (euthanasia not related to this study) were used for three-dimensional reconstruction and visualization of mainly the myocardial sleeve and neuron cell bodies.

209

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Light microscopy

Samples were fixed in 4% phosphate buffered formaldehyde for minimum 48 hours. Before the samples were excised, Indian ink was injected into the pulmonary vein wall at the level of the veno- atrial junction. Longitudinal and transverse tissue samples, from the veno-atrial junction to approximately 5 cm more distally, were dehydrated and embedded in paraffin by use of a Microm tissue processor STP 420D (Prosan, Merelbeke, Belgium) and Microm embedding station EC 350-1 (Prosan, Belgium), respectively. The sections were cut transversely at 5 µm thickness and subsequently stained with HE.

3D reconstruction

The equine samples were fixed in 4% formaldehyde solution for 48 hours and processed routinely as described before. Sections of 10 µm were transversely cut on a Micron H360 microtome (Prosan, Belgium) of which only each 10th section was subsequently stained with haematoxylin and eosin (HE). The sections were evaluated and photographed using an Olympus BX61 light microscope and Olympus DP50 camera (Olympus, Belgium).

The sets of images were 3D reconstructed using Amira 6.1 3D software (FEI, USA). The different tissues or structures, such as myocardial sleeve, fat, neuron cell bodies and the complete wall were identified and segmented accordingly.

The porcine samples were fixated and processed as described before but serially sectioned at 8 µm thickness. Every 50th section from the part containing the ostium was used for three-dimensional reconstruction. From the two other parts every 10th section was used. A three-dimensional image visualizing the lumen of both ostia, the pulmonary veins wall, fat, neuron cell bodies and the myocardial sleeve was reconstructed using the Amira 6.1 3D software (FEI, USA).

Transmission electron microscopy

The prelevated samples were fixed in Karnovsky’s fixative (24h) and, subsequently, washed in 0.1M sodium cacodylate buffer after which the samples were postfixated in 1% OsO4 (1.30h), dehydrated in series of alcohol and embedded in epon 812 (Leica EM TP, Leica Microsystems GmbH, Wetzlar, Germany). Firstly, semithin sections were cut (0.5-1µm) by use of a Leica EMUC6 ultramicrotome (Germany), stained with toluidine blue and observed under the light microscope. Finally, ultrathin sections (80nm) were cut, mounted on formvar coated grids, contrasted with 1% uranyl acetate and 1.33% lead citrate and examined under the Jeol JEM 1400 Plus transmission electron microscope (Jeol Ltd, Tokyo, Japan). The identification of telocytes was based on the typical characteristics of this

210

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

cell type, described by Popescu and Nicolescu (2013) and summarized in our previous study by Vandecasteele et al. (2017).

Results

Horse

Light microscopy

Each histologically examined pulmonary vein sample (n=11) contained a myocardial sleeve, although some sleeves were very short and others much longer. The high variability of the measured sleeve length resulted into an irregular cuff organization of all sleeve tissue of one ostium, which ends distally in an erratic pattern. The muscle fibres of the myocardial sleeve were arranged in longitudinal and transverse layers (Fig. 1). Serial transverse histological sections demonstrated the course of the myocardial sleeve in the pulmonary vein wall of ostium I (Fig. 1). The width of the myocardial sleeve was reduced as it elongates into the pulmonary veins wall towards the lungs. Figure 1 gives an indication of the range of sleeve width whereby at 1.75 mm from the ostium the myocardial sleeve was 5.2 mm wide, whereas at 4 mm from the ostium, the sleeve measured 1.1 mm in width. The length of the sleeve may differ to a great extent between different animals, between different ostia and even between different sample spots inside an ostium but a larger ostium demonstrates in most cases a longer myocardial sleeve. Overall, the length of the sleeve varied between 5 mm and 30 mm but in one sample of ostium III, the sleeve length was 43.5 mm. Fig. 2 demonstrates a 12.4 mm long myocardial sleeve of a pulmonary vein draining into ostium IV.

211

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Fig. 1. Left panel: Serial transverse sections of an equine pulmonary vein wall draining into ostium I with a myocardial sleeve encircled in black. (a) 1.75 mm distal to the ostium, the myocardial sleeve is 5.2 mm wide; (b) 2.5 mm from the ostium, 3.2 mm wide; (c) 3.25 mm from the ostium, 1.7 mm wide; (d) 4 mm from the ostium, 1.14 mm wide; bars = 0.70 mm. Right panel: Longitudinal section of an equine pulmonary vein at the veno- atrial junction. Muscle fibres cut longitudinally (1), muscle fibres cut transversally (2). *Indian ink injected at the veno-atrial junction, HE staining, bar = 1 mm.

Fig. 2. Top side: Macroscopic overview, with the atrial side located on the right, of a longitudinal section of the equine left cranial pulmonary vein with the myocardial sleeve, draining into ostium IV. Bottom side: Microscopic detail of the box with an equine myocardial sleeve measuring 12.4 mm between the injection spot (veno-atrial junction) and the most distal point of the sleeve. 1: Indian ink, HE staining, bar = 1 mm.

No neuron cell bodies were found alongside the common proximal part of confluenting pulmonary veins, which is called the antrum. However, clustered ganglia were recovered at the level of the dorsal atrial wall and the veno-atrial junction. The histological samples of the veno-atrial junction contained several ganglia with an average diameter of approximately 250 µm in the vicinity of thick layers of myocardial tissue (Fig. 3).

212

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Fig. 3. Overview of an equine pulmonary vein wall with ganglia (1) surrounded by fat (2), cardiomyocytes (3) and nerve bundles containing axons (4), HE staining, bar = 500 µm.

3D reconstruction of the myocardial sleeve

At the level of ostium IV, the myocardial sleeve demonstrated also an irregular pattern as it did not completely surround the pulmonary vein. Distally, the sleeve fanned out around the three branches of the vein and ended “pointy shaped”. In contrast to the investigated porcine pulmonary veins, no neuron cell bodies were found (Fig. 4). The most proximal blue part changes to red more distally to visualize specifically the sleeve pattern of myocardial tissue on its own (Fig. 4).

Fig. 4. Panel a: Macroscopic overview of a transverse section through the three branches of the pulmonary vein draining the left cranial lung lobe of a horse. Lumen (1), myocardial sleeve tissue (2), HE staining, bar = 8 mm. Panel b: three-dimensional reconstruction of ostium IV of a horse with the lumen (yellow) of the branches of the pulmonary vein draining the cranial lung lobe, along which the myocardial sleeve (red) is fanning out. A part of the left atrial dorsal wall surrounding ostium IV is indicated in blue.

213

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Transmission electron microscopy

Cells, displaying the characteristic properties of telocytes defined by Popescu and Nicolescu (2013), were also found in the wall of the pulmonary veins veno-atrial junction in the horse (Fig. 5). Long slender prolongations (approximately 22 µm in Fig. 5a), called telopodes, demonstrating thicker parts at a regular interval, called podoms, were observed in contact with myocardial sleeve tissue (distance between both < 20nm) (Fig. 5).

Fig. 5. Panel a: TEM image of a telocyte prolongation (tpr) in contact (<20 nm) with equine myocardial sleeve tissue (myo). Podom (po). Panel b: TEM image of a telocyte cell body (tcb) with a long telopode (tpr) in close vicinity of equine myocardial sleeve tissue (myo). Podom (po).

214

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Pig

Both ostia demonstrated a myocardial sleeve which fans out irregularly (Fig. 6, 7). The sleeve of ostium I (largest ostium) is noticeably longer compared to the sleeve of ostium II. The myocardial sleeve was mostly accompanied by fatty tissue. As such, fat tissue does not only occur at the veno- atrial junction but was also recovered alongside the pulmonary vein and even more distal than the myocardial tissue (Fig. 6, 7). The pulmonary veins draining through both ostia demonstrated neuron cell bodies irregularly spread over the full length of the myocardial sleeve except for certain zones without ganglion cells. Some neuron cell bodies were found just distally of the myocardial sleeve, thus in the region towards the lungs. The neuron cell bodies were present in close vicinity to the myocardial sleeve or surrounded by fat tissue or even localized in a layer of connective and smooth muscle tissue (Fig.6).

Fig. 6. 3D reconstruction of the two porcine pulmonary veins ostia (I and II); panel a: the lumen (yellow) is surrounded by a combination of connective tissue, smooth muscle tissue and cardiac muscle tissue (blue). The myocardial sleeve (red) fans out distally and ends in a tip. Panel b: Fat is also present at the veno-atrial junction and alongside the pulmonary veins antrum, intermingling with the other tissues or present in fat pads. Panel c and d: The lumen is also surrounded by neuron cell bodies (purple), which are irregularly distributed.

215

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Fig. 7. 3D reconstruction of the two porcine pulmonary veins ostia (I and II); panel a: Only the distal part of the myocardial sleeve is visualized, enclosing the lumen. Groups of neuron cell bodies follow the myocardial sleeve and ganglion cells are present until the distal end of the sleeve. Panel b: The irregular distribution of fat tissue which contains the majority of the neuron cell bodies.

Discussion

This study gives a complete 3D image of the build-up of the pulmonary veins in the pig and the horse, completed with a demonstration of telocytes and the myocardial sleeve tissue in the horse. Previously we demonstrated telocytes and cardiac tissue in the porcine pulmonary veins (Vandecasteele et al., 2015 and 2017). These data, together with the visualization of neuron cell bodies and ganglia in pigs and horses, complete the existing literature about the presence of histological structures which are thought to play a role in cardiac arrhythmias such as atrial fibrillation.

Overall, the build-up of the pulmonary veins wall and the presence of myocardial sleeve tissue and telocytes in pigs and horses is comparable with humans. However, the distribution of neuron cell bodies alongside the myocardial sleeve in horses differs from humans and pigs.

The rich innervated pulmonary veins wall and especially the presence of neuron cell bodies which may interact with cardiac muscle, was visualized in humans by Vaitkevicius et al. (2008) with whole mount histochemical staining. A close vicinity between neuron cell bodies and cardiac muscle and the fact that neuron cell bodies follow the myocardial sleeves might indicate a relation between these tissue types. Hence, several studies such as Mehall et al. (2007) demonstrated that epicardial mapping in humans allows examination and identification of ganglionic plexi at the level of the upper half of the heart, more specifically, the highest concentration was located at the Marshall ligament area and at the superior part of the interatrial groove. Further, this study indicated that electrical isolation of these plexi may stop AF in certain patients. Nakagawa et al. (2004) showed that co- stimulation of ganglionic plexi and pulmonary veins may lead to AF initiation by pulmonary vein foci, but stimulation of only the pulmonary vein did not. As such, there seems to be an important

216

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

interaction between the atrial autonomic nerves and the pulmonary veins in the context of arrhythmogenesis (Patterson et al., 2005) whereby autonomic nerve stimulation increases AF susceptibility. On might therefore question whether the initiation of AF in the PV is really spontaneous as stated by Haïssaguerre et al. (1998) or not. An epicardial approach to electrophysiological studies and ablation of the pulmonary veins is therefore useful to assess the influence of nerve tissue and the effect of an ablation procedure (Mehall et al., 2007). Nevertheless, in literature, there is still much debate about the need or effect of epicardial ablation. Indeed, the highly irregular distribution of nerve tissue in the pulmonary veins wall, and the fact that a perfect circumferential lesion is still required to completely isolate the pulmonary vein, make the interpretation of the results difficult. Further research regarding ablation techniques with and without involvement of nerve tissue is needed to elucidate the contribution of the autonomic nerves to AF pathophysiology.

Also the presence of fat is important as Deneke et al. (2005) indicated that a non-ablated myocardial fiber was insulated by fat and our own study (unpublished data) proved that fat can be an insulator for both cardiac muscle and neuron cell bodies while cardiac muscle itself facilitates heat transfer during an ablation procedure. The variability of tissue in which neuron cell bodies are present may also be a possible reason why a diversity of treatment results can be seen in different patients after ablation procedures.

Although pigs are frequently used as a model for cardiovascular research they are unlikely to develop AF under natural circumstances. The advantage of horses as model for research is that this species is very susceptible to both natural and induced AF. In pigs, ganglia were shown along the length of the pulmonary veins which corresponds with humans. However, in horses, thick nerve bundles, parallel with the pulmonary veins, were present but the ganglia were only found at the veno-atrial junction. The nerve bundle distribution in these samples is in accordance with pulmonary vein samples, of other horses and other ostia, which were investigated before (unpublished data) and corresponds with samples of men and pigs. As ablation is typically performed at the level of the pulmonary veins ostium, it is probably not important in the context of AF treatment whether or not ganglia are located distally from the ablation point as these structures are electrically isolated after an ablation procedure, assuming a circumferential injury is applied.

217

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

References

Armour J.A., Murphy D.A., Yuan B.-X., Macdonald S. and Hopkins D.A. (1997). Gross and microscopic anatomy of the human intrinsic cardiac nervous system. Anat Rec. 247:289-298.

Bode F., Blanck O., Gebhard M., Hunold P., Grossherr M., Brandt S., Vonthein R., Thiele H., Dunst J. and Rades D. (2015). Pulmonary vein-isolation by radiosurgery: implications for non-invasive treatment of atrial fibrillation. Europace. 17:1868-1874.

De Clercq D., van Loon G., Tavernier R., Duchateau L. and Deprez P. (2008). Atrial and Ventricular Electrical and Contractile Remodeling and Reverse Remodeling Owing to Short-Term Pacing-Induced Atrial Fibrillation in Horses. J Vet Internal Med. 22:1353-1359.

Decloedt A., Schwarzwald C., De Clercq D., Van Der Vekens N., Pardon B., Reef V. and van Loon G. (2015). Risk factors for recurrence of atrial fibrillation in horses after cardioversion to sinus rhythm. J Vet Intern Med. 29:946-953.

Haïssaguerre M., Lim K.T., Jacquemet V., Rotter M., Dang L., Hocini M., Matsuo S., Knecht S., Jais P. and Virag N. (2007). Atrial fibrillatory cycle length: computer simulation and potential clinical importance. Europace. 9:64-70.

Haïssaguerre M. (2010). In search of the sources of cardiac fibrillation. EMBO Mol Med. 2:117-119.

Ho S.Y., Cabrera J.A., Tran V.H., Farre J., Anderson R.H. and Sanchez-Quintana D. (2001). Architecture of the pulmonary veins: relevance to radiofrequency ablation. Heart. 86:265-270.

218

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Mehall J.R., Kohut R.M., Schneeberger E.W., Taketani T., Merrill W.H. and Wolf R.K. (2007). Intraoperative epicardial electrophysiologic mapping and isolation of autonomic ganglionic plexi. Ann Thorac Surg. 83:538-541.

Morel E., Meyronet D., Thivoet-Bejuy F. and Chevalier P. (2008). Identification and distribution of interstitial Cajal cells in human pulmonary veins. Heart Rhythm. 5:1063-1067.

Nakagawa H., Scherlag B.J., Aoyama H., Foresti S., Yamanashi W.S., Pitha J.J., Lazzara R. and Jackman W.M. (2004). Catheter ablation of cardiac autonomic nerves for prevention of atrial fibrillation in a canine model. Heart Rhythm. 1:S10 (abstract).

Patterson E., Po S.S., Scherlag B.J. and Lazzara R. (2005). Triggered firing in pulmonary veins initiated by in vitro autonomic nerve stimulation. Heart Rhythm. 2:624–631.

Pauza D.H., Skripka V., Pauziene N. and Stropus R. (2000). Morphology, distribution, and variability of the epicardiac neural ganglionated subplexuses in the human heart. Anat Rec. 259:353–382.

Peng J., Madrid A.H., Palmeiro A., Rebollo J.M., Limón L., Nannini S., Marín I., Bernal E., Escobar C., Viana M. and Moro C. (2004). Saline irrigated catheter ablation for pulmonary vein isolation in pigs: an experimental model. PACE. 27:495-501.

Roux N., Havet E. and Mertl P. (2004). The myocardial sleeves of the pulmonary veins: potential implications for atrial fibrillation. Surg Radiol Anat. 26:285–289.

219

Chapter 10: 3D reconstruction of porcine and equine pulmonary veins

Vaitkevicius R., Saburkina I., Zaliunas R., Pauziene N., Vaitkeviciene I., Schauerte P. and Pauza D.H. (2008). Innervation of pulmonary veins: morphologic pattern and pathways of nerves in the human fetus. Ann Anat. 190:158-166.

Vandecasteele T., van Loon G., Vandevelde K., De Pauw B., Simoens P. and Cornillie P. (2016). Topography and ultrasonographic identification of the equine pulmonary vein draining pattern. Vet J. 210:17-23.

