Cardiac Pathology: A Guide to Current Practice [2nd ed. 2019] 978-3-030-24559-7, 978-3-030-24560-3

Table of contents :
Front Matter ....Pages i-ix
The Normal Adult Heart and Methods of Investigation (S. Kim Suvarna)....Pages 1-23
Cardiac Electrophysiology (Paul D. Morris, Jonathan Sahu)....Pages 25-48
Cardiac Imaging (Abdallah Al-Mohammad, Peter W. G. Brown)....Pages 49-74
Current Therapeutics for Cardiac Disease (Abdallah Al-Mohammad)....Pages 75-91
The Heart at Autopsy, Including Radiological Autopsy of the Heart (S. Kim Suvarna)....Pages 93-126
Embryology of the Heart (Michael T. Ashworth)....Pages 127-135
Ischaemic Heart Disease (Katarzyna Michaud)....Pages 137-151
Myocarditis (Martin J. Goddard)....Pages 153-165
Valvular Heart Disease (Stephen D. Preston)....Pages 167-184
Transplant Pathology (Desley A. H. Neil)....Pages 185-204
Cardiomyopathies (Clare R. Bunning, S. Kim Suvarna)....Pages 205-225
Cardiac Tumours (Doris M. Rassl)....Pages 227-254
Congenital Heart Disease (Michael T. Ashworth)....Pages 255-276
Sudden Cardiac Death (S. Kim Suvarna)....Pages 277-311
Back Matter ....Pages 313-323

Citation preview

S. Kim Suvarna  Editor

Cardiac Pathology A Guide to Current Practice Second Edition

123

Cardiac Pathology

S. Kim Suvarna Editor

Cardiac Pathology A Guide to Current Practice Second Edition

Editor S. Kim Suvarna Sheffield Teaching Hospitals Royal Hallamshire Hospital Sheffield Teaching Hospitals Sheffield South Yorkshire UK

ISBN 978-3-030-24559-7    ISBN 978-3-030-24560-3 (eBook) https://doi.org/10.1007/978-3-030-24560-3 © Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

To the family of Lucy Atherton, and all similar families, who have allowed the use of autopsy images for the education of all those practicing cardiac pathology.

Preface to the Second Edition

It is some 8 years since the first edition was published. Since then, I have been grateful to receive helpful comments and suggestions as to how to improve the publication. It is maintained that good cardiac pathology knowledge is always to be balanced alongside awareness of embryological and developmental matters, current therapeutics, modern imaging techniques, and a reasonable understanding of cardiac electrophysiology. These subjects have kept their place in the publication, along with the other standard cardiac pathology chapters. The rising use of postmortem radiology prompts its appearance within the autopsy chapter. As always, thanks go to the chapter authors for their unfaltering dedication to the project. They have built upon the previous edition and the work of other practitioners. Likewise, I am indebted to my colleagues who have shared their interesting cases, thereby adding to the photomicrographs and macroscopic images in the publication. As previously, the support and understanding from the publishers have been valued, and my grateful thanks are also given to Seonaid Ashby for the secretarial support. Lastly, my ongoing thanks go to Grace, Miranda, and Elara who have given me the time to work on this project. Sheffield, UK October 2019

S. Kim Suvarna

vii

Contents

The Normal Adult Heart and Methods of Investigation�������������������������������������������������   1 S. Kim Suvarna Cardiac Electrophysiology�������������������������������������������������������������������������������������������������  25 Paul D. Morris and Jonathan Sahu Cardiac Imaging�����������������������������������������������������������������������������������������������������������������  49 Abdallah Al-Mohammad and Peter W. G. Brown Current Therapeutics for Cardiac Disease�����������������������������������������������������������������������  75 Abdallah Al-Mohammad The Heart at Autopsy, Including Radiological Autopsy of the Heart�����������������������������  93 S. Kim Suvarna Embryology of the Heart��������������������������������������������������������������������������������������������������� 127 Michael T. Ashworth Ischaemic Heart Disease����������������������������������������������������������������������������������������������������� 137 Katarzyna Michaud Myocarditis ������������������������������������������������������������������������������������������������������������������������� 153 Martin J. Goddard Valvular Heart Disease������������������������������������������������������������������������������������������������������� 167 Stephen D. Preston Transplant Pathology��������������������������������������������������������������������������������������������������������� 185 Desley A. H. Neil Cardiomyopathies��������������������������������������������������������������������������������������������������������������� 205 Clare R. Bunning and S. Kim Suvarna Cardiac Tumours ��������������������������������������������������������������������������������������������������������������� 227 Doris M. Rassl Congenital Heart Disease��������������������������������������������������������������������������������������������������� 255 Michael T. Ashworth Sudden Cardiac Death������������������������������������������������������������������������������������������������������� 277 S. Kim Suvarna Index������������������������������������������������������������������������������������������������������������������������������������� 313

ix

The Normal Adult Heart and Methods of Investigation S. Kim Suvarna

1

Introduction

The heart is a complex, folded and hollow muscular structure situated just to the left side of the mid-low sternum when viewed from the front of the body, being enclosed by the pericardial sac and joined to the great vessels [1–3]. It develops its shape from embryological folding of the cardiac tissues (see chapter “Embryology of the Heart”) [4]. The surface/external landmarks of the cardiac tissues, viewed from the front of the body, are the right and left parasternal second intercostal spaces down to the right sixth costal cartilage with the apex of the heart being in the fifth left intercostal space mid-clavicular line. To appreciate any cardiac disorder one must first appreciate normality. This simple statement is deceptive since the heart is an organ with a complex 3-dimensional architecture, along with mechanical functionality. It is made mostly of muscle, although the muscular tissue varies in format across the chambers, with a dynamic microscopic layout and an elegant electrophysiological function. There are non-­muscular tissues are also present within the heart, namely blood vessels, nerves, connective tissues and fat. These are normal, but excessive amounts of fat and fibrous tissues may point to a pathological process, as discussed later in this book. In this chapter the normal heart is considered purely from the macroscopic and histological perspective, along with attention to specialised cardiac tissues and structures. It should be appreciated that the images have been derived following standard autopsy and referred examinations [5, 6]. This often involves separating the mid-ventricular tissues in transverse section through to the apex. Other views in this chapter, and elsewhere in the book, show cardiac tissues with the ventricles intact.

S. K. Suvarna (*) Sheffield Teaching Hospitals, Royal Hallamshire Hospital Sheffield Teaching Hospitals Sheffield, South Yorkshire, UK e-mail: [email protected]

There is no single, or perfect, method to consider and assess cardiac tissue. Indeed, all examinations should be guided by the anatomical or clinical query to be considered by the pathologist and/or other medical practitioner. However, any cardiac sample examination and assessment requires a clear understanding of relevant clinical data— which may often include results of a variety of tests (see chapters on “Cardiac Imaging” and “Electrophysiology”) along with “Medication History”.

2

Pericardium

The pericardium is a serous cavity/sac-like structure (Figs. 1 and 2), with slight distensibility, encasing the heart and its large vascular connections. It measures about 1 mm in thickness although peripheral mediastinal fat may make it appear more prominent. At the superior boundary it may abut the thymus. It has a well-defined dense fibrous wall with collagenised tissue. As stated, a variable amount of fat lies immediately adjacent. There is normally only a scanty amount of clear (serous) fluid within the pericardial sac, generally less than 0.5 ml. This fluid allows for heart movement freely during contraction (systole) and relaxation (diastole), as well as movement of the chest. The pericardial sac (containing the heart) is in direct continuity with the adjacent mediastinal soft tissue structures of the oesophagus, thymus and the lungs on either side [2]. The inferior aspect of the pericardial sac is bounded by the diaphragm. The superior aspect, comprising the great vessel tributaries and soft tissues, runs up to the thoracic inlet [2]. Histologically the pericardium, and the outer surface of the heart, is lined by a monolayer of bland cuboidal mesothelial cells (Fig.  3). The wall of the pericardial tissues comprises dense fibrous connective tissue with some adjacent fatty connective tissue. There is a limited blood and lymphatic vascular network present with scanty nerve fibres. There is normally no inflammatory cell component.

© Springer Nature Switzerland AG 2019 S. K. Suvarna (ed.), Cardiac Pathology, https://doi.org/10.1007/978-3-030-24560-3_1

1

2

S. K. Suvarna

Fig. 3  Histology of the pericardium demonstrates mesothelium either side, reflecting the pericardial and pleural aspects. There are scattered vascular channels and some fibrous and fatty connective tissues. The high magnification (low right inset) shows detail from the regular monolayer of mesothelial cells (Haematoxylin & eosin)

3 Fig. 1  The pericardium has been opened and shows the heart within the smooth-surfaced pericardial sac

Fig. 2  The back of the pericardium is seen, after heart tissue removal. The openings of the venae cavae, aorta, pulmonary artery and pulmonary veins are visible

External Cardiac Morphology

The external (epicardial) surface has a smooth aspect with some fat usually being evident (Fig. 4). Rarely, for those with significant systemic disease/malnutrition, these may be no appreciable fat. There may be minor fibrous thickening on the anterior epicardial surface of the right ventricle where the heart ‘rubs’ against the pericardium/chest wall (Fig.  5). A moderate and variable amount of fat is usually seen running parallel/alongside the coronary blood vessels and in the grooves between chambers. There is a small lymphatic vascular network present along with a few lymphoid cells often to be found in the fat. The heart is generally regarded about the size of the individual’s fist in health, growing from childhood through to adult status. However, it may be considerably enlarged in varying states of disease. However, the concept of ‘fist size’ is unreliable various and measurements may give a better method for assessment. Indeed, the weight of a normal heart may be assessed only if empty of blood and detached from adjacent tissue. There exist tables to compare cardiac weight against body mass, which may assist analysis—with predicted weights and a population range [7, 8]. Alternatively, one may make a basic calculation against the body mass, with expected values of 0.45% and 0.4% for males and females respectively [9]. However, the standardised charts and calculation derivations might be argued to be dated and thereby not reflective of current diets and exercise habits. Some caution always needs to be exercised when considering the cardiac weight. Indeed, there is some evidence to

The Normal Adult Heart and Methods of Investigation

Fig. 4  The heart is seen from the anterior aspect with two auricles at the top. The ventricular chambers are seen with coronary vessels coursing across the surface

suggest that a higher figure (possibly 0.51% of body mass) might be more appropriate [10, 11]. In simple terms, there are four chambers with great vein and artery connections evident when viewed from the outside. The two atrial and two ventricle compartments are separated at the level of the coronary sulcus by the central fibrous body. There is internal septation into right and left halves by the inter-atrial and inter-ventricular tissues. Whilst often displayed in planar two-dimensional view, it should be appreciated that, when the individual is standing, the thoracic anatomy points the heart partially downwards and towards the left side of the body [2, 12]. Thus, when the individual is standing upright, the base of the heart is mostly the right atrium and right ventricle. The chambers are best considered in sequential format, and convention has it that all analysis follows “the flow of the blood” [1, 5, 13]. One ‘starts’ in the right atrium, then considers the right ventricle, left atrium and left ventricle in turn.

4

3

Fig. 5  The pericardium often has a small area of semi-opaque fibrous tissue. This reflects rubbing of the heart against the pericardium during contractions in life

Right Atrium and Tricuspid Valve

Externally, this roughly round chamber is noted to have an external slightly triangular appendage, or auricle (Fig. 6).

Fig. 6  The heart specimen is seen from the upper right oblique aspect, allowing inspection of the right atrial appendage in detail. The appendage has a roughly triangular structure

4

The right atrium derives venous blood flow from the superior and inferior venae cavae, the coronary sinus and other minor cardiac veins. The venae cavae enter the posterior and basal aspect of the right atrial chamber. There is some returning cardiac venous blood inflow from the coronary sinus, usually covered in part by a flap of tissue (Eustachian valve) (Fig. 7). The sinus may have a vestigial ‘web’ present, being a normal anatomy variant (Fig. 8). This sinus receives blood from the great cardiac vein, but there are small anterior (or minor) cardiac veins draining the anterior wall of the right atrium (Fig. 7).

S. K. Suvarna

The chamber internally consists of three parts, termed venous, trabeculated and vestibular, as indicated (Fig.  9). There is a smooth walled posterior part (venous inlet), joining the superior and inferior venae cavae, which also derives blood from the coronary sinus. On the anterior aspect and fully involving the appendage are parallel, trabeculated muscle bands. The auricular appendage itself consists of the triangular projection running anteriorly across the upper anterior cardiac tissues (Fig. 6). There is an ill-defined groove externally (sulcus terminalis) that highlights the division of the venous (posterior) and trabeculated (anterior) zones internally with a ridge (crista terminalis). The other part of the chamber is the smooth surfaced vestibule (at the base of the chamber) being the support tissue for the tricuspid valve and the outflow for this chamber (Fig. 9). The inter-atrial septum shows the fossa ovalis, being an oval depression/residuum from closed-off gestational/ante-­ natal blood flow connection (foreman ovale) (Figs. 7 and 9). In about 15–20% of the population there is probe patency of part of the septal tissues (Fig. 10). This is a normal anatomical variation and usually without functional deficit—unless the hole remains patent into childhood/adulthood, allowing ‘shunting of blood’ across the septum (see chapter “Congenital Heart Disease”). The coronary sinus/Eustachian valve tissues are noted posterior-laterally in the right atrium (Figs. 7 and 9).

Fig. 7  The right atrium has been opened and laid almost flat revealing the fossa ovalis at the top with the coronary sinus below (∗). The trabeculated atrial appendage is seen, opened onto both sides of the dissection. One can also see several small coronary veins (v) entering from the septal tissue

Fig. 8  The coronary sinus may have a congenital web of fibrous tissue within the opening. This has no functional deficit with regard to the heart and is not normally a focus for infective endocarditis

Fig. 9  The right atrium has been opened, from the posterior aspect with a cut between the superior and inferior venae cavae and with a further cut running alongside the atrial and ventricular septum. Focally there is the trabeculated appendage/auricular architecture. There are smooth endothelial surfaces elsewhere. The trabeculated (Trab) and vestibular areas (Vest) are marked. The venous part sits next to the entrance points of the superior and inferior vena cava vessels. On the left side one can see the atrial septum and fossa ovalis. The tricuspid valve is present at the base. The valve chordae are noted to run from edge of the valve to the papillary muscle tissues of the ventricle

The Normal Adult Heart and Methods of Investigation

5

5

Fig. 10  This autopsy specimen shows forceps demonstrating the probe patency of the atrial septum, passing through the foramen ovale

Right Ventricle and Pulmonary Valve

The second cardiac chamber is the right ventricle (Fig. 12). This makes up the anterior and inferior part of the heart (when standing erect) and has a vaguely crescentic architecture on transverse section (Fig. 13) with some features suggesting an inverted cone shape, in the longitudinal axis. There is a variable fat (adipocytic) component, often best appreciated by histology (Fig. 14). Internally, there are three parts evident, described as inflow, trabecular and outflow components (based on the path the blood takes passing through the chamber). There are coarse trabeculations across the chamber evenly, apart from the outflow tract (conus), that is relatively smooth walled (Fig. 15). The right ventricle is thinner than the left (Fig.  13). Mature fat (adipose tissue) in variable amounts is normally present in the free wall of the chamber (Fig.  14). This should not be confused with the fat/fibrous tissue replacement of arrhythmogenic (right ventricular) cardiomyopathy (see chapters “Cardiomyopathies” and “Sudden Cardiac Death”). Given that the ventricular wall varies markedly in thickness around the circumference and has variable fat content, it is often found that different pathologists, assessing the same right ventricle tissues, will derive different wall thicknesses. Consequently, it may be ­pragmatic to assess right ventricle wall thickness in the outflow tract,

Fig. 11  The unopened tricuspid valve, seen from above, shows the flaccid valve leaflets with chordae radiating downwards into the ventricle tissues. The mitral valve is seen towards the left side of the image

Blood flows out from the right atrium across the atrio-­ ventricular orifice/tricuspid valve (Figs. 7 and 11). This usually has a circumference of 10–12  cm. It allows blood to enter the posterior aspect of the right ventricle. The valve is generally defined as three parts (anterior, posterior and septal). The tricuspid valve (Fig. 11) has a thin pliable tri-leaflet structure (less than 1  mm thickness), anchored onto the myocardial wall tissues and by chordae tendinae onto the right vestibular wall (Fig. 9). The right side of the atrio-ventricular orifice, from above, is part of the fibrous skeleton of the heart.

Fig. 12  The right ventricle is seen from the posterior aspect with the atrium above. The coarse trabeculations are noted and the chordae/papillary muscles are also evident. The inlet is seen immediately below the tricuspid valve. The ventricular trabeculated part is clearly demarcated and the blood passes towards the outlet/conus

6

S. K. Suvarna

Fig. 13  A transversely sliced view of right and left ventricle tissues. The variably trabecular architecture can be fully appreciated in this view along with comparative detail of the two ventricular wall thicknesses. This view of the cardiac tissues (standard in autopsy examinations) allows the cross section of the chamber diameters can be defined. The coronary arteries are seen running through the epicardial fat peripherally

Fig. 15  The right ventricle outflow tract has been opened anteriorly allowing inspection of the pulmonary valve, smooth aspect conus tissue and proximal pulmonary artery. Measurement of the right ventricle outflow tract at this point is recommended (arrow) as this thickness is easy to define and repeat—particularly as no trabeculations are present

Fig. 14  The right ventricle can be very fatty and with relatively few myocytes. This should not be regarded as indicative of arrhythmogenic cardiomyopathy (see Chap. “Cardiomyopathies”)

approximately 10 mm below the pulmonary valve—being a smooth area, which is easy to define and measure (Fig. 15). This area also has less peripheral fat, along with a smooth inner surface allowing for ease of assessment. The normal right ventricle outflow tract measures 2–4 mm. There are three papillary muscles, running from the walls of the right ventricle that anchor into the free margins and ventricular undersurfaces of the tricuspid valve by fibrous cords (chordae tendinae) (Fig. 12). The septomarginal trabeculum (Fig. 16) is a notable/well-­ defined band of muscle crossing the cavity of the right ventricle from the septum to the anterior papillary muscle, importantly carrying part of the ventricular bundle of conducting tissue. This ensures that the papillary muscles have already contracted (tensioning the chordae tendinae) when

Fig. 16  The right ventricle and right atrial tissues has been opened to allow definition of the septomarginal band of myocardial parenchyma (arrowed) as it runs from the high septal tissue towards the right ventricle trabeculations and papillary muscles. This carries right bundle branch fibres to assist with effective right ventricular contraction

right ventricular contraction commences. This permits efficient closure of the tricuspid valve during systole. At the apex of the outflow tract is the pulmonary valve (Figs.  15, 17, and 18). This valve has three, roughly equal

The Normal Adult Heart and Methods of Investigation

7

6

Fig. 17  This is the pulmonary valve seen from above in a freshly dissected heart showing the thin and translucent tri-leaflet architecture without coronary ostia. The aortic valve is seen at the top of the image

Left Atrium and Mitral Valve

Oxygenated blood returns from the lungs to the left atrium through the four pulmonary veins, two from each lung. These enter the venous inlet (akin to the right atrium anatomy) high/posteriorly in the chamber. For the left atrial chamber itself, there are no specific landmarks externally—apart from the oblong (rather ‘dog-ear’) appendage (Fig. 19). Otherwise, the left atrium has a broadly rounded architecture. Internally, the left atrium (Fig. 20) has a mainly smooth endocardial surface, although the left side appendage is trabeculated. The only other landmark aside from the veins and appendages, is the closed fossa ovalis. This is seen on the inter-atrial septal aspect, being the closed foramen ovale. The left arterial chamber (Fig. 20) can also be divided into the three parts, analogous to the right atrium. These regions are termed venous inlet or posterior, appendicular and vestibular, the latter supporting the adjacent valve. After atrial contraction, blood passes across the left atrioventricular orifice/mitral valve into the left ventricle. The mitral valve (Figs. 20 and 21) has a normal adult ring circumference of 8–10  cm. Contrasting with the right atrium, there are only two mitral (bicuspid) valve leaflets present. The anterior cusp/leaflet is longer in circumference and larger than the posterior, with the valve opening having a slightly curved architecture (Fig.  21) when viewed from above [6]. These

Fig. 18  The pulmonary artery root shows three normal valve cusps similar in quality with a bland intimal aspect for the pulmonary artery/ trunk. One notes the absence of coronary ostia. The trabeculations of the right ventricle are coarse and thick, as compared with the left side

semi-lunar leaflets joined minimally, by commissures, to each other at the wall. The ring circumference is generally 5.5–7.5 cm. The valve function permits unidirectional deoxygenated blood to flow into the pulmonary artery and thence towards the alveolated tissues. The pulmonary artery is discussed later in this chapter.

Fig. 19  The left atrium is seen from the cardiac specimen from the upper left oblique aspect. The appendage has a somewhat square architecture

8

S. K. Suvarna

Fig. 20  The left atrium has been opened by cuts interlinking the pulmonary veins and with a further cut running alongside the atrial and ventricular septum. The two leaflets of the mitral valve are seen with adjacent chordae and papillary muscles. The auricle (a) can be inspected directly. The left atrium is largely smooth surfaced

(Figs.  1, 4 and 22). Readily appreciable fat within the left ventricle wall and septum often points to a pathological process, such as cardiomyopathy or prior infarction. The left ventricular chamber is longer than the right, and has a rounded transverse cross section, with an inverted cone architecture. The chamber has trabeculations throughout, apart from the very apical part of the outflow tract (making this area useful for estimating wall thickness and possible hypertrophy). However, some prefer to estimate left ventricle wall thickness by measurements taken in the posterior, lateral, anterior free wall and septum. The left ventricle trabeculations are somewhat finer than those of the right ventricle, but are most marked towards the apex. There are two large papillary muscles, attached via chordae tendinae to the mitral valve (Fig. 22). The chordae insert into the free margins and ventricular surfaces of the mitral valve (Fig. 23). Within the chamber (Fig.  24) there is the comparable inlet, trabecular and outflow ventricular compartment subdivisions. The aortic outflow tract/vestibule is thin muscle and, in part, fibrous. The left ventricle outflow tract passes behind the right ventricle outflow tract, approximately in perpendicular fashion, to finish at the aortic valve. The aortic valve ring has an average circumference of 5.5–8.5 cm. It is centrally positioned within the fibrous body/skeleton of the heart. Indeed, it is sometime described as wedged (centrally within a triangle) between the pulmonary valve

Fig. 21  The dissected heart is seen with the mitral valve, from above. The tricuspid valve is seen adjacent, towards the right side of the image

valve leaflets also have a similar structure and attachments as that of the tricuspid valve, but with two papillary muscle anchors.

7

Left Ventricle and Aortic Valve

The last chamber to be considered is the left ventricle, on the left side of the heart (Fig. 4). There is normally only scanty fat present peripherally, generally distributed on the epicardial aspect of the heart mostly alongside vascular channels

Fig. 22  The left ventricle has been opened from the back to demonstrate the atrial and ventricular chambers. As with the right side, there is the inlet, ventricular trabeculated and outlet components evident. The trabeculations are slightly finer than those on the right side. Papillary muscles are seen at the ventricular base/side walls

The Normal Adult Heart and Methods of Investigation

Fig. 23  The mitral valve undersurface (ventricularis) is seen in detail, folded towards the side, from the ventricular aspect. Firstly, its close alignment to the aortic valve tissues can be appreciated. Secondly, it should be noted that the chordal tissues insert, not only at the edge of the valve, but also at other positions along the valve undersurface

Fig. 24  The left ventricle outflow tract has been opened, by running a cut up the anterior wall of the left ventricle, alongside the left anterior descending artery, and then turning between the left main stem artery and left auricle across the aortic valve. The smooth inner aspect of the outflow tract can be appreciated along with the left ventricle wall thickness. This is a good point to measure ventricular wall thickness, as it is easy to define and standardise. The proximal aortic tissues can also now be examined for arterial disease and the coronary ostia are also seen

(anterior) and the atrio-ventricular valves (posteriorly) when viewed from above and if one has the atria removed (Fig. 25).

9

Fig. 25  The dissected heart, if the atrial tissues are removed en-bloc, will demonstrate the four valves sitting largely within a single plane of tissue in the mid-section of the cardiac parenchyma. In this view, the pulmonary artery (PA) is towards the front (lower part of image) with the aorta/aortic valve (AV) in the centre and the mitral (MV) and tricuspid (TV) valves towards the left and right side respectively at the back of the heart

The muscular wall of the left ventricle is comparatively thick (Figs. 13 and 24), on average being 14–20 mm in the outflow tract (growing from childhood to adulthood). The same limitations of wall thickness assessment exist for this chamber, as do for the right. Indeed, getting different pathologists to agree where to measure a single wall thickness is often difficult, even if they use multiple measurement points (e.g. anterior, lateral, posterior, etc.). This is clearly an unsatisfactory solution, given the pivotal role this chamber has for systemic circulation realities. It is recommended that measurements are taken 10 mm below the outflow tract anteriorly, as a simple standard position—unless the case involves asymmetric hypertrophy or a local mass lesion. It should be noted that those involved in regular heavy sports training, may have a considerably greater left ventricle muscle mass, than average members of the population. It should also be noted that some similar hypertrophy of the left ventricle is seen at the end of a normal pregnancy. Both of these physiological situations will return normal to the average ventricular mass, after the end of training or following the birth of the child respectively. Furthermore, it should be appreciated that, as one passes to old age some general/ mild cardiac tissue mass reduction may occur, particularly in the left ventricle. One of the most important components of the left ventricle is the septum, being the wall between the two ventricles. This is mostly of relatively uniform thickness muscle tissue. However, it is thinner at the top, with apex of the septum comprising a slender band of fibrous tissue (membranous septum), which can be transilluminated (Fig.  26). This

10

Fig. 26  The dissected heart is held up to the light, allowing centrally the membranous septum to be transilluminated. This may help identify the area of the atrio-ventricular node and His bundle to be blocked histologically

S. K. Suvarna

Fig. 28  The opened aortic outflow tract has the right and left coronary artery ostia in the right and left coronary sinus positions. The posterior sinus has no coronary artery ostium

Fig. 29  The opened aortic valve shows two coronary sinuses with coronary artery ostia, together with the three-leaflet valve architecture. The left coronary artery emerges from one of the cardiac valve cusps, which has been cut in half. Note the proximity of the aortic valve tissue to the mitral valve elements

Fig. 27  The aorta close-up view, tilted en-face, shows the tri-leaflet architecture without the leaflets being closely opposed. This is a normal feature of the post-mortem heart and should not be regarded as evidence for regurgitation

s­ tructure is in communication with the tricuspid valve ring, adjacent to which passes the His bundle (see section “The Conduction System” below). This membranous portion is derived from the same tissue that forms the valves. It develops separately to the muscular part of the septum. Blood leaving the left ventricle is at high pressure as it passes across the three semi-lunar valve leaflets of the aortic valve. The aortic valve (Figs. 27 and 28) structure is similar

to the right side pulmonary counterpart. However, immediately above the aortic valve, two of the three valve sinuses (left, right) have the orifices of the left main stem and the right coronary artery (Fig.  29). The third (posterior) sinus has no coronary artery orifice, being often defined as the ‘non-coronary’ sinus. Occasionally, the small conus artery has origin adjacent to the right coronary artery, appearing as a separate small vessel. This should not be confused with a coronary anomaly—being a normal anatomy variant (see chapter “The Heart at Autopsy, Including Radiological Autopsy of the Heart”). Occasionally the left anterior descending and circumflex arteries originate separately from the same cusp. The local aortic wall has a firm but pliable quality being generally 0.5–2 mm thickness.

The Normal Adult Heart and Methods of Investigation

8

Aorta and Pulmonary Artery

11

These are large muscular and elastic arteries, emerging from the top of the heart, similar to each other in calibre and architecture (Figs. 15 and 29). However, the wall of the aorta is thicker as it has to cope with a higher blood pressure, compared with the right side. Macroscopically the normal lumen aspect shows a bland and smooth endothelial-lined surface. The basic histological structure (Fig.  30) is that of an inner thin loose fibrous intimal lining, surfaced by a bland monolayer of endothelial cells. Enclosing this layer is the media, which comprises the bulk of the artery wall. There is an inner elastic lamina between intima and media. The outer layer of the vessel, termed the adventitia, is vascularised fibrous connective tissue and fibro-adipose parenchyma.

There is an outer elastic lamina at the interface of the media and adventitia. Small arteriolar and capillary blood vessels (vasa vasorum) penetrate the walls of these large arteries (in common with all such vessels) to supply the vessel wall tissues. Looking specifically at the aorta, even in young apparently very fit individuals there is often a minimal amount of non-specific/intimal focal fibrous thickening histologically approximately rarely amounting to more than 5–10% lumen stenosis, although larger amounts develop with increasing age. This is, by contrast, not the same situation for atheroma in the pulmonary artery, which might point to pulmonary hypertension particularly if seen after the third artery division onwards.

Fig. 30  Histological examination of the aortic and pulmonary arterial tissue shows a thicker aortic wall, reflecting the greater pulse pressures, with otherwise similar content and structure. Note that the intima is

very thin, being almost impossible to define. The bulk of the wall comprises the muscular media and elastic tissue. The outer fibrous adventitia is seen peripherally (Elastic van Gieson)

12

9

S. K. Suvarna

 he Histology and Ultrastructure T of the Myocardial Wall

Histologically, the epicardium (outer cardiac layer) has a thin layer of fibrous tissue. The outer surface of the heart is bounded by a monolayer of mesothelial cells—akin to the pericardium (see Fig.  2). As reported, there is variable fat, accentuated adjacent to the coronary vessels. The majority of the heart is muscle, enough there also is a fibrous tissue framework, rich vascular network with some innervation. Whilst the tissues are similar across the whole organ, the chambers and the walls of each of the four chambers are histologically distinct. The cardiac myocytes [14– 16] in the left ventricle and septum are somewhat brick-like and are closely applied to one another, with little other tissue (Figs.  31, 32, and 33). The myocytes in the right ventricle have a variable amount of admixed adipocytes (Figs. 14 and 34). With regard to the fatty tissue component of the right ventricle, it should be noted that this varies between the sexes—with females having more adipose tissue than males. The amount of fat also generally increases with age [17]. Myocytes of the atria are often more slender and supported in a fibrous and variably fatty stroma (Fig. 35). The ventricular cardiac myocytes (Figs. 31 and 32), generally of the order of 125  ×  30  μm (length  ×  width), have central nuclei and abundant cytoplasm—the latter element mainly containing eosinophilic myofibres [15, 16]. There is usually a single central nucleus, although this may not always be present in the plane of section. In most areas of the myocardium the cells are arranged in longitudinal format, akin to a brick wall. However, at the front and back of the heart, at the confluence of the right and left ventricles there is often a degree of cellular disarray, not to be confused with hypertrophic cardiomyopathy (Fig. 36).

Fig. 32  Autopsy left ventricle tissue showing autolytic/degenerative qualities and some separation between the cells. Many autopsy myocardial tissues show slight interstitial widening/oedema, potentially reflecting the circumstances surrounding the death of the individual. This is common in hospital cases and should not be over interpreted as a pathological process in its own right. Capillary blood vessels are evident. Within the bands of fibrous tissue are darkly nucleated cells. These are uncommitted interstitial cells, often misinterpreted as lymphocytes (Haematoxylin & eosin)

Fig. 33  Resin-embedded thin section of cardiac myocytes demonstrating the banded muscular structure. Intercalated discs can be seen at the longitudinal ends of the myocytes (arrowed) (Toluidine blue)

Fig. 31  Left ventricle tissue is seen with closely applied cardiac myocytes and minimal interstitial tissues

Rarely a small number of cardiac myocytes may have more than one nucleus. The striated quality of the muscle can be appreciated with standard histology, although phosphotungstic acid haematoxylin, Masson’s trichrome or toluidine blue (resin thin section) histochemistry (Fig. 33) are better at demonstrating fine cell detail. The cells may often exhibit focal lipochrome accumulation, often immediately adjacent to the nucleus (Fig. 37). The yellow-brown pigmented material comprises effete cell membrane and organelle matter.