Vandecasteele T., Cornillie P., Vandevelde K., Logothetidou A., Couck L., van Loon G. and Van den Broeck W. (2017). Presence of Ganglia and Telocytes in Proximity to Myocardial Sleeve Tissue in the Porcine Pulmonary Veins Wall. Anat Histol Embryol. 46:325-333. van Loon G., Duytschaever M., Tavernier R., Fonteyne W., Jordaens L. and Deprez P. (2002). An equine model of chronic atrial fibrillation: Methodology. Vet J. 164:142–150. https://doi.org/10.1053/tvjl.2001.0668.

Young LE. (2003). Equine athletes, the equine athlete’s heart and racing success. Exp Physiol. 88: 659–663.

220

Chapter 11

Echocardiographic identification of atrial-related structures and vessels validated by CT images of equine hearts

Adapted from Vandecasteele T., Cornillie P., Van Steenkiste G., Vandevelde K., Gielen I., Vanderperren K. and van Loon G. (2018). Echocardiographic identification of atrial-related structures and vessels validated by CT images of equine hearts. Equine Vet J. DOI: 10.1111/evj.12969.

221

222

Chapter 11: Echocardiographic identification of equine pulmonary veins

Summary

Echocardiography is the imaging technique of choice for the equine heart. Nevertheless, knowledge about ultrasonographic identification of dorsally located structures and vessels, related to the atria, is scarce. To describe the ultrasound approach and the identification of structures and vessels in relation to the atria in healthy horses, CT images from two equine hearts, casted with self-expanding foam, were segmented and used to identify atrial-related structures and vessels. These images were compared with standard and non-standard ultrasound images from ten healthy horses obtained from a left and right parasternal view optimized to visualize the dorsal cardiac area. On new ultrasound views, specific atrial anatomical landmarks such as vena cava, pulmonary arteries, intervenous tubercle and oval fossa were identified in all horses. In addition, ultrasound views were defined to visualize the brachiocephalic trunk, pulmonary veins and their ostia. The 3D segmented CT images were used to reconstruct slices that corresponded with the ultrasound images and allowed correct identification of specific structures. Ultrasound exams and casts were from different animals. A small number of casts and horses were used so anatomical variation or individual differences in identifying structures on ultrasound could not be assessed. Important cardiac structures and vessels, even the different pulmonary veins, could be identified on standard and non-standard ultrasound images in adult horses. This knowledge is important to guide and develop interventional cardiology and might be useful for diagnostic and therapeutic purposes.

223

Chapter 11: Echocardiographic identification of equine pulmonary veins

Introduction

While fluoroscopy, MRI and CT are widely used in all aspects of human cardiology (Bluemke et al., 2008; Farré et al., 2010; Schuetz et al., 2010), these techniques have limited value (fluoroscopy) or cannot be applied in adult horses (MRI, CT). These techniques are a mainstay in human and small animal cardiology and often the gold standard for diagnosis and prognosis of cardiac diseases and provide essential guidance for invasive cardiac procedures. Although in equine cardiology fluoroscopy may be helpful to some extent, echocardiography is the principal imaging tool. For routine cardiac assessment, the echocardiographic exam is performed in a standard way with predefined long and short axis image planes from the left and right thorax that allow identification and measurements of common structures such as atria, ventricles, myocardial walls, valves and the origin of aorta (Ao) and pulmonary artery (PA) (Feigenbaum, 1986). While these rather ventrally located structures have been described in detail, limited information is available regarding more dorsally located structures such as dorsal atrial walls, cranial (CrVC) and caudal vena cava (CaVC), brachiocephalic trunk (BT) and pulmonary veins (PV) because they are partially covered by air-filled lungs, and therefore more difficult to visualize (Patteson, 1999; Marr and Patteson, 2010; Palgrave and Kidd, 2014). Recently, based upon a topographical study, the ultrasonographic identification of PV in horses was described (Vandecasteele et al., 2016).

Invasive cardiac procedures are of increasing interest in equine cardiology with new techniques being developed but navigation of equipment through vessels and cardiac chambers relies on good ultrasonographic anatomical landmarks. More and more centres offer transvenous electrical cardioversion which requires accurate positioning of catheters in right atrium (RA) and left pulmonary artery (lPA) (Levy et al., 1992; Alt et al., 1997; McGurrin et al., 2005; van Loon et al., 2005; De Clercq et al., 2008; Schauvliege et al., 2009). Pacemaker lead implantation, cardiac biopsy procedures and occluder implantation are performed with ultrasound guidance (van Loon et al., 2001; van Loon et al., 2002; Javsicas et al., 2010; Decloedt et al., 2016). Electrophysiological studies in horses are performed both for research and for diagnostic purposes (De Clercq et al., 2014; Decloedt et al., 2014; Broux et al., 2016). Recently, the first successful 3D electro-anatomical cardiac mapping has been reported in adult horses (van Loon et al., 2017). Ultrasound guided electro-anatomical mapping from within the RA, left atrium (LA) and PV would be of great interest in horses treated for atrial fibrillation to assess PV ectopy (Young, 2003) and could lead to development of new treatment strategies. The importance of the PV in atrial fibrillation pathophysiology has been well established in humans (Haïssaguerre et al., 1998) but information is currently lacking in horses. Visualization of the PV would also be of interest to assess PV size and flow.

224

Chapter 11: Echocardiographic identification of equine pulmonary veins

The aim of this study was to identify left and right atrial-related anatomical landmarks and blood vessels on ultrasound. In order to do so, CT images from casted equine hearts were segmented, 3D reconstructed and subsequently used for comparison with ultrasound images to allow correct identification of the different structures.

Materials and Methods

The current study (EC 2015/96) was performed following the guidelines of the local Ethical Committee. Cardiopulmonary sets of four adult horses (2 Warmblood geldings, 1 Thoroughbred mare and 1 Quarter horse stallion; body weight 530-600kg; age 2.5-6 years), euthanized for non- cardiovascular reasons, were removed immediately after death with preservation of the proximal part of CrVC, CaVC and Ao. Lungs were partially removed and via an incision in the left and right auricle, after removal of blood clots, one-component self-expanding polyurethane construction foam (Hubo, Belgium) was injected through a tube, placed into the right (RV) and left ventricle (LV). Whilst the foam was injected, the tube was slowly retracted towards the atria. Two hours after foam injection the cardiopulmonary set was subjectively evaluated to represent a close to normal expansion of all cardiac chambers and vessels (Fig. 1). Of the two best casted hearts, distal parts of the lungs were removed and computed tomographic (CT) scans were made with a four-slice helical CT device at 120 kV and 140 mA (CT scanner, LightSpeed, Qx/I, GE Medical Systems, Milwaukee, WI).

Fig. 1. Casted heart after injection of polyurethane foam into the heart and major vessels. Panel A: the foam cast with the surrounding cardiac tissue still present was used to make the CT images. Panel B: foam cast without the surrounding cardiac tissue.

The obtained CT DICOM images from both hearts were imported in dedicated computer software (Amira 6.1 3D software, Thermo Fisher, MA, USA) and 3D-reconstructed. Subsequently, within the

225

Chapter 11: Echocardiographic identification of equine pulmonary veins

software, cardiac chambers and associated vessels were segmented with specific colour codes (Table 1) to allow quick and easy orientation of the 3D reconstruction.

Pulmonary vein ostium I Pulmonary vein ostium II Pulmonary vein ostium III Pulmonary vein ostium IV Left atrium (LA) Right atrium (RA) Left ventricle (LV) Right ventricle (RV) Aorta (AO) Pulmonary trunk/arteries (PT/PA) Cranial vena cava (CrVC) and azygos vein Caudal vena cava (CaVC) Myocardium Table 1. Colour codes for the different chambers and associated vessels, used for segmentation of the CT images.

In ten horses (9 Warmbloods and 1 Thoroughbred) echocardiography (GE Vivid 7 Dimension with 3S phased array transducer at 1.7/3.4 MHz, GE Healthcare, Diegem, Belgium) was performed from a left and right parasternal approach focussing on the dorsal cardiac region to identify left and right atrial- related anatomical structures and blood vessels. Different non-standard views were used to identify important structures and vessels. Most of these views had been used by our research group during previous invasive cardiac procedures (G. van Loon, unpublished data). For each echocardiographic view, the transducer handling was accurately described by defining probe position and rotation (Table 2). Probe position refers to the spot where the transducer was placed on horse’s thorax. Probe rotation describes how many degrees the probe was rotated clockwise (e.g. +90°) or counter clockwise (e.g. -90°) in regard to the neutral position (0°, index mark of the probe dorsally), which was close to the long axis view of the heart. Finally, probe angulation was reported following the standardized nomenclature used for radiographic projections (Smallwood et al., 1985), approved by the American College of Veterinary Radiology. This nomenclature system describes the direction of the beam from the point where it enters the body to the point where it would exit the body, separated by a hyphen. Standard abbreviations, such as left (Le), right (Rt), cranial (Cr), caudal (Ca), dorsal (D) and ventral (V), are used and may be combined. A horizontal probe position, approximately perpendicular to the chest of the standing horse, is regarded as a standard, non- oblique view. For oblique views, angles of obliquity can be inserted in addition to the term ‘oblique’ (O). Angles were subjectively assessed. For example, a right parasternal four-chamber view requires

226

Chapter 11: Echocardiographic identification of equine pulmonary veins

the probe to point about 10° caudally compared to the standard probe direction and would be described as a right cranial to left caudal oblique view, or more precisely a Rt10Cr–LeCaO view. For all images, the probe index mark was displayed right on screen.

Right Rt5Cr10V-LeCaDO, 0° rotation RA, RV, LA, LV, III, IV Rt30V-LeDO, -60° rotation CaVC, RA, RV, LA, AV, II Rt45V-LeDO, +80° rotation CrVC, CaVC, IT, RA, BT, Ao, rPA, III, II, OF, limbus Rt60V-LeDO, +80° rotation RAA, TC, CrVC, CaVC, IT Rt60V-LeDO, -45° rotation III, RA, IV, LA Rt10Ca45V-LeCrDO, +80° rotation Caudal part of CrVC Rt70V-LeDO, +80° rotation Vena azygos Left Le10Ca-RtCrO, 0° rotation LV, LA, III, CaVC, rPA Le10Cr10V-RtCaDO, 0° rotation LV, LA, II, IV Le15Ca60V-RtCrDO, +80° rotation PT, lPA, rPA, Ao, CrVC, RAA Le60V-RtDO, +60° rotation PT, lPA, rPA, LA, Ao, IV, PT Le10Cr45V-RtCaDO, +80° rotation LA, III, II, rPA Le10Ca45V-RtCrDO, +80° rotation LA, III, II, rPA Le45Ca10V-RtCrDO, 0° rotation RA, RV, PT, pulmonary valve Table 2. Probe angulation and rotation with corresponding visible structures on ultrasound from a right and left parasternal view.

Results

Most important structures and vessels related to the atria were identified on CT and ultrasound images in all horses. Below the ultrasound images are explained, using the CT image as reference and to identify different structures. Table 2 summarizes most important findings.

Right parasternal view

From the standard four-chamber view (Rt10Cr-LeCaO) with slightly more cranial and dorsal angulation (Rt5Cr10V-LeCaDO), ostium III could be easily identified by its pulmonary veins which travel from right to left and which are found adjacent to the more ventrally located RA and the more dorsally located right pulmonary artery (rPA). Close to ostium III, the oval fossa (OF) was visible as a small, thinned area in the interatrial septum or, depending on the ultrasonographic view, as a round anechoic structure in the atrial septum. The limbus of the oval fossa, which is the prominent craniodorsal margin, could also be identified. From the same view, ostium IV was visible at the left dorsal wall of the left atrium (LA) (Fig. 2A&B). With slightly more caudal angulation (Rt20Cr10V- LeCaDO), ostium III and IV, and the right pulmonary artery (rPA) disappeared from screen while the

227

Chapter 11: Echocardiographic identification of equine pulmonary veins

opening of ostium II in the LA could be identified. This view also showed part of the debouchment of the CaVC, especially when angulating even more caudally (Rt30Cr10V-LeCaDO).

On a short-axis view at the aortic valve level (Rt30V-LeDO and -60° rotation), displaying CaVC, RA, RV, LA and aortic valves, PV ostium II was visualized at the caudodorsal border of the LA (Fig. 2C&D). Pointing more dorsally (Rt60V-LeDO) and with -45° rotation showed ostium III, adjacent to the RA, and ostium IV debouching at the left dorsal wall of the LA.

Fig. 2. Panels A and B: right parasternal long axis four-chamber view on ultrasound (A) and 3D-reconstructed CT image (B) visualizing right atrium (RA), right ventricle (RV), left atrium (LA), left ventricle (LV), ostium III (III) and ostium IV (IV). From the 4th intercostal space, the ultrasound beam points slightly from right cranioventral to left caudodorsal (Rt5Cr10V-LeCaDO) with about 0° rotation. Panels C (ultrasound) and D (CT image): right parasternal short axis view at aortic valve level (Rt30V-LeDO, rotation -60°) displaying caudal vena cava (CaVC), right atrium (RA), right ventricle (RV), left atrium (LA), aortic valve (AV) and ostium II (II).

From the four-chamber view, the ultrasound beam was subsequently rotated clockwise to +80°, and angled dorsally and slightly more cranially (Rt45V-LeDO) until the characteristic intervenous tubercle (IT) was identified as a triangular myocardial structure originating from the right atrial wall (Fig. 3A&B). This view showed the debouchment of the CrVC and CaVC into the RA. Deep to the RA, the bifurcation between Ao and BT was visible. Caudal to the IT, the thin wall of the OF with its limbus was seen, adjacent to the debouchment of ostium III. Craniomedial to ostium III, adjacent to the Ao, the rPA was found while deep to ostium III, ostium II was visible (Fig. 3A&B). When angulating more cranially (Rt10Ca45V-LeCrDO), the most caudal part of the CrVC could be identified. Placing the

228

Chapter 11: Echocardiographic identification of equine pulmonary veins

probe in the 3rd intercostal space with dorsal angulation (Rt60V-LeDO) and +80° rotation also allowed visualisation of the terminal crest (TC) and right atrial appendage (RAA) in 3 horses (Fig. 3C&D). For this view, the horse’s right foot had to be placed forward, and substantial pressure with the probe against the triceps muscle was needed. In only 1 horse, the vena azygos could be seen by maximal dorsal angulation (Rt70V-LeDO).

Fig. 3. Right parasternal short axis view from the 4th (A) and 3rd (C) intercostal space with corresponding CT images (B & D). Panel A: From the right 4th intercostal space, with +80° rotation and dorsal angulation (Rt45V- LeDO), the cranial (CrVC) and caudal vena cava (CaVC), separated by the intervenous tubercle (IT), are seen entering the right atrium (RA). Deep to the RA, the brachiocephalic trunk (BT), aorta (Ao), right pulmonary artery (rPA), ostium III, ostium II and oval fossa (OF) with limbus can be identified. In this horse, the limbus is more clearly visible at a different time point during the cardiac cycle (insert on the left). Panel C: from the 3rd intercostal space and with dorsal and cranial angulation (Rt60V-LeDO) and +80° rotation, the right atrial appendage (RAA) and terminal crest (TC) can be seen, in addition to the cranial vena cava (CrVC), caudal vena cava (CaVC), intervenous tubercle (IT) and ostium III (III).

Left parasternal view

From a left parasternal long-axis two-chamber view (Le-Rt, rotation 0°), LA, LV and mitral valve were visualised. With slight cranial angulation (Le10Ca-RtCrO), a longitudinal section through ostium III was identified on the right side of the heart, close to the mitral valve annulus, between the CaVC and rPA (Fig. 4A&B). By positioning and/or pointing the probe slightly more dorsally (Le10Ca10V-RtCrDO), the LA was centred on the image and the dorsal aspects of the LA became better visible.

229

Chapter 11: Echocardiographic identification of equine pulmonary veins

Subsequently, slight caudal angulation (Le10Cr10V-RtCaDO) displayed ostium IV and a longitudinal section through II (Fig. 4C&D). Compared to ostium III, ostium II was located more distant from the mitral valve annulus. Ostium IV originated more dorsally and to the left of ostium II (Fig. 4C&D).

Fig. 4. Left parasternal long- axis views with echocardiographic (A&C) and corresponding 3D-reconstructed (B&D) images. Panel A: 4th intercostal space with slight cranial angulation (Le10Ca-RtCrO) and no rotation (0°), visualizing left ventricle (LV), left atrium (LA) and ostium III (III) flanked by caudal vena cava (CaVC) and right pulmonary artery (rPA). Panel C: compared to Fig. 4A, the probe is now pointing slightly more dorsally and caudally (Le10Cr10V-RtCaDO); besides left ventricle (LV) and left atrium (LA), ostium II (II) and ostium IV (IV) are now visualized. The dotted line indicates the connection of ostium II with the LA.

From the third intercostal space with cranial and slight dorsal angulation (Le45Ca10V-RtCrDO) and no rotation (0°), first, the RA, RV, pulmonary valve and the base of the pulmonary trunk (PT) were visualised. For this image, the horse’s left foot was placed more forward and substantial pressure against the triceps muscle was needed. Subsequently, the course of the PT was followed by gradually rotating the probe clockwise and angulating it at the same time more dorsally and caudally. With the probe rotated at about +80° and with dorsal and slight cranial angulation (Le15Ca60V-RtCrDO), the bifurcation of the PT into lPA and rPA was visible (Fig. 5A&B). On this image, the aorta was centrally located with the PT caudolaterally curved around it, while the CrVC and RAA located deep to the Ao.

230

Chapter 11: Echocardiographic identification of equine pulmonary veins

Fig. 5. Ultrasound (A) and corresponding CT image (B). Panel A: left parasternal oblique view from the 3rd intercostal space with +80° rotation and dorsal angulation (Le15Ca60V-RtCrDO), visualizing the bifurcation of the pulmonary trunk (PT) into left (lPA) and right pulmonary artery (rPA). The aorta (Ao) is centrally located while the cranial vena cava (CrVC) and the right atrial appendage (RAA) are deep to the Ao.

From this cranial PT bifurcation image, the probe was swapped to the 4th intercostal space, slightly more caudally angulated (Le60V-RtDO) and slightly counter clockwise rotated to +60°. On this image, the left atrium was found lateral to the longitudinally sectioned PT and its bifurcation in rPA and lPA (Fig. 6A&B), while the transverse section of the Ao was seen craniomedial to the PT. During the cardiac cycle the PT often moved in and out of the image. Adjacent to the PT bifurcation, near the lPA, a transverse section through ostium IV could be identified, entering into the LA. rPA and lPA did not always appear simultaneously on screen and rPA visualisation often required minimal beam adjustment. From this caudal PT bifurcation image, dorsal probe angulation was slightly reduced and slightly angled caudally (Le10Cr45V-RtCaDO) and rotated towards +80° until the PT disappeared and the left atrium was in the centre of the image. From this view pulmonary veins could be seen entering the left atrium. Minimal change in dorsal to ventral angulation and rotation, allowed to visualise (simultaneously or separately) a longitudinal section through ostium II and ostium III (Fig. 6C&D), whereby ostium II was found caudal to ostium III, and the rPA in between both vessels. A similar image could be obtained from the 5th intercostal space with slight cranial angulation (Le10Ca45V-RtCrDO) and the same rotation (+80°). When the beam was further angulated ventrally, the mitral valve became visible.

231

Chapter 11: Echocardiographic identification of equine pulmonary veins

Fig. 6. Left parasternal 4th intercostal echocardiographic (A&C) and corresponding CT (B&D) images from oblique views with dorsal angulation and +60° to +80° rotation. Panels A (Le60V-RtDO, rotation +60°) and B show the pulmonary trunk (PT) and the bifurcation into left (lPA) and right (rPA) pulmonary artery, caudolaterally flanked by left atrium (LA) and craniomedially by aorta (Ao). A cross section through ostium IV (IV) is found adjacent and lateral from the PT bifurcation. Panel C is taken with slightly less dorsal angulation (Le10Cr45V-RtCaDO) and a rotation of +80°, and shows the left atrium (LA) with ostium III (III) and the more caudodorsally located ostium II (II). Between both ostia the rPA is found.

232

Chapter 11: Echocardiographic identification of equine pulmonary veins

Discussion

In human medicine, transoesophageal and transthoracic (apical) views allow fairly good visualisation of the dorsal atria and pulmonary veins (Huang et al., 2008; Baer et al., 2018). The latter have been extensively studied because of their importance in the pathophysiology and therapy of atrial fibrillation, however, visualisation, measurements and therapeutic procedure guidance are almost all exclusively based on CT or MR images.

New standard views were created to visualize specific structures and anatomical landmarks and many views have become part of routine cardiac interventions in our research group. Although for small animals the Echocardiography Committee of The Specialty of Cardiology has recommended a short axis image display orientation with the cranial part of the heart right on screen (Thomas et al., 1993), we deliberately choose to have exceptions to this rule. First, in the standing horse, many equine cardiologists prefer a different probe rotation to obtain short axis images, whereby the caudal heart is displayed right on screen (Long et al., 1992). More importantly, our main goal was to allow for guidance of catheters or devices during their transvenous insertion for cardiac interventions, without suddenly having to rotate the probe almost 180°. For a right parasternal view this means starting with a probe rotation of +90° to visualize a device entering from CrVC into RA and subsequently following it towards the RV by counter clockwise probe rotation towards the four- chamber view at 0°. From this view, it was relatively easy to continue ultrasound guidance towards the pulmonary artery on the right ventricular inflow-outflow image. For ultrasound guidance from the left parasternal view, the device was visualized from a cranial view of the pulmonary valve and PT (0°). By gradually pointing more dorsally, at the same time rotating clockwise and then positioning it in the 4th intercostal space, an intracardiac device could be followed nicely along the PT towards the bifurcation. On that final image, the caudal part of the heart was positioned on the right of the screen. From this view, minimal ventral angulation displayed the left atrium with its pulmonary veins, again with the caudal aspect of the heart right on screen.

In standard echocardiography, probe handling, especially angulation, is generally not described in much detail. The heart is easy to find on ultrasound and fine-tuning of probe angulation is mainly based upon the reproduction of a standard image on screen. We noticed that, even experienced equine cardiologists, initially did find some views challenging to acquire and interpret. Indeed, the cardiac window for these images is narrow, difficult to find and on the edge of the lung field while probe manipulation is very subtle. We therefore choose to provide more detail regarding probe manipulation based upon the standardized nomenclature used for radiographic projections [25],

233

Chapter 11: Echocardiographic identification of equine pulmonary veins

approved by the American College of Veterinary Radiology. It is an easy to use system to describe oblique views in detail and the nomenclature is well known amongst veterinarians and in veterinary literature. We assessed angles of obliquity subjectively. We did not make attempts to measure these angles more accurately because due to substantial individual variation, absolute angle values are not applicable for each individual horse. We rather wanted to provide guidance on how to obtain an image and how to manoeuvre from one image to another.

Identifying specific landmarks such as IT, OF, limbus, vena cava, and the position and side branches of PV, PA and Ao has so far proven to be extremely helpful to achieve a well-controlled insertion of devices into the heart (G. van Loon, unpublished data). Visualization of OF and limbus could be very useful to develop a technique for ultrasound-guided transseptal puncture. With such a technique, access to the left heart without the need of retrograde arterial catheterisation could be obtained, which would allow pressure monitoring, electrophysiological studies, biopsy or device insertion.

Visualisation and correct identification of the PV and their relation to other vessels was also one of the aims of our study. These vessels are of particular interest to study atrial tachyarrhythmia pathophysiology in horses and might become a target for future treatment strategy. In addition, as in small animal medicine, PV diameter or flow measurements could be useful to assess specific heart disease such as mitral regurgitation severity (Tabata et al., 2003; Brewer et al., 2012; Merveille et al., 2015).

Except for the azygos vein and the PV ostium I, and in some horses the RAA and TC, most important atrial related structures could be identified in all horses. Due to individual variation and body condition, some views were more challenging to take than others and required some practicing. Imaging of the PV ostia, for example, was sometimes more difficult as these orifices are located at the border of the acoustic window. The visibility of ostium I depends on its location and shows anatomical variation in horses (Vandecasteele et al., 2016).

Equine cardiovascular imaging by radiography or MRI is very limited or impossible due to the size of the animal, making ultrasound the most important imaging modality. Our study provides new standard views for improved visualisation and 3D understanding of important anatomical landmarks and vessels in relation to the heart. Indeed, in the future, additional approaches and views may be found helpful to visualize specific regions. Thorough ultrasound knowledge of the entire heart is desired to achieve a more accurate diagnosis and to develop advanced invasive cardiac procedures in adult horses.

234

Chapter 11: Echocardiographic identification of equine pulmonary veins

References

Alt E., Ammer R., Schmitt C., Evans F., Lehmann G., Pasquantonio J. and Schömig A. (1997). A comparison of treatment of atrial fibrillation with low-energy intracardiac cardioversion and conventional external cardioversion. Eur Heart J. 18:1796-1804.

Baer J., Wyatt M.M. and Kreisler K.R. (2018). Utilizing transesophageal echocardiography for placement of pulmonary artery catheters. Echocardiogr doi: 10.1111/echo.13812.

Bluemke D.A., Achenbach S., Budoff M., Gerber T.C., Gersh B., Hillis L.D., Hundley W.G., Manning W.J., Printz B.F., Stuber M. and Woodard P.K. (2008). Noninvasive coronary artery imaging: magnetic resonance angiography and multidetector computed tomography angiography: a scientific statement from the American Heart Association Committee on Cardiovascular Imaging and Intervention of the Council on Cardiovascular Radiology and Intervention, and the Councils on Clinical Cardiology and Cardiovascular Disease in the Young. Circulation. 118:586-606.

Brewer F.C., Moïse N.S., Kornreich B.G. and Bezuidenhout A.J. (2012). Use of computed tomography and silicon endocasts to identify pulmonary veins with echocardiography. J Vet Cardiol. 14:293-300.

Broux B., De Clercq D., Decloedt A., De Baere S., Devreese M., Van der Vekens N., Ven S., Croubels S. and van Loon G. (2016). Pharmacokinetics of intravenously and orally administered sotalol hydrochloride in horses and effects on surface electrocardiogram and left ventricular systolic function. Vet J. 208:60-64.

De Clercq D., Decloedt A., Sys S.U., Verheyen T., Van Der Vekens N. and van Loon G. (2014). Atrial fibrillation cycle length and atrial size in horses with and without recurrence of atrial fibrillation after electrical cardioversion. J Vet Intern Med. 28:624-629.

Decloedt A., De Clercq D., Van Der Vekens N., Verheyen T. and van Loon G. (2014). Noninvasive determination of atrial fibrillation cycle length by atrial colour tissue Doppler imaging in horses. Equine Vet J. 46:174-179.

Decloedt A., de Clercq D., Ven S., van der Vekens N., Chiers K. and van Loon G. (2016). Right atrial and right ventricular ultrasound-guided biopsy technique in standing horses. Equine Vet J. 48:346-351.

235

Chapter 11: Echocardiographic identification of equine pulmonary veins

Farré J., Anderson R.H, Cabrera J.A., Sánchez-Quintana D., Rubio J.M., Benezet-Mazuecos J., Del Castillo S. and Macia E. (2010). Cardiac anatomy for the interventional arrhythmologist: I. terminology and fluoroscopic projections. Pacing Clin Electrophysiol. 33:497-507.

Feigenbaum H. (1986) Echocardiography. 4th edn., Lea & Febiger, Philadelphia.

Huang X., Huang Y., Huang T., Huang W. and Huang Z. (2008) Individual pulmonary vein imaging by transthoracic echocardiography: an inadequate traditional interpretation. Eur J Echocardiogr. 9:655- 660.

Javsicas L.H., Giquère S., Maisenbacher H.W., Schmidt M., Frederick J.D., Conway J.A. and Estrada A.H. (2010). Percutaneous transcatheter closure of an aorta-cardiac fistula in a Thoroughbred stallion using an Amplatzer occluder device. J Vet Intern Med. 24:994-998.

Levy S., Lauribe P., Dolla E., Kou W., Kadish A., Calkins H., Pagannelli F., Moyal C., Bremondy M. and Schork A. (1992). A randomized comparison of external and internal cardioversion of chronic atrial fibrillation. Circulation. 86:1415-1420.

Long K.J., Bonagura J.D. and Darke P.G. (1992). Standardised imaging technique for guided M-mode and Doppler echocardiography in the horse. Equine Vet J. 24:226-235.

Marr C.M. and Patteson M. (2010). Echocardiography. In: Cardiology of the Horse, 2nd edn., Ed: C.M. Marr and M. Bowen, Saunders Elsevier, Amsterdam, pp 3-21.

McGurrin M.K., Physick-Sheard P.W. and Kenney D.G. (2005). How to perform transvenous electrical cardioversion in horses with atrial fibrillation. J Vet Cardiol. 7:109-119.

Merveille A.C., Bolen G., Krafft E., Roels E., Gomart S., Etienne A.L., Clercx C. and Mc Entee K. (2015). Pulmonary Vein-to-Pulmonary Artery Ratio is an Echocardiographic Index of Congestive Heart Failure in Dogs with Degenerative Mitral Valve Disease. J Vet Intern Med. 29:1502-1509.

Palgrave K. and Kidd J.A. (2014). Introduction. In: Atlas of equine ultrasonography, 1st edn, Ed: J.A. Kidd, K.G. Lu and M.L. Frazer, John Wiley & Sons, New Jersey, pp 1-22.

Patteson M. (1999). Two-dimensional and M-mode echocardiography. In: Cardiology of the horse, Ed: C.M. Marr, W.B. Saunders, London, pp 93-116.

236

Chapter 11: Echocardiographic identification of equine pulmonary veins

Schauvliege S., van Loon G., De Clercq D., Devisscher L., Deprez P. and Gasthuys F. (2009). Cardiovascular responses to transvenous electrical cardioversion of atrial fibrillation in anaesthetized horses. Vet Anaesth Analg. 36:341-351.

Schuetz G.M., Zacharopoulou N.M., Schlattmann P. and Dewey M. (2010). Meta-analysis: noninvasive coronary angiography using computed tomography versus magnetic resonance imaging. Ann Intern Med. 152:176-177.

Smallwood J.E., Shively M.J., Rendano V.T. and Habel R.E. (1985). A standardized nomenclature for radiographic projections used in veterinary medicine. Vet Radiol. 26:2-9.

Tabata T., Thomas J.D. and Klein A.L. (2003) Pulmonary venous flow by Doppler echocardiography: Revisited 12 years later. JACC. 41:1243-1250.

Thomas W.P., Gaber C.E., Jacobs G.J., Kaplan P.M., Lombard C.W., Moise N.S. and Moses B.L. (1993). Recommendations for standards in transthoracic two-dimensional echocardiography in the dog and cat. J Vet Intern Med. 7:247-252.

Vandecasteele T., van Loon G., Vandevelde K., De Pauw B., Simoens P. and Cornillie P (2016) Topography and ultrasonographic identification of the equine pulmonary vein draining pattern. Vet J. 210:17-23. van Loon G., Fonteyne W., Rottiers H., Tavernier R., Jordaens L., D'Hont L., Colpaert R., De Clercq T. and Deprez R. (2001). Dual-chamber pacemaker implantation via the cephalic vein in healthy equids. J Vet Intern Med. 15:564-571. van Loon G., Fonteyne W., Rottiers H., Tavernier R. and Deprez P. (2002). Implantation of a dual- chamber, rate-adaptive pacemaker in a horse with suspected sick sinus syndrome. Vet Rec. 151:541- 545. van Loon G., De Clercq D., Tavernier R., Amory H. and Deprez P. (2005). Transient complete atrioventricular block following transvenous electrical cardioversion of atrial fibrillation in a horse. Vet J. 170:124-127. van Loon G., Boussy T., Vera L., De Clercq D., Schauvliege S., Vandecasteele T., Decloedt A., Van Steenkiste G. and Van Langenhove G. (2017). Electroanatomical cardiac mapping in an adult horse in sinus rhythm. J Vet Intern Med. 31:1572-1604.

237

Chapter 11: Echocardiographic identification of equine pulmonary veins

Young L.E. (2003). Equine athletes, the equine athlete’s heart and racing success. Exp Physiol. 88: 659-663.

238

Chapter 12

General discussion

239

240

Chapter 12: General discussion

Nowadays, numerous studies are conducted to investigate atrial fibrillation (AF) and to improve the treatment of AF in men. This research remains very important, not only because of the increasing importance of AF in our ever-aging population, but also because the mechanism of AF induction is still not fully understood.

In this context, our study started with the elementary research questions associated with the design of a novel device to enhance the treatment of AF in men. This led to a feasibility study in which we tested whether a wirelessly heatable stent could be implanted inside the PVs in order to obtain fast and efficient circumferential ablation of the PV wall. The most important advantage of such a device is the fact that circumferential lesions could be made much faster and with a higher success rate compared to current techniques, which are still technically demanding. By creating ablative lesions, the PVs are electrically isolated from the left atrium to prevent AF initiation from PV triggers. Moreover, an implanted device can remain in place, be overgrown and if necessary, be used for a second ablation procedure without any surgical intervention. The development and application of the device required an animal model. The pig was chosen as a cardiovascular model as the heart of pigs is comparable with that of humans and pigs are often used as an animal model for human research. However, in veterinary practice, pigs are never diagnosed with AF. The absence of AF in pigs poses no problem to evaluate an ablation procedure as PVI can be diagnosed by an electrophysiological study. On the other hand, AF is a well-known problem in the horse. As such, together with the studies conducted in pigs as animal model, the research was also extended towards horses as animal patients, and this specific part of the research will be discussed further on.