The Normal Adult Heart and Methods of Investigation

Fig. 34  High magnification of the right ventricle shows cardiac myocytes with significant amounts of mature adipose connective tissue interspersed. It should be noted that there is no significant fibrous tissue associated (Haematoxylin & eosin)

13

Fig. 37  Light microscopy of ventricular tissue showing brown pigment in the cytoplasm (lipochrome/lipofuscin). This is seen commonly in those of advanced age (Haematoxylin & eosin)

Fig. 35  Atrial tissue myocytes are often more slender and are set within a fibrous stroma Fig. 38  Ultrastructural view of a cardiac myocyte showing a central nucleus (N) with associated lipofuscin (arrowed) along with the banded muscular tissue. Abundant mitochondria are evident

Fig. 36  There is often a degree of fibre disarray at the confluence of the right and left ventricles, with no disarray elsewhere. This should not be confused with hypertrophic cardiomyopathy (Haematoxylin & eosin)

The nucleus itself often varies in shape and size, being enlarged ovoid in the young and fit, and roughly square in hypertrophic states. The terminal junctions of the myocytes have dense eosinophilic bands, known as intercalated discs (Fig. 33). These are sites of cell:cell adhesion and electrical communication, the latter effectively allowing the heart to act electrically and functionally akin to a syncytial mass. The ultrastructure of the heart largely recapitulates that seen at light microscopy. The cardiac myocytes classically are seen as having a block-like architecture with a well-­ defined nucleus and myofibrils (Figs.  38 and 39). Electron microscopy also allows detailed inspection of the banded myofibril structure. These bands reflect the different protein filaments, which vary in architecture according to whether

14

Fig. 39  Electron microscopy of a cardiac myocyte showing nucleus (N) and banded muscular tissue immediately adjacent. Small tubular (T tubules) and cisternal spaces are arrowed. Scattered mitochondria are also present

Fig. 40  Electron microscopy high magnification detail of the banded myofilament architecture allows some better appreciation of the arrangement of the contractile protein apparatus. Overlaid in schematic form are representations of the thin actin and related protein filaments (black) and thick myosin/related protein filaments (white). The contraction of cardiac myocytes is accomplished by the ATP-driven ratcheting of the proteins to allow greater/lesser interdigitation, and thereby cardiac myocyte shortening before later relaxation

the tissue fixation occurs when there is relaxation or contraction. The portion of tissue between the two consecutive Z lines is called a sarcomere. There are two main types of interdigitating filaments (Figs. 40 and 41), being designated thick and thin filaments. The thin actin filaments overlap variably the thick myosin filaments. During contraction, the ATP-driven energy-dependant ‘ratcheting’ of the filaments occurs and the two Z lines are drawn closer together, resulting in cellular contraction. This shortens the cell along the long axis. The

S. K. Suvarna

Fig. 41  An electron microscopic view of the heart cell in cross-section showing part of the end-plate centre. The myofibrils are seen with the actin and myosin fibres in close interlocking elements. Sarcoplasmic reticulum and mitochondrial elements are also noted adjacent

tightly bunched myofibrillar pattern is appreciated in transverse view (Fig. 41). Higher magnification also allows inspection of the cytosolic, or non-myofibril, compartment. There are usually abundant mitochondria, along with glycogen stores and lipid droplets (Fig.  42). There is some smooth and rough endoplasmic reticulum as would be expected. These parallel the standard features seen in other tissues. Special attention to the mitochondria, with their relatively uniform folded membranous architecture is recommended, in order to exclude mitochondrial myopathy. At the longitudinal poles of the cells are the intercalated discs. These are best appreciated in schematic form (Fig. 43). They have a complex folded structure and have several types of junction. Firstly, there are fascia adherens (linking different cells longitudinally in line with myofibrillar apparatus and cell skeleton elements). Then, there are the gap junctions [18, 19]. These are sometimes known as nexi, and contain the connexins (gating domains) which allow cell:cell depolarisation, akin to a syncytium. The cell:cell depolarisation wave to propagates rapidly due to this electrical cell:cell coupling. After childhood, the intercalated discs are only found at the ends of the long axis of the cells, but are more widely distributed in infancy. There are also undifferentiated regions usually present between the desmosome/myofibrillar insertion sites of the intercalated discs (Fig. 43). Contraction follows the cell membrane depolarisation wave being swept inwards along the invaginated membrane elements. These are called the T-tubule system (Figs. 39 and 41), allowing superficial/external and deep cellular depolarisation in uniform fashion. The T-tubules pass directly towards the Z-bands of the myofibres, but interact with the sarcoplas-

The Normal Adult Heart and Methods of Investigation Fig. 42 Three-dimensional diagrammatic reconstruction of cardiac muscle cells in the region of an intercalated disc, a junctional complex between neighbouring cells. The interdigitating transverse parts of the intercalated disc form a fascia adherens, with numerous desmosomes; gap junctions are found in the longitudinal parts of the disc. The organisation of the transverse (T) tubules and the sarcoplasmic reticulum is also shown. This figure was published in Gray’s Anatomy 40th ed., Susan Standring editor, page 140, copyright Elsevier 2009

15

Myofibril

Sarcoplasmic reticulum

Dyad

Terminal cisterna of sarcoplasmic reticulum

T-tubule Sarcomore Sarcolemma

Gap junction

Z-disc

Fascia adherens

Fig. 43  Part of an intercalated disc is seen with a gap junction (GJ) being defined as a thin dense structure running between two adjacent myocytes. There is some ‘undifferentiated’ cell membrane (U) and insertion points of the muscular apparatus of each cell (∗)

mic reticulum closely, at regions called dyads. The sarcoplamic reticulum (Fig. 42) is a complicated folded cistern-­like network, layered around the bundles of myofibrils that serve to store calcium. At depolarisation, the sarcoplasmic reticulum activation allows calcium to be liberated resulting in myocyte contraction. The filament contraction involves adenosine triphosphate (ATP) lysis, increasing the protein interdigitation, thereby shortening the cell. Relaxation is also dependant on new ATP production—thus, also being energy dependent. In hypoxic tissues at autopsy, the contracted elements may remain, being known as contraction bands. Of note, within the atrial tissues there are dense membrane-­ bound granules, containing atrial natriuretic factor (ANF) . These granules after secretion are responsible for diuresis following atrial distension. The granules are found close the nucleus/Golgi apparatus (Figs. 44 and 45). The interstitial tissues of the myocardium comprise a small amount of fibrous (collagenised) connective tissue and

16

Fig. 44  Membrane-bound atrial natriuretic factor (ANF) granules (∗) can be seen in the cytoplasm of atrial cells in the upper right atrium, lying against the nuclear membrane

S. K. Suvarna

tissues are best appreciated with Masson’s trichrome histochemistry, and the tissues may be investigated for inflammatory components and the vasculature by means of immunohistochemistry. The use of trichrome stains is of particular value in the right ventricle, highlighting the enhanced fibrous tissue of an arrhythmogenic cardiomyopathic status (see chapters “The Heart at Autopsy, Including Radiological Autopsy of the Heart” and “Cardiomyopathies”). Such stains also reveal areas of scarring in the left ventricle—often in reaction to ischaemic damage or prior myocarditis (see chapters “Ischaemic Heart Disease” and “Myocarditis”). The interstitium ultrastructurally contains collagen and elastic tissues together with “adhesion fibronectin molecules”. In addition, there is some proteoglycans and loose amorphous matrix. The collagen (principally types 1 and 3) is required to maintain cell: cell alignment and structural integrity during contraction and relaxation. The collagens connect individual myocytes together, along with linkage to capillaries and the adjacent interstitial tissues. Fibronectin links the cardiac myocyte with local extracellular matrix.

10

Fig. 45  High magnification electron microscopy of atrial natriuretic factor (ANF) granules

a vascular network of blood and lymphatic vessels. Scattered mast cells are occasionally seen in the background, but no other significant inflammatory cell population (beyond that seen in the circulatory compartment) should normally be present. Nevertheless, a few lymphocytes can often be observed in autopsy samples (see chapter “The Heart at Autopsy, Including Radiological Autopsy of the Heart”). Fibroblastic elements may be seen in the interstitial compartment, and their nuclei can occasionally be confused with lymphocytes at light microscopy (Fig.  32). The interstitial

Coronary Blood Vessels

The coronary vasculature is divided principally into outer arteries and veins with some lymphatics. The coronary arteries run from the root of the aorta over the heart, progressively subdividing on the epicardial surface (Fig. 46). They only later pass perpendicularly inwards to supply the myocardial tissues, with progressive branching of the arterial vessels. The arteries normally have a smooth intimal/lumen aspect and muscular walls measuring less than 1 mm thickness. They are accompanied by the venous circulation, with similar branching patterns, and have fatty connective tissue adjacent (Fig. 47). The right coronary artery arises from the right aortic sinus just above the aortic valve leaflet, at the sino-tubular junction, passing in a groove (sulcus) between the appendage of the right atrium and upper part of the right ventricle. It turns inferiorly and then round onto the posterior-inferior aspect of the heart. In its mid-section, it gives rise to the acute marginal artery (passing perpendicularly down along the lateral wall of the ventricle). The right coronary artery continues and terminates at the top of the posterior interventricular septum, merging with the terminal part of the circumflex artery. At this point the combined vessel runs down along the midline of the heart posteriorly, being known as the posterior inter-ventricular descending artery. The left coronary artery arises from the left aortic sinus, likewise just above the valve, akin to the left main stem. This short artery (left main stem) runs downwards and forwards between the pulmonary trunk and left auricle where it divides (after about 5–15 mm) into the left anterior descending and (left) circumflex arteries.

The Normal Adult Heart and Methods of Investigation

Fig. 46  The anterior surface of the heart is displayed with a line representation of the main coronary artery branches. The dotted lines represent coronary vessels on the posterior aspect of the heart

17

onto the back of the heart. The circumflex vessel can end in two manners, with the first merging with the right coronary artery, as above. Alternately it may terminate as a set of obtuse marginal artery branches (usually one, but up to three can be seen: termed OM1, 2 and 3). The obtuse marginal artery (or group of arteries) runs downwards towards the apex along the lateral and posterior wall of the left ventricle. The left anterior descending artery runs anteriorly over the anterior septal tissues (between the ventricles) with branches penetrating into the septal tissues. There is also often a variable calibre diagonal artery starting in the mid-­ section of the left anterior descending artery running obliquely left and downwards across the left ventricular tissue. In most people (about 65% population) the right coronary system is larger than the left, referred to as right-dominant. There is some evidence that this pattern equalises with age as more blood is needed for the harder-working left ventricle. Aside from the main arteries described above, there are also small radicular arteries evident that are quite variable, but still part of normality. The coronary arteries anastomose poorly, and can be considered as end-arteries (i.e. lacking significant inter-artery anastomosis). Since the arteries blood supply is derived from the aortic root, it should be noted that the bulk of coronary blood flow occurs during diastole— which also blends with the reality that blood flow is better through the relaxation phase for ventricular tissue. The arteries have a similar histology (Fig. 48) to the aorta, with thin intimal tissues surfaced by endothelium. The medial smooth muscular/medial tissues have an internal and

Fig. 47  The epicardial surface of the heart shows both coronary veins as well as coronary arteries running alongside each other within the fatty tissue parenchyma

The circumflex artery curves towards the left around the atrio-ventricular sulcus towards the left side passing round

Fig. 48  A coronary artery is seen in cross section with well-defined elastic laminae and muscular walled structure. The peripheral fibrous adventitial tissue is seen. There is no significant thickening of the intimal layer (Elastic van Gieson)

18

S. K. Suvarna

Fig. 49  A cardiac vein is seen with a valve on the left-hand side of the lumen

external elastic lamina on the inner/outer aspect. The adventitia is likewise bland vascularised fibrous tissue. The arteries described progressively subdivide, become progressively smaller and eventually connect into a rich capillary network. The capillaries drain into venules and thence veins, broadly following the pathways of the coronary arterial system. The main venous system comprises the great cardiac vein, running in the posterior/left sided atrio-ventricular groove with drainage from the anterior, lateral and posterior aspects of the left ventricle. The small cardiac vein (draining the right side of the heart and right atrial tissue) joins close to the great vein termination point at the coronary sinus. The veins have valves present, akin to systemic veins (Fig. 49). It should not be forgotten there is a moderate network of lymphatic vessels within all the tissues of the heart. These lymphatic vessels have a monolayer of endothelium and have a simple basement membrane as the outer layer.

11

The Valves

There are two types of cardiac valve. The atrio-ventricular (tricuspid and mitral) valves separate the atrial and ventricular chambers, preventing reflux of blood into the atria upon ventricular systole (Figs. 9 and 20). The semilunar (pulmonary and aortic) valves, situated at the ventricular outflows, prevent reflux of blood back into the ventricles after ventricular systole ends (Figs. 15 and 28). Their gross anatomy is described alongside that of the chambers earlier in the chapter. Histologically, the atrio-ventricular valves (Fig. 50) have several parts best visualised by elastic van Gieson connective tissue stains. The central part (zona fibrosa) comprises the

Fig. 50  Composite diagram of the tricuspid valve showing the ventricular wall attachment and the valve arching across the centre and top of the image towards the left side. Chordae (c) are noted running from the under surface of the valve tissue. The high magnification cross section of one of the chordae is noted in the bottom left corner (h). High magnification (inset ** low right) of the valve tissues shows a variably collagenised framework with thin endothelial surfaces and some elastic parenchyma (Elastic van Gieson)

strength of the valve, being composed of a plate of dense collagenous matrix. The fibrosa, accounting for the valve structural function, is tied into with the annulus (wall interface) of the valve tissue, being composed largely of collagen. There are very sparse nerve fibres and lymphatics present. The ventricular aspect valve tissue (ventricularis), is formed of a monolayer endothelium covering a thin connective tissue a substructure containing elastic. Within this tissue there are also the fibrous anchor points of the chordal tissues. The chordae tendinae are similar to joint tendons with parallel collagen fibres running along the long axis of the fibre. They enmesh with the adjacent papillary muscles (Figs. 22 and 50). The atrial aspect (atrialis and spongiosa) is composed of loose connective tissue with minor elastic tissue. The amount of loose spongy connective tissue (mainly proteoglycans with some collagen and elastic fibrous tissue) varies along the length of the valve cusp with scanty fibroblast-­like cells being present. There is normally no blood vessel or inflammatory component present within the valve. The semilunar valves (Fig. 51) also show similar layers, being covered with endothelial cells. Proceeding from the ventricle into the artery there is ventricularis, spongiosa, fibrosa and arterialis compartments respectively. The fibrosa again shows dense collagen with some elastic tissues. There may be occasional fibroblasts, but there normally is no vasculature. The fibrosa blends with the adjacent annulus/wall tissues. The ventricularis contains elastic tissue with some collagen. The spongiosa is only partially present towards the outer rim of the valve. The ventricularis layer is focally

The Normal Adult Heart and Methods of Investigation

19

Fig. 52  The right atrium is seen from above, with some tension applied upwards to the auricular appendage. At the root of the superior vena cava is a representation of the sino-atrial node (yellow). This cannot be specifically identified from the external or internal aspect and histological sampling of the tissues is required to completely block this area if it is to be considered

Fig. 51  The pulmonary valve tissues are seen here with the outflow tract of the right ventricle and proximal pulmonary artery root running along the left hand border. A high magnification inset of the valve tissues is seen allowing appreciation of the valve substructure (Elastic van Gieson)

thickened on the semilunar valves producing nodular bulges (aortic: noduli Arantii; pulmonary: noduli Morgagni). The arterialis layer has some loose collagen and elastic tissue.

12

The Conduction System

All cardiac myocytes have pacemaker (spontaneous rhythmic myocyte depolarisation) capability. They spontaneously beat/contract even in isolation (see chapter “Cardiac Electrophysiology”). However, to have an efficient heart beat, coordinated organ contraction is required, with this being managed by the cardiac conduction system. As none of the components of the cardiac conduction are visible by naked eye it is necessary to block the tissue around the appropriate landmarks and then section diligently through the tissues. Often one will need multiple levels, in order to fully identify all the appropriate structures histologically (see chapter “The Heart at Autopsy, Including Radiological Autopsy of the Heart”). Fixation of tissues prior to tissue blocking (i.e. not sampled fresh in the mortuary) is often an

advantage for providing appropriately orientated blocks of the conduction system parenchyma. The sino-atrial node (SAN) tissue acts as the pacemaker for the heart (see chapter “Cardiac Electrophysiology”). It is positioned at the apex of the crista terminalis at the top of the right atrium (Fig. 52). Depolarisation from the SAN tissue spreads steadily in a wave-like fashion across the atrial muscle towards the atrio-ventricular node. The node itself is a banana-shaped structure (Fig. 52) lying immediately in the subepicardial tissues at the junction between the superior vena cava and right atrium. Its position can be highlighted histologically to an extent by identifying the sino-atrial artery (Figs. 53 and 54). The SAN comprises a meshwork of irregularly orientated and rather slender myocytes (Fig. 55) within a poorly defined, connective tissue matrix. Another clue to the location of this tissue is the local sympathetic and parasympathetic autonomic nerve tissue (Fig. 56). There are three bundle of Purkinje (fast conducting) fibres that pass from the SAN to the atrioventricular node (AVN). Bachmann’s bundle (anterior internodal tract) is a small bundle of muscle fibre running on the inner aspect of the right atrium, passing from the top of the left atrium. It is thought to have a role in atrial synchrony, but it is rarely sought in terms of histology. There are also two other intra-­atrial pathways described, being the Wenckebach (middle internodal) and Thorel (posterior internodal) pathways. These are also rarely dissected for histology and analysis, but are considered important in terms of atrial dysrhythmias.

20

S. K. Suvarna

Fig. 54  The sino-atrial node tissue is seen immediately adjacent to the sino-atrial artery (∗). The right atrial (RA) and superior vena cava tissues are seen adjacent (Haematoxylin & eosin)

Fig. 53  Macroscopically, in blocks taken along the line of blood flow (see chapter “The Heart at Autopsy, Including Radiological Autopsy of the Heart”), the normal sino-atrial node is present within an ill-defined zone of fatty and fibrous tissue, adjacent to the sino-atrial artery

Atrial depolarisation is not transmitted directly to the ventricle, since the central fibrous body and local fat act as an insulator, stopping ventricular depolarisation. There is a delay in conduction of the depolarisation to the ventricles mediated via the atrio-ventricular nodal (AVN) tissues. These are also supplied by both sympathetic and parasympathetic nerve parenchyma, akin to the SAN. The atrio-ventricular node AVN (Figs. 57 and 58) is found at the base of the right atrium at the apex of the triangle of Koch, in the subendocardial tissue (see chapter “The Heart at Autopsy, Including Radiological Autopsy of the Heart”). It lies below the tendon of Todaro (running from the superior limb of the Eustachian valve/coronary sinus towards the membranous septum, and above the annulus of the tricuspid valve). The membranous septum can be readily seen by

Fig. 55  High magnification of the nodal tissue shows a somewhat haphazard myocyte arrangement set within fibrous stroma (Haematoxylin & eosin)

transillumination (Fig.  26) The AVN, like the SAN, has a rather haphazard histological myocyte architecture within

The Normal Adult Heart and Methods of Investigation

21

Fig. 56  High magnification view of the nerve and ganglionic tissues of the sino-atrial node, which are part of the autonomic nervous system

Fig. 58  Histological view of the atrio-ventricular node sitting immediately adjacent to the central fibrous body and part of the membranous septal tissues. The nodal tissue is arrowed (Haematoxylin & eosin)

Fig. 57  Right atrial tissues are seen with a schematic representation of the atrio-ventricular node (yellow) and the proximal pathways for the bundle branch tissues. Given the absence of clear anatomical landmarks of the nodal tissue one needs to block this entire area when considering nodal and bundle histology

fibrous stroma. The node tissue tapers to enter the membranous septum, and then runs as a slender tract of muscle (His bundle) fibre through this fibrous boundary (Fig.  59). Occasional adipocytes can be seen, but there should be no other tissue present in the bundle, as these might impede cardiac conduction. The base of the His bundle is just above the membranous septum (Fig. 26), where it splits into two bundle branches,

which supply the right and left ventricles. The bundles (Fig.  60) are conceptualised electrically as well-defined tracts, but the reality is that they run in small fascicles in a mesh-like array across the ventricular subendocardial parenchyma. The bundle branch conducting cells appear paler in sections than standard cardiac myocytes, and often have a slender disposition. These specialised bundle branch muscle cells also pass in the subendocardial tissues. They serve to deliver depolarisation impulses rapidly to the entirety of the ventricles. The papillary muscles often contract marginally prior to the bulk of the ventricular myocardium, tensioning the chordae.

13

Final Comment

Armed with knowledge of the normal anatomy and histology, one can now make an assessment of cardiac disease.

22

Fig. 59  Membranous septal tissue is seen with tricuspid valve and high ventricular septal tissues in the plane of section. The base of the His bundle is noted (∗) and there is clear fibrous tissue separation of the atrial and ventricular component requiring electrical depolarisation to pass along the bundle of His for ventricular excitation (Elastic van Gieson) Acknowledgements  Grateful thanks are expressed to Mr. B. Wagner, Senior Electron Microscopist, Sheffield Teaching Hospitals for his expertise and photography of ultrastructural histology in this chapter.

References 1. Anderson RH, Becker AE. Normal cardiac anatomy. In: Anderson RH, Becker AE, Roberts WB (eds). The cardiovascular system. Part A: general considerations and congenital malformations. Edinburgh, Churchill Livingstone, 1993; pp 3-26. 2. Abrahams PH, Spratt JD, Loukas M, van Schoor AN.  McMinn and Abrahams clinical atlas of human anatomy. 7th ed. London: Elsevier Mosby; 2013. p. 184–96. 3. Tortura GJ, Davidson B.  Principles of anatomy and physiology. 11th ed. Hoboken, NJ: John Wiley Sons; 2006. p. 695–735. 4. Sadler TW. Langman’s medical embryology. 12th ed. Hagerstown, MD: Lippincott, Williams & Wilkins; 2012. p. p162–200. 5. Suvarna SK. National guidelines for adult autopsy cardiac dissection: are they achievable? Histopathology. 2008;53:97–112. 6. Suvarna SK.  Thorax, heart, lungs, mediastinum, and pleura. In: Atlas of adult autopsy. New York, NY: Springer; 2016. p. 65–160.

S. K. Suvarna

Fig. 60  The base of the membranous septum has the bottom part of the His bundle displayed (∗) and part of the left bundle branch group radiating downwards and into the endocardial tissue adjacent (arrow) (Haematoxylin and eosin) 7. Kitson DW, Scholz DG, Hagen PT, Ilstrup DM, Edwards WD. Age-­ related changes in normal human heart during the first ten decades of life. Part II.  Maturity: a quantitative anatomical study of 765 specimens from subjects 20–99 years old. Mayo Clin Proc. 1988;63:137–46. 8. Sheppard M, Davies MJ. Cardiac examination and normal cardiac anatomy. In: Practical cardiovascular pathology. 2nd ed. London: Hodder-Arnold; 2011. p. 1–23. and 133–9. 9. Silver MM, Silver MD.  Examination of the heart and cardiovascular specimens in surgical pathology. In: Gotlieb AI, Schoen FJ, Silver MD, editors. Cardiovascular pathology. 3rd ed. New York, NY: Churchill Livingstone; 2001. p. 1–29. 10. Lucas SL.  Derivation of new reference tables for human heart weights in light of increasing body mass index. J Clin Pathol. 2011;64:279–80. 11. Gaitskell K, Perera R, Soilleux EJ.  Derivation of new reference tables for human heart weights in light of increasing body mass index. J Clin Pathol. 2011;64:358–62. 12. Ellis H, Logan BM, Duran AK, Bowden DJ. Human sectional anatomy. 4th ed. Boca Raton, FL: CRC Press; 2015. p. 104–34. 13. Becker AE, Anderson RH.  Cardiac adaptation and its sequelae. In: Cardiac pathology. An integrated text and color atlas, vol. 1. Edinburgh: Churchill Livingstone; 1982. p. 2–1.8. 14. Borley NR, Collins P, Crossman AR, Gatzoulis MA, Healy J, Johson D, Mahadevan V, Newell RML Smooth muscle and the

The Normal Adult Heart and Methods of Investigation c­ ardiovascular and lymphatic systems. Gray’s anatomy. Standring S Wigley CB The anatomical basis of clinical practice (40th edn). Barcelona, Churchill Livingstone Elsevier, 2008, pp 127-143. 15. Ross MH, Pawsha W.  Histology. A text and atlas. 6th ed. Philadelphia, PA: Lippincott, Williams & Wilkins; 2011. p. 400–39. 16. Veinot JP, Ghadially FN, Walley VM. Light microscopy and ultrastructural of the blood vessels and heart. In: Silver MD, Gotlieb AI, Schoen FJ, editors. Cardiovascular pathology. 3rd ed. New  York, NY: Churchill Livingstone; 2001. p. 30–53.

23 17. Tansey DK, Aly Z, Sheppard MN. Fat in the right ventricle of the normal heart. Histopathology. 2005;46:98–104. 18. Yeager M.  Structure of gap junction intercellular channels. Gen Struct Biol. 1998;121:231–54. 19. Severs MJ, Coppen SR, Dupont E, Yeh HI, Co YS, Matsushita T.  Gap junction alterations in human cardiac disease. Cardiovasc Res. 2004;62:368–77.

Cardiac Electrophysiology Paul D. Morris and Jonathan Sahu

1

Introduction

To serve its primary function of pumping blood, the heart depends upon a continually cycling sequence of complex electrophysiological processes. These processes rely on specialised molecular, cellular and anatomical adaptations. These generate and propagate electrical impulses through the heart in an organised fashion, ultimately resulting in coordinated myocardial contraction. The key components in this sequence involve: • A resting electrochemical gradient across the myocyte cell membrane. • Automaticity, which allows certain myocytes to spontaneously discharge electrical impulses and act as cardiac pacemakers. • Gap junctions between cells, which convey impulses to adjacent cells. • Specialised conduction tissues, which rapidly propagate the electrical impulses across the myocardium in an organised fashion. • Excitation-contraction coupling, which links electrical stimulation to myofibril conformational changes resulting in cellular contraction. • Repolarisation, which returns the myocyte membrane to the resting state before the depolarisation process restarts. This process commences in utero and cycles continually until death. A derangement at any level may result in arrhythmia and/or uncoordinated cardiac contraction, both of which can impair cardiac output. This chapter focuses on the normal electrophysiological and cellular contraction processes of cardiac electrophysiolP. D. Morris (*) University of Sheffield, Sheffield, UK e-mail: [email protected] J. Sahu Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK e-mail: [email protected]

ogy from the molecular level, up to the tissue adaptations of the human heart. Later in the chapter, clinically relevant topics such as the electrocardiogram, pacemaker therapy and basic electrophysiological interventions are also reviewed.

2

Cellular Electrophysiology

2.1

Resting Membrane Potential

The concentrations of electrically charged ions across the cardiac myocyte cell membrane are unbalanced, due in large part to the cell membrane being “semi-permeable”, with selective preferential permeability to particular ions. This differential loading results in electrical and chemical gradients across the myocyte cell membrane. These gradients are generated by energy-utilising, active transport mechanisms, which drive ions across the membrane against their concentration gradients. The resulting equilibrium state results in the inner surface of the myocyte membrane being negatively charged, relative to the extracellular fluid. This is known as the polarised state. Inside the cell, the cytoplasm contains a far higher concentration of potassium ions (K+) than the extracellular fluid. In the extracellular fluid, the predominant cations are sodium (Na+) and calcium (Ca2+) and the main anion is chloride (Cl−). These concentration gradients are generated and maintained by the active transport of ions, counter to their concentration gradients. This process consumes energy in the form of ATP (Fig. 1). In order to maintain the ion concentrations across the cell membrane and thus the resting potential, three membrane-bound, energy-utilising transport proteins are present spanning the cell membrane: The Na+-K+ ATP-ase The Na+-Ca2+ ATP-ase The Ca2+ pumps

© Springer Nature Switzerland AG 2019 S. K. Suvarna (ed.), Cardiac Pathology, https://doi.org/10.1007/978-3-030-24560-3_2

Transports 3 Na+ out for each 2K+ ions into the cell Transports 3 Na+ out for each 1 Ca2+ ion into the cell Transport Ca2+ from the cell and into the sarcoplasmic reticulum (SR)

25

26

P. D. Morris and J. Sahu

Fig. 1  Three of the membrane-spanning pumps are seen. These consume energy in the form of ATP in the generation of electrochemical gradients across the membrane. Approximate concentrations of the important cations are also shown in proportional format

To understand the development of the resting membrane potential one must consider the three cations (K+, Na+ and Ca2+) and their permeability across the cell membrane. The cell permeability to K+ ions is high and, as a result of the chemical concentration gradient, K+ ions diffuse out of the cell through potassium channels. Intracellular negatively charged ions such as proteins and sulphates cannot cross the selectively permeable membrane and therefore, remain inside the cell. The passage of positively charged ions out of the cell creates a potential gradient across the cell membrane with the development of an increasing negative charge on the inner surface of the cell membrane. Movement of potassium ions is determined by a balance of forces, between the developing electrical charge, which oppose K+ ion efflux and the concentration gradient which encourages K+ efflux. At the point where these opposing forces balance, equilibrium is achieved and the potential difference at this point is called the equilibrium potential (E). For K+ ions the ‘EK’ is around −96 mV, for Na+ ions the ENa is +52 mV, and for Ca2+ ions the ECa is +134 mV. The contribution of these cations towards the overall membrane potential is not equal and is determined by the permeability or ‘conductance’ (g) of the membrane to particular ions. Conductance is not fixed, but is dynamic, and so the conductance of individual ions can vary significantly depending upon the state of the membrane (resting or active). The membrane potential (Em) is the sum of the products of the conductance of a particular ion multiplied by the equilibrium potential of the ion, expressed as:

E m   g K E K    g Na E Na    g Ca E Ca    g Cl E Cl 

Under resting conditions the ionic conductance to potassium, gK, is far greater than the ionic conductance to sodium, calcium or chloride ions, and thus the membrane potential approximates the equilibrium potential of K+ (Ek). A small

background, inward Na current (Ib) results in the Em being slightly higher (less negative) than the EK. In most myocytes, this is balanced by the outward cation current (IKir) holding the resting membrane potential stable. The magnitude of the resting membrane potential is not uniform. For ventricular myocytes it is approximately –90  mV, whereas in sino-atrial (SA) and atrio-ventricular (AV) nodal cells it is more positive, around −60 mV due, in part, to a relative deficiency of a particular subset of K+ channels (inwardly rectifying Kir channels). The direction of ionic transport depends upon the physiologic state of the cardiac myocytes. For example, there is inwards movement of Ca2+ ions during phase 2 of the action potential but the ions are actively pumped out of the cell during other phases.

2.2

Ion Channels

Ionic transport across the myocyte cell membrane occurs through ion-specific membrane-spanning proteins. The structure and therefore selectivity of these channels is dynamic. Alterations in the intra- or extracellular electrolyte concentrations or the potential difference (voltage, mV) across the membrane may alter the conformation (structure) of channels, thus affecting conductance. This property is called gating. Cardiac channels can be divided into three classes depending upon the type of gating. • Voltage-gated channels (the majority) • Ligand-dependent channels • Receptor-coupled channels Voltage-gated channels undergo a conformational change in response to changes in the trans-membrane potential.

Cardiac Electrophysiology

Typically this takes the form of either activation (open) or deactivation (closed). The Na+ channel is typical of an ‘inactivating’ channel, wherein application of a test potential results in opening of the Na+ channel and a rapid influx of Na+ ions. Even with the maintenance of the potential, the sodium channel becomes inactivated and so the conductance falls to zero. This inactivated state is not the same as the closed state. In order for the channel conductance to increase once again, recovery from inactivation has to occur. For a non-inactivating channel the conductance remains high for the duration of the potential. On removal of the test potential the conductance falls to zero. An example of non-­inactivating channels are the inwardly rectifying potassium channels (Kir). Ligand-dependent channels are highly selective. Ligands can either activate (increase conductance) or inactivate channels. An example is the acetylcholine (ACh) dependent K+ channel (KAch). Acetyl choline binds to muscarinic receptors (M2) in the myocyte cell membrane resulting in the activation of G-proteins. Dissociation and interaction of the G-protein subunits with the K+ channel results in a configuration change and an increase in conductance to K+ ions. Receptor-coupled gating requires the translation of a physical stimulus into a conformational change in channel structure and conductance. Examples include stretch-­ activated chloride and cation channels. The former is important in regulation of cell volume and the latter important in cardiac dilatation.

Fig. 2  The sequence of activation and inactivation of the ‘m’ and ‘h’ gates in a cardiac myocyte Na+ channel. (a) The ‘m’ gate is closed during the resting state, which corresponds to action potential phase 4. (b) The ‘m’ gate opens in response to a depolarising stimulus, which brings

27

The Na+, K+ and Ca2+ channels and their corresponding currents are responsible for the generation of a cardiac action potential that propagates through the cardiac tissue. The organised electrical wave of depolarisation that occurs is closely followed by, and coupled to, the mechanical contraction of the heart. Although ion channels share a common basic structure with several transmembrane protein domains encircling a central aperture, there is heterogeneity in the ultrastructure and specific arrangement of the subunits and domains within the myocyte cell membrane. In short, ion channels vary in their level of complexity and this determines the individual characteristics of the channel.

2.3

Sodium Channel/Current

The Na+ channel is a good example of a voltage-gated, inactivating channel. Application of a small test potential results in a small increase in Na+ conductance and the generation of a small current. By increasing the amplitude of the test potential, the Na+ conductance and current increases. When a threshold potential (−40 to −50  mV) is reached the Na+ channels are open and the conductance and current is maximal. At more depolarised states, inactivation of the Na+ channel occurs. This can be seen as two ‘gates’ within channel, designated ‘m’ and ‘h’ gates, with these represented as a central ‘m’ and outer ‘h’ gates (Fig. 2). In the resting state

the potential to a threshold level, and this corresponds to phase 0. (c) After only milliseconds, the ‘h’ gate closes inactivating the channel. (d) On repolarisation, the ‘m’ gate closes and the ‘h’ gate opens

28

the channel is functionally closed by the ‘m’ gate. Partial depolarisation opens the ‘m’ gate, allowing inward Na+ ion flow and further depolarization of the cell membrane. The channel remains open for only milliseconds, since at higher potentials the ‘h’ gate closes, now rendering the channel inactive. On repolarisation, the ‘m’ gate closes and the ‘h’ gate opens (Fig. 2). These channels can be blocked in either the active or inactivated states to treat abnormal cardiac rhythms (see chapter “Current Therapeutics for Cardiac Disease”). This model of activation and inactivation similarly applies to other currents such as the Ca2+ current and the transient outward K+ current.

P. D. Morris and J. Sahu

conductance through this channel is blocked by polyamines and magnesium ions. This is a useful adaptation since it conserves the intracellular K+ and thus reduces the overall energy consumption of the myocyte. A separate group of delayed rectifiers (KV) are voltage gated and activate slowly during depolarisation. They, therefore, activate towards the end of the action potential (phase 2, see below) and bring about repolarisation back towards the resting membrane potential.

2.6

Calcium Channels/Current

Four types of Ca2+ channels have been identified in cardiac tissue. The two main types are the long acting (L-type) and 2.4 Potassium Channel/Current the transient (T-type) channels. The L-type channels are triggered by higher (depolarised) This is a diverse group. Eight types (distributed variably potentials and help sustain the action potential (AP) in phase across cardiac tissues) have been described with a variety of 2. The L-type membrane-associated channel is the channel modes of gating including activation/inactivation and non-­ responsible for the majority of the calcium entry into the inactivating, voltage-gating and ligand-dependent gating. myocyte. The consequent rise in intracellular Ca2+ concenOutward potassium currents are important in the phases 1, tration induces calcium release from the sarcoplasmic reticu2, 3 and 4 of the cardiac action potential (Fig. 3). lum (SR), which is vital in linking electrical excitation to the Transient outward K+ channels (Ito) occur in a relatively interaction of actin and myosin and thus myofibril shortenhigh density in the atria, nodal tissue, Purkinje fibres and the ing  – known as ‘excitation-contraction coupling’. These epicardial myocytes. This channel contributes to some of the L-type channels are the target for the dihydropyridine class regional variation in the action potential duration in the heart of calcium channel blocking drugs (e.g. amlodipine and (especially the phase 1 ‘spike’, see below). The other major nifedipine). group of K+ channels, which are important in the developT-type calcium channels affect the rise in the resting ment of the cardiac action potential, are the delayed rectifier potential (towards the depolarisation threshold potential) K+ channels (KV). This group includes IKr (rapid) and IKs observed in cells capable of spontaneously depolarising (slow) channels. These channels are non-inactivating. They without the need for an external stimulus. Thus, these chanare termed delayed because their activation times are rela- nels are important in initiating depolarisation. T-type Ca2+ tively slow with time constants of around one second. channels are well represented in tissues with increased automaticity like the sino-atrial node (SAN) and atrio-ventricular (AVN). The T-type Ca2+ channel is an inactivating voltage-­ + 2.5 The Rectifier K Currents gated channel, which is active at negative potentials and is inactivated by depolarisation. The activation and inactivation Inwardly rectifying K+ channels (Kir) conduct K+ inwardly at time constants of this channel are very short. Since the conpotentials more negative than EK and outwardly at potentials ductance of the channels is relatively poor, the resulting curhigher than EK. Therefore, they maintain (‘rectify’) the intra- rent is relatively small. cellular K+ concentration at EK, irrespective of the membrane The third type of Ca2+ channel occurs on the internal potential. The name is slightly confusing since under physi- membrane of the SR. Ryanodine receptors on the SR detect ological conditions and during the action potential cycle, K+ an increase in cytosolic Ca2+ concentration and this induces a ion movement is outwards. It is during voltage patch clamp release of Ca2+ from the SR, the so-called ‘calcium-induced experiments, with the application of strongly negative calcium release’ (CICR). The CICR is crucial in the (hyperpolarised) voltages when inward K+ ion movement ­development of cardiac contraction. The final mechanism for occurs. The weak Na+ background current (Ib) holds the Ca2+ release into the myocyte is an indirect mechanism utilismembrane potential (Em) slightly higher than EK and so the ing Inositol triphosphate (IP3). IP3 receptors are present in Kir channels normally allow K+ ion efflux in the resting state, smooth muscle and the conduction system. Increased levels along the steep concentration gradient, generating the resting of IP3 arise with both sympathetic and parasympathetic stimmembrane potential. During depolarisation (phase 2 plateau) ulation and paracrine substances such as angiotensin II.

Cardiac Electrophysiology Fig. 3  Timing and contribution of Na+, Ca2+ and K+ currents through an action potential cycle. The thickness of the bars approximates the magnitude of the currents. Na+ current; INa, Ca2+ current; ICa, Inwardly rectifying K+ channel current; IKir, Delayed K+ rectifier current; IKv, transient outward K+ current; IKto

29

0 mV

–95 mV

/ Na

/ Ca

/ Kir

/ Kv

/ Kto

30

2.7

P. D. Morris and J. Sahu

The Action Potential

The myocardial tissues continually cycle through depolarisation and repolarisation, with each cycle corresponding to a heartbeat. This cycle, or action potential (AP), is quickly conducted through the specialized cardiac conduction tissues in an organised and sequential manner to bring the wave of depolarisation to the whole heart and trigger a coordinated myocardial contraction. The AP follows five well characterised phases, designated as phases 0, 1, 2, 3 and 4 (Fig.  4). The resting membrane potential corresponds with phase 4. The action potential sequence is broadly similar in all cardiac cells. Depolarisation is followed by a transient partial repolarisation, followed by a plateau phase before repolarisation returns the membrane back to its polarised resting potential. Different anatomical areas of the myocardium serve different roles. For example, atrial myocyte function is subtly different to a ventricular myocyte, which, in turn, is different to an atrio-­ventricular nodal or Purkinje cell function. Localised differences in the AP sequence reflect this. Such differences account for variability in AP duration and properties such as automaticity, seen in nodal cells. Similar to the neuronal and skeletal muscle fibres, cardiac myocyte depolarisation is a very rapid process, taking the cell membrane from a negative membrane potential to a slightly positive membrane potential. However, the cardiac action potential is unique in its considerably longer duration (200–400 ms vs 1–4 ms) (Fig. 5). This specialised adaptation is driven by prolonged inward Ca2+ and then Na+ currents which are balanced against outward K+ rectification currents. This accounts for the plateau in phase 2 of the action potential. The delay allows time for myocardial contraction to occur.

Fig. 5  Comparison between a neuronal/skeletal-muscle action potential (red dotted line) and a much longer cardiac myocyte (black solid line)

2.7.1 Phase 4 Phase 4 is the so-called resting phase where the cardiac myocyte is held in a negative, polarised state with a resting membrane potential between −90 to −95  mV.  This phase corresponds to atrial and ventricular diastole. In atrial and ventricular myocytes, this phase is relatively stable (flat profile) and depolarisation (usually) occurs if a depolarising electrical stimulus is conducted to the myocyte from adjacent cells. If such a stimulus brings the myocyte membrane potential to a threshold level of ~–60 mV then fast Na+ channels open and phase 0 commences. 2.7.2 Phase 0 Once the threshold potential is reached, fast Na+ channels open and Na+ rapidly diffuses into the cell down the steep concentration gradient. This inward Na+ current (INa) raises the intracellular membrane potential (becoming less negative) to approximately +20 mV. Once depolarized, the voltage gated Na+ channels quickly inactivate and close. The Na+ current is thus abruptly terminated. 2.7.3 Phase 1 Phase 1 is a phase of rapid but only partial repolarisation back to a membrane potential of about zero (0  mV). It is driven by the transient outward current of K+ ions (Ito). This current is very short-lived being activated by depolarisation; and inactivated very promptly by the partial repolarisation.

Fig. 4 A single action potential sequence, with the five phases illustrated

2.7.4 Phase 2 During this phase, there are a number of competing, yet balanced, ion currents. The net result is a plateau in the membrane potential in the depolarised state. The early part of the plateau is driven by the inward Ca2+ current (ICa) through L-type Ca2+ channels. These are activated by depolarisation and, as their name suggests, remain activated for a prolonged time before slow inactivation. The latter part of the plateau is

Cardiac Electrophysiology

31

driven by the inward Na+ current (INa) which is, in turn, driven by the Na+: Ca2+ exchange pump. This pumps Na+ into the cell in exchange for Ca2+ ions out and is aided in its function by the relatively high intra-cellular calcium levels during myocyte depolarisation. Balancing these currents, the major outward current is of K+ ions through a variety of channels (Ito, IKr and IKs).

2.7.5 Phase 3 Towards the end of the plateau phase, the slow K+ channels, also known as delayed rectifier channels, open. K+ efflux currents (IKv comprising IKr and IKs) begin to increase in magnitude and eventually overpower the late inward Na+ and Ca2+ currents (INa and ICa) active during the plateau phase. This results in the membrane potential becoming increasingly negative and eventually becoming polarised once again (phase 4), back to a resting membrane potential of −90 to −95 mV. The heart is ‘wired’ in such a way as to allow these waves of depolarisation to efficiently and rapidly spread across the entire endomyocardial surface followed by the myocardial wall, resulting in efficient ventricular filling (diastole) and ejection (systole). The SAN is a collection of myocytes found in the posterior wall of the right atrium close to the entry point of the superior vena cava (see chapter “The Normal Adult Heart and Methods of Investigation”). These cells spontaneously generate action potentials which, via gap junctions, are propagated to the remainder of the myocardium. The wave of depolarisation rapidly moves across the right and (via Bachmann’s bundle) the left atria, resulting in a near synchronous atrial contraction. The atria are electrically insulated from the ventricles by the annulus fibrosis, which corresponds to the fibrous mitral and tricuspid valve annuli. The only way for impulses to reach the ventricles is via the AVN, which is situated in the inferior inter-atrial septum (see chapter “The Normal Adult Heart and Methods of Investigation”). Here, the wave of depolarisation is held up, allowing optimal ventricular filling before being transmitted down to the inter-ventricular septum through the bundle of His. The bundle of His rapidly transmits the impulse down the septum towards the ventricular apex. The His bundle divides at the lower end into a right bundle branch and a left bundle branch. The left bundle branch separates into a left anterior and left posterior bundle. These insulated bundle fibres terminate into a vast network of sub-endocardial Purkinje fibres. The Purkinje fibres rapidly disseminate the wave of depolarisation to the entire right and left ventricles in an endocardial to epicardial direction. Although the ventricles depolarise almost synchronously, the apex to base direction of conduction allows the apex to depolarise and contract very slightly earlier than the base. This tensions the chordae of the

Fig. 6  Representation of the arrangement of the conduction tissues (‘wiring’) in the heart. (A) SAN, (B) Bachmann’s bundle, (C) AVN, (D) bundle of His dividing into right and left anterior and left posterior bundles, (E) Purkinje fibres, (F) insulating fibrous annular ring around the mitral and tricuspid valves

atrio-ventricular valves allows the heart to expel blood in an efficient manner (Fig. 6).

2.8

Nodal Action Potential

Clusters of specialised cardiac myocytes arranged into the SAN and the AVN differ from other excitable tissue in that they have an intrinsic and spontaneous electrical instability (automaticity). In the healthy heart, it is these cells that act as the cardiac pacemakers, dictating the frequency of depolarisation and repolarisation of all the other myocytes and hence the heart rate. In health, the SAN cells serve this function but, as a fail-safe mechanism, a hierarchy of other areas (AVN, His bundle and Purkinje cells) can act as subsidiary pacemakers to prevent asystole, should the sinus node fail. However, these subsidiary pacemakers have slower rates of discharge. As a general rule, the anatomical distance from the SA node is associated, inversely, with the frequency of action potential generation. Consequently, subsidiary pacemakers, if required to perform this function, do not always pace at a sufficient rate or in such a coordinated fashion.

32

The action potential profile of nodal cells is distinct from that of other myocytes. Polarisation is not as prominent in these cells, since there is a relative deficiency of the rectifier K+ channels (Kir). The “resting” membrane potential is therefore less negative than previously described, at approximately –60  mV, and is thus held closer to the threshold potential. The ‘resting’ membrane potential does not really exist in such cells since the potential in phase 4 is constantly rising towards depolarisation. Instead of the relatively stable, flat profile of phase 4 previously described, the membrane potential is constantly drifting upwards (thereby becoming less negative) and bringing the myocytes towards the threshold potential which will trigger depolarisation (Fig. 7). The quicker the rise in membrane potential in phase 4, the quicker the rate of discharge and, the faster the heart rate. These phase 4 ‘pacemaker potentials’ are generated by several unbalanced ion currents. In addition to the slow inward background Na+ current (Ib), the ‘hyperpolarisation-activated’ or ‘funny’ current (If) allows Na influx at potentials more negative than −50 mV. The channel and its corresponding current are known as ‘funny’ since it is, rather unusually, activated by hyperpolarisation and not depolarisation. The Ib and If currents result in the initial rise in membrane potential in these cells. When a threshold of −50  mV is reached, activation of transient (T-type) Ca2+ channels occurs. Nodal cells possess a high density of these channels. The resulting Ca2+ ion influx further depolarises the membrane potential and brings the cell to approximately −40 mV. At this point, the L-type Ca2+ channels become activated. This is the major current of phase 0 in nodal myocytes and is distinct from atrial and ventricular myocytes. The kinetics of these currents are relatively slow, in comparison to the fast Na+ current which accounts for phase 0  in non-nodal myocytes. As a consequence the

P. D. Morris and J. Sahu

slope of the phase 0 depolarisation in nodal cells is much gentler and the conduction velocity slower in these cells. The currents that cause repolarisation in nodal tissues are similar to the non-nodal cells (i.e. the delayed rectifier K+ currents, IKv). The origin of the cardiac depolarisation arises from the site with the fastest rate of spontaneous depolarisation. In the normal, healthy heart this occurs in the SAN. If this node, the natural pacemaker site of the heart, fails due to disease, then subsidiary pacemaker sites such as the AVN may take over as the site of origin of cardiac depolarisation. A good example of this is in complete heart block, where there is no electrical communication between the atria and the ventricles. In this situation, the heart relies on cells below the atria to generate action potentials, otherwise, ventricular asystole would occur. Depending on the precise level of the block, cells in the AVN, His bundles, or Purkinje fibres can, therefore, take over the role of pacemaker. However, the ‘escape’ rate of spontaneous depolarisation in these tissues is slower, often resulting in symptomatic bradycardia and the requirement of an medical device pacemaker.

2.9

Modulation of the Pacemaker Function

The heart is well innervated. The intrinsic rate of SAN depolarisation is approximately 100–110 beats/min (bpm). However, at rest the adult heart normally beats at around 60 bpm and during significant exercise the rate can be closer to 200 bpm. The pacemaker rate is therefore modulated by a variety of factors, the most important of which is the autonomic nervous system. The parasympathetic nervous system, via the vagus (cranial X) nerve, has the effect of slowing the pacemaker rate, the speed of conduction, and the duration of the action potential, via acetylcholine (ACh) nerve innervation. Its nerve fibres are mainly concentrated in the nodal and conduction tissues. The sympathetic nervous system (T1–T5, sympathetic chain) innervates relatively more of the myocardial tissues including the ventricular myocardium and activation has the opposite effect. Since the inherent rate of the SAN is 100  bpm, at rest, the parasympathetic nervous system is dominant. Noradrenaline (NA) released from sympathetic nerve fibres and adrenaline (A) from the adrenal medulla bind to beta adrenoreceptors, of which the beta-1 adrenoreceptor predominates. Receptor binding results in activation of adenylyl cyclase via a G protein-coupled signal transduction pathway. Adenylyl cyclase catalyses the conversion of ATP into cyclic adenosine mono-phosphate (cAMP). cAMP has Fig. 7  A comparison between a ventricular or atrial action potential the effect of activating the ‘funny’ channel in nodal cells and (black line) and a nodal action potential (blue line). Note the less-­ negative and upwardly drifting phase 4 membrane potential, the slower also of activating protein kinase A (PKA). Activation of the funny current (If) accelerates the rise in phase 4 of nodal cells phase 0 and lack of plateau in the nodal action potential

Cardiac Electrophysiology

and so accelerates ‘pacemaker potentials’ and increases heart rate (positive chronotropic effect). Activation of PKA has several desirable effects. PKA activates L-type Ca2+ channels, which also accelerate the phase 4 pacemaker potentials in nodal cells. L-type Ca2+ channel activation results in an amplification of the depolarisation and plateau magnitude currents. The resulting increased inward Ca2+ flux increases intracellular Ca2+ concentration and this increases contractility (positive inotropic effect), which is usually appropriate under sympathetic activation. PKA also augments the function of the delayed rectifier currents and this shortens phase 2 and 3 of the action potential. Another effect of PKA is a faster clearance of Ca2+ back into the sarcoplasmic reticulum following systole. This results in quicker relaxation of the myofibrillar contractile apparatus, this being known as a positive lusitropic effect). With predominant parasympathetic activity, ACh binds to muscarinic M2 receptors and, via the release of inhibitory G protein subunits, inhibits adenylyl cyclase. This results in a reduction in cAMP.  The effects of parasympathetic activation are therefore, to oppose the mechanisms and currents described above. The rate of rise of the membrane potential in phase 4 is therefore reduced, thus delaying the development of the next action potential. Spontaneous pacemaker activity is thereby reduced with a consequent reduction in heart rate (Fig. 8). Once a cardiac myocyte has depolarised, further stimulation cannot generate an action potential for a period of time. The absolute refractory period (ARP) includes phase 0, 1, 2 and the first part of phase 3. The short, remaining period of time to completion of repolarisation is the relative refractory

Fig. 8  The effect of the autonomic nervous system on nodal pacemaker potentials. A ‘normal’ nodal action potential is shown in black. Sympathetic nervous stimulation increases pacemaker potentials in phase 4 and causes a steeper rise in the phase 4 potentials. The threshold potential is therefore reached sooner and thus, an action potential is triggered earlier (blue dotted line) resulting in a more frequent rate of discharge. Parasympathetic stimulation has the opposite effect and causes a slower rise in potential in phase 4. The threshold is reached later and so the action potential is delayed (green dotted line) resulting in a slower rate of discharge

33

Fig. 9  The absolute refractory period (ARP) and the relative refractory period (RRP)

period (RRP) (Fig. 9). During the RRP, only a stimulus of sufficient energy may generate another cardiac action potential. This is one of the mechanisms which may cause polymorphic ventricular tachycardia (PVT).

2.10

 ffects of Electrolyte Disturbance E and Disease

Disturbances in intra- or extracellular electrolyte concentrations can have deleterious effects on the action potential and therefore the rhythm of the heart. Since the resting membrane potential is influenced more by K+ (EK) than any other ion, alterations in the concentration of this ion have more impact than any other. Significant hyperkalaemia influences the EK and this increases the resting membrane potential (less negative). At less negative potentials, Na+ channels are more likely to be inactivated and, although the membrane potential is ‘closer’ to the threshold potential, this results in a weaker and more delayed phase 0. Hyperkalaemia also augments the function of the delayed K+ rectifier channels, which results in stronger repolarisation currents and ultimately in more rapid repolarisation. The effect of hyperkalaemia is therefore a reduction in myocardial excitability and a reduction in the speed of onset, duration and magnitude of phase 0 of the action potential, along with a brisker repolarisation phase. Overall, the action potential duration is shortened. Profound hyperkalaemia can result in cardiac arrest due to bradycardia or asystole. Conversely, hypokalaemia induces hyperpolarisation and this results in an increase in the number of Na+ channels held in the activated state. This results in more excitable myocardium. Hypokalaemia inhibits the function of Kir channels leading to a delayed repolarisation phase and a prolongation of action potential duration. This can predispose to re-entrant tachycardias such as ventricular tachycardia (VT), which may degrade into VF or asystole.

34

P. D. Morris and J. Sahu

Magnesium is required for Na-K ATPase function and for the partial blockade of the rectifying K+ channels during plateau phase 2. Consequently, hypomagnesaemia results in reduced polarisation (less negative in phase 4), intracellular K+ depletion and prolonged repolarisation. In extreme cases hypomagnesaemia can result in re-entrant arrhythmias like PVT also known as Torsade de Pointes (TdP). Profound hypocalcaemia results in a prolonged plateau phase whereas profound hypercalcaemia shortens the plateau phase. Hypercalcaemia also reduces the action potential conduction velocity and reduces the ARP. Thus, the net effect of hypercalcaemia is a shortening of the action potential duration and predisposition to AVN block. The net effect of hypocalcaemia is a prolongation of the action potential duration with a corresponding lengthening of the refractory period, which may have an anti-arrhythmic effect. In chronic heart failure and conditions associated with myocardial fibrosis (e.g. after a myocardial infarction) myocytic expression of certain K+ channels (predominantly Kto) may be reduced, leading to a delayed repolarisation phase. The prolongation of the plateau phase results in a prolongation of the action potential duration and higher intracellular Ca+ levels, which may predispose to after-depolarisations. After-depolarisations and prolongation of repolarisation (and thus refractory period) may both result in malignant ventricular arrhythmia. This helps explain the higher rates of sudden arrhythmic death observed in this population. Acute ischaemia induces the KATP channel, which is a subtype of Kir channel. This increases rectifying currents and overpowers the inward cation currents during the phase 2 plateau. This results in a shortening of phase 2, more rapid repolarisation and less Ca2+ influx. This has a negative inotropic effect. This is desirable since it helps, at least in part, limit any ischaemic burden.

3

 ardiac Conduction and Excitation-­ C Contraction Coupling

3.1

Conduction of Electrical Impulses

Cardiac myocytes are arranged and connected longitudinally. The junction between two cardiac myocytes is called the intercalated disc. Adjacent myocytes are connected by anchoring proteins called desmosomes. Efficient cardiac contraction (systole) and relaxation (diastole) relies upon the atria and subsequently the ventricles contracting sequentially as single units. The wave of depolarisation and repolarisation must therefore propagate across the myocardium rapidly with little resistance to ionic flow between cells. Ion currents and therefore, impulse transmission, is made possible by the presence of gap junctions between myocytes.

Gap junctions are membrane-spanning conduits that allow ions to flow between adjacent myocytes freely. Each gap junction or ‘connexon’ is constructed of six protein subunits (‘connexins’) arranged into a membrane spanning tubule. Connexons join end-to-end with connexons from adjacent cells. This arrangement results in the cytoplasm of each myocyte being in direct continuity with the cytoplasm of, not only adjacent myocytes, but the remainder of the myocardium. The density of these gap junctions varies in the adult, with a relatively high density along the orientation of the fibre but are relatively sparse orthogonally. As a result, the conduction velocity along the fibre orientation is rapid, with conduction velocities in other angles being slow. Certain loci, such as the SAN and AVN have a relatively low density of gap junctions with a consequent slowing of action potential conduction. Within healthy myocardium, there is variability in conduction velocity between different myocardial regions. Conduction velocity is proportional to the density of gap junctions and relates to the predominance of either fast (non-nodal myocytes with Na+-dependent phase 0) or slow (nodal myocytes with Ca2+-dependent phase 0) response fibres. In pathological situations, for example when scar is present within the myocardium, the conduction velocity may be significantly impeded, or even blocked, which may encourage an arrhythmia. The impulse conduction velocity may also be influenced by other factors such as the autonomic nervous system, hormones, ischaemia and drugs.

3.2

Excitation-Contraction Coupling

The wave of cardiac depolarisation is quickly followed by contraction of the myocardium, being termed ‘excitation-­ contraction coupling’. The importance of rapid impulse conduction across the myocardium, in a normal heart, cannot be over-emphasised. Near simultaneous depolarisation in the atria, and later, the ventricles results in synchronous contraction of the atrium, and subsequently, the ventricles. The wave of depolarisation is delayed through the AVN and this allows time for atrial contraction to augment and optimally fill the ventricles (the atrial ‘kick’) prior to ventricular systole. This adds around 10% of the final volume of the ventricle, prior to contraction. The action potential duration in cardiac myocytes is much longer than that of nerve cells and skeletal muscle fibres. However, there is another key difference between these tissues. Nerve and skeletal muscle cells release Ca2+ in response to the high intracellular Na+, which occurs with depolarisation. However, in cardiac myocytes it is the rise in intracellular Ca2+ concentration, which occurs during phase 2 of the action potential, that is the stimulus for calcium release from the sarcoplasmic

Cardiac Electrophysiology

reticulum, and thence myocardial contraction. This is termed calcium induced calcium release (CICR). Contraction of the myocardium occurs during phase 2 of the cardiac action potential. Ca2+ enters the myocyte during this phase via Na+/Ca2+ exchange and the L-type Ca2+ channel. The L-type Ca2+ channels are clustered in the transverse tubules, close to the junctional regions of the sarcoplasmic reticulum (SR) (see chapter “The Normal Adult Heart and Methods of Investigation”). The L-type channels are activated during phase 2 (plateau) of the action potential and the resulting inward Ca2+ current causes a relatively small increase in the local Ca2+ concentration around the junctional SR.  The rise in Ca2+ concentration, in turn, activates the ryanodine channels (also known as calcium release channels) in the junctional SR membrane, leading to a release of Ca+ from the SR into the cytoplasm with a corresponding large increase in the cytoplasmic Ca2+ concentration. Towards the end of systole, the myocyte expels the Ca2+ back into the SR via Ca-ATPase pumps and back into the extracellular space via Na+/Ca2+ exchange proteins, a process known as restitution. The Ca2+ released into the cytoplasm interacts with the troponin-tropomyosin binding on the thin actin filaments causing a conformational change, which exposes the myosin binding site on the actin filaments. Cross-bridges can therefore form between the myosin heads and the actin binding sites. This is an energy utilising process with a single molecule of ATP being used to “cock” the myosin head prior to the force being applied and contraction. The contraction involves troponin-mediated movement of the actin over the thick myosin filament, a mechanism known as the ‘sliding filament mechanism’ (Fig. 10). Disengagement of the myosin head from the actin binding site also requires energy, utilising another molecule of ATP.  Thus, both contraction and relaxation are both energy dependant. Under resting conditions, the rise in intracellular Ca+ is sufficient to activate approximately 30–40% of the available actin-myosin cross-bridge binding sites. Under increased physiological stress, more Ca+ is released from the SR. This results in more actin-myosin cross-bridges being recruited, which has a positive inotropic effect (increased contractility). The contractile force is therefore increased in proportion to the rise in cytoplasmic Ca2+. Beta-adrenergic receptor stimulation by NA or A achieves this effect by increasing the amplitude and duration of phase 2 Ca2+ currents. Myocardial stretch increases the myocyte’s sensitivity to Ca2+ ions and this forms the basis of the Frank-­ Starling mechanism, whereby increased myocardial stretch (during diastolic filling) results in increased contractile force. Therefore, cardiac output is maintained under times of physiological stress and when there is an increase in diastolic filling pressures or prolonged filling times.

35

4

The Electrocardiogram (ECG)

The ECG provides a wealth of diagnostic information and has become a part of routine clinical assessment. Entire textbooks are dedicated to the subject. Here, the basics are provided, with further reading suggested later. The ECG non-invasively evaluates cardiac electrical activity via external transducers placed across the body surface. This is made possible by the fact that cardiac electrical signals are conducted through the body’s fluid and tissue compartments. It is important to note from the outset that the electrical signals recorded at the body surface onto an ECG tracing represent a summation of all the electrical impulses from all the tissues of the atria and the ventricles. This means that not all individual currents are seen on the ECG trace. Larger currents may ‘cancel out’ or ‘hide’ the effects of any smaller currents as the wave of depolarisation sweeps across the surface of the heart. By using 10 electrodes (1 of which acts as earth) on the surface of the body, 12 different representations of the electrical activity of the heart can be displayed (i.e. the ‘12-lead’ ECG). The 12 ‘leads’ do not refer to 12 individual transducers but to the nine unipolar and three bipolar ‘views’ of the heart which are represented in graphical form on a 12-lead ECG print-out. The limb leads comprise three bipolar recordings (leads I, II, III) and three augmented (a) unipolar recordings (aVR, aVL and aVF). These six leads look at the heart from different positions in the coronal plane to give a 2-D electrical representation of the heart. The recordings are such that an electrical wave moving towards the electrode is represented by a positive (upward) deflection, and a wave travelling in the opposite direction by a negative (downward) deflection. The electrical activity is delineated as an electrical potential plotted against time, measured against time (seconds). The opposite deflections are true for a wave of repolarisation. Thus, a single complex (cardiac cycle) with atrial depolarisation (P wave) followed by ventricular depolarisation (QRS waves) and repolarisation (T wave) is shown in (Fig. 11) with the component parts. Leads I, II and III are bipolar recordings. Lead I records the electrical activity of the heart as ‘seen’ between the left arm and right arm. The positive vector is from right to left. Lead II records between the right arm and the left leg with a positive vector from the right arm to the left leg. Lead III records between the left arm and the right leg. The positive vector is from the left arm to the right leg. The augmented (a) leads, aVR, aVL and aVF, are unipolar recordings. They display the summated signals from a single point, the right shoulder, left shoulder and symphysis pubis respectively. The limb lead tracings may be useful in deducing the overall electrical axis of the heart, which can be influenced by various disease states, and are also useful in

36

Fig. 10  A representation of the sliding filament theory of myofibril contraction. The diagram shows a simplified representation of a sarcomere with central myosin thick filaments surrounded by actin thin filaments which attach peripherally at the Z line. (a) The myosin binding sites on the actin thin filaments are blocked by tropomyosin (yellow). (b) Calcium (purple) causes a conformational change which exposes the myosin bind-

P. D. Morris and J. Sahu

ing sites allowing cross-bridge formation between actin and myosin. (c) The myosin heads are ‘cocked’ back which is an energy utilising process consuming a single molecule of ATP. Note that this has the overall effect of sliding the actin along relative to the myosin and shortening the sarcomere. (d) The myosin heads have ‘ratcheted’ on to the next binding site. Disengagement of the myosin heads also requires energy via ATP

Cardiac Electrophysiology

37

‘looking’ at the inferior surface of the heart (typically leads II, III and aVF), for example in diagnosing an inferior myocardial infarction. Figure 12 demonstrates the six limb leads and the angles from which they electrically ‘view’ the heart. In addition to the limb leads, there are six unipolar chest leads. These leads ‘view’ the heart in a slightly tilted transverse plane and can be used to localize abnormalities and disease states in the right ventricular, septal, anterior or lateral

walls of the heart. If required, transducers can be moved onto the right side of the chest or to the posterior chest wall to aid the diagnosis of diseases in other non-standard territories (e.g. right ventricular or posterior territory myocardial infarction). Figure 13 demonstrates the arrangement of the six chest leads. Taken together, the 12 lead ECG can thus be appreciated to illustrate the equivalent of a 3-D view of the electrical activity of the heart. A standard ECG is performed at a paper speed of 25 mm/s and calibrated with a 1 cm deflection per 1 mV. The P wave represents total atrial depolarisation and the QRS complex represents ventricular depolarisation. The

Fig. 11 (a) A single ECG complex. The P, Q, R, S, T and R waves are labelled, as are the PR, QRS and QT intervals. (b) Multiple complexes are shown in a rhythm strip

Fig. 13  The six chest leads. Note V1–V2 correspond to the RV, V3–V4 correspond to the septal region and V5–V6 correspond to the left ventricle and lateral wall. Transverse section

Fig. 12  The six limb leads and the angles from which they ‘view’ the heart’s electrical activity. Coronal section

38

P. D. Morris and J. Sahu

T wave represents ventricular repolarisation. Atrial repolarisation does occur but this is usually hidden by the ventricular depolarisation which occurs at the same time. In health, a small Q wave represents septal depolarisation (from left to right). A larger (pathological) Q wave mostly represents an established myocardial infarction in a particular territory denoted by the lead involved. For a healthy individual, at rest, the ECG should be within the following normal limits: PR interval QRS interval QT intervala Should be corrected for rate

a

Fig. 14  An example of first-degree heart block with a prolongation of the PR interval. The start of the P wave occurs 320 ms (8 mm) before the the R wave. The normal PR interval is up to 200 ms

Fig. 15  An example of second-degree heart block, Mobitz type 1 (Wenckebach). Note the progressively prolonging PR interval culminating in a dropped beat

Fig. 16  An example of second-degree heart block, Mobitz type 2. Not every P wave is conducted to the ventricles. This example shows 2:1 atrioventricular block

120–200 ms 50% of capillaries, although lesser degrees of staining warrant discussion with the clinicians and testing for a donor-specific antibody, particularly when the C4d is ­performed on formalin-fixed tissues. A positive CD68 is considered when intravascular macrophages are present in more than 10% of capillaries. Any clinical or histological concern for antibody-mediated rejection should prompt testing of more biopsies than the minimum and many centres have adopted routine staining of all biopsies.