As a result, the collected histological and anatomical data was mainly used as a basis to perform preclinical interventions in the pig as a model for humans. Moreover, this information may also be used in the future for research in other fields, such as for a biomedical approach or an embryonic study into the importance and the specific way certain structures are formed. The presence of various cell types, such as ganglia, nerves, telocytes, fat and connective tissue as well as myocardial tissue, with importance in the complex mechanism of AF induction and maintenance, was cited. The function of these structures and interaction with each other could already be a histological research topic on itself. This is already the case for telocytes in cardiovascular research. However, in-depth histological research is often lacking. This is illustrated by the absence of large-scale histopathological studies of ablated human pulmonary veins. An embryological study into the formation and distribution of the intracardiac conduction tissue could yield interesting data which may help to understand the

241

Chapter 12: General discussion induction of AF or the failure of certain procedures to treat AF. Such histological and embryological data in men and horses, concerning the occurrence of the aforementioned cell types at the atriovenous junction and along the antrum of the pulmonary veins, could be a valuable addition to the electrophysiological data that is comprehensively collected. This may also provide an explanation why extensive individual variation, influencing the treatment outcome, is noticed. The combination of embryological, histological and interventional data of several species and the comparison between species such as horses, pigs and men, may clarify the role of certain cell or tissue types in the complex mechanism of AF. In addition, the collected histological data serve not only the unravelment of the AF mechanisms but may be used as basic information for further physiological studies in pigs as an animal model and in horses as cardiovascular patients.

Pigs as an animal model

Composition of the PV wall

As the PVs are important structures in the induction mechanism of AF, detailed information about the topography of the left atrium and PVs was gathered. The number and position of the PVs ostia in pigs were macroscopically investigated. Dissection of cardiopulmonary sets of pigs demonstrated the presence of two ostia but substantial variation was noticed in the branching pattern of the PV itself. This is in line with the existing literature (Barone, 1976; Constantinescu and Schaller, 2012). As a result, morphological variation may hamper the development of a standardized approach of the PVs ostia during a clinical intervention. In our porcine implantation studies, we mostly aimed for the larger ostium I. However, this ostium I and its draining PV demonstrated more variation compared to the smaller ostium II, which sometimes required us to opt for ostium II.

242

Chapter 12: General discussion

Fig. 1. Scheme visualizing AF mechanisms.

The collected anatomical data of the PVs was also supplemented with a histological approach as the histological structure of the PV wall and the coexistence of different cell types required some specification. Histological data about the composition of the PV wall is important as the wall is partially destructed during an ablation procedure. The question may arise “which tissue and cell types are affected while conducting such an intervention?” Especially the presence and extent of the myocardial sleeves are of interest. The myocardial sleeve is considered as an important structure for impulse formation and conduction of abnormal stimuli from the PVs towards the left atrium. During ablation, the myocardial tissue is destructed to induce fibrosis at the level of the applied lesions in an

243

Chapter 12: General discussion attempt to block abberant impulses from reaching the heart. In accordance with humans, myocardial sleeve tissue has also been described in pigs. In contrast with the numerous data that is available regarding the PV wall composition in man, only one paper (Nathan and Gloobe, 1970) gives detailed information about the average myocardial sleeve length in pigs, ranging between 6 and 10 mm. However, this study was based on observations in only a small number of animals. In this view, chapters 3 and 4 complement the literature regarding the histological composition of the PVs in pigs. In addition, Chapter 4 indicates the presence of cell types such as neuron bodies, organized in ganglia, and telocytes at the level of the porcine PVs, both of which may also be involved in the induction of AF. In men, fat pads containing ganglia are described at the level of the veno-atrial junction. In pigs, ganglia are found intra-and extramurally in the same region. Earlier research describing the distribution of these ganglia in dogs and men did not reveal any correlation between the distribution of the ganglia and the presence of myocardial tissue throughout the PV wall. However, we observed a similar pattern between the localisation of nervous and cardiac tissue in the porcine PVs.

Regarding the controversy about the presence of nodal-like cell types in the PVs, Mahida et al. (2015) mentioned arguments against the existence of any specialized conductive cell type. Indeed, a histological study of the PVs in dogs revealed the absence of any conduction cell type and all investigated cells were histologically identified as atrial myocytes (Hocini et al., 2002). Also connexins, proteins composing gap junctions, play a role in cell coupling and electrical conduction. As such, these might also be worth investigating in the scope of atrial arrhythmias. Literature reveals that in canine hearts, connexin distribution and morphology would be identical in the atrium and the PVs (Verheule et al., 2002). These studies demonstrate that the morphological basis and pathophysiology of AF are still not completed elucidated, which require further research.

The role of these ganglia is highly discussed in literature. Several studies pointed out that the cardiac autonomic nervous system, consisting of intrinsic and extrinsic autonomic nerves, is involved in the elicitation and persistence of AF (Lu et al., 2015). One can assume that these ganglia act as a connection between the nerve system and the conduction seen through myocardial sleeve tissue. In the general introduction, it was mentioned that studies in men and dogs prove that epicardial ganglia are innervating the PVs by sending thin nerves towards the ganglia at the veno-atrial junction of the veins. A counter-argument for this conduction pathway could be that no impulses are recovered during electrophysiological analysis and mapping procedures. However, such procedures are only performed endocardially while the nerves are located epicardially. This may be a reason why hybrid ablation procedures, which are executed both endo- and epicardially, may offer a better result in AF treatment. Debate is still ongoing about the efficacy of the autonomic ablation as an extra treatment

244

Chapter 12: General discussion strategy next to catheter ablation in case of AF (Tan et al., 2007). Some researchers demonstrated positive results with the combination strategy (Pappone et al., 2004; Nakagawa et al., 2006) whereas others found no positive difference in the treatment of AF between patients with and without denervation (Cummings et al., 2004; Lemery et al., 2006). However, further analysis shows that the results they present are very variable and that the studies are difficult to compare as the study design may be different. Moreover, the results are also influenced by the type of AF patients (paroxysmal versus persistent) and the used ablation technique. In persistent AF patients, the success rate is lower as it is expected that in those patients, the ganglia play a minor role as fibrosis and atrial remodeling will have a more prominent role in the progress of AF stabilization. Moreover, there is still no consensus on what technique will give the best results to ablate ganglionic plexi (Stavrakis and Po, 2017). One can assume that, based on these data, the link between AF and the autonomic nerve system is much more complicated than is currently expected. Tan et al. (2007) suggested individual variability as a possible explanation for these discrepancies between different studies, as some patients demonstrated more distinct autonomic triggers than others. To date, it is suggested that autonomic modulation has a supportive role in AF ablation (Tan et al., 2007). Differences in enhanced automaticity, which is the spontaneous depolarization of myocardial cells, may also contribute to individual variability in AF development.

As mentioned before, the mechanism leading to AF is not fully understood. We do know that the incidence increases with age. This could be related to age-related changes in nervous and cardiac tissue. With increasing age, both morphological and physiological changes occur at the level of the motor units of skeletal muscle cells. These structures consist of a single peripheral neuron and its associated muscle fibers (Deschenes, 2011; Kandel et al., 2012). In man, motor units disappear gradually but slowly until an age of 60 years, after which this process accelerates (Gordon et al., 2004). At higher age, the conduction velocity of efferent axons also reduces (Di Iorio et al., 2006). Literature proves that muscle weakness at higher age is not only linked with atrophy of the muscle but that changes in the nervous system might also influence this pathological condition (Manini et al., 2013). Additionally, fewer neurons are noticed in epicardiac ganglia in older people compared to children (Pauza et al., 2000). Since skeletal muscle is affected by this process at a higher age, one might wonder whether the same processes may affect cardiac muscle as well. Cardiac muscle tissue is also strongly influenced by nerve tissue and cardiac disorders such as AF show a higher prevalence in older men. One can speculate if due to the partial loss of the autonomic nerve influence on the heart, ectopic foci may arise and increase the risk of AF.

The role of telocytes is extensively studied in literature but no clear conclusions are postulated. According to current theories, it is supposed that telocytes are a particular type of fibroblasts based

245

Chapter 12: General discussion on some characteristics they have in common. However, many different cell subpopulations exist under the heading of “fibroblast” as a result of the extensive range of immunohistochemical properties that were granted to fibroblasts. On the other hand, the question arises whether telocytes are one cell type or whether the term refers to an entire cell population, also with a wide range of immunohistochemical characteristics depending on the cell function or localization. To conclude, when looking at mesenchymal stem cells as an undifferentiated cell type from which a range of differentiated cells with different functions but certain similar immunological characteristics arise during adult life (Gilbert, 2010), we pose the question if telocytes can be categorized likewise or arise as a subpopulation of fibroblasts.

Despite the present investigations, the answer to whether pigs are the ideal model to develop ablation techniques at the level of the PVs, remains inconclusive. A valuable animal model requires that the histological structure and anatomical organization of the PVs and the distribution of different cell types in the PV walls are similar to what is found in men. This should be investigated by studying human and porcine PVs histologically in one and the same study, in which the same histologal parameters are being evaluated in both species by using identical methods.

Acute ablation lesion identification

After an ablation procedure in humans, success is defined as the detection of a conduction block at the level of the applied ablation lesion, by use of electrophysiology. In an animal model however, evaluation of the lesions can also be performed by post-mortem anatomical dissection and histological visualization. The immunohistochemical identification of acute ablation lesions in chapter 7 was based on the heat denaturation of the MYBPC3 protein. This method was chosen above the direct visualization of typical tissue damage and cell death markers. In addition, one could assume that destruction of muscle tissue due to a heat or cold source may also induce several histopathological changes on a cellular level. In this context, Dos Santos et al. (2012) indicated that radiofrequency ablation of myocardial tissue in rats did not induce an apoptotic process. This means that the frequently used apoptosis-specific-stainings such as TUNEL and cleavage-caspase-3 staining would not offer any added value during an immunohistochemical investigation of ablated myocardial tissue. Therefore, we opted to visualize the ablation lesions indirectly by use of a marker for the cardiac specific myosin binding protein C (MYBPC3).

Chronic ablation lesion identification

Chronic ablation lesions were histologically identified by use of a paragon staining, based on the combination of fuchsin and toluidine blue. In general, paragon staining is used to visualize cartilage

246

Chapter 12: General discussion damage. Histological sections of implanted stents can only be made after embedding of the stents in a resin. In case this resin is still present on the sections, a regular histological staining such as H&E staining cannot be used, contrary to paragon staining. Therefore, in the field of implanted stent sectioning, paragon staining is frequently used. In the articles by Spurlock et al. (1966) and Schulz et al. (2000), the technique was described to light microscopically examine tissues embedded in an epoxy resin by means of a paragon staining. Nowadays, commercial histological embedding resins are available that can be resolved after cutting the sections, enabling the use of other staining techniques. Another histological staining, i.e. Azan-Mallory, was also used to visualize the ingrown fibrous tissue, replacing the ablated and thus denaturing tissue. Azan-Mallory is a variant of the Mallory trichrome stain, typically used for the staining of connective tissue and collagen fibers. Literature does not present data related to the use of the Azan-Mallory staining to specifically visualize ablation lesions but is very suitable to distinguish the ingrowing fibrous tissue from the rest of the heart muscle tissue. Azan-Mallory and Masson's trichrome stainings are very similar since both are trichrome stains. Masson's trichrome staining is often used during histological investigation of heart pathologies such as myocardial infarction. In the article by Suvik and Effendy (2012), it is indicated that a modified Masson's trichrome staining offers advantages over a H&E staining to visualize wound healing. On H&E sections, the collagen deposition could not be measured and differentiated, which was possible with the adapted Masson's trichrome staining, since with this latter staining a clear picture was obtained of the deposition of the collagen fibers and the re- organization of the tissue. Trichrome stainings such as Masson's trichrome and Azan-Mallory will also visualize the cross-striations more clearly than a regular H&E staining.

Superior (cranial) vena cava

While the PVs are the main focus of this thesis in the context of AF eliciting foci, literature indicates that in certain cases, the vena cava may play a similar role in the induction of AF. Morphologically, the PVs demonstrate myocardial sleeves, as myocardial tissue fans out into the proximal part of the PV wall. The cranial vena cava (superior vena cava in man), however, demonstrates a similar structure. Apart from the electrophysiological consequences, this arrangement was also from a pure morphological point of view beneficial to our ablation studies. In fact, it was sometimes difficult to deliver a stent inside the porcine PVs because of the short antrum. In these cases, the implanted stent is likely to dislodge and migrate to the left atrium. The superior vena cava could also be used as implantation site since its wall demonstrates similar myocardial sleeves and the vessel itself is a simple elongated tubular structure lengthy enough to contain the stent. Hence, it was concluded that the vena cava could be used as a worthy alternative.

247

Chapter 12: General discussion

Stent ablation complications

Our researched procedure includes the implantation of a self-expanding nitinol stent. As this stent may remain at the same place in the PVs for a long time, tissue reaction can be expected. Total coverage of the stent by neo-intima is desired as long as the neo-intima formation is not too excessive. The induction of excessive neo-intima formation can lead to pulmonary vein stenosis (PVS) in certain cases. However, the implants used in our experiments did stretch the PV wall and created an enlarged lumen. This provides a larger buffer before neo-intima formation can lead to complete stenosis of these veins. In that way, our custom-made implants compose a major advantage compared to the stents used today in the treatment of PVS as these stents do not provide any enlargement of the antral lumen. Moreover, the next step in the stent development is to make the supporting part of the next generation of custom-made stents biodegradable so that only the burning ring will remain at the level of the PV antrum. This may also reduce the amount of reactive tissue on the implanted stent and thus reduce the risk for stenosis.

PVS is not the only possible complication which may occur during such invasive cardiopulmonary interventions. In humans, radiofrequency ablation procedures at the level of the PV may induce coagulum formation. Coagulum is caused by heat-induced denaturation of blood plasma proteins. Since thrombin is not involved in this process, the use of heparin will not prevent this reaction. Coagulum formation at the surface of a catheter may also pose a risk for thromboembolism (Demolin et al., 2002). Therefore, human patients are checked for thromboembolism in the brain with a MRI scan after this type of intervention (Anselmino et al., 2013). In our study, coagulum or thromboembolism formation was also a major concern. However, no MRI scan of the porcine brain was performed. Such investigation would also be void because of a particular morphological feature of the arterial blood supply to the porcine brain: the retia mirabilia. Before reaching the circulus arteriosus (Willis), arterial blood on the way towards the brain must pass a rete mirabile which spreads out over the larger confluence area of the intercranial branches of the internal carotid and maxillary arteries. This maze-like network of very fine arterial vessels may preclude thrombi from embolizing the brain. A better alternative to review for thromboembolizations is the gross pathological inspection of the porcine kidneys in the sacrificed animals.

Not only the possible complications which may arise during the ablation procedure itself were taken into account. In certain human patients, the morphology of the dorsal part of the left atrium may pose a problem to implant a device into the PVs. Humans demonstrate variation in PV organization up to the extent that even a single, common ostium can be observed in certain cases. These patients may be difficult to treat with our described technique as it is an important requirement that the

248

Chapter 12: General discussion implanted stents exactly fit the PVs. Therefore, in a first step, preprocedural CT images of the patient’s heart should be made to analyze the 3D-morphology of the PVs and to determine whether, based on these data, the patient is eligible for this treatment. This would be in accordance with the patient selection for cryoballoon therapy.

During the implantation trials, it was clear that the 3D topography, localization and orientation of the ostia and PVs in relation to the thorax orientation were very important. However, this 3D information was not available in our study which made it difficult to determine the ideal position of the coil around the pig. The coil needs to be precisely positioned around the pig’s thorax as the orientation of the coil and the implanted stent, relative to one another, has a major impact on the energy development of the system. This 3D data could have been obtained by placing a pig with an implanted stent into a CT scan.

Genetic predisposition

Another possibly important contributing factor in AF is genetic predisposition, and especially the variation at the level or in close vicinity of the PITX2 gene seems to be involved. This gene plays a major role during the development of myocardial sleeve tissue (Mommersteeg et al., 2007; Wang et al., 2010). In several population genetic studies, specific genetic variations located in the proximity of the PITX2 locus showed a consistent and strong association with AF susceptibility (Gudbjartsson et al., 2007; Ellinor et al., 2012). A few studies were conducted with mouse and rat models as well, showing the same link between altered expression of the PITX2 gene and susceptibility for arrhythmias (Kirchhof et al., 2011; Scridon et al., 2015). PITX2 deficiency studies resulted in electrical and structural remodeling in a murine model (Syeda et al., 2017). These studies indicate the need for more research to clarify the genetic link of arrhythmias induced by the PVs (Mahida et al., 2015).

Horses as AF patients

It is assumed that the PVs play a similar role as in human patients. The large size of the equine atria is a predisposing factor for the induction and maintenance of AF. Indeed, larger atria can contain multiple re-entry waves, which reduce the chance that all these waves disappear simultaneously. If an impuls needs to travel further to reach the same spot, delineated by an anatomical structure, the chance gets higher that this spot is already re-excitable. In that way, this abnormal conduction pathway can be sustained.