7

Contraction Band Artefact

Contraction bands (Fig.  9) are readily identified in cardiac transplant biopsies and are not an indicator of ‘contraction band necrosis’. They occur as a result of the procedure and placing in the cold fixative [7].

8

Quilty Lesion

This lesion, named after the first patient, is a mononuclear cell infiltrate (Fig. 10), which can be limited to the endocardium (Quilty A lesion) or involve the underlying myocardium, generally with quite conspicuous muscle damage

Fig. 9  Thick and thin transverse contraction bands are seen across the myocytes (H&E)

(Quilty B lesion). These first became recognised with the introduction of cyclosporine, not being present in the azathioprine and prednisolone era [41]. There has been some change in the balance between A and B lesions with the advent of tacrolimus [42]. Cyclosporine use in non-heart (i.e. renal/liver transplant recipients) does not produce Quilty lesions (in native hearts examined post-mortem) [43]. Quilty lesions are composed of predominantly T lymphocytes, which comprise more than 50% of the infiltrate with B lymphocytes and macrophages being present in lesser amounts. They also have a follicular dendritic cell framework which becomes increasingly more extensive with larger lesions [44]. There is an organisation to Quilty lesions (Fig. 11), which have a central B lymphocyte component with supporting follicular dendritic cells surrounded by T lymphocytes. Small capillary sized blood vessels, with high endothelial venule (HEV) features, are present within a Quilty lesion. The Quilty lesion has features of a tertiary lymphoid tissue suggesting an attempt to mount a local response to persistent alloantigen stimulation [45]. This is in contrast to rejection where the same cells are present, but without organisation of the cells into compartments. Macrophages make up more than 50% of the infiltrate and follicular dendritic cells and a vascular component are not present [44].

Transplant Pathology

193

Fig. 10 (a) Low power of multiple biopsies showing prominent Quilty lesions (effect). Some can be seen to be confined to the endocardium (Quilty A) (solid arrow) and some extending into the myocardium

(Quilty B) (dashed arrow) (H&E). (b) High power of a Quilty B lesion showing the typical dense inflammatory infiltrate. Entrapment of muscle fibres with apparent damage is readily apparent (H&E)

Fig. 11 (a) Low power of a Quilty lesion used for the immunohistochemistry stains (H&E). (b) Higher power of this quilty lesion showing the extensive vascularity within the lesion (H&E). (c) CD3 T lymphocyte immunostain, when compared to d, the T lymphocytes predominantly are forming a cuff around the central area where there are fewer

T lymphocytes. (d) CD20 B lymphocyte immunostain showing the central part of the Quilty has a more dense B lymphocyte population. (e) CD21 immunostain to demonstrate the follicular dendritic cell framework within this well developed Quilty lesion

194

D. A. H. Neil

10

Fig. 11 (continued)

Early studies found no adverse outcome related to either Quilty A or B lesions, so required no augmentation of immunosuppression. Thus, it is vital to differentiate the ­ Quilty lesion from true cellular rejection. It is generally accepted that most grade 2 rejections are tangentially cut Quilty B lesions [10, 46] and further levels should be cut or spares stained to try to confirm an endocardial component. Immunohistochemistry for CD3, CD68 and CD21 may also be helpful in identification of Quilty lesions. The most recent ISHLT guidelines no longer require distinction between A and B lesions. More recently studies have questioned the lack of adverse outcome of Quilty lesions [47–49], with variable findings on an association with the development of coronary vasculopathy [50, 51].

9

Previous Biopsy Site

These are common occurring in around 65% of all biopsies [52]. They vary in appearance depending on the time since the biopsy. Very recent previous biopsy sites often contain fibrin, a mixed inflammatory cell infiltrate and quite conspicuous muscle damage (Fig.  12a, b) and an overlying mural thrombus may be identified. With increasing time the thrombus undergoes organisation. The inflammation subsides and later the damaged muscle is removed by phagocytosis (Fig. 12c). The presence of haemosiderin-laden macrophages (Fig. 12d) is often the best indication that an area of inflammation in a previous biopsy site. Necrotic muscle fibres can be highlighted with C9 immunohistochemistry [24, 25]. If early post-transplant cases, the differentiation from peri-­ transplant injury and previous biopsy site may not be possible. Healed previous biopsy sites form fibrous scars indistinguishable from scars related to peri-transplant injury or ischaemia.

Peri-transplant Injury

There are numerous mechanisms of injury to the donor heart in the peri-transplant period. Injury can occur to the donor heart prior to transplantation as a result of the catecholamine surge (“storm”) during donor brain death. This situation produces contraction band necrosis and microinfarcts similar to those seen with inotrope use for haemodynamic maintenance. The catecholamine surge is most pronounced in donor death from head injury and it has long been recognised that patients with head injuries can die from a cardiac event [53– 56]. Contrastingly periods of hypotension, prior to organ procurement and hypothermic ischaemia during cold storage after procurement, can also result in varying degrees of reversible and irreversible myocyte injury. After transplantation a reperfusion injury occurs as a result of the release of free radicals. The patterns of the myocyte injury from these various entities are slightly different and poorly understood. The catecholamine surge produces contraction band necrosis. This cannot be assessed in early post-transplant biopsies as contraction bands may occur artefactually. Microinfarcts occur as a result of vasospasm related to the catecholamine storm or inotrope usage and are identified initially by coagulative necrosis. These manifest as hypereosinophilic, often wavy, myocytes with loss of nuclei. They are readily seen in early post-transplant biopsies for variable periods of time but usually most pronounced in the first few weeks (Fig. 13). Depending on the degree of immunosuppression there is a variable inflammatory response. Unlike rejection the muscle damage/necrosis is out of proportion to the inflammatory infiltrate (see Fig.  13), whereas in rejection the muscle damage is very subtle. C9 highlights the necrotic muscle fibres [24, 25] which are often single cells or small groups of ­ myocytes (see Fig. 13). This makes peri-transplant injury readily identifiable, as pure cellular rejection does not result in C9 positive necrotic myocytes. Neutrophil infiltration is part of the reperfusion injury, though this is not usually conspicuous in cardiac transplant biopsies.

11

 roblems Related to Specimen P Orientation

As these tiny 2–3 mm biopsies are randomly orientated there may be difficulties in distinguishing previous biopsy sites and Quilty lesions from rejection. A careful search for pigmented (haemosiderin laden) macrophages over the levels and serial sections helps confirm a previous biopsy site, whilst further levels or staining of spare sections may help confirm endocardial attachment for Quilty lesions. However, it should be recognised that a rejection infiltrate can also involve the endocardium.

Transplant Pathology

195

Fig. 12 (a) Low power view of a recent previous biopsy site with marked inflammation and muscle damage and overlying thrombus (arrow) (H&E). (b) Higher power view of the mixed infiltrate in a recent previous biopsy site (H&E) (c) An older previous biopsy site in

the region of 2–4 weeks of age in which the inflammation has largely subsided and much of the damaged myocytes have been cleared (H&E). (d) A higher power view of the biopsy site in Fig.  12c showing the brownish tinge of the haemosiderin laden macrophages (H&E)

Fig. 13 (a) Haematoxylin and eosin stained section showing scattered necrotic myocytes which are hypereosinophilic with granular cytoplasm and a surrounding light inflammatory cell infiltrate—the muscle

damage out of proportion to the inflammatory infiltrate. (b) C9 stained section staining scattered necrotic myocytes brown

196

12

D. A. H. Neil

Infection

Post transplant infections are decreasing, reflecting better anti-microbial prophylaxis and refinements in immunosuppression, allowing a lesser immunosuppressive burden, with 0.6 infective episodes per patient in the current era [57]. Infections involving the transplanted heart were rare, but are now very infrequent [7]. The most likely organisms to potentially cause histological changes in endomyocardial biopsies are viruses, with a viral myocarditis being indistinguishable from rejection in the majority of cases. With decreasing frequency Parvovirus B19, Epstein-Barr virus (EBV), cytomegalovirus (CMV) and adenovirus viral genome have been identified by nested polymerase chain reaction (PCR) in post-transplant endomyocardial biopsies [58, 59]. Parvovirus B19 is being increasingly reported, being often associated with an anaemia and myocarditis in both the immunocompetent and immunosuppressed [60]. CMV viral titres are generally monitored serologically and, if raised, a careful hunt for the characteristic inclusions within myocytes, endothelial or stromal cells should be undertaken. CMV immunohistochemistry should also be performed. In both North and South America Chagas’ myocarditis should be considered. Chagas’ disease can occur by transmission from the donor or reactivation in the recipient [61]. Amastigote nests can be seen on H&E and Giemsa stains [61]. Toxoplasmosis is endemic in Europe and some tropical areas. Previous exposure of both the donor and recipient to toxoplasmosis is tested prior to transplantation and routine prophylaxis is instituted using trimethoprim and sulphamethoxazole [62, 63]. Latent infection in the myocardium, in which there are cysts containing bradyzoites is the most common method of donor transmission. However, reactivation is possible, and a primary infection can also occur in those not previously exposed [62, 63]. Whilst the p­ rophylaxis has largely prevented toxoplasmosis, case reports do occur in patients not given prophylaxis because of allergy to either of the drugs [64]. A chronic donor toxoplasma cyst is the most likely way of finding toxoplasmosis in an endomyocardial biopsy.

13

incidence in children around 13% [68]. PTLD cases occur most commonly within the first year post-transplant [65, 66, 69], but have been reported to develop as long as 20 years post-transplantation [67]. There is a variable (30–60%) associated mortality rate [69]. Increased immunosuppression and EBV infection are two of the most important risk factors for the development of a PTLD [65, 66, 68, 69]. The lesions range from lymphocyte hyperplasia through to frank lymphoma, which have been divided into four basic histological types in the 2008 WHO classification of tumours of haematopoietic and lymphoid tissues [65, 66, 69, 70]. 1. Early lesions: There is preservation of underlying tissue architecture with mixed B and T lymphocytes and plasma cells. There are two histological patterns seen—plasmacytic hyperplasia and infectious mononucleosis-like. They are usually polyclonal and usually positive for EBV proteins by immunohistochemistry (LMP1) or EBV encoded RNA (EBERs) by in-situ hybridisation. 2. Polymorphic PTLD: Destruction of underlying tissue architecture, full spectrum of B lymphocytes. Most express EBER or LMP1. 3. Monomorphic PTLD: There is architectural and cytological atypia of a degree readily classified as lymphoma on morphological features. It is usually B-cell lineage, but some T-cell and NK cell lymphomas also occur. Most of the B cell lymphomas are diffuse large B cell lymphomas, but Burkitt lymphoma, plasma cell myeloma and plasmacytoma-­like lesions also occur. The B cell and NK cell lymphomas are usually positive for EBV. 4. Classical Hodgkin lymphoma-type PTL. In this the majority of Reed Sternberg cells are positive for EBERs. The presentation is usually extranodal with involvement of the transplanted heart rare (Fig.  14) [71]. Cardiac

Post Transplant Lymphoproliferative Disorder

Post transplant lymphoproliferative disorder (PTLD) is the most common malignancy following solid organ transplantation in children and second to skin cancers in adults [65]. Heart transplant recipients have one of the highest incidences of PTLD, in the vicinity of 2–5% [66, 67], with a higher

Fig. 14  Explanted/postmortem heart with a PTLD seen as the whitish nodule in the wall of the atrium involving the suture line

Transplant Pathology

involvement usually presents with heart dysfunction with the ­clinical suspicion of rejection. Endomyocardial biopsy will permit diagnosis usually.

14

Recurrent Disease

Recurrent disease is recurrence of the original disease process, which necessitated transplantation, in the transplanted heart. Histopathological assessment of the explanted heart is necessary to accurately diagnose as many histologically characterised causes of heart failure as possible. Pre-­transplant clinical diagnosis is incorrect in 30% of patients transplanted for conditions other than ischaemic heart disease [72]. Amyloidosis of all types can recur in the transplanted heart giving an overall lower survival rate that cardiac transplantation for other causes [73]. Patients currently transplanted for AL amyloid (light chain amyloid) should also be worked up for a bone marrow/stem cell transplant. Together with chemotherapy, this may provide complete remission in more than 60% of patients [73–79]. Combined kidney and cardiac transplantation are undertaken for non-AL variant Apo-A1 variants with recurrence occurring at 5  years, but the progression is slow with normal cardiac and renal function after 10 years [73, 75]. Combined liver and heart transplantation is performed for non-AL variant and wild-types of transthyretin amyloid. Recurrence does not occur as the liver was the source of the amyloid [73, 75]. Approximately 25% of patients transplanted for giant cell myocarditis recur, at a mean of 3 year post-transplant, with no impact on survival rates at 3  years post-transplant. Detection is by identification of giant cells in the routine endomyocardial biopsy. The earliest reported recurrence was at 3 weeks post-transplantation. Treatment is by initial steroid bolus and tapering course over several months [73, 80]. Sarcoidosis has been reported to recur in four patients, generally being identified on routine endomyocardial biopsy by the identification of non-caseating granulomas and exclusion of infective causes. They have been treated successfully with increased steroid dose [73, 81–84].

15

The Role of Molecular Assessment

Molecular assessment of transplant biopsies in not routine, but is likely to become increasingly utilised with the progress towards personalised medicine, molecular finding being incorporated within the Banff Renal Transplant criteria [85]. Molecular analysis of transplant endomyocardial biopsies started with an extra biopsy being obtained and sent to a central centre for molecular assessment [86, 87]. This method is at risk of sampling error with non-representative fragments

197

consisting predominantly of fibrous tissue, Quilty lesions, previous biopsy sites or peritransplant injury being sampled, the latter 2 only recently recognised as producing a “new molecular “injury” signature” [88]. This is evolving to new techniques allowing assessment of sections cut from formalin fixed paraffin embedded sections, such that the morphology can be assessed initially, representative areas identified and sampled/non-representative areas excluded followed by RNA extraction for molecular analysis [89]. This technique is amenable to being performed locally rather than in a large central laboratory minimising delays related to shipping, and is likely to be the method adopted with time. Transcripts associated with T cell-mediated rejection relate to activated effector T cells, or activated or interferon gamma (IFγ) induced macrophages whilst endothelial injury or activated NK cell transcripts are associated with antibody mediated rejection [90].

16

Graft Vasculopathy

Graft failure/patient survival after the first year has remained relatively unchanged [91] despite improvements in immunosuppressant treatments with a steady 5–6% annual loss. The major cause for this is the development of graft vasculopathy (GV)—also known as graft vascular disease, coronary artery vasculopathy, chronic vascular rejection, accelerated graft atherosclerosis. GV produces painless ischaemia of the denervated graft, presenting as sudden death, silent myocardial infarcts or heart failure [6]. There is diffuse involvement of all arteries, and veins may also be involved, with the ­second and third order intramyocardial branch arteries being the most affected [92–94]. The process is predominantly an intimal proliferative process and is considered to be a response to injury to the endothelium, although injury to the media may also contribute. The cause of the endothelial injury is multifactorial [91, 95], starting with systematic effects in the donor with brain death [96–98], endothelial and smooth muscle injury during the preservation period [99– 102] and subsequent reperfusion injury [100, 103]. After that, there is continuing post-transplant related to antibody mediated rejection and cell mediated vascular rejection (endothelialitis), viral infections [59], hyperlipidaemia and more. The balance between the various injurious agents will vary from patient to patient. The arteries involved are too large to be sampled in an endomyocardial biopsy and the diffuse nature of the disease makes GV difficult to identify on angiography [104]. Identification by angiography requires comparison of annual angiograms looking for a subtle decrease in calibre of the vessels. Intravascular ultrasound allows identification of intimal thickness and, if used in conjunction with

198

other software, can produce a picture of the area of the intima (called virtual histology) [104]. Allowing GV lesions to be detected in early stages means potentially more patients might be amenable to treatment [91]. Examination of the explanted heart either at the time of retransplantation or post-mortem allows identification of the arterial changes. Differentiation from pre-existing “normal” intimal thickening (Fig.  15) and pre-existing atherosclerosis (Fig.  16), sometimes severe and requiring coronary bypass grafting at the time of transplantation, is important. Coronary arteries in adults have an adaptive layer of fibroelastic thickening between the endothelium and internal elastic lamina [105].

D. A. H. Neil

Atherosclerosis is focal and eccentric with destruction of the internal elastic lamina and involves the epicardial coronary arteries often with a lipid core and calcification (see Fig. 16). Graft vascular disease is diffuse and concentric, often most prominent in the intramyocardial arteries, with only focal disruption to the internal elastic lamina and rarely contains lipid or calcification (Fig. 17) [92–94, 106]. However, there is some evidence that progression of pre-existing atherosclerotic plaques may also be accelerated. Graft vasculopathy initially is loose and has a thin rim (Fig. 17a), although with time it becomes thicker and more fibromuscular (Fig. 17b). There may be associated acute vascular rejection component (Fig. 17c, d) in which there is an

Fig. 15 (a) Low power Elastic van Giesson (EVG) stained section showing the normal adaptive layer of concentric fibrointimal thickening in coronary arteries. (b) Higher power view showing the internal elastic lamina (arrow) with adaptive fibroelastic intimal thickening (EVG)

Fig. 16 (a) Low power section showing naturally occurring atherosclerosis with eccentric plaques, lipid cores (∗) and destruction of the underlying internal elastic lamina. An intact area of internal elastic

lamina (arrow) is seen (EVG). (b) Higher power showing lipid core (∗) and destruction of the internal elastic lamina with fragments of the internal elastic lamina (arrow) still visible (EVG)

Transplant Pathology

199

Fig. 17 (a) Low power of a relatively early graft vasculopathy with loose intimal thickening and intact internal elastic lamina (EVG). (b) More advanced graft vasculopathy with marked concentric narrowing of the lumen by fibromuscular intimal thickening (EVG). (c)

Haematoxylin and eosin stained section of the same vessel as in Fig.  17b showing that there is transmural inflammation “acute on chronic vascular rejection”. (d) Acute vascular rejection with intimal inflammation (H&E)

intimal inflammatory cell infiltrate, sometimes with transmural inflammation. In this situation a rejection component can be implied in the pathogenesis and the term “acute on chronic vascular rejection” may be used. However, other non-“rejection” causes are still possible.

mismatch between donor and recipient, particularly if there is an element of pulmonary hypertension in the recipient and if the right heart is unable to pump against these pressures. The findings at autopsy can be a constellation of these different injuries, but often a mixture of contraction band necrosis and coagulative/ischaemic necrosis are often seen (Fig. 18). Contraction band necrosis is related to hypercontraction and occurs during brain death, particularly when there is a rapid marked increase in intracranial pressure such as in closed head injuries [96, 109, 110], and as part of the reperfusion injury [111, 112]. Coagulative necrosis occurs during the various periods of ischaemia in the donor and during storage [96, 109, 110], and mainly stain with C9 [23–25]. Cells with contraction band necrosis can secondarily progress to C9 positive necrosis [113], and so if the patient survives several days there might be some C9 positive cells with contraction bands still apparent (Fig. 19).

17

Primary Graft Failure

Primary graft failure is a term used when the transplanted heart never functioned without medical support and results in patient death or the need for an urgent re-transplant. The underlying problem is peri-transplant injury which encompasses injury related to hypotension in the donor, brain death [96], preservation injury (hypothermic ischaemia in preservation solutions) and reperfusion injury related to free radicals [103, 107, 108]. This can be exacerbated by a size

200

Fig. 18 (a) Contraction band necrosis with transverse contraction bands seen across many myocytes (H&E). (b) An ischaemic necrotic fibre stains with C9 whilst adjacent fibres with contraction band necrosis do not stain. (c) Coagulative necrosis of myocytes with granular

D. A. H. Neil

cytoplasm and a moth eaten appearance (∗). Cross striations can be seen in normal myocytes (arrows) (H&E). (d) C9 stains multiple necrotic fibres undergoing coagulative necrosis

18

Fig. 19  A myocyte with contraction bands may become C9 positive with time

 ampling of the Explanted or Post-­ S mortem Cardiac Transplant

Cardiac transplants either obtained at post-mortem or explanted at the time of re-transplantation should be examined diligently, similarly to a native heart in cases of sudden death [114–117] with the addition of an assessment of the anastomoses for evidence of dehiscence or stenosis. The usual anastomoses are the aorta, pulmonary artery, superior and inferior vena cava and left atrium (patch/cuff containing all four pulmonary veins). If the transplant is more than approximately 1 month old, there is usually fibrous obliteration of the pericardial space making it difficult to remove, which complicates identification and examination of the coronary arteries. A suture in the apex is usual from venting air at the end of the procedure and there is often also another

Transplant Pathology

suture in the right auricular appendage from cardiopulmonary bypass. The coronary artery walls should be assessed macroscopically for discrete lesions of atherosclerosis. Inspection for diffuse thickening related to graft vascular disease should follow, with samples being taken throughout the length looking for both types of lesions. Areas of thrombosis should also be noted and sampled. As with autopsy heart protocols, complete transverse sections through the ventricles from the apex to the midventricular level are made and focal lesions identified. Any discrete lesions should be sampled, together with a minimum of five samples from the transverse section at mid-ventricular level. Macroscopic photographs of the heart whole; from the stages during dissection of any focal lesion/s and those taken at the mid-ventricular section may be of particular use for case review with clinicians. Ideally, all cases should be managed by, or referred to, a specialist cardiac pathologist with transplant experience. The histological assessment will follow the above standards for cardiac graft histology review.

References 1. Capoccia M.  Mechanical circulatory support for advanced heart failure: are we about to witness a new “gold standard”? J Cardiovasc Dev Dis. 2016;3:35. 2. Phan K, Huo YR, Zhao DF, Yan TD, Tchantchaleishvili V.  Ventricular recovery and pump explantation in patients supported by left ventricular assist devices: a systematic review. ASAIO J. 2016;62:219–31. 3. Khush K, Cherikh WS, Chambers DC, et  al. The International Thoracic Organ Transplant Registry of the International Society for Heart and Lung Transplantation: thirty-fifth adult heart transplantation report-2018; focus theme: multiorgan transplantation. J Heart Lung Transplant. 2018;37:1155–68. 4. Hogg R, Rushton S, Cardiothoracic Advisory Group Clinical Audit Group. Annual report on cardiothoracic organ transplantation 2017-18. NHSBT and NHSE [2017-18]. Watford: NHSBT; 2018. p. 1–150. 5. Hamour IM, Khaghani A, Kanagala PK, Mitchell AG, Banner NR.  Current outcome of heart transplantation: a 10-year single centre perspective and review. QJM. 2010;104:335–43. 6. Crespo-Leiro MG, Barge-Caballero E, Marzoa-Rivas R, Paniagua-­ Martin MJ.  Heart transplantation. Curr Opin Organ Transplant. 2010;15:633–8. 7. Tan CD, Baldwin WM III, Rodriguez ER. Update on cardiac transplantation pathology. Arch Pathol Lab Med. 2007;131:1169–91. 8. Billingham ME. Some recent advances in cardiac pathology. Hum Pathol. 1979;10:367–86. 9. Billingham ME, Cary NR, Hammond ME, et al. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: heart rejection study group. The International Society for Heart Transplantation. J Heart Transplant. 1990;9:587–93. 10. Stewart S, Winters GL, Fishbein MC, et  al. Revision of the 1990 working formulation for the standardization of nomenclature in the diagnosis of heart rejection. J Heart Lung Transplant. 2005;24:1710–20.

201 11. Spiegelhalter DJ, Stovin PG. An analysis of repeated biopsies following cardiac transplantation. Stat Med. 1983;2:33–40. 12. Sharples LD, Cary NR, Large SR, Wallwork J.  Error rates with which endomyocardial biopsy specimens are graded for rejection after cardiac transplantation. Am J Cardiol. 1992;70:527–30. 13. Kobashigawa J, Crespo-Leiro MG, Ensminger SM, et al. Report from a consensus conference on antibody-mediated rejection in heart transplantation. J Heart Lung Transplant. 2011;30:252–69. 14. Berry G, Angelini A, Burke M, et al. The ISHLT Working formulation for pathological diagnosis of antibody-mediated rejection in heart transplantation: evolution and current status (2005-2011). J Heart Lung Transplant. 2011;30:601. 15. Behr TM, Feucht HE, Richter K, et al. Detection of humoral rejection in human cardiac allografts by assessing the capillary deposition of complement fragment C4d in endomyocardial biopsies. J Heart Lung Transplant. 1999;18:904–12. 16. Feucht HE, Schneeberger H, Hillebrand G, et al. Capillary deposition of C4d complement fragment and early renal graft loss. Kidney Int. 1993;43:1333–8. 17. Rodriguez ER, Skojec DV, Tan CD, et  al. Antibody-mediated rejection in human cardiac allografts: evaluation of immunoglobulins and complement activation products C4d and C3d as markers. Am J Transplant. 2005;5:2778–85. 18. Reed EF, Demetris AJ, Hammond E, et  al. Acute antibody-­ mediated rejection of cardiac transplants. J Heart Lung Transplant. 2006;25:153–9. 19. Stewart S, Cary NRB. The pathology of heart and lung transplantation. Curr Diagn Pathol. 1996;3:69–79. 20. Suvarna SK, Kennedy A, Ciulli F, Locke TJ. Revision of the 1990 working formulation for cardiac allograft rejection: the Sheffield experience. Heart. 1998;79:432–6. 21. Maleszewski JJ, Kucirka LM, Segev DL, Halushka MK. Survey of current practice related to grading of rejection in cardiac transplant recipients in North America. Cardiovasc Pathol. 2011;20:261. 22. Hook S, Caple JF, McMahon JT, Myles JL, Ratliff NB. Comparison of myocardial cell injury in acute cellular rejection versus acute vascular rejection in cyclosporine-treated heart transplants. J Heart Lung Transplant. 1995;14:351–8. 23. Howie AJ. C9 immunohistology in detection of myocardial infarction. J Pathol. 2001;193:421. 24. Robert-Offerman SR, Leers MP, van Suylen RJ, Nap M, Daemen MJ, Theunissen PH. Evaluation of the membrane attack complex of complement for the detection of a recent myocardial infarction in man. J Pathol. 2000;191:48–53. 25. Yasojima K, Schwab C, McGeer EG, McGeer PL. Human heart generates complement proteins that are upregulated and activated after myocardial infarction. Circ Res. 1998;83:860–9. 26. Fishbein MC, Kobashigawa J. Biopsy-negative cardiac transplant rejection: etiology, diagnosis, and therapy. Curr Opin Cardiol. 2004;19:166–9. 27. Hammond EH, Hansen JK, Spencer LS, Jensen A, Yowell RL.  Immunofluorescence of endomyocardial biopsy specimens: methods and interpretation. J Heart Lung Transplant. 1993;12:S113–24. 28. Hammond EH, Yowell RL, Nunoda S, et al. Vascular (humoral) rejection in heart transplantation: pathologic observations and clinical implications. J Heart Transplant. 1989;8:430–43. 29. Olsen SL, Wagoner LE, Hammond EH, et al. Vascular rejection in heart transplantation: clinical correlation, treatment options, and future considerations. J Heart Lung Transplant. 1993;12:S135–42. 30. Bonnaud EN, Lewis NP, Masek MA, Billingham ME. Reliability and usefulness of immunofluorescence in heart transplantation. J Heart Lung Transplant. 1995;14:163–71. 31. Crespo M, Pascual M, Tolkoff-Rubin N, et  al. Acute humoral rejection in renal allograft recipients: I. Incidence, serology and clinical characteristics. Transplantation. 2001;71:652–8.

202 32. Chantranuwat C, Qiao JH, Kobashigawa J, Hong L, Shintaku P, Fishbein MC.  Immunoperoxidase staining for C4d on paraffin-­ embedded tissue in cardiac allograft endomyocardial biopsies: comparison to frozen tissue immunofluorescence. Appl Immunohistochem Mol Morphol. 2004;12:166–71. 33. Miller DV, Roden AC, Gamez JD, Tazelaar HD. Detection of C4d deposition in cardiac allografts: a comparative study of immunofluorescence and immunoperoxidase methods. Arch Pathol Lab Med. 2010;134:1679–84. 34. Lones MA, Czer LS, Trento A, Harasty D, Miller JM, Fishbein MC.  Clinical-pathologic features of humoral rejection in cardiac allografts: a study in 81 consecutive patients. J Heart Lung Transplant. 1995;14:151–62. 35. Ratliff NB, McMahon JT.  Activation of intravascular macrophages within myocardial small vessels is a feature of acute vascular rejection in human heart transplants. J Heart Lung Transplant. 1995;14:338–45. 36. Takemoto SK, Zeevi A, Feng S, et  al. National conference to assess antibody-mediated rejection in solid organ transplantation. Am J Transplant. 2004;4:1033–41. 37. Wu GW, Kobashigawa JA, Fishbein MC, et  al. Asymptomatic antibody-mediated rejection after heart transplantation predicts poor outcomes. J Heart Lung Transplant. 2009;28:417–22. 38. Tan CD, Sokos GG, Pidwell DJ, et  al. Correlation of donor-­ specific antibodies, complement and its regulators with graft dysfunction in cardiac antibody-mediated rejection. Am J Transplant. 2009;9:2075–84. 39. Kfoury AG, Hammond ME, Snow GL, et  al. Cardiovascular mortality among heart transplant recipients with asymptomatic antibody-­mediated or stable mixed cellular and antibody-­mediated rejection. J Heart Lung Transplant. 2009;28:781–4. 40. Berry GJ, Burke MM, Andersen C, et al. The 2013 International Society for Heart and Lung Transplantation Working Formulation for the standardization of nomenclature in the pathologic diagnosis of antibody-mediated rejection in heart transplantation. J Heart Lung Transplant. 2013;32:1147–62. 41. Forbes RD, Rowan RA, Billingham ME. Endocardial infiltrates in human heart transplants: a serial biopsy analysis comparing four immunosuppression protocols. Hum Pathol. 1990;21:850–5. 42. Gajjar NA, Kobashigawa JA, Laks H, Espejo-Vassilakis M, Fishbein MC. FK506 vs. cyclosporin. Pathologic findings in 1067 endomyocardial biopsies. Cardiovasc Pathol. 2003;12:73–6. 43. Barone JH, Fishbein MC, Czer LS, Blanche C, Trento A, Luthringer DJ.  Absence of endocardial lymphoid infiltrates (Quilty lesions) in nonheart transplant recipients treated with cyclosporine. J Heart Lung Transplant. 1997;16:600–3. 44. Sattar HA, Husain AN, Kim AY, Krausz T.  The presence of a CD21+ follicular dendritic cell network distinguishes invasive Quilty lesions from cardiac acute cellular rejection. Am J Surg Pathol. 2006;30:1008–13. 45. Di Carlo E, D’Antuono T, Contento S, Di NM, Ballone E, Sorrentino C. Quilty effect has the features of lymphoid neogenesis and shares CXCL13-CXCR5 pathway with recurrent acute cardiac rejections. Am J Transplant. 2007;7:201–10. 46. Marboe CC, Billingham M, Eisen H, et al. Nodular endocardial infiltrates (Quilty lesions) cause significant variability in diagnosis of ISHLT Grade 2 and 3A rejection in cardiac allograft recipients. J Heart Lung Transplant. 2005;24:S219–26. 47. Hiemann NE, Knosalla C, Wellnhofer E, Lehmkuhl HB, Hetzer R, Meyer R. Quilty in biopsy is associated with poor prognosis after heart transplantation. Transpl Immunol. 2008;19:209–14. 48. Hiemann NE, Knosalla C, Wellnhofer E, Lehmkuhl HB, Hetzer R, Meyer R.  Quilty indicates increased risk for microvasculopathy and poor survival after heart transplantation. J Heart Lung Transplant. 2008;27:289–96.

D. A. H. Neil 49. Chantranuwat C, Blakey JD, Kobashigawa JA, et  al. Sudden, unexpected death in cardiac transplant recipients: an autopsy study. J Heart Lung Transplant. 2004;23:683–9. 50. Yamani MH, Ratliff NB, Starling RC, et  al. Quilty lesions are associated with increased expression of vitronectin receptor (alphavbeta3) and subsequent development of coronary vasculopathy. J Heart Lung Transplant. 2003;22:687–90. 51. Zakliczynski M, Nozynski J, Konecka-Mrowka D, et  al. Quilty effect correlates with biopsy-proven acute cellular rejection but does not predict transplanted heart coronary artery vasculopathy. J Heart Lung Transplant. 2009;28:255–9. 52. Sibley RK, Olivari MT, Ring WS, Bolman RM. Endomyocardial biopsy in the cardiac allograft recipient. A review of 570 biopsies. Ann Surg. 1986;203:177–87. 53. Connor RC. Heart damage associated with intracranial lesions. Br Med J. 1968;3:29–31. 54. Connor RC. Focal myocytolysis and fuchsinophilic degeneration of the myocardium of patients dying with various brain lesions. Ann N Y Acad Sci. 1969;156:261–70. 55. Connor RC. Myocardial damage secondary to brain lesions. Am Heart J. 1969;78:145–8. 56. Connor RC.  Fuchsinophilic degeneration of myocardium in patients with intracranial lesions. Br Heart J. 1970;32:81–4. 57. Haddad F, Deuse T, Pham M, et  al. Changing trends in infectious disease in heart transplantation. J Heart Lung Transplant. 2010;29:306–15. 58. Breinholt JP, Moulik M, Dreyer WJ, et  al. Viral epidemiologic shift in inflammatory heart disease: the increasing involvement of parvovirus B19 in the myocardium of pediatric cardiac transplant patients. J Heart Lung Transplant. 2010;29:739–46. 59. Moulik M, Breinholt JP, Dreyer WJ, et al. Viral endomyocardial infection is an independent predictor and potentially treatable risk factor for graft loss and coronary vasculopathy in pediatric cardiac transplant recipients. J Am Coll Cardiol. 2010;56:582–92. 60. Kotton CN. Update on infectious diseases in pediatric solid organ transplantation. Curr Opin Organ Transplant. 2008;13:500–5. 61. Casadei D.  Chagas’ disease and solid organ transplantation. Transplant Proc. 2010;42:3354–9. 62. Kotton CN, Lattes R. Parasitic infections in solid organ transplant recipients. Am J Transplant. 2009;9(Suppl 4):S234–51. 63. Derouin F, Pelloux H. Prevention of toxoplasmosis in transplant patients. Clin Microbiol Infect. 2008;14:1089–101. 64. Sanchez MA, Debrunner M, Cox E, Caldwell R. Acquired toxoplasmosis after orthotopic heart transplantation in a sulfonamide-­ allergic patient. Pediatr Cardiol. 2011;32:91–3. 65. Parker A, Bowles K, Bradley JA, et al. Diagnosis of post-transplant lymphoproliferative disorder in solid organ transplant recipients BCSH and BTS Guidelines. Br J Haematol. 2010;149:675–92. 66. Tsao L, Hsi ED.  The clinicopathologic spectrum of posttransplantation lymphoproliferative disorders. Arch Pathol Lab Med. 2007;131:1209–18. 67. Grivas PD.  Post-transplantation lymphoproliferative disorder (PTLD) twenty years after heart transplantation: a case report and review of the literature. Med Oncol. 2011;28:829. 68. Manlhiot C, Pollock-Barziv SM, Holmes C, et al. Post-transplant lymphoproliferative disorder in pediatric heart transplant recipients. J Heart Lung Transplant. 2010;29:648–57. 69. Mucha K, Foroncewicz B, Ziarkiewicz-Wroblewska B, Krawczyk M, Lerut J, Paczek L.  Post-transplant lymphoproliferative disorder in view of the new WHO classification: a more rational approach to a protean disease? Nephrol Dial Transplant. 2010;25:2089–98. 70. Swerdlow SH, Campo E, Harris NL, et al. WHO classification of tumours of haematopoietic and lymphoid tissues. Geneva: World Health Organisation; 2008.

Transplant Pathology 71. Turillazzi E, Pennella A, Di GG, Neri M, Fineschi V.  Post-­ transplant lymphoproliferative disorder in the heart late after heterotopic transplantation: autopsy findings. J Heart Lung Transplant. 2010;29:904–6. 72. Luk A, Metawee M, Ahn E, Gustafsson F, Ross H, Butany J. Do clinical diagnoses correlate with pathological diagnoses in cardiac transplant patients? The importance of endomyocardial biopsy. Can J Cardiol. 2009;25:e48–54. 73. Neil D.  Recurrent and de novo disease in kidney, heart, lung, pancreas and intestinal transplants. Curr Opin Organ Transplant. 2006;11:289–95. 74. Alloni A, Pellegrini C, Ragni T, et  al. Heart transplantation in patients with amyloidosis: single-center experience. Transplant Proc. 2004;36:643–4. 75. Dubrey SW, Burke MM, Hawkins PN, Banner NR. Cardiac transplantation for amyloid heart disease: the United Kingdom experience. J Heart Lung Transplant. 2004;23:1142–53. 76. Kristen AV, Meyer FJ, Perz JB, et al. Risk stratification in cardiac amyloidosis: novel approaches. Transplantation. 2005;80:S151–5. 77. Kristen AV, Sack FU, Schonland SO, et  al. Staged heart transplantation and chemotherapy as a treatment option in patients with severe cardiac light-chain amyloidosis. Eur J Heart Fail. 2009;11:1014–20. 78. Mignot A, Varnous S, Redonnet M, et al. Heart transplantation in systemic (AL) amyloidosis: a retrospective study of eight French patients. Arch Cardiovasc Dis. 2008;101:523–32. 79. Sattianayagam PT, Gibbs SD, Pinney JH, et al. Solid organ transplantation in AL amyloidosis. Am J Transplant. 2010;10:2124–31. 80. Moloney ED, Egan JJ, Kelly P, Wood AE, Cooper LT Jr. Transplantation for myocarditis: a controversy revisited. J Heart Lung Transplant. 2005;24:1103–10. 81. Luk A, Lee A, Ahn E, Soor GS, Ross HJ, Butany J. Cardiac sarcoidosis: recurrent disease in a heart transplant patient following pulmonary tuberculosis infection. Can J Cardiol. 2010;26:e273–5. 82. Oni AA, Hershberger RE, Norman DJ, et al. Recurrence of sarcoidosis in a cardiac allograft: control with augmented corticosteroids. J Heart Lung Transplant. 1992;11:367–9. 83. Strecker T, Zimmermann I, Wiest GH.  Pulmonary and cardiac recurrence of sarcoidosis in a heart transplant recipient. Dtsch Med Wochenschr. 2007;132:1159–62. 84. Yager JE, Hernandez AF, Steenbergen C, et  al. Recurrence of cardiac sarcoidosis in a heart transplant recipient. J Heart Lung Transplant. 2005;24:1988–90. 85. Adam B, Mengel M. Transplant biopsy beyond light microscopy. BMC Nephrol. 2015;16:132. 86. Halloran PF, Potena L, Van Huyen JD, et  al. Building a tissue-­ based molecular diagnostic system in heart transplant rejection: the heart molecular microscope diagnostic (MMDx) System. J Heart Lung Transplant. 2017;36:1192–200. 87. Loupy A, Duong Van Huyen JP, Hidalgo L, et  al. Gene expression profiling for the identification and classification of antibody-­ mediated heart rejection. Circulation. 2017;135:917–35. 88. Halloran PF, Reeve J, Aliabadi AZ, et  al. Exploring the cardiac response to injury in heart transplant biopsies. JCI Insight. 2018;3:123674. 89. Afzali B, Chapman E, Racape M, et al. Molecular assessment of microcirculation injury in formalin-fixed human cardiac allograft biopsies with antibody-mediated rejection. Am J Transplant. 2017;17:496–505. 90. Halloran PF, Venner JM, Madill-Thomsen KS, et  al. Review: the transcripts associated with organ allograft rejection. Am J Transplant. 2018;18:785–95. 91. Vassalli G, Gallino A, Weis M, et al. Alloimmunity and nonimmunologic risk factors in cardiac allograft vasculopathy. Eur Heart J. 2003;24:1180–8.

203 92. Demetris AJ, Zerbe T, Banner B.  Morphology of solid organ allograft arteriopathy: identification of proliferating intimal cell populations. Transplant Proc. 1989;21:3667–9. 93. Billingham ME.  Cardiac transplant atherosclerosis. Transplant Proc. 1987;19:19–25. 94. Billingham ME.  Graft coronary disease: the lesions and the patients. Transplant Proc. 1989;21:3665–6. 95. Eisen HJ. Pathogenesis and management of cardiac allograft vasculopathy. Curr Opin Organ Transplant. 2004;9:448–52. 96. Pratschke J, Volk HD. Brain death-associated ischemia and reperfusion injury. Curr Opin Organ Transplant. 2004;9:153–8. 97. Mehra MR, Uber PA, Ventura HO, Scott RL, Park MH.  The impact of mode of donor brain death on cardiac allograft vasculopathy: an intravascular ultrasound study. J Am Coll Cardiol. 2004;43:806–10. 98. Cohen O, De La Zerda DJ, Beygui R, Hekmat D, Laks H. Donor brain death mechanisms and outcomes after heart transplantation. Transplant Proc. 2007;39:2964–9. 99. Neil DA, Lynch SV, Hardie IR, Effeney DJ. Cold storage preservation and warm ischaemic injury to isolated arterial segments: endothelial cell injury. Am J Transplant. 2002;2:400–9. 100. Neil DA, Maguire SH, Walsh M, Lynch SV, Hardie IR, Effeney DJ. Progression of changes in arteries following cold storage preservation in UW and Collins solution in a syngeneic aortic transplant model. Transplant Proc. 1997;29:2561–2. 101. Hiemann NE, Musci M, Wellnhofer E, Meyer R, Hetzer R. Light microscopic biopsy findings after heart transplantation and possible links to development of graft vessel disease. Transplant Proc. 1999;31:149–51. 102. Koch A, Bingold TM, Oberlander J, et  al. Capillary endothelia and cardiomyocytes differ in vulnerability to ischemia/reperfusion during clinical heart transplantation. Eur J Cardiothorac Surg. 2001;20:996–1001. 103. Kupiec-Weglinski JW. Ischemia and reperfusion injury. Curr Opin Organ Transplant. 2004;9:130–1. 104. Logani S, Saltzman HE, Kurnik P, Eisen HJ, Ledley GS.  Clinical utility of intravascular ultrasound in the assessment of coronary allograft vasculopathy: a review. J Interv Cardiol. 2011;24:9–14. 105. Stary HC, Blankenhorn DH, Chandler AB, et al. A definition of the intima of human arteries and of its atherosclerosis-prone regions. A report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation. 1992;85:391–405. 106. Gao SZ, Schroeder JS, Alderman EL, et al. Clinical and laboratory correlates of accelerated coronary artery disease in the cardiac transplant patient. Circulation. 1987;76:V56–61. 107. Becker LB. New concepts in reactive oxygen species and cardiovascular reperfusion physiology. Cardiovasc Res. 2004;61:461–70. 108. Turer AT, Hill JA.  Pathogenesis of myocardial ischemia-­ reperfusion injury and rationale for therapy. Am J Cardiol. 2010;106:360–8. 109. Nijboer WN, Schuurs TA, Van der Hoeven JAB, Ploeg RJ. Effect of brain death and donor treatment on organ inflammatory response and donor organ viability. Curr Opin Organ Transplant. 2004;9:110–5. 110. Van der Hoeven JAB, Ploeg RJ. Effects of brain death on donor organ viability. Curr Opin Organ Transplant. 2001;6:75–82. 111. Rodriguez-Sinovas A, Abdallah Y, Piper HM, Garcia-Dorado D. Reperfusion injury as a therapeutic challenge in patients with acute myocardial infarction. Heart Fail Rev. 2007;12:207–16. 112. Inserte J, Garcia-Dorado D, Ruiz-Meana M, et  al. Effect of inhibition of Na(+)/Ca(2+) exchanger at the time of myocardial reperfusion on hypercontracture and cell death. Cardiovasc Res. 2002;55:739–48.

204 113. Miller DL, Li P, Dou C, Armstrong WF, Gordon D.  Evans blue staining of cardiomyocytes induced by myocardial contrast echocardiography in rats: evidence for necrosis instead of apoptosis. Ultrasound Med Biol. 2007;33: 1988–96. 114. Boyle JJ, Rassl DM, Neil D, Suvarna K, Doran H. Heart dissection  - explants post cardiac transplantation. In: Boyle JJ, editor. Tissue Pathways for cardiovascular pathology. London: Royal College of Pathologists; 2008. p. 20–1.

D. A. H. Neil 115. Boyle JJ, Rassl DM, Neil D, Suvarna K, Doran H. Autopsy cardiac dissection. In: Boyle JJ, editor. Tissue pathways for cardiovascular pathology. London: Royal College of Pathologists; 2008. p. 22–7. 116. Basso C, Burke M, Fornes P, et al. Guidelines for autopsy investigation of sudden cardiac death. Virchows Arch. 2008;452:11–8. 117. Suvarna SK.  National guidelines for adult autopsy cardiac dissection and diagnosis--are they achievable? A personal view. Histopathology. 2008;53:97–112.

Cardiomyopathies Clare R. Bunning and S. Kim Suvarna

1

Introduction

Cardiomyopathies are disorders of the myocardium resulting in structural and functional abnormality, unexplained by coronary artery narrowing or abnormal ventricular loading. They have a variable prevalence and may be misdiagnosed under other cardiac conditions. They most commonly demonstrate an autosomal dominant pattern of inheritance, although other patterns of inheritance such as autosomal recessive, X linked recessive and mitochondrial patterns are seen. Acquired forms of cardiomyopathy are also seen. The most recently proposed classification system is the 2013 WHF-MOGE(S) system which proposes a descriptive genotype phenotype classification system. The parameters are: morphofunctional phenotype (M), organ(s) involvement (O), genetic inheritance pattern (G), aetiological annotation (E) including genetic defect or underlying disease/substrate, and the functional status (S) of the disease [1]. The aim of this system is to address the main attributes of cardiomyopathy in lieu of a future genetic classification system. This chapter will focus primarily on the main morphofunctional phenotypes of cardiomyopathy. These are dilated cardiomyopathy (DCM), isolated left ventricular non compaction cardiomyopathy (LVNC), hypertrophic cardiomyopathy (HCM), arrhythmogenic right ventricular cardiomyopathy (ARVC) and restrictive cardiomyopathy (RCM). It will address how the pathologist can make an accurate diagnosis by performing a detailed macroscopic and microscopic examination of the heart and appreciate the genetic factors associated with these disorders that apply to surviving family members.

C. R. Bunning Royal Hallamshire Hospital, Sheffield, UK e-mail: [email protected] S. K. Suvarna (*) Sheffield Teaching Hospitals, Royal Hallamshire Hospital Sheffield Teaching Hospitals Sheffield, South Yorkshire, UK e-mail: [email protected]

2

Dilated Cardiomyopathy (DCM)

Dilated cardiomyopathy is the most common pathology creating congestive heart failure in the adult population worldwide. It is best regarded as a pattern of disease which can have a multitude of genetic, acquired or idiopathic aetiologies. It manifests as ventricular dilatation, with contractile dysfunction of the left, or both ventricles (Fig. 1). Clinically, it is associated with progressive congestive heart failure, being diagnosed in life. Genetic causes of DCM involve defects in genes encoding myocyte sarcomere, cytoskeleton, nuclear membrane or mitochondrial proteins [2]. Titin (TTN) is the largest human protein and is required for sarcomere assembly, function and stability. At a genetic level mutations in TTN predominate with truncating mutations in TTN accounting for up to 25% of primary DCM cases [3]. These truncating mutations in TTN are also implicated in peripartum-related cardiomyopathy [4] and may be associated with other acquired forms of DCM. Truncating mutations are present in 1.5% of the population [5]. Some sarcomere encoding genes such as MYH7 and TNNT2 can cause DCM, although these genes may also be mutated in HCM. These are less common causes than TTN mutations and alter different residues to the HCM mutations [5]. DCM may be associated with conduction defects and arrhythmias, particularly with mutations in the LMNA gene. This encodes the nuclear membrane protein lamin A/C, being involved in approximately 6–8% of DCM cases [6]. Fatal cardiac arrhythmia occurs in approximately half of LMNA cardiomyopathy genotypes and can occur before DCM symptoms [7], meaning it will be an autopsy diagnosis. Mutations in the ABCC9 gene encoding a cardiac potassium channel subunit have likewise been associated with DCM and atrial fibrillation [8]. Mutations in cytoskeletal proteins such as cardiac actin, desmin, beta-sarcoglycan, delta-sarcoglycan, dystrophin and vinculin affect force transmission during cell contraction. Desmin associated mutations are implicated in less than 1% of DCM cases [9], but also affect force transmission. They

© Springer Nature Switzerland AG 2019 S. K. Suvarna (ed.), Cardiac Pathology, https://doi.org/10.1007/978-3-030-24560-3_11

205

206

C. R. Bunning and S. K. Suvarna

Fig. 1  Dilated cardiomyopathy in a case of prior myocarditis. There is generalised ventricular dilatation (bar marker 3  cm). This pattern of chamber enlargement is the standard format seen in many toxic, inflammatory and degenerative disorders as an end point Fig. 3  A case of sudden death in myotonic dystrophy. The macroscopy was minimally dilated, but there was histological diffuse fibrous and fatty tissue replacement seen across the ventricular tissues, more easily discerned on the left side (haematoxylin and eosin stain)

Fig. 2  Left ventricle tissues in a case of Duchenne-type muscular dystrophy. The trichrome highlights the scarred/fibrous element (green) as compared with the viable muscular tissue (red). Clearly electrical depolarisation and cardiac contractility will be degraded by this amount of fibrous tissue replacement

can also cause an accumulation of misfolded proteins resulting in cardiac cell toxicity, especially if mutations of the alpha-beta-crystallin chaperone protein are present [10]. Dystrophin mutations are X linked recessive and associated with dystrophinopathies (Figs. 2 and 3). Similarly mitochondrial TAZ gene mutation related DCM (Barth Syndrome) also shows X linked recessive inheritance [11]. Sarcoglycan mutations exhibit an autosomal recessive or dominant pattern of inheritance and are associated with muscular dystrophy. One of the more recent developments, as a result of genetic linkage and gene mapping studies, is the identification of heterozygous missense mutation of a spliceosome protein (pre-mRNA splicing is catalysed by a spliceosome ribonucleoprotein complex composed of five SnRNPs and proteins), the ribonucleic acid binding motif protein 20

(RBM20) in DCM [12]. This has been found to regulate premRNA splicing of multiple genes, some of which affect TTN [13, 14]. Acquired causes of DCM account for the majority (about two thirds) of congestive cardiac failure cases seen in life, having no overt gene linkage. In such cases, the DCM phenotype may reflect coronary artery disease, valvular disease, hypertension, post-viral myocarditis and in association with Lyme disease, HIV infection or Chagas disease. Likewise, the DCM phenotype can be seen in peripartum/postpartum states [15] and infiltrative diseases of the heart (e.g., sarcoidosis, amyloidosis and haemochromatosis). It can also occur as a consequence of toxins such as alcohol, cocaine and following chemotherapy (in particular, doxorubicin). Finally, it can also occur in the context of other disease processes such as end-stage renal failure, obstructive sleep apnoea, and autoimmune conditions such as systemic lupus erythematosus, celiac disease and various endocrine dysfunction. Alcoholic cardiomyopathy is perhaps the most common cause of acquired DCM, often underappreciated by clinician and pathologist. The adverse effects of ethanol misuse on the cardiac muscle is believed to be the result of several mechanisms including apoptosis, oxidative stress and free radical generation, impaired mitochondrial function, accelerated protein catabolism and deranged fatty acid metabolism and transport [16]. The macroscopic and microscopic phenotypes of DCM are similar in the vast majority of cases, regardless of the underlying cause (Fig.  3). DCM is characterised by an increase in heart weight along with relative thinning of the left ventricular wall. There is enlargement of both ventricular cavities (Figs.  1 and 4). The normal trabecular pattern is partly flattened outwards/effaced and diffuse endocardial

Cardiomyopathies

Fig. 4  A mid-ventricular transverse slice of heart tissue in an autopsy case of dilated cardiomyopathy needs to be positioned carefully in order to fully appreciate the degree of ventricular dilatation (bar marker 1 cm)

207

Fig. 6  Photograph of an autopsy transverse slice of the heart at mid-­ septal level from a patient with familial HCM. There is disproportionate left ventricular hypertrophy affecting the interventricular septum. The right ventricle is also involved by the hypertrophic process

sparse lymphocytic infiltrate accompanies some cases, but is not specific and should not be automatically taken to suggest resolving myocarditis. Some histiocytes, associated with myocyte loss can be seen in some cases. As the macroscopic and microscopic features are similar irrespective of the aetiology, a detailed scrutiny of the history is required to determine the likely underlying cause. If acquired causes of DCM have been excluded, then it is important to consider the referral of first-degree relatives of potential familial DCM for cardiac screening.

3

Fig. 5  There is often a diffuse fibrosis involving the myocardium in cases of dilated cardiomyopathy. Such changes should not be confused with fibrosis from ischaemic heart disease or fatty/fibrous tissue replacement in cases of arrhythmogenic cardiomyopathy. The cardiac myocytes may be of variable size, although they are generally enlarged in the long axis

thickening may be seen. Mural thrombus due to non-laminar blood flow is occasionally present, surrounding the ventricular trabeculations and/or within the atrial appendages. DCM shows non-specific microscopic features and some, but not all, may be seen with each case. Classically, the myocytes are sometimes thinned, although many are of normal shape and size. There may be albeit nuclear hypertrophy and some pleomorphism. There may be loss of myofibrils resulting in myocyte pallor. There are varying degrees of either coarse, or diffuse, interstitial fibrosis within the ventricular myocardium depending on disease duration—chronic cases have greater levels of fibrosis (Figs. 3 and 5). A patchy and

Hypertrophic Cardiomyopathy (HCM)

Hypertrophic cardiomyopathy is a common cardiovascular disease with an estimated prevalence ranging from 1 in 200– 500 of the population. It has an autosomal dominant pattern of inheritance in most cases [17]. In adults, it classically shows non-dilated left ventricular hypertrophy (LVH) (Figs. 6 and 7) with asymmetric left ventricular myocardial thickness of more than 15  mm on imaging studies. At autopsy, it is associated with myocardial and myofibrillar disarray (Fig. 8) and an increase in interstitial fibrosis unexplained by other systemic or cardiac causes of LVH.  The adult phenotype is variable, some with concentric ventricular walls or only minimal thickening. It is an important cause of SCD in young adults, and in particular athletes, with sudden death most commonly occurring in asymptomatic adults under 35 years of age [18]. HCM is caused by mutations in the genes encoding sarcomeric proteins. The most frequently encountered mutations are in cardiac myosin binding protein C (MYBPC3), which result in mutated mRNAs and truncated proteins lacking the myosin and/or titin binding sites. These impair sarcomere

208

Fig. 7  Two transverse slices at mid- and low level across the ventricles in another case of hypertrophic cardiomyopathy showing the very pronounced left ventricle hypertrophy. The differential diagnosis in such cases would have to include hypertension and various infiltrative disorders

Fig. 8  Haematoxylin and eosin stained section showing florid myocyte disarray and fibrosis in familial HCM. Disarray is characterized by hypertrophic myocytes with enlarged and pleomorphic nuclei aligned at odd angles to one another

C. R. Bunning and S. K. Suvarna

structure and function. HCM mutations are also seen with myosin heavy chain 7 (MYH7) which encodes the beta-myosin heavy chain. These mutations account for approximately 50% of cases [19] with MYPBC3 mutations accounting for up to 40% of cases [20]. Genes for thin filaments, sarcomereassociated proteins, and genes encoding plasma membrane proteins such as Junctophilin 2, mitochondrial proteins and Z-discs have also been identified as less common causes of HCM. Sporadic non-familial cases of HCM may also arise as a result of de-novo mutations. A further cause of HCM has been identified resulting from mutations in the ‘four and a half’ LIM domain 1 (FHL1) with an X linked pattern of inheritance [21]. LIM domains mediate protein-protein interactions critical to cellular processes. In the heart FHL1 is associated with TTN with FHL1 upregulation seen in cardiac hypertrophy [22]. The clinical spectrum of HCM as shown by genotypephenotype correlation analyses is highly variable reflecting gene penetrance. Patients with more than one mutation may have a more severe form of the disease presenting with symptoms or SCD at an earlier age. Genes may be affected by mutations at various sites. Furthermore, different mutations in the same gene may be associated with different disease severity and/or prognosis. The same genotype in one individual may be expressed as a different phenotype in another, indicating both genetic and environmental factors contribute to the ultimate phenotype expressed in an ­individual [5]. Due to this heterogeneity HCM mutations are considered to be pathogenic if there is familial co-segregation of the HCM phenotype previously reported as a cause of HCM. The mutation must be absent from unrelated and ethnic matched controls. Likewise, protein structure and function must be detrimentally altered with amino acid change in a highly conserved protein region [23]. Highly pathogenic mutations, with early manifestation of LVH and an increased risk of SCD, are found with R403Q, R453C, G716R and R719W mutations resulting in malignant defects in MYH7. R92W mutation in the cardiac troponin T has also been associated with an increased risk of SCD [24]. One should be aware that cardiac hypertrophy may occur as a consequence of valvular heart disease, essential hypertension, non-compaction cardiomyopathy, athletic training, haemochromatosis and infiltrative cardiomyopathies (discussed below). It is therefore unwise to reach a diagnosis of HCM simply on macroscopic assessment. One should also be aware of the effects of agonal contraction. In such cases the systolic contraction results, in apparent increased left ventricular wall thickness, this can potentially lead to misdiagnosis of HCM. Cardiac mass correlated with body weight is required to give a more accurate indicator of genuine LVH, instead of measurements of left ventricular wall thickness in

Cardiomyopathies

isolation (see chapter “The Normal Adult Heart and Methods of Investigation”). However as stated HCM does not always present with marked ventricular hypertrophy. Indeed, isolated papillary muscle hypertrophy is thought to be another phenotypic variant. Cardiac troponin T mutations associated with SCD are also associated with minimal hypertrophy [18, 25]. HCM is characterised by several different morphological patterns which may be asymmetrical or symmetrical. The most common phenotype is asymmetrical septal hypertrophy with or without subaortic obstruction. Systolic anterior motion (SAM) of the mitral valve is seen on echocardiogram in approximately 30–60% of patients, with HCM thought to be due to septal hypertrophy causing outflow tract obstruction at high flow velocities. Mitral valvular apparatus abnormalities are also common in patients with HCM, which may also be a cause of SAM [26]. Endocardial fibrosis over the septum leads to a sub-aortic mitral impact lesion due to contact of the anterior cusp of the mitral valve with the hypertrophied basal septum (Fig. 9). SAM and this lesion may occur in HCM as well as in hypertrophied hearts due to hypertension, in hyperdynamic states or hypovolaemia [26]. Thus, progression of any HCM diagnosis requires wide-ranging consideration of the macroscopic phenotype in conjunction with histology and other data. The symmetrical form of HCM is characterised by concentric thickening of the left ventricle. Macroscopically it is indistinguishable from hypertrophy due to essential hypertension, aortic stenosis or indeed athletic training. Adequate histological sampling is therefore essential. In HCM, often the right ventricle may also be involved by the hypertrophic process [27]. Various morphological variants of HCM are seen. These comprise apical HCM, asymmetric posterior

Fig. 9  Photograph of a sub-aortic mitral impact lesion, which resembles a mirror-image of the anterior cusp of the mitral valve in a patient with familial HCM

209

left ventricular wall hypertrophy, mid-ventricular obstructive HCM with hypertrophy confined to the middle portion of the left ventricle and involvement of the papillary muscles [26]. A further sub-group of HCM patients with left ventricular apical aneurysms has also been observed [26]. Some HCM sub-types are of relevance since patients with mid-ventricular HCM and asymmetric left ventricular wall hypertrophy are generally severely symptomatic. In endstage HCM v­ entricular dilatation occurs due to myocyte loss with extensive replacement-type fibrosis. This may simulate DCM. The classic microscopic features of HCM are myocyte hypertrophy and disarray (Fig. 8), although many variations are recognised (Figs.  10, 11, and 12). Myocyte disarray occurs due to the loss of the normal parallel arrangement of

Fig. 10  A case of HCM with marked enlargement of the haphazard myocytes with enlarged nuclei

Fig. 11  Some cases of HCM show relatively mild disarray, even if the macroscopic features are characteristic

210

Fig. 12  Late stage HCM often has pronounced interstitial fibrosis. This may be associated with a dilated or restrictive cardiomyopathy

Fig. 13  Masson’s trichrome stain confirming the marked increase in interstitial collagen, which is an integral part of the disease process and a hallmark of familial HCM (Courtesy of Dr. S. Hughes)

myocytes within the myocardium. The myocytes may assume a cartwheel or herringbone pattern due to abnormal cell-to-cell contacts. There is often disruption of the myofibrillary architecture, with criss-crossing of the myofibrils, which can be best demonstrated by phosphotungstic acid-haematoxylin (PTAH) staining and/or electron microscopy. Coarse or fine interstitial fibrosis, or a combination of these patterns, due to increased amounts of collagen, may be seen depending on the chronicity of the disease. Special stains such as Masson’s trichrome (or other stain) can be used to stain collagen and permit assessment of fibrosis (Fig.  13). Fibrosis is a hallmark of HCM and is associated with an increased risk of arrhythmias and progression to heart failure.