AF research and device development in horses requires detailed information about the morphology of the PVs. In accordance with the research conducted in pigs, the left atrium and PVs were investigated in horses, mainly via silicon casting of cardiopulmonary sets. The topography of the

249

Chapter 12: General discussion equine PVs can be described as each part of both lungs is drained by its own vein into the left atrium. In general, four ostia were observed but more variation was seen in this species compared to pigs. In certain cases, only three ostia were seen as ostium I can be absent. Although, a significant number of hearts were assessed, it cannot be excluded that no other branching patterns may occur.

It is indispensable to recognize the ostia of the PVs on ultrasound. Therefore, the next step was to visualize these structures in the living horse as this could potentially lead to the development of new diagnostic or treatment modalities. The equine PVs are difficult to study as imaging techniques such as CT and MRI cannot be used to visualize the thoracic content in horses. Recently, a robot CT imaging technique was developed for horses. However, image quality was insufficient to get 3D insight into the heart and surrounding blood vessels. Therefore, echocardiography remains the only valuable tool to visualize the equine PVs in vivo.

Extensive expertise on echocardiography is present in our research group. Therefore, a study was set up to describe ultrasonographic imaging of this dorsal heart region in unsedated, standing, healthy horses. The dorsal region of the heart is characterized by a tangle of vessels which makes it difficult to identify all the different structures on ultrasound. Based on our morphological findings, a better insight in the dorsal heart region and drainage area of the PVs was obtained. This data served as a basis to study the equine PVs echocardiographically and to describe the position of the PVs in relation to its surrounding structures. In general, three out of four ostia (II, III and IV) are recognizable on ultrasound whereas ostium I can only be observed exceptionally. This is mainly due to the anatomical variation and small dimensions of ostium I, as well as the fact that this ostium is sometimes absent.

The silicone casts were useful to unravel the anatomy and drainage pattern of the PVs. However, it remained very difficult to obtain a 3D insight into the ultrasound images. Therefore, we concluded that a 3D model would provide the missing information. In order to identify the unknown atrial- related structures on echocardiographic images of a horse, 3D CT-segmented images of equine casted hearts were used. The casting was performed by injecting self-expanding polyurethane foam into a post-mortem dissected heart. The CT-slices were obtained by placing the casted hearts directly into a CT scan. All CT-slices were labeled and 3D reconstructed to build up a 3D model. Only minor incongruences could be noticed between the echo and CT images, caused by the manipulation and removal of the cardiopulmonary set from the thorax and the amount of expansion by the foam. However, these slight differences were not problematic to allow proper identification of the specific structures. An in-situ casting of the lungs and heart could be a solution, especially in a formalized specimen, as this would minimize the movement of the organs when opening the thorax. In addition,

250

Chapter 12: General discussion the technique revealed just how difficult it is to expand the heart chambers to their in vivo volume. Pressures, comparable to the in vivo situation, could not be verified, however, the self-expanding foam seemed the preferred product whilst still being easy to use. With this information, an experienced operator can clearly visualize the PVs and its ostia, although, it remains a technically difficult procedure. In the future, these echocardiographic data may serve as a guide for measurement of PV size and flow, and in order to perform left atrial and PV electroanatomical mapping. This technique could provide essential information about PV involvement in AF pathophysiology. On long term, this visualization and mapping of the PVs could allow to explore ablation procedures in horses.

Diameter of the equine PVs ostia

Detailed data about the pulmonary venous anatomy and dimensions is important during an intervention at that level. To study the changes of the pulmonary venous dimensions during the cardiac cycle, it is necessary to be able to define the ostia of the pulmonary veins during ultrasound investigation. Moreover, the exact delineation of the atrio-venous transition will allow the measurement of the diameter of the different PVs ostia. In humans, this diameter is used as a predictor for PV isolation success as Hauser et al. (2015) demonstrated that a larger PV size is correlated with a higher incidence of AF recurrence after PV isolation. In horses, we studied the PVs ostia both by anatomical dissection and ultrasound investigation. However, the anatomical organization of the equine PVs makes it difficult to measure the diameter as the delineation of the ostia is complicated by the gradual transition from the left atrium towards the veins. This is in contrast to the situation in humans where the PVs show a more abrupt transition. During equine echocardiography, ostium III was indicated as a good candidate for diameter measurements as this ostium is consistenly present and lies right next to the right pulmonary artery. This ultrasound image can be an ideal reference position for measuring the diameter of this ostium in a consistent way.

Atrial conduction in the equine heart

Compared to the major atrial conduction pathways in humans, it is assumed that similar internodal tracts constitute an important part of the electrical conduction between the sinoatrial and atrioventricular node in horses. However, less research was conducted in horses and thus detailed physiological and morphological data is lacking. In horses, the Bachmann’s bundle is one of the four conduction tracts which constitute the atrial conduction system and the only one which innervates the left atrium. The atrial conducting tracts or internodal pathways are of utmost importance to transmit rapidly the electrical pulses through the atria and towards the AV node (Muir et al., 2008).

251

Chapter 12: General discussion

Histological investigation of the equine PVs

The equine PVs were also histologically investigated. Little histological data about the PVs is available. In one article based upon one animal, the length of the myocardial sleeve was described, varying between 9 and 13 mm (Nathan and Gloobe, 1970). This data is essential to develop ablation techniques and to interpret ablation success or failure. In all horses, myocardial sleeve tissue was observed and the sleeves were longer than in pigs. This was likely due to the animal size and the longer distance between heart and lungs. In horses, thick nerve bundles, parallel with the PVs, were observed but ganglia were only found at the veno-atrial junction. This is not in accordance with the distribution of ganglia in pigs and men as in both, ganglia were found along the length of the PVs. The nerve bundle distribution in our samples corresponds with samples of men and pigs.

Future prospects

Chapters 5, 6 and 8 describe the development of a new ablation device with the purpose to treat AF patients in a fast and efficient way in the future. On its own, the developed technology to heat up a stent wirelessly provides interesting data and opportunities and raises the question whether this device could also be used for other purposes. In the case of chronic hypertension or chronic kidney disease, renal denervation is carried out to stop the sympathetic hyperactivation of the kidney. This is done by renal nerve ablation at the level of the renal artery as alongside this artery renal efferent pre-ganglionic nerves and afferent nerves are found. Indeed, the kidney is strongly innervated by a network of renal afferent and efferent sympathetic fibers (Barajas et al., 1992; Papademetriou et al., 2014). Implantation of the newly developed stent inside the renal arteries and subsequently heating of this stent to ablate the sympathetic nerves may lead to an efficient denervation of the kidney.

In case of refractory idiopathic pulmonary hypertension, no cure is available to date. However, ablation of the pulmonary trunk was performed in a case study and provided promising results (Kiuchi et al., 2015). With regard to this pathology, a stent could also be implanted inside the pulmonary trunk to perform an ablation procedure to destroy certain structures inside the pulmonary trunk wall, as the hypertension is thought to be caused by an imbalance between vasoconstrictors and vasodilators which are locally released (Galiè et al., 2009). Additionally, vascular wall remodeling would also be involved in increased pulmonary vascular resistance (Hoeper et al., 2009). Furthermore, there are several other diseases like oesophageal stenosis, tracheomalacie, bile tract stenosis, prostate hypertrophy and diseases that involve narrowing of a tract of any sort, in which the developed implantable stent may offer an added value regarding the treatment of these pathologies.

252

Chapter 12: General discussion

As mentioned before, in certain porcine cases, the length of the tested stents posed a problem to fit completely into the antrum of the PVs, allowing the distal part of the stent to partly cover the debouchment of a PV. During the clinical trials, it became clear that in most cases, the antrum in pigs is shorter than the PVs antrum in humans, thus this is less likely to occur in the latter. PVs of which the blood flow is partly blocked by a stent would not induce a problem on a short term but this can lead to a complete stenosis after a couple of months. To solve this problem, biodegradable stents could be used in the future. As a result, only the proximal burning ring would remain after some time and the distal support part would have disappeared. Biodegradable stents have also been tested in patients with a stricture of the pancreatic duct and may be a promising treatment in the future (Cahen et al., 2017).

In the next step of our stent development, the current stent design may also be adapted by use of covered or treated implants instead of bare-metal stents to limit or control the formation of excessive neo-intima. As mentioned before, polytetrafluoroethylene covered stents were already used to treat refractory PVS in men. It was indicated that these treated stents offer new options to tackle the problem of neo-intima formation by creation of a circumferential boundary between the venous lumen and the wall (Gordon and Moore, 2010). These covered stents may eventually be loaded with anti-arrhythmic drugs which can be distributed locally to stop abnormal conduction as well as help in maintaining the sinus rhythm.

In addition to the optimalization of stent design and ablation procedure, there are still other challenges discussed in this manuscript. Especially on a histological or ultrastructural level, further research needs to be done, for example to unravel the role of different cells or structures in the induction of AF. It has already been mentioned that literature is not unambiguous about the identification of telocytes. The immunohistochemical identification of telocytes on a light microscopic level remains a very difficult task. Numerous articles state to differentiate telocytes from fibroblasts but this appears to be very ambiguous. Bei et al. (2015) mentioned cardiac telocytes in primary culture to be CD34 positive and fibroblast cultures which are CD34 negative. However, other studies point out that CD34 is expressed by multipotent mesenchymal stromal cells and non- hematopoietic cell types such as interstitial cells (Sidney et al., 2014). A possible explanation could lie in the fact that this marker may be highly tissue-specific and that different studies do not always examine the same tissue. To conclude whether CD34 only binds telocyte-like cells or a range of different cardiac cell types, an immunogold staining of myocardial tissue with this antibody should be performed.

253

Chapter 12: General discussion

Overall, the aim of an ablation procedure in the context of AF is to induce a permanent PVI. As can be concluded from all previous data, the ideal treatment development is still hampered by the incomplete understanding of the exact induction mechanism of AF. As a result, Bunch and Cutler (2015) posed, in a review article, the question “Is PVI still the cornerstone in atrial fibrillation ablation?” This research describes cases of AF ablation successes, even without the presence of a complete conduction block. This may be explained as in these particular cases extrapulmonary vein tissue, such as ganglia, was sufficiently ablated. This leads to the question whether in the future, ablation strategy development needs to focus on the modification of PVI methods alone or also on the adaptation of these methods in combination with the inclusion of extrapulmonary vein ablation targets.

254

Chapter 12: General discussion

References

Anselmino M., Matta M., Toso E., Ferraris F., Castagno D., Scaglione M., Cesarani F., Faletti R. and Gaita F. (2013). Silent cerebral embolism during atrial fibrillation ablation: pathophysiology, prevention and management. J Atr Fibrillation. 6:796.

Barajas L., Liu L. and Powers K. (1992). Anatomy of the renal innervation: Intrarenal aspects and ganglia of origin. Can J Physiol Pharmacol. 70:735-749.

Barone R. (1976). Poumons. In: Anatomie compare des mammifères domestiques, tome 3, splanchnologie, appareil digestif - appareil respiratoire. Tixier & Fils, Lyon, pp.757-799.

Bei Y., Zhou Q., Fu S., Ly D., Chen P., Chen Y., Wang F. and Xiao J. (2015). Cardiac telocytes and fibroblasts in primary culture: different morphologies and immunophenotypes. PLoS One. 10:e0115991. doi:10.1371/journal.pone.0115991.

Bugge E., Nicholson I.A. and Thomas S.P. (2005). Comparison of bipolar and unipolar radiofrequency ablation in an in vivo experimental model. Eur J Cardiothorac Surg. 28:76-80.

Bunch T.J. and Cutler M.J. (2015). Is pulmonary vein isolation still the cornerstone in atrial fibrillation ablation? J Thorac Dis. 7:132-141.

Cahen D.L., van der Merwe S.W., Laleman W., Poley J.W. and Bruno M.J. (2017). A biodegradable non-covered self-expandable stent to treat pancreatic duct strictures in chronic pancreatitis: a proof of principle. Gastrointest Endosc. doi: 10.1016/j.gie.2017.08.018.

Chen P.-S., Maruyama M. and Lin S.-F. (2010). Arrhythmogenic foci and the mechanisms of atrial fibrillation. Circ Arrhythm Electrophysiol. 3:7–9.

Constantinescu G.M. and Schaller O. (eds.) (2012). Angiologia. In: Illustrated Veterinary Anatomical Nomenclature, Enke, Stuttgart, pp.234–413.

Copley D.J. and Hill K.M. (2016). Atrial fibrillation: a review of treatments and current guidelines. AACN Adv Crit Care. 27:120-128.

Cummings J.E., Gill I., Akhrass R., Dery M., Biblo L.A. and Quan K.J. (2004). Preservation of the anterior fat pad paradoxically decreases the incidence of postoperative atrial fibrillation in humans. J Am Coll Cardiol. 43:994–1000.

255

Chapter 12: General discussion

Danias P.G., Caulfield T.A., Weigner M.J., Silverman D.I. and Manning W.J. (1998). Likelihood of spontaneous conversion of atrial fibrillation to sinus rhythm. J Am Coll Cardiol. 31:588-592.

Datino T., Macle L., Qi X.-Y., Maguy A., Comtois P., Chartier D., Guerra P.G., Arenal A., Fernández- Avilés F. and Nattel S. (2010). Mechanisms by which adenosine restores conduction in dormant canine pulmonary veins. Circulation. 121:963–972.

Demolin J.M., Eick O.J., Munch K., Koullick E., Nakagawa H. and Wittkampf F. (2002). Soft thrombus formation in radiofrequency catheter ablation. Pacing Clin Electrophysiol. 25:1219–1222.

Deschenes M.R. (2011). Motor unit and neuromuscular junction remodeling with aging. Curr Aging Sci. 4:209–220.

Di Iorio A., Cherubini A., Volpato S., Sparvieri E., Lauretani F., Franceschi C., Senin U., Abate G., Paganelli R., Martin A., Andres-Lacueva C. and Ferrucci L. (2006). Markers of inflammation, vitamin E and peripheral nervous system function: the InCHIANTI study. Neurobiol Aging. 27:1280–1288.

Dos Santos L.F., Antonio E., Serra A., Venturini G., Okada M., Araújo S., Tucci P., De Paola A. and Fenelon G. (2012). Radiofrequency ablation does not induce apoptosis in the rat myocardium. PACE. 35:449-455.

Ellinor P.T., Lunetta K.L., Albert C.M., Glazer N.L., Ritchie M.D., Smith A.V., Arking D.E., Müller- Nurasyid M., Krijthe B.P., Lubitz S.A., Bis J.C., Chung M.K., Dörr M., Ozaki K., Roberts J.D., Smith J.G., Pfeufer A., Sinner M.F., Lohman K., Ding J., Smith N.L., Smith J.D., Rienstra M., Rice K.M., Van Wagoner D.R., Magnani J.W., Wakili R., Clauss S., Rotter J.I., Steinbeck G., Launer L.J., Davies R.W., Borkovich M., Harris T.B., Lin H., Völker U., Völzke H., Milan D.J., Hofman A., Boerwinkle E., Chen L.Y., Soliman E.Z., Voight B.F., Li G., Chakravarti A., Kubo M., Tedrow U.B., Rose L.M., Ridker P.M., Conen D., Tsunoda T., Furukawa T., Sotoodehnia N., Xu S., Kamatani N., Levy D., Nakamura Y., Parvez B., Mahida S., Furie K.L., Rosand J., Muhammad R., Psaty B.M., Meitinger T., Perz S., Wichmann H.E., Witteman J.C., Kao W.H., Kathiresan S., Roden D.M., Uitterlinden A.G., Rivadeneira F., McKnight B., Sjögren M., Newman A.B., Liu Y., Gollob M.H., Melander O., Tanaka T., Stricker B.H., Felix S.B., Alonso A., Darbar D., Barnard J., Chasman D.I., Heckbert S.R., Benjamin E.J., Gudnason V. and Kääb S. (2012). Meta-analysis identifies six new susceptibility loci for atrial fibrillation. Nat Genet. 44:670-675.

Ehrlich J.R., Cha T.-J., Zhang L., Chartier D., Melnyk P., Hohnloser S.H. and Nattel S. (2003). Cellular electrophysiology of canine pulmonary vein cardiomyocytes: action potential and ionic current properties. J Physiol. 551:801–813.

256

Chapter 12: General discussion

Falk R.H. (1992). Proarrhythmia in patients treated for atrial fibrillation or flutter. Ann Intern Med. 117:141-150.

Falk R.H. (2001). Atrial fibrillation. N Engl J Med. 344:1067-1078.

Galiè N., Manes A., Negro L., Palazzini M., Bacchi-Reggiani M.L. and Branzi A. (2009). A meta-analysis of randomized controlled trials in pulmonary arterial hypertension. Eur Heart J. 30:394–403.

Gilbert S.F. (2010). The stem cell concept. In: Developmental Biology, 9th edition. MA: Sinauer Associates, pp 330-331.

Gillinov A.M. and Saltman A.E. (2007). Ablation of atrial fibrillation with concomitant cardiac surgery. Semin Thorac Cardiovasc Surg. 19:25-32.