C. R. Bunning and S. K. Suvarna

Fig. 14  The small intramural branches of coronary arteries exhibit luminal narrowing and medial hypertrophy (dysplasia) in HCM, contributing to myocardial ischaemia and myocyte loss leading to increased fibrosis (Courtesy of Dr. S. Hughes)

Provided an entire circumferential left ventricular slice of myocardium at mid-septal level is sampled in HCM cases (large blocks or standard multiple blocks), disarray will usually be evident in more than 20% of the myocardium in at least two blocks. Overdiagnosis of fibre disarray is possible in normal hearts, where the left and right ventricles interdigitate anteriorly and posteriorly (see chapter “The Normal Adult Heart and Methods of Investigation”). The myocytes are not usually parallel at these points in normal hearts, providing a risk of a misdiagnosis of HCM.  Mild degrees of fibre irregularity can also be seen in normal hearts around trabeculations, adjacent to blood vessels, near areas of fibrosis and where large muscle bundles converge. Abnormal myocardial vasculature is seen with HCM. The small intramural branches of coronary arteries often show ‘dysplasia’, which is characterized by luminal narrowing and medial hypertrophy (Fig. 14). These vascular changes likely contribute to ischaemia and myocyte loss with replacement fibrosis, although some of the changes may reflect abnormal myocardial contraction around the vessels. The vascular changes may contribute to electrical instability, resulting in arrhythmia and/or SCD.  The end-point effect may also explain the DCM phenotype of HCM culminating in congestive heart failure. In summary HCM is a common genetic disorder and should always be considered in cases of sudden death, particularly in young adults and those with a family history of cardiac disease. Accurate diagnosis is essential to identify those families who will need cardiac screening, mutation analysis, and genetic counseling. However, if recognised and treated, HCM may have a good prognosis, with patients being asymptomatic with a near normal life expectancy.

Cardiomyopathies

4

Arrhythmogenic Right Ventricular Cardiomyopathy/Arrhythmogenic Cardiomyopathy

Arrhythmogenic cardiomyopathy (AC) is a form of cardiomyopathy which characteristically affects the right ventricle (Arrhythmogenic right ventricle cardiomyopathy, ARVC), although left ventricle formats (Arrhythmogenic left ventricle cardiomyopathy, ALVC) are also known. AC has an e­ stimated prevalence of 1  in 1000 to 1  in 10,000 and a predominantly autosomal dominant pattern of inheritance [28]. It is an important cause of SCD in young people and athletes. ARVC is characterised by progressive myocardial loss and fibro-adipose replacement primarily affecting the right ventricle (Figs. 15, 16, and 17), although some co-involvement of the left ventricle is also often present. ALVC (where the disease is mainly left ventricular biased) has also been described with similar fatty tissue replacement and fibrous tissue replacement of the myocardium [29]. In AC, the fibrous and fatty changes tend to be centripetal, spreading from the epicardium inwards (Figs. 18 and 19). They may be seen macroscopically and histologically. There are at least 12 subtypes of ARVC (Fig. 20) with different genetic mutations in desmosomal and non-desmosomal proteins attributed to the different subtypes [30]. In life, the diagnosis of ARVC is based on the revised 2010 task force criteria that includes ventricular functional and structural changes, ECG abnormalities, arrhythmias, familial and genetic factors [31]. There is, however, currently no single diagnostic test for AC. The pathogenesis of AC is associated predominantly with mutations in desmosomal proteins, structures which provide cell:cell adhesion and mediators of intra- and intercellular signalling pathways (Fig.  20). These are localised to the

Fig. 15  A case of arrhythmogenic cardiomyopathy, biased to the right side (ARVC). This shows florid fatty change on the right ventricle cut surface. Some minor changes may be present on the left side, but this would need histology to confirm the disorder

211

intercalated disc in cardiac myocytes (see chapter “The Normal Adult Heart and Methods of Investigation”). The most common mutation observed affects plakophilin 2 (ARVC Type 9), a linker between desmoplakin and the cadherin tails [32, 33]. The incidence of this mutation within individuals with ARVC is between 10% and 45%. These individuals present earlier than other subtypes and arrhythmia free survival is low [32, 34]. Mutations in desmoplakin, desmoglein 2, desmocollin 2 and plakoglobin are desmosomal mutations seen in other ARVC subtypes, with desmoplakin being the second most common mutation found [32]. Desmoplakin (ARVC Type 8) mutations are associated with earlier left ventricular involvement. Recessive mutations cause Carvajal syndrome. Desmoglein 2

Fig. 16  Arrhythmogenic cardiomyopathy can show very focal fatty tissue replacement and one needs to be alert to patchy replacement of ventricular tissue both in the right, as well as the left, ventricle parenchyma. In this case, focal right ventricle fatty tissue replacement is indicated by the “f”

Fig. 17  The characteristic ARVC dilated right ventricle with a pronounced fatty wall is seen (Courtesy of Dr. C. Howitt)

212

Fig. 18  Masson trichrome histology demonstrating the right ventricle fibrous and fatty tissue replacement that has swept inwards from the epicardial (epi) aspect to the endocardial (endo) aspect of the ventricle chamber

Fig. 19  This photomicrograph shows myocardial substitution and replacement by fibro-adipose tissue involving the outer third of the posterior wall of the left ventricle in a patient with AC with predominant left ventricular involvement. The fibroadipose replacement is advancing from the epicardium (top of image) towards the endocardium

(ARVC Type 10) mutations are mainly associated with left ventricular involvement. Desmocollin 2 (ARVC Type 11) mutations are associated with woolly hair and mild palmoplantar keratoderma, in addition to ARVC if homozygous. Plakoglobin (ARVC Type 12) mutations, if recessive, cause Naxos disease and, if dominant, more typical ARVC [34]. Mutations in non-desmosomal genes are also attributed to rarer subtypes of ARVC. These mutations include genes which encode proteins such as phospholamban. This regu-

C. R. Bunning and S. K. Suvarna

lates calcium movement within cardiac myocytes and transmembrane protein 43 (ARVC Type 5, a highly lethal subtype) [35]. Transmembrane protein 43 may have a role in maintaining nuclear envelope structure and also has a response element for an adipogenic transcription factor. Mutations in the cardiac ryanodine receptor gene (ARVC Type 2), which encodes a sarcoplasmic reticulum calcium release channel protein, are associated with catecholaminergic polymorphic ventricular tachycardias without overt cardiac structural abnormality. However, areas of fibrofatty replacement of the sub-epicardial layer of the right ventricle may be seen on histology in this subtype [36]. Transforming growth factor beta-3 mutations (ARVC Type 1) are believed to increase myocardial fibrosis and modulate the expression of desmosomal genes [37]. Desmin mutations (ARVC Type 7) are seen in association with skeletal muscle myopathies. Titin mutations (ARVC type 4) and lamin A/C mutations are also associated with an AC phenotype. It is thought that the common pathway of pathology involves repeated mechanical stress causing myocyte detachment/damage and apoptosis. It has to be appreciated that mechanical integrity of the intercalated discs is also lost as a consequence of mutations. There may be accompanying inflammation and induction of adipogenic and fibrogenic genes in the stromal cells. Repair by fibrous and adipose tissue replacement occurs as a response to myocyte damage (Fig. 21). There is the loss of normal cell:cell depolarisation and interspaced fat and fibrous tissue resulting in a risk for re-entry ventricular arrhythmias. A broad spectrum of structural and functional abnormalities has been described in AC depending upon the stage of the disease. In some cases, the features may be marked and blatant (Fig. 15) or difficult to appreciate even if you know the histology is characteristic (Fig. 22). The condition starts as a primary concealed phase, with little structural or functional change, although there is still a risk of SCD from arrhythmias—especially during/immediately after extreme exertion. The secondary overt arrhythmic phase with structural and functional cardiac changes and arrhythmias. The tertiary phase shows global right ventricular dysfunction and dilatation, with a quaternary phase showing left ventricular involvement often with bi-ventricular cardiac failure. The changes of early ARVC are often localised, with a predilection for the apical, subtricuspid, and pulmonary outflow regions of the right ventricle: the so-called ‘triangle of dysplasia’ which manifests as segmental aneurysms or wall thinning [38–40]. As the disease progresses, right ventricular dilation and involvement of the left ventricle can occur (Fig. 16). Left ventricular wall involvement usually begins as subepicardial posterior wall fibrosis, which becomes diffuse

Cardiomyopathies

213

Type 8: Desmoplakin mutation – associated with early left ventricular involvement & Carvajal syndrome if recessive

Type 10: Desmoglein mutation – associated with left ventricular involvement

PP D

Type 12: Plakoglobin mutation – if dominant causes typical ARVC, if recessive results in Naxos disease

DG DC

PG

DP

DP PG

Type 7: Desmin mutation are seen in association with skeletal muscle myopathies

DC DG

Type 11: Desmocollin mutation – associated with woolly hair & mild palmoplantar keratoderma plus ARVC if homozygous

DG=Desmoglein, DC=Desmocollin, PP=Plakophilin, PG=Plakoglobin, DP=Desmoplakin, D=Desmin

D

PP

Type 9: Plakophilin mutation - the most common causative mutation of ARVC ARVC Subtype

Mutation

Type 1

Transforming growth factor beta 3 mutation

Type 2

Cardiac ryanodine receptor gene mutation

Type3

Unknown

Type 4

Titin mutation

Type 5

Transmembrane protein 43

Type 6

Unknown

Fig. 20  A diagrammatic representation of the genes and proteins involved in arrhythmogenic cardiomyopathy

Fig. 21  High magnification histology of AC showing the fat and fibrous tissue replacement of ventricular tissues

Fig. 22  Multiple slices of the right ventricular wall are seen in a case of AC. The degree of fatty tissue replacement may be subtle and overlooked unless one takes appropriate histology samples

214

C. R. Bunning and S. K. Suvarna

resulting in end-stage AC. One should be aware that this can simulate DCM in appearance [40–44]. Microscopically classical ARVC is characterized by fibrous and adipose tissue infiltration of the right ventricle starting in the outer epicardial layer and progressing towards the endocardium (Figs. 17, 18, and 19). As described the left ventricle can also be variably involved with fibrosis being the diagnostic hallmark [45, 46]. A variant characterised by subepicardial and mediomural fibrous and adipose tissue replacement confined to the left ventricle has been identified as a cause of SCD in the young [42, 43, 47, 48]. Despite extensive disease involvement of the left ventricle, SCD was the initial presenting symptom. A subepicardial distribution pattern of fibrous and adipose tissue replacement at autopsy should alert the pathologist to the possibility of ARVC with predominant or exclusive left ventricular involvement. Left ventricular subepicardial myocardial lesions are rare in other cardiac diseases [49], but frequent in ARVC. Alternate conditions for the pathologist when considering a diagnosis of ARVC are Brugada syndrome, ‘athletes’ heart and the normal adipose tissue replacement of the right ventricle with advancing age (especially in women and/or obese individuals). Myocardial fibrosis is also a consequence of other cardiac pathology, such as hypertensive heart disease, myocardial ischaemia, toxins, some illicit drugs (amphetamines and cocaine) and post-viral myocarditis. Such non-AC fibrosis contrastingly favours the subendocardial zone, with sparing of the epicardial region of the heart with hypertensive heart disease and myocardial ischaemia. In such cases the fibrosis is patchy and randomly distributed throughout the entire ventricular wall of both ventricles. This pattern of fibrosis is also seen with postviral myocarditis.

5

I solated Left Ventricular Noncompaction (LVNC)

Isolated left ventricular non-compaction (LVNC) is a rare congenital myocardial disorder with a poor prognosis that can present in either childhood or adulthood. It may present with systolic and diastolic dysfunction, arrhythmia or thromboembolic complications [50]. It occurs due to the persistence of the non-compacted endocardial layer, characteristic of the early fetal period before myocardial compaction is complete within the left ventricle [51]. Diagnosis is generally made via echocardiography or magnetic resonance imaging (MRI) (Figs. 23 and 24). LVNC shows both an X linked and autosomal dominant pattern of inheritance. Mutations in several genes have been associated with isolated LVNC, with these reflecting alterations to the Z band alternatively spliced PDZ-motif

Fig. 23  Photograph of left ventricular myocardium from the autopsy of a patient with LVNC. The non-compacted endocardial layer is composed of a complex meshwork of elongated and thinned trabeculations with deep intertrabecular recesses imparting a sponge-like appearance to the left ventricular wall

Fig. 24  Another case of LVNC showing the classic spongy left ventricle wall (Courtesy of Dr. A. Warfield)

protein (ZASP) (a protein involved with sarcomere stability), alpha-dystrobrevin (DTNA; a component of the dystrophin associated protein complex which localises to the sarcolemma), tafazzin (TAZ-G4.5; an enzyme involved in the metabolism of cardiolipin) and genes encoding alphacardiac actin, troponin T and beta-myosin with sarcomere protein mutations, being more common in adults [52]. Mutations in hyperpolarization-activated cyclic nucleotide channel 4 (HCN4) have also been linked with bradycardia and LVNC [53]. Isolated LVNC was detected in 9.2% of children with cardiomyopathy, being the third most common type [54]. In the paediatric population it can be present with other cardiac abnormalities, neuromuscular diseases and mitochondriopa-

Cardiomyopathies

215

Fig. 25  Low power haematoxylin and eosin stained photograph of LVNC showing that the non-compacted layer is composed of ‘finger-­ like’ projections

Fig. 26  A Masson’s trichrome stain confirms the prominent endocardial and subendocardial fibrosis, which is a feature of this disease due to abnormal myocardial microperfusion

thy [55]. By contrast, its prevalence in the adult population is estimated between 0.01% and 0.26% [56, 57]. LVNC may be encountered as an incidental finding at autopsy, but is more commonly seen in the setting of heart failure. In adults, it can present as SCD without preceding cardiac or family history. Macroscopically, a sponge-like appearance to the wall of the left ventricle is seen (once called ‘spongy left ventricular myocardium’) (Figs. 23 and 24). This complex pattern of trabeculation is seen in association with deep intertrabecular recesses. It is often associated with overlying mural thrombus due to stasis. The diagnosis of LVNC should be confined to the assessment of trabeculations in the left ventricle, where the compaction during development is normally greatest. The right ventricle is less compacted and could simulate pathological non-compaction. Histologically, LVNC is characterized by prominent and thinned trabeculations, manifest as finger-like processes. Subendocardial fibrosis is usually evident, due to the impaired coronary microcirculation associated with this disease (Figs. 25 and 26).

and idiopathic RCM. The secondary causes of RCM include a multitude of infiltrative and non-infiltrative cardiac pathologies, often as a component of a multisystem disease process, with the best example being amyloid (see below). TEMF is found in tropical regions, encompassing areas of Africa, Asia and South America and is rarely encountered in the industrial world. It affects children and adolescents resulting in heart failure and death. It is characterised by fibrous tissue deposition within the endomyocardium. It is believed that a multitude of factors conspire in a susceptible individual to result in TEMF, with these including dietary, environmental, infectious and genetic factors [58]. There is often an initial active phase with acute inflammation of the heart resulting in myocardial oedema, an eosinophilic infiltrate, subendocardial myofibre necrosis and vasculitis. Mural thrombi may develop in the ventricular apices with a risk of thromboembolic events [58]. RCM develops in the chronic phase with atrial dilatation and ventricular restriction, due to fibrosis which usually ceases just below the ventricular outflow tract. Microscopically endocardial thickening is seen with hyalinised collagen deposition. Lymphocytes may be present along the endocardial myocardial interface. Eosinophil infiltration is not seen in the chronic phase. Marked ascites is often the consequence of the progressive cardiac failure. Loeffler’s endomyocarditis is an infrequently encountered form of RCM (Fig. 27). This condition is found in conjunction with hypereosinophilic disease states, such as Churg-Strauss, eosinophilic leukaemia, parasitic infection, drug reactions and myeloproliferative disorders. Microscopically the initial stages of the disease show an acute inflammatory predominantly eosinophilic myocarditis, principally involving the endocardium and myocardium. Endocardial thrombus (Fig. 28), associated with underlying

6

Restrictive Cardiomyopathy (RCM)

Restrictive cardiomyopathy (RCM) encompasses a broad group of cardiomyopathy disorders, characterised by impaired filling of the ventricles due to stiffness and noncompliance (inelasticity) of the heart as a result of infiltrative processes or fibrosis (myocardial or endomyocardial). The contractile function and myocardial thickness may appear normal. RCM has both primary and secondary disorders, with primary causes being less common. Examples include tropical endomyocardial fibrosis (TEMF), Loeffler’s endomyocarditis

216

C. R. Bunning and S. K. Suvarna

disorders where intracellular deposits are characteristic. These include haemochromatosis, some glycogen storage disorders and lysosomal disorders which will be discussed later. Rare non-infiltrative secondary causes of RCM include iatrogenic causes such as radiotherapy, chemotherapy, changes following heart transplantation and rarely carcinoid heart disease.

7

Fig. 27  Photograph of a transverse slice of the heart at mid-septal level from a patient with hypereosinophilic syndrome and Loeffler’s endomyocarditis. A shaggy coat of thrombus is seen coating the right ventricle and there is fibrosis and white endocardial thickening of the left ventricle

Fig. 28  Haematoxylin and eosin stained photomicrograph showing that in the acute phase the endocardial thickening is due to granulation tissue covered by more recently formed thrombus (top) imparting a layered appearance

granulation tissue, may also be seen. Extensive fibrosis of the endocardium is later seen, ultimately causing a restrictive syndrome. Studies into the genetics of idiopathic RCM show that it is primarily a genetic disease. A disease causing mutation was found in more than 50% of cases of individuals involved in the studies [59, 60]. The mutations were found in ­sarcomeric and cytoskeletal proteins, ion channel genes and mitochondrial proteins including genes such as MYH7, DES, MYBPC3, TNNT and TTN [59, 60]. The more commonly encountered RCMs reflect secondary diseases mainly including infiltrative pathologies. The best example is amyloid. Other less common causes are storage

Amyloid Heart Disease

Cardiac amyloidosis occurs as part of a multisystem or localised disorder resulting from the deposition of extracellular insoluble fibrils of amyloid beta-pleated sheet protein in the interstitium of the myocardium and/or walls of blood vessels. There are several forms of amyloidosis that can involve the heart. These include senile cardiac amyloidosis, immunoglobulin in origin (AL amyloidosis), familial transthyretin related amyloidosis (TTR amyloidosis) or secondary to chronic infection as part of the acute phase serum amyloid A (amyloid AA). Senile cardiac amyloidosis (also known as wild type ATTR amyloidosis) is a common incidental autopsy finding in the elderly (usually in those aged 75 or more years), being an under-diagnosed condition. The disorder is caused by the deposition of wild type transthyretin within the heart and is thought to affect 10–25% of patients with heart failure, albeit with preserved ejection fraction [61]. There is often substantial left ventricular wall thickening at diagnosis. Carpel tunnel syndrome is commonly associated, preceding the cardiac symptoms by 10–15 years [62]. Another form of age-related amyloidosis is isolated atrial amyloidosis resulting from overproduction of atrial natriuretic factor (see chapter “The Normal Adult Heart and Methods of Investigation”). Amyloid deposits are limited to the atria with left atrial predominance and there is an association with atrial fibrillation [63]. AL amyloidosis (monoclonal immunoglobulin light chain amyloidosis) is the most common generalised type of amyloidosis to affect the heart. It is associated with plasma cell dyscrasias, such as B cell lymphoma, Waldenstrom’s macroglobulinaemia and multiple myeloma. AL amyloid fibrils are composed of monoclonal immunoglobulin kappa or lambda light chain fragments. Extracardiac involvement can also be seen, although the heart is the main organ affected. Familial transthyretin related amyloidosis (FTTR) is caused by the deposition of mutant TTR. In the majority of cases this is caused by a single point mutation in the TTR gene, resulting in a single amino acid substitution [62]. It is associated with an autosomal dominant pattern of inheritance, but different phenotypes are seen depending on the TTR mutation involved. Macroscopically cardiac amyloidosis with ventricular involvement may appear as normal to thickening of the ven-

Cardiomyopathies

tricular wall with a hypertrophic appearance which could be confused with HCM. The heart has a firm rubbery feel with a reportedly ‘waxy’ appearance to the cut surface of the myocardium. Limited amyloid in the heart may be seen in the right atrium, sometimes termed isolated atrial amyloidosis. Rarely this is seen as small translucent nodules coating the atrial surface which may appear brown after formalin fixation. Microscopically, amyloid is seen as interstitial infiltrates of homogeneous eosinophilic material surrounding individual myocytes forming a honeycomb pattern and/or nodules of amyloid due to coalescence of amyloid following myocyte death (Figs. 29 and 30). Nodular amyloid deposits may also be seen within the media and adventitia of the intramural coronary vessels, these may cause luminal narrowing [64, 65]. Amyloid deposition can occur in all layers of the heart. The pattern of deposition can be nodular, pericellular or mixed type with or without vascular involvement [66] (Figs. 30 and 31). This is visible with a standard H&E stain. Amyloid material can be highlighted by a Sirius or Congo red stain (Fig. 30). The amyloid deposits classically appear red/orange when viewed by light microscopy (see chapter “The Heart at Autopsy, Including Radiological Autopsy of the Heart”) and display apple-green birefringence when viewed under polarized light. Electron microscopy and immunohistochemistry can occasionally be used for further diagnostic confirmation and analysis of the amyloid fibril sub-type. Electron microscopy reveals the presence of linear non-branching fibrils with a diameter of 7.5–10  nm externally coating myocytes (Figs.  31 and 32). Immunohistochemistry can be used to identify the major protein component of the amyloid fibril, which may be pertinent to familial cases.

217

Fig. 30  Sirius red histochemistry highlights the case of amyloid deposition around individual myocytes

Fig. 31  Ultrastructural view of cardiac myocytes, externally coated by amyloid fibrils (marked by arrows)

Fig. 29  The myocardium is heavily infiltrated by eosinophil matrix around cells and throughout the interstitium. The inset macroscopic view of cardiac amyloidosis shows a semi-rigid structure that is self-­ supporting, even in the fresh post-mortem state. This macroscopic qual- Fig. 32  Electron micrograph of amyloid fibrils which typically have a ity often provides a clue to the ultimate histology diameter of 7.5–10 nm

218

8

C. R. Bunning and S. K. Suvarna

Cardiomyopathy Associated with Storage Disorders

This is a diverse group of cardiomyopathies. Glycogen storage disease and lysosomal storage disorders are hereditary conditions characterised by various disorders of carbohydrate and glycolipid metabolism resulting in an intracellular accumulation of substances. Some conditions are rapidly fatal from multiple organ system dysfunction soon after birth, but others present solely with cardiomyopathy later. Some storage disorders will produce a very restrictive pattern.

9

Anderson-Fabry Disease

Anderson-Fabry (often abbreviated to Fabry) disease is a lysosomal storage disorder with an X linked recessive pattern of inheritance, resulting in deficient/absent lysosomal alpha-galactosidase A activity [67]. This results in the accumulation of glycophospholipid in multiple organs leading to angiokeratoma, renal failure, cardiomyopathy, arrhythmias and peripheral and nervous system disorders. Macroscopically the most common appearance of cardiac Fabry’s disease is concentric ventricular hypertrophy (Fig. 33), although it may mimic HCM [68]. On histological examination prominent myocyte sarcoplasmic vacuolation is seen (Fig. 34), although in some cases this may not be present. Electron microscopy shows classic concentric lamellar bodies of alternating dense and pale material (‘myelinoid’ bodies) within the myocyte sarcoplasm [69] (Figs. 35 and 36).

Fig. 34  Haematoxylin and eosin stained photomicrograph showing cytoplasmic vacuolization which is characteristic of Anderson-Fabry disease

Fig. 35  Electron micrograph of ‘myelinoid’ figures within the myocyte sarcoplasm in a case of Anderson-Fabry disease

10

Danon Disease

This cardiomyopathy is a rare X linked lysosomal/glycogen storage disorder resulting in cardiomyopathy of hypertrophic type together with skeletal myopathy. The causative mutation occurs in the lysosomal associated membrane protein 2 gene (LAMP2). This results in decreased/complete absence of LAMP 2 protein depending on the mutation present and impairment of lysosomal transport and degradation of cellular material. The result of this is an accumulation of autophagic material and glycogen in cardiac and skeletal muscle Fig. 33  Macroscopic photograph of the cardiac variant of Anderson-­ cells [70]. On histology/electron microscopy of skeletal Fabry disease with asymmetrical ventricular hypertrophy. This can mimic HCM and needs histology and other tests to confirm the muscle/cardiac biopsy intracytoplasmic vacuoles are seen diagnosis within the myocytes (Figs. 37 and 38).

219

Cardiomyopathies

Fig. 36  High magnification of the ultrastructure of the Fabry’s cardiac myocyte showing detail of the myelinoid bodies

Fig. 37  The histology of Danon disease (re-processed from a previous paraffin embedded piece of tissue), showing mild disarray and minor cytoplasmic vacuolar change

Fig. 39  Haematoxylin and eosin stained section in an autopsy case of infantile Glycogen Storage Disorder (type II) (Courtesy of Dr. M. Ashworth)

Fig. 40  Buffy coat lymphocyte from a 60 year old woman with enzymatically proven adult-onset Pompe disease showing membrane bound glycogen (Courtesy of Dr. M. Ashworth)

11

Pompe Disease

Pompe disease is an autosomal recessive disorder resulting in alpha-glucosidase deficiency and the accumulation of glycogen in the heart, skeletal muscle, liver and nervous system [71]. There is often asymmetric left ventricular hypertrophy, which may have a very similar macroscopic appearance to HCM. The use of a periodic-acid Schiff (PAS) stain highlights the PAS positive glycogen within vacuoles (Figs. 39 and 40).

12

Fig. 38  The ultrastructure of Danon disease shows the autophagocytic vacuole containing cell debris, but little glycogen

Haemochromatosis

This genetic condition reflects an inherited autosomal recessive iron storage disorder. There is also the similar process of haemosiderosis as a consequence of iron overload—follow-

220

C. R. Bunning and S. K. Suvarna

ing multiple blood transfusions, increased iron absorption or exogenous iron treatment. Haemochromatosis manifests as iron deposition within multiple organs and tissues such as the liver, pancreas, joints and heart. Cardiac involvement is initially characterised by diastolic dysfunction (RCM) with arrhythmias, and in later stages by dilated cardiomyopathy due to pressure effects on the left ventricle. Iron deposition occurs from the epicardial surface to the endocardial surface, resulting in myocyte damage and death as a consequence of free radical generation. A Perls (Prussian blue) stain is best to assess the amount and distribution of iron (Figs.  41 and 42). In normal hearts no stainable iron should be seen [72].

Fig. 43  The dilated and hypertrophic heart (602 g) in a case of a significantly obese individual (BMI 38.9) is seen with both ventricular wall thickening and some chamber dilatation. Other factors such as obstructive sleep apnoea and obesity-hypoventilation syndrome often may aggravate any cardiac dysfunction alongside diabetic vasculopathy and other metabolic consequences of obesity

13

Fig. 41  Haemochromatosis seen on haematoxylin & eosin histology. There is subtle brown pigmentation in the myocytes (Courtesy of Dr. D. Rassl)

This condition is becoming more likely to be encountered by the pathologist with the rising general body mass index (BMI) of the population. It is believed to arise as a result of increased total blood volume and decreased systemic vascular resistance, resulting in a high cardiac output state maintained by an increased stroke volume. The persistence of this high cardiac output state results in adaptive cardiac structural changes such as left ventricular dilatation or concentric or eccentric left ventricular hypertrophy [73, 74] (Fig. 43). The mass of hearts in the morbidly obese is difficult to evaluate, with the heart mass not being directly proportional to the body mass. However, hearts of more that 500–550 g, are recognised to be prone increasingly to failure and/or sudden death.

14

Fig. 42  Haemochromatosis reveals its iron overload with Perl’s histochemistry staining of myocytes (Courtesy of Dr. D. Rassl)

Obesity-Related Cardiomyopathy

Alcoholic Cardiomyopathy

This entity has been previously discussed in the DCM section however it is fairly commonly encountered and is therefore worthy of further discussion. Alcohol is a direct cardiac toxin as well as adversely affecting other organs, notably the liver. Alcoholic cardiomyopathy is the result of chronic effects and can present with a hypertrophic, dilated or mixed morphology.

Cardiomyopathies

15

 eripartum and Postpartum (PPCM) P Cardiomyopathy

This is a rare form of cardiomyopathy occurring towards the end of pregnancy or several months postpartum. It is an idiopathic condition presenting with left ventricular systolic dysfunction and heart failure, sharing many similarities with DCM (Fig. 44). There are multiple hypotheses as to the aetiology of PPCM.  These include a possible genetic predisposition to the condition, an inflammatory background, an angiogenic imbalance during pregnancy, oxidative stress and the proapoptotic properties of prolactin during pregnancy [75]. PPCM is more common in older women and women who have had multiple previous pregnancies, as well as those with cardiovascular risk factors such as obesity, smokers and hypertensive patients [76].

16

221

disorder [80, 81]. Ventricular tachyarrhythmias and fibrillation can also be seen. These arrhythmias are thought to reflect conduction block/re-entrant loops as a direct result of sarcoid granulomas and scar tissue. Macroscopically, the heart may appear normal, although on section patchy infiltrates may be seen on the myocardial wall (Fig. 45). The left ventricle and interventricular septum are the most commonly involved sites. On histology, the classical appearance is of ‘naked’ non-caseating granulomas comprising epithelioid histiocytes, with some multinucleated giant cells surrounded by a scanty lymphocytic infiltrate (Figs. 46 and 47). The Langhans type giant cells may have occasional Schaumann and asteroid bodies, but triangulation

Sarcoid Heart Disease

Sarcoid is a systemic granulomatous disorder of unknown aetiology. Along with other tissues, especially the respiratory tract, the myocardium can be affected by the granulomas. In addition, the heart may be affected by sarcoid-associated fibrotic lungs resulting in pulmonary hypertension. Cardiac sarcoidosis occurs clinically in 5–10% of patients with known sarcoid [77]. Indeed, clinically silent cardiac sarcoid has been observed in approximately 20% of Caucasians and black Americans, and potentially up to 70–80% of Japanese patients at autopsy [78, 79]. Cardiac involvement is an unfavourable prognostic factor, with cardiac sarcoid being associated with atrial and ventricular arrhythmogenic activity. Atrioventricular block is the most commonly encountered

Fig. 44  A case of postpartum cardiomyopathy is seen with some cardiac hypertrophy (often seen in pregnancy). There may be DCM features, although these are not seen in this case

Fig. 45  Macroscopic view in a case of sarcoidosis. The chambers are mildly dilated and there is also some wall thickening. Focal scarring is seen mainly in the septum and left ventricle wall towards the base of the image (Courtesy of Dr. P. Gallagher)

Fig. 46  Histological view of left ventricle tissues affected focally by non-caseating granulomatous inflammation, in a known case of sarcoidosis (Courtesy of Dr. P. Gallagher)

222

C. R. Bunning and S. K. Suvarna

genetic studies. These are considered in detail in chapter “Sudden Cardiac Death”.

18

Conclusion

There are many forms of cardiomyopathy that should be known to the pathologist, with a multitude of aetiologies and presentations as described in, but not limited to, this chapter (Table 1). Research into these disease processes is continuing to identify numerous gene mutations, associated with the various presentations and forms of cardiomyopathy. This is driving increasing knowledge and new vectors for screening and treatment. Ongoing interest in this dynamic group of pathologies is recommended. Fig. 47  Electron microscopy in a case of a dilated mixed pattern cardiomyopathy clinically identified a sarcoid granuloma, in a case of unexpected sarcoidosis. The Langhans giant cell has a peripheral rim of nuclei

Fig. 48  Mediastinal node tissue with non-caseating granulomatous features, supporting the diagnosis of sarcoid

against mediastinal nodal granulomas may assist (Fig. 48). The granulomas become burnt-out with time and undergo fibrosis. The granulomas may be seen on endomyocardial biopsy, but sensitivity for non-caseating granuloma identification at endomyocardial biopsy is low at less than 20% [82].