Gordon T., Hegedus J. and Tam S.L. (2004). Adaptive and maladaptive motor axonal sprouting in aging and motoneuron disease. Neurol Res. 26:174–185.

Gordon B.M. and Moore J.W. (2010). Treatment of pulmonary vein stenosis with expanded polytetrafluoroethylene covered stents. Cath Cardiovasc Interv. 75:263-267.

Gourraud J.-B., Andrade J.G., Macle L. and Mondésert B. (2016). Pharmacological tests in atrial fibrillation ablation. Arrhythm Electrophysiol Rev. 5:170-176.

Gudbjartsson D.F., Arnar D.O., Helgadottir A., Gretarsdottir S., Holm H., Sigurdsson A., Jonasdottir A., Baker A., Thorleifsson G., Kristjansson K., Palsson A., Blondal T., Sulem P., Backman V.M., Hardarson G.A., Palsdottir E., Helgason A., Sigurjonsdottir R., Sverrisson J.T., Kostulas K., Ng M.C., Baum L., So W.Y., Wong K.S., Chan J.C., Furie K.L., Greenberg S.M., Sale M., Kelly P., MacRae C.A., Smith E.E., Rosand J., Hillert J., Ma R.C., Ellinor P.T., Thorgeirsson G., Gulcher J.R., Kong A., Thorsteinsdottir U. and Stefansson K. (2007). Variants conferring risk of atrial fibrillation on chromosome 4q25. Nature. 448:353-357.

Hauser T.H., Essebaq V., Baldessin F., McClennen S., Yeon S.B., Manninq W.J. and Josephson M.E. (2015). Prognostic value of pulmonary vein size in prediction of atrial fibrillation recurrence after pulmonary vein isolation: a cardiovascular magnetic resonance study. J Cardiovasc Magn Reson. 17:49.

Hocini M., Ho S.Y., Kawara T., Linnenbank A.C., Potse M., Shah D., Jaïs P., Janse M.J., Haïssaguerre M. and De Bakker J.M. (2002). Electrical conduction in canine pulmonary veins: electrophysiological and anatomic correlation. Circulation. 105:2442-2448.

257

Chapter 12: General discussion

Hoeper M.M., Barberia J.A., Channick R.N., Hassoun P.M., Lang I.M., Manes A., Martinez F.J., Naeije R., Olschewski H., Pepke-Zaba J., Redfield M.M., Robbins I.M., Souza R., Torbicki A. and McGoon M. (2009). Diagnosis, assessment, and treatment of non-pulmonary arterial hypertension pulmonary hypertension. J Am Coll Cardiol. 54:85–96.

January C.T., Wann L.S., Alpert J.S., Calkins H., Cigarroa J.E., Cleveland J.C. Jr., Conti J.B., Ellinor P.T., Ezekowitz M.D., Field M.E., Murray K.T., Sacco R.L., Stevenson W.G., Tchou P.J., Tracy C.M. and Yancy C.W. (2014). 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol. 64:1-76.

Kandel E.R., Schwartz J.H., Jessell T.M., Siegelbaum S.A. and Hudspeth A.J. (2012). Principles of Neural Science. 5. McGraw-Hill Companies, Inc.

Kirchhof P., Kahr P.C., Kaese S., Piccini I., Vokshi I., Scheld H.H., Rotering H., Fortmueller L., Laakmann S., Verheule S., Schotten U., Fabritz L. and Brown N.A. (2011). PITX2c is expressed in the adult left atrium, and reducing Pitx2c expression promotes atrial fibrillation inducibility and complex changes in gene expression. Circ Cardiovasc Genet. 4:123-133.

Kiuchi M.G., Andrea B.R., da Silva G.R., Coelho S.B.P., Paz L.M.R., Shaojie C. and Souto G.L.L. (2015). Pulmonary artery ablation to treat pulmonary arterial hypertension: a case report. J Med Case Rep. 9:284.

La Meir M. (2013). New technologies and hybrid surgery for atrial fibrillation. Rambam Maimonides Med J. 4:e0016.

Laupacis A., Albers G., Dalen J., Marvin I.D., Jacobson A.K. and Singer D.E. (1998). Antithrombotic therapy in atrial fibrillation. Chest. 114:579S-589S.

Lee S.-H., Tai C.-T., Hsieh M.-H., Tsao H.M., Lin Y.J., Chang S.L., Huang J.L., Lee K.T., Chen Y.J., Cheng J.J. and Chen S.A. (2005). Predictors of non-pulmonary vein ectopic beats initiating paroxysmal atrial fibrillation: implication for catheter ablation. J Am Coll Cardiol. 46:1054–1059.

Lemery R., Birnie D., Tang A.S., Green M. and Gollob M. (2006). Feasibility study of endocardial mapping of ganglionated plexuses during catheter ablation of atrial fibrillation. Heart Rhythm. 3:387– 396.

258

Chapter 12: General discussion

Liang B.T., Frame L.H. and Molinoff P.B. (1985). Beta 2-adrenergic receptors contribute to catecholamine-stimulated shortening of action potential duration in dog atrial muscle. Proc Natl Acad Sci. 82:4521–4525.

Lu Y., Sun J., Zhou X., Sun Q., Sun B.S. and Tang B. (2015). Atrial Fibrillation Electrical Remodeling via Ablation of the Epicardial Neural Networks and Suprathreshold Stimulation of Vagosympathetic Nerve. Med Sci Monit. 21: 82–89.

Luik A., Radzewitz A. Kieser M., Walter M., Bramlage P., Hörmann P., Schmidt K., Horn N., Brinkmeier- Theofanopoulou M., Kunzmann K., Riexinger T., Schymik G., Merkel M. and Schmitt C. (2015). Cryoballoon Versus Open Irrigated Radiofrequency Ablation in Patients With Paroxysmal Atrial Fibrillation: The Prospective, Randomized, Controlled, Noninferiority FreezeAF Study. Circulation. 132:1311-1319.

Mahida S., Sacher F., Derval N., Berte B., Yamashita S., Hooks D., Denis A., Amraoui S., Hocini M., Haïssaguerre M. and Jais P. (2015). Science linking pulmonary veins and atrial fibrillation. Arrhythm Electrophysiol Rev. 4:40-43.

Manini T.M., Hong S.L. and Clark B.C. (2013). Aging and muscle: a neuron’s perspective. Curr Opin Clin Nutr Metab Care. 16:21-26.

Miyazaki S., Taniguchi H., Hachiya H., Nakamura H., Takagi T., Hirao K. and Iesaka Y. (2016). Durability of cryothermal pulmonary vein isolation - Creating contiguous lesions is necessary for persistent isolation. Int J Cardiol. 220:395-399.

Mommersteeg M.T., Brown N.A., Prall O.W., de Gier-de Vries C., Harvey R.P., Moorman A.F. and Christoffels V.M. (2007). Pitx2c and Nkx2-5 are required for the formation and identity of the pulmonary myocardium. Circ Res. 101:902-909.

Muir W.W. and Hubbell J.A.E. (2008). Equine anesthesia: monitoring and emergency therapy, The cardiovascular system, pp. 37-61, Elsevier Health Sciences, Edinburgh.

Nakagawa H., Scherlag B.J., Wu R., Po S., Lockwood D., Yokoyama K., Herring L., Lazzara R. and Jackman W.M. (2006). Addition of selective ablation of autonomic ganglia to pulmonary vein antrum isolation for treatment of paroxysmal and persistent atrial fibrillation (Abstract) Circulation. 110:III- 459.

Nathan H. and Gloobe H. (1970). Myocardial atrio-venous junctions and extensions (sleeves) over the pulmonary and caval veins. Anatomical observations in various mammals. Thorax. 25:317.

259

Chapter 12: General discussion

Ninet J., Roques X., Seitelberger R., Deville C., Pomar J.L., Robin J., Jegaden O., Wellens F., Wolner E., Vedrinne C., Gottardi R., Orrit J., Billes M.A., Hoffmann D.A., Cox J.L. and Champsaur G.L. (2005). Surgical ablation of atrial fibrillation with off-pump, epicardial, high-intensity focused ultrasound: results of a multicenter trial. J Thorac Cardiovasc Surg. 130:803-809.

Oral H., Souza J.J., Michaud G.F., Knight B.P., Goyal R., Strickberger S.A. and Morady F. (1999). Facilitating Transthoracic Cardioversion of Atrial Fibrillation with Ibutilide Pretreatment. N Engl J Med. 340:1849-1854.

Papademetriou V., Doumas M., Anyfanti P., Faselis C., Kokkinos P. and Tsioufis C. (2014). Renal Nerve Ablation for Hypertensive Patients with Chronic Kidney Disease. Curr Vasc Pharmacol. 12:47-54.

Pappone C., Santinelli V., Manguso F., Vicedomini G., Gugliotta F., Augello G., Mazzone P., Tortoriello V., Landoni G., Zangrillo A., Lang C., Tomita T., Mesas C., Mastella E. and Alfieri O. (2004). Pulmonary vein denervation enhances long-term benefit after circumferential ablation for paroxysmal atrial fibrillation. Circulation. 109:327–334.

Pauza D.H., Skripka V., Pauziene N. and Stropus R. (2000). Morphology, distribution, and variability of the epicardiac neural ganglionated subplexuses in the human heart. Anat Rec. 259:353–382.

Prasad S.M., Maniar H.S., Camillo C.J., Schuessler R.B., Boineau J.P., Sundt T.M. 3rd, Cox J.L. and Damiano R.J. Jr. (2003). The Cox maze III procedure for atrial fibrillation: long-term efficacy in patients undergoing lone versus concomitant procedures. J Thorac Cardiovasc Surg. 126:1822-1828.

Providencia R., Defaye P., Lambiase P.D., Pavin D., Cebron J.P., Halimi F., Anselme F., Srinivasan N., Albenque J.P. and Boveda S. (2017). Results from a multicentre comparison of cryoballoon vs. radiofrequency ablation for paroxysmal atrial fibrillation: is cryoablation more reproducible? Europace. 19:48-57.

Schulz C., Herrmann R.A., Beilharz C., Pasquantonio J. and Alt E. (2000). Coronary stent symmetry and vascular injury determine experimental restenosis. Heart. 83 :462-467.

Scridon A., Fouilloux-Meugnier E., Loizon E., Rome S., Julien C., Barrès C. and Chevalier P. (2015). Long-standing arterial hypertension is associated with Pitx2 down-regulation in a rat model of spontaneous atrial tachyarrhythmias. Europace. 17:160-165.

Sidney L.E., Branch M.J., Dunphy S.E., Dua H.S. and Hopkinson A. (2014). Concise review: evidence for CD34 as a common marker for diverse progenitors. Stem Cells. 32:1380–1389.

260

Chapter 12: General discussion

Spurlock B.O., Skinner M.S. and Kattine A.A. (1966). A simple rapid method for staining epoxy- embedded specimens for light microscopy with the polychromatic stain paragon. Am J Clin Pathol. 46:252-258.

Stambler B.S., Wood M.A., Ellenbogen K.A., Perry K.T., Wakefield L.K. and Vanderlugt J.T. (1996). Efficacy and safety of repeated intravenous doses of ibutilide for rapid conversion of atrial flutter or fibrillation. Ibutilide Repeat Dose Study Investigators. Circulation. 94:1613-1621.

Stavrakis S. and Po S. (2017). Ganglionated plexi ablation: physiology and clinical applications. Arrhythm Electrophysiol Rev. 6:186-190.

Suvik A. and Effendy A.W.M. (2012). The use of modified Masson’s trichrome staining in collagen evaluation in wound healing study. Mal J Vet Res. 3:39-47.

Syeda F., Kirchhof P. and Fabritz L. (2017). PITX2-dependent gene regulation in atrial fibrillation and rhythm control. J Physiol. 595:4019-4026.

Tan A.Y., Chen P.-S., Chen L.S. and Fishbein M.C. (2007). Autonomic nerves in pulmonary veins. Heart Rhythm. 4:57–S60.

Verheule S., Wilson E.E., Arora R., Engle S.K., Scott L.R. and Olgin J.E. (2002). Tissue structure and connexin expression of canine pulmonary veins. Cardiovasc Res. 55:727-738.

Wasserlauf J., Pelchovitz D.J., Rhyner J., Verma N., Bohn M., Li Z., Arora R., Chicos A.B., Goldberger J.J., Kim S.S., Lin A.C., Knight B.P. and Passman R.S. (2015). Cryoballoon versus radiofrequency catheter ablation for paroxysmal atrial fibrillation. Pacing Clin Electrophysiol. 38:483-489.

Wolf P.A., Abbott R.D. and Kannel W.B. (1991). Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 22:983-988.

Xu J., Luc J.G.Y. and Phan K. (2016). Atrial fibrillation: review of current treatment strategies. J Thorac Dis. 8:886-900.

261

262

Summary

263

264

Summary

Atrial fibrillation (AF) is the most important cardiac arrhythmia in men. Although the mechanism behind the elicitation of this pathology has not been fully elucidated, different treatment strategies were developed in the last decades. Ablation of the PVs remains the cornerstone in AF treatment in men; however, due to varying results, more research is still needed to improve the outcome of those ablation strategies. The development of a new AF treatment method for humans requires an animal model. Therefore, the pig was selected as the porcine heart is comparable with the heart in humans. Yet, at the start of this thesis, little information was available on the porcine PVs histology, morphology and variability. This information is indispensable to develop and implant an ablation device in the porcine PV.

Horses, on the other hand, are also prone to develop AF but a treatment method to prevent long- term recurrence is not yet available. Due to the size of a horse, visualization of the heart of an adult horse is limited to echocardiography. This complicates the development of an AF treatment strategy. Moreover, detailed cardiac ultrasound information on the equine PVs was lacking at the beginning of this study.

The general introduction (Chapter 1) of this doctoral thesis provides a comprehensive overview of the literature on the PV morphology in pigs and horses, atrial fibrillation in men and echocardiography in horses. This chapter focuses, specifically, on the limited data about the number and debouchment patterns of the porcine and equine PVs. In addition, the pathophysiology, prevalence and some treatment strategies of AF in men were described more extensively. Finally, Chapter 1 summarizes the limited echocardiographic data of the equine PV and points out the value of pigs as a cardiovascular model.

Chapter 2 describes the objectives of this thesis. The first aim of this manuscript was to study the detailed morphology of the porcine and equine PVs. Those veins were also histologically investigated to demonstrate the presence of ganglia and telocytes and to build-up a 3D model. A second objective was to develop a new ablation treatment for AF patients which was tested and adapted in pigs as an animal model. The final aim of this thesis was to visualize immunohistochemically acute ablation lesions.

In Chapter 3, the morphology and variability of the PVs of the pig is demonstrated, which acted as a route map during the catheterization procedures (Chapters 6 and 8). Two ostia are described through which the PVs drain oxygenated blood into the left atrium. Each lung lobe has its corresponding PV of which the blood of the right cranial and middle lung lobe is drained through a common ostium. Histologically, myocardial tissue fanning out from the myocardium into the PV wall,

265

Summary indicated as myocardial sleeve, was assessed. The myocardial sleeve length can measure over 1 centimetre at its most distal point.

In Chapter 4, the porcine PVs are further histologically investigated. At the atrio-pulmonary junction, ganglia and telocytes were found near myocardial sleeve tissue inside the PV wall. In addition, intra- and extramural ganglia containing telocytes were demonstrated. Those telocytes constitute a network near neuron cell bodies.

Chapters 5, 6 and 8 describe the process in which a new device is developed to treat AF in men. In chapter 5, an implant prototype was placed inside an ovine PVs antrum via open heart surgery. The heat was generated wirelessly outside the body and subsequently transferred through wires towards the implanted device to apply an ablation lesion at the PV wall. Afterwards, the acute ablation lesions were histologically evaluated. Chapter 6 demonstrates the implantation of self-expanding nitinol stents inside an antrum of the PVs via catheterization of pigs. The stents were placed through a transfemoral or transatrial approach and the stents remained in place during a 3 months follow-up phase. After this period in which no abnormal events were noticed, all samples were macroscopically and histologically evaluated. This revealed that all stents were almost completely overgrown without excessive neo-intima and thrombus formation. Chapter 8 finalizes the development process with the implantation of stents via transatrial access inside the PVs in pigs. Ablation of these stents was performed by use of a contactless energy transfer from an extracorporal coil to a secondary ring embedded in the proximal part of the implant. Clear ablation lesions were seen acutely and 3 months after the ablation procedure with mild neointimal proliferation. Bidirectional conduction block was electrophysiologically demonstrated which proved the efficacy of this ablation strategy.

Chapter 7 visualizes acute ablation lesions in the porcine cranial vena cava and PVs by immunohistochemistry. The MYBPC3 antibody indicated indirectly the spots with denatured myocardial tissue, caused by an ablation procedure. In certain samples, no ablation lesions were noticed due to the application of too low energy levels or induction of extreme high temperatures. Interestingly, myocardial tissue layers separated by fat tissue showed no ablation lesions as fat tissue acts as an insulator for the penetrating heat. The immunohistochemical technique offers the possibility to objectively measure acute ablation lesions.

In Chapter 9, the venous drainage of both equine lungs and the position of the different PV ostia are investigated to determine whether those ostia can be identified on cardiac ultrasound. Mostly, four ostia are present of which three out of the four ostia can be observed echocardiographically. Little variability was observed in the ostium draining the most caudal parts of both lungs. However, much more variation was seen in the drainage pattern of the other PVs. The ostium draining the cranial and

266

Summary middle part of the right lung was found to be an ideal landmark for ultrasound orientation. It was concluded that ostia II, III and IV can be identified on ultrasound.

In Chapter 10, the myocardial sleeves of the porcine and equine PVs are histologically studied and reconstructed three-dimensionally. In addition, neuron cell bodies at the veno-atrial junction were light microscopically visualized. A histological overview of the organization of nervous, myocardial and adipous tissue at the level of the PVs was provided. Finally, the presence of telocytes inside the equine PVs was visualized by use of transmission electron microscopy.

Chapter 11 explains an ultrasonographic approach in horses to identify cardiac structures and vessels of the dorsal heart region. Several cardiac ultrasound views were defined on which specific anatomical atrial landmarks such as vena cavae, pulmonary trunk and arteries, intervenous tubercle, terminal crest and oval fossa were identified. Moreover, echocardiographic procedures were defined to visualize the brachiocephalic trunk, PVs and their ostia. This information provides essential anatomical landmarks useful during invasive cardiac procedures. In the future, this information may allow to develop advanced diagnostics or treatment options of specific cardiac diseases.

Finally, the general discussion and conclusions follow in Chapter 12, providing additional information regarding the topics discussed in previous chapters to tackle the remaining questions. This chapter highlights the advantages of our developed ablation device but also discusses issues that will need to be improved. Possible prototype adaptations, such as a biodegradable stent supporting structure, were suggested which may improve the outcome of an implantation procedure. Additionally, several stent ablation complications such as pulmonary vein stenosis, coagulum or thromboembolism formation were discussed. Ganglia and telocytes may play a role in the development of AF. Therefore extrapulmonary tissue ablation, next to the current ablation techniques, could be beneficial. This chapter ends with an overview of the various treatment methods for AF in humans, with a reference to the interesting opportunities of a hybrid ablation technique.

267

268

Samenvatting

269

270

Samenvatting

Atriale fibrillatie is de belangrijkste aritmie bij de mens, maar toch is het ontstaansmechanisme van deze pathologie nog steeds niet volledig opgehelderd. Verschillende behandelingsmethodes werden tijdens het laatste decennium ontwikkeld waarvan ablatie van de pulmonale venen de hoeksteen is in de behandeling van atriale fibrillatie bij de mens. Nochtans blijft, wegens de sterk variërende resultaten, de nood naar meer onderzoek bestaan om de succescijfers van de huidige ablatieprocedures te kunnen verhogen. Om een nieuwe behandelingsmethode voor atriale fibrillatie te kunnen ontwikkelen, is een diermodel nodig. Het varken werd gekozen als model aangezien het varkenshart vergelijkbaar is met het hart van de mens. Bij de start van deze thesis was er weinig informatie beschikbaar omtrent de histologie, morfologie en variabiliteit van de pulmonale venen van het varken. Deze informatie is onmisbaar om een stent, waarmee geableerd wordt, te kunnen ontwikkelen en implanteren in de pulmonale venen van het varken.

Paarden zijn sterker gepredisponeerd om atriale fibrillatie te ontwikkelen dan het varken. Atriale fibrillatie kan vaak succesvol behandeld worden, maar recidief komt nog vaak voor. Om behandeling van atriale fibrillatie te verbeteren is een betere visualisatie van het hart nodig waarvoor echocardiografie momenteel de belangrijkste techniek is. Bij de start van dit onderzoek ontbrak echter gedetailleerde echocardiografische informatie over de pulmonale venen van het paard.

De algemene inleiding (Hoofdstuk 1) van deze doctoraatsthesis geeft een overzicht van de wetenschappelijke literatuur omtrent de morfologie van de pulmonale venen van het varken en het paard, atriale fibrillatie bij de mens en echocardiografie bij het paard. Dit luik focust zich specifiek op de informatie die beschikbaar is omtrent het aantal pulmonale venen van het varken en het paard en de wijze waarop deze uitmonden in het linker atrium. Daarnaast werden de pathofysiologie, prevalentie en enkele behandelingsmethodes van atriale fibrillatie bij de mens uitgebreid beschreven. Tot slot vat hoofdstuk 1 de beperkte echocardiografische gegevens van de pulmonale venen van het paard samen en wordt het belang van het varken als cardiovasculair proefdiermodel benadrukt.

De doelstellingen van de experimenten binnen dit doctoraat werden opgelijst in Hoofdstuk 2. Het eerste doel van deze thesis was gedetailleerd inzicht verwerven in de morfologie van de pulmonale venen van het varken en het paard. Deze venen werden ook histologisch onderzocht om de aanwezigheid van ganglia en telocyten aan te tonen en een 3D model op te bouwen. Een tweede doel van dit manuscript was de ontwikkeling van een nieuwe ablatietechniek, uitgetest en aangepast in een varkensmodel, ter behandeling van humane patiënten met atriale fibrillatie. Bijkomend werd getracht om acute ablatie letsels immunohistochemisch te visualiseren.

271

Samenvatting

In Hoofdstuk 3 wordt de morfologie en variabiliteit van de pulmonale venen bij het varken onderzocht. Deze informatie kon gebruikt worden als een leidraad gedurende de katheterisatie procedures, beschreven in Hoofdstuk 6 en 8. Bij het varken werden twee ostia beschreven waarlangs de pulmonale venen zuurstofrijk bloed draineren in het linker atrium. Iedere longkwab heeft zijn eigen pulmonale vene. Het bloed van de rechter craniale en middelste longkwab wordt gedraineerd via een gemeenschappelijk ostium. De pulmonale venen werden ook histologisch onderzocht waarbij hartspierweefsel, dat vanuit het linker atrium uitwaaiert in de pulmonale venenwand, beschreven werd. Dit hartspierweefsel wordt als “myocardial sleeve” aangeduid. De lengte van deze strook hartspierweefsel kan, gemeten op het meest distale punt, meer dan 1 centimeter bedragen.

In Hoofdstuk 4 worden de pulmonale venen verder histologisch onderzocht. De atrio-pulmonale vene overgang werd in detail bekeken waarbij ganglia en telocyten aangetroffen werden in de nabijheid van “myocardial sleeve” weefsel. Bijkomend beschrijft dit hoofdstuk de aanwezigheid van intra- en extramurale ganglia, die telocyten bevatten. Deze telocyten vormen een netwerk in de buurt van het cellichaam van neuronen.

Hoofdstuk 5, 6 en 8 omvat de studie waarin een nieuw ablatiesysteem ontwikkeld werd met het uiteindelijke doel atriale fibrillatie bij de mens te behandelen. Hoofdstuk 5 beschrijft de procedure waarin een prototype geïmplanteerd werd in de pulmonale venen van schapen door middel van openhartchirurgie. De ablatie werd uitgevoerd door middel van hitte. Deze kon, buiten het dier, draadloos opgewekt en vervolgens via draden overgebracht worden naar de geïmplanteerde stent. Nadien werden de acute ablatieletsels histologisch onderzocht. Hoofdstuk 6 demonstreert de implantatie van zelf-expanderende nitinol stents in de antra van de pulmonale venen van varkens. De implantatie gebeurde via katheterisatie van de varkens. De stents werden geplaatst via een transfemorale of transatriale benadering waarna deze stents ter plaatse bleven gedurende 3 maanden. Na deze periode, waarin er zich geen abnormale situaties voordeden, werden alle stalen van de pulmonale venen zowel macroscopisch als histologisch beoordeeld. Dit toonde aan dat alle stents vrijwel volledig overgroeid waren zonder de vorming van overmatig veel neo-intima weefsel en thrombus ontwikkeling. Hoofdstuk 8 beschrijft de laatste stap van het ontwikkelingsproces waarin stents geïmplanteerd worden in de pulmonale venen van varkens via een transatriale benadering. De ablatie werd uitgevoerd door middel van een coil aangebracht rondom de thorax van het varken en een tweede coil ingebed in de geïmplanteerde stent. Tussen beide coils kon een energieveld opgewerkt worden dat voldoende hitte produceerde om de ablatieletsels aan te brengen. Duidelijke ablatieletsels werden opgemerkt, zowel in het acute stadium als 3 maanden na de ablatieprocedure. Hierbij werd enkel een milde neo-intima proliferatie geobserveerd. Door middel van

272

Samenvatting elektrofysiologisch onderzoek kon een bidirectionele geleidingsblokkade aangetoond worden wat de efficaciteit van de ablatieprocedure bewijst.

In Hoofdstuk 7 wordt de visualisatie van acute ablatieletsels in de craniale vena cava en pulmonale venen beschreven. Deze letsels werden gevisualiseerd door middel van een immunohistochemische kleuring op basis van het MYBPC3-antilichaam. Dit antilichaam zorgde ervoor dat de plaatsen met gedenatureerd myocardweefsel, geïnduceerd door een ablatieprocedure, indirect konden gelokaliseerd worden. In bepaalde stalen werden geen ablatieletsels opgemerkt aangezien te lage energiewaarden of extreem hoge temperaturen gebruikt werden. Ter hoogte van myocardweefsel dat geïsoleerd was door vetweefsel, werden geen ablatieletsels teruggevonden. Vetweefsel werkt namelijk als een goede isolator voor de penetrerende hitte. Deze immunohistochemische techniek biedt de mogelijkheid om objectief acute ablatieletsels te meten.

In Hoofdstuk 9 wordt de veneuze drainage van beide longen van het paard onderzocht. Daarnaast werden de verschillende ostia bestudeerd zodat bepaald kon worden of deze ostia geïdentificeerd kunnen worden door middel van echocardiografie. Weinig variatie kon waargenomen worden betreffende het ostium waarlangs de meest caudale delen van beide longen gedraineerd worden. Daarentegen werd er veel meer variatie opgemerkt in het drainage patroon van de pulmonale venen die uitmonden in het linker atrium via de overige ostia. Er werd geconcludeerd dat zowel ostia II, III en IV geïdentificeerd kunnen worden door middel van echocardiografie.

In Hoofdstuk 10 wordt de myocardial sleeve van de pulmonale venen van het varken en het paard histologisch bestudeerd en driedimensioneel gereconstrueerd. Bovendien konden neuronale cellichamen ter hoogte van de atrio-veneuze overgang lichtmicroscopisch gevisualiseerd worden. Een histologisch overzicht van de organisatie van zenuw, hartspier en vetweefsel ter hoogte van de pulmonale venen werd voorzien. Finaal werd de aanwezigheid van telocyten in de pulmonale venen van het paard aangetoond door middel van een transmissie-elektronen microscoop.

De echocardiografische benadering om alle hartstructuren en bloedvaten van de dorsale hartregio te identificeren, wordt beschreven in Hoofdstuk 11. Diverse beelden werden gedefinieerd waarop specifieke anatomische herkenningspunten zoals de holle aders, longslagaders, tuberculum intervenosum, crista terminalis en fossa ovalis geïdentificeerd werden. Bovendien werden ook echocardiografische procedures beschreven om de truncus brachiocephalicus, de pulmonale venen en hun ostia te visualiseren. Deze informatie kan als een handleiding gebruikt worden tijdens het katheteriseren en genereren van een 3D-map van diverse hartstructuren en bloedvaten, zoals de pulmonale venen. In de toekomst zou deze informatie ook een belangrijke bijdrage kunnen leveren aan de ontwikkeling van een nieuwe behandeling van atriale fibrillatie bij het paard.

273

Samenvatting

Ten slotte worden de algemene discussie en conclusies weergeven in Hoofdstuk 12, dat bijkomende informatie voorziet betreffende de onderwerpen, besproken in voorgaande hoofdstukken, om zo de overblijvende vragen te beantwoorden. Nochtans worden hierdoor ook nieuwe vragen en speculaties opgesteld. Dit hoofdstuk benadrukte de voordelen van het ontwikkelde ablatiesysteem. Echter, verschillende nadelen werden ook besproken. Zo konden er enkele mogelijke aanpassingen aan het ablatieprototype, zoals een biodegradeerbaar stentonderdeel, gesuggereerd worden om de efficaciteit van de ablatieprocedures te kunnen verhogen. Bovendien werden verschillende stent ablatiecomplicaties zoals pulmonale vaatvernauwing, stolsel of trombo-embolie vorming toegelicht.

De rol van ganglia en telocyten in de ontwikkeling van atriale fibrillatie werd besproken. Vervolgens werd er gewezen op de mogelijke interessante toegevoegde waarde die bekomen kan worden door het ableren van extra-pulmonale weefsels, als aanvulling op de huidige ablatietechnieken. Dit hoofdstuk eindigt met een overzicht van de verschillende behandelingsmethodes van atriale fibrillatie bij de mens. Hierbij worden de interessante opportuniteiten van een hybride ablatiesysteem benadrukt.

274

Curriculum vitae

275

276

Curriculum vitae

Tim Vandecasteele werd geboren op 7 september 1983 te Menen, België. Na het beëindigen van het secundair onderwijs aan het Mater Amabilis instituut te Wervik, richting Wetenschappen-Wiskunde, startte hij in 2001 aan de studie Diergeneeskunde aan de Universiteit Gent waar het diploma van Master in de Diergeneeskunde in 2008 werd behaald.

Onmiddellijk na zijn studies, startte hij in 2008 als zelfstandig varkensdierenarts bij de dierenartsenassociatie CVBA Vartos te Roeselare. Deze functie werd uitgeoefend tot 2012 waarna hij een voltijds assistentenmandaat opnam aan de vakgroep Morfologie van de Gentse faculteit Diergeneeskunde. Naast zijn onderzoek was hij betrokken bij de practica anatomie, histologie en dierkunde.

Tim is promotor van 10 masterproeven en fungeerde ook als leescommissaris. Hij is auteur van 9 en mede-auteur van 10 wetenschappelijke publicaties in internationale tijdschriften. Hij trad op als reviewer voor verschillende wetenschappelijke tijdschriften in het vakgebied anatomie en histologie. Hij nam actief deel aan diverse nationale en internationale congressen. In 2012 won hij de “FMV ULg award” voor beste poster en in 2014 behaalde hij de “ Simic-Grau EAVA award” voor beste orale presentatie.

277

278

Bibliography

279

280

Bibliography

Publications

Vandecasteele T., Cornillie P., Van Steenkiste G., Vandevelde K., Gielen I., Vanderperren K. and van Loon G. (2018). Echocardiographic identification of atrial-related structures and vessels validated by CT images of equine hearts. Equine Vet J. DOI: 10.1111/evj.12969.

Vandecasteele T., Van den Broeck W., Tay H., Couck L., van Loon G. and Cornillie P. (2018). 3D reconstruction of the porcine and equine pulmonary veins, supplemented with the identification of telocytes in the horse. Anat Histol Embryol. 47:145-152.

Boussy T., Vandecasteele T., Vera L., Schauvliege S., Philpott M., Clement E., van Loon G., Willenz U., Granada J.F., Stone G.W., Reddy V.Y. and Van Langenhove G. (2018). Isolation of Pulmonary Veins using a Thermo Reactive Implantable Device with External Energy Transfer: Evaluation in a Porcine Model. Pacing Clin Electrophysiol. DOI:10.1111/pace.13345.

Vandecasteele T., Schauvliege S., Philpott M., Clement E., van Loon G., Vera L., Boussy T., Van Bergen T., Van den Broeck W., Cornillie P. and Van Langenhove G. (2018). A Preliminary Study of Pulmonary Vein Implant Applicability and Safety as a Potential Ablation Platform in a Follow-up Study in Pigs. Pacing Clin Electrophysiol. 41:167-171.

Peeters G., Debbaut C., Friebel A., Cornillie P., De Vos W., Favere K., Vander Elst I., Vandecasteele T., Johann T., Van Hoorebeke L. UGent, Monbaliu D., Drasdo D., Hoehme S., Laleman W. and Segers P. (2017). Quantitative analysis of hepatic macro- and microvascular alterations during cirrhogenesis in the rat. J Anat. DOI:10.1111/joa.12760.

Tay H., Vandecasteele T. and Van den Broeck W. (2017). Identification of telocytes in the porcine heart. Anat Histol Embryol. DOI:10.1111/ahe.12296.

Logothetidou A., De Spiegelaere W., Van den Broeck W., Vandecasteele T., Couck L., Simoens P. and Cornillie P. (2017). Stereological and immunogold studies on TIE1 and TIE2 localization in glomeruli indicate angiopoietin signaling in podocytes. Micron. 97:6-10.