17

Channelopathies

These gene-defect lesions may often present as sudden deaths, but are generally considered within the cardiomyopathy families. They generally appear as normal hearts in autopsy studies and require reservation of spleen tissue for

Table 1  Classification of main cardiomyopathies (modified from European Heart Journal https://doi.org/10.1093/eurheartj/ehm342) DCM Genetic • Unknown gene • Sarcomeric protein mutations (see also HCM) • Z-band (Muscle LIM protein, TCAP) • Cytoskeleton genes (dystrophin, desmin, metavinculin, epicardin, sarcoglycan complex, etc.) • Nuclear membrane (lamin A/C, emerin) • Intercalated disc protein (see also AC) • Mitochondrial disorders Acquired • Myocarditis (infective/toxic/immune) • Kawasaki disease • Eosinophilic syndrome (Churg-Strauss, etc.) • Viral persistence • Drugs • Pregnancy-related (PPCM) • Nutritional (thiamine, selenium, hypocalcaemia, etc.) • Alcohol HCM Genetic • Unknown gene • Sarcomeric gene mutations (beta-myosin heavy chain, cardiac myosin binding protein C, cardiac troponin I, troponin T, alpha-­tropomyosin, cardiac actin, alpha myosin heavy chain, titin, troponin C, muscle LIM protein, etc.) • Glycogen storage disease (Pompe, PRKAG2, Danon, etc.) • Lysosomal storage disease (Anderson-Fabry, Hurler’s syndrome, etc.) • Disorders of fatty acid metabolism • Carnitine deficiency • Mitochondrial cytopathies • Phosphorylase B kinase deficiency • Syndromic HCM (Noonan’s, LEOPARD, Friedreich’s ataxia, Beckwith-Wiedemann, Swyer syndromes) • Other (familial amyloid, phospholamban promoter) Acquired • Obesity cardiomyopathy • Offspring of diabetic mothers

Cardiomyopathies Table 1 (continued) • Athletes • Amyloid AC Genetic • Familial, unknown gene • Intercalated disc protein mutations (Plakoglobin, Desmoplakin, Plakophilin 2, Desmoglein 2, Desmocollin 2) • Cardiac ryanodine receptor (RyR2) mutations • Transforming growth factor beta Acquired • None formally recognised RCM Genetic • Unknown gene • Sarcomeric gene mutations (troponin I, myosin light chain) • Familial amyloid (TTR = RCM and neuropathy) • Apolipoprotein (RCM and nephropathy) • Desminopathy • Anderson-Fabry • Glycogen storage disease • Haemochromatosis Acquired • Amyloid • Scleroderma • Endomyocardial fibrosis (ergotamine, serotonin, busulphan, etc.) • Carcinoid heart disease • Radiation • Drug toxicity (anthracycline) Unclassified Genetic, unknown gene • Left ventricular non-compaction • Barth syndrome • Lamin A/C • alpha-dystrobrevin • ZASP Acquired • Takotsubo cardiomyopathy Acknowledgements  Grateful thanks are expressed to Mr. B. Wagner, Senior Electron Microscopist, Sheffield Teaching Hospitals for his expertise and photography of ultrastructural histology in this chapter.

References 1. Arbustini E, Narula N, Dec WG, et al. The MOGE(S) Classification for a phenotype–genotype nomenclature of cardiomyopathy. Endorsed by the World Heart Federation. J Am Col Cardiol. 2013;62:2046–72. 2. McNally E, Golbus J, Puckelwatz M. Genetic mutations and mechanisms in dilated cardiomyopathy. J Clin Invest. 2013;123(1):19–26. 3. Vikhorev PG, Smoktunowicz N, Munster AB, et  al. Abnormal contractility in human heart myofibrils from patients with dilated cardiomyopathy due to mutations in TTN and contractile protein genes. Sci Rep. 2017;7(1):14829. 4. Ware JS, Li J, Mazaika E, et  al. Shared genetic predisposition in peripartum and dilated cardiomyopathies. N Engl J Med. 2016;374(3):233–41.

223 5. Marian AJ, van Rooij E, Roberts R.  Genetics and genomics of single-gene cardiovascular diseases: common hereditary cardiomyopathies as prototypes of single-gene disorders. J Am Coll Cardiol. 2016;68(25):2831–49. 6. Parks SB, Kushner JD, Nauman D, et al. Lamin A/C mutation analysis in a cohort of 324 unrelated patients with idiopathic or familial dilated cardiomyopathy. Am Heart J. 2008;156:161–9. 7. Van Berlo JH, De Voogt WG, Van Der Kooi AJ, et al. Meta-analysis of clinical characteristics of 299 carriers of LMNA gene mutations: do lamin A/C mutations portend high risk of sudden death? J Mol Med. 2005;83:79–83. 8. Bienengraeber M, Olson T, Selivanov V, et al. ABCC9 mutations identified in human dilated cardiomyopathy disrupt catalytic KATP channel gating. Nat Genet. 2004;36:382–7. 9. Taylor MR, Slavov D, Ku L, Di Lenarda A, et  al. Prevalence of desmin mutations in dilated cardiomyopathy. Circulation. 2007;115:1244–51. 10. Goldfarb LG, Dalakas MC. Tragedy in a heartbeat: malfunctioning desmin causes skeletal and cardiac muscle disease. J Clin Invest. 2009;119(7):1806–13. 11. Bione S, Dadamo P, Maestrini E, et al. A novel X-linked gene, G4.5., is responsible for Barth syndrome. Nat Genet. 1996;12:385–9. 12. Brauch KM, Karst ML, Herron KJ, et al. Mutations in ribonucleic acid binding protein gene cause familial dilated cardiomyopathy. J Am Coll Cardiol. 2009;54(10):930–41. 13. Guo W, Schafer S, Greaser M, et  al. RBM20, a gene for hereditary cardiomyopathy, regulates titin splicing. Nat Med. 2012;18(5):766–73. 14. Guo W, Zhu C, Yin Z, et al. Splicing factor RBM20 regulates transcriptional network of titin associated and calcium handling genes in the heart. Int J Biol Sci. 2018;14(4):369–80. 15. Bollen IAE, Van Deel ED, Kuster DWD, Der Velden JV. Peripartum cardiomyopathy and dilated cardiomyopathy: different at heart. Front Physiol. 2014;5:531. 16. Piano MR, Phillips SA.  Alcoholic cardiomyopathy: pathophysiologic insights. Cardiovasc Toxicol. 2014;14(4):291–308. 17. Semsarian C, Ingles J, Maron MS, Maron BJ. New perspectives on the prevalence of hypertrophic cardiomyopathy. J Am Coll Cardiol. 2015;65(12):1249–54. 18. Young L, Smedira NG, Tower-Rader A, Lever H, Desai MY. Hypertrophic cardiomyopathy: a complex disease. Cleve Clin J Med. 2018;85(5):399–411. 19. Konno T, Chang S, Seidman JG, Seidman CE. Genetics of hypertrophic cardiomyopathy. Curr Opin Cardiol. 2010;25(3):1–8. 20. Berrens-Gawlik V, Mearini G, Gedicke-Hornung C, et  al. MYBPC3 in hypertrophic cardiomyopathy: from mutation identification to RNA-based correction. Pflugers Arch. 2014;466:215–23. 21. Hartmannova H, Kubanek M, Sramko M, et al. Isolated X-linked hypertrophic cardiomyopathy caused by a novel mutation of the four-and-ahalf LIM domain 1 gene. Circ Cardiovasc Genet. 2013;6(6):543–51. 22. Liang Y, Bradford WH, Zhang J, Sheikh F.  Four and a half LIM domain protein signaling and cardiomyopathy. Biophys Rev. 2018;10(4):1073–85. 23. Roma-Rodrigues C, Fernandes AR. Genetics of hypertrophic cardiomyopathy: advances and pitfalls in diagnosis and therapy. Appl Clin Genet. 2014;7:195–208. 24. Ackerman MJ, VanDriest SL, Ommen SR.  Prevalence and agedependence of malignant mutations in the beta-myosin heavy chain and troponin T genes in hypertrophic cardiomyopathy: a comprehensive outpatient perspective. J Am Coll Cardiol. 2002;39(12):2042–8. 25. Moolman JC, Corfield VA, Posen B, Ngumbela K.  Sudden death due to troponin T mutations. J Am Coll Cardiol. 1997;29(3):549–55.

224 26. Parto VM, Antoncecchi V, Sozzi F, et al. Echocardiogenic diagnosis of the different phenotypes of hypertrophic cardiomyopathy. Cardiovasc Ultrasound. 2016;14(1):30. 27. Mozaffarian D, Caldwell JH.  Right ventricular involvement in hypertrophic cardiomyopathy: a case report and literature review. Clin Cardiol. 2001;24:2–8. 28. Raju H, Alberg C, Sagoo GS, et  al. Inherited cardiomyopathies. BMJ. 2011;343:d6966. 29. Sen-Chowdhry S, Syrris P, Prasad SK, et al. Left-dominant arrhythmogenic cardiomyopathy: an under-recognized clinical entity. J Am Coll Cardiol. 2008;52:2175–87. 30. Christien KH, Bazoukis G, Liu T, et al. Arrhythmogenic right ventricular cardiomyopathy/dysplasia (ARVC/D) in clinical practice. J Arrhythm. 2017;34(1):11–22. 31. Marcus FI, McKenna WJ, Sherrill D, Basso C, Bauce B, Bluemke DA, et al. Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the Task Force criteria. Eur Heart J. 2010;31:806–14. 32. Sen‐Chowdhry S, Syrris P, Ward D, Asimaki A, Sevdalis E, Mckenna WJ. Clinical and genetic characterization of families with arrhythmogenic right ventricular dysplasia/cardiomyopathy provides novel insights into patterns of disease expression. Circulation. 2007;115:1710–20. 33. Bhonsale A, Groeneweg JA, James CA, et  al. Impact of genotype on clinical course in arrhythmogenic right ventricular dysplasia/cardiomyopathy-associated mutation carriers. Eur Heart J. 2015;36(14):847–55. 34. Iyer VR, Chin AJ. Arrhythmogenic right ventricular cardiomyopathy/Dysplasia (ARVC/D). Am J Med Genet C Semin Med Genet. 2013;163C(3):185–97. 35. Merner ND, Hodgkinson KA, Haywood AF, et al. Arrhythmogenic right ventricular cardiomyopathy type 5 is a fully penetrant, lethal arrhythmic disorder caused by a missense mutation in the TMEM43 gene. Am J Hum Genet. 2008;82:809. 36. Rampazzo A, Nava A, Erne P, et al. A new locus for arrhythmogenic right ventricular cardiomyopathy (ARVD2) maps to chromosome 1q42-q43. Hum Mol Genet. 1995;4:2151–4. 37. Leask A, Abraham DJ.  TGF-beta-signaling and the fibrotic response. FASEB J. 2004;18:816–27. 38. Thiene G, Nava A, Corrado D, et  al. Right ventricular cardiomyopathy and sudden death in young people. N Engl J Med. 1988;318:129–33. 39. Marcus FI, Fontaine G, Guiraudon G, et  al. Right ventricular  dysplasia. A report of 24 adult cases. Circulation. 1982;65:384–99. 40. Corrado D, Basso C, Thiene G, et  al. Spectrum of clinicopathologic manifestations of arrhythmogenic right ventricular cardiomyopathy/dysplasia: a multicenter study. J Am Coll Cardiol. 1997;30:1512–20. 41. Corrado D, Basso C, Thiene G.  Arrhythmogenic right ventricular cardiomyopathy: diagnosis, prognosis and treatment. Heart. 2000;83:588–95. 42. Gallo P, D’Amati G, Pellicia F.  Pathologic evidence of extensive left ventricular involvement in arrhythmogenic right ventricular cardiomyopathy. Hum Pathol. 1992;23:948–52. 43. Pinamonti B, Sinagra GF, Salvi A, et al. Left ventricular involvement in right ventricular dysplasia. Am Heart J. 1992;123:711–24. 44. Miani D, Pimamonti B, Bussani R, et al. Right ventricular dysplasia: a clinical and pathological study of two families with left ventricular involvement. Br Heart J. 1993;69:151–7. 45. D’Amati G, Leone O, di Gioa CR, et  al. Arrhythmogenic right ventricular cardiomyopathy: clinicopathologic correlation based on a revised definition of pathologic patterns. Hum Pathol. 2001;32:1078–86. 46. Burke AP, Farb A, Tashko G, et  al. Arrhythmogenic right ventricular cardiomyopathy and fatty replacement of the right ven-

C. R. Bunning and S. K. Suvarna tricular myocardium: are they different diseases? Circulation. 1998;97:1571–80. 47. De Pasquale CG, Heddle WF. Left sided arrhythmogenic ventricular dysplasia in siblings. Heart. 2001;86:128–30. 48. Michalodimitrakis M, Papadomanolakis A, Stiakakis J, et  al. Left side right ventricular cardiomyopathy. Med Sci Law. 2002;42:313–7. 49. Shirani J, Roberts WC.  Subepicardial myocardial lesions. Am Heart J. 1993;125:1346–52. 50. Oechslin EN, Attenhofer Jost CH, Rojas JR, et al. Long-term follow-­up of 34 adults with isolated left ventricular noncompaction: a distinct cardiomyopathy with poor prognosis. J Am Coll Cardiol. 2000;36:493–500. 51. Sedmera D, Pexieder T, Vuillemin M, et al. Developmental patterning of the myocardium. Anat Rec. 2000;258:319–37. 52. Ichida F. Left ventricular noncompaction. Circ J. 2009;73:19–26. 53. Milano A, Vermeer A, Lodder E, et al. HCN4 mutations in multiple families with bradycardia and left ventricular noncompaction cardiomyopathy. J Am Coll Cardiol. 2014;64(8):745–56. 54. Andrews RE, Fenton MJ, Ridout DA, Burch M. British Congenital Cardiac Association. New-onset heart failure due to heart muscle disease in childhood: a prospective study in the United Kingdom and Ireland. Circulation. 2008;117:79–84. 55. Stollberger C, Finsterer J, Blazek G.  Left ventricular hypertrabeculation/noncompaction and association with additional cardiac abnormalities and neuromuscular disorders. Am J Cardiol. 2002;90:899–902. 56. Oeschlin EN, Attenhoffer Jost CH, Rojas JR, et al. Long-term follow-­up of 34 adults with isolated left ventricular noncompaction: a distinct cardiomyopathy with poor prognosis. J Am Coll Cardiol. 2000;36:493–500. 57. Stollberger C, Winkler-Dworak M, Blazek G, Finsterer J. Prognosis of left ventricular hypertrabeculation/noncompaction is dependent on cardiac and neuromuscular comorbidity. Int J Cardiol. 2007;121:189–93. 58. Grimaldi A, Mocumbi AO, Freers J.  Tropical endomyocardial fibrosis natural history, challenges and perspectives. Circulation. 2016;133:2503–15. 59. Gallego-Delgado M, Delgado JF, Brossa-Loidi V, et al. Idiopathic restrictive cardiomyopathy is primarily a genetic disease. J Am Coll Cardiol. 2016;67(25):3021–3. 60. Kostareva A, Kiselev A, Gudkova A, et  al. Genetic spectrum of idiopathic restrictive cardiomyopathy uncovered by next generation sequencing. PLoS One. 2016;11(9):e0163362. https://doi. org/10.1371/journal.pone.0163362. 61. Halatchev IG, Zheng J, Ou J.  Wild-type transthyretin cardiac amyloidosis (ATTRwt-CA), previously known as senile cardiac amyloidosis: clinical presentation, diagnosis, management and ­ emerging therapies. J Thorac Dis. 2018;10(3):2034–45. 62. Patel KS, Hawkins PN. Cardiac amyloidosis: where are we today? J Intern Med. 2015;278:126–44. 63. Goette A, Röcken C.  Atrial amyloidosis and atrial fibrillation: a gender-dependent “arrhythmogenic substrate”? Eur Heart J. 2004;25(14):1185–6. 64. Roberts WC, Waller BF.  Cardiac amyloidosis causing cardiac dysfunction: analysis of 54 necropsy patients. Am J Cardiol. 1983;52:137–46. 65. Booth DR, Tan SY, Hawkins PN, et  al. A novel variant of transthyretin, 59Thr-Lys, associated with autosomal dominant cardiac amyloidosis in an Italian family. Circulation. 1995;91:962–7. 66. Smith TJ, Kyle RA, Lie JT. Clinical significance of histopathologic patterns of cardiac amyloidosis. Mayo Clin Proc. 1984;59:547–55. 67. Kint JA.  The enzyme defect in Fabry’s disease. Nature. 1970;227(5263):1173. 68. Linhart A, Palecek T, Bultas J, et  al. New insights in cardiac structural changes in patients with Fabry’s disease. Am Heart J. 2000;139:1101–8.

Cardiomyopathies 6 9. Germain DP. Fabry disease. Orphanet J Rare Dis. 2010;5:30. 70. D’souza RS, Levandowski C, Slavov D, et  al. Danon disease: clinical features, evaluation and management. Circ Heart Fail. 2014;7:843–9. 71. Hirschhorn R, Reuser A.  Glycogen storage disease type II: acid alpha-glucosidase (acid maltase) deficiency. In: Scriver C, Beaudet A, Sly W, Valle D, editors. The metabolic and molecular bases of inherited disease, vol. 3. New  York, NY: McGraw-Hill; 2001. p. 3389–420. 72. Gujja P, Rosing DR, Tripodi DJ, Shizukuda Y. Iron overload cardiomyopathy, better understanding of an increasing disorder. J Am Coll Cardiol. 2011;56(13):1001–12. 73. Alpert MA. Obesity cardiomyopathy: pathophysiology and evolution of the clinical syndrome. Am J Med Sci. 2001;321(4):225–36. 74. Peterson LR, Waggoner AD, Schechtman KB, Meyer T, Gropler RJ, Barzilai B, Davila-Roman VG.  Alterations in left ventricular structure and function in young healthy obese women: assessment by echocardiography and tissue Doppler imaging. J Am Coll Cardiol. 2004;43(8):1399–404. The end result is congestive cardiac failure.

225 75. Kim MJ, Shin MS. Practical management of peripartum cardiomyopathy. Korean J Intern Med. 2017;32(3):393–403. 76. Sliwa K, Fett J, Elkayam U. Peripartum cardiomyopathy. Lancet. 2006;368:687–93. 77. Newman LS, Rose CS, Maier LA.  Sarcoidosis. N Engl J Med. 1997;336:1224–34. 78. Baughman RP, Teirstein AS, Judson MA, et  al. Case Control Etiologic Study of Sarcoidosis (ACCESS) Research Group. Clinical characteristics of patients in a case control study of sarcoidosis. Am J Respir Crit Care Med. 2001;164:1885–9. 79. Iwai K, Sekiguti M, Hosoda Y, et  al. Racial difference in cardiac sarcoidosis incidence observed at autopsy. Sarcoidosis. 1994;11:26–31. 80. Kim JS, Judson MA, Donnino R, et  al. Cardiac sarcoidosis. Am Heart J. 2009;157:9–21. 81. Sekhri V, Sanal S, Delorenzo LJ, et al. Cardiac sarcoidosis: a comprehensive review. Arch Med Sci. 2011;7:546–54. 82. Uemura A, Morimoto S, Hiramitsu S, et al. Histologic diagnostic rate of cardiac sarcoidosis evaluation of endomyocardial biopsies. Am Heart J. 1999;138:299–302.

Cardiac Tumours Doris M. Rassl

1

Introduction

Cardiac tumours are rare and account for only 0.45–0.85% of cardiac surgical procedures [1]. Metastatic tumours are much more frequent than primary neoplasms [2], although in surgical series metastatic disease of the heart is still uncommon (3.3%) [1]. The vast majority of primary cardiac tumours are benign, although some can mimic malignant entities. In all cases, careful histological examination is required. In cases of cardiac tumours, clinical presentation is myriad, but is generally related to the presence of a mass lesion affecting the heart. Both benign and malignant cardiac tumours may present with sudden death, chest pain, cardiac failure, superior vena cava syndrome, valvular abnormalities and/or arrhythmias. The presentation depends mainly upon the site and size of the lesion. However, many tumours are completely asymptomatic, being discovered incidentally. Systemic or pulmonary emboli, caused by detached tumour tissue or mobilization of thrombotic deposits, can be seen with both benign and malignant intra-cavity lesions. Pericardial effusions, often haemorrhagic, tend to be associated with malignancy. Occasionally, rather protean systemic symptoms, including fever, cachexia and malaise may arise, being thought to be related to the release of cytokines. However, a systemic response may also occur secondary to embolus or infective vegetation related to the tumour. Abnormal laboratory findings that may develop include leukocytosis, anaemia, thrombocytosis or thrombocytopenia, polycythaemia, hypergammaglobulinaemia and an elevated erythrocyte sedimentation rate [3].

D. M. Rassl (*) Histopathology, Royal Papworth Hospital, Cambridge, UK e-mail: [email protected]

2

Tumours Metastatic to the Heart

Although metastases represent the commonest tumours of the heart, they are relatively rare with a reported incidence of 1.5–20% in autopsies of cancer patients, but are anticipated to increase with extended survival of oncology patients. Metastases may originate from any malignant tumour, but leading causes for cardiac metastasis are melanoma (Fig. 1), carcinomas (including lung, breast, oesophagus, kidney, ovary and rarely colorectal), sarcomas and haematological neoplasms (leukaemia and lymphoma). Tumour spread can occur, in descending order of frequency, to the pericardium, epicardium, myocardium, endocardium and less frequently to intracavitary regions, with a predominance of right-sided cardiac involvement [4]. Metastatic deposits may be diffuse, multifocal (Fig.  2), or consist of a single mass lesion. The diagnostic criteria reflect those of the primary tumour. The presentation depends on the site of cardiac involvement and the volume of the tumour; many secondary tumours of the heart are clinically silent. There is a predictable poor prognosis. The differential diagnosis for cardiac metastases includes thrombi, vegetations and primary cardiac tumours such as myxoma [4].

Fig. 1  Low power view of a subendocardial deposit of metastatic malignant melanoma (Haematoxylin & Eosin)

© Springer Nature Switzerland AG 2019 S. K. Suvarna (ed.), Cardiac Pathology, https://doi.org/10.1007/978-3-030-24560-3_12

227

228

D. M. Rassl

Fig. 2  Macroscopic view of the heart at autopsy showing disseminated mesothelioma metastases with no features of direct extension from the pleural primary site. This patient suffered ventricular tachycardias and a final ventricular fibrillation arrest

3

Cardiac Myxoma

Although some have previously suggested that they may be reactive lesions or hamartomas, myxomas are now accepted as representing true neoplasms. DNA analysis of myxomas has shown most to be diploid, although 13% are aneuploid [5, 6]. The cell of origin has been postulated to be either from mesenchymal ‘embryonic rests’, possibly entrapped foregut derivatives, or subendothelial reserve cells present within the fossa ovalis and/or surrounding endocardium, capable of divergent differentiation. This explains the presence of ‘heterologous’ elements in some myxomas and the eclectic immunohistochemical profile [7]. More recent molecular studies have revealed protein expression profiles similar to those seen in endocardial-mesenchymal transformation of the endocardial cushion [8]. Myxomas account for 36–76% of cases in surgical series, and are often located in the atria [1, 9]. The highest incidence is in women (F:M ratio 1.9:1) [10, 11], in the age range 30–70 years, but have been reported in all age groups. Less than 10% of cardiac myxomas occur in familial clusters as part of Carney’s complex (myxoma syndrome). This has an autosomal dominant mode of inheritance, is a multiple neoplasia syndrome and includes cardiac, cutaneous and breast myxomas, cutaneous and mucosal pigmented lesions, Sertoli cell testicular tumours, thyroid tumours and endocrine abnormalities (Cushing’s syndrome and acromegaly) [8, 12]. Different combinations of simultaneously occurring lesions may be found and these syndromes have been demonstrated

to be due to a genetically heterogenous mutation of the tumour suppressor gene PRKAR1A (protein kinase A regulatory subunit-1-alpha gene) on chromosome 17q22-24 [2]. To date, 122 different PRKAR1A mutations have been identified and this gene may play a role in both cardiac development and myxoma genesis [13]. About 70% of patients with Carney’s syndrome a have germline mutation in the PRKAR1A gene [8]. Cardiac myxomas occurring in the context of Carney’s syndrome tend to occur in a younger age group than sporadic cases, show no sex predominance, may be multifocal [14, 15] and often show atypical localization outside the left atrium [8]. Myxomas may be entirely asymptomatic (10–20%) [8]. Symptoms, when they occur, are usually related to the location of the lesion within the cardiac chambers. They can result in obstruction of the mitral or tricuspid valves, leading to cardiac failure or palpitations. The classical clinical signs on auscultation are of a murmur which changes with time and position, and/or an early diastolic ‘tumour plop’, caused by prolapse of the lesion through the valve orifice. Pulmonary or systemic emboli can occur, depending on the primary site of the tumour, and may be due to surface thrombus or embolization of the tumour itself. Villous tumours and those with abundant myxoid stroma are more frequently associated with embolization. Overexpression of matrix metalloproteinases and associated tumour degeneration may increase the risk of embolization [8]. Systemic emboli to the brain, coronary arteries, kidney, spleen and extremities have been documented. Tumour emboli may infiltrate the arterial wall, causing weakening and aneurysm formation [8]. Constitutional symptoms may occur, including fever, weight loss and arthralgias. Clinical signs, such as an elevated erythrocyte sedimentation rate, anaemia and leukocytosis may also be present. Many of these findings are thought to be the result of cytokine release by the tumour, including interleukins (IL6, IL4, IL12p70), interferon gamma and tumour necrosis factor [8]. Secondary infection of the tumour may also cause systemic symptoms. Paraneoplastic syndromes (demyelinating neuropathy, epistaxis, pancreatitis, vasculitis/vasculopathy) have also been reported in association with myxomas [8]. Cardiac myxomas most commonly occur in the left atrium on the intra-atrial septum in the region of the foramen ovale [16] or the mitral annulus (approximately 75%). Of the remainder, 18% are found in the right atrium and approximately 6% in the ventricles, with left ventricular myxomas accounting for about 2.5% of all cases. Valvular myxomas are very rare (less than 1%), with atrioventricular valves, particularly the mitral valve being affected [13]. Myxomas arise from the endocardium, and project into the cardiac chamber as a sessile or pedunculated mass. Involvement of the underlying myocardium is not a feature.

Cardiac Tumours

Fig. 3  Cardiac myxoma (37  mm maximal diameter) macroscopic view, showing an irregular surface with variable areas of haemorrhage and myxoid change (white zones). This type of myxoma can fragment and embolise

The tumours are usually rounded polypoid masses, but may have a villous architecture with multiple, thick papillary fronds. They range in size from about 1–15 cm [17]. Generally, they have a variegated appearance, with opaque grey and yellow areas, as well as red/brown areas, corresponding to zones of haemorrhage (Fig. 3). They have a gelatinous consistency. The cut surface may be solid or contain cystic spaces. Foci of calcification may be present, with extensive calcification being accompanied by osseous metaplasia [13]. It is important for the pathologist to identify and sample the base of the lesion, as the surgeon will generally take a cuff of myocardium along with the tumour to ensure complete excision. This should be inked and processed in its entirety, in order to confirm excision and also to exclude the differential diagnosis of an infiltrating malignancy. Microscopically, myxomas are usually covered by a single layer of flattened endothelial cells. The bulk of the lesion is composed of bland stellate, fusiform or polygonal myxoma (or lepidic) cells, in a copious metachromatic myxoid stroma rich in mucopolysaccharides with a variable amount of proteoglycan, collagen and elastin, staining strongly for alcian blue (resistant to hyaluronidase) with patchy reactivity for mucicarmine and periodic acid-Schiff (PAS) stains (diastase resistant) [3, 8]. The cells have small round, oval, sometimes spindled nuclei and a moderate amount of eosinophilic cytoplasm. They are scattered singly, but are also seen as nests and cords, and surrounding vascular channels within the stroma (Fig. 4). This latter feature is characteristic, being helpful in excluding the differential diagnosis of organising thrombus. Mitoses are not normally seen and, if present, should alert the pathologist to an alternative malignant diagnosis.

229

Fig. 4  Cardiac myxoma demonstrating perivascular cuffing by lepidic cells, set within a background myxoid matrix (Haematoxylin & Eosin)

Fig. 5  Glandular variant of cardiac myxoma. Note the layer of columnar epithelium towards the left hand side of the image (Haematoxylin & Eosin)

There is usually prominent haemorrhage with haemosiderin deposition within the stroma. On occasion, haemosiderin and calcium-encrusted elastic fibres, known as Gamna-Gandy bodies, may be observed. Secondary degenerative changes, which can be seen to a variable degree, include fibrosis, cystic change, necrosis, thrombosis, calcification and metaplastic bone formation [8]. A stromal inflammatory infiltrate, which includes lymphocytes and plasma cells, as well as macrophages and rare giant cells, is usually present. This inflammatory component often extends into the underlying myocardium. At the base of the lesion, there are often thick-walled vessels. In less than 3% of myxomas glandular structures lined by a single layer of bland cuboidal to columnar, mucin-secreting or ciliated epithelial cells are found (Fig.  5) [3]. Some myxomas contain aggregates of

230

D. M. Rassl

cells with the morphological and immunostaining characteristics of fibroblasts, myofibroblasts or smooth muscle. Foci of extramedullary haematopoiesis may be seen in 7% of myxomas. Thymic epithelial rests and thymoma-like elements have been observed in some cases [3, 8]. Myxomas have an eclectic immunohistochemical profile, being variably positive for markers of neural differentiation (PGP9.5, S100 protein) [7], synaptophysin, smooth muscle actin, desmin, endothelial markers, vimentin and calretinin [3, 8]. Calretinin is the most useful marker, staining the myxoma cells in most cases. Epithelial cells within the glandular variant are positive for cytokeratin and carcinoembryonic antigen [3]. The differential diagnosis of myxoma includes organizing thrombus, papillary fibroelastoma, myxoid intimal fibroplasia and primary or metastatic sarcoma, including myxofibrosarcoma. A myxomatous neoplasm that does not conform to the histological pattern described above or tumours showing mitotic activity, cytological atypia or invasion should not be diagnosed as benign myxoma. The concept of ‘malignant myxoma’ is rejected by many [18], although very occasional reports of a sarcoma arising within an otherwise typical myxoma are noted [19].

4

Papillary Fibroelastoma

Papillary fibroelastomas are probably the most common primary cardiac valvular tumour-like lesions. Due to higher resolution imaging techniques they are now detected more frequently and may be more prevalent than previously thought, with recent studies suggesting that they are the most common cardiac tumour [20–23]. Echolucencies with speckled appearance and stippled pattern near the edges, which shimmer are suggestive of these lesions [24]. They can occur in all age groups, but are most often observed between the fourth and eighth decade, with a mean age of 60 years and no obvious sex predilection [22]. They do not tend to recur following excision, which is usually performed as a valve-­sparing shave excision of the lesion [22]. Although it is generally accepted that papillary fibroelastomas are non-­neoplastic, there is lack of agreement about whether the lesions are hamartomatous, represent a peculiar kind of reactive phenomenon or are even simply organised thrombi [22, 25]. A hamartomatous derivation has been suggested on the basis of the similarity of the entity to normal chordae tendineae [3], although the rarity of this lesion in childhood argues against this concept. On the basis of the similarity of papillary fibroelastoma to Lambl’s excrescences, which are composed of fibrin, others have suggested the lesion is caused by a reactive response in endothelial cells, which undergo hyperplasia, with excessive basement membrane formation, possibly secondary to trauma [3]. The increased incidence of

these lesions in structurally abnormal hearts, where there is disruption to the normal flow of blood potentially leading to repetitive haemodynamic trauma, supports this theory [26]. Abnormal blood flow might also explain the presence of organised thrombus. Papillary fibroelastoma has been shown to contain fibrin, hyaluronic acid and elastic fibres, in keeping with organised thrombus [22]. Detailed cytogenetic studies have been difficult [23], but instead of being driven by mutational changes and genomic instability, cardiac papillary fibroelastomas are thought to arise from a combination of non-genetic causes, including endothelial damage due to haemodynamic or iatrogenic stress, viral-induced growth, microthrombi aggregation and hamartomatous origins [20]. The clinical presentation varies from asymptomatic to heart failure, syncope or sudden death [23], and papillary fibroelastomas can cause considerable morbidity as a result of systemic or pulmonary emboli. Emboli are a more common complication than with myxoma (occurring in 34% of papillary fibroelastomas, compared to 24% of myxomas) [22]. This is probably due to their location and high blood flow rate, as left-sided valves are most commonly involved, with the aortic valve being the usual site [22, 23]. It is believed that emboli originate from thrombus associated with the lesion, as much as from the lesion itself [22, 27], leading to the proposal that anticoagulation be considered as a therapeutic option in patients unfit for surgery [27, 28]. Tumour mobility has been reported as the only independent predictor of death and non-fatal embolization associated with cardiac papillary fibroelastomas [24]. There is also the risk of obstruction of the coronary artery ostia in those lesions situated on the aortic valve [3], as well as a risk of valvular dysfunction [22]. However, many are asymptomatic and are discovered incidentally. Interestingly, mortality rates may be higher with fibroelastomas than with myxomas. This is postulated to be due to the overall older age at which they occur, and the increased risk of embolism [22]. Papillary fibroelastomas tend to occur in structurally abnormal hearts, and associations have been described with rheumatic heart disease, hypertrophic cardiomyopathy, valvular heart disease, and congenital heart disease [23]. Most are found on the valvular endocardium, the aortic valve being the usual site, followed by the mitral valve. Less commonly they involve the right-sided valves and non-valvular sites in the atria and ventricles, including papillary muscles [22, 23, 26]. About 23% arise from the atrial or ventricular endocardial surface [22]. Rarely, they may be multiple in the same or different locations of the heart [22, 24, 26]. Papillary fibroelastomas generally are around 10 mm in diameter, but range in size from 2 to 70 mm [22, 25]. They are cream in colour and have a characteristic appearance when immersed in liquid, with multiple fine strand-like fronds emanating

Cardiac Tumours

Fig. 6  Papillary fibroelastoma: macroscopic appearance resembling a sea-anemone (11  mm diameter). The lesion is best appreciated when suspended in a Petri dish with formalin

Fig. 7  Papillary fibroelastoma showing a villous architecture with bland fibrous stromal cores (Haematoxylin & Eosin)

from a central core at the base, often described as resembling a sea anemone (Fig. 6). The lesions may act as a nidus for thrombus formation and are sometimes associated with infective vegetations [22, 25], which may alter the classical macroscopic appearance. The fine strands seen macroscopically are composed of a central avascular hypocellular fibro-collagenous core, often showing hyaline change, covered by a monolayer of endothelial cells (Figs.  7 and 8). In some cases, the hyalinised core is surrounded by a peripheral layer of loose connective tissue rich in mucopolysaccharide and proteoglycan matrix. Stains for elastin reveal the presence of elastic fibres, from which the lesion derives its name. There is no infiltration of the underlying valve leaflet. The differential diagnosis includes cardiac myxoma, Lambl’s excrescence, vegetations and thrombus. Although

231

Fig. 8  Papillary fibroelastoma at high magnification showing a bland endothelial surface (Haematoxylin & Eosin)

cardiac myxoma can sometimes have a fronded appearance, the fronds are much thicker than those of papillary fibroelastoma. Myxomas arise from the atrial, or occasionally from the ventricular, wall and only a small number of case reports describe myxomas arising from a cardiac valve. Myxomas differ histologically from fibroelastomas with the presence of myxoma cells and blood vessels, although there may be some confusion between the two entities macroscopically, especially if there is associated thrombus [29]. Calretinin staining, which is positive in myxoma cells, may help distinguish between myxoma and papillary fibroelastoma [24]. Lambl’s excrescences are 1–2 mm in size and occur at the line of closure of the semilunar valves as fine thread-like strands, whereas fibroelastomas occur anywhere on the valve surface. Although fibroelastomas are morphologically similar (hence the synonym ‘giant Lambl’s excrescence’), Lambl’s excrescences are composed of fibrin and lack the acid-mucopolysaccharide matrix of fibroelastomas [25]. However, many transitional forms exist. Papillary fibroelastomas may be associated with thrombus, and it is important to exclude an underlying fibroelastoma in any valvular/mural thrombus or in embolic material. This requires processing all the material submitted for histological examination.

5

Rhabdomyoma

Rhabdomyomas are benign striated muscle tumours, regarded as hamartomas, without malignant potential [30]. They are usually classified as cardiac or extra-cardiac, based on their location and unique histology [31]. Extra-cardiac rhabdomyomas are very rare (60% of cases) [43]. Most cases are asymptomatic, but depending on their location these lesions can cause intractable arrhythmias and even sudden death. Surgical excision, even if incomplete, is effective [43]. Hamartomas of mature cardiac myocytes predominantly arise in the left ventricle, although they can be found in the atria and are usually solitary, but multiple lesions have been reported. Grossly they are poorly defined, firm, pale, fibrotic appearing areas, usually less than 5 cm in size. The margins may merge with the surrounding normal myocardium or there may be a relatively abrupt transition [43]. Microscopically the lesions are characterized by localized disorganized and hypertrophied mature myocytes with large, irregular nuclei. Myocyte vacuolization and venular dilatation may be seen, more often in paediatric cases [44]. Associated fibrosis, adipose tissue, smooth muscle and nerves can be present to a variable degree. If significant amounts of such components are apparent, the term mesenchymal hamartoma is used [45]. The tumours express troponin, desmin, actin and myosin, and are non-proliferative on Ki-67 staining. The differential diagnosis includes cardiac rhabdomyoma (but spider cells are not seen), hypertrophic cardiomyopathy (which does not form a discrete mass) and cardiac fibroma [43].

8

Cardiac Fibroma

Cardiac fibromas are benign congenital mesenchymal tumours of fibroblasts, or myofibroblasts, in variably collagenized stroma. These are rarely observed in adults, but are

234

the second most common tumour in the paediatric population following rhabdomyomas [46, 47]. One third of cases present under 1 year of age [48], but they can occur in all age groups [49]. Only 15% occur in adulthood and since the lesion is considered to be congenital [50], those discovered in older patients may actually have been present since birth. There is a slight male predominance and an association with Gorlin syndrome (naevoid basal cell carcinoma syndrome), an autosomal dominant disease caused by mutations of the tumour suppressor gene PTCH1 at chromosome 9q22.3. This gene encodes a tumour suppressor protein, which inhibits the Hedgehog signalling pathway [51]. The typical clinical triad of Gorlin syndrome consists of multiple basal cell carcinomas, keratocystic odontogenic tumours and cerebral calcifications, although there is a wide variety of phenotypic manifestations [34]. Approximately 3–5% of cardiac fibromas are found in people with Gorlin syndrome and conversely 3–5% of individuals with this syndrome have cardiac fibromas [34]. A suppressor gene role for PTCH1 is also implicated in non-syndromic cardiac fibromas, with therapeutic implications [46]. As stated, the tumour is benign and may be asymptomatic. However, sudden death can occur secondary to arrhythmia and conduction disturbances. There may be serious morbidity if the tumour is large enough to form a significant space-occupying lesion. Cardiac fibromas are not associated with tumour emboli and do not spontaneously regress [52], but often cease to grow [53], so that the enlarging, developing heart favourably alters the heart to tumour size ratio, possibly obviating the need for surgery. A ‘watch and wait’ approach is therefore valid in symptomatic but stable paediatric patients. Partial excision of symptomatic lesions is possible, to preserve the integrity of the heart. Due to their infiltrative nature surgical excision is not always an option and defibrillator implantation or cardiac transplantation may be necessary in some patients [46]. Cardiac fibromas are intramural and the most common location is within the myocardium of the left ventricular free wall [46, 53], although cases do occur in the right ventricle, septum and rarely the atria. The lesions range in size from a few millimetres to over 12 cm [53]. Multiple fibromas have been reported, but these must be distinguished from outgrowths at the periphery of the main tumour mass [53]. Careful sectioning is required to demonstrate continuity of ‘satellite’ lesions with the main tumour. Macroscopically cardiac fibromas are well circumscribed and usually solid, white and fibrous, although cystic degeneration may be apparent at the centre of larger lesions. When examined microscopically, fibromas are unencapsulated and usually show myocardial infiltration at the periphery, with entrapment of myocytes, which are sometimes seen quite deeply within the lesion [46].

D. M. Rassl

Fig. 12  Low power view of a cardiac fibroma showing bland spindle cells in a collagenous background (Haematoxylin & Eosin)

Fig. 13  Higher magnification of a cardiac fibroma demonstrating relatively uniform spindle cells, without mitotic activity (Haematoxylin & Eosin)

Histologically cardiac fibromas are composed of monomorphic, cytologically bland spindle cells in a collagenous background (Figs. 12 and 13). The cellularity of the lesion varies with age. Fibromas in neonates tend to be highly ­cellular, and there may even be rare mitoses, whereas those occurring in adulthood are predominantly composed of collagen, with only sparse bland spindle cells. Foci of dystrophic calcification are common, especially in tumours from older individuals. The more cellular tumours sometimes contain foci of extramedullary haematopoiesis [46]. Myxoid change and foci of chronic inflammation may also occur, occasionally with perivascular aggregates of lymphocytes and histiocytes around vessels or at the junction between the lesion and the myocardium [46, 47]. Stains for elastin reveal the presence of elastic fibres. Cardiac fibromas are positive

Cardiac Tumours

for vimentin and smooth muscle actin, and negative for desmin, CD34, HMB45 and S100 [46]. The differential diagnosis includes sarcoma, inflammatory myofibroblastic tumour and rhabdomyoma. Sarcoma may come into the differential, especially in more cellular lesions. Mitoses are extremely rare in fibromas, and their presence warrants consideration of malignancy. Inflammatory myofibroblastic tumours are very rare in the heart and their diagnosis is often suggested by significant numbers of inflammatory cells [48]. Entrapped myocytes within a fibroma may show degenerative vacuolation, and the lesion could be mistaken for a rhabdomyoma [48].

9

Cystic Tumour of the Atrioventricular (AV) Node

Cystic tumour of the atrioventricular node is a benign cystic mass at the base of the atrial septum, in the region of the atrioventricular node [54]. It is considered a form of endodermal heterotopia with a rest of endoderm becoming incorporated into the heart during cardiac folding and development, and differentiating towards an upper foregut phenotype [55, 56]. It is believed to reflect alterations in the development of the cardiac neural crest, which could explain the presence of ultimobranchial rests in the AV node region [54]. In 10% of cases the lesions occur in association with other midline defects and there is evidence that the lesions are capable of slow proliferation, the presence of mitoses indicating cell renewal, and the imbalance between cell proliferation and cell loss may cause alteration in  the size and shape of the mass [57]. Possible familial occurrence, as well as the association with other congenital and midline defects, may suggest a genetic defect involving the migration of embryologic tissues [54]. Cystic tumours of the AV node are rare, with a mean age at presentation of 38 years (age range from birth to 95 years), being more commonly found in women (F:M  =  3:1) [56]. The precise incidence is difficult to determine as in most cases the lesions are asymptomatic until presentation, or sudden death. There are rare macroscopic clues to their presence at autopsy, but in some cases a slightly elevated lesion at the base of the atrial septum is apparent. Awareness of the possible presence of this lesion in patients with heart block limited to the AV node or sudden cardiac death warrants detailed examination of the cardiac conduction system [58]. Due to their location, the lesions can lead to conduction defects causing variable degrees of heart block, including complete heart block in over 65% of patients and partial AV block in 15% of patients [58]. Ventricular tachycardia or fibrillation is experienced by some patients but there appears to be no relationship between size and the occurrence of lethal arrhythmia [59]. In a small number of cases which presented with atrial fibrillation and atrioventricular block, the

235

lesions have been successfully resected, usually requiring subsequent pacemaker support [59, 60]. Cystic tumours of the AV node may be visualized by transoesophageal echocardiography, computed tomography or cardiac magnetic resonance imaging (CMR). With the latter modality the lesions have been reported to be of high ­intensity on T1-weighted images and isointense with myocardium on T2-weighted images [59]. The lesions are located in the right atrium near the base of the atrial septum, in the region of the AV node and vary in size, ranging from 2 to 20  mm [59]. They are multicystic, although cysts may be barely visible macroscopically, and not infrequently contain a small volume of yellow/brown, semi-solid material [59, 61]. Microscopically, tumour cell nests and variably sized cystic spaces lined by tumour cells are set within a fibrous stroma devoid of smooth muscle cells, and which may show foci of chronic inflammation (Figs. 14 and 15). The tumour

Fig. 14  Low power view of a cystic tumour of the atrioventricular node (arrowed) above the membranous septum (S) (Haematoxylin & Eosin)

236

D. M. Rassl

cysts. In comparison to metastatic carcinoma, cystic tumour of the atrioventricular node is a benign lesion (albeit with fatal electrophysiological potential) with histologically bland epithelial cells, lacking significant mitotic activity or pleomorphism. Teratomas by contrast should show elements derived from the different embryonic cell layers (ecto-, endoand mesoderm). Bronchogenic cysts are typically larger single cysts with a muscular wall and tend to occur on the epicardial surface. Similarly mesothelial cysts are usually larger, unilocular and localized to the surface of the heart [57].

10 Fig. 15  Cystic tumour of the atrioventricular node with variably sized cystic spaces lined by tumour cells set within fibrous stroma (Haematoxylin & Eosin)

cells may have a cuboidal, transitional or squamoid morphology, and occasionally show features of sebaceous-type differentiation [56]. A minor population of neuroendocrine cells may also be apparent [54]. The cysts often form two cell layers, the inner (luminal) one being cuboidal and the outer consisting of transitional cells, and may contain amorphous material and keratinous debris, sometimes with evidence of calcification [57, 61]. Generally, cystic tumours of the atrioventricular node respect the boundaries of the central fibrous body, with no involvement of the ventricular myocardium or the valves [57]. The main tumour cells express many cytokeratins (CAM5.2, 34βE12, AE1/AE3, CK7), as well as EMA, B72.3, CEA, CA19.9, p63, bcl-2 and galectin 3 [54]. Scattered neuroendocrine cells, described in some reports, may be present which stain for calcitonin, chromogranin, synaptophysin and thyroid transcription factor 1 (TTF1), as well as cytokeratins, EMA and CEA [58]. The tumour cells are reported to be negative for mesothelial (thrombomodulin, HBME, calretinin, Wilms tumour 1 (WT1)), endothelial and lymphatic markers (CD31, factor VIII related antigen), as well as CK20, p53, cyclin D1, thyroglobulin, vimentin, estrogen and progesterone receptor [54, 58], although one case report found focal positivity for progesterone and strong staining for estrogen receptor [61], suggesting that some lesions might be hormone sensitive. On electron microscopic examination the solid cell nests demonstrate a well-formed basement membrane, desmosomes and cytoplasmic tonofilaments. Cells lining the spaces also have short microvilli and some may contain electron dense material [62], consistent with the finding that neuroendocrine markers stain some of the cells. The differential diagnosis includes metastatic carcinoma, intra-cardiac teratoma, and mesothelial and bronchogenic

Mesothelial/Monocytic Incidental Cardiac Excrescence (MICE)

An awareness of MICE is important to avoid confusion with neoplasms such as metastatic carcinoma, perivascular epithelioid tumour, Langerhans cell histiocytosis (LCH) and histiocytoid haemangioma [63]. Cardiac MICE is a rare benign, tumour-like lesion, which is usually an incidental histological finding and is composed of aggregates of mesothelial cells forming small groups, tubules, cords or micropapillary structures, admixed with monocytes/histiocytes, other inflammatory cells (neutrophils, occasional lymphocytes and eosinophils) and fibrin. Sparse adipocyte-like vacuoles and foreign material can also be seen. The lesions lack a vascular network or supporting stroma [3, 64]. MICE have been found in the cardiac chambers, on cardiac valves (especially the aortic and mitral valve), in the pericardial space, the ascending aorta, mediastinum and pleural space [3, 64]. There is an equal sex incidence, with an age range of 5–80 years, but most patients are above 60 years of age. Although the exact pathogenesis is unclear, these lesions are thought to be reactive or iatrogenic, possibly initiated by instrumentation, mechanical irritation or inflammation [63, 64]. Various reports have lead to the idea that MICE may be part of a spectrum of benign lesions, all composed of histiocytes and mesothelial cells, which may be termed “histiocytosis with rasinoid nuclei (HR)” [65].

11

 ardiac Lipoma and Lipomatous C Hypertrophy

Lipomatous hypertrophy and lipoma represent 6–10% of benign cardiac lesions. CT and MRI scans are helpful in establishing the fatty nature of these lesions and the exact anatomical location. Lipomatous hypertrophy is found almost exclusively in older (generally 50 or more years), obese patients [66]. It is defined as fatty infiltration (usually greater than 2  cm in

Cardiac Tumours

thickness), is usually brown and firm in consistency and generally affects the interatrial septum, although it may involve the entire atrial wall [67, 68]. It typically spares the foramen ovale, giving the lesion a characteristic dumb-bell shape [67]. It is usually located anterior to the foramen ovale, but may extend into the region of the atrio-ventricular node. The histogenesis of lipomatous hypertrophy is uncertain. It may result from a metabolic disturbance associated with obesity, increasing age, starvation or anaemia [69]. Lipomatous hypertrophy is usually asymptomatic and most often an incidental finding at autopsy. It may uncommonly be the cause of atrial arrhythmias (atrial fibrillation and atrioventricular block) [66], congestive cardiac failure and, if large, superior vena cava obstruction [70]. Lipomatous hypertrophy does not represent a neoplastic process and is a non-encapsulated lesion composed of varying proportions of mature adipocytes and brown fat cells, which contain multiple small vacuoles with a central nucleus. Myocytes are usually entrapped within the mass, especially at the periphery. These may show bizarre hypertrophic and degenerative changes, but mitoses are absent [70–73]. Often areas of fibrosis and focal collections of chronic inflammatory cells are present (Fig. 16). Cardiac lipomas, by contrast, are well-defined, benign masses composed of mature, white adipocytes. Cardiac lipomas are relatively rare and account for only 0.5–3% of excised heart tumours, with a reported incidence of 2.9–8% [70, 74, 75]. They can occur at any age, but are most prevalent between 40 and 60 years, with equal frequency in both sexes [67, 76]. The presentation of cardiac lipomas is varied and depends on their location and size. Many are incidental findings, but they can cause arrhythmias, syncope, rarely embolization

Fig. 16  Lipomatous hypertrophy including fetal fat cells, entrapped myocytes and a collection of chronic inflammatory cells (Haematoxylin & Eosin)

237

and even compression of the coronary arteries or blood flow within the heart [77]. Echocardiographic abnormalities may also be encountered [70]. Excision of cardiac lipomas is indicated if clinical symptoms develop or it is not possible to confidently exclude liposarcoma. Surgical excision generally provides complete cure and a good long-term prognosis [67]. Lipomas are encapsulated and can occur throughout the heart, including the visceral and parietal pericardium [66]. Most are subendocardial (50%), and the remaining tumours have a pericardial (25%) or intramyocardial (25%) position. Typical locations include the right atrium and left ventricle [67, 76]. Small sclerosed lipomas (fibrolipomas) may occur as small nodules on cardiac valves [74]. The size of cardiac lipomas varies and those that are left in situ when they are asymptomatic may grow to large dimensions [72]. Although typically solitary, multiple cardiac lipomas may be found in association with tuberous sclerosis [71], and an association with Cowden syndrome has also been reported [74]. Macroscopically lipomas are yellow, well-circumscribed, encapsulated masses which may be polypoid or pedunculated [74]. Similar to extra-cardiac lipomas, microscopically those in the heart are circumscribed and composed of bland, mature adipocytes with minimal or no atypia. A few myocytes may be entrapped at the margin, but a thin fibrous capsule is usually present [73]. Fat necrosis with associated macrophages may be seen [74]. The adipocytes of lipomas are S100 protein-positive. Unusual histologic variants of lipoma are generally not encountered within the heart [70]. The differential diagnosis includes liposarcoma, intramuscular variant of haemangioma [70] and hibernoma. The brown fat cells of lipomatous hypertrophy should not be mistaken for the lipoblasts of liposarcoma, which have indented, hyperchromatic, atypical nuclei [73]. Liposarcomas usually grow quickly and infiltrate into surrounding cardiac and thoracic structures. It should be noted that the intramuscular variant of haemangioma may contain variable numbers of adipocytes [70], and could be confused with a true lipomatous lesion. Hibernoma is a rare, benign tumour of brown adipose tissue, composed of multivacuolated adipocytes. Hibernomas are encapsulated and slow growing. They possibly arise from vestiges of brown fat that can persist around the great vessels, or at unusual sites. Aberrant differentiation of mesenchymal cells, ectopic growth or migration of adipose tissue may account for their occurrence. Four variants have been identified: the typical hibernoma, a mixture of hibernoma cells and white fat cells; the lipoma-like variant, with only scattered hibernoma cells amongst mature adipocytes; the myxoid variant and the spindle cell variant [78]. A diagnosis of hibernoma should only be considered where the macroscopic appearances do not fit lipomatous hypertrophy, and where microscopically the lesion is predominantly composed of multivacuolated cells [71].

238

12

D. M. Rassl

Cardiac Haemangioma

Cardiac haemangiomas are benign vascular tumours or malformations [79]. They represent a benign proliferation of endothelial cells, which form vascular channels, often separated by varying amounts of supporting connective tissue, in a non-encapsulated but relatively well-demarcated structure without necrosis [80]. Primary cardiac haemangiomas are infrequent and usually solitary. They account for about 2–3% of all primary cardiac tumours and up to 5% of all benign cardiac tumours [81]. They can occur at any age, with a mean age at diagnosis in the fifth to sixth decade [79]. There is no significant sex predominance [79, 81]. The clinical presentation of cardiac haemangiomas depends primarily on the tumour location, size and extension. The lesions may involve any part of the heart, including the endocardium and heart valves [82], myocardium or epicardium/pericardium, but most commonly arise from the ventricles followed by the atria [83]. Congenital haemangiomas frequently involve the right atrium [79]. Cardiac haemangiomas are usually asymptomatic and discovered incidentally, either radiologically, during cardiac surgery or at autopsy. In symptomatic cases, the most common symptoms include shortness of breath, atypical chest pain, palpitations and arrhythmias [81], but pericardial effusion or haemopericardium, cardiac tamponade, congestive heart failure, right ventricular outflow tract obstruction, coronary insufficiency, systemic embolization, valvular stenosis and sudden death have also been reported [84–86]. Intra-­ cavitary tumours may produce a triad of obstruction, embolization and constitutional symptoms [81]. Genetic susceptibility to cardiac haemangiomas has not been described and cardiac haemangiomas are usually sporadic, without associated extra-cardiac lesions. Large cardiac haemangiomas can show associated thrombosis with thrombocytopenia and consumptive coagulopathy (Kasabach-Merritt syndrome) [79]. The diagnosis is made by transthoracic echocardiography, computed tomography (CT), magnetic resonance imaging, or cardiac catheterization [83]. Coronary angiography may reveal the feeding vessel to the tumour and a characteristic tumour blush [87]. CT shows a heterogenous appearance, sometimes with foci of calcification [79]. In many cases a definitive diagnosis is only made after surgical excision and histological examination. The natural history of haemangiomas is unpredictable. They can regress, remain static or enlarge over time. Surgical resection is the treatment of choice for symptomatic lesions and the outcome and prognosis is usually good [88]. Follow-up is recommended to allow for the detection of tumour recurrence, although the rate of recurrence is unknown [81]. Most haemangiomas have been reported to represent small, red to purple endocardial nodules ranging from 0.2 to

Fig. 17  Low power view of a cavernous haemangioma composed of multiple dilated vascular channels (Haematoxylin and Eosin)

3.5 cm in diameter, either polypoid or sessile, and without evidence of infiltration [80]. Intramural haemangiomas tend to be poorly circumscribed masses with a haemorrhagic or congested appearance [82]. Pericardial haemangiomas typically are solitary and well defined. They arise from the visceral pericardium, and range in size from 1 to 14 cm [84]. Histologically, haemangiomas have been subdivided into: cavernous haemangiomas composed of multiple dilated, thin-walled vascular channels (Fig.  17); capillary haemangiomas comprising small vessels resembling capillaries in a lobular or grouped arrangement, often with a myxoid background with an associated feeder vessel; and arterio-venous haemangiomas composed of malformed arteries and veins [85, 87]. Infantile haemangiomas are considered a type of capillary haemangioma with densely packed capillaries, no intervening fibrous tissue and sometimes a high mitotic rate in the proliferative stage. GLUT1 expression is characteristic in this subtype of capillary haemangioma [79]. Most of the lesions have overlapping features, and many contain variable amounts of interspersed adipose and fibrous tissue [82, 89], especially the intramuscular lesions. They parallel similar haemangiomas elsewhere in the body. Intramural lesions are also often surrounded by hypertrophied myocytes. The tumour cells stain positively for standard endothelial markers such as CD31, CD34 and factor VIII. The pericytic framework of cells surrounding the endothelial lined vascular channels stains for smooth muscle actin. The differential diagnosis includes congenital endothelial-­ lined cysts found along the lines of closure of heart valves, cardiac varices, cardiac myxoma and angiosarcoma. Congenital endothelial lined cysts lack a vascular wall and are not connected to a feeder vessel. Cardiac varices tend to

Cardiac Tumours

occur in the right atrium and represent dilated, thrombosed veins, and are considered by some a form of venous malformation/haemangioma [80]. In haemangiomas, the classic myxoma cells are absent, there is a well-developed pericytic framework of cells surrounding the endothelial structures, and cellular areas with numerous capillaries are usually present. Some low-grade angiosarcomas may be difficult to differentiate from haemangiomas, but lack of cellular pleomorphism, mitotic activity, necrosis and cellularity may aid in distinguishing haemangioma from angiosarcoma [85].

239

pathogenesis of extracardiac IMTs, their role has not been substantiated in cardiac IMTs [94, 96, 97]. Although cardiac and extra-cardiac IMTs share many features, there are differences in that cardiac IMTs usually behave in a benign fashion, lacking metastatic potential, whereas atypical and aggressive cases of extra-cardiac IMTs have been reported [98]. The clinical manifestations of cardiac IMT are primarily related to the size and location of the tumour, and include heart failure, syncope, arrhythmias, inflow/outflow obstruction, myocardial infarction and even sudden death [96] when there is involvement or prolapse into the coronary arteries. Peripheral embolization of IMT has also been described 13 Granular Cell Tumour [95]. Up to one third of patients with IMT (irrespective of origin/location) can present with a constitutional syndrome and Schwannoma of fever, malaise and weight loss, in addition to microcytic Cardiac granular cell tumours are thought to be of neuroec- hypochromic anaemia, thrombocytosis, polyclonal hypertodermal origin, usually arise epicardially in the region of the gammaglobulinaemia and elevated serum acute phase reacsinus node, form firm, well-circumscribed masses and histo- tants, thought to stem from IMT-mediated cytokine logically are composed of collections of cells with abundant expression (IL-6) [94, 99]. These symptoms often resolve eosinophilic, finely granular cytoplasm, which stain posi- once the lesion is excised. IMTs may be identified by echotively for S100 protein, CD68, inhibin-alpha and neuron-­ cardiography and appear bright on ultrasound. specific enolase (NSE) [90]. Treatment options depend on the location of the tumour Cardiac schwannomas occur in adults, in any chamber of and associated symptoms, although complete surgical resecthe heart, although an origin in the right atrium near the sep- tion is the favoured option, with a good prognosis. Tumour tum is most frequent [91]. Pericardial schwannomas may recurrence is rare (8% overall), but may be more likely if arise on rare occasions from cardiac branches originating resection is incomplete [96]. Steroids, chemotherapy and from the vagus nerve or cardiac plexus [92]. They arise from radiotherapy have been used, mainly in patients with unreSchwann cells of nerve fibres and share the same variety of sectable or incompletely resected tumours [96]. gross and histological appearances as their soft tissue Up to 60% of extra-cardiac IMTs show a clonal cytogecounterparts. netic aberration that activates the ALK receptor kinase gene (chromosome 2p23), leading to ALK protein o­ verexpression, which is detectable immunohistochemically with cytoplas14 Inflammatory Myofibroblastic mic staining [98]. Several ALK fusion products involving tropomyosin genes and the clathrin heavy-chain gene have Tumour been reported [97]. In extra-cardiac IMTs ALK reactivity has Inflammatory myofibroblastic tumour (IMT) is a low grade been associated with a more favourable prognosis. However, neoplasm of mesenchymal cells with smooth muscle and ALK expression appears different in cardiac and extra-­ myofibroblastic differentiation, and an associated chronic cardiac IMTs, with little or no ALK expression and no eviinflammatory infiltrate [93]. IMTs can be located in essen- dence of a prognostic role in cardiac cases. No other specific tially any organ system, but are primarily found in soft tis- cytogenetic abnormality, to correlate with biological behavsues and viscera of children and young adults [94]. A cardiac iour or clinical presentation, has been described. origin is exceedingly rare and the majority of these lesions Either side of the heart can be affected [95, 98, 100] and have been described in children and young adults, although although some reports have suggested a right-sided predomithere have been infrequent reports in older adolescents [95]. nance, these may have included a heterogenous group of There is no apparent gender predominance. The aetiology tumours [95]. IMTs tend to arise from endocardial surfaces, and pathogenesis of cardiac IMTs are still unclear. including the valves [100], to form polypoid, broad-based It has been suggested that IMTs arise secondarily to lesions, generally with a smooth surface, which may show immune or inflammatory dysregulation, or represent an exag- overlying fibrin [95]. Non-luminal masses are very rare [98]. gerated response to tissue injury [96]. Post-viral cell cycle The lesions range from approximately 1.5–6 cm in size [95]. dysregulation has also been suggested as a possible aetiology. Grossly, IMTs exhibit textural variation, including Although rare reports of Epstein-Barr virus (EBV), human fleshy and fibrous portions [100]. The cut surface usually herpes virus 8 (HHV8), and IgG4 have been postulated in the has a tan-­white, sometimes whorled appearance and may

240

D. M. Rassl

Fig. 18  Inflammatory myofibroblastic tumour with relatively bland spindle cells set within a fibrous stroma, alongside other areas with more dense cellularity (Haematoxylin & Eosin)

Fig. 19  Inflammatory myofibroblastic tumour showing a more cellular background with a prominent inflammatory cell infiltrate (Haematoxylin & Eosin)

show myxoid areas, focal punctuate haemorrhage and softer, yellow foci, the latter being sometimes related to areas of coagulative necrosis or infarction, possibly due to torsion [98]. The histological appearance of cardiac IMT is heterogenous with a variable degree of cellularity and collagen deposition, including myxoid areas, areas with densely collagenized stroma and focal areas of compact cellularity (Fig. 18) [95, 100]. The tumours contain spindle cells showing fibroblastic and myofibroblastic differentiation, arranged in fascicles or exhibiting a storiform architecture. The spindle cells possess vesicular, oval nuclei, fine chromatin and mostly inconspicuous nucleoli, although occasional prominent nucleoli may be seen. Cytological atypia is low grade and mitotic activity infrequent (70% stenosis) with, or sometimes without, myocardial ischaemic changes • Defined cardiomyopathy capable of dysrhythmic death (e.g. HCM, ARVC, DCM) • Muscular dystrophy (known history) with histological myocardial alterations • Myocarditis (only with myocyte degeneration) • Known (or later confirmed) channelopathy • Significant (i.e.: widespread) amyloid deposition • Widespread granulomatous inflammatory foci, in keeping with sarcoid Possibly has caused the sudden death (providing there is no significant other pathology): • Myocardial fibrosis, in keeping with prior infarction but without acute changes (note: this would require significant coronary atheroma to be present) • Floppy mitral valve (supported by ballooning, atrial dilatation, jet lesions etc.) • Coronary artery anomaly (ideally with some myocardial ischaemic fibrosis) • Moderate coronary atheroma (40–60% stenosis), but would require need cardiac hypertrophy and/or old infarction scarring in the heart to substantiate Having sorted this broad ‘batting order’ of probability in one’s mind, the converse question needs to be addressed, “what is inappropriate to list as a cardiac cause of sudden death?” This can conceivably cover virtually everything else, but problems that are regularly debated at autopsy include the following: • Minor histological patchy fibrosis of the myocardium (i.e. without appreciable coronary disease or other antecedent ECG or clinical history, etc.)

S. K. Suvarna

• Minor cardiac hypertrophy alone (no more than 10% greater than the expected cardiac mass) • Coronary calcification without significant narrowing (