Logothetidou A., Vandecasteele T., Van Mulken E., Vandevelde K. and Cornillie P. (2017). Intussusceptive angiogenesis and expression of Tie receptors during porcine metanephric kidney development. Histol Histopathol. 32:817-824.

281

Bibliography

Gasthuys E., Schauvliege S., van Bergen T., Millecam J., Cerasoli I., Martens A., Gasthuys F., Vandecasteele T., Cornillie P., Van den Broeck W., Boyen F., Croubels S. and Devreese M. (2017). Repetitive urine and blood sampling in neonatal and weaned piglets for pharmacokinetic and pharmacodynamic modelling in drug discovery : a pilot study. Lab Anim. DOI:10.1177/0023677217692372.

Vandecasteele T., Schauvliege S., Boussy T., Philpott M, Clement E., Vera L., Cornillie P., De Spiegelaere W., Van Langenhove G., van Loon G. and Van den Broeck W. (2017). Immunohistochemical identification of stent-based ablation lesions in the superior vena cava and pulmonary veins. J Histol Histopathol. 4:14.

Peeters G., Debbaut C., Laleman W., Monbaliu D., Vander Elst I., Detrez J.R., Vandecasteele T., De Schryver T., Van Hoorebeke L. UGent, Favere K., Verbeke J., Segers P. UGent, Cornillie P. and De Vos W. (2017). A multilevel framework to reconstruct anatomical 3D models of the hepatic vasculature in rat livers. J Anat. 230:471-483.

Vandecasteele T., Philpott M., Boussy T., van Loon G., Cornillie P. and Van Langenhove G. (2016). A preclinical study of an implanted device in the pulmonary veins, intended for the treatment of atrial fibrillation in an ovine model. Pacing Clin Electrophysiol. 39:822-829.

Saey V., Vandecasteele T., van Loon G., Cornillie P., Ploeg M., Delesalle C, Gröne A., Gielen I., Ducatelle R. and Chiers K. (2016). Friesian horses as a possible model for human acquired aortopulmonary fistulation. BMC Res Notes. 9:405-410.

Gasthuys E., Vandecasteele T., De Bruyne P., Vande Walle J., De Backer P., Cornillie P. UGent, Devreese M. and Croubels S. (2016). The potential use of piglets as human pediatric surrogate for preclinical pharmacokinetic and pharmacodynamic drug testing. Curr Pharm Des. 22:4069-4085.

Saey V., Vandecasteele T., Cornillie P., van Loon G., Ducatelle R. and Chiers K. (2016). Three Dimensional Replication of Aortopulmonary Fistulas in Friesian Horses Using Vascular Casting. J Comp Pathol. 154:60.

282

Bibliography

Van Loon G., Vandecasteele T., Vandevelde K., Decloedt A., De Clercq D. and Cornillie P. (2015). Ultrasonographic Identification of the Pulmonary Veins in Adult Horses. Eq Vet J. 47:26-27.

Conference contributions

Vandecasteele T., Cornillie P., Van Steenkiste G., Gielen I., Vanderperren K. and van Loon G. (2018). Echocardiography of the atrium, based on a 3D model of the equine heart. Oral abstract presentation, Proceedings 1st Vet2030 Symposium, Ghent, Belgium.

Van Loon G., Vandecasteele T., Van Steenkiste G., Gielen I., Vanderperren K. and Cornillie P. (2017). Ultrasonographic identification of atrial-related structures, validated by 3D CT-segmented imaging of equine casted hearts. Proceedings 10th ECEIM Congress, Budapest, Hungary.

Tay H., Vandecasteele T. and Van den Broeck W. (2017). Identification of telocytes in the porcine heart. Proceedings 9th Meeting of Young Generation of Veterinary Anatomists, Brno, Czech Republic.

Vandecasteele T., van Loon G., Van Langenhove G., Boussy T., Philpott M., Clement E., Willentz U., Walser U. and Cornillie P. (2016). Ablation of the pulmonary veins with a new implantable device technique in a pig model. Poster presentation, Proceedings 31st EAVA Congress, Vienna, Austria, Anat Histol Embryol. 45:87-88.

Vandevelde K., Van den Broeck W., Wang N., Vandecasteele T. and van Loon G. (2016). Connexines in the horse heart. Proceedings 31st EAVA Congress, Vienna, Austria, Anat Histol Embryol. 45:88-89.

Peeters G., Debbaut C., Laleman W., Monbaliu D., Detrez J.R., Vandecasteele T., De Schryver T., Favere K., Verbeke J., Cornillie P. , De Vos W. and Segers P. (2016). A multilevel framework to reconstruct anatomical 3D models of the hepatic vasculature in rat livers. Proceedings 22nd Congress of ESB, Lyon, France.

Peeters G., Debbaut C., De Vos W., Cornillie P., Detrez J., Vandecasteele T., De Schryver T., Monbaliu D., Laleman W. and Segers P. (2016). A multiscale framework for studying vascular morphology alterations during liver cirrhogenesis: a feasibility study. Proceedings 43rd Annual Congress of the European Society for Artificial Organs, Warsaw, Poland, Int J Artif Organs. 39.

283

Bibliography

Saey V., Vandecasteele T., Cornillie P., van Loon G., Ducatelle R. and Chiers K. (2015). Three dimensional replication of aortopulmonary fistulas in Friesian horses using vascular casting. Proceedings 33rd ESVP/ 26th ECVP Annual meeting, Helsinki, Finland.

Van Loon G., Vandecasteele T., Vandevelde K., Decloedt A., De Clercq D. and Cornillie P. (2015). Ultrasonographic identification of the pulmonary veins in adult horses. Proceedings 54th BEVA Congress, Liverpool, UK, p.263.

Vandecasteele T., Vandevelde K., Simoens P., van Loon G. and Cornillie P. (2014). Anatomical description and ultrasonographic identification of the venous pulmonary circulation in horses. Oral abstract presentation, Proceedings 30th EAVA Congress, Cluj-Napoca, Roemenia, Anat Histol Embryol. 43:93-94.

Vandecasteele T., Vandevelde K., Doom M., Van Mulken E. and Cornillie P. (2013). Researching atrial fibrillation in animal models or veterinary patients: major role for the anatomist. Oral abstract presentation, Proceedings 7th Meeting of Young Generation of Veterinary Anatomists, Leipzig, Germany, p.44.

Vandecasteele T., Doom M., Van Mulken E., Schwagten B., Van Langenhove G., De Beule M., Mortier P. UGent, Segers P. and Cornillie P. (2012). Renewed interest in porcine and horse heart and pulmonary vein anatomy in an experimental model for atrial fibrillation treatment. Poster presentation, Proceedings 2nd scientific meeting of the Faculty of Veterinary Medicine, Liège, Belgium.

284

Dankwoord

285

286

Dankwoord

Terugkeren naar de faculteit was eigenlijk niet mijn bedoeling... Het was meer een samenloop van omstandigheden. Terugkijkend op de voorbije 6 jaar, kan ik enkel maar concluderen dat deze periode voorbij gevlogen is en ik altijd met heel veel passie aan mijn doctoraat gewerkt heb. Het voelde zelfs nooit echt als werken aan, wat enkel maar een goed teken kan zijn. Ik gebruik hier de woorden “mijn doctoraat” maar dit werk kon natuurlijk nooit tot stand gekomen zijn zonder de hulp en medewerking van een hele groep andere mensen die ik hierbij wil bedanken.

Startend met mijn promotoren, wil ik als eerste Prof. dr. Pieter Cornillie bedanken. Samenwerken met u was altijd aangenaam. Ik kon zomaar uw kantoor binnenlopen om mijn grote en kleine ideeën of vragen voor te leggen. Vooral de vrijheid die u me gaf en het vertrouwen die u in mijn stelde om zelf mijn pad uit te stippelen apprecieër ik enorm. De zinsnede “ik kan u een hoop bomen geven maar het bos door de bomen moet je zelf kunnen zien” zal ik nooit vergeten en vat goed samen hoe een doctoraat uiteindelijk dient aangepakt te worden.

Prof. dr. Wim Van den Broeck, bij u kon ik terecht als ik histologische vragen had. Maar vooral de manier waarop u wetenschappelijk onderzoek benadert, blijft me bij. Met een positieve ingesteldheid en door stap voor stap te werk te gaan was het toch mogelijk om uiteindelijk resultaten te halen uit zaken die in eerste instantie onmogelijk leken te zijn.

Prof. dr. Gunther van Loon, ik moet u zeker bedanken voor de vele uren die u uittrok om samen data te overlopen, te discussiëren en te redeneren en vooral samen in de stallen te staan op zoek naar goeie echobeelden van pulmonale venen. Het blijft me nog steeds zelf verbazen dat we zoveel jaren geleden zaten te kijken naar een hele hoop structuren waar we niks over wisten en dat u nu zo vlot in staat bent om alles te benoemen. Vooral uw passie voor de cardiologie werkte heel aanstekelijk voor mij, uw gave om dergelijke complexe materie gestructureerd en overzichtelijk te vertellen was altijd super verhelderend. Dankzij u zal het hart altijd mijn favoriete onderzoeksonderwerp blijven.

Ook wil ik Dr. Glenn Van Langenhove bedanken. Eenvoudig gezegd zou een groot stuk van deze thesis gewoon niet mogelijk geweest zijn zonder u, waarvoor dank. Uw out-of-the-box thinking werkte verfrissend en inspirerend, altijd klaar met een antwoord of nieuw idee zorgde dit voor vertrouwen en uw no-nonsense mentaliteit creërde een motiverende atmosfeer. Tijdens de vele uren waarin u vertelde over uw avonturen in de humane cardiologie zat ik altijd gebiologeerd te luisteren, wat vaak leidde tot een nieuwe brainstormsessie. Bovendien apprecieer ik het vertrouwen dat u in mij stelde.

287

Dankwoord

Hier wil ik ook onmiddelijk Prof. dr. Simoens bedanken. U was zovele jaren mijn vakgroepvoorzitter maar voor vele van uw werknemers waaronder mezelf meer een vaderfiguur. Altijd klaar om te luisteren en te helpen waar mogelijk. Uw passie voor de anatomie was continu aanwezig en werkte aanstekelijk, een betere ambassadeur is er niet te vinden. Uw oog voor detail en correctheid bleef me verbazen maar verhoogde ook de kwaliteit van het werk van uw onderzoekers. Ik hoop dat u nu volop kunt genieten van uw vrije tijd. Prof. dr. De Spiegelaere, we hebben in het kader van mijn onderzoek minder samengewerkt maar in het voorbije jaar was het interessant voor mij om de mening te horen van iemand die het geheel van op enige afstand bekeek.

Bovendien wil ik ook alle leden van de examencommisie: Prof. dr. Jeroen Dewulf, Prof. dr. Pascale Smets, Prof. dr. Claudia Wolschrijn, Prof. dr. Annelies Decloedt en dr. Veronique Saey bedanken voor hun tijd, moeite en constructieve bijdrage aan deze doctoraatsthesis.

Dan ons kantoortje dat door de vele jaren bezet werd door verschillende mensen. Els, jij was de eerste en ook enige die in ons bureau zat toen ik aankwam. Bedankt voor de gesprekjes en eerste introductie in de vakgroep. Daarnaast wil ik zeker Tassie en Hanna vermelden. Tassie, baas van ons kantoor en stem tijdens stille momenten ;-), heel erg bedankt voor de unieke hilarische momenten waar het vaak net niet “te laat” was. Dankzij jou kennen we een aantal uiterst nuttige Griekse zinnen... uhum... Hanna, we blijven sowieso telocyten pioniers maar misschien toch goed dat je nu een andere weg bent ingeslagen. Maar ik apprecieer dat je er altijd was om te discussiëren over onze “god-cells”. Veel succes met het onderzoek in de komende jaren.

En ja, Kimberley, waar moet ik beginnen, ook een belangrijk deel geweest van onze bureau gedurende zovele jaren maar uiteindelijk ben je natuurlijk zoveel meer voor mij. Het klikte heel vlug tussen ons tijdens het werk waar ik je sowieso ook wil voor bedanken. Want naast het feit dat we nu toch al enkele jaren lief en leed delen besef ik ook heel goed dat jij me heel erg geholpen hebt tijdens dit onderzoek. De vele uren die we samen doorbrachten in de snijzaal, aan het casten en het dissecteren, waarvan ik toen nog niet wist dat het je ding niet was. Maar naast die vele uren, is het vooral je inzicht, vlug redeneren, kortom je intellect dat me vooruit hielp en vertrouwen gaf. Dit is trouwens een kenmerk dat ik nog elke dag mag bewonderen. Net zoals je gave om te spelen met woorden waardoor je meermaals, niet alleen mij maar vele anderen met mij, verbaasde als je weer vooraan een zaal, groot of klein, complexe materie, in zoals je zelf zegt jip en janneke taal, overbracht naar studenten of collega’s. Getuigen daarvan zijn de talloze studenten die je ieder jaar opnieuw kwamen bedanken omdat ze jou als assistent of zelfs jou als examinator hadden. Chapeau!

Er zijn natuurlijk nog veel meer mensen op de vakgroep Morfologie die me geholpen hebben bij het realiseren van dit doctoraatswerk. Liesbeth, Kleine, amai wat een leven jij bracht toen je als

288

Dankwoord youngster hier toekwam. Sterk, hoe je je de voorbije jaren dergelijke moeilijke technieken eigen hebt gemaakt. De toewijding die je nu dagelijks toont en het samen blijven zoeken naar oplossingen maken van jou een droomlaborant voor elke doctoraatsstudent. Maar ik zal vooral onze dude- gesprekjes missen ;-), zo nuchter, rechtuit in je verwoording en met de voetjes-op-de-grond, het was telkens een verademing. Ik kan alleen maar zeggen, Kleine, doe zo voort en laat maar zien wat je kunt ;-).

Ik wil ook Bart DP, Jürgen, Patrick, Evelyne en Lobke, de vaste waarden van onze vakgroep, bedanken. Elk in jullie eigen domein waren jullie altijd bereid om te helpen, mee te denken of iets te verduidelijken. Patrick, merci om al het materiaal te voorzien en handige tips te geven als er iets praktisch moest gedaan worden. Lobke, zonder jou zou ik nooit de 3D reconstructies kunnen maken hebben en bedankt voor de vele immuno’s. Bart, dankzij jou zag ieder filmpje er piekfijn uit en zijn mijn pc-skills verder aangescherpt. Jürgen, je nuchtigere kijk op experimenten werkte vaak verhelderend. Evelyne, bedankt voor alle dagelijkse praktische zaken die je in orde bracht zodat wij met ons hoofd tussen de experimenten en de boeken konden zitten.

Een hele resem van doctoraatsstudenten, Annelies, Marlien, Maarten, Charis, Marieke, Evelien, Khan, Martine, Marjan, Karlijn hebben hun periode op de vakgroep met mij gedeeld, met de ene had ik al meer contact dan met de andere maar ik wil ook jullie allemaal bedanken voor de leuke gesprekjes.

Tijdens de voorbije jaren werkte ik met verschillende praktijkassistenten samen om de vele practica in goede banen te leiden. Velen van jullie hadden veel meer ervaring dan ik en bijgevolg heb ik van elk van jullie bijgeleerd. Winny, Mariella, Bart VDV, Sofie, Martine, Christophe, Isabelle, Monique, Jan, Ymke, Eva, Jenske en waarschijnlijk vergeet ik nog mensen, bedankt! Een speciale vermelding voor Bart, ik kon alleen maar jaloers toekijken hoe goed jij een demo viscera kan geven.

Sandra, ik wil jou hier ook nog vermelden. Ik wil jou zeker bedanken voor de amusante gesprekjes en de vaak oprechte en bemoedigende woorden.

Behalve de vakgroep Morfologie, vergeet ik natuurlijk niet het ganse team van de vakgroepen Inwendige Ziekten, Heelkunde en Medische Beeldvorming. Ik wil jullie ook allemaal bedanken om mee te helpen bij alle proeven op de varkentjes en om die proeven überhaupt mogelijk te maken. Een speciaal dankwoordje toch voor Lisse en Glenn. Merci voor alle uren die jullie meegeholpen hebben en voor het vaak achterblijven met mij tot in de late uurtjes, jullie zijn een top-duo.

Ik wil sowieso ook het ganse team van Fulgur bedanken. Eli en Matthew, vaak begreep ik niet waarover jullie bezig waren, maar toch had ik telkens bewondering voor de creativiteit waarmee

289

Dankwoord jullie alle obstakels aanpakten en een oplossing bedachten. Hierbij wil ik ook nog mijn naamgenoot Tim bedanken om toch een deel van uw kostbare tijd te investeren in ons project.

Tot slot wil ik mijn familie bedanken, met in het bijzonder mijn ouders en mijn zus. Zonder jullie zou ik hier nooit geraakt zijn en zou ik vandaag dit werk niet kunnen voorleggen. Bedankt voor alles!

